Disaggregating chaperones: an unfolding story.
ABSTRACT Stress, molecular crowding and mutations may jeopardize the native folding of proteins. Misfolded and aggregated proteins not only loose their biological activity, but may also disturb protein homeostasis, damage membranes and induce apoptosis. Here, we review the role of molecular chaperones as a network of cellular defenses against the formation of cytotoxic protein aggregates. Chaperones favour the native folding of proteins either as "holdases", sequestering hydrophobic regions in misfolding polypeptides, and/or as "unfoldases", forcibly unfolding and disentangling misfolded polypeptides from aggregates. Whereas in bacteria, plants and fungi Hsp70/40 acts in concert with the Hsp100 (ClpB) unfoldase, Hsp70/40 is the only known chaperone in the cytoplasm of mammalian cells that can forcibly unfold and neutralize cytotoxic protein conformers. Owing to its particular spatial configuration, the bulky 70 kDa Hsp70 molecule, when distally bound through a very tight molecular clamp onto a 50-fold smaller hydrophobic peptide loop extruding from an aggregate, can locally exert on the misfolded segment an unfolding force of entropic origin, thus destroying the misfolded structures that stabilize aggregates. ADP/ATP exchange triggers Hsp70 dissociation from the ensuing enlarged unfolded peptide loop, which is then allowed to spontaneously refold into a closer-to-native conformation devoid of affinity for the chaperone. Driven by ATP, the cooperative action of Hsp70 and its co-chaperone Hsp40 may thus gradually convert toxic misfolded protein substrates with high affinity for the chaperone, into non-toxic, natively refolded, low-affinity products. Stress- and mutation-induced protein damages in the cell, causing degenerative diseases and aging, may thus be effectively counteracted by a powerful network of molecular chaperones and of chaperone-related proteases.
Science 08/1973; 181(4096):223-30. · 31.20 Impact Factor
[show abstract] [hide abstract]
ABSTRACT: Amyloid aggregates are associated with a number of mammalian neurodegenerative diseases. Infectious aggregates of the mammalian prion protein PrP(sc) are hallmarks of transmissible spongiform encephalopathies in humans and cattle (Griffith, 1967; Legname et al., 2004; Prusiner, 1982; Silveira et al., 2004). Likewise, SDS-stable aggregates and low-n oligomers of the Abeta peptide (Selkoe et al., 1982; Walsh et al., 2002) cause toxic effects associated with Alzheimer's disease (Selkoe, 2004). The discovery of prions in lower eukaryotes, for example, yeast prions [PSI(+)], [PIN(+)], and [URE3] suggested that prion phenomena may represent a fundamental process that is widespread among living organisms (Chernoff, 2004; Uptain and Lindquist, 2002; Wickner, 1994; Wickner et al., 2004). These protein structures are more stable than other cellular protein complexes, which generally dissolve in SDS at room temperature. In contrast, the prion polymers withstand these conditions, while losing their association with their non-prion partners. These bulky protein particles cannot be analyzed in polyacrylamide gels, because their pores are too small to allow the passage and acceptable resolution of the large complexes. This problem was first circumvented by Kryndushkin et al. (2003), who used Western blots of protein complexes separated on agarose gels to analyze the sizes of SDS-resistant protein complexes associated with the yeast prion [PSI(+)]. Further studies have used this approach to characterize [PSI(+)] (Allen et al., 2005; Bagriantsev and Liebman, 2004; Salnikova et al., 2005), and another yeast prion [PIN(+)] (Bagriantsev and Liebman, 2004). In this chapter, we use this method to assay amyloid aggregates of recombinant proteins Sup35NM and Abeta42 and present protocols for Western blot analysis of high molecular weight (>5 MDa) amyloid aggregates resolved in agarose gels. The technique is suitable for the analysis of any large proteins or SDS-stable high molecular weight complexes.Methods in Enzymology 02/2006; 412:33-48. · 2.04 Impact Factor
[show abstract] [hide abstract]
ABSTRACT: Aggregation and subsequent development of protein deposition diseases originate from conformational changes in corresponding amyloidogenic proteins. The accumulated data support the model where protein fibrillogenesis proceeds via the formation of a relatively unfolded amyloidogenic conformation, which shares many structural properties with the pre-molten globule state, a partially folded intermediate first found during the equilibrium and kinetic (un)folding studies of several globular proteins and later described as one of the structural forms of natively unfolded proteins. The flexibility of this structural form is essential for the conformational rearrangements driving the formation of the core cross-beta structure of the amyloid fibril. Obviously, molecular mechanisms describing amyloidogenesis of ordered and natively unfolded proteins are different. For ordered protein to fibrillate, its unique and rigid structure has to be destabilized and partially unfolded. On the other hand, fibrillogenesis of a natively unfolded protein involves the formation of partially folded conformation; i.e., partial folding rather than unfolding. In this review recent findings are surveyed to illustrate some unique features of the natively unfolded proteins amyloidogenesis.Current Alzheimer Research 07/2008; 5(3):260-87. · 3.95 Impact Factor
Disaggregating Chaperones: An Unfolding Story
Current Protein and Peptide Science, 2009, 10, 432-446
1389-2037/09 $55.00+.00 © 2009 Bentham Science Publishers Ltd.
Sandeep K. Sharma1, Philipp Christen2 and Pierre Goloubinoff1,*
1Département de Biologie Moléculaire Végétale, Université de Lausanne, CH-1015 Lausanne, Switzerland;
2Biochemisches Institut, Universität Zürich, CH-8057 Zürich, Switzerland
Abstract: Stress, molecular crowding and mutations may jeopardize the native folding of proteins. Misfolded and aggre-
gated proteins not only loose their biological activity, but may also disturb protein homeostasis, damage membranes and
induce apoptosis. Here, we review the role of molecular chaperones as a network of cellular defences against the forma-
tion of cytotoxic protein aggregates. Chaperones favour the native folding of proteins either as “holdases”, sequestering
hydrophobic regions in misfolding polypeptides, and/or as “unfoldases”, forcibly unfolding and disentangling misfolded
polypeptides from aggregates. Whereas in bacteria, plants and fungi Hsp70/40 acts in concert with the Hsp100 (ClpB) un-
foldase, Hsp70/40 is the only known chaperone in the cytoplasm of mammalian cells that can forcibly unfold and neutral-
ize cytotoxic protein conformers. Owing to its particular spatial configuration, the bulky 70 kDa Hsp70 molecule, when
distally bound through a very tight molecular clamp onto a 50-fold smaller hydrophobic peptide loop extruding from an
aggregate, can locally exert on the misfolded segment an unfolding force of entropic origin, thus destroying the misfolded
structures that stabilize aggregates. ADP/ATP exchange triggers Hsp70 dissociation from the ensuing enlarged unfolded
peptide loop, which is then allowed to spontaneously refold into a closer-to-native conformation devoid of affinity for the
chaperone. Driven by ATP, the cooperative action of Hsp70 and its co-chaperone Hsp40 may thus gradually convert toxic
misfolded protein substrates with high affinity for the chaperone, into non-toxic, natively refolded, low-affinity products.
Stress- and mutation-induced protein damages in the cell, causing degenerative diseases and aging, may thus be effec-
tively counteracted by a powerful network of molecular chaperones and of chaperone-related proteases.
Keywords: Hsp70, DnaK, holdase, unfoldase, unfolding, disaggregation, Hsp100, ClpB, GroEL, small Hsps, Hsp90, protein
Stress-Unfolded Proteins can Misfold and Form Stable
In his determining experiments, Anfinsen demonstrated
that under optimal in vitro conditions, some artificially un-
folded proteins can spontaneously refold into their native
conformation without requiring any external assistance from
other macromolecules. This implied that the amino acid se-
quence of proteins should, in principle, suffice to determine
their native, thermodynamically most stable, biologically
active conformation . Yet, Anfinsen also noted that the
yield of recovered native proteins after in vitro refolding
decreased as the protein concentration or the temperature
increased. Presently, most protein biochemists are aware that
the higher the denaturing temperature, or the longer the ex-
posure of a native protein to a heat-stress, the higher will be
the oligomeric state, stability, compactness, SDS-resistance,
overall ?-sheet content and hydrophobic exposure of the re-
sulting protein aggregates [2-4].
