Biological and Chemical Approaches to Diseases of Proteostasis Deficiency - 2009
36
Biological and Chemical Approaches to Diseases of Proteostasis Deficiency Evan T. Powers, 1 Richard I. Morimoto, 3 Andrew Dillin, 4 Jeffery W. Kelly, 1 and William E. Balch 2 1 Departments of Chemistry and Molecular and Experimental Medicine and the Skaggs Institute for Chemical Biology, 2 Departments of Cell Biology and Chemical Physiology and the Institute for Childhood and Neglected Diseases, The Scripps Research Institute, La Jolla, California 92037; email: [email protected], [email protected], [email protected] 3 Department of Biochemistry, Molecular Biology, and Cell Biology, Rice Institute for Biomedical Research, Northwestern University, Evanston, Illinois 60208–3500; email: [email protected] 4 The Howard Hughes Medical Institute and Molecular and Cell Biology Laboratory, The Salk Institute for Biological Studies, La Jolla, California 92037; email: [email protected] Annu. Rev. Biochem. 2009. 78:959–91 First published online as a Review in Advance on March 19, 2009 The Annual Review of Biochemistry is online at biochem.annualreviews.org This article’s doi: 10.1146/annurev.biochem.052308.114844 Copyright c 2009 by Annual Reviews. All rights reserved 0066-4154/09/0707-0959$20.00 Key Words aging, amyloid, chaperones, heat shock response, protein folding and misfolding, unfolded protein response Abstract Many diseases appear to be caused by the misregulation of protein main- tenance. Such diseases of protein homeostasis, or “proteostasis,” include loss-of-function diseases (cystic fibrosis) and gain-of-toxic-function dis- eases (Alzheimer’s, Parkinson’s, and Huntington’s disease). Proteostasis is maintained by the proteostasis network, which comprises pathways that control protein synthesis, folding, trafficking, aggregation, disag- gregation, and degradation. The decreased ability of the proteosta- sis network to cope with inherited misfolding-prone proteins, aging, and/or metabolic/environmental stress appears to trigger or exacer- bate proteostasis diseases. Herein, we review recent evidence support- ing the principle that proteostasis is influenced both by an adjustable proteostasis network capacity and protein folding energetics, which to- gether determine the balance between folding efficiency, misfolding, protein degradation, and aggregation. We review how small molecules can enhance proteostasis by binding to and stabilizing specific proteins (pharmacologic chaperones) or by increasing the proteostasis network capacity (proteostasis regulators). We propose that such therapeutic strategies, including combination therapies, represent a new approach for treating a range of diverse human maladies. 959 Annu. Rev. Biochem. 2009.78:959-991. Downloaded from www.annualreviews.org by Universidad Autonoma Metropolitana on 10/20/11. For personal use only.
description
Biological and Chemical Approaches to Diseases of Proteostasis Deficiency
Transcript of Biological and Chemical Approaches to Diseases of Proteostasis Deficiency - 2009
ANRV378-BI78-33 ARI 5 May 2009 15:3
Biological and ChemicalApproaches to Diseasesof Proteostasis DeficiencyEvan T. Powers,1 Richard I. Morimoto,3
Andrew Dillin,4 Jeffery W. Kelly,1
and William E. Balch2
1Departments of Chemistry and Molecular and Experimental Medicine and the SkaggsInstitute for Chemical Biology, 2Departments of Cell Biology and Chemical Physiologyand the Institute for Childhood and Neglected Diseases, The Scripps Research Institute,La Jolla, California 92037; email: [email protected], [email protected],[email protected] of Biochemistry, Molecular Biology, and Cell Biology, Rice Institute forBiomedical Research, Northwestern University, Evanston, Illinois 60208–3500;email: [email protected] Howard Hughes Medical Institute and Molecular and Cell Biology Laboratory,The Salk Institute for Biological Studies, La Jolla, California 92037; email: [email protected]
Annu. Rev. Biochem. 2009. 78:959–91
First published online as a Review in Advance onMarch 19, 2009
The Annual Review of Biochemistry is online atbiochem.annualreviews.org
This article’s doi:10.1146/annurev.biochem.052308.114844
Copyright c© 2009 by Annual Reviews.All rights reserved
0066-4154/09/0707-0959$20.00
Key Words
aging, amyloid, chaperones, heat shock response, protein folding andmisfolding, unfolded protein response
AbstractMany diseases appear to be caused by the misregulation of protein main-tenance. Such diseases of protein homeostasis, or “proteostasis,” includeloss-of-function diseases (cystic fibrosis) and gain-of-toxic-function dis-eases (Alzheimer’s, Parkinson’s, and Huntington’s disease). Proteostasisis maintained by the proteostasis network, which comprises pathwaysthat control protein synthesis, folding, trafficking, aggregation, disag-gregation, and degradation. The decreased ability of the proteosta-sis network to cope with inherited misfolding-prone proteins, aging,and/or metabolic/environmental stress appears to trigger or exacer-bate proteostasis diseases. Herein, we review recent evidence support-ing the principle that proteostasis is influenced both by an adjustableproteostasis network capacity and protein folding energetics, which to-gether determine the balance between folding efficiency, misfolding,protein degradation, and aggregation. We review how small moleculescan enhance proteostasis by binding to and stabilizing specific proteins(pharmacologic chaperones) or by increasing the proteostasis networkcapacity (proteostasis regulators). We propose that such therapeuticstrategies, including combination therapies, represent a new approachfor treating a range of diverse human maladies.
959
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
Contents
INTRODUCTION . . . . . . . . . . . . . . . . . . 960PROTEIN FOLDING IN VITRO . . . 961BIOLOGICALLY ASSISTED
PROTEIN FOLDING . . . . . . . . . . . . 963THE PROTEOSTASIS
NETWORK . . . . . . . . . . . . . . . . . . . . . . 963INTEGRATING FOLDING
ENERGETICS WITH THEPROTEOSTASIS NETWORKCAPACITY . . . . . . . . . . . . . . . . . . . . . . . 966
THE PROTEOSTASISBOUNDARY . . . . . . . . . . . . . . . . . . . . . 967
PROGRAMMING THEPROTEOSTASIS BOUNDARY . . . 969
FOLDING DISEASES AND THEPROTEOSTASIS BOUNDARY . . . 971
THE PROTEOSTASIS BOUNDARYAND PHARMACOLOGICCHAPERONES. . . . . . . . . . . . . . . . . . . 973
MOVING THE PROTEOSTASISBOUNDARY WITHPROTEOSTASISREGULATORS . . . . . . . . . . . . . . . . . . . 975
THE PROTEOSTASISNETWORK IN AGING . . . . . . . . . . 979
OBESITY AND THEPROTEOSTASIS NETWORK. . . . 979
CANCER AND THEPROTEOSTASIS NETWORK. . . . 980
AGE-RELATED DEGENERATIVEDISEASES ANDPROTEOSTASIS . . . . . . . . . . . . . . . . . 982
THE FUTURE OFPHARMACOLOGICMODULATION OFPROTEOSTASIS . . . . . . . . . . . . . . . . . 983
INTRODUCTION
Understanding how the unfolded ensembleof polypeptides resulting from the process oftranslation arrives at their native structure(s)for function and how these structures are
maintained and ultimately turned over is a ma-jor challenge. We need to understand how thecell interprets protein energetics and influencesprotein folding in both human health and dis-ease. Unlike protein folding in the test tube, aprotein in a eukaryotic cell must fold and func-tion in a crowded environment, as well as ina variety of distinct environments defined bythe cell’s compartmentalized organization, e.g.,the cytoplasm, exocytic and endocytic compart-ments, the mitochondria, the nucleus, and theextracellular space. Moreover, proteins are sub-ject to extensive changes in structure as they cy-cle between inactive and active conformationsin response to posttranslational modification(s)and/or as they engage in the protein-proteininteractions that enable their biology.
Proteins face many challenges to normalfolding, refolding, and function owing to a con-stant barrage of physical, metabolic, and envi-ronmental stresses. These include changes inthe concentration and composition of small-molecule metabolites that strongly influencefolding by binding-induced modulation of pro-tein stability, changes in the concentrationof osmolytes that influence protein stabilitythrough modulating the hydrophobic effect andhydrogen bonding, changes in temperature thatinfluence energetics, and changes in the con-centration of reactive oxygen species (ROS)or oxidized small molecules that result fromreactions with ROS that alter the structureand function of proteins by aberrantly mod-ifying them. These modifications can dena-ture the folded protein to render it inactiveand/or trigger folded or natively unstructuredproteins to aggregate and become toxic to thecell.
We now recognize that the chemical andenergetic properties specified by the aminoacid sequence of each polypeptide (the primarystructure) encoded by the genome, while veryimportant in partially determining the foldingenergy landscape, are only part of how proteinsevolve biological function. Numerous macro-molecular assistants exist in cells to influencethe folding of the proteome. There is also an
960 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
intrinsically unstructured proteome that, owingto the sequences of its components, probablynever adopts a permanent three-dimensionalstructure but likely requires macromolecularassistance to prevent aggregation and pro-mote function. In the case of the foldedproteome, such assistance controls the rate ofprotein synthesis; influences the rate of fold-ing of the proteome; effects membrane traf-ficking patterns responsible for compartmen-tal localization, which influences stability; andmitigates aggregation and mediates degrada-tion, enabling protein turnover. These compet-ing biological pathways comprising hundreds ofcomponents make up the proteostasis (proteinhomeostasis) network (1).
The proteostasis network and its pathwaysare controlled by numerous integrated signal-ing pathways. These signaling pathways, a sub-set of which are stress responsive, includingthose that respond to misfolding proteins, op-timize the capacity of the proteostasis networkto maintain protein function. This occurs bymore efficiently folding the natively folded pro-teome and maintaining the natively unstruc-tured proteome in a nonaggregated state inresponse to challenges from the inherited pro-teome (e.g., mutated misfolding-prone pro-teins), somatic mutations that result in loss offunction or disruption of proteostasis capacity(e.g., due to misfolding or aggregation), andthe environment (including oxidative stress).The proteostasis network is present in all ex-tant life and is therefore ancient. It necessar-ily coevolved with the remarkable diversity ofpolypeptide sequences to enable the evolutionof a wide variety of organisms, by expanding thecapacity of proteins to function in increasinglycomplex cellular and subcellular environments,and to perform increasingly specialized cellulartasks.
When a protein fails to fold properly owingto an alteration in its sequence (e.g., a mutationor an aberrant posttranslational modificationcaused by oxidative stress), and/or possibly as aconsequence of a change in the concentration,and/or the composition of the components of
the proteostasis network, there is a breakdownin biology. Loss-of-protein function or gain-of-toxic function (the latter often being associatedwith aggregation) can trigger disease by inter-fering with cell function (2–4). The multipleroles of the proteostasis network in maintain-ing the proteome and, therefore, normal phys-iology in response to challenges during devel-opment, aging, and stress are only beginning tobe appreciated (1).
To understand the operation of the pro-teostasis network in health and disease, wepresent a hypothesis about biologically as-sisted folding that integrates folding energet-ics. Through mathematical modeling, whereinfolding energetics and proteostasis networkcomponent composition and concentration arevariables, we propose a framework for how theproteostasis network interprets and influencesprotein folding energetics to control the effi-ciency of folding. We introduce the notion ofa minimal “proteostasis boundary,” that is, thefolding energetics required to achieve foldingof a protein at a given proteostasis network ca-pacity. Using modeling approaches, we illus-trate that, although the proteostasis networkcan make up for deficiencies in folding ener-getics and is thus remarkably adaptable, thereare limits set by the folding energetics. We buildon the proteostasis boundary hypothesis to il-lustrate what can go wrong in proteostasis dis-eases. We review the demonstrated biologicaland chemical strategies to ameliorate loss- andgain-of-function misfolding maladies by adjust-ing the stability of the fold or the capacity of theproteostasis network, or both, leading us to pro-pose a view of how enhanced proteome main-tenance can be used to ameliorate numerousdiseases of complex etiology facing humanityin the twenty-first century.
PROTEIN FOLDING IN VITRO
The challenges associated with transforming alargely unordered ensemble of conformationsresulting from translation (the unfolded state)into a few closely related three-dimensional
www.annualreviews.org • Diseases of Proteostasis Deficiency 961
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
F F
U U
U U
U U
M
M
M
b
Conformatio
nal coordinate #1 Conformational coordinate #2Conform
ational coordinate #1 Conformational coordinate #2
In vitro protein folding In vivo protein folding: the effect ofthe proteostasis network
Energy
High
Low
Energy
a
Figure 1The folding energy landscape for a hypothetical polypeptide. The major conformational ensembles are labeled U (unfolded state), M(misfolded state), and F (folded state). (a) A given energy landscape in vitro. (b) The same energy landscape in vivo influenced by theproteostasis network, which is envisioned to minimize the aggregated/misfolded population, thereby improving folding efficiency.
structures that enable function (the folded state)is as ancient as life itself. In its simplest form, itcan be described by the fundamental principlesput forward by Anfinsen (5) that the native orfolded state of a protein is the most thermo-dynamically stable structural ensemble—onethat is determined by the amino acid sequenceand the solution conditions. A contemporaryview of the energetic component of proteinfolding is illustrated by folding energy land-scapes (Figure 1) (6). Here, the unfolded state,represented by the high-entropy state at thetop of the funnel, folds by multiple parallelpathways and sometimes detectable intermedi-ates to achieve its most thermodynamically sta-ble (lowest energy) ensemble of closely relatedstructures, the natively folded state.
Although there are many energeticallyaccessible paths down the folding funnel(Figure 1) (7), some with local energy min-ima that could trap the protein in partiallyfolded, misfolded, and/or aggregated states(Figure 1a), the chemistry and energetics ofthe polypeptide chain impose constraints thatlimit the number of pathways that are actuallyused in a given aqueous solvent and at a given
temperature (8). Even largely unordered con-formers can have hydrophobic clusters and/orgroups of electrostatic interactions that restrictthe available pathways a given protein can useto get to the folded state ensemble (9). Fold-ing transition states are often associated withthe formation of reverse turns or hydrophobicclusters that act as kinetic “gatekeepers” in ac-quisition of the folded state (8, 10).
We now appreciate that the folded state isnot simply the snapshot obtained by X-ray crys-tallography; rather, the folded state is a dynamicensemble of closely related conformers. The ex-treme extension of this concept is the realiza-tion that a fraction of the human proteome isintrinsically disordered and that this high de-gree of flexibility enables interactions (some-times facilitated by posttranslational modifica-tions) with multiple partners (11). Although therules directing the physical chemistry of proteinfolding remain to be fully elucidated, rapid ad-vances in this area now allow us to predict withincreasing accuracy the folds of small proteins(<100 kDa) based solely on the chemical andphysical properties of the polypeptide chain(8, 12).
