THE ULTRASTRUCTURE OF THE Z DISC IN SKELETAL MUSCLE

This electron microscopic study deals with the structure of the Z disc of frog's skeletal muscle, with special regard to the I filaments—whether they pass through the Z disc or terminate at it. In most longitudinal sections the I filaments terminate as rod-like projections on either side of the Z disc, one I filament on one side lying between two I filaments on the opposite side. This indicates that the I filaments are not continuous through the Z disc. The rod-like projections are often seen to consist of filaments (denoted as Z filaments) which meet at an angle. In cross-sections through the Z region the I filaments and Z filaments form tetragonal patterns. The I filaments are situated in the corners of the squares; the oblique Z filaments form the sides of squares. The tetragonal pattern formed by the Z filaments is rotated 45 degrees with respect to the tetragons formed by the I filaments on both sides of Z. This structural arrangement is interpreted to indicate that each I filament on one side of the Z disc faces the center of the space between four I filaments on the opposite side of Z and that the interconnection is formed by four Z filaments.     

Arrangement of actin filaments and myosin-like filaments in the contractile ring and of actin-like filaments in the mitotic spindle of dividing HeLa cells

We used a glutaraldehyde-tannic acid-saponin fixative to improve the preservation of actin filaments in dividing HeLa cells during preparation for thin sectioning. The contractile ring in the cleavage furrow is composed of a parallel array of actin filaments that circle the equator. We show that many of these actin filaments are arranged in small bundles. These bundles consist of about 25 filaments throughout cytokinesis. For comparison, filopodia on these cells have about 23 actin filaments packed at a higher density than the filaments in the contractile ring bundles. Some of the contractile ring actin filaments appear to radiate out from electron-dense sites on the plasma membrane. The contractile ring also has a large number of short filaments 13 nm in diameter that closely resemble filaments formed from purified human cytoplasmic myosin. These thick filaments are aligned circumferentially and interdigitate with the actin filaments, as expected for a sliding filament mechanism of tension generation. There are no long actin filaments in the mitotic spindle, but there are a large number (400 to 1000 per μm ³) of very short filaments identical in appearance to actin filaments in other parts of these cells. These short filaments may account for the reported staining of the mitotic spindle with fluorescent antibodies to actin and with fluorescent myosin fragments.     

Actin filaments, stereocilia, and hair cells of the bird cochlea. IV. How the actin filaments become organized in developing stereocilia and in the cuticular plate

In 8-day-old embryos stereocilia can be identified on the hair cells of the chick cochlea; within each is a small population of actin filaments which extend from the tip of the stereocilium to the apical cytoplasm of the cell. These filaments are not ordered in a regular way, however, and tend to be found near the lateral margins of the stereocilia with large spaces between adjacent filaments. By 9 days the spaces between adjacent filaments are reduced and there are regions where the crossover points of adjacent actin helices are in register even though in cross section the actin filaments do not lie on a regular lattice. By 10-11 days the actin filaments become progressively more crossbridged together and we can recognize in longitudinal section horizontal stripes caused by the periodicity of the crossbridges. In transverse section the filaments begin to lie on a hexagonal lattice. Each stereocilium, however, contains less than 100 actin filaments. Evidence is presented that once crossbridging is maximal and the filaments hexagonally packed (Days 11-12), the stereocilia increase in width by the orderly addition of actin filaments to the lateral margins of the existing filament bundle so that by Day 16 we find up to 400 filaments all packed on a hexagonal lattice. Thus there are two stages in bundle formation. In the first a small number of filaments condense into a hexagonally packed, crosslinked bundle. In the second, the bundle increases in diameter by addition of filaments to the periphery of the bundle in a process akin to crystal growth. From observations on the elongation of filaments in the rootlets and stereocilia, we conclude that rootlets grow by addition of subunits at the nonpreferred end while stereocilia elongate by addition to the preferred end. What makes this interesting is that these two modes of addition occur at different developmental times.     

Interaction of C-protein with myosin

The effect of C-protein on the assembly reaction of myosin was studied by flow birefringence, electron microscopy, and ultracentrifugation. Myosin filaments were formed by dilution to a lower ionic strength. Thinner filaments of 70-110 A in diameter were formed in the presence of C-protein. When dilution was effected by moderately slow dilution (dilution time of 0.5-2 min) or by stepwise dilution, C-protein favored the formation of longer filaments. When dilution was effected by even slower dilution (dilution time above 2 min), C-protein favored the formation of shorter filaments. Longer filaments formed by slow dilution incorporated more C-protein than shorter ones formed by faster dilution. Addition of C-protein to a solution of myosin filaments caused association of the filaments into longer filaments. The elongation effect was slower and stronger for longer filaments.     

