- Email: [email protected]
Contribution of Myosin Rod Protein to the Structural Organization of Adult and Embryonic Muscles in Drosophila
doi:10.1016/S0022-2836(03)00827-1
J. Mol. Biol. (2003) 331, 1077–1091
Contribution of Myosin Rod Protein to the Structural Organization of Adult and Embryonic Muscles in Drosophila Erzse´bet Polya´k, David M. Standiford*, Vladimir Yakopson Charles P. Emerson Jr and Clara Franzini-Armstrong Department of Cell and Developmental Biology University of Pennsylvania 245 Anatomy-Chemistry Building, 36th Street and Hamilton Walk, Philadelphia PA 19104-6058, USA
Myosin rod protein (MRP) is a naturally occurring 155 kDa protein in Drosophila that includes the myosin heavy chain (MHC) rod domain, but contains a unique 77 amino acid residue N-terminal region that replaces the motor and light chain-binding domains of S1. MRP is a major component of myofilaments in certain direct flight muscles (DFMs) and it is present in other somatic, cardiac and visceral muscles in adults, larvae and embryos, where it is coexpressed and polymerized into thick filaments along with MHC. DFM49 has a relatively high content of MRP, and is characterized by an unusually disordered myofibrillar ultrastructure, which has been attributed to lack of cross-bridges in the filament regions containing MRP. Here, we characterize in detail the structural organization of myofibrils in adult and embryonic Drosophila muscles containing various MRP/MHC ratios and in embryos carrying a null mutation for the single MHC gene. We examined MRP in embryonic body wall and intestinal muscles as well as in DFMs with consistent findings. In DFMs numbers 49, 53 and 55, MRP is expressed at a high level relative to MHC and is associated with disorder in the positioning of thin filaments relative to thick filaments in the areas of overlap. Embryos that express MRP in the absence of MHC form thick filaments that participate in the assembly of sarcomeres, suggesting that myofibrillogenesis does not depend on strong myosin –actin interactions. Further, although thick filaments are not well ordered, the relative positioning of thin filaments is fairly regular in MRP-only containing sarcomeres, confirming the hypothesis that the observed disorder in MRP/MHC containing wild-type muscles is due to the combined action between the functional behavior of MRP and MHC myosin heads. Our findings support the conclusion that MRP has an active function to modulate the contractile activity of muscles in which it is expressed. q 2003 Elsevier Ltd. All rights reserved.
*Corresponding author
Keywords: myosin rod protein; myosin; Drosophila; myofibrillogenesis; sarcomere
Introduction Myosin heavy chain (MHC), the motor protein of muscle thick filaments, has two distinct domains: the head and the tail. The head has a globular region with ATPase and actin-binding functions Abbreviations used: MRP, myosin rod protein; MHC, myosin heavy chain; DFM, direct flight muscles; LMM, light meromyosin; GFP, green fluorescent protein. E-mail address of the corresponding author: [email protected]
and extends into a single alpha-helical segment with binding sites for the two light chains. Cyclic interactions of the MHC head, or myosin crossbridge, with the actin filament, coupled to hydrolysis of ATP, results in muscle contraction by a sliding filament mechanism, producing either tension or shortening. The myosin heads are seen by electron microscopy as “cross-bridges” projecting from the surface of the thick filament and attaching to the thin filaments when the muscle is fixed either during active contraction or in the rigor state, in which all cross-bridges are attached.
0022-2836/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
1078
The tail is fully alpha-helical; it forms a coiled-coil with a second MHC, whose light meromyosin (LMM) region is responsible for the spontaneous polymerization of myosin into filaments at physiologic ionic strength.1 The properties of mutant LMMs suggest that charge distributions in the coiled-coil may influence the detailed axial stagger of the interacting molecules, thus determining the axial periodicities of myosin heads over the surface of the thick filaments.2 A specific sequence near the C terminus of the coiled-coil region of the MHC is necessary for the formation of ordered LMM paracrystals and thus presumably for the formation of the thick filament backbone.3 In Drosophila melanogaster, a number of musclespecific isoforms are generated through alternative mRNA splicing from a single MHC gene (DMhc).4 – 6 However, within the DMhc gene, a second transcriptional unit encodes a 155 kDa protein termed the myosin rod protein (MRP).7,8 The MRP transcript initiates within DMhc intron 12 and encodes a 77 amino acid residue N-terminal domain that is specific to MRP before continuing in-frame through exon 13 and the remainder of the exons that encode the rod domain of MHC. Thus, MRP has the same rod domain as MHC, but the motor and light chain-binding regions are replaced by an MRP-specific N-terminal domain. The proline-rich MRP N-terminal domain is similar to the N-terminal extensions of MLC2 in Drosophila7,9 and the essential light chain (ELC) of vertebrate cardiac myosins,10,11 which have been proposed to mediate thick and thin filament interactions through the formation of molecular tethers.10 Myosin splice variants with a truncated N-terminal domain, similar to MRP, are also found in several smooth muscles of molluscs,12,13 suggesting that such a role might be broadly important for contractile function. Previous work in Drosophila showed that MRP is widely expressed in somatic muscles but is also found in cardiac and visceral muscles and in nonmuscle tissues.7,8 MRP was found to be particularly abundant in a set of direct flight muscles (DFM) in the adult thorax, and an analysis of these muscles showed that MRP associates with MHC in the thick filament as a homodimer and can constitute a significant fraction of the thick filament protein relative to MHC. The inclusion of MRP at a high level into the thick filament resulted in the detectable loss of MHC-mediated cross-bridges in the rigor filament and is believed to contribute to several distinct ultrastructural features observed in these muscles, including the presence of longitudinally disordered thick and thin filament and a remarkable “wandering” behavior of thin filaments. However, the extent of these phenotypes was not uniform among the various MRPexpressing muscles, suggesting that MRP might have muscle-specific effects on sarcomeric ultrastructure. Here, we extend the previous observations through characterization of the sarcomeric organiz-
Contribution of MRP to Sarcomere Organization
ation in muscles that contain different levels of MRP. In addition, we address the relative contribution of MRP and MHC to the muscle phenotype through the analysis of myofibers in mutant animals that express MRP, but not MHC.
Results The content of MRP in direct flight muscles The analysis of a lacZ reporter gene under the regulation of the Mrp promoter showed previously that a subset of DFMs in the adult thorax express MRP, with the most prominent of these being DFM49, 53 and 55 (previously identified as DFM 54).8 MRP-specific antibodies used for immunolabeling showed some differences in labeling intensity among these muscles (not shown), suggesting varying levels of MRP. These levels appeared highest in DFM49 and were consistent with this muscle having the greatest degree of ultrastructural “disorder”. To directly determine the relative level of MRP expression in the DFM muscle, individual muscles were dissected and subject to quantitative immunoanalysis to establish the ratio of MRP to MHC in each muscle (Figure 1). These data show that the level of MRP is high in DFM49, 53 and 55, where it exceeds an 8:1 ratio with MHC in all cases. Thus, in each of these muscles MRP is expected to be a major thick filament protein. DFM51 and 54 and the tergal depressor of the trochanter (TDT) were also assayed and, consistent with our earlier data, were found not to express MRP. Structure of the sarcomere in MRP-expressing DFMs of wild-type adult fly DFM51 and 54, which do not express MRP, have a very similar structure. In particular, in the overlap region of the A band thin and thick filaments have an ordered parallel arrangement (Figure 2(A), (C), and (D)). DFM53 and 55, which do express MRP, also have structural features common to each other, but differ from DFM51 and 54. The spacing between thick filaments is variable and thin filaments wander back and forth between thick filaments, sometimes approaching one and sometimes its neighbor (Figure 2(B) and (E) – (H)). The disorder is most apparent in DFM55, which has a high relative content of MRP and resembles the highly disordered DFM49, which had been selected for an initial study8 (Figure 2(H)). When muscles that do not express Mrp are fixed in rigor, cross-bridges are connected to the actin filaments at evenly spaced target zones with a marked periodicity. Each target zone is occupied by a single or a doublet of cross-bridges (Figure 2(C) and (D)). In MRP-expressing muscles, cross-bridges are fewer and less periodically arranged. In DFM53 the content of cross-bridges varies between small domains that have an almost
Contribution of MRP to Sarcomere Organization
1079
Figure 1. (A) The myosin rod protein (Mrp) gene is encoded by a separate transcription unit embedded within the Drosophila 36B muscle myosin heavy chain (Mhc) gene. The Mrp transcript initiates in Mhc intron 12 and encodes a unique 77-residue N-terminal peptide that is fused in-frame with the rod domain of the Mhc gene. Thus, the MRP N terminus replaces the MHC motor and light chain-binding domain (LCB), but is identical with MHC in the filament forming rod domain. (B) Cartoon of adult lateral thorax showing the position of the direct flight muscles (DFMs; nomenclature according to Millar.45) Muscles-expressing MRP are filled. A transgenic GFP-reporter construct using the Mrp promoter domain was used to identify and isolate individual MRP-expressing DFM49, 53 and 55 from adult flies. (C) Immunoblot analysis of MHC and MRP (upper and lower bands, respectively) isolated from individual DFMs and the tergal depressor of the trochanter (TDT) muscle, showing that MRP is heavily expressed in DFM49, 53 and 55, while it is not detected in DFM51, 54 or the TDT. Quantification of the bands from MRP-expressing muscles indicates that the MHC-to-MRP ratio is at least 1:8 in each muscle.
full complement of cross-bridges and others where cross-bridges are mostly missing (Figure 2(E) –(G)). DFM55 has no domains with ordered cross-bridge arrangement (Figure 2(H)). Measurements of the average distance between actin target zones shows that the spacing is shorter and considerably more uniform in DFM51 and 54, which have no MRP, than in DFM53, 49 and 55, which contain MRP (Table 1). It should be noted that the positioning of cross-bridge occupied target zones in the highly disordered DFM is less accurately measurable than in DFM51 and 54. Conversely, the occupancy of actin filaments by cross-bridges is higher in DFM51 and 54 than in DFM53 (Table 1). The disposition of thin filaments is directly related to the frequency of cross-bridges; in crossbridge-rich regions the filaments are straight and parallel to the thick filaments, while where crossbridges are missing the filaments are wavy and irregular. Based on previous observations,8 is it logical to assume that regions with lower frequency of cross-bridges are those in which a
portion of the thick filament backbone is constituted by MRP. On this basis, the structural data provided here indicate that sarcomere organization is strongly dependent on the MRP content and that the content may vary within domains of the same muscle fiber and even within the same sarcomere (Figure 2(G)). Observations on cross-sections of DFM muscles are in agreement with predictions from longitudinal sections, in the sense that the ordered arrangement of thin and thick filaments is affected by the presence of MRP. In muscles that lack MRP, thick filaments form a hexagonal array with only occasional discontinuities in the pattern (Figure 3(A)). The thin filaments are precisely disposed in an equidistant (dyadic) position between two adjacent thick filaments. Either a single crossbridge or a doublet of cross-bridges occupies each dyadic position. The number of thin filaments surrounding each thick filament varies between seven and ten in DFM51 (Figure 3(B)) and six to eight in DFM54 (image not shown). The variability in the thin/thick filament ratio gives an appearance of
1080
Contribution of MRP to Sarcomere Organization
Figure 2. Sarcomere structure of DFMs. Comparison of the non-MRP-containing DFM51 (A) and the MRP-containing DFM53 (B), showing considerably higher disorder in the latter in the disposition and spacing of both thin and thick filaments. At higher magnification, muscles containing no MRP (C) and (D) have evenly spaced rigor crossbridges and the thin filaments are straight and equidistant from the two adjacent thick filaments. In muscles with MRP content (E) – (H), the frequency of cross-bridges varies in different areas and the spacing between thin and thick filaments is highly variable.
