Suppression of Muscle Hypercontraction by Mutations in the Myosin Heavy Chain Gene of Drosophila melanogaster

Corresponding author: John C. Sparrow, University of York, York YO10 5YW, United Kingdom., jcs1{at}york.ac.uk (E-mail)

Communicating editor: T. SCHÜPBACH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The indirect flight muscles (IFM) of Drosophila melanogaster provide a good genetic system with which to investigate muscle function. Flight muscle contraction is regulated by both stretch and Ca²⁺-induced thin filament (actin + tropomyosin + troponin complex) activation. Some mutants in troponin-I (TnI) and troponin-T (TnT) genes cause a "hypercontraction" muscle phenotype, suggesting that this condition arises from defects in Ca²⁺ regulation and actomyosin-generated tension. We have tested the hypothesis that missense mutations of the myosin heavy chain gene, Mhc, which suppress the hypercontraction of the TnI mutant held-up² (hdp²), do so by reducing actomyosin force production. Here we show that a "headless" Mhc transgenic fly construct that reduces the myosin head concentration in the muscle thick filaments acts as a dose-dependent suppressor of hypercontracting alleles of TnI, TnT, Mhc, and flightin genes. The data suggest that most, if not all, mutants causing hypercontraction require actomyosin-produced forces to do so. Whether all Mhc suppressors act simply by reducing the force production of the thick filament is discussed with respect to current models of myosin function and thin filament activation by the binding of calcium to the troponin complex.

THE indirect flight muscles (IFM) of Drosophila melanogaster provide a powerful genetic system with which to understand muscle function, structure, and development. As flight is not required for survival under laboratory conditions, many mutants have been obtained in the genes for the major sarcomeric proteins by selection for flightlessness or for a "wings-up" phenotype (see BERNSTEIN et al. 1993 ; VIGOREAUX 2001 ). A number of Drosophila muscle protein gene mutations that give a dominant or recessive flightless phenotype produce an additional recessive phenotype in which IFM fibers undergo an auto-destructive contraction after the muscles have developed normally (KRONERT et al. 1995 ). This leads to separation and accumulation of fiber material to one or both attachment sites or to fiber bunching with detachment from both ends. Intracellularly, the phenotype is very variable but is characterized by disruption of the myofibrillar lattice as well as by bulging and shortening of individual sarcomeres. In some mutants (e.g., Mhc¹³ and fln⁰) the phenotype includes muscle protein proteolysis (KRONERT et al. 1995 ; REEDY et al. 2000 ).

As the term "hypercontraction" has been used in muscle pathology to describe similar types of muscle damage arising from excessive contraction in mutant Caenorhabditis elegans (KORSWAGEN et al. 1997 ; GARCIA-ANOVEROS et al. 1998 ), in reperfused rat hearts (DUNCAN 1987 ; BHATTI et al. 1989 ; MONTICELLO et al. 1996 ), in Duchenne muscular dystrophy (VALENTINE et al. 1989 ; TAY et al. 1992 ; COZZI et al. 2001 ), and in human muscle injuries (ROTH et al. 2000 ; FINOL et al. 2001 ), we have defined this Drosophila phenotype as "hypercontraction." We make the important distinction, implicit in the above, that the term is used only where the muscles develop normally, or nearly so, before the muscle damage occurs.

Extant mutants exhibiting the phenotype include held-up² [hdp²; an allele of the wings-upA gene, which encodes Troponin-I (TnI)], up¹⁰¹ [an allele of the upheld gene, which encodes Troponin-T (TnT)], some alleles of the IFM-specific actin gene, Act88F (AN and MOGAMI 1996 ), three alleles of the myosin heavy chain gene, Mhc⁶, Mhc¹³, and Mhc¹⁹ (KRONERT et al. 1995 ), and fln⁰, an allele of the flightin gene, fln (REEDY et al. 2000 ). The fact that mutant alleles of the TnI and TnT genes produce hypercontraction suggests that defects in Ca²⁺ regulation can produce this phenotype. The recovery of three missense mutants of the Mhc gene (KRONERT et al. 1995 ) and, more recently, of an allele of flightin (REEDY et al. 2000 ), all of which cause hypercontraction, suggests that a single explanation for hypercontraction is unlikely. The extant hypercontracting Mhc alleles are restricted to a small, five-amino-acid region of the light meromyosin domain. It is through this domain that myosin dimerizes by the assembly of -helical coiled-coil rods, which subsequently polymerize to form thick filaments. Flightin is likely a thick filament protein associated with the myosin rod domain (VIGOREAUX et al. 1993 ; REEDY et al. 2000 ). These latter mutants suggest that structural defects within the sarcomere can also lead to hypercontraction. Since actin is the major component of the thin filament, the Act88F alleles that cause hypercontraction could produce either defects in the troponin-tropomyosin (Tn-Tm) complex regulation or, like the two thick filament proteins, structural defects of the sarcomere.

hdp² is a point mutation (A116V) within exon 5 of the wupA gene (BEALL and FYRBERG 1991 ) and likely affects Ca²⁺ regulation. It corresponds to a highly conserved residue, alanine 25, of vertebrate skeletal muscle TnI, which is part of the N-terminal -helix that interacts with TnC (VASSYLYEV et al. 1998 ). To investigate the protein interactions occurring within the Tn-Tm complex during Ca²⁺ regulation, PRADO et al. 1995 recovered six suppressors of hdp² by their suppression of the wings-up phenotype. One suppressor is an intragenic mutation within the wupA gene (PRADO et al. 1995 ), another is a missense mutation within the Tm2 gene (NAIMI et al. 2001 ), and the remaining four are Mhc gene mutations within the myosin head (KRONERT et al. 1999 ).

The recovery of Mhc alleles as suppressors of a mutation affecting thin filament regulation might seem surprising, unless the forces produced by the interaction of myosin with actin are normally required to produce the hypercontraction phenotype. Support for this comes from experiments of BEALL and FYRBERG 1991 that removed all the IFM myosin from hdp² flies using an IFM-specific Mhc null mutation, Ifm(2)2 (now known as Mhc⁷). The hdp²/Y; Mhc⁷ muscles did not hypercontract but also myofibrils did not develop. The investigators proposed that actomyosin interactions exacerbate the structural or functional defect resulting from the troponin-I mutation. We further propose that it is the actomyosin force generation that is largely responsible for the hypercontraction either in response to aberrant regulation of contraction or when sarcomeric structure is compromised. This proposal further predicts that Mhc suppressors will be neither allele nor gene specific in their interactions with hypercontracting alleles in other genes. We have examined this proposal using genetic approaches to reduce the concentration of myosin heads in the sarcomere and by characterizing new Mhc missense suppressors of hdp².

