Muscle physiology and biochemistry

Troponin I: Inhibitor or facilitator Key words: troponin I, troponin C, troponin T, troponin, tropomyosin, actin, actomyosin, calcium activated MgATPase, calciumsensitivity, skeletal, ca

MUSCLE PHYSIOLOGYAND BIOCHEMISTRY

Muscle Physiology and Biochemistry

Reprinted from Molecular and Cellular Biochemistry, Volume 190 (1999)

Springer-Science+Business Media, B.V

Library of Congress Cataloging-in-Publication Data

Musc1e physiology and biochemistry/edited by Shoichi Imai, Makoto

Endo, Iwao Ohtsuki

p cm (Developments in molecular and cellular

biochemistry )

ISBN 978-1-4613-7534-0 ISBN 978-1-4615-5543-8 (eBook) DOI 10.1007/978-1-4615-5543-8

1 Musc1es Physiology 2 Musc1e contract ion 3 Musc1es-

-Molecular aspects 4 Musc1es Metabolism 1 Imai, Shoichi,

1931- II Endo, Makato 1933- III Ohtsuki, Iwao

IV Series

QP321.M8917 1998

ISBN 978-1-4613-7534-0

Printed an acid-free paper

Ali rights reserved

© 1999 Springer Science+Business Media Dordrecht

Originally published by K1uwer Academic Publishers in 1999

Softcover reprint of the hardcover 1 st edition 1999

No part ofthe material protected by this copyright notice may be reproduced or utilized in any farm or by any means, electronic or mechanical,

inc1uding photocopying, recarding ar by any information storage and

retrieval system, without written permis sion from the copyright owner

Molecular and Cellular BiochelDistry:

An International Journal for Chemical Biology in Health and Disease

CONTENTS VOLUME 190, Nos 1 & 2, January (I) 1999

MUSCLE PHYSIOLOGY AND BIOCHEMISTRY

Shoichi Imai, Makoto Endo and Iwao Ohtsuki

Preface

M Endo: Dedication

J.Gergely: Professor Ebashi' s impact on the study ofthe regulation of striated muscle contraction

S.Y Perry: Troponin I: Inhibitor or facilitator

I Ohtsuki: Calcium ion regulation ofmuscle contraction: The regulatory role oftroponin T

K Yamada: Thermodynamic analyses ofcalcium binding to troponin C, calmodulin and parvalbumins by using microcalorimetry

M Yazawa, K.-i Nakashima and K Yagi: A strange calmodulin of yeast

A.G Szent-Gyorgyi, Y.N Kalabokis and C.L Perreault-Micale: Regulation by molluscan myosins

Y Yazawa and M Kamidochi: The properties and function of invertebrate new muscle protein

A Weber: Actin binding proteins that change extent and rate of actin monomer-polymer distribution by different mechanisms

M Tanokura andY Suzuki: A phosphorus-31 nuclear magnetic resonance study on the complex ofchicken gizzard myosin subfragment

I with adenosine diphosphate

DJ Hartshorne and K Hirano: Interactions of protein phosphatase type 1, with a focus on myosin phosphatase

K Fujita, L.-HYe, M Sato,T.Okagaki, Y Nagamachi and K Kohama: Myosin light chain kinase from skeletal muscle regulates

anATP-dependent interaction between actin and myosin by binding to actin

T.Murahashi,A Fujita and T Kitazawa: Ca 2 -induced Ca 2 desensitization ofmyosin light chain phosphorylation and contraction in

phasic smooth muscle

T.Masuda, K Ohmi, H Yamaguchi, K Hasegawa,T.Sugiyama,Y.Matsuda, M lino andY.Nonomura: Growing and differentiating

characterization ofaortic smooth muscle cell line, p53LMACO 1 obtained from p53 knock out mice

K Sobue, K Hayashi and W Nishida: Expressional regulation of smooth muscle cell-specific genes in association with phenotypic

modulation

I Niki and H Hidaka: Roles of intracellular Ca 2+receptors in the pancreatic~-cellin insulin secretion

Y.Soeno, H Yajima,Y.Kawamura, S Kimura, K Maruyama andT.Obinata: Organization ofconnectinltitin filaments in sarcomeres

of differentiating chicken skeletal muscle cells

K.-i Kusano, H Abe andT.Obinata: Detection of a sequence involved in actin-binding and phosphoinositide-binding in the

N-terminal side ofcofilin

E Ozawa,Y.Hagiwara and M Yoshida: Creatine kinase, cell membrane and Duchenne muscular dystrophy

T.Masaki, H Ninomiya,A Sakamoto and Y Okamoto: Structural basis of the function of endothelin receptor

Y.Yoshida,A Toyosato, M.O Islam, T Koga, S Fujita and S Imai: Stimulation ofplasma membrane Ca 2+-pump ATPase ofvascular

smooth muscle by cGMP-dependent protein kinase: Functional reconstitution with purified proteins

H Yamamoto and M Kawakita: Chemical modification ofan arginine residue in theATP-binding site ofCa 2+-transportingATPase

of sarcoplasmic reticulum by phenylglyoxal

M Hirata, M Yoshida,T.Kanematsu and H Takeuchi: Intrinsic inhibitor of inositol I,4,5-trisphosphate binding

M lino: Dynamic regulation of intracellular calcium signals through calcium release channels

Y Ogawa,T.Murayama and N Kurebayashi: Comparison of properties ofCa H release channels between rabbit and frog skeletal

Molecular and Cellular Biochemistry190: 1, 1999.

Preface

The papers in this issue were contributed by close friends,

coworkers and pupils of Professor Setsuro Ebashi They are

dedicated to him to commemorate his great and pioneering

contribution to the advancement of muscle physiology and

biochemistry, which in course of time exerted a great

in-fluence on the whole field of life science We would like to

express our cordial thanks to an the contributors who made

the publication of this issue possible Owing to some expected troubles ofone ofthe editors (M E.) the publicationofthis issue has been greatly delayed, for which he sincerelyapologizes to all the contributors and other editors Webelieve that this issue reveals the present state of research onmuscle and/or calcium that had been opened up by ProfessorEbashi

un-Shoichi Imai, Niigata, JapanMakoto Endo, Saitama, JapanIwao Ohtsuki, Fakuoka, Japan

Setsuro Ebashi was born in Tokyo on 31 st August, 1922

There is a Japanese saying that 'Sandalwood is fragrant even

in seed leaf.' Genius displays itself even in childhood

Finishing the six-year course ofprimary school in five years

and the five-year course of middle school in four years, he

entered the First High School, the most prestigious high

school in Japan, at the age of only 15

In July 1942, when he was an undergraduate student of

Faculty of Medicine, Tokyo Imperial University (now called

University ofTokyo), he by chance visited the laboratory of

Dr Hiroshi Kumagai, at that time Lecturer in Pharmacology,

to have a practical training during the summer vacation This

was the beginning of an admirable relationship of love and

kindness between a pupil and a teacher as well as the start of

Dr Ebashi' s muscle research However, the War severed the

relationship: the teacher went to Indonesia to teach in Jakarta

Medical School, and the pupil received his M.D degree in

1944 and served in the war as a naval surgeon When Dr

Ebashi was demobilized in 1946, he went straight to Dr

Kumagai's laboratory again

Dr Ebashi's research was at first electrophysiology of

smooth muscle in which Dr Kumagai had a deep interest

However, in 1950 Dr Ebashi was deeply impressed with a J

Physiol paper by Hodgkin and Katz (1949) which completely

elucidated the mechanism of excitation as he felt At about

the same time he was also deeply inspired by a book

'Chemistry of Muscular Contraction' byA Szent-Gyorgyi

(1949) These readings led him to change the subject of his

research to the contractile mechanisms

He raised the following question Although Szent-Gyorgyi

demonstrated thatATP added to the actin-myosin system such

as actomyosin thread or glycerinated muscle induces

con-traction, removal ofATP does not cause relaxation, which is

quite different from, for example, acetylcholine-induced

contraction of living muscle, where the removal of

acetyl-choline causes relaxation His idea was that there must be

something in living muscle to cause relaxation, which was

lost and absent in the actomyosin systems He started to

search for the relaxing factor in homogenized muscle and

soon he found the factor and reported to a meeting of a

Japanese muscle physiology group in 1952 Sometime after

this Dr Kumagai found a paper by Marsh in Nature (1951)

that had already reported the same factor However, this was

not a disappointment for young Dr Ebashi but rather an

encouragement because it proved that his direction of

research was right Having inquired further into the relaxingfactor, he demonstrated in 1955 that the essential component

of the relaxing factor was in the particulate fraction, againstthe general beliefat that time that it may beATP-regeneratingsoluble enzyme(s)

As for the mechanism ofrelaxation by the relaxing factor,once again against the general belief at that time that therelaxing factor might produce some (organic) substancewhich in tum acts on the actomyosin system to cause relaxa-tion, Dr Ebashi showed in the early 60s that removal ofCa2+ion from the medium by the relaxing factor is the cause ofrelaxation His evidence consisted oftwo important discoveriesthat the particulate relaxing factor strongly accumulates Ca2+ion from the medium in the presence ofATP, and that a minuteamount of Ca2+ion is necessary for the contractile reaction

of well-washed Ca2+-free natural actomyosin system though physiologists had recognized the contraction-inducing action of Ca2+ ion, it had not been recognized bymuscle biochemists before Dr Ebashi, because all thebiochemical experiments were done in the presence ofsufficient amount of Ca2+ion contaminated from reagents orexuded from glasswares Dr Ebashi further demonstratedelectronmicroscopically that the relaxing factor has a vesi-cular structure, indicating that it is the fragment ofthe sarco-plasmic reticulum (SR) Since relaxation is the reverse of

Al-Molecular and Cellular Biochemistry 190:5-8, 1999.

© 1999 Kluwer Academic Publishers.

regulation of striated muscle contraction

Key words: troponin, tropomyosin, thin filament regulation, Ca2+

Introduction

I am indeed greatly pleased and honored to be able to join in

the celebration of the remarkably productive and influential

contribution of Professor Ebashi's life in science This gives

me particular pleasure since so much of what my colleagues

and I were able to do in the last four decades has depended

on his contributions

Dr Ebashi became known very early in his career by

establishing the nature ofthe so-called relaxing factor This was

followed by the identification of calcium ions as key

mes-sengers in the activation process of muscle contraction and

the discovery of the troponin complex which, together with

tropomyosin, was identified as the actin-bound regulatory

system of striated muscle (See [I]) In this brief review (For

detailed reviews see: [1,52-57]) I should like to trace some

of the developments concerning the regulation of striated

muscle that were sparked by the ground-breaking work ofDr

Ebashi and his colleagues and to point to some questions that

currently await answers

The discovery of troponin

The new component discovered in Dr Ebashi's laboratory

was first known as native tropomyosin [2,3].Itrendered the

interaction of actin and myosin in the presence ofATP Ca2+

sensitive Native tropomyosin was soon separated into twocomponents: tropomyosin and a new entity that becameknown as troponin Troponin was identified as the receptorfor Ca2+, whose role in actomyosin activation had earlierbeen established [4, 5] with four Ca-binding sites in troponin[6] There were indications that troponin is a multi-componentsystem [7] The work initiated in the Ebashi laboratorystarted a new era in muscle research involving manylaboratories all over the world It became clear that therewas a calcium binding component and another one thatinhibited the Mg2+stimulatedATPase ofpurified actomyosin[8,9] By 1972 a general consensus was reached that troponinconsists of three subunits [10-13] whose names indicate theirroles: viz troponin C (TnC), troponin I (Tnl), and troponin T(TnT) for calcium binding, inhibition, and troponin binding,respectively

The availability of purified components oftroponin made

it possible to build on the findings that emerged in the earlierstages Calcium binding studies [14], utilizing the calciumbuffer system originally developed by Dr Ebashi and hiscolleagues [5], located four calcium binding sites in TnC andshowed that there are two classes each containing twocalcium binding sites Two sites bind calcium with highaffinity, as well as Mg2+ although with lower affinity, whilethe other two sites ofTnC are essentially specific for calcium

Address/or offprints: J.Gergely, Muscle Research Group, Boston Biomedical Research Institute, 220 Staniford Street, Boston, MA 02114-2500, USA

Troponin C - The Ca2+ receptor

While research on troponin began to flourish studies on a

related protein led to important results Parvalbumin, which

had first been found in fish muscle as a cytoplasmic rather

than a myofibrillar protein, was characterized both in terms

of primary structure and crystal structure as having two Ca2+

binding sites [15] This work led to the concept of the

so-called EF hands suggested by a bent middle finger, the

thumb and index finger as depicting a calcium binding loop

flanked by twoahelices as the model ofcalcium binding sites

originally found in parvalbumin and by now known to occur

in large super families of Ca2+binding proteins

When the amino acid sequence ofTnC became known [16]

a high degree ofhomology with parvalbumin was recognized

and the Ca2+binding sites were identified A variety ofstudies

have shown that sites I and II in the N-terminal domain are

Ca-specific sites; sites III and IV are the high affinity Ca-Mg

sites in the C domain The former are recognized as the

func-tionally important triggering sites (see [17]) and references

therein)

The similarities between the TnC and parvalbumin

struc-tures led to speculations about how two parvalbumin-like

halves could be fitted into the structure ofTnC [18] When the

structure ofTnC was solved by x-ray crystallography [19, 20],

it showed two domains each a parvalbuminlike structure

-connected, however, by a singleahelix instead ofa compact

molecule essentially containing two paravalbumin-like

structures

The molecular switch in troponin C

An important step toward our current understanding of the

chain ofevents initiated by calcium binding to the triggering

sites inTnC came from insights ofHerzberg et al [21] gained

in comparing the structure of the N- and C-terminal

homo-logous domains in TnC Owing to the conditions of

crystal-lization the former contained no bound calcium, while the

latter had two sites occupied by Ca2+ Thus the difference

between the two domains would give a clue to the

con-formational changes brought about by calcium when it

becomes bound to the N terminal sites This led to the

suggestion that the connector between helices Band C

together with the link between them moves away from helix

D which is part ofthe long helix connecting the two domains,

exposing a hydrophobic area which was presumed to become

an interacting site with Tn! Soon thereafter various pieces

ofevidence emerged for this view Site directed mutagenesis

of charged residues [22] or disulfide formation between

genetically engineered Cys residues [23], in segments whose

separation was expected to change upon Ca2+ binding

according to the model, led to changes in Ca2+-binding and

ATPase activity Distance determinations by resonanceenergy transfer between probes on appropriately placedengineered Cys residues showed a Ca2+-induced changecorresponding to the expectations based on the model [24].Finally, solution ofthe high resolution NMR structure ofTnCwith four Ca2+bound [25] brought definitive proof for thepostulated structure The opening of the N terminal domain

of TnC may be considered as the molecular switch in TnC.The NMR structure revealed some differences between helix

B in the N-terminal domain and the corresponding helix inthe C terminal domain even when both sites in each domainwere occupied by Ca2+.Italso pointed to some flexibility ofthe central helix in solution A recent comparison ofthe highresolution NMR structures ofcardiac and skeletal TnC in the4-Ca2+ state shows that the extent of opening of the hydro-phobic surface is much less in the case of cardiac muscle, afinding whose full implications are yet to be explored [26].The opening of the N terminal domain of TnC may beconsidered as the molecular switch in TnC

The molecular switch in troponin I

The next question that has received some partial answers overthe years is the status of the molecular switch in Tn! In theabsence ofhigh resolution structures for TnI and its complexessuch answers must remain tentative There is evidence thatportions of Tnl move under the influence of activation fromTnC to actin, and under conditions corresponding to relaxationthey return to TnC One of the sites that has been used iscysteine 133 [27] and current studies are further exploringmovements in Tnl by labeling cysteine residues introduced bygenetic engineering as has been done in the case ofTnC Thereare some not fully answered questions concerning the relation

of the region that comes into close contact with actin and theso-called inhibitory region that emerged in earlier studies andcontains the stretch of residues 96-116 in Tnl [28] Evidence

is accumulating that this inhibitory region is indeed acting with both domains of TnC [29-31] but its mode ofinteraction with actin needs further elucidation During recentyears a reasonable consensus has emerged concerning theoverall arrangement of the polypeptides in TnC- and Tnlrelative to each other Both crosslinking [32, 33] and fragmentbinding [34] studies suggest that the two chains run inopposite directions; that is, the N terminus of TnC interactsmainly with the C terminus ofTnI and vice versa However,evidence is also at hand indicating that within Tnl certainstretches may run locally in opposite directions while theoverall trend is preserved Recent work on troponin with Tnlcontaining only Cys 133 and Cys 48 thiols for placement ofprobes for resonance energy transfer studies has shown metaldependent conformational changes in Tnl modulated by theinteraction oftroponin with actin [35]

inter-Research on certain aspects of TnI/TnC interaction has

been stimulated by studies on calmodulin, which is an

activator of a large number of enzymes In light of the close

similarities between TnI and calmodulin with respect to their

chemical and crystallographic structure the question arises

whether the structural changes occurring when TnI binds to

TnC are similar to those taking place on the interaction of

calmodulin with one of its target proteins In the case of

calmodulin, both x-ray diffraction [36] and multidimensional

NMR [37] studies showed that at least with the M13 peptide

derived from myosin light chain kinase - a well known target

of calmodulin playing a role in the activation of smooth

muscle contraction - is accompanied by a large structural

change in calmodulin bringing the two globular domains

homologous to those in TnC close together As far as the

TnI'TnC complex is concerned, evidence points in the

opposite direction indicating an essentially extended

structure for TnC based both on resonance energy transfer

distance determinations [38] and low angle x-ray and

neutron diffraction studies [39,40] The latter studies also

point to the existence of masses derived from TnI beyond

the Nand C terminal domains of TnC, a picture whose

details remain to be filled in in terms of the course of the

polypeptide chain in Tn! Students of this field are eagerly

awaiting a more definitive x-ray crystallographic study of

the intact troponin complex Very recently crystallization of

a complex between an N-terminal fragment of TnI and TnC

has been achieved and the high resolution structure derived

from x-ray diffraction reported [41] This work provides some

interesting interim results pending the availability of crystals

of the full complex

Troponin T

Although the shape ofTnT has been long established starting

with the immuno-electron microscopic demonstration that the

globular C-terminal portion and the highly a-helical

N-terminal portion of TnT occupy distinguishable sites along

tropomyosin [42], little is known about the role ofTnT in the

mechanism of regulation While early work has considered

TnT mainly as an anchor tying the rest of the complex to

tropomyosin, hence the suffix T to troponin, recent evidence

assigns a more active role to TnT serving as a signal transmitter

between TnC and TnI [43] as well as modulating the effect

ofmyosin heads on the activity ofthe troponin complex [44)

Filament regulation and cooperativity

From the earliest days ofthe identification ofthe participants

in the regulatory machinery, viz troponin and tropomyosin,

7the question of how changes in solution are related to those

in the actin filament itself have been intriguing (see [45] andreferences therein) Recent x-ray diffraction studies onreconstituted actomyosin gels and on muscle fibers havethrown new light on tropomyosin movement associated with

Ca2+activation [46-48).Itappears that in the regulated thinfilament Ca2 +binding to troponin is accompanied by amovement about 30° azimuthally towards the central groovefrom a position where it would block strong myosin binding,according to the current model of myosin-actin interactions(see [49]) Binding of myosin, which would take place to asite partially unblocked by calcium, causes a small butsignificant further change in the position of tropomyosin,consistent with a cooperative role of myosin - first pointed

to by A Weber and her colleagues [50] - in the full activation

of the thin filament This two step model ofactivation seems

to be in harmony with the three state model based on kineticstudies in solution, the third state being the Ca2+ free,'blocked' state [51]

adenosine-5'-5 Ebashi S: Calcium binding and relaxation in the actomyosin system J Biochem 48: 150 -151, 1960

6 Ebashi S, Ebashi F, KodamaA: Troponin as the Ca 2 +-receptive protein

in the contractile system J Biochem 62: 137-138, 1967

7 Ebashi S, Wakabayashi T, Ebashi E: Troponin and its components J Biochem 69: 441-445, 1971

8 Hartshorne OL, Perry SV, Schaub MC: A protein factor inhibiting the magnesium-activated adenosine triphosphatase of desensitized actomyosin Biochem J 104: 907-913,1967

9 Hartshorne OJ, Mueller H: Fractionation oftroponin into two distinct proteins Biochem Biophys Res Comm 31: 647 653, 1968

10 Greaser ML, Gergely J: Reconstitution oftroponin activity from three protein components J Bioi Chern 246: 4226-4233, 1971

II Greaser ML, Gergely J: Purification and properties of the components from troponin J Bioi Chern 248: 2125-2133,1973