962 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
BIOLOGICALLY ASSISTEDPROTEIN FOLDING
Although small, single-domain proteins foldwith amazing speed and efficiency at low con-centrations in aqueous buffers, achieving andmaintaining the folds of larger, multidomainsoluble or membrane proteins is far more chal-lenging, if not impossible in vitro. The fold-ing of soluble and membrane-associated mul-tidomain proteins is also challenging in vivo, ashighlighted below.
The rate of folding in a cell is affected bythe highly regulated rate of translation and,therefore, by several factors including thecomposition of the translation components,including aminoacyl tRNA pools. Cellularfolding kinetics are also affected by bothsynonymous (single nucleotide polymorphismsthat alter the nucleic acid but not the amino acidsequence) and nonsynonymous (nonsense andmissense mutations that change the amino acidsequence) substitutions in the genome that alterthe rate of translation and, in the case of the lat-ter, the sequence of the nascent chain (13, 14).
The environment of the cell is very crowded(15) and, therefore, has high protein concentra-tions that promote aggregation. Cellular pro-tein aggregation is associated with proteotoxi-city and must be minimized, especially duringchemical, physical, and metabolic stress. Thecellular folding of multidomain proteins is gen-erally slow and involves significant populationsof folding intermediates. This is especially truefor transmembrane proteins that must fold inthe context of multiple local environments: thelipid bilayer for the transmembrane domainsand the aqueous environment for the lumenaland cytoplasmic domains (16).
Protein folding, unfolding, and refolding areconstantly occuring throughout the lifetime ofnearly all proteins. Thus, the fold is highly dy-namic and must be protected as it proceedsthrough rapid changes in conformation, simplyas a consequence of its folding equilibrium orin response to posttranslational modificationsand/or interactions with a variety of proteinpartners required for its function (11).
Finally, proteins have distinct turnover ratesthat are integrated with function, and this fine-tuning can become defective, contributing todisease. Protein turnover rates are generallylinked to the life span of the cell, suggestingthat this is an adaptable feature of biologicallyassisted folding.
THE PROTEOSTASIS NETWORK
It is the job of the cellular proteostasis networkto enable cellular protein folding and functionin the face of all of the challenges discussedabove. The proteostasis network consists of nu-merous biological pathways. The macromolec-ular components of these pathways compriseover 1000 general and specialized chaperones,folding enzymes, and degradation componentsas well as trafficking components, the latter in-fluencing compartmental localization. Controlof the proteostasis network is accomplished bysignaling pathways that directly regulate theconcentration, distribution, and activities of thecomponents that make up the proteostasis net-work and, therefore, the relative activities ofthe biological pathways in the network (17, 18)(Figure 2).
The folding function of the proteostasisnetwork is accomplished by smoothing theenergy landscape of proteins (Figure 1b) usingchaperones and folding enzymes which bind tointermediates and transition states, respectively.Chaperones promote folding and maintenancewithin the cell largely by minimizing mis-folding and aggregation, and chaperones arespecialized for various compartments of thecell. Different cells have varying proteostasiscapacities reflected in the composition and con-centrations of their proteostasis components,presumbly to handle the different folding chal-lenges that arise in response to differentiationduring development. Cytosolic chaperones in-clude the ribosome-associated chaperones, theHsp40/Hsp70 and Hsp90 chaperone systems,and the chaperonins, such as TRiC, that facil-itate folding by encapsulation (19–21). Thereare analogous Hsp40/Hsp70/Hsp90 cognatechaperones found in the endoplasmic reticulum
www.annualreviews.org • Diseases of Proteostasis Deficiency 963
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
Figure 2Managing proteostasis. Illustrated are the layers of interactions that facilitate the function of the proteostasisnetwork to generate and maintain functional proteins. The proteostasis network is composed of thecomponents outlined in the first layer (in red font), including the ribosome, chaperones, aggregases, anddisaggregases that direct folding, as well as pathways that select proteins for degradation [e.g., theubiquitin-proteasome system (UPS), endoplasmic reticulum (ER)-associated degradation (ERAD) systems,proteases, autophagic pathways, lysosomal/endosomal targeting pathways, and phagocytic pathways, thelatter are responsible for the recognition, uptake, and degradation of extracellular proteins]. The secondlayer includes signaling pathways (in green font) that influence the activity of components found in the firstlayer. The third layer (in blue font) includes genetic and epigenetic pathways, physiologic stressors, andintracellular metabolites that affect the activities defined by the second and first layers.
(ER), as well as a number of compartment-specific folding specialists (22, 23). Chaperonescollaborate with folding enzymes, includingredox enzymes that promote oxidative folding(disulfide bond formation) in the ER (24)and peptidyl-prolyl isomerases that catalyzecis-trans amide bond isomerizations in proteinfolding. Many chaperones specializing in the
folding of a given protein or a group of relatedproteins exist, including, for example, Hsp47for collagen (25) and microsomal triglyceridetransfer protein (MTP) for apolipoproteinB-containing chylomicron particles (26). Onthe basis of our current understanding, chap-erones are best thought of as macromoleculesthat bind to exposed hydrophobic surfaces in
964 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
misfolded or aggregated states and, in a nu-cleotide hydrolysis-dependent fashion, changeconformations affording the previously boundprotein another opportunity to fold (27–29).Chaperones and folding enzymes facilitatethe folding of multidomain proteins, likelythrough transient sequestration of foldingintermediates, enabling putatively well-choreographed events that would otherwise bedifficult to achieve on a biological timescale inthe complex environment of the cell. Strongevidence for an extracellular chaperone systemis currently lacking, although it is conceivablethat the immune system recognizes and clearsmisfolded extracellular proteins in the contextof the proteostasis program. Numerous factorsinfluence the properties of the proteostasisnetwork including changes in cellular ATPlevels, amino acid pools, metabolites, lipidhomeostasis, and ion balance. These not onlyalter folding capacity, but also modulate the ac-tivity of a number of other proteostasis networkpathways such as degradative pathways, in-cluding ubiquitin-based proteasome-mediatedpathways, and lysosomal and autophagic traf-ficking pathways (Figure 2) that are integralto maintenance of proteostasis (30–33).
Signaling pathways that regulate proteinsynthesis, folding, trafficking, aggregation, dis-aggregation, and degradative pathways of theproteostasis network include: the unfolded pro-tein response (UPR), which principally influ-ences ER folding capacity (34–36); the heatshock response (HSR), which balances pro-teostasis capacity with demand in the cytosol(17, 37); pathways that influence subcellularCa2+ concentrations, such as ER Ca2+ con-centrations, which can increase the ability ofthe cell to handle the folding of N-linked gly-coproteins in the ER (about one third of thehuman proteome) via Ca2+-sensitive foldingchaperones (calnexin, calreticulin, and BiP) (38,39); inflammatory responses, which regulatecell defense and death pathways (40, 41); andhistone deacetylase (HDAC) pathways, whichmay integrate proteostasis capacity throughepigenetic pathways (Figure 2). Such signal-ing circuits control the capacity and composi-
tion of the proteostasis network through tran-scriptional, translational, and posttranslationalmechanisms to balance or rebalance proteosta-sis by reducing demand, enhancing folding andrepair processes such as disaggregation, and/orby mediating degradation. When proteostasisis severely compromised, as can occur in someearly- and late-onset genetic diseases associatedwith the inheritance of a misfolding-prone pro-tein, unsuccessful signaling attempts to rebal-ance proteostasis can activate cell death path-ways, targeting the cell for destruction (34–36).In multicellular organisms, proteostasis path-ways can be under cell-nonautonomous globalcontrol by neuronal and possibly nonneuronalsignaling pathways (42, 43).
The proteostasis network is not only highlyadaptable, as discussed above (enabled by theinfluence of multiple stress-responsive signal-ing pathways), but it also can be quite distinctin each cell type; see Figure 3 and Supple-mental Figure 1 (follow the SupplementalMaterial link from the Annual Reviews homepage at http://www.annualreviews.org) (44).This is not surprising given the central role ofproteins in cell physiology and the diversity offunction(s) in distinct cells entrained by devel-opmental pathways. The response to diversi-fication includes proteostasis network compo-nents that are conserved throughout evolution(42) and those that are specialized (25, 26), re-flecting the many different proteostasis chal-lenges. For example, hepatocytes, plasma cells,and β-cells must produce high levels of dis-tinct secreted proteins, whereas a fibroblast hasless secretory activity and, therefore, a less-specialized ER proteostasis network capacity.Indeed, the compositional complexity of theproteostasis network scales with the complexityof the organism, consistent with the hypothe-sis that the proteostasis network influences theevolution of protein sequences. In other words,a protein’s ability to achieve its functional state(which, in part, directs phenotypic selection) isdependent on the proteostasis network. Thus,we suggest that any effort to understand pro-tein folding, and therefore protein function,in vivo will ultimately need to consider the
www.annualreviews.org • Diseases of Proteostasis Deficiency 965
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
Expression profilesComplete imageG
rouped proteostasis network com
ponentsZoomed section of image
Figure 3Expression profiling of human proteostasis network components in 80 human tissues (44). Hierarchical clustering was used to groupproteostasis network components (y-axis on right) on basis of the similarity of their expression profiles across the 80-tissue array (x-axison top). A dendrogram is provided where lengths of the branches leading up to each node directly reflect the degree of correlationbetween the expression profiles as assessed by the pair-wise similarity function described previously (171). To see the high-resolutionimage that can be zoomed to identify the individual components and tissues see Supplemental Figure 1. The heat map illustratesabundance as higher (red ) or lower ( green) relative to the mean (black) value across all tissues.
interdependence of folding energetics and theproteostasis network (1, 8, 12). The degreeto which the proteostasis network biases fold-ability and function remains to be determined,but it is clear that many, if not most, pro-teins cannot fold without assistance within thecell.
INTEGRATING FOLDINGENERGETICS WITH THEPROTEOSTASIS NETWORKCAPACITY
Although the energy landscape view of proteinfolding is a useful framework for interpretingexperimental folding studies carried out in vitro
(Figure 1a), it does not directly illustrate thecritical interplay between the energetics of theprotein (defined by the dynamics of the popula-tions of the unfolded, intermediate, and foldedstates) and the biology of folding. Qualitatively,we know that chaperones and folding enzymesbind to folding intermediates and transitionstates, resculpting the folding free-energy land-scape (Figure 1b). While useful, it does not in-form us on how the proteostasis network main-tains proteostasis.
One semiquantitative approach towardunderstanding the interdependence of proteinfolding energetics and biological foldinginfluenced by proteostasis network capacity hasbeen developed using mathematical modeling
966 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
(45, 46). Specifically, we have modeled aspectsof the proteostasis network to show how dif-ferent levels of proteostasis network capacityhandle the folding and trafficking of proteinsin the exocytic pathway as a function of kineticand thermodynamic parameters. The FoldExmodel (46) describes how the inherent ener-getics of the polypeptide chain are interpretedby and then influenced by the proteostasis net-work within the ER, a compartment that plays acentral role in the biogenesis and export of pro-teins to multiple cellular compartments and thecell surface (Figure 4a). In this model, trans-lation, chaperoning, export, and degradationpathways were each treated as single entities,using a Michaelis-Menten kinetic formalismgenerally reserved for studying enzyme kinetics(Figure 4a). The mathematical modeling en-abled us to predict how distinct secretoryprotein energetics can be influenced by dif-fering proteostasis network activities. TheFoldEx model revealed that the energeticsof the polypeptide chain and the capacity ofindividual pathways of the proteostasis networktogether determine the extent to which a desta-bilized protein will fold and be exported or bedegraded (Figure 4a) (46). Importantly, thismodel can qualitatively fit experimental data,suggesting that, even with its simplifications, itis able to rationalize and predict experimentalobservations. The FoldEx model was used todefine a “minimal export threshold,” a bound-ary in the three-dimensional space defined byprotein stability, folding rate, and misfoldingrate. The location of this boundary dependson the proteostasis network capacity, i.e., theconcentration and activities of the translation,chaperoning, export, and degradation pathwaysof the proteostasis network and the proteinfolding and misfolding energetics. Proteinswith energetics within the export thresholdat a defined proteostasis network capacity areexported at levels sufficient for function; thosewith energetics outside the boundary are not.
We want to emphasize that the positionof the minimal export threshold, defined bythe variables in the FoldEx model, is stronglyinfluenced by the capacity of the proteostasis
network to generate and maintain the foldedstate of a protein in the ER for export. Con-sequently, whether a protein is exported effi-ciently enough for it to function depends onboth protein folding energetics and the pro-teostasis network capacity. Although simplified,the Michaelis-Menten treatment of these in-teractions revealed that there is no single pa-rameter, such as the thermodynamic stabil-ity, the folding rate, or the misfolding rate,that determines whether a given protein willfold, traffic, and generate sufficient activity inits destination environment to enable func-tion. Thus, the FoldEx model provides a usefultheoretical framework that semi-quantitativelyrecapitulates experimental data to rationalizethe general operation of the exocytic path-way in the context of folding energetics andthe adjustable proteostasis network capacity(46).
THE PROTEOSTASIS BOUNDARY
The basic understanding of the function of theER revealed by the FoldEx model should also beuseful for understanding the interplay betweenthe chemistry and the biology of protein fold-ing in other cellular compartments, each withdistinct environments and proteostasis networkcomposition and capacity. Other cellular com-partments include the many post-ER exocyticand endocytic trafficking compartments, thecytoplasm, the mitochondria, and the nucleus.The concept of the minimal export thresholdfrom the FoldEx model should be extendableto cover any situation in which protein foldingenergetics and the proteostasis network com-bine to determine whether a protein can achieveadequate levels of folding for function. We re-fer to this more general mathematical modelas Folding for the Function of protein x orFoldFx (Figure 4b). FoldFx, like FoldEx, en-ables the interdependence of folding energeticsand the proteostasis network capacity to be vi-sualized. As in FoldEx, we treat the translationmachinery (T in Figure 4b), chaperones (C inFigure 4b) and degradation pathways (D inFigure 4b) as single entities. The export process
www.annualreviews.org • Diseases of Proteostasis Deficiency 967
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
T
TU
TUC1 C1U
C1
C1
C1
C2U U
M
F EF
E
E
R
RR
R
E
R
R
RU
RM
T
T
C1ERAF ERAD
Export
ExportD
egradationT: Sec61 translocon
C1: Chaperone in high-affinity state
C2: Chaperone in low-affinity state
E: Export machinery
U: Unfolded protein
F: Folded (export-competent) protein
M: Misfolded protein
R: Retrotranslocation machinery
FoldEx
Minimal exportthreshold
T
TU
TUC1 C1U
C1
C1
C1
C2U U
M
F
D
DD
D
D
D
DU
DM
T
T
C1
Chaperoning
Degradation
Function
FunctionD
egradation
T: Translation machinery
C1: Chaperone in high-affinity state
C2: Chaperone in low-affinity state
U: Unfolded protein
F: Folded, function protein
M: Misfolded protein
D: Degradation machinery
FoldFx
Proteostasisboundary
a
b
Slow
Foldingrate
Foldingrate
Slow
Stable
Unstable
Stability
Stability
Misfoldingrate
Fast
Slow
Fast
Fast
Slow
Stable
Unstable
Misfoldingrate
Fast
Figure 4Models for proteostasis. (a) (left) The FoldEx model for protein folding and export from the endoplasmic reticulum (ER) (46). Themodel includes ER-assisted folding (ERAF), ER-associated degradation (ERAD), and export. (right) A plot showing the minimal exportthreshold as a surface in a space defined by protein thermodynamic stability, folding kinetics, and misfolding kinetics. Proteins withfolding energetics within the boundary are exported efficiently enough to ensure adequate function in their destination environment(46). (b) (left) The FoldFx model for protein folding and function. FoldFx is largely analogous to FoldEx, except that the export step isreplaced simply by protein function, and it is not limited to describing protein folding in a single cellular compartment. (right) A plotsimilar to the one in panel a, except that it shows the proteostasis boundary instead of the minimal export threshold. Proteins withfolding energetics that lie within the proteostasis boundary fold efficiently enough to ensure adequate levels of function.
in FoldEx (Figure 4a) is replaced in theFoldFx model with function (F in Figure 4b),which now simply reflects the generation ofa functional protein. Thus, in the FoldFx
view, translation, chaperones, and degradativepathways are the factors that determine thelevels of functional protein by interpretingand influencing folding energetics. Signaling
968 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
pathways (e.g., UPR, HSR, Ca2+, etc.) thatcontrol the composition and concentration ofproteostasis components through transcrip-tional and posttranslational modifications areaccommodated in this simplified view by virtueof the adjustable concentrations of the pro-teostasis network components (T, C, and D inFigure 4b) (46).
