Observations on naturally occurring gaps in the axonal neurofilament array of cultured neurons have demonstrated that neurofilament polymers move along axons in a rapid, intermittent, and highly asynchronous manner. In contrast, studies on axonal neurofilaments using laser photobleaching have not detected movement. Here, we describe a modified photobleaching strategy that does permit the direct observation of neurofilament movement. Axons of cultured neurons expressing GFP-tagged neurofilament protein were bleached by excitation with the mercury arc lamp of a conventional epifluorescence microscope for 12-60 s. The length of the bleached region ranged from 10 to 60 m. By bleaching thin axons, which have relatively few neurofilaments, we were able to reduce the fluorescent intensity enough to allow the detection of neurofilaments that moved in from the surrounding unbleached regions. Time-lapse imaging at short intervals revealed rapid, intermittent, and highly asynchronous movement of fluorescent filaments through the bleached regions at peak rates of up to 2.8 m/s. The kinetics of movement were very similar to our previous observations on neurofilaments moving through naturally occurring gaps, which indicates that the movement was not impaired by the photobleaching process. These results demonstrate that fluorescence photobleaching can be used to study the slow axonal transport of cytoskeletal polymers, but only if the experimental strategy is designed to ensure that rapid asynchronous movements can be detected. This may explain the failure of previous photobleaching studies to reveal the movement of neurofilament proteins and other cytoskeletal proteins in axons.
INTRODUCTIONSlow axonal transport is the mechanism by which cytoskeletal and cytosolic proteins are transported along axons from their site of synthesis in the nerve cell body. During the past decade there have been numerous efforts to observe this movement directly in living cells, but these studies (Baas and Brown, 1997;Hirokawa et al., 1997) have yielded conflicting results. The two techniques that have been used the most widely are fluorescence photobleaching and photoactivation. In these related approaches, fluorescent or caged fluorescent cytoskeletal subunit proteins are injected into nerve cells and then a laser is used to bleach or activate the fluorescence in a short segment of axon. The bleached or activated proteins then are observed by time-lapse imaging to detect their movement.The first photobleaching study (Keith, 1987) on slow axonal transport reported slow and synchronous movement of tubulin in cultured PC12 cells, but attempts to reproduce this result in cultured PC12 cells and cultured chick and mouse sensory neurons were unsuccessful (Lim et al., 1989;Lim et al., 1990; Hirokawa, 1990, 1992). Subsequent photobleaching and photoactivation studies (Sabry et al., 1995;Takeda et al., 1995) on tubulin in the motor neurons of developing grasshopper and zebrafish embryos, as well as photobleaching studies (Okabe and Hirokawa, 1990;Takeda ...