Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations

1993 Molecular cloning, expression, and protein interaction of avian muscle titin Kuan Onn Tan Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Biochemistry Commons, and the Molecular Biology Commons

Recommended Citation Tan, Kuan Onn, "Molecular cloning, expression, and protein interaction of avian muscle titin " (1993). Retrospective Theses and Dissertations. 10555. https://lib.dr.iastate.edu/rtd/10555

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. U'M'I MICROFILMED 1994 INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

University Microfilms International A Bell & Howell Information Company 300 North Zeeb Road. Ann Arbor, Ml 48106-1346 USA 313/761-4700 800/521-0600

Order Number 9414024

Molecular cloning, expression, and protein interaction of avian muscle titin

Tan, Kuan Onn, Ph.D.

Iowa State University, 1993

UMI 300 N. Zeeb Rd. Ann Arbor, MI 48106

Molecular cloning, expression, and protein interaction of avian muscle titin

by

Kuan Onn Tan

A Dissertation Submitted to the

Graduate Faculty in Partial Fulfillment of the

Requirements for the Degree of

DOCTOR OF PHILOSOPHY

Department: Biochemistry and Biophysics Major: Biochemistry

Approved: Members of the Committee

Signature was redacted for privacy.

In Charge of Major Work

Signature was redacted for privacy.

For the Major De^tment Signature was redacted for privacy.

Signature was redacted for privacy.

Iowa State University Ames, Iowa

1993 11

TABLE OF CONTENTS

GENERAL INTRODUCTION 1

LITERATURE REVIEW 3

PAPER I. MOLECULAR CHARACTERIZATION OF AVIAN MUSCLE TTTIN 24

ABSTRACT 26

INTRODUCTION 27

MATERIALS AND METHODS 30

RESULTS 35

DISCUSSION 58

ACKNOWLEDGMENTS 63

REFERENCES 64

PAPER II. INTERACTION OF CLONED FRAGMENTS FROM THE A-I JUNCTION REGION OF CHICKEN MUSCLE TITIN WITH MYOSIN AND ACTIN 70

ABSTRACT 72

INTRODUCTION 73

MATERIALS AND METHODS 77

RESULTS 81 iii

DISCUSSION 109

ACKNOWLEDGMENTS 114

REFERENCES 115

OVERALL SUMMARY 121

LITERATURE CITED 129

ACKNOWLEDGMENTS 141 1

GENERAL INTRODUCTION

Titin, a megadalton myofibrillar protein (Maruyama et al., 1976; Wang et al.,

1979), is a major muscle protein that makes up 8-10% of the myofibrillar protein (Trinick et al., 1984; Wang, 1985; Maruyama, 1986). In the muscle, titin may provide structural integrity to the sarcomere, which is the basic repeating unit of striated muscle. In addition, due to its elasticity, titin could confer elastic properties to myofibrils as well as modulate the passive force generation of muscles (Wang et al., 1991; Horowits, 1992),

Furthermore, in developing muscle, titin may play a role in the organization of muscle proteins by serving as scaffolding (Fulton and Isaacs, 1991). In order to better understand the titin molecule, the major objectives of my graduate work were to determine a portion of the primary structure of titin molecule through molecular cloning and to determine the function(s) of the cloned titin fragment. Molecular cloning of titin cDNAs required construction of a cDNA library, screening of the cDNA libarary, sequence analysis, and characterization of the cDNA clones by immunological and molecular methods. To determine the function(s) of cloned titin fragment, restriction fragments of titin cDNAs were expressed in E.coli and individual titin fragments were purified. The cloned titin fragments were then used for interaction studies, in which solid phase binding assays were used to determine possible interactions of the cloned titin fragments with other muscle proteins. 2

Explanation of Dissertation Format

This dissertation contains two manuscripts. The first manuscript was prepared according to the guidelines of The Journal of Biological Chemistry and this manuscript has been published (Tan et al., 1993). The second manuscript was prepared according to guidelines of a second journal for possible submission in the future. The majority of the work in the first paper was done by myself, with the exception of immunofluorescence experiments, which were performed by Dr. Ted W. Huiatt, and the initial isolation of a

0.2 kb titin cDNA clone, which was done by Dr. Greg R. Sater while working as a graduate student in Dr. Huiatt's laboratory. The work reported in the second paper was done solely by myself. The organization of the dissertation includes this General

Introduction, a brief Literature Review covering muscle proteins in general and a specific review of the literature related to titin, Papers I and II, and an Overall Summary.

Literature cited in this General Introduction, the Literature Review, and the Overall

Summary follows the Overall Summary. 3

LITERATURE REVIEW

Myofibrillar Proteins

In the myofibrils of striated muscle cells, actin and myosin are the major components of the thin and thick filaments, respectively, which are involved in force generation during muscle contraction according to the generally accepted sliding filament

model of contraction (Huxley and Hanson, 1954; Huxley and Niedergerke, 1954). In striated muscle, such as skeletal muscle, thick and thin filaments are organized into a

regular repeating pattern that gives the muscle a striated appearance in the phase-contrast or electron microscope. The basic repeating unit of myofibrils is called a sarcomere.

The dark band in the center of each sarcomere is the A-band, which is the region containing the thick, myosin-containing filaments. The region from the end of the A-band in one sarcomere to the start of the A-band in the next sarcomere is the I-band, which is

made up of the thin, actin-containing filaments. The actin filaments are anchored at the

Z-line, which is at the center of I-band. The thin filaments run in parallel from the Z- line, through the I-band, and into the A-band, where they interdigitate with the thick

filaments. Other muscle proteins are closely associated with either thick or thin filaments at certain regions along the filaments. Although some of these additional proteins have been purified and cloned, in many cases their functions remain elusive. Other than myosin, the functions of thick filament-associated proteins are not known. The functions of some of the proteins that are associated with thin filaments, such as tropomyosin and troponins I, T and C, have been determined. 4

From the amino acid sequence, actin is a molecule of 43 kDa (Elzinga et al.,

1973). Each molecule of G-actin (monomeric actin) contains a bound ATP and a bound cation (Mg^^ or Ca^^). Actin molecules (G-actin) can undergo polymerization to form actin filaments (F-actin) in the presence of salt (KCl or MgCy. F-actin consists of a two stranded helix, with crossover points of the two strands every 35 to 38 nm along the filaments (Hanson and Lowy, 1963). This arrangement of the actin monomers in F-actin forms two long grooves running along the F-actin filaments. In muscle thin filaments, these grooves are occupied by the elongated molecule, tropomyosin (Ebashi, et al., 1969),

Actin filaments exhibit polarity as shown by binding of myosin heavy meromysin (HMM) to the filaments, resulting in an arrowhead appeareance in micrographs of negatively stained preparations (Huxley, 1963). The thin filaments in muscle all have the same orientation, with the so-called "barbed end" of F-actin located at the Z-line end.

At least three other proteins are associated with F-actin in muscle thin filaments.

Tropomyosin is present in the long grooves of F-actin as mentioned before. This molecule consists of two a-helical chains and that twist around each other to form a coiled-coil structure. The structure is stabilized by hydrophobic interactions of nonpolar amino acid residues present at points of contact between the two a-helical chains. The amino acid sequence of tropomyosin shows the presence of two hydrophobic residues in each seven-residue repeat or heptad (Hitchcock-DeGregori and Varnell, 1990). These residues occur at a regular interval of three and then four amino acids throughout the molecule; therefore, two hydrophobic residues from each heptad in one chain can come in contact with two hydrophobic residues from each heptad in the other chain along the 5 intertwined tropomyosin molecule. The tropomyosin molecule can be further divided into seven pseudo-repeats that fulfill the requirement for the binding of each tropomyosin molecule to seven actin monomers (Stewart and McLachlan, 1975). According to the steric blocking model for Ca^^ regulation of contraction, the function of tropomyosin is to block the interaction of actin and myosin, and thus block active muscle contraction, in the absence of calcium ion (Moore et al., 1970; Parry and Squire, 1973).

Troponin, which consists of three subunits, troponin-C (TN-C), troponin-I (TN-I), and troponin-T (TN-T), is also associated with thin filaments. TN-C has four calcium binding sites and interacts with TN-I and TN-T (See Squire, 1986 for review). In addition, TN-T and TN-I also interact with tropomyosin and actin, respectively. Binding of TN-I to actin inhibits the actin-myosin ATPase in vitro', however, the inhibitory effect seems to be modulated through tropomyosin in vivo. In the presence of calcium ion, TN-

C binds Ca^+, releasing the inhibitory effects of TN-I and tropomyosin on actin, either directly through binding to TN-I or indirectly through the participation of TN-T which interacts with TN-C and tropomyosin. Once the inhibitory effects are removed, myosin heads can attach to actin filaments to generate active muscle contraction.

Thin filaments are also associated longitudinally with a large protein, nebulin

(Wang and Williamson, 1980; Wang, 1981; Wang and Wright, 1988), which has an estimated molecular weight of 800 kDa (Stedman et al., 1988; Hu et al., 1989). Nebulin is a family of proteins with different size isoforms in different skeletal muscle fiber types.

The function of the protein may be to serve as a template for the elongation of the F-actin filaments and thus, the length of muscle thin filaments may be determined by the size of nebulin. Correlations have been observed between the size of nebulin and length of thin filaments from different muscles (Kruger et al., 1991; Labeit et al., 1991). In cardiac

muscle, where nebulin is absent, the lengths of the thin filaments appear to be heterogenous (Robinson and Winegard, 1979). The partial cDNA sequence of nebulin shows a repeating pattern of motifs consisting of 35 amino acid residues (Wang et al.,

1990; Labeit et al., 1991). In addition, self comparision of the sequence has identified super-repeats of seven motifs. The seven motif super-repeat may bind to seven actin monomers along the actin filament. The cDNA sequence of nebulin predicts secondary structure rich in a-helices, which could allow nebulin to interact with the long-pitch helix of F-actin filaments (Trinick, 1992). This idea is supported by results showing the interactions of cloned nebulin fragments with actin in vitro (Jin and Wang, 1991a, b).

In the I-band, F-actin filaments are anchored at the Z-line. The major known protein component of Z-line is a-actinin, which is a dimer of 102 kDa subunits (Suzuki et al., 1976) and is likely to be the protein component of the Z-filaments (Yamaguchi et al.,

1983, 1985). Since a-actinin cross-links F-actin in vitro (Goll et al., 1972), it is likely that a-actinin also cross-links F-actin filaments from adjacent sarcomeres at the Z-line

(Suzuki et al., 1976; Yamaguchi et al., 1985). However, other protein(s) may also be involved in the interaction between a-actinin and F-actin (Goll et al., 1991). Several other proteins, such as zeugmatin (Maher et al., 1985), may also be Z-line components, but the exact composition of the entire Z-line structure is not known.

The major component of thick filaments is myosin. Myosin is an elongated molecule with a tail and two globular heads. A single myosin molecule has a molecular 7

mass of approximately 500 kDa and is made up of six subunits, two heavy chains (230

kDa each) and four light chains (about 20 kDa each) (Lowey, 1979; Harrington, 1979).

The heavy chains make up the myosin tail and most of the two heads. The light chains

are bound to the heads of myosin in such a way that two light chains are associated with

each myosin head. The light chains can be classified in two classes, regulatory and

essential, and each head contains two light chains, one member of each class. The light

chains serve to modulate the basic function of myosin, which is both an ATPase and an

actin binding protein. The domains that are responsible for these functions are localized

to the head regions of the myosin molecules. The rod or tail domain of myosin is

involved in formation of myosin filaments.

Proteins that are associated with thick filaments include skelemin (Price, 1987), M-

protein (Masaki and Takaiti, 1974), myomesin (Grove et al., 1984), creatine kinase

(Wegmann, 1992), 86 kDa protein (MyBP-H) (Bahler et al., 1985a, 1985b; Vaughan et

al., 1993a, 1993b), C-protein (MyBP-C) (Offer et al., 1973; Craig and Offer, 1976;

Vaughan et al., 1993a), and titin (Furst et al., 1988, 1989b). Skelemin is located at the

periphery of the M-line in mammalian striated muscle and may be involved in linking

adjacent myofibrils at the M-line. M-protein, creatine kinase, and myomesin are all

located in the M-line at the center of the A-band. All three of these proteins appear to be

structural components of the M-line, but their exact arrangement in the M-line structure is

not clearly known. The M-line appears to be involved in linking together adjacent thick

filaments. The 86 kDa protein (MyBP-H) and C-protein (MyBP-C) are located in a series of 9 and 7 stripes, respectively, spaced 43 nm apart, in the middle one-third of each half 8 of the A-band, in the region of the thick filaments known as the C-zone. In general, C- protein (MyBP-C) and 86 kDa protein (MyBP-H) are co-localized in 7 of the 9 stripes, and 2 of the stripes contain only 86 kDa protein (MyBP-H), but the exact distribution is dependent on both species and muscle type. Titin is located along the length of the thick filaments, as described in the following section of this review. Although all of these proteins are localized at defined regions of thick filaments, their precise functions are not known. However, cDNA sequences have shown that, with the exception of creatine kinase, all of these myosin associated proteins, including C-protein, 86 kDa protein, M- protein, skelemin, and smooth muscle and non-muscle myosin light chain kinases, contain repeating patterns of the class I and class II motifs described below for the protein titin.

Characteristics of Titin

Titin, an extremely large myofibrillar protein, was first reported by Wang et al.

(1979) and Maruyama et al. (1976). Titin, also called connectin (Maruyama et al.,

1981), is relatively abundant, comprising approximately 10% of myofibrillar mass

(Trinick et al., 1984; Wang, 1985; Maruyama, 1986). Therefore, titin is the third most abundant muscle protein after myosin (43%) and actin (22%) (Maruyama, 1986). Titin appears as two closely spaced bands (designated T1 and T2) on low percentage acrylamide

SDS-PAGE. T2 is considered to be a proteolytic product of T1 (Wang, 1985; Maruyama et al., 1986). Due to its large size, the molecular weight of titin has not been precisely determined; however, based on equilibrium sedimentation measurements, Kurzban and

Wang (1988) have estimated the molecular weights of T1 and T2 to be 2.4-2.6 x 10® Da and 2.1-2.4 x 10® Da, respectively. Similar estimates have also been reported by 9

Maruyama et al. (1984b). Intact titin is highly insoluble and requires denaturing solvents

for extraction (Wang, 1982); however, T2 has been purified under nondenaturing

conditions (Trinick et al., 1984; Maruyama et al., 1984b; Wang et al., 1984). Purified

T2 has an appearance similar to a string of beads, where the spacing between successive

beads is about 4 nm in electron micrographs (Trinick et al., 1984). Further studies of

purified T2 showed that the molecules are extremely long with a length up to 1 ixm and a diameter of 4-5 nm (Wang et al., 1984). In addition, Nave et al. (1989) have reported

the isolation of T2 molecules that appear, in rotary shadowed preparations, to be extended

molecules with a globular head at one end, and have suggested that the globular head may be composed of M-line proteins. In combination with other antibody localization studies,

these results show that the breakdown product of titin, T2, is anchored at or near the M- line in myofibrils. Circular dichroism has estimated 60% P-sheet and 30% P-tum for

both T1 and T2 (Kimura et al., 1992; Maruyama et al., 1986). Raman spectroscopy and infrared dichroism studies also showed similar results (Uchida et al., 1991).

Both immunological and physical studies have indicated that titins isolated from cardiac and skeletal muscles have similar properties (Pierobon-Bormioli et al., 1992; Itoh et al., 1986). Minor differences in the amino acid composition between cardiac and skeletal titin have been noted (Itoh et al., 1986). In addition, some monoclonal antibodies

were able to differentiate cardiac and skeletal titin or titins of different animals (Hill and

Weber, 1986). The significance of these differences is not known. More importantly, different size isoforms of titin have been identified (Wang et al., 1991; Horowits, 1992), and the authors have shown that titins isolated from different muscles differed in mobility 10 on SDS-PAGE in low percentage polyacrylamide gels. In addition, the authors have found possible relationships between the size of isolated titin and tension developments of muscles (see below on "Possible Functions of Titin in Muscle").

Titin is a phosphoprotein containing up to 12 moles of phosphate per mole of titin

(Somerville and Wang, 1987). In vivo phosphorylation studies have shown that two moles of serine residues of titin could be phosphorylated (Somerville and Wang, 1988). In contrast, in vitro phosphorylation of purified titin with several purified kinases has not been successful; however, autophosphorylation of titin has been reported (Takano-Ohmuro et al., 1992). Autophosphorylation of arthropod projectin, a ti tin-related protein, was also reported (Maroto et al., 1992).

Other than the breakdown product, T2, titin can be easily degraded into smaller fragments during postmortem storage. Matsuura et al. (1991) have reported a 1200 kDa degradation product of titin. Polyclonal antibodies generated against this degradation product recognized titin epitopes near the N, line, which is a structural feature of the I- band often seen in phase-contrast or electron micrographs. Therefore, the 1200 kDa polypeptide represents the portion of titin near or at the Z-line. In addition, Takahashi et al. (1992) have reported that degradation of titin could be induced by incubating myofibrils with calcium to obtain a 1200 kDa degradation product along with P-connectin

(T2). They also reported the binding of calcium to P-connectin and proposed that titin could change its elasticity by responding to the physiological concentration of calcium ions. 11

Localization Studies of Titin

Most of the studies on localization of titin in the sarcomere have been accomplished by the use of antibodies at the light and electron microscopy levels. These studies were made possible by purification of titin, and the subsequent production of monoclonal and polyclonal antibodies directed against the purified molecules.

Immunofluorescence studies have shown that most polyclonal antibodies to titin strongly labeled myofibrils at A-I junction (Fulton and Isaacs, 1991). Labeling of the gap between thin and thick filaments in overstretched muscle was also reported (LaSalle et al., 1983;

Maruyama et al., 1984a; Maruyama et al., 1985). At the ultrastructural level, both polyclonal and monoclonal antibodies specific for titin labeled titin from Z-line to regions close to M-line (Maruyama et al., 1985; Wang, 1985; Furst et al., 1988; Itoh et al.,

1988; Trombitas et al., 1991). Furst et al. (1988, 1989b) have generated a map of titin localization in the sarcomere by using monoclonal antibodies that labeled ten nonrepetitive epitopes starting at the Z- line, and covering both the I- and A-bands. In the A-band, one monoclonal antibody labeled 0.2 ixm before the M-line (Furst et al., 1988). These results demonstrated that titin is located at regions in the A-band, I-band, and Z-line of the sarcomere. A single titin molecule appears to extend over the entire distance from M-line to the Z-line. In addition, monoclonal antibodies have also detected seven repetitive epitopes of titin in the A-band region (Furst el al., 1989b) in which the 42 nm spacings of the repetitive epitopes coincided with the known positions of C-protein (MyoBP-C) (Craig and Offer, 1976) and 86 kDa protein (MyoBP-H) (Bahler et al, 1985b). Therefore, the repetitive epitopes of titin may be required for interactions with these myosin-associated 12 proteins. Besides this immunological evidence, additional evidence also points to the association of titin with myosin filaments and the extension of titin from the thick filaments to the Z-line. The association of titin with myosin filaments was demonstrated by partial removal of myosin filaments with a high concentration of KCl (Higuchi et al.,

1992). In these experiments, titin antibodies specifically labeled at a position of 0.8 /tm from the M-line before extraction. The position of labeling moved to 0.3 /am from the Z- line when thick filaments were depolymerized to a length of less than 0.4 fim; therefore, the association of titin and myosin was destroyed by the removal of myosin filaments as reflected by the movement of the titin epitopes from the A-band to the Z-line. Additional evidence for extension of titin from the thick filaments into the I-band region was provided by Funatsu et al. (1990). These investigators used plasma gelsolin to remove thin filaments to uncover titin in the I-band region. After removal of thin filaments, titin molecules appeared to connect the edges of thick filaments through the I-band and into the

Z-line region. More recently, detailed organization of titin (connectin) in cardiac myofibrils was visualized with deep-etch replica electron microscopy after removal of both actin and myosin filaments by gelsolin and potassium acetate, respectively (Funatsu et al., 1993). Cardiac myofibrils were used to avoid confusion of nebulin for titin or vice versa because nebulin is not present in cardiac muscle. In this study, titin filaments were reported to project out of the M-line and to run in parallel into the I-band region. At a region corresponding to the N; line of the I-band, individual titin filaments branched into thinner strands. These branched titin filaments were also joined adjacently along certain length and then separated at regions that ran into the Z-lines (Funatsu et al., 1993). The 13 authors also suggested that titin filaments may be weakly and randomly associated with actin filaments in the I-band region in intact myofibrils. Interactions of titin and actin filaments have been reported (Kimura et al., 1984a; Maruyama et al., 1987). Together, these results confirm early observations of "gap filaments", which were found in the gap region between thick and thin filaments of highly-stretched muscle fibers (Huxley and

Peachey, 1961; Sjostrand, 1962; Locker and Leet, 1975, 1976), and "end filaments", located at the ends of isolated thick filaments (Trinick, 1981).

Since titin appears to connect thick filaments to the Z-line, titin may be responsible

I for the elastic behavior of myofibrils. To demonstrate that behavior, several investigators have used monoclonal antibodies specific for certain parts of titin, such as A-band and I- band epitopes, to label stretched myofibrils and control myofibrils that were not stretched.

In stretched myofibrils, antibody labeling of titin in the I-band moved as myobrils were stretched (Itoh et al., 1988; Horowits et al., 1989; Trombitas et al., 1990). The labeling position returned to its original position when stretched myofibrils were released to resting length (Maruyama et al., 1989). In comparison, antibody labeling of titin epitopes in the

A-band was not affected by stretching the myofibrils (Itoh et al., 1988). The movement of antibody labeling positions in the I-band provides conclusive evidence for the elasticity of the I-band portion of titin. Although these results do not support elasticity of titin in the A-band, elasticity of titin was reported when myosin filaments were partially depolymerized with a high concentration of KCl (Higuchi et al., 1992). In these experiments, the authors have demonstrated changes of titin antibody labeling positions at the A-band as sarcomere length changed when thick filaments were partially 14

depolymerized. They suggested that titin molecules are elastic throughout their entire

length. This suggestion may be in general agreement with the observation that T2 showed

a homogeneous appearance of beads on a string in electron micrographs (Trinick et al.,

1984; Maruyama et al., 1984b, Wang et al., 1984). It is likely that the elasticity of titin cannot manifest itself when titin is tightly associated with myosin filaments in the A-band

region (Whiting et al., 1989). The elasticity of titin molecules in the I-band region, from

the A-I junction to N, line (Trombitas et al., 1990), may be responsible for the functional

implication of elasticity. Recently, the portion of titin located from N,-line to Z-line (N,-

Z region) was found to be elastic (Trombitas and Pollack, 1993); however, this elasticity

was revealed only when sarcomere was extremely stretched. Thus, the N,-Z region of

titin is functionally stiff and may not contribute to elasticity of titin in muscle, like the A-

band portion of titin.

Possible Functions of Titin in Muscle

In order to study the function of titin molecules, several investigators have tried to destroy titin molecules in myofibrils by means of ionizing radiation or enzymatic digestion. In experiments where titin and nebulin were predominantly destroyed by radiation (Horowits et al., 1986), both active tension in response to calcium and passive

tension in response to stretch were greatly reduced. Further studies have shown that, in isometrically contracted muscle when sarcomere lengths were less than 2,8 fim, A-bands normally moved toward Z-lines (Horowits and Podolsky, 1987). In addition, titin antibody that labeled at the I-bands also moved in the same direction as the A-band during

A-band movement (Horowits et al., 1989). However, at sarcomere lengths greater than 15

2.8 fim, when titin was supposedly stretched, the A-band was perfectly centered during contraction (Horowits and Podolsky, 1987). Together, these results lend support to titin's role in providing resting tension to the sarcomere as well as in keeping the A-band centered during contraction. Selective digestion of titin with low concentrations of trypsin also provided similar results, in that both resting and active tensions were reduced

(Higuchi, 1992). However, the relative resting tension was shown to decrease faster than the active tension. Since muscles differ in degree of elasticity and resting tension upon stretch, Wang et al. (1991) have studied the possible correlation between titin isoforms

(titin molecules that differ in size) and passive force generation in muscles. By isolating titin molecules from different rabbit muscles and measuring resting tension at different sarcomere length, Wang el al. (1991) have shown that muscles that express larger titin molecules tend to initiate tension and reach elastic limits at higher sarcomere lengths.

