Membrane tethers are extracted at constant velocity from neuronal growth cones using a force generated by a laser tweezers trap. A thermodynamic analysis shows that as the tether is extended, energy is stored in the tether as bending and adhesion energies and in the cell body as "nonlocal" bending. It is postulated that energy is dissipated by three viscous mechanisms including membrane flow, slip between the two monolayers that form the bilayer, and slip between membrane and cytoskeleton. The analysis predicts and the experiments show a linear relation between tether force and tether velocity. Calculations based on the analytical results and the experimental measurements of a tether radius of ～0.2 μm and a tether force at zero velocity of ～8 pN give a bending modulus for the tether of 2.7×10-19 N-m and an extraordinarily small "apparent surface tension" in the growth cone of 0.003 mN/m, where the apparent surface tension is the sum of the far-field, in-plane tension and the energy of adhesion. Treatments with cytochalasin B and D, ethanol, and nocodazole affect the apparent surface tension but not bending. ATP depletion affects neither, whereas large concentrations of DMSO affect both. Under conditions of flow, data are presented to show that the dominant viscous mechanism comes from the slip that occurs when the membrane flows over the cytoskeleton. ATP depletion and the treatment with DMSO cause a dramatic drop in the effective viscosity. If it is postulated that the slip between membrane and cytoskeleton occurs in a film of water, then this water film has a mean thickness of only ～10 A.

The plasma membrane of most cells is drawn tightly over the cytoskeleton of the cell, resulting in a significant tension being developed in the membrane. The tension in the membrane can be calculated fi-om the force required to separate it ffom the cytoskeleton; and the force itself can be measured rapidly by using laser tweezers. Recent observations indicate that decreasing membrane tension stimulates endocytosis and increasing tension stimulates secretion. Thus, membrane tension provides a simple physical mechanism to control the area of the plasma membrane. Here, we speculate that tension is a global parameter that the cell uses to control physically plasma membrane dynamics, cell shape and cell motility.

During the growth of axons, the surface area of the neuron increases dramatically. Membrane addition as well as exchange could contribute to rapid membrane dynamics or flow. Using diffusing latex beads to monitor membrane flow, we find that axonal membrane flows rapidly (7 pm/mm) from growth cone to cell body during axon growth and that flow is inhibited by brefeldin A. To power this flow, there is a membrane tension gradient from growth cone to cell body that could draw the membrane overtheaxon at that rate. Further, when an artificial flow is induced to the center of the axon by use of laser tweezers, the primary source of the membrane is from the growth cone. We suggest that during neuron growth, there is excess membrane added at the growth cone in chick dorsal root ganglia (DRGs) that undergoes endocytosis at the cell body, thereby creating a flow that can rapidly alter the content of the axon membrane.

Many cell phenomena involve major morphological changes, particularly in mitosis and the process of cell migration. For cells or neuronal growth cones to migrate, they must extend the leading edge of the plasma membrane as a lamellipodium or filopodium. During extension of filopodia, membrane must move across the surface creating shear and flow. Intracellular biochemical processes driving extension must work against the membrane mechanical properties, but the forces required to extend growth cones have not been measured. In this paper, laser optical tweezers and a nanometer-level analysis system were used to measure the neuronal growth cone membrane mechanical properties through the extension of filopodia-like tethers with IgG-coated beads. Although the probability of a bead attaching to the membrane was constant irrespective of treatment; the probability of forming a tether with a constant force increased dramatically with cytochalasin B or D and dimethylsulfoxide (DMSO). These are treatments that alter the organization of the actin cytoskeleton. The force required to hold a tether at zero velocity (F0) was greater than forces generated by single molecular motors, kinesin and myosin; and F0 decreased with cytochalasin B or D and DMSO in correlation with the changes in the probability of tether formation. The force of the tether on the bead increased linearly with the velocity of tether elongation. From the dependency of tether force on velocity of tether formation, we calculated a parameter related to membrane viscosity, which decreased with cytochalasin B or D, ATP depletion, nocodazole, and DMSO. These results indicate that the actin cytoskeleton affects the membrane mechanical properties, including the force required for membrane extension and the viscoelastic behavior.

Development of the nervous system requires that neuronal growth cones, in coordination with growing axons, migrate along precise paths defined by specific extracellular matrix cues until they ncounter their targets. Laminin promotes growth cone migration through receptors such as the integrins, but the underlying physical mechanism is poorly understood. We have investigated the cytoskeletal associations and surface dynamics of endogenous β1 integrins in chick dorsal root ganglion growth cones migrating on laminin. A single-beam optical gradient trap was used to place 0.5-µm-diameter polystyrene beads conjugated with anti-β1 integrin monoclonal antibodies at desired locations on the growth cone surface. We found a substantial increase in the stable attachment of these beads, with subsequent slow rearward motion, on the front periphery of the growth cone compared to the ase. The surface dynamics of smaller aggregates of integrin were explored by monitoring the temporal and spatial displacements of 40-nm-diameter gold particles coated with anti-β1 integrin antibodies. The small particles were transported preferen tially to the growth cone periphery by brief directed xcursions interspersed with periods of diffusion. In addition, the leading edge of the growth cone was supported to a greater extent by an actin-dependent cytoskeleton that resisted mechanical tether formation. Such a regional differentiation of the growth cone has not been documented previously and has implications for the mechanism of growth cone migration and guidance.