During the intracellular de novo folding of nascent pro-
teins, or the import of polypeptides across organellar mem-
branes, short hydrophobic segments sequentially emerge
from the ribosome, or the membrane pore, preventing im-
proper trans associations with other hydrophobic segments
from the same polypeptide chain. In contrast, during heat
*Address correspondence to this author at the Département de Biologie
Moléculaire Végétale, Université de Lausanne, CH-1015 Lausanne, Swit-
zerland; Tel: +41 21 692 42 32; Fax: +41 21 692 41 95;
stress, whole native thermolabile polypeptides may partially
unfold and concomitantly expose multiple hydrophobic re-
gions on the same polypeptide chain, readily interacting with
one another and thus leading to the stabilization of misfolded
intramolecular structures. Similarly, wrong hydrophobic
interactions between nearby stress-misfolded polypeptides
will lead to the stabilization of non-functional inter-
molecular structures, dubbed “aggregates”. Thus, depending
on the temperature, the protein concentration and the dura-
tion of the stress, stable, active native protein oligomers,
made of a discrete number of native subunits, will spontane-
ously convert into nearly as stable, inactive aggregates made
of a continuum of misfolded subunits. Following stress ex-
posure, despite the intrinsic ability of individual misfolded
subunits to reach a thermodynamically more stable native
state, heat-induced aggregates often remain insoluble and
inactive even under conditions favouring oligomer dissocia-
tion, such as extensive dilutions and long incubations at low
temperatures . The surprising stability of such aggregates
is to be attributed to the formation of many small hydropho-
bic ?-sheet-enriched surfaces in the misfolding polypeptides,
all seeking less exposure to water by cooperatively associat-
ing with one another, thus maintaining the misfolded poly-
peptides tightly entangled. The highly cooperative nature of
intermolecular hydrophobic associations is thought to re-
strain the local random molecular motions that might other-
wise, with time, release individual misfolded polypeptides
from the aggregate and thus allow their native spontaneous
Disaggregating Chaperones: An Unfolding StoryCurrent Protein and Peptide Science, 2009, Vol. 10, No. 5 433
Cellular Crowding and Aggregate Formation
In addition to temperature, mutations and chemical modi-
fications, cellular crowding is believed to be a major cause
for protein aggregation . Indeed, the protein concentration
in the cytoplasm approximates 350 mg ml?1, a significant
molecular crowding thought to strongly favour unproductive
associations between de novo folding or stress-unfolding
polypeptides . Yet, proteins at high concentration often
better withstand heat-denaturation than diluted proteins. For
example, we find that mammalian mitochondrial malate de-
hydrogenase inactivates at 47 oC twice faster at 0.5 ?M than
at 5 ?M (not shown). Thus, the very high crowding and the
elevated viscosity of the cytoplasm could have a significant
protective effect on the stability of the thermolabile proteins
of the cell . However, once partially heat-unfolded, the
aggregation propensity of such thermolabile proteins would
expectedly increase with the cellular concentration of stress-
Predicting Misfolding and Aggregation Propensity in
The propensity of a protein to form stable aggregates in
vitro and in the cell correlates, in part, with the length of the
polypeptide chain . The misfolding propensity further
depends on the relative abundance of hydrophobic and
charged residues and on their distribution (clustered versus
scattered) along the sequence. Polypeptide segments en-
riched in charged residues and with few scattered hydropho-
bic residues are thus unlikely to form cohesive secondary
structures under in vitro conditions and are sometimes de-
scribed as “natively unfolded proteins” . Yet, in the
crowded cellular environment many “natively unfolded pro-
teins” may interact with other proteins and membranes and
acquire secondary structures. Hence, proteins such as ?-
synuclein, tau or LEA protein, that were initially described
as “natively unfolded”, acquire ?-helical structures in the
presence of other proteins, such as actin, calmodulin, or of
Protein Aggregates May Become Cytotoxic
A worrisome observation is that in aging tissues, perhaps
because of accumulating mutations and deficient chaperone
expression , some "natively unfolded proteins", such as
tau protein and ?-synuclein, fold into highly stable ?-sheet
structures that further aggregate into toxic fibrils and form
Lewy bodies or amyloid plaques as observed in many Alz-
heimer and Parkinsonian pathologies (for a review, see Ref.
). Moreover, regular heat-induced misfolding proteins
can in vitro seed, accelerate and even propagate the aggrega-
tion of other misfolding proteins . In Alzheimer disease,
cytotoxic aggregates of A? peptides can cause the co-
aggregation of ?-synuclein, normally more related to Parkin-
son disease [16, 17]. Similarly, the expression of polyglu-
tamine tract aggregates in Caenorhabditis elegans strongly
affects the misfolding fate of other thermolabile mutant pro-
teins , demonstrating a vast pleiotropic effect of aggre-
gating proteins on protein homeostasis in general, leading to
apoptosis and tissue loss in aging animals [14, 19-21]. A
remarkable inhibitory effect by substoichiometric amounts of
heavy metal ions on the in vitro refolding of proteins was
recently observed . Because a few toxic protein con-
formers can accelerate and propagate the aggregation of
other stress-labile proteins in the cell [23, 24], metal poison-
ing could lead to major pleiotropic disturbances of protein
homeostasis, especially in aging chaperone-defective cells.
Protein Misfolding and Aggregation Cause Important
The formation and accumulation of protein aggregates
may result both in loss of function and toxicity. Accounting
in part for cellular toxicity, small soluble aggregates of ?-
synuclein, A? peptides, PrpSc or polyglutamine tracts can
spontaneously assemble into membrane pores, causing ion
leakage [25-27]. Failure of the cell to prevent the formation
of misfolded proteins, and to rescue or eliminate them, re-
sults in the accumulation of intracellular aggregates in the
form of stable inclusion bodies in prokaryotes  and of
inclusion bodies and tangles in animal cells. Many mutation-
induced or age-related diseases correlate with the occurrence
of intra- and extracellular protein aggregates, such as amy-
loid deposits, fibers or tangles that specifically stain with
Congo red or thioflavin-T, due to their high content in ?-
sheet structures [20, 29]. Neurodegenerative diseases, includ-
ing Alzheimer, Parkinson, Huntington and Creutzfeld-Jacob
diseases are associated with specific protein misfolding
events leading to the formation of amyloid fibrils, tangles or
other forms of pathological protein aggregates within and
outside cells . Although amyloids and fibrils are his-
tological hallmarks of abnormal protein folding and assem-
bly, they are rather to be considered as the successful out-
come of an active detoxification mechanism of the cell,
whereby highly toxic smaller oligomers are actively trans-
ported and compacted by the aggresome [21, 31]. The most
common acquired muscle disorder in older persons is spo-
radic inclusion-body myositis (s-IBM). In s-IBM, abnormal
accumulation of amyloid ?, phosphorylated tau protein and
of other proteins accumulating in the brain of Alzheimer
patients, together with the aging intracellular milieu of the
muscle fiber, appear as key upstream pathogenic events [32,
33]. Retinal dystrophies (RD), which comprise a group of
clinically and genetically heterogeneous retinal disorders,
typically result in loss of vision due to the degeneration of
photoreceptors . Abnormal protein accumulation and
aggregation resulting from protein misfolding and/or from
proteasome inhibition have been implicated in the patho-
genesis of RD [35, 36].