Muscles with larger titin isoforms also developed the lowest tension. Similarly, Horowits

(1992) has reported that rabbit soleus muscle, which has larger titin isoform than psoas muscle, developed lower resting tension than psoas muscle. These results show that passive force generation in muscles may be modulated by expressing different titin isoforms.

Titin may be also important in myofibrillogenesis, which is the de novo assembly of myofibrils in developing muscle. Although the exact sequence of events of myofibrillogenesis has not been clarified, the early presence of titin during myofibrillogenesis has prompted its consideration for an important role (see Fulton and

Isaacs, 1991; Epstein and Fischman, 1991 for review). The possible roles of titin in 16 myofibrillogenesis include serving as scaffolding structure for organization of other muscle proteins in early myofibrillogenesis and integrating the thick and thin filaments in later myofibrillogenesis. Myofibrillogenesis studies have tried to establish the association of titin molecules with other thick and thin filament proteins. However, due to the low resolution of light microscopy used in these studies, conflicting results have been reported.

Early expression of titin has been detected in round postmitotic myoblasts of embryonic chick skeletal muscle in culture (Colley et al., 1990). Round postmitotic myoblasts represent an early stage of myoblast differentiation in comparison to spindle- shaped myoblasts (Colley et al., 1990). Expression of titin preceded the expression of myosin heavy chain, and titin molecules were found in a perinuclear distribution along with myosin heavy chain, zeugmatin, and a-actinin in immunofiuorescent studies. The colocalization patterns of these proteins were not affected by Triton X-100 extraction; therefore, these proteins might form a stable complex. Titin molecules were also found near membranes or in the cytoplasm of round post-mitotic myoblasts; however, the presence of titin molecules at these sites was found predominantly in spindle-shaped postmitotic myoblasts. It is likely that a transition from perinuclear distribution to distribution near the membrane or in the cytoplasm occurred as the number of spindle- shaped myoblasts increased, along with increased expression of muscle specific proteins.

Other in vitro studies of myofibrillogenesis have focused on postmitotic spindle-shaped myoblasts or mature myotubes. By using multiple antibodies specific to muscle proteins, the location of titin molecules with respect to other muscle proteins was studied during skeletal and cardiac myofibrillogenesis (Handel et al. 1989, 1991; Komiyama et al., 1990; 17

Schultheiss et al., 1990). Information obtained from these studies suggest an association of titin with either myosin heavy chains (MHC) or I-Z-I complexes in early myofibrillogenesis.

Association of titin with MHC has been reported during myofibrillogenesis in postmitotic mononucleated myoblasts in chick skeletal muscle cell cultures. Both synthesis of titin and MHC are simultaneously initiated and the two proteins are colocalized in nonstriated and later, striated myofibrils (Hill et al., 1986; Huiatt et al.,

1989). The observation was further supported by immunofluorescence studies of colcemid and taxol treated myoblasts. Myoblasts treated with colcemid delayed orderly integration of thick and thin filaments, and individual I-Z-I complexes without associated 1.6 /xm long thick filaments were frequently observed. On the other hand, taxol induces formation of laterally aligned 1.6 urn long thick filaments without thin filaments or Z bands. Titin molecules were colocalized with MHC in these chemically treated myoblasts. These observations lend support to the hypothesis that myofibrillogenesis involves the formation of two semi-autonomous organizing centers, the Z-line and the A-band. In addition, biochemical studies of titin by chemical crosslinking, immuno-precipitation, and pulse- labeling have demonstrated the association of titin and myosin (Isaacs et al., 1989; Isaacs et al., 1992) in developing chick skeletal muscle. Komiyama et al. (1990) have also reported the appearance of I-Z-I complexes positive for a-actinin, troponin C and a-actin prior to appearance of MHC and its associated proteins, including titin. Titin was then assumed to play a major role in integrating myosin filaments with the pre-existing I-Z-I complexes. On the other hand, Schultheiss et al. (1990) have localized the Z-line epitope 18 of titin in I-Z-I-like complexes in nonstriated myofibrils and ectopic patches containing a- actin and troponin I. However, A-I junction epitope of titin was localized in regions adjacent to, rather than part of, the I-Z-I-like complexes. These observations were interpreted to suggest that titin molecules are linked to I-Z-I-like complexes through their

Z-line epitopes, whereas the A-I junction epitopes are free from association with I-Z-I like complexes. The free end of titin molecules may interact with or capture nearby thick filaments. Localization of MHC on non-striated myofibrils are variable in that MHC is absent or located many micrometers away from the I-Z-I-like complexes.

Other reports of myofibrillogenesis emphasized the importance of SLFS (stress fiber-like structures) for early myofibrillogenesis and the involvement of titin in SLFS for possible organization of other muscle proteins. Results from both skeletal and cardiac myofibrillogenesis in culture (Handel et al., 1989, 1991) have demonstrated the early association of thin filament components such as actin (a- and y-actin), and tropomyosin with titin on SFLS. In skeletal myofibrillogenesis, MHC exhibited variable association with the SFLS and the Z-line structure appeared later on the SFLS. Myofibrillogenesis in the mouse embryos has shown the sequential appearance of desmin, titin, muscle-specific actin and MHC, and nebulin (Furst et al., 1989a). Therefore, the appearance of titin preceded that of other muscle proteins other than desmin. Early expression of desmin and titin in mouse embryos was also reported by Schaart et al. (1989). It is not clear whether the discrepancies between in vivo and in vitro results were due to differences in animals or to developmental conditions of the muscle cells. In addition, SFLS have only been observed in cultured cells. In combination, these results suggest an important role for 19 titin in myofibrillogenesis, but the exact function of titin in this process remains to be determined.

Molecular Structure of Titin

Before the cDNA sequence of titin was known, the gene organization and DNA sequence of twitchin, the unc-22 gene product in C.elegans, was reported (Benian et al.,

1989). Twitchin is associated with myosin filaments. Mutants lacking this protein do not develop normal body muscle contraction in that the worms are constantly twitching in the absence of active muscle contraction (Benian et al., 1989). The unc-22 gene encodes a protein of 753,494 daltons (Benian et al., 1989, 1993) with repetitive domains I and II that are homologous to fibronectin type III and immunoglobulin superfamily C-2 set domains, respectively. The repetitive pattern of these domains is I-I-II. The presence of these motifs in twitchin and titin explained the cross-reactivity of a titin monoclonal antibody to twitchin (Matsuno et al., 1989).

Domains or motifs I and II were later found in the amino acid sequence derived from the cDNAs of rabbit skeletal muscle titin (Labeit et al., 1990). The reported sequence of titin was from the A-band portion of titin molecules and the sequence showed a repeating pattern of motifs I and II, I-I-I-II-I-I-I-II-I-I-II (Labeit et al, 1990, 1992;

Trinick, 1991). For rabbit titin, copies of motif I shared an average of 40% identity. In comparison, copies of motif II shared 30% identity (Trinick, 1991). The conservation of amino acid residues in motif I is evident by the presence of two conserved regions nearer to N- and C-terminals of the motif. These conserved regions are separated by a variable region in the middle of the motifs. Amino acid residues in motif II are not as conserved 20 as those of motif I. In addition to motifs I and II, both twitchin and titin have a kinase domain at or near the carboxyl terminus. The significance of the kinase domain is not yet entirely clear. Recently, additional sequence from the C-terminus of titin molecules shows the presence of KSP (amino acid sequence, lysine-serine-proline) motifs that could be phosphorylated in muscle cells (Gautel et al., 1993).

In addition to twitchin, other titin or titin-related proteins have been reported in other invertebrate muscles (Nave et al., 1990, 1991). Lakey et al. (1990) have isolated a

800 kDa protein, p800, from muscle of Lethocerus indiens and obtained a partial DNA sequence of this protein. The derived amino acid sequence shows repeating motifs I and

II similar to those found in twitchin and titin. pSOO has been localized to ends of thick filaments and Z-discs of asynchronous flight muscle; therefore, it is likely to connect thick filaments to Z-discs. In synchronous muscle such as leg muscle, pSOO is localized along the thick filaments, similar to the localization of twitchin in C. elegans. Projectin, a titin- related protein in Drosophila, has similar distribution in both synchronous and asynchronous muscles. Due to sequence homology, primers generated from DNA sequences of p800 and twitchin have been used successfully in PGR cloning of either the cDNA or gene of projectin (Ayme-Southgate et al., 1991; Fyrberg et al., 1992). As expected, motifs I and II, with the repeating pattern of I-I-II, were also found in projectin.

An additional feature found in projectin and twitchin is the presence of a leucine residue in position 10 of every other copy of motif I. In comparison, copies of motif I in vertebrate titin lack the leucine residue. Since it is conserved in the invertebrates, the 21

leucine residue may play an important role in the invertebrate titins. Other than in

insects, projectin was also found in crayfish claw muscle (Hu et al., 1990).

Although motifs I and H are found in titin or titin-related molecules of vertebrates and invertebrates, these motifs are not limited to these high molecular weight proteins.

These motifs are also found in proteins that are much smaller than titin. A common

feature of these proteins is their association with myosin filaments. To date, motifs I and

II have been found in the myosin-associated proteins C-protein (MyBP-C) (Einheber and

Fischman, 1990, Furst et al., 1992; Weber et al., 1993), 86 kDa protein (MyBP-H)

(Vaughan et al, 1993a and 1993b), M-protein (Noguchi et al., 1992), skelemin (Price and

Gomer, 1993), smooth muscle and non-muscle myosin light chain kinases (Olson et al.,

1990; Shoemaker et al., 1990), and 165 and 190 kDa proteins (165 kDa protein is most likely to be M-protein, Vinkemeier et al., 1993). The arrangement of motifs I and II in

these proteins are different from each other. The different arrangement of motifs may be due to the different association of these proteins with myosin filaments. The function of

motifs I and II may also be structural, because smooth muscle myosin light chain kinase

has both catalytic and regulatory domains flanked by the motifs. Motifs I and II are also found in extracellular proteins. These proteins are present in many cell types, including

muscle cells. Among those extracellular proteins are N-CAM (neural cell adhesion

molecule) and fasciclin (Edelman, 1987; Harrelson and Goodman, 1988). These extracellular proteins are probably involved in cell-cell recognition and participate in

protein-protein interactions at cell surfaces. 22

Since motifs I and II are present in muscle proteins that are associated with

myosin, it is likely that these motifs can interact with myosin. Consistent with this idea,

cloned rabbit titin fragments derived from the A-band portion of titin have been shown to

interact with LMM (light meromyosin) of myosin by a solid phase binding assay (Labeit

et al., 1992). The ability of the cloned titin fragments to bind myosin is cumulative in

that stronger binding is found with an increasing number of motifs. The cloned titin

fragments from the A-band region of titin also interact with C-protein (Labeit et al.,

1992). The titin fragments used in the interaction studies likely contained both motifs I

and II, although this could not be determined from the data presented. Thus, it was not

clear from these studies that the interactions required both types of motifs or whether both

motifs contributed equally to the interactions. In addition, these interaction studies did not

provide answers to the significance of 11-motif super-repeat, I-I-I-II-I-I-I-II-I-I-II, present

in titin molecules (Labeit et al, 1990, 1992; Trinick, 1991), However, since the super-

repeats could generate a periodicity of 44 nm (length of 11 motifs) and this periodicity

coincides with the 43 nm periodicity of myosin filaments (Trinick, 1992), it is possible

that the 11-motif super-repeats interact with myosin filaments in vivo. Interactions of

purified titin with myosin and myosin filaments, AMP-deaminase, C-protein, actin and

actin filaments have been reported (Kimura et al., 1984a, b; Maruyama et al., 1987;

Soteriou et. al., 1993).

Although both localization studies and molecular cloning have contributed to

understanding of titin, a more complete understanding of titin requires complete

information on the primary structure of the whole titin molecule as well as biochemical 23 studies on possible interactions of titin molecules with other muscle proteins colocalized with titin along the M-line, I-band and Z-line regions. Currently, it is not clear how much the A-band and I-band regions of titin differ. Data from electron microscopy studies have suggested a homogeneous structure of beads on a string along the entire purified T2, a breakdown product of titin (Trinick et al., 1984). In addition, it is not known whether titin can regulate the length of thick filaments which have defined lengths in muscle cells. If so, what are the structural domains of titin involved? Equally intriguing is what proteins are involved in anchorage of titin at both ends of the molecule located at the M- and Z-line regions. What are the domains of titin that might interact with these proteins? Molecular cloning of the titin portion located at the junction between

A-band and I-band and biochemical studies of cloned titin fragments, as reported in the following two papers, provided some structural details of the molecule as well as the importance of certain motifs involved in interactions with other muscle proteins. 24

PAPER I. MOLECULAR CHARACTERIZATION OF AVIAN MUSCLE TITIN 25

Molecular Characterization of Avian Muscle Titin*

Kuan O. Tan , Greg R. Sater §, Alan M. Myersi, Richard M. Robson , and Ted W. Huiatt I

From the Muscle Biology Group, Departments of Animal Science and Biochemistry and Biophysics, and ^the Department of Biochemistry and Biophysics, Iowa State University, Ames, Iowa 50011.

Running Title: Cloning of avian muscle titin cDNA

* This work was supported in part by grants from the USDA Competitive Grants Program (Awards 89-37265-4441 and 92-37206-8051), the Muscular Dystrophy Association, the American Heart Association, Iowa Affiliate, and the Beef Industry Council of the National Live Stock and Meat Board, This is Journal Paper No. J-15277 of the Iowa Agriculture and Home Economics Experiment Station, Ames, Iowa. Project Nos. 2127,2921, 2930, and 3169.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBankl EMBL Data Bank with accession number(s) LI9140.

§ Present address: Department of Otolaryngology, Barnes Hospital, St. Louis, Missouri 63110.

U To whom correspondence should be addressed: Muscle Biology Group, 3112 Molecular Biology Building, Iowa State University, Ames, lA 50011. Tel.: 515-294-8289; Fax: 515-294-0453. 26

ABSTRACT

Titin is an approximately 3000 kDa polypeptide that constitutes a set of elastic filaments that connect thick filaments to the Z-line in vertebrate striated muscle myofibrils. To characterize the primary structure of titin, three overlapping cDNA clones comprising 2.4 kilobases of avian muscle titin coding sequence were obtained from a cDNA library constructed from embryonic chick cardiac muscle RNA size selected for large transcripts. Expression of one cDNA clone in E. coli produced a fusion protein that reacted specifically with titin antibodies, and titin antiserum affinity-purified against this fusion protein reacted specifically with titin on immunoblots of chicken cardiac and skeletal muscle myofibrils. Indirect immunofluorescence localization with the fusion protein-specific antibodies demonstrated that the cDNA sequence was from the region of titin located in the myofibrillar A-band adjacent to the A/I junction. Derived amino acid sequences demonstrated a repeating pattern of fibronectin type III and immunoglobulin C2 motifs, as shown previously for a portion of rabbit skeletal muscle titin located in the central region of the A-band and for other myofibrillar proteins that bind to myosin.

Differences between the rabbit and chicken titin sequence included a unique, serine-rich region in one motif, which represents a potential phosphorylation site. This is the first report of sequence information for avian titin, from a previously uncharacterized portion of the titin molecule. 27

INTRODUCTION

The megadalton protein titin (1), also called connectin (2), is a single polypeptide

chain of -2800-3000 kDa (3, 4) that constitutes 8-10% (5, 6) of the myofibrillar proteins in striated muscle cells. Current evidence (see [6-10] for review) indicates that titin constitutes a third set of longitudinal filaments in the sarcomere. Antibody localization studies have shown that titin is a component of the "gap filaments" observed in overstretched muscle (6, 11, 12), and titin (connectin) filaments also have been observed extending from the thick filaments to the Z-line in cardiac muscle following depolymerization of the thin filaments (13). The location and orientation of titin in the sarcomere has been mapped by immunoelectron microscopy with monoclonal antibodies to a series of unique epitopes of titin (6, 14, 15). These studies suggest that single titin molecules extend from the M-line to the Z-line in each half of the sarcomere and that titin molecules are oriented in opposite polarities in the two halves of the same sarcomere.

Epitopes of titin in the A-band appear to be rigidly attached to the thick filaments, whereas epitopes in the I-band behave elastically in response to stretch or contraction (14-

16).

Suggested functions for titin include: (1) regulating the length of the thick, myosin filaments by acting as a "molecular ruler" or template for assembly (10, 17); (2) acting as an elastic element in the myofibril involved in the regulation and maintenance of muscle elasticity, stiffness, and resistance to stretch (18, 19); (3) maintaining the positional stability of thick filaments during contraction or stretch (20); and (4) involvement in the 28 organization of myofibrils during the de novo assembly of myofibrils in developing muscle cells (8, 21). While evidence exists to support each of these possible functions, a complete understanding of the role of titin in muscle cells will require elucidation of the structure of the titin molecule.

The molecular weight of the titin polypeptide chain suggests that the titin message contains a coding region of ~75 kb and thus, obtaining cDNA sequences for the entire molecule has been difficult. Partial cDNA sequences for human and rabbit skeletal muscle titin, from the portion of the molecule located in the central region of the A-band, have been reported (22, 23). Derived amino acid sequences show a repeating pattern of two motifs, each about 100 amino acids long. Motif I is similar to fibronectin type III domains, and motif II belongs to the C2 set of immunoglobulin domains. These motifs have been implicated in binding of titin to myosin and C-protein (23), and an 11-motif super repeat may correspond to the 43-nm structural repeat in the thick filament (10, 23).

A repeating pattern of motifs I and II is also found in the myosin-associated, titin-like proteins twitchin, from C. elegans (24), and projectin, from Drosophila (25, 26), and both motifs are found in other vertebrate-muscle myosin-associated proteins (27-34). Both human titin and twitchin also contain a putative kinase domain near their C-terminus, the region of the molecule located adjacent to the M-line in myofibrils (23, 24).

To characterize further the primary structure of titin, we have cloned and sequenced titin cDNAs from an embryonic chick cardiac cDNA library. The resulting cDNA sequence codes for a portion of the titin molecule located near the A/I junction.

Thus, this cDNA sequence provides not only information on the structure of an additional 29 isoform of titin, but on an as yet uncharacterized region of the titin molecule. The A/I junction marks the transition between the rigid association of titin with the thick filament and the elastic portion of titin in the I-band. The adjacent portion of the thick filament is the region in which the packing of myosin molecules is altered to accommodate the tapered ends of the thick filaments, and the region of the thick filament that does not contain C-protein. In addition, the majority of polyclonal antibodies to titin recognize an antigenic "hot spot" at the A/I junction (8). Autoantibodies to titin have been found in patients with both myasthenia gravis and thymoma, and one such autoantibody recognizes the A/I junction region (35). Therefore, this region of titin seems to contain an antigenically and physiologically important epitope. A preliminary report of this work has been presented in abstract form (36). 30

MATERIALS AND METHODS

Construction and screening of cDNA libraries. Total RNA was prepared by the

guanidinium isothiocyanate/CsCl method (37) from hearts of 16-d incubated chick

embryos (Hy-Vac Laboratory Eggs, Gowrie, lA). The initial Xgtll cDNA library was

constructed from heart RNA by the procedure of Huynh et al. (38), modified to increase

the frequency of messages coding for all portions of the titin molecule. cDNA was

prepared from total heart RNA, a mixture of random tetra- and hexadeoxynucleotides

(Pharmacia LKB Biotechnology) was used to prime both first and second strand synthesis,

and E. coli DNA ligase (Pharmacia LKB) was added to the reaction mixture. Before

addition of EcoRL linkers, cDNA was digested with 10 units each of Pstl and Hindlll for

60 min at 37 °C. DNA was ligated into the EcoRl site of phage X cloning vector Xgtll.

The Xgtll library was screened as described (38) with polyclonal antibodies to titin, and a

single, positive clone (E-2) was isolated and plaque purified. Both the 0.2 kb' EcoRI

fragment inserted in this clone and a larger, Kpnl/Sstl fragment containing regions of the

cloning vector on each side of the cDNA insert were subcloned into pUCllS, forming

plasmid clones E-2a and E-2b, respectively.

A second cDNA library, in phage X cloning vector A,ZAP, was constructed from

embryonic chick heart RNA size-selected for large RNA molecules by gel filtration.

Total heart RNA was dissolved in 8 M urea, 0.5% SDS, 0.1 M NaCl, 0.01 M EDTA,

0.01 M Tris-HCl, pH 7,6, and fractionated on a Sepharose CL-2B (Pharmacia) column

(0.5 X 40 cm) as described elsewhere (39, 40). Fractions eluted in the first peak were 31 pooled, precipitated twice with ethanol, and used to synthesize cDNA with a cDNA

Synthesis Kit (Pharmacia). For first strand synthesis, 5 ng of RNA was primed with both random hexamers (Promega) and poly-dT. The cDNAs were ligated to the EcoRI site of the XZAP II vector (Stratagene) and packaged with an in vitro packaging system

(Gigapack II Gold, Stratagene). The 0.2 kb EcoRI insert from clone E-2 was ^^P-labeled with a random priming kit (Boehringer Mannheim) and used to screen approximately 3.5

X 10® clones from the amplified A,ZAP library as described elsewhere (41) with minor modifications. Three positive clones (TZ-3, TZ-6, and TZ-7) were plaque purified and the cDNA inserts subcloned into pBluescript SK- (Stratagene) propagated in E. coli strain

TG 1.

Antibody preparation and characterization. Titin was purified by Biogel A-50

(Bio-Rad) chromatography of an SDS extract of chicken leg muscle myofibrils according to Wang (42) and further purified by preparative SDS-PAGE (43). Rabbit polyclonal antibodies to purified titin were prepared as described previously (43). The titin antiserum reacted specifically with titin on immunoblots of chicken cardiac and skeletal muscle myofibrils. Antiserum was used for immunofluorescence; for all other procedures, IgG was isolated by (NHJ^SO^ precipitation (44). For screening of cDNA libraries and characterization of cDNA clones, titin antibodies were purified by affinity chromatography (45) to remove cross-reactivity with E. coli and Xgtl 1 proteins. The p- galactosidase monoclonal antibody was purchased from Promega.

Expression and analysis of fusion protein. Plasmid clone TZ-3 was digested with

Pstl, and the resulting 1.3 kb fragment of the cDNA insert, designated TP3, was 32 subcloned into pUR vectors (pUR 290, 291, and 292) (46) propagated in E. coli strain

TG 1. Cells were grown on nitrocellulose overlaid on LB agar medium containing 2 mM

IPTG for 6 h to induce production of P-galactosidase fusion proteins, and then screened with titin antibodies as described (38). Plasmid was isolated from positive clones and the orientation was verified by restriction digests. For immunoblot analysis, small-scale liquid cultures of positive clones were incubated with 2 mM IPTG for 3 h, and the insoluble fraction of the bacterial lysate was isolated (47).

Affinity purification offiision protein specific antibodies. TG 1 cells transformed with pUR292 containing the TP3 insert were induced with IPTG, transferred to nitrocellulose membrane, and lysed (37, 38). The membrane was incubated with polyclonal titin antibodies at a concentration of 8 /xg/ml for 2-3 h at room temperature, and then washed with TBS with several changes of buffer over a period of 2 h. Bound antibodies were eluted with 100 mM glycine-HCl, pH 2.6, 75 mM NaCl for 10 min at room temperature, and the eluate was immediately neutralized with 1 M Tris-HCl, pH 8.0

(32, 44). Eluted antibodies were concentrated 5-10 fold by dialysis against polyethylene glycol before use.

