Journal of Muscle Research and Cell Motility 13, 608-618 (1992)

Isolation, purification and partial characterization of tropomyosin and troponin subunits from the lobster tail muscle ANDREA

MIEGEL, ~ TOMOYOSHI

KOBAYASHI

2 and Y U I C H I R O

MAI~DA

TM

1European Molecular Biology Laboratory at DESY, NotkestraJ~e 85, D-W2000 Hamburg 52, Germany 2Department of Biological Chemistry, University of Maryland School of Medicine, 660 West Redwood Street, Baltimore, MD 2120L USA Received 27 June I991; revised and accepted 20 February 1992

Summary In a search for an invertebrate muscle from which the muscle regulatory proteins could be obtained in a great quantity and at high homogeneity, the regulatory proteins, tropomyosin (Tm) and three subunits of troponin (Tn), have been isolated from the lobster tail muscle, purified and partially characterized. The calcium-sensitive ATPase of lobster myofibril was restored when purified lobster Tm and lobster Tn were added to actin. Quantitative SDS-polyacrylamide gel electrophoresis showed that the lobster muscle contains actin, Tm, Tn with a molar ratio 7:1:1 and that lobster Tn consists of three subunits, one of each I, C and T. Each subunit was identified according to its effect on the acto-S1 ATPase rate. The isomer composition in each fraction of purified Tn subunit and in Tm are different from the rabbit skeletal muscle proteins; Tm consists of a single species of polypeptide of M r 38 000; the TnT fraction appears to be homogeneous with M r 43 000; the TnI fraction contains five isomers, all showing similar isoelectric pH, differing in M r in the range from 28 000 to 31 000; two TnC fractions contain three isomers in total with a range of M r from I8 500 to 19 000. Further study of the lobster Tm elucidated that digestion by carboxypeptidase A gave rise to a homogeneous preparation of truncated and non-polymerizable Tm which is devoid of i1 residues at the C-terminus of the molecule. The C-terminal amino acid sequence of 11 residues is homologous to the thoracic isomer generated from Drosophila me]anogaster Tm-I gene. The present study indicated that, despite heterogeneities owing to the occurrence of isomers, the lobster regulatory proteins serve as an invertebrate source of the proteins for structural and biophysical studies, alternative to vertebrate counterparts.

Introduction Vertebrate skeletal muscle contraction is regulated by binding of calcium ions to the regulatory protein troponin (Tn) which together with tropomyosin (Tm) is located on the actin filaments (see reviews; Ebashi & Endo, 1968; Ohtsuki et al., 1986; Leavis & Gergely, 1984). The regulatory proteins and the mode of regulation are fairly well understood in the vertebrate skeletal muscle, although detailed molecular mechanism of the calcium regulation remains unknown. On the other hand, in invertebrate muscles the T m - T n system is less known. In the present report, we have studied the regulatory proteins of lobster tail muscle as an example of the T m - T n system of invertebrate muscles. *To whom correspondence should be addressed. 0142-4319 9

1992 Chapman & Hall

This study had two specific questions in mind. First, it was of interest to know if the T m - T n system of the invertebrate works in the same way as the mammalian counterpart, although the invertebrate is evolutionary distantly related to the mammal. The X-ray diffraction pattern from the lobster muscle did not show a significant intensity increase in the second actin layer-line (D. Popp & Y. Ma6da, unpublished data), although the intensity increase is believed to shift the Tm strands on the actin filaments and this shift is believed to be associated with the regulatory mechanism (Haselgrove, 1973; Huxley, I973; Parry & Squire, 1973; Vibert et al., 1972). In order to know the reasons for the absence of the intensity increase in the lobster muscle, we decided to see if there are any differences in the mode of regulation of the T m - T n system of this muscle compared with its counterpart in the vertebrate skeletal muscle.

Lobster muscle regulatory proteins Second, for biophysical and structural studies, especially for crystallographic studies of the regulatory proteins, an invertebrate muscle could be a useful source of regulatory proteins if individual proteins can be obtained in a great quantity (20-100 mg in a single batch) and at high homogeneity ( > 95% of a single isomer). In general, a protein from invertebrate animals would serve as a useful alternative to the mammalian counterpart, provided that the invertebrate proteins, although chemically differentiated, function in the same way as the mammalian regulatory system. In particular, it is of great interest if the isomer composition of individual proteins is much simpler in an invertebrate muscle. As is well known, rabbit skeletal muscle Tm consists of 0r and fl-isomers (Bronson & Schachat, 1982) which are not easily separated from each other to a high purity. TnT from the rabbit skeletal muscle also consists of a number of isomers, resulting from the alternative splicing of mRNA (Breitbart & Nadal-Ginard, 1986; Fujita et al., 1991), and the isomers are not separated from each other (Briggs eta]., 1984). Moreover, in the case of rabbit skeletal muscle a-Tm, carboxypeptidase digestion does not give rise to a homogeneous preparation of a truncated Tm (Walsh et al., 1984). Tm which is devoid of its C-terminal 11 residues is non-polymerizable and has been used for the study of the head-to-tail interaction between Tm molecules (Tawada et al., 1975; Johnson & Smillie, 1977; Mak & Smillie, 1981; Walsh et a[., 1984; Pan et al., 1989). These heterogeneities of rabbit preparations are among the technical obstacles for further studies of the T m - T n system. The lobster tail muscle is superior to any other invertebrate muscle; live animals are commercially available throughout the year from which flesh meat of 300 g is easily obtained. Previously published works on t h e T m - T n system of this muscle (Lehman et aI., 1973, 1976, Lehman & Szent-Gy6rgyi, 1975; Regenstein & Szent-Gy6rgyi, 1975; Mykles, 1985) have left some of the above questions unanswered. The regulatory proteins from crayfish muscle, another crustacean, have also been studied in some detail (Benzonana et al., 1974; Wnuk et al., 1984; Shinoda et al., 1988; Kobayashi et al., 1989a, b; Wnuk, 1989; Miyazaki eta]., 1990). The muscle used is the abdominal flexor of the American lobster Homarus americanus, the major tail muscle of which dominates in cross-sections of the tail. The mechanical properties of this muscle have not been studied, probably because individual muscle fibres are long and bundles of fibres are highly twisted making physiological studies extremely difficult. From its function, serving as the major muscle used for quick bending of the tail for escaping in alarm, the muscle is highly likely to be a fast muscle, as is the deep lateral abdominal extensor muscle (Jahromi & Atwood, 1969). Preliminary results of part of this work have been reported (Miegel & Ma6da, I990).

