PolyQDependent Misfolding and Aggregation

The presence of an expanded polyQ tract invariably results in protein mis-folding, but protein context modulates both repeat threshold and the kinetics of aggregation. For example, the most common TBP allele in Caucasians contains 38 polyQ-encoding d(CAG) repeats, while a polyQ stretch of this length would be conducive to aggregation as well as pathogenic in five of the eight remaining polyQ disease proteins (Reid et al. 2003). PolyQ aggregation has been investigated intensively in vitro by use of synthetic polyQ peptides and recombinant polyQ proteins as well as cellular models of polyQ disease (Pe-rutz et al. 1994; Scherzinger et al. 1997; Hackam et al. 1999; Poirier et al. 2005). Max Perutz, the esteemed structural biologist, provided some of the earliest theoretical insight as well as empirical data regarding the nature of polyQ interactions. On the basis of molecular modeling, he suggested that polyQ domains might self-associate as antiparallel P-strands that are connected by an elaborate array of hydrogen bonds involving both main chain and side chain amide groups. Thus, by analogy to leucine zippers that link a-helices, Perutz envisioned a polar zipper structure for polyQ aggregates (Perutz et al. 1994).

PolyQ aggregates resemble amyloids in appearance in electron micrographs and display some of the same histochemical and kinetics properties in vitro (Scherzinger et al. 1997, 1999; Chen et al. 2001, 2002a,b). Aggregation of synthetic polyQ peptides occurs by nucleation-dependent polymerization. Specifically, an initial nucleation event, which may actually involve the mis-

folding of a single polyQ monomer rather than the formation of an unstable oligomer, is followed by the rapid addition of polyQ monomer in the elongation phase (Chen et al. 2002b). Whereas fast elongation ensures that a single aggregate forms in vitro, multiple nucleation events may occur in the neurons of patients or mouse models. Also, the elongation process is markedly protracted in the context of the cellular environment. In neuronal nuclei, individual aggregates, after slowly polymerizing as separate entities, may eventually fuse to form a single, large inclusion. Notably, this progression would be consistent with the histochemical time course of nuclear polyQ accumulation detailed already (Michalik and Broeckhoven 2003). Furthermore, as the rate of nucleus formation is directly related to the length of the synthetic polyQ tract, it has been suggested that polyQ aggregation kinetics may underlie the correlation between repeat length and the age of onset in the polyQ diseases (Chen et al. 2002b).

In a cellular model of HD, the cytoplasmic and nuclear environments are not differentially conducive to aggregate formation. Moreover, the subcellu-lar localization of aggregates, which can be manipulated by the attachment of either a nuclear localization signal or a nuclear export signal to mutant htt fragments, does not modulate the toxicity of expanded polyQ in cultured cells (Hackam et al. 1999). Nevertheless, there are some differences in the nature of nuclear and cytoplasmic aggregates. The latter are generally smaller in size, at least when present in neuronal processes (Li et al. 1999). Additionally, whereas NII in all of the polyQ diseases colocalize with ubiquitin, perikaryal aggregates in SCA2 (Huynh et al. 2000) and SCA6 (Ishikawa et al. 1999) neurons and neuropil htt aggregates lack this decoration (Gutekunst et al. 1999; Li et al. 1999). The presence of ubiquitin as well as proteaso-mal subunits in NII probably indicates the involvement of the ubiquitin-proteasome system (UPS) in aggregate clearance (Everett and Wood 2004). However, markedly reduced nuclear aggregation is observed in SCA1 trans-genic mice that lack the E6-AP ubiquitin ligase (Cummings et al. 1999). This paradoxical finding suggests that ubiquination may actually stabilize aggregates in some fashion.

Whereas polyQ expansion in nuclear proteins can be sufficient to produce NII, the same is not true in large, cytoplasmic polyQ proteins, like htt and atrophin-1. Rather, proteolytic processing may be a prerequisite for nuclear accumulation as well as intraceullular aggregate formation by the latter. Loss of htt and atrophin-1 carboxy-terminal antigenicity in NII is consistent with the occurrence of potentially extensive processing (Schilling et al. 1999a,b; Gutekunst et al. 1999), and cleavage sites for various proteases have been identified in vitro (Kim et al. 2001; Gafni and Ellerby 2002; Wellington et al. 2002; Lunkes et al. 2002; Zhou et al. 2003; Nucifora et al. 2003; Gafni et al. 2004). Perikaryal and neuropil htt aggregates also consist of polyQ-containing fragments (DiFiglia et al. 1997; Gutekunst et al. 1999). Similarly, in vitro experiments (Wellington 1998) and immunhistochemical examination of NII in brains of patients have revealed evidence of proteolytic cleavage of other polyQ-expanded proteins, including ataxin-3 (Goti et al. 2004), ataxin-7 (Garden et al. 2002), and androgen receptor (Li et al. 1998). It is possible that conformational differences between soluble and aggregated proteins may contribute to some of the discrepancies in immunoreactivity. Nevertheless, comparison of transgenic mice expressing full-length mutant htt (YAC 46, 72, and 128) and N-terminal htt fragments (N171 or R6/2 lines) demonstrates that truncated polyQ proteins with an expanded polyQ tract are not only sufficient to induce aggregation and neuropathology but may actually be more toxic than their unprocessed counterparts (Mangiarini et al. 1996; Schilling et al. 1999a,b; Hodgson et al. 1999; Slow et al. 2003).