Immuiwblotting. SDS-PAGE was done according to Laemmli (48). For samples that contained titin, a discontinuous separating gel consisting of equal layers of 5% (100:1 acrylamide:bis), 8% (37.5:1), and 12% (37.5:1) acrylamide, with no stacking gel, was used to resolve both titin and lower molecular weight proteins on the same gel. Proteins were electrophoretically transferred to nylon membrane (Zeta-Probe, Bio-Rad) in a Bio-

Rad Trans-Blot cell at 55 V for 20 h in 25 mM Tris-glycine, pH 8.0, 0.01% SDS transfer 33 buffer. Antibody incubations were done by standard procedures (44) with a 1:40 dilution of titin IgG or a 1:50,000 dilution of monoclonal P-galactosidase antibodies, 1% nonfat dry milk (Blotto) in TBS for blocking, alkaline-phosphatase-conjugated second antibodies

(Sigma), and colorimetric detection with nitro blue tetrazolium and 5-bromo-4-chloro-3- indolyl-phosphate (Promega).

Southern and northern analysis. Genomic DNA was prepared from embryonic chick thigh (49), digested with endonucleases, separated on a 0.7% agarose gel, and alkaline blotted on Zeta-Probe membrane (Bio-Rad). RNA was fractionated on formaldehyde agarose gels (37) and blotted on the same membrane. Both blotting and hybridization were done according to the manufacturer's protocols supplied with the Zeta-

Probe membrane.

Indirect immunofluorescence labeling of myofibrils. Myofibrils were prepared from heart and skeletal muscles (leg muscles) of adult chickens as described by Wang et al. (50), using a rigor buffer consisting of 75 mM KCl, 1 mM MgClj, 2 mM EGTA, 10 mM K-phosphate, pH 7.0, 1 g/1 pennicillin, 1 mM phenylmethylsuifonylchloride, 25 mg/1 soybean trypsin inhibitor (Sigma), and 0.5 mg/1 leupeptin (Boehringer Mannheim).

Myofibrils were stored at -20 "C in rigor buffer containing 50% glycerol until use.

For indirect immunofluorescence, a drop of glycerinated myofibrils was placed on a slide and the myofibrils were allowed to settle for 5 min at room temperature. Slides were washed several times with PBS, glycerol, incubated with primary antibody (1:100 dilution of titin antibodies in PBS or undiluted, epitope-specific antibodies) for 30 min at

37 °C, washed 3 times with PBS, incubated with rhodamine-conjugated goat anti-rabbit 34

IgG (Cappel) for 30 min at 37 "C, washed 3 times with PBS, and mounted in 9:1 glycerol:PBS, pH 8.0, containing 1% (w/v) n-propyl gallate (51). Slides were examined with a Zeiss Photomicroscope HI equipped with epifluorescence optics and a 63X planapochromat objective and photgraphed on Kodak T-Max 400 film. Measurements were made on prints at a nominal magnification of 1900x with a measuring magnifier with

0.1 mm divisions. Exact magnifications for each set of prints were determined from printed images of a stage micrometer.

DNA sequence analysis. For clones E-2a and E-2b, in vector pUCllS, single- stranded phagemid DNA was prepared as described elsewhere (52). For clones TZ-3 and

TZ-6, a series of nested deletions was prepared with an Erase-a-Base System (Promega).

Deleted clones were analyzed for cDNA size, and single-stranded DNA was generated from appropriate clones with VCSM13 helper phage (Stratagene). Nucleotide sequence analysis by the chain termination method (53) was done at the Iowa State University

Nucleic Acid Center with an Applied Biosystems 373A DNA Sequencer. Both strands of clones TZ-6 and TZ-3 (regions not overlapping with TZ-6) were sequenced. For clone E-

2b, a 15 base synthetic nucleotide complimentary to the sequence of the lacZ gene immediately adjacent to the EcoRI site was synthesized at the Iowa State University

Nucleic Acid Center and used as a primer for manual dideoxy sequencing with a commercial sequencing kit (Gibco/BRL).

Sequence analysis was done with the University of Wisconsin Genetics Computer

Group (GCG) sequence analysis software package (54). BESTFIT was used for all sequence comparisons. FASTA was used for database searching. 35

RESULTS

Isolation of titin cDNA clones. Polyclonal antibodies to chicken muscle titin were used to screen a Xgtll cDNA library constructed from embryonic chick heart RNA.

Screening of ~3.5 x 10® independent clones in the unamplified library identified one positive clone, designated clone E-2, which was plaque purified. This clone contained an

~0.2 kb cDNA insert. Western blots of plaque lysates from this positive clone demonstrated the presence of a fusion protein recognized by both titin and P-galactosidase antibodies, and sequence analysis demonstrated the presence of an open reading frame in frame with the lacZ gene (results not shown).

To identify larger titin cDNA clones, the insert from clone E-2 was used to screen a second, embryonic heart cDNA library, constructed in the vector XZAP. The XZAP library was constructed from total RNA that had been fractionated by gel filtration chromatography to select for large RNAs. Dot blot hybridization of selected column fractions showed that clone E-2 hybridized strongly with fractions eluted in the void volume, and thus, only RNA eluted in this first peak was used for cDNA synthesis. A total of 3.25 X 10® independent clones was obtained from this library. The library was amplified, a small number of clones were screened, and three positive clones were isolated and purified: TZ-3 (1.5 kb cDNA insert), TZ-6 (1.8 kb cDNA insert), and TZ-7

(1.0 kb cDNA insert). Restriction mapping (Fig. 1) demonstrated extensive overlap between all three clones. Together, clones TZ-3 and TZ-6 contain a contiguous sequence corresponding to ~2.4 kb of titin mRNA. Figure 1. Restriction map of titin cDNAs

A partial restriction map of the cloned cDNA fragments was obtained from restriction digests and was confirmed by

DNA sequencing. Clone E-2 was obtained by titin antibody screening of an embryonic heart, Xgtll cDNA library.

Clones TZ-3, TZ-6, and TZ-7 were obtained by screening a second cDNA library with the insert of clone E-2 as a

probe. a g s î 0) On

-Sst I -Xba I fk-

p oo '

- Ar I M -&0 RI - Eco RI

at -

-Hind m oM

-5am HI ro k, Co

z-e 38

Characterization of positive clones. To confirm that these cDNA clones specify titin, the peptide encoded by a portion of TZ-3 was characterized for reactivity with titin antibodies. Plasmid clone TZ-3 was digested with Pstl to generate a 1.3 kb fragment

(TP3), which was ligated to the insertion site near the 3' end of the lacZ gene in vectors pUR290, 291, and 292. The resulting p-galactosidase fusion protein was expressed in E. coli. All positive clones identified by screening with titin antibodies contained TP3 inserted in pUR292 in the same orientation. Immunoblot analysis of bacterial lysates

(Fig. 2) showed that the fusion protein coded for by TP3 reacted specifically with both titin and P-galactosidase antibodies. The lower molecular weight band recognized by the titin antibodies is likely to be a degradation product of the fusion protein, because it was not observed in controls, and degradation of fusion proteins produced with this vector has been reported previously (46). No immunoreactivity with titin antibodies was observed in controls, including lysates from TGI cells alone and TGI cells transformed with pUR292 vector with no insert (Fig. 2).

To further demonstrate that the cloned cDNA codes for titin, bacterial lysates from

TGI cells transformed with TP3 in pUR292 were used to affinity purify antibodies specific for the fusion protein from the polyclonal titin antibodies, via a blot- immunoaffinity procedure. The affinity purified, fusion protein-specific antibodies specifically labeled the titin band in immunoblots of purified chicken skeletal muscle titin and adult chicken cardiac and skeletal muscle myofibrils (Fig. 3). Antibodies purified against lysates of TGI cells transformed with pUR292 alone did not label myofibrils on immunoblots. Figure 2. Immunoblot analysis of cDNA encoded titin fusion protein

Clone TZ-3 was digested with Pstl to generate a 1.3 kb fragment (TPS), which was subcloned into pUR292 and

propagated in E. coli strain TG 1. Proteins in cell lysates from cells transformed with plasmid containing TP3 (TPi),

cells transformed with the pUR292 vector with no insert (pUR292) or host cells with no plasmid (TGI cells) were

separated by SDS-PAGE on an 8% acrylamide gel and electrophoretically transferred to nylon membrane. Identical

blots were stained for total protein, incubated with monoclonal P-galactosidase antibodies, or incubated with polyclonal

rabbit antibodies to chicken skeletal muscle titin, as indicated. (+) or (-) indicates cells incubated with or without

IPTG induction, respectively. V) «/) CN JS cs g; o o •© u cs u CO a CO Qg Q_ 5 Q. 3 5 1— Q. IPTG + - + - + - - + - + fusion protein' p-gal-

Li M

itLSuZs^ Protein stain Anti-p-gal Anti-titin Figure 3. Specificity of titin polyclonal antibodies affinity purified against the

polypeptide encoded by TP3

Samples of purified chicken skeletal muscle titin, chicken cardiac myofibrils {Ca

Mj), and chicken skeletal muscle myofibrils {Sk Mj) were separated by SDS-PAGE

on a discontinuous 5%/8%/12% acrylamide gel and electorophoretically blotted.

a, Coomassie blue stained gel. b, blot incubated with titin antibody affinity

purified against TG 1 cells containing pUR292 vector only, c, blot incubated with

titin antibody affinity purified against TG 1 cells containing TP3 in vector

pUR292. Migration positions of titin (7), nebulin {N), and myosin heavy chain

(M) are indicated. I # Titin Q Co Mf Sk Mf

Titin Ca Mf Sk Mf

: Titin i Ca Mf Sk Mf 43

Indirect immunofluorescence. Immunofluorescence localization on myofibrils was used to determine the portion of the titin molecule coded for by the cloned cDNA. Both polyclonal anti-titin antiserum and the affinity-purified, fusion protein-specific antibodies labeled chicken skeletal muscle myofibrils at the A/I junction (Fig. 4). However, the titin antiserum labeled a broad band near the A/I junction, with some additional faint fluorescence in portions of both the A and I-bands (Fig. 4b), whereas the fusion protein antibodies specifically labeled a narrow band near each end of the A-band (Fig. 4d). The center-to-center distance between the two fluorescent bands within single sarcomeres (i.e. the distance measured through the M-line) in skeletal muscle myofibrils labeled with the fusion-protein antibodies averaged 1.4 ± 0.01 /tm (mean + S.E., n=198). The average length of the A-bands in the corresponding phase contrast images was 1.6 ± 0.01 /xm.

For chicken cardiac myofibrils labeled with the fusion-protein antibodies (not shown), the center-to-center distance between the fluorescent bands within the same sarcomere also averaged 1.4 ± 0.01 ^tm (n=107) and the corresponding A-band length was 1.6 ± 0.01

(itn. The distance between the two fluorescent bands at each end of the A-band in a single sarcomere did not change with decreasing or increasing sarcomere length, but each band was located closer to the Z-line in contracted sarcomeres than in relaxed sarcomeres for either skeletal or cardiac muscle myofibrils labeled with the affinity-purified antibodies.

Northern and southern analysis. A Kpnl-Sstl restriction fragment of TZ-6 cDNA was used to probe Northern blots of total RNA isolated from embryonic and adult thigh and heart muscles (Fig. 5). A single RNA species of "limited mobility" (>23 kb) was Figure 4. Indirect inununofluorescence localization of titin antibodies on chicken

skeletal muscle myofibrils

Paired phase contrast (a,c) and fluorescence (b,d) images of myofibrils incubated

with control, polyclonal titin antibodies (a and 6), or with antibodies affinity

purified against the polypeptide coded for by TP3 (c and d). Rhodamine-

conjugated goat anti-rabbit IgG was used as the second antibody. Arrows indicate

the position of the M-line (center of the A-band), Bar = 5 fim. mmmsmmwÈmsi

«g## M#* #W0 Figure 5. Northern blots of chicken muscle RNA hybridized with a titin cDNÂ

probe

Total cellular RNA was isolated from skeletal and cardiac muscle by the

guanidinium isothiocyanate/CsCl method, fractionated on 0.8% agarose gels,

blotted, and hybridized with [^^P]-labeled TZ-6. E Sk and E Ca, 16-d embryonic

chick skeletal and cardiac muscles, respectively. A Sk and A Ca, adult chicken

skeletal and cardiac muscles, respectively. Migration positions of the ribosomal

RNAs are indicated for each of the two blots, and the position of size markers is

included for the blot on the right. The migration distance for the two gels differs

because they were run at different times. I

0 tn CO u LU LU LU , ' ".- •

^23.1 k

^9.4 k

•<-6.6 k

28 s ^18 S 48

detected in both cardiac and skeletal muscle. The very large titin mRNA showed some

degradation even though the ribosomal RNAs appeared to be intact in the gel.

Preliminary results from ribonuclease protection assays with clone TZ-3 (data not shown)

indicated that mRNA homologous to this cDNA clone was expressed in striated muscles

(heart, leg, and breast) of 16-d and 21-d chick embryos and adult chickens, but not in

brain, liver, lung, or gizzard. Only a single protected fragment of the expected size was

found with RNA from either the skeletal or cardiac muscles.

Southern analysis of chicken genomic DNA digested with Sstl, EcoRl, and Hindlll

probed with the insert from clone E-2 showed only a single hybridizing fragment in each

of the restriction digests (Fig. 6). This result suggests that the chicken genome contains a single gene coding for titin.

Sequence analysis. The complete 2.4 kb contiguous sequence contained within clones TZ-3 and TZ-6 is shown in Fig. 7, along with the derived amino acid sequence.

This entire sequence comprises 2452 base pairs, coding for 817 amino acids. Only a single open reading frame was detected. This open reading frame is the same as that found for the fusion protein coded for by Xgtl 1 clone E-2, determined by sequencing through the EcoRI site of a KpnllSstl subclone (subclone E-2b) that contained a portion of the lacX gene (data not shown). Analysis of the amino acid sequence demonstrated the presence of two classes of repeating motifs, each approximately 100 amino acid residues long. Titin motif I is homologous to fibronectin type III motifs, and titin motif II is homologous to immunoglobulin C2 motifs. Beginning at the 5' end, the chicken titin sequence consists of a partial copy of motif I, five complete copies of motif I and two of Figure 6. Southern blot of chick genomic DNA hybridized with titin cDNA probe

Embryonic chicken genomic DNA was digested with Sstl, EcoRl, and Hindlll,

separated on an agarose gel and transferred to a nylon membrane. The blot was

hybridized at high stringency with the 0.2 kb insert from cDNA clone E-2, labeled

with Hindlll digested X fragments were used as size markers, as indicated. ^^23.1 Icb

<-9.4 kb

^6.6 kb

-*-4.4 kb ::'éW ..Sr

•2.3 kb 2.0 kb

•0.6 kb

^0.1 kb Figure 7. Nucleotide and derived amino acid sequence of the contiguous 2.4 kb portion of chicken titin cDNÂ

Nucleotide residues are numbered in the 5' to 3' direction beginning with the first nucleotide in the TZ-6 insert,

Amino acid residues in the derived sequence are numbered beginning with the first complete codon in the single, open

reading frame predicted from the DNA sequence. The beginning of each repeating motif in the sequence is shown in

shadowed type. Members of motif I are enclosed in boxes. Regions not boxed are members of motif II. The unique,

serine-rich sequence found in the fifth complete copy of motif I (amino acid residues 702-712) is underlined. The

double underline indicates the sequence of the E-2 clone. A 2 CGT GAA Tcr GAC CAT AGA c,cc TGG ACT CCA OTA AGC TAT ACA GTT ACA AGA CAO 1 R E s D H R A W T P V S Y T V T R Q 56 AAT GCT GTT GTT CAG GOG CTO ACT GAA GOG AAA OCC TAC ITT TTC COA ATT GCA 19 N A V V Q O L T E O K A Y F F R 1 A 110 GCA GAG AAT ATA ATC GGC ATO GOT CCT TTC ACA GAG ACO ACA AAA GAG CTC ATC 37 A E N 1 I G M O P F T E T T K E L I 164 ATT AGA OAT CCA ATA ACC GTT CCT GAC CGT CCT GAA GAC CTA GAG GTT AAA GCA 55 I R D P 1 T V »' U k P G Ù L E V K A 218 GTT ACO AAO AAC TCT GIT ACA TTA ACT TGG AAT CCT CCA AAA TAT AAT GGA GGC 73 V T K N S V T L T W N P P K Y N O G 272 TCA GAT ATC ACC ATO TAT GTT CFA GAG AGT CGT CTC ATT GGA AAA GAA AAA TTC 91 S D I T M Y V L E S R L 1 G K E K F 326 CAC AGA GTT ACT AAO GAG AAA ATO CTT GAT AGO AAA TAC ACT GTA GAA GGC TTO 109 H R V T K E K M L D R K Y T V E G L 380 AAA GAG GOT GAC ACC TAT GAG TAT CGT OTG AGT OCT TOC AAT ATT OTG GOG CAA 127 K E G D T Y E Y R V S A C N I V O Q 434 GGC AAO CCA TCA TTT TOC ACA AAA CCA ATC ACA TOC AAO GAT GAA ATT GCT CCT 145 G K P S F C T K P I T C K D E 1 A P 488 CCA ACA.. CTC. OAC ...qu.. GAC TTC AGA GAC AAG CTA ATT GTT COO GTT GOT GAA GCC 163 P T D D F R D K L I V R V G E A 542 ITC AGC CTO Acr OOA TOT TAC TCA GGA AAO CCC ACA CCA AAG ATT ACT TGG CTA INI I- S 1, 1 (1 K Y S G K P T V K I 1 W I, 596 AGO GAT GAT AIT GIT ITG AAA GAA GAT OAC CGT ACA AAG AIT AAG ACA ACT CCT 199 R D D 1 V L K E D D R T K 1 K T T P 650 ACA ACA CTT TOT CTO GOG ATA CFA AAA TCT GTC COT GAA GAT TCA GGC AAA TAT 217 T T L C L O I L K S V R E D S O K Y 704 TOT GTT ACT OTA GAG AAC AOC ACA OOA TCO AGA AAA GOG TTT TOT CAA GTT ACT 235 C V T V E N S T G S R K O F C Q V T 758 GTT GTC OAT CGC CCA GTT ATA TTT GAT GAA GTT CAT 253 V V D R 1i É# «i m T # P V I F D E V H 812 AAA GAT CAT ATO ATA GTT TCC TGG AAA CCT CCA CTC GAT GAC GGA GGC AGT GAA 271 K D H M I V S W K P P L D D G G S E 866 AIT ACA AAT TAC ATT ATr GAG AAO AAO GAT ACA CAC AGA GAC CTO TOO ATG CCA 289 1 T N Y 1 1 E K K D T H R D L W M P 920 GTT ACA TCT GCC ACA GTT AAO ACA ACA TOC AAA ATT CCC AAO CTO CTT GAA GGA 307 V T S A T V K T T C K 1 P K L L E G 974 AGA GAA TAT CAA GTT COA ATr TAT OCT GAA AAT CTC TAC GGC ATA AGT GAT CCT 325 R E Y Q V R I Y A E N L Y G 1 S D P 1028 CTC ATC TCT OAT GAG ATO AAA OCT AAG OAC CGT TTC AGC GTT 343 L I S D E M K A K D R F S V 11 ## # 1082 GAA CAA CCA GTT ATT AAG GAT GTT ACA AAA GAC TCT GCA GTA GTA GTT TGG AAC 361 •::v:0 P V I K D V T K D S A V V V W N 1136 AAA CCA TAT GAT OOA OGC AAG CCC ATA ACA AAC TAT ATT OTG GAA AAG AAO GAA 379 K P Y D G G K P 1 T N Y 1 V E K K E 1190 ACA ATG GCT ACT AGA rCÎG GTC AGA GIT ACC AAA GAC CCT AIT ne CCC AGT ACT 397 1 M A T R W V R V T K I) P l F P S T 1244 CAG TTC AAA OTA CCT GAC CTC en GAA GGC TOC CAG TAT GAA TTC CCT GTT TCA 415 Q F K V P D L L E G C Q Y E F R V S 1298 GCA GAA AAT GAA AIT GOT GTT GOT AAT CCT AGT CCA CCA TCA AAO CCA ATr TTT 433 A E N E 1 G V O N P S P P S K P I F 1352 GCT AGA GAT CCA AIT GTA AAA CCA....AOX ...CCA,...CCT. qy ...AAT CCG GAA GCA GTA GAT 451 A R D P 1 V K S r V N P E A V D 1406 AAA ACT AAO AAT TCT GTT OAC TTA ACT TOO CAA CCA CCA CGC CAT GAT GOT AAT 469 K T K N S V D L T W Q P P R H D G N — - tAiin rs/rr A A /-t « T-T. * — — • — — - -

1 D X A A D CI 118 VXD DDD VDV JXD XXO XDX DVD zet-z 1 W d A X Ù a yi X M a CI V D CI a b A E6i. VIO oxv VDD DXD VDV WD DJX DVV VDV DDX DVD XVD ODD XOO XVO WD DVD XVX 8i£Z H A T A O X 1 a V O O Cl y a d y N M SU VVO DXD DXX XVX VOD JOV VXV DVO XDD XOO DDD XVO XDD XVD VDD VOD XVV DOX KEZ A 1 S A a X a A a I H H I pss 51 Mil iÇZ, JJ.0 XXD DDX XXD WD XDV OVV DVO XXO XVD VXV DVD XVD DXD XVV DVV VDD VDD OiZZ ## ma d H 1 d a M V M 1 d T X a A A 6£6 DÛD XDD XDD XVD DXD VDD XVO OXV XDD VVV XXV VDD DVV DXD VDV VVO DXD XVX 91ZZ d D y D a 1 N a V V A M a 1 A b X O IIL VOO VDD VOV XOO XXX DXD xvv OVD XDD VDD XXO VDD DX.L JXD XVX DVD XDV DDD Z9IZ a M 1 A V M H S H X N S S s s A S £0/. WD oxv DJZ) XVX VVV XDD oxv DVD XDV DDD VVV DVV DDX DDX DDX VDX DXD XDX 801Z A M b V s V A a 51 y a A A A N X I A Ç89 VXD DOX DVD VDD XDX XDD VXD XVD VVV DDV WD DXD DXD XVX DVV DDV XXV XXO t'SOZ S O 0 a a a d a M A\ X 1 A D S a S a L99 XDX DOD XDD XVD XVD VVO VDD VVO DVV DOX VDV DXD DVX DDX XDX XVO XDV VDV OOOZ A a S A ô 1 N , H d J O M a 1 A 1 A 6W XXD XVD XDV DXD WD DXD XW VOV DÙD VDD XÛÛ JDD DVV XVO V.LX DXD XXD VXD 9fr6l M A S A X M S D V d N a A X 1 s A I l£9 WV VXD VDX XXD JOV DVV VDX XOO XDD VDD XVV WD XXO VDV DXD DDX DVX X.LV Z681 D D a a H A X N y I a M S O A D X £19 VOD XDD XVD VVO DVD D.1.D .ir)v DVV .I.OD .I.I.V VVO .I.IO DVV .1.DV DDD .U.O DDD VDV 8£81 A a 1 (I V X d S A a M CI a M A M S A S6Ç XXD DVD X.LV XVO XDD VVV DDD VDX DXD VVO OVV XVD WD VVV OVV DDX XDX XXO MAI M d a d A O M A V s 1 y 1 X a D V UÇ VW VDD DX.L VDD XXD XDD VW XXV XXO XDD XDV XXD VOV DXD VDV XVO VDD VDD 0£LI H A H 0 a a V n N V d a 6ÇÇ VOV VXD XVD OVD WD ODD DDD oxv XVV XDD w 'wv' O/D 3% XDD VVO 9Z.91 1 H a M A a A d . a d A A V d a s a o It-Ç XXD VOV DVD OVV VXD WD VXD VDD XVD XDD XXD DVX DDD VDD WD XDX WD XDD ZZ9I V V N A V Vi A y a )l A X 1 D a X 1 0 £ZÇ VDD VDD XVV DXD VDD OXV DXD DDV DXX DVV XVX DDV DXX XOO DVD XDV XXD VOD 89SI X A 31 A M X a d D S a d X 1 N V y. 51 SOS XDV DXD VVV XVX VVV VDV DVO XDD XDX XDX XVO VDD VDV X.IO xvv DDD DVV VVV MSI AV a a fl O A )l b A a A 'I A D I 1 N D HP DOX WD VVO XVO VOD XXD OVV OVD XVX VVO XXO X.LD DVX DDD XXV VXV ovv XOO 09M N D a H y d d b AV X 1 a A S N M X M 69^ XW XDD XVD XVD DOD VDD VDD WD DDX XDV vxx DVO XXO XDX XW DVV XDV VW 90M a A V a d N (t S tt M A I d a a V IS# XVO VXD VDD VVO ODD XVV xxo XDD VDD XDV VDD VVV VXD XXV VDD XVD vov XDD ZS£I j 1 d •A S d d S d N 0 A O I a N a V ££»> XXX XXV VDD DVV VDX VDD VDD XDV XDD XVV XDD XID XOO XXV WD XVV WD VDD 86Z1 s A M a a A b D O a 1 1 a d A M a b SI*' VDX XIO XDD DXX VVO XVX DVD DOX DDD WD XXD DXD DVO XDD VXD VVV DXX DVD pm X S d a I d (1 ÎJ X A M A M M X V M .1. L6C XDV XDV DDD DU. J.I.V XDD DVD VVV DDV XI.D VOV DXD DO.I. VDV .LDV XDD oxv VDV 0611 0 M a A 1 A N X 1 d )I O D CI A d )1 6i£ VVO DVV OVV VVO DXD XXV XVX DVV VDV VXV DDD DVV DOD VOD XVO XVX VDD VVV 9£1I N M A A A V S a •A X A a M I A d b H I9£ DW DOX XXD VXD VXD VDD XDX DVO VVV VDV XXO XVD OVV XXV XXO VDD WD VVO Z801 V a A S a H a V M W a a S 1 'I £»-£ XDD VDD XVD XDD XXD DDV DXX XDD DVO OVV XDD VVV OXV DVO XVD XDX DXV DXD 8Z0I d a S I D A 1 N a V A I H A b A a a SZ£ XDD XVO XDV vxv DOD DVX DXD XVV WD XDD XVX XXV VDD XXD WD XVX WD vov PL6

53

motif n arranged in the order I-II-I-I-I-II-I, and another partial copy of motif I (Fig. 7).