Chemical Chaperones May Prevent Stress-Induced Un-
folding of Labile Proteins
First in the line of cellular defences against stress-
induced toxic protein aggregations are the chemical chaper-
ones, which are osmolytes that mostly accumulate in dehy-
drated or salt-stressed organisms . Salt- or osmotic-
stress- induced accumulation of small organic compounds,
such as trehalose, potassium glutamate, glycine betaine or
proline, which may reach up to 0.5 M in the cytoplasm, can
significantly stabilize thermolabile proteins against heat de-
naturation [38, 39]. For example, we find that the initial in
vitro rate of inactivation of porcine mitochondrial malate
dehydrogenase (0.5 ?M) at 47oC is at least 15 times faster
without, than in the presence of 15% glycerol (data not
434 Current Protein and Peptide Science, 2009, Vol. 10, No. 5 Sharma et al.
shown). Yet, the accumulation of osmolytes is primarily ob-
served alongside with growth arrest caused by salt or dehy-
dration stress, suggesting that the accumulation of protective
chemical chaperones in the cell might have a high fitness
cost (for a review, see Ref. ).
Molecular Chaperones and Proteases Neutralize and De-
toxify Protein Aggregates
Molecular chaperones are a second line of cellular de-
fence against stress-induced protein misfolding and aggrega-
tion in the cell. Chaperones belong to several conserved
families of proteins that generally control protein homeosta-
sis in the cell. Prokaryotes and eukaryotes have evolved a
battery of different molecular chaperones, many of which are
stress-inducible heat shock proteins (Hsps). All chaperone
families share the common ability to screen proteins for ab-
normal structural elements . They may act either as pas-
sive “holdases”, binding misfolding intermediates and pre-
venting their aggregation, and/or as active “unfoldases”,
forcibly unfolding stable protein aggregates into natively
refoldable, or protease-degradable species . The five
major classes of sequence-conserved chaperones are:
Hsp100 (ClpB), Hsp90 (HtpG), Hsp70 (DnaK), Hsp60
(GroEL), and the small Hsps (IbpA/B) (Escherichia coli
orthologues shown in Parentheses (Table 1). All classes are
ATPases, with the exception of the small Hsps.
In addition to their ability to prevent and repair stress-
damaged proteins, chaperones perform significant house-
keeping functions under physiological conditions that un-
likely generate protein misfolding. For example, Hsp27,
Hsp90 and Hsp70 control the oligomeric state and thus the
activity of many native proteins, such as clathrin cages [42-
43], SNARES  and complexes involved in the signal
transduction of steroids and of heat-shock [45, 46]. More-
over, Hsp70, Bip, and possibly ClpC serve as motor compo-
nents of the cellular machinery, translocating pre-proteins
from the cytoplasm into the mitochondria, the endoplasmic
reticulum and chloroplasts, respectively [47-49]. In animals,
the massive accumulation of molecular chaperones, in par-
ticular of Hsp70 and Hsp27, can inhibit the activity of vari-
ous native complexes, both in the inflammatory and apop-
totic pathways, likely by actively maintaining the Nf?B oli-
gomers in a disassembled “alter-native” state . RNAi-
mediated inhibition of Hsp27 correspondingly activates in-
flammation and the Nf?B pathway in keratinocytes .
Another example of a physiological chaperone function is
the case of adenovirus-mediated overexpression of Hsp70,
which arrests sepsis-induced apoptosis of pulmonary alveo-
lar cells type 1, inhibits the unchecked division of pulmonary
alveolar cells and promotes their differentiation into func-
tional type 2 cells during recovery from an acute respiratory
distress syndrome . Many cancer cells resisting chemo-
therapy often display significantly higher than usual levels of
molecular chaperones [52, 53]. Similarly, young tissues that
are generally highly potent at overexpressing molecular
chaperones show an increased resistance to chemotherapy,
stress-induced apoptosis and heat- or ischemia-induced dam-
age. Conversely, old tissues that are deficient in expressing
molecular chaperones show increased sensitivity to chemo-
therapy, inflammation, apoptosis and ischemia-induced dam-
The Molecular Mechanism of Unfoldase Chaperones
To mediate proper unfolding/refolding of many different
proteins in the cell, ATP-driven unfoldase chaperones must
have the ability: (1) to specifically bind their misfolded pro-
tein substrates with a higher affinity than their natively re-
folded products; (2) to recruit the energy of ATP hydrolysis
to favour unfolding of the bound misfolded substrate; (3) to
dissociate timely from the newly unfolded product, allowing
the latter to refold spontaneously into a low-affinity chaper-
Under stringent in vitro conditions devoid of significant
spontaneous protein refolding, “holdase” chaperones, such as
the small Hsps , Hsp90  or GroEL-mini chaperones
, may serve as an efficient reservoir for stably misfolded,
but least aggregated protein substrates for subsequent active
unfolding by ATP-driven unfoldase chaperones, such as
Hsp70/Hsp40, Hsp100 and GroEL/GroES [58, 59]. Some
AAA+ ATPase proteases, such as Lon , ClpX/P, ClpA/P
[61, 62], the 26S proteasome , HslU/V or FtsH (for a
review, see Ref. ) also use the energy of ATP-hydrolysis
to forcibly unfold and feed aggregated or tagged proteins
into the narrow entry of their proteolytic chamber.
The molecular chaperones Hsp60 (GroEL, Escherichia
coli orthologue) , Hsp70 (DnaK) and Hsp100 (ClpB) 
may effectively “promote” the native folding of hundreds of
different proteins in the same cell. It is neither conceivable
nor necessary that a chaperone will have to provide specific
folding instructions to each different substrate. Indeed, the
spontaneous native refolding of artificially unfolded proteins
demonstrates that, in principle, the amino acid sequence of
proteins contains all the necessary instructions to reach the
native state . Thus, to promote the correct refolding of a
multitude of different proteins, all that a given chaperone
must be able to accomplish is: (1) to specifically recognise
and bind its many different misfolded substrates without
interacting with the much more abundant native proteins of
the cell, and (2) to apply an unfolding force on its bound
misfolded substrates, giving them a renewed “Anfinsen”
opportunity to refold spontaneously into the native state.
Hsp70/Hsp40 as Holdase and Unfoldase Chaperones
A large body of data now shows that Hsp70 chaperones
use the free energy from ATP binding and hydrolysis to re-
duce actively the concentration of misfolded and aggregated
proteins, as well as to regulate the oligomeric state and activ-
ity of various house-keeping “alter-native” proteins in the
cell. In this section, we discuss the particular role of Hsp70
as an efficient “unfoldase” chaperone that can actively un-
wind and disaggregate and thus detoxify cells from poten-
tially harmful protein aggregates. To fulfil its unfoldase role
in protein homeostasis, Hsp70 must collaborate with a J-
domain co-chaperone, typically Hsp40 (DnaJ, CpbA or DjlA
in E. coli), that targets the chaperone onto its various protein
substrates and, together with a nucleotide exchange factor
(NEF, GrpE in E. coli), controls the coupling between ATP
consumption and the unfolding work of the Hsp70 chaper-
DnaK is the major representative of the Hsp70 family in
E. coli. In eukaryotes, members of the Hsp70 protein family
take part in numerous cellular processes beyond defences
Disaggregating Chaperones: An Unfolding StoryCurrent Protein and Peptide Science, 2009, Vol. 10, No. 5 435
Table 1. Major Families of Molecular Chaperones and Related Proteases
ATP Dependence Function
Hsp10 Mitochondria GroES
? Co-chaperone of Hsp60
Cooperates with Hsp70,
Holdase, co-chaperone of
GroEL + Holdase and unfoldase
Holdase and unfoldase,
Holdase, signaling, unfol-
Hsp100 Mitochondria ClpB +
Holdase (?), unfoldase,
Co-chaperone of Hsp70,
nucleotide exchange factor
- - - +
Co-chaperone of Hsp70,
nucleotide exchange factor
Proteasome lid (19S) Cytoplasm HslU +
Unfoldase, cofactor of 20S
proteasome (or of HslV)
ClpA/C Chloroplasts ClpA +
Unfoldase, cofactor of
ClpX Mitochondria ClpX +
Unfoldase, cofactor of
Lon + Unfoldase, protease
FtsH + Unfoldase, protease
? Unfoldase, protease
against stress damage, such as de novo protein folding, pro-
tein translocation, vesicular trafficking and cellular signal-
ling (for reviews, see Refs. [66, 67]). DnaK is eightfold more
abundant than any other chaperone in E. coli. During heat
shock, the expression level of DnaK and its co-chaperones is
strongly enhanced, presumably to cope with the rising
amount of misfolded proteins in the cell. DnaK knockout
mutants are hardly viable. Even at low growth temperature
they accumulate hundreds of insoluble misfolded proteins
. Overproduction of GroEL/GroES, ClpB, HtpG or
IbpA/B in ?dnaK cells may partially suppress protein aggre-
gation at a non-permissive temperature (42°C), but does not
prevent temperature sensitivity . The high cellular concen-
tration of DnaK, combined with its promiscuous and effi-
cient substrate-binding capacity, qualifies it as one of the
major “holding” and unfolding chaperones in E. coli.