Structure and paramyosin content of tarantula thick filaments

Muscle fibers of the tarantula femur exhibit structural and biochemical characteristics similar to those of other long-sarcomere invertebrate muscles, having long A-bands and long thick filaments. 9-12 thin filaments surround each thick filament. Tarantula muscle has a paramyosin:myosin heavy chain molecular ratio of 0.31 +/- 0.079 SD. We studied the myosin cross-bridge arrangement on the surface of tarantula thick filaments on isolated, negatively stained, and unidirectionally metal-shadowed specimens by electron microscopy and optical diffraction and filtering and found it to be similar to that previously described for the thick filaments of muscle of the closely related chelicerate arthropod, Limulus. Cross-bridges are disposed in a four-stranded right-handed helical arrangement, with 14.5-nm axial spacing between successive levels of four bridges, and a helical repeat period every 43.5 nm. The orientation of cross-bridges on the surface of tarantula filaments is also likely to be very similar to that on Limulus filaments as suggested by the similarity between filtered images of the two types of filaments and the radial distance of the centers of mass of the cross-bridges from the surfaces of both types of filaments. Tarantula filaments, however, have smaller diameters than Limulus filaments, contain less paramyosin, and display structure that probably reflects the organization of the filament backbone which is not as apparent in images of Limulus filaments. We suggest that the similarities between Limulus and tarantula thick filaments may be governed, in part, by the close evolutionary relationship of the two species.     

Effects of DMSO, pH, stretch and calcium on the thick filaments in an amphibian smooth muscle.

Conventional fixation with glutaraldehyde fixatives at pH 7.4 did not preserve/promote thick filaments in Bufo smooth muscle. However, if pH was lowered to 6.0, thick filaments were present and addition of dimethyl sulfoxide (DMSO) led to a greater number of thick filaments. Stretch alone had little effect on the presence of thiber seen. Calcium had little effect on the numbers of thick filaments, but it affected the appearance of the thick filaments. Low calcium (10(-7) M) caused a higher proportion of rod-shaped filaments, while in high calcium (10(-3) M) most thick filaments were ribbons. The great lability of the thick filaments in amphibian smooth muscles makes them ideal for studying factors which affect the appearance of thick filaments in smooth muscle, but it also raises the question of the degree of aggregation of myosin in the living cell.     

Reorganization of contractile elements in the platelet during clot retraction

Electron microscopic studies have been carried out on human platelets in the clot retraction. In the early stage of clot formation, platelets send out filopodia, in which thin filaments run longitudinally. The thin filaments are often observed to attach to the cell membrane where fibrin strands bind from the extracellular surface. In the later stage of clot formation, thick filaments become observable, mainly in the cell body of the platelets. These thick filaments are arranged to form an ordered array, and thin filaments run parallel to them. The thin filaments often attach to the end of the thick filaments. However, thin filaments are not seen between the arrays of thick filaments. Similar structures are also observed in the cytoskeleton of the contracted platelet. These filaments closely resemble the purified myosin aggregates formed under low ionic strength. Thus, during clot retraction, both actin and myosin in platelets are reorganized into thin and thick filaments, respectively.     

The variable twist of actin and its modulation by actin-binding proteins

Previous studies demonstrated that actin filaments have variable twist in which the intersubunit angles vary by approximately +/- 10 degrees within a filament. In this work we show that this variability was unchanged when different methods were used to prepare filaments for electron microscopy. We also show that actin-binding proteins can modulate the variability in twist. Three preparations of actin filaments were photographed in the electron microscope: negatively stained filaments, replicas of rapidly frozen, etched filaments, and frozen hydrated filaments. In addition, micrographs of actin + tropomyosin + troponin (thin filaments), of actin + myosin S1 (decorated filaments), and of filaments frayed from the acrosomal process of Limulus sperm (Limulus filaments) were obtained. We used two independent methods to measure variable twist based on Fourier transforms of single filaments. The first involved measuring layer line intensity versus filament length and the second involved measuring layer line position. We measured a variability in the intersubunit angle of actin filaments of approximately 12 degrees independent of the method of preparation or of measurement. Thin filaments have 15 degrees of variability, but the increase over pure actin is not statistically significant. Decorated filaments and Limulus filaments, however, have significantly less variability (approximately 2 and 1 degree, respectively), indicating a torsional stiffening relative to actin. The results from actin alone using different preparative methods are evidence that variable twist is a property of actin in solution. The results from actin filaments in the presence of actin-binding proteins suggest that the angular variability can be modulated, depending on the biological function.     