A unique, 11-residue, serine-rich sequence (VSVSSSSKKRS), which represents a potential phosphorylation site (see Discussion) is found in the final (3') complete copy of motif L

A comparison of the derived amino acid sequences of each of the two motifs is shown in Fig. 8. The five copies of motif I in the chicken titin sequence (Fig. 8a), are, on average, 35% identical. The lowest identity, 25%, was found between the two complete copies closest to the N-terminal (T2 vs. T4). The highest identity was 45% (T4 vs. T5). The three highly conserved amino acid residues characteristic of fibronectin type

III motifs (55) are found in all five copies of motif I. Overall, motif I consists of two highly conserved regions separated by a variable region near the center of the motif. The serine-rich sequence described above is located in this variable region of motif T8. The consensus sequence of chicken titin motif I is similar to the consensus sequence for this motif in rabbit titin (22) and in the related nematode protein twitchin (24). Two differences between the chicken and rabbit titin sequences occur near the C-terminal end of motif I. A charged amino acid (Asp or Lys) is found at one position (position 152 in

T2) in all copies of motif I in chicken titin as well as in the corresponding position in the incomplete copy of motif I at the N-terminal end of the sequence (position 51 in Fig. 7); in rabbit titin, only 5 of the 16 copies of motif I have a charged residue in this position.

A basic residue (Arg or Lys) is found at a second position (residue 157 in T2, Fig. 8a) in

4 of the 5 copies of motif I and in the incomplete N-terminal motif (residue 56 in Fig. 7), but in only 5 of the 16 copies of motif I in the rabbit sequence; this position corresponds Figure 8. Alignments of motifs I and H in the derived amino acid sequence of

chicken titin and related proteins

Sequence alignment was performed witli BESTFIT and visually inspected. T2-T8

are the repeating motifs in the chicken titin sequence numbered in order beginning

with the first motif found in the partial cDNA sequence. Residue numbers denote

the beginning and end of each motif in the overall sequence (Fig. 7). A dash

indicates a gap in the sequence, an asterisk indicates a nonconserved amino acid in

the consensus sequences, and identical residues are shaded. Panel a, alignment of

motif I (the fibronectin type III motifs). Residues found in at least three of the

five copies of motif 1 are included in the consensus sequence {Ck. Ti. Con.I).

Also included for comparison are consensus motif I sequences from rabbit titin

(Ra. Ti. Con.I) (9) and nematode twitchin (Twit. Con. I) (24), motif I sequences

from chicken M-protein (Ck. M-pro.) (30) and chicken C-protein (Ck. C-pro.)

(27), and the single copy of motif I in chicken smooth muscle myosin light chain

kinase (Ck. MLCK) (33). Brackets indicate the regions of the chicken titin

sequence that are highly conserved in copies of motif I, and arrows indicate

specific residues that differ between chicken and rabbit titin as noted in the text.

Arrowheads below the sequences indicate the positions of the three highly

conserved residues in fibronectin type III repeats. Panel b, alignment of motif II

(the immunoglobulin C2 motifs). Residues found in both copies of motif II (T3

and 77) are included in the consensus sequence (Ck. Ti. Con. 11), and are compared

with copies of motif II in other proteins as in panel a. Arrowheads indicate the

conserved residues characteristic of Ig C2 domains; open arrowheads indicate the

position of the conserved cysteine residues in the Ig C2 sequences. 55

a

T2 62 |DRiEDI£VKA-VTOSVTl.TWNiiKYNi||])ITMyVLESRLIGKEKFHR--VTK T4 257 '---VT T5 357 Idaîeqpvikd-vtîossavvvwnt^-yIIIkpitîïvivekIetmatrîvrytkdp T6 458 |si#^li|EAv|-KTKNSVDLTWQi|RHÎiNGKIIGYLmQKVGDEE|p^^ T8 655 #:G0RHLQVSD - mSDSC YLTWKEPEDDSGgVlimrVVERKDVASAQ#VS - - - VS Ck. Ti. Con.I Ra. Ti. Con.I Twit. Con. I Ck. C-pro. PI>ieiPQSyRVTS -|GE|WAViS|pA|:PF|i|MPITGYI2£ER|KKGSMRi^ - Ck. MLCK |D||AGTPCASDIRSS|LTls|YGSSTiii|AVQSiTiEiraSVDNI#DDL- - - T Ck. M-pro. |GAiMDVKCHi-ANRi^IViteiNTTSQNPVIG|F|DKCEVGIZNivQCNDAP

T3 163 EmDim - - FRDKLI^iGEAFSlTGRYSSEETiRRITKLRDDilVLKEDDRTKlK'Kr T7 561 VS|KKE|KEV - - s PKADgE# Ck. Ti. Con.II ÈA * * * * •-A-R+TWrB* * ******* *0*+$: Ra. Ti. Con.II Twit. Con.II -•VKAG-AHHh Ck. C-pro. --PAEVQTRTS: Ck. MLCK Ck. M-pro. GKVIGG

llSRLIGKEKFHR--VTKEKMU)mCYT#GM#T:^EY#SACNIVi#:QMSFCT##rC#EIA#: 162 llKRDTHRDLiMP- --VTSATVK$:TCKIPKW#RE0QVEIYj^NLY&SD#:-LISDEM#kFSV 356 0#TMATRgVRVTKDP-IFPSTQFKVPDmJBGCQYEmVSA#EI(3VGNfSPPS0gFH?IVK 457 iEYQKVGDEEiKKANLTPDSCPETKYKVTGLTEGLTYKFRVM&VSAAGESEPAYVPDiVEViBRLEi 560 ÇERKDVASAQÏVS---VSSSSKKRSmEKYM&Q#LF&VAAmLF&#WETI#mAfLH§: 754 ********** ||EK-

LEI)|IVIJŒDDRTIfKI|PTTLC|G|IJCS|-B||S|l«CV|BiST||EKGFCQiT|V|R 256 iCKE6)ŒV--SPKADiEViGVGSKiEiSBTl-HE®GiIiSLli|llPASSKTVSVKiLMK 654

•KDG *-*L+'A~*'TR*r+iË*'S3; -A-ASk-T* * * * *R+B,*^:*r!Yr*-*-3;L^N-**^*-KS •*-V +V:'*%IJD.*'

SREGGAL--PAEVQTRTSDVDSVFF:IRSAARP-LS|NiEMRiRIDNMEDCATLRLE^VER MKFRKQIQENEYIK|ENAENSSK|TisSTKQE - Hc|c|TL\jËiKlJSRQAQVNLT|viK FKNDKALELNEHYKVSLEQGKYASUriKGlTSÈislKisiYpiKYiGETVDVTVsiYliî

56 to a Lys residue in the twitchin consensus sequence. Sequences for motif I in the fast skeletal muscle isoform of C-protein, chicken gizzard myosin light chain kinase, and chicken pectoralis muscle M-protein are also shown for comparison. For M-protein and

C-protein, the copy of the motif in the sequence that gave the highest identity with chicken titin is shown (i.e., motif CV in [27] for C-protein and motif M F1 in [30] for

M-protein). The majority of the conserved residues in chicken titin motif I are found in motif I in these three other proteins.

The two copies of motif II in the chicken titin sequence are 30% identical (Fig.

8b). Of the 10 highly conserved positions characteristic of the Ig C2 superfamily (56), 5 are conserved in the titin sequence, and two others contain conservative substitutions.

The two conserved Cys residues found in Ig C2 domains are not found in chicken titin, as is true for other non-immunoglobulin members of this superfamily, including rabbit titin and twitchin. Although the chicken titin sequence contains only two copies of motif II for comparison, it appears that motif II exhibits more variability than motif I.

When the entire derived amino acid sequence obtained for chicken titin was compared with the partial rabbit titin sequence reported by Labeit et al. (23), 41% identity and 58% similarity were found. The amino acid sequence of chicken titin also exhibited sequence homology with twitchin (37% identity), chicken C-protein (26% identity), chicken M-protein (25 % identity) and chicken smooth muscle myosin light chain kinase

(20% identity). Both the chicken and rabbit titin sequences showed about the same identity to chicken C-protein and chicken smooth muscle myosin light chain kinase, but 57 fewer gaps were used in chicken titin sequence for sequence alignment than for the rabbit titin (22) sequence. 58

DISCUSSION

We have obtained cDNA clones coding for a total of ~2.4 kb of the chicken

muscle titin message. This sequence codes for 817 amino acids, corresponding to a

polypeptide with a molecular weight of -94,000. Neither cDNA sequences for any avian

muscle titin nor sequences from this portion of the titin molecule have been reported

previously. That these cDNAs code for titin has been shown both by immunological

evidence and by amino acid sequence homology with rabbit and human titin (22, 23).

Indirect immunofluorescence with antibodies affinity purified against the polypeptide

encoded by the 1.3 kb fragment TP3, which is the 3' end of the contiguous sequence,

demonstrated that this sequence codes for the region of the titin molecule located in the A-

band near the A/I junction in myofibrils. The fluorescent labeling is clearly in the A-

band, as the location with respect to the A-band did not change with changes in sarcomere

length. Although localization at this region could also conceivably occur if the affinity-

purified antibodies exhibited crossreactivity between epitopes in the TP3-encoded

polypeptide and epitopes in this specific A-band region of titin, unless the antibodies also

had significantly higher affinity for a crossreacting epitope than for the correct epitopes,

one would expect to see localization at both the correct site and at the location of the

crossreacting epitope. In fact, labeling was observed only in a single, well-defined band

(Fig. 4d). Furthermore, we have recently prepared mouse polyclonal antibodies to an expressed protein corresponding to -2.1 kb of the entire cDNA sequence, including TP3,

and these antibodies also label myofibrils specifically in the A-band near the A/I junction, 59 by indirect immunofluorescence^. Thus, it is evident that the TP3 encoded region of titin is indeed located 0.1 ^m from the ends of the thick filaments. This region of the titin molecule is at least 100 nm closer to the free ends of the thick filaments than the rabbit skeletal muscle titin sequence (23). The C-terminus of titin is at the end of the molecule adjacent to the M-line (23), and thus, the portion of titin encoded by the remainder of the chicken titin sequence, which includes -1.1 kb at the 5' end of clone TZ-6 (see Fig.l), extends from the location of the fluorescent labeling towards the free ends of the thick fllaments.

Although the avian titin cDNAs were isolated from a heart library, these cDNAs are not specific for the cardiac isoform of titin, as both Northern blots (Fig. 5) and

RNase-protection assays (not shown) demonstrated hybridization of clone TZ-6 to RNA from embryonic and adult chicken cardiac and skeletal muscles. Either the cDNAs are specific for an isoform of titin that is expressed in all of these muscles, or this cDNA sequence represents a region of the molecule that is highly homologous among several different isoforms. There is, at present, no evidence for the expression of more than one isoform of titin in adult chicken cardiac muscle. Isoforms of titin with different molecular weights are expressed differentially during embryonic and neonatal development in chicken skeletal muscle (57). Rabbit skeletal and cardiac muscle express titin isoforms of different molecular size (19), and this difference is found in the I-band part of the molecule. Hill and Weber (58) described two monoclonal antibodies that reacted with skeletal but not cardiac muscle titin in rats or mice, but recognized both in chickens, and these antibodies labeled epitopes in the I-band. On the other hand, the majority of the 60 polyclonal and monoclonal titin antibodies that react with the A-band or A/I junction region of titin, including those used herein, recognize both skeletal and cardiac muscle titins (14, 15, 59-61). Taken together, these results suggest that the major differences between cardiac and skeletal muscle titin, and probably between other isoforms of titin, are in the region of the titin molecule located in the I-band, whereas the A-band regions of different titin isoforms expressed in different muscles of the same species seem to be similar. Thus, it is likely that our cloned titin sequence is derived from the avian cardiac isoform, but represents a region of the titin molecule that is highly homologous, if not identical, between different chicken titin isoforms. Because titin appears to be coded for by a single gene in chicken (Fig. 6) as well as in human and mouse (22), it is likely that isoforms are generated by alternate splicing. Both the XZAP library and the titin cDNA probes described herein should prove useful in extending the cDNA sequence into the region coding for the I-band portion of titin, and thus, in identification and characterization of differentially spliced isoforms.

The amino acid sequence derived from the chicken muscle cDNA clones clearly demonstrates the presence of motifs characteristic of fibronectin type III domains (motif I) and immunoglobulin C2 domains (motif II). These motifs have been described in regions of rabbit and human titin located in the central portion of the A-band in the myofibril (22,

23), in the invertebrate, thick filament-associated, titin-like proteins twitchin (24) and projectin (25, 26) and in several other myosin-associated proteins, including chicken fast muscle C-protein (27), bovine and human slow muscle C-protein (28), chicken Myosin- binding protein H (29), chicken fast muscle M-protein (30), murine skelemin (31), and 61 chicken smooth muscle and nonmuscle myosin light chain kinases (33, 34). In rabbit titin, these two motifs repeat in an 11-motif super repeat with the order I-I-I-II-I-I-I-II-I-I-

II (9), and this super repeat is thought to correspond to the 43 nm periodicity of the myosin filament. The repeating pattern of the seven complete and two partial motifs observed in the chicken titin sequence, I-I-II-I-I-I-II-I-I, appears to fit this same pattern, and the chicken titin sequence data suggests that this super repeat pattern extends to the region of titin located closer to the ends of the thick filaments than the previously reported

(23) rabbit titin sequence. In combination, the chicken and rabbit titin sequences suggest that this repeat pattern is conserved among vertebrate muscle titins. The repeating motif pattern appears to be important in the location of titin along the entire length of the vertebrate muscle thick filament, as other proteins that bind to discrete locations on the thick filament, such as M-protein, C-protein, and skelemin, have very different repeating patterns of the two types of motif. Differences observed between our chicken titin sequence and the rabbit titin sequence could be due to a difference in location, species differences, or both. In the rabbit sequence, copies of motif I were about 40% conserved

(9), and sequences from two different regions of rabbit titin, the CE12 and MS2 regions in the EMBL database, shared 50% identity. The homology between copies of motif I in chicken titin was somewhat lower (35%), and the lowest identity (25%) was observed between the two complete copies of motif I nearest to the 5' end of the sequence, corresponding to the most distal portion of titin in the A-band. This lower homology may be due to changes in the titin sequence as one moves farther towards the end of the thick filament. 62

A unique, Ser-rich sequence was found in one copy of motif I (T8, Fig. 8; see also Fig. 7), in the variable region near the center of the motif. This sequence was not found in any other portion of our chicken titin sequence, or in the sequence of rabbit titin from the central portion of the A-band (23). This unique sequence (VSVSSSSKKRS) contains six Ser residues as well as three adjacent basic residues. The concentration of

Ser residues in this region suggests the possibility that this sequence could be a phosphorylation site. Comparison with a compilation of consensus phosphorylation sites

(62) shows that this region of the titin sequence contains the consensus phosphorylation sites for cGMP-dependent protein kinase (KXS, KXXS, and KKXS), glycogen synthase kinase 3 (SXXXS), and protein kinase C (SXK, KXXS, and KXS). Titin is known to be a phosphoprotein, containing up to 12 mol of phosphate per mole of titin, mostly on Ser residues, as isolated from mouse diaphragm (63), and titin is phosphorylated in vivo both in frog skeletal muscle (64) and mouse diaphragm (63). Phosphorylation of this Ser-rich site in the titin molecule by other protein kinases could have important functional implications, such as altering the interaction with myosin. The isolation of the cDNA encoding this region will allow expression and further characterization of this unique potential phosphorylation site. 63

ACKNOWLEDGEMENTS

We thank Jo Black Philips and Janet Stephenson for expert technical assistance, and Dr. John E. Mayfield for advice on the molecular cloning of titin. 64

REFERENCES

1. Wang, K., McClure, J., and Tu, A. (1979) Proc. Natl. Acad. ScL USA 76, 3698- 3702

2. Maruyama, K., Kimura, S., Ohashi, K., and Kawano, Y. (1981) J. Biochem. 89, 701-709

3. Kurzban, G.P., and Wang, K. (1988) Biochem. Biophys. Res. Commun. 150, 1155- 1161

4. Maruyama, K., Kimura, S., Yoshidomi, H., Sawada, H., and Kikuchi, M. (1984) J. Biochem. 95, 1423-1493

5. Trinick, J., Knight, P., and Whiting, A. (1984) J. Mol. Biol. 180, 331-356

6. Wang, K. (1985) Cell and Muscle Motility (Shay, J.W., ed) Vol. 6, pp. 315-369, Plenum Press, New York

7. Maruyama, K. (1986) Int. Rev. Cytol. 104, 81-114

8. Fulton, A.B., and Isaacs, W.B. (1991) BioEssays 13. 157-161

9. Trinick, J. (1991) Curr. Opin. Cell Biol. 3, 112-119

10. Trinick, J. (1992) FEBS Lett. 307, 44-48

11. LaSalle, P., Robson, R.M., Lusby, M.L., Parrish, P.C., Stromer, M.H., and Huiatt, T.W. 1983. J. Cell Biol. 97, 258a (abstr.)

12. Maruyama, K., Yoshioka, T., Higuchi, H., Ohashi, K., Kimura, S., and Natori, R. (1985)7. Cell Biol. 101, 2167-2172

13. Punatsu, T., Kono, E., Higuchi, H., Kimura, S., Ishiwata, S., Yoshioka, T., Maruyama, K., and Tsukita, S. (1993) J. Cell Biol. 120, 711-724

14. Fûrst, D.O., Osborn, M., Nave, R., and Weber, K. (1988)/. Cell Biol. 106, 1563- 1572

15. Fiirst, D.O., Nave, R., Osborn, M., and Weber, K. (1989) /. Cell Sci. 94, 119-125 65

16. Itoh, Y., Suzuki, T., Kimura, S., Ohashi, K., Higuchi, H., Sawada, H., Shimizu, T., Shibata, M., and Maruyama, K. (1988) J. Biochem. 104, 504-508

17. Whiting, A., Wardale, J., and Trinick, J. (1989) J. Mol. Biol. 205, 263-268

18. Horowits, R. (1992) Biophys. J. 61, 392-398

19. Wang, K., McCarter, R., Wright, J., Beveriy, J., and Ramirez-Mitchell, R. (1991) Proc. Natl. Acad. Sci. USA 88, 7101-7105

20. Horowits, R., and Podolsky, R.J. (1987) J. Cell Biol. 105, 2217-2223

21. Epstein, H.F., and Fischman, D.A. (1991) Science 251, 1039-1044

22. Labeit, S., Barlow, D.P., Gautel, M., Gibson, T., Holt, J., Hsieh, C-L., Francke, U., Leonard, K., Wardale, J., Whiting, A., and Trinick, J. (1990) Nature 345, 273-276

23. Labeit, S., Gautel, M., Lakey, A., and Trinick, J. (1992) EMBO J. 11, 1711-1716

24. Benian, G.M., Kiff, I.E., Neckelmann, N., Moerman, D.G., and Waterston, R.H. (1989) Nature 342, 45-50

25. Ayme-Southgate, A., Vigoreaux, J., Benian, G., and Pardue, M.L. (1991) Proc. Natl. Acad. Sci. U.S.A. 88, 7973-7977

26. Fyrberg, C.C., Labeit, S., Bullard, B., Leonard, K., and Fyrberg, E. (1992) Proc. R. Soc. Lond. Ser. B 249, 33-40

27. Einheber, S., and Fischman, D.A. (1990) Proc. Natl. Acad. Sci. USA 87, 2157- 2161

28. Furst, D.O., Vinkemeier, U., and Weber, K. (1992) J. Cell Sci. 102, 769-778

29. Vaughan, K.T., Weber, F.E., Einheber, S., and Fischman, D.A. (1993) J. Biol. Chem. 268, 3670-3676

30. Noguchi, J., Yanagisawa, M., Imamura, M., Kasuya, Y., Sakurai, T., Tanaka, T., and Masaki, T. (1992) J. Biol. Chem. 267, 20302-20310

31. Price, M.G., and Gomer, R.H. (1993) J. Biol. Chem. 268, 21800-21810 66

32. Guemero, V., Jr., Russo, M.A., Olson, N.J., Putkey, J.A., and Means, A.R. (1986) Biochemistry 25, 8372-8381

33. Olson, N.J., Pearson, R.B., Neadleman, D.S., Hurwitz, M.Y., Kemp, B.E., and Means, A. R. (1990) Proc. Natl. Acad. Sci. U.S.A. 87, 2284-2288

34. Shoemaker, M.O., Law, W., Shattuck, R.L., Kwiatkowski, A.P., Matrisian, P.E., Guerra-Santos, L., Wilson, E., Lukas, T.J., Van Eldik, L.J., and Watterson, D.M. (1990)7. Cell Biol. Ill, 1107-1125

35. Williams, C.L., Hay, J.E., Huiatt, T.W., and Lennon, V.A. (1992) Lab. Invest. 66, 331-336

36. Tan, K.O., Myers, A.M., Robson, R.M., and Huiatt, T.W. (1992) Mol. Biol. Cell 3, 1467 (abstr.)

37. Maniatis, T., Fritsch, E. P., and Sambrook, J. (1982) Molecular cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

38. Huynh, T.V., Young, R.A., and Davis, R.W. (1985) in DNA Cloning: A Practical Approach (Glover, D.M., ed) Vol. 1, pp. 49-78, IRL Press, Oxford

39. Rydlander, L., Pigon, A., and Edstrom, J.-E. (1980) Chromosoma 81, 101-113

40. Hôôg, C., Engberg, C,, and Wieslander, L. (1986) Nucleic Acids Res. 14, 703-719

41. Benton, W.D., and Davis, R.W. (1977) Science 196, 180-182

42. Wang, K. (1982) Methods Enzymol. 85, 264-274

43. Richardson, F.L., Stromer, M.H., Huiatt, T.W., and Robson, R.M. (1981) Eur. J. cell Biol. 26, 91-101

44. Harlow, E., and Lane, D. (1988) Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

45. De Wet, J.R., Fukushima, H., Dewji, N.N., Wilcox, E., O'Brien, J.S., and Helinski, D. R. (1984) DNA 3, 437-447

46. Ruther, U., and Muller-Hill, B. (1983) EMBO J. 2, 1791-1794 47. Kleid, D. G., Yansura, D., Small, B., Dowbenko, D., Moore, D.M., Grubman, M.J., Mckercher, P.D., Morgan, D.O., Robertson, B.H., and Bachrach, H.L. (1981) Science 214, 1125-1129

48. Laemmli, U.K. (1970) Nature 227, 680-685

49. Ausubel, P.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., and Struhl, K. (1987, 1988) Current Protocols in Molecular Biology, John Wiley and Sons, New York

50. Wang, S.-M., Greaser, M.L., Schultz, E., Bulinski, J.C., Lin, J.J.-C., and Lessard, J.L. (1988)7. Cell Biol 107, 1075-1083

51. Giloh, H., and Sedat, J.W. (1982) Science 217, 1253-1255

52. Vieira, J., and Messing, J. (1987) Methods Enzymol. 153, 3-11

53. Sanger, F., Nicklen, S., and Coulsen, A.R. (1977) Proc. Natl. Acad. Sci. USA 74, 5463-5467

54. Genetics Computer Group (1991) Program Manual for the GCG Package, Version 7, Genetics Computer Group, 575 Science Dr., Madison, WI

55. Petersen, T.E., Skorstengaard, K., and Vibe-Pedersen, K. (1989) in Fibronectin (Mosher, D.F., ed.) pp. 1-24, Academic Press, Inc., San Diego, CA

56. Williams, A.P., and Barclay, A.N. (1988) Annu. Rev. Immunol. 6, 381-405

57. Yoshidomi, H., Ohashi, K., and Maruyama, K. (1985) Biomed. Res. 6, 207-212

58. Hill, C., and Weber, K. (1986)7. Cell Biol. 102, 1099-1108

59. Wang, S.-M., and Greaser, M.L. (1985) 7. Muscle Res. Cell. Motil. 6, 293-312

60. Itoh, Y., Kimura, S., Suzuki, T., Ohashi, K., and Maruyama, K. (1986) 7. Biochem. 100, 439-447

61. Pierobon-Bormioli, S., Biral, D., Betto, R., and Salviati, G. (1992) 7. Muscle Res. Cell Motil. 13, 35-38

62. Pearson, R.B., and Kemp, B.E. (1991) Methods Enzymol. 200, 62-81

63. Somerville, L.L., and Wang, K. (1988) Arch. Biochem. Biophys. 262, 118-129 68

Somerville, L.L., and Wang, K. (1987) Biochem. Biophys. Res. Commun. 147, 986-992 69

FOOTNOTES

'The abbreviations used are: kb, kilobase(s); PAGE, polyacrylamide gel electrophoresis, IPTG, isopropyl-P-thiogalactopyranoside; PBS, phosphate-buffered saline;

TBS, Tris-buffered saline.

