Posttranslational modifications of tubulin in teleost photoreceptor cytoskeletons

1. Posttranslational modifications of tubulin by acetylation and detyrosination have been correlated previously with microtubule stability in numerous cell types. 2. In this study, posttranslational modifications of tubulin and their regional distribution within teleost photoreceptor cones and rods are demonstrated immunohistochemically using antibodies specific for acetylated, detyrosinated, or tyrosinated tubulin. 3. Immunolocalization was carried out on isolated whole cones and mechanically detached rod and cone inner/outer segments. 4. Acetylated tubulin within rods and cones is found only in microtubules of the ciliary axoneme of the outer segment. Detyrosinated tubulin is also enriched in axonemes of both rod and cone outer segments. 5. Distributions of tyrosinated and detyrosinated cytoplasmic microtubules differ within cones and rods. In cones, detyrosinated and tyrosinated tubulins are both abundant throughout the cell body. In rods, the ellipsoid and myoid contain much more tyrosinated tubulin than detyrosinated tubulin. Comparisons between whole cones and cone fragments suggest that detyrosinated microtubules are more stable than tyrosinated microtubules in teleost photoreceptors. 6. Our findings provide further evidence that microtubules of teleost cones differ from rod microtubules in their stabilities and rapidity of turnover within the photoreceptor inner segment.     

Phosphorylation by Rho kinase regulates CRMP-2 activity in growth cones

Collapsin response mediator protein 2 (CRMP-2) enhances the advance of growth cones by regulating microtubule assembly and Numb-mediated endocytosis. We previously showed that Rho kinase phosphorylates CRMP-2 during growth cone collapse; however, the roles of phosphorylated CRMP-2 in growth cone collapse remain to be clarified. Here, we report that CRMP-2 phosphorylation by Rho kinase cancels the binding activity to the tubulin dimer, microtubules, or Numb. CRMP-2 binds to actin, but its binding is not affected by phosphorylation. Electron microscopy revealed that CRMP-2 localizes on microtubules, clathrin-coated pits, and actin filaments in dorsal root ganglion neuron growth cones, while phosphorylated CRMP-2 localizes only on actin filaments. The phosphomimic mutant of CRMP-2 has a weakened ability to enhance neurite elongation. Furthermore, ephrin-A5 induces phosphorylation of CRMP-2 via Rho kinase during growth cone collapse. Taken together, these results suggest that Rho kinase phosphorylates CRMP-2, and inactivates the ability of CRMP-2 to promote microtubule assembly and Numb-mediated endocytosis, during growth cone collapse.     

The control of microtubule stability in vitro and in transfected cells by MAP1B and SCG10

In neurons, the regulation of microtubules plays an important role for neurite outgrowth, axonal elongation, and growth cone steering. SCG10 family proteins are the only known neuronal proteins that have a strong destabilizing effect, are highly enriched in growth cones and are thought to play an important role during axonal elongation. MAP1B, a microtubule-stabilizing protein, is found in growth cones as well, therefore it was important to test their effect on microtubules in the presence of both proteins. We used recombinant proteins in microtubule assembly assays and in transfected COS-7 cells to analyze their combined effects in vitro and in living cells, respectively. Individually, both proteins showed their expected activities in microtubule stabilization and destruction respectively. In MAP1B/SCG10 double-transfected cells, MAP1B could not protect microtubules from SCG10-induced disassembly in most cells, in particular not in cells that contained high levels of SCG10. This suggests that SCG10 is more potent to destabilize microtubules than MAP1B to rescue them. In microtubule assembly assays, MAP1B promoted microtubule formation at a ratio of 1 MAP1B per 70 tubulin dimers while a ratio of 1 SCG10 per two tubulin dimers was needed to destroy microtubules. In addition to its known binding to tubulin dimers, SCG10 binds also to purified microtubules in growth cones of dorsal root ganglion neurons in culture. In conclusion, neuronal microtubules are regulated by antagonistic effects of MAP1B and SCG10 and a fine tuning of the balance of these proteins may be critical for the regulation of microtubule dynamics in growth cones.     