slight disorder to the cross-sectional pattern of DFM muscles that do not contain MRP. However, close examination (Figure 3(B)) shows that the thin filaments are approximately equidistant between thick filaments, suggesting equal interaction between a thin filament and two adjacent thick filaments. Thick and thin filament patterns are not as regular in MRP-expressing muscles. First, the hexagonal array of thick filaments is distorted and this seems to be dependent on MRP content: DFM53, a muscle with an intermediate level of MRP, shows more distortion than DFM51. Second, even in muscles where the thin/thick filament ratio is approximately the same, the thin to
thick filament distance varies more strongly in MRP-expressing muscles than in those where MRP is lacking (Figure 3(C) and (D)). In particular, some thin filaments are very close to the surface of thick filaments (arrows, Figure 3(D) and (F)) and some are at large distances, two positions that are not seen in the absence of MRP (Figure 3(B)). DFM49 (see Ref. 8) and 55 (Figure 3(E) and (F)) show the highest degree of disorder. The structural characteristics for DFM49 have been detailed by Standiford et al.8 In DFM55 the thick filaments do not form a hexagonal array and thin filaments are not only variably positioned relative to the surface of the thick filaments, but also are very unevenly
1081
Contribution of MRP to Sarcomere Organization
Table 1. Measurements of DFM cross-bridge distances and filament densities Distance between cross-bridgeoccupied target zones (average (range) nm) DFM51 DFM54 DFM53 DFM49 DFM55
32.6 ^ 2.6(27.9– 38.5) n ¼ 76c 32.8 ^ 3.0(27.5– 38.7) n ¼ 48c 43.6 ^ 12.2(25.6– 103.2) n ¼ 151c 43.5 ^ 14.6(25.9–91.0)d n ¼ 27c 45.2 ^ 15.7(20.6–91.0)d n ¼ 56c
Cross-bridge no./100 nm (average (range)a)
Thick filament density (average no. (mm2)b)
Thin/thick filament ratio (average no. b)
4.7 ^ 0.7(3.1– 5.9) n ¼ 55 4.7 ^ 0.6(3.7– 6.3) n ¼ 36 3.2 ^ 0.9(1.7– 5.2) n ¼ 61
711 ^ 38.5 n ¼ 21 658 ^ 53 n ¼ 21 823 ^ 64 n ¼ 21 994 ^ 107 n ¼ 21 908 ^ 68 n ¼ 21
3.1 ^ 0.2 n ¼ 21 2.4 ^ 0.2 n ¼ 21 1.8 ^ 0.2 n ¼ 21 1.8 ^ 0.2 n ¼ 21 1.6 ^ 0.3 n ¼ 21
a
All cross-bridges occupying actin segment were counted and referred to the length of the filament. Mean ^ SD measured in 21 different 4.7 £ 1022 mm2 areas for each muscle. Data are not corrected for shrinkage and compression artifacts. Average distance in the DFM51 and 54 is expected to be 38.7 nm, which is the axial repeat in rigor IFM.46 c Mean ^ SD, n ¼ number of actin segments measured, from three flies for each measurement. d In DFM49 and DFM55 there is a very large variation in the distances between cross-bridge-occupied sites along filament segments and there is usually very little overlap in the fixed muscles. This limited the available sites for measurements and the data are less accurate. b
Figure 3. Comparison of rigor DFM muscles in transverse sections. In the non-MRP containing DFM51 (A) and (B) the thick filaments are in fairly regular hexagonal arrangement. Seven to ten thin filaments surround each thick filament. Each dyadic position along a line joining the center of two adjacent thick filaments is occupied by either one or two symmetrically disposed thin filaments that are equidistant from the two nearest thick filaments. In the MRP-containing DFM53 (C) and (D), the hexagonal arrangement of thick filaments is not as regular. Four to seven thin filaments surround each thick filament; their position is not always symmetric relative to the dyadic position; and their distance from the thick filament surface is variable. Some filaments actually are very close to the thick filament’s surface (arrows). In DFM55 (E) and (F), which have a higher amount of MRP, the thick filaments are quite disordered; two to eight thin filaments surround each thick filament and the thin –thick distances are even more variable than in DFM53.
1082
distributed: some areas totally lack thin filaments so that adjacent thick filaments are very close to each other. Overall, the ratio of thin to thick filaments is reduced compared to DFMs that do not express MRP (Table 1) and the complement of thin filaments serving each thick filament profile varies between 2 and 8. The thick filament density for the different DFM muscles was also measured as the number of thick filaments per unit crosssectional area (Table 1). This analysis showed a somewhat higher density of thick filaments in MRP-expressing muscles, suggesting that the inclusion of this protein can effect filament packing in the DFMs. Structure of body wall and intestinal muscles in wild-type and Mhc 1 mutant embryos The above results suggest that the level of MRP relative to MHC can have a differential, dosedependent effect on the structure of the sarcomere. In order to examine this further, we studied the embryonal body wall muscles from wild-type fly, that express both MHC and MRP, and from genetically mutant embryos that express MRP, but not MHC (Mhc 1/Mhc 1) in order to determine the ability of MRP alone to construct a thick filament and to assess the affect that this might have on sarcomeric structure. In Drosophila embryos the body wall muscles are segmentally organized, with each body segment containing about 30 different muscle fibers.14 Of these, approximately half express MRP.8 Although animals that are null for MHC die at hatching due to a failure to break from the egg membranes, muscle development in Mhc 1 homozygotes is essentially complete and thus can be examined for the contribution of MRP alone to sarcomeric structure.
Contribution of MRP to Sarcomere Organization
In order to investigate the results of MRP expression on sarcomeric structure in the absence of myosin, we established a line that contained the Mhc 1 null allele in combination with a balancer second chromosome that contained a green fluorescent protein (GFP) gene under the regulation of the Act5C promoter for expression in the early embryo. Embryos that are homozygous for Mhc 1 can then be selected for the absence of GFP staining under the fluorescence dissecting microscope. As seen in Figure 4(A) and (B), immunostaining of wild-type embryos with antibodies directed against the rod domain of myosin/MRP shows the organization of the embryonic musculature. Western blot of total protein extract prepared from stage 18 embryos (B, inset) shows that overall MRP is expressed at much lower level than MHC. However, the MRP content of individual groups of muscles varies. In MHC-null embryos (Figure 4(C)) some muscles are well-delineated by the anti-myosin rod antibody, indicating a relatively high level of MRP expression. The fibers in the MHC-null embryos are less ordered and shorter than those in wild-type. These observations are consistent with the data presented by O’Donnell & Bernstein,15 who previously noted the existence of thick filaments in Mhc 1 homozygotes. Ultrastructure of embryonic body wall muscles from wild-type and MHC-null embryos Thin sections cut transversely to the long axis of the embryo were used to examine the ultrastructure of the musculature in wild-type and mutant animals. In these studies, direct comparisons between specific body wall muscles were not made due to the difficulty in identifying individual
Figure 4. Immunoanalysis of muscle patterns in Drosophila embryos. (A) Wild-type embryonic muscles are outlined by an anti-MHC rod antibody and peroxidase-labeled secondary antibody. The posterior end is at right and dorsal is up. Note variable orientations of small bundles of cells (B) and (C). Fluorescent immunolabeling of wild-type (B) and MHC-null (C) embryos using an antibody against the rod portion of the molecule. Muscle fibers in the MHC-null body wall express readily detectable amounts of MRP, but in general the fibers are less organized and shorter. (B, inset) Western blot of total embryo extract using anti-MHC rod antibody shows that MRP (lower band, single arrow) is present in addition to MHC (upper band, double arrow). The blot combines all embryonic muscles, some of which do not contain any MRP. (B) and (C) are at the same magnification.
Contribution of MRP to Sarcomere Organization
muscles in the sectioned embryos. However, all observations were made at the level of abdominal segments 2 – 7, where internal markers such as gut profiles and the ventral nerve chord could be used as landmarks for orientation and for confirming approximate segment level. Due to the variable orientations of the muscle, transverse and longitudinal views of the muscle were obtained in the same sections. In general, the embryonic body wall muscles are small and have a somewhat variable ultrastructure. Some cells have several small diameter myofibrils, others only a single or at most two myofibrils of larger diameter. We have identified the cells with larger myofibrils as belonging to A2 – A7, but the small myofibril fibers may be from more anterior or posterior segments. Solid core thick filaments of uniform diameter are arranged in a hexagonal pattern that is quite ordered in the small myofibrils, but tends to be disordered in the larger ones (Figure 5(A)). Local thin-to-thick filament ratio is variable, with a ring of six/seven to ten thin filaments surrounding each thick filament. Most thin filaments are equidistant between two thick filaments and in the dyadic position (Figure 5(B)), but some thin filaments do not belong to this pattern and are found at odd positions and either close to the surface of the thick filament or at relatively large distances from it (arrows Figure 5(A) and (B)). This occurs more frequently in the myofibrils that have a less regular arrangement of thick filaments. The muscle depicted in Figure 5 is from a dorsal muscle group, which has relatively low MRP expression levels, and lateral muscle groups have similar appearance. The arrangement of thin and thick filaments in these muscles resembles that seen in DFM53, suggesting that the slight degree of disorder observed is due to the presence of MRP. None of the embryonic muscles have the high degree of thin and thick filament disorder detected in the DFM muscles with high content of MRP, such as DFM55 and 49. In the body wall muscles of MHC-null embryos, thick filaments are present and they overlap with thin filaments. The profiles of thin and thick filaments are equally sharp in some regions of each fiber, indicating that the filaments run parallel to each other (Figure 5(D)). The orientation of groups of filaments, however, may change abruptly within the same fiber (upper border of Figure 5(C)). There is no noticeable hexagonal arrangement of thick filaments and the thin/thick filaments ratio is very high. Long rows and large groups of thin filaments are visible (Figure 5(C) and (D), arrowheads and arrows). In longitudinal sections, the thick filaments are fairly parallel to each other, they overlap with thin filaments and are interposed between dense bodies (arrowheads, Figure 5(F)). Thus the filaments are obviously part of sarcomeres, but they have different sizes and may abruptly change direction. There is a large variation from fiber to fiber in
1083
the frequency and size of thick filaments. Figure 5 illustrates one of the fibers with the highest frequency of thick filaments; others have fewer and some have none. In the latter type of cells, the entire cytoplasm is filled by thin filaments. Although we were not able to relate the occurrence of thick filaments to specific Mrp-expressing muscles, their presence in certain fibers of the Mhc 1 embryo is expected to result from the expression of MRP and its assembly around a paramyosin core to form myosin-free thick filaments. Further, variations in the frequency of thick filaments in the muscles of the Mhc 1 embryo is predicted to be directly relative to the level of MRP in individual muscles. One feature of the MHC-null muscles is worth noting: the distance between thin and thick filaments is more constant than in the wild-type muscles containing a mixture of MHC and MRP. A halo of fairly uniform size separates the thin and thick filament profiles. In the wild-type embryonic body muscles, the cross-sectional diameter of thick filaments is fairly uniform (Figure 5(A) and (B)), except for the end regions, where the filaments taper off (not shown). The average thick filament diameter in the area of overlap with thin filaments is 18.9(^ 2.7) nm (mean ^ SD, from 35 measurements in five fibers). In the MHC-null mutants on the other hand, the diameter of the thick filaments is highly variable and may be quite large, measuring between 14 nm and 60 nm (Figures 5(C) and (D), and 6(A)–(D)). In sections parallel with the filament long axis, the larger filaments show the periodic striations characteristic of paramyosin paracrystals (Figure 6(E) and (F)) and of the core of paramyosin-containing thick filaments that have been stripped of their surface coating of myosin.16 – 18 Drosophila muscles normally contain paramyosin.19,20 Therefore, it is likely that the fibers in the MHC-null embryos express approximately normal levels of paramyosin and that large size of the thick filaments results from the high ratio of paramyosin to MRP. Some of the larger filaments may be composed purely of paramyosin. Ultrastructure of gut muscles from wild-type and MHC-null embryos The wild-type gut muscles have a higher thin/ thick filament ratio than adult DFM and embryonic body wall muscles (Figure 7(A) and (B)). A doublet of thin filaments occupies the majority of the dyadic positions, so that a total of 11 –12 thin filaments surround each thick filament. As a result, the overall appearance in cross-section is quite well ordered. Interestingly, even though the wild-type gut muscles contain MRP,8 we do not observe the variations in thin filament positioning that are present in other MRP-containing muscles. The gut muscles of the MHC-null embryo have one major feature in common with the MHC-null body muscles: a higher thin/thick filament ratio than in
1084
Contribution of MRP to Sarcomere Organization
Figure 5. EM structure of body wall muscles from wild-type and Mhc 1 mutant embryos. In transverse sections of wild-type muscle (A) and (B) some of the same elements of disorganization seen in the MRP-containing DFMs are visible. In particular, the position of thin filaments is variable in the pattern, so that some are close to the thick filaments (arrow) and others are asymmetrically disposed relative to the adjacent thick filaments (double arrows). Nine to 12 thin filaments surround each thick filament, but with local variations. In the MHC-null mutants (C)– (F),
Contribution of MRP to Sarcomere Organization
1085
Figure 6. Examples of large diameter filaments (A)– (E) in the MHC-null embryonic body wall muscles. These filaments are thought to have a large paramyosin core and a surface array of myosin. Some very large filaments indeed appear as typical, banded paramyosin paracrystals, and may have no myosin on their surface (E) and (F).
the wild-type embryo (Figure 7(C) and (D)). However, differently from the body wall muscles, the MHC-null gut muscles have thick filaments of fairly uniform diameter, and large filaments are rare (Figure 7(C), arrow). One possibility is that this is due to a lower content of paramyosin, or lower ratio of paramyosin to MRP. In addition, the MHC-null gut muscles (Figure 7(D)) have a fairly regular disposition of thin and thick filaments in cross-section and parallel thin and thick filaments in longitudinal section (Figure 7(E)). In these two details they differ from the adult wild-type DFM muscles. It is possible that in the gut muscle of wild-type embryos, MHC to MRP ratio is relatively low so that the absence of MHC in Mhc 1 mutant embryos has less of an affect on muscle structure in the gut than on body wall muscles. Interestingly, gut muscle in the Mhc 1 mutants has a higher degree of order than the MRP-expressing DFMs in the wild-type fly. The implications of this observation are considered in the Discussion. When the distance between the edge of the thick filaments and the edge of the thin filaments were measured in gut muscle, we found that in both wild-type and MHC-null fibers, the distance is the same: 6.7(^ 2.6) nm (n ¼ 33), indicating that the distance between the nearest thin filaments and the thick filaments in the MHC-null gut muscles is not different from the wild-type.
