Flagellar hook structures of Caulobacter and Salmonella and their relationship to filament structure

Native flagellar hooks from a polarly flagellated bacterium, Caulobacter crescentus, and polyhooks from a peritrichously flagellated bacterium, Salmonella typhimurium. have been studied by densitometry of electron micrographs of negatively stained specimens, followed by computerized Fourier analysis and three-dimensional reconstruction. The two structures are remarkably similar. In both cases, the subunits are arranged along a right-handed basic helix of 2.3 nm pitch with successive subunits separated by an azimuthal angle of 64 to 65 °, and there is a pronounced system of continuous 6-start grooves and ridges on the surface of the structures. The subunit of Salmonella (M_r 42,000, versus 70,000 for Caulobacter) is somewhat thinner and yields a smaller overall hook diameter. The “bent finger” subunit shape and orientation in both cases suggests that the hook could bend readily by a sliding motion in the 11-start direction at inner radii, with the 6-start groove preventing collision at outer radii. The basic helical pitch of the Salmonella hook structure, and the number of subunits per basic helical turn (5.56) makes it highly compatible with the Salmonella flagellar filament (2.6 nm pitch. 5.51 subunits per turn); so also does the elongated shape and tilt angle of the hook and flagellin subunits in the respective structures. The two structures may therefore conjoin directly in the intact flagellum, although participation of a minor protein is not ruled out by the data.     

Three-dimensional reconstruction of the flagellar hook from Caulobacter crescentus

The structure of the bacterial flagellar hook produced by a mutant of Caulobacter crescentus was studied by electron microscopy, optical diffraction, and digital image processing techniques. The helical surface lattice of the hook is defined by a single, right-handed genetic helix having a pitch of about 23 Å, an axial rise per subunit of 4 Å and an azimuthal angle between subunits of 64·5 °. The lattice is also characterized by intersecting families of 5-start, 6-start and long-pitch 11-start helices. These helical parameters are remarkably similar to those determined for the flagellar filaments from several strains of gram-negative bacteria. The technique of three-dimensional image reconstruction (DeRosier & Klug, 1968) was applied to nine of the better preserved specimens and the diffraction data from five of these were correlated and averaged and used to generate an average three-dimensional model of the hook. The pattern of density modulations in the three-dimensional model is suggestive of an elongated, curved shape for the hook subunit (100 Å × 25 Å × 25 Å). The subunits are situated in the lattice of the polyhook such that their long axes are tilted about 45 ° with respect to the hook axis. The subunits appear to make contact with each other along the 6-start helices at a radius of 80 Å and also along the 11-start helices at a radius of 65 Å. Few structural features are revealed at radii between 15 å and 45 Å and, therefore, we are unable to decide to what extent the hook subunits extend into this region. The most striking characteristic of the model is the presence of deep, broad, continuous 6-start helical grooves extending from an inner radius of about 50 Å to the perimeter of the particle at 105 Å radius. Normal hooks usually appear curved in electron micrographs and sometimes so are the mutant hooks; the prominent 6-start grooves appear to allow for bending with minimal distortion of matter in the outer regions of the hook. A round stain-filled channel about 25 Å in diameter runs down the center of the polyhook. Such a channel supports a model for flagellar assembly in which flagellin subunits travel through the interior of the flagellum to the growing distal end of the filament.     

Evidence for a mixed lattice in microtubules reassembled in vitro

An extensive structural analysis of microtubules assembled in vitro has been carried out using electron microscopy in conjunction with computer analysis based on Fourier transforms and helical diffraction theory. Microtubules assembled in vitro displayed a range of numbers of protofilaments from 12 to 16, with 14 the most abundant (84% in one large sampling). In almost all structures observed protofilaments are staggered to form a characteristic 3-start shallow helix. The presence of the 3-start helix was confirmed by fiber tilting experiments to correct the effects of microtubule flattening. Since α and β tubulin subunits alternate along the protofilaments, continuous helical lattices can be constructed with interactions between adjacent protofilaments involving unlike subunits (type A lattice) or like subunits (type B lattice). However, the 14-protofilament, 3-start microtubules are incompatible with either the A or B-type continuous helical lattice. Evidence is presented which indicates that lattice discontinuities are present which generate features of both the A and B-types, with the latter predominating.     

Interpretation of the X-ray scattering profiles of chromatin at various NaCl concentrations by a simple chain model.