DnaK consists of a 44-kDa NH2-terminal ATPase do-
main and a 25-kDa COOH-terminal substrate-binding do-
main . Hydrolysis of ATP and ADP/ATP exchange in
the nucleotide-binding domain, control the functional proper-
ties of the substrate-binding domain by allosteric crosstalk.
ATP-liganded DnaK exhibits low affinity for misfolded sub-
strates and fast rates of substrate binding and release [69,
70]. In contrast, ADP.DnaK is characterized by a 100-fold
higher affinity for substrate and consequently by a very slow
rate of substrate release [71, 72]. During the chaperone cy-
436 Current Protein and Peptide Science, 2009, Vol. 10, No. 5 Sharma et al.
cle, DnaK alternates between the low and the high affinity
state, under the control of the co-chaperones DnaJ and GrpE
Until recently, a rather passive chaperone mechanism
was assigned to Hsp70: Hsp70 would sequester exposed hy-
drophobic segments in the misfolding substrate thereby
counteracting wrong intra-molecular misfolding events and
intermolecular aggregations, thus slowly and only indirectly
promoting the proper native refolding of the substrate .
Binding of DnaK to protein substrates, such as denatured
firefly luciferase and rhodanese, or “alter-native” proteins
, such as the replication initiator protein RepA or the
RNA polymerase subunit ?32, accelerates ATP hydrolysis by
DnaK up to 2 orders of magnitude [74,75], a substrate effect
that is further amplified by DnaJ-triggering . DnaJ stimu-
lates ATP hydrolysis in the substrate-bound ATP·DnaK
molecules and thus promotes the formation of considerably
more stable [ADP·DnaK·substrate] complexes. These long-
lived chaperone-substrate complexes then act as entropic
pulling species, the dangling bound chaperone molecule ac-
tively disentangling the misfolded regions that flank the
chaperone-binding sites in the substrate . GrpE acceler-
ates the release of ADP, triggering the “unlocking” of DnaK
from its substrate and rebinding of ATP (Fig. 1). The final
result of this cycle is the productive transient unfolding of a
stably misfolded polypeptide region, leading to its gradual
disentanglement and spontaneous native refolding. In the
cytoplasm of animal cells, disaggregation may be achieved
by Hsp70/Hsp40/NEF alone. The unfoldase reaction is, how-
ever, much less efficient in animals than in plants, yeast and
bacteria, where an additional powerful ATPase unfoldase,
Hsp104 (ClpB), renders unfolding of large stable protein
aggregates significantly more efficient than by the sole
Hsp70/Hsp40/NEF system [77, 78].
All known chaperone activities of Hsp70 in the cell ap-
pear to rely on the transient interaction of Hsp70 with short
extended peptide segments of the substrate, made of about
seven non-bulky hydrophobic residues, ideally flanked by
positive charges . In all instances, a co-chaperone with a
conserved J-domain is also to be associated to the same pro-
tein substrate, to co-stimulate ATP-hydrolysis in a nearby
substrate-bound DnaK molecule [75, 80]. DnaJ thus triggers
the conversion of an Hsp70 molecule, which is loosely asso-
ciated to its misfolded substrate, into one that is very
strongly associated to its substrate and can thus apply an
unfolding force on misfolded segments flanking the chaper-
one-binding site. Finally, upon GrpE-accelerated ADP/ATP
exchange, the unlocked chaperone may dissociate from a
resulting newly enlarged unfolded peptide loop, which may
spontaneously refold into a more native structure devoid of
exposed hydrophobic residues and therefore also of chaper-
one binding sites (Fig. 1).
As yet, it is unclear whether the association of DnaK and
DnaJ with the substrate protein follows an obligatory or ran-
dom order. Fluorescence-labelled peptides proved to bind to
ATP·DnaK and DnaJ with similar rates and affinity .
These values diverge among themselves by orders of magni-
tude and overlap with the few known values for proteins
(72). At any rate, the formation of ternary [(ATP·DnaK)·
substrate·DnaJ] complexes is essential for feeding a substrate
Fig. (1). DnaK/DnaJ/GrpE unfoldase cycle. In Step 1, hydrophobic surfaces abnormally exposed on a misfolded and aggregated substrate
bind to the substrate-binding domain of DnaJ and ATP·DnaK binds to a nearby extended hydrophobic segment of the substrate protein. Step
2 completes the formation of a ternary [(ATP·DnaK)·substrate·DnaJ] complex, in which the substrate and the J-domain conjointly trigger
ATP hydrolysis and (Step 3) cause the high-affinity “locking” of DnaK onto its substrate. (Step 4) The [(ADP·DnaK)·substrate] complex is
relatively long-lived, allowing the Brownian motion of the bound DnaK molecule to apply a pulling force of entropic origin on the misfolded
segments flanking the DnaK-bound segment. GrpE then catalyzes the ADP/ATP exchange and the consequent “unlocking” of DnaK from a
newly enlarged unfolded loop in the substrate, which, once chaperone-free, may spontaneously refold into a more native structure (Step 5)
with less exposed hydrophobic segments, or may misfold again (Step 6) and re-enter the chaperone cycle.
Disaggregating Chaperones: An Unfolding StoryCurrent Protein and Peptide Science, 2009, Vol. 10, No. 5 437
into the chaperone cycle, as the ternary complex is the pre-
requisite for efficient acceleration of the hydrolysis of DnaK-
bound ATP (75, 80). Thus, to qualify as a substrate for a
DnaK/DnaJ/GrpE system, a protein must possess at least one
DnaJ-binding site and one DnaK-binding site differing from
one another, which, however, must be located close enough
to allow the J-domain of the substrate-bound DnaJ to trigger
ATP hydrolysis in a vicinal substrate-associated ATP·DnaK
molecule. The consequent “locking” of DnaK onto the sub-
strate, with a remarkable 100-fold increase in affinity, is the
committed step of the chaperone cycle, providing the struc-
tural basis for the subsequent step of substrate unfolding.
Hsp70S Unfold Proteins by Entropic Pulling
The energy of ATP hydrolysis serves to “lock” DnaK
onto a loop at the surface of stable protein aggregates. Multi-
ple ADP-liganded Hsp70 molecules tightly attached to loops
of the same aggregated substrate polypeptide cooperate in
applying stretching forces by entropic pulling . A single
Hsp70, or more effectively several Hsp70 molecules, locked
to loops in substrate polypeptide chains recruit random
Brownian motions to pull apart aggregated proteins and thus
to distend the loop segments caught up in aggregates. The
gain in entropy resulting from the increased motility of the
Hsp70·loop complexes diffusing away from the aggregate
may thus overcome the aggregate-stabilizing energy [83, 84].