609 Materials and methods

Chemicals Chemicals of analytical grades were purchased from Fluka, Merck, Serva or Sigma. Stock solution of 10 M LiC1 and 8 M urea were cleaned by passing through Chelex 100 resin (BioRad) or through a AG50I-X8 (BioRad) mixed bed resin column, respectively. The urea solution was then used within 1-2 days.

Muscle Live American lobsters Homarus americanus were purchased from a local supplier. The bulky part of the tail muscle, the abdominal flexor, was used, the superficial and deep abdominal extensors being excluded.

Myofibrils With a meat grinder the muscle was minced and then homogenized with a shear force homogenizer in washing solution (50 mM KC1, 5 mM Tris HC1, pH 8.0, 0.1 mM 1,4-dithio-DLthreitol (DTT) with 1% Triton X-100. The washed muscle was collected by centrifugation. The washing was repeated three times followed by two more times of washing without Triton X-IO0.

Isolation of lobster T m - T n complex The methods of Ojima and Nishita (1986a, b), based on the method of Ebashi and colleagues (1971), were employed with some modifications. The washed myofibril was subjected to isoelectric precipitation at pH 5.5 in 0.4 M LiC1, 25 mM Tris HC1, 0.1 mM CaC12, 0.1 mM DTT. After adjusting the pH of the supematant to 8.0, ammonium sulphate was added and the fraction 40-60% saturation was collected by centrifugation. The fraction contained the complex of Tm-Tn.

Isolation of lobster Tn The Tm-Tn complex, dispersed with 1.0 M LiC1, 25 mM Tris HC1, 0.1 mM CaC12, 0.1 mM DTT, pH 7.5, was subjected to isoelectric precipitation at pH 4.6-4.7. The precipitate contained mainly Tm which was heavily contaminated with Tn. To obtain more Tn, occasionally the isoelectric precipitation at pH 4.6-4.7 was repeated again. To the supernatant after the isoelectric precipitation, pH being readjusted to 8.0, ammonium sulphate was added, the fraction 45-65% being collected by centrifugation. The precipitate (containing Tn) was dialysed against 10 mM Tris HC1, pH 8.0, 0.1 mM DTT, 1 mM NaN 3 and stored frozen.

Isolation and purification of lobster Tn subunits The Tn preparation in 6 M urea, 20 mM Tris HC1, pH 7.8, 0.3 mM DTT, I m M EDTA was loaded on a S-sepharose fast flow (Pharmacia-LKB) column (Fig. 1). On applying a gradient Of 0--0.6 M NaC1, TnT was eluted at approximately 0.25 M and TnI at 0.4 M. The unadsorbed fractions (containing TnC, contaminated Tm and others) were pooled and directly applied on a Q-sepharose fast flow column pre-equilibrated with the same 6 M urea solution (Fig. 2). On applying a gradient of 0-0.4 M NaC1, the contaminated Tm (at 0.2 M NaC1), copurified myosin light chain-1 (at 0.26 M NaC1), TnC-1 (at 0.3 M NaC1) and TnC-2 (at 0.34 M NaC1) were eluted in this order. The TnT fraction was dialysed against 0.6 M KCI, 10 mM Tris

610

MIEGEL, K O B A Y A S H I and MA12DA precipitation at pH 4.5 in the presence of 1 M LiC1. For further purification, the lobster Tm fraction was heated to denature Tn and other contaminating proteins. The crude Tm after the isoelectric precipitation was suspended with the solution containing 0.4 M LiC1, followed by a fivefold dilution with water to adjust the LiC1 concentration below 0.I M. After EDTA was added to I mM, the preparation was heated to 85 ~ C for 3 rain and denatured proteins were removed by centrifugation.

Carboxypeptidase A cleavage of Tm

B

0.6

-30 i oo C',I

II

s;'/

2.5

2.0 1.5

C

1.0

The method of Mak and Smillie (i98I) was employed with some modifications. MgC12 and Zn(CH3COO) 2 were added to the purified Tm preparation (which contained I mM EDTA) to

0.5

SS

.

see

0.4

"L

T

,4'*s

0.3

, ~ ~

0.2 i

0.5

0.1

0 0

A 20 40 60 fraction numbers

0 80

Fig. 1. Column chromatography of the lobster Tn complex into three subunits in the presence of 6 M urea. (A) SDS-PAGE patterns of the fractions. (B) Elution profile from a column of cation exchanger S-Sepbarose fast flow; 300 mg of Tn complex in 20 mM Tris HCI, pH 7.8, 0.3 mM DTT, I mM EDTA with 6 M urea was loaded on a column 2.5 cm x 20 cm long which is pre-equilibrated with the same solution. Total 800 ml of a linear gradient 0-0.6 M NaC1 was applied and fractions of 10 ml each were collected at a flow rate of 3.5 ml rain -~ at 4 ~ C. In A, lane M shows molecular weight markers; lane S, the Tn complex applied; other lanes are the fractions numbered in B. In B, the fractions indicated by the bars were pooled for further fractionations. C denotes the fraction consisting mainly of TnC and Tm; T, TnT; I, TnI. HC1, pH 7.8, 0.1 mM DTT, i mM NaN 3, and stored frozen. The TnI, TnC-1 and TnC-2 fractions were dialysed against the same solution without KC1, and were stored frozen. The yields from I00 g minced muscle were (on average) 250 mg crude Tin, 100 mg crude Tn, 30 mg TnT, 15 mg TnI, 6 mg TnC-1, 3 mg TnC-2.

Fractionation of lobster TnI isomers TnI was further fractionated by column chromatography on a Q-sepharose fast flow column (Fig. 3). On applying a NaC1 gradient of 0-0.3 M, TnI-2, TnI-4, TnI-1 and TnI-3 were eluted in this order.