Kinetics of the formation and dissociation of actin filament branches mediated by Arp2/3 complex

The actin filament network at the leading edge of motile cells relies on localized branching by Arp2/3 complex from "mother" filaments growing near the plasma membrane. The nucleotide bound to the mother filaments (ATP, ADP and phosphate, or ADP) may influence the branch dynamics. To determine the effect of the nucleotide bound to the subunits of the mother filament on the formation and stability of branches, we compared the time courses of actin polymerization in bulk samples measured using the fluorescence of pyrene actin with observations of single filaments by total internal reflection fluorescence microscopy. Although the branch nucleation rate in bulk samples was nearly the same regardless of the nucleotide on the mother filaments, we observed fewer branches by microscopy on ADP-bound filaments than on ADP-P(i)-bound filaments. Observation of branches in the microscope depends on their binding to the slide. Since the probability that a branch binds to the slide is directly related to its lifetime, we used counts of branches to infer their rates of dissociation from mother filaments. We conclude that the nucleotide on the mother filament does not affect the initial branching event but that branches are an order of magnitude more stable on the sides of new ATP- or ADP-P(i) filaments than on ADP-actin filaments.     

Interaction of bacterial flagella with myosin

The interaction of isolated flagellar filaments of Bacillus brevis var. G.-B. P+ with skeletal muscle myosin has been investigated. Bacterial flagellar filaments co-precipitate with myosin at low ionic strength (at the conditions of myosin aggregation). Addition of bacterial flagellar filaments to myosin led to inhibition of its K+-EDTA- and Ca2+-ATPase activity, but had no influence on Mg2+-ATPase. Monomeric protein of bacterial flagella filaments (flagellin) did not co-precipitate with myosin and had no influence on its ATPase activity. The flagella filaments did not co-precipitate with myosin in the presence of F-actin if it was mixed with myosin before the filaments. If the flagella filaments were added to myosin solution before the addition of F-actin the amount of filaments and actin in myosin precipitate were comparable. In this case the presence of flagella filaments decreased activation of myosin Mg2+-ATPase by actin to 25-30%. Thus the bacterial flagellar filaments are able to interact with myosin and modify its ATPase activity. Probably, these properties of filaments are caused by resemblance of flagellin and actin. For instance, the unique origin of these proteins may be the reason of such resemblance.     

Control of actin filament length and turnover by actin depolymerizing factor (ADF/cofilin) in the presence of capping proteins and ARP2/3 complex.

The effect of Arabidopsis thaliana ADF1 and human ADF on the number of filaments in F-actin solutions has been examined using a seeded polymerization assay. ADF did not sever filaments in a catalytic fashion, but decreased the steady-state length distribution of actin filaments in correlation with its effect on actin dynamics. The increase in filament number was modest as compared with the large increase in filament turnover. ADF did not decrease the length of filaments shorter than 1 micrometer. ADF promoted the rapid turnover of gelsolin-capped filaments in a manner dependent on the number of pointed ends. To explain these results, we propose that, as a consequence of the cooperative binding of ADF to F-actin, two populations of energetically different filaments coexist in solution pending a flux of subunits from one to the other. The ADF-decorated filaments depolymerize rapidly from their pointed ends, while undecorated filaments polymerize. ADF also promotes rapid turnover of gelsolin-capped filaments in the presence of the pointed end capper Arp2/3 complex. It is shown that the Arp2/3 complex steadily generates new barbed ends in solutions of gelsolin-capped filaments, which represents an important aspect of its function in actin-based motility.     