In this context, it is pertinent to address the role of DnaJ
as sensor for non-native (or alter-natively folded) proteins in
the cell. Using peptide libraries immobilized on membranes,
DnaJ and DnaK were shown to share similar hydrophobic
binding motives when interacting with small unstructured
peptides . However, the crystal structure of [Hsp70· pep-
tide] complexes  showed a fundamental difference in the
way Hsp70 and Hsp40 may bind to the substrate protein: In
order to fit into the narrow substrate-binding groove of the
locked, ADP-liganded conformation of DnaK, the bound
substrate segment must be fully extended and devoid of
bulky structural elements. In contrast, the co-chaperone DnaJ
has been reported to bind native  and denatured proteins
as well as L- and D- peptides [85, 88]. The presumed surface
of the protein-binding domain of DnaJ does not dictate a
significant constraint on the bulkiness of the substrate and
may thus interact directly with large and bulky hydrophobic
regions on the surface of misfolded substrates. DnaJ is thus
best suited to scan for and bind to hydrophobic surfaces cov-
ering voluminous misfolded structures, whereas DnaK may
scan for and bind only to fully unfolded hydrophobic seg-
ments in the same substrate.
Thus, instead of competing for similar binding sites, the
chaperone and the co-chaperone may complement each
other: whereas DnaJ might preferably bind onto the mis-
folded structures that inter-space the loops protruding from
an aggregate, DnaK would preferably “lock” onto to the
loops that inter-space the misfolded structures in the same
misfolded polypeptide. The Brownian motions of an Hsp70
molecule locked onto a protruding loop, would exert a pull-
ing and unfolding force on the flanking misfolded regions of
the loop, thereby unthreading DnaJ-binding sites made of
inter-loop assemblies (Fig. (1), Step 3). This targeting
mechanism of DnaJ would explain how in chaperone refold-
ing assays EC50 values are reached with as little as one DnaJ
dimer for 20 molecules of DnaK . Once the distended
unfolded loop is finally released from Hsp70’s grip, it must
be given time to refold spontaneously into a more native-like
structure devoid of affinity for both Hsp70 chaperone and
Could direct conformational work be exerted by the
Hsp70 molecule onto the specific peptide segments to which
the chaperone is locked? DnaK has been shown to have a
modest catalytic activity as a peptide bond cis-trans
isomerase, accelerating the rate-limiting formation of rare
cis-peptide bonds in the native conformations of certain pro-
teins . Yet, unlike the Hsp70 unfoldase/disaggregase and
translocase activities, the isomerase activity is neither ATP
nor ADP-dependent, indicating that the isomerase activity is
not necessarily connected with the global, ATP-driven chap-
erone mechanism . Nevertheless, ATP hydrolysis, ATP
re-binding with a possible “catapult effect” , and/or the
isomerisation of the peptide bonds in the chaperone-bound
segments, could, in principle also unfold a misfolded seg-
ment. But Hsp70 can bind only to short extended segments
of 8 to 10 mostly non-bulky hydrophobic residues devoid of
structure, which occur only about once every 40 residues in
average proteins . Thus, even if in a collapsed aggregate
all the Hsp70-binding sites were fully available for binding,
Hsp70 binding would still be limited to about 20% of the
total sequence. How then, would the chaperone still be able
to drive the unfolding of the remaining the 80 % of the sub-
strate that contain all the misfolded ?-sheet enriched struc-
tures, with whom it may not have any direct physical interac-
The only plausible explanation is that, when locked on
small unstructured loops in the substrate, the Hsp70 chaper-
one uses the energy of ATP hydrolysis to recruit its own
Brownian motions to exert a pulling force on the remaining
80 % of the polypeptide, composed of structured inter-loop
regions containing bulky misfolded ?-sheets. By way of dis-
tal pulling, multiple concomitantly bound Hsp70 molecules
) may thus forcibly convert a stably misfolded polypep-
tide into a globally unfolded, natively refoldable polypeptide.
To demonstrate Hsp70/Hsp40 mediated unfolding activ-
ity (Fig. 2), we exploited the fact that heat-aggregated glu-
cose-6-phosphate dehydrogenase contains significantly more
wrong ?-sheets than its basal 31 % level of native ?-sheets
and that thioflavin-T can specifically bind ?-sheet structures
without affecting the chaperone activity. We monitored the
effect of DnaK/DnaJ/GrpE chaperones on preformed stable
G6PDH aggregates by online fluorescence spectroscopy.
During the first minutes of chaperone action, a sharp ATP-
dependent decrease of fluorescence was observed. Because a
loss of thioflavin-T binding is best interpreted as a simple
loss of ?-sheet structures, the rapid decrease in fluorescence
implies that the Hsp70 chaperone system readily unfolds ?-
sheet structures in the aggregates. Following this rapid un-
folding phase, a delayed and slower native refolding phase
was observed, which was later paralleled by a minor regain
of thioflavin-T binding, interpreted as a regain of some na-
tive ?-sheet structures. These observations strongly suggest
that the misfolded G6PDH substrate is first locally and
gradually unfolded by the chaperones, then disentangled, and
438 Current Protein and Peptide Science, 2009, Vol. 10, No. 5 Sharma et al.
then globally unfolded into a natively refoldable chaperone
Fig. (2). DnaK/DnaJ/GrpE actively unfolds stable protein ag-
gregates. G6PDH (1.5 ?M)was first denatured at 52°C to an activ-
ity of 2% of the initial; no spontaneous refolding was observed.
Preformed G6PDH aggregates were then incubated with DnaK (10
?M), DnaJ (0.8 ?M), GrpE (0.5 ?M) and ATP (5 mM) during 40
min at 30°C. Thioflavin T fluorescence (circles) and G6PDH activ-
ity (triangles) were continuously measured.
Hsp70 and Hsp40 are also Holdases
The first biochemical assay for a chaperone ever, ad-
dressed both the ability of GroEL/GroES to prevent the ag-
gregation, and to “assist” the native refolding of a denatured
RubisCO substrate, in a strict ATP-dependent manner .
Yet, chaperones have since been described in the literature
mostly as factors that merely prevent the aggregation of
other proteins. The binding of chaperone molecules to “cli-
ent” proteins, indeed lowers the concentration of aggrega-
tion-prone hydrophobic segments and thus reduces their ag-
gregation . The importance of such “holdase” activity
has been demonstrated in vitro, with various degrees of sig-
nificance, in the case of small Hsps (sHsps), Hsp40, Hsp70,
GroEL and Hsp90 [95-98]. Whereas Hsp70 and GroEL dis-
play both unfoldase and holdase activities, the sHsps are
exclusive holdases. One report suggests that Hsp100 mono-
mers can prevent the in vitro aggregation of ?-synuclein
. Hsp100 oligomers, however, are generally considered
to be ineffective holdases in vitro and in the cell .
Even without assistance from Hsp100 (ClpB), the
Hsp70/40 chaperone system can unquestionably unfold sta-
ble preformed protein aggregates . What is then the
possible contribution of a “holding” function, to the central
unfolding mechanism of Hsp70/40 chaperones? Fig. (3) il-
lustrates the relative and specific contribution of a preceding
“holding” step by DnaK and/or DnaJ during the heat-induced
aggregation of a model substrate, to the effectiveness of sub-
sequent ATP-driven unfolding/refolding action by the full
Hsp70/40/GrpE system. When the thermosensitive substrate
luciferase (350 nM) was heat-denatured for 7 min at 40°C in
the presence or absence of either, or both, thermoresistant
DnaK (3.5?M) and DnaJ (0.7 ?M) chaperones, it lost 99 %
of its activity. Only 2.5 % of the luciferase that had been
heat-denatured without holdases chaperones, did spontane-
ously refold thereafter during 1 h at 25°C (Fig. (3), Lane 1).