Like the minimal export threshold used tounderstand the joint role of folding energet-ics and the proteostasis network capacity inthe exocytic pathway in the FoldEx model(Figure 4a) (46), the combined role of energet-ics and the proteostasis network in folding ef-ficiency in other compartments is best viewedin terms of a minimal proteostasis boundary,which we refer to simply as the proteostasisboundary (Figure 4b). The proteostasis bound-ary defines the folding energetics that a proteinmust have to achieve adequate levels of functionin the context of a given proteostasis networkcapacity in a given cell. Like the minimal exportthreshold in FoldEx, the proteostasis bound-ary is best illustrated as a boundary in three-dimensional space, defined by protein foldingthermodynamics (from unstable to stable—thex-axis), folding kinetics (from slow to fast—they-axis) and misfolding kinetics (from slow tofast—the z-axis), a convention used through-out this review (Figure 4b). Like the minimalexport threshold, the location and shape of theproteostasis boundary is highly dependent onboth the composition and concentration of pro-teostasis network components.
FoldFx simplifies, with rather broad strokes,the interdependence of folding energetics andproteostasis network capacity in generating apopulation of protein x that is folded and func-tional. Of course, the composition of the var-ious proteostasis networks, dictating the posi-tion of the proteostasis boundary, is expected tobe different for each of the different compart-ments in a cell (e.g., ER, cytosol, mitochon-dria, nucleus, etc.), reflecting their specializedfunctions. Moreover, we envision that a spe-cific protein would only require a subset of theproteostasis network components, which inturn suggests that the proteostasis boundary
could be different for proteins that utilize a dif-ferent set of proteostasis network components,all else being constant. Expanding the FoldFxmodel to encompass the actual complexity ofdistinct groups of proteins would certainly addmuch needed detail. However, we believe thatit is not necessary to create multiple proteosta-sis boundaries for individual or groups of pro-teins comprising the proteome of an organismin order to illustrate how the model can helpus understand the role of folding energeticsand proteostasis network capacity in health anddisease, and the pharmacologic managementthereof.
PROGRAMMING THEPROTEOSTASIS BOUNDARY
In Figure 5, the utility of the FoldFx model is il-lustrated with a hypothetical biological networkof interacting proteins of a proteome. A bio-logical network is defined by nodes or proteins(indicated by spheres) and edges or protein in-teractions within the network (indicated by thelines between the spheres). A set of hypotheti-cal nodes is positioned in Figure 5 according tothe corresponding proteins’ folding energetics:their stabilities, folding rates, and misfoldingrates at a given proteostasis network capac-ity. The energetic parameters for the nodes inFigure 5 were randomly generated with theconstraint that they be physically plausible (e.g.,it was assumed that folding could not take placeon a timescale faster than 1 μs). The proteosta-sis boundary was determined by arbitrarily set-ting the concentrations of the proteostasis net-work components to reasonable values (46) andsolving the FoldFx model (mathematically anal-ogous to the FoldEx model). This solution wasthen used to determine which region of thespace, defined by the protein stability, fold-ing rate, and misfolding rate, yielded a morethan minimal fraction of functional protein (thisfraction was set to 0.1 for Figure 5). The pro-teostasis boundary is the surface that surroundsthis region.
In a healthy cell, nodes are positioned withinthe limits defined by the proteostasis boundary
www.annualreviews.org • Diseases of Proteostasis Deficiency 969
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
a b Aggregation
Loss of function
Slow
Fast
Slow
Stable
Unstable
Misfoldingrate
Fast
Foldingrate
Stability
Slow
Fast
Slow
Stable
Unstable
Misfoldingrate
Fast
Foldingrate
Stability
Figure 5A hypothetical network of interacting proteins as viewed in the FoldFx model. Each node (or sphere) in thenetwork represents a corresponding protein’s folding energetics, and each connection (line) represents aphysical or functional interaction. The surface represents a proteostasis boundary, which, for the sake ofsimplicity, is shown as being the same for all of the proteins in the network. (a) All of the nodes are within theproteostasis boundary in a healthy cell. (b) Mutations or aberrant posttranslational modifications can alterthe folding energetics of stable proteins, making their corresponding nodes fall outside the proteostasisboundary. This can lead to either loss of function (red node) or aggregation (black node).
(Figure 5a, green nodes). Some nodes fall wellwithin the proteostasis boundary, indicatingthat they are fast folding and thermodynam-ically stable proteins. These proteins may notrequire significant biological assistance and willlikely tolerate major changes to the capacity ofthe proteostasis network. In contrast, less sta-ble, slowly folding, and/or rapidly misfoldingproteins are closer to the proteostasis bound-ary, and their ability to achieve biological func-tion is thus likely to be much more sensitive toproteostasis network capacity.
When the folding energetics of a given pro-tein are adversely affected by mutation or byoxidative stress (e.g., a lipid aldehyde modi-fication), it is reasonable to hypothesize thatthe protein of interest would move outside thelimits defined by proteostasis boundary. In thiscase, the protein can be prone to either degra-dation (Figure 5b, red node) or aggregation(Figure 5b, black node). Of particular inter-est are those proteins lying close to or at theproteostasis boundary. We propose that theseproteins will be particularly sensitive to mod-est changes in the capacity and compositionof the components of the proteostasis network
that influence the shape of the proteostasisboundary.
In the FoldFx view, the position and shapeof the proteostasis boundary suggests that manysequences that are competent to fold in vitro(in buffer at low protein concentration) maynot be useful for function in vivo, as has beenobserved for many mutant proteins and pro-tein fragments, especially if a relatively highconcentration of protein is required for func-tion as it would exacerbate aggregation (45). Al-though the actual distribution of nodes defininga biological network relative to the proteosta-sis boundary is influenced by both the fold-ing energetics and the biology of the network,we suggest that proteins that are marginallystable and lie on or close to the proteostasisboundary may participate in the reversible as-sembly of dynamic protein complexes (11, 47).Moreover, given the high variability of the pro-teostasis network capacity in different cell types(Figure 3), housekeeping enzymes that are crit-ical for all cells may be largely insensitive to theshape of the proteostasis boundary because theyare energetically stable and fall deep within itsembrace (48).
970 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
The apparent plasticity of the proteostasisboundary illustrates how critical the proteosta-sis network is to protein folding efficiency andalso to the life of a cell. This hypothesis is con-sistent with recent evidence that suggests theproteostasis network does not possess signifi-cant excess capacity; rather, it is finely tunedand offers just enough capacity for the pro-tein folding load (17, 49, 50). Although at firstglance confounding, this is consistent with thekey role of the proteostasis network in definingfunctionality in diverse cellular environments.Thus, by setting the proteostasis boundary as athreshold for generating folded and functionalproteins, the proteostasis network can createand maintain functional proteins in response tothe local environment.
Of course, the chaperone and folding en-zyme components that contribute to the loca-tion and shape of the proteostasis boundary arethemselves part of the biological network. Weexpect that they would generally lie well withinthe proteostasis boundary and would act as hubs(highly connected nodes in the biological net-work) that have numerous, but weak, interac-tions with multiple proteins that can be alteredin response to stress. As such, they have beensuggested to promote stability of the biologi-cal network, buffer noise, and regulate signal-ing pathways that would otherwise spuriouslymodify the network (see Reference 51 for fur-ther details).
FOLDING DISEASES AND THEPROTEOSTASIS BOUNDARY
Unlike in vitro assessment(s) of folding andfunction, biology needs to distinguish patho-logical disorder from functional disorder en-coded in the polypeptide sequence. It is notyet understood how this is accomplished. TheFoldFx model and the proposed concept of aproteostasis boundary provide a framework toassess the impact of folding defects on pro-teostasis. These ideas allow us to better under-stand new pharmacologic approaches for main-taining health and countering disease by either
adjusting the energetics of the protein fold, thecapacity of the proteostasis network, or both.
The first question we need to ask is: Whenis a protein misfolded from a functionalperspective? Here, we must consider thestability of the fold and proteostasis networkcapacity in light of the shape and position ofthe proteostasis boundary that will be requiredfor activity. Thus, we would suggest that akey concern in human health is how the localenvironment, a mutation, or a posttransla-tional modification alters a protein’s ability tofunction in a biological network defined by theproteostasis boundary.
In inherited diseases associated with a fold-ing deficiency, it is well established that a changein the amino acid sequence of a protein cansignificantly alter folding energetics and, there-fore, the position of the protein relative to theproteostasis boundary. For conservative mis-sense mutations, the substitution may be neu-tral. In other words, it has little influence onfolding kinetics or thermodynamics. In thiscase, it would have little effect on function un-less it eliminates a catalysis-critical or binding-critical residue. A less conservative missensemutation or an amino acid deletion often altersthe kinetics or the thermodynamics of folding.Such a mutation could move a protein outsidethe proteostasis boundary, where it may becomesusceptible to substantial misfolding, aggrega-tion, and/or degradation. The effect of the pro-tein being excessively degraded could lead sim-ply to loss of that protein’s function, in whichcase only one node in the network would be af-fected (Figure 6a, red node). The effect couldalso be more far reaching if, for example, theprotein was involved in interactions that stabi-lized other proteins. In the latter case, some orall of the other proteins with which the mutantprotein interacts may now fall outside the pro-teostasis boundary in response to destabiliza-tion (Figure 6b, multiple red nodes), furthercompromising cell or organism function.
Alterations in the composition or concen-tration of proteostasis network componentsin disease could have a global effect on thefolding of the proteome, depending on how
www.annualreviews.org • Diseases of Proteostasis Deficiency 971
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
Slow
Fast
Slow
Stable Unstable
Misfoldingrate
Fast
Slow
Misfoldingrate
Fast
FoldingrateSlow
Fast
Foldingrate
StabilityStable Unstable
Stability
a bDestabilizedmutant,
loss of function
Originalmutant
Interactionpartnersdestabilized
Figure 6Plots illustrating the effect of a destabilizing mutation on protein function. (a) The destabilized protein falls outside the proteostasisboundary and is degraded, but the loss of function is limited to the destabilized protein itself. (b) As in panel a, degradation of theoriginal mutant destabilizes the proteins it interacts with, leading to further loss of function and multiple nodes being repositionedoutside the proteostasis boundary.
many proteins are close to the proteostasisboundary in a given cell type. We predict thata decline in the capacity of a core proteostasisnetwork component, such as Hsp70, Hsp90, orproteasome subunits, could have far-reachingand toxic effects. In contrast, we posit thesphere of influence of disrupting more specificcomponents (e.g., Hsp90 cochaperones, suchas immunophilin isoforms) will be restrictedto those proteins whose folding pathwaysdepend specifically on the activity of thosecomponents. This interpretation is consistentwith the considerable expansion of cochap-erone complexity with increasing eukaryoticcell complexity. If folding capacity, as definedby the proteostasis boundary, is increasedby overexpressing proteostasis network coreor accessory chaperone/cochaperone com-ponents (as occurs normally but transiently,for example, in the UPR and HSR), the newproteostasis network is likely to be protective,as is observed in stress responses (34–36). Wepropose that manipulating degradative activ-ities (e.g., ubiquitin-mediated proteasomal,lysosomal, and autophagy pathways) is poten-tially a double-edged sword. Degradation is a
normal activity that is used to control a varietyof activities, including cell cycle and develop-mental programs, or removal of misfolded oraggregated toxic proteins. Thus, decreasingdegradation could, in principle, increase theconcentration of folded, functional protein;alternatively, it could also increase the concen-tration of misfolded, toxic species. Conversely,increasing the activity of degradation pathwayscould promote removal of toxic, pathology-associated misfolded and aggregated protein,but it could also lead to global destabilization ofthe biological network by overzealous removalof one or more proteins, which are close to theproteostasis boundary owing to their sequencesbut are critical for normal function and survival.
Given the FoldFx view, we now need to con-sider the possibility that the clinical features ofmany inherited and sporadic diseases, as well asresponses to stress encountered during aging,are consequences of changes in the proteosta-sis network (1). We suggest that such changesin the proteostasis network with aging andthe corresponding changes in the shape of theproteostasis boundary, although variable fordifferent compartments and cell types, alter
972 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
the ability of a cell to handle its proteostasisload. Thus, we propose that pharmacologic up-or downregulation of the capacity of the pro-teostasis network provides an avenue for inter-vention in diseases of proteostasis, as discussedbelow, where we summarize the increasing ev-idence that supports this hypothesis.
THE PROTEOSTASIS BOUNDARYAND PHARMACOLOGICCHAPERONES
The primary challenge in ameliorating loss-of-function misfolding diseases is to identifysmall-molecule pharmacologic agents that se-lectively or specifically enhance protein stabil-ity and function. One could envision movinga node from outside the proteostasis bound-ary to inside by stabilizing the borderline pro-tein or by expanding the proteostasis networkcapacity and thus expanding the proteostasisboundary to envelop the destabilized protein,i.e., the problematic node. Herein, we focuson two classes of small molecules that arebecoming increasingly recognized in the lit-erature and that can accomplish these goals:
pharmacologic chaperones (PCs), which movethe node by stabilizing the destabilized proteinof interest, and proteostasis regulators (PRs),which expand the proteostasis boundary to sur-round the borderline protein by enhancing thecapacity of the proteostasis network.
