We examined the respective roles of dynein and kinesin in axonal transport of neurofilaments (NFs). Differentiated NB2a/d1 cells were transfected with green fluorescent protein-NF-M (GFP-M) and dynein function was inhibited by co-transfection with a construct expressing myc-tagged dynamitin, or by intracellular delivery of purified dynamitin and two antibodies against dynein's cargo domain. Monitoring of the bulk distribution of GFP signal within axonal neurites, recovery of GFP signal within photobleached regions, and real-time monitoring of individual NFs/punctate structures each revealed that pertubation of dynein function inhibited retrograde transport and accelerated anterograde, confirming that dynein mediated retrograde axonal transport, while intracellular delivery of two anti-kinesin antibodies selectively inhibited NF anterograde transport. In addition, dynamitin overexpression inhibited the initial translocation of newly-expressed NFs out of perikarya and into neurites, indicating that dynein participated in the initial anterograde delivery of NFs into neurites. Delivery of NFs to the axon hillock inner plasma membrane surface, and their subsequent translocation into neurites, was also prevented by vinblastine-mediated inhibition of microtubule assembly. These data collectively suggest that some NFs enter axons as cargo of microtubues that are themselves undergoing transport into axons via dynein-mediated interactions with the actin cortex and/or larger microtubules. C-terminal NF phosphorylation regulates motor association, since anti-dynein selectively coprecipitated extensively phosphorylated NFs, while anti-kinesin selectively coprecipitated less phosphorylated NFs. In addition, however, the MAP kinase inhibitor PD98059 also inhibited transport of a constitutively-phosphorylated NF construct, indicating that one or more additional, non-NF phosphorylation events also regulated NF association with dynein or kinesin.

Overexpression of tau compromises axonal transport and induces retraction of growing neurites. We tested the hypothesis that increased stability provided by neurofilaments (NFs) may prevent axonal retraction. NB2a/d1 cells were differentiated for 3 days, at which time phosphorylated NFs appear and for 14 days, which induces continued neurite elongation and further phospho-NF accumulation. Cultures were transfected with a construct that expresses full-length, 4-repeat tau. Consistent with prior studies, overexpression of tau induced retraction of day three axonal neurites even following treatment with the microtubule-stabilizing drug taxol. Axonal neurites of day 14 cells were more resistant to tau-mediated retraction. To test whether or not this resistance was derived from their additional NF content, day 3 cultures were co-transfected with constructs expressing tau and NF-M (which increases overall axonal NFs). Overexpression of NF-M attenuated tau-mediated retraction of day 3 axonal neurites. By contrast, co-transfection with constructs expressing tau and vimentin (which increases axonal neurites length) did not attenuate tau-mediated neurite retraction. Co-precipitation experiments indicate that tau is a cargo of kinesin, and that tau overexpression may displace other kinesin-based cargo, including both critical cytoskeletal proteins and organelles. However, cultures simultaneously transfected with constructs expressing NF-M and tau, the level of examined vesicles was maintained. These collectively indicate that NFs stabilize developing axonal neurites and can counteract the destabilizing force resulting from overexpression of tau, and underscore that the development and stabilization of axonal neurites is dependent upon a balance of cytoskeletal elements.

Vimentin (Vm) is initially expressed by early neuronal precursors in situ and in culture. Vm is essential for neuritogenesis at least in culture and is gradually replaced by neurofilaments (NFs) because of down-regulation of Vm expression. This period is accompanied by a slowing of axonal elongation. We examined whether continued expression of Vm could foster continued axonal elongation. NB2a/d1 cells differentiated with dibutyryl cAMP were transfected with constructs expressing Vm or the middle-molecular-weight NF subunit (NF-M) each conjugated to green fluorescent protein (GFP). Axonal neurites of cells expressing GFP-Vm were 30% longer than those of nonexpressing cells, or cells expressing GFP-M, and exhibited a decrease in neurite caliber. Expression of GFP-M did not enhance axonal neurite length but significantly increased caliber. These findings provide further evidence of a role for Vm in axonal outgrowth. Culturing of nontransfected cells on laminin increased neurite length, but cells expressing GFP-Vm demonstrated an equivalent increase whether cultured on laminin or culture plastic. Axonal neurites of cells expressing GFP-Vm turned to avoid a nonfavorable substrate (nitrocellulose), but culturing of these cells on nitrocellulose did not impair axonal outgrowth. These latter findings indicate that the more robust outgrowth following reexpression of Vm is independent of a favorable or nonfavorable substrate but that axonal neurites of these cells still interact with the substrate to the extent that the substrate can influence directionality.

Multi-electrode array systems have enabled the in vitro electrophysiological study of neuronal networks. The data processing component of these systems consists of an advanced computer system and data acquisition electronics that collectively cost more than the multi-electrode arrays and amplifiers. Considering that these elaborate systems may be cost-prohibitive for many laboratories, we have developed a simple but novel method for recording groups of related multi-electrode array channels with a low-cost data acquisition system.

No abstract