Long continuous actin bundles in Drosophila bristles are constructed by overlapping short filaments

The actin bundles essential for Drosophila bristle elongation are hundreds of microns long and composed of cross-linked unipolar filaments. These long bundles are built from much shorter modules that graft together. Using both confocal and electron microscopy, we demonstrate that newly synthesized modules are short (1-2 microm in length); modules elongate to approximately 3 microm by growing over the surface of longitudinally adjacent modules to form a graft; the grafted regions are initially secured by the forked protein cross-bridge and later by the fascin cross-bridge; actin bundles are smoothed by filament addition and appear continuous and without swellings; and in the absence of grafting, dramatic alterations in cell shape occur that substitutes cell width expansion for elongation. Thus, bundle morphogenesis has several components: module formation, elongation, grafting, and bundle smoothing. These actin bundles are much like a rope or cable, made by overlapping elements that run a small fraction of the overall length, and stiffened by cross-linking.     

Cargo properties play a critical role in myosin Va-driven cargo transport along actin filaments

High-resolution experiments revealed that a single myosin-Va motor can transport micron-sized cargo on actin filaments in a stepwise manner. However, intracellular cargo transport is mediated through the dense actin meshwork by a team of myosin Va motors. The mechanism of how motors interact mechanically to bring about efficient cargo transport is still poorly understood. This study describes a stochastic model where a quantitative understanding of the collective behaviors of myosin Va motors is developed based on cargo stiffness. To understand how cargo properties affect the overall cargo transport, we have designed a model in which two myosin Va motors were coupled by wormlike chain tethers with persistence length ranging from 10 to 80 nm and contour length from 100 to 200 nm, and predicted distributions of velocity, run length, and tether force. Our analysis showed that these parameters are sensitive to both the contour and persistence length of cargo. While the velocity of two couple motors is decreased compared to a single motor (from 531 ± 251 nm/s to as low as 318 ± 287 nm/s), the run length (716 ± 563 nm for a single motor) decreased for short, rigid tethers (to as low as 377 ± 187 μm) and increased for long, flexible tethers (to as high as 1.74 ± 1.50 μm). The sensitivity of processive properties to tether rigidity (persistence length) was greatest for short tethers, which caused the motors to exhibit close, yet anti-cooperative coordination. Motors coupled by longer tethers stepped more independently regardless of tether rigidity. Therefore, the properties of the cargo or linkage must play an essential role in motor-motor communication and cargo transport.     

Regulation of actin filament cross-linking and bundle shape in Drosophila bristles

Previous studies demonstrate that in developing Drosophila bristles, two cross-linking proteins are required sequentially to bundle the actin filaments that support elongating bristle cells. The forked protein initiates the process and facilitates subsequent cross-linking by fascin. Using cross-linker-specific antibodies, mutants, and drugs we show that fascin and actin are present in excessive amounts throughout bundle elongation. In contrast, the forked cross-linker is limited throughout bundle formation, and accordingly, regulates bundle size and shape. We also show that regulation of cross-linking by phosphorylation can affect bundle size. Specifically, inhibition of phosphorylation by staurosporine results in a failure to form large bundles if added during bundle formation, and leads to a loss of cross-linking by fascin if added after the bundles form. Interestingly, inhibition of dephosphorylation by okadaic acid results in the separation of the actin bundles from the plasma membrane. We further show by thin section electron microscopy analysis of mutant and wild-type bristles that the amount of material that connects the actin bundles to the plasma membrane is also limited throughout bristle elongation. Therefore, overall bundle shape is determined by the number of actin filaments assembled onto the limited area provided by the connector material. We conclude that assembly of actin bundles in Drosophila bristles is controlled in part by the controlled availability of a single cross-linking protein, forked, and in part by controlled phosphorylation of cross-links and membrane actin connector proteins.     