Small-molecule PCs function by binding tothe destabilized target protein and thus sta-bilizing it. PCs, in general, should be effica-cious within the context of most, but not all,existing proteostasis networks depending onhow destabilized the metastable protein is. Formany misfolding-prone proteins, PCs shouldbe able to move the protein within the pro-teostasis boundary, thereby increasing functionby elevating the concentration of folded func-tional protein. In principle, there are three waysto correct a misfolding disease with PCs: Themisfolding-prone protein can be thermody-namically stabilized (Figure 7a, point is movedalong the stability axis); or the folding rate canbe increased by a PC that stabilizes the fold-ing transition state (Figure 7a, point is movedalong the folding rate axis); or the misfoldingrate can be decreased by stabilizing the na-tive state (Figure 7a, point is moved along the
a b
Pharmacologicchaperonestabilizesmutant
Increasestability
Increasefolding rate
Decreasemisfolding rate
Slow
Misfoldingrate
Fast
Slow
Misfoldingrate
Fast
SlowFast
Foldingrate
Slow
Fast
Foldingrate
StableUnstable
StabilityStable
Unstable
Stability
Figure 7The effect of pharmacologic chaperones on protein folding in the FoldFx model. (a) Pharmacologic chaperones bind to a protein,directly affecting its folding energetics. This binding can help rescue a protein with defective folding energetics by increasing its foldingrate, decreasing its misfolding rate, increasing its stability, or any combination thereof. (b) Increasing protein stability, illustrated in thisrotated plot, is the mechanism of action of most known pharmacologic chaperones.
www.annualreviews.org • Diseases of Proteostasis Deficiency 973
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
misfolding rate axis). To date, only the firstand third mechanisms have been demonstratedexperimentally, although the potential remainsfor other classes of PCs to be developed.
Known classes of mutant proteins that canbe corrected by PCs include G protein–coupledreceptors (52), neurotransmitter receptors (53),glycosidases, and analogous enzymes associatedwith lysosomal storage diseases (LSDs) (54–56), and, potentially, the cystic fibrosis trans-membrane conductance regulator (CFTR) (57,58). In cystic fibrosis, the �F508 deletion inCFTR disrupts its folding and targets the pro-tein for degradation in the ER. Putative PCsthat act on the CFTR include polyaromaticcompounds (59) that have been shown to sta-bilize the transmembrane helices (60) in or-der to stabilize CFTR for export from the ER.This activity would be visualized in terms of theFoldFx model as shown in Figure 7b, in which ared node (representing a loss-of-function mu-tant protein) that lies outside the proteostasisboundary is moved inside the barrier by ther-modynamic stabilization of the �F508 CFTRfold. A second class of PCs, referred to as po-tentiators, cannot stabilize the fold sufficientlyfor export from the ER but, once CFTR is atthe cell surface, will promote stabilization ofthe channel and thereby increase conductance,perhaps through decreased degradation and/orallosteric mechanisms favoring interaction withregulatory kinases and phosphatases (61, 62).The Vertex 770 potentiator has just completedhighly successful phase II clinical trials wherea marked recovery in sweat chloride, nasal po-tential, and lung capacity was observed for theG551E CFTR variant that is exported from theER normally but is inactive at the cell surface(http://www.cff.org/). It has been suggestedthat a combination of a corrector PC, to pro-mote folding and export, and a potentiator PC,to enhance activity of the partially destabilizedchannel, could be the key to effectively treatingcystic fibrosis patients (63). Similarly, numer-ous antagonists and agonists for ion channelsand neurotransmitter receptors that affect theirfunction may do so partly by functioning as PCsto favor altered folding and trafficking.
The use of PCs is on firm mechanisticground and is particularly promising for thetreatment of LSDs, including Fabry’s, Pombe’s,Tay-Sachs, and Gaucher’s diseases (54, 55,64). Several PCs that bind to and demonstra-bly stabilize the mutant lysosomal enyzmessusceptible to excessive degradation are invarious phases of clinical trials for differentLSDs. LSDs are largely loss-of-function dis-eases caused by the excessive ER-associateddegradation of destabilized variant lysosomalenzymes. The resulting loss of function leadsto substrate accumulation in the lysosome and,hence, pathology. Like the �F508 CFTR vari-ant, most misfolding-prone LSD-related vari-ant proteins do not trigger the UPR because theproteins are removed by ER-associated degra-dation, leaving the proteostasis network unper-turbed. However, it is now appreciated thatnumerous ligands, many of which are in-hibitors, that bind to the active sites of specificdestabilized lysosomal enzymes stabilize theirfolds in the ER (Figure 7b). Stabilization of β-glucocerebrosidase variants leads to enhancedfolding and trafficking of the protein to the lyso-some where it is more stable and functional (al-beit with reduced specific activity) owing to thelower pH of this organelle relative to the neu-tral pH/environment of the ER and the highconcentration of substrate (55, 65, 66). As men-tioned above, it is expected and observed thatnot all β-glucocerebrosidase variants respondto PC therapy. Highly destabilized mutants ofthe enzyme are degraded so efficiently that theconcentration of folded enzyme available forPC binding is simply too low for a PC to havean observable effect on the folding equilibriumconstants. Destabilized glucocerebrosidase mu-tants require a different approach to improvetheir export from the ER and downstream func-tion, which involves increasing proteostasis net-work capacity (see below).
A special class of PCs has proven to beuseful for stabilizing proteins against mis-folding and amyloid fibril formation in de-generative diseases by differentially stabiliz-ing the non-amyloidogenic native state overthe misfolding transition state. These so-called
974 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
kinetic stabilizers function by significantly de-creasing the misfolding rate. Two chemicalclasses of kinetic stabilizers are currently be-ing used in human clinical trials to amelioratethe transthyretin (TTR) amyloidoses (66, 67;http://www.clinicaltrials.gov). TTR is syn-thesized as a monomer and tetramerizes be-fore ER export. Mutations that destabilize thetetramer result in its disassembly to monomersin the serum. These monomers can partially un-fold and undergo amyloid formation at a ratethat increases with increasing TTR monomerconcentration. Kinetic stabilizers (66, 67) bindto TTR tetramers, preventing dissociation un-der physiological conditions, thereby decreas-ing the monomer pool in the serum, substan-tially slowing and, in some scenarios, prevent-ing protein aggregation. Thus, TTR kineticstabilizers show how gain-of-toxic function canbe avoided by a special class of PCs that kinet-ically stabilize the native non-amyloidogenicstate of proteins.
MOVING THE PROTEOSTASISBOUNDARY WITHPROTEOSTASIS REGULATORS
Proteostasis regulators (PRs) include smallmolecules that offer the advantage that onecompound can be used to expand the proteosta-sis boundary for numerous misfolding-proneproteins that use a common set of proteosta-sis network components associated with a par-ticular cell type, compartment, or activity (1).PRs can favor folding by adjusting the com-position, concentration, and, thus, the capacityof the proteostasis network by expanding theproteostasis boundary (Figure 8). We specu-late that PRs could also be discovered that col-lapse the proteostasis boundary by upregulatingcomponents of the proteostasis network that fa-vor unfolding and degradation (Figure 8), al-though we are unaware of any such compounds.
In general, we hypothesize that the pro-teostasis boundary could be expanded bychanging the levels and/or activities of core(Hsp70/Hsp90) proteostasis network pathways.Expansion can occur, for example, by induc-
ing cellular Ca2+ distribution changes (68) orby inducing the UPR or HSR signaling path-ways (34–36), all of which are known to in-crease proteostasis network capacity (see be-low). A selective expansion of the proteostasisboundary could also be achieved by PRs that di-rectly target one or a few of the diverse Hsp40isoforms that modulate the folding of specificmisfolding-prone proteins. Instead of inducingthe UPR or HSR, the Hsp70 or Hsp 90 coresystems could be more selectively altered by tar-geting one or a few of the many cochaperonesthat differentially regulate interactions with de-sired misfolding-prone proteins. We proposethat in this way PRs could function as rheostatsto fine-tune the proteostasis boundary in a par-ticular compartment in the cell with respect toa subset of proteins.
PRs could also potentially be used to pre-condition the proteostasis network to more ef-fectively handle metabolic stress (69) and ag-ing by increasing the protective capacity of theproteostasis network prior to an insult. Gen-erally speaking, PRs could be used to upregu-late a pathway within the proteostasis networkto increase proteostasis capacity, just as the cellnormally uses transient proteostasis signalingpathways such as the UPR and HSR to respondto stress and reestablish proteostasis. Below webriefly highlight the potential utility of a few ofthe compounds in the growing PR toolbox.
The drug salubrinal (70) uses aspects of onearm of the UPR to protect cells against proteinmisfolding stress. Phosphorylation of eukary-otic initiation factor 2 (eIF2) is accomplished byat least four kinases, including the UPR kinasePERK in response to unfolded protein accumu-lation in the ER (35, 36). Phosphorylated eIF2protects cells by blocking translation, therebyreducing the folding load. Salubrinal activatesthe PERK-ATF4 arm of the UPR downstreamof PERK by inhibiting dephosphorylation ofeIF2, thereby easing the folding load and avoid-ing the apoptosis that is typically associated withsustained activation of the PERK-AFT4 arm ofthe UPR (70, 71).
PRs that activate the HSR include trip-tolide, quercetin, and celastrol, all heat shock
www.annualreviews.org • Diseases of Proteostasis Deficiency 975
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
a
b
Slow
Misfoldingrate
Fast
Slow
FastFoldingrate
Stable
Unstable
Stability
PR favoringfolding
(expands PB)
PR favoringdegradation
(constricts PB)
Figure 8The effect of proteostasis regulators (PRs) on the proteostasis boundary (PB). (a) PRs can either expand theproteostasis boundary, favoring protein folding, or constrict the proteostasis boundary, favoring degradation.The expanded and contracted proteostasis boundaries resulted from increasing or decreasing, respectively,the concentrations of chaperones or degradation machinery. (b) Influence of select PRs (outer layer, in purplefont) on managing proteostasis. The listed drugs and their effects on the indicated pathways and componentsdescribed in Figure 2 are highlighted in the text.
976 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
transcription factor 1 (HSF1) enhancers (72–74). Celastrol was originally identified in ascreen for compounds that were neuroprotec-tive in Huntington’s disease models (75). Ev-idence now suggests that HSF1 production isincreased by reduced insulin growth factor-1receptor (IGF1-R) signaling, which pro-tects against neurodegeneration and improveslongevity (see below) (1). Although celas-trol’s mechanism of action is not fully eluci-dated, recent studies suggest it to be a thiol-reactive molecule that triggers stress responsesthrough this chemical activity (72). Reflectingthis chemical activity, celastrol has been foundto inhibit the activity of the HSF1 repressorHsp90 (76), to inhibit NFκB (77), and to in-hibit the proteasome (78) as well as to mod-ulate Ca2+-signaling pathways (72). Celastrolalso appears to be useful as an adjuvant forarthritis (79) and for preventing cancer throughHsp90-dependent steps (see below). The dif-ferent responses of various cell types to celas-trol could be a consequence of the differentdemands that cells place on their unique pro-teostasis networks and the composition of thesignaling pathways that control proteostasis. Asimilar possibility could apply to nonsteroidalantiinflammatory drugs (NSAIDs), includingaspirin and acetaminophen, which inhibit thecyclooxygenases COX 1 and 2. NSAIDs haveantioxidant properties (80) and have been sug-gested to modulate Hsp70 and Hsp90 chaper-ones, as well as HSF1 signaling pathways (81).Given the emerging complexity of their targetsand the broad range of indications exhibited bythese compounds, we speculate that NSAIDs,like HSF1 pathway modulators, may have unex-pected influences on the position and/or shapeof the proteostasis boundary.
Ca2+-sensitive signaling pathways controlcell function at multiple levels (38, 39). Re-cent evidence suggests that alterations ofintracellular Ca2+ levels can regulate pro-teostasis network capacity in the ER and,possibly, in other cellular compartments andhence alter the position and shape of the pro-teostasis boundary (68). Inhibition of l-typeCa2+ channels in the plasma membrane using
either diltiazem or verapamil partially restoredβ-glucocerebrosidase folding, trafficking andfunction in Gaucher’s disease patient-derivedfibroblasts (68). Structure-activity relationshipstudies of analogs showed a clear correlationbetween Ca2+ antagonism and PR function.This observation led to the suggestion that thebeneficial effects of diltiazem and verapamilare mediated by increased ER Ca2+ concen-trations, likely influencing the activity of Ca2+-dependent ER chaperones like calnexin and cal-reticulin, and, indirectly, the function of BiP,the major Hsp70-related chaperone, throughCa2+-linked mechanisms. Because diltiazemand verapimil increase the ER proteostasisnetwork capacity and expand the proteostasisboundary, these PRs have the ability to enhancethe folding, trafficking, and function of non-homologous mutant lysosomal enzymes asso-ciated with other distinct LSDs, emphasizingthe feasibility of discovering one PR that wouldbe useful for the treatment of multiple diseases(68). Moreover, recent studies have establisheda link between Ca2+ signaling and redox activitywithin the ER, suggesting that protection fromoxidative stress may also be a factor (82).
PRs that are known to activate the UPRare also useful in rescuing the folding, traf-ficking, and function of unrelated, misfolding-prone lysosomal enzymes (83). Co-applicationof MG-132 (a PR that functions as a proteasomeinhibitor and as a UPR activator) or celastrol(a PR that is both a UPR and a HSR activa-tor) with RNAi to separately inhibit each of thethree arms of the UPR and the HSR revealed astrong dependence of the action of these PRs ontwo or more arms of the UPR, but not the HSR.Enhancement of enzyme folding, trafficking,and activity by these PRs was further increasedby β-glucocerebrosidase-targeting PCs, evenfor a variant that cannot normally be rescuedby PCs (83). In fact, the activities of PCs andPRs exhibited synergy in several lysosomal stor-age disease contexts. Notably, such results sug-gest that these PRs improve the proteostasisnetwork capacity, enabling production of morefolded β-glucocerebrosidase for PC binding,thereby significantly increasing folded enzyme
www.annualreviews.org • Diseases of Proteostasis Deficiency 977
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
concentration and subsequent export to thelysosome (83). It is likely that this synergisticrescue of proteostasis arises from expanding theproteostasis boundary with the PR, while simul-taneously stabilizing the increased populationof the folded state by PC binding to the na-tive state of the folding-compromised enzyme,moving it even further within the proteosta-sis boundary. We anticipate that PRs and PCs,which offer mechanistically distinct solutions toenhancing proteostasis capacity, can be com-bined to address challenges to the proteostasisnetwork that could not be met as well by eithertype of agent alone.