^Tan, K.O., and Huiatt, T.W., unpublished results. 70

PAPER II. INTERACTION OF CLONED FRAGMENTS FROM THE A-I JUNCTION

REGION OF CHICKEN MUSCLE TITIN WITH MYOSIN AND ACTIN 71

Interaction of Cloned Fragments from the A-I Junction Region of Chicken Muscle Titin with Myosin and Actin

Kuan O. Tan, Richard M. Robson, and Ted W. Huiatti

From the Muscle Biology Group, Departments of Animal Science and of Biochemistry and Biophysics, Iowa State University, Ames, Iowa 50011.

Running Title: Interactions of cloned titin fragments

This work was supported in part by USDA Competitive Grants Program Award, 92- 37206-8051. This is Journal Paper No J- of the Iowa Agriculture and Home Economics Experiment Station, Ames, lA, Project Nos. 2127 and 3169.

1 To whom correspondence should be addressed: Muscle Biology Group, 3112 Molecular Biology Bldg., Iowa State University, Ames, lA 50011. Tel: 515-294-8289; Fax: 515- 294-0453 72

ABSTRACT

Four cloned titin fragments, with sizes ranging from 35-78 kDa, were expressed in

E. coli and purified. In solid phase binding assays, all four fragments interacted with

myosin and weakly with actin. The interactions were stronger at lower concentrations of

KCl; however, significant interactions were observed at conditions close to physiological ionic strength (20 mM phosphate buffer containing 100-200 mM KCl). Competition assays demonstrated that titin fragments with a larger number of motifs have higher affinity for myosin. In addition, competition assays showed that titin fragments that are similar in molecular weight, but differ in motif arrangement and completeness of motif II, exhibited different affinity for myosin when tested in buffer containing 200 mM KCl.

These results suggest that motif II in particular is likely to be important in contributing to interaction with myosin. Although cloned titin fragments interact with monomeric myosin in 100-200 mM KCl, no significant interaction was observed between titin fragments and myosin filaments under the same conditions. However, lower KCl concentrations (20

mM KCl) allowed the latter interaction to occur. Together, these results suggest that the interactions of titin fragments with both myosin and actin are at least in part electrostatic, and that the interactions with myosin are limited by the number of binding sites available on the myosin filaments. Thus, significant interaction between titin and myosin in muscle cells appears to require a large number of interacting domains from the titin molecules. 73

INTRODUCTION

Titin (Wang et al., 1979), also called connectin (Maruyama et al., 1981), is a single polypeptide with a molecular weight of 2800-3000 kDa (Maruyama et al., 1984;

Kurzban and Wang, 1988), which spans each half of the thick filaments and links the thick filaments, through the I-band region, to the Z-line (Maruyama et al., 1985; Wang,

1985; Furst et al., 1988, 1989; Funatsu et al., 1993). The localization of titin at the junction between A- and I-bands in overstretched muscle (LaSalle et al., 1983; Itoh et al.,

1988) confirmed early observations of "gap filaments" (Huxley and Peachey, 1961;

Sjostrand, 1962; Locker and Leet, 1975), or "end filaments" (Trinick, 1981), both of which are likely to be parts of titin molecules. Localization studies with antibodies to single epitopes of the long titin molecules (Itoh et al., 1988; Furst et al., 1988, 1989) suggest that the portion of titin located in the A-band of myofibrils is rigidly attached to the thick filaments. However, the molecular interactions responsible for this attachment are not known.

Myosin is the major protein component of the thick filaments, and makes up the backbone of these filaments. Several other proteins are known to be associated with the thick filaments, and these proteins have been localized to defined regions of the A-band.

M-protein (Masaki and Takaiti, 1974; Woodhead and Lowey, 1982), creatine kinase

(Wegmann et al., 1992), myomesin (Grove et al., 1984), and skelemin (Price, 1987) are all located at the M-line, in the center of the A-band. C-protein (Offer et al., 1973) and

86 kDa protein (Bahler et al., 1985a, b) are located on a series of stripes spaced 43 nm 74 apart, in the central 1/3 of each half of the A-band. The enzyme AMP-deaminase has been localized at the ends of thick filaments, at the A-I junction (Ashby et al., 1979).

The exact location of AMP-deaminase was determined to be 60 nm beyond where the myosin filaments end by electron microscopy methods (Cooper and Trinick, 1984). These electron microscopy results also seemingly challenge the significance of the findings that

AMP-deaminase interacts with the S-2 or heavy meromyosin parts of myosin in vitro

(Ashby et al., 1979; Barshop and Frieden, 1989; Marquetant et al., 1989). Recently,

Soteriou et al. (1993) have reported interactions of purified titin with AMP-deaminase in solid phase binding assays. In addition, the purified titin also interacts with myosin

(specifically with the light meromyosin portion of the myosin rod), actin, X-protein (also known as slow muscle C-protein), and C-protein (Soteriou et al., 1993). However, the interaction with actin is weak (Soteriou et al., 1993). Interactions of purified titin with myosin and actin detected by turbidity or cosedimentation assays in addition to electron microscopy methods were also reported earlier (Kimura et al., 1984, Maruyama et al.,

1987). Furthermore, titin is likely to be associated with M-line proteins as evident from the appearance of purified titin that showed a globular head, believed to be made up of

M-line proteins (Nave et al., 1989; Vinkemeier et al., 1993), at one end of the titin molecules.

Partial cDNA sequences of titin have shown the presence of two repeating motifs,

I and II in titin (Labeit et al, 1990, 1992; Trinick, 1991; Tan et al., 1993). Motifs I and

II are members of the fibronectin type III domains and the C2 set of the immunoglobulin superfamily, respectively, and each motif is about 100 amino acids long. In the region of 75

titin located in the A-band in myofibrils, these motifs are arranged in a 11-motif super-

repeat, I-I-I-n-I-I-I-II-I-I-II, which may correspond to the 43 nm structural repeat in the

thick filament (Labeit et al., 1992; Trinick, 1992). Motifs I and II are also present in other proteins associated with myosin, but the motifs are arranged in a different repeating pattern. These proteins include C-protein, also called MyBP-C (Einheber and Fischman,

1990, Furst et al., 1992; Weber et al., 1993), 86-kDa protein or MyBP-H (Vaughan et al,

1993a, b), M-protein (Noguchi et al., 1992), skelemin (Price and Gomer, 1993), and smooth muscle and non-muscle myosin light chain kinases (Olson et al., 1990; Shoemaker et al., 1990). In invertebrates, these motifs are present in the titin-like proteins twitchin of C. elegans (Benian et al., 1989, 1993), projectin of Drosophila (Ayme-Southgate,

1991; Fyrberg et al., 1992), and pSOO of Lethocerus indicus (Lakey et al., 1990). In addition, copies of motif II are present in kettin, a large protein found at the Z-disc of insect muscle (Lakey et al., 1993). Besides motifs I and II, unique sequences of titin include a kinase domain close to the C-terminus (M-line end) of the molecule (Labeit et al., 1992), sequences in the I-band region that show homology to neurofilament subunits

(Maruyama et al., 1993), and KSP (amino acid sequence, Lys-Ser-Pro) motifs, which are

potential phosphorylation sites (Gautel et al., 1993).

Although much is known about the sequence of these motifs, the functions of these

motifs are not well understood at the molecular level. cDNA-encoded peptides from the portion of titin located in the central region of the A-band showed interactions with C- protein and myosin in solid phase binding assays (Labeit et al., 1992). These investigators suggested that the interaction with myosin is additive in that increasing the 76 number of motifs in the expressed peptides increased the interactions. Although motifs I and n were implicated in the interactions, it is not known whether motifs I and II contribute equally to the interactions or whether a certain type of motif has higher affinity for a certain protein. Furthermore, it is not known if motifs from other portions of titin can interact with myosin.

We have recently reported the isolation and characterization of cDNAs coding for the region of chicken titin located in the A-band near the A-I junction, near the ends of the thick filaments (Tan et al., 1993). In this paper, we examined the interaction of expressed titin fragments, containing different numbers and arrangements of the two motifs, with myosin and actin. 77

MATERIALS AND METHODS

Construction of expression vectors. Chicken muscle titin cDNA clones TZ-3 and

TZ-6 in pBluescript (SK-) were digested with Pst I and EcoR l-Pst I, respectively. The resulting 1.3 kb Pst I fragment of TZ-3 (TP-3) and 1.2 kb EcoR l-Pst I fragment of TZ-6

(TEP-6) were gel purified. TEP-6 was subcloned into pUC-118 plasmid vector and the purified plasmid was linearized with Pst I. TP-3 was then ligated to TEP-6 to generate clone EPP in pUC-118 (pUC-EPP). The orientation of cDNA fragments in EPP was verified by restriction digests and sequencing. Individual expression vectors were constructed in pQE-30 plasmid vector (Qiagen). Three restriction fragments of EPP, Sst

1-Bam HI (SB), Sst l-Hind III (SH), and Sst l-Pst I (SP), were subcloned into the plasmid to generate expression vectors E-SB ,E-SH, and E-SP, respectively. In addition, two restriction fragments, Hind Ill-Hind III and Hind IIWvo I, of clone TZ-3 in pBluescript

(pBL-TZ3) were also subcloned into pQE-30 to yield expression vectors E-HH and E-HA, respectively. The identity of restriction fragments in pQE-30 was verified by sequencing.

Addition of C-terminal epitope tag. Proteins expressed from the vectors E-SH, and

E-SP were tagged with a peptide (HSV-Tag) consisting of 11 amino acids of the following sequence: Gln-Pro-Glu-Leu-Ala-Pro-Glu-Asp-Pro-Glu-Asp. This epitope tag is derived from glycoprotein D of herpes simplex virus (HSV) and is recognized by HSV-Tag monoclonal antibody (HSV.Tag antibody, Novagen). To add the tag to the proteins, double stranded oligo-nucleotide (HSV-Tag 50 mer, Novagen) with the sequence 5'-A-

CAG-CCT-GAA-CTC-GCT-CCA-GAG-GAT-CCG-GAA-GAT-TAA-GTG-A1T-AAC- 78

TCA-T-3' was first phosphorylated with T4 polynucleotide kinase, and then ligated to the

3' end of restriction fragments in pQE-30 according to standard methods (Maniatis et al.,

1982).

Expression and purification of proteins. E. coli (SG130009[pREP4]) containing the above expression vectors was grown in LB medium supplemented with ampicillin (100/xg/ml) and Kanamycin (25^ig/ml) until O.D.6oo=0.9 and induced with 2 mM IPTG for 5 hr at 37°C. Cells were collected by centrifugation at 4000 x g for 20 min and the pellet was suspended in sonication buffer (50 mM Na-phosphate, pH 8.0, 500 mM NaCl, 5 ^M E64, and 0.1 mM PMSF). The resuspended pellet was stored at -20°C overnight and subsequently thawed and sonicated. Cellular debris was pelleted by centrifugation at 10,000 x g and the supernatant was incubated with Ni-NTA resin

(Qiagen) on ice for 1 hr. The resin was then loaded into a small column and the column was washed sequentially with sonication buffer and sonication buffer with added 40 mM imidazole-HCl, pH 7.0, until the A^go of the eluant was < 0.01 for each wash. Proteins were eluted from the column by sequential application of 100 mM and 250 mM imidazole-HCl, pH 7, in sonication buffer. Fractions were collected and stored at -20°C until used.

Actin was purified from porcine skeletal muscle as described (Spudich and Watt,

1971) and further purified on a Sephacryl S-300 HR (Pharmacia) column. Troponin C was purified from beef skeletal muscle by use of DEAE chromatography (Arakawa et al.,

1970; Margossian and Cohen, 1973, Eisenberg and Kielley, 1974). Myosin was prepared from chicken leg muscle and separated from C- and M-proteins by DEAE-Sephadex 79

chromatography according to Pastra-Landis et al. (1983). Chicken muscle tropomyosin

was purchased from Sigma (T 1646).

Gel electrophoresis and Western blotting. SDS-PAGE was done essentially according to Laemmli et al. (1970). After electrophoresis, proteins were electrophoretically transferred to nylon membrane (Zeta-Probe, Bio-Rad) in a Bio-Rad

Trans-Blot cell operated at 55 V overnight. Transfer buffer consisted of 25 mM Tris-

glycine, pH 8.0, 0.01% SDS, with no methanol. Antibody incubations were done

essentially as described elsewhere (Tan et al., 1993).

Dot-blot binding assay. Myosin, actin, bovine serum albumin (BSA), tropomyosin

(TM), and troponin C (TN-C) were serially diluted with coating buffer (50 mM Tris-HCl,

pH 8, 500 mM KCl, 1 mM EDTA, and 0.3 mM DTT [Soteriou et al., 1993]), and 2 /il of each dilution was spotted on nitrocellulose membranes. The membranes were dried at

37°C for 30 min and then blocked with blocking buffer (20 mM sodium phosphate, pH 7,

100 mM KCl, 1 % BSA, and 1 % Tween 20) for 1 hr with three changes. Membranes

were then incubated with expressed titin fragments E-SH-HSV or E-SP-HSV, in blocking buffer, at 4°C overnight. After 3 washes with the blocking buffer, membranes were

incubated with HSV-Tag monoclonal antibody (1:10,000) diluted with blocking buffer containing 0,2% Tween 20 (Sigma) at room temperature for 1 hr. Membranes were then

washed 3 times and incubated with alkaline phosphatase-conj ugated second antibody

(1:10(X)) at room temperature for 1 hr. After 3 washes, bound antibodies were detected

with NBT (nitro blue tetrazolium) and BCIP (5-bromo-4-chloro-3-indolyl-phosphate)

(Promega). Detection of myosin with monoclonal antibody MF-20, prepared against 80 chicken skeletal muscle myosin, was carried out as described for Western blots (Tan et al., 1993). Monoclonal antibody MF-20 (Bader et al., 1982) was obtained from the

Developmental Studies Hybridoma Bank (Johns Hopkins University School of Medicine,

Baltimore, MD).

Microliter plate assay. Plates were coated with proteins diluted in buffer S (50 mM Tris-HCl, pH 8, 500 mM KCl, 1 mM EDTA, and 0.3 mM DTT [Soteriou et al.,

1993]) or for actin, in buffer A (10 mM Tris-HCl, pH 8, 100 mM KCl) at a concentration of 4/xg/ml for 1 hr at 37 °C and then overnight at 4°C. The plates were then incubated with blocking buffer (20 mM sodium phosphate, pH 7, 100 or 200 mM

KCl, 1 % BSA, and 1 % Tween 20) at room temperature for 2 hr with 5-6 changes of buffer. Cloned titin fragments diluted in blocking buffer were then added and the plates incubated overnight at 4°C. After incubation, plates were washed three times with the blocking buffer. Detection of cloned titin fragments with monoclonal antibody specific for the HSV-tag (1:10,(XX)) was carried out essentially as described above for the dot-blot assay. Signals were developed with alkaline phosphatase substrate 104 (Sigma) and quantitated with a microplate reader. Values reported are the average of readings from 3-

4 wells. 81

RESULTS

Bacterial expression of titin polypeptides. Two overlapping chicken muscle titin cDNA clones, TZ-6 and TZ-3, in vector pBluescript, were obtained as described previously (Tan et al., 1993). Together, these clones contain 2.45 kb of contiguous titin sequence, from a region of the titin molecule located in the A-band ~ 0.1 ixm from the ends of the thick filaments in myofibrils. Restriction fragments of these clones were ligated and cloned in pUC-118 to obtain a single cDNA clone, clone EPP, of 2.45 kb.

The arrangement of the class I and II motifs in this clone is shown in Fig. 1, along with restriction fragments generated from this clone. Restriction fragments of EPP coding for approximately 7, 5, and 3 motifs (SB, SH, and SP, respectively) were subcloned in expression vector pQE-30 to obtain a series of clones coding for polypeptides of different length. In addition, restriction fragments of TZ-3 were subcloned in pQE-30 to obtain clones HH and HA, which code for approximately 3 and 2 1/2 motifs, respectively.

From the sizes of the restriction fragments, SB (2.098 kb), SH (1.703 kb), SP (0.967 kb),

HH (0.939 kb), and HA (0.795 kb), the expected sizes of the proteins expressed from these fragments are 78 kDa (E-SB), 64 kDa (E-SH), 36 kDa (E-SP), 35 kDa (E-HH), and

30 kDa (E-HA), respectively. Two proteins, E-SH-HSV and E-SP-HSV were tagged C- terminally with HSV-tag, an 11-amino acid epitope tag that allows detection with a specific monoclonal antibody. The addition of the HSV-tag (11 residues) increased the expected molecular weight of the proteins by approximately 1.2 kDa. Figure 1. Arrangement of motifs I and H and generation of deletion clones

Overlapping chicken muscle titin cDNA clones TZ-6 and TZ-3 obtained previously (Tan et al., 1993) were ligated at

the Pst I site to generate clone EPF. Restriction fragments SB, SH, and SP were obtained by restriction digests of

EPP. Restriction fragments HH and HA were obtained by restriction digests of TZ-3. Ed represents the 5'-end of

TZ-3. Furthermore, a double stranded oligonucleotide encoding 11 amino acid residues (HSV-Tag, Novagen) of

glycoprotein D from the herpes simplex virus was ligated to the 3' end of restriction fragments, SH and SP. +/-HSV,

with and without the HSV-Tag, respectively. The expected size of the protein coded for by each restriction fragment

is indicated at the right. EcoR l-Not I adaptor Sst I Xba I

Ed

PSt I

EcoR I

EcoR I

Ava I Hind III

BamH I

EcoR l-Not I adaptor 84

An agarose gel of the restriction fragments excised from clones EPF and TZ-3 is shown in Fig. 2. Expression of the cDNA fragments from pQE-30 vector provided for inclusion of 6 histidine residues at the N-terminus of the expressed proteins, and this His- tag allowed subsequent purification of the proteins on Ni-NTA column. All proteins were purified under non-denaturing conditions, and eluted with 100 mM imidazole-HCl, pH 7.

Purified proteins were analyzed on SDS-PAGE (Fig. 3). Multiple bands were observed for all of the purified fragments except for E-HH. The presence of more than one band could due to several factors, including degradation, even though the protease inhibitors

E64 and PMSF were added during the protein preparation, read-through events that could result in added amino acids at the C-terminus of the proteins, or anomalous mobility due to addition of the 6x-His tag. The E-SB fragment appeared to be degraded from the expected size of 78 kDa to 66 kDa, while E-SH-HSV migrated as a doublet near the expected size of 64 kDa. Fragment E-HH exhibited a single band with a higher than expected size (40 instead of 35 kDa). The major band of both E-SP and E-SP-HSV was at near the expected molecular weight. The presence of HSV-tag at C-terminus of E-SH-

HSV and E-SP-HSV was detected with HSV-Tag monoclonal antibody on a Western blot

(Fig. 3).

Dot-blot binding assay. A nitrocellulose dot-blot assay was used to examine possible interactions of fragments E-SH-HSV (5 motifs) and E-SP-HSV (3 motifs) with selected myofibrillar proteins (Fig. 4). Cloned titin fragments, E-SH-HSV and E-SP-

HSV, showed binding to both myosin and actin (Fig. 4a and 4b). The interactions of both fragments were stronger with myosin than with actin. No significant interaction of either Figure 2. Agarose gel of restriction digests of titin clones

Restriction fragments (SB, SH, SP, HH, and HA) from restriction digests (see

Fig.l) of EPF in pUC 118 (pUC-EPP) and TZ-3 in pBluescript (pBL-TZ3) were

gel purified and subcloned into pQE-30 expression vector (Qiagen). Lanes a-d and

lanes e-f are restriction digests of pUC-EPP and pBL-TZ3, respectively, a. EPF

was liberated from pUC-EPP with Not L b. SB fragment was generated by

restriction digests of pUC-EPP with Sst I and BamH L c and d. pUC-EPP was

first digested with Sst I and Hind III (c and d), and subsequently with Pst I (d). e.

pBL-TZ3 digested with Hind III. f. pBL-TZ3 digested simultaneously with Hind

III and Ava I. [

pUC-EPP PBL-TZ3 , , 1 KW

23.1- 9.41- 655- 4.36-

2.32-; 202-

SP HH

0.56n

a b c d e f Figure 3. Characterization of expressed titin fragments by SDS-PAGE and

inununoblotting

Cloned titin fragments E-SB, E-SH-HSV, E-HH, E-SP-HSV, and E-SP were

expressed in E.coli, purified on Ni-NTA columns, and eluted with 100 mM

imidazole-HCl. At the left is SDS-PAGE of purified fragments on a 10%

acrylamide gel stained with Coomassie blue. At the right is an immunoblot of a

duplicate gel incubated with monoclonal antibody to the HSV-tag. Both E-SH-

HSV and E-SP-HSV reacted with the HSV-tag antibody. lO w in f ? ? i 0 Marker I E-SB E-SH-HSV 9 E-HH E-SP-HSV I E^P A/lyosin

E-SB E-SH-HSV E-HH E-SP-HSV E-SP Myosin Figure 4. Dot-blot binding assay of cloned titin fragments

Myosin, bovine serum albumin (BSA), actin, troponin C (TN-C), and tropomyosin

(TM) were serially diluted with coating buffer (see Materials and Methods) and 2

ixl of each dilution were spotted on nitrocellulose membranes. The membranes

were incubated with cloned titin fragments, E-SH-HSV or E-SP-HSV in blocking

buffer containing 100 mM KCl. a. Filter incubated with E-SH-HSV (10 nM). b.