Purification of lobster Tm The Tm fraction contained a substantial amount of Tn, only a part of which can be removed by repeating the isoelectric

B TI~

"•

C-1 C-2

0.5 0.4-

0.4

soo~ ~

0.3

0.3"~

LC-2

0.2-

',~

, ~,~-~'

0.2

< 0.1.

t0

""

40

I

6o

'

~

80

'

,0'0

I

120

0

'

14'0

fraction numbers Fig. 2. Column chromatography of the lobster TNC fraction into subunits. (A) SDS-PAGE patterns of the fractions. (B) Elution profile from a column of anion exchanger Q-Sepharose fast flow. The unadsorbed fraction from the S-sepharose fast flow column (Fig. 1) was loaded on a column 1.5 cm x 20 cm long which was pre-equilibrated with the same solution as in Fig. 1. Total 400 ml of a linear gradient 0-0.4 M NaC1 was applied and fractions 3.5 ml each were collected at a flow rate of 1.2 ml min 1 at 4 ~ C. In A, lane M shows molecular weight markers; other lanes are the fractions numbered in B. In B, the fractions are marked by major constituents; LC-I and LC-2 denote myosin light chain I and 2, respectively.

611

Lobster muscle r e g u l a t o r y proteins

At preset times a 20 gl aliquot was removed and evaporated. The released amino acids were analysed by a Waters PICO-Tag amino acid analysis system which employs the pre-column derivatization of the amino acids with phenylisothiocyanate (PITC), followed by analysis of the derivatives by HPLC.

Analysis of the amino acid composition After hydrolysis of a protein for 24 h in 6 M HC1 containing I% phenol vapour at 106 ~ C, the amino acid composition was analysed by the Waters system. To determine cysteine content, the protein dissolved in 6 M guanidine-HC1, 0.25 M Tris HC1, l m M EDTA, pH 8.5, was reduced with 2-mercaptoethanol, followed by modification by 4-vinylpyridine to measure the amount of pyridylethyl cysteine.

B 0.8

1-2 r

1-4 I-1

ATPase assay

1-3

0.6.

= o

11 ca

0.4

.,~

0.2

0.3

o

g

0.2

+

Z 0.1

I !

0

0 20

30

40

50

60

70

80

fraction numbers Fig. 3. Column chromatography of the lobster TnI fraction into subunits. (A) SDS-PAGE patterns of the fractions. (B) Elution profile from a column of anion exchanger Q-Sepharose fast flow. The TnI fraction (182 mg) from the S-sepharose fast flow column (Fig. 1) was loaded on a column 2.5 cm x 20 cm long which was pre-equilibrated with the same solution as in Fig. I. Total 800 ml of a linear gradient 0--0.3 M NaC1 was applied and fractions 9 ml each were collected at a flow rate of 3 ml min at 4 ~ C. In A, lane M shows molecular weight markers; S is the material loaded; other lanes are the fractions numbered in B. In B, the fractions are marked by major constituents. a final concentration of 2 mM and 0.1 mM, respectively. Carboxypeptidase A (Sigma, C9762) was added to 1/40 (w/w) of Tin, as well as, 0.1 mM phenylmethanelsulphonyl fluoride (PMSF) to a final concentration of 0.1raM, and the reaction mixture was incubated at 37 ~ C for 90 min. The reaction was terminated by heating the mixture to 85 ~ C for 3 min. If necessary, a cycle of enzyme reaction followed by heating was repeated once again. The yield of the purified and truncated Tm was 80-I00 mg from 100 g minced muscle.

Analysis of amino acid sequence at the C-terminus The C-terminal peptide sequence was determined by following the time course of amino acid released from lobster Tm by carboxypeptidase A digestion. The enzyme to a final concentration of 0.04 mg ml 1, as well as PMSF to 0.1raM, were added to lobster Tm at 2 mg ml ; in 0.1 M 4-ethylmorpholine, pH 8.0, 0.1 mM NaN 3 and the mixture was incubated at 37~

Solution conditions employed: 60mM KC1, 4 mM MgCl 2, 13raM 4-(2-hydroxyethyl)-l-piperazineethanesulphonic acid (HEPES), 0.5 mM DTT, 1ram NaN 3, 1ram ATP, ethylene glycol-O,O'-bis(2-aminoethyl)-N,N,N',N'-tetraacetic acid (EGTA)-CaC12 buffer with 1ram EGTA for appropriate free Ca 2+ concentrations, pH 7.0-7.2, at 17.5 ~ C. For myofibrils, myosin with actin, the colorimetric measurements of ATP hydrolysis after White (1982) was employed with minor modifications. The 1 [,tM of inorganic phosphate in the total reaction mixture of 0.375 ml gave rise to Ass0 1.50. For $1 with actin-Tm-Tn, the reaction was followed by measuring the absorbance of NADH (Imamura et al., 1966). The reaction mixture of i ml contained pyruvate kinase 0.05 mg ml 1, lactate dehydrogenase 0.063 mg ml 1, phosphoenolpyruvate 0.2 raM, NADH 0.04 mM and proteins. The molar extinction coefficient of NADH r = 6.2 x 1 0 3 per molar per cm at 2 = 340 nm was used. Actin and $1(A2) were prepared from rabbit skeletal muscle.

Gel electrophoresis Polyacrylamide gel electrophoresis in the presence of 0.1% sodium dodecyl sulphate (SDS) was carried out by employing the method of Laemmli (1970) with some minor modifications. The slab gels were either 1.0 mm thick, 80 x 80 mm or 1.5 mm thick, 120 x 140 ram. Two dimensional gel electrophoresis was carried out according to either the isoelectric focusing (IEF) (O'Farrell, 1975) or the non-equilibrium pH gradient electrophoresis (NEPHGE) (O'Farrell eta]., 1977) method as the first dimension.