In order to interpret the change in the X-ray scattering profiles from rat thymus chromatin, extensive model calculation was carried out. Chromatin is modelled as a string of subunits (nucleosomes) in which disorder is introduced into the positions of adjacent subunits. Disposition parameters characterizing the arrangement of subunits were estimated for various states of chromatin, so that the main feature of the scattering profiles is described. The result indicated that the structure of chromatin changes, as the NaCl concentration increases, from the extended "beads-on-a string" structure to the condensed helical structure. The latter has an outer diameter of about 26 nm with 3-4 nucleosomes per turn. In the intermediate state, it has a loose helical structure. The estimation of disorder suggested that the arrangement of subunits is appreciably disordered even in the condensed helical filament at 50 mM NaCl. Our model for chromatin condensation seems to support models of the "crossed linker" type.     

Secondary structure of the r(CUUCGG) tetraloop.

The oligonucleotide r(GGACUUCGGUCC) has been observed to adopt a hairpin conformation by solution NMR and a double helical conformation by X-ray diffraction. In order to understand this apparent conflict, we used time-resolved fluorescence depolarization and 19fluorine NMR to follow the secondary structure of this dodecamer as the solution composition was changed stepwise from the NMR experimental conditions to those used for crystallization. Calculation of the dodecamer concentration in the crystal (180 mM strands) and the cation concentration needed for neutrality (>2 M) prompted investigation of a tethered species, in which two dodecamers are connected by a string of 4 nt, geometrically equivalent to approximately 100 mM strands, in 2.5 M NaCl. The RNA tetraloop and its DNA analog maintain a single-strand hairpin conformation in solution, even under the conditions used to grow the crystal. Under high salt conditions, the tethered RNA and DNA analogs of this sequence yield secondary components which could be the double helical conformation. Crystal contacts in addition to solvent changes and high RNA concentrations are needed to obtain the double helix as the predominant species.     

Protein-RNA interactions during TMV assembly

A review of the structural studies of tobacco mosaic virus (TMV) is given. TMV is essentially a flat helical microcrystal with 16 1/3 subunits per turn. A single strand of RNA runs along the helix and is deeply embedded in the protein. The virus particles form oriented gels from which high-resolution X-ray fiber diffraction data can be obtained. This may be interpreted by the use of six heavy-atom derivatives to give an electron density map at 0.4 nm resolution from which the RNA configuration and the form of the inner part of the protein subunit may be determined. In addition, the protein subunits form a stable 17-fold two-layered disk which is involved in virus assembly and which crystallizes. By the use of noncrystallographic symmetry and a single heavy-atom derivative, it has been possible to solve the structure of the double disk to 0.28 nm resolution. In this structure one sees that an important structural role is played by four alpha-helices, one of which (the LR helix) appears to form the main binding site for the RNA. The main components of the binding site appear to be hydrophobic interactions with the bases, hydrogen bonds between aspartate groups and the sugars, and arginine salt bridges to the phosphate groups. The binding site is between two turns of the virus helix or between the turns of the double disk. In the disk, the region proximal to the RNA binding site is in a random coil until the RNA binds, whereupon the 24 residues involved build a well-defined structure, thereby encapsulating the RNA.     

Conformational studies of anionic melittin analogues: effect of peptide concentration, pH, ionic strength, and temperature--models for protein folding and halophilic proteins.

Melittin (MLT), a 26-residue cationic (net charge +5 at pH 7.2) peptide from bee venom, is well known to be a monomeric, approximately random coil; but when its charges are reduced by titration, by acetylation (net charge +2) or succinylation (net charge -2), or by screening by salt, it goes over to tetrameric alpha-helix. The conversion is promoted by raising the peptide concentration. The tetramer is held together by hydrophobic forces. We have changed the net charge to -6 by acylation with acetylcitric anhydride (a new acylating agent); this anionic derivative forms tetrameric helix at neutral pH, without salt, and at relatively low concentration, conditions under which the cationic MLT does not become helical. Thus, a high net charge is not sufficient to prevent association and helix formation. We have synthesized an anionic melittin analogue of MLT (E-MLT; net charge -4) in which all five lysine and arginine residues are replaced with glutamate, and acetyl and succinyl derivatives of E-MLT (net charges -5 and -6). All three of these are resistant to helix formation. They require much higher NaCl or NaClO4 concentration for helix formation than does MLT. Even CaCl2, MgCl2, and spermine.4HCl are less effective in helicizing E-MLT than MLT. MLT, at pH 7.2, shows increasing helix as the peptide concentration increases (8-120 microM), but E-MLT and its acyl derivatives do not. MLT and acylated MLTs in the helical tetramer show both cold- and heat-induced unfolding, with maximum stability near room temperature. At high temperature, a significant amount of residual structure remains. Heating (to 100 degrees C) monomeric MLT (i.e., MLT at low concentration) or E-MLT results in a monotonic increase in negative ellipticity. In 1.0 M NaCl, E-MLT (at sufficiently high concentration) also shows cold and hot unfolding. The results are discussed in respect to charge-charge and charge-dipole interactions, and hydrophobic effects. E-MLT is also discussed in relation to proteins of halophilic bacteria, which have higher proportions of anionic residues than do corresponding proteins of nonhalophiles.     