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Fly strains:
All chromosome and gene symbols unless specifically mentioned are as described in FlyBase (http://flybase.bio.indiana.edu/). Canton-S was used as the control in all the experiments unless specified. The Y57 and Y97 transgenic lines express a myosin heavy chain polypeptide lacking the head (or motor) domain and were gifts from S. I. Bernstein and R. M. Cripps (CRIPPS et al. 1999 ). The Mhc suppressors of hdp², Mhc^D1, Mhc^D41, Mhc^D45, and Mhc^D62 were obtained from A. Ferrus. fln⁰ is described in REEDY et al. 2000 . All flies used for experiments are 2–4 days old unless otherwise indicated. All stocks and crosses were maintained at 25° on a yeast-sugar-agar medium.

Isolation, mapping, and sequencing of dominant Mhc suppressor mutations:
Ethyl methanesulfonate (EMS) mutagenesis and isolation of dominant suppressors of the hdp² wings-up phenotype were as described in PRADO et al. 1995 . Mutants mapping to the second chromosome were assigned as Mhc suppressors by noncomplementation with lethal Mhc alleles, including Df(2L)H20 (except Mhc^Su(2)D), and designated as Mhc^Su(2)X, where X is an allele identifier. They are referred to as Su(2)X mutations in the text. Oligonucleotide primers used for PCR and sequencing were as described in KRONERT et al. 1999 .

Hemithorax mounts for polarized light microscopy:
IFMs were prepared for polarized light microscopy as described by NONGTHOMBA and RAMACHANDRA 1999 . Briefly, fly thoraces were frozen in liquid nitrogen, longitudinally bisected with a razor blade, dehydrated through an alcohol series and then cleared in methyl salicylate and mounted in Canada balsam/DPX. Photographs were taken with a Leica microscope using polarized light optics.

Transmission electron microscopy:
Fly half thoraces were prepared following the protocol of KRONERT et al. 1995 and embedded in Epon E218. Sections stained with lead citrate and counterstained with uranyl acetate were examined using a JEOL 1200 EX transmission electron microscope.

Behavioral studies:
Measurements of adult flight, walking, larval crawling, and feeding behaviors were conducted as described in NAIMI et al. 2001 . The flight index is calculated as the percentage of flies that flew up or horizontally, rather than down or not at all.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The hypercontracted IFM phenotype:
Hypercontraction can take a number of forms. Table 1 summarizes the thick and thin filament protein mutants showing hypercontraction and the developmental stage at which this phenotype occurs. In many mutants hypercontraction ends with the muscles parted, or seemingly so, in the middle with the bulk of the fiber bunched at one or both attachment sites (Fig 1B). In other mutants the fibers separate from the attachment sites and bunch in the middle of the fiber (Fig 1C and Fig D). On the basis of these characteristics, only a handful of IFM mutants are classified as hypercontracted. In some mutants, Act88F^R28C and Act88F^E334Q (Fig 1D), the phenotype is less extreme in terms of either the degree of muscle shortening or the number of fibers showing the phenotype (Fig 1E), and we refer to this as partial hypercontraction (NAIMI et al. 2001 ).

View larger version (127K): In this window In a new window Download PPT slide   Figure 1. Polarized light micrographs of hypercontracting IFM. (A) IFM of wild-type fly. A star indicates one of the DLMs. (B) hdp2 showing IFM hypercontraction; fibers are bunched to the cuticle. Arrow indicates the bunched fibers at one end of the thorax. (C) DLMs of the Act88FE334Q fly just before the eclosion showing partially hypercontracted phenotype. Note the initiation of hypercontraction at the attachment site of the fibers (arrowhead). (D) Late Act88FE334Q showing the DLMs pulled to the center (arrows). (E) up101 showing the partially pulled DLMs (arrowhead). (F) hdp2/Y; Y97 partial suppression of the hdp2 phenotype with a copy of the myosin headless construct. (G) hdp2/Y; Mhc10/+ showing that a reduction in MHC partially suppresses the hdp2 phenotype. (H) hdp2/Y; Mhc10/+, Y97, a copy of Mhc10, and the headless myosin construct completely suppress the hdp2 hypercontraction (star). In all frames the anterior fly thorax is at the left corner, and the dorsal side is toward the top. All flies are 2–4 days old unless otherwise indicated and at the same magnification. Bar, 0.125 mm.

For Mhc⁶, Mhc¹³, Mhc¹⁹, and fln⁰, the hypercontraction phenotype develops progressively during the first 24 hr after adult eclosion (KRONERT et al. 1995 ; REEDY et al. 2000 ); for hdp², we have shown (NAIMI et al. 2001 ) that the IFM form normally up to 78 hr after puparium formation (APF), when twitching of the pupal legs is first observed and the IFM begin to shorten and break, a process that is complete at, or just before, eclosion.

Why is this hypercontraction phenotype produced at different developmental stages? Although the development and maturation of myofibrils continue for a few hours after eclosion, it is likely that functional myofibrils are formed by 75 hr APF (REEDY and BEALL 1993 ; REEDY et al. 2000 ). Thus the TnI mutation, hdp², causes a pupal development of the phenotype while the four mutants of two thick filament proteins, MHC and FLN, lead to the progressive adult phenotype. We have investigated whether this correlation extends to other hypercontracting muscle protein mutations. The up¹⁰¹ mutation shows normal development until 78 hr APF when, like hdp² (Table 1), the IFM begin to hypercontract, a process that is complete by eclosion with the exception of a few flies that still remain partially hypercontracted (Fig 1E; Table 2). Actin Act88F^R28C and Act88F^E334Q mutants show almost normal myofibril development until 78 hr APF with the initiation of fiber detachment from one end (Fig 1C), followed by fibers bunching to one end of the thorax (not shown) or in the middle (Fig 1D). Act88F^G268D develops "wavy" fibers but partially hypercontracts only after 78 hr APF, while in Act88F^R95C partial hypercontraction develops only after eclosion (Table 1).

Hypercontraction is suppressed by reducing the amount of functional myosin heads:
The appearance of hypercontraction in muscles that have developed normally suggests a priori that forces developed by actomyosin crossbridges cause the damage. We began by extending the experiment of BEALL and FYRBERG 1991 on the hypercontraction of hdp², but rather than removing all the myosin (which also removes the thick filament and myofibrillar lattice), we used two "headless" Mhc-expressing transgenic lines, Y57 (on the first chromosome) and Y97 (on the third chromosome; CRIPPS et al. 1999 ) to reduce the myosin head concentration in the thick filaments. The headless myosin molecules retain the regulatory light chain binding site and all domains C-terminal of this. They coassemble with endogenous full-length myosin in the thick filaments of IFM (CRIPPS et al. 1999 ).