12 Ebashi S: Separation oftroponin into its three components J Biochem 72: 787-90, 1972

13 Perry SV, Cole HA, Head JF, Wilson FL: Localization and mode of action of the inhibitory protein component of the troponin complex Cold Spring Harbor Symp Quant Bioi 37: 251-262, 1972

14 Potter JO, Gergely J: The calcium and magnesium binding sites on troponin and their role in the regulation of myofibrillar adenosine triphosphatase J BioI Chern 250: 4628-4633, 1975

15 Kretsinger RH, Nockolds CE: Carp muscle calcium binding protein.

II Structure determination and general description J Bioi Chern 248: 3313-3326, 1973

16 Collins lH, Potter lD, Hom Ml, Wilshire G, lackman N: The amino

acid sequence of rabbit skeletal muscle troponin C: Gene replication

and homology with calcium-binding proteins from carp and hake

muscle FEBS Lett 36: 268-272, 1973

17 Grabarek Z, Drabikowski W, Leavis PC, Rosenfeld SS, Gergely L:

Proteolytic fragments oftroponin C Interactions with the other troponin

subunits and biological activity 1 Bioi Chern 256: 13121-13127, 1981

18 Kretsinger RH, Barry CD: The predicted structure ofthe calcium-binding

component oftroponin Biochim Biophys Acta 405: 40-52,1975

19 Sundaralingam M, Bergstrom R, Strasburg G, Rao ST, Roychowdhury

P,el al.:Molecular structure oftroponin C from chicken skeletal muscle

at 3-angstrom resolution Science 227: 945-948, 1985

20 Herzberg 0, lames MN: Structure of the calcium regulatory muscle

protein troponin-C at 2.8 A resolution Nature 313: 653-659, 1985

21 Herzberg 0, Moult 1, lames MN: A model for the Ca'+-induced

conformational transition of troponin C A trigger for muscle

contraction 1 Bioi Chern 261: 2638-2644, 1986

22 Fujimori K, Sorenson M, Herzberg 0, Moult 1, Reinach FC: Probing

the calcium-induced conformational transition of troponin C with

site-directed mutants Nature 345: 182-184, 1990

23 Grabarek Z, Tan RY, Wang 1, Tao T, Gergely J: Inhibition of mutant

troponin C activity by an intra-domain disulphide bond Nature 345:

132-135,1990

24 Wang Z, Gergely 1, Tao T: Characterization of the Ca'+-triggered

conformational transition in troponin C Proc Natl Acad Sci USA 89:

11814-11817, 1992

25 Siupsky CM, Sykes BD: NMR solution structure ofcalcium-saturated

skeletal muscle troponin C Biochemistry 34: 15953-15964, 1995

26 Gagne SM, Li MX, McKay RT, Sia SK, Spyracopoulos L,el al.:The

calcium induced structural change in that triggers skeletal and cardiac

muscle contraction Biophys 1 72: A332, 1997

27 Tao T, Gong BJ, Leavis PC: Calcium-induced movement oftroponin-l

relative to actin in skeletal muscle thin filaments Science 247:

1339-1341,1990

28 Syska H, Wilkinson 1M, Grand RJ, Perry SV: The relationship between

biological activity and primary structure of troponin I from white

skeletal muscle of the rabbit Biochem J 153: 375-387, 1976

29 Leszyk 1, Grabarek Z, Gergely J, Collins JH: Characterization of

zero-length cross-links between rabbit skeletal muscle troponin C and

troponin I: Evidence for direct interaction between the inhibitory region

oftroponin I and the NH,- terminal, regulatory domain oftroponin C.

Biochemistry 29: 299-304, 1990

30 Pearlstone lR, Smillie LB: Evidence for two-site binding oftroponin

I inhibitory peptides to the Nand C domains of troponin C

Bio-chemistry 34: 6932-6940, 1995

31 Pearlstone JR, Sykes BD, Smillie LB: Interactions of structural

C-domain and regulatory N-C-domains of troponon C with repeated

sequence motifs in troponin I Biophys J 72: A331, 1997

32 Kobayashi T, Tao T, Gergely J, Collins 1: Structure of the troponin

complex Implications of photocross-linking oftroponin I to troponin

C thiol mutants 1 Bioi Chern 269: 5725-5729, 1994

33 lha PK, Mao C, Sarkar S: Photo-cross-linking of rabbit skeletal

troponin I deletion mutants with troponin C and its thiol mutants: The

inhibitory region enhances binding oftroponin I fragments to troponin

C Biochemistry 35: 11026-11035, 1996

34 Farah CS, Miyamoto CA, Ramos C, Dasilva A, Quaggio RB,el al.:

Structural and regulatory functions of the NH,- and COOH-terminal

regions of skeletal muscle troponin I J Bioi Chern 269: 5230-5240,

1994

35 Luo Y, Wu l-L, Gergely 1, Tao T: Troponin T and Ca2+ dependence of

the distance between Cys48 and Cys 133 of troponin I in the ternary

troponin complex and reconstituted thin filaments Biochemistry 36:

11027-11035,1997

36 Meador WE, Means AR, Quiocho FA: Target enzyme recognition by calmodulin: 2.4Astructure of a calmodulin-peptide complex Science 257: 1251-1255,1992

37 Ikura M, Clore GM, Gronenborn AM, Zhu G, Klee CB, Bax A: Solution structure of a calmodulin-target peptide complex by multi- dimensional NMR Science 256: 632-638, 1992

38 Gong B-1, Wang Z, Tao T, Gergely J: Troponin C remains extended in the ternary trroponin complex Biophys 1 66: A346, 1994

39 Olah GA, Rokop SE, Wang C-LA, Blechner SL, Trewhella J: Troponin

I encompasses an extended troponin C in the Ca'+-bound complex - a small-angle X-ray and neutron scattering study Biochemistry 33: 8233-8239, 1994

40 Olah GA, Trewhella J: A model structure ofthe muscle protein complex 4-Ca'+-troponin C-troponin I derived from small-angle scattering data

- implications for regulation Biochemistry 33: 12800-12806, 1994

41 Vassylyev DG, Takeda S, Wakatsuki S, Maeda K, Maeda, Y: Crystal structure of troponin C in complex with troponin I fragment at 2.3A resolution Proc Natl Acad Sci (USA) 95: 4747 4852, 1998

42 Ohtsuki I: Molecular arrangement oftroponin-T in the thin filament J Biochem 86: 491 497,1979

43 Potter JD, Sheng Z, Pan BS, Zhao J: A direct regulatory role for troponin T and a dual role for troponin C in the Ca'+ regulation of muscle contraction 1 BioI Chern 270: 2557-2562, 1995

44 Schaertl S, Lehrer SS, Geeves MA: Separation and characterization

of the two functional regions of troponin involved in muscle thin filament regulation Biochemistry 34: 15890-15894, 1995

45 Potter JD, Gergely 1: Troponin tropomyosin and actin interactions in the Ca2+ regulation ofmuscle contraction Biochemistry 13:2697-2703, 1974

46 Lorenz M, Poole KJV, Popp D, Rosenbaum G, Holmes KC:An atomic model of the unregulated thin filament obtained by X-ray fiber diffraction on oriented actin-tropomyosin gels J Mol Bioi 246: 108- 119,1995

47 Poole KJV, Evans G, Rosenbaum G, Lorenz M, Holmes KC: Thc effect

of crossbridges on the calcium sensitivity of the structural change in the regulated thin filament Biophys J 68: A365, 1995

48 Holmes KC: The actomyosin interaction and its control by myosin Biophys 168: S2-S7, 1995

tropo-49 Rayment 1, Holden HM, Whittaker M, Yohn CB, Lorenz M et al.:

Structure ofthe actin-myosin complex and its implications for muscle contraction Science 261: 58-65, 1993

50 Bremel RD, Murray JM, Weber A: Manifestations of cooperative behavior in the reglated actin filament during actin activated ATP hydrolysis in the presence ofcalcium Cold Spring Harbor Symp Quant BioI 37: 267-275,1972

51 McKillop D, Geeves MA: Regulation of the interaction between actin and myosin subfragment-I - evidence for 3 states ofthe thin filament Biophys J 65: 693-701,1993

52 Leavis PC, Gergely 1: Thin filament proteins and thin filament-linked regulation of vertebrate muscle contraction CRC Crit Rev Biochem 16: 235-305,1984

53 Ohtsuki 1, Maruyama K, Ebashi S: Regulatory and cytoskeletal proteins

of vertebrate skeletal muscle Adv Prot Chern 38: 1-67, 1986

54 Zot AS, Potter lD: Structural aspects of troponin-tropomyosin regulation of skeletal muscle contraction Ann Rev Biophys Biophys Chern 16: 535-559, 1987

55 Chalovich 1M: Actin mediated regulataion of muscle contraction Pharmac Ther 55: 95-148,1992

56 Farah CS, Reinach FC: The troponin complex and regulation of muscle contraction [Review] FASEB 19: 755-767, 1995

57 Tobacman LS: Thin Filament-mediated Regulation of Cardiac Contraction [Review] Ann Rev Physiol58: 447 481,1996

Molecular and Cellular Biochemistry190: 9-32, 1999.

©1999 Kluwer Academic Publishers.

Troponin I: Inhibitor or facilitator

Key words: troponin I, troponin C, troponin T, troponin, tropomyosin, actin, actomyosin, calcium activated MgATPase, calciumsensitivity, skeletal, cardiac muscle, muscle regulation, protein kinase A, protein kinase C, phosphorylation, phosphorylationsite, inhibitory peptide, actin binding site, binding site

Introduction

I was probably one ofthe first muscle scientists from the West

to meet Dr Setsuro Ebashi in Japan after World War II The

occasion was not a scientific conference but during a tour of

Japan in 1953 by the Cambridge University rugby team, for

which at the time I was acting as manager I met Dr Ebashi

and his mentor, Professor Kumagai, during a diversion from

my sporting responsibilities when I gave a lecture at the

University ofTokyo about our work on the muscle proteins

From that time we have maintained a friendship and my

respect for him has grown steadily with his many

achieve-ments.Itis a very special pleasure for me to contribute to this

volume dedicated to my long standing friend

Our research careers have been dominated by similar

interests In 1955 he reported [1] the association of relaxing

factor activity with the particulate fraction from muscle At

about the same time we had shown [2, 3] that the MgATPaseactivity ofcrude preparations ofactomyosin, euphemisticallycalled 'natural actomyosin' was sensitive to low concentra-tions of EDTA and glycolcomplexon, the name then used forEGTA, and which we had obtained from Schwarzenbachbefore it was commercially available Thus it was clear to us

at that time that a trace of calcium was required for theMgATPase of 'natural actomyosin' On the other hand wereported that the MgATPase ofactomyosin prepared from thepurified proteins was insensitive to calcium chelators.Unfortunately we were not smart enough to show why thesetwo preparations differed in behaviour.Itwas left to Ebashi

in 1963 [4] to report that 'natural tropomyosin' (tropomyosin

+troponin) was responsible for the effect and thus open upthe whole field of calcium regulation in striated muscle As

a consequence ofhis discovery oftroponin my colleagues and

I were able later to identify a component of the complex,

Address for offprints:S.V Perry, Department of Physiology, Medical School, University of Birmingham, Birmingham, B15 2TT, UK

troponin I Over the years troponin I has become one of my

favourite proteins and this review of its properties, the

development ofideas on its function and thoughts on its mode

of action represents my contribution to this commemorative

volume

Discovery and early studies on troponin I

During the development of'desensitized actomyosin' for the

assay of troponin (or EGTA sensitising activity, as it was

called at that time) it was noted that occasionally, and

especially after ageing, the troponin extracts developed

inhibitory activity that was not calcium sensitive [5] The fact

that the inhibitory factor was specific for the MgATPase of

actomyosin and inhibited its superprecipitation in the absence

ofEGTA suggested that it might be derived from the troponin

complex or was a modified form of it [6, 7] This hypothesis

was confirmed when it was later shown that troponin could

be fractionated into inhibitory (troponin B) and calcium

sensitising (troponin A) factors [8, 9] Later it became clear

that some inhibitory protein fractions also contained another

basic protein of higher molecular weight, the '37000

component', later called troponin T, [10], which was shown

to form a viscous complex with tropomyosin In 1972 the

nomenclature proposed by Greaser et al.[ll] for the

com-ponents of the troponin complex, troponin C, I and Twas

adopted Before the nomenclature was rationalised the

various research groups working on the fractionation of

troponin used their own names to distinguish their fractions

and the inhibitory factor was variously known as troponin B,

troponin 2, troponin II and component II (see [12] for details)

Isoforms of troponin I and their

distribution

Muscle is the only tissue that has been shown to contain

significant amounts of TN-I In vertebrates it is restricted to

striated muscle and thin filament regulation in smooth muscle

involves another actin-binding protein, caldesmon, which

possesses some properties similar to those ofTN-I Isoforms

of TN-I have been reported to be present in invertebrate

smooth muscle in a few instances These include the adult

body wall ofthe ascidian, Halocynthia roretzi [13], adductor

muscle of the scallop [14] and the oviduct myoepithelial

sheath of Caenorhabitis elegans [15]

Three isoforms, fast skeletal, slow skeletal and cardiac

TN-I , each the product of a separate gene, are present in

mam-malian striated muscle [16] As yet there is no evidence of a

distinct foetal form and the earliest isoform detected in

myocytes growing in culture and early embryos is the slow

skeletal form [17].Itis ofinterest in this respect that the foetalheart contains the slow skeletal isoform which is slowlyreplaced during immediate post natal development by thecardiac isoform [18, 19] This process is under the controlofthe transcription factor GATA-4 [20] Replacement of theendogenous cardiac isoform in myocytes by adenovirus-mediated skeletal slow TN-I transfer increased the Ca2

sensitivity ofthe tension development ofpermeabilized singlemyocytes [21] Although mammalian skeletal TN-I appears

to be under the control oftwo genes a recent report providesevidence for the expression ofthree genes in whole myotomalmuscle ofsalmon fry [22] During the later foetal stages bothfast and slow skeletal isoforms are present in mammalianskeletal muscle muscle cells [23]

The genes responsible for encoding the isoforms of TNIand TNT are organised in pairs Those for the fast skeletalisoform of TNI (TNNI2) and the fast skeletal form of TNT(TNNT3) are both located on chromosome llp15.5 [23a].The genes for the slow skeletal muscle TNI and cardiac TNTare on chromosome lq32 whereas those for cardiac TNI(TNNI3) and slow TNT are located on chromosome 19q13.4[23b] This organisation contrasts with that ofother sarcomericprotein genes and could be explained if the troponin geneswere derived by triplication of an ancestral TNI/TNT genepair [23a]

With further postnatal development the expression of thegene not appropriate for the cell type is suppressed with theresult that the mature adult skeletal muscle cell usuallycontains only the isoform characteristic for its function [24].The detection ofTN-I isoforms by specific antibody stainingprovides a convenient and reliable method of muscle celltyping [25] and the presence of TN-I in serum can be used forthe detection ofdisease in skeletal and cardiac muscle [26] Theimmunological detection ofserum cardiac TN-I is widely used

in cardiology as an index ofmyocardial damage [27] Slow andfast isoforms are present in the same skeletal muscle cell as aconsequence of cross innervation [28], hormone intervention[29], and in pathological conditions [30]

Invertebrate troponins have not been as widely studied astheir vertebrate counterparts but it is clear that they do notrepresent such a homogeneous group as the latter Attentionhas been directed particularly to striated arthropod musclewhere there were early reports of troponin-like systems ininsect flight muscles [31], crayfish [32], lobster [33], andhorseshoe crab [34]

Components similar to TN-I, C and T ofvertebrate musclehave been identified in many invertebrates but they frequentlydiffer in molecular mass from their vertebrate counterparts.Often the component identified as troponin I has a highermolecular mass than the vertebrate form due to the addition

of residues at the N-terminus.Itis of interest that TN-I fromthe crayfish, Astacus leptodactylus, possesses an additionalN-terminal sequence ofabout 30 residues, similar to vertebrate

cardiac TN-I This region, however, only exhibits a sequence

identity score of22% with the vertebrate cardiac isoform and

does not possess an equivalent phosphorylation site

Com-parison of the sequence of the whole molecule with rabbit

cardiac troponin I reveals a sequence identity of 26% [35]

Sequence analysis does not reveal any polymorphism in this

species of crayfish whereas tail muscle of another crayfish,

Procambarus c1arkii, is reported to contain two isoforms of

troponin I with molecular masses of 25 and 23 kDa [36]

The 52kDa subunit of the troponin system of the striated

adductor of the Akazara scallop, Chlamys nipponensis

akazara, that has been provisionally identified as a TN-I, can

be cleaved into two major fragments [37] The C-terminal

fragment of 17 kDa exhibits 39% sequence homology with

crayfish troponin I [38] On the basis ofthis and the increased

inhibition of actomyosin ATPase obtained with scallop

tropomyosin alone it has been concluded that the 52 kDa

protein is indeed a troponin I.Itis reported in this study that

scallop tropomyosin alone inhibits the actomyosin ATPase

by 88% which is increased to 95% in the presence of TN-I

The N-terminal 35 kDa fragment does not have inhibitory

activity.Ithas a unique amino acid composition with glutamic

acid, arginine and alanine accounting for approximately 75%

ofthe total In the ascidian, Halocynthia roretzi, the troponin

I isoform present in the striated cardiac and the smooth body

wall muscles is similar in polypeptide length to vertebrate

striated muscle TN-I isoforms The ascidian larvae express

two isoforms with truncation ofabout 30 amino acid residues

at the C-terminus indicating that at least three genes are

responsible for encoding TN-I in this organism [39]

The original report of the presence of troponin in

Letho-cerus flight muscle [31] suggested that the system consisted

of components similar to those of the vertebrate system

Nevertheless reinvestigation, taking precautions to minimise

proteolysis [40], has so far failed to identify a protein that is

strictly analogous to vertebrate TN-I.Itis concluded that the

troponin system in flight muscle consists of TN-C, a TN-T

ofhigher molecular mass, 53 kDa, than its vertebrate

counter-part, and troponin H, molecular mass 80 kDa Unlike TN-I,

TN-H did not inhibit actomyosin MgATPase alone but did

in the presence ofTN-T and tropomyosin The inhibition was

relieved in a calcium-sensitive manner in the presence

ofTN-C Lethocerus TN-H can be proteolytically digested to give

a peptide with sequence homologies to vertebrate TN-I(B

Bullard, personal communication) Itis immunologically

similar to a protein ofcomparable molecular mass present in

Drosophila muscle that is a fusion protein oftropomyosin and

a hydrophobic sequence rich in proline [41] In view of the

fact that locus of the TN-I gene family has been identified in

Drosophila it is surprising that a similar TN-I isoform has not

been isolated from Lethocerus Drosophila also expresses

TN-H, a protein which apparently replaces TN-I in

Letho-cerus muscles Barbas et at [42] have studied a region

11identified as HL I which corresponds to one of the comple-mentation groups that constitute the haplolethal region oftheShaker gene complex of Drosophila This region encodes afamily of TN-I proteins that are expressed in a develop-mentally regulated manner As judged from the gene sequencethe protein sequence of Drosophila isoform p6a10 has 65%identity with the well-characterised TN-I of crayfish [35].It

is of particular interest that Barbas et at [42] also report the

the existence of mutant phenotypes in which the expressionofcertain TN-I isoforms is impaired or prevented In addition

to defects confined to specific muscles in these mutants,aberrant neurogenesis is observed This is the first time thatmutations in a TN-I gene have been reported to produce aneffect in nervous tissue These findings suggest that TN-I mayhave a role in nerve cell development in addition to its welldefined function in I filament regulation of striated muscle

Structure of troponin I

The isoforms of TN-I represent a homologous group ofproteins with the molecular ratios ofthose ofvertebrate originlying in the range of about 20000-24000 All consist of asingle polypeptide chain and rabbit fast skeletal muscle TN-

1, which the most widely studied, contains 181 amino acidresidues and possesses a M, of about 21073 (Fig I) Theamino acid sequences of fast skeletal, slow skeletal andcardiac TN-I ofthe rabbit are 60% identical [43] The identity

of sequence is even greater between a given isoform type indifferent species, for example the amino acid residues in fastskeletal muscle isoforms of rabbit and chicken are 85%identical As is the case with TN-T the cardiac forms of TN-

I have a slightly higher molecular ratio, about 24000, thanthe skeletal isoforms In the case of TN-I this is due to anadditional, strongly conserved, N-terminal peptide of about

30 residues in which is located an important phosphorylationsite (see section on phosphorylation) TN-I from invertebratemuscle is more variable in size than the vertebrate protein.It

ranges from the smallest reported to date isolated fromascidian larval muscle consisting of 142 residues [39], to verymuch larger molecules ofmolecular ratio greater than 50000

In some insect flight muscles itmay occur as a fusion proteinwith a tropomyosin-like sequence (see above)