In addition to various components of the UPS, polyQ aggregates are also immunoreactive for a number of molecular chaperones, including, most notably, Hsp40 and Hsp70 family members. The former, which are considered cochaperones, recognize and deliver misfolded proteins to the latter. Hsp70 chaperones have an intrinsic ATPase activity that facilitates refolding; however, recalcitrant proteins are ubiquitinated and targeted to the proteasome for degradation. In sum, the colocalization data, which have been collected from cellular (Wyttenbach et al. 2000; Chai et al. 1999a,b; Stenoien et al. 1999; Jana et al. 2001) and mouse models (Jana et al. 2001; Hay et al. 2004; Adachi et al. 2003; Cummings et al. 1998) as well as brain tissue of patients (Chai et al. 1999a,b; Cummings et al. 1998), indicate that polyQ aggregates trigger the normal cellular response to misfolded protein (Fig. 2). Screens for genetic modifiers of polyQ-induced neurodegeneration have substantiated the involvement of the protein folding machinery (Fernandez-Funez et al. 2000; Kazemi-Esfarjani and Benzer 2000). Although biochemical purification of polyQ aggregates is indicative of sequestration (Suhr et al. 2001), live cell imaging has demonstrated that the interaction between Hsp70 and these structures can be dynamic (Kim et al. 2002).

The refolding and clearance of misfolded polyQ proteins by chaperones and the UPS, respectively, may impact the subcellular distribution of mutant polyQ. Biochemical analysis of brains from HD repeat knockin mice, in which a 150 d(CAG) repeat is present in the endogenous mouse htt (Hdh) gene, indicates that a collection of truncated htt fragments accumulate in neuronal nuclei in association with an age-dependent decrease in proteasomal function (Zhou et al. 2003). Most of these N-terminal htt fragments are smaller than the size threshold for passive diffusion through the nuclear pore complex, and recent evidence suggests that their entry into the nucleus occurs by a Ran GTPase-independent process. PolyQ expansion decreases the interaction of N-terminal htt with a component of the nuclear export machinery, which can explain the accumulation and concomitant aggregation of mutant htt fragments in the nucleus (Cornett et al. 2005). It is unclear if this mechanism of nuclear accumulation applies to other polyQ disease proteins, particularly those that normally localize to the nucleus.

Fig. 2 Intracellular aggregation of expanded polyQ proteins. PolyQ aggregates can form in both the cytoplasm and the nucleus, depending on the polyQ disease protein. The initial step in aggregate formation is polyQ-mediated protein misfolding (black lines). Molecular chaperones recognize misfolded polyQ proteins and attempt to reintroduce the proper conformation. Chaperone substrates that cannot be refolded are targeted to the protea-some for degradation (light dashed line). Importantly, misfolded polyQ proteins that are refractory to refolding can aggregate (dark dotted line) if not degraded. Moreover, age-dependent decline in proteasome function would result in increased aggregation over time. Proteolytic processing often precedes polyQ-mediated aggregate formation in both the nucleus and the cytoplasm and may be a prerequisite for the nuclear accumulation of mutant atrophin-1 and mutant huntingtin (htt). Nuclear entry (solid line) can be facilitated by classic nuclear localization signals or, at least in the case of htt fragments, may occur by a Ran GTPase-independent process (see text for details). NPC nuclear pore complex, HSP heat shock protein

Fig. 2 Intracellular aggregation of expanded polyQ proteins. PolyQ aggregates can form in both the cytoplasm and the nucleus, depending on the polyQ disease protein. The initial step in aggregate formation is polyQ-mediated protein misfolding (black lines). Molecular chaperones recognize misfolded polyQ proteins and attempt to reintroduce the proper conformation. Chaperone substrates that cannot be refolded are targeted to the protea-some for degradation (light dashed line). Importantly, misfolded polyQ proteins that are refractory to refolding can aggregate (dark dotted line) if not degraded. Moreover, age-dependent decline in proteasome function would result in increased aggregation over time. Proteolytic processing often precedes polyQ-mediated aggregate formation in both the nucleus and the cytoplasm and may be a prerequisite for the nuclear accumulation of mutant atrophin-1 and mutant huntingtin (htt). Nuclear entry (solid line) can be facilitated by classic nuclear localization signals or, at least in the case of htt fragments, may occur by a Ran GTPase-independent process (see text for details). NPC nuclear pore complex, HSP heat shock protein

Was this article helpful?

0 0
Confident Kids

Confident Kids

Although nobody gets a parenting manual or bible in the delivery room, it is our duty as parents to try to make our kids as well rounded, happy and confident as possible. It is a lot easier to bring up great kids than it is to try and fix problems caused by bad parenting, when our kids have become adults. Our children are all individuals - they are not our property but people in their own right.

Get My Free Ebook


Post a comment