When luciferase was denatured in the presence of DnaK and
DnaJ (albeit without ATP), only 1.8 % more luciferase was
recovered after the stress, on top of the low 2.5 % baseline
(Fig. (3), Lane 2). Thus, although “holding” is often de-
scribed in the chaperone literature as being the only possible
mechanism by which all molecular chaperones may “assist”
the native folding of “client” proteins, we observed here no
significant net holdase activity for DnaK and DnaJ.
Fig. (3). DnaK and DnaJ are both holdase and unfoldase chap-
erones. Luciferase (350 nM) was incubated for 7 min at 40°C in the
absence or presence of either DnaK (K; 3.5 ?M), or DnaJ (J; 0.7
?M), or both (KJ), as indicated. After heat treatment, the denatured
luciferase was supplemented with K and/or J, and with 1.4 ?M
GrpE (E) and 5 mM ATP as indicated, and allowed to fold for 60
min at 25°C. The recovered luciferase activity is expressed in per-
cent of the activity before denaturation.
ones, then treated with the complemented whole chaperone
system (DnaK, DnaJ, GrpE and ATP) after the stress, a net
amount of 20 % native luciferase was recovered. This clearly
confirms that the Hsp70/Hsp40/NEF system can effectively
act on stably misfolded substrates preformed without previ-
ous “holding”, and actively convert a significant fraction of
them into natively refoldable chaperone products.
Remarkably, when luciferase was heat-denatured in the
presence of DnaK, DnaJ, and only after the stress supple-
mented with the GrpE and ATP, about 70% of the luciferase
was recovered (Fig. (3), Lane 6). This clearly demonstrates
that DnaK and DnaJ did serve as “holdases” during the dena-
turing stress; the holdase effect, however, became manifest
only during the subsequent ATP-fuelled unfoldase action.
Although holding in itself is not productive and not obliga-
tory for the subsequent ATP-driven unfolding activity of this
chaperone system, it increased the chaperone efficiency by
300% ! Interestingly, the presence of ATP with DnaK and
DnaJ during the denaturing stress did not improve the effi-
ciency of DnaK and DnaJ as a holdases. Apparently, DnaK
is already an optimally effective holdase when it interacts
with the substrate in its unlocked, nucleotide-free or ATP-
bound state. Noticeably, although DnaJ was 5 times less
abundant than DnaK, it proved to be as a powerful holdase
When luciferase was first heat-denatured without chaper-
Disaggregating Chaperones: An Unfolding StoryCurrent Protein and Peptide Science, 2009, Vol. 10, No. 5 439
as DnaK (Fig. (3), Lanes 4 and 5). This suggests that, besides
DnaK, DnaJ carries much of the pre-holdase activity in the
In the cell, not only Hsp40 (and Hsp70) may serve as
efficient holding chaperones to increase the efficiency of
subsequent unfoldase chaperones. The small heat shock pro-
teins IbpB from E. coli and Hsp17 from Synechocystis 
have been shown to serve as effective surrogate holdases
during heat-denaturation of malate dehydrogenase (MDH)
. Indeed, pre-binding of denaturing MDH to thermosta-
ble IbpB during heat-stress significantly increases the effi-
ciency of subsequent native MDH refolding by DnaK/DnaJ/
GrpE after the stress. IbpB is thought to have the ability to
reduce passively the misfolding damages in the denaturing
substrate [59, 102] and thus optimally present smaller mis-
folded MDH aggregates to the DnaK/DnaJ/GrpE unfoldase
system. The latter chaperone system is more efficient at dis-
aggregating small misfolded aggregates than at globally un-
folding misfolded monomers. It thus allows misfolded
monomers to bind to the GroEL/GroES system, which is
unable to accommodate aggregates in its cavity but is much
more efficient than DnaK/DnaJ/GrpE at globally unfolding a
misfolded 30-kDa MDH polypeptide into a natively refold-
able enzyme [59, 103].
Small Heat Shock Proteins: A Paradigm for Holdase
The small heat shock proteins (sHsps) are a family of
ubiquitous chaperones found in bacteria and most of the cel-
lular compartments of plants, fungi and mammals. The ex-
pression of many sHsps is very sensitive to variations in the
environment. For example, in the moss Physcomitrella pat-
ens, a mild non-noxious rise of temperature within physio-
logical range, can induce a more than 1000-fold accumula-
tion of sHsps, leading to acquired thermotolerance of the
plant [104, 105] (for a review, see Ref. ). As also re-
flected by their variable generic names, small Hsps, which
include mammalian Hsp27 and ?-crystallins, organellar
Hsp23, Hsp17 and Hsp16, as well as bacterial IbpA/B (E.
coli) and Hsp17 (Synechocystis), vary in size from 15 to
30 kDa and in oligomeric state. The part most variable in
sequence and length is the NH2-terminal domain. They share
a conserved 80 to 100 amino acid long domain, known as the
?-crystallin domain, which is often located at the COOH-
terminus and is thought to convey stability and solubility to
the quaternary structure of the chaperone . The sHsps
share common biochemical properties, such as the ability to
bind damaged misfolded polypeptides and aggregates with
high capacity. Some sHsps can also bind and stabilize mem-
branes under heat shock conditions . The affinity of
small Hsps for misfolded proteins and membranes may de-
pend on their oligomeric state, which in turn may be modu-
lated by temperature and phosphorylation.
Mammalian Hsp27 is abundantly expressed in many cell
types and tissues at specific stages of differentiation and de-
velopment. The Hsp27 dimer appears to be the building
block for dynamic multimeric complexes, depending on
phosphorylation and exposure to various forms of stresses,
including heat, hydrogen peroxide and other oxidants,
ischemia, mitogens and inflammatory cytokines, such as
TNF? and IL-1? . In tissues of several protein misfold-
ing diseases, Hsp27 and ?B-crystallin are up-regulated. Inef-
fective induction of sHsp expression in old unhealthy tissues
in response to stress suggests that impaired sHsp action un-
derlies diseases such as cataract and cancer . There is
strong evidence for a protective role of sHsps against protein
conformational diseases. Several hereditary neurodegenera-
tive diseases are associated with mutations in sHsp, namely
Hsp27, ?B-crystallin, ?A-crystallin and HspB8/Hsp22 [110-
113] (for a review, see Ref. ).
Some sHsps have been suggested to consolidate the in-
tracellular redox homeostasis  and to stabilise stressed
membranes . Yet, the most compelling body of in vitro
and in vivo evidence point at their central role in protein ho-
meostasis, in particular at preventing stress- or mutation-
induced protein misfolding and aggregation. Because sHsps
do not hydrolyse ATP, they may not act as unfoldases but
rather as typical holdases, engaging in strong hydrophobic
interactions with exposed hydrophobic surfaces in misfold-
ing proteins and thus reducing protein aggregation. Small
Hsps often compensate their relative ineffectiveness as pas-
sive holdases by being under the control of exquisitely sensi-
tive, stress-responsive promoters. They may accumulate at so
elevated concentrations that they prevent the aggregation of
most misfolding proteins that might form during the life-
span of a cell. Thus, in the human eye lens lacking an active
protein clearance mechanism, the same non-renewable ?-
crystallin molecules may successfully prevent light scatter-
ing of stress-damaged proteins during 70 to 80 years of con-
tinuous environmental injuries by excess light, UV, patho-
gens and pollutants [55, 58, 116, 117]. In contrast to the inert
eye lens, metabolically active cells may efficiently couple
the strong holdase activity of the sHsps to the forcible ATP-
driven unfoldase activity of the Hsp70/40, GroEL/S and
Hsp100 chaperones [59, 102].
The Hsp100 Powerstroke Unfoldases
The Hsp100 chaperones are members of the AAA+
(ATPase associated with various cellular activities) superfa-
mily. They include bacterial ClpB, mitochondrial Hsp78 and
chloroplast ClpC, as well as eukaryotic orthologues in the
cytoplasm of fungi, yeast (Hsp104) and plants (Hsp101).