F-actin bundles in Drosophila bristles are assembled from modules composed of short filaments

The actin bundles in Drosophila bristles run the length of the bristle cell and are accordingly 65 microns (microchaetes) or 400 microns (macrochaetes) in length, depending on the bristle type. Shortly after completion of bristle elongation in pupae, the actin bundles break down as the bristle surface becomes chitinized. The bundles break down in a bizarre way; it is as if each bundle is sawed transversely into pieces that average 3 microns in length. Disassembly of the actin filaments proceeds at the "sawed" surfaces. In all cases, the cuts in adjacent bundles appear in transverse register. From these images, we suspected that each actin bundle is made up of a series of shorter bundles or modules that are attached end-to-end. With fluorescent phalloidin staining and serial thin sections, we show that the modular design is present in nondegenerating bundles. Decoration of the actin filaments in adjacent bundles in the same bristle with subfragment 1 of myosin reveals that the actin filaments in every module have the same polarity. To study how modules form developmentally, we sectioned newly formed and elongating bristles. At the bristle tip are numerous tiny clusters of 6-10 filaments. These clusters become connected together more basally to form filament bundles that are poorly organized, initially, but with time become maximally cross-linked. Additional filaments are then added to the periphery of these organized bundle modules. All these observations make us aware of a new mechanism for the formation and elongation of actin filament bundles, one in which short bundles are assembled and attached end-to-end to other short bundles, as are the vertical girders between the floors of a skyscraper.     

Identification of actin filaments in the rhabdomeral microvilli of Drosophila photoreceptors

The phototransductive microvilli of arthropod photoreceptors each contain an axial cytoskeleton. The present study shows that actin filaments are a component of this cytoskeleton in Drosophila. Firstly, actin was detected in the rhabdomeral microvilli and in the subrhabdomeral cytoplasm by immunogold labeling with antiactin. Secondly, the rhabdomeres were labeled with phalloidin, indicating the presence of filamentous actin. Finally, the actin filaments were decorated with myosin subfragment-1. The characteristic arrowhead complex formed by subfragment-1 decoration points towards the base of the microvilli, so that the fast growing end of each filament is at the distal end of the microvillus, where it is embedded in a detergent-resistant cap. Each microvillus contains more than one actin filament. Decorated filaments extend the entire length of each microvillus and project into the subrhabdomeral cytoplasm. This organization is comparable to that of the actin filaments in intestinal brush border microvilli. Similar observations were made with the photoreceptor microvilli of the crayfish, Procambarus. Our results provide an indication as to how any myosin that is associated with the rhabdomeres might function.     

Behaviour of microtubules and actin filaments in living Drosophila embryos

We describe the preparation of novel fluorescent derivatives of rabbit muscle actin and bovine tubulin, and the use of these derivatives to study the behaviour of actin filaments and microtubules in living Drosophila embryos, in which the nuclei divide at intervals of 8 to 21 min. The fluorescently labelled proteins appear to function normally in vitro and in vivo, and they allow continuous observation of the cytoskeleton in living embryos without perturbing development. By coinjecting labelled actin and tubulin into the early syncytial embryo, the spatial relationships between the distinct filament networks that they form can be followed second by second. The dynamic rearrangements of actin filaments and microtubules observed confirms and extends results obtained from previous studies, in which fixation techniques and specific staining were used to visualize the cytoskeleton in the Drosophila embryo. However, no tested fixation method produces an exact representation of the in vivo microtubule distribution.     

The role of actin filaments and microtubules in hepatocyte spheroid self-assembly

Cultured rat hepatocytes self-assemble into three-dimensional structures or spheroids that exhibit ultrastructural characteristics of native hepatic tissue and enhanced liver-specific functions. The spheroid formation process involves cell translocation and changes in cell shape, indicative of the reorganization of the cytoskeletal elements. To elucidate the function of the cytoskeleton, hepatocytes undergoing spheroid formation were treated with drugs that disrupt the different cytoskeletal components. Cytochalasin D, which targets the actin filaments, caused inhibition of spheroid formation. The role of microtubules in this process was assessed by incubating the cells with taxol or nocodazole. Perturbation of microtubules had minimal effects on spheroid assembly. Scanning electron micrographs showed no morphological differences between spheroids formed in control cultures and those formed in the presence of taxol or nocodazole. In addition, the effects of those agents on hepatocyte functions were investigated. Albumin secretion and cytochrome P450 2B1/2 activities of hepatocytes were comparable in spheroids formed in the presence of taxol or nocodazole to those formed in control cultures. The levels of these liver-specific activities were lower in cytochalasin D--treated cultures where only dispersed cells or cell clumps were found but spheroids had not found. Thus, hepatocytes require an intact actin network to self-assemble efficiently into functional tissue-like structures. Perturbation of the microtubule lattice does not impair the formation process. Events that transpire during hepatocyte spheroid self-assembly exhibit striking similarities to processes commonly observed in tissue morphogenesis. The results provide insight into the mechanisms that cells employ to organize into tissues and can contribute to our understanding of how to control the cellular assembly in tissue engineering and clinical applications.     