Discussion The MRP is the first example of a naturally occurring protein with the ability to substitute for
myosin in the assembly of the thick filament. MRP clearly contributes to distinct and unexpected structural features in muscles that express it. Here, we have characterized this contribution and have defined three central features of MRP inclusion in sarcomeric assembly and structure. First, we show that the levels of MRP inclusion in normal muscle fibers can exceed that of MHC and that the sarcomeric disorder among flight muscles that express MRP is variable. Second, we determine that MRP alone, in the absence of MHC, can support thick filament and sarcomere formation. Finally, we demonstrate that some characteristics of the disorder in MRP-expressing muscles are dependent on the presence of both MRP and MHC, indicating that these features are a result of the functional interactions between these two components of the thick filament. We find that high levels of MRP content correlate with two structural details of the sarcomere in adult DFM muscles: (1) the degree of order in the hexagonal arrays of filaments is diminished in muscles that have high levels of MRP; and (2) the variability of thin – thick filament distances increases with MRP content. These observations can be described by a model in which MRP is copolymerized with MHC into the thick filament and effectively replaces myosin motor domains with MRP N-terminal domains. An essential concept of the model is that the MRP N-terminal domain can act as a tether, providing a myosin to actin link that is considerably weaker than an active cross-bridge. Although the conformation and length to the MRP N terminus has not been directly determined, this concept is based on the
the thin/thick filament ratio is obviously higher, since clusters of thin filaments are seen between the thick ones. The cross-sectional diameter of the thick filaments is highly variable (C). (E) and (F) show the A-band (E) and I-Z-I regions and parts of A-bands (F) of MHC-null muscles, in longitudinal sections. Typical elements of the sarcomere are present: thin and thick filaments overlap, I-band and interposed dense bodies (arrowheads). However, there is a scarcity of thick filaments and there is some disorder of thin and thick filaments (E).
1086
Contribution of MRP to Sarcomere Organization
Figure 7. Structure of gut muscles from wild-type and MHC-null embryos. Transverse sections show a well-ordered double hexagonal arrangement of thick and thin filaments (A, arrow) in the wild-type embryo (A) and (B). The almost complete complement of 12 thin filaments around each thick filament and the constant thick filament diameter contribute to the symmetry of the system. In the mutant embryos (C) and (D) thick filament diameter is variable, but very large thick filaments, such as shown in (C) and (D) (arrows) are rare. The clusters of thin filament profiles are due to the high actin/myosin ratio in the absence of MHC. (E) A longitudinal section: two A-bands are separated by Z bodies (Z), and the I-band is very short (arrow). Thin and thick filaments are well aligned and are less disordered than in the body wall muscles (Figure 5(E)).
Contribution of MRP to Sarcomere Organization
similarity of this domain with the N-terminal extensions found in vertebrate cardiac light chains, where both structural and functional data exist to support a tethering function.10 Even if the MRP N-terminal domain is tethered to the actin it would be too small to be easily detectable in thin sections, so that areas where MRP is present would appear empty. The disorder in the thick-to-thick and thick-to-thin arrays in MRP-containing muscles is likely to result from the uneven pull between thin and thick filaments due to the uneven distribution of cross-bridges and MRP tethers on the surface of filaments that contain a large proportion of MRP. Rigor crossbridges impose a specific inter-filament spacing in the area of thin and thick filament overlap.21 In insect muscles where MRP is not expressed, the evenly distributed cross-bridges impose a very precise position to the thin filaments, so that the thin filaments stray only minimally from the perfect dyadic position, equidistant between the filaments. In MRP-containing muscles, we assume that regions of strong pull alternate with regions of weak attachments and this would easily distort the filament disposition, as observed. Interestingly, although MRP content is high in DFM49, 53 and 55, we observe that the level of disorder is consistently less in DFM53. This could represent the effects of a slightly greater MHC content, or potentially represents a structural and perhaps functional difference that is distinct to this muscle. If the former is correct, then the extent and frequency of thick filament segments that show gaps in the array of rigor cross-bridges would be expected to increase with higher levels of MRP. Experiments to manipulate the MRP/MHC ratio in muscle fibers will examine this possibility. In MRP-expressing DFMs, we can define two components of this disorder: one is the fact that some thin filaments are located very close to the surface of the thick filament and the other is that some thin filaments seem to freely float away from the thick filaments. Rigor cross-bridges not only strongly hold actin, but also do not allow it to come too close to the thick filament.21 Thus, the short regions of close approximation between thin and thick filaments in muscles that contain MRP may be sites of MRP tethering. The thin filament segments that freely float away from the thick filaments, on the other hand, may have had their MRP tethers pulled off during cross-bridge action, perhaps due to buckling of the thin filament, allowing the filaments to move freely during fixation and embedding. The same uneven distribution of forces that disturbs the symmetry of thin filament disposition around thick filaments would disturb the hexagonal arrangement of thick filaments, as we observe. Muscles in MHC-null embryos give a direct confirmation of the hypothesis that filament lattice disorder in MHC-MRP-containing muscle is due to the combined action of the two molecules. The body wall and gut muscles of MHC-null embryos
1087
have uneven distribution of thick filaments and a high thin-to-thick filament ratio, but show a quite constant spacing between thick and thin filaments. We assume that the equally weak forces of the MRP tethers from two thick filaments interacting with one thin filament in the absence of myosin hold the thin filament in place between them. The expression of headless myosin (MRP) offers the further opportunity to consider two questions in myofibril assembly: what factors are necessary for the assembly of thick filaments and what factors control the formation of a sarcomere with overlapping thin and thick filaments?22 MHC-null muscles contain thick filaments that overlap with thin filaments in the context of fairly well developed sarcomeres. Thus, the tail region of the myosin is sufficient to form filaments and a strong actin – myosin interaction is not a requirement for the basic steps in sarcomere formation and for the overlap of thin and thick filaments in the overlap zone. These conclusions are not surprising, since the LMM region of the myosin molecule has the self-assembly properties that are required to form thick filaments1 and assemblies of either Z line and thin filaments or of aligned thick filaments with M lines (A-bands) can form independently of each other.22,23 Further evidence that actomyosin interactions are not required for sarcomeric formation is the fact that overlap of thin and thick filaments was also seen in a study of the expression of headless myosin molecules in a myosin null background24 and that myosin filaments seem capable of overlapping with microtubules, with which they do not directly interact.25 In addition to its contribution to sarcomere formation, MRP is also predicted to interact with paramyosin in a manner similar to MHC. It is known that paramyosin is contained in the thick filaments of a variety of invertebrate muscles. If the paramyosin –myosin ratio is fairly high, the former constitutes the core of the filament and the latter is helically arranged on the surface. The higher the paramyosin/myosin ratio, the larger is the filament diameter.18,26,27 In muscles with small amounts of paramyosin, the protein is associated with myosin subfilaments and thus it is distributed throughout the thick filament.28 Even though the molecular packing of paramyosin in the filament core is paracrystalline, the fine periodicity due to this arrangement is masked by the myosin on the surface.18 In the body wall and gut muscles of MHC-null embryos, we observe thick filaments of variable diameter with no noticeable periodicity, as well as rarer periodically striped paramyosin crystals. We conclude that MRP interacts with paramyosin in the same manner as MHC and that MRP is most likely to be located at the surface of the thick filaments. Since the overall level of myosin in these muscles is low in the absence of MHC, some paramyosin forms myosin-free paracrystals. It is clear that the inclusion of MRP into the thick
1088
filament is associated with significant structural characteristics in the sarcomere that are not found in fibers that lack MRP. The possibility of MRPmediated tethering interaction between the thick and thin filaments suggests that the inclusion of MRP may also confer significant functional properties to muscle as well. Several other contractile proteins with domains similar to the N terminus of MRP have been shown to mediate thick –thin filament interactions. For instance the MLC2 from Drosophila has an N-terminal extension not found in other regulatory light chains29 that is similar in length and composition to the MRP N terminus. The function of this domain has been studied in the indirect flight muscle (IFM) where its removal from MLC2 leads to a reduction in the dynamic stiffness and elastic modulus in the IFM and an associated reduced power output.30,31 A similar N-terminal extension is found in the A1 isoform of the vertebrate ELC, and has been shown to interact with the actin filament.32 – 35 Functional analysis of this domain has demonstrated its ability to affect myosin kinetics, including MgATPase activity and actin-binding affinities (reviewed by Sanbe et al.10), showing that N-terminal extensions similar to that found on MRP can have regulatory properties. Further, mutations in the N-terminal extension of the cardiac ELC are linked to familial cardiac hypertrophies associated with the loss of stretch activation in the papillary muscles.36 Together, these observations are consistent with N-terminal extensions forming a molecular tether that acts in parallel to the cross-bridge to enhance the resonant properties of the muscle. A diversity of muscles in Drosophila express MRP, ranging from the fast DFMs to the slow muscles of the gut, and these muscles are likely to be connected through physiological parameters that benefit from the inclusion of MRP into the thick filament. As yet, this commonality is not understood, but several possibilities exist and might include the need to generate a fiber with low power, but a high resting tension to resist stretch. The b1 muscle in the blowfly, Calliphora, is equivalent to DFM49 in Drosophila and has been shown to do primarily negative work when stimulated at a frequency of the wing beat.37 – 39 Thick-to-thin filament tethering mediated by MRP in the DFMs of Drosophila might be predicted to participate in a similar function. Alternatively, gut fibers expand at the front of peristaltic contractile waves, where MRP-tethering may serve as an “energy dissipater”40 to protect fibers when subjected to potentially injurious stretch. In contrast to such passive activities, MRP may have a more active function and serve a role similar to that of the MLC2 extension to enhance the oscillatory work performance of the fiber.31 The genetic and molecular tools available in Drosophila are likely to provide the tools required to determine the nature and role of this novel class of contractile proteins.
Contribution of MRP to Sarcomere Organization
Materials and Methods Drosophila methods All strains were grown on standard molasses, cornmeal and yeast medium at room temperature.41 Strain yw1118 was used as the wild-type reference strain in all experiments. For analysis of muscle structure in animals that are null for MHC, flies carrying the Mhc 1 mutation,15 which removes a 1 kb interval at the 5-prime end of the Mhc gene and is homozygous lethal, were crossed to a second balancer chromosome that is wild-type for MHC, but contains a copy of the gene encoding the green fluorescent protein (GFP) under the control of the Actin5C promoter (CyO-pAct-GFP; Bloomington Stock Center). Embryos from this strain that are homozygous for this mutation can be identified by the absence of GFP expression (see below). For animals expressing GFP under the regulation of the Mrp promoter, the coding region of enhanced GFP (EGFP; Clontech), was amplified by PCR used to exactly replace the lacZ coding sequence from the gD1142 construct.8 For examination of embryonic musculature in wildtype flies, eggs were collected for three hours at 22 8C onto molasses/agar plates and embryos were aged to Stage 18 to complete embryonic muscle formation and maturation.42 Embryos were processed for immunostaining by dechorionating with bleach and fixing for 15 minutes in a solution of TBS-formaldehyde (100 mM Tris (pH 7.5), 150 mM NaCl, 0.01 M EGTA, 3.7% formaldehyde), that was mixed in equal parts with heptane. The aqueous solution was then replaced with methanol to remove the vitelline membrane, the heptane was removed and embryos were washed with several changes of methanol and stored at 220 8C. Embryos genetically null for MHC were selected from the progeny of flies of the Mhc 1/Cyo Act-GFP genotype by placing them in 1% (w/v) agarose after dechorionation and viewing them in a Leica MZ12 dissecting scope equipped for epifluorescence. Immunostaining For immunostaining, embryos were rehydrated in a buffer containing 100 mM Tris (pH 7.5), 150 mM NaCl, 0.1% Triton X-100 (TBX), incubated in TBX þ 2% (w/v) non-fat dried milk (blocking) solution for 15 minutes and in the primary antibody (see below) for one hour. This was followed by several washes in TBX and incubation with the secondary antibody in TBX þ 2% milk for one hour. Embryos were washed extensively with TBX and viewed directly for fluorescently labeled secondary antibodies or developed colorimetrically for alkaline phosphatase conjugated secondary antibodies using AP staining buffer (0.1 M Tris (pH 8.5), 0.1 M NaCl, 0.05 M MgCl, 0.1% Tween 20) supplemented with 0.17 mg/ml 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and 0.33 mg/ml nitro blue tetrazolium (NBT). AP-stained embryos were cleared in TBX solution containing 70% (v/v) glycerol and observed with DIC or bright-field optics using a Leica DMRe research microscope. Fluorescently labeled embryos were observed either under epifluorescence illumination using a Leica DMRe microscope and a Princeton Instruments cooled CCD camera, or with a Zeiss LSM 150 confocal microscope. The digitized images were processed using Openlab (Improvision) and/or Photoshop (Adobe) software.