Two regions corresponding to residues 17-23 and 97-121

in the rabbit fast skeletal muscle protein that are considered

to be of functional importance are particularly stronglyconserved in all isoforms The region represented by residues97-121 includes the smallest sequence that possesses in-hibitory activity and is known to interact with actin Theregion of residues 137-144 has some common features ofsequence with the inhibitory peptide region [44] In extension

of these observations it has been pointed out that the terminal portion ofthe polypeptide chain exhibits a conserved

Lys Leu

Me~

Ala

120

ASp _A_r_g A_r_g v_a_I_Ar g,~ Se r Al a

Met Phe Glu Ser GI uSer

Fig I. Sequence of rabbit fast skeletal TN-I derived from a cDNA clone expressed inE coli[97] This corresponds to a protein of 182 amino acid residues with a molecular weight of 21162 kDa The sequence corresponding to the inhibitory peptide [60] is underlined and that corresponding to the minimum inhibitory peptide [62] is enclosed in a box Note: This sequence is slightly different from that originally proposed by Wilkinson and Grand [43] from amino acid analysis of the isolated protein in that a methionine residue replaces the N-terminal acetyl group and argl53, asp 154 and leu 155 are inserted The evidence would suggest that the N-terminal of rabbit fast skeletal TN-I is acylated as are other myofibrillar proteins, M, 21073 In most of the work published the original numbering of residues [43] has been used and has been retained (up to glul52) in this review on the assumption that the naturally expressed protein in the fast skeletal muscle of the rabbit is acylated and consists of 181 residues Authors should make clear, particularly in mutation studies, which sequence is used When results are quoted in this review the residue numbers used by the authors are retained.

repeat motifin three positions ofthe sequences ofmammalian

and avian TN-I sequences currently available [45] Using the

residue numbering ofthe rabbit fast skeletal isoform these are

in the region of residues 101-114 (designated ex), residues

121-132(p)and residues 135-146(y)

TN-I is a basic protein with a high isoelelectric point and

in the isolated form tends to aggregate under physiological

pH values and ionic strength For this reason it has beendifficult to study the isolated protein in solution and, as it hasnot so far been crystallised, a high resolution structure is not

available Despite its tendency to aggregate the isolated rabbit

fast skeletal isoform is readily degraded by proteolysis and

phosphorylated at residues thrll and serll7 by kinases

indicating that most of the molecule is available to solvent

Evidence that suggests the polypeptide chain in the isolated

molecule is in an extended and flexible form Structural

predictions based on the amino sequence of the region

represented by residues 96-148 indicate that it consists ofa

series of alternating coil and helix [45]

TN-I is designed to function as a complex in association

with the other myofibrillar proteins and an intrinsic ability

ofits polypeptide chain to adapt to interacting proteins is vital

for its function in the troponin complex As the interaction

with TN-C is central to its function, the conformation it adopts

in the TN-I1TN-C complex is of particular interest Despite

the availability of the high resolution structure of TN-C,

similar detailed information about the structure ofTN-I in the

free form or complexed with other proteins is not yet

avail-able Some progress has been made towards solving this

problem by the determination of the conformation of the

peptides corresponding to residues 104-115 [46,47] by NMR

methods and the N-terminal region by X-ray crystallography

[48a] In the latter study crystals of the complex oftroponin

C with the N-terminal fragment ofTN-I consisting ofresidues

1-47 were examined From neutron and low angle X-ray

scattering data the so-called 'dumbell structure' model has

been proposed for the TN-I1TN-C complex (see section on

complex below) The more recent neutron scattering patterns

obtained with deuterated TN-I reconstituted in the skeletal

trononin complex suggest that the TN-I in the complex is

elongated and consists of two subdomains [49] The bulk of

the molecular mass (65%) exists as highly oblate ellipsoid of

revolution with the remainder of the molecule present as a

highly prolate ellipsoid ofrevolution The best fit of the data

was obtained when the axes of the ellipsoids were sharply

inclined to each other Unlike the TN-I, in these studies the

TN-C was not observed to undergo a measureable global

change in shape on addition of calcium Italso appeared to

be elongated as in crystals of the isolated protein

Functions of troponin I

TN-I can interact with all the major proteins ofthe I filament,

actin, tropomyosin, TN-C and TN-T, properties that clearly

indicate its central role in the regulatory process in striated

muscle A great deal ofinformation, some ofwhich is at times

confusing, exists about the amino acids and peptide regions

involved in these interactions There is a great need for a high

resolution three dimensional structure of TN-I to which this

data can be related The most striking property of TN-I is its

ability to inhibit the magnesium activated ATPase of

acto-13myosin, but not the calcium activated ATPase (CaATP assubstrate) ofactomyosin or myosin [6] This clearly indicatesthat in some way TN-I blocks the interaction of actin withmyosin that is responsible for activation of the MgATPase

With in vitro systems using purified TN-I inhibition of the

MgATPase of actomyosin can be obtained with a molar ratio

of actin monomer to TN-I of I: I [12] In the presence oftropomyosin the inhibitory action is much enhanced andvalues approaching 90-95%, depending on the conditions,are obtained With molar ratios of actin monomer to TN-I of

3-4: I and higher, inhibition is obtained in in vitro conditions.

The results of immunochemicallocalisation studies and theknown protein composition ofthe myofibril indicate that TN-

I is located at every seventh actin monomer.Itis consideredthat all the actin monomers of the thin filament, excludingthose that may possibly be blocked even in stimulated muscle

by the troponin complex itself, have the capability of acting with myosin This implies that one TN-I molecule exertsits inhibitory action over seven actin monomers occupying adistance of 38.5 nm along the I filament The mechanism ofthis remarkable cooperative property is still uncertain

inter-Tropomyosin and troponin I function

The current dogma of the field is that tropomyosin is thecomponent that blocks the sites on actin involved in inter-action with myosin in resting muscle.Itis postulated that onstimulation the tropomyosin moves, leaving actin free tointeract with myosin, the MgATPase is stimulated and themuscle contracts, the so-called 'steric hypothesis' Tropo-myosin lies in the groove of the actin double helix and isresponsible for the intensification of the 2nd and 3rd layerline reflections of actin in the muscle X-ray diffractionpattern This hypothesis was proposed by Haselgrove [50]and Huxley [51] to explain changes in the layer line re-flections on contraction of frog and toad muscles Thesechanges were interpreted as arising from movement of thetropomyosin molecule in the filament groove from theblocking position to that which permits the MgATPase ofmyosin to be activated by interaction with actin Due to anincorrect assumption about the polarity of the actin filament

in the original study, the position of the tropomyosin in thegroove has been changed [52] Nevertheless the basic tenets

of the hypothesis still stand The hypothesis is supported byelectron microscope image reconstruction studies that showtropomyosin to move when the thin filament is activated bycalcium [53] and the fact that the X-ray diffraction studiesindicate tropomyosin movement can be detected beforetension development in intact muscle

The problem is to demonstrate clearly that in resting muscletropomyosin blocks the site(s) on actin, interaction of which

with an as yet undefined site(s) on myosin leads to activation

ofthe MgATPase.Itis difficult to postulate with any precision

on this matter whilst there is uncertainty about the detailed

nature of the interaction of actin with myosin that is

respon-sible for the contractile process In recent years the concept

of strong and weak binding states corresponding to the

relaxed and contracting states respectively has been proposed

Ifthis actually is the case the steric hypothesis would require

that tropomyosin is involved in the change from weak to the

strong binding state of the actomyosin interaction In its

current form the hypothesis demands that tropomyosin can

interact with actin in a manner that inhibits the actomyosin

MgATPase

There is experimental evidence from enzymic studies

suggesting that this can occur, but the effect very much

depends on the conditions of assay [54] For this reason it is

very difficult to relate the results to events occurring in the

myofibril which is essentially a protein gel system Both

inhibitory and potentiating effects oftropomyosin have been

reported with systems reconstituted from the component

proteins [55] and often with those containing myosin

frag-ments to produce soluble enzyme systems [54, 56, 57] In

addition to the ionic conditions and the relative

concen-trations of the protein components being factors in

de-termining the effect of tropomyosin on the actomyosin

MgATPase, the physical state ofthe actomyosin system may

also be important This was first indicated by Katz [55] who

reported in his study of the superprecitation of actomyosin

that tropomyosin inhibited the clearing (solution) phase but

enhanced the MgATPase during superprecipitation A

signifi-cant observation suggesting that tropomyosin may have a

direct effect on the myosin itself was that treatment of the

myosin with a sulphydryl reagent destroyed the inhibitory

effect on superprecipitation, but left the enhancement of the

ATPase unchanged

Some what different findings have been reported when the

effects oftropomyosin are studied on the enzymic activity of

actomyosin extracted directly from myofibrils This

prepara-tion, 'desensitized actomyosin', from which endogenous

troponin and tropomyosin has been removed may

corre-spond more closely to the in vivo situation At low ionic

strength, under which conditions desensitized actomyosin

has MgATPase activity comparable to that of the intact

myofibrils, no inhibitory activity could be detected with

tropomyosin [58, 59] Some activation ofthe MgATPase was

evident but this was reduced or eliminated ifthe tropomyosin

was further purified On the other hand the CaATPase of

'desensitised actomyosin' was inhibited under these

condi-tions suggesting that when CaATP was the substrate,

tropo-myosin modifies the enzymic process, but not with MgATP

as the substrate This complements the evidence ofKatz [55]

indicating that tropomyosin may have some direct effect on

the enzymic activity of myosin

From the results of the enzymic studies carried out insolution it has been concluded that tropomyosin inhibition

is correlated with complex formation between it and actin [54].This correlation is far from complete, for smooth muscletropomyosin shows maximal potentiation of skeletal acto-myosinATPase at high ionic strength when actin-tropomyosininteraction is considered to be at a maximum [56] To explainthese inconsistencies yet another hypothesis has been in-troduced This postulates that the actin- tropomyosin complexexists in strong and weak myosin-binding states, the pro-portion of which determining the ATPase activity of thesystem under any given conditions

The maximal inhibition obtained with tropomyosin insystems using myosin fragments at higher ionic strengths,under which conditions the intrinsic MgATPase activity islow, about 60% The much higher levels of inhibitionobtained in the presence of TN-I are more comparable tothose that exist in resting muscle This could be explained bythe steric hypothesis on the assumption that TN-I increasesthe binding constant of actin for tropomyosin Itis difficult

to see how this achieves greater inhibition according to thehypothesis ofWilliamset al [56] unless it is considered that

TN-I converts all of the actin-tropomyosin complex into theweak myosin binding state

Even if it could be clearly shown that in resting muscletropomyosin blocks the interaction site on actin there is nodirect information about the nature ofthe process that causestropomyosin to move when muscle is stimulated Onesuggestion has been that the conformational changes occurringwhen TN-C binds calcium are transmitted to tropomyosinthrough TN-lor TN-T, both of which can be demonstrated

to interact with TN-C and tropomyosin Nevertheless it isdifficult to visualize how interaction with the troponincomplex restricted to one region of the tropomyosin, wouldcause sideways movement of the molecule along the whole

of its length The fact that tropomyosin in the absence of TN-C andTN-T can extend the inhibitory activity of one molecule ofTN-I to actin monomers with which it is not in contact,suggests that tropomyosin has an important role in controllingthe relationship between the actin monomers in the filament.The intact molecule ofTN-I is not essential for the inhibitoryactivity in the presence of tropomyosin to be extended overmore than one actin monomer The effect is obtained with theinhibitory peptide representing residues 96-116 of the rabbitfast skeletal isoform [60] Inhibition can also be obtained withresidues 101-115 [61] and a synthetic peptide corresponding

to residues 105-114, which is the minimum length required toproduce this effect [62] The evidence from affinity chroma-tography [63] and from fluorescence studies on pyrene-labelledtropomyosin [64] is that any direct interaction between TN-Iand tropomyosin is weak In the presence of TN-T and itstropomyosin binding fragments, TN-I is bound more strongly

in the ternary complex [64] This presumably reflects the

ability of TN-T to link TN-I to tropomyosin rather than any

change in affinity of the latter protein for tropomyosin

The association of inhibitory activity with small peptides

in the presence of tropomyosin suggests that the interaction

of TN-I at a relatively small region of the actin molecule is

responsible for the cooperative effect There are a number of

possible explanations, for example

(I) Binding at the site induces a conformational change in

the actin monomer which is transmitted to neighbouring

monomers with which TN-I is not bound The conformational

change might be responsible for the following

(a) The affinity of tropomyosin for the site on actin

that interacts with myosin and which is responsible for the

activation of the MgATPase, is increased The result would

be that all the actins in the thin filament are unable to interact

with myosin There are steric difficulties with this

explana-tion, particularly in the relation oftropomyosin to TN-I at the

actin monomer(s) where the troponin is bound All the

evidence suggests that this is at or close to the activation site

on the actin There is also the in vitro enzymic evidence in

which correlation between tropomyosin binding and

in-hibition is not as close as might be expected This model gives

tropomyosin an active role

(b) The conformational change on binding TN-I

transmitted to neighbouring actins with which TN-I is not

directly associated, is such to render them unable to interact

with myosin The tropomyosin in this model does not play

an active role but acts as a kind of template that supports the

actin filament in a manner that permits the conformational

changes induced by TN-I binding to spread to adjacent

monomers It is not unreasonable to expect that if

con-formational change occurs in one monomer in the actin

filament the neighbouring monomers must undergo change

to maintain the symmetry of the I filament structure In this

model the tropomyosin does not have a blocking role but its

movement is purely an adjustment on the surface of the I

filament in response to the conformational changes occurring

in the actin monomers

(2) In this model it is the binding ofthe myosin head to actin,

once the sites on actin are rendered available for myosin

interaction, that induces cooperative activity between actin

monomers in the filaments There is already some evidence for

this type of cooperative behaviour [65] Tropomyosin could

play active or passive roles as outlined above in such a model

Conformational changes that occur in the actin monomers

during the contraction-relaxation cycle do not lead to any

marked changes in the helical parameters of the actin

fila-ments and the subunit repeat remains constant [51] From a

recent reanalysis of the low angle X-ray data it has been

concluded that in addition to the tropomyosin movement

there are small but plausible actin subdomain movements

The data cannot be explained by a tropomyosin shift on its

15own [66] Nevertheless the fact that there is evidence thatconformational changes occur in actin indicates that thesetake place without disrupting the periodicity of the doublehelical structure Presumably this is stabilised by the tropo-myosin molecules in the large grooves ofthe filament ProtonNMR studies indicate that the binding of TN-I to the N-terminal sites of actin produce changes at the C-terminuswhere it is not bound As a result of these changes the alkalilight chain (ALC I) of myosin no longer binds to the C-terminal region of actin [67] From the analysis of threedimensional images reconstructed from cryo-electron micro-graphs Ishikawa and Wakabayashi [68] have reported changes

in the structure ofreconstituted actin filaments in the presence

of calcium ions Although the results of this study aresuggestive of conformational changes in the actin filamentthe authors were not able to conclude whether they were due

to tropomyosin movement or conformational change in theactin itself

The interaction of troponin I with actin

The fact that the inhibitory peptide, residues 96-116, is theonly fragment ofTN-I present in the cyanogen bromide digestwith inhibitory activity emphasises the special significance

of this region for interaction with actin Its propertiesresemble those of the intact molecule in that in addition toits inhibitory activity being enhanced by tropomyosin, itseffect is neutralised by troponin C Indeed cardiac fibrebundles reconstituted with the cardiac inhibitory peptide areable to undergo sequential contraction- relaxation cycles [69]

On a molar basis it, and the shorter synthetic duodecapeptide,residues 104-115, is 45-70% as effective as TN-I [60,62].The shorter sequence represents less than half of the highlyconserved region of rabbit fast skeletal muscle TN-I, whichextends from residues 97-121 (Fig I) The correspondingminimum length inhibitory peptide from rabbit cardiac TN-

I is slightly less effective as an inhibitor of the actomyosinATPase than the skeletal peptide It differs from the latter inthat pro II 0 is replaced by threonine and arg 113 by leucine.Studies with hybrid peptides indicate that substitution ofresidue 110 had little effect on activity and the substitutionofarg 113 was probably largely responsible for the difference

in activity of the two forms [70] The importance of thearginine residues, which are principally located in theC-terminal moiety of the inhibitory peptide, for the inter-action with actin is indicated by proton NMR studies of thebinding of the inhibitory peptide to defined cleavage frag-ments of actin [71] The inhibitory region represented byresidues 96-116 in the rabbit fast skeletal isoform is stronglyconserved in vertebrates with an identity score of85% In thecorresponding segment of crayfish TN-I only 57% of theresidues are identical or functionally conserved [35]

(1021 20

, 30

40 BINDS

L- 80 90

Two N-terminal regions of actin, residues1-7and23-27,

were identified as interaction sites with the inhibitory peptide

[72] Similar regions have been identified for interaction of

the inhibitory peptide from cardiac TN-l with actin (J.P

Trayer, personal communication) Evidence for aN-terminal

TN-I binding site on actin has also been obtained by linking studies using the 'zero-length' carbodiimide reagentthat is specific for lysine-carboxylate contacts [73] Bycomparison of peptide patterns after proteolysis in theabsence and presence ofthe cross linker it was concluded that

~/////)/

Fig. 3 Scheme indicating the interactions of TN-I with TN-C and actin/

tropomyosin as proposed by Tripet et al [81] Regions involved in

interactions are indicated by arrows The shaded regions of TN-I are those essential for full inhibitory activity Hatched regions are those presumed to interact with TN-C only Figures indicate residue numbers in the TN-I polypeptide chain.

Actin-Tm

TN-I with the inhibitory region, residues 96-116, deleted iscompletely inactive as an inhibitor [79] Another mutant withresidues 105-115 deleted is partially active as an inhibitorsuggesting that the whole ofthe inhibitory peptide region andnot simply the minimal inhibitory peptide region, residues104-115, is required for function [79] The minimum sequence

to obtain inhibitory activity comparable to the intact rabbitfast skeletal TN-I molecule is residues 98-148 [80, 81].Removal of residues 140-148 reduces the inhibitory action

to that ofthe inhibitory peptide This observation, and the factthat a synthetic peptide corresponding to residues 128-148bound to the actin-tropomyosin filament and induced a weakinhibitory activity, has lead to the suggestion that the regionconsisting of residues 140-148 is a second actin binding site(Fig 3) [81]

Similar findings about the role of the C-terminal part ofthe molecule in the inhibitory activity of TN-I have beenobtained with the cardiac isoform [82] Using mutants ofmouse cardiac TN-I (211 residues) it has been concluded thatresidues 152-199 are required to obtain inhibition equal tothat of the wild type protein These correspond to residues120-167 of rabbit fast skeletal TN-I and it is suggested thattheir results can be interpreted to imply that there are two actinbinding sites in the corresponding region of cardiac TN-I.Until the nature ofthe interaction in this region, which is verysimilar in amino acid sequence in the skeletal and cardiacisoforms, can be defined precisely in terms ofthe amino acidresidues involved, e.g by NMR studies, judgement on thisinterpretation must be reserved

the region represented by residues 1-12 was cross linked to

TN-I Positions 98, 105 and 107 are occupied by lysine

residues in the inhibitory peptide In view ofthe fact that the

minimum peptide required for interaction with actin

corre-sponds to residues 105-114 and in the light of the NMR

results [72], it is likely that lysines 105 or 107 are involved

in the cross linking (Fig 2)

Thus the evidence suggests that the regions represented by

residues 104-115 of the TN-I molecule and the N-terminus

of actin are involved in the interaction between the two

proteins that is functionally significant This interaction is

essential for inhibition and in initiating events that lead to

amplification of the effect by tropomyosin In the myofibril

at the moment of stimulation induced by the rise in calcium

concentration some movement of the TN-I must take place

to enable the actin to activate the actomyosin MgATPase

Resonance energy transfer measurements on addition ofCa2+

to reconstituted thin filaments indicate that both the

N-terminal (cys48 [m]) and the C-N-terminal (cysI33) regions of

TN-I move away from the C-terminus of actin (cys374)

towards TN-C The results suggest that the movement ofthe

C-terminal of TN-I, about 15A,is more extensive than that

of the N-terminal region [74, 75] Similar movement is

observed when the reconstituted filament system is activated

by binding myosin S I to the filament in the absence of Ca2+.