Structure of an F-actin trimer disrupted by gelsolin and implications for the mechanism of severing

Stable oligomers of filamentous actin were obtained by cross-linking F-actin with 1,4-N,N'-phenylenedimaleimide and depolymerization with excess segment-1 of gelsolin. Segment-1-bound and cross-linked actin oligomers containing either two or three actin subunits were purified and shown to nucleate actin assembly. Kinetic assembly data from mixtures of monomeric actin and the actin oligomers fit a nucleation model where cross-linked actin dimer or trimer reacts with an actin monomer to produce a competent nucleus for filament assembly. We report the three-dimensional structure of the segment-1-actin hexamer containing three actin subunits, each with a tightly bound ATP. Comparative analysis of this structure with twelve other actin structures provides an atomic level explanation for the preferential binding of ATP by the segment-1-complexed actin. Although the structure of segment-1-bound actin trimer is topologically similar to the helical model of F-actin (1), it has a distorted symmetry compared with that of the helical model. This distortion results from intercalation of segment-1 between actin protomers that increase the rise per subunit and rotate each of the actin subunits relative to their positions in F-actin. We also show that segment-1 of gelsolin is able to sever actin filaments, although the severing activity of segment-1 is significantly lower than full-length gelsolin.     

On the surface lattice of microtubules: helix starts, protofilament number, seam, and handedness

The tubulin monomers of brain microtubules reassembled in vitro are arranged on a 3-start helix, irrespective of whether the number of protofilaments is 13 or 14. The dimer packing is that of the B-lattice described for flagellar microtubules. This implies that the tubulin core of microtubules contains at least one helical discontinuity. Neither 5-start nor 8-start helices have a physical significance and thus cannot be implicated in models of microtubule elongation, but the structure is compatible with elongation of protofilaments by dimers or protofilamentous oligomers. The inner and outer surfaces of the microtubule wall can be visualized by propane jet freezing, freeze fracturing, and metal replication, at a resolution of at least 4 nm. The 3-start helix is left-handed, in contrast to a previous study based on negative staining and shadowing. The reasons for this discrepancy are discussed.     

Low pH-Induced Pore Formation by the T Domain of Botulinum Toxin Type A is Dependent upon NaCl Concentration

Botulinum neurotoxins (BoNTs) undergo low pH-triggered membrane insertion, resulting in the translocation of their light (catalytic) chains into the cytoplasm. The T (translocation) domain of the BoNT heavy chain is believed to carry out translocation. Here, the behavior of isolated T domain from BoNT type A has been characterized, both in solution and when associated with model membranes. When BoNT T domain prepared in the detergent dodecylmaltoside was diluted into aqueous solution, it exhibited a low pH-dependent conformational change below pH 6. At low pH the T domain associated with, and formed pores within, model membrane vesicles composed of 30 mol% dioleoylphosphatidylglycerol/70 mol% dioleoylphosphatidylcholine. Although T domain interacted with vesicles at low (50 mM) and high (400 mM) NaCl concentrations, the interaction required much less lipid at low salt. However, even at high lipid concentrations pore formation was much more pronounced at low NaCl concentrations than at high NaCl concentration. Increasing salt concentration after insertion in the presence of 50 mM NaCl did not decrease pore formation. A similar effect of NaCl concentration upon pore formation was observed in vesicles composed solely of dioleoylphosphatidylcholine, showing that the effect of NaCl did not solely involve modulation of electrostatic interactions between protein and anionic lipids. These results indicate that some feature of membrane-bound T domain tertiary structure critical for pore formation is highly dependent upon salt concentration.     