1089
Contribution of MRP to Sarcomere Organization
The following primary antibodies were used: rabbit polyclonal anti-MHC rod domain7 diluted 1:1000; and rabbit anti-GFP antibodies, diluted 1:200. Secondary antibodies were goat anti-rabbit conjugated to Texas red (Cappel Research Products, Durham, NC), diluted 1:100; and alkaline phosphatase conjugated goat anti-rabbit (Vector Labs), diluted 1:1000. Immunoblot analysis The ratio of MHC to MRP in individual DFM muscles was determined by immunoblot analysis. Adult female flies were either dehydrated in 100% EtOH or acetone at room temperature, and then dissected using fine forceps and tungsten needles to select individual DFMs. For direct visualization of MRP-expressing muscles, lacZ from the gD1142 construct8 was exactly replaced with the coding sequence for the green fluorescence protein and used to transform wild-type embryos. Muscles expressing the MRP-GFP reporter (Figure 1) were dissected under UV illumination using a Leica MZFL dissecting scope. Eight to ten muscles of each type were selected and transferred to TBX solution containing proteinase inhibitors (Sigma) at 4 8C. Muscle containing samples were centrifuged at 15,000g and pellets dissolved in sample buffer,43 boiled for ten minutes and proteins were separated on SDS-8% PAGE. Resolved proteins were transferred to nitrocellulose and MHC and MRP were simultaneously labeled using polyclonal antibodies directed against the MHC/MRP rod domain7 (see also immunolabeling) diluted at 1:5000. Proteins were detected with HRP-conjugated anti-rabbit secondary antibodies (Vector) and blots were developed using chemiluminescence detection (Supersignal; Pierce). Films were digitized and MHC and MRP levels were quantified using the Scion Image software package. Ratios from a typical experiment are presented in Figure 1. Due to the extreme differences in expression levels, it was not possible to record signals from both MRP and MHC within the linear range of the film using this method, and MRP ratios are potentially underestimated. Electron microscopy For analysis of adult muscles, CO2 anaesthetized animals were glued to the bottom of Petri dishes with nitrocellulose. Thoraces were opened in rigor solution (75 mM potassium acetate, 5 mM magnesium acetate, 5 mM EGTA, 15 mM potassium phosphate (pH 7.0), 30% glycerol, 0.05 –0.1% Triton X-100), dissected at room temperature to remove the majority of the IFM, and stored overnight at 4 8C. Following several washes with 0.1 M sodium cacodylate buffer, muscles were treated with 0.2% tannic acid in cacodylate buffer for one hour at room temperature, then fixed in 3.5% glutaraldehyde and 0.2% tannic acid in buffer for one hour. After further cleaning of unwanted muscles, the thoraces were kept in cacodylate buffer overnight at 4 8C. Stage 18 wild-type and mutant embryos were prepared for electron microscopy by dechorionation as for immunolabeling followed by hand devitellinization and fixation in glutaraldehyde. Embryos and adult thoraces were post-fixed in 2% osmium tetraoxide in buffer (one hour); washed in water (3 £ for five minutes), stained en bloc with saturated aqueous uranyl acetate (at room temperature overnight); dehydrated in acetone and infiltrated in
Epon. Prior to Epon polymerization, individual DFMs were dissected out of the thorax using fine tungsten needles. Ultrathin sections were stained in uranyl acetate and in a mixture of lead salts44 and observed in a Philips 410 EM.
Acknowledgements Supported by NIH grant 5-PO1-HL15835 to the Pennsylvania Muscle Institute. Our thanks to N. Glaser and M. Lai for technical assistance and K. Kozopas (Univ. Alabama) for indicating the correct DFM nomenclature.
References 1. Huxley, H. E. (1963). Electron microscope studies on the structure of natural and synthetic protein filaments from striated muscle. J. Mol. Biol. 7, 281 –306. 2. Atkinson, S. J. & Stewart, M. (1991). Molecular basis of myosin assembly: coiled-coil interactions and the role of charge periodicities. J. Cell Sci.—Suppl. 14, 7 – 10. 3. Sohn, R. L., Vikstrom, K. L., Strauss, M., Cohen, C., Szent-Gyorgyi, A. G. & Leinwand, L. A. (1997). A 29 residue region of the sarcomeric myosin rod is necessary for filament formation. J. Mol. Biol. 266, 317 –330. 4. George, E. L., Ober, M. B. & Emerson, C. P., Jr (1989). Functional domains of the Drosophila melanogaster muscle myosin heavy-chain gene are encoded by alternatively spliced exons. Mol. Cell. Biol. 9, 2957 –2974. 5. Hastings, G. A. & Emerson, C. P., Jr (1991). Myosin functional domains encoded by alternative exons are expressed in specific thoracic muscles of Drosophila. J. Cell Biol. 114, 263– 276. 6. Bernstein, S. I., Mogami, K., Donady, J. J. & Emerson, C. P., Jr (1983). Drosophila muscle myosin heavy chain encoded by a single gene in a cluster of muscle mutations. Nature, 302, 393 –397. 7. Miedema, K., Hanske, M., Akhmanova, A., Bindels, P. & Hennig, W. (1995). Minor-myosin isoform synthesized preferentially in Drosophila testis is encoded by the muscle myosin heavy chain gene. Mech. Dev. 51, 67 –81. 8. Standiford, D. M., Davis, M. B., Miedema, K., Franzini-Armstrong, C. & Emerson, C. P., Jr (1997). Myosin rod protein: a novel thick filament component of Drosophila muscle. J. Mol. Biol. 265, 40 – 55. 9. Toffenetti, J., Mischke, D. & Pardue, M. L. (1987). Isolation and characterization of the gene for myosin light chain two of Drosophila melanogaster. J. Cell Biol. 104, 19 –28. 10. Sanbe, A., Gulick, J., Fewell, J. & Robbins, J. (2001). Examining the in vivo role of the amino terminus of the essential myosin light chain. J. Biol. Chem. 276, 32682 – 32686. 11. Schaub, M. C., Hefti, M. A., Zuellig, R. A. & Morano, I. (1998). Modulation of contractility in human cardiac hypertrophy by myosin essential light chain isoforms. Cardiovasc. Res. 37, 381– 404. 12. Yamada, A., Yoshio, M., Oiwa, K. & Nyitray, L.
1090
13.
14.
15.
16. 17.
18.
19.
20.
21.
22. 23.
24.
25.
26.
27.
(2000). Catchin, a novel protein in molluscan catch muscles, is produced by alternative splicing from the myosin heavy chain gene. J. Mol. Biol. 295, 169– 178. Shelud’ko, N., Permjakova, T., Tuturova, K., Neverkina, O. & Drozdov, A. (2001). Myorod, a thick filament protein in molluscan smooth muscles: isolation, polymerization and interaction with myosin. J. Muscle Res. Cell Motil. 22, 91 – 100. Bate, M. (1993). The mesoderm and its derivatives. In The Development of Drosophila melanogaster (Bate, M. & Martinez-Arias, A., eds), pp. 1013– 1090, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. O’Donnell, P. T. & Bernstein, S. I. (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. Elfvin, M., Levine, R. J. & Dewey, M. M. (1976). Paramyosin in invertebrate muscles. I. Identification and localization. J. Cell Biol. 71, 261– 272. Cohen, C., Szent-Gyorgyi, A. G. & Kendrick-Jones, J. (1971). Paramyosin and the filaments of molluscan “catch” muscles. I. Paramyosin: structure and assembly. J. Mol. Biol. 56, 223– 227. Szent-Gyorgyi, A. G., Cohen, C. & Kendrick-Jones, J. (1971). Paramyosin and the filaments of molluscan “catch” muscles. II. Native filaments: isolation and characterization. J. Mol. Biol. 56, 239– 258. Maroto, M., Arredondo, J., Goulding, D., Marco, R., Bullard, B. & Cervera, M. (1996). Drosophila paramyosin/miniparamyosin gene products show a large diversity in quantity, localization, and isoform pattern: a possible role in muscle maturation and function. J. Cell Biol. 134, 81 – 92. Maroto, M., Arredondo, J. J., San Roman, M., Marco, R. & Cervera, M. (1995). Analysis of the paramyosin/miniparamyosin gene. Miniparamyosin is an independently transcribed, distinct paramyosin isoform, widely distributed in invertebrates. J. Biol. Chem. 270, 4375– 4382. Matsubara, I., Goldman, Y. E. & Simmons, R. M. (1984). Changes in the lateral filament spacing of skinned muscle fibres when cross-bridges attach. J. Mol. Biol. 173, 15 – 33. Fyrberg, E. & Beall, C. (1990). Genetic approaches to myofibril form and function in Drosophila. Trends Genet. 6, 126– 131. Holtzer, H., Hijikata, T., Lin, Z. X., Zhang, Z. Q., Holtzer, S., Protasi, F., Franzini-Armstrong, C. & Sweeney, H. L. (1997). Independent assembly of 1.6 microns long bipolar MHC filaments and I-Z-I bodies. Cell Struct. Funct. 22, 83 – 93. Cripps, R. M., Suggs, J. A. & Bernstein, S. I. (1999). Assembly of thick filaments and myofibrils occurs in the absence of the myosin head. EMBO J. 18, 1793– 1804. Antin, P. B., Forry-Schaudies, S., Friedman, T. M., Tapscott, S. J. & Holtzer, H. (1981). Taxol induces postmitotic myoblasts to assemble interdigitating microtubule-myosin arrays that exclude actin filaments. J. Cell Biol. 90, 300–308. Levine, R. J., Elfvin, M., Dewey, M. M. & Walcott, B. (1976). Paramyosin in invertebrate muscles. II. Content in relation to structure and function. J. Cell Biol. 71, 273– 279. Castellani, L., Vibert, P. & Cohen, C. (1983). Structure
Contribution of MRP to Sarcomere Organization
28.
29. 30.
31.
32.
33.
34.
35.
36.
37. 38.
39. 40.
41. 42. 43. 44.
of myosin/paramyosin filaments from a molluscan smooth muscle. J. Mol. Biol. 167, 853– 872. Castellani, L. & Vibert, P. (1992). Location of paramyosin in relation to the subfilaments within the thick filaments of scallop striated muscle. J. Muscle Res. Cell Motil. 13, 174– 182. Parker, V. P., Falkenthal, S. & Davidson, N. (1985). Characterization of the myosin light-chain-2 gene of Drosophila melanogaster. Mol. Cell. Biol. 5, 3058– 3068. Moore, J. R., Dickinson, M. H., Vigoreaux, J. O. & Maughan, D. W. (2000). The effect of removing the N-terminal extension of the Drosophila myosin regulatory light chain upon flight ability and the contractile dynamics of indirect flight muscle. Biophys. J. 78, 1431– 1440. Irving, T., Bhattacharya, S., Tesic, I., Moore, J., Farman, G., Simcox, A. et al. (2001). Changes in myofibrillar structure and function produced by N-terminal deletion of the regulatory light chain in Drosophila. J. Muscle Res. Cell Motil. 22, 675– 683. Henry, G. D., Winstanley, M. A., Dalgarno, D. C., Scott, G. M., Levine, B. A. & Trayer, I. P. (1985). Characterization of the actin-binding site on the alkali light chain of myosin. Biochim. Biophys. Acta, 830, 233– 243. Timson, D. J., Trayer, H. R., Smith, K. J. & Trayer, I. P. (1999). Size and charge requirements for kinetic modulation and actin binding by alkali 1-type myosin essential light chains. J. Biol. Chem. 274, 18271– 18277. Trayer, I. P., Trayer, H. R. & Levine, B. A. (1987). Evidence that the N-terminal region of A1-light chain of myosin interacts directly with the C-terminal region of actin. A proton magnetic resonance study. Eur. J. Biochem. 164, 259– 266. Hayashibara, T. & Miyanishi, T. (1994). Binding of the amino-terminal region of myosin alkali 1 light chain to actin and its effect on actin – myosin interaction. Biochemistry, 33, 12821 –12827. Poetter, K., Jiang, H., Hassanzadeh, S., Master, S. R., Chang, A., Dalakas, M. C. et al. (1996). Mutations in either the essential or regulatory light chains of myosin are associated with a rare myopathy in human heart and skeletal muscle. Nature Genet. 13, 63 – 69. Tu, M. S. & Dickinson, M. (1994). Modulation of negative work output from a steering muscle of the blowfly Calliphora vicina. J. Expt. Biol. 192, 207– 224. Tu, M. S. & Dickinson, M. H. (1996). The control of wing kinematics by two steering muscles of the blowfly (Calliphora vicina). J. Comp. Physiol. [A], 178, 813– 830. Dickinson, M. H. & Tu, M. S. (1997). The function of Dipteran flight muscle. Comp. Biochem. Physiol. 116, 223– 238. Maughan, D., Moore, J., Vigoreaux, J., Barnes, B. & Mulieri, L. A. (1998). Work production and work absorption in muscle strips from vertebrate cardiac and insect flight muscle fibers. Advan. Expt. Med. Biol. 453, 471– 480. Ashburner, M. (1989). Drosophila: A Laboratory Handbook, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Campos-Ortega, J. (1997). The Embryonic Development of Drosophila melanogaster, Springer, New York. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680– 685. Sato, T. (1968). A modified method for lead staining of thin sections. J. Electron Microsc. 17, 158– 159.
Contribution of MRP to Sarcomere Organization
45. Miller, A. (1950). The internal anatomy and histology of the imago of Drosophila melanogaster. In The Biology of Drosophila (Demerec, M., ed.), pp. 420– 534, Wiley, New York.
1091
46. Reedy, M. K. (1968). Ultrastructure of insect flight muscle. I. Screw sense and structural grouping in the rigor cross-bridge lattice. J. Mol. Biol. 31, 155 –176.