Quantitative SDS gel electrophoresis The amount of dye bound to proteins within the gels was determined either by densitophotometry of the gels (Potter, 1974) or by extraction of the dye with pyridine/water mixture followed by measuring absorbance of the extract (Murakami & Uchida, 1984). For densitophotometry, slab gels were stained with Fast Green (I.C.42053, Sigma) and scanned with either a Joyce-Loebl microdensitometer type IIIc or Beckmann CDS-200 gel scanner at • = 570 nm. For extraction, from slab gels which were stained with Coomassie Brilliant Blue R-250 (C.I. 42660, Sigma), the bands were cut out and dyes were extracted in 25% pyridine in water. Absorbance of the extract was measured at 605 nm. For each experiment, purified proteins of known weight were subjected to the same procedure as the complex of interest to obtain the colour factor (absorbance per unit

612

MIEGEL, KOBAYASHI and MAIqDA Results

Tm and Tn reconstituted the calcium sensitivity of the actomyosin Lobster Tm and lobster Tn were isolated and purified separately (Figs 1 and 4A). The calcium sensitivity of the ATPase activity of lobster myofibrils (Fig. 5A) was reconstituted when both lobster Tm and Tn were added to rabbit skeletal muscle F-actin (Fig. 5B). The half maximum of the ATPase activity of the lobster myofibril and of the reconstituted actomyosin were at pCa (=-log[Ca2+]) 5.9 and 5.6, respectively, which were close to each other. The calcium sensitivity was saturated at the molar ratio actin to Tm to Tn of approximately 7:1:1 (Fig. 6A and B). The same ratio was obtained by quantitative analysis of patterns of polyacrylamide gel electrophoresis in the presence of SDS (SDS-PAGE) (Table 1). Adding lobster Tm to actin decreased the ATPase rate when the molar ratio myosin subfragment-1 ($1) to actin is low (1:40 to 1:20) as shown in Fig. 6A. With the ratio as high as 3:1, the ATPase rate increased slightly as more Tm was added (data not shown), as shown for the rabbit skeletal muscle proteins (Lehrer & Morris, 1982). Adding lobster Tn to the actin-Tm complex suppresses the actomyosin ATPase, which is recovered on raising free [Ca2"+](Fig. 6A and B). The present results indicate that the lobster muscle is regulated by Tn and Tm in the same way as in rabbit skeletal muscle. Isomer composition of lobster Tn subunits Three Tn subunits were isolated by ion-exchange column chromatography in the presence of 6 M urea. TnI and TnC are mixtures of isomers, while the TnT preparation appears to be homogeneous (but see Discussion). TnCs were separated into two fractions by an anion exchange column chromatography (Fig. 2); fraction-1 (TnC-1), the more basic fraction, gave rise to a single band, while fraction-2 (TnC-2), the more acidic fraction, contains two isomers TnC-2a and TnC-2b. Further isolation of 2a and 2b were only possible on a reverse phase HPLC column (Garone et al., 1991). All the isomers have a range of apparent Mr from 18 500 to 19000 and Fig. 4. SDS-polyacrylamide gel electrophoresis of lobster

Fig. 4.

weight of protein) of each component. To check and ensure the linearity of the system, at least four data points were taken both for purifed proteins and for the mixture of interest within each data set. The system was linear up to 6 t,tg protein, and 4 ~g of lobster Tm was used to correct for possible differences between gels.

muscle proteins, B and C, two dimensional gel electrophoresis of lobster Tn-Tm complex. (A) Lanes are: a-c, myofibrils; d, Tm-Tn complex; e, Tin; f, Tn; g, TnT; h, This; i, TnC-1; j, TnC-2. On the right hand of each gel, the bars indicate the positions of the proteins. Ac stands for actin. Electrophoresis was carried out in 12.5% polyacrylamide slab gel containing 0.1% SDS (0.5 mm thick, 95 x 95 mm). In B, the first dimension was non-equilibrium pH gradient electrophoresis followed by the SDS-polyacrylamide gel electrophoresis as the second dimension (O'Farrell et al., 1977). TnCs did not sufficiently enter the gel owing to extreme acidity. In C, the first dimension was isoelectric focusing followed by the SDS-polyacrylamide gel electrophoresis as the second dimension (O'Farrell et al., 1975). Tnls did not enter the gel because these are too basic. In B and C, the preparation contained small amounts of LC-1.

Lobster muscle regulatory proteins

0.5

613 on the SDS-gels of fleshly-prepared lobster myofibrils and of pieces of live muscle, and the identical pattern has also been reported from other lobster muscles (Mykles, 1985). The TnI fraction was further separated on a Q-sepharose column (Fig. 3) into four fractions each of which contains essentially a single isomer of TnI (at a purity of about 90%), I-2, I-4, I-1 or I-3, 1-5 not being recovered. The TnT fraction collected from a major peak from the S-sepharose fast flow chromatography showed a single band at an apparent Mr 43 000 on SDS-PAGE, although on two-dimensional gels (especially when IEF was employed in the first separation, Fig. 4C) the main spot at isoelectric pH 6.5 was always associated with four to five minor spots in the more basic region. Peptide mapping of proteins extracted from individual spots, being cleaved with either lysyl endopeptidase (Wako Pure Chemical, Osaka) or 0r (Boehringer Mannheim), did not show any significant difference between these polypeptides, as far as longer peptides seen on ordinary SDS-PAGE are concerned (unpublished data). Therefore we conclude tentatively that the lobster tail muscle TnT is homogeneous. The multiple spots on the two-dimensional gel may

A

0.4

o

~o ea~

0.3

o,,,~

~D O

"~

0.2 0.1

.....

J"l

.......

!

.......

'1

.......

I ' ' .......

!

.......

7

6 5

~

:::I.

---- "

4

1 9

9

....

"'1

8

9

,

.....

I

7

9

, .....

"1

6

. . . . . . .

I

9

5

9 ','"'1

. . . .

4

.,=

10

B

3

- log [Ca 2+ ] (pCa)

C

6 4

Fig. 5. Reconstitution of Caz+ sensitivity of actomyosin ATPase with lobster Tm and lobster Tn. (A) Lobster myofibril 0.343 mg ml -I. The rate of the Pi release was measured by the colorimetry (White, 1982). (B) Rabbit skeletal muscle $1(A2) 0.034rag ml -~ (0.3 ~M) with rabbit skeletal muscle actin 0.25 mg ml-~ (6 [.tM) plus lobster Tm 0.06 mg ml -~ (1.2 [.tM) and lobster Tn 0.08 mg ml ~ (1.25 ~M). The ATPase rate of $1 alone has been subtracted from each measurement. The ATPase rate was followed by measuring absorbance of NADH (Imamuraet al., 1966). The solution condition employed is given under Materials and methods. The free Ca2+ concentration was calculated using the apparent Ca2+ binding constant of EGTA pK~ = 6.45 (Allen et aI., 1976).

isoelectric pH 4.0-4.5. The weight ratio of the three isomers differed from batch to batch. The one-dimensional (Fig. 4A) and the twodimensional gels (Fig. 4B and C) indicated that the TnI fraction contains five isomers, all showing similar isoelectric pH at 9.0 or higher, differing in M~ in the range 28 000 to 31 000. The weight ratio of the five isomers TnI-1 to TnI-2 to TnI-3 to TnI-4 to TnI-5 (descending order of apparent molecular weight) was 25:2:10:62:I, which is fairly constant from one batch to another. Each band probably corresponds to a distinct TnI isomer, not a proteolytic product of another; these bands were seen with the same proportion

o -~

2 0

9 ' Tn/Tm

1.0 7 xTm/Ac

1"0.