Remarkably, they are absent from the cytoplasm of animal
cells. Hsp100 chaperones share close sequence, structural
and functional relationships with gated proteases, in particu-
lar with the AAA+ hexameric complexes serving as co-
protease lids to organellar or bacterial proteases, such as
HslU/V and ClpA/P and Lon or the eukaryotic proteasome
(Table 1). Like other AAA+, and by analogy to sequence-
wise and structurally similar co-protease ClpA, Hsp100
(ClpB) chaperones are organised in hexameric rings with a
narrow central cavity that can allow the translocation only of
a single unfolded polypeptide, and in certain cases of an un-
folded polypeptide loop . The cavity spans about 90 Å
, from the NH2-terminal substrate-binding ring of the
chaperone, through two consecutive ATPase rings, to the
exit on the COOH-terminal side of the cylinder. Interest-
ingly, at the COOH-terminal exit of the ClpA co-protease,
four specific anchoring residues can specifically bind a hep-
tameric cylinder of the ClpP protease, leading to the degra-
dation of substrates that were previously forcibly unfolded
440 Current Protein and Peptide Science, 2009, Vol. 10, No. 5 Sharma et al.
and translocated. When artificially deprived of ClpP, ClpA
can act as an unfolding chaperone, using the energy of ATP
hydrolysis to unfold monomeric ssrA-tagged green fluores-
cent protein . Symmetrically, when artificially engi-
neered to bind ClpP, ClpB can act as an unfolding protease,
using the energy of ATP-hydrolysis to unfold and degrade
misfolded substrates . This strongly indicates that a
unidirectional translocation of unfolded substrates across the
ClpB cylinder is central to the ClpB chaperone mechanism.
Yet, in the cell, ClpB does not naturally associate with a pro-
tease, but rather works in close collaboration with the bacte-
rial Hsp70/40 chaperone system (DnaK/DnaJ) at unfolding
stable protein aggregates [77, 78]. Although DnaK/DnaJ/
GrpE alone can already unfold some types of stable protein
aggregates, converting them into natively refoldable products
(Fig. (2)), the presence of ClpB can significantly increase the
efficiency of the unfolding reaction . Dose responses
suggest that ClpB activity increases the apparent affinity of
DnaK for the aggregated substrates . The specific inhi-
bition of ClpB by addition of low concentrations of ammo-
nium sulphate at different times of the coupled ClpB +
DnaK/DnaJ/GrpE chaperone reaction, indicated that ClpB
activity is essential only during the early phases of the reac-
tion, when active disaggregation occurs but not yet native
refolding . This suggests a sequential mechanism,
whereby ClpB, together with DnaK/DnaJ/GrpE, primarily
performs the initial disaggregation of large stable aggregates
into smaller ones, and argues against an obligate unidirec-
tional translocation of all unfolded substrates across the
ClpB cylinder. Indeed, DnaK/DnaJ/GrpE alone suffices to
complete the unfolding and disaggregation of small aggre-
gates into natively refoldable polypeptides .
A first mechanism was suggested to explain how ClpB
may increase the apparent affinity of DnaK for the stably
aggregated substrates without completely translocating them;
during the initial disaggregation phase of the reaction prior to
native refolding, the NH2-terminal side of the ClpB hexamer
might apply to the surface of compact aggregates some itera-
tive “scraping” strokes and, upon partial “swallowing” into
ClpB’s central cavity of detached polypeptide loops and their
regurgitation, gradually detach and unfold compact mis-
folded segments otherwise inaccessible to DnaK/DnaJ bind-
ing [78, 121]. Yet, other experimental evidence also indi-
cated that both ClpB and DnaK/DnaJ/GrpE act upon stable
protein aggregates by continuously extracting individual
unfolded polypeptides, and not by first fragmenting large
aggregates . This supports a mechanism whereby indi-
vidual misfolded polypeptides are first unfolded and indi-
vidually disentangled from compact aggregates by DnaK/
DnaJ/GrpE, then completely threaded through ClpB and then
natively refolded with or without renewed assistance from
DnaK/DnaJ/GrpE [122, 123]. It is, however, unclear why
ClpB alone in the presence of ATP remains a very inefficient
unfoldase chaperone  and moreover generates larger
aggregates [78, 125]. This raises the possibility that the
DnaK/DnaJ/GrpE chaperones have first to disaggregate large
compact aggregates and unfold them into individual mis-
folded polypeptides, which only then can enter into the ClpB
cylinder and unfold into natively refoldable products .
It is possible that, for the same reasons that unfolded poly-
peptides exiting the ribosome or import pores need assis-
tance by Hsp70/40 to natively refold, unfolded polypeptides
exiting from the ClpB cavity may also require assistance
Recently, however, analysis of the processing by ClpB +
DnaK/DnaJ/GrpE of aggregates consisting of fusion proteins
of misfolded flanked by tightly folded native domains, dem-
onstrated that a partial threading of the substrate through
ClpB suffices to mediate the proper native refolding of the
misfolded domains. This implied that complete unfolding by
unidirectional translocation across the chaperone’s cavity is
not absolutely obligatory to the chaperone mechanism .
As a rather lame explanation, it has been suggested that dur-
ing stalled threading, ClpB and Hsp104 rings, which are la-
bile hexamers in vitro, could “facilitate”, the dissociation of
the trapped unfolded substrates by naturally falling apart. If
this were the case, however, a mechanism of obligate oli-
gomer dissociation would have first to be demonstrated in
vivo for ClpB and Hsp104, as well as for paralogous AAA+
unfoldases, such as ClpA, ClpX, Lon and the 19S lid of the
proteasome. Alternatively, the results by Haslberger et al.,
(2008) revive the initial idea that ClpB might have to merely
“scrape” the surfaces of some compact aggregates and, upon
partially “swallowing” and “regurgitating” detached mis-
folded loops from the aggregates, significantly increases the
substrate’s affinity for subsequent unfolding by DnaK/
DnaJ/GrpE. This does not exclude the possibility that some
aggregates may have to be first unfolded by DnaK/
DnaJ/GrpE, then completely unfolded by threading through
ClpB, then possibly unfolded again by DnaK/DnaJ/GrpE, to
become at last, natively refoldable (for a review, see Ref.
GroEL: A Probable Unfoldase
The chaperonin GroEL and its co-chaperonin GroES are
abundant heat shock proteins in eubacteria, mitochondria and
chloroplasts, where they are required for viability under
physiological and stress conditions. Overproduction in bacte-
ria of both GroEL and GroES suppresses temperature-
sensitive mutations in a large number of genes, apparently by
promoting the correct folding of mutant polypeptides at non-
permissive temperatures . This observation, which was
later confirmed with Hsp90 chaperones , demonstrates
that molecular chaperones can serve as evolutionary buffers:
Without stress, they maintain newly mutated, aggregation-
prone proteins in a mostly functional state. Under stress,
however, chaperones may become recruited by an excess of
stress-induced misfolded proteins, thereby revealing new
phenotypes, possibly with novel solutions to ever-evolving
The GroEL monomer is composed of an ATP-binding
base domain, a flexible hinge domain, topped by an apical
substrate- and GroES-binding domain. The GroEL oligomer
is a tetradecamer composed of two base-to-base assembled
heptameric toroids with two unconnected central cavities that
can accommodate each a single folding substrate polypeptide
of up to 55 kDa . Larger polypeptides, protein oli-
gomers or aggregates are very poor GroEL/GroES substrates
because they can neither fit into the cavities of GroEL14, nor
be fully covered by the GroES7 caps during the chaperone
cycle . GroEL/GroES can only act on individual mis-
folded polypeptides that must first bind to a circular hydro-
Disaggregating Chaperones: An Unfolding StoryCurrent Protein and Peptide Science, 2009, Vol. 10, No. 5 441
phobic binding surface bordering the inner entry of the cen-
tral cavity. ATP binding induces the seven apical domains of
each GroEL7 ring, at least three of them binding the mis-
folded substrate, to rise outward and then to turn aside (for a
review, see Ref. ). These conformational changes first
apply an unfolding stroke to the bound misfolded substrate
and, second, release the partially unfolded substrate into the
central cavity of the chaperone. As ATP binding concomi-
tantly causes the binding of a GroES7 co-chaperone lid, on
top of the seven reoriented apical domains of each GroEL7
toroid, the two entries of the two cavities of GroEL14 become
transiently covered [130-133], compelling the newly released
unfolded substrates to refold natively inside the protected
environment of the central cavities. Then, in a carefully co-
ordinated allosteric mechanism between the two opposite
rings, one GroES7 timely dissociates on one end to allow the
release of a natively refolded product from one toroid and
the subsequent rebinding of another misfolded substrate.