Edited by J. Karn (Received 25 March 2003; received in revised form 20 June 2003; accepted 24 June 2003)
J. Mol. Biol. (2003) 331, 1077–1091
Contribution of Myosin Rod Protein to the Structural Organization of Adult and Embryonic Muscles in Drosophila Erzse´bet Polya´k, David M. Standiford*, Vladimir Yakopson Charles P. Emerson Jr and Clara Franzini-Armstrong Department of Cell and Developmental Biology University of Pennsylvania 245 Anatomy-Chemistry Building, 36th Street and Hamilton Walk, Philadelphia PA 19104-6058, USA
Myosin rod protein (MRP) is a naturally occurring 155 kDa protein in Drosophila that includes the myosin heavy chain (MHC) rod domain, but contains a unique 77 amino acid residue N-terminal region that replaces the motor and light chain-binding domains of S1. MRP is a major component of myofilaments in certain direct flight muscles (DFMs) and it is present in other somatic, cardiac and visceral muscles in adults, larvae and embryos, where it is coexpressed and polymerized into thick filaments along with MHC. DFM49 has a relatively high content of MRP, and is characterized by an unusually disordered myofibrillar ultrastructure, which has been attributed to lack of cross-bridges in the filament regions containing MRP. Here, we characterize in detail the structural organization of myofibrils in adult and embryonic Drosophila muscles containing various MRP/MHC ratios and in embryos carrying a null mutation for the single MHC gene. We examined MRP in embryonic body wall and intestinal muscles as well as in DFMs with consistent findings. In DFMs numbers 49, 53 and 55, MRP is expressed at a high level relative to MHC and is associated with disorder in the positioning of thin filaments relative to thick filaments in the areas of overlap. Embryos that express MRP in the absence of MHC form thick filaments that participate in the assembly of sarcomeres, suggesting that myofibrillogenesis does not depend on strong myosin –actin interactions. Further, although thick filaments are not well ordered, the relative positioning of thin filaments is fairly regular in MRP-only containing sarcomeres, confirming the hypothesis that the observed disorder in MRP/MHC containing wild-type muscles is due to the combined action between the functional behavior of MRP and MHC myosin heads. Our findings support the conclusion that MRP has an active function to modulate the contractile activity of muscles in which it is expressed. q 2003 Elsevier Ltd. All rights reserved.
*Corresponding author
Keywords: myosin rod protein; myosin; Drosophila; myofibrillogenesis; sarcomere
Introduction Myosin heavy chain (MHC), the motor protein of muscle thick filaments, has two distinct domains: the head and the tail. The head has a globular region with ATPase and actin-binding functions Abbreviations used: MRP, myosin rod protein; MHC, myosin heavy chain; DFM, direct flight muscles; LMM, light meromyosin; GFP, green fluorescent protein. E-mail address of the corresponding author: [email protected]
and extends into a single alpha-helical segment with binding sites for the two light chains. Cyclic interactions of the MHC head, or myosin crossbridge, with the actin filament, coupled to hydrolysis of ATP, results in muscle contraction by a sliding filament mechanism, producing either tension or shortening. The myosin heads are seen by electron microscopy as “cross-bridges” projecting from the surface of the thick filament and attaching to the thin filaments when the muscle is fixed either during active contraction or in the rigor state, in which all cross-bridges are attached.
0022-2836/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
1078
The tail is fully alpha-helical; it forms a coiled-coil with a second MHC, whose light meromyosin (LMM) region is responsible for the spontaneous polymerization of myosin into filaments at physiologic ionic strength.1 The properties of mutant LMMs suggest that charge distributions in the coiled-coil may influence the detailed axial stagger of the interacting molecules, thus determining the axial periodicities of myosin heads over the surface of the thick filaments.2 A specific sequence near the C terminus of the coiled-coil region of the MHC is necessary for the formation of ordered LMM paracrystals and thus presumably for the formation of the thick filament backbone.3 In Drosophila melanogaster, a number of musclespecific isoforms are generated through alternative mRNA splicing from a single MHC gene (DMhc).4 – 6 However, within the DMhc gene, a second transcriptional unit encodes a 155 kDa protein termed the myosin rod protein (MRP).7,8 The MRP transcript initiates within DMhc intron 12 and encodes a 77 amino acid residue N-terminal domain that is specific to MRP before continuing in-frame through exon 13 and the remainder of the exons that encode the rod domain of MHC. Thus, MRP has the same rod domain as MHC, but the motor and light chain-binding regions are replaced by an MRP-specific N-terminal domain. The proline-rich MRP N-terminal domain is similar to the N-terminal extensions of MLC2 in Drosophila7,9 and the essential light chain (ELC) of vertebrate cardiac myosins,10,11 which have been proposed to mediate thick and thin filament interactions through the formation of molecular tethers.10 Myosin splice variants with a truncated N-terminal domain, similar to MRP, are also found in several smooth muscles of molluscs,12,13 suggesting that such a role might be broadly important for contractile function. Previous work in Drosophila showed that MRP is widely expressed in somatic muscles but is also found in cardiac and visceral muscles and in nonmuscle tissues.7,8 MRP was found to be particularly abundant in a set of direct flight muscles (DFM) in the adult thorax, and an analysis of these muscles showed that MRP associates with MHC in the thick filament as a homodimer and can constitute a significant fraction of the thick filament protein relative to MHC. The inclusion of MRP at a high level into the thick filament resulted in the detectable loss of MHC-mediated cross-bridges in the rigor filament and is believed to contribute to several distinct ultrastructural features observed in these muscles, including the presence of longitudinally disordered thick and thin filament and a remarkable “wandering” behavior of thin filaments. However, the extent of these phenotypes was not uniform among the various MRPexpressing muscles, suggesting that MRP might have muscle-specific effects on sarcomeric ultrastructure. Here, we extend the previous observations through characterization of the sarcomeric organiz-
Contribution of MRP to Sarcomere Organization
ation in muscles that contain different levels of MRP. In addition, we address the relative contribution of MRP and MHC to the muscle phenotype through the analysis of myofibers in mutant animals that express MRP, but not MHC.
Results The content of MRP in direct flight muscles The analysis of a lacZ reporter gene under the regulation of the Mrp promoter showed previously that a subset of DFMs in the adult thorax express MRP, with the most prominent of these being DFM49, 53 and 55 (previously identified as DFM 54).8 MRP-specific antibodies used for immunolabeling showed some differences in labeling intensity among these muscles (not shown), suggesting varying levels of MRP. These levels appeared highest in DFM49 and were consistent with this muscle having the greatest degree of ultrastructural “disorder”. To directly determine the relative level of MRP expression in the DFM muscle, individual muscles were dissected and subject to quantitative immunoanalysis to establish the ratio of MRP to MHC in each muscle (Figure 1). These data show that the level of MRP is high in DFM49, 53 and 55, where it exceeds an 8:1 ratio with MHC in all cases. Thus, in each of these muscles MRP is expected to be a major thick filament protein. DFM51 and 54 and the tergal depressor of the trochanter (TDT) were also assayed and, consistent with our earlier data, were found not to express MRP. Structure of the sarcomere in MRP-expressing DFMs of wild-type adult fly DFM51 and 54, which do not express MRP, have a very similar structure. In particular, in the overlap region of the A band thin and thick filaments have an ordered parallel arrangement (Figure 2(A), (C), and (D)). DFM53 and 55, which do express MRP, also have structural features common to each other, but differ from DFM51 and 54. The spacing between thick filaments is variable and thin filaments wander back and forth between thick filaments, sometimes approaching one and sometimes its neighbor (Figure 2(B) and (E) – (H)). The disorder is most apparent in DFM55, which has a high relative content of MRP and resembles the highly disordered DFM49, which had been selected for an initial study8 (Figure 2(H)). When muscles that do not express Mrp are fixed in rigor, cross-bridges are connected to the actin filaments at evenly spaced target zones with a marked periodicity. Each target zone is occupied by a single or a doublet of cross-bridges (Figure 2(C) and (D)). In MRP-expressing muscles, cross-bridges are fewer and less periodically arranged. In DFM53 the content of cross-bridges varies between small domains that have an almost
Contribution of MRP to Sarcomere Organization
1079
Figure 1. (A) The myosin rod protein (Mrp) gene is encoded by a separate transcription unit embedded within the Drosophila 36B muscle myosin heavy chain (Mhc) gene. The Mrp transcript initiates in Mhc intron 12 and encodes a unique 77-residue N-terminal peptide that is fused in-frame with the rod domain of the Mhc gene. Thus, the MRP N terminus replaces the MHC motor and light chain-binding domain (LCB), but is identical with MHC in the filament forming rod domain. (B) Cartoon of adult lateral thorax showing the position of the direct flight muscles (DFMs; nomenclature according to Millar.45) Muscles-expressing MRP are filled. A transgenic GFP-reporter construct using the Mrp promoter domain was used to identify and isolate individual MRP-expressing DFM49, 53 and 55 from adult flies. (C) Immunoblot analysis of MHC and MRP (upper and lower bands, respectively) isolated from individual DFMs and the tergal depressor of the trochanter (TDT) muscle, showing that MRP is heavily expressed in DFM49, 53 and 55, while it is not detected in DFM51, 54 or the TDT. Quantification of the bands from MRP-expressing muscles indicates that the MHC-to-MRP ratio is at least 1:8 in each muscle.
full complement of cross-bridges and others where cross-bridges are mostly missing (Figure 2(E) –(G)). DFM55 has no domains with ordered cross-bridge arrangement (Figure 2(H)). Measurements of the average distance between actin target zones shows that the spacing is shorter and considerably more uniform in DFM51 and 54, which have no MRP, than in DFM53, 49 and 55, which contain MRP (Table 1). It should be noted that the positioning of cross-bridge occupied target zones in the highly disordered DFM is less accurately measurable than in DFM51 and 54. Conversely, the occupancy of actin filaments by cross-bridges is higher in DFM51 and 54 than in DFM53 (Table 1). The disposition of thin filaments is directly related to the frequency of cross-bridges; in crossbridge-rich regions the filaments are straight and parallel to the thick filaments, while where crossbridges are missing the filaments are wavy and irregular. Based on previous observations,8 is it logical to assume that regions with lower frequency of cross-bridges are those in which a
portion of the thick filament backbone is constituted by MRP. On this basis, the structural data provided here indicate that sarcomere organization is strongly dependent on the MRP content and that the content may vary within domains of the same muscle fiber and even within the same sarcomere (Figure 2(G)). Observations on cross-sections of DFM muscles are in agreement with predictions from longitudinal sections, in the sense that the ordered arrangement of thin and thick filaments is affected by the presence of MRP. In muscles that lack MRP, thick filaments form a hexagonal array with only occasional discontinuities in the pattern (Figure 3(A)). The thin filaments are precisely disposed in an equidistant (dyadic) position between two adjacent thick filaments. Either a single crossbridge or a doublet of cross-bridges occupies each dyadic position. The number of thin filaments surrounding each thick filament varies between seven and ten in DFM51 (Figure 3(B)) and six to eight in DFM54 (image not shown). The variability in the thin/thick filament ratio gives an appearance of
1080
Contribution of MRP to Sarcomere Organization
Figure 2. Sarcomere structure of DFMs. Comparison of the non-MRP-containing DFM51 (A) and the MRP-containing DFM53 (B), showing considerably higher disorder in the latter in the disposition and spacing of both thin and thick filaments. At higher magnification, muscles containing no MRP (C) and (D) have evenly spaced rigor crossbridges and the thin filaments are straight and equidistant from the two adjacent thick filaments. In muscles with MRP content (E) – (H), the frequency of cross-bridges varies in different areas and the spacing between thin and thick filaments is highly variable.