An electrodynamic (moving field) theory of muscular contraction

It is proposed that muscular contraction is the result of electrostatic attraction between oppositely charged areas on actin and myosin filaments. On the latter charged areas are assumed to be moving, always a step ahead of stationary charged areas on actin filaments, the moving charges pulling the stationary charges, hence the actin filaments, with them. It may be noted that electric motors in human technology work on a similar moving field principle. On myosin filaments minute charged areas are assumed to spiral along the surface of the filament on 2 or 3-start helical paths, probably the latter, thus engaging with adjacent actin filaments in a screw-like fashion. The spiralling charges follow each other like peristaltic waves, engaging with an increasing number of static fields on actin filaments as interdigitation proceeds. The source of the electrostatic charges are assumed to be minute voltaic cells, one associated with every myosin head. It is suggested that they could be calcium-magnesium cells, calcium adsorbed by troponin complexes on actin filaments constituting one electrode, and magnesium complexed with ATP on myosin filaments the other. The potential difference that has to exist between actin and myosin filaments, if muscles are to be capable of developing a maximum force of 20 N per cm2, is calculated at about 50 mV.     

Alternative flagellar filament types in the haloarchaeon Haloarcula marismortui

Many Archaea use rotation of helical flagellar filaments for swimming motility. We isolated and characterized the flagellar filaments of Haloarcula marismortui, an archaeal species previously considered to be nonmotile. Two Haloarcula marismortui phenotypes were discriminated--their filaments are composed predominantly of either FlaB or FlaA2 flagellin, and the corresponding genes are located on different replicons. FlaB and FlaA2 filaments differ in antigenicity and thermostability. FlaA2 filaments are distinctly thicker (20-22 nm) than FlaB filaments (16-18 nm). The observed filaments are nearly twice as thick as those of other characterized euryarchaeal filaments. The results suggest that the helicity of Haloarcula marismortui filaments is provided by a mechanism different from that in the related haloarchaeon Halobacterium salinarum, where 2 different flagellin molecules present in comparable quantities are required to form a helical filament.     

Ca2+- and S1-induced conformational changes of reconstituted skeletal muscle thin filaments observed by fluorescence energy transfer spectroscopy: structural evidence for three States of thin filament

Rabbit skeletal muscle alpha-tropomyosin (Tm) and the deletion mutant (D234Tm) in which internal actin-binding pseudo-repeats 2, 3, and 4 are missing [Landis et al. (1997) J. Biol. Chem. 272, 14051-14056] were used to investigate the interaction between actin and tropomyosin or actin and troponin (Tn) by means of fluorescence resonance energy transfer (FRET). FRET between Cys-190 of D234Tm and Gln-41 or Cys-374 of actin did not cause any significant Ca2+-induced movement of D234Tm, as reported previously for native Tm [Miki et al. (1998) J. Biochem. 123, 1104-1111]. FRET did not show any significant S1-induced movement of Tm and D234Tm on thin filaments either. The distances between Cys-133 of TnI, and Gln-41 and Cys-374 of actin on thin filaments reconstituted with D234Tm (mutant thin filaments) were almost the same as those on thin filaments with native Tm (wild-type thin filaments) in the absence of Ca2+. Upon binding of Ca2+ to TnC, these distances on mutant thin filaments increased by approximately 10 A in the same way as on wild-type thin filaments, which corresponds to a Ca2+-induced conformational change of thin filaments [Miki et al. (1998) J. Biochem. 123, 324-331]. The rigor binding of myosin subfragment 1 (S1) further increased these distances by approximately 7 A on both wild-type and mutant thin filaments when the thin filaments were fully decorated with S1. This indicates that a further conformational change on thin filaments was induced by S1 rigor-binding (S1-induced or open state). Plots of the extent of S1-induced conformational change vs. molar ratio of S1 to actin showed that the curve for wild-type thin filaments is hyperbolic, whereas that for mutant thin filaments is sigmoidal. This suggests that the transition to the S1-induced state on mutant thin filaments is depressed with a low population of rigor S1. In the absence of Ca2+, the distance also increased on both wild-type and mutant thin filaments close to the level in the presence of Ca2+ as the molar ratio of S1 to actin increased up to 1. The curves are sigmoidal for both wild-type and mutant thin filaments. The addition of ATP completely reversed the changes in FRET induced by rigor S1 binding. For mutant thin filaments, the transition from the closed state to the open state in the presence of ATP is strongly depressed, which results in the inhibition of acto-myosin ATPase even in the presence of Ca2+. The present FRET measurements provide structural evidence for three states of thin filaments (relaxed, Ca2+-induced or closed, and S1-induced or open states) for the regulation mechanism of skeletal muscle contraction.     