Somewhat less movement is observed in the presence ofCa2+

[76] In association with these changes, when the low affinity

calcium binding sites of TN-C are occupied, cys98 of the

latter protein moves closer to the inhibitory peptide [77]

From binding and ATPase studies on the effects of TN-l

on the actomyosin - S I ATPase in the presence of the

regulatory proteins it has been confirmed that the TN-I

inhibitory activity is independent ofthe presence ofTN-C or

TN-T [78] Although Ca2+dissociates intrinsic TN-l from an

actin-tropomyosin site it remains bound to the complete I

filament system suggesting that another region in addition to

the inhibitory peptide region is involved in actin binding The

inhibitory peptide, residues 98-116, is able to maximally

inhibit to the same extent as the intact molecule but is about

50% as effective as equimolar amounts of intact TN-I in

producing 50% inhibition ofthe actomyosin MgATPase [60]

This implies that another region of the molecule that may

interact with actin is required for full inhibitory activity

Proton NMR studies have so far been largely restricted to the

interaction of the inhibitory peptide with the N-terminus of

actin and less is known about other regions of the molecule

that may be involved

Because the sequence of residues 121-146 of rabbit fast

TN-I has some common features with residues 108 115 of

the inhibitory region, it has been suspected that regions

C-terminal to the inhibitory peptide may be involved in actin

interaction Additional interaction in this region would be

expected to have a modulating role, for a mutant of skeletal

Interaction of troponin I with troponin C

The properties and functions ofTN-I and TN-C are relatively

well defined compared with those of the third component of

the troponin complex, TN-T The fact that TN-I has an

inhibitory property that is neutralised by TN-C implies that

they interact together and this interaction has a central role

in the regulatory process The stability ofthis complex in the

presence ofcalcium is clearly illustrated by the demonstration

that when the two proteins are present in equimolar amounts

it migrates as a single band on electrophoresis in 6-8 M urea

[12,83] In the presence of EGTA the complex dissociates

and the TN-C migrates as a single band ofdifferent mobility

The stability of this complex and its calcium dependence

enabled the development ofa simple method for the isolation

oftroponin-I in one stage from whole muscle homogenates

using a TN-C affinity column [16] On application of this

technique to the cyanogen bromide digest of rabbit fast

skeletal muscle TN-I only peptides consisting ofresidues

1-21,1-47 and 96-116 are bound to theTN-C affinity column

[60] The implication that these two regions are involved in

TN-C binding and are important for the function of the

complex is supported by the fact that they contain the two

regions of troponin I that are strongly conserved in all

vertebrate isoforms Subsequent investigations by a number

ofworkers have supported these findings (for reviews see [44,

84,85])

Inhibitory peptide region

The fact that TN-C neutralises inhibition by residues

105-114 also indicates that it will interact with this region to

displace the actin from TN-I so that it can activate the myosin

MgATPase TN-C interacts with lysine, leucine and

phenyl-alanine, residues that are more abundant in the N-terminal

region of the inhibitory peptide [71] The NMR evidence

indicates that TN-C interacts with both N- and C-terminal

parts ofthe inhibitory peptide region ofTN-I in the presence

of calcium Cross linking studies confirm that this is the case

for they demonstrate that in the TN-I/TN-C complex the

central helical region and both the Nand C-terminal domains

ofTN-C all interact with the inhibitory region of TN-I (Fig

2) The interaction appears to be antiparallel in so far as TN-C

cys98 cross links with residues 103-111 [86] and cys57 in

TN-C cross links to residues 113-121 in TN-I [87] Cross

linking experiments with mutants in which single cysteine

residues are inserted at positions 6, 48, 89, 104, 133 or 179

are also consistent with an antiparallel arrangement of the

polypeptide chains of the two proteins [88] Gly I04 of the

inhibitory peptide also cross links with metl55 in a region

consisting of residues 154-159 ofTN-C, at the end of helix

H ([89], Fig 2)

Two dimensional NMR study of the interaction of theC-terminal part of the inhibitory peptide using a syntheticpeptide, N-acetyl TN-I (104-115) amide has enabled theconformational changes that occur when it complexes withtroponin C to be described [46] The sequence of this part ofthe inhibitory peptide is somewhat unusual in that it consists

of six basic residues alternating with hydrophobic residues,interrupted in the centre by two adjacent proline residues.It

is concluded from the data that when bound the peptide forms

an helical amphiphilic structure bent round the central prolineresidues so that the hydrophobic residues are brought closertogether to form a hydrophobic face Campbell and Sykes[46] suggest this interacts with an exposed hydrophobicregion of troponin C The interaction of the complete in-hibitory peptide with TN-C must be more extensive than thisfor earlier NMR studies [71] indicated that its N-terminalresidues were also perturbed in the presence ofTN-C Theseresidues are presumably less important in actin binding forthe shorter peptide acts as an inhibitor of the actomyosinATPase that is neutralised byTN-C Nevertheless in view of

the studies of Zang et at[79] with mutants they are of somesignificance for inhibitory function

The inhibitory peptide region in isolated TN-I must beexposed to solvent for it is readily susceptible to limitedproteolysis by chymotrypsin, cleavage occurring at asp I0 I

or Iys I07 [90] Despite the evidence for the involvement ofthis region in interaction with TN-C it is equally susceptible

to chymotryptic attack in the TN-I/TN-C complex

Comparison between the skeletal and cardiac systems hasbeen made using the 104-115 residue peptide in whichproline 110 is replaced with glycine as an analogue of thecardiac inhibitory peptide [47] In rabbit cardiac TN-I prollO

is replaced by threonine Itis concluded that on interactionwith bovine cardiac TN-C the peptide analogue undergoessimilar conformational changes to those occurring in theskeletal system, but the cardiac peptide appears more flexibleabout the glycine residue Itis a little difficult to be certainwhether this is a real difference between the fast skeletal andcardiac systems or simply due to the fact that comparison hasnot been made with strictly homologous peptides

The interaction ofTN-I or the inhibitory peptide with actin

is independent of calcium whereas that with TN-C is muchstrengthened in the presence of this cation With isolatedproteins TN-C can neutralise the inhibitory activity of TN-I

in the presence of EGTA [91] indicating that calcium is notessential for the interaction In other words the interactionbetween the two proteins is sufficiently strong in the absence

of calcium to neutralise inhibitory activity Neverthelessproton NMR investigations indicate that TN-C interactionwith the inhibitory peptide is less strong than with the intactprotein for it is modulated by calcium binding at con-centrations that would suggest the high affinity sites ofTN-Care filled [71]

N-terminal region oftroponin I

In the original studies on the digests of fast skeletal TN-I

obtained by a variety of specific cleavage methods the

terminal peptide representing residues 1-47 was the most

strongly bound to the TN-C affinity columns A shorter

peptide obtained by cyanogen bromide digestion consisting

ofresidues 1-21 was also bound, but less strongly [60] These

observations indicate that in addition to the centrally located

inhibitory peptide sequence, the N-terminus ofTN-I interacts

strongly with TN-C Further substantiation to this view was

provided by the finding that two cyanogen bromide fragments

representing different regions of rabbit fast TN-C form

calcium dependent complexes with TN-I [92] One of these

consisting of residues 83-134 was shown to neutralise the

inhibitory activity and inhibit the phosphorylation of ser117

ofTN-I much more effectively than thrll Clearly it interacts

with the inhibitory region Leavis et al [93] employing a wide

range of specific cleavage fragments also came to the

conclusion that there were two sites of attachment between

TN-C and TN-I

Recent photocrosslinking studies with single cysteine

mutants have identified a specific cross link between

TN-C 158 and met21 of TN-I and a range of crosslinks between

TN-C21 and residues 9frl31 of TN-I [93a] Proton NMR

studies [71] of the interaction of peptides representing

residues 1-21 and 9frl16 of TN-I with TN-C indicate they

do not compete on binding implying that separate sites are

involved and are close together By applying a spin label to

cysteine 98 ofTN-C it was concluded that the sites were both

within 15 Aof cys98 In both cases the interactions were

calcium sensitive although dissociation at the N-terminal site

ofTN-I occurred at higher calcium concentrations than were

required for dissociation at the inhibitory peptide site

Although Katayama and Nozaki [94] agree on the basis of

electrophoresis studies that the binding of TN-I fragment

consisting of residues 1-21 to TN-C is calcium dependent

they conclude that their inhibitory fragment, residues

lOl-llS, is not This is surprising for the NMR data indicates that

the slightly longer intact inhibitory peptide, consisting of

residues 9fr116, does require a trace of calcium, sufficient

to saturate the high affinity sites, to bind to TN-C [71].Itis

reported that the mutant of mouse cardiac TI with 53

N-terminal residues deleted does not bind to cardiac TN-C or

restore calcium activation to the myofibrillar ATPase

Somewhat surprisingly the mutant with additional N-terminal

residues deleted, TN-IsQ-211' binds weakly to TN-C and

partially restores calcium activation [94a]

Although the function ofthe inhibitory region ofTN-I has

been clear for some time that ofthe other calcium -dependent

binding site at the N-terminus has received less attention The

evidence with peptide fragments indicates that its presence

is not essential for inhibition Recent studies on mutant forms

19

of human cardiac TN-I have confirmed that TN-C interactswith the homologous N-terminal site on this protein [95] Thestrength ofTN-C binding to the N-terminus ofTN-I has beenconfirmed by Ngai and Hodges [96] who have reported that

a synthetic peptide consisting of residues 1-40 of TN-I candispiace TN-I from a preformed TN-I1TN-C complex.Italsoprevents TN-C from neutralising the inhibitory activity oftheinhibitory peptide on the MgATPase of actomyosin In thelight ofthese results it has been suggested that the N-terminalregion of TN-I plays an important role in modulating thecalcium sensitive control ofthe actomyosin MgATPase Iftheresults obtained with the fragments and synthetic peptides can

be applied to the intact proteins, binding of TN-C to theN-terminus of TN-I would be expected to occur later as thecalcium concentration rises on stimulation Ngai and Hodgesresults imply that this should lead to detachment of TN-Cfrom the inhibitory region and inhibition of the ATPase, iespeed up relaxation Nevertheless the N-terminus does not

appear to be essential for function, at least in in vitro, systems.

A mutant in which the first 57 N-terminal residues are deleted[97] has inhibitory properties similar to wild type TN-I andretains calcium dependent interaction with TN-C

From an extensive study with deletion mutants Farah et al.

[98] conclude that the N-terminal region does not have aregulatory role but ascribe such a function to the C-terminalregion of the molecule They consider that residues 1-102

of TN-I playa structural role in the troponin complex, beingresponsible for binding to the C-terminal domain of TN-Cand for stabilising the incorporation ofTN-T into the ternarytroponin complex The mutant TN-I10J_1R2 despite containingonly 44% of the residues of the intact molecule and only part

of the TN-C binding region of the inhibitory peptide region,inhibits the actomyosin ATPase as well as the intact molecule

Itinteracts with the N- and C-terminal domains ofTN-C in acalcium dependent manner and can be reconstituted into afunctional calcium sensitive filament

C-terminal region oftroponin 1.

No evidence was obtained by affinity chromatography forTN-C binding to specific cleavage fragments of TN-I pro-duced from the region C-terminal to the inhibitory peptide,one ofwhich corresponded to the C-terminal46 residues [60].Nevertheless the sequence in regions of the C-terminalportion of TN-I is strongly conserved between isoforms,suggesting this region is of functional importance The use

of deletion mutants of TN-I has thrown new light on thepossible role of the C-terminal domain [98] TN-I I2Q-1S2' aswould be expected, had no significant inhibitory or activatingeffect on actomyosinATPase whereas the inhibition obtainedwith TN-I10J-1S2 was 80% relieved by TN-C Mutants in whichparts ofthe C-terminal region are removed possess inhibitory

activity which is neutralised when complexed with TN-C in

the presence or absence of calcium, i.e the system is not

calcium sensitive This suggests that the C-terminal regions

have a role in determining the calcium sensitivity ofthe

TN-I/TN-C complex These results must be interpreted in the light

ofthe observation that neutralisation ofthe inhibitory activity

of isolated rabbit fast skeletal TN-I with its homologous

TN-C is also virtually complete at equimolar ratios whether

calcium is present or not [91] Farah et al [98] conclude that

the COOH domain ofTN-I has a regulatory role and interacts

with the N-terminal domain ofTN-C

The region in the vicinity of cysl33 of TN-I appears to be

of significance for the function of the TN-I/TN-C complex

This residue is located at about 16 residues on the COOH side

of the strongly conserved inhibitory region occupied by actin

and/or TN-C The phosphorylation of serl17 in rabbit fast

skeletal TN-I is blocked by TN-C This fits in well with the

proposal for a 'second TN-C binding site' in the regions of

residues 115-131 [81] Nevertheless the adjacent cys 133

appears to be exposed on the surface of TN-I when it is

complexed with other components of the troponin complex

[99] which suggests that the binding ofTN-C in the region of

serl17 does not extend up to cys 133 The conclusion that there

is a binding site in this region is supported by fluorescent

energy transfer experiments [100, 101] indicating that when

TN-I binds to TN-C the distance between cys98 ofTN-C and

cys 133 oftroponin I decreases In the complex the N-terminal

of TN-C is also close to this region for a mutant in which

residue 12 ofTN-C is replaced by cysteine a cross link between

this residue occurs at or near metl 34 ofTN-1 [102] With

aTN-C mutant in which residue 98 is replaced by leucine and residue

89 by cysteine, cross-linking occurred with residues in the

region of 108-113 of troponin I These results are further

evi-dence of an antiparallel arrangement of the peptide chains of

the two proteins in the region ofthe inhibitory peptide (Fig 2)

From the results of a number of investigations involving

binding, enzymic and crosslinking studies it is emerging that

the region ofTN-I represented by residues 96-131 is involved

in binding to the Nand C terminal domains of TN-C ([45,

81,89, 93a, 98] Fig 3) An additional site(s) for binding actin

has been proposed on the C-terminal side of this site (see

section on actin binding)

The C-terminal region of cardiac TN-I is essential for the

full development of Ca2+-sensitive force in reconstituted

skinned cardiac fibre bundles Replacement ofwild type

TN-I with the mutant TN-TN-I1-151reduced force development by two

thirds [103]

Troponin I-troponin C complex

There is much to be learnt about the nature of the calcium

regulation of the inhibition by TN-I in the troponin system

The interaction ofTN-C with TN-I is clearly complex and atthe heart of the regulatory process Even in the absence ofcalcium the interaction is strong enough in the isolated proteinsystem for the inhibitory activity of TN-I to be neutralised.Values for the equilibrium binding constants vary [104-107]but generally accepted values would be approximately 106

M-I and 109M-I in the absence and presence of calciumrespectively [84] This approximately 1000 fold increase inaffinity on binding calcium is responsible for initiatingchanges that transform the muscle from the resting to thecontracted state The binding of TN-I also increases theaffinity for calcium of both the high affinity and low affinitydomains ofTN-C [106-108]

The N-terminal and inhibitory regions of skeletal TN-I arefairly well defined as sites of interaction with TN-C but thenature of the interaction and location of the interactionC-terminal to the inhibitory peptide region is less welldefined Recently NMR spectroscopy studies with the peptiderepresenting residues 115-131 of skeletal TN-I indicate thatthis region binds at the N-terminal hydrophobic pocket ofTN-

C [I 08a] This conclusion is supported by studies withcysteine mutuants of the skeletal protein demonstratingcrosslinking of met 121 with the hydrophobic region [1 08b].The corresponding region of human cardiac TN-I [residues148-164] also binds to the regulatory domain ofcardiac TN-

C but there are differences from the skeletal system in theconformational changes induced and the affinity of theinteraction [1 08c, 108d] Much effort has gone into definingthe complementary sites on TN-C (for reviews see [44, 84,98]) The interaction interface of TN-C would appear to beextensive as judged by the regions in the primary sequencethat have been reported to be involved in interacting with TN-

I (Fig 2) This may be more apparent than real for despitethe availability ofthe crystal structure ofTN-C it is uncertainifthe molecule in the troponin complex is as extended as thecrystallographic evidence would suggest Some information

on this point has been provided by small angle X-ray andneutron scattering This data can be interpreted by anextended 'dumbbell like' structure for the skeletal TN-I/TN-Ccomplex In this structure the TN-I winds round the extendedTN-C and makes contact with the hydrophobic patchesconsidered to be present in the Nand C domains ofthe TN-C[109, 110] This model is supported by the results of floures-cence lifetime, acrylamide quenching and photocrosslinkingstudies [111].Itmight be expected with such a model that thedistance between the Nand C terminal domains of TN-I,about 40A,would change when it complexed with TN-C.This has been shown not to be the case using fluorescentprobes on cys48 and cys133 ofTN-I In the presence ofTN-Tand in the reconstituted thin filament this distance increased

to about 50A. Removal of calcium from the thin filamentsystem caused a further increase to about 60A[112] This

fits in well with the observations of Stone et al [49] that

TN-I in the troponin complex is less elongated when calcium is

bound to TN-C

In contrast to the above evidence suggesting that the

TN-C molecule is in an extended form in theTN-C/TN-I

com-plex, a recent crystal structure ofTN-C (two Ca2+bound state)

complexed with the 47 residue N-terminal fragment of

TN-I indicates that the TN-C is in a compact globular form [48a]

The amphiphilic C terminal end of this a-helical peptide is

bound in the hydrophobic pocket ofthe N-terminal regulatory

domain ofTN-C.Itshould be noted that there is evidence with

isolated peptides that a region C-terminal to the inhibitory

peptide of TN-I also binds to this hydrophobic pocket [81,

108a, 108c] Itremains to be demonstrated whether both

regions of TN-I can interact with this pocket when the

complex between the intact proteins is formed.Itis possible

that some isolated peptides from TN-I can interact with this

hydrophobic pocket on TN-C in a non-specific manner

There are two fairly well defined sites on TN-C involved

in interaction with the inhibitory region of TN-I The first is

the region of the E helix ofTN-C (residues 93-103), which

has been shown to be implicated by NMR [71] and cross

linking studies involving residues 89 and 98 [102, 113, 114]

The second involves the N-terminal calcium binding domains

Cross links have been demonstrated with the C helix in

calcium binding site II ofTN-C [115] and with the region of

alanine 57 ofTN-C [87, 116, 117] This latter interaction fits

in well with that suggested by Herzberget ai.[118] in their

model for the mechanism of action of TN-C Itis proposed

that on binding calcium at sites I and II the B/C pair ofhelices

move away from the AIDpair to expose a patch of

hydro-phobic residues that becomes a binding site for TN-I

Hydro-phobic side chains are known to be involved in the interaction

ofthe inhibitory peptide with TN-C [71, 119] as are the acidic

side chains of glu84 and asp85 of TN-C [120] Despite the

evidence for a range of groups being involved in the TN-I

interaction, only a sub-region ofthe TN-C N-terminal domain

would appear to be sensitive to bulky covalent adducts

Covalently linked peptide or biotin at residues 45, 81, 84, and

85 ofcardiac C only had a major effect on the transmission of

the calcium signal, as measured by the ATPase level of the

activated myofibril, in the case of residue 81 [121]

The use of mutant forms, and synthetic peptides

corre-sponding to fragments of TN-I, in reconstituted regulated

actomyosin and skinned muscle fibre systems has enable the

assignment of distinct functions to different regions of the

molecule By combining the biochemical and biophysical

evidence of the sites of interaction with the results of studies

on the properties of deletion mutants of TN-I Farahet ai [98]

have proposed a scheme for the interaction ofthe two proteins

In this scheme the three domains of each protein are aligned

in an antiparallel manner with direct interaction between the

Nand C-terminals of both proteins.Itis clear that the central

region of the TN-I molecule, residues 96-148 in the rabbit

21fast skeletal isoform, is ofparticular importance for its function.The interactions ofthis region with other proteins ofthe myo-fibril are illustrated in the schemeofTripetet ai.(Fig 3) Thisregion must be intact for full inhibitory activity and is essentialfor Ca2+sensitivity of the actomyosin MgATPase [80, 82].Although the region ofthe molecule N-terminal to residue 96

is not essential for inhibitory activity it is required for maximumATPase activity in the regulated actomyosin sytem [80]

Interaction of troponin I with troponin T

Ithas long been known that TN-T is an essential component

of the troponin complex for the inhibitory action of TN-I to

be regulated by calcium, but the precise relationship betweenthe two proteins is not as well defined as that between TN-Iand TN-C (see Perry [12Ia] for review) The fact that in theearly studies on TN-I preparations often contained TN-T [10]suggested that there might be some interaction between thetwo proteins The affinity ofTN-I for TN-T, however, wouldappear to be rather low for interaction has not been demon-strated under the conditions of electrophoresis as is the casewith the complexes of TN-C with TN-I and TN-T Thetendency to copurify may also reflect the similar physicalproperties of the two proteins, as both, unlike TN-C, arestrongly basic with high isoelectric points

Complexing TN-T with TN-I produces changes in thereactivities of TN-I Iysines in the region of residues 40-98[122] These changes were similar to those observed whenthe reactivities ofthe Iysines in isolated TN-I were comparedwith those ofTN-I in the troponin complex.Itwas concludedthat this region ofTN-I is involved in interaction with TN-T