In combination with the IFM-specific Mhc¹⁰ null allele, the Y97 construct acts as a dose-dependent suppressor of hdp² (Fig 1, F–H; Fig 2, E–J). In hdp² flies, although the muscle and myofibrils develop normally before 78 hr APF, by eclosion the fibers are completely pulled apart (Fig 1B). Hypercontraction completely disrupts the myofibrillar lattice, leaving fields of disordered thick and thin filaments (Fig 2D) in which misaligned Z-bands and M-lines are seen. Few sarcomeres are seen and these are short (see Table 3) and often have a bulging appearance (Fig 2C). The presence of one copy of Y97 partially suppresses the hdp² phenotype at gross fiber (Fig 1F) and sarcomeric (Fig 2E and Fig F) levels. A slight suppression of the hypercontraction phenotype of hdp² occurs due to hdp²; Mhc¹⁰/+ (Fig 1G and Fig 2G and Fig H). In Mhc¹⁰/+ flies the removal of one functional Mhc gene copy will lead to 60% of wild-type myosin accumulation (O'DONNELL and BERNSTEIN 1988 ; CRIPPS et al. 1994 ). Suppression of hypercontraction increases as the gene dose of Y97 increases and that of Mhc⁺ decreases until in hdp²/Y; Mhc¹⁰/+; Y97 flies suppression is almost complete (Table 2; Fig 1H). Myofibrillar organization is partially restored with sarcomere lengths approaching those of wild type (Fig 2I). A remaining difference is the increased gap between neighboring myofibrils (Fig 2J) compared to wild-type controls (Fig 2B). Apart from the wild-type controls (Canton-S), all of the genotypes are flightless. This is hardly surprising as both hdp² and a reduction in Mhc gene dosage cause flightlessness. The headless myosin cannot replace the power generation of wild-type MHC.

View larger version (89K): In this window In a new window Download PPT slide   Figure 2. Electron micrographs of hdp2 myofibrils suppressed by combinations of Mhc10 and the Y97 headless myosin transgenic insert. (A) Wild-type, longitudinal section (LS). (B) Wild-type, transverse section (TS). Myofibrils showing highly ordered myofibrillar lattices and borders. (C) hdp2 LS from newly eclosed flies. Sarcomere length is greatly reduced with streaming of Z- and M-bands (arrows). (D) hdp2 TS disruption of the myofibrillar borders and lattices (arrow); star indicates the region where thick and thin filament integrity is still preserved. (E and F) hdp2/Y; Y97 partial suppression of the hdp2 phenotype with a copy of the Y97 headless construct. Sarcomere structure is slightly improved. (G and H) hdp2/Y; Mhc10/+. Partial suppression of the hdp2 phenotype by reduction of MHC. Breaks within the sarcomeric lattice are evident (arrow) and sarcomere length is shorter than normal. (I and J) hdp2/Y; Mhc10/+, Y97. One copy of Mhc10 and one copy of the Y97 headless myosin construct completely suppress the hdp2 hypercontraction. Sarcomere length is nearly normal and there are minimal breaks and disruptions of the myofibrils; however, increased gaps filled with sarcoplasmic material separate the myofibrils (arrowheads). M, M-band. Z, Z-Band. Myo, myofibril. Bar, 1 µm for all the LS (all are of same magnification) and 0.5 µm for all the TS (all are at same magnification).

Overall genotypes that suppress the wings-up phenotype also suppress IFM hypercontraction, but in individual flies that is not invariably the case. Flies with hdp²/Y; Mhc¹⁰/+ or hdp²/Y; +/+; Y97 genotypes with partially suppressed hypercontraction show the full range of wing position from "wings up" to "wings held beside the abdomen" to "normal wing position" (Table 2), as do Mhc¹⁰/Mhc¹⁰ homozygotes, which never show hypercontraction. In general we find in these and other studies that the correlation between the wings-up and IFM hypercontraction phenotypes is poor.

Mhc suppressor mutations of hdp² hypercontraction:
Six dominant EMS-induced suppressor mutations of the hdp² wings-up phenotype were recovered on the second chromosome from a screen of 25,000 progeny. Five mapped to the region between black (48.5) and cinnabar (57.5), which includes the Mhc gene. Four suppressors, Su(2)A, Su(2)B, Su(2)C, and Su(2)F, are Mhc alleles by their failure to complement the recessive lethality of Mhc¹ and Df(2)H20. Mhc¹ is a null allele due to a 1-kb internal deletion (O'DONNELL and BERNSTEIN 1988 ); Df(2)H20 spans from 36A8-9 to 36F1 (STEWARD and NUSSLEIN-VOLHARD 1986 ), which includes the Mhc gene. Lethality of these four Mhc suppressors as homozygotes or in heterozygous combination with either Mhc¹ or Df(2)H20 occurs early in development. Some die shortly after egg hatch after some crawling or in late first/early second instar at the first larval molt (double sets of mouthparts are seen in most larvae so molting is incomplete). All four mutants are dominant suppressors, which fully suppress hdp² hypercontraction (Table 4) but not the flightless phenotype. Each is dominant flightless in the absence of the hdp² mutation. This suggests that each Mhc suppressor mutation severely affects myosin function. Since they affect embryonic muscles and effect suppression of IFM hypercontraction, they are likely in constitutive exons of the Mhc gene since embryonic and IFM cDNAs differ in all alternative exons with the exception of exon 3, which is used in some embryonic muscles (GEORGE et al. 1989 ; BERNSTEIN and MILLIGAN 1997 ; KRONERT et al. 1999 ; ZHANG and BERNSTEIN 2001 ). Genomic DNA from Mhc homozygous larvae was obtained by selecting yellow embryos/larvae from sib-mated y/y or y/Y; CyO, y⁺/Mhc^x flies (see KRONERT et al. 1999 ) and the constitutive and IFM-specific exons were sequenced following PCR. The Mhc-coding region of each mutant strain contained a single amino acid change encoded within a constitutive exon. The changes are given with the Drosophila codon numbers and the chicken skeletal muscle MHC homolog (in parentheses) to allow direct comparisons with Mhc mutants in other species.

Suppressors Su(2)A, Su(2)B, and Su(2)C were found to have the same point mutation (C T) causing amino acid 401 (404 in chicken skeletal muscle myosin) to change from proline to serine in constitutive exon 8. This encodes part of the actin-binding region of the myosin head (Fig 3). These mutants could have arisen from a single mutational event and we have assumed that they did so. Suppressor Su(2)F is also a missense mutation (G A) changing alanine at 462 (465) to threonine, a position close to the ATP-binding site (Fig 3).