Microtubule polymer assembly and transport during axonal elongation

As axons elongate, tubulin, which is synthesized in the cell body, must be transported and assembled into new structures in the axon. The mechanism of transport and the location of assembly are presently unknown. We report here on the use of tubulin tagged with a photoactivatable fluorescent group to investigate these issues. Photoactivatable tubulin, microinjected into frog embryos at the two-cell stage, is incorporated into microtubules in neurons obtained from explants of the neural tube. When activated by light, a fluorescent mark is made on the microtubules in the axon, and transport and turnover can be visualized directly. We find that microtubules are generated in or near the cell body and continually transported distally as a coherent phase of polymer during axon elongation. This vectorial polymer movement was observed at all levels on the axon, even in the absence of axonal elongation. Measurements of the rate of polymer translocation at various places in the axon suggest that new polymer is formed by intercalary assembly along the axon and assembly at the growth cone in addition to transport of polymer from the cell body. Finally, polymer movement near the growth cone appeared to respond in a characteristic manner to growth cone behavior, while polymer proximally in the axon moved more consistently. These results suggest that microtubule translocation is the principal means of tubulin transport and that translocation plays an important role in generating new axon structure at the growth cone.     

Regulated actin cytoskeleton assembly at filopodium tips controls their extension and retraction.

The extension and retraction of filopodia in response to extracellular cues is thought to be an important initial step that determines the direction of growth cone advance. We sought to understand how the dynamic behavior of the actin cytoskeleton is regulated to produce extension or retraction. By observing the movement of fiduciary marks on actin filaments in growth cones of a neuroblastoma cell line, we found that filopodium extension and retraction are governed by a balance between the rate of actin cytoskeleton assembly at the tip and retrograde flow. Both assembly and flow rate can vary with time in a single filopodium and between filopodia in a single growth cone. Regulation of assembly rate is the dominant factor in controlling filopodia behavior in our system.     

Interrelationships of tubulin-GDP and tubulin-GTP in microtubule assembly

We previously reported that direct incorporation of GDP (i.e., without an initial hydrolysis of GTP) into microtubules occurs throughout an assembly cycle in a constant proportion. The exact proportion varied with reaction conditions, becoming greater under all conditions in which tubulin-GDP increased relative to tubulin-GTP (low Mg2+ and GTP concentrations, high tubulin concentrations, and in the presence of exogenous GDP). These findings led us to explore further interrelationships of tubulin-GDP and tubulin-GTP in microtubule assembly. We have now determined the minimum amount of tubulin-GTP required for the initiation of microtubule assembly and the relative efficiency with which tubulin-GDP participates in microtubule elongation. When GTP, GDP, and tubulin concentrations were varied at a constant Mg2+ concentration (0.2 mM), initiation of assembly required that 35% of the nucleotide-bearing tubulin be in the form of tubulin-GTP, and incorporation of tubulin-GDP into microtubules during elongation was only 60% as efficient as would be predicted on the basis of its proportional concentration in the reaction mixtures. Very different results were obtained when the Mg2+ concentration was varied. Even though Mg2+ enhances the binding of GTP to tubulin (the equilibrium constant for the exchange of GTP for GDP was 0.2 in the absence of exogenous Mg2+, 3 with 0.2 mM Mg2+, 5 with 0.5 mM Mg2+, and 11 with 2 and 4 mM Mg2+), as Mg2+ was increased the proportion of tubulin-GTP required for the initiation of microtubule assembly rose greatly, and the direct incorporation of tubulin-GDP into microtubules during elongation became progressively more efficient. In the absence of exogenous Mg2+, only 20% tubulin-GTP was required for initiation, and tubulin-GDP was directly incorporated into microtubules half as efficiently as would be predicted on the basis of its concentration in the reaction mixture. At the highest Mg2+ concentration examined (4 mM), 80% tubulin-GTP was required for initiation of assembly, and tubulin-GDP was incorporated into microtubules as efficiently as tubulin-GTP.     

Evidence that calcium may control neurite outgrowth by regulating the stability of actin filaments