The binding of calcium to the complex produced changes inthe reactivities of some Iysines in this region suggesting thatthe TN-I1TN-T interaction is modified when TN-C bindscalcium Further evidence of TN-T interaction at this region

is that cysteine residues 48 and 64, which are accessible toacetamide labelling in TN-I and the TN-I/TN-C complex, arenot accessible in the TN-I/TN-T complex or in wholetroponin [99] The C-terminal peptide ofTN-T, consisting ofresidues 159-259, has been shown to be the region involved

in interaction with TN-I,23_,26' In the light of the fact that thereactivity of Iysines 223 and 226 of TN-T were reduced inthe TN-I/TN-T complex it has been suggested that the region

of residues 223-227 of TN-T is directly involved in theinteraction with TN-I The report that the deletion mutant TN-

TI_201 does not bind TN-I supports this conclusion [127].Earlier conclusions about the involvement of the regionrepresented by residues 40-98 of TN-I in interaction withTN-T is compatible with more recent studies with deletionmutants of the former protein [98, 127] From these it isconcluded that the N-terminal consisting ofresidues 1-98 isnecessary for the incorporation of TN-T in the complex but

is not necessary for the calcium regulation of the inhibitory

activity ofTN-I in a reconstituted filament This observation

with the skeletal isoform correlates with recent studies on the

cardiac protein which has about 30 additional residues at the

N-terminus Whereas mutant mouse cardiac TN-I with residues

I-53 deleted binds to cardiacTN-T, with residues 1-79 deleted

it does not [94a] The TN-T binding region of skeletal TN-I

probably does not extend as far as residue 98 for a deletion

mutant missing residues I-57 is unable to bind TN-T [128]

Analysis of the sequences of TN-I and TN-T from widely

different species confirms the conclusions about the

inter-action sites on the two proteins made by more direct studies

Ithas been proposed that the evolutionary conserved heptad

repeat motif with the potential for a-helical coiled coil

formation that is observed in the amino acid sequences

ofTN-I and TN-T represents regions where the proteins interact

[128a] In the case of human fast skeletal muscle similarities

in sequence exist between residues 20-110 of TN-I and

residues 160-240 ofTN-T.Both regions contain heptad repeats

that could be involved in a-helical coiled coil formation

The study ofthe properties ofthe isolated proteins suggests

that an important role ofTN-T is to change the TN-C/TN-I

complex from a calcium insensitive to a calcium sensitive

form in the MgATPase system Itcould do this by reducing

the binding constant for the TN-I/TN-C complex in the

absence of calcium The results obtained with mutant forms

of TN-I, however, exclude TN-T having this effect by

interaction in the region of TN-I represented by residues

1-98 This mutant exhibits normal calcium sensitivity when

incorporated into a reconstituted thin filament [98] Other

mutant studies with the N-terminal region of TN-I deleted

also emphasise the importance of this region and its

inter-action with TN-T for the calcium sensitivity of the troponin

regulated MgATPase of actomyosin [121 a, 129]

Malnic et al [129a] have confirmed the interaction of the

C-terminal region ofTT (residues 216-263) with the

N-terminal region ofTN-I (residue 1-98) In their revised model

for the calcium switch of the troponin complex in the thin

filament they postulate that TN-T interacts with the N-terminal

region of TN-I and not directly with TN-C Itis difficult to

reconcile this model with the convincing evidence that TN-C

interacts with the C-terminal region ofTN-T (for a review see

[12Ia]) and studies that have shown that TN-I with the first

N-terminal 57 residues deleted can function in the troponin

complex with the calcium regulation unchanged [128]

Phosphorylation of troponin I

Skeletal muscle

The original report [130] that the troponin complex could be

phosphorylated by protein kinase A was followed up by a

number of workers [131-133] Both phosphorylase andprotein A kinases were used as phosphorylating agents and

as the composition of the troponin complex was not welldefined at the time there was some confusion as to whichprotein component was the major target In due course itbecame clear that both enzymes phosphorylated TN-I andonly phosphorylase kinase phosphorylated TN-Tat a signi-ficant rate [132-134] With the isolated protein thrll ofrabbitfast skeletal TN-I is the major site for phosphorylase kinase,with serll7 phosphorylated much more slowly In contrastphosphorylation by protein kinase A is largely restricted toserll7 [135, 136] These sites would appear to be partially

phosphorylated in vivo for when TN-I is isolated from rabbit

fast skeletal muscle under conditions designed to minimiseendogenous enzymic activity it usually contains about 0.5mole phosphate per mole of protein This phosphate doesnot appear to be in rapid equilibrium with the intracellular

pools for Ribulow et al [137] were unable demonstrate

incorporation of32P in TN-I as a consequence of contractileactivity in stimulated frog muscle The functional signi-ficance of phosphorylation at these sites is not clear butserine or threonine residues are present in homologouspositions in the sequences of all the isoforms of TN-I TheN-terminal site is close to a site of interaction with TN-Cand the other is adjacent to the inhibitory peptide regionwhere both actin and TN-C are known to interact In view ofthe presence of charged residues in the interaction sites ofboth proteins it is likely that electrostatic forces playa part

in these interactions The introduction of negatively chargedphosphate residues at or close to the interaction sites would

be expected to affect the binding constants In this respect it

is of interest that if ser43/ser45 of mouse cardiac TN-I aremutated to alanine the calcium sensitivity and Ca2+activatedMgATPase are both reduced [138] In the presence ofTN-Cthe phosphorylation of fast skeletal TN-I is markedlyinhibited suggesting that on complex formation the phos-phorylation sites are blocked Hydroxy amino acids that arepotential targets for protein kinases are present in homo-logous positions in skeletal and cardiac TN-I isoforms Inthe case of the mouse cardiac TN-I ser43/ser45 and thrl44are in homologous positions to thrll and serl16 ofthe rabbitfast skeletal isoform respectively The mouse cardiac siteshave been shown be phosphorylated by protein kinase C[138]

The TN-C peptide consisting of residues 83-134 alsoinhibits phosphorylation of serine 117 [92] These observa-tions fit in well with current views on the regions of TN-Iconsidered to be involved in interaction with TN-C (seeearlier section and Fig 2) The lack of incorporation of 32pinto TN-I of intact contracting skeletal muscle indicates that

in the myofibril the sites of phosphorylation are not available

to the endogenous kinases presumably because they areblocked by other components of the troponin complex

Cardiac muscle

N-terminai site

Following the report [139] that cardiac troponin was

phos-phorylated by endogenous protein kinase A it was shown that

rabbit cardiac troponin was phosphorylated 5-10 times more

rapidly than the skeletal complex and that virtually all the

phosphate was incorporated in TN-I [134, 140] Rabbit

cardiac TN-I differs from its skeletal isoform in that when

isolated by affinity chromatography under conditions that

inhibit endogenous enzymes, its phosphate content is much

higher The amount depends on the conditions of slaughter

and varies from about 1-2 moles per mole of protein

The rate of phosphorylation of the isolated protein by

protein kinase A is about 30 times faster than that of the

skeletal isoform [141] Also it is not significantly inhibited

by C On the other hand phosphorylation of cardiac

TN-I by phosphorylase kinase is strongly inhibited by TN-C This

suggests that the site homologous with thrll in fast skeletal

muscle, which is the preferred site for phosphorylase kinase,

is located in a TN-C binding site on cardiac troponinI.Thus

it would appear that unlike the sites close to the TN-C

interaction sites, the preferred phosphorylation site on cardiac

TN-I for protein kinase A, is freely available to enzymes even

when incorporated in the troponin complex

The fact that the phosphate content depends on the

con-ditions of isolation, being lowest when the rabbit was

slaughtered under anaesthesia [141], suggested that the

phosphorylation state might be related to the physiological

state of the heart Further evidence for such an association

was provided by the correlation between the contractile

force developed in the perfused rat heart on application of

adrenaline and the extent of TN-I phosphorylation [142]

Similar observations were made with the perfused rabbit

heart in which the phosphorylation site was found to be

located in the N-terminal cyanogen bromide fragment

comprising residues 1-48 [143] In contrast to an earlier

claim that phosphorylation increased the sensitivity of the

myofibrillar ATPase to calcium [144] it was later shown to

cause a decrease in calcium sensitivity i.e the calcium

concentration required for 50% activation was increased

[143, 145] These findings have been confirmed and

ex-tended by many workers (for review see [85]) The change

in calcium sensitivity is independent of the TN-C isoform

for it can be obtained in skinned cardiotrabeculae with

cardiac or skeletal TN-C [146] Surprisingly, in the light of

these findings, phosphorylation of recombinant human

cardiac TN-I introduced into rat and rabbit skinned soleus

fibres did not result in a change in the calcium sensitivity

of force production [147] If both of these findings are

confirmed this would suggest another cardiac specific

myofibrillar protein(s) other than TN-C, possibly TN-T, is

essential for the N-terminal phosphorylation of cardiac

in response to the rise in the calcium transients induced byintervention with adrenaline By reducing the sensitivity ofthe myofibrillarATPase to calcium it has been suggested thatphosphorylation of cardiac TN-I contributes to the speeding

up of relaxation that is associated with adrenaline vention There is no direct experimental evidence for this,indeed recent studies employing laser flash photolysis inskinned guinea pig trabeculae suggest no change in relaxationrate occurs on phosphorylation of the TN-I [148] Experi-ments with phospholamban knock-out mice also indicate thatphosphorylation of cardiac TN-I plays a minor role in PKAmediated accelerated relaxation [148a]

inter-The major part ofthe amino acid sequence of cardiac

TN-I shows strong homology with the skeletal isoforms butdiffers from them in that the N-terminus is extended by about

30 residues [43] This additional N-terminal sequence isstrongly conserved between species and contains the site that

is further phosphorylated on adrenaline intervention in vivo When rabbit cardiac TN-I is phosphorylated in vitro with

protein kinase A the majority of phosphate is incorporatedinto the N-terminal site with a few percent ofthe total at serine

146 which is homologous with the protein kinase A specificsite of rabbit fast skeletal TN-I [149]

The N-terminal site was originally considered to correspond

to serine 20 from the sequence studies that were available atthe time [43].Itis now known from the work of Mittman et

ai.[150] that there were errors in the original sequence and thatthe phosphorylation site consists oftwo residues, serine 22 and

23 in the rabbit, both of which can be phosphorylated This

report explained the inability of Moir et ai [151] to identify a

second site ofphosphorylation in the perfused rabbit heart afteradrenaline intervention In these studies despite the fact thatvirtually all the phosphate was located in the N-terminal peptideand associated with 'serine 20' the value rose to approximately

2 mole per mole after adrenaline intervention

Ithas been long known that interaction ofTN-I with TN-Cmarkedly increases the affinity of the latter protein forcalcium.Itis therefore not surprising that covalent modifi-cation of TN-I, particularly with highly charged phosphategroups should modify the effects of the interaction In thisrespect it may be significant that cardiac TN-C differs fromthe fast skeletal isoform in possessing only one calciumspecific binding site for regulating the contractile response.TN-I isolated under conditions to restrict endogenous enzymeactivity from the normal beating heart in the anaesthetisedrabbit usually contains about one mole ofphosphate per moleofprotein, mainly associated with the ser22/ser23 site [151].The decrease in calcium sensitivity ofthe actomyosin ATPase

to calcium accompanies a rise to approximately two moles

ofphosphate per mole ofprotein This would suggest that the

phosphorylation ofboth serine residues is required to produce

the increase in calcium sensitivity Recent studies with mutant

forms of cardiac TN-I introduced into skinned fibres have

indicated that this is the case [152]

The state of the TN-I in the normal unstimulated rabbit

heart is uncertain but four species could be present, namely

the ser22ser23, serP22ser23, ser22serP23 and serP22serP23

forms, with the total amount of phosphate not exceeding

about one mole per mole of protein To produce a significant

change in calcium sensitivity after adrenaline the

diphos-phorylated form must be absent or present as a very small

proportion ofthe total In that event the TN-I would be present

as the serP22 and serP23 forms, the proportion depending on

the relative susceptibility of the two serine residues to the

endogenous protein kinases.Itis not yet clear as to the precise

phosphorylation state of the normal beating heart in the

absence of adrenaline, although there are suggestions that

both monophosphorylated forms are present [150, 153, 154]

In a recent study of the distribution of the phosphorylated

forms ofTN-I in different regions of rabbit and bovine hearts

the diphosphate was the main form present [154a] In contrast,

the monophosphate was the predominant species present in the

human heart samples analysed The physiological states ofthe

hearts used in these studies were not defined

The fact that ordered phosphorylation of the two serine

residues occurs [155, 156] suggests that if protein kinase A is

the only enzyme involved in TN-I phosphorylation, serP23 will

be the major form present in the unstimulated rabbit heart This

conclusion follows from enzymic studies on synthetic peptides

corresponding to the N-terminal region ofhuman cardiac

TN-I In this cardiac isoform, with the phosphorylation site at ser23/

24, ser24 is almost completely phosphorylated by protein

kinase A before traces ofthe diphosphorylated peptide appear

[156] In this system little evidence ofphosphorylation ofser23

was obtained before ser24 was phosphorylated This implies

that ser23, whether in the non or monophosphorylated forms

ofthe N-terminal peptide, is phosphorylated much more slowly

than ser24 This would appear to be a general property of

cardiac TN-I phosphorylation by protein kinase A Enzymic

studies on mouse cardiac mutants in which each ofthe serines

of the phosphorylation site (in this species ser22 and ser23)

were replaced in tum by alanine, suggested that in the wild

type molecule ser23 was phosphorylated by protein kinase A

more rapidly than ser22 [152, 157] These results strongly

suggest that before adrenaline intervention the C-terminal

serine of the phosphorylation site is the residue that is

phos-phorylated This might be expected in view of its position in

the peptide sequence and the known specificity ofthe enzyme

The sequence RRRSS, which is the N-terminal

phosphoryla-tion site of all vertebrate cardiac TN-I so far sequenced, is

rather unusual and has presumably evolved to accommodate

the special needs of cardiac muscle regulation The arginineresidue immediately N-terminal to the serine residue is ofspecial significance Replacement by alanine or methionineleads to a reversal of the rate of phosphorylation of the twoserines [158] A feature ofthe sequence is that there is anotherserine, N-terminal to the phosphorylation site, that is notphosphorylated

Ifindeed the basal form is mainly ser23serP24, in thehuman heart for example, it is questionable whether proteinkinase A is sufficiently active after adrenaline intervention

to complete phosphorylation of the site or whether anotherenzyme may be involved In the case of phospholambanwhich also possesses an N-terminal dual phosphorylation site,two different enzymes are involved After adrenaline inter-vention the TN-I phosphate reverts to the basal level implyingthat ordered dephosphorylation has taken place There ismuch yet to be learnt about the enzymic mechanisms involved

in the phosphorylation and dephosphorylation ofTN-I in thefunctioning heart

The N-terminal peptide characteristic of cardiac troponin

I would appear to be a somewhat flexible structure comparedwith the rest of the molecule NMR studies suggest thatregions ofcardiac TN-I not interacting with cardiac TN-C arehighly flexible [158a] The phosphorylation site is available

to kinases and phosphatases in the functioning heart and anyinteractions that occur with other components ofthe troponincomplex are such as to enable the site to be available to theseenzymes In view ofthe marked changes in properties ofTN-Ion conversion from the mono to the diphosphorylated form

it is important to distinguish the structural effects associatedwith these two forms Most ofthe results so far reported havebeen obtained with the fully phosphorylated TN-I

The nature ofthe interaction ofcardiac TN-I and TN-C hasbeen less intensively studied than the skeletal system Asmight be expected in view of the sequence homology of thetwo systems, NMR studies [159] indicate that the con-formation of the two complexes are similar In both casescomplex formation with TN-I decreases interdomain flex-ibility and maintains TN-C in an extended conformation.Evidence has been obtained that the two proteins bind in anantiparallel fashion and that binding occurs between theinhibitory region and the region ofthe C-terminus oftroponin

C represented by residues metI20 and met157 [160,161] Ahint of the role of the cardiac N-terminal peptide is provided

by indications that by binding to the C-terminal ofTN-C itcan modulate the flexibility of the interdomain linker [159].Phosphorylation leads to conformational changes in theTN-I itself[156, 158, 161-163], in the interaction with TN-C[164-167] and in the binding of calcium to TN-C in thecomplex [168] Effects that might be expected in view of thechange in calcium sensitivity of the troponin system thatoccurs as a consequence ofphosphorylation By studying thechanges in distance between trp 192 and cysteine residues

introduced at different positions in mutant cardiac TN-I

molecules it has been concluded from flourescent resonance

energy transfer measurements that the structural changes

induced by PKA phosphorylation are confined to the

N-terminal half of the molecule [168a] The binding constant

was shown to be little changed by removing the N-terminal

32 residues of mouse cardiac TN-I [163] The results of

Al-Hillawi et al [165] indicate that calcium strengthens the

binding of the cardiac isoforms of TN-I to TN-C much less

than is the case with the skeletal isoforms Surprisingly

phosphorylation desensitised the interaction to calcium to the

extent that the cardiac complex no longer migrated as a

complex on electrophoresis under conditions in which the

25skeletal complex is stable Unlike the skeletal system thebinding ofcardiac TN-I to actin/tropomyosin (skeletal) showsmarked cooperativity This is lost on phosphorylation [165].NMR studies on the phosphorylation by protein kinase A

of a synthetic peptide corresponding to the region of theN-terminal phosphorylation site ofhuman cardiac TN-I havethrown some light on the conformational changes that occur[158] Phosphorylation of ser24 leads to interaction betweenser24 and arg 22 and some decrease in flexibility in thisregion The major change occurs on phosphorylating ser23which leads to a marked strengthening of the interactionbetween the phosphate groups and arginine residues 20, 21and 22 The result is that, unlike the monophosphate, the

Fig.4 Conformation about the phosphorylation site of the N-terminal region of human cardiac TN-I deduced from NOE data obtained from a synthetic peptide corresponding to residues 17-30 [158].

diphosphate is unable to bind a paramagnetic anionic probe

due to screening ofthe arginine side chains Phosphorylation

also abolishes the binding ofthe peptide to TN-C presumably

because the arginine side chains are no longer available for

electrostatic interactions Itshould be pointed out that any

interactions that occur in vivo between the N-terminal region

of cardiac TN-I and TN-C cannot involve tight binding as is

the case with other regions of TN-I for the N-terminal site

must be freely available to kinase and phosphatase The

structure of the peptide chain in the region of the

phos-phorylation site as computed from NOE spectrum data is

illustrated in Fig 4 The looped constrained structure is

compatible with the conclusions made from fluorescence

studies on a cys5 mutant ofmouse cardiac TN-I labelled with

the sulphydryl reagent IAANS [162)

Phosphorylation at sites other than N-terminal

Cardiac TN-I possesses potential phosphorylation sites in

homologous positions to thr11 and ser117 in the rabbit skeletal

muscle isoform These sites are partially phosphorylated in

skeletal muscle but as the phosphate does not rapidly

equili-brate with the intracellular phosphate pools they are probably

involved in long term modulation ofthe TN-IffN-C interaction

[169] The role of phosphorylase and protein A kinases in

phosphorylating homologous sites in cardiac troponin has not

been investigated in any detail Serl46 of rabbit cardiac TN-I

has been shown to be phosphorylated by protein kinase A in

vitroalthough at a much slower rate than ser22/23 [149, 151)

Cardiac TN-I contains a number of hydroxy amino acids

that are potential sites for phosphorylation of the isolated

protein In the rabbit there are 10 serines and 11 threonine

residues and all the vertebrate cardiac TN-Is so far

se-quenced (Swissprot) have a similar number with the

hy-droxy amino acids in homologous positions Whether these

residues are available for rapid phosphorylation and

de-phosphorylation, as is required for a dynamic modulatory

role, will depend on how exposed the site is in the troponin

complex in the living muscle Sites available for

phos-phorylation in all isoforms will be limited as much of the

surface must be blocked by the three proteins, TN-C, TN-T

and actin, with which it must interact to function This is

amply born out by the inhibition by TN-C of the

phos-phorylation by phosphorylase and protein A kinases ofrabbit

fast skeletal TN-I [134] Significantly the major regulatory

site is located on the flexible exposed extension of the

polypeptide chain at the N-terminus of the cardiac isoform

and is not blocked by TN-C

Protein kinase C can phosphorylate mouse cardiac TN-I

at ser43/ser45 [138, 170-172) Phosphorylation at these sites

does not alter the calcium sensitivity but results in a decrease

in the Ca2+activated MgATPase of actomyosin [138, 174]

