For Protein Misassembly, It's the “I” Decade

(Mitraki and King, 1989). For example, David Brems and colleagues at Upjohn showed that a folding intermediate Insoluble aggregates of normally well-behaved proteins is involved in the aggregation of growth hormone in vitro are featured in a variety of human disease states, includand that the hydrophobic face of a particular helix is ing various forms of amyloidosis (Sipe, 1992) and the important to this interaction. The Jonathan King group prion diseases (Prusiner and DeArmond, 1995). Protein at MIT identified a series of temperature sensitive folding aggregation and precipitation is also a commonplace mutations that function not by destabilizing the protein’s observation in biotechnology, in both the inclusion body native state butby placing a particular folding intermediformation that can occur in the cell during heterologous ate in jeopardy, leading to the irreversible formation of expression of cDNAs and in attempts to refold these inclusion bodies. proteins in vitro (Mitraki and King, 1989). In addition, it In this article I review the highlights of work over the is now well-accepted that a primary function of many past few years that provides further support of misasmolecular chaperones is to thwart misassembly and agsembly mechanisms involving structured folding intergregation during the protein folding process (Hartl, mediates, especially with respect to human diseases 1996). There is a voluminous literature on the interaction involving protein deposition. Many of these studies also of chaperones with proteins in various states of folding. begin to address key questions suggested by the There has been generally less interest in the characterinvolvement of partially folded states: structural details ization of aggregation processes or products in nonon the aggregation intermediates and the aggregates chaperoned folding, perhaps in part because of historithemselves, specificity of aggregate formation, and cal assumptions about the nature of the phenomenon. strategies for inhibiting these processes. The conventional wisdom regarding protein aggregaConformational Changes in Protein tion and precipitation has occupied two extreme posiMisassembly In Vivo tions. Some processes, like aggregation during folding In principle, the formation of aggregate from a folding of a protein in vitro, or inclusion body formation in bacteintermediate can take place either in the folding or unria, are usually regarded as being driven by nonspecific, folding direction (Figure 2, top). Interestingly, both pathhydrophobic interactions operating on random coil ways areprobably represented in cases of protein depostates or on collapsed, molten globule states. Other sition in vivo (Wetzel, 1992). In the case of cytoplasmic processes, such as the extracellular aggregation of inclusion body formation in bacteria, off-pathway aggresome proteins into amyloid fibrils, have been visualized gation probably most often occurs in the folding direcaccording to the sickle hemoglobin model, in which mution, with the extent of deposition depending on a kinetic tations alter the local surface properties of the native, competition between productive folding and aggregafolded state to introduce new packing interactions for tion for a poorly soluble, transient intermediate. Amyloid noncovalent polymerization. The first view tends to disformation, in which fibrils are deposited outside the cell, courage attempts at mechanistic understanding and presumably occurs at some point after proteins have therapeutic intervention; the second suggests that the had an opportunity to complete most or all of the folding straightforward route to both lies in the determination process. In this case, aggregate formation seems to of the high resolution structures of native states. involve unfolding intermediates that can be populated, Folding Intermediates and Protein Misassembly in equilibrium with the native state, under physiological Recent results suggest, however, that many examples conditions. The in vitro correlate for the former process of protein aggregation occur by mechanisms involving is the lossof molecules due to aggregationduring refoldstructured folding intermediates. There are several iming of denatured proteins; the correlate for the latter portant implications of such mechanisms. One is that is the loss of molecules to aggregation during thermal there exists a class of diseases involving aberrant prounfolding of native proteins. tein folding. Another is that such processes can involve Amyloid fibril formation associated with different huinteractions that exhibit significant structural specificity man disease states occurs with a variety of different which cannot be deduced by the examination of native proteins and peptides that have no obvious common states. properties in amino acid sequence, three-dimensional This is not a new idea. In 1974, Michel Goldberg and structure, or function (Sipe, 1992). Despite these differcolleagues put forward a model for the unfoldingences, amyloid fibrils are remarkably similar in size and induced aggregation of multidomain proteins in vitro shape as viewed in the electron microscope, and also (Mitraki and King, 1989). This model, shown in Figure 1, share dye-binding and optical properties that suggest invokes structured folding intermediates with domains significant structural similarity (Sipe, 1992). There is now or subdomains folded as they are in the native state, substantial evidence that unfolding intermediates are but which undergo intermolecular, rather than intramothe building blocks for amyloid fibril formation from globlecular, interactions with each other during folding or ular proteins. For example, the kinetics of amyloid fibril unfolding—leading to the formation of polymers of par-