View larger version (36K): In this window In a new window Download PPT slide   Figure 3. The locations of the new suppressor mutations (blue) and those previously reported (red) within the atomic structure (2MYS.PDB) of chicken myosin S1 (RAYMENT et al. 1993B ) depicted using Protein Explorer (http://www.umass.edu/microbio/chime/explorer). Suppressors Su(2)A, Su(2)B, and Su(2)C show the same amino acid change P401S (404 in chicken skeletal muscle myosin). Suppressor Su(2)F is the mutation A462T (465) in the same actin-binding loop. Su(2)D leads G413S (416). Mutations D1 (D625G) and D45 (A261T) are as described in KRONERT et al. 1999 and occur in the actin-binding loop and near the ATP-binding pocket, respectively. Exon 7 (orange) is alternatively spliced in D41 as a result of a 2-bp insertion. D62 causes an eight-amino-acid deletion near the actin-binding loop and is not shown, as the loop is not seen in this atomic structure (see KRONERT et al. 1999 for more details).

Su(2)D is homozygous viable in combination with hdp² and on its own. Su(2)D heterozygotes without hdp² can fly as well as wild type [flight index of 79.6 ± 2.8 (SD), n = 54; in comparison to wild type, 78 ± 6.7, n = 51]. Homozygous Su(2)D flies, though, fly (flight index 63.9 ± 3.2) significantly less well (at 5% level) than wild type. Su(2)D heterozygotes partially suppress the hdp² wing position and muscle phenotypes but do so completely in homozygotes (Table 4). Results from complementation analysis of Su(2)D and Mhc¹ are ambiguous because Mhc¹/+ flies are viable and show partial IFM hypercontraction (NONGTHOMBA and RAMACHANDRA 1999 ). Su(2)D maps to position 54.2 on chromosome 2, i.e., within the Mhc region. Sequencing of the constitutive and IFM-specific Mhc exons of homozygous flies following PCR showed a single base pair change (G A) leading to an amino acid change from glycine to serine in codon 413 (416), close to the Su(2)B suppressor mutant (Fig 3). Su(2)E is a very weak suppressor for both wing and muscle phenotypes of hdp², making it difficult to map. We have not included more detailed data on this mutation.

Abnormal myofibrillar structure in hdp² IFM suppressed by myosin suppressor mutations:
Su(2)B completely suppresses fiber hypercontraction as seen in polarized light, except for some thinning of the dorsal-longitudinal muscle (DLM) fibers (Fig 4A) in a few flies. Electron micrographs of hdp²/Y; Su(2)B/+ flies (Fig 4B and Fig C) show a complete recovery of wild-type myofibrillar structure, although the sarcomere length remains slightly, but significantly, shorter than that of wild type (Table 3). hdp² hypercontraction is completely suppressed by Su(2)F; sarcomere structure is comparable to wild type (Fig 4E) except that mean sarcomere length remains significantly shorter than that of wild type (Table 3) and a few days after eclosion muscle fibers become thin in many areas and contract (Fig 4D). In cross sections the periphery of the myofibrils show loosely packed thick and thin filaments (Fig 4F), quite similar to hdp² myofibrils when they start to hypercontract (Fig 2D).

View larger version (157K): In this window In a new window Download PPT slide   Figure 4. Suppression of hdp2 hypercontraction by new Mhc mutations. (A) Polarized light micrograph of 2-day-old hdp2/Y; Mhc2B/+. Fibers appear completely normal except for some damage (star) in one or two fibers of some of the flies. (B and C) Electron micrographs of hdp2/Y; Mhc2B/+ myofibrils. Sarcomere length appears slightly shorter than that of wild type and myofibrils (Myo) are packed very close to each other. (D) Polarized light micrograph of 10-day-old hdp2/Y; Mhc2F/+. Fibers (star) show breaks; 2- to 3-day-old flies show completely normal fibers (not shown). (E) LS of 2-day-old hdp2/Y; Mhc2F/+ showing completely normal myofibrils. Sarcomere lengths are quite normal; compare with Fig 2A. (F) TS of 10-day-old hdp2/Y; Mhc2F/+ showing the aged-related disruption of the myofibrils (Myo). Anterior-posterior axis of the thoraces runs from right to left. Bar, 0.125 mm for A and D; 1 µm for B and E; and 0.5 µm for C and F.

Su(2)D heterozygotes partially suppress hdp². Half of the flies still show a wings-up phenotype and >80% of them have a partially suppressed muscle phenotype (Table 4). The fibers are thin and hypercontract from the posterior ends of the thorax (Fig 5A). Sarcomere structure is improved compared to hdp² (Fig 5B), but sarcomere length is barely half that of wild type (Table 3) and at the periphery the myofibrillar lattice is perturbed, suggesting that hdp² suppression is only partial at this level (Fig 5C). Homozygous Su(2)D completely suppresses hdp² hypercontraction (Fig 5D; Table 4) but still fails to restore wild-type sarcomere length (Table 3); homozygous Su(2)D sarcomeres are significantly shorter than those of wild type. For 6–7 days after eclosion the muscle structure remains completely normal, but after this the central myofibrillar lattice becomes disordered (Fig 5E and Fig F), with characteristic Z-band streaming and gaps in the lattice. We have previously reported on a similar age-related progressive myopathy with hdp² and the Tm2 suppressor mutant, D53 (NAIMI et al. 2001 ). Electron micrographs of myofibrils from the Mhc suppressors as hetero- or homozygotes (in the case of 2D) without hdp² have a completely wild-type appearance (data not shown).

View larger version (158K): In this window In a new window Download PPT slide   Figure 5. Suppression of hdp2 hypercontraction by mutations in Mhc S1. (A) Polarized light micrograph of 2-day-old hdp2/Y; Mhc2D/+ flight muscles. Fibers are broken at many places, particularly in E and F of DLM fibers (star). (B and C) Electron micrographs of hdp2/Y; Mhc2D/+ myofibrils. Sarcomeres appear disrupted and shorter; the periphery of the myofibrils (Myo) is loosely packed with disrupted thick and thin filament lattices (arrowhead). (D) Polarized light micrograph of 10-day-old hdp2/Y; Mhc2D/Mhc2D showing completely normal fibers. (E and F) LS and TS of 10-day-old hdp2/Y; Mhc2D/Mhc2D showing age effects [2- to 3-day-old flies show completely normal myofibrils (not shown)]. Normal length sarcomeres show disruptions and gaps (arrows) that are clearly visible in the centers (arrows) of myofibrils (Myo), which is different from the hdp2/Y; Mhc2F/+ suppressor. Anterior-posterior axis running from left to right for the thoraces. Bar, 0.134 mm for A and D; 1 µm for B and E; and 0.5 µm for C and F.