'

"~ 6 ~

F

2

0

0.0

9

0.5

7 x Tm/ Ac

1.0

0.0

0.5 TnI / Tm

1.0

0.0

0.5

1.0

TuC-2 /Tm

1.5

0.0

'

. 0.5

9

. 1.0

'

1.5

TnT / Tm

Fig. 6. Effects on the Mg-ATPase activity of reconstituted actomyosin of A, lobster Tin; B, lobster Tn; and C-F, lobster Tn subunits. (A) Lobster Tm was added to rabbit actin with the molar ratio indicated on the abscissa. (B) Lobster Tn was added to a 4:1 mixture of rabbit actin plus lobster Tm with the molar ratio indicated on the abscissa. (C) Lobster Tm was added to actin with variable molar ratios. (D) Various amounts of lobster TnI were added to actin-Tm complex (with a molar ratio 6.6:1). (E) Various amounts of lobster TnC-2 were added to actin-Tm-TnI complex (6.6:1:1). (F) Various amounts of lobster TnT were added to actin-Tm-TnI-TnC-2 complex (6.6:1: I:1). The protein preparations used in A and B were of a different batch from those used in C-F. In each panel, the ATPase rate of $1 alone has been subtracted. The ATPase rate was measured in the presence of i mM EGTA (C)) or i mM CaC12 (0). Otherwise the measuring condition was the same as for Fig. 5B. Each reaction mixture contained 0.017 mg ml -I (0.I5 laM) rabbit $1 (A2), 0.25 mg ml ~ (6 I.tM) rabbit actin.

MIEGEL, KOBAYASHI and MA/~DA

614 Table I. Quantitative SDS-polyacrylamide gel electrophoresis of lobster Tm-Tn complex

A, Experiment 1 Total absorbance* Colouring factor:l: Weight ratio Molecular mass (Mr) Mol per tool Tm B, Experiment 2 Scan areaw Colouring factor:l: Weight ratio Molecular mass (M r) Mol per tool Tm

Tm

Tn T

TnI

TnC

1.0 1.0 1.0 76000 1.0

0.720 0.83I 0.866 43 000 1.53

0.408 1.18I 0.345 30000 0.88

0.095 0.434 0.219 18 500 0.90

1.0 1.0 1.0 76000 1.0

1.027 1.60 0.642 43 000 1.13

0.686 1.71 0.401 30000 1.02

O.137 0.81 0.169 18 500 0.7

*Absorbance at 605 nm of the pyridine extract from the bands cut out from the slab gels stained with Coomassie Brilliant Blue R-250. :l:Absorbance per unit weight of purified constituent protein was expressed as a ratio with respect to Tin. w areas determined by the densitometric scan were expressed as a ratio with respect to Tin. The slab gels after being stained with Fast Green were scanned with a Beckmann CDS-200 gel scanner at ;~ = 570. be explained in terms of remaining interactions between TnT and the gel. The tailing of the major spot towards the basic region cannot be from phosphorylation which would cause a shift of spots to a more acidic region.

Identification of three Tn subunits Each subunit has been identified according to its effect on the acto-S1 ATPase rate (Fig. 6): (I) TnI alone inhibited the ATPase (Fig. 6D), the inhibition was removed on adding TnC (Fig. 6E), and only after adding TnT was the ATPase Ca z+ sensitive (Fig. 6F). In this series of experiments, the TnI used was a mixture of isomers. As the effect of TnI was saturated at a molar ratio of 1:1 to Tm, and as all the isomers have similar isoelectric pH, it is likely that each is a TnI isomer. Both fractions TnC-1 and TnC-2 (the mixture of TnC-2a and TnC-2b) showed the identical effect on the acto-S1 ATPase. The stoichiometry of Tn subunits was T to total I to total C = I : I : 1 (Table 1), as in the rabbit skeletal T m - T n system (Potter, I974).

Lobster Tm is homogeneous Both one-dimensional (Fig. 4A) and two-dimensional gels (Fig. 4B and C) showed that the lobster Tm preparation consists of a single species of polypeptide with apparent Mr of 38 000 and apparent isoelectric pH 5.5 (measured on a two-dimensional gel with NEPHGE as the first dimension). This is in contrast to the rabbit skeletal muscle which contains two isomers ~ and ft.

Carboxypeptidase A digestion results in homogeneous, nonpolymerizable Tm Like rabbit skeletal ~-Tm, 11 residues were removed from the C-terminus of lobster Tm by carboxypeptidase A,

B

21

s s

~

1-[ YL " /

L~' ~"

::v

E

~: ~ , 1 " ~ t " ~ F ~ w

~F "OW

/ //A/,r

.

0

30

,

.

.

60

,

.

90

F T

.

120

time (rain) Fig. 7. Digestion of the lobster Tm by carboxypeptidase A. (A) SDS-PAGE of the lobster Tin; lane a, before digestion; lane b, mixture of the samples loaded on lane a and c; and, lane c after digestion. For each lane, 10 I.tg of each protein was loaded. (B) Time course of amino acids released from the lobster Tm by carboxypeptidase A. Each data point is marked by a letter which represents the one-letter code of the amino acid. Inset shows the C-terminal sequence which is consistent with the time course and is very similar to that of Drosophila Tm-I gene product (the thoracic isomer) indicated underneath.