Ca(2+)- and s1-induced movement of troponin T on mutant thin filaments reconstituted with functionally deficient mutant tropomyosin

The deletion mutant (D234Tm) of rabbit skeletal muscle alpha-tropomyosin, in which internal actin-binding pseudo-repeats 2, 3, and 4 are missing, inhibits the thin filament activated myosin-ATPase activity whether Ca(2+) ion is present or not [Landis et al. (1997) J. Biol. Chem. 272, 14051-14056]. Fluorescence resonance energy transfer (FRET) showed substantial changes in distances between Cys-60 or 250 of troponin T (TnT) and Gln-41 or Cys-374 of actin on wild-type thin filaments corresponding to three states of thin filaments [Kimura et al. (2002) J. Biochem. 132, 93-102]. Troponin T movement on mutant thin filaments reconstituted with D234Tm was compared with that on wild-type thin filaments to understand from which the functional deficiency of mutant thin filaments derives. The Ca(2+)-induced changes in distances between Cys-250 of TnT and Gln-41 or Cys-374 of F-actin were smaller on mutant thin filaments than on wild-type thin filaments. On the other hand, the distances between Cys-60 of TnT and Gln-41 or Cys-374 of F-actin on mutant thin filaments did not change at all regardless of whether Ca(2+) was present. Thus, FRET showed that the Ca(2+)-induced movement of TnT was severely impaired on mutant thin filaments. The rigor binding of myosin subfragment 1 (S1) increased the distances when the thin filaments were fully decorated with S1 in the presence and absence of Ca(2+). However, plots of the extent of S1-incuced movement of TnT against molar ratio of S1 to actin in the presence and absence of Ca(2+) showed that the S1-induced movement of TnT was also impaired on mutant thin filaments. The deficiency of TnT movement on mutant thin filaments causes the altered S1-induced movement of TnI, and mutant thin filaments consequently fail to activate the myosin-ATPase activity even in the presence of Ca(2+).     

Cytoskeletal reorganization of human platelets after stimulation revealed by the quick-freeze deep-etch technique

We studied the cytoskeletal reorganization of saponized human platelets after stimulation by using the quick-freeze deep-etch technique, and examined the localization of myosin in thrombin-treated platelets by immunocytochemistry at the electron microscopic level. In unstimulated saponized platelets we observed cross-bridges between: adjoining microtubules, adjoining actin filaments, microtubules and actin filaments, and actin filaments and plasma membranes. After activation with 1 U/ml thrombin for 3 min, massive arrays of actin filaments with mixed polarity were found in the cytoplasm. Two types of cross-bridges between actin filaments were observed: short cross-bridges (11 +/- 2 nm), just like those observed in the resting platelets, and longer ones (22 +/- 3 nm). Actin filaments were linked with the plasma membrane via fine short filaments and sometimes ended on the membrane. Actin filaments and microtubules frequently ran close to the membrane organelles. We also found that actin filaments were associated by end-on attachments with some organelles. Decoration with subfragment 1 of myosin revealed that all the actin filaments associated end-on with the membrane pointed away in their polarity. Immunocytochemical study revealed that myosin was present in the saponin-extracted cytoskeleton after activation and that myosin was localized on the filamentous network. The results suggest that myosin forms a gel with actin filaments in activated platelets. Close associations between actin filaments and organelles in activated platelets suggests that contraction of this actomyosin gel could bring about the observed centralization of organelles.     

Interaction of microtubule-associated proteins with actin filaments. Studies using the fluorescence-photobleaching recovery technique

The interaction of microtubule-associated proteins with actin filaments has been investigated by measuring the diffusion coefficient of either the filament or the microtubule-associated proteins. Experiments were performed using the technique of fluorescence photobleaching recovery with actin labeled with iodoacetamidotetramethyl rhodamine or microtubule-associated proteins labeled with iodoacetamidofluorescein. Actin filaments composed of pure rhodamine-labeled actin are not immobilized under a variety of conditions (Tait, J. F., and Frieden, C. (1982c) Biochemistry 21, 6046-6053). We find that addition of microtubule-associated proteins to rhodamine-labeled actin in a ratio as low as 1:1000 can cause immobilization, presumably cross-linking actin into a network of nondiffusible filaments. Immobilization occurs after polymerization is complete, suggesting either a length redistribution of actin filaments, a redistribution of the cross-links between filaments, or the slow addition of actin filaments to other filaments via the microtubule-associated protein. Experiments using fluorescein-labeled microtubule-associated proteins show that these proteins are bound to actin filaments as they are formed and that binding depended on actin concentration, indicating that there are a number of binding sites on the actin filaments. However, while the actin filaments become completely immobilized, the microtubule-associated proteins become only partially immobilized suggesting at least two different classes of binding affinities. The large peptide obtained from trypsin-treated fluorescein-labeled microtubule-associated proteins is not able to immobilize actin filaments since it does not bind to the filaments.     