The role of actin filaments in the gravitropic response of snapdragon flowering shoots

The involvement of the actin and the microtubule cytoskeleton networks in the gravitropic response of snapdragon ( Antirrhinum majus L.) flowering shoots was studied using various specific cytoskeleton modulators. The microtubule-depolymerizing drugs tested had no effect on gravitropic bending. In contrast, the actin-modulating drugs, cytochalasin D (CD), cytochalasin B (CB) and latrunculin B (Lat B) significantly inhibited the gravitropic response. CB completely inhibited shoot bending via inhibiting general growth, whereas CD completely inhibited bending via specific inhibition of the differential flank growth in the shoot bending zone. Surprisingly, Lat B had only a partial inhibitory effect on shoot bending as compared to CD. This probably resulted from the different effects of these two drugs on the actin cytoskeleton, as was seen in cortical cells. CD caused fragmentation of the actin cytoskeleton and delayed amyloplast displacement following gravistimulation. In contrast, Lat B caused a complete depolymerization of the actin filaments in the shoot bending zone, but only slightly reduced the amyloplast sedimentation rate following gravistimulation. Taken together, our results suggest that the actin cytoskeleton is involved in the gravitropic response of snapdragon shoots. The actin cytoskeleton within the shoot cells is necessary for normal amyloplast displacement upon gravistimulation, which leads to the gravitropic bending.     

The receptor tyrosine phosphatase Dlar and integrins organize actin filaments in the Drosophila follicular epithelium.

BACKGROUND: Regulation of actin structures is instrumental in maintaining proper cytoarchitecture in many tissues. In the follicular epithelium of Drosophila ovaries, a system of actin filaments is coordinated across the basal surface of cells encircling the oocyte. These filaments have been postulated to regulate oocyte elongation; however, the molecular components that control this cytoskeletal array are not yet understood. RESULTS: We find that the receptor tyrosine phosphatase (RPTP) Dlar and integrins are involved in organizing basal actin filaments in follicle cells. Mutations in Dlar and the common beta-integrin subunit mys cause a failure in oocyte elongation, which is correlated with a loss of proper actin filament organization. Immunolocalization shows that early in oogenesis Dlar is polarized to membranes where filaments terminate but becomes generally distributed late in development, at which time beta-integrin and Enabled specifically associate with actin filament terminals. Rescue experiments point to the early period of polar Dlar localization as critical for its function. Furthermore, clonal analysis shows that loss of Dlar or mys influences actin filament polarity in wild-type cells that surround mutant tissues, suggesting that communication between neighboring cells regulates cytoskeletal organization. Finally, we find that two integrin alpha subunits encoded by mew and if are required for proper oocyte elongation, implying that multiple components of the ECM are instructive in coordinating actin fiber polarity. CONCLUSIONS: Dlar cooperates with integrins to coordinate actin filaments at the basal surface of the follicular epithelium. To our knowledge, this is the first direct demonstration of an RPTP's influence on the actin cytoskeleton.     

F actin bundles in Drosophila bristles. I. Two filament cross-links are involved in bundling

Transverse sections though Drosophila bristles reveal 7-11 nearly round, plasma membrane-associated bundles of actin filaments. These filaments are hexagonally packed and in a longitudinal section they show a 12-nm periodicity in both the 1.1 and 1.0 views. From earlier studies this periodicity is attributable to cross-links and indicates that the filaments are maximally cross-linked, singed mutants also have 7-11 bundles, but the bundles are smaller, flattened, and the filaments within the bundles are randomly packed (not hexagonal); no periodicity can be detected in longitudinal sections. Another mutant, forked (f36a), also has 7-11 bundles but even though the bundles are very small, the filaments within them are hexagonally packed and display a 12-nm periodicity in longitudinal section. The singed-forked double mutant lacks filament bundles. Thus there are at least two species of cross-links between adjacent actin filaments. Hints of why two species of cross-links are necessary can be gleaned by studying bristle formation. Bristles sprout with only microtubules within them. A little later in development actin filaments appear. At early stages the filaments in the bundles are randomly packed. Later the filaments in the bundles become hexagonally packed and maximally cross-linked. We consider that the forked proteins may be necessary early in development to tie the filaments together in a bundle so that they can be subsequently zippered together by fascin (the singed gene product).     