615

Lobster muscle regulatory proteins Table 2. Amino acid compositions of tropomyosins

Amino acids

Lobster tail muscle Tin*

Thoracic isomer Drosophila Tm-Iq:

Rabbit skeletal muscle o~-Tmw

Asp (D) Glu (E) Ser (S) Gly (G) His (H) Arg (R) Thr (T) Ala (A) Pro (P) Tyr (Y) Val (V) Met (M) Ile (I) Leu (L) Phe (P) Lys (K) Trp (W) Cys (C) Total

36.2 73.1 13.9 4.9 0 23.1 9.5 35.4 0 4.1 10.6 7.2 3.6 30.2 2.8 26.8 0 0 (284)

36 70 14 3 0 16 15 25 0 2 18 7 6 29 5 35 0 3 284

29 70 15 3 2 14 8 36 0 6 9 6 12 33 1 39 0 1 284

*This study.The valuespresentedwere calculatedby assumingthe total residues of 284. The values were obtained from one of six analyseswhich gave rise to almost identical results. :l:From the nucleic acid sequence (Basi & Storti, 1986). w the peptide sequence (Stone & Smillie, 1978).

resulting in non-polymerizable Tm, the truncated Tm shows low viscosity even at low salt concentrations. Unlike rabbit skeletal muscle ~-Tm, however, the digested preparation is homogeneous. The species lacking C-terminal 11 residues accounts for > 95% as assessed by SDS-PAGE (Fig. 7A).

C-terminal amino acid sequence of Tm Sequence information was obtained by following the time course of amino acid release by carboxypeptidase A (Fig. 7B). Although the time course alone did not enable us to determine the sequence unambiguously, the time course was consistent with the sequence -LDQTFSELSGY(-COOH), which would be very similar to the sequence -LDQTFAELTGY(-COOH) of the corresponding part of the thoracic isomer of Drosophila melanogaster Tm-I gene product (Basi & Storti, 1986). The two sequences would differ from each other solely in two residues which are conservatively replaced (T to S, A to S). Amino acid composition of Tm In Table 2, the amino acid composition of the lobster Tm is compared with those of rabbit skeletal muscle ~-Tm and the thoracic isomer of Drosophila Tm-I gene product. The lobster Tm is slightly more acidic than others. This protein contains no cysteine.

Discussion

The most interesting finding of the present study is that Tm from the lobster tail muscle consists of a single species and that homogeneous preparations of truncated and non-polymerizable Tm can be easily obtained in great quantity (200-300 mg per batch). This is in contrast to the rabbit skeletal muscle Tm preparation in two ways. First, the rabbit preparation consists of two isomers, ~and fl-Tm which cannot be easily isolated from each other in high purity. Although the rabbit cardiac muscle contains exclusively a-Tm, it is not practical for an ordinary laboratory to use rabbit hearts as a source of Tm in great quantity. Second, digestion of rabbit a-Tm by carboxypeptidase A results in a heterogeneous population of molecules differing in the number of C-terminal residues removed by the enzyme. Even when employing an improved protocol of enzyme reaction (Mak & Smillie, 1981), and taking precautions to dephosphorylate serine283 (Mak et al., 1978), the truncated rabbit Tm is still substantially heterogeneous (Walsh eta]., 1984), as confirmed in the present study (data not shown). The homogeneous C-terminal degradation of lobster Tm may be accounted for partly by a C-terminal amino acid sequence which differs substantially from its rabbit muscle counterpart (-LDHALNDMTSI-COOH), and partly by the absence of phosphorylated residues in the C-terminal segment. Apart from improved homogeneity, the truncated lobster Tm is hardly distinguishable from the

616 truncated rabbit 0~-Tm; both have low viscosity even at low salt concentrations and low affinity to F-actin in the absence of Tn. Native lobster Tm also shares characteristics with the rabbit skeletal muscle c~-Tm; the paracrystals formed in the presence of 50 mM MgC12 show striations 40 nm apart and the radius of gyration of the cross-section of the molecules, 0.67 nm, was obtained by small angle X-ray scattering (data not shown). These results indicated that lobster Tm molecules, like rabbit skeletal muscle ~-Tm, are about 40 nm long and consist of two identical polypeptides forming an ~-helical coiled coil. The lobster tail muscle Tm is therefore a useful alternative to rabbit skeletal muscle ~-Tm, especially when the protein, whether native or truncated, is required in a great quantity at a high purity. Thus we have recently obtained a new crystal form of truncated lobster Tm (Miegel et al., 1992). It is worth noting that the lobster tail muscle Tm apparently has a C-terminal sequence of 1I amino acids which is very similar to that of the thoracic isoform generated from the Tm-I gene of Drosophila melanogaster (Basi & Sorti, I986). It is of interest to know if the overall sequence of lobster Tin, as well as sequences of lobster Tn subunits, are similar to those of Drosophila, both insects and crustacea being arthropods. In the present study methods have also been established to isolate the Tn complex, TnT, TnI-l, TnT-2, TnI-3, TnI-4 and TnC-1. These purified Tn subunits are now routinely prepared in large quantities, 10 mg or more, although not as large an amount as 200-300 mg for Tm, from 300 g muscle processed at one time. In the present study we have presented evidence that contraction of the lobster tail muscle is regulated through the Tm-Tn system which is located on the thin filaments. The molar ratio of T to I to C within the Tn complex is, like in the rabbit skeletal muscle, 1: I: I, not 1:2 : 1 as previously estimated for the lobster muscle (Lehman et al., 1976). The mode of regulation by Tm-Tn, the role of each subunit, and the overall physicochemical properties of Tm and Tn subunits are also similar to those of the vertebrate skeletal muscle. The structural changes of the thin filaments induced by the Ca 2+ binding also appears to be the same between the lobster and the vertebrate muscle. The X-ray diffraction pattern from the capillary-orientated thin filaments which are reconstituted from rabbit actin and lobster Tm-Tn showed the intensity increase of second actin layer-line in a [Ca2+]-dependent manner (Popp & Ma6da, unpublished data). In spite of the overall similarity of the Tm-Tn system, individual lobster proteins differ from the rabbit skeletal counterparts in details of physicochemical properties and, in particular, in the isomer composition. In the rabbit skeletal muscle, Tm and TnT are heterogeneous, while TnI and TnC consist of isomers of five and three, respectively.