We investigated the effects of calcium removal and calcium ionophores on the behavior and ultrastructure of cultured chick dorsal root ganglia (DRG) neurons to identify possible mechanisms by which calcium might regulate neurite outgrowth. Both calcium removal and the addition of calcium ionophores A23187 or ionomycin blocked outgrowth in previously elongating neurites, although in the case of calcium ionophores, changes in growth cone shape and retraction of neurites were also observed. Treatment with calcium ionophores significantly increased growth cone calcium. The ability of the microtubule stabilizing agent taxol to block A23187-induced neurite retraction and the ability of the actin stabilizing agent phalloidin to reverse both A23187-induced growth cone collapse and neurite retraction suggested that calcium acted on the cytoskeleton. Whole mount electron micrographs revealed an apparent disruption of actin filaments in the periphery (but not filopodia) of growth cones that were exposed to calcium ionophores in medium with normal calcium concentrations. This effect was not seen in cells treated with calcium ionophores in calcium-free medium or cells treated with the monovalent cation ionophore monensin, indicating that these effects were calcium specific. Ultrastructure of Triton X-100 extracted whole mounts further indicated that both microtubules and microfilaments may be more stable or extraction resistant after treatments which lower intracellular calcium. Taken together, the data suggest that calcium may control neurite elongation at least in part by regulating actin filament stability, and support a model for neurite outgrowth involving a balance between assembly and disassembly of the cytoskeleton.     

Tubulin cofactor B plays a role in the neuronal growth cone

Tubulin cofactors, initially identified as alpha-, beta-tubulin folding proteins, are now believed to participate in the complex tubulin biogenesis and degradation routes, and thus to contribute to microtubule functional diversity and dynamics. However, a concrete role of tubulin cofactor B (TBCB) remains to be elucidated because this protein is not required for tubulin biogenesis, and it is apparently not essential for life in any of the organisms studied. In agreement with these data, here we show that TBCB localizes at the transition zone of the growth cones of growing neurites during neurogenesis where it plays a role in microtubule dynamics and plasticity. Gene silencing by means of small interfering RNA segments revealed that TBCB knockdown enhances axonal growth. In contrast, excess TBCB, a feature of giant axonal neuropathy, leads to microtubule depolymerization, growth cone retraction, and axonal damage followed by neuronal degeneration. These results provide an important insight into the understanding of the controlling mechanisms of growth cone microtubule dynamics.     

ULTRASTRUCTURE AND FUNCTION OF GROWTH CONES AND AXONS OF CULTURED NERVE CELLS

Dorsal root ganglion nerve cells undergoing axon elongation in vitro have been analyzed ultrastructurally. The growth cone at the axonal tip contains smooth endoplasmic reticulum, vesicles, neurofilaments, occasional microtubules, and a network of 50-A in diameter microfilaments. The filamentous network fills the periphery of the growth cone and is the only structure found in microspikes. Elements of the network are oriented parallel to the axis of microspikes, but exhibit little orientation in the growth cone. Cytochalasin B causes rounding up of growth cones, retraction of microspikes, and cessation of axon elongation. The latter biological effect correlates with an ultrastructural alteration in the filamentous network of growth cones and microspikes. No other organelle appears to be affected by the drug. Removal of cytochalasin allows reinitiation of growth cone-microspike activity, and elongation begins anew. Such recovery will occur in the presence of the protein synthesis inhibitor cycloheximide, and in the absence of exogenous nerve growth factor. The neurofilaments and microtubules of axons are regularly spaced. Fine filaments indistinguishable from those in the growth cone interconnect neurofilaments, vesicles, microtubules, and plasma membrane. This filamentous network could provide the structural basis for the initiation of lateral microspikes and perhaps of collateral axons, besides playing a role in axonal transport.     

Role of tubulin-associated proteins in microtubule nucleation and elongation

Previous experiments have shown that a fraction of microtubule-associated proteins is essential for the self-assembly of microtubules in vitro. When tubulin was titrated with increasing concentrations of these non-tubulin accessory factors, both the rate and extent of polymerization increased in a sigmoidal as opposed to a stoichiometric fashion. The non-tubulin proteins promoted the nucleation of microtubules as determined from the analysis of the kinetics of tubulin selfassembly and the examination of the microtubule length distribution following polymerization. The effect of the non-tubulin factors on microtubule elongation was determined by kinetic experiments in which purified tubulin subunits were added to microtubule seeds and the initial rate of polymerization was measured under conditions where spontaneous self-assembly was below detectable levels. In addition, microtubule growth was also observed when isolated flagellar axonemes were incubated with purified tubulin subunits indicating that the non-tubulin factors were not an absolute requirement for elongation. Analysis of the data in terms of the condensation mechanism of microtubule assembly indicated that the non-tubulin proteins stimulated the growth of microtubules not by increasing the rate of polymerization but by decreasing the rate of depolyerization. The mechanism by which these accessory factors promote tubulin assembly may be summarized as follows: under the conditions employed, they are required for tubulin initiation but not for elongation; the factors affect the extent and net rate at which polymer is formed by binding to the polymer, thereby stabilizing the formed microtubules and consequently shifting the equilibrium to favor assembly.     