slight disorder to the cross-sectional pattern of DFM muscles that do not contain MRP. However, close examination (Figure 3(B)) shows that the thin filaments are approximately equidistant between thick filaments, suggesting equal interaction between a thin filament and two adjacent thick filaments. Thick and thin filament patterns are not as regular in MRP-expressing muscles. First, the hexagonal array of thick filaments is distorted and this seems to be dependent on MRP content: DFM53, a muscle with an intermediate level of MRP, shows more distortion than DFM51. Second, even in muscles where the thin/thick filament ratio is approximately the same, the thin to
thick filament distance varies more strongly in MRP-expressing muscles than in those where MRP is lacking (Figure 3(C) and (D)). In particular, some thin filaments are very close to the surface of thick filaments (arrows, Figure 3(D) and (F)) and some are at large distances, two positions that are not seen in the absence of MRP (Figure 3(B)). DFM49 (see Ref. 8) and 55 (Figure 3(E) and (F)) show the highest degree of disorder. The structural characteristics for DFM49 have been detailed by Standiford et al.8 In DFM55 the thick filaments do not form a hexagonal array and thin filaments are not only variably positioned relative to the surface of the thick filaments, but also are very unevenly
1081
Contribution of MRP to Sarcomere Organization
Table 1. Measurements of DFM cross-bridge distances and filament densities Distance between cross-bridgeoccupied target zones (average (range) nm) DFM51 DFM54 DFM53 DFM49 DFM55
32.6 ^ 2.6(27.9– 38.5) n ¼ 76c 32.8 ^ 3.0(27.5– 38.7) n ¼ 48c 43.6 ^ 12.2(25.6– 103.2) n ¼ 151c 43.5 ^ 14.6(25.9–91.0)d n ¼ 27c 45.2 ^ 15.7(20.6–91.0)d n ¼ 56c
Cross-bridge no./100 nm (average (range)a)
Thick filament density (average no. (mm2)b)
Thin/thick filament ratio (average no. b)
4.7 ^ 0.7(3.1– 5.9) n ¼ 55 4.7 ^ 0.6(3.7– 6.3) n ¼ 36 3.2 ^ 0.9(1.7– 5.2) n ¼ 61
711 ^ 38.5 n ¼ 21 658 ^ 53 n ¼ 21 823 ^ 64 n ¼ 21 994 ^ 107 n ¼ 21 908 ^ 68 n ¼ 21
3.1 ^ 0.2 n ¼ 21 2.4 ^ 0.2 n ¼ 21 1.8 ^ 0.2 n ¼ 21 1.8 ^ 0.2 n ¼ 21 1.6 ^ 0.3 n ¼ 21
a
All cross-bridges occupying actin segment were counted and referred to the length of the filament. Mean ^ SD measured in 21 different 4.7 £ 1022 mm2 areas for each muscle. Data are not corrected for shrinkage and compression artifacts. Average distance in the DFM51 and 54 is expected to be 38.7 nm, which is the axial repeat in rigor IFM.46 c Mean ^ SD, n ¼ number of actin segments measured, from three flies for each measurement. d In DFM49 and DFM55 there is a very large variation in the distances between cross-bridge-occupied sites along filament segments and there is usually very little overlap in the fixed muscles. This limited the available sites for measurements and the data are less accurate. b
Figure 3. Comparison of rigor DFM muscles in transverse sections. In the non-MRP containing DFM51 (A) and (B) the thick filaments are in fairly regular hexagonal arrangement. Seven to ten thin filaments surround each thick filament. Each dyadic position along a line joining the center of two adjacent thick filaments is occupied by either one or two symmetrically disposed thin filaments that are equidistant from the two nearest thick filaments. In the MRP-containing DFM53 (C) and (D), the hexagonal arrangement of thick filaments is not as regular. Four to seven thin filaments surround each thick filament; their position is not always symmetric relative to the dyadic position; and their distance from the thick filament surface is variable. Some filaments actually are very close to the thick filament’s surface (arrows). In DFM55 (E) and (F), which have a higher amount of MRP, the thick filaments are quite disordered; two to eight thin filaments surround each thick filament and the thin –thick distances are even more variable than in DFM53.
1082
distributed: some areas totally lack thin filaments so that adjacent thick filaments are very close to each other. Overall, the ratio of thin to thick filaments is reduced compared to DFMs that do not express MRP (Table 1) and the complement of thin filaments serving each thick filament profile varies between 2 and 8. The thick filament density for the different DFM muscles was also measured as the number of thick filaments per unit crosssectional area (Table 1). This analysis showed a somewhat higher density of thick filaments in MRP-expressing muscles, suggesting that the inclusion of this protein can effect filament packing in the DFMs. Structure of body wall and intestinal muscles in wild-type and Mhc 1 mutant embryos The above results suggest that the level of MRP relative to MHC can have a differential, dosedependent effect on the structure of the sarcomere. In order to examine this further, we studied the embryonal body wall muscles from wild-type fly, that express both MHC and MRP, and from genetically mutant embryos that express MRP, but not MHC (Mhc 1/Mhc 1) in order to determine the ability of MRP alone to construct a thick filament and to assess the affect that this might have on sarcomeric structure. In Drosophila embryos the body wall muscles are segmentally organized, with each body segment containing about 30 different muscle fibers.14 Of these, approximately half express MRP.8 Although animals that are null for MHC die at hatching due to a failure to break from the egg membranes, muscle development in Mhc 1 homozygotes is essentially complete and thus can be examined for the contribution of MRP alone to sarcomeric structure.
Contribution of MRP to Sarcomere Organization
In order to investigate the results of MRP expression on sarcomeric structure in the absence of myosin, we established a line that contained the Mhc 1 null allele in combination with a balancer second chromosome that contained a green fluorescent protein (GFP) gene under the regulation of the Act5C promoter for expression in the early embryo. Embryos that are homozygous for Mhc 1 can then be selected for the absence of GFP staining under the fluorescence dissecting microscope. As seen in Figure 4(A) and (B), immunostaining of wild-type embryos with antibodies directed against the rod domain of myosin/MRP shows the organization of the embryonic musculature. Western blot of total protein extract prepared from stage 18 embryos (B, inset) shows that overall MRP is expressed at much lower level than MHC. However, the MRP content of individual groups of muscles varies. In MHC-null embryos (Figure 4(C)) some muscles are well-delineated by the anti-myosin rod antibody, indicating a relatively high level of MRP expression. The fibers in the MHC-null embryos are less ordered and shorter than those in wild-type. These observations are consistent with the data presented by O’Donnell & Bernstein,15 who previously noted the existence of thick filaments in Mhc 1 homozygotes. Ultrastructure of embryonic body wall muscles from wild-type and MHC-null embryos Thin sections cut transversely to the long axis of the embryo were used to examine the ultrastructure of the musculature in wild-type and mutant animals. In these studies, direct comparisons between specific body wall muscles were not made due to the difficulty in identifying individual
Figure 4. Immunoanalysis of muscle patterns in Drosophila embryos. (A) Wild-type embryonic muscles are outlined by an anti-MHC rod antibody and peroxidase-labeled secondary antibody. The posterior end is at right and dorsal is up. Note variable orientations of small bundles of cells (B) and (C). Fluorescent immunolabeling of wild-type (B) and MHC-null (C) embryos using an antibody against the rod portion of the molecule. Muscle fibers in the MHC-null body wall express readily detectable amounts of MRP, but in general the fibers are less organized and shorter. (B, inset) Western blot of total embryo extract using anti-MHC rod antibody shows that MRP (lower band, single arrow) is present in addition to MHC (upper band, double arrow). The blot combines all embryonic muscles, some of which do not contain any MRP. (B) and (C) are at the same magnification.
Contribution of MRP to Sarcomere Organization
muscles in the sectioned embryos. However, all observations were made at the level of abdominal segments 2 – 7, where internal markers such as gut profiles and the ventral nerve chord could be used as landmarks for orientation and for confirming approximate segment level. Due to the variable orientations of the muscle, transverse and longitudinal views of the muscle were obtained in the same sections. In general, the embryonic body wall muscles are small and have a somewhat variable ultrastructure. Some cells have several small diameter myofibrils, others only a single or at most two myofibrils of larger diameter. We have identified the cells with larger myofibrils as belonging to A2 – A7, but the small myofibril fibers may be from more anterior or posterior segments. Solid core thick filaments of uniform diameter are arranged in a hexagonal pattern that is quite ordered in the small myofibrils, but tends to be disordered in the larger ones (Figure 5(A)). Local thin-to-thick filament ratio is variable, with a ring of six/seven to ten thin filaments surrounding each thick filament. Most thin filaments are equidistant between two thick filaments and in the dyadic position (Figure 5(B)), but some thin filaments do not belong to this pattern and are found at odd positions and either close to the surface of the thick filament or at relatively large distances from it (arrows Figure 5(A) and (B)). This occurs more frequently in the myofibrils that have a less regular arrangement of thick filaments. The muscle depicted in Figure 5 is from a dorsal muscle group, which has relatively low MRP expression levels, and lateral muscle groups have similar appearance. The arrangement of thin and thick filaments in these muscles resembles that seen in DFM53, suggesting that the slight degree of disorder observed is due to the presence of MRP. None of the embryonic muscles have the high degree of thin and thick filament disorder detected in the DFM muscles with high content of MRP, such as DFM55 and 49. In the body wall muscles of MHC-null embryos, thick filaments are present and they overlap with thin filaments. The profiles of thin and thick filaments are equally sharp in some regions of each fiber, indicating that the filaments run parallel to each other (Figure 5(D)). The orientation of groups of filaments, however, may change abruptly within the same fiber (upper border of Figure 5(C)). There is no noticeable hexagonal arrangement of thick filaments and the thin/thick filaments ratio is very high. Long rows and large groups of thin filaments are visible (Figure 5(C) and (D), arrowheads and arrows). In longitudinal sections, the thick filaments are fairly parallel to each other, they overlap with thin filaments and are interposed between dense bodies (arrowheads, Figure 5(F)). Thus the filaments are obviously part of sarcomeres, but they have different sizes and may abruptly change direction. There is a large variation from fiber to fiber in
1083
the frequency and size of thick filaments. Figure 5 illustrates one of the fibers with the highest frequency of thick filaments; others have fewer and some have none. In the latter type of cells, the entire cytoplasm is filled by thin filaments. Although we were not able to relate the occurrence of thick filaments to specific Mrp-expressing muscles, their presence in certain fibers of the Mhc 1 embryo is expected to result from the expression of MRP and its assembly around a paramyosin core to form myosin-free thick filaments. Further, variations in the frequency of thick filaments in the muscles of the Mhc 1 embryo is predicted to be directly relative to the level of MRP in individual muscles. One feature of the MHC-null muscles is worth noting: the distance between thin and thick filaments is more constant than in the wild-type muscles containing a mixture of MHC and MRP. A halo of fairly uniform size separates the thin and thick filament profiles. In the wild-type embryonic body muscles, the cross-sectional diameter of thick filaments is fairly uniform (Figure 5(A) and (B)), except for the end regions, where the filaments taper off (not shown). The average thick filament diameter in the area of overlap with thin filaments is 18.9(^ 2.7) nm (mean ^ SD, from 35 measurements in five fibers). In the MHC-null mutants on the other hand, the diameter of the thick filaments is highly variable and may be quite large, measuring between 14 nm and 60 nm (Figures 5(C) and (D), and 6(A)–(D)). In sections parallel with the filament long axis, the larger filaments show the periodic striations characteristic of paramyosin paracrystals (Figure 6(E) and (F)) and of the core of paramyosin-containing thick filaments that have been stripped of their surface coating of myosin.16 – 18 Drosophila muscles normally contain paramyosin.19,20 Therefore, it is likely that the fibers in the MHC-null embryos express approximately normal levels of paramyosin and that large size of the thick filaments results from the high ratio of paramyosin to MRP. Some of the larger filaments may be composed purely of paramyosin. Ultrastructure of gut muscles from wild-type and MHC-null embryos The wild-type gut muscles have a higher thin/ thick filament ratio than adult DFM and embryonic body wall muscles (Figure 7(A) and (B)). A doublet of thin filaments occupies the majority of the dyadic positions, so that a total of 11 –12 thin filaments surround each thick filament. As a result, the overall appearance in cross-section is quite well ordered. Interestingly, even though the wild-type gut muscles contain MRP,8 we do not observe the variations in thin filament positioning that are present in other MRP-containing muscles. The gut muscles of the MHC-null embryo have one major feature in common with the MHC-null body muscles: a higher thin/thick filament ratio than in
1084
Contribution of MRP to Sarcomere Organization
Figure 5. EM structure of body wall muscles from wild-type and Mhc 1 mutant embryos. In transverse sections of wild-type muscle (A) and (B) some of the same elements of disorganization seen in the MRP-containing DFMs are visible. In particular, the position of thin filaments is variable in the pattern, so that some are close to the thick filaments (arrow) and others are asymmetrically disposed relative to the adjacent thick filaments (double arrows). Nine to 12 thin filaments surround each thick filament, but with local variations. In the MHC-null mutants (C)– (F),
Contribution of MRP to Sarcomere Organization
1085
Figure 6. Examples of large diameter filaments (A)– (E) in the MHC-null embryonic body wall muscles. These filaments are thought to have a large paramyosin core and a surface array of myosin. Some very large filaments indeed appear as typical, banded paramyosin paracrystals, and may have no myosin on their surface (E) and (F).
the wild-type embryo (Figure 7(C) and (D)). However, differently from the body wall muscles, the MHC-null gut muscles have thick filaments of fairly uniform diameter, and large filaments are rare (Figure 7(C), arrow). One possibility is that this is due to a lower content of paramyosin, or lower ratio of paramyosin to MRP. In addition, the MHC-null gut muscles (Figure 7(D)) have a fairly regular disposition of thin and thick filaments in cross-section and parallel thin and thick filaments in longitudinal section (Figure 7(E)). In these two details they differ from the adult wild-type DFM muscles. It is possible that in the gut muscle of wild-type embryos, MHC to MRP ratio is relatively low so that the absence of MHC in Mhc 1 mutant embryos has less of an affect on muscle structure in the gut than on body wall muscles. Interestingly, gut muscle in the Mhc 1 mutants has a higher degree of order than the MRP-expressing DFMs in the wild-type fly. The implications of this observation are considered in the Discussion. When the distance between the edge of the thick filaments and the edge of the thin filaments were measured in gut muscle, we found that in both wild-type and MHC-null fibers, the distance is the same: 6.7(^ 2.6) nm (n ¼ 33), indicating that the distance between the nearest thin filaments and the thick filaments in the MHC-null gut muscles is not different from the wild-type.