Time-resolved solution X-ray scattering of tobacco mosaic virus coat protein: kinetics and structure of intermediates

The kinetics of assembly and disassembly of tobacco mosaic virus coat protein (TMVP) following temperature jumps have been studied by small-angle X-ray scattering and turbidimetry. The structures of the principal aggregates of TMVP oligomers (A protein), intermediate size (helix I) and large size helical rods (helix II), have been characterized by their average radii of gyration of thickness, cross section, and shape obtained from the corresponding regimes of the small-angle scattering pattern. This structural information was obtained within seconds after the temperature-induced initiation of either polymerization or depolymerization and allowed us to detect transient intermediates. This methodology made it possible to observe and characterize the structure of a principal intermediate. Taken together with other kinetic information, these data suggest that polymerization of TMVP under virus self-assembly conditions may proceed via a single-layered helical nucleus that contains about 20 subunits. Previous studies have shown that overshoot polymerization of TMVP can occur and results in metastable long helical viruslike rods which subsequently depolymerize and then form short helical rods, depending on the conditions of the final equilibrium state. The longer rods (helix II) are overshoot polymers which form within seconds and contain 17 1/3 subunits per turn (helix IIB), in contrast to the subunit packing arrangement of 16 1/3 subunits per turn found in the shorter helical rods (helix IA). The latter packing arrangement is the one found in TMV. An overall polymerization scheme is proposed for the formation of these two helical forms of TMVP.     

Large-scale opening of A + T rich regions within supercoiled DNA molecules is suppressed by salt.

Large-scale cooperative helix opening has been previously observed in A + T rich sequences contained in supercoiled DNA molecules at elevated temperatures. Since it is well known that helix melting of linear DNA is suppressed by addition of salt, we have investigated the effects of added salts on opening transitions in negatively supercoiled DNA circles. We have found that localised large-scale stable melting in supercoiled DNA is strongly suppressed by modest elevation of salt concentration, in the range 10 to 30 mM sodium. This has been shown in a number of independent ways: 1. The temperature required to promote cruciform extrusion by the pathway that proceeds via the coordinate large-scale opening of an A + T rich region surrounding the inverted repeat (the C-type pathway, first observed in the extrusion of the ColE1 inverted repeat) is elevated by addition of salt. The temperature required for extrusion was increased by about 4 deg for an addition of 10 mM NaCl. 2. A + T rich regions in supercoiled DNA exhibit hyperreactivity towards osmium tetroxide as the temperature is raised; this reactivity is strongly suppressed by the addition of salt. At low salt concentrations of added NaCl (10 mM) we observe that there is an approximate equivalence between reducing the salt concentration, and the elevation of temperature. Above 30 mM NaCl the reactivity of the ColE1 sequences is completely supressed at normal temperatures. 3. Stable helix opening transitions in A + T rich sequences may be observed with elevated temperature, using two-dimensional gel electrophoresis; these transitions become progressively harder to demonstrate with the addition of salt. With the addition of low concentrations of salt, the onset of opening transitions shifts to higher superhelix density, and by 30 mM NaCl or more, no transitions are visible up to a temperature of 50 degrees C. Statistical mechanical simulation of the data indicate that the cooperativity free energy for the transition is unaltered by addition of salt, but that the free energy cost for opening each basepair is increased. These results demonstrate that addition of even relatively low concentrations of salt strongly suppress the large-scale helix opening of A + T rich regions, even at high levels of negative supercoiling. While the opening at low salt concentrations may reveal a propensity for such transitions, spontaneous opening is very unlikely under physiological conditions of salt, temperature and superhelicity, and we conclude that proteins will therefore be required to facilitate opening transitions in cellular DNA.     

Interaction of chromatin with NaCl and MgCl2: Solubility and binding studies,transition to and characterization of the higher-order structure

Chicken erythrocyte chromatin containing histones H1 and H5 was carefully separated into a number of well-characterized fractions. A distinction could be made between chromatin insoluble in NaCl above about 80 mm, and chromatin soluble at all NaCl concentrations. Both chromatin forms were indistinguishable electrophoretically and both underwent the transition from the low salt “10 nm” coil to the “30 nm” higher-order structure solenoid by either raising the MgCl₂ concentration to about 0.3 mm or the NaCl concentration to about 75 mm. The transitions were examined in detail by elastic light-scattering procedures. It could be shown that the 10 nm form is a flexible coil. For the 30 nm solenoid, the assumption of a rigid cylindrical structure was in good agreement with 5.7 nucleosomes per helical turn. However, disagreement of calculated frictional parameters with values derived from quasielastic light-scattering and sedimentation introduced the possibility that the higher-order structure, under these conditions, is more extended, flexible, or perhaps a mixture of structures. Values for density and refractive index increments of chromatin are also given.To understand the interaction of chromatin with NaCl and with MgCl₂, a number of experiments were undertaken to study solubility, precipitation, conformational transitions and binding of ions over a wide range of experimental conditions, including chromatin concentration.     