The role of actin filaments in the organization of the endoplasmic reticulum in honeybee photoreceptor cells

Close to the bases of the photoreceptive microvilli, arthropod photoreceptors contain a dense network of endoplasmic reticulum that is involved in the regulation of the intracellular calcium concentration, and in the biogenesis of the photoreceptive membrane. Here, we examine the role of the cytoskeleton in organizing this submicrovillar endoplasmic reticulum in honeybee photoreceptors. Immunofluorescence microscopy of taxol-stabilized specimens, and electron-microscopic examination of high-pressure frozen, freeze-substituted retinae demonstrate that the submicrovillar cytoplasm lacks microtubules. The submicrovillar region contains a conspicuous F-actin system that codistributes with the submicrovillar endoplasmic reticulum. Incubation of retinal tissue with cytochalasin B leads to depolymerization of the submicrovillar F-actin system, and to disorganization and disintegration of the submicrovillar endoplasmic reticulum, indicating that an intact F-actin cytoskeleton is required to maintain the architecture of this domain of the endoplasmic reticulum. We have also developed a permeabilized cell model in order to study the physiological requirements for the interaction of the endoplasmic reticulum with actin filaments. The association of submicrovillar endoplasmic reticulum with actin filaments appears to be independent of ATP, Ca²⁺ and Mg²⁺, suggesting a tight static anchorage.     

Centrosome separation: respective role of microtubules and actin filaments

In mammalian cells, the separation of centrosomes is a prerequisite for bipolar mitotic spindle assembly. We have investigated the respective contribution of the two cytoskeleton components, microtubules and actin filaments, in this process. Distances between centrosomes have been measured during cell cycle progression in Xenopus laevis XL2 cultured cells in the presence or absence of either network. We considered two stages in centrosome separation: the splitting stage, when centrosomes start to move apart (minimum distance of 1 microm), and the elongation stage (from 1 to 7 microm). In interphase, depolymerisation of microtubules by nocodazole significantly inhibited the splitting stage, while the elongation stage was, on the contrary, facilitated. In mitosis, while nocodazole treatment completely blocked spindle assembly, in prophase, we observed that 55% of the centrosomes separated, versus 94% in the control. Upon actin depolymerisation by latrunculin, splitting of the interphase centrosome was blocked, and cells entered mitosis with unseparated centrosomes. Cells compensated for this separation delay by increasing the length of both prophase and prometaphase stages to allow for centrosome separation until a minimal distance was reached. Then the cells passed through anaphase, performing proper chromosome separation, but cytokinesis did not occur, and binuclear cells were formed. Our results clearly show that the actin microfilaments participate in centrosome separation at the G2/M transition and work in synergy with the microtubules to accelerate centrosome separation during mitosis.     

ADP-ribosylated actin caps the barbed ends of actin filaments

The mode of action on actin polymerization of skeletal muscle actin ADP-ribosylated on arginine 177 by perfringens iota toxin was investigated. ADP-ribosylated actin decreased the rate of nucleated actin polymerization at substoichiometric ratios of ADP-ribosylated actin to monomeric actin. ADP-ribosylated actin did not tend to copolymerize with actin. Actin filaments were depolymerized by the addition of ADP-ribosylated actin. The maximal monomer concentration reached by addition of ADP-ribosylated actin was similar to the critical concentration of the pointed ends of actin filaments. ADP-ribosylated actin had no effect on the rate of polymerization of gelsolin-capped actin filaments which polymerize at the pointed ends. The results suggest that ADP-ribosylated actin acts as a capping protein which binds to the barbed ends of actin filaments to inhibit polymerization. Based on an analysis of the depolymerizing effect of ADP-ribosylated actin, the equilibrium constant for binding of ADP-ribosylated actin to the barbed ends of actin filaments was determined to be about 10(8) M-1. As actin is ADP-ribosylated by perfringens iota toxin and by botulinum C2 toxin, it appears that conversion of actin into a capping protein by ADP-ribosylation is a pathophysiological reaction catalyzed by bacterial toxins which ultimately leads to inhibition of actin assembly.     