Discussion The MRP is the first example of a naturally occurring protein with the ability to substitute for
myosin in the assembly of the thick filament. MRP clearly contributes to distinct and unexpected structural features in muscles that express it. Here, we have characterized this contribution and have defined three central features of MRP inclusion in sarcomeric assembly and structure. First, we show that the levels of MRP inclusion in normal muscle fibers can exceed that of MHC and that the sarcomeric disorder among flight muscles that express MRP is variable. Second, we determine that MRP alone, in the absence of MHC, can support thick filament and sarcomere formation. Finally, we demonstrate that some characteristics of the disorder in MRP-expressing muscles are dependent on the presence of both MRP and MHC, indicating that these features are a result of the functional interactions between these two components of the thick filament. We find that high levels of MRP content correlate with two structural details of the sarcomere in adult DFM muscles: (1) the degree of order in the hexagonal arrays of filaments is diminished in muscles that have high levels of MRP; and (2) the variability of thin – thick filament distances increases with MRP content. These observations can be described by a model in which MRP is copolymerized with MHC into the thick filament and effectively replaces myosin motor domains with MRP N-terminal domains. An essential concept of the model is that the MRP N-terminal domain can act as a tether, providing a myosin to actin link that is considerably weaker than an active cross-bridge. Although the conformation and length to the MRP N terminus has not been directly determined, this concept is based on the
the thin/thick filament ratio is obviously higher, since clusters of thin filaments are seen between the thick ones. The cross-sectional diameter of the thick filaments is highly variable (C). (E) and (F) show the A-band (E) and I-Z-I regions and parts of A-bands (F) of MHC-null muscles, in longitudinal sections. Typical elements of the sarcomere are present: thin and thick filaments overlap, I-band and interposed dense bodies (arrowheads). However, there is a scarcity of thick filaments and there is some disorder of thin and thick filaments (E).
1086
Contribution of MRP to Sarcomere Organization
Figure 7. Structure of gut muscles from wild-type and MHC-null embryos. Transverse sections show a well-ordered double hexagonal arrangement of thick and thin filaments (A, arrow) in the wild-type embryo (A) and (B). The almost complete complement of 12 thin filaments around each thick filament and the constant thick filament diameter contribute to the symmetry of the system. In the mutant embryos (C) and (D) thick filament diameter is variable, but very large thick filaments, such as shown in (C) and (D) (arrows) are rare. The clusters of thin filament profiles are due to the high actin/myosin ratio in the absence of MHC. (E) A longitudinal section: two A-bands are separated by Z bodies (Z), and the I-band is very short (arrow). Thin and thick filaments are well aligned and are less disordered than in the body wall muscles (Figure 5(E)).
Contribution of MRP to Sarcomere Organization
similarity of this domain with the N-terminal extensions found in vertebrate cardiac light chains, where both structural and functional data exist to support a tethering function.10 Even if the MRP N-terminal domain is tethered to the actin it would be too small to be easily detectable in thin sections, so that areas where MRP is present would appear empty. The disorder in the thick-to-thick and thick-to-thin arrays in MRP-containing muscles is likely to result from the uneven pull between thin and thick filaments due to the uneven distribution of cross-bridges and MRP tethers on the surface of filaments that contain a large proportion of MRP. Rigor crossbridges impose a specific inter-filament spacing in the area of thin and thick filament overlap.21 In insect muscles where MRP is not expressed, the evenly distributed cross-bridges impose a very precise position to the thin filaments, so that the thin filaments stray only minimally from the perfect dyadic position, equidistant between the filaments. In MRP-containing muscles, we assume that regions of strong pull alternate with regions of weak attachments and this would easily distort the filament disposition, as observed. Interestingly, although MRP content is high in DFM49, 53 and 55, we observe that the level of disorder is consistently less in DFM53. This could represent the effects of a slightly greater MHC content, or potentially represents a structural and perhaps functional difference that is distinct to this muscle. If the former is correct, then the extent and frequency of thick filament segments that show gaps in the array of rigor cross-bridges would be expected to increase with higher levels of MRP. Experiments to manipulate the MRP/MHC ratio in muscle fibers will examine this possibility. In MRP-expressing DFMs, we can define two components of this disorder: one is the fact that some thin filaments are located very close to the surface of the thick filament and the other is that some thin filaments seem to freely float away from the thick filaments. Rigor cross-bridges not only strongly hold actin, but also do not allow it to come too close to the thick filament.21 Thus, the short regions of close approximation between thin and thick filaments in muscles that contain MRP may be sites of MRP tethering. The thin filament segments that freely float away from the thick filaments, on the other hand, may have had their MRP tethers pulled off during cross-bridge action, perhaps due to buckling of the thin filament, allowing the filaments to move freely during fixation and embedding. The same uneven distribution of forces that disturbs the symmetry of thin filament disposition around thick filaments would disturb the hexagonal arrangement of thick filaments, as we observe. Muscles in MHC-null embryos give a direct confirmation of the hypothesis that filament lattice disorder in MHC-MRP-containing muscle is due to the combined action of the two molecules. The body wall and gut muscles of MHC-null embryos
1087
have uneven distribution of thick filaments and a high thin-to-thick filament ratio, but show a quite constant spacing between thick and thin filaments. We assume that the equally weak forces of the MRP tethers from two thick filaments interacting with one thin filament in the absence of myosin hold the thin filament in place between them. The expression of headless myosin (MRP) offers the further opportunity to consider two questions in myofibril assembly: what factors are necessary for the assembly of thick filaments and what factors control the formation of a sarcomere with overlapping thin and thick filaments?22 MHC-null muscles contain thick filaments that overlap with thin filaments in the context of fairly well developed sarcomeres. Thus, the tail region of the myosin is sufficient to form filaments and a strong actin – myosin interaction is not a requirement for the basic steps in sarcomere formation and for the overlap of thin and thick filaments in the overlap zone. These conclusions are not surprising, since the LMM region of the myosin molecule has the self-assembly properties that are required to form thick filaments1 and assemblies of either Z line and thin filaments or of aligned thick filaments with M lines (A-bands) can form independently of each other.22,23 Further evidence that actomyosin interactions are not required for sarcomeric formation is the fact that overlap of thin and thick filaments was also seen in a study of the expression of headless myosin molecules in a myosin null background24 and that myosin filaments seem capable of overlapping with microtubules, with which they do not directly interact.25 In addition to its contribution to sarcomere formation, MRP is also predicted to interact with paramyosin in a manner similar to MHC. It is known that paramyosin is contained in the thick filaments of a variety of invertebrate muscles. If the paramyosin –myosin ratio is fairly high, the former constitutes the core of the filament and the latter is helically arranged on the surface. The higher the paramyosin/myosin ratio, the larger is the filament diameter.18,26,27 In muscles with small amounts of paramyosin, the protein is associated with myosin subfilaments and thus it is distributed throughout the thick filament.28 Even though the molecular packing of paramyosin in the filament core is paracrystalline, the fine periodicity due to this arrangement is masked by the myosin on the surface.18 In the body wall and gut muscles of MHC-null embryos, we observe thick filaments of variable diameter with no noticeable periodicity, as well as rarer periodically striped paramyosin crystals. We conclude that MRP interacts with paramyosin in the same manner as MHC and that MRP is most likely to be located at the surface of the thick filaments. Since the overall level of myosin in these muscles is low in the absence of MHC, some paramyosin forms myosin-free paracrystals. It is clear that the inclusion of MRP into the thick
1088
filament is associated with significant structural characteristics in the sarcomere that are not found in fibers that lack MRP. The possibility of MRPmediated tethering interaction between the thick and thin filaments suggests that the inclusion of MRP may also confer significant functional properties to muscle as well. Several other contractile proteins with domains similar to the N terminus of MRP have been shown to mediate thick –thin filament interactions. For instance the MLC2 from Drosophila has an N-terminal extension not found in other regulatory light chains29 that is similar in length and composition to the MRP N terminus. The function of this domain has been studied in the indirect flight muscle (IFM) where its removal from MLC2 leads to a reduction in the dynamic stiffness and elastic modulus in the IFM and an associated reduced power output.30,31 A similar N-terminal extension is found in the A1 isoform of the vertebrate ELC, and has been shown to interact with the actin filament.32 – 35 Functional analysis of this domain has demonstrated its ability to affect myosin kinetics, including MgATPase activity and actin-binding affinities (reviewed by Sanbe et al.10), showing that N-terminal extensions similar to that found on MRP can have regulatory properties. Further, mutations in the N-terminal extension of the cardiac ELC are linked to familial cardiac hypertrophies associated with the loss of stretch activation in the papillary muscles.36 Together, these observations are consistent with N-terminal extensions forming a molecular tether that acts in parallel to the cross-bridge to enhance the resonant properties of the muscle. A diversity of muscles in Drosophila express MRP, ranging from the fast DFMs to the slow muscles of the gut, and these muscles are likely to be connected through physiological parameters that benefit from the inclusion of MRP into the thick filament. As yet, this commonality is not understood, but several possibilities exist and might include the need to generate a fiber with low power, but a high resting tension to resist stretch. The b1 muscle in the blowfly, Calliphora, is equivalent to DFM49 in Drosophila and has been shown to do primarily negative work when stimulated at a frequency of the wing beat.37 – 39 Thick-to-thin filament tethering mediated by MRP in the DFMs of Drosophila might be predicted to participate in a similar function. Alternatively, gut fibers expand at the front of peristaltic contractile waves, where MRP-tethering may serve as an “energy dissipater”40 to protect fibers when subjected to potentially injurious stretch. In contrast to such passive activities, MRP may have a more active function and serve a role similar to that of the MLC2 extension to enhance the oscillatory work performance of the fiber.31 The genetic and molecular tools available in Drosophila are likely to provide the tools required to determine the nature and role of this novel class of contractile proteins.
Contribution of MRP to Sarcomere Organization
Materials and Methods Drosophila methods All strains were grown on standard molasses, cornmeal and yeast medium at room temperature.41 Strain yw1118 was used as the wild-type reference strain in all experiments. For analysis of muscle structure in animals that are null for MHC, flies carrying the Mhc 1 mutation,15 which removes a 1 kb interval at the 5-prime end of the Mhc gene and is homozygous lethal, were crossed to a second balancer chromosome that is wild-type for MHC, but contains a copy of the gene encoding the green fluorescent protein (GFP) under the control of the Actin5C promoter (CyO-pAct-GFP; Bloomington Stock Center). Embryos from this strain that are homozygous for this mutation can be identified by the absence of GFP expression (see below). For animals expressing GFP under the regulation of the Mrp promoter, the coding region of enhanced GFP (EGFP; Clontech), was amplified by PCR used to exactly replace the lacZ coding sequence from the gD1142 construct.8 For examination of embryonic musculature in wildtype flies, eggs were collected for three hours at 22 8C onto molasses/agar plates and embryos were aged to Stage 18 to complete embryonic muscle formation and maturation.42 Embryos were processed for immunostaining by dechorionating with bleach and fixing for 15 minutes in a solution of TBS-formaldehyde (100 mM Tris (pH 7.5), 150 mM NaCl, 0.01 M EGTA, 3.7% formaldehyde), that was mixed in equal parts with heptane. The aqueous solution was then replaced with methanol to remove the vitelline membrane, the heptane was removed and embryos were washed with several changes of methanol and stored at 220 8C. Embryos genetically null for MHC were selected from the progeny of flies of the Mhc 1/Cyo Act-GFP genotype by placing them in 1% (w/v) agarose after dechorionation and viewing them in a Leica MZ12 dissecting scope equipped for epifluorescence. Immunostaining For immunostaining, embryos were rehydrated in a buffer containing 100 mM Tris (pH 7.5), 150 mM NaCl, 0.1% Triton X-100 (TBX), incubated in TBX þ 2% (w/v) non-fat dried milk (blocking) solution for 15 minutes and in the primary antibody (see below) for one hour. This was followed by several washes in TBX and incubation with the secondary antibody in TBX þ 2% milk for one hour. Embryos were washed extensively with TBX and viewed directly for fluorescently labeled secondary antibodies or developed colorimetrically for alkaline phosphatase conjugated secondary antibodies using AP staining buffer (0.1 M Tris (pH 8.5), 0.1 M NaCl, 0.05 M MgCl, 0.1% Tween 20) supplemented with 0.17 mg/ml 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and 0.33 mg/ml nitro blue tetrazolium (NBT). AP-stained embryos were cleared in TBX solution containing 70% (v/v) glycerol and observed with DIC or bright-field optics using a Leica DMRe research microscope. Fluorescently labeled embryos were observed either under epifluorescence illumination using a Leica DMRe microscope and a Princeton Instruments cooled CCD camera, or with a Zeiss LSM 150 confocal microscope. The digitized images were processed using Openlab (Improvision) and/or Photoshop (Adobe) software.