Structure of flexible filamentous plant viruses

Flexible filamentous viruses make up a large fraction of the known plant viruses, but in comparison with those of other viruses, very little is known about their structures. We have used fiber diffraction, cryo-electron microscopy, and scanning transmission electron microscopy to determine the symmetry of a potyvirus, soybean mosaic virus; to confirm the symmetry of a potexvirus, potato virus X; and to determine the low-resolution structures of both viruses. We conclude that these viruses and, by implication, most or all flexible filamentous plant viruses share a common coat protein fold and helical symmetry, with slightly less than 9 subunits per helical turn.     

Stoichiometric analysis of the flagellar hook-(basal-body) complex of Salmonella typhimurium

The stoichiometries of components within the flagellar hook-(basal-body) complex of Salmonella typhimurium have been determined. The hook protein (FlgE), the most abundant protein in the complex, is present at approximately 130 subunits. Hook-associated protein 1 (FlgK) is present at approximately 12 subunits. The distal rod protein (FlgG) is present at approximately 26 subunits, while the proximal rod proteins (FlgB, FlgC and FlgF) are present at only approximately six subunits each. The stoichiometries of the proximal rod proteins and hook-associated protein 1 are, within experimental error, consistent with values of 5 or 6, and 11, respectively. Such values would correspond to either one or two turns of a helical structure with a basic helix of approximately 5.5 subunits per turn, which is the geometry of both the hook and the filament and, one supposes, the rod and hook-associated proteins. These stoichiometries may derive from rules for the heterologous interactions that occur when a helical structure consists of successive segments constructed from different proteins; the stoichiometries within the hook and the distal portion of the rod must, however, be set by different mechanisms. The stoichiometries for the ring proteins are approximately 26 subunits each for the M-ring protein (FliF), the P-ring protein (FlgI), and the L-ring protein (FlgH); the protein responsible for the S-ring feature is not known. The rings presumably have rotational rather than helical symmetry, in which case the stoichiometries would be directly constrained by the intersubunit bonding angle. The ring stoichiometries are discussed in light of other information concerning flagellar structure and function.     

Solution structure and dynamics of ADF from Toxoplasma gondii

Toxoplasma gondii ADF (TgADF) belongs to a functional subtype characterized by strong G-actin sequestering activity and low F-actin severing activity. Among the characterized ADF/cofilin proteins, TgADF has the shortest length and is missing a C-terminal helix implicated in F-actin binding. In order to understand its characteristic properties, we have determined the solution structure of TgADF and studied its backbone dynamics from ¹⁵N-relaxation measurements. TgADF has conserved ADF/cofilin fold consisting of a central mixed β-sheet comprised of six β-strands that are partially surrounded by three α-helices and a C-terminal helical turn. The high G-actin sequestering activity of TgADF relies on highly structurally and dynamically optimized interactions between G-actin and G-actin binding surface of TgADF. The equilibrium dissociation constant for TgADF and rabbit muscle G-actin was 23.81 nM, as measured by ITC, which reflects very strong affinity of TgADF and G-actin interactions. The F-actin binding site of TgADF is partially formed, with a shortened F-loop that does not project out of the ellipsoid structure and a C-terminal helical turn in place of the C-terminal helix α4. Yet, it is more rigid than the F-actin binding site of Leishmania donovani cofilin. Experimental observations and structural features do not support the interaction of PIP2 with TgADF, and PIP2 does not affect the interaction of TgADF with G-actin. Overall, this study suggests that conformational flexibility of G-actin binding sites enhances the affinity of TgADF for G-actin, while conformational rigidity of F-actin binding sites of conventional ADF/cofilins is necessary for stable binding to F-actin.     

Salt bridges do not stabilize polyproline II helices

Interactions between side chains, and in particular salt bridges, have been shown to be important in the stabilization of secondary structure. Here we investigate the contribution of a salt bridge formed between a lysine and a glutamate to the polyproline II (P(II)) helical content of proline-rich peptides. Since this structure has precisely three residues per turn, charged residues spaced three residues apart are on the same side of the helix and are best situated to interact. By contrast, computer simulations show that charged residues spaced four residues apart are both too far apart to interact strongly and are oriented such that interactions are unlikely. We have measured the P(II) content of peptides containing a lysine and glutamate pair spaced three or four residues apart using circular dichroism spectroscopy. Somewhat surprisingly we find that the P(II) content is insensitive to both the spacing and the pH. These findings indicate that i --> i + 3 salt bridges do not stabilize the P(II) helical conformation. The implications of these observations for both P(II) helix formation and denatured protein conformations are discussed.     