Drosophila quail, a villin-related protein, bundles actin filaments in apoptotic nurse cells

Drosophila Quail protein is required for the completion of fast cytoplasm transport from nurse cells to the oocyte, an event critical for the production of viable oocytes. The abundant network of cytoplasmic filamentous actin, established at the onset of fast transport, is absent in quail mutant egg chambers. Previously, we showed that Quail is a germline-specific protein with sequence homology to villin, a vertebrate actin-regulating protein. In this study, we combined biochemical experiments with observations in egg chambers to define more precisely the function of this protein in the regulation of actin-bundle assembly in nurse cells. We report that recombinant Quail can bind and bundle filamentous actin in vitro in a manner similar to villin at a physiological calcium concentration. In contrast to villin, Quail is unable to sever or cap filamentous actin, or to promote nucleation of new actin filaments at a high calcium concentration. Instead, Quail bundles the filaments regardless of the calcium concentration. In vivo, the assembly of nurse-cell actin bundles is accompanied by extensive perforation of the nurse-cell nuclear envelopes, and both of these phenomena are manifestations of nurse-cell apoptosis. To investigate whether free calcium levels are affected during apoptosis, we loaded egg chambers with the calcium indicator Indo-1. Our observations indicate a rise in free calcium in the nurse-cell cytoplasm coincident with the permeabilization of the nuclear envelopes. We also show that human villin expressed in the Drosophila germline could sense elevated cytoplasmic calcium; in nurse cells with reduced levels of Quail protein, villin interfered with actin-bundle stability. We conclude that Quail efficiently assembles actin filaments into bundles in nurse cells and maintains their stability under fluctuating free calcium levels. We also propose a developmental model for the fast phase of cytoplasm transport incorporating findings presented in this study.     

Drosophila bristles and the nature of quantitative genetic variation

Numbers of Drosophila sensory bristles present an ideal model system to elucidate the genetic basis of variation for quantitative traits. Here, we review recent evidence that the genetic architecture of variation for bristle numbers is surprisingly complex. A substantial fraction of the Drosophila genome affects bristle number, indicating pervasive pleiotropy of genes that affect quantitative traits. Further, a large number of loci, often with sex- and environment-specific effects that are also conditional on background genotype, affect natural variation in bristle number. Despite this complexity, an understanding of the molecular basis of natural variation in bristle number is emerging from linkage disequilibrium mapping studies of individual candidate genes that affect the development of sensory bristles. We show that there is naturally segregating genetic variance for environmental plasticity of abdominal and sternopleural bristle number. For abdominal bristle number this variance can be attributed in part to an abnormal abdomen-like phenotype that resembles the phenotype of mutants defective in catecholamine biosynthesis. Dopa decarboxylase (Ddc) encodes the enzyme that catalyses the final step in the synthesis of dopamine, a major Drosophila catecholamine and neurotransmitter. We found that molecular polymorphisms at Ddc are indeed associated with variation in environmental plasticity of abdominal bristle number.     

Two general classes of cytoplasmic actin filaments in tissue culture cells: The role of tropomyosin

In the assembly of actin filaments that takes place during the spreading of a polulation of human lung cells, after trypsin detachment off the substratum and replating, tropomyosin exhibits a considrable lag in its association with the newly forming filament bundles; it begins to associate with them during the later stages of cell spreading as the actin filament bundles normally seen in interphase cells begin to organize. This lag is evident in a number of cell types that are spreading onto a substratum; it does not appear to be due to a selective degradation of this molecule during rounding up of the cells, since tropmyosin associates with the actin filament bundles after this lag even under conditions where the protein synthetic activity of the cell is inhibited to more than 95% by cycloheximide. The preferential binding of tropomyosin to fully assembled filament bundles but not to newly formed bundles of actin filaments suggests therefore the existence of two classes of action filaments: those that bind tropomyosin and those that do not. This selective localization of tropomyosin and those that do not. This selective localization of tropomyosin on actin filaments was further pursued by examining the localization of this molecule in membrane ruffles. The immunofluorescent results indicate that ruffling is an actin-filament-dependent, microtubule-independent phenomenon. Tropomyosin is absent from membrane ruffles under a variety of circumstances where ruffling is expressed and, more generally, from any other cellular activity where actin filaments are expected to be in a dynamic state of reorganization or are required to be in a flexible configuraion. It is concluded that in tissue culture cells tropomyosin binds preferentially to actin filaments involved in structural support to confer rigidity upon them as well as aid them in maintaining a stretched phenotype. The absence of tropomyosin from certain motile phenomena where actin filaments are involved indicates that these classes of actin filaments are regulated by cytoplasmic mechanisms distinct from that by which tropomyosin (and troponin) mediates contractility in skeletal mulscle; it opens the possibility that different types of actin filaments enagaged in different cellular motile phenomenon in tissue culture cells may be regulated by a host of coexisting regulatory mechanisms, some as yet undetermined.     