The holdase activity of GroEL14 was initially shown with
urea-denatured RubisCO : upon removal of denaturing
urea, the sole presence of GroEL14 (without ATP) generated
inactive, albeit native-gel-soluble RubisCO species. In con-
trast, the absence of GroEL14 allowed the formation of inac-
tive insoluble RubisCO aggregates that did not enter the na-
tive gels . Unlike the favourable but non-obligate hol-
dase activity of DnaK and DnaJ, the holdase activity of
GroEL14 was essential to the success of subsequent unfol-
dase activity. Indeed, only misfolded protein that was suc-
cessfully pre-bound to GroEL14 and thus prevented from
forming GroEL-resistant aggregates, was fully natively re-
foldable, in a strict ATP- and GroES7-dependent manner
It has initially been suggested that GroEL/GroES drives
native refolding simply by allowing the native refolding of
unfolded substrate monomers to occur in the protected
“cages” of the GroEL/GroES complex, thus by merely
avoiding wrong associations with neighbouring misfolded
species . However, the discovery that ClpB, together
with Hsp70/Hsp40/NEF, can clearly unfold already formed,
stable protein aggregates , paved the way to a new con-
cept in chaperone science: some chaperones may not only
prevent protein aggregation but also use the energy of ATP
hydrolysis to forcibly unfold stably misfolded proteins, lead-
ing to disaggregation and subsequent native refolding. An
unfoldase activity of the GroEL+GroES+ATP system was
initially substantiated by proton exchange experiments .
Yet, the unfoldase activity of GroEL/GroES remained there-
after mostly ignored for more than a decade [136, 137], until
it was independently confirmed by several groups [138,
Because GroEL/GroES can only unfold misfolded
monomers that are smaller than 55 kDa, but not bulkier pro-
tein aggregates, the successful pre-binding to GroEL of a
substrate during denaturation is essential for its subsequent
native refolding . Remarkably, as compared to larger
aggregates, misfolded monomers are the least efficient sub-
strates of the DnaK/DnaJ/GrpE system: only a molar excess
of DnaK can globally unfold them by entropic pulling into
natively refoldable species . It is therefore not surprising
that the two chaperone systems best collaborate when large
stable aggregates are first efficiently converted by
DnaK/DnaJ/GrpE into misfolded monomers, which in turn
are efficiently converted by GroEL/GroES into natively re-
foldable species . This highly cooperative bi-chaperone
system can be further improved by a third partner, such as
ClpB, and a fourth partner, such as the exclusive holdase
chaperone IbpB, whose presence during heat-denaturation,
reduces the degree of misfolding and aggregation of a ther-
molabile substrate, thereby increasing the efficiency of sub-
sequent unfolding and disaggregation by DnaK/DnaJ/GrpE
and global unfolding of misfolded monomers by GroEL/
GroES [59, 102].
Hsp90: An Intensively Studied, Yet Most Mysterious
Hsp90 (Grp94 in the endoplasmic reticulum, HptG in E.
coli) is one of the most conserved chaperone families (Table
1). Although members of this family, including bacterial
HtpG, are strongly induced by heat shock or oxidative stress,
the role of Hsp90 as a chaperone expected to prevent and/or
repair structural damages in heat-labile proteins, remains to
be demonstrated experimentally. Hsp90 is an extremely slow
ATPase with an atypical binding site for a bended ATP
molecule, similar to Type II DNA topoisomerases, with
which it shares many inhibitors, such a novobiocin and radi-
cicol [140, 141]. Despite their broader-than-generally-belie-
ved spectrum of action, geldanamycin and radicicol are con-
sidered to be highly specific inhibitors of Hsp90, leading to a
probable overestimation of the various biological functions
specifically assigned to Hsp90. Geldanamycin and radicicol
strongly bind to the ATP-binding domain of the Hsp90
chaperone, causing the dissociation of chaperone-bound pro-
teins, which are called “clients”, for lack of evidence that
Hsp90 acts as an enzyme capable of converting high-affinity
substrates into low-affinity products. In vitro, Hsp90 has
been shown to reduce the light scattering signal of artificially
aggregating proteins [97, 142], and Hsp90/topoisomerase
inhibitors often cause some protein aggregation in cells .
The unique experimental tool of presumably specific inhibi-
tors, together with the high affinity of eukaryotic Hsp90 for a
multitude of native “clients”, as well as for a dozen of co-
chaperones , has generated massive information on the
supposed physiological functions of Hsp90s. In eukaryotes,
several hundred of Hsp90 clients have been identified, the
large majority of which are natively folded transcription fac-
tors and signalling kinases . Their blockage by inhibi-
tors produced strong pleiotropic effects on cell signalling,
proliferation and survival. Whereas other ATPase chaper-
ones, such as Hsp70, ClpB and GroEL may use the energy of
ATP hydrolysis to convert forcibly stable misfolded protein
substrates into unfolded, natively refoldable, or protease-
degradable chaperone products, there is no information on
the precise nature of the ATP-driven mechanism by which
Hsp90s presumably control structural changes in such a
plethora of client proteins.
Recently, important progress was made regarding the
structure of eukaryotic Hsp90 and bacterial HtpG .
Hsp90 contains three major domains: the NH2-terminal do-
main containing the ATP-binding site that also binds gel-
danamycin and radicicol, the middle domain that has high
affinity for most co-chaperones and client proteins, and the
442 Current Protein and Peptide Science, 2009, Vol. 10, No. 5 Sharma et al.
COOH-terminal base domain, mediating interactions be-
tween Hsp90 subunits and with scaffolding co-chaperones
such as Hop . The active chaperone is probably a dimer
in which the two Hsp90 subunits are connected by their
COOH-terminal domains, with the middle and distal do-
mains facing but standing apart away from each other, up to
40 Å in the case of the distal NH2-terminal domains. Upon
ATP binding and hydrolysis, the two middle domains tightly
bind each other, resulting in a significant compaction of the
Hsp90 dimer . Clients most likely bind to the middle
domains, but unlikely between the two subunits for lack of
space, not only in the compacted, but also in the relaxed V-
shaped state of the dimer. Thus, current knowledge does not
support an intuitive model whereby a bound Hsp90 client
could be forcibly unfolded consequent to a transient moving
apart of the two middle domains in the dimer driven by ATP
binding, ATP hydrolysis or ADP release . To increase
complexity, eukaryotic Hsp90 can sequentially form stable
complexes with a plethora of clients and co-chaperones that
activate or inhibit its ATPase activity. Other scaffolding pro-
teins, such as Hop, mediate the association of Hsp90 with
other chaperones, such as Hsp70, Hsp40, FKBP peptidylpro-
lyl isomerases and E3 ubiquitin ligases. To extract a general
common mechanism for all Hsp90s, particular cases must
therefore be clearly singled out. This might be achieved with
the study of the prokaryotic HtpG, which appears to act pre-
dominantly as a stress-induced chaperone involved in basic
protein homeostasis, independently from myriads of co-
chaperones and scaffolding proteins.
Protection Against and Rescue from Misfolding Versus
Proteolysis or Compaction
The role of holdase and unfoldase chaperones in the gen-