Tubulin subunit sorting: Analysis of the mechanisms involved in the segregation of unique tubulin isotypes within microtubule copolymers

Summary Vertebrate cells contain biochemical and genetic isotypes of tubulin which are expressed in unique combinations in different tissues and cell types. To determine if mixtures of tubulin isotypes assemblein vitro to form different classes of microtubules, we analyzed the composition of microtubule copolymers assembled from mixtures of chicken brain and erythrocyte tubulin. During microtubule elongation brain tubulin assembled onto the ends of microtubules faster than erythrocyte tubulin, resulting in copolymers with continually changing ratios of isotypes along their lengths. Unlike examples of microtubule assembly where the rate of polymerization depends on the association rate constant (k⁺) and the subunit concentration, the rate and extent of sorting in copolymers appear to depend on the dissociation rate constant (k^–), which governs the rate at which subunits are released from tubulin oligomers and microtubules and thereby made available for reassembly into copolymers. The type of microtubule seed used to initiate elongation was also found to influence the composition of copolymers, indicating that polymerization favors association of subunits of the same isotype.     

Microtubule behavior in the growth cones of living neurons during axon elongation

To understand how microtubules are generated in the growth cone, we have imaged fluorescently tagged microtubules in living frog embryonic neurons. The neurons were labeled by injecting rhodamine-labeled tubulin into the fertilized egg and explanting the neurons from the neural tube. Microtubules extend deep into the growth cone periphery and adopt three characteristic distributions: (a) dispersed and splayed throughout much of the growth cone; (b) looped and apparently contorted by compression; and (c) bundled into tight arrays. These distributions interconvert on a time scale of several minutes and these interconversions are correlated with the behavior of the growth cone. We observed microtubule growth and shrinkage in growth cones, but are unable to determine their contribution to net assembly. However, translocation of polymer form the axon appears to be a major mechanism of generating new polymer in the growth cone, while bundling of microtubules in the growth cone appears to be the critical step in generating new axon. Neurons that were about to turn spontaneously generated microtubules in the future direction of growth, suggesting that orientation of microtubules might be an important early step in neuronal pathfinding.     

Erythrocyte microtubule assembly in vitro. Determination of the effects of erythrocyte tau, tubulin isoforms, and tubulin oligomers on erythrocyte tubulin assembly, and comparison with brain microtubule assembly

Two tubulin variants, isolated from chicken brain and erythrocytes and known to have different peptide maps and electrophoretic properties, are demonstrated to exhibit different assembly properties in vitro: 1) erythrocyte tubulin assembles with greater efficiency (lower critical concentration, greater elongation rate) but exhibits a lower nucleation rate than brain tubulin, and 2) erythrocyte tubulin readily forms oligomers whose presence significantly retards the rate of elongation, suggesting that tubulin oligomers may also be important for determining the rate of assembly and the length of microtubules in erythrocytes. Erythrocyte tubulin isolated by cycles of in vitro assembly-disassembly is also demonstrated to contain a 67-kDa tau factor that greatly enhances microtubule nucleation but has little effect on elongation rates or critical concentration. Immunofluorescence microscopy with tau antibody indicates that tau is specifically associated with marginal band microtubules, suggesting that it may be important for determining microtubule function in vivo.     

Microtubule dynamics in interphase cells

The sites of microtubule growth and the kinetics of elongation have been studied in vivo by microinjection of biotin-labeled tubulin and subsequent visualization with immunocytochemical probes. Immunofluorescence and immunoelectron microscopy demonstrate that injected biotin-labeled subunits are incorporated into new segments of growth which are contiguous with unlabeled microtubules. Rapid incorporation occurs by elongation of existing microtubules and new nucleation off the centrosome. The growth rate is 3.6 micron/min and is independent of the concentration of injected labeled tubulin. This rate of incorporation together with turnover of existing microtubules leads to approximately 80% exchange in 15 min. The observed kinetics and pattern of microtubule turnover allow for an evaluation of the relevance of several in vitro models for steady-state dynamics to the in vivo situation. We have also observed a substantial population of quasi-stable microtubules that does not exchange subunits as rapidly as the majority of microtubules and may have specialized functions in the cell.     