An emerging category of PRs that pre-sumably function through regulation of tran-scription are the HDAC inhibitors (HDACi)that modulate the epigenome (84). HDACi ap-pear to modulate transcription, and therebyproteostasis, by preventing histone deacetyla-tion (85–88). HDACs comprise a group of 18enzymes. They are divided into three majorclasses, the Zn2+-dependent classes I and II,and the NAD+-dependent class III enzymes(sirtuins) (89–93), the latter being sensitiveto the natural product, resveratrol (94, 95).HDACs function posttranslationally to regu-late the level of acetylation and hence the ac-tivity of transcription factors and other pro-teins, such as the chaperone Hsp90 (85, 96,97) and HSF1, the latter illustrating the in-tegral role of HDACs in proteostasis (98).Increasing evidence suggests that HDACis,such as tubacin, appear to strongly influence theproteostasis network. Tubacin targets HDAC6.HDAC6, through its ability to regulate theacetylation status and activity of Hsp90, hasbeen proposed to control the cellular responseto stress through the interaction of Hsp90 withHSF1, which in turn controls the cellular re-sponse to multiple cytosolic stress events. Simi-larly, 4-phenylbutyrate (4-PBA), a putative low-affinity HDACi, has been shown in mousemodels to provide benefit for numerous mis-folding diseases that challenge the proteostasisnetwork, including metabolic syndrome (99),cystic fibrosis (100, 101) and α-1-antitrypsin(α1AT) deficiency (102), possibly by expanding
the proteostasis boundary (Figure 8). Changesin the proteostasis boundary in response toHDACi may also protect organisms from agingand aging-associated neurodegenerative dis-eases (see below).
Autophagy, a proteostasis pathway enablingprotein degradation, plays an important role incell maintenance by removing intracellular ag-gregates and misfolded proteins by deliveringthem to lysosomes. The autophagic pathwayhas three branches to facilitate this process, re-ferred to as macroautophagy, microautophagy,and chaperone-mediated autophagy (103, 104).Macroautophagy is negatively regulated bythe mTOR pathway, which is inhibited byrapamycin (105–107). Upregulation of mTORis protective against neurodegenerative disease(see below) (108, 109). Thus, we proposethat PRs, such as rapamycin, that regulatethe proteostasis boundary via their effects onautophagy pathways to remove unfolded, mis-folded, or aggregated protein could be usefulfor a broad range of diseases involving theaccumulation of aggregation-prone proteins inthe cytosol and trafficking compartments. Thelatter includes Z-variant α1AT aggregates,which accumulate in the ER and are eliminatedby macroautophagy (102, 110, 111). Recentstudies reveal that Ca2+-sensitive signalingrepresents another pathway by which PRscould control autophagy (112).
Although we have focused on PCs andPRs that modify the proteostasis network in acell-autonomous fashion, organismal biologyoperates in the context of interconnected tissueand organ systems. Thus, the responses ofproteostasis networks need to be globally coor-dinated. Recent work suggests that organismalcontrol of the proteostasis network can becell nonautonomous and centrally regulatedby neuronal signaling pathways. Exposure ofCaenorhabditis elegans to heat stress revealedthat the response of somatic cells requiredonly one thermosensory neuron, AFD (42,113). Thus, we speculate that coordinatedregulation of the proteostasis boundary withinand between organ systems serves to integratebehavioral, metabolic, and stress-related
978 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
pathways. PRs that mimic the activity of thesecell-nonautonomous pathways remain to bedeveloped, although incretin analogs may havePR properties (see below).
THE PROTEOSTASISNETWORK IN AGING
Processes associated with aging have directsignaling-mediated connections to the pro-teostasis network and are, therefore, capableof moving the proteostasis boundary, poten-tially explaining why so many loss- and gain-of-function diseases are triggered or exacer-bated by aging (Figure 8). Substantial progresshas been made in understanding factors con-tributing to aging through the IGF1-R signal-ing pathways, dietary restriction, mitochondrialrespiration, (114–117) and, more recently, theElt3, -5, and -6 signaling pathways, the lat-ter having genetic links to IGF1-R pathways(118). Longevity is markedly extended by re-ducing the level or activity of IGF1-R signal-ing in worm, fly, and mouse models (114–117).IGF1-R is a negative regulator of both FOXOand HSF1 signaling pathways that directly in-fluence the proteostasis network. One of sev-eral current models for the effects of aging onorganisms is that the proteostasis network be-comes burdened by the accumulation of pro-teins modified by reactive oxygen species or ox-idative metabolites, particularly in nondividingcells such as the neuron. Such modified proteinshave a tendency to misfold and/or aggregate,placing a significant demand on the proteostasisnetwork. To meet this challenge, it has recentlybeen demonstrated that reduction of IGF1-Rsignaling results in HSF1 activation and upreg-ulation of protective features of the proteosta-sis network. Thus, reduction in IGF1-R sig-naling appears to compensate for what wouldotherwise be a gradual, aging-related collapseof proteostasis owing to an increased misfold-ing/aggregation load. Intriguingly, a reductionin IGF1-R signaling could even “precondi-tion” aging cells for later proteostasis challenges(1). Thus, we suggest that modulation of theproteostasis network by PRs that reduce IGF1-
R signaling or upregulate HSF1 or FOXO arelikely to be effective for aging-associated degen-erative diseases linked to protein aggregationby maintaining or expanding the proteostasisboundary (Figure 8) and should, therefore, beof high interest to biotechnology and pharma-ceutical industries.
OBESITY AND THEPROTEOSTASIS NETWORK
Obesity is a major problem in industrializedsocieties and negatively influences the responseof the normally adaptable proteostasis net-work to stress, including that associated withmetabolic disease (99, 119–121). A high-fatdiet, a lack of exercise, and genetic influencesstrongly impact the ability of the proteostasisnetwork to maintain the function of β-cells inresponse to the increased demand for insulinand in response to dysregulation of insulin sig-naling (36, 122, 123). A large body of evidencenow suggests that when the rate of insulinproduction in the ER exceeds the cell’s capacityto fold and secrete it, numerous signaling path-ways, including the UPR, attempt to preventtoxicity and cell death. Thus, not only doesinsulin folding fail, but we speculate that theproteostasis boundary eventually collapses inresponse to the unmanageable misfolding load.Thus, the β-cell consequently loses its ability togenerate and maintain the structure and func-tion of other proteins essential for survival. Thisis illustrated in terms of the FoldFx model inFigure 9. The collapse of proteostasis capacityin the β-cell is exemplified not only by depo-sition of aggregates of the peptide hormone,amylin (124, 125), but also by β-cell death, pre-sumably resulting from a sustained activationof the PERK-ATF4 arm of the UPR, which islinked to death pathways through the transcrip-tion factor CHOP (34–36, 126). Although themost common diabetes drugs provide tempo-rary relief, they do so by further increasing in-sulin production in β-cells and, therefore, couldexacerbate the problem in the long term. Incontrast, HDACi, including 4-PBA (99, 127),and activators of sirtuins, such as resveratrol
www.annualreviews.org • Diseases of Proteostasis Deficiency 979
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
ba
Slow
Misfoldingrate
Fast
Slow
Misfoldingrate
Fast
Slow FastSlow Fast Foldingrate
Foldingrate
Stable
UnstableStability
Stable
UnstableStability
Increasedinsulin load Insulin load collapses
proteostasis boundary
Loss of function,aggregation,cell death, andtype 2 diabetesresults
Figure 9A potential mechanism for the pathogenesis of type 2 diabetes visualized in terms of the FoldFx model using a hypothetical proteininteraction network. (a) Increased synthesis of insulin (represented as a large node) challenges the proteostasis network. The insulin loadsaturates the proteostasis network, leaving little capacity for maintaining the normal protein folding load. (b) The proteostasis boundaryconstricts, leading to loss of β-cell function, protein aggregation (of the peptide hormone amylin in particular, black node), β-cell death,and disease. Because amylin is an intrinsically disordered peptide, the assignment of values for its folding kinetics and thermodynamicsreflects its dynamic equilibrium between degradation-competent and -incompetent states.
and its analogs (94, 128–130), restore β-cellhomeostasis in mouse models of type 2 dia-betes, perhaps by modulation of the highlycompromised proteostasis network. Intrigu-ingly, mimics of the secreted incretin classof hormones, including the glucagon-likepeptide-1 (GLP-1) receptor agonist exenatide(131) or inhibitors of dipeptidyl peptidase-IVthat block GLP-1 degradation (132), mayfunction by rebalancing the proteostasis net-work in the β-cell. Moreover, recent studiesrevealed that cell-nonautonomous inter-organcommunication between the liver and thepancreas via neuronal pathways controls β-cellfunction and insulin secretion (133). Thus,consistent with the response of worms to heatstress through neuronal signaling pathways(42), compounds such as exenatide and otherpolypeptide hormones may fall into the broadcategory of cell-nonautonomous PRs andpresent a more global biological solution foraugmentation of the proteostasis network topreserve insulin synthesis, protect β-cells fromapoptosis, and promote β-cell proliferation.
CANCER AND THEPROTEOSTASIS NETWORK
The unregulated cell division exhibited bycancer cells requires exceptionally high rates ofprotein synthesis and maintenance that are outof balance relative to the normal, fully differ-entiated state, taxing the proteostasis network(134–136). Hsp90 is a highly abundant chap-erone that has >300 clients, including manykinases involved in cell proliferation (http://www.picard.ch/downloads/downloads.htm).Unlike Hsp70, which is involved in cotrans-lational protein folding, Hsp90 is largelyinvolved in late events in protein foldingpathways, where it maintains proteins innear-native states for functional purposes.Hsp90 is sensitive to a number of inhibitorsthat belong to the ansamycin class of antibioticsincluding geldanamycin and its derivatives,radicicol, and a series of derivatives basedon purine and pyrazole scaffolds (137–139).Interestingly, cancer cells are highly sensitiveto PRs inhibiting Hsp90. Hsp90 inhibitorsare currently the focus of multiple anticancer
980 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
Slow
Misfoldingrate
Fast
Slow
FastFolding
rate
Stable
Unstable
Stability
Slow
Misfoldingrate
Fast
Slow
FastFolding
rate
Stable
Unstable
Stability
Hsp 90inhibitor
Cancer cell network;proteostasis boundary expanded
to handle high protein synthesis load
Cancer cell network with Hsp90 inhibited;
proteostasis boundary collapses
Cancercell death
Figure 10Treatment of cancer using Hsp90 inhibitors visualized in terms of the FoldFx model.
clinical trials (136, 140). We suggest thatdisrupting Hsp90 function in a cancer cellwould likely contract the proteostasis boundaryin the FoldFx model (Figure 10). This wouldlead to loss of function for many proteinswith marginal protein folding energetics,thereby killing the cancer cell, possibly withoutsignificantly impacting normal cell physiology.
Compounds that modulate oxidative fold-ing and metabolic pathways, the latter being acentral feature in proteome maintenance dur-ing stress and aging, are also likely candi-dates as PRs to treat cancer. Arsenicals foundin traditional Chinese medicine and used fortreatment of cancer (141) function as PRs pre-sumably by compromising the redox poten-tial of the cytosolic folding environment. Asa consequence, we predict that they signifi-cantly challenge the capacity of the proteostasisnetwork to fold proteins with disufide bondsand, hence, the ability of the cancer cell tomaintain viability (142). Although modula-tion of oxidative stress remains a major chal-lenge for PR development, compounds influ-encing S-nitrosylation, and/or inflammatorysignaling pathways, including NF-κB or PERKsignaling via activation of ATF4 and Nrf2transcription factors (41), and Ca2+-signaling
pathways that are linked to redox pathways mayrepresent major opportunities to reduce or pre-vent disease (143).
PRs that inhibit degradation are now partof an established clinical approach to preventgrowth of specific classes of cancer. Here, PRsthat are proteasome inhibitors illustrate thevalue of manipulating the proteostasis bound-ary to promote cell death (30, 144, 145).Proteasome-inhibiting PRs decrease the degra-dation rate of misfolded proteins. The result-ing increase in the concentration of misfoldedproteins likely saturates the proteostasis net-work, collapsing the proteostasis boundary be-low a threshold that can sustain cell viability (seeFigure 10). For example, in multiple myeloma,the proteasome inhibitor bortezomib selec-tively kills the cancerous plasma cells that pro-duce large amounts of immunoglobulin, some-times resulting in remission of the cancer (146,147). In terms of the FoldFx model, we specu-late that this selectivity arises because myelomacells are already challenged with a high proteinfolding load, and proteasome inhibition furtherdiminishes the capacity of the proteostasis net-work to maintain the proteostasis boundary ata functional level. The effect could be similarto that of treating other types of cancer cells
www.annualreviews.org • Diseases of Proteostasis Deficiency 981
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
with Hsp90 inhibitors, although different com-ponents of the proteostasis network are targeted(Figure 10).
AGE-RELATED DEGENERATIVEDISEASES AND PROTEOSTASIS
Although we know little about maintenanceof the intrinsically disordered proteome, it isclearly important as intrinsically disorderedproteins form the aggregates that trigger or ex-acerbate the clinically most important neurode-generative diseases, including the aggregationof Aβ associated with Alzheimer’s disease, α-synuclein associated with Parkinson’s disease,and the exon1 fragment of huntingtin associ-ated with Huntington’s disease. Longevity sig-naling pathways that influence cytosolic pro-teostasis have been shown to have strong effectson the progression of gain-of-toxic-functionneurodegenerative diseases in animal models(1).
In a C. elegans model of Aβ proteotoxic-ity, inhibiting the IGF1-R signaling pathway,which negatively regulates HSF1, revealed twoways in which a cell can use the proteostasisnetwork to protect itself (148). The first ac-tivity, enabled by upregulation of HSF1 tran-scription factor activity, is a disaggregation
activity that appears to reduce aggregate load.A second pathway, also downstream of IGF1-R signaling, utilizes the FOXO transcriptionprogram, apparently to enhance the forma-tion of large aggregates, thereby protectingthe cell from smaller, more toxic oligomersby placing these out of reach of the func-tional environment defined by the proteosta-sis boundary (148, 149). The putative effects ofthese pathways on proteostasis are illustratedin terms of the FoldFx model in Figure 11. Itwill be interesting to discern whether the pro-teostasis network generally uses aggregation-promoting activities to mitigate the effect ofacute and/or chronic failures of proteostasiswhen the capacity of the protective disaggrega-tion/degradative pathways becomes ineffectiveor saturated (150).
A second example of the proteostasisnetwork protecting the cell from aggregation-associated proteotoxicity is observed in C.elegans models harboring huntingtin polyQexpansions (17, 50, 151, 152). Here, the cyto-toxicity triggered by a polyQ aggregate loadis exacerbated by decreased HSF1 activity andameliorated by reduced IGF1-R signaling thatincreases HSF1-mediated transcription. More-over, introduction of another misfolding-proneprotein into C. elegans harboring huntingtin
Slow
Mis-foldingrate
Fast
Slow
Mis-foldingrate
Fast
Slow
Mis-foldingrate
Fast
Slow
FastFoldingrate Slow
FastFoldingrateSlow
FastFoldingrate
Stable
Unstable
Stability
Stable
Unstable
Stability
Stable
Unstable
Stability
HSF1pathways
HSF1 signaling expandsproteostasis boundary
FOXO signaling detoxifiesaggregates
FOXOpathways
Figure 11The effect of activating HSF1 pathways on protein aggregation diseases like Alzheimer’s and Huntington’s disease visualized in terms ofthe FoldFx model. HSF1 signaling expands the proteostasis boundary, allowing it to recognize the aggregation-prone protein (blacknode) and alleviating the gain of toxic function. In contrast, FOXO-based pathways promote a protective form of aggregation that yieldsless toxic aggregates. The detoxified aggregate is indicated in the plot on the left side of the figure by the white node. As in Figure 10,the assigned values for folding stability and kinetics for intrinsically disordered proteins are meant to suggest the partitioning betweenstates that are either susceptible to degradation or not.