Chloroplast ultrastructure and thylakoid polypeptide composition are affected by different salt concentrations in the halophytic plant Arthrocnemum macrostachyum

The effect of different external salt concentrations, from 0 mM to 1030 mM NaCl, on photosynthetic complexes and chloroplast ultrastructure in the halophyte Arthrocnemum macrostachyum was studied. Photosystem II, but not Photosystem I or cytochrome b6/f, was affected by salt treatment. We found that the PsbQ protein was never expressed, whereas the amounts of PsbP and PsbO were influenced by salt in a complex way. Analyses of Photosystem II intrinsic proteins showed an uneven degradation of subunits with a loss of about 50% of centres in the 0 mM NaCl treated sample. Also the shape of chloroplasts, as well as the organization of thylakoid membranes were affected by NaCl concentration, with many grana containing few thylakoids at 1030 mM NaCl and thicker grana and numerous swollen thylakoids at 0 mM NaCl. The PsbQ protein was found to be depleted also in thylakoids from other halophytes.     

Application of the method of phage T4 DNA ligase-catalyzed ring-closure to the study of DNA structure. II. NaCl-dependence of DNA flexibility and helical repeat

In this work, we demonstrate that it is possible to determine the molar cyclization factor jM from single ligation reactions in which both circular and linear dimer DNA species are formed concurrently from linear monomers. This approach represents a significant improvement over previous methods, in which jM is evaluated from the ratio of the rate constants for two separate processes; namely (1) the cyclization of linear DNA and (2) the association of two linear molecules to form linear dimers. Determination of jM for a 366 base-pair molecule yields 5.8 X 10(-8) M, in close agreement with the value of 5.6 X 10(-8) M determined by Shore et al. for the same molecule. Using the current approach for the determination of jM, we have investigated the dependence on NaCl concentration (0 to 162 mM-NaCl, 1 mM-MgCl2) of both the lateral and torsional flexibilities of DNA. The principal observation is that both quantities are essentially constant over the above range of NaCl concentrations, with the persistence length P approximately 450 (+/- 15) A, and the torsional elastic constant C approximately 2.0 (+/- 0.2) X 10(-19) erg cm. These observations are in accord with the previous theoretical prediction that P becomes essentially independent of NaCl concentration above 10 to 20 mM. We have examined the dependence of the helical repeat of DNA on NaCl concentration over the above range, and have found the value of 10.44 base-pairs per turn to be essentially constant over that range. This last result suggests that earlier studies have overestimated the dependence of DNA helical twist on salt concentration.     

The formation of actin oligomers studied by analytical ultracentrifugation

The small oligomers formed from Mg-G-actin under favorable conditions were studied by sedimentation velocity ultracentrifugation. The critical concentration of actin at pH 7.8 in the presence of 100 microM MgCl2 and 200 microM ATP was 12.5 +/- 2.8 microM. Under these conditions, about 15% of 7.5 microM Mg-actin was converted to oligomers of subunit size four to eight in 5 h at 20 degrees C. In 100 microM MgCl2 and no free ATP, the critical concentration was about 6.5 microM, and about 22% of 7.5 microM Mg-actin was converted to dimers in 80 min. There were no detectable higher oligomers or F-actin present in either case. As determined by the analysis of ATP hydrolysis, most, if not all, of the oligomer subunits contained ATP. When 28.5 microM actin was polymerized to steady state in 100 microM MgCl2 and 200 microM ATP, about 50% of the actin was present as F-actin, consistent with the critical concentration (approximately 12.5 microM), about 50% as oligomers as large as seven subunits, and only about 5% as monomers. When solutions containing oligomers were diluted the oligomers dissociated. Alternatively, when the MgCl2 concentration was raised to 1 mM, the solutions containing oligomers polymerized more rapidly than monomeric Mg-G-actin and to the same final steady state. These data are entirely consistent with the condensation-elongation model for helical polymerization proposed by Oosawa and Kasai (Oosawa, F., and Kasai, M. (1962) J. Mol. Biol. 4, 10-21) according to which, under certain conditions, substantial amounts of short linear and helical oligomers should be formed below the critical concentration and linear oligomers should coexist with monomers and F-actin at steady state.