Filter incubated with E-SP-HSV (10 nM). C. Control filter without titin fragment.

Binding was detected with HSV-tag monoclonal antibody and alkaline phosphatase

second antibody, d. Identical blot incubated with myosin monoclonal antibody

MF-20, to reveal any myosin contaiminations in the proteins spotted on the

nitrocellulose. ng 500 250 125 63 31 500 250 125 63 31 Myosin

Myosin

Rem E-SH-HSV or E-SP-HSV was observed with troponin C (TN-C) or tropomyosin (TM).

The weak interaction of both expressed fragments with actin was not due to myosin contamination because the myosin-specific monoclonal antibody, MF-20, did not react with actin spotted on the nitrocellulose (Fig. 4d). On SDS-PAGE (not shown), no myosin was observed in any of the other protein samples.

Microliter plate binding assay. A more quantitative, microtiter plate binding assay was used to further characterize the interactions of expressed titin fragments with myosin and actin. Quantitative analysis of the interaction of expressed titin fragment E-SH-HSV with selected myofibrillar proteins on a microtiter plate assay (Fig. 5) shows that the interaction with myosin was several fold greater than the interaction with actin.

Absorbance values obtained for myosin were several times greater than that for actin, as much as 5 times in some instances. Again, no significant binding of E-SH-HSV to troponin C (TN-C) or tropomyosin (TM) was observed.

This initial assay was done in 100 mM KCl. To examine the effect of KCl on the binding, microtiter plate assays of binding of E-SH-HSV and myosin or actin was done at different concentrations of KCl (Fig. 6). Results demonstrated that the interactions with both myosin and actin were affected by the KCl concentration used in the binding buffer.

Greater interactions of titin fragment E-SH-HSV with either myosin or actin occurred when concentration of KCl was lowered. To study the interactions at physiological ionic strength, binding buffer containing 20 mM Na-phosphate, pH 7, and 100 or 200 mM of

KCl was used in subsequent experiments. Figure S. Microtiter plate assay of the interaction of cloned titin fragment E-SH-

HSV with selected myofibrillar proteins

Myosin, actin, tropomyosin (TM), and troponin-C (TN-C) coated on a microtiter

plate were incubated with increasing concentrations of E-SH-HSV in blocking

buffer containing 100 mM KCl (see Materials and Methods). Binding was

detected with HSV-tag monoclonal antibody and alkaline phosphatase-conj ugated

second antibody and quantitated by using a microtiter plate reader as described in

Materials and Methods. The absorbance values were corrected for background

absorbance from wells coated with bovine serum albumin. Absorbance (405 nm) p p o ^ b) ôo b ro

*

0)

CO ^1. S

4 H Z I O Figure 6. Effect of KCl concentration on binding of cloned titin fragment E-SH-HSV

to myosin and actin

Myosin and actin were coated on microtiter plate as described in Materials and

Methods. Wells were incubated with E-SH-HSV diluted to 30 nM with blocking

buffer containing different concentrations of KCl. 95

0.8

0.6

0.2

0 100 200 300 400 [KCI] (mM) 96

To establish that the binding of the expressed titin fragments to myosin was not due to the HSV-tag added to the C-terminus of the proteins, increasing concentrations of

E-SP (no HSV-tag present at the C-terminus) or lysozyme was used to compete with 30 nM E-SP-HSV. Increased concentrations of E-SP, but not lysozyme, effectively reduced the binding of E-SP-HSV to myosin coated on a microtiter plate (Fig. 7). Approximately

50-60 % reduction in signal was observed when 30 nM of E-SP was used to compete with

30 nM of E-SP-HSV.

When the binding of titin fragment E-SH-HSV to myosin was compared to the binding of E-SP-HSV to myosin, E-SP-HSV generally produced stronger signals than E-

SH-HSV on microtiter plate assays (Fig. 8). Since the difference in binding could be due to differences in affinity for myosin or to differences in the number of binding sites available on the myosin molecules for each titin fragment (assuming that the HSV-tag monoclonal antibody has the same affinity for the HSV-tag at the C-terminus of E-SH-

HSV or E-SP-HSV), a competition assay was subsequently used to compare the affinity of the titin fragments for myosin. Increasing concentrations of E-SB or E-HH were mixed with 30 nM of E-SH-HSV or E-SP-HSV and incubated with myosin coated on microtiter plate. Fig. 9 shows that E-SB (7 motifs) competed better than E-HH (3 motifs) with either E-SH-HSV or E-SP-HSV, suggesting that titin fragments with a greater number of motifs have higher affinity for myosin. Although E-SH-HSV (5 motifs) competed better than E-SP-HSV (3 motifs) with either E-SB or E-HH, the difference in competition efficiency between E-SH-HSV and E-SP-HSV is quite small. Figure 7. Binding specificity of cloned titin fragment to myosin

Increasing concentrations of E-SP or lysozyme were mixed with 30 nM E-SP-HSV

in blocking buffer containing 100 mM KCl and added to wells coated with myosin.

Competition was measured as the percentage of maximum absorbance obtainable in

the presence of E-SP-HSV (30 nM) alone, without E-SP or lysozyme. 98

120

100

0 O) 80 (0 £ 3

1 60 E o

40

0 20 40 60 80 100 [E-SP] or [Lysozyme] (nM)

[-e-E-SP -•-Lysozyme) Figure 8. Interaction of E-SH-HSV and E-SP-HSV with myosin as measured by

microtiter plate assay

Increasing concentrations of E-SH-HSV or E-SP-HSV were incubated with myosin

coated on microliter plate in the absence of other competing titin fragments. The

binding assays were carried out in blocking buffer containing 100 mM KCl. 100

1.4

1.2

1.0

0.8

0.6 n 0.4

0.2

0.0 0 10 20 30 40 50 60 70 80 90 100 [E-SH-HSV] or [E-SP-HSV] (nM)

E-SH-HSV -e-E-SP-Hs"^ Figure 9. Competition assay of cloned titin fragments

Increasing concentrations of E-SB or E-HH were mixed with 30 nM of E-SH-HSV

or E-SP-HSV in blocking buffer containing 100 mM KCl and added to wells

coated with myosin. Competitions were measured as the percentage of maximum

values obtainable in the presence of 30 nM E-SH-HSV or E-SP-HSV, without the

competing titin fragments (E-SB or E-HH). Squares and circles represent binding

of E-SH-HSV and E-SP-HSV, respectively, in the presence of E-HH (thin line) or

E-SB (heavy line). Triangles represent competition assays of E-SH-HSV and E-

SP-HSV in the presence of lysozyme. 102

120 r

100

(D •e§ I ! e a

20 40 60 80 [E-SB] or [E-HH] or [Lysozyme] (nM)

^E-SH-HSV (E-SB) -e-E-SP-HSV (E-SB)

-e- E-SH-HSV (E-HH) ^ E-SP-HSV (E-HH) E-SH/SP-HSV (Lyso) 103

To compare titin fragments of similar molecular weight but with different arrangements of motifs, E-SP and E-HH were used to compete with E-SH-HSV. E-SP contains the motif pattern I-II-I, and the motif pattern of E-HH is I,/2-I-I-n,/2. Thus, E-SP possesses a complete copy of motif n, while motif II of E-HH is incomplete, truncated by approximately 50 %. On the other hand, E-HH has 2 1/2 copies of motif I in comparison to 2 copies of motif I in E-SP. E-SH-HSV contains all the motifs I and n present in both

E-SP and E-HH (see Fig. 1). In competition assays, increasing concentrations of E-HH and E-SP were used to compete with 30 nM of E-SH-HSV (Fig. 10). As shown in Fig.

10a, both titin fragments competed equally well with E-SH-HSV in buffer containing 1(X) mM KCl. However, when KCl concentration was increased to 200 mM, E-SP competed with E-SH-HSV better than E-HH (Fig. 10b).

All of the interaction studies of expressed titin fragments with myosin described above were carried out with monomeric myosin, because plates were coated with myosin in 0.5 M KCl, using a procedure similar to that described by Labeit et al. (1992). This was done to provide direct comparison with titin fragments from A-I junction region to the reported results with titin fragments from the central A-band region. To determine any difference in interaction of titin fragments with myosin and actin filaments, increasing concentrations of myosin filaments and F-actin were mixed with E-SH-HSV (30 nM) in microtiter plate coated with myosin. In binding buffer containing 200 mM KCl, neither myosin nor F-actin competed significantly with myosin (Fig 11a). However, when the

KCl concentration was lowered to 20 mM, myosin filaments competed with myosin coated Figure 10. Cloned titin fragments differing in motif arrangements exhibit

differential binding to myosin

Increasing concentrations of E-HH (I,/2-I-I-IIi/2) or E-SP (I-II-I) were mixed with

E-SH-HSV (30 nM) in microtiter plate coated with myosin. Based on cDNA

sequence, the molecular weight of E-HH and E-SP are approximately 35 kDa;

however, E-HH and E-SP differ in the completeness of motif II. a. Competition

assay carried out in blocking buffer containing 100 mM KCl. b. Same assay was

carried out in blocking buffer containing 200 mM KCl. % of maximum absorbance % of maximum absorbance

o> 00 o o o o

% % o

m 05 "D 0> Figure 11. Preferential binding of cloned titin fragment to myosin over myosin

filaments at high KCl concentration

Myosin filaments were prepared by dialyzing myosin against 20 mM K"*"-

phosphate, pH 7, 100 mM KCl overnight. F-actin was prepared by dilution of

actin into buffer A (10 mM Tris-HCl, pH 8, 100 mM KCl) containing 10 mM

Mg^"^. Increasing concentrations of myosin filaments or F-actin were mixed with

30 nM of E-SH-HSV in phosphate buffer containing 200 or 20 mM KCl (plus 10

mM Mg^"*" for actin filament) and incubated with myosin coated on microtiter

plate, a. Competition assay conducted in buffer containing 200 mM KCl. b.

Same assay conducted in buffer containing 20 mM KCl. Myosin-F, myosin

filaments; Actin-F, actin filaments. 107

120

E 1001

80

60 -

40

20 0 20 40 60 80 100 [Myosin or Actin Filaments] or [Lysozyme] (nM) (-Q-Myosin-F *-Actin-F T^rLysozyrne)

1001

8

§ 60 I E Af\ o

20 0 20 40 60 80 100 [Myosin or Actin Filaments] or [Lysozyme] (nM) [•e-myosin-F -«-Actin-F -a-Lysozyme] 108 on the microtiter plate (Fig. 1 lb). Much weaker competition was observed for F-actin filaments. No competition was observed when lysozyme was used. 109

DISCUSSION

We reported earlier the isolation of titin cDNAs encoding an A-I junction portion of titin (Tan et al., 1993). To determine whether this A-I junction portion of titin could interact with myofibrillar proteins, cloned titin fragments of different size were expressed in E.coli and purified, and possible interactions of expressed titin fragments with selected myofibrillar proteins were determined by solid phase binding assays.

Both dot-blot binding assays and microtiter plate assays have shown the binding of titin fragment E-SH-HSV to myosin and actin. The interaction is stronger with myosin than with actin. Interaction of central A-band portion of titin with myosin has been reported (Labeit et al. 1992). Thus, both the central A-band and the A-I junction regions of titin are capable of binding to myosin. However, it remains to be determined whether the two different parts of titin could bind equally well to myosin. Comparison of the derived amino acid sequences of titin have shown that copies of motif I at the A-I junction portion of titin (Tan et al., 1993) are not as conserved as copies of motif I at the central

A-band region of titin (Labeit et al., 1992); therefore, it is likely that the two different parts of titin could bind to myosin with different affinities.

Titin fragment E-SH-HSV, from the A-I junction region, also exhibits lesser but significant binding to actin. The actin binding capability was not found in expressed fragments from the central A-band portion of titin (Labeit et al., 1992). Thus, the A-I junction region of titin is unique in that it interacts with both myosin and actin. However, the significance of actin binding is not clear. The ability to interact with myosin or actin 110 is not restricted to titin fragments, as purified native titin also binds to either myosin or actin in a solid phase binding assay (Soteriou et al. 1993), and the observed binding of native titin to myosin was stronger than the binding to actin. In addition, the purified native titin also interacted with C-protein, X-protein, and AMP-deaminase (Soteriou et al.,

1993). Furthermore, the T2 portion of titin (P-connectin) also interacts with myosin filaments and F-actin (actin filaments) as judged by electron microscopy and a combination of other methods (Kimura et al., 1984; Maruyama et al., 1987).

Interactions of titin fragments with myosin and actin are affected by the concentration of KCl (Fig. 6). The interaction with myosin or actin increased when KCl concentration was decreased. These results provide an indication that electrostatic forces are involved in the interactions. Similar effect of KCl on the interaction of native titin with myosin filaments or F-actin was also reported (Kimura et al., 1984; Maruyama et al., 1987). To carry out interaction studies near physiological ionic strength, buffers made up of 20 mM sodium phosphate, pH 7, containing 100 or 200 mM KCl were used.

Smaller titin fragments may be able to interact with myosin at more than one binding site on the myosin molecule. This idea is supported by results from microti ter plate assays (Fig. 8), which showed that titin fragment E-SP-HSV (3 motifs) interacts with myosin better than E-SH-HSV (5 motifs). It is possible that the small size of E-SP-HSV allows the molecule to interact with more than one binding site or similar binding sites on the myosin molecules. Although the binding capability of E-SP-HSV is better than E-SH-

HSV, E-SP-HSV has lower affinity for myosin than E-SH-HSV in competition assays

(Fig. 9) which show that larger titin fragments have higher affinity for myosin than Ill smaller titin fragments. In the competition assays, E-SB (7 motifs) competed much better than E-HH (3 motifs) with titin fragments E-SH-HSV (5 motifs) and E-SP-HSV (3

motifs). These results indicate that E-SB has higher affinity for myosin than E-HH. On the other hand, the difference in binding affinity between E-SH-HSV (5 motifs) and E-SP-

HSV (3 motifs) is small which may be due to the smaller size difference between these

two titin fragments (about 2 motifs). In comparison, E-SB and E-HH differ in size by approximately 4 motifs.

When comparing binding efficiency of E-HH (I,/2-I-I-II,/2) and E-SP (I-II-I), which have similar molecular weight but differ in motif arrangement and completeness of motif

II, E-SP showed stronger binding to myosin than E-HH at higher concentration of KCl

(200 mM as opposed to 100 mM, Fig. 10a and 10b). These results suggest that intact

motif n may be involved in binding to myosin with high affinity. Although important,

motif II is not likely to be sufficient for a strong interaction between titin and myosin. It is likely that a strong interaction between titin and myosin requires contribution of both

motifs I and II from the titin molecule. This idea is supported by the fact that titin fragment E-HA (Fig. 1), lacking motif II, can also compete for binding to myosin with

titin fragment E-SH-HSV (data not shown). Besides the individual contribution of motifs

I and II of the titin fragments to the interaction with myosin, the overall motif arrangement in the titin fragments may also contribute to the interaction. It is conceivable

that the motif arrangement of titin fragments could affect interaction with myosin

filaments, which have a structural repeat of 43-nm. Trinick (1992) has suggested that the

11-motif super-repeat (I-I-I-II-I-I-I-II-I-I-II) of titin (Labeit et al., 1992) may correspond 112 to the structural repeat of the myosin filament. The contribution of motif arrangement to the binding of myosin filament is further supported by the fact that the myosin-associated proteins C-protein (Einheber and Fischman, 1990, Furst et al., 1992; Weber et al., 1993),

86-kDa protein (Vaughan et al., 1993a, b), M-protein (Noguchi et al., 1992), and skelemin (Price and Gomer, 1993) all have different arrangements of motifs I and n and are all colocalized with myosin filaments at different regions of the filaments in the A- band of myofibrils.

Although cloned titin fragments interact with myosin in buffer containing fairly high concentrations of KCl (100-200 mM), titin fragment E-SH-HSV did not interact with myosin filaments under the same conditions (Fig 11a). However, significant interactions occurred when the KCl concentration was lowered to 20 mM (Fig. lib). Interactions were seen at KCl concentrations as high as 50 mM (data not shown). It is very likely that the formation of filaments could lead to structural changes of the myosin or that some binding sites could be covered in the filaments. Such changes could reduce the number of binding sites available on myosin as well as reduce the interaction between the proteins due to higher packing order of the filaments that would generate steric hindrance.

Although these changes may be affected by concentration of KCl, the electrostatic interaction between titin fragment and myosin is likely to be affected the most. Thus, in order for titin fragments to interact with possibly a limited number of binding sites on

myosin, low KCl concentration is required to enhance the interaction. Since larger titin fragments have higher affinity for myosin (see figure 9), it is possible that larger titin fragments could tolerate higher concentration of KCl better than smaller titin fragments. 113

Taken together, these results suggest that significant interaction of titin with myosin filaments in muscle cells require participation of many interacting domains, motifs I and

II, from titin molecules. 114

ACKNOWLEDGMENTS

We thank Jean Schmidt and Jinwen Zhang for providing a generous amount of actin, and Kim Ho for donating troponin-C. We would also thank Jo Philips for expert technical assistance in purification of myosin and preparation of MF-20 monoclonal antibody. 115

REFERENCES

Arakawa, N., R. M. Robson, and D. E. Goll. 1970. An improved method for the preparation of a-actinin from rabbit striated muscle. Biochim. Biophys. Acta 200:284-295.

Ashby, B., C. Frieden, and R. Bischoff. 1979. Immunofluorescent and histochemical localization of AMP deaminase in skeletal muscle. J. Cell Biol. 81:361-373.

Ayme-Southgate, A., J. Vigoreaux, G. Benian, and M. L. Pardue. 1991. Drosophila has a twitchin/titin related gene that appears to encode projectin. Proc. Natl. Acad. Sci. USA 88:7973-7977.

Bader, D., T. Masaki, and D. A. Fischman. 1982. Immunochemical analysis of myosin heavy chain during avian myogenesis in vivo and in vitro. J. Cell Biol. 95:763- 770.

Bahler, M., H. M. Eppenberger, and T. Wallimann. 1985a. Novel thick filament protein of chicken pectoralis muscle: the 86K protein. I. Purification and characterization. J. Mol. Biol. 186:381-391.

Bahler, M., H. M. Eppenberger, and T. Wallimann. 1985b. Novel thick filament protein of chicken pectoralis muscle: the 86K protein, II. Distribution and localization. J. Mol. Biol. 186:393-401.

Bar shop, B. A., and C. Frieden. 1989. Analysis of the interaction of rabbit skeletal muscle adenylate deaminase with myosin subfragments. J. Biol. Chem. 259:60- 66.

Benian, G. M., J. E. Kiff, N. Neckelmann, D. G. Moerman, R. H. Waterston. 1989. Sequence of an unusually large protein implicated in regulation of myosin activity in C. elegans. Nature 342:45-50.

Benian, G. M., S. W. L'Hernault, and M. E. Morris. 1993. Additional sequence complexity in muscle gene, unc-22, and its encoded protein, Twitchin, of Caenorhabditis elegans. Genetics 134:1097-1104,

Cooper J, and J, Trinick, 1984, Binding and location of AMP deaminase in rabbit psoas muscle myofibrils, J, Mol, Biol. 177:137-152. 116

Einheber, S., and D. A. Fischman. 1990. Isolation and characterization of a cDNA clone encoding avian skeletal muscle C-protein: An intracellular member of the immunoglobulin superfamily, Proc. Natl. Acad. Sci. USA 87:2157-2161.

Eisenberg, E., and W. W. Kielley. 1974. Troponin-tropomyosin complex. Column chromatographic separation and activity of the three active troponin components with and without tropomyosin present. J. Biol. Chem. 249:4742-4748.

Funatsu, T., E. Kono, H. Higuchi, S. Kimura, S. Ishiwata, T. Yoshioka, K. Maruyama, and S. Tsukita. 1993. Elastic filaments in situ in cardiac muscle: Deep-etch replica analysis in combination with selective removal of actin and myosin filaments. J. Cell Biol. 120:711-724.

Furst, D. O., M. Osborn, R. Nave, and K. Weber. 1988. The organization of titin filaments in half-sacromere revealed by monoclonal antibodies in immunoelectron microscopy: a map of ten nonrepetitive epitopes starting at the Z line extends close to the M line. J. Cell Biol. 106:1563-1572.

Furst, D. O., R. Nave, M. Osborn, and K. Weber. 1989. Repetitive titin epitopes with a 42 nm spacing coincide in relative position with known A-band striations also identified by major myosin-associated proteins. J. Cell Sci. 94:119-125.

Furst, D. O., U. Vinkemeier, and K. Weber. 1992. Mammalian skeletal muscle C- protein: Purification from bovine muscle, binding to titin and the characterization of a full-length human cDNA. J. Cell Sci. 102:769-778.

Fyrberg, C. C., S. Labeit, B. Bullard, K. Leonard, and E. Fyrberg. 1992. Drosophila projectin: relatedness to titin and twitchin and correlation with lethal(4)102 CDa and bent-Dominant. Proc. R. Soc. Lond. B 249:33-40.

Gautel, M., K. Leonard, and S. Labeit. 1993. Phosphorylation of KSP motifs in the C- terminal region of titin in differentiating myoblasts. EMBO J. 12:3827-3834.

Grove, B. K., V. Kurer, C. Lehner, T. C. Doetschman, J.-C. Perriard, and H. M. Eppenberger. 1984. A new 185,000-dalton skeletal muscle protein detected by monoclonal antibodies. J. Cell Biol. 98:518-524.

Huxley, A. F., and L. D. Peachey. 1961. The maximum length for contraction in vertebrate striated muscle. J.Physiol. 156:150-165.

Huxley, H. E. 1963. Electron microscope studies on the structure of natural and synthetic protein filaments from striated muscle. J. Mol. Biol. 7:281-308. 117

Itoh, Y., T. Suzuki, S. Kimura, K. Ohashi, H. Higuchi, H. Sawada, T. Shimizu, M. Shibata, and K. Maruyama. 1988. Extensible and less extensible domains of connectin filaments in stretched vertebrate skeletal muscle sacromeres as detected by immunofluorescence and immunoelectron microscopy using monoclonal antibodies. J. Biochem. 104:504-508.

Kimura, S., K. Maruyama, and Y. P. Huang. 1984. Interactions of muscle 6-connectin with myosin, actin, and actomyosin at low ionic strengths. J. Biochem. 96:499- 506.

Kurzban, G. P., and K. Wang. 1988. Giant polypeptides of skeletal muscle titin: sedimentation equilibrium in guanidine hydrochloride. Biochem. Biophy. Res. Comm. 150:1155-1161.

Labeit, S., D. P. Barlow, M. Gautel, T. Gibson, J. Holt, C-L. Hsieh, U. Francke, K. Leonard, J. Wardale, A. Whiting, and J. A. Trinick. 1990. Regular pattern of two types of 100-residue motif in the sequence of titin. Nature 345:273-276.

Labeit, S., M. Gautel, A. Lakey, J. Trinick. 1992. Towards a molecular understanding oftitin. EMBO J. 11:1711-1716.

Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685.

Lakey, A., C. Ferguson, S. Labeit, M. Reedy, A. Larkin, G. Butcher, K. Leonard, and B. Bullard. 1990. Identification and localization of high molecular weight proteins in insect flight and leg muscle. EMBO J. 9:3459-3467.

Lakey, A., S. Labeit, M. Gautel, C. Ferguson, D. P. Barlow, K. Leonard, and B. Bullard. 1993. Kettin, a large modular protein in the Z-disc of insect muscles. EMBO J. 12:2863-2871.

LaSalle, F., R. M. Robson, M. L. Lusby, F. C. Parrish, M. H. Stromer, and T. W. Huiatt. 1983. Localization of titin in bovine skeletal muscle by immunofluorescence and immunoelectron microscope labeling. J. Cell Biol. 97: 258a (abstr.)