982 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
polyQ expansions further sensitizes the organ-ism to polyQ proteotoxicity (50), presumably asa consequence of consumption of proteostasisnetwork capacity. This result illustrates thepoint that the proteostasis boundary for anyone protein is dependent on the overall burdenplaced on the proteostasis network by otherproteins. It also leads us to propose that any ge-netic predisposition toward protein misfolding/aggregation and/or disabling a component(s)of the proteostasis network could contributeto the sporadic forms of neurodegenera-tive diseases like Huntington’s, Alzheimer’s,Parkinson’s diseases, and amyotrophic lateralsclerosis (ALS), simply by placing an additionalload on the proteostasis network (50, 153).
Given the FoldFx model, it is apparent thataggregate load in neurodegenerative diseasecould be adjusted using either PCs or PRs. Oneapproach to reduce aggregate accumulation isto utilize kinetic stabilizers, as described above,to stabilize the native non-amyloidogenic stateof a given aggregation-prone protein, such astransthyretin. Another approach is to utilizesmall molecules that inhibit the aggregationof proteins at the self-assembly stage, such as4,5-dianilinophthalimide (154), phenolsulfon-phthalein (155), and others that appear to pre-vent the formation of mature (fibrillar) aggre-gates from nascent aggregates (156, 157). Apotential limitation of aggregation inhibitors isthat they may lead to the generation of smaller,more toxic oligomers, challenging the protec-tive aggregation-promoting pathways of theproteostasis network.
Other strategies for using PRs to amelio-rate neurodegenerative diseases beyond HSF1or FOXO activation include overexpressionof HDAC6, which has been shown to rescueneurodegeneration by way of the ubiquitin-proteasome system (UPS) and autophagy path-ways (158), and use of HDACi to protect againstParkinson’s (159) and Huntington’s diseases(160). These observations suggest that thesecompounds may have PR activities that arelinked to epigenetic features controlling theexpression of degradative and aging-associatedsignaling pathways, thereby regulating the
position of a node or the proteostasis boundary.A second possibility is suggested by the etiologyof Parkinson’s disease. A mutation in parkin,an E3 ubiquitin ligase (161), is associated withan early-onset form of Parkinson’s disease,suggesting that Parkinson’s disease involves adefect in the UPS. Therefore, PRs that poten-tiate parkin activity could ameliorate Parkin-son’s disease by repositioning the proteostasisboundary to favor degradation in general (162).Finally, in ALS (163), mutation of superoxidedismutase 1 leads to accumulation of reactiveoxygen species in both cell-autonomous (164)and cell-nonautonomous (165) fashions that ap-pear to contribute to protein misfolding, mito-chondrial dysfunction, and oxidative damage,thereby challenging the proteostasis network.PRs that maintain the proteostasis boundaryby dampening the impact of oxidative insult(163, 166) could be effective therapeutics forALS.
The examples discussed illustrate only afew of the many ways in which the increasingevidence in the literature suggests that manipu-lation of the proteostasis network and the pro-teostasis boundary in a compartment- and cell-type-specific fashion with PRs may be a usefulapproach to expand proteostasis network capac-ity and mitigate disease progression.
THE FUTURE OFPHARMACOLOGICMODULATION OFPROTEOSTASIS
The dependence of the function of distinct celltypes on their proteostasis network capacities,and hence the position and shape of their pro-teostasis boundaries, is becoming increasinglyevident. FoldFx describes a way to integratefolding energetics with proteostasis networkcapacity and provide a basis to think aboutadapting the proteostasis network for diseaseintervention. While the simplifications em-ployed in our mathematical model undoubtedlymask additional approaches that may apply to agiven cellular compartment or tissue environ-ment, it can be expanded to account for more
www.annualreviews.org • Diseases of Proteostasis Deficiency 983
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
complexity. It should be evident from this re-view that proteostasis is not as much about qual-ity control as it is about using the combinationof protein folding energetics and the adjustableproteostasis network capacity to optimally sup-port function (45, 46). Because the proteostasisnetwork and, hence, the proteostasis boundaryare adjusted by signaling pathways such as theUPR or the HSR, these pathways mesh proteinfunction with changing cellular and organismalneeds (167–169). Given the fundamental
interplay between folding energetics andproteostasis network capacity in maintainingthe proteome, it is likely that numerous classesof compounds will be found that function aseither PCs or PRs to ameliorate diseases ofproteostasis deficiency. These new classes ofpharmacologic agents (1, 170) should be widelyapplicable to the amelioration of human diseaseowing to the central roles that protein biogen-esis, folding, maintenance, and degradationpathways play in physiology and pathology.
DISCLOSURE STATEMENT
R.I.M., A.D., and J.W.K. are cofounders, shareholders, and paid consultants for Proteostasis Ther-apeutics Inc. and FoldRx pharmaceuticals ( J.W.K. only), which are developing drugs that could bePRs and PCs. W.E.B. is a shareholder in Proteostasis Therapeutics Inc. and a paid consultant forProteostasis Therapeutics Inc. and FoldRx pharmaceuticals. E.T.P. receives milestone paymentsfrom FoldRx pharmaceuticals.
ACKNOWLEDGMENTS
This work was supported by NIH grants GM033301, GM042336, HL079442, AG031097, andDK05187 to W.E.B.; NIH grants DK046335, AG018917, AG031097, and DK07529 to J.W.K.;the Lita Annenberg Hazen Foundation, the Bruce Ford and Anne Smith Bundy Foundationand the Skaggs Institute for Chemical Biology to J.W.K.; NIH, the McKnight Foundation forNeuroscience, and HHMI to A.D.; and NIH grants GM038109, GM081192, AG026647, andNS047331to R.I.M. We acknowledge the Cystic Fibrosis Foundation for support of W.E.B. Wethank Dr. Andrew Su, Genomics Institute of the Novartis Foundation, La Jolla, CA, for assistancein generating Figure 3, and Colleen Fearns for critically reading this manuscript.
LITERATURE CITED
1. Balch WE, Morimoto RI, Dillin A, Kelly JW. 2008. Adapting proteostasis for disease intervention.Science 319:916–19
2. Braun P, Rietman E, Vidal M. 2008. Networking metabolites and diseases. Proc. Natl. Acad. Sci. USA105:9849–50
3. Goh KI, Cusick ME, Valle D, Childs B, Vidal M, Barabasi AL. 2007. The human disease network. Proc.Natl. Acad. Sci. USA 104:8685–90
4. Ideker T, Sharan R. 2008. Protein networks in disease. Genome Res. 18:644–525. Anfinsen CB. 1973. Principles that govern the folding of protein chains. Science 181:223–306. Oliveberg M, Wolynes PG. 2005. The experimental survey of protein-folding energy landscapes. Q. Rev.
Biophys. 38:245–887. Udgaonkar JB. 2008. Multiple routes and structural heterogeneity in protein folding. Annu. Rev. Biophys.
37:489–5108. Dill KA, Ozkan SB, Shell MS, Weikl TR. 2008. The protein folding problem. Annu. Rev. Biophys.
37:289–3169. Auer S, Miller MA, Krivov SV, Dobson CM, Karplus M, Vendruscolo M. 2007. Importance of metastable
states in the free energy landscapes of polypeptide chains. Phys. Rev. Lett. 99:178104
984 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
10. Luheshi LM, Crowther DC, Dobson CM. 2008. Protein misfolding and disease: from the test tube tothe organism. Curr. Opin. Chem. Biol. 12:25–31
11. Uversky VN, Oldfield CJ, Dunker AK. 2008. Intrinsically disordered proteins in human diseases: intro-ducing the D2 concept. Annu. Rev. Biophys. 37:215–46
12. Das R, Baker D. 2008. Macromolecular modeling with rosetta. Annu. Rev. Biochem. 77:363–8213. Powers ET, Balch WE. 2008. Costly mistakes: translational infidelity and protein homeostasis.
Cell 134:204–614. Drummond DA, Wilke CO. 2008. Mistranslation-induced protein misfolding as a dominant constraint
on coding-sequence evolution. Cell 134:341–5215. Zhou HX, Rivas G, Minton AP. 2008. Macromolecular crowding and confinement: biochemical, bio-
physical, and potential physiological consequences. Annu. Rev. Biophys. 37:375–9716. Kelly JW, Balch WE. 2006. The integration of cell and chemical biology in protein folding. Nat. Chem.
Biol. 2:224–2717. Morimoto RI. 2008. Proteotoxic stress and inducible chaperone networks in neurodegenerative disease
and aging. Genes Dev. 22:1427–3818. Young JC, Agashe VR, Siegers K, Hartl FU. 2004. Pathways of chaperone-mediated protein folding in
the cytosol. Nat. Rev. Mol. Cell Biol. 5:781–9119. Saibil HR. 2008. Chaperone machines in action. Curr. Opin. Struct. Biol. 18:35–4220. Bukau B, Weissman J, Horwich A. 2006. Molecular chaperones and protein quality control. Cell 125:443–
5121. Brodsky JL, Chiosis G. 2006. Hsp70 molecular chaperones: emerging roles in human disease and iden-
tification of small molecule modulators. Curr. Top. Med. Chem. 6:1215–2522. Ni M, Lee AS. 2007. ER chaperones in mammalian development and human diseases. FEBS Lett.
581:3641–5123. Shimizu Y, Hendershot LM. 2007. Organization of the functions and components of the endoplasmic
reticulum. Adv. Exp. Med. Biol. 594:37–4624. Appenzeller-Herzog C, Ellgaard L. 2008. The human PDI family: versatility packed into a single fold.
Biochim. Biophys. Acta 1783:535–4825. Nagata K. 2003. HSP47 as a collagen-specific molecular chaperone: function and expression in normal
mouse development. Semin. Cell Dev. Biol. 14:275–8226. Hussain MM, Rava P, Pan X, Dai K, Dougan SK, et al. 2008. Microsomal triglyceride transfer protein
in plasma and cellular lipid metabolism. Curr. Opin. Lipidol. 19:277–8427. Horwich AL, Fenton WA, Chapman E, Farr GW. 2007. Two families of chaperonin: physiology and
mechanism. Annu. Rev. Cell Dev. Biol. 23:115–4528. Tang YC, Chang HC, Chakraborty K, Hartl FU, Hayer-Hartl M. 2008. Essential role of the chaperonin
folding compartment in vivo. EMBO J. 27:1458–6829. Sharma S, Chakraborty K, Muller BK, Astola N, Tang YC, et al. 2008. Monitoring protein conformation
along the pathway of chaperonin-assisted folding. Cell 133:142–5330. Konstantinova IM, Tsimokha AS, Mittenberg AG. 2008. Role of proteasomes in cellular regulation.
Int. Rev. Cell Mol. Biol. 267:59–12431. Tasaki T, Kwon YT. 2007. The mammalian N-end rule pathway: new insights into its components and
physiological roles. Trends Biochem. Sci. 32:520–2832. Levine B, Kroemer G. 2008. Autophagy in the pathogenesis of disease. Cell 132:27–4233. Kundu M, Thompson CB. 2008. Autophagy: basic principles and relevance to disease. Annu. Rev. Pathol.
3:427–5534. Malhotra JD, Kaufman RJ. 2007. The endoplasmic reticulum and the unfolded protein response. Semin.
Cell Dev. Biol. 18:716–3135. Ron D, Walter P. 2007. Signal integration in the endoplasmic reticulum unfolded protein response.
Nat. Rev. Mol. Cell Biol. 8:519–2936. Lin JH, Walter P, Yen TS. 2008. Endoplasmic reticulum stress in disease pathogenesis. Annu. Rev. Pathol.
3:399–42537. Shamovsky I, Nudler E. 2008. New insights into the mechanism of heat shock response activation.
Cell Mol. Life Sci. 65:855–61
www.annualreviews.org • Diseases of Proteostasis Deficiency 985
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
38. Petersen OH, Michalak M, Verkhratsky A. 2005. Calcium signalling: past, present and future. Cell Calcium38:161–69
39. Burdakov D, Petersen OH, Verkhratsky A. 2005. Intraluminal calcium as a primary regulator of endo-plasmic reticulum function. Cell Calcium 38:303–10
40. Medzhitov R. 2008. Origin and physiological roles of inflammation. Nature 454:428–3541. Zhang K, Kaufman RJ. 2008. From endoplasmic-reticulum stress to the inflammatory response. Nature
454:455–6242. Prahlad V, Cornelius T, Morimoto RI. 2008. Regulation of the cellular heat shock response in Caenorhab-
ditis elegans by thermosensory neurons. Science 320:811–1443. Murphy KG, Bloom SR. 2006. Gut hormones and the regulation of energy homeostasis. Nature 444:854–
5944. Su AI, Wiltshire T, Batalov S, Lapp H, Ching KA, et al. 2004. A gene atlas of the mouse and human
protein-encoding transcriptomes. Proc. Natl. Acad. Sci. USA 101:6062–6745. Sekijima Y, Wiseman RL, Matteson J, Hammarstrom P, Miller SR, et al. 2005. The biological and
chemical basis for tissue-selective amyloid disease. Cell 121:73–8546. Wiseman RL, Powers ET, Buxbaum JN, Kelly JW, Balch WE. 2007. An adaptable standard for protein
export from the endoplasmic reticulum. Cell 131:809–2147. Haynes C, Oldfield CJ, Ji F, Klitgord N, Cusick ME, et al. 2006. Intrinsic disorder is a common feature
of hub proteins from four eukaryotic interactomes. PLoS Comput. Biol. 2:e10048. Hillenmeyer ME, Fung E, Wildenhain J, Pierce SE, Hoon S, et al. 2008. The chemical genomic portrait
of yeast: uncovering a phenotype for all genes. Science 320:362–6549. Brignull HR, Morley JF, Morimoto RI. 2007. The stress of misfolded proteins: C. elegans models for
neurodegenerative disease and aging. Adv. Exp. Med. Biol. 594:167–8950. Gidalevitz T, Ben-Zvi A, Ho KH, Brignull HR, Morimoto RI. 2006. Progressive disruption of cellular
protein folding in models of polyglutamine diseases. Science 311:1471–7451. Palotai R, Szalay MS, Csermely P. 2008. Chaperones as integrators of cellular networks: changes of
cellular integrity in stress and diseases. IUBMB Life 60:10–1852. Conn PM, Ulloa-Aguirre A, Ito J, Janovick JA. 2007. G protein–coupled receptor trafficking in health
and disease: lessons learned to prepare for therapeutic mutant rescue in vivo. Pharmacol. Rev. 59:225–5053. Millar NS, Harkness PC. 2008. Assembly and trafficking of nicotinic acetylcholine receptors (Review).