Motion of actin filaments in the presence of myosin heads and ATP.

We measured, by fluorescence correlation spectroscopy, the motion of actin filaments in solution during hydrolysis of ATP by acto-heavy meromyosin (acto-HMM). The method relies on the fact that the intensity of fluorescence fluctuates as fluorescently labeled actin filaments enter and leave a small sample volume. The rapidity of these number fluctuations is characterized by the autocorrelation function, which decays to 0 in time that is related to the average velocity of translation of filaments. The time of decay of the autocorrelation function of bare actin filaments in solution was 10.59 +/- 0.85 s. Strongly bound (rigor) heads slowed down the diffusion. Direct observation of filaments under an optical microscope showed that addition of HMM did not change the average length or flexibility of actin filaments, suggesting that the decrease in diffusion was not due to a HMM-induced change in the shape of filaments. Rather, slowing down of translational motion was caused by an increase in the volume of the diffusing complex. Surprisingly, the addition of ATP to acto-HMM accelerated the motion of actin filaments. The acceleration was the greatest at the low molar ratios of HMM:actin. Direct observation of filaments under an optical microscope showed that in the presence of ATP the average length of filaments did not change and that the filaments became stiffer, suggesting that acceleration of diffusion was not due to an ATP-induced increase in flexibility of filaments. These results show that some of the energy of splitting of ATP is impaired to actin filaments and suggest that 0.06 +/- 0.02 of HMM interferes with the diffusion of actin filaments during hydrolysis of ATP.     

Tomographic reconstruction of treponemal cytoplasmic filaments reveals novel bridging and anchoring components

An understanding of the involvement of bacterial cytoplasmic filaments in cell division requires the elucidation of the structural organization of those filamentous structures. Treponemal cytoplasmic filaments are composed of one protein, CfpA, and have been demonstrated to be involved in cell division. In this study, we used electron tomography to show that the filaments are part of a complex with a novel molecular organization that includes at least two distinct features decorating the filaments. One set of components appears to anchor the filaments to the cytoplasmic membrane. The other set of components appears to bridge the cytoplasmic filaments on the cytoplasmic side, and to be involved in the interfilament spacing within the cell. The filaments occupy between 3 and 18% of the inner surface of the cytoplasmic membrane. These results reveal a novel filamentous molecular organization of independent filaments linked by bridges and continuously anchored to the membrane.     

RecA-ssDNA filaments supercoil in the presence of single-stranded DNA-binding protein

Using atomic force microscopy (AFM), we find that RecA-single-stranded DNA (RecA-ssDNA) filaments, in the presence of single-stranded DNA-binding (SSB) protein, organize into left-handed bundles, which differ from the previously reported disordered aggregates formed when SSB is excluded from the reaction. In addition, we see both left- and right-handedness on bundles of two filaments. These two-filament supercoils, individual filaments, and other smaller bundles further organize into more complicated bundles, showing overall left-handedness which cannot be explained by earlier arguments that presumed supercoiling is absent in RecA-ssDNA filaments. This novel finding and our previous results regarding supercoiling of RecA-double-stranded DNA (RecA-dsDNA) filaments are, however, consistent with each other and can possibly be explained by the intrinsic tendency of RecA-DNA filaments, in their fully coated form, to order themselves into helical bundles, independent of the DNA inside the filaments (ssDNA or dsDNA). RecA-RecA interactions may dominate the bundling process, while the original conformation of DNA inside filaments and other factors (mechanical properties of filaments, concentration of filaments, and Mg(2+) concentration) could contribute to the variation in the appearance and pitch of supercoils. The tendency of RecA-DNA filaments to form ordered supercoils and their presence during strand exchange suggest a possible biological importance of supercoiled filaments.