Antagonistic forces generated by cytoplasmic dynein and myosin-II during growth cone turning and axonal retraction

Cytoplasmic dynein transports short microtubules down the axon in part by pushing against the actin cytoskeleton. Recent studies have suggested that comparable dynein-driven forces may impinge upon the longer microtubules within the axon. Here, we examined a potential role for these forces on axonal retraction and growth cone turning in neurons partially depleted of dynein heavy chain (DHC) by small interfering RNA. While DHC-depleted axons grew at normal rates, they retracted far more robustly in response to donors of nitric oxide than control axons, and their growth cones failed to efficiently turn in response to substrate borders. Live cell imaging of dynamic microtubule tips showed that microtubules in DHC-depleted growth cones were largely confined to the central zone, with very few extending into filopodia. Even under conditions of suppressed microtubule dynamics, DHC depletion impaired the capacity of microtubules to advance into the peripheral zone of the growth cone, indicating a direct role for dynein-driven forces on the distribution of the microtubules. These effects were all reversed by inhibition of myosin-II forces, which are known to underlie the retrograde flow of actin in the growth cone and the contractility of the cortical actin during axonal retraction. Our results are consistent with a model whereby dynein-driven forces enable microtubules to overcome myosin-II-driven forces, both in the axonal shaft and within the growth cone. These dynein-driven forces oppose the tendency of the axon to retract and permit microtubules to advance into the peripheral zone of the growth cone so that they can invade filopodia.     

Microtubule elongation and guanosine 5'-triphosphate hydrolysis. Role of guanine nucleotides in microtubule dynamics

The tubulin concentration dependence of the rates of microtubule elongation and accompanying GTP hydrolysis has been studied over a large range of tubulin concentration. GTP hydrolysis followed the elongation process closely at low tubulin concentration and became gradually uncoupled at higher concentrations, reaching a limiting rate of 35-40 s-1. The kinetic parameters for microtubule growth were different at low and high tubulin concentrations. Elongation of microtubules has also been studied in solutions containing GDP and GTP in variable proportions. Only traces of GTP present in GDP were necessary to confer a high stability (low critical concentration) to microtubules. Pure GDP-tubulin was found unable to elongate microtubules in the absence of GTP but blocked microtubule ends with an equilibrium dissociation constant of 5-6 microM. These data were accounted for by a model within which, in the presence of GTP-tubulin at high concentration, microtubules grow at a fast rate with a large GTP cap; the GTP cap may be quite short in the region of the critical concentration; microtubule stability is linked to the strong interaction between GTP and GDP subunits at the elongating site; dimeric GDP-tubulin does not have the appropriate conformation to undergo reversible polymerization. These results are discussed with regard to possible role of GDP and GTP and of GTP hydrolysis in microtubule dynamics.     

Differential localisation of tyrosinated, detyrosinated, and acetylated alpha-tubulins in neurites and growth cones of dorsal root ganglion neurons

The comparative distribution of tyrosinated, detyrosinated, and acetylated alpha-tubulins was examined in neurites of rat dorsal root ganglion neurones in culture using immunofluorescence microscopy. Phase contrast observations of single neurones revealed that the neurites were actively motile, and rhodamine phalloidin staining of actin filaments showed the extent of lamellopodia and microspike projections from the growth cones. From double-labelling experiments using antibodies against tyrosinated, detryrosinated, or acetylated alpha-tubulin, it was found that the three different isoforms were differentially localised in neurites and growth cones. Detyrosinated and acetylated forms of alpha-tubulin were in the main restricted to the neurites extending no further than the base of the growth cones. Tyrosinated alpha-tubulin was, however, distributed throughout the body of the growth cone and into the base of some microspikes. Following treatment with taxol to promote microtubule assembly, detyrosinated and acetylated alpha-tubulins were found to be colocalised with tyrosinated alpha-tubulins throughout the growth cones of all cells examined. These results would be consistent with axonal transport of tyrosinated alpha-tubulin followed by assembly in the growth cone and subsequent detyrosination and acetylation. In addition the presence of unmodified alpha-tubulin in the growth cone may be necessary for the provision of labile microtubules for growth cone motility and extension.     