Suppression by the myosin suppressors of other behaviors affected by hdp²:
The hdp² mutation is in the constitutively expressed exon 5 of the TnI gene. Its effects on other muscle groups have been observed as changes in behavior (adult jumping and walking and larval crawling and feeding), including an age-dependent myopathy of the legs, associated with ultrastructural defects (NAIMI et al. 2001 ). All the new myosin suppressors fully suppress the effects of hdp² on walking (Fig 6) and all other behaviors (data not shown), consistent with their position within constitutive Mhc gene exons. In Su(2)D heterozygotes the hdp² walking behavior is partially suppressed while in homozygotes the suppression is complete. Suppressor Su(2)B itself shows a progressive myopathy of the leg muscles, but hdp²/Y; Su(2)B/+ walks as well as wild type (Fig 6).

View larger version (26K): In this window In a new window Download PPT slide   Figure 6. Suppression of the hdp2 walking behavior. hdp2 flies show age-related myopathy in their walking behavior, which is partially or completely suppressed by all the suppressors. As in the fiber morphology, Mhc2D/+ partially suppress the hdp2 myopathy. All the suppressor genotypes show normal walking behavior that is not significantly different from wild type except Mhc2B/+, which shows significant (5%) levels of age-related myopathy. "Seconds" is the mean time (10 tests/sample) taken for 50% of the flies to walk upward >80 mm.

Headless transgenic construct and newly isolated myosin suppressors suppress other hypercontracting alleles:
If the IFM hypercontraction phenotype requires actomyosin force production, then a priori all myosin suppressor mutations, including the transgenic headless myosin constructs, should suppress all those mutations that can generate the phenotype. We have tested this by making genotypes containing hypercontracting alleles and the Mhc suppressors and/or the headless myosin constructs. Both headless myosin constructs (Y97 and Y57) suppress the up¹⁰¹, Mhc¹³, and fln⁰ genes in a dose-dependent manner (Table 2). A single copy of Y57 is enough to suppress fln⁰ hypercontraction in the absence of any full-length endogenous MHC. Fiber morphology looks normal although with less birefringence, possibly a result of the highly disrupted myofibrillar organization at eclosion (data not shown). Such myofibrillar disruption is usually seen a few days after eclosion in fln⁰ mutants and then only after hypercontraction has occurred (REEDY et al. 2000 ). The fln⁰ fiber phenotype is also suppressed by a copy of the Mhc mutation Su(2)B (Table 4), indicating that force is required for fln⁰ hypercontraction.

Other myosin suppressors of hdp² also suppress up¹⁰¹:
KRONERT et al. 1999 described four Mhc suppressors of hdp² in which the mutants localized within the myosin head domain. These mutations were dominant flightless in combination with hdp² (PRADO et al. 1995 ) but fully or partially flighted in heterozygous condition without hdp², indicating that the mutant myosins can assemble into normal myofibrils. We find that these mutations completely suppress the hypercontraction phenotype in flies up¹⁰¹ and up¹⁰¹,+/+, hdp² (Table 5). The degree of suppression appears to relate to how severely a mutation affects the molecule; the intragenic deletion mutations (D41, D62) suppress more strongly than either point mutation (D1 or D45). Similar results were obtained with fln⁰. All four myosin alleles suppress hypercontraction of the flightin mutant but D62 showed the strongest effect (data not shown).

View this table: In this window In a new window   Table 5. Suppression phenotypes of hdp2 and up101 with D series mutant suppressors of KRONERT et al. 1999

Mhc suppressors cause hypercontraction in the absence of a functional troponin complex:
The wupA allele hdp³ causes a missplicing of the IFM-specific transcript and no functional TnI is produced (BARBAS et al. 1993 ). The result is that the IFM fail to form and only small muscle remnants are seen in the thoraces of later pupal and adult stages (Fig 7A; BEALL and FYRBERG 1991 ; BARTHMAIER and FYRBERG 1995 ). A priori an absence of TnI is expected to prevent thin filament inhibition of muscle contraction. Unregulated contraction during myogenesis will lead to the observed phenotype. We have used this circumstance to explore whether the MHC produced in myosin suppressor homozygotes produces sufficient force in vivo to cause the muscle destruction previously seen in hdp³ flies. These experiments were performed to resolve the issue (see below) as to whether the Mhc suppressor mutations affect only force production or the role of the myosin in the muscle activation processes itself, since some current models (reviewed in GORDON et al. 2000 ) suggest that binding of the myosin head to the actin of the thin filament plays a role in muscle activation. The results (Table 6; Fig 7) show that in hdp³ flies lacking myosin, hdp³/Y; Mhc¹²/Mhc¹², the muscles do not hypercontract, nor do they do so in flies heterozygous, Mhc^X/Mhc¹², for any of the suppressors (Mhc^X) and the Mhc¹² null mutant. However, in hdp³ genotypes containing Mhc suppressors heterozygous with Mhc¹², i.e., hdp³/Y; Mhc^X/Mhc¹², hypercontraction occurs (Fig 7C), although in some cases it is only partial (alleles Su(2)B, Su(2)F, and D62; see Fig 7B). These latter results suggest that these alleles show the largest reduction in force production so that, even in the absence of normal muscle regulation, in a single gene dose they can barely produce sufficient force to cause hypercontraction.

View larger version (37K): In this window In a new window Download PPT slide   Figure 7. Mhc suppressors cause hypercontraction in the absence of a functional troponin complex. (A) Polarized light micrograph of 2-day-old hdp3/Y; thorax appears completely empty without IFMs and tergal depressor of trochanter (star). Occasionally skeins of fibers are seen (arrow). (B) hdp3/Y; Mhc2B/Mhc12. The bulk of IFM fibers develop (squares) but show partially pulled fibers from attachment sites (arrowheads), indicating partial contact between thick and thin filaments with reduced force. (C) hdp3/Y; Mhc2D/Mhc12. Only a few bits of fiber remain, indicating hypercontraction caused by the actomyosin force. hdp3 requires very little functional myosin to cause hypercontraction (U. NONGTHOMBA and J. C. SPARROW, unpublished results). Anterior-posterior axis running from left to right for the thoraces. Bar, 0.125 mm.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Our aim was to test the following proposals: that actomyosin force generation is required to produce the hypercontraction phenotype and that force reduction explains the action of Mhc suppressors. Additionally, we intended to distinguish between these proposals, also made by KRONERT et al. 1999 , and their alternative explanation that Mhc suppressors may indicate direct interactions between the myosin head and the TnI component of the troponin complex.

If the role of the myosin head in hypercontraction is to produce the forces that destroy the fibers in response to aberrant regulation of contraction or when the sarcomeric structure is compromised, then all hypercontracting mutants should be suppressed by reductions in myosin head concentration. In addition, suppression by Mhc alleles should be neither allele nor gene specific. The headless Mhc gene construct suppresses the hypercontraction muscle phenotypes of the hdp², up¹⁰¹, Mhc¹³, and fln⁰ mutants in an Mhc gene dose-dependent manner consistent with this expectation. In addition, all of the Mhc suppressors described previously (KRONERT et al. 1999 ) and newly described here suppress the hypercontraction of all these hypercontracting mutants. There is no evidence for gene or allele specificity in the known Mhc suppressors. Although the fiber morphology shows complete suppression by the headless myosin constructs, normal sarcomeric structure is not restored. This is not surprising since the headless myosin does not produce wild-type myofibrils when expressed alone or in combination with wild-type myosin (CRIPPS et al. 1999 ). Incomplete suppression of structural aspects suggests that reduced force production is not sufficient to allow normal myofibrillogenesis or to prevent microdamage within the sarcomeres.