MIEGEL, KOBAYASHI and MAI~DA In the present study it was not possible to conclude decisively if lobster tail muscle TnT is homogeneous. Although the two-dimensional gel electrophoresis showed multi-spots, no other data were obtained which were taken as solid evidence for the occurrence of isomers. It is likely that TnT consists of a single species, or isomers which are almost identical to each other. Nearly complete protein sequence analysis has identified several points of microheterogeneity (Kobayashi, Miegel, Ma6da & Collins, unpublished data). The isomer compositions obtained in the present study are generally consistent with the previous results, if not in every detail. In the present study the protein of Mr 43 000 was conclusively identified as TnT, although Regenstein and Szent-Gy6rgyi (1975) tentatively identified as TnT a 52 000 component which was not seen in the present study as a major component. After completion of the present study, Nishita and Ojima (1990) confirmed our results by identifying a protein of Mr 42 000 from the lobster tail muscle as TnT. A TnT isomer of Mr 55 000 was reported to be found in a slow muscle (the superficial abdomenal extensor muscle) (Mykles, 1985). A protein component of apparent Mr 55 000 was often observed as a constituent of the minor peak which leads the major TnT peak on the elution profile from the S-sepharose column (Fig. IB). The minor component is likely to originate from slow muscles which accidentally contaminated our preparation. Therefore TnT in the present study corresponds to TnT-2 of Mykles (1985), although we do not know if TnT-2 consists of a single species as discussed above. The isomer compositions of TnI and TnC in the present study are also not inconsistent with previous work (Mykles, 1985) in which isomer compositions were studied in various fast and slow muscles, but not the bulk muscle used by us. Mykles (1985) distinguished five TnI isomers in the fast muscles investigated, although the ratios between the isomers were different from those reported here. If these five protein components are all distinct isomers of TnI or some are proteolytic products of others will remain unsolved until amino acid sequences are obtained. Two TnC isomers in the fast muscles and one more in a slow muscle were distinguished in the previous study, while in this study all the three isomers occurred in the (fast) muscle. The differentiated distributions of isomers are most likely to arise from divergence among fast muscles, or to slow muscle fibres which co-exist in the muscle. TnC isomers in the present study, TnC-i, 2a and 2b, may correspond to TnC-1, 2 and 3, respectively, of the previous report (Mykles, I985). The amino acid sequences of these three proteins are now known (Garone e~ al., I991), indicating that these are all distinct isomers of TnC. Sequencing of lobster TnT and TnI isomers, as well as myosin light chains, purified in the present study are in progress (J. H. Collins et al.).

Lobster muscle r e g u l a t o r y proteins

Acknowledgements W e thank D r J. H. Collins and D. M y k l e s for discussions and critical reading of the manuscript, and Dr R. S. G o o d y for the rabbit $1 preparations. This w o r k was partly s u p p o r t e d b y the Muscular D y s t r o p h y Association (USA) and the National Science Foundation (USA).

References ALLEN, D. G., BLINKS,J. R. & PRENDERGAST,F. G. (1976) Aequorin luminescence: relation of light emission to calcium concentration--a calcium-independent component. Science 195, 996-8. BASI, G. S. & STORTI, R. V. (1986) Structure and DNA sequence of the tropomyosin I gene from Drosophila melanogaster. ]. Biol. Chem. 261, 817-27. BENZONANA, G., KOHLER, L. & STEIN, E. A. (1974) Regulatory proteins of crayfish tail muscle. Biochim. Biophys. Acta 368, 247-58. BREITBART,R. E. & NADAL-GINARD,B. (1986) Complete nucleotide sequence of the fast skeletal troponin T gene, Alternative spliced exons exhibit unusual interspecies divergence. J. Mol. Biol. 188, 313-24. BRIGGS,M. M., KLEVIT,R. E. & SCHACHAT,F. H. (1984) Heterogeneity of contractile proteins. Purification and characterization of two species of troponin T from rabbit fast skeletal muscle. J. Biol. Chem. 259, 10369-75. BRONSON, D. D. & SCHACHT, F. H. (1982) Heterogeneity of contractile proteins. Differences in tropomyosin in fast, mixed, and slow skeletal muscles of the rabbit. J. Biol. Chem. 257, 3937-44. EBASHI, S. & ENDO M. (1968) Calcium ions and muscle contraction. Prog. Biophys. Mol. Biol. 18, 125-83. EBASHI,S., WAKABAYASHI,T. & EBASHI,F. (1971) Troponin and its components. J. Biochem. (Tokyo) 69, 441-5. FUJITA, S., MAEDA, K. & MAI~DA, Y. (1991) Complete coding sequences of cDNAs of four variants of rabbit skeletal muscle troponin-T. J. Muscle Res. Cell Motil. 12, 560-5. GARONE, L., THEIBERT, J. L., MIEGEL,A., MAI~DA,Y., MURPHY, C. & COLLINS, J. H. (1991) Lobster troponin C: amino acid sequences of three isomers. Arch. Biochem. Biophys. 291, 89-91. HASELGROVE, J. C. (1973) X-ray evidence for a conformational change in the actin-containing filaments of vertebrate striated muscle. Cold Spring Harbor Symp. Quant. Biol. 37, 341-52. HUXLEY, H. E. (1973) Structural changes in the actin- and myosin-containing filaments during contraction. Cold Spring Harbor Syrup. Quant. Biol. 37, 361-76. IMAMURA,K., TADA, M. & TONOMURA, Y. (1966) The pre-steady state of the myosin-adenosine triphosphate system, IV. Liberation of ADP from the myosin-ATP system and effects of modifiers on the phosphorylation of myosin. J. Biochem. (Tokyo) 59, 280-9. JAHROMI, S. S. & ATWOOD, H. L. (1969) Correlation of structure, speed of contraction, and total tension in fast and slow abdominal muscle fibres of the lobster (Homarus americanus). ]. Exp. Zool. 171, 25-38. JOHNSON, P. & SMILLIE, L. (1977) Polymerizability of rabbit skeletal tropomyosin: effects of enzyme and chemical modifications. Biochemistry 16, 2264-9.