Apart from their possible role as phosphorylation sites these

residues have functional significance If both are mutated

to alanine the calcium sensitivity and the Ca2+ activatedMgATPase is reduced Unlike the situation with the wild typeisoform phosphorylation of ser23/ser24 of the mutant doesnot further decrease the calcium sensitivity [138, 174).Ithasbeen suggested that phosphorylation at these sites might beresponsible for the negative inotropic effect of phorbol esters

on various cardiac preparations [173] Evidence that thesesites have been shown to be phosphorylated in cardiacmyocytes has been taken to support such a role.Itshould benoted that phosphorylation of such sites is not unique tocardiac muscle The phosphorylation of ser146 of rabbitcardiac TN-I by protein kinase A in not markedly inhibited

by TN-C, unlike phosphorylation of other sites by phorylase kinase [141, 175]

phos-The protein kinases are not completely specific for thepotential phosphorylation sites and under some conditions,for example, protein kinase C will phosphorylate ser23/ser24ofmouse cardiac TN-I [138,176).Itis clear that the modula-tion of calcium sensitivity by phosphorylation of the N-terminal extension is unique to cardiac muscle and it is very

probable that in the response in vivo to adrenaline, protein

kinase A is the enzyme involved It is more likely thatphosphorylation of the homologous potential phosphoryla-tion sites in TN-I isoforms, if indeed they have a modulatoryfunction, has a role that is common to the contractile process

in striated muscle in general, rather one than specific forcardiac tissue Much has to be learnt about the detailed role

of protein kinase C in regulating cardiac contractility in vivo

for it will also phosphorylate TN-T In the latter case thepattern of phosphorylation depends on the isoforms ofproteinkinase C present [177] and presumably the same will apply

to TN-I The protein kinase C gene family is complex andthe pattern of isoform expression is tissue specific Most of

the in vitro phosphorylation studies on cardiac TN-I have

been carried out with porcine brain protein kinase C whichconsists principally of the ex isoform Confirmation withisoforms specific for cardiac muscle would indicate whetherphosphorylation of TN-I by protein kinase C has a uniquemodulatory role in that tissue Some progress in this directionhas been made by the indication that there is decreasedmyofilament responsiveness to calcium in hearts fromtransgenic mice in which the protein kinase C~2 isoform isover expressed [1 77a)

resulting from the binding ofcalcium on TN-C to the enzymic

site on the myosin head where the MgATP is hydrolysed This

unique role requires that it must interact with each of the

proteins involved in this process, directly with actin,

TN-C, TN-T and possibly indirectly with tropomyosin and

myosin Its extended, apparently flexible, structure must be

important for this role Considerable progress has been

made in identifying regions of the TN-I molecule concerned

with its function With the present state of knowledge there

may, however, be dangers in being too positive about the

assignment of TN-I properties to precisely defined regions

of the polypeptide chain It is likely that some of the

pro-perties ofthe wild type protein depend on the entire molecule

being intact Mutation or chemical modification at one

position could cause changes elsewhere in the molecule that

are important for the property under investigation This may

explain some of the discrepancies that exist in assigning

properties to certain regions e.g the role of N-terminal

regions in regulating the Ca2+sensitization ofthe actomyosin

MgATPase [cf 80, 97, 98, 129, 138]

TN-I has derived its name from the property that led to its

discovery but there must be some doubt whether its role in the

myofibril is to act directly as an inhibitor According to the

steric hypothesis tropomyosin has this role Perhaps its most

significant property is its ability to induce different

con-formational changes in each of the proteins with which it

interacts in order to facilitate their function With TN-C it

increases the affinity for calcium so that this cation effectively

triggers contraction In cardiac muscle, by virtue of an

addi-tional N-terminal phosphorylation site, it is able to modulate

the process by changing the affinity ofTN-C for calcium in a

dynamic way in response to hormonal influences One

mole-cule of TN-I also interacts with actin and apparently induces

cooperativity of function over a stretch of seven actin

monomers Whatever the explanation of this effect, whether

tropomyosin has an active or passive role, the crucial event is

the binding ofTN-I to actin A more apt description ofthe role

of TN-I might be to call it a facilitator in that by interaction

with associated proteins it endows them with properties of

functional importance that they do not possess in its absence

Acknowledgements

I am grateful to Val Patchell and Barry Levine for their helpful

comments on this review and their help in its preparation

References

I Kumagi H, Ebashi S, Takeda F: Essential relaxing factor in muscle other

than myokinase and creatine phosphokinase Nature 176: 166, 1955

27

2 Perry SV, Grey TC: A study of the effects of substrate concentration and certain relaxing factors on the magnesium activated myofibrillar ATPase Biochem J 64: 184-192, 1956

3 Perry SV, Grey TC: Ethylenediaminetetracetate and the ATPase activities of actomyosin systems Biochem J 64: 5, 1956

4 Ebashi S: Third component participating in the superprecipitation of natural actomyosin Nature 200: 1010,1963

5 Perry SV, Davies V, Hayter D: 'Natural actomyosin' and the factor sensitising actomyosin adenosine triphosphatase to ethylene bis (ethyleneamine) tetraacetate Biochem J 99: Ic, 1966

6 Hartshorne DJ, Perry SV, Davies V: A factor inhibiting the adenosine triphosphatase activity and superprecipitation of actomyosin Nature 209: 1352, 1966

7 Hartshorne DJ, Perry SV, Schaub MC: A protein factor inhibiting the magnesium activated adenosine triphosphatase of desensitized actomyosin Biochem J 104: 907-913, 1967

8 Hartshorne DJ, Mueller H: Fractionation of troponin into two components Biochem Biophys Res Com 31: 647 -653, 1968

9 Schaub MC, Perry SV: The relaxing factor system of striated muscle Resolution of the troponin complex into inhibitory and calcium ion-sensitizing factors and their relationship to tropomyosin Biochem

J 115: 993-1004,1969

10 Wilkinson JM, Perry SV, Cole HA, Trayer IP: The regulatory proteins

of the myofibril Characterisation and biological activity of the components of inhibitory factor Biochem J 127: 215-224, 1972

II Greaser ML, Yamaguchi M, Brekke C, Potter J, Gergely J: Troponin subunits and their interactions Cold Spring Harbour Sym Quant BioI 37: 235-244,1972

12 Perry SV, Cole HA, Head JF, Wilson FJ: Localization and mode of action of the inhibitory protein component of the troponin complex Cold Spring Harbour Sym Quant Bioi 37: 251-262, 1972

13 Endo T, Obinata T: Troponin and its components from ascidian muscle.

Develop-16 Syska H, Perry SV, Trayer IP: A new method for the isolation of troponin I (inhibitory protein) using affinity chromatography Evidence for three different forms oftroponin 1 in striated muscle FEBS Letts 41: 253-257,1974

17 Sunderland CJ, Esser KA, Eisum KA, Gordon MC, Hardeman EC: Identification of a program of contractile protein gene expression initiated upon skeletal muscle differentiation Dev Dyn 196: 25-36, 1993

18 Sabry MA, Dhoot GK: Identification and pattern of expression of a developmental isoforrn oftroponin I in chicken and rat cardiac muscle.

J Muscle Res Cell Mot 10: 85-91,1989

19 Saggin L, Gorza L, Ausoni S, Schiaffino S: Troponin I switching in developing heart J BioI Chern 264: 16299-16302, 1989

20 Murphy AM, Thompson WR, Peng FL, Jone " L: Regulation of the cardiac troponin I gene by the transcription factor GATA-4 Biochem

J 322: 393-401, 1997

21 Westfall MV, Rust EM, Metzger JM: Slow skeletal troponin I gene transfer, expression and myofilament incorporation enhances adult cardiac myocyte contractile function Proc Nat! Acad Sci 94: 5444-

23a Barton PJR, Townsend PJ, Brand NJ, Yacoub MH: Localization of

the fast skeletal muscle troponin I gene (TNNT2) to IlpI5.5: Genes

for troponin I and troponin T are organised in pairs Hum Genet 61:

519-523, 1997

23b Mogensen J, KruseTA, BorglumAD: Assignment ofthe human cardiac

troponin I gene (TNNI3) to chromosome 19q13.4 by radiation hybrid

mapping Cytogenet Cell Genet 79: 272-273, 1997

24 Dhoot GK, Perry SV: Distribution ofthe polymorphic forms oftroponin

components and tropomyosin in skeletal muscle Nature 278:

714-718,1979

25 Dhoot GK, Gell PGH, Perry SV: Localisation of the different forms

of troponin I in skeletal and cardiac muscle Exp Cell Res 117:

357-370, 1978

26 Cummins P, Perry SV: Troponin I from human skeletal and cardiac

muscles BiochemJ 171: 251-259,1978

27 Mair J: Cardiac troponin I and troponin T: Are enzymes still relevant

as markers? Clin Chern Acta 257: 99-115, 1997

28 Dhoot GK, Perry SV, Vrbova G: Changes in the distribution of the

components of the troponin complex in muscle fibres after cross

innervation Exp Neur 72: 513-530, 1981

29 Dhoot GK, Perry SV: The effect of thyroidectomy on the distribution

of slow forms of troponin I in rat soleus muscle FEBS Letts 133:

225 229, 1981

30 Dhoot GK, Perry SV: The use of antibodies to the regulatory proteins

in the study of phenotype expresion in normal and diseased muscle.

In: S Ebashi (ed) Muscular Dystrophy University of Tokyo Press,

Tokyo,1982,pp.467-470

31 Bullard B, Dabrowska R, Winkelman L: The contractile and regulatory

proteins of insect flight muscle Biochem J 135: 277-286, 1973

32 Benzonana G, Kohler L, Stein EA: Regulatory proteins of crayfish

tail Biochem Biophys Acta 368: 247-258, 1974

33 Regenstein JM, Szent-Gyorgyi AG: Regulatory proteins of lobster

striated muscle Biochemistry 14: 917-925,1975

34 Lehman W, Regenstein JM, Ransom AL: The stoichiometry of the

components of arthropod thin filaments Biochem Biophys Acta 434:

215-222,1976

35 Kobayashi T, Takagi T, Konishi K, Cox JA: Amino acid sequence of

crayfish troponin I J BioI Chern 264: 1551-1557, 1989

36 Shinoda Y, Yamada A, Yagi, K: Identification oftroponin I of crayfish

myofibrils J Biochem 103: 636-Q40, 1988

37 Ojima T, Tanaka H, Nishita K: Cyanogen bromide fragments of

Akazara scallop M, 52000 troponin I J Biochem 108: 519-521, 1990

38 Ojima T, Tanaka H, Nishita K: Amino acid sequence ofC-terminal 17

kDa CNBR fragment of Akazara scallop troponin I J Biochem 117:

158-162,1995

39 Yuasa HJ, Sato S, Yamamoto H, Takagi T: Primary structure oftroponin

I isoforms from Halocynthia roretzi J Biochem 123: 374-380, 1997

40 Bullard B, Leonard K, Larkins A, Butcher G, Karlik C, Fyrberg E:

Troponin of asynchronous flight muscle J Mol Bioi 204: 621-637,

1988

41 Karlik CC, Mahaffey JW, Coutu MD, Fyrberg EA: Organisation of

contractile genes within the 88F subdivision of the D rnelanogaster

third chromosome Cell 37: 469-481, 1984

42 Barbas JA, Galceran J, Krah-Jentgens I, de la Pompa JL, Canal I,

Pongs 0, FerrusA: Troponin I is encoded in the haplolethal region of

the Shaker gene complex of Drosophila Genes Dev 5: 132-140,1991

43 Wilkinson JM, Grand RJA: Comparison of amino acid sequence of

troponin I from different striated muscles Nature 271: 31-35, 1978

44 Farah CS, Reinach FC: The troponin complex and regulation of muscle

contraction FASEB J: 755-767,1995

45 Pearlstone JR, Sykes BD, Smillie LB: Interactions of structural C and

N regulatory domains oftroponin C with repeat motifs in troponin 1.

Biochemistry 36: 7601-7606, 1997

46 Campbell AP, Sykes BD: Interaction oftroponin I and troponin C Use oftwo-dimensional nuclear magnetic resonance transferred nuclear Over- hauser effect to determine the structure ofthe inhibitory troponin I peptide when bound to skeletal troponin C J Mol Bioi 222: 405-421,1991

47 Campbell AP,Van Eyk JE, Hodges RS, Sykes BD: Interaction of troponin I and troponin C; use of the two dimensional transferred nuclear Overhauser effect to determine the structure of a gly-II 0 inhibitory troponin I peptide analogue when bound to cardiac troponin

C Biochem Biophys Acta 1160: 35 54, 1992

48 Saijo Y, Takeda S, Scherer A, Kobayashi T, Maeda Y, Taniguchi H, Yao M, Wakatsuki S: Production, crystallisation and preliminary X-ray anaysis of rabbit skeletal muscle troponin C and fragment (1-47) of troponin I Protein Sci 6: 916 918,1997

48a Vassylyev DG, Tokeda S, Wakatsuki S, Maeda K, Maeda Y: Crystal structure of troponin C in complex with troponin I fragment 2.3A resolution Proc Natl Acad Sci USA 95: 4847-4852, 1998

49 Stone DB, Timmins PA, Schneider DK, Krylova I, Ramos CHI,

Reinach FC, Mendelsohn RA: The effect of regulatory Ca'· on the in

situstructures oftroponin C and troponin I: A neutron scattering study.

J Mol BioI 281: 689-704, 1998

50 Haselgrove JC: X-ray evidence for conformational change in the containing filaments of vertebrate striated muscle Cold Spring Harbor Symp Quant Bioi 37: 341-352, 1972

actin-51 Huxley HE: Structural changes in actin- and myosin-containing filaments during contraction Cold Spring Harbor Symp Quant Bioi 37: 361-378,1972

52 Taylor K,Amos LA: A new model for the geometry ofbinding ofmyosin crossbridges to muscle thin filaments J Mol Bioi 147: 297-324,1981

53 Lehman W, Craig R, Vibert P: Cal+ induced tropomyosin movement

in Limulus thin filament revealed by three dimensional reconstruction Nature 368: 65-67, 1994

54 Eaton BL, Kominz DR, Eisenberg E: Correlation between the inhibition

of the acto-heavy meromyosin ATPase and the binding of tropomyosin

to F-actin; effects of Mg'., KCI, troponin I and troponin C chemistry 14: 2718-2725,1975

Bio-55 Katz AM: Influence of tropomyosin upon the reactions of actomyosin

at low ionic strength J BioI Chern 239: 3304-3311, 1964

56 Williams Jr DL, Greene LE, Eisenberg E: Comparison of effects of smooth and skeletal muscle tropomyosins on interactions of actin and myosin subfragment I Biochemistry 23: 4150-4155, 1984

57 Heeley DH, Watson MH, Mak AS, Dubord P, Smillie LB: Effect of phosphorylation on the interaction and functional properties of rabbit skeletal muscle aa-tropomyosin J BioI Chern 264: 2424-2430, 1989

58 Schaub MC, Hartshorne DJ, Perry SV: The adenosine triphosphatase activity of desensitised actomyosin Biochem J 104: 263-269, 1967

59 Schaub MC, Ermini M: Effect of bivalent cations on the adenosine triphosphatase activity of actomyosin and its modification by tropomyosin and troponin Biochem Jill: 777-783, 1969

60 Syska H, Wilkinson JM, Perry SV: The relationship between biological activity and primary structure oftroponin I from white skeletal muscle

of the rabbit Biochem J 153: 375-387,1976

61 Nozaki S, Kobayashi K, Katayama E, Matsumura I: Synthetic studies

on troponin I active site Preparation of a pentadecapeptide with inhibitory activity towards actomyosin adenosine triphosphatase Chern Letts 3: 345 348, 1980

62 Talbott JA, Hodges RS: Synthetic studies on the inhibitory region of rabbit skeletal troponin I J Bioi Chern 256: 2798-2802, 1981

63 Pearlstone JR, Smillie LB: Effects oftroponin I plus -C on the binding oftroponin T and its fragments to a-tropomyosin J Bioi Chern 258: 2534-2542, 1983

64 Zhou X, Morris EP, Lehrer SS: Troponin I and troponin-C binding to actin tropomyosin and dissociation by myosin S I Biophys J 68: A167, 1995

65 Bremel RD, Weber A: Cooperative behaviour within the functional

unit of the actin filament in vertebrate striated muscle Nature 238:

97-101,1977

66 AI-Khayat HA, Vagi N, Squire JM: Structural changes in

actin-tropomyosin during muscle regulation: computor modelling of low

angle X-ray diffraction data J Mol Bioi 252: 611-{j32, 1995

67 Grand RJA, Henry DG, Moir AJG, Perry SV, Trayer IP, Dalgarno,

DC, Levine BA, Parker SB: Modulation by troponin C of the troponin

I inhibition of skeletal actomyosin interactions A PMR study, In: B.

de Barnard, G.L Sottocasa, G Sandri, E Carafoli, A.N Taylor,T.C.

Vanaman and RJ.P Williams (eds).Calcium Binding Proteins Elsevier/

North Holland, Amsterdam, 1983, pp 379-380

68 Ishikawa T, Wakabayashi T: Calcium induced change in the three

dimensional structure of thin filaments of rabbit skeletal muscles as

revealed by cryoelectron microscopy Biochem Biophys Res Com 203:

951-958,1994

69 Van Eyk JE, Strauss JD, Hodges RS, Ruegg JC: A synthetic peptide

mimics troponin I function in the calcium regulation of muscle

contraction FEBS Lett 323: 223-228, 1993

70 Talbot JA, Hodges RS: Comparative studies on the inhibitory region of

selective species oftroponin I J Bioi Chern 256: 12374-12378,1981

71 Dalgarno DC, Grand RJA, Levine BA, Moir MG, Scott GMM, Perry

SV: Interaction between troponin I and troponin C Definition of

topography by proton magnetic resonance spectroscopy FEBS Letts

150: 54-58, 1982

72 Levine BA, Moir MG, Perry SV: The interaction of troponin I with

the N-terminal region of actin Eur J Biochem 172: 389-397, 1988

73 Grabarek Z, Gergely J: Troponin I binds to the N-terminal twelve

residue segment of actin Biophys J 51: 331 a, 1987

74 Jiang J, Wu J-L, Gergely J, Tao T: Ca'· induced movement of the

N-terminal region of troponin I relative to troponin C and actin Biophys

J 68: A165, 1995

75 Tao T, Gong BJ, Leavis PC: Calcium induced movement oftroponin I

relative to actin in skeletal muscle thin filaments Science 247:

1339-1441,1990

76 Li HC, Palm T, Sale K, Fajer PG: Crossbridges-induced movement of

troponin I relative to actin in skeletal muscle thin filaments Biophys

J 72: A248, 1997

77 Li Z, Gergely J, Tao T: Resonant energy transfer and photocrosslinking

studies of the interaction between troponin C and the inhibitory region

of a monocysteine mutant oftroponin I Biophys J 68: A 165, 1995

78 Lehrer SS, Chai M, Geeves MA: Effects of troponin I on actin S I

ATPase and S 1 binding kinetics in the absence and presence of rabbit

skeletal tropomyosin and troponin Biophys J 72: A60, 1997

79 Zang R, Sheng Z, Jones M, Parsons B, Potter JD: Effect of mutations

in the inhibitory region of skeletal muscle troponin I on its biological

activity Biophys J 70: A 165, 1996

80 Van Eyk JE, Thomas LT, Tripet, B, Wiesner RJ, Pearlstone JR, Farah

CS, Reinach FC, Hodges RS: Distinct regions of troponin I regulate

Ca'·-dependent activation and Ca'· sensitivity of the

acto-SI-TM-ATPase activity ofthin filaments J Bioi Chern 272: 10529-10537,1997

81 Tripet B, Van Eyk JE, Hodges RS: Mapping of a second

actin-tropomyosin and a second troponin C binding site within the C terminus

oftroponin I and their importance in the Ca'·-dependent regulation of

muscle contraction J Mol Bioi 271 : 728-750, 1997

82 Rarick HM, Tu X-H, Solaro RJ, Martin A: The C terminus of cardiac

troponin 1 is essential for full inhibitory activity and Ca'· sensitivity

of rat myofibrils J BioI Chern 272: 26887-26892, 1997

83 Head JP, Perry SV: The interaction of the calcium binding protein

(troponin C) with bivalent cations and the inhibitory protein (troponin

I) Biochem J 137: 145-154,1974

84 Grabarek Z, Tao T, Gergely J: Molecular mechanism of troponin C

function J Muscle Res Cell Mot 13: 383-393, 1992

87 Kobayashi T, Tao T, Grabarek Z, Gergely J, Collins JH: Cross-linking

of residue 57 in the regulatory domain of mutant rabbit skeletal muscle troponin C to the inhibitory region of troponin I J BioI Chern 266: 13746-13751,1991

88 Luo J, Qian J, Wu J-L, Tao T, Gergely J: Inter-subunit crosslinking

of single cys mutant troponin I in binary and ternary complexes Biophys J 72: A59, 1997