If suppression of hypercontraction by Mhc alleles is by reduced force production, then the stronger suppressors should be those Mhc alleles with more extreme phenotypes. Suppression by the new Mhc alleles is complete by the criterion of fiber structure in each case and most are stronger suppressors than the earlier ones (KRONERT et al. 1999 ) where suppression was incomplete. Some of these earlier suppressors now fly and homozygotes survive, a feature originally true only for D1 (KRONERT et al. 1999 ). Stronger hdp² suppressors have a more extreme phenotype with respect to myosin function. So Su(2)B and Su(2)F have embryonic lethal and dominant flightless phenotypes, whereas Su(2)D is viable and flighted on its own, but only partially suppresses hdp². The D1 suppressor is at odds with this relationship between suppression and myosin dysfunction; it is homozygous viable and flighted, even allowing the flight of some hdp²/Y; D1/+ flies (Table 5).

The Mhc suppressor mutations all occur in the head domain (Fig 3). D1 (D625G, chicken myosin S1 numbering system) and D62 (a 24-bp in-frame deletion) are in the actin-binding loop (KRONERT et al. 1999 ); Su(2)B (P404S) and Su(2)D (G416S) are located fairly close together in a region known to be involved in actin binding (RAYMENT et al. 1993A ; UYEDA et al. 1994 ; ROVNER et al. 1995 ). D45 (A261T) and D41 (a 2-bp insertion that affects splicing) cause changes near the ATP-entry and the ATP-binding sites (KRONERT et al. 1999 ), while Su(2)F (A465T) is also close to the ATP-binding site. The mutant residues do not form the single cluster expected if they affect a specific binding of the myosin head to a component of the Tm-Tn complex, perhaps TnI, as suggested by KRONERT et al. 1999 , nor has any such interaction been detected in the large volume of research on thin filament regulation (see review by GORDON et al. 2000 ). However, all the mutant residues, including the new Mhc suppressors, are in head regions important for actin binding and nucleotide exchange/hydrolysis, supporting the argument that all the mutants affect the crossbridge cycle and force production, changes sufficient to explain suppression.

Su(2)B (P404S) is next to residue R405, a hot spot for mutations causing human hypertrophic cardiomyopathy (HCM). Arginine 405 is part of a myosin loop that could directly interact with actin (RAYMENT et al. 1995 ). Myosins from myopathy patients move actin filaments with decreased velocity in an in vitro motility assay (CUDA et al. 1993 ) and muscle fibers display diminished power output (LANKFORD et al. 1995 ), although more recent in vitro studies of the same HCM myosins showed enhanced myosin activity (PALMITER et al. 2000 ; YAMASHITA et al. 2000 ). The partial suppression of hdp³ hypercontraction by Su(2)B suggests that mutations in this region of the myosin molecule can reduce force in vivo, but do not directly address changed functions in the HCM mutations.

An absence of clustering of suppressor mutations and the fact that their effects can be explained by effects on myosin ATP hydrolysis and actin interactions reducing force production is not consistent with the proposed direct interaction between myosin and TnI (KRONERT et al. 1999 ). However, current models (reviewed in GORDON et al. 2000 ) suggest that myosin is involved in both force production and thin filament regulation. GEEVES and LEHRER 1998 have developed a model of thin filament regulation based on the kinetic studies of MCKILLOP and GEEVES 1993 in which the Tm-Tn complex can exist in three states on the F-actin thin filament core. In the absence of calcium, TnI binding to actin holds the Tm-Tn complex in the "closed" state in which the myosin-binding site is occluded, preventing myosin from binding actin. Calcium released into the muscles following neural stimulation binds to TnC, which undergoes a conformational change that alters its relationship with TnI, resulting in release of TnI binding from actin. This represents the "blocked" state, but in this state small movements of the Tm/Tn across the F-actin surface allow small numbers of myosin heads to bind F-actin, leading to the displacement of this complex to the "open" state. In this state myosin heads can bind to any available actin "target" site and muscle activation is achieved. In this model, therefore, a small fraction of myosin heads play an important role in the blocked to open transition. In doing so they bind to F-actin. There is no reason to assume that this binding is in any way different from the binding of a myosin head that also produces force. It is thus formally very difficult to determine whether the Mhc mutations that suppress hdp² do so by affecting regulation rather than force production. The reduced ability of the myosin suppressor mutants to hypercontract the IFM in the absence of a functional regulatory system (hdp³) and any wild-type myosin certainly suggests that the suppressor mutations produce less force, but does not allow us to deduce that this is the primary effect of these mutants when acting as suppressors. At present we cannot perform with Drosophila proteins the type of sophisticated biochemical and biophysical experiments needed to resolve this issue. However, considerable progress has been made recently with the purification and assay of Drosophila IFM actin (RAZZAQ et al. 1999 ; SCHMITZ et al. 2000 ) and myosin (SWANK et al. 2001 , SWANK et al. 2002 ), including ATPase assays, in vitro motility, single molecule studies, and rapid kinetics.

IFM hypercontraction appears to develop from either misregulation of muscle contraction (hdp², up¹⁰¹) or structural defects arising from reduced sarcomeric integrity (Mhc⁶, Mhc¹³, fln⁰). It may be significant that in the former mutant group hypercontraction occurs during late pupal stages and is complete shortly after eclosion (NAIMI et al. 2001 ), while in the latter the phenotype arises during the first day or so of adult life (KRONERT et al. 1995 ). On this basis different hypercontracting Act88F alleles may affect regulation or sarcomeric integrity. Despite the different etiology the same Mhc suppressors suppress both groups. This does not allow us to distinguish between force production and regulation of muscle contraction as the primary effect of the mutant myosins. With few exceptions (see NAIMI et al. 2001 ) hypercontraction phenotypes are restricted to the IFM. It is intriguing that this phenotype, which presumably arises from excessive shortening, shows up in the IFM, muscles where contraction is typically isometric.

