“The CAP-Gly of p150: one domain, two diseases, and a function at the end”

Abstract

In this issue of Neuron, two manuscripts reveal a novel role for p150 in initiation of retrograde transport at synaptic terminals. Moreover, insights are revealed on how mutations of p150's CAP-Gly domain lead to both Perry Syndrome and HMN7B disease.

Although most cells are measured in microns, neurons, especially peripheral neurons, can be a meter long and therefore make extreme demands on our molecular motors. Small wonder that mutations in ubiquitous motor proteins give rise to specifically neurological diseases. Two such diseases, Perry Syndrome and the distal Hereditary Motor Neuropathy VIIB (HMN7B), are examples of that phenomenon and their cell biological basis has been examined by two papers in this issue (Moughamian & Holzbaur and Lloyd et al., 2012). Although their symptoms are quite different, both diseases are caused by mutations in the same domain of the dynactin subunit p150/Glued. By approaching the function of this domain in Drosophila neurons and mouse dorsal root ganglion (DRG) neurons, the present studies illuminate the function of p150/Glued in axonal transport.

Axonal microtubules are uniformly polarized with their (+) ends away from the soma. Two classes of motor, kinesins and cytoplasmic dynein, move along these microtubule tracks to transport cargo between the soma and nerve terminals. Retrograde, (-) end-directed transport is performed by dynein. Two important functions of retrograde transport are escorting aggregated/misfolded proteins back to the soma for degradation (Johnston et al., 2002) and communicating synaptic and trophic signals to the soma to regulate gene expression (reviewed by Cosker et al., 2008). The dynein motors are multi-subunit complexes and much of the complex remains poorly understood. Moreover, dynein does not act alone; it acts in a complex with a second multimeric protein assembly known as dynactin. The largest subunit of dynactin is p150, the mammalian homolog of the Drosophila Glued gene (Holzbaur et al., 1991). Dynactin is mainly thought to be required for attaching cargo to dynein with p150 forming the dynein-dynactin link (Karki and Holzbaur, 1995; Vaughan and Vallee, 1995). Additional dynein-independent functions of p150 have been reported that involve organizing microtubule arrays and anchoring microtubules at the centrosome (Askham et al., 2002;Quintyne et al., 1999).

The cytoskeletal functions of p150 rely on its N-terminal, cytoskeleton-associated protein/glycine-rich (CAP-Gly) domain (Figure 1A). Those interactions suggested that p150 anchors dynein to microtubules and thereby increases processivity-- the number of consecutive steps a motor takes before falling off the microtubules. Purified dynein was much less processive in vitro when either p150 was absent or the CAP-Gly domain was inhibited (Ross et al., 2006; and references therein). In vivo, however, dynein's processivity was unperturbed when p150's CAP-Gly domain was deleted (Kim et al., 2007). What then is the purpose of p150's CAP-Gly domain? One possibility was that it was required only at the (+) ends of microtubules and not for processivity along their tracks. A small population of p150 localizes to the (+) ends and p150's (+)-end binding is regulated by phosphorylation of a serine within the CAP-Gly domain (Vaughan et al., 2002). Moreover, p150 directly interacts with the (+)-end binding proteins EB1, EB3, and CLIP170 (Lansbergen et al., 2004;Ligon et al., 2003). In this issue, both Moughamian & Holzbaur and Lloyd et al. examine the requirement of the CAP-Gly domain in retrograde axonal transport. Knockdown of p150 in both fly and mouse neurons disrupted axonal transport and provided systems in which to restore a deleted p150. Both groups report that wild-type p150 and p150 lacking the CAP-Gly domain (ΔCAP-Gly) could equally rescue much of the p150 knockdown phenotype; the CAP-Gly domain was not required for axonal transport or dynein processivity. However, the large accumulations of p150 that normally occur at the (+) ends of wild type axons, in tips of distal neurites or in terminal synaptic boutons, was dependent on the presence of the CAP-Gly domain and, at least in the DRG neurons, required EB1 and EB3. They hypothesized that p150 might act in these neurite tips to recruit dynein and thereby initiate retrograde transport. Retrograde flux, or the net movement of cargo from the (+) end to the soma, was therefore examined by fluorescently labeling lysosomes in DRG neurons and endosomes in the fly. By photobleaching a region close to the neurite tip and then watching the transit of those cargoes through the bleached region, both groups observed that, in the absence of CAP-Gly domain, these organelles were not leaving the endings in appropriate numbers, although they were correctly delivered to the distal tips. Promoting the initiation of retrograde transport represents a new neuronal function for p150's CAP-Gly domain.

Figure 1.