Mol. Membr. Biol. 25:279–9254. Grabowski GA. 2008. Treatment perspectives for the lysosomal storage diseases. Expert Opin. Emerg.
Drugs 13:197–21155. Sawkar AR, Cheng W-C, Beutler E, Wong C-H, Balch WE, Kelly JW. 2002. Chemical chaperones
increase the cellular activity of N370S beta-glucosidase: a therapeutic strategy for Gaucher disease. Proc.Natl. Acad. Sci. USA 99:15428–33
56. Fan J-Q, Ishii S, Asano N, Suzuki Y. 1999. Accelerated transport and maturation of lysosomal alpha-galactosidase A in Fabry lymphoblasts by an enzyme inhibitor. Nat. Med. 5:112–15
57. Yoo CL, Yu GJ, Yang B, Robins LI, Verkman AS, Kurth MJ. 2008. 4′-Methyl-4,5′-bithiazole-basedcorrectors of defective �F508-CFTR cellular processing. Bioorg. Med. Chem. Lett. 18:2610–14
58. Amaral MD, Kunzelmann K. 2007. Molecular targeting of CFTR as a therapeutic approach to cysticfibrosis. Trends Pharmacol. Sci. 28:334–41
59. Pedemonte N, Lukacs GL, Du K, Caci E, Zegarra-Moran O, et al. 2005. Small-molecule correctors ofdefective �F508-CFTR cellular processing identified by high-throughput screening. J. Clin. Investig.115:2564–71
60. Loo TW, Bartlett MC, Clarke DM. 2008. Correctors promote folding of the CFTR in the endoplasmicreticulum. Biochem. J. 413:29–36
61. Noel S, Strale PO, Dannhoffer L, Wilke M, DeJonge H, et al. 2008. Stimulation of salivary secretion invivo by CFTR potentiators in Cftr+/+ and Cftr−/− mice. J. Cyst. Fibros. 7:128–33
62. Wang Y, Bartlett MC, Loo TW, Clarke DM. 2006. Specific rescue of cystic fibrosis transmembraneconductance regulator processing mutants using pharmacological chaperones. Mol. Pharmacol. 70:297–302
986 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
63. Wang Y, Loo TW, Bartlett MC, Clarke DM. 2007. Additive effect of multiple pharmacological chap-erones on maturation of CFTR processing mutants. Biochem. J. 406:257–63
64. Yu Z, Sawkar AR, Kelly JW. 2007. Pharmacologic chaperoning as a strategy to treat Gaucher disease.FEBS J. 274:4944–50
65. Yu Z, Sawkar AR, Whalen LJ, Wong CH, Kelly JW. 2007. Isofagomine- and 2,5-anhydro-2,5-imino-d-glucitol-based glucocerebrosidase pharmacological chaperones for Gaucher disease intervention. J.Med. Chem. 50:94–100
66. Hammarstrom P, Wiseman RL, Powers ET, Kelly JW. 2003. Prevention of transthyretin amyloid diseaseby changing protein misfolding energetics. Science 299:713–16
67. Johnson SM, Wiseman RL, Sekijima Y, Green NS, Adamski-Werner SL, Kelly JW. 2005. Native statekinetic stabilization as a strategy to ameliorate protein misfolding diseases: a focus on the transthyretinamyloidoses. Acc. Chem. Res. 38:911–21
68. Mu T-W, Fowler DM, Kelly JW. 2008. Partial restoration of mutant enzyme homeostasis in threedistinct lysosomal storage disease cell lines by altering calcium homeostasis. PLoS Biol. 6:e26
69. Obrenovitch TP. 2008. Molecular physiology of preconditioning-induced brain tolerance to ischemia.Physiol. Rev. 88:211–47
70. Boyce M, Bryant KF, Jousse C, Long K, Harding HP, et al. 2005. A selective inhibitor of eIF2α dephos-phorylation protects cells from ER stress. Science 307:935–39
71. Boyce M, Py BF, Ryazanov AG, Minden JS, Long K, et al. 2008. A pharmacoproteomic approachimplicates eukaryotic elongation factor 2 kinase in ER stress-induced cell death. Cell Death Differ. 15:589–99
72. Trott A, West JD, Klaic L, Westerheide SD, Silverman RB, et al. 2008. Activation of heat shock andantioxidant responses by the natural product celastrol: transcriptional signatures of a thiol-targetedmolecule. Mol. Biol. Cell 19:1104–12
73. Corson TW, Crews CM. 2007. Molecular understanding and modern application of traditionalmedicines: triumphs and trials. Cell 130:769–74
74. Westerheide SD, Morimoto RI. 2005. Heat shock response modulators as therapeutic tools for diseasesof protein conformation. J. Biol. Chem. 280:33097–100
75. Wang J, Gines S, MacDonald ME, Gusella JF. 2005. Reversal of a full-length mutant huntingtinneuronal cell phenotype by chemical inhibitors of polyglutamine-mediated aggregation. BMC Neurosci.6:1
76. Zhang T, Hamza A, Cao X, Wang B, Yu S, et al. 2008. A novel Hsp90 inhibitor to disrupt Hsp90/Cdc37complex against pancreatic cancer cells. Mol. Cancer Ther. 7:162–70
77. Jung HW, Chung YS, Kim YS, Park YK. 2007. Celastrol inhibits production of nitric oxide and proin-flammatory cytokines through MAPK signal transduction and NF-κB in LPS-stimulated BV-2 microglialcells. Exp. Mol. Med. 39:715–21
78. Yang H, Shi G, Dou QP. 2007. The tumor proteasome is a primary target for the natural anticancercompound Withaferin A isolated from “Indian winter cherry”. Mol. Pharmacol. 71:426–37
79. Li H, Zhang YY, Tan HW, Jia YF, Li D. 2008. Therapeutic effect of tripterine on adjuvant arthritis inrats. J. Ethnopharmacol. 118:479–84
80. Merrill GF, Goldberg E. 2001. Antioxidant properties of acetaminophen and cardioprotection. Basic Res.Cardiol. 96:423–30
81. Shertzer HG, Schneider SN, Kendig EL, Clegg DJ, D’Alessio DA, Genter MB. 2008. Acetaminophennormalizes glucose homeostasis in mouse models for diabetes. Biochem. Pharmacol. 75:1402–10
82. Gorlach A, Klappa P, Kietzmann T. 2006. The endoplasmic reticulum: folding, calcium homeostasis,signaling, and redox control. Antioxid. Redox Signal. 8:1391–418
83. Mu T-W, Ong DST, Wang Y-J, Balch WE, Yates JR, et al. 2008. Chemical and biological approachessynergize to ameliorate protein-folding diseases. Cell 134:769–81
84. Lee KK, Workman JL. 2007. Histone acetyltransferase complexes: One size doesn’t fit all. Nat. Rev. Mol.Cell Biol. 8:284–95
85. Sadoul K, Boyault C, Pabion M, Khochbin S. 2008. Regulation of protein turnover by acetyltransferasesand deacetylases. Biochimie 90:306–12
www.annualreviews.org • Diseases of Proteostasis Deficiency 987
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
86. Echaniz-Laguna A, Bousiges O, Loeffler JP, Boutillier AL. 2008. Histone deacetylase inhibitors: thera-peutic agents and research tools for deciphering motor neuron diseases. Curr. Med. Chem. 15:1263–73
87. Xu WS, Parmigiani RB, Marks PA. 2007. Histone deacetylase inhibitors: molecular mechanisms ofaction. Oncogene 26:5541–52
88. Martin M, Kettmann R, Dequiedt F. 2007. Class IIa histone deacetylases: regulating the regulators.Oncogene 26:5450–67
89. Guarente L. 2007. Sirtuins in aging and disease. Cold Spring Harb. Symp. Quant. Biol. 72:483–8890. Feige JN, Johan A. 2008. Transcriptional targets of sirtuins in the coordination of mammalian physiology.
Curr. Opin. Cell Biol. 20:303–991. Chen D, Guarente L. 2007. SIR2: a potential target for calorie restriction mimetics. Trends Mol. Med.
13:64–7192. Fraga MF, Esteller M. 2007. Epigenetics and aging: the targets and the marks. Trends Genet. 23:413–1893. Westphal CH, Dipp MA, Guarente L. 2007. A therapeutic role for sirtuins in diseases of aging? Trends
Biochem. Sci. 32:555–6094. Milne JC, Lambert PD, Schenk S, Carney DP, Smith JJ, et al. 2007. Small molecule activators of SIRT1
as therapeutics for the treatment of type 2 diabetes. Nature 450:712–1695. Milne JC, Denu JM. 2008. The Sirtuin family: therapeutic targets to treat diseases of aging. Curr. Opin.
Chem. Biol. 12:11–1796. Valenzuela-Fernandez A, Cabrero JR, Serrador JM, Sanchez-Madrid F. 2008. HDAC6: a key regulator
of cytoskeleton, cell migration and cell-cell interactions. Trends Cell Biol. 18:291–9797. Matthias P, Yoshida M, Khochbin S. 2008. HDAC6 a new cellular stress surveillance factor. Cell Cycle
7:7–1098. Westerheide SD, Anckar J, Stevens SM, Sistonen L, Morimoto RI. 2009. Stress-inducible regulation of
heat shock factor 1 by the deacetylase SIRT1. Science 323:1063–6699. Ozcan U, Yilmaz E, Ozcan L, Furuhashi M, Vaillancourt E, et al. 2006. Chemical chaperones reduce
ER stress and restore glucose homeostasis in a mouse model of type 2 diabetes. Science 313:1137–40100. Singh OV, Pollard HB, Zeitlin PL. 2008. Chemical rescue of �F508-CFTR mimics genetic repair in
cystic fibrosis bronchial epithelial cells. Mol. Cell Proteomics 7:1099–110101. Pruliere-Escabasse V, Planes C, Escudier E, Fanen P, Coste A, Clerici C. 2007. Modulation of epithelial
sodium channel trafficking and function by sodium 4-phenylbutyrate in human nasal epithelial cells.J. Biol. Chem. 282:34048–57
102. Perlmutter DH. 2006. The role of autophagy in α-1-antitrypsin deficiency: a specific cellular responsein genetic diseases associated with aggregation-prone proteins. Autophagy 2:258–63
103. Ding WX, Yin XM. 2008. Sorting, recognition and activation of the misfolded protein degradationpathways through macroautophagy and the proteasome. Autophagy 4:141–50
104. Dice JF. 2007. Chaperone-mediated autophagy. Autophagy 3:295–99105. Pattingre S, Espert L, Biard-Piechaczyk M, Codogno P. 2008. Regulation of macroautophagy by mTOR
and Beclin 1 complexes. Biochimie 90:313–23106. Sarkar S, Ravikumar B, Floto RA, Rubinsztein DC. 2009. Rapamycin and mTOR-independent au-
tophagy inducers ameliorate toxicity of polyglutamine-expanded huntingtin and related proteinopathies.Cell Death Differ. 16:46–56
107. Zemke D, Azhar S, Majid A. 2007. The mTOR pathway as a potential target for the development oftherapies against neurological disease. Drug News Perspect. 20:495–99
108. Reiling JH, Sabatini DM. 2008. Increased mTORC1 signaling UPRegulates stress. Mol. Cell 29:533–35109. Ozcan U, Ozcan L, Yilmaz E, Duvel K, Sahin M, et al. 2008. Loss of the tuberous sclerosis complex
tumor suppressors triggers the unfolded protein response to regulate insulin signaling and apoptosis.Mol. Cell 29:541–51
110. Perlmutter DH. 2009. Autophagic disposal of the aggregation-prone protein that causes liver inflamma-tion and carcinogenesis in α-1-antitrypsin deficiency. Cell Death Differ. 16:39–45
111. Granell S, Baldini G. 2008. Inclusion bodies and autophagosomes: Are ER-derived protective organellesdifferent than classical autophagosomes? Autophagy 4:375–77
112. Hetz C, Glimcher L. 2008. The daily job of night killers: alternative roles of the BCL-2 family inorganelle physiology. Trends Cell Biol. 18:38–44
988 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
113. Garcia SM, Casanueva MO, Silva MC, Amaral MD, Morimoto RI. 2007. Neuronal signaling modulatesprotein homeostasis in Caenorhabditis elegans postsynaptic muscle cells. Genes Dev. 21:3006–16
114. Giannakou ME, Partridge L. 2007. Role of insulin-like signalling in Drosophila lifespan. Trends Biochem.Sci. 32:180–88
115. Mair W, Dillin A. 2008. Aging and survival: the genetics of life span extension by dietary restriction.Annu. Rev. Biochem. 77:727–54
116. Hansen M, Hsu AL, Dillin A, Kenyon C. 2005. New genes tied to endocrine, metabolic, and dietaryregulation of lifespan from a Caenorhabditis elegans genomic RNAi screen. PLoS Genet. 1:119–28
117. Tullet JM, Hertweck M, An JH, Baker J, Hwang JY, et al. 2008. Direct inhibition of the longevity-promoting factor SKN-1 by insulin-like signaling in C. elegans. Cell 132:1025–38
118. Budovskaya YV, Wu K, Southworth LK, Jiang M, Tedesco P, et al. 2008. An elt-3/elt-5/elt-6 GATAtranscription circuit guides aging in C. elegans. Cell 134:291–303
119. Kahn SE, Hull RL, Utzschneider KM. 2006. Mechanisms linking obesity to insulin resistance and type2 diabetes. Nature 444:840–46
120. Gregor MG, Hotamisligil GS. 2007. Adipocyte stress: the endoplasmic reticulum and metabolic disease.J. Lipid Res. 48:1905–14
121. Scheuner D, Kaufman RJ. 2008. The unfolded protein response: a pathway that links insulin demandwith β-cell failure and diabetes. Endocr. Rev. 29:317–33
122. Gregor MF, Hotamisligil GS. 2007. Thematic review series: Adipocyte Biology. Adipocyte stress: theendoplasmic reticulum and metabolic disease. J. Lipid Res. 48:1905–14
123. Eizirik DL, Cardozo AK, Cnop M. 2008. The role for endoplasmic reticulum stress in diabetes mellitus.Endocr. Rev. 29:42–61
124. Haataja L, Gurlo T, Huang CJ, Butler PC. 2008. Islet amyloid in type 2 diabetes, and the toxic oligomerhypothesis. Endocr. Rev. 29:303–16
125. Hull RL, Westermark GT, Westermark P, Kahn SE. 2004. Islet amyloid: a critical entity in the patho-genesis of type 2 diabetes. J. Clin. Endocrinol. Metab. 89:3629–43
126. Song B, Scheuner D, Ron D, Pennathur S, Kaufman RJ. 2008. CHOP deletion reduces oxidative stress,improves β-cell function, and promotes cell survival in multiple mouse models of diabetes. J. Clin.Investig. 118:3378–89
127. Dashwood RH, Ho E. 2007. Dietary histone deacetylase inhibitors: from cells to mice to man. Semin.Cancer Biol. 17:363–69
128. Metoyer CF, Pruitt K. 2008. The role of sirtuin proteins in obesity. Pathophysiology 15:103–8129. Elliott PJ, Jirousek M. 2008. Sirtuins: novel targets for metabolic disease. Curr. Opin. Investig. Drugs
9:371–78130. Bordone L, Guarente L. 2007. Sirtuins and β-cell function. Diabetes Obes. Metab. 9(Suppl. 2):23–27131. Salehi M, Aulinger BA, D’Alessio DA. 2008. Targeting β-cell mass in type 2 diabetes: promise and
limitations of new drugs based on incretins. Endocr. Rev. 29:367–79132. Halimi S. 2008. DPP-4 inhibitors and GLP-1 analogues: for whom? Which place for incretins in the
management of type 2 diabetic patients? Diabetes Metab. 34(Suppl. 2):S91–95133. Imai J, Katagiri H, Yamada T, Ishigaki Y, Suzuki T, et al. 2008. Regulation of pancreatic β-cell mass by
neuronal signals from the liver. Science 322:1250–54134. Neckers L. 2007. Heat shock protein 90: the cancer chaperone. J. Biosci. 32:517–30135. Brown MA, Zhu L, Schmidt C, Tucker PW. 2007. Hsp90—from signal transduction to cell transforma-
tion. Biochem. Biophys. Res. Commun. 363:241–46136. Pearl LH, Prodromou C, Workman P. 2008. The Hsp90 molecular chaperone: an open and shut case
for treatment. Biochem. J. 410:439–53137. Chiosis G. 2006. Discovery and development of purine-scaffold Hsp90 inhibitors. Curr. Top. Med. Chem.