Not all Mhc alleles that cause myosin dysfunction are suppressors. Two exceptions are Mhc⁶ and Mhc¹³, which have hypercontraction phenotypes themselves. In both, the mutant amino acids are within the myosin -helical coiled-coil domain that associates to form the thick filaments. Mhc⁵ acts as an enhancer of the hdp² phenotype as hdp²/Y; Mhc⁵/+ males are lethal as young larvae. However, this allele on its own produces only a hypercontraction phenotype in adults and the synthetic lethality is almost certainly due to epistatic interactions. The Mhc⁵ mutation substitutes G200 with aspartate (G200D); residue G200 is at the beginning of a helix that interacts with bound nucleotide (KRONERT et al. 1999 ). Mhc⁵ myosin must be able to produce sufficient force for hypercontractive destruction of the muscle, but also affect regulation so that the phenotype occurs during late pupal/early adult life. The occurrence of hypercontraction at this time is consistent with an effect on regulation. How can Mhc⁵ affect regulation? One possibility is that it does so by altering myosin kinetics so that a fraction of heads remains bound, keeping the thin filament in the "open" state in the absence of calcium. The Mhc⁸ mutation (Y832H, a mutation in the myosin lever arm) is recessive lethal and, like Mhc⁵, its survival as a wild-type heterozygote is severely reduced in combination with hdp² (KRONERT et al. 1999 ), so it is an enhancer of hdp². How a lever arm mutation achieves this is not clear. The lever arm binds the essential and regulatory light chains, so Mhc⁸ may affect regulation rather than compromising its force-producing capacity.

Since most Mhc missense mutants that reduce the efficacy of myosin function suppress hypercontraction, they will not be very informative about troponin/tropomyosin complex function. However, the small number of mutations with unexpected interactions with troponin mutations (e.g., Mhc⁵ and Mhc⁸) are likely to be informative about troponin/tropomyosin complex function. Clearly, selection for hypercontraction suppressors efficiently isolates Drosophila Mhc mutations with important effects on myosin function.

Many human familial myofibrillar myopathies of skeletal and cardiac muscle have been linked with sarcomeric proteins (reviewed in COONAR and MCKENNA 1997 ; REDWOOD et al. 1999 ; SEIDMAN and SEIDMAN 2001 ) including myosin, actin, tropomyosin, and the troponins. The hypertrophic cardiomyopathies show very variable penetrance and expressivity, which are likely to have a genetic component (COONAR and MCKENNA 1997 ). The study of interacting mutations in Drosophila, where muscle structure and sarcomeric proteins are very similar to their vertebrate counterparts, should shed light on the direct effects of muscle dysfunction and on the genetic interactions that are important in the occurrence of human muscle disease.

	ACKNOWLEDGMENTS

We thank Sandy Bernstein and Richard Cripps for providing the transgenic lines containing the headless myosin constructs, Alberto Ferrus for the original myosin hdp² suppressor lines, and Meg Stark for her excellent EM work. This research was supported by the BBSRC (UK; J.C.S.) and the National Science Foundation (J.O.V.).

Manuscript received November 13, 2002; Accepted for publication January 24, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AN, H. and K. MOGAMI, 1996  Isolation of 88F actin mutants of Drosophila melanogaster and possible alterations in the mutant actin structures. J. Mol. Biol. 260:492-505.[Medline]

BARBAS, J. A., J. GALCERAN, L. TORROJA, A. PRADO, and A. FERRUS, 1993  Abnormal muscle development in the heldup³ mutant of Drosophila melanogaster is caused by a splicing defect affecting selected troponin I isoforms. Mol. Cell. Biol. 13:1433-1439.[Abstract/Free Full Text]

BARTHMAIER, P. and E. FYRBERG, 1995  Monitoring development and pathology of Drosophila indirect flight muscles using green fluorescent protein. Dev. Biol. 169:770-774.[Medline]

BEALL, C. J. and E. A. FYRBERG, 1991  Muscle abnormalities in Drosophila melanogaster heldup mutants are caused by missing or aberrant troponin-I isoforms. J. Cell Biol. 114:941-951.[Abstract/Free Full Text]

BERNSTEIN, S. I. and R. A. MILLIGAN, 1997  Fine tuning a molecular motor: the location of alternative domains in the Drosophila myosin head. J. Mol. Biol. 271:1-6.[Medline]

BERNSTEIN, S. I., P. T. O'DONNELL, and R. M. CRIPPS, 1993  Molecular genetic analysis of muscle development, structure and function in Drosophila.. Int. Rev. Cytol. 143:63-152.[Medline]

BHATTI, S., G. ZIMMER, and J. BEREITER-HAHN, 1989  Enzyme release from chick myocytes during hypoxia and reoxygenation: dependence on pH. J. Mol. Cell. Cardiol. 21:995-1008.[Medline]

COONAR, A. S. and W. J. MCKENNA, 1997  Molecular genetics of familial cardiomyopathies. Adv. Genet. 35:285-324.[Medline]

COZZI, F., M. CERLETTI, G. C. LUVONI, R. LOMBARDO, and P. G. BRAMBILLA et al., 2001  Development of muscle pathology in canine X-linked muscular dystrophy. II. Quantitative characterization of histopathological progression during postnatal skeletal muscle development. Acta Neuropathol. 101:469-478.[Medline]

CRIPPS, R. M., K. D. BECKER, M. MARDAHL, W. A. KRONERT, and D. HODGES et al., 1994  Transformation of Drosophila melanogaster with the wild type myosin heavy chain gene: rescue of mutant phenotypes and analysis of defects caused by overexpression. J. Cell Biol. 126:689-699.[Abstract/Free Full Text]

CRIPPS, R. M., J. A. SUGGS, and S. I. BERNSTEIN, 1999  Assembly of thick filaments and myofibrils occurs in the absence of the myosin head. EMBO J. 18:1793-1804.[Medline]

CUDA, G., L. FANANPAZIR, W.-S. ZHU, J. R. SELLERS, and N. D. EPSTEIN, 1993  Skeletal muscle expression and abnormal function of ß-myosin in hypertrophic cardiomyopathy. J. Clin. Invest. 91:2861-2865.

DUNCAN, C. J., 1987  Role of calcium in triggering rapid ultrastructural damage in muscle: a study with chemically skinned fibres. J. Cell Sci. 87:581-594.[Abstract/Free Full Text]

FINOL, H. J., A. MARQUEZ, E. NAVAS, and N. R. DE-NAVAS, 2001  Extraocular muscle ultrastructural pathology in the paraneoplastic phenomenon associated with retinoblastoma. J. Exp. Clin. Cancer Res. 20:281-285.[Medline]

GARCIA-ANOVEROS, J., J. A. GARCIA, J. D. LIU, and D. P. COREY, 1998  The nematode degenerin UNC-105 forms ion channels that are activated by degeneration- or hypercontraction-causing mutations. Neuron 20:1231-1241.[Medline]

GEEVES, M. A. and S. LEHRER, 1998  The muscle thin filament as a classical cooperative/allosteric regulatory system. J. Mol. Biol. 277:1081-1089.[Medline]