The CAP-Gly domain of p150 is required for initiation of dynein loading at (+) ends

If this is the main function of the CAP-Gly domain and anterograde transport is unaffected, one would expect distal accumulations of dynein and its cargo. In fact, Lloyd et al. noticed gross accumulations of endosome components, neuronal membranes, and dynein in the distal boutons of fly neurons when the CAP-Gly domain was lacking. Moughamian & Holzbaur also looked for such accumulations in DRG neurons, but didn't see them. This phenotypic distinction is a curious difference between the studies but may not reflect a species difference in the function of the domain so much as the conditions studied. The fly neuromuscular junction has differentiated terminal boutons in which the cargo piles up but the ongoing axonal growth in DRG cultures may have allowed dynein and its cargoes to be dispersed as the neurite extended. Significant differences may nonetheless exist in the manner in which cells handle the initiation of retrograde transport. In mammalian neurons, although p150 is enriched at (+) ends, little dynein accumulates. p150 at the (+) ends may capture and rapidly tether arriving dynein for the immediate initiation of cargo loading and retrograde transport. However, in fungi, not just p150, but all of the dynactin/dynein complex and LIS1 are enriched at hyphal (+) end tips through an interaction of the EB1-like fungal protein, Peb1, and p150's CAP-Gly domain. In those cells, dynactin and dynein are delivered but are not released for retrograde transport until triggered by the separate delivery of early endosomes (Lenz et al., 2006). Thus some cell types may elect only to keep dynactin on hand at the (+) end (through the EB1/EB3/p150 interaction-Figure 1B), while other cells store dynein there as well. The mechanism regulating initiation of motor activity will likely differ between cell types.

Both the Perry Syndrome and HMN7B mutations occur within the CAP-Gly domain of p150 (Figure 1A) and both are autosomal dominant diseases, but whereas HMN7B, like Amyotrophic Lateral Sclerosis, causes degeneration specifically of motor neurons, Perry Syndrome most prominently affects the substania nigra and brainstem and causes Parkinsonian symptoms. To study Perry Syndrome and HMN7B, both groups expressed transgenes containing the pathogenic point mutations. By examining retrograde flux, both groups found that the disease mutations perturbed the ability of p150 to associate with microtubules and observed problems with the initiation of retrograde transport. Why then do they cause such different symptoms in humans? Both groups noted that protein aggregates formed when these alleles were expressed, but that this tendency, particularly in neurons, was more pronounced for the HMN7B mutations. This distinction correlates with the histopathology of affected individuals. Potentially more enlightening, however, were biochemical studies by Moughamian & Holzbaur. Although both Perry and HMN7B mutations allow p150 to dimerize and incorporate into the dynactin complex, the HMN7B mutation alone prevents the dynactin complex from binding to dynein. Whereas the Perry Syndrome mutations lie on the surface, in or very close to the site of microtubule and EB1 binding, the HMN7B mutation is in the core of the domain and likely to interfere with its folding. Thus, although CAP-Gly domain is far from the known dynein-interacting portion of p150, the likely severe misfolding of this domain may promote its aggregation and prevent proper incorporation into the motor. These biochemical changes are reflected in phenotypic differences observed in these studies. In DRG neurons, the HMN7B mutation seriously perturbed both anterograde and retrograde transport and decreased the processivity of cargo, as might be expected if dynein was operating without its dynactin partner. This defect did not arise when the Perry Syndrome allele was expressed. In Drosophila, only the HMN7B mutation caused dynein heavy chain to accumulate substantially in the terminal boutons, as might be expected if the dynein motor is bereft of dynactin association. Thus HMN7B may be understood as a dominant negative that compromises the entire function of the dynactin complex, while Perry Syndrome selectively impairs retrograde initiation while leaving other functions of dynactin intact.

Of course several questions remain unanswered. Most particularly, we don't yet know why the broader disruption of dynactin function is most manifest in the substantia nigra and brainstem while the motor neurons are most sensitive to the subtler impairment of retrograde initiation. That puzzle vexes most discussions of neurodegenerative disease. The specificities may arise from differences in the dependence of neuronal subtypes on retrograde transport of survival signals or in their sensitivity to protein aggregates. Mechanistic questions also remain: how does p150 regulate transport at terminal boutons? Does the phosphorylation of p150's CAP-Gly domain trigger the release of the entire dynein/dynactin complex from the (+) ends? Several microtubule (+) end-binding proteins (EB1, EB3, CLIP170 and LIS1) may regulate initiation of transport in conjunction with p150 perhaps by allowing p150 to bind (+) ends, to capture dynein, or even to be released from the (+) ends once the dynactin/dynein complex is formed. The present papers, however, by examining rigorously the cell biology of these mutations and the CAP-Gly domain itself have opened doors to further understanding retrograde movement and stress the importance of maintaining a finely tuned axonal transport system.