Locker, R. H., and N. G. Leet. 1975. Histology of highly-stretched beef muscle. I. The fine structure of grossly stretched single fibers. J. Ultrastruct. Res. 52:64-75.

Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 118

Margossian, S. S., and C. Cohen. 1973. Troponin subunit interactions. J. Mol. Biol. 81:409-413.

Marquetant, R., R. L. Sabina, and E. W. Holmes. 1989. Identification of a non- catalytic domain in AMP deaminase that influences binding to myosin. Biochemistry 28:8744-8749.

Maruyama, K., S. Kimura, K. Ohashi, and Y. Kawano. 1981. Connectin, an elastic protein of muscle. Identification of " titin" with connectin. J. Biochem. 89, 701- 709.

Maruyama, K., S. Kimura., H. Yoshidomi, H. Sawada, and M. Kikuchi. 1984. Molecular size and shape of P-connectin, an elastic protein of striated muscle. J. Biochem. 95:1423-1493.

Maruyama, K., T. Yoshioka, H. Higuchi, K. Ohashi, S. Kimura, and R. Natori. 1985. Connectin filaments link thick filaments and Z lines in frog skeletal muscle as revealed by immunoelectron microscopy. J. Cell Biol. 101:2167-2172.

Maruyama, K., D. H. Hu, T. Suzuki, and S. Kimura. 1987. Binding of actin filaments to connectin filaments. J. Biochem. (Tokyo) 101:1339-1346.

Maruyama, K., T. Endo, H, Kume, Y. Kawamura, N. Kanzawa, Y. Nakauchi, S. Kimura, S. Kawashima, and K. Maruyama. 1993. A novel domain sequence of connectin localized at the I-band of skeletal muscle sarcomeres: Homology to neurofilament subunits. Biochem. Biophys. Res. Commun. 194:1288-1291.

Masaki, T and O. Takaiti. 1974. M-protein. J. Mol. Biochem, 75:367-380.

Nave, R., D. O. Furst, and K. Weber. 1989. Visualization of the polarity of isolated titin molecules: A single globular head on a long thin rod as the M-band anchoring domain? J. Cell Biol. 109:2177-2187.

Noguchi, J., M. Yanagisawa, M. Imamura, Y. Kasuya, T. Sakurai, T. Tanaka, and T. Masaki. 1992. Complete primary structure and tissue expression of chicken pectoralis M-protein. J. Biol. Chem. 267:20302-20310.

Offer, G., C. Moos, and R. Starr. 1973. A new protein of the thick filaments of vertebrate skeletal myofibrils. J. Mol. Biol. 74:653-676.

Olson, N. J., R. B. Pearson, D. S. Neadleman, M. Y. Hurwitz, B, E. Kemp, and A. R. Means. 1990. Regulatory and structural motifs of chicken gizzard myosin. Proc. Natl. Acad. Sci. USA 87:2284-2288. 119

Pastra-Landis, S. C., T. Huiatt, and S. Lowey. 1983. Assembly and kinetic properties of myosin light chain isozymes from fast skeletal muscle. J. Mol. Biol. 170:403- 422.

Price, M. G. 1987. Skelemins: Cytoskeletal proteins located at the periphery of M-discs in mammalian striated muscle. J. Cell Biol. 104:1325-1336.

Price, M. G., and R. H. Gomer. 1993. Skelemin, a cytoskeletal M-disc periphery protein, contains motifs of adhesion/recognition and intermediate filament proteins. J. Biol. Chem. 268:21800-21810.

Shoemaker, M. O., W. Lau, R. L. Shattuck, A. P. Kwiatkowski, P. E. Matrisian, L, Guerra-Santos, E. Wilson, T. J. Lukas, L. J. Van Eldik, and D. M. Watterson. 1990. Use of DNA sequence and mutant analyses and antisense oligodeoxynucleotides to examine the molecular basis of non-muscle myosin light chain kinase autoinhibition, calmodulin recognition, and activity. J. Cell Biol. 111:1107-1125.

Sjostrand, F. S. 1962. The connections between A- and I- band filaments in striated frog muscle. J. Ultrastruct. Res. 7:225-246.

Soteriou, A., M. Gamage, and J. Trinick. 1993. A survey of interactions made by the giant protein titin. J. Cell Sci. 104:119-123.

Spudich, J. A., and S. Watt. 1971. The regulation of rabbit skeletal muscle contraction. I. Biochemical studies on the interaction of the tropomyosin-troponin complex with actin and the proteolytic fragments of myosin. J. Biol. Chem. 246:4866-4871.

Tan, K. O., G. R. Sater, A. M. Myers, R. M. Robson, and T. W. Huiatt. 1993. Molecular characterization of avian muscle titin. J. Biol. Chem. 268:22900- 22907.

Trinick, J. A. 1981. End-filaments: a new structural element of vertebrate skeletal muscle thick filaments. J. Mol. Biol. 151:309-314.

Trinick, J., P. Knight, and A. Whiting. 1984. Purification and properties of native titin J. Mol. Biol. 180:331-356.

Trinick, J. 1991. Elastic filaments and giant proteins in muscle, Curr. Opinion Cell Biol. 3:112-119.

Trinick, J. 1992. Understanding the functions of titin and nebulin. FEBS Lett. 307:44- 48. 120

Vaughan, K. T., F. E. Weber, S. Einheber, and D. A. Fischman. 1993a, Molecular cloning of chicken myosin-binding protein (MyBP) H (86-kDa protein) reveals extensive homology with MyBP-C (C-protein) with conserved immunoglobulin C2 and fibronectin type III motifs. J. Biol. Chem. 268:3670-3676

Vaughan, K. T., F. E. Weber, F. C. Reinach, T. Ried, D. Ward, and D. A. Fischman. 1993b. Human myosin binding protein H (MyBP-H): complete primary sequence, repeat structure, genomic organization and chromosomal localization. Genomics 16:34-40.

Vinkemeier, U., W. Obermann, K. Weber, and D. O. Furst. 1993. The globular head domain of titin extends into the center of the sarcomeric M band: cDNA cloning, epitope mapping and immunoelectron microscopy of two titin-associated proteins. J. Cell Sci. 106:319-330.

Wang, K., J. McClure, and A. Tu. 1979. Titin: Major myofibrillar components of striated muscle. Proc. Natl. Acad. Sci. USA 76:3698-3702.

Wang , K. 1985. Sarcomere-associated cytoskeletal lattices in striated muscle. Review and hypothesis. In Cell and Muscle Motility, vol. 6 (J. W, Shay, ed.) Plenum Publishing Corp., New York, pp.315-369.

Weber, F. E., K. T. Vaughan, F. C. Reinach, and D. A. Fischman. 1993. Complete sequence of human fast-type and slow-type muscle myosin-binding protein C (MyBP-C): Differential expression, conserved domain structure and chromosomal assignment. Eur. J. Biochem. 216:661-669.

Wegmann, G., E. Zanolla, H. M. Eppenberger, and T. Wallimann. 1992, In situ compartmentation of creatine kinase in intact sarcomeric muscle: the acto-myosin overlap zone as a molecular sieve. J. Muscle Res. Cell Motil. 13: 420-435.

Whiting, A., J, Wardale, and J. Trinick. 1989. Does titin regulate the length of muscle thick filaments. J. Mol. Biol. 205:263-268.

Woodhead, J. L., and S. Lowey. 1982. Size and shape of skeletal muscle M-protein. J. Mol. Biol. 157:149-154. 121

OVERALL SUMMARY

The overal goals of the my dissertation project were to use molecular cloning

methods to determine a portion of the primary structure of the major myofibrillar protein

titin and to examine possible functions of the cloned region of the titin molecule. To accomplish the first of these goals, a cDNA library, in cloning vector X-Zap II, was

constructed from embryonic chick heart RNA size selected for large RNAs. From

screening of the cDNA library with a 0.2 kb cDNA probe previously isolated by antibody

screening of a X-gtll cDNA library, three cDNA clones encompassing a total of 2.5 kb of the coding region of titin were obtained. Immunological methods demonstrated immunoreactivity of the cDNAs encoded protein with titin polyclonal antibodies.

Immunofluorescence microscopy showed that titin polyclonal antibodies affinity purified against the cDNA encoded protein labeled near the A-I junction of chicken skeletal and cardiac muscle myofibrils. Thus, these cDNAs encode the region of titin located in the

A-band, adjacent to the A-I junction, in myofibrils. The affinity-purified antibodies also

reacted with purified skeletal and cardiac muscle titins on Western blots, confirming the immunofluorescence studies demonstrating that the cDNAs code for titin. These results are further supported by an RNase protection assay showing that the titin message encoded by the cDNA is present in both embryonic and adult skeletal (thigh and breast) and cardiac muscles, but not in smooth muscle, brain, or lung. However, it is not clear

whether the cDNA encoded message is specific for these muscles or if there are other titin isoforms that shared the sequence encoded by these cDNAs and are also expressed in 122

these muscles. The presence of titin isoforms that differ in molecular weight have been

reported (Wang et al., 1991; Horowits, 1992). In addition, some monoclonal antibodies are able to distinguish titins from heart and skeletal muscles (Hill and Weber, 1986).

However, these studies have indicated that titin isoforms differ in the I-band region of the

molecule. Furthermore, monoclonal and polyclonal antibodies that react with the A-band and/or A-I junction portion of titin recognize both skeletal and cardiac titins (Itoh et al.,

1986; Furst et al., 1988, 1989b; Pierobon-Bormioli et al., 1992). Thus, these results suggest that titin isoforms share the same or a very similar primary sequence at the A-

band and A-I junction regions.

Molecular methods have shown that titin is encoded by a large mRNA that has

"limited mobility" on agarose gels, and that this mRNA is specifically expressed in striated muscles. Furthermore, titin cDNA hybridized to a single band on a Southern blot of chick genomic DNA, suggesting that titin is encoded by a single copy gene. In human and mouse, titin is also encoded by a single gene (Labeit et al., 1990). Thus, it is very likely that the reported isoforms of titin that differ in molecular weights (Wang et al.,

1991; Horowits, 1992) are generated by alternative splicing of titin mRNA.

Analysis of the derived amino acid sequence of the chicken titin cDNA has shown

the presence of repeating motifs I and II, each about 100 amino acids long. These motifs are homologous to the type III domains of fibronectin and the C2 immunoglobulin superfamily, respectively. Motifs I and II are repeated in the sequence with a repeating pattern of I-I-II-I-I-I-II-I-I. Members of motif I shared a range of 25-45 % identity with an average of 35% identity. This identity is less than the average identity of 40% in 123 copies of motif I found in the A-band region of the reported rabbit titin sequence (Trinick,

1991). In the chicken sequence, the identity was lowest in the two complete copies of motif I nearest to the 5' end (N-terminus). It is possible that this difference in homology is due to the location of these motifs nearer to the I-band region. In general, motif I can be divided into two conserved regions that are separated by a non-conserved region in the middle of the motif. Three amino acid residues, tryptophan, leucine, and tyrosine are conserved through out all copies of motif I; therefore, these residues are likely to be functionally important. In addition, an unique serine-rich region (VSVSSSSKKRS) is found in the copy of motif I located at the 3' end (C-terminus) of the chicken sequence.

A similar serine-rich region is not found in the reported sequence of rabbit titin (Labeit et al., 1990, 1992; Trinick, 1991). The sequence includes the consensus phosphorylation site for at least one protein kinase (Pearson and Kemp, 1991). It is possible that phosphorylation of titin at this sequence could alter interactions between titin and myosin.

The two copies of motif II in chicken titin sequence shared 30% identity, which is similar to the percent identity found in motif lis of rabbit titin (Trinick, 1991). In comparison to motif I, the sequence of motif II is more variable. Furthermore, motif II shares only five out of the ten conserved residues found in the C2 set of the immunoglobulin superfamily (Tan et al., 1993). Although motif II, like motif I, is found in proteins associated with myosin, it is also found in at least one intracellular protein that is not associated with myosin. Kettin, a large protein (500-700 kDa) found in the Z- disc of insect muscles, also contains multiple copies of motif II ; however, each copy of motif n is separated from adjacent copies of motif II by a linker sequence made up of 35 124 amino acid residues (Lakey et al., 1993). In addition, multiple copies of motif II are also found in the portion of titin located in the myofibrillar I-band, which is the portion of titin that is not associated with myosin (Maruyama et al., 1993). It is likely that divergence in the sequence of motif II allows proteins with motif II or sequences similar to motif II to interact with different proteins. Interactions of motif n in kettin with actin and a-actinin have been reported (Lakey et al., 1993).

The second part of this study involved characterization of the interactions of cloned titin fragments with myosin and actin. Solid phase binding assays demonstrated interactions of cloned titin fragments with myosin and actin. The interactions of the titin fragments are stronger with myosin than with actin. In contrast, cloned titin fragments from the A-band portion of titin interact with myosin, but not with actin (Labeit et al.,

1992). In addition, the A-band titin fragments also interact with C-protein (Labeit et al.,

1992). Similar solid phase binding assays carried out with purified titin also showed interactions with myosin and actin, as well as C-protein, and AMP-deaminase (Soteriou et al., 1993). However, the reported interaction of titin with actin was weak (Soteriou et al., 1993). The studies in this dissertation have shown that the interactions are affected by the concentration of KCl and, in general, a lower KCl concentration allows stronger interactions. Therefore, the interactions are probably electrostatic. Additional interaction studies reported here were carried out in 20 mM sodium phosphate buffer containing 100 or 200 mM KCl to provide ionic strengths similar to physiological conditions.

When interactions of titin fragments that differed in number of motifs, I-II-I and I-

II-I-I-I-II,/2 (i.e. 3 motifs versus 5 motifs), with myosin were compared, the 3-motif titin 125 fragment generated a stronger signal than the 5-motif titin fragment on the microtiter plate assay. However, results from competition assays showed that the 3-motif titin fragment

has lower affinity for myosin than the 5-motif titin fragment. In addition, a 7-motif titin fragment (I-H-I-I-I-II-I) also produced stronger interactions with myosin than a 3-motif

titin fragment ) in competition assays. These results suggest that the 3-motif

titin fragment may interact with greater number of binding sites on myosin than the 5-

motif titin fragment. Taken together, these results suggest that the capability of titin

fragments to interact with myosin is determined by the number of motifs in each titin fragment that contribute to interactions, as well as the number of binding site available on

myosin. Although Labeit et al. (1992) have reported that the interaction with myosin is additive in that increasing the number of motifs I and II in the cloned titin fragments

would increase the interactions, the authors did not provide the identity of motifs that

made up the titin fragments that they used. Thus, it is not clear whether the additive effect was due to number of motifs alone or to the number of a certain type of motifs.

The latter could determine the number of binding sites on myosin molecules.

When binding to myosin of titin fragments of similar molecular weight that

differed in the completeness of motif II (i.e. I-II-I and I1/2-I-I-II1/2), were compared, the I-

II-I fragment bound equally well or better than the I„2-I-I-IIi/2 fragment to myosin in

competition assays (i.e., both competed against titin fragment made up of I-II-I-I-I-II,/2),

depending on the concentration of KCl used for the interaction studies. No significant difference in efficiency of binding between the two titin fragments was observed when the competition assay was conducted in phosphate buffer containing 100 mM KCl; however. 126

in phosphate buffer containing 200 mM KCl, the I-II-I fragment consistently competed

better than I,,2-1-1-111/2 fragment. These results suggest that intact motif II in E-SP has

high affinity for myosin. However, in the absence of motif II, titin fragment (I1/2-I-I) can

also compete with titin fragment (I-n-I-I-I-n,/2) for binding to myosin in phosphate buffer containing 100 mM KCl, suggesting that motif I can interact with myosin. Thus, it is likely that both motifs I and II are required for a strong interaction with myosin and motif

II may have higher affinity than motif I.

Although cloned titin fragments interact with myosin and actin at high concentrations of KCl (100-200 mM KCl), the fragments could only interact with myosin

filaments at low concentration of KCl (such as 20 mM KCl). It is likely that the packing of myosin into filaments reduces the number of binding sites available on myosin;

therefore, a lower concentration of KCl is required for the electrostatic interactions to occur. These results also suggest that meaningful interaction of titin with myosin

filaments in muscle cells requires large numbers of domains from motifs I and II in titin

molecules. Interactions of T2, the breakdown product of titin, with myosin and actin

filaments also have been reported (Kimura et al., 1984a; Maruyama et al., 1987). These

authors reported that the interactions occurred at low ionic strengths, in agreement with

the results described herein.

Although the current studies have shown interactions of myosin and actin with cloned titin fragments from the A-I junction portion of titin, it is not known whether the

cloned titin fragments could also interact with AMP-deaminase or C-protein (MyBP-C).

AMP-deaminase is located at or close to the A-I junction region (Cooper and Trinick, 127

1984) and its interaction with purified titin have been reported (Soteriou et al., 1993). In addition, interaction of cloned titin fragments from the A-band portion of titin with C- protein was also reported (Labeit et al., 1992). In the future, similar solid phase binding assays could be used to determine interaction between the A-I junction titin fragments with

AMP-deaminase and C-protein (MyBP-C).

The results presented in both papers have provided information on the primary structure of titin as well as the importance of each structure, namely motif I and motif H.

From the information provided, we now know that the A-I junction region of titin is also make up of motifs I and II, like the A-band portion of the titin molecule. However, sequence variations, such as changes in amino acid residues and the introduction of potential phosphorylation site(s), in the A-I junction sequence were found and these variations could provide functional flexibility to titin molecule as it encounters different proteins along each half sarcomere. For example, the titin portion at the A-I junction can interact with both myosin and actin. In contrast, the central A-band portion of titin can reportedly only interact with myosin, not actin. The interaction studies also provided information on the important factors that contribute to the interaction of titin fragments with myosin or actin. Among the contributing factors are the size of titin fragments and type of motif that make up the fragments as well as the number of binding sites from myosin. In addition, results from studies on the effect of KCl concentration on the interaction of titin fragments with myosin or myosin filaments suggest that in a muscle cell, interaction between titin and myosin filaments requires multiple interacting domains 128

(motifs I and II) from the titin molecules. Thus, both papers reported in this dissertation have contributed significant new knowledge about the structure and function of titin. 129

LITERATURE CITED

Ayme-Southgate, A., J. Vigoreaux, G. Benian, and M. L. Pardue. 1991. Drosophila has a twitchin/titin related gene that appears to encode projectin. Proc. Natl. Acad. Sci. USA 88:7973-7977.

Bahler, M., H. M. Eppenberger, and T. Wallimann. 1985a. Novel thick filament protein of chicken pectoralis muscle: the 86K protein. I. Purification and characterization. J. Mol. Biol. 186:381-391.

Bahler, M., H, M. Eppenberger, and T. Wallimann. 1985b. Novel thick filament protein of chicken pectoralis muscle: the 86K protein. II. Distribution and localization. J. Mol. Biol. 186:393-401.

Benian, G. M., J. E. Kiff, N. Neckelmann, D. G. Moerman, and R. H. Waterston, 1989. Sequence of an unusually large protein implicated in regulation of myosin activity in C. elegans. Nature 342:45-50.

Benian, G. M., S. W. L'HernauIt, and M. E. Morris. 1993. Additional sequence complexity in muscle gene, unc-22, and its encoded protein, twitchin, of Caenorhabditis elegans. Genetics 134:1097-1104.

Colley, N. J., K. T. Tokuyasu, and S. J. Singer. 1990, The early expression of myofibrillar proteins in round postmitotic myoblasts of embryonic skeletal muscle. J. Cell Sci. 95:11-22.

Cooper, J., and J. Trinick. 1984. Binding and location of AMP deaminase in rabbit psoas muscle myofibrils. J. Mol. Biol. 177:137-152.

Craig, R., and G., Offer. 1976. The location of C-protein in rabbit skeletal muscle. Proc. R. Soc. Lond. B 192:451-461.

Ebashi, S., M. Endo, and I. Ohtsuki. 1969. Control of muscle contraction. Q. Rev. Biophys. 2:351-384.

Edelman, G. M. 1987. CAM and Igs: Adhesion and the evolutionary origins of immunity. Immunol. Rev. 100:11-45.

Einheber, S., and D. A. Fischman. 1990. Isolation and characterization of a cDNA clone encoding avian skeletal muscle C-protein: An intracellular member of the immunoglobulin superfamily. Proc. Natl. Acad. Sci. USA 87:2157-2161. 130

Elzinga, M,, J. H. Collins, W. M. Kuehl, and R. S. Adelstein. 1973, Complete amino- acid sequence of actin of rabbit skeletal muscle. Proc. Natl. Acad. Sci. USA 70:2687-2691.

Epstein, H. P., and D. A. Fischman. 1991. Molecular analysis of protein assembly in muscle development. Science 251:1039-1044.

Fulton, A. B., and W. B. Isaacs. 1991. Titin, a huge, elastic sarcomeric protein witii a probable role in morphogenesis. BioEssays 13:157-161.

Funatsu, T., H. Higuchi, and S. Ishiwata. 1990. Elastic filaments in skeletal muscle revealed by selective removal of thin filaments with plasma gelsolin. J. Cell Biol. 110:53-62.

Funatsu, T., E. Kono, H. Higuchi, S. Kimura, S. Ishiwata, T. Yoshioka, K. Maruyama, and S. Tsukita. 1993. Elastic filaments in situ in cardiac muscle: Deep-etch replica analysis in combination with selective removal of actin and myosin filaments. J. Cell Biol. 120:711-724.

Furst, D. O., M. Osborn, R. Nave, and K. Weber. 1988. The organization of titin filaments in the half-sacromere revealed by monoclonal antibodies in immunoelectron microscopy: A map of ten nonrepetitive epitopes starting at the Z line extends close to the M line. J. Cell Biol. 106:1563-1572.

Furst, D. O., M. Osborn, and K. Weber. 1989a. Myogenesis in the mouse embryo: Differential onset of expression of myogenic proteins and the involvement of titin in myofibril assembly. J. Cell Biol. 109:517-527.

Furst, D. O., R. Nave, M. Osborn, and K. Weber. 1989b. Repetitive titin epitopes with a 42 nm spacing coincide in relative position with known A band striations also identified by major myosin-associated proteins. J. Cell Sci. 94:119-125.

Furst, D. O., U. Vinkemeier, and K. Weber. 1992. Mammalian skeletal muscle C- protein: Purification from bovine muscle, binding to titin and the characterization of a full-length human cDNA. J. Cell Sci. 102:769-778.

Fyrberg, C. C., S. Labeit, B. Bullard, K. Leonard, and E. Fyrberg. 1992. Drosophila projectin: relatedness to titin and twitchin and correlation with lethal(4)102 CDa and bent-Dominant. Proc. R. Soc. Lond. B 249:33-40.

Gautel, M., K. Leonard, and S. Labeit. 1993. Phosphorylation of KSP motifs in the C- terminal region of titin in differentiating myoblasts. EMBO J. 12:3827-3834. 131

Goll, D. E, A. Suzuki, J. Temple, and G. R. Holmes. 1972. Studies on purified a- actinin. I. Effect of temperature and tropomyosin on the a-actinin/F-actin interaction. J. Mol. Biol. 67:469-488.

Goll, D. E., W. R. Dayton, I. Singh, and R. M. Robson. 1991. Studies of the a- actinin/actin interaction in the Z-disk by using calpain. J. Biol. Chem. 266:8501- 8510.

Grove, B. K., V. Kurer, C. Lehner, T. C. Doetschman, J.-C. Perriard, and H. M. Eppenberger. 1984. A new 185,000-dalton skeletal muscle protein detected by monoclonal antibodies. J. Cell Biol. 98:518-524.

Handel, S. E., S.-M. Wang, M. L. Greaser, and E. Schultz. 1989. Skeletal myofibrillogenesis as revealed with a monoclonal antibody to titin in combination with detection of the a- and y-isoforms of actin. Dev. Biol. 132:35-44.

Handel, S. E., M. L. Greaser, E. Schultz, S.-M. Wang, J. C. Bulinski, J. J.-C. Lin, and J. L. Lessard. 1991. Chicken cardiac myofibrillogenesi studied with antibodies specific for titin and the muscle and nonmuscle isoforms of actin and tropomyosin. Cell Tiss. Res. 263:419-430.