The organization of actin filaments in human placental villi

The surface of the syncytial trophoblast of the human placenta is covered by a microvillous (brush) border that is in direct contact with maternal blood. Because of this location, it is the site of a variety of transport, enzymatic and receptor activities vital to many placental functions. The organization of the brush border as well as other features of placental villus organization may well be influenced by the distribution of cytoplasmic actin filaments. In order to determine the distribution of actin filaments in human placenta, small pieces of villi were briefly fixed in glutaraldehyde, permeabilized with saponin, and incubated in solutions containing subfragment 1 of myosin (S1). After S1 decoration of actin filaments, tissue was fixed in glutaraldehyde containing tannic acid in order to better visualize the polarity of the filaments, and prepared for electron microscopic examination. The microvilli each contained a core of actin filaments running from the tip of the microvillus to the apical cytoplasm. Most of the actin filaments displayed a distinct polarity, with the S1 arrowheads pointing away from the microvillar tips. These filaments extended only a short distance into the apical cytoplasm. There appeared to be another group of actin filaments in a matlike arrangement in the apical cytoplasm. Coated pits and vesicles were often observed between the microvilli. There appeared to be no clear association between the coated pits and decorated actin filaments, but this was difficult to establish with certainty because of the close proximity of the microvilli. Bundles of actin filaments were sometimes observed near the basal cell surface of the syncytial trophoblast, and in pericytes and capillary endothelial cells in the cores of the villi.(ABSTRACT TRUNCATED AT 250 WORDS)     

Barrier role of actin filaments in regulated mucin secretion from airway goblet cells

Airway goblet cells secrete mucin onto mucosal surfaces under the regulation of an apical, phospholipase C/G_q-coupled P2Y₂ receptor. We tested whether cortical actin filaments negatively regulate exocytosis in goblet cells by forming a barrier between secretory granules and plasma membrane docking sites as postulated for other secretory cells. Immunostaining of human lung tissues and SPOC1 cells (an epithelial, mucin-secreting cell line) revealed an apical distribution of - and -actin in ciliated and goblet cells. In goblet cells, actin appeared as a prominent subplasmalemmal sheet lying between granules and the apical membrane, and it disappeared from SPOC1 cells activated by purinergic agonist. Disruption of actin filaments with latrunculin A stimulated SPOC1 cell mucin secretion under basal and agonist-activated conditions, whereas stabilization with jasplakinolide or overexpression of - or -actin conjugated to yellow fluorescent protein (YFP) inhibited secretion. Myristoylated alanine-rich C kinase substrate, a PKC-activated actin-plasma membrane tethering protein, was phosphorylated after agonist stimulation, suggesting a translocation to the cytosol. Scinderin (or adseverin), a Ca²⁺-activated actin filament severing and capping protein was cloned from human airway and SPOC1 cells, and synthetic peptides corresponding to its actin-binding domains inhibited mucin secretion. We conclude that actin filaments negatively regulate mucin secretion basally in airway goblet cells and are dynamically remodeled in agonist-stimulated cells to promote exocytosis. lung; mucus; exocytosis     

Kinetics of the cooperative association of actin to actin filaments.

The cooperative formation of actin filaments from monomers was followed by light scattering and electron microscopy. The results are well described by a simple model mechanism in which the growth and destruction of filaments occurs by stepwise addition or dissociation of protomers. All steps except the dimerisation step are assumed to have identical rate constants. These were found to be 5 X 10(3) M-1 - sec-1 and 3 X 10(-2) sec-1 for the association and dissociation, respectively (at pH 7.5 and in the presence of 10(-3) M calcium chloride). The equilibrium constant of elongation as obtained from the critical concentration is 1.7 X 10(5) M-1. The corresponding equilibrium constant of dimerisation is about 10 million times smaller (cooperativity parameter sigma = 2 X 10(-7)). This makes the nucleation extremely difficult and cooperativity very high. A best fit of the model to the experimental data is achieved when the destruction of a dimer is much faster than the addition of a third protomer (fast monomer- dimer pre-equilibrium). The size of the nucleus from which propagation becomes faster than dissociation is 3.