89 Ngai SM, Sonnichensen FD, Hodges RS: Photochemical cross-linking between native rabbit skeletal troponin C and benzylbenzoyl inhibitory peptide, residuesI04-115 J Bioi Chern 269: 2165-2172, 1994

90 Tao T, Gong B-J, Gergely J: The conformation oftroponin I studied

by chymotryptic digestion Biophys J 72: A59, 1997

91 Amphlett GW, Vanaman TC, Perry SV: Effect of the troponin C-like protein from bovine brain (brain modulator protein) on the Mg 2._

stimulated ATPase of skeletal muscle actomyosin FEBS Letts 72: 162-168, 1976

92 Weeks RA, Perry SV: Characterisation of a region of the primary sequence oftroponin C involved in calcium ion dependent interaction with troponin I Biochem J 173: 449 457, 1978

93 Leavis PC, Rosenfeld S, Gergely J, Grabarek Z, Drabikowski W: Proteolytic fragments of troponin C Localisation of high and low affinity Ca 2• binding sites and interactions with troponin I and troponin T J BioI Chern 253: 5452-5459, 1978

93a Leszyk J, Tao T, Nuwaysir LM, Gergely J: Identification of photocrosslinking sites in troponin I with 4-maleimidobenzophenone labelled mutant troponin-Cs having single cysteines at positions 21 and 158 J Muse Res Cel Motil 19: 479 490, 1998

94 Katayama E, Nozaki S: Ca 2• dependent binding of synthetic peptides corresponding to some regions oftroponin I to troponin C J Biochem 91: 1449-1452,1982

94a Rarick HM, Tang H-P, Guo X-D, Tu X-H, Martin AF, Solaro RJ: Interations of the N-terminal region ofcardiac TNI with the structural cTNC site and cTNT are involved in the Ca 2• dependent activation

of MgATPase activity Biophys J 74: A52, 1998

95 Bingham RP, Trayer IP: Characterisation ofa troponin C binding site

on the N-terminal oftroponin I J Muscle Res Cell Mot 18: 260, 1997

96 Ngai SM, Hodges RS: Biologically important interactions between synthetic peptides of the N-terminal region oftroponin I and troponin

C J BioI Chern 267: 15715-15720, 1992

97 Sheng Z, Pan B-S, Miller TE, Potter JD: Isolation, expression and mutation of a rabbit muscle eDNA clone for troponin I J BioI Chern 267: 25407-25413, 1992

98 Farah CS, Miyamoto CA, Ramos CH, da Silva ACR, Quaggio RB, Fujimori K, Smillie LB, Reinach FC: Structural and regulatory functions of the NH2and COOH-terminal regions of skeletal muscle troponin I J Bioi Chern 269: 5230-5240, 1994

99 Chong PSC, Hodges RS: Proximity of sulphydryl groups to the sites

of interaction between components of the troponin complex from rabbit skeletal muscle J Bioi Chern 257: 2549-2555,1982

100 Wang CK, Cheung HC: Proximity relationship in binary complex formed between troponin I and troponin C J Mol Bioi 190: 509-521, 1984

101 Tao T, Gowell E, Strasbourg GM, Gergely J, Leavis PC: Ca 2• pendence of the distance between cys98 oftroponin C and cysl33 of troponin I in the ternary complex Biochemistry 28: 5902-5908, 1989

de-102 Kobayashi T, Tao T, Gergely J, Collins JH: Structure of the troponin complex Implications of photo crosslinking oftroponin I to troponin

C thiol mutants J BioI Chern 269: 5725-5729, 1994

103 Keller RS, Rarick HM, Martin AF, Solaro RJ: Effect of truncated

and a C-terminal mutant of cardiac TN-Ion myofilament force.

Biophys J 72: A379, 1997

104 Ingraham RH, Swenson CA: Binary interactions oftroponin subunits.

J Bioi Chern 259: 9544-9548,1984

105 Leavis PC, Gowell E, Tao T: Flourescence lifetimes and acrylamide

quenching studies of the interaction between troponin subunits.

Biochemistry 23: 4156-4161,1984

106 Wang CK, Cheung HC: Energetics of the binding of calcium and

troponin I to troponin C Biophys J 48: 727-739, 1985

107 Cheung HC, Wang CK, Malik NA: Interactions oftroponin subunits:

Free energy of binary and tertiary complexes Biochemistry 26:

5904-5907, 1987

108 Potter JD, Gergely J: The calcium and magnesium binding sites on

troponin and their role in the regulation of myofibrillar adenosine

triphosphatase J BioI Chern 250: 4628-4633, 1975

108a McKay RT, Tripet BP, Hodges RS, Sykes BD: Interaction of the

second binding region of troponin I with the regulatory domain of

troponin C as determined by NMR Biophys J 74: A298, 1998

I08b Luo Y, Qian Y, Lesyk J, Gergley G, Tao T, Residues 48 and 82 at the

N-terminal hydrophobic patch of troponin-C photocrosslink to

methionine 121 oftroponin-I Biophys J 74: A143, 1998

I08c Li MX, Spyrocoupolos L, Sykes BD: Interaction of human cardiac

Troponin-I (residues 148-161) wi th the regulatory domain of human

cardiac Troponin C Biophys J 74: A51, 1998

I08d Davis B, Liu W, Putkey JA: Effect of troponin I association of the

calcium-bound conformation of cardiac troponin C Biophys J 74:

A143,1998

109 Olah GA, Trewewhella J: A model structure of the muscle protein

complex 4Ca H troponin C troponin I derived from small angle

scattering data: Implications for regulation Biochemistry 33:

12800-12806,1994

110 Olah GA, Rokop SE, Wang CLA, Blechner SL, Trewhella J: Troponin

I encompasses an extended troponin C in the Ca 2' bound complex: A

small angle X-ray and neutron scattering study Biochemistry 33:

8233-8239, 1994

Ill Tao T, Qian Y, BoldoghI, Gergely J: Flourescent lifetime, acrylamide

quenching and photo crosslinking studies between troponin I and

troponin T and monocysteine mutants of troponin C Biophys J 68:

A166,1995

112 Luo Y, Wu J-L, Gergely J, Tao T: Troponin T and Ca 2' dependence

on the distance between cys48 and cys 133 in the ternary troponin

complex and reconstituted thin filaments Biochemistry 36:

11027-11035,1997

113 Leszyk J, Collins JH, Leavis PC, Tao T: Cross-linking of rabbit

skeletal muscle troponin subunits: Labelling of cys 98 of troponin C

with 4-maleimidobenzophenone and analysis of products formed in

the binary complex with troponin T and the ternary complex with

troponins I and T Biochemistry 27: 6983-6987,1988

114 Leszyk J, Grabarek Z, Gergely J, Collins JH: Characterisation

ofzero-length cross links between rabbit skeletal muscle troponin C and

troponin I: Evidence for a direct interaction between the inhibitory

region of troponin I and the NH2-terminal, regulatory domain of

troponin C Biochemistry 29: 299-304, 1990

115 Leszyk J, Grabarek Z, Gergeiy J, Collins J: Characterisation of zero

cross links between rabbit skeletal muscle troponin C and troponin I:

Evidence for a direct interaction between the inhibitory region of

troponin I and the NH2-terminal regulatory domain of troponin C.

Biochemistry 29: 299-304,1990

116 Wang Z, Sarkar S, Gergely J, Tao T: Ca 2 + dependent interactions

between the C helix oftroponin C and troponin I Photo crosslinking

and fluorescent studies using a recombinant troponin C J Bioi Chern

119 Grand RJA, Levine BA, Perry SV: Proton magnetic resonance studies

on the interaction of rabbit skeletal muscle troponin I with troponin

C and actin Biochem J 203: 61-68,1982

120 Kobayashi T, Zhao X, Wade R, Collins JH: Effect of the acidic residues

of the N-terminal domain of troponin C on the calcium dependent interaction with the inhibitory peptide Biophys J 70: A38, 1996

121 Lin X, Dotson DG, Putkey JANA: Covalent binding of peptides to the N terminal hydrophobic region of cardiac troponin C has limited effects on function J Bioi Chern 271: 244-249, 1996

12 Ia Perry SV: Troponin T: Genetics, properties and function J Muse Res Cell Motil 19: 575-602, 1998

122 Hitchcock-Gregori SE: Study of the structure of troponin I by measuring the relative reactivities oflysines with acetic anhydride J BioI Chern 257: 7372-7380, 1982

123 Ohtsuki I: Molecular arrangement oftroponin T in the thin filament.

127 Jha Pk, Leavis PC, Sarkar S: Interaction of deletion mutants of troponins I and T: COOH-terminal truncation oftroponin T abolishes troponin I binding and reduces Ca 2 + sensitivity of the reconstituted regulatory system Biochemistry 35: 16573-16580,1996

128 Potter JD, Sheng Z, Pan B-S, Zhao: A direct regulatory role for troponin T and a dual role for troponin C in the Ca 2 + regulation of muscle contraction J Bioi Chern 270: 2557-2562, 1995

128a Stefancsik R, Jha PK, Sarkar P: Identification and mutagenesis of a highly conserved domain in troponin T responsible for troponin I binding: Potential for coiled coil interaction Proc Nat! Acad Sci USA 95: 957-962, 1998

129 Thomas L, Van Eyk JE, Farah CS, Reinack FC, Hodges RS, Smillie LB: The role oftroponin I and troponin T in the calcium dependent regulation of reconstituted thin filaments Biophys J 70: A 166, 1996 129a Malnic B, Farah CS, Reinach FC: Regulatory properties of the NH2-and COOH-terminal domains oftroponin T J BioI Chern 273: 10594- 10601,1998

130 Bailey C, Villar-Palasi C: Cyclic AMP dependent phosphorylation oftroponin Feb Proc 30: 1147, 1971

131 Stull JT, Brostrom CO, Webs EG: Phosphorylation of the inhibitor component of troponin by phosphorylase kinase J BioI Chern 247: 5272-5274, 1972

132 Pratje E, Heilmeyer LMG: Phosphorylation ofrabbit muscle troponin and actin by 3'-5'cAMP dependent kinase FEBS Lett 27: 89-73,1972

133 Perry SV, Cole HA: Phosphorylation of the '37000 component' of the troponin complex (troponin T) Biochem J 131: 425-428,1973

134 Perry SV, Cole HA: Phosphorylation of troponin and the effects of interactions between the components Biochem J 141: 733-744, 1974

135 Moir AJG, Wilkinson JM, Perry SV: The phosphorylation sites of troponin I from white skeletal muscle of the rabbit FEBS Lett 42: 253-256,1974

136 HuangTS, Bylund DB, Stuil JT, Krebs EJ: The amino acid sequences

of the phosphorylated sites in troponin I from skeletal muscle FEBS Lett 42: 249-252, 1974

137 Ribulow H, Barany K, Steinschneider A, Barany M: Lack of

phosphate incorporation in troponin I in living frog muscle Arch

Biochem Biophys 179: 81-88, 1977

138 Noland Jr TA, Guo X, Raynor RL, Jideama NM, Averyhart-Fullard

V, Solaro RJ, Kuo JF: Cardiac I mutants Phosphorylation by protein

kinases C and A and regulation of the Ca 2 ·-stimulated MgATPase of

reconstituted actomyosin-S-I J Bioi Chern 270: 25445-25454, 1997

139 Reddy YS, Ballard D, Giri NY, Schwartz A: Phosphorylation of

cardiac native tropomyosin and troponin, inhibitory effect of

actomyosin and possible presence ofendogenous myofibrillar located

cyclic AMP dependent protein kinase J Mol Cell Cardiol 5:

461-471,1973

140 Reddy YS, Schwartz A: Phosphorylation ofcardiac contractile proteins

by cyclic AMP dependent protein kinase Fed Proc 33: 1294, 1974

141 Cole HA, Perry SV: The phosphorylation oftroponin I from cardiac

muscle Biochem J 149: 525-533, 1975

142 England PJ: Correlation between contraction and phosphorylation

of the inhibitory subunit oftroponin in perfused rat heart FEBS Lett

50: 57 60, 1975

143 Solaro J, Moir AJG, Perry SV: Phosphorylation oftroponin I and the

ionotropic effect of adrenaline in the perfused rabbit heart Nature

262:615 617,1976

144 Rubio R Bailey C, Villar-Palasi C: Effects of cyclic AMP dependent

protein kinase on cardiac actomyosin: increase in Ca 2• sensitivity

and possible phosphorylation oftroponin I J Cyclic Nucleotide Res

1: 143-150,1975

145 Ray KP, England PJ: Phosphorylation of the inhibitory subunit of

troponin and its effect on the calcium dependence ofcardiac myofibril

adenosine triphosphatase FEBS Lett 70: 11-16, 1976

146 Ding X-L, Sonnenblick EH, Gulat J: Pathways of communication

between cardiac TN-I and troponin C during acidosis and TN-I

phosphorylation by protein kinase Biophys J 72: A59, 1997

147 Kogler H, Plathow C, AI-Hillawi E, Trayer IP, Ruegg JC: Effects of

reconstitution with wild type and mutant cardiac troponin I in skinned

soleus fibres Eur J Physiol433: SS P263, 1997

148 Johns EC, Simmet SJ, Muffigan IP, Ashley CC: Troponin

phos-phorylation does not increase the rate of relaxation following laser

flash photolysis of diazo-2 in guinea pig skinned trabeculae Eur J

Physiol 433: 842-844, 1997

148a Li L, Chu X, Kranias EG, Bers DM: Phosphorylation of

phos-pholamban is more crucial than troponin I phosphorylation in PKA

mediated acceleration of relaxation in cardiac myocytes Biophys J

74: A248, 1998

149 Moir AJG, Perry SV: The sites of phosphorylation of rabbit cardiac

troponin I by adenosine 3'5'-cyclic monophosphate dependent protein

kinase Effect of interaction with troponin C Biochem J 167:

333-343,1977

150 Mittman K, Jaquet K, Heilmeyer Jr LMG: A common motif of two

adjacent phosphoserines in bovine, rabbit and human cardiac troponin

I FEBS Lett 273: 41-45,1990

151 Moir AJG, Solaro J, Perry SV: The site ofphosphorylation oftroponin

I in the perfused heart: the effect of adrenaline Biochem J 185:

505-513,1980

152 Zhang R, Zhao J, Potter JD: Phosphorylation of both serine residues

in cardiac troponin I is required to decrease the Ca 2• affinity ofcardiac

troponin C J Bioi Chern 270: 30773-30780, 1995

153 Jaquet K, Thieleczek R, Heilmeyer Jr LMG: Pattern formation on

cardiac troponin I by consecutive phosphorylation and

dephos-phorylation Eur J Biochem 231: 486-490, 1995

154 Reiffert SU, Jaquet K, Heilmeyer Jr LMG Ritchie M, Geeves MA:

Bisphosphorylation of cardiac troponin I modulates the Ca 2•

dependent binding of myosin subfragment S I to reconstitued thin

filaments FEBS Lett 384: 43-47,1996

31

154a Ardell P, Dorka P, Jaquet K, Heilmeyer LMG Jr, Kortke H, Korer R, Notohamiprodjo G: Microanalysis and distribution ofcardiac troponin

I phosphate species in heart areas Bioi Chern 379: 341-347, 1998

155 Mittman K, Jaquet K, Heilmeyer Jr LMG: Ordered phosphorylation

of a minimal recognition motif for cAMP dependent protein kinase present in cardiac troponin I FEBS Lett 302: 133-137,1992

156 Quirk PG, Patchell VB, Gao Y, Levine BA, Perry SV: Sequential phosphorylation ofadjacent serine residues on the N-terminal region

of cardiac troponin I: Structure activity implications of ordered phosphorylation FEBS Lett 370: 175-178, 1995

157 Malhotra A, Nakouzi A, Bowman J, Buttrick PM: Role of protein kinase A (PKA) and protein kinase C (PKC) phosphorylation of troponin I in cardiac muscle Biophys J 70: A52, 1996

158 Keane NE, Quirk PG, Gao Y, Patchell VB, Perry SV, Levine BA: The ordered phosphorylation of cardiac troponin I by the cAMP- dependent protein kinase Structural consequences and functional implication Eur J Biochem 248: 329-337, 1997

158a Gasmi-Seabrook G, Howarth JW, Finley N, Abbott MB, Britto RM, Rosevear PR: Structure ofthe C-terminal domain ofcardiac troponin

C complexed with the N-terminal of cardiac troponin I Biophys J 74: A299, 1998

159 Kleerekoper 0, Howarth J, Guo X, Solaro Rj, Rosevear PR: Cardiac troponin I induced conformational changes in cardiac troponin C as monitored by NMR using site-directed spin and isotope labelling Biochemistry 34: 13343-13352, 1995

160 Krudy GA, Kleerekoper 0, Guo X, Howarth JW, Solaro RJ, Rosevear PR: NMR studies delineating space relationships within the cardiac troponin I troponin C complex J Bioi Chern 269: 23731-

23735, 1994

161 Howarth JW, Krudy GA, Putkey JA, Rosevear PR: An NMR and spin label study of the effects of binding calcium and troponin I inhibitory peptide to cardiac troponin C Protein Sci 4: 671 680, 1995

162 Dong W-J, Chandra M, Xing J, Solaro JD, Cheung HC: Conformation

of the N-terminal segment of a monocysteine mutant oftroponin 1of cardiac muscle: Biochemistry 36: 6745 6753, 1997a

163 Dong W-J, Chandra M, Xing J, She M, Solaro JD, Cheung HC: Phosphorylation induced distance change in a cardiac muscle troponin

I mutant Biochemistry 36: 6754 6761, 1997b

164 Liao R, Wang CK, Cheung HC: Effect of phosphorylation of cardiac troponin I on the fluorescence properties of its single tryptophan as determined by pico second spectroscopy SPIE Proceedings 204: 699-

705, 1990

165 AI-Hillawi E, Bhandari DG, Trayer HR, Trayer IP: The effect of phosphorylation of cardiac troponin I on its interaction with actin and cardiac troponin C Eur J Biochem 228: 962-970, 1995

166 Dong M, Xing J, She M, Chandra M, Solaro RJ, Cheung HC: Phosphorylation increased distance in cardiac troponin I Biophys J 70: A38, 1996

167 Chandra M, Pan B-S, Solaro JS: Effect of PKA dependent phorylation of cardiac troponin I with Nand C-terminal domains of cardiac troponin C Biophys J 70: A53, 1996

phos-168 Robertson SP, Johnson JD, Hoiroyde MJ, Kranias EG, Potter JD: The effect of troponin I phosphorylation on the Ca 2 ·-binding properties of the Ca 2• regulatory site ofbovine troponin J Bioi Chern 257: 260-263, 1982

168a Dong W-J, Chandra M, Xing J, Solaro RJ: Phosphorylation induced conformational change in single cysteine mutants ofcardiac troponin

I Biophys J 74: A51, 1998

169 Perry SV: Phosphorylation and the modulation of the calcium regulation of contraction In: K Maruyama, Y Nonomura, K Kohama (eds) Calcium as Cell Signal Igaku-Shion, Tokyo, 1995,

pp 49-59

170 Katoh N, Wise BC, Kuo JF: Phosphorylation of cardiac troponin

inhibitory subunit (troponin I) and tropomyosin binding unit (troponin

T) by cardiac lipid-sensitive Ca2+ dependent protein kinase Biochem

J 209: 189-195,1983

171 Mazzei GF, Kuo JF: Phosphorylation of skeletal muscle troponin I

and troponin T by phospholipid-sensitve Ca2+ dependent protein

kinase Biochem J 218: 361-369,198

172 Noland Jr TA, Raynor RL, Kuo JF: Identification of sites

phos-phorylated in bovine cardiac troponin I and troponin T by protein kinase

C and comparative substrate activity of synthetic peptides containing

phosphorylation sites J BioI Chern 264: 20778-20785, 1989

173 Noland TA, Kuo JF: Protein kinase C phosphorylation of cardiac

troponin I or troponin T inhibits Ca'+-stimulated actomyosin

MgATPase activity J Bioi Chern 266: 4974-4978, 1991

174 Noland JrTA, Guo X, Raynor RL, Jideama, NM,Averyhardt-Fullard

V, Solaro J, Kuo JF: Cardiac troponin I mutants J Bioi Chern 270:

177 Jideama NM, NolandTA, Raynor RL, Blobe GC, Fabbro D, Kazanietz

MG, Blumberg PM, Hannun YA, Kuo JF: Phosphorylation ficities of protein kinase isozymes for bovine cardiac troponin I and troponin T and sites within these proteins and regulation of myo- filament properties J Bioi Chern 271: 23277-23283, 1996 177a Takeishi Y, Chu G, Kirkpatrick DM, Li Z, Wakasaki H, Kranias EG,

speci-King GL, Walsh RA: In vitro phosphorylation of cardiac troponin I by

protein kinase C132 decreases cardiomyocyte calcium responsiveness and contractility in transgenic mouse hearts J Clin Invest 102: 72-78, 1998

Molecular and Cellular Biochemistry190: 33-38, 1999.