The sequence of a region of bacteriophage phiX174 DNA coding for parts of genes A and B

The kinetics of microtubule assembly were investigated by monitoring changes in turbidity which result from the scattering of incident light by the polymer. These studies indicated that assembly occurred by a pathway involving a nucleation phase, followed by an elongation phase as evidenced by a lag in the polymerization kinetics, followed by a psuedo-first-order exponential increase in turbidity. Analytical ultracentrifugation of solutions polymerized to equilibrium showed that 6 S tubulin was the only species detectable in equilibrium with microtubules. Investigation of the elongation reaction in mixtures of 6 S tubulin and microtubule fragments demonstrated that: (1) the net rate of assembly was the sum of the rates of polymerization and depolymerization; (2) the rate of polymerization was proportional to the product of the microtubule number concentration and the 6 S tubulin concentration; and (3) the rate of depolymerization was proportional to the number concentration of microtubules. These results demonstrate that microtubule assembly occurs by a condensation polymerization mechanism consisting of distinct nucleation and elongation steps. Microtubules are initiated in a series of protein association reactions in a pathway that has not been fully elucidated. Elongation proceeds by the consecutive association of 6 S tubulin subunits onto the ends of existing microtubules. Similarly, depolymerization occurs by dissociation of 6 S subunits from the ends of microtubules. The rate constants measured for polymerization and depolymerization at 30 °C were 4 × 10⁶m^?1 s^?1 and 7 s^?1, respectively.     

A reassessment of the factors affecting microtubule assembly and disassembly in vitro

Current models of microtubule assembly from pure tubulin involve a nucleation phase followed by microtubule elongation at a constant polymer number. Both the rate of microtubule nucleation and elongation are thought to be tightly influenced by the free GTP-tubulin concentration, in a law of mass action-dependent manner. However, these basic hypotheses have remained largely untested due to a lack of data reporting actual measurements of the microtubule length and number concentration during microtubule assembly.Here, we performed simultaneous measurements of the polymeric tubulin concentration, of the free GTP-tubulin concentration, and of the microtubule length and number concentration in both polymerizing and depolymerizing conditions. In agreement with previous work we find that the microtubule nucleation rate is strongly dependent on the initial GTP-tubulin concentration. But we find that microtubule nucleation persists during microtubule elongation. At any given initial tubulin-GTP concentration, the microtubule nucleation rate remains constant during polymer assembly, despite the wide variation in free GTP-tubulin concentration. We also find a remarkable constancy of the rate of microtubule elongation during assembly. Apparently, the rate of microtubule elongation is intrinsic to the polymers, insensitive to large variations of the free GTP-tubulin concentration. Finally we observe that when, following assembly, microtubules depolymerize below the free GTP-tubulin critical concentration, the rate-limiting factor for disassembly is the frequency of microtubule catastrophe. At all time-points during disassembly, the microtubule catastrophe frequency is independent of the free GTP-tubulin concentration but, as the microtubule nucleation rate, is strongly dependent on the initial free GTP-tubulin concentration. We conclude that the dynamics of both microtubule assembly and disassembly depend largely on factors other than the free GTP-tubulin concentration. We propose that intrinsic structural factors and endogenous regulators, whose concentration varies with the initial conditions, are also major determinants of these dynamics.     

Polarized compartmentalization of organelles in growth cones from developing optic tectum

We have used computer-assisted reconstructions of continuous serial sections to study the cytoplasmic organization of growth cones in vivo. Optic tecta from 6.25-6.5-d-old chicken embryos were quick-frozen and then freeze-substituted in acetone-osmium tetroxide or, for comparison, prepared by conventional fixation. Images of eight freeze-substituted and two conventionally fixed growth cones were reconstructed from aligned serial micrographs. After freeze-substitution, numerous lumenless membrane-bound sacs arrayed in multilamellar stacks appear to replace the abundant smooth endoplasmic reticulum found after chemical fixation. Microtubule fascicles progressively diverge from their typical fascicular organization in the initial segment of the growth cone and are absent in the varicosity and the more distal segment. Mitochondria, in contrast, are concentrated in the proximal segment of the varicosity; multilamellar stacks and endosome-like vacuoles are in the distal segment; and coated pits and vesicles are concentrated near the terminal filopodium, which is the most distal and organelle-poor domain of the growth cone. These observations suggest that dilation and fusion of the lumenless, membrane-bound sacs that occurs during chemical fixation give rise to the network of smooth endoplasmic reticulum. The three-dimensional reconstructions show that the cytoplasmic components of growth cones, including the membrane-bound sacs and multilamellar stacks revealed by freeze substitution, are polarized along the axis of these growth cones, which suggests that they have a role in recycling of membrane during elongation of the growth cone.