GEORGE, E. L., M. B. OBER, and C. P. EMERSON, 1989  Functional domains of the Drosophila melanogaster muscle myosin heavy chain gene are encoded by alternatively spliced exons. Mol. Cell. Biol. 9:2957-2974.[Abstract/Free Full Text]

GORDON, A. M., E. HOMSHER, and M. REGNIER, 2000  Regulation of contraction in striated muscle. Physiol. Rev. 80:853-924.[Abstract/Free Full Text]

KORSWAGEN, H. C., J. H. PARK, Y. OHSHIMA, and R. H. PLASTERK, 1997  An activating mutation in Caenorhabditis elegans Gs protein induces neural degeneration. Genes Dev. 11:1493-1503.[Abstract/Free Full Text]

KRONERT, W. A., P. T. O'DONNELL, A. FIECK, A. LAWN, and J. O. VIGOREAUX et al., 1995  Defects in the Drosophila myosin rod permit sarcomere assembly but cause flight muscle degeneration. J. Mol. Biol. 249:111-125.[Medline]

KRONERT, W. A., A. ACEBES, A. FERRUS, and S. I. BERNSTEIN, 1999  Specific myosin heavy chain mutations suppress troponin I defects in Drosophila muscles. J. Cell Biol. 144:989-1000.[Abstract/Free Full Text]

LANKFORD, E., N. EPSTEIN, L. FANANAPAZIR, and H. L. SWEENEY, 1995  Abnormal contractile properties of muscle fibers expressing beta-myosin heavy chain gene mutations in patients with hypertrophic cardiomyopathy. J. Clin. Invest. 95:1409-1414.

MCKILLOP, D. F. and M. A. GEEVES, 1993  Regulation of the interaction between actin and myosin subfragment 1: evidence for three states of the thin filament. Biophys. J. 65:693-701.[Abstract/Free Full Text]

MONTICELLO, T. M., C. A. SARGENT, J. R. MCGILL, D. S. BARTON, and G. J. GROVER, 1996  Amelioration of ischemia/reperfusion injury in isolated rat hearts by the ATP-sensitive potassium channel opener BMS-180448. Cardiovasc. Res. 31:93-101.[Medline]

NAIMI, B., A. HARRISON, M. CUMMINS, U. NONGTHOMBA, and S. CLARK et al., 2001  A tropomyosin-2 mutation suppresses a troponin I myopathy in Drosophila.. Mol. Biol. Cell 12:1529-1539.[Abstract/Free Full Text]

NONGTHOMBA, U. and N. B. RAMACHANDRA, 1999  A direct screen identifies new flight muscle mutants on the Drosophila second chromosome. Genetics 153:261-274.[Abstract/Free Full Text]

O'DONNELL, P. T. and S. I. BERNSTEIN, 1988  Molecular and ultrastructural defects in a Drosophila myosin heavy chain mutant: differential effects on muscle function produced by similar thick filament abnormalities. J. Cell Biol. 107:2601-2612.[Abstract/Free Full Text]

PALMITER, K. A., M. J. TYSKA, J. R. HAEBERLE, N. R. ALPERT, and L. FANANAPAZIR et al., 2000  R403Q and L908V mutant beta-cardiac myosin from patients with familial hypertrophic cardiomyopathy exhibit enhanced mechanical performance at the single molecule level. J. Muscle Res. Cell Motil. 21:609-620.[Medline]

PRADO, A., I. CANAL, J. A. BARBAS, J. MOLLOY, and A. FERRUS, 1995  Functional recovery of troponin-I in a Drosophila heldup mutant after a second site mutation. Mol. Biol. Cell 6:1433-1441.[Abstract]

RAYMENT, I., H. M. HOLDEN, M. WHITTAKER, C. B. YOHN, and M. LORENZ et al., 1993a  Structure of the actin-myosin complex and its implications for muscle contraction. Science 261:58-65.[Abstract/Free Full Text]

RAYMENT, I., W. R. RYPNIEWSKI, K. SCHMIDT-BASE, R. SMITH, and D. R. TOMCHICK et al., 1993b  Three-dimensional structure of myosin subfragment-1: a molecular motor. Science 261:50-58.[Abstract/Free Full Text]

RAYMENT, I., H. M. HOLDEN, J. R. SELLERS, L. FANANAPAZIR, and N. EPSTEIN, 1995  Structural interpretation of the mutations in the beta-cardiac myosin that have been implicated in familial hypertrophic cardiomyopathy. Proc. Natl. Acad. Sci. USA 92:3864-3868.[Abstract/Free Full Text]

RAZZAQ, A., S. SCHMITZ, C. VEIGEL, J. E. MOLLOY, and M. A. GEEVES et al., 1999  Actin residue Glu[93] is identified as an amino acid affecting myosin binding. J. Biol. Chem. 274:28321-28328.[Abstract/Free Full Text]

REDWOOD, C. S., J. C. MOOLMAN-SMOOK, and H. WATKINS, 1999  Properties of mutant contractile proteins that cause hypertrophic cardiomyopathy. Cardiovasc. Res. 44:20-36.[Abstract/Free Full Text]

REEDY, M. C. and C. BEALL, 1993  Ultrastructure of developing flight muscle in Drosophila.. Dev. Biol. 160:443-465.[Medline]

REEDY, M. C., B. BULLARD, and J. O. VIGOREAUX, 2000  Flightin is essential for thick filament assembly and sarcomere stability in Drosophila flight muscles. J. Cell Biol. 151:1483-1500.[Abstract/Free Full Text]

ROTH, S. M., G. F. MARTEL, and M. A. ROGERS, 2000  Muscle biopsy and muscle fiber hypercontraction: a brief review. Eur. J. Appl. Physiol. 83:239-245.[Medline]

ROVNER, A. S., Y. FREYZON, and K. M. TRYBUS, 1995  Chimeric substitutions of the actin-binding loop activate dephosphorylated but not phosphorylated smooth muscle heavy meromyosin. J. Biol. Chem. 270:30260-30263.[Abstract/Free Full Text]

SCHMITZ, S., J. CLAYTON, U. NONGTHOMBA, C. VEIGEL, and M. GEEVES et al., 2000  Drosophila ACT88F indirect flight muscle-specific actin is not N-terminally acetylated: a mutation in N-terminal processing affects actin function. J. Mol. Biol. 295:1201-1210.[Medline]

SEIDMAN, J. G. and C. SEIDMAN, 2001  The genetic basis for cardiomyopathy: from mutation identification to mechanistic paradigms. Cell 104:557-567.[Medline]

STEWARD, R. and C. NUSSLEIN-VOLHARD, 1986  The genetics of the dorsal-Bicaudal-D region of Drosophila melanogaster.. Genetics 113:665-678.[Abstract/Free Full Text]

SWANK, D. M., M. L. BARTOO, A. F. KNOWLES, C. I