©1999 Kluwer Academic Publishers.

Calcium ion regulation of muscle contraction: The regulatory role of troponin T

Itwas in the early summer of 1962 that Professor Setsuro

Ebashi first isolated a protein factor from minced muscle that

sensitized actomyosin to Ca2+[1-3] This finding is the real

dawn of the molecular biology ofliving muscle contraction

The Ca2+-sensitizing factor called native tropomyosin was

thereafter separated into two proteins One was tropomyosin,

a fibrous protein which had been found, while the other was

a new globular protein, troponin [4, 5]

Troponin is the sole Ca2+-receptive protein in the

con-traction of vertebrate striated muscles [5] and is regularly

distributed along the entire length of the thin filament[6, 7]

The contractile interaction of myosin and actin is regulated

by troponin and tropomyosin through Ca2+ In the absence of

Ca2+,troponin, in collaboration with tropomyosin, depresses

the contractile interaction of actin with myosin The

depres-sion by troponin-tropomyosin is removed through the action

of Ca2+on troponin and the contraction is then activated [8]

(Fig la)

Troponin consists of three different components, namely;

Ca2+-binding component (troponin C), inhibitory component

(troponin I) and tropomyosin-binding component (troponin

T) [9-12] The presence of all three components oftroponin

is required for Ca2+-regulation of the contractile interaction

of myosin-actin-tropomyosin, which itself stays in the

activated state irrespective ofthe Ca2+-concentration (Fig 1b)

[13] The essential roles of the troponin components in the

Ca2+-regulation are as follows Troponin I alone inhibits thecontractile interaction ofactin-tropomyosin with myosin andthis inhibition by troponin I is then removed by the additionoftroponin C, regardless ofthe Ca2+-concentration The Ca2+-sensitivity is conferred on the contractile interaction only inthe concomitant presence of troponins I, C and T In thisrespect, troponin I and troponin C are the inhibitory andde-inhibitory (activating) components on the contraction,respectively, whereas troponin T is the regulatory component

of the troponin complex

In this article, the Ca2+-regulatory mechanisms of skeletalmuscle contraction are discussed with special reference to theregulatory role of troponinT

Troponin T subfragments

Troponin T is a tropomyosin-binding component and itspresence is essential for the Ca2+-regulation of contractionthrough troponin-tropomyosin, even though troponin T itselfdoes not significantly affect the contractile interaction ofmyosin-actin-tropomyosin or the inhibitory action oftroponin

I [13] Troponin T binds to tropomyosin, troponin I andtroponin C The interaction with tropomyosin is highlysensitive to ionic strength, while the interaction with troponins

C and I is relatively stable in relation to the ionic strength

Address for offprints:1 Ohstuki, Department of Pharmacology, Faculty of Medicine, Kyushu University, Fukoka 812 82, Japan

Relaxation 1 ======= _

Ca2 +-regulating action of troponin T

101 residues (Fig 2) Both subfragments are highly soluble,although troponin T is insoluble at low ionic strengths Theproperties ofthese subfragments can thus be examined underphysiological conditions (13,18,19]

Although both troponin T subfragments bind to myosin, the binding stability oftroponin T1is much higher

tropo-than that oftroponin T z.This indicates that troponin T binds

to tropomyosin mainly through the troponin TIregion Thebinding region within the troponin T, is in the a-helical region

of residues 71-150 (CB2 fragment) which can thus form thestable triple stranded binding structure with the coiled-coiltropomyosin [20, 21] Regarding the binding of troponin Tz

to tropomyosin, studies using troponinTzfragments indicatethat the main binding region to tropomyosin is located at thesmall C-terminal region of 17 residues This binding totropomyosin though the troponinT zregion is necessary forthe Caz+-regulating action of troponinT.

Troponin Tzalso binds to troponin I and troponin C, whiletroponin T, shows no affinity to these troponin components

Ithas been demonstrated by the use oftroponinTzfragmentsthat the small region in troponinT zincluding residues 222-

227, which are highly conserved among several kinds oftroponin T [22], is critical for the binding to troponin I,

although the broad region oftroponin T zalso weakly interactswith troponinI.Troponin C interacts with the broad regionoftroponinT z,in which the N-terminal side region (troponin

Tz~III)is mainly related to the Caz+- dependent interactionand the remaining C-terminal region is involved in theCaz+-independent interaction.Itis, however, uncertain as to

whether or not the troponin C-troponin T zinteraction isactually present and operates in the ternary troponin complex

A water soluble fragment of troponin T called the 26 K

fragment is produced by endogenous protease in the musculartissue This fragment is devoid ofthe N-terminal45 residues

of troponin T and has the same properties as the intacttroponin T with regard to both the Caz+-regulating action andtropomyosin-binding [23]

The most significant finding obtained from the studies on thechymotryptic subfragments of troponin T is that the Caz+-regulating action of troponin T is almost fully retained in

troponin T z.But the Caz+-regulating activity of the troponin

T zsubfragment disappears by the removal of the C-terminal

17 residues with a concomitant decrease in the affinity totropomyosin, while the affinity to troponin C and troponin Iremains mostly unchanged [24, 25] Tropomyosin-bindingthrough the C-terminal region oftroponinT ztherefore plays

a critical role in the Ca2+-regulating action oftroponin T (Fig.3) The strong binding of the range of residues 71-150 oftroponin TIto tropomyosin, at the same time, stably fixes the

[Ca2+]

TN·I+TN·CContraction ~ -'' ' !.~' ' !.~ -:: '-'''==

Contraction ~ -:; -""""'==

[Ca2+]

'Two domain structure of troponin T was first indicated by the finding that

antitroponin T antibody formed a pair of striations in each troponin period

along the thin filament bundle [16, 17] Study of chymotryptic troponin T

subfragments actually demonstrated that the striation by antitroponin T on

the Z-band side was formed by the N-terminal(T,) region of troponin T

and the other striation on the filament-top side by the C-terminal (T,) region

[15].

Troponin T from the rabbit skeletal muscle is a single peptide

of 259 amino acid residues, which contains 130 charged

residues, and its N-terminal region is highly rich in acidic

residues, while the C-terminal region is rich in basic residues

[14]

Troponin T is split into two subfragments, troponin T, and

troponin T z,by mild treatment with chymotrypsin [15]1

Troponin T1is an acidic fragment of the N-terminal 158

residues and troponin T zis a basic fragment ofthe C-terminal

Fig I. Ca'+-regulatory mechanism of troponin and tropomyosin (A)

Ca'+-regulation of the contractile interaction of myosin-actin in the absence

and presence oftroponin (TN) and tropomyosin (TM) (8) Ca2+-regulation

of the contractile interaction of myosin-actin-tropomyosin in the absence

and presence oftroponin components Cited from Ohtsuki et al [13] TN'C,

I and T - troponins C, I and T.

b)

a)

Fig 2 Distribution of the interacting properties of rabbit skeletal troponin T along its amino acid sequence Cited from Ohtsuki et al [13].

position of the troponin CI·T complex to the filamentous

tropomyosin-actin and hence supports the regulatory

inter-action oftroponin T2with tropomyosin and troponin CI

The N-terminal region oftroponin T, which has no affinity

to tropomyosin, is not related to the Ca2+- regulating activity

of troponin T In fact, the 26 K fragment devoid of the

N-terminal residues from troponin T shows exactly the same

Ca2+-activation profiles of contraction as those seen in intact

troponin T [23, 26] This conclusion has also been further

confirmed under more physiological conditions as follows

There are several isoforms oftroponin T in the muscle fibers,

in which the variety is restricted in the N-terminal region [27,

28] Regarding the significance of the variable N-terminal

region of troponin T isoforms, the relative content of the

specific isoforms of t.oponin T (troponin T2f) as well astropomyosin (aa-tropomyosin) has been indicated to correlatewith the higher cooperativity ofthe Ca2+-activated contraction

of the skinned fibers [29] Based on this finding, it has beenproposed that the N-terminal region oftroponin T affects thecooperative property of the Ca2+-activation of contractionthrough the interaction with the C-terminal region of tropo-myosin including the end-to-end connecting portion, for theend-to-end interaction oftropomyosin is known to be involved

in the cooperativity of the Ca2+-activation [30] Detailedstudies using the replacement technique specific for troponin

within the myofibrillar lattice in situ [31,32], however, have

revealed that the Ca2+-activation profiles, in terms ofboth the

Ca2+-sensitivity and its cooperativity, do not change even after

Fig.3 Arrangement oftroponin T, and troponin T, in the troponin-tropomyosin complex Cited from Nagano and Ohtsuki [39] Two domains oftroponin

T (TI' T,) bind to tropomyosin (TM) antiparallelly in relation to the amino acid sequence; the N-terminal troponin T, region occupies the portion oftroponin

T at the C-terminal side of tropomyosin and the C-terminal troponin T, region at the N-terminal side of tropomyosin [7, 15, 17] The shape and size of this model approximately coincide with those observed in the fresh troponin preparations [40].

by weakening the interaction of troponin I with troponin C

in the absence ofCa2+[13]

The essential features ofCa2+-regulation in the thin filamentare the depression by the inhibitory action troponin I and the

Ca2+-dependent removal of this depression by the activating(de-inhibitory) action oftroponinC.However, the contraction

is activated regardless of the Ca2+-concentration in theabsence of troponin T (Fig Ib) Troponin T is the in-dispensable component for the inhibition of contraction bytroponin complex-tropomyosin in the absence ofCa2+ Therole of troponin T is thus thought to integrate the two

Ca2+-independent actions oftroponin I and C into the Ca2dependent function

+-Fig.4 Ca 2 '-reguJation of the interaction of the troponin components and actin-tropomyosin (A) The thin filament (actin-tropomyosin-troponin C·IT filament) Actin-tropomyosin is inhibited by troponin I in the absence of

Ca 2' and the inhibition by troponin I is removed in the presence of Ca 2',

(B) The actin-tropomyosin-troponin C·I filament (without troponin T) Actin

is not inhibited by troponin I in the absence and presence of Ca 2', though troponin I binds to actin-tropomyosin only in the absence of Ca 2'.

Abbreviations: TM - tropomyosin; C, I and T - troponin C, I and T.

Conclusion

the exchange of the endogenous troponin T2fwith another

isofonn (troponin T1f),and vice versa, in the myofibrils [33]

and also after the exchange ofthe whole endogenous troponin

T with its fragments devoid ofthe N-tenninal residues in the

myofibrils [32] and the skinned fibers [34] The N-tenninal

region oftroponin T is therefore by no means involved in the

Ca2+-activation profiles ofcontraction including cooperativity

The troponin T , subfragment itself shows the strong

inhibitory action on actomyosin ATPase activity in the

presence of tropomyosin-troponin I·C regardless of Ca2

+-concentrations [25] But this depression by troponin TIis not

affected by the addition of troponin T2and would thus be

caused through the interaction with tropomyosin, which does

not exist in the thin filament Actually troponin TI

sub-fragment interacts with the C-tenninal region oftropomyosin,

which shows no affinity to the troponin complex [13, 35]

This strongly suggests that the C-tenninal region of

tropo-myosin is out of the range for the troponin T-binding under

physiological conditions

Ca2 +-regulation in the thin filament

In the absence of Ca2+,actin in the thin filament is inhibited

from interacting with myosin for contraction This relaxed

state is caused by the inhibitory action of troponin I on

actin-tropomyosin and then is released for contraction on

Ca2+-binding to troponinC.The inhibitory activity is localized

in the relatively small region (inhibitory region; residues

96-116) oftroponin I of 178 residues from rabbit skeletal muscle

[36] This inhibitory region fragment also binds to troponin

C and its affinity is potentiated by the Ca2+-binding to

troponinC.The inhibitory interaction ofthe inhibitory region

in troponin I with actin-tropomyosin predominates in the

absence of Ca2+,while, in the presence of Ca2+, the affinity

of the inhibitory region of troponin I to troponin C prevails

over its affinity to actin-tropomyosin and hence actin

mole-cules are released from the inhibited state for the contractile

interaction with myosin (Fig 4)

The Ca2+-dependent competition oftroponin C and

actin-tropomyosin to the inhibitory region of troponin I thus is

considered to be the essential mechanism of the Ca2

+-regulatory pathway in the thin filament These competitive

processes, however, do not operate when troponin T is

removed from the thin filament The binary complex of

troponin C·I (without troponin T) does not confer Ca2

+-sensitivity on actin-tropomyosin, and the contractile

inter-action is activated (de-inhibited) even in the absence ofCa2+

in this situation, though the binary troponin Col complex binds

to actin-tropomyosin in the absence ofCa2+and is dissociated

from actin-tropomyosin in the presence ofCa2+[36, 37] (Fig

4) Troponin T thus makes the inhibitory region oftroponin

I fully available for the interaction with actin-tropomyosin,

Studies on chymotryptic troponin T subfragments have

clarified that the Ca2+-regulating activity is localized in the

troponin T2region, which contains the C-terminus Although

the involvement of the N-terminal region has also been

postulated in relation to the cooperative property ofthe Ca2

+-activated contraction, this possibility was later ruled out

based on more detailed examinations under physiological

conditions The tight coupling ofthe Ca2+-regulating action

of troponin T with its binding to tropomyosin through the

troponin T2region also suggests that the steric arrangement

of troponin components and tropomyosin in the thin

fila-ment is delicately regulated through troponin T, in which

the C-terminal T2region is situated at the pivotal position

and thus plays a crucial role in the Ca2+-regulation of

contraction

Although the contraction of most striated muscles is

regulated by troponin-tropomyosin through Ca2+, the Ca2

+-regulation of contraction in some striated muscles of

in-vertebrates such as molluscs is myosin-linked, where the

action of Ca2+ on the essential light chain (Ca2+-binding

subunits) is mediated to the active site ofheavy chain within

the myosin molecule [37] The contraction is activated

regardless of Ca2+concentration when the regulatory light

chain (RLC) is removed from myosin, indicating that RLC

is the inhibitory subunit The myosin-linked Ca2+-regulatory

system, however, does not contain a subunit, of which

function corresponds to that of troponin T No protein

factors with high affinity to tropomyosin are also apparently

involved in the Ca2+-regulation of smooth muscle

con-traction Troponin T is the truely regulatory component

specific for the troponin-linked regulation The detailed

examination of the regulatory properties oftroponin Twill

certainly lead to the clarification of the characteristic

mechanisms ofthe Ca2+-regulation by troponin-tropomyosin

under physiological conditions

Acknowledgement

Itis my great pleasure to dedicate this article to Professor S

Ebashi, who discovered troponin and inaugurated a new era

of the molecular biology of the Ca2+-regulation of muscle

contraction I also express my sincere gratitude to Professor

S Ebashi for his thoughtful advice and encouragement

throughout my scientific life

References

I Ebashi S: Third component participating in the superprccipitation of

'natural actomyosin' Nature 200: 1010,1963

2 Ebashi S, Ebashi F: A new protein component participating in the

superprecipitation of myosin B J Biochem 55: 604-613, 1964

37

3 Ebashi S: Calcium ion and muscle contraction In: K Maruyamaet.

ai (eds) Calcium as cell signal Igaku-shoin Tokyo-New York, 1995,

II Greaser ML, Gergely J: Reconstitution oftroponin activity from three protein components J BioI Chern 246: 4226 4233, 1971

12 Ebashi S: Separation oftroponin into its three components J Biochem 72: 787-790, 1972

13 Ohtsuki I, Maruyama K, Ebashi S: Regulatory and cytoskeletal proteins

of vertebrate skeletal muscle Adv Prot Chern 38: 1 67,1986

14 Pearlstone JR, Carpenter MR, Johnson P, Smillie LB: Amino acid sequence oftropomyosin-binding component ofrabbit skeletal muscle troponin Proc Natl Acad Sci 73: 1902-1906,1976

15 Ohtsuki I: Molecular arrangement oftroponin T in the thin filament J Biochem 86: 491-497,1979

16 Ebashi S, Ohtsuki I, Mihashi K: Regulatory proteins of muscle with special reference to troponin Cold Spring Harbor Symp Quant Bioi 37: 215-224, 1973

17 Ohtsuki I: Distribution of troponin components in the thin filament studied by immunoelectron microscopy J Biochem 77: 633 639, 1975

18 Tanokura M, Tawada Y, Ohtsuki I: Chymotryptic subfragments of troponin T from rabbit skeletal muscle I Determination ofthe primary structure J Biochem 91: 1257-1265,1982

19 Tanokura M, Tawada Y, OnoA, Ohtsuki I: Chymotryptic subfragments oftroponin T from rabbit skeletal muscle Interaction with tropomyosin, troponin I and troponin C J Biochem 93: 331-337, 1983

20 Jackson P, Amphlett GW, Perry SV: The primary structure oftroponin

T and the interaction with tropomyosin Biochem J 151: 85-97, 1975

21 Nagano K, Miyamoto S, Matsumura M, Ohtsuki I: Possible formation

of a triple-stranded coiled-coil region in tropomyosin-troponin T binding complex J Mol Bioi 141: 217-222, 1981

22 Endo T, Matsumoto K, Hama T, Ohtsuka Y, Katura G, Obinata T:

Distinct troponin T genes are expressed in embryonic larval tail striated muscle and adult body wall smooth muscle of ascidian J Bioi Chern 271: 27855-27862,1996

23 Ohtsuki I, Shiraishi F, Suenaga N, Miyata T, Tanokura M: A 26K fragment of troponin T from rabbit skeletal muscle J Biochem 95: 1337-1342,1984

24 Ohtsuki I, Yamamoto K, Hashimoto K: Effect of two C-terminal side chymotryptic troponin T subfragments on the Ca 2 +-sensitivity of superprecipitation and ATPase activities of actomyosin J Biochem 90: 259-261, 1981

25 OnoyamaY, Ohtsuki I: Effect of chymotryptic troponin T subfragments

on the calcium ion-sensitivity of ATPase and superprecipitation of actomyosin J Biochem 100: 517-519, 1986

26 Pan B-S, Gordon AM, Potter JD: Deletion ofthe first 45 NH2-terminal residues of rabbit skeletal troponin T strengthens binding of troponin

to immobilized tropomyosin J Bioi Chern 266: 12432-12438, 1991

27 Wilkinson JM, Moir AJ, Waterfield MD: The expression of multiple

forms of troponin T in chicken fast skeletal muscle may result from

differential splicing ofa single gene Eur J Biochem 143: 47-56, 1984

28 Briggs MM, Lin J J-C, Schachat FH: The extent of amino-terminal

heterogeneity in rabbit fast skeletal muscle troponin T J Muscle Res

Cell Mot 8: 1-12,1987

29 Schachat FH, Diamond M, Brandt PW: Effect of different troponin

T-tropomyosin combinations thin filament activation J Mol Bioi 198:

551-554,1987

30 Tawada Y, Ohara H, Ooi T, Tawada K: Non polymerizable tropomyosin

and control of the superprecipitation of actomyosin J Biochem 78:

65-72, 1975

31 Hatakenaka M, Ohtsuki 1: Effect of removal and reconstitution of

tro-ponins C and I on the Ca2+-activated tension development of single

glycerinated rabbit skeletal muscle fibers Eur J Biochem 205:

985-993,1992

32 Shiraishi F, Kambara M, Ohtsuki I: Replacement of troponin

com-ponents in myofibrils J Biochem III: 61-65, 1992

33 Hatakenaka M: The Ca'+-activation profile of rabbit fast skeletal

myofibrils is not affected by troponin T isoforms Biomed Res 17:

95-100, 1996

34 Morimoto S, Yanaga F: Troponin T and its fragment devoid of the N-terminal 51 residues from chicken skeletal muscle show the same Ca'+-regulating action in skinned fiber contraction In preparation.

35 Suenaga N: Effect oftroponin and Troponin T, on the iodination of tyrosyl residues of a-tropomyosin Fukuoka Acta Medica 79: 493-

496, 1988

36 Syska H, Wilkinson JM, Grand RJA, Perry SV: The relationship between biological activity and primary structure oftroponin I from white skeletal muscle of rabbit Biochem J 153: 375-387,1976

37 Potter JD, Gergely J: Troponin, tropomyosin, and actin interactions in the Ca'+-regulation of muscle contraction Biochemistry 13: 2697-

2703, 1974

38 Szent-Gyorgyi AG, Chantler PD: Control of contraction by calcium binding to myosin In: A.G Engel, C Franzini-Armstrong (eds) Myology McGraw-Hill, New York, 1994, pp 506-528

39 Nagano K, Ohtsuki 1: Prediction of approximate quaternary structure oftroponin complex Proc Japan Acad 58: 73-77, 1982

40 Ohtsuki I, Onoyama Y, Shiraishi F: Electron microscopic study of troponin J Biochem 103: 913-919, 1988