617 KOBAYASHI,T., TAKAGI,T., KONISHI,K. & COX, J. A. (1989a) Amino acid sequences of crayfish troponin I. ]. Biol. Chem. 264, 155I-7. KOBAYASHI, T., TAKAGI, T., KONISHI, K. & WNUK, W. (1989b) Amino acid sequences of the two major isoforms of troponin C from crayfish. J. Biol. Chem. 264, 18247-59. LAEMMLI,U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227, 680-5. LEAVIS,P. C. & GERGELY,J. (1984) Thin filament proteins and thin filament-linked regulation of vertebrate muscle contraction. CRC Crit. Rev. Biochem. 16, 235-305. LEHMAN, W. & SZENT-GYORGYI,A. G. (1975) Regulation of muscular contraction: distribution of actin control and myosin control in the animal kingdom. 1.. Gen. Physiol. 66, 1-30. LEHMAN, W., KENDRICK-JONES,]. & SZENT-GYORGYI,A. G. (1973) Myosin-linked regulatory systems: comparative studies. Cold Spring Harbor Symp. Quant. Biol. 37, 319-30. LEHMAN, W., REGENSTEIN, R. M. & RANSOM, A. L. (1976) The stoichiometry of the components of arthropod thin filaments. Biochim. Biophys. Acta 434, 215-22. LEHRER,S. S. & MORRIS, E. P. (1982) Dual effects of tropomyosin and troponin-tropomyosin on actomyosin subfragment 1 ATPase. J. Biol. Chem. 287, 8073-80. MAK, A. & SMILLIE,L. B. (1981) Non-polymerizable tropomyosin: preparation, some properties and F-actin binding. Biochem. Biophys. Res. Comm. 101, 208-14. MAK, A., SMILLIE,L. B. & BA.P,~NY,M. (1978) Specific phosphorylation at serine-283 of ~-tropomyosin from frog skeletal and rabbit skeletal and cardiac muscle. Proc. Natl Acad. Sci. USA 75, 3588-92. MIEGEL,A. & MAI~DA,Y. (1990) Regulation of lobster tail muscle. Biophys. J. 57, 150a. MIEGEL,A., LEE,L., DAUTER,Z. & MAI~DA,Y. (1992) A new crystal form of tropomyosin. In Mechanism of Myofilament Sliding in Muscle Contraction (edited by SUGI,H. & POLLACK,G. H.) New York: Plenum Press (in press). MIYAZAKI,J., HOSOYA, M., ISHIMODA-TAKAGI,T. & HIRABAYASHI, T. (1990) Tissue specificity of tropomyosin from the crayfish, Cambarus clarki. ]. Biochem. (Tokyo) 108, 59-65. MURAKAMI,U. & UCHIDA, K. (1984) Two-dimenmsional electrophoresis of troponin complex with non-equilibrium pH gradient-sodium dodecyl sulphate polyacrylamide slab gel. J. Biochem. (Tokyo) 95, 1577-84. MYKLES, D. (1985) Heterogeneity of myofibrillar proteins in lobster fast and slow muscles: variants of troponin, paramyosin and myosin light chains comprise four distinct protein assemblages. ]. Exp. ZooL 234, 23-32. NISHITA, K. & OJIIMA, T. (1990) American lobster troponin. J. Biochem. (Tokyo) 108, 677-83. O'FARRELL,P. H. (1975) High resolution two-dimensional electrophoresis of proteins. ]. Biol. Chem. 250, 4007-21. O'FARRELL,P. Z., GOODMAN, H. M. & O'FARRELL,P. H. (1977) High resolution two-dimensional electrophoresis of basic as well as acidic proteins. Cell 12, 1133-42. OHTSUKI, I., MARUYAMA,K. & EBASHI,S. (1986) Regulatory and cytoskeletal proteins of vertebrate skeletal muscle. Adv. Protein Chem. 38, 1-67. OJIMA, T. & NISHITA, K. (1986a) Isolation of troponins from striated and smooth adductor muscles of Akazara scallop. ]. Biochem. (Tokyo) 100, 821-4.

618 OJIMA, T. & NISHITA,K. (1986b) Troponin from Akazara scallop striated adductor muscles. ]. Biol. Chem. 261, 16749-54. PAN, B-S., GORDON, A. M. & LUO, Z. (1989) Removal of tropomyosin overlap modifies cooperative binding of myosin S-1 to reconstituted thin filaments of rabbit striated muscle. ]. Biol. Chem. 264, 8495-8. PARRY, D. A. D. & SQUIRE, J. M. (1973) Structural role of tropomyosin in muscle regulation: analysis of the X-ray diffraction patterns from relaxed and contracting muscles. ]. Mol. Biol. 75, 33-55. POTTER, J. D. (1974) The content of troponin, tropomyosin, actin and myosin in rabbit skeletal muscle myofibrils. Arch. Biochem. Biophys. 162, 436-41. REGENSTEIN, J. M. & SZENT-GYORGYI, A. G. (1975) Regulatory proteins of lobster striated muscle. Biochemistry 14, 917-25. SHINODA, Y., YAMADA, A. & YAGI, K. (1988) Identification of troponin-I of crayfish myofibrils. ]. Biochem. (Tokyo) 103, 636-40. STONE, D. & SMILLIE,L. B. (1978) The amino acid sequence of rabbit skeletal ~-tropomyosin. J. Biol. Chem. 253, 1137-48.

MIEGEL, K O B A Y A S H I and MAI~DA TAWADA,Y., OHARA,H., OOI, T. & TAWADA,K. (1975) Non-polymerizable tropomyosin and control of the superprecipitation of actomyosin. J. Biochem. (Tokyo) 78, 65-72. VIBERT, P. J., HASELGROVE,J. C., LOWY, J. & POULSEN, F. R. (1972) Structural changes in actin-containing filaments of muscle. J. Mol. Biol. 71, 757-67. WALSH, T. P., TRUEBLOOD, C. E., EVANS, R. & WEBER, A. (1984) Removal of tropomyosin overlap and the co-operative response to increasing calcium concentrations of the acto-subffagment-1 ATPase. ]. MoI. Biol. 182, 265-9. WHITE, H. D. (1982) Special instrumentation and techniques for kinetic studies of contractile systems. In Methods in Enzymology, Vol. 85 (edited by FREDERIKSEN,D. W. & CUNNINGHAM, L. W.) pp. 698--708. New York: Academic Press. WNUK, W. (1989) Resolution and calcium-binding properties of the two major isoforms of troponin C from crayfish. ]. Biol. Chem. 264, 18240-6. WNUK, W., SCHOECHLIN, M. & STEIN, E. A. (1984) Regulation of actomyosin ATPase by a single calcium-binding site on troponin C from crayfish. ]. Biol. Chem. 14, 9017-23.