6:1183–91138. Solit DB, Chiosis G. 2008. Development and application of Hsp90 inhibitors. Drug Discov. Today 13:38–
43139. Taldone T, Gozman A, Maharaj R, Chiosis G. 2008. Targeting Hsp90: small-molecule inhibitors and
their clinical development. Curr. Opin. Pharmacol. 8:370–74
www.annualreviews.org • Diseases of Proteostasis Deficiency 989
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
140. Powers MV, Workman P. 2007. Inhibitors of the heat shock response: biology and pharmacology. FEBSLett. 581:3758–69
141. Wang L, Zhou GB, Liu P, Song JH, Liang Y, et al. 2008. Dissection of mechanisms of Chinese medicinalformula Realgar-Indigo naturalis as an effective treatment for promyelocytic leukemia. Proc. Natl. Acad.Sci. USA 105:4826–31
142. Dilda PJ, Hogg PJ. 2007. Arsenical-based cancer drugs. Cancer Treat. Rev. 33:542–64143. Nakamura T, Lipton SA. 2008. Emerging roles of S-nitrosylation in protein misfolding and neurode-
generative diseases. Antioxid. Redox Signal. 10:87–101144. Yang H, Landis-Piwowar KR, Chen D, Milacic V, Dou QP. 2008. Natural compounds with proteasome
inhibitory activity for cancer prevention and treatment. Curr. Protein Pept. Sci. 9:227–39145. Bossi G, Sacchi A. 2007. Restoration of wild-type p53 function in human cancer: relevance for tumor
therapy. Head Neck 29:272–84146. Chauhan D, Hideshima T, Anderson KC. 2008. Targeting proteasomes as therapy in multiple myeloma.
Adv. Exp. Med. Biol. 615:251–60147. Mateos MV, San Miguel JF. 2007. Bortezomib in multiple myeloma. Best Pract. Res. Clin. Haematol.
20:701–15148. Cohen E, Bieschke J, Perciavalle RM, Kelly JW, Dillin A. 2006. Opposing activities protect against
age-onset proteotoxicity. Science 313:1604–10149. Haass C, Selkoe DJ. 2007. Soluble protein oligomers in neurodegeneration: lessons from the Alzheimer’s
amyloid β-peptide. Nat. Rev. Mol. Cell Biol. 8:101–12150. Kaganovich D, Kopito R, Frydman J. 2008. Misfolded proteins partition between two distinct quality
control compartments. Nature 454:1088–95151. Morimoto RI. 2006. Stress, aging, and neurodegenerative disease. N. Engl. J. Med. 355:2254–55152. Brignull HR, Moore FE, Tang SJ, Morimoto RI. 2006. Polyglutamine proteins at the pathogenic thresh-
old display neuron-specific aggregation in a pan-neuronal Caenorhabditis elegans model. J. Neurosci.26:7597–606
153. Shao J, Diamond MI. 2007. Polyglutamine diseases: emerging concepts in pathogenesis and therapy.Hum. Mol. Genet. 16(Spec. 2):R115–23
154. Wang H, Duennwald ML, Roberts BE, Rozeboom LM, Zhang YL, et al. 2008. Direct and selectiveelimination of specific prions and amyloids by 4,5-dianilinophthalimide and analogs. Proc. Natl. Acad.Sci. USA 105:7159–64
155. Levy M, Porat Y, Bacharach E, Shalev DE, Gazit E. 2008. Phenolsulfonphthalein, but not phenolph-thalein, inhibits amyloid fibril formation: implications for the modulation of amyloid self-assembly.Biochemistry 47:5896–904
156. Gan L. 2007. Therapeutic potential of sirtuin-activating compounds in Alzheimer’s disease. Drug NewsPerspect. 20:233–39
157. Selkoe DJ. 2007. Developing preventive therapies for chronic diseases: lessons learned from Alzheimer’sdisease. Nutr. Rev. 65:S239–43
158. Pandey UB, Nie Z, Batlevi Y, McCray BA, Ritson GP, et al. 2007. HDAC6 rescues neurodegenerationand provides an essential link between autophagy and the UPS. Nature 447:859–63
159. Outeiro TF, Marques O, Kazantsev A. 2008. Therapeutic role of sirtuins in neurodegenerative disease.Biochim. Biophys. Acta 1782:363–69
160. Thomas EA, Coppola G, Desplats PA, Tang B, Soragni E, et al. 2008. The HDAC inhibitor 4b amelio-rates the disease phenotype and transcriptional abnormalities in Huntington’s disease transgenic mice.Proc. Natl. Acad. Sci. USA 105:15564–69
161. Winklhofer KF. 2007. The parkin protein as a therapeutic target in Parkinson’s disease. Expert Opin.Ther. Targets 11:1543–52
162. Lim KL, Tan JM. 2007. Role of the ubiquitin proteasome system in Parkinson’s disease. BMC Biochem.8(Suppl. 1):S13
163. Valdmanis PN, Rouleau GA. 2008. Genetics of familial amyotrophic lateral sclerosis. Neurology 70:144–52
990 Powers et al.
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
ANRV378-BI78-33 ARI 5 May 2009 15:3
164. Gruzman A, Wood WL, Alpert E, Prasad MD, Miller RG, et al. 2007. Common molecular signature inSOD1 for both sporadic and familial amyotrophic lateral sclerosis. Proc. Natl. Acad. Sci. USA 104:12524–29
165. Yamanaka K, Chun SJ, Boillee S, Fujimori-Tonou N, Yamashita H, et al. 2008. Astrocytes as determinantsof disease progression in inherited amyotrophic lateral sclerosis. Nat. Neurosci. 11:251–53
166. Winyard PG, Moody CJ, Jacob C. 2005. Oxidative activation of antioxidant defence. Trends Biochem. Sci.30:453–61
167. Sangster TA, Salathia N, Undurraga S, Milo R, Schellenberg K, et al. 2008. HSP90 affects the expressionof genetic variation and developmental stability in quantitative traits. Proc. Natl. Acad. Sci. USA 105:2963–68
168. Rutherford S, Hirate Y, Swalla BJ. 2007. The Hsp90 capacitor, developmental remodeling, and evolution:the robustness of gene networks and the curious evolvability of metamorphosis. Crit. Rev. Biochem. Mol.Biol. 42:355–72
169. Cowen LE, Lindquist S. 2005. Hsp90 potentiates the rapid evolution of new traits: drug resistance indiverse fungi. Science 309:2185–89
170. Hopkins AL. 2008. Network pharmacology: the next paradigm in drug discovery. Nat. Chem. Biol. 4:682–90
171. Gurkan C, Lapp H, Alory C, Su AI, Hogenesch JB, Balch WE. 2005. Large-scale profiling of RabGTPase trafficking networks: the membrane. Mol. Biol. Cell 16:3847–64
www.annualreviews.org • Diseases of Proteostasis Deficiency 991
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
AR378-FM ARI 7 May 2009 15:43
Annual Review ofBiochemistry
Volume 78, 2009
ContentsPreface � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �v
Prefatory Articles
FrontispieceE. Peter Geiduschek � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �xii
Without a License, or Accidents Waiting to HappenE. Peter Geiduschek � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 1
FrontispieceJames C. Wang � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �30
A Journey in the World of DNA Rings and BeyondJames C. Wang � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �31
Biochemistry and Disease Theme
The Biochemistry of Disease: Desperately Seeking SyzygyJohn W. Kozarich � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �55
Biosynthesis of Phosphonic and Phosphinic Acid Natural ProductsWilliam W. Metcalf and Wilfred A. van der Donk � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �65
New Antivirals and Drug ResistancePeter M. Colman � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �95
Multidrug Resistance in BacteriaHiroshi Nikaido � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 119
Conformational Pathology of the Serpins: Themes, Variations,and Therapeutic StrategiesBibek Gooptu and David A. Lomas � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 147
Getting a Grip on Prions: Oligomers, Amyloids, and PathologicalMembrane InteractionsByron Caughey, Gerald S. Baron, Bruce Chesebro, and Martin Jeffrey � � � � � � � � � � � � � � � � � 177
Ubiquitin-Mediated Protein Regulation
RING Domain E3 Ubiquitin LigasesRaymond J. Deshaies and Claudio A.P. Joazeiro � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 399
Regulation and Cellular Roles of Ubiquitin-SpecificDeubiquitinating EnzymesFrancisca E. Reyes-Turcu, Karen H. Ventii, and Keith D. Wilkinson � � � � � � � � � � � � � � � � � � � � 363
vi
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
AR378-FM ARI 7 May 2009 15:43
Recognition and Processing of Ubiquitin-Protein Conjugatesby the ProteasomeDaniel Finley � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 477
Degradation of Activated Protein Kinases by UbiquitinationZhimin Lu and Tony Hunter � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 435
The Role of Ubiquitin in NF-κB Regulatory PathwaysBrian Skaug, Xiaomo Jiang, and Zhijian J. Chen � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 769
Biological and Chemical Approaches to Diseases of ProteostasisDeficiencyEvan T. Powers, Richard I. Morimoto, Andrew Dillin, Jeffery W. Kelly,and William E. Balch � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 959
Gene Expression
RNA Polymerase Active Center: The Molecular Engineof TranscriptionEvgeny Nudler � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 335
Genome-Wide Views of Chromatin StructureOliver J. Rando and Howard Y. Chang � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 245
The Biology of Chromatin Remodeling ComplexesCedric R. Clapier and Bradley R. Cairns � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 273
The Structural and Functional Diversity of Metabolite-BindingRiboswitchesAdam Roth and Ronald R. Breaker � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 305
Lipid and Membrane Biogenesis
Genetic and Biochemical Analysis of Non-Vesicular Lipid TrafficDennis R. Voelker � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 827
Cholesterol 24-Hydroxylase: An Enzyme of Cholesterol Turnoverin the BrainDavid W. Russell, Rebekkah W. Halford, Denise M.O. Ramirez, Rahul Shah,and Tiina Kotti � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �1017
Lipid-Dependent Membrane Protein TopogenesisWilliam Dowhan and Mikhail Bogdanov � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 515
Single-Molecule Studies of the Neuronal SNARE Fusion MachineryAxel T. Brunger, Keith Weninger, Mark Bowen, and Steven Chu � � � � � � � � � � � � � � � � � � � � � � � 903
Mechanisms of EndocytosisGary J. Doherty and Harvey T. McMahon � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 857
Contents vii
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
AR378-FM ARI 7 May 2009 15:43
Recent Advances in Biochemistry
Motors, Switches, and Contacts in the ReplisomeSamir M. Hamdan and Charles C. Richardson � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 205
Large-Scale Structural Biology of the Human ProteomeAled Edwards � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 541
Collagen Structure and StabilityMatthew D. Shoulders and Ronald T. Raines � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 929
The Structural and Biochemical Foundations of Thiamin BiosynthesisChristopher T. Jurgenson, Tadhg P. Begley, and Steven E. Ealick � � � � � � � � � � � � � � � � � � � � � � � � 569
Proton-Coupled Electron Transfer in Biology: Results fromSynergistic Studies in Natural and Model SystemsSteven Y. Reece and Daniel G. Nocera � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 673
Mechanism of Mo-Dependent NitrogenaseLance C. Seefeldt, Brian M. Hoffman, and Dennis R. Dean � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 701
Inorganic Polyphosphate: Essential for Growth and SurvivalNarayana N. Rao, Marıa R. Gomez-Garcıa, and Arthur Kornberg � � � � � � � � � � � � � � � � � � � � � 605
Essentials for ATP Synthesis by F1F0 ATP SynthasesChristoph von Ballmoos, Alexander Wiedenmann, and Peter Dimroth � � � � � � � � � � � � � � � � � � 649
The Chemical Biology of Protein PhosphorylationMary Katherine Tarrant and Philip A. Cole � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 797
Sphingosine 1-Phosphate Receptor SignalingHugh Rosen, Pedro J. Gonzalez-Cabrera, M. Germana Sanna, and Steven Brown � � � � 743
The Advent of Near-Atomic Resolution in Single-Particle ElectronMicroscopyYifan Cheng and Thomas Walz � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 723
Super-Resolution Fluorescence MicroscopyBo Huang, Mark Bates, and Xiaowei Zhuang � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 993
Indexes
Cumulative Index of Contributing Authors, Volumes 74–78 � � � � � � � � � � � � � � � � � � � � � � � � � �1041
Cumulative Index of Chapter Titles, Volumes 74–78 � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �1045
Errata
An online log of corrections to Annual Review of Biochemistry articles may be found athttp://biochem.annualreviews.org/errata.shtml
viii Contents
Ann
u. R
ev. B
ioch
em. 2
009.
78:9
59-9
91. D
ownl
oade
d fr
om w
ww
.ann
ualr
evie
ws.
org
by U
nive
rsid
ad A
uton
oma
Met
ropo
litan
a on
10/
20/1
1. F
or p
erso
nal u
se o
nly.