1089
Contribution of MRP to Sarcomere Organization
The following primary antibodies were used: rabbit polyclonal anti-MHC rod domain7 diluted 1:1000; and rabbit anti-GFP antibodies, diluted 1:200. Secondary antibodies were goat anti-rabbit conjugated to Texas red (Cappel Research Products, Durham, NC), diluted 1:100; and alkaline phosphatase conjugated goat anti-rabbit (Vector Labs), diluted 1:1000. Immunoblot analysis The ratio of MHC to MRP in individual DFM muscles was determined by immunoblot analysis. Adult female flies were either dehydrated in 100% EtOH or acetone at room temperature, and then dissected using fine forceps and tungsten needles to select individual DFMs. For direct visualization of MRP-expressing muscles, lacZ from the gD1142 construct8 was exactly replaced with the coding sequence for the green fluorescence protein and used to transform wild-type embryos. Muscles expressing the MRP-GFP reporter (Figure 1) were dissected under UV illumination using a Leica MZFL dissecting scope. Eight to ten muscles of each type were selected and transferred to TBX solution containing proteinase inhibitors (Sigma) at 4 8C. Muscle containing samples were centrifuged at 15,000g and pellets dissolved in sample buffer,43 boiled for ten minutes and proteins were separated on SDS-8% PAGE. Resolved proteins were transferred to nitrocellulose and MHC and MRP were simultaneously labeled using polyclonal antibodies directed against the MHC/MRP rod domain7 (see also immunolabeling) diluted at 1:5000. Proteins were detected with HRP-conjugated anti-rabbit secondary antibodies (Vector) and blots were developed using chemiluminescence detection (Supersignal; Pierce). Films were digitized and MHC and MRP levels were quantified using the Scion Image software package. Ratios from a typical experiment are presented in Figure 1. Due to the extreme differences in expression levels, it was not possible to record signals from both MRP and MHC within the linear range of the film using this method, and MRP ratios are potentially underestimated. Electron microscopy For analysis of adult muscles, CO2 anaesthetized animals were glued to the bottom of Petri dishes with nitrocellulose. Thoraces were opened in rigor solution (75 mM potassium acetate, 5 mM magnesium acetate, 5 mM EGTA, 15 mM potassium phosphate (pH 7.0), 30% glycerol, 0.05 –0.1% Triton X-100), dissected at room temperature to remove the majority of the IFM, and stored overnight at 4 8C. Following several washes with 0.1 M sodium cacodylate buffer, muscles were treated with 0.2% tannic acid in cacodylate buffer for one hour at room temperature, then fixed in 3.5% glutaraldehyde and 0.2% tannic acid in buffer for one hour. After further cleaning of unwanted muscles, the thoraces were kept in cacodylate buffer overnight at 4 8C. Stage 18 wild-type and mutant embryos were prepared for electron microscopy by dechorionation as for immunolabeling followed by hand devitellinization and fixation in glutaraldehyde. Embryos and adult thoraces were post-fixed in 2% osmium tetraoxide in buffer (one hour); washed in water (3 £ for five minutes), stained en bloc with saturated aqueous uranyl acetate (at room temperature overnight); dehydrated in acetone and infiltrated in
Epon. Prior to Epon polymerization, individual DFMs were dissected out of the thorax using fine tungsten needles. Ultrathin sections were stained in uranyl acetate and in a mixture of lead salts44 and observed in a Philips 410 EM.
Acknowledgements Supported by NIH grant 5-PO1-HL15835 to the Pennsylvania Muscle Institute. Our thanks to N. Glaser and M. Lai for technical assistance and K. Kozopas (Univ. Alabama) for indicating the correct DFM nomenclature.
References 1. Huxley, H. E. (1963). Electron microscope studies on the structure of natural and synthetic protein filaments from striated muscle. J. Mol. Biol. 7, 281 –306. 2. Atkinson, S. J. & Stewart, M. (1991). Molecular basis of myosin assembly: coiled-coil interactions and the role of charge periodicities. J. Cell Sci.—Suppl. 14, 7 – 10. 3. Sohn, R. L., Vikstrom, K. L., Strauss, M., Cohen, C., Szent-Gyorgyi, A. G. & Leinwand, L. A. (1997). A 29 residue region of the sarcomeric myosin rod is necessary for filament formation. J. Mol. Biol. 266, 317 –330. 4. George, E. L., Ober, M. B. & Emerson, C. P., Jr (1989). Functional domains of the Drosophila melanogaster muscle myosin heavy-chain gene are encoded by alternatively spliced exons. Mol. Cell. Biol. 9, 2957 –2974. 5. Hastings, G. A. & Emerson, C. P., Jr (1991). Myosin functional domains encoded by alternative exons are expressed in specific thoracic muscles of Drosophila. J. Cell Biol. 114, 263– 276. 6. Bernstein, S. I., Mogami, K., Donady, J. J. & Emerson, C. P., Jr (1983). Drosophila muscle myosin heavy chain encoded by a single gene in a cluster of muscle mutations. Nature, 302, 393 –397. 7. Miedema, K., Hanske, M., Akhmanova, A., Bindels, P. & Hennig, W. (1995). Minor-myosin isoform synthesized preferentially in Drosophila testis is encoded by the muscle myosin heavy chain gene. Mech. Dev. 51, 67 –81. 8. Standiford, D. M., Davis, M. B., Miedema, K., Franzini-Armstrong, C. & Emerson, C. P., Jr (1997). Myosin rod protein: a novel thick filament component of Drosophila muscle. J. Mol. Biol. 265, 40 – 55. 9. Toffenetti, J., Mischke, D. & Pardue, M. L. (1987). Isolation and characterization of the gene for myosin light chain two of Drosophila melanogaster. J. Cell Biol. 104, 19 –28. 10. Sanbe, A., Gulick, J., Fewell, J. & Robbins, J. (2001). Examining the in vivo role of the amino terminus of the essential myosin light chain. J. Biol. Chem. 276, 32682 – 32686. 11. Schaub, M. C., Hefti, M. A., Zuellig, R. A. & Morano, I. (1998). Modulation of contractility in human cardiac hypertrophy by myosin essential light chain isoforms. Cardiovasc. Res. 37, 381– 404. 12. Yamada, A., Yoshio, M., Oiwa, K. & Nyitray, L.
1090
13.
14.
15.
16. 17.
18.
19.
20.
21.
22. 23.
24.
25.
26.
27.
(2000). Catchin, a novel protein in molluscan catch muscles, is produced by alternative splicing from the myosin heavy chain gene. J. Mol. Biol. 295, 169– 178. Shelud’ko, N., Permjakova, T., Tuturova, K., Neverkina, O. & Drozdov, A. (2001). Myorod, a thick filament protein in molluscan smooth muscles: isolation, polymerization and interaction with myosin. J. Muscle Res. Cell Motil. 22, 91 – 100. Bate, M. (1993). The mesoderm and its derivatives. In The Development of Drosophila melanogaster (Bate, M. & Martinez-Arias, A., eds), pp. 1013– 1090, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. O’Donnell, P. T. & Bernstein, S. I. (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. Elfvin, M., Levine, R. J. & Dewey, M. M. (1976). Paramyosin in invertebrate muscles. I. Identification and localization. J. Cell Biol. 71, 261– 272. Cohen, C., Szent-Gyorgyi, A. G. & Kendrick-Jones, J. (1971). Paramyosin and the filaments of molluscan “catch” muscles. I. Paramyosin: structure and assembly. J. Mol. Biol. 56, 223– 227. Szent-Gyorgyi, A. G., Cohen, C. & Kendrick-Jones, J. (1971). Paramyosin and the filaments of molluscan “catch” muscles. II. Native filaments: isolation and characterization. J. Mol. Biol. 56, 239– 258. Maroto, M., Arredondo, J., Goulding, D., Marco, R., Bullard, B. & Cervera, M. (1996). Drosophila paramyosin/miniparamyosin gene products show a large diversity in quantity, localization, and isoform pattern: a possible role in muscle maturation and function. J. Cell Biol. 134, 81 – 92. Maroto, M., Arredondo, J. J., San Roman, M., Marco, R. & Cervera, M. (1995). Analysis of the paramyosin/miniparamyosin gene. Miniparamyosin is an independently transcribed, distinct paramyosin isoform, widely distributed in invertebrates. J. Biol. Chem. 270, 4375– 4382. Matsubara, I., Goldman, Y. E. & Simmons, R. M. (1984). Changes in the lateral filament spacing of skinned muscle fibres when cross-bridges attach. J. Mol. Biol. 173, 15 – 33. Fyrberg, E. & Beall, C. (1990). Genetic approaches to myofibril form and function in Drosophila. Trends Genet. 6, 126– 131. Holtzer, H., Hijikata, T., Lin, Z. X., Zhang, Z. Q., Holtzer, S., Protasi, F., Franzini-Armstrong, C. & Sweeney, H. L. (1997). Independent assembly of 1.6 microns long bipolar MHC filaments and I-Z-I bodies. Cell Struct. Funct. 22, 83 – 93. Cripps, R. M., Suggs, J. A. & Bernstein, S. I. (1999). Assembly of thick filaments and myofibrils occurs in the absence of the myosin head. EMBO J. 18, 1793– 1804. Antin, P. B., Forry-Schaudies, S., Friedman, T. M., Tapscott, S. J. & Holtzer, H. (1981). Taxol induces postmitotic myoblasts to assemble interdigitating microtubule-myosin arrays that exclude actin filaments. J. Cell Biol. 90, 300–308. Levine, R. J., Elfvin, M., Dewey, M. M. & Walcott, B. (1976). Paramyosin in invertebrate muscles. II. Content in relation to structure and function. J. Cell Biol. 71, 273– 279. Castellani, L., Vibert, P. & Cohen, C. (1983). Structure
Contribution of MRP to Sarcomere Organization
28.
29. 30.
31.
32.
33.
34.
35.
36.
37. 38.
39. 40.
41. 42. 43. 44.
of myosin/paramyosin filaments from a molluscan smooth muscle. J. Mol. Biol. 167, 853– 872. Castellani, L. & Vibert, P. (1992). Location of paramyosin in relation to the subfilaments within the thick filaments of scallop striated muscle. J. Muscle Res. Cell Motil. 13, 174– 182. Parker, V. P., Falkenthal, S. & Davidson, N. (1985). Characterization of the myosin light-chain-2 gene of Drosophila melanogaster. Mol. Cell. Biol. 5, 3058– 3068. Moore, J. R., Dickinson, M. H., Vigoreaux, J. O. & Maughan, D. W. (2000). The effect of removing the N-terminal extension of the Drosophila myosin regulatory light chain upon flight ability and the contractile dynamics of indirect flight muscle. Biophys. J. 78, 1431– 1440. Irving, T., Bhattacharya, S., Tesic, I., Moore, J., Farman, G., Simcox, A. et al. (2001). Changes in myofibrillar structure and function produced by N-terminal deletion of the regulatory light chain in Drosophila. J. Muscle Res. Cell Motil. 22, 675– 683. Henry, G. D., Winstanley, M. A., Dalgarno, D. C., Scott, G. M., Levine, B. A. & Trayer, I. P. (1985). Characterization of the actin-binding site on the alkali light chain of myosin. Biochim. Biophys. Acta, 830, 233– 243. Timson, D. J., Trayer, H. R., Smith, K. J. & Trayer, I. P. (1999). Size and charge requirements for kinetic modulation and actin binding by alkali 1-type myosin essential light chains. J. Biol. Chem. 274, 18271– 18277. Trayer, I. P., Trayer, H. R. & Levine, B. A. (1987). Evidence that the N-terminal region of A1-light chain of myosin interacts directly with the C-terminal region of actin. A proton magnetic resonance study. Eur. J. Biochem. 164, 259– 266. Hayashibara, T. & Miyanishi, T. (1994). Binding of the amino-terminal region of myosin alkali 1 light chain to actin and its effect on actin – myosin interaction. Biochemistry, 33, 12821 –12827. Poetter, K., Jiang, H., Hassanzadeh, S., Master, S. R., Chang, A., Dalakas, M. C. et al. (1996). Mutations in either the essential or regulatory light chains of myosin are associated with a rare myopathy in human heart and skeletal muscle. Nature Genet. 13, 63 – 69. Tu, M. S. & Dickinson, M. (1994). Modulation of negative work output from a steering muscle of the blowfly Calliphora vicina. J. Expt. Biol. 192, 207– 224. Tu, M. S. & Dickinson, M. H. (1996). The control of wing kinematics by two steering muscles of the blowfly (Calliphora vicina). J. Comp. Physiol. [A], 178, 813– 830. Dickinson, M. H. & Tu, M. S. (1997). The function of Dipteran flight muscle. Comp. Biochem. Physiol. 116, 223– 238. Maughan, D., Moore, J., Vigoreaux, J., Barnes, B. & Mulieri, L. A. (1998). Work production and work absorption in muscle strips from vertebrate cardiac and insect flight muscle fibers. Advan. Expt. Med. Biol. 453, 471– 480. Ashburner, M. (1989). Drosophila: A Laboratory Handbook, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Campos-Ortega, J. (1997). The Embryonic Development of Drosophila melanogaster, Springer, New York. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680– 685. Sato, T. (1968). A modified method for lead staining of thin sections. J. Electron Microsc. 17, 158– 159.
Contribution of MRP to Sarcomere Organization
45. Miller, A. (1950). The internal anatomy and histology of the imago of Drosophila melanogaster. In The Biology of Drosophila (Demerec, M., ed.), pp. 420– 534, Wiley, New York.
1091
46. Reedy, M. K. (1968). Ultrastructure of insect flight muscle. I. Screw sense and structural grouping in the rigor cross-bridge lattice. J. Mol. Biol. 31, 155 –176.
Edited by J. Karn (Received 25 March 2003; received in revised form 20 June 2003; accepted 24 June 2003)