Hanson, E. J., and J. Lowy. 1963. The structure of F-actin and of actin filaments isolated from muscle. J. Mol. Biol. 6:46-60.

Harrelson, A. L., and C. S. Goodman. 1988. Growth cone guidance in insects: Fasciclin II is a member of the immunoglobulin superfamily. Science 242:700- 708.

Harrington, W. F. 1979. Contractile proteins of muscle. In The proteins IV., pp 245- 409, Academic Press, New York and London.

Higuchi, H., T. Suzuki, S. Kimura, T. Yoshioka, K. Maruyama, and Y. Umazume. 1992. Localization and elasticity of connectin (titin) filaments in skinned frog muscle fibres subjected to partial depolymerization of thick filaments. J. Muscle Res. Cell Motil. 13:285-294.

Higuchi, H. 1992. Changes in contractile properties with selective digestion of connectin (titin) in skinned fibers of frog skeletal muscle. J. Biochem. Ill: 291-295.

Hill, C., and K. Weber. 1986. Monoclonal antibodies distinguish titins from heart and skeletal muscle. J. Cell Biol. 102:1099-1108. 132

Hill, C. S., S. Duran, Z. Lin, K. Weber, and H. Holtzer. 1986. Titin and myosin, but not desmin, are linked during myofibrillogenesis in postmiotic mononucleated myoblasts. J. Cell Biol. 103:2185-2196.

Horowits, R. 1992. Passive force generation and titin isoforms in mammalian skeletal muscle. Biophys. J. 61:392-398.

Horowits, R., E. S. Kempner, M. E. Bisher, and R. J. Podolsky. 1986. A physiological role for titin and nebulin in skeletal muscle. Nature 323:160-164.

Horowits R., and R. J. Podolsky. 1987. The positional stability of thick filaments in activated skeletal muscle depends on sarcomere length: Evidence for the role of titin filaments. J. Cell Biol. 105:2217-2223.

Horowits, R., K. Maruyama, and R. J. Podolsky. 1989. Elastic behavior of connectin filaments during thick filament movement in activated skeletal muscle. J. Cell Biol. 109:2169-2176.

Hu, D. H., S. Kimura, and K. Maruyama. 1989. Myosin oligomers as the molecular mass standard in the estimation of molecular mass of nebulin (~800 kD) by SDS- polyacrylamide gel electrophoresis. Biomed. Res. 10:165-168.

Hu, D. H., A. Matsuno, K. Terakado, T. Matsuura, S. Kimura., and K. Maruyama. 1990. Projectin is an invertebrate connectin (titin): Isolation from crayfish claw muscle and localization in crayfish claw muscle and insect flight muscle. J. Muscle Res. Cell Motil. 11:497-511.

Huiatt, T. W., M. A. Kurpakus, R. M. Robson. 1989. Immunofluorescence localization studies on the organization of titin and nebulin during myofibril assembly in embryonic chick skeletal muscle cell cultures. J. Anim. Sci. 67 (Suppl.2):99.

Huxley, H. E. 1963. Electron microscope studies on the structure of natural and synthetic protein filaments from striated muscle. J. Mol. Biol. 7:281-308.

Huxley, H., and J. Hanson. 1954. Changes in the cross-striations of muscle during contraction and stretch and their structural interpretation. Nature 173:973-976.

Huxley, A. F., and R. Niedergerke. 1954. Structural changes in muscle during contraction. Nature 173:971-973.

Huxley, A. P., and L. D. Peachey. 1961. The maximum length for contraction in vertebrate striated muscle. J, Physiol. (Lond.) 156:150-165. 133

Isaacs, W. B., I. S. Kim, A. Struve, and A, B. Fulton. 1989. Biosynthesis of titin in cultured skeletal muscle cells. J. Cell Biol. 109:2189-2195.

Isaacs, W. B., I. S. Kim, A. Struve, and A. B. Fulton. 1992. Association of titin and myosin heavy chain in developing skeletal muscle. Proc. Natl. Acad. Sci. USA 89:7496-7500.

Itoh, Y., S. Kimura, T. Suzuki, K. Ohashi, and K. Maniyama. 1986. Native connectin from porcine cardiac muscle. J. Biochem. 100:439-447.

Itoh, Y., T. Suzuki, S. Kimura, K. Ohashi, H. Higuchi, H. Sawada, T. Shimizu, M. Shibata, and K. Maniyama. 1988. Extensible and less extensible domains of connectin filaments in stretched vertebrate skeletal muscle sarcomeres as detected by immunofluorescence and immunoelectron microscopy using monoclonal antibodies. J. Biochem. 104:504-508.

Jin, J-P, and K. Wang. 1991a. Nebulin as a giant actin-binding template protein in skeletal muscle sarcomere: Interaction of actin and cloned human nebulin fragments. FEBS Lett. 281:93-96.

Jin, J-P., and K. Wang. 1991b. Cloning, expression and protein interaction of human nebulin fragments composed of varying numbers of sequence modules. J. Biol. Chem. 266:21215-21223.

Kimura, S., K. Maruyama, and Y. P. Huang. 1984a. Interactions of muscle P-connectin with myosin, actin, and actomyosin at low ionic strengths. J. Biochem. 96:499- 506.

Kimura, S., H. Yoshidomi, and K. Maruyama. 1984b. Proteolytic fragments of connectin cause aggregation of myosin filaments but not of actin filaments. J. Biochem. 96:1947-1950.

Kimura, S., T. Matsuura, S. Ohtsuka, Y. Nakauchi, A. Matsuno, and K. Maruyama. 1992. Characterization and localization of a-connectin (titin 1): An elastic protein isolated from rabbit skeletal muscle. J. Muscle Res. Cell Motil. 13:39-47.

Komiyama, M., K. Maruyama, and Y. Shimada. 1990. Assembly of connectin (titin) in relation to myosin and a-actinin in cultured cardiac myocytes. J. Muscle Res. Cell Motil. 11:419-428.

Kruger, M., J, Wright, and K. Wang. 1991, Nebulin as a length regulator of thin filaments of vertebrate skeletal muscles: Correlation of thin filament lengths, nebulin size, and epitope profile. J. Cell Biol. 115:97-107. 134

Kurzban, G. P., and K. Wang. 1988. Giant polypeptides of skeletal muscle titin: Sedimentation equilibrium in guanidine hydrochloride. Biochem. Biophys. Res. Commun. 150:1155-1161.

Labeit, S., D. P. Barlow, M. Gautel, T. Gibson, J. Holt, C-L. Hsieh, U. Francke, K. Leonard, J. Wardale, A. Whiting, and J. A. Trinick. 1990. Regular pattern of two types of 100-residue motif in the sequence of titin. Nature 345:273-276.

Labeit, S., T. Gibson, A. Lakey, K. Leonard, M. 2kviani, P. Knight, J. Wardale, and J. Trinick. 1991. Evidence that nebulin is a protein-ruler in muscle thin filaments. FEBS Lett. 282:313-316.

Labeit, S., M. Gautel, A. Lakey, and J. Trinick. 1992. Towards a molecular understanding of titin. EMBO J. 11:1711-1716.

Lakey, A., C. Ferguson, S. Labeit, M. Reedy, A. Larkin, G. Butcher, K. Leonard, and B. Bullard. 1990. Identification and localization of high molecular weight proteins in insert flight and leg muscle. EMBO J. 9:3459-3467.

Lakey, A., S. Labeit, M. Gautel, C. Ferguson, D. P. Barlow, K. Leonard, and B. BuUard. 1993. Kettin, a large modular protein in the Z-disc of insect muscles. EMBO J. 12:2863-2871.

LaSalle, F., R. M. Robson, M. L. Lusby, F. C. Parrish, M. H. Stromer, and T. W. Huiatt. 1983. Localization of titin in bovine skeletal muscle by immunofluorescence and immunoelectron microscope labeling. J. Cell Biol. 97: 258a.

Locker, R. H,, and N. G. Leet. 1975. Histology of highly-stretched beef muscle. I. The fine structure of grossly stretched single fibers. J. Ultrastruct. Res. 52:64-75.

Locker, R. H., and N. G. Leet. 1976. Histology of highly-stretched beef muscle. II. Further evidence on the location and nature of gap filament. J. Ultrastruct. Res. 55:157-172.

Lowey, S. 1979. Thick filament proteins in muscle. In Fibrous proteins: Scientific, Industrial, and Medical Aspects., pp 1-25, Academic Press, New York.

Maher, P. A., G. F. Cox, and S. J. Singer. 1985. Zeugmatin: A new high molecular weight protein associated with Z lines in adult and early embryonic striated muscle. J. Cell Biol. 101:1871-1883. 135

Maroto, M., J. Vinos, R. Marco, and M. Cervera. 1992. Authophosphorylating protein kinase activity in titin-like arthropod projectin. J. Mol. Biol. 224:287-291.

Maruyama, K. 1986. Connectin, an elastic filamentous protein from striated muscle. Int. Rev. Cytol. 104:81-114.

Maruyama, K., R. Natori, and Y. Nomomura. 1976, New elastic protein from muscle. Nature 262:58-60.

Maruyama, K., S. Kimura, K. Ohashi, and Y. Kawano. 1981. Connectin, an elastic protein of muscle. Identification of "titin" with connectin. J. Biochem. 89, 701- 709.

Maruyama K,, H. Sawada, S. Kimura, K. Ohashi, H. Higuchi, and Y. Umazume. 1984a. Connectin filaments in stretched skinned fibers of frog skeletal muscle. J. Cell Biol. 99:1391-1397.

Maruyama, K., S. Kimura., H. Yoshidomi, H. Sawada, and M. Kikuchi. 1984b. Molecular size and shape of P-connectin, an elastic protein of striated muscle. J. Biochem. 95:1423-1493.

Maruyama, K., T. Yoshioka, H, Higuchi, K. Ohashi, S. Kimura, and R. Natori. 1985. Connectin filaments link thick filaments and Z-lines in frog skeletal muscle as revealed by immunoelectron microscopy. J, Cell Biol. 101:2167-2172.

Maruyama, K., Y. Itoh, and F. Arisakam. 1986. Circular dichroism spectra show abundance of P-sheet structure in connectin, a muscle elastic protein. FEBS Lett. 202:353-355.

Maruyama, K., D. H. Hu, T. Suzuki, and S. Kimura. 1987. Binding of actin filaments to connectin filaments. J. Biochem. 101:1339-1346.

Maruyama, K., A. Matsuno, H. Higuchi, S. Shimaoka, S. Kimura, and T. Shimizu. 1989. Behaviour of connectin (titin) and nebulin in skinnned muscle fibres released after extreme stretch as revealed by immunoelectron microscopy. J. Muscle Res. Cell Motil, 10:350-359.

Maruyama, K,, T. Endo, H, Kume, Y. Kawamura, N. Kanzawa, Y. Nakauchi, S. Kimura, S. Kawashima, and K. Maruyama. 1993. A novel domain sequence of connectin localized at the I band of skeletal muscle sarcomeres: Homology to neurofilament subunits. Biochem. Biophys. Res. Commun. 194:1288-1291.

Masaki, T., and O. Takaiti. 1974. M-protein. J. Biochem. 75:367-380. 136

Matsuno, A., H. Takano-Ohmuro, Y. Itoh, T. Matsuura, M. Shibata, H. Nakae, T. Kaminuma, and K. Maruyama. 1989. Anti-connectin monoclonal antibodies that react with the unc-22 gene product bind dense bodies of caenorhabditis (nematode) bodywall muscle cells. Tissue Cell 21:495-505.

Matsuura, T., S. Kimura, S. Ohtsuka, and K. Maruyama. 1991. Isolation and characterization of 1200 KDa peptide of a-connectin. J. Biochem. 110:474-478.

Moore, P. B., H. E. Huxley, and D, J. DeRosier. 1970. Three-dimensional reconstruction of F-actin, thin filaments and decorated thin filaments. J. Mol. Biol. 50:279-292.

Nave, R., D. O. Furst, and K. Weber. 1989. Visualization of the polarity of isolated titin molecules: A single globular head on a long thin rod as the M-band anchoring domain? J. Cell Biol. 109:2177-2187.

Nave, R., and K. Weber. 1990. A myofibrillar protein of insect muscle related to vertebrate titin connects Z-band and A-band: Purification and molecular characterization of invertebrate mini-titin. J. Cell Sci. 95:535-544.

Nave, R., D. Furst, U. Vinkemeier, and K. Weber. 1991. Purification and physical properties of nematode mini-titins and their relation to twitchin. J. Cell. Sci. 98:491-496.

Noguchi, J., M. Yanagisawa, M. Imamura, Y. Kasuya, T. Sakurai, T. Tanaka, and T. Masaki. 1992. Complete primary structure and tissue expression of chicken pectoralis M-protein. J. Biol. Chem. 267:20302-20310.

Offer, G., C. Moos, and R. Starr. 1973. A new protein of the thick filaments of vertebrate skeletal myofibrils. Extraction, purification, and characterization. J. Mol. Biol. 74:653-676.

Olson, N. J., R. B. Pearson, D. S. Neadleman, M. Y. Hurwitz, B. E. Kemp, and A. R, Means. 1990. Regulatory and structural motifs of chicken gizzard myosin light chain kinase. Proc. Natl. Acad. Sci. USA 87:2284-2288.

Parry, D. A. D., and J. M. Squire. 1973. Structural role of tropomyosin in muscle regulation: Analysis of the X-ray diffraction patterns from relaxed and contracting muscles. J. Mol. Biol. 75:33-55.

Pearson, R. B., and B. E. Kemp. 1991. Protein kinase phosphorylation site sequences and consensus specificity motifs: tabulation. Methods Enzymol. 200:62-81. 137

Pierobon-Bormioli, S., D. Biral, R. Betto, and G. Salviati. 1992. Immunoelectron microscopic epitope locations of titin in rabbit heart muscle. J. Muscle Res. Cell Motil. 13:35-38.

Price, M. G. 1987. Skelemins: Cytoskeletal proteins located at the periphery of M-discs in mammalian striated muscle, J. Cell Biol. 104:1325-1336.

Price, M. G., and R. H. Gomer. 1993. Skelemin, a cytoskeletal M-disc periphery protein, contains motifs of adhesion/recognition and intermediate filament proteins. J. Biol. Chem. 268:21800-21810.

Robinson, T. P., and S. Winegard. 1979. The measurement and dynamic implications of thin filaments length in heart muscle. J. Physiol. (Lond) 286:607-619.

Schaart, G., C. Viebahn, W. Langmann, and F. Ramaekers. 1989. Desmin and titin expression in early postimplantation mouse embryos. Development 107:585-596.

Schultheiss T., Z. Lin, M-H, Lu, J. Murray, D. A. Fischman, K. Weber, T. Masaki, M. Imamura, and H. Holtzer. 1990. Differential distribution of subsets of myofibrillar proteins in cardiac nonstriated and striated myofibrils. J. Cell Biol. 110:1159-1172.

Shoemaker, M. O., W. Lau, R. L. Shattuck, A. P. Kwiatkowski, P. E. Matrisian, L. Guerra-Santos, E, Wilson, T. J. Lukas, L. J. Van Eldik, and D. M. Watterson. 1990. Use of DNA sequence and mutant analyses and antisense oligodeoxynucleotides to examine the molecular basis of non-muscle myosin light chain kinase autoinhibition, calmodulin recognition, and activity. J. Cell Biol. 111:1107-1125.

Sjostrand, F. S. 1962. The connections between A- and I- band filaments in striated frog muscle. J. Ultrastruct. Res. 7:225-246.

Somerville, L. L., and K. Wang. 1987. In vivo phosphorylation of titin and nebulun in frog skeletal muscle. Biochem. Biophys. Res. Commun. 147:986-992.

Somerville, L. L., and K. Wang. 1988. Sarcomere matrix of striated muscle: In vivo phosphorylation of titin and nebulin in mouse diaphragm muscle. Arch. Biochem. Biophys. 262:118-129.

Soteriou, A., M. Gamage, and J. Trinick. 1993. A survey of interactions made by the giant protein titin. J. Cell Sci. 104:119-123. 138

Squire, J. M. 1986. In Muscle: Design, Diversity, and Disease., pp 125-129, Benjamin/Cummings Publishing Co., Menlo Park, CA.

Stedman H., K. Browning, N. Oliver, M. Oronzi-Scott, K. Fischbeck, S. Sarkar, J. Sylvester, R. Schmickel, and K. Wang. 1988. Nebulin cDNAs detect a 25- kilobase transcript in skeletal muscle and localize to human chromosome 2. Genomics 2:1-7.

Stewart, M., and A. D. McLachlan. 1975. Fourteen actin binding sites on tropomyosin? Nature 257:331-333.

Stone, D. et al. 1974. Tropomyosin: Correlation of amino acid sequence and structure. FEES Lett. 31:125-136.

Suzuki, A., D. E. Goll, I. Singh, R. E. Allen, R. M. Robson, and M. H. Stromer. 1976. Some properties of purified skeletal muscle a-actinin. J, Biol. Chem. 251:6860-6870.

Takahashi, K., A. Hattori, R. Tatsumi, and K. Takai. 1992. Calcium-induced splitting of connectin filaments into P-connectin and a 1,200-kDa subfragment. J. Biochem. 111:778-782.

Takano-Ohmuro, H., Y. Nakauchi, S. Kimura, and K. Maruyama. 1992. Authophosphorylation of P-connectin (titin 2) in vitro. Biochem. Biophys. Res. Commun. 183:31-35.

Tan, K. O., G. R. Sater, A. M, Myers, R. M. Robson, and T. W. Huiatt. 1993. Molecular characterization of avian muscle titin. J. Biol. Chem. 268:22900- 22907.

Trinick, J. 1991. Elastic filaments and giant proteins in muscle. Curr. Opinion Cell Biol. 3:112-119.

Trinick, J. 1992. Understanding the functions of titin and nebulin. FEES Lett. 307:44- 48.

Trinick, J. A. 1981. End-filaments: A new structural element of vertebrate skeletal muscle thick filaments. J. Mol. Biol. 151:309-314.

Trinick, J., P. Knight, and A. Whiting. 1984. Purification and properties of native titin. J. Mol. Biol. 180:331-356. 139

Trombitas, K., G. H. Pollack, J. Wright, and K. Wang. 1990. Elastic behavior and arrangement of titin filaments in the I-band. Proc. XII Intern. Congr. Electron Microscopy., pp 478-479, San Francisco Press, Inc. San Francisco, CA.

Trombitas, K., P. H. W. W. Baatsen, M. S. Z. Kellermayer, and G. H. Pollack. 1991. Nature and origin of gap filaments in striated muscle. J. Cell Sci. 100:809-814.

Trombitas, K., and G. H. Pollack. 1993. Elastic properties of the titin filament in the Z- line region of vertebrate striated muscle. J. Muscle Res. Cell Motil. 14:416-422.

Uchida, K., I. Harada, Y. Nakauchi, and K. Maruyama. 1991. Structural properties of connectin studied by ultraviolet resonance raman spectroscopy and infrared dichroism. FEES Lett. 295:35-38.

Vaughan, K. T., F. E. Weber, S. Einheber, and D. A. Fischman. 1993a. Molecular cloning of chicken myosin-binding protein (MyBP) H (86-kDa protein) reveals extensive homology with MyBP-C (C-protein) with conserved immunoglobulin C2

and Fibronectin type in motifs. J. Biol. Chem. 268:3670-3676

Vaughan, K. T., F. E. Weber, F. C. Reinach, T. Ried, D. Ward, and D. A. Fischman. 1993b. Human myosin binding protein H (MyBP-H): Complete primary sequence, repeat structure, genomic organization and chromosomal localization. Genomics 16:34-40.

Vinkemeier, U., W. Obermann, K. Weber, and D. O. Furst. 1993. The globular head domain of titin extends into the center of the sarcomeric M band: cDNA cloning, epitope mapping and immunoelectron microscopy of two titin-associated proteins. J. Cell Sci. 106:319-330.

Wang, K. 1981. Nebulin, a giant protein component of Nj-line of striated muscle. J. Cell Biol. 91:355a.

Wang, K. 1982. Purification of titin and nebulin. Methods Enzymol. 85:264-274,

Wang , K. 1985. Sarcomere-associated cytoskeletal lattices in striated muscle. Review and hypothesis. In Cell and Muscle Motility., vol.6, pp 315-369, (J. W. Shay, ed.) Plenum Publishing Corp., New York.

Wang, K., and C. L. Williamson. 1980. Identification of an N2-line protein of striated muscle. Proc. Natl. Acad. Sci. USA 77:3254-3258. 140

Wang, K., and J. Wright. 1988. Architecture of the sarcomere matrix of skeletal muscle: Immunoelectron microscopic evidence that suggest a set of parallel inextensible nebulin filaments anchored at the Z-line. J. Cell Biol. 107:2199-2212.

Wang, K., J. McClure, and A. Tu. 1979. Titin: Major myofibrillar components of striated muscle. Proc. Natl. Acad. Sci. USA 76:3698-3702.

Wang, K., R. Ramirez-Mitchell, and D. Palter. 1984. Titin is an extraordinarily long, flexible, and slender myofibrillar protein Proc. Natl. Acad. Sci. USA 81:3685- 3689.

Wang, K., M. Knipper, Q.Q. Huang, L. Hsu, A. Van Heerden, K. Browning, X. Quian, and H. J. Stedman. 1990. Structural motifs of nebulin as predicted by sequences of human nebulin cDNA. J. Cell Biol. 111:428a.

Wang, K., R. McCarter, J. Wright, J. Beverly, and R. Ramirez-Mitchell. 1991. Regulation of skeletal muscle stiffness and elasticity by titin isoforms: A test of the segmental extension model of resting tension. Proc. Natl. Acad. Sci. USA 88:7101-7105.

Weber, F. E., K. T. Vaughan, F. C. Reinach, and D. A. Fischman. 1993. Complete sequence of human fast-type and slow-type muscle myosin-binding protein C (MyBP-C): Differential expression, conserved domain structure and chromosome assignment. Eur. J. Biochem. 216:661-669.

Wegmann, G., E. Zanolla, H. M. Eppenberger, and T. WalUmann. 1992. In situ compartmentation of creatine Idnase in intact sarcomeric muscle: The acto-myosin overlap zone as a molecular sieve, J. Muscle Res. Cell Motil. 13: 420-435.

Whiting, A., J. Wardale, and J. Trinick. 1989. Does titin regulate the length of muscle thick filaments? J. Mol. Biol. 205:263-268.

Yamaguchi, M., R. M. Robson, M. H. Stromer, N. R. Cholvin, and M. Izumimoto. 1983. Properties of soleus muscle Z-lines and induced Z-line analogs revealed by dissection with Ca^'''-activated neutral protease. Anat. Rec. 206:345-362.

Yamaguchi, M., M, Izumimoto, R. M. Robson, and M. H. Stromer. 1985. Fine structure of wide and narrow vertebrate muscle Z-lines. A proposed model and computer simulation of Z-line architecture. J. Mol. Biol. 184:621-643. 141

ACKNOWLEDGMENTS

I would like to express a special thanks to Dr. Alan M. Myers for his helpful advice and suggestions, and opportunity to learn helpful research techniques from his lab.

I would also like to thank other members of my committee, Drs. Richard M. Robson,

John E. May Geld, and Bonita A. Glatz for expert advice and discussion. I would like to extend my appreciation to those who worked in the lab for their encouragement and help, especially Jo Philips for her expert technical assistance in production of antibodies and purification of proteins, John Fuller for help on computer work, and Janet Stephenson for her knowledge on muscle cell culture.

I would like to express my gratitude to my major professor. Dr. Ted W. Huiatt, for the opportunity to work on very interesting projects. My sincere thanks for his expert advice, comments, and guidance throughout my graduate study. I would also like to express my appreciation for the many hours he spent on helping me complete this work and preparing the first manuscript for submission to The Journal of Biological Chemistry.

Finally, my deepest appreciation goes to my wife, Fang, and daughter, Emily, for their support, patience, and understanding. My appreciation also goes to my parents and grandparents for enduring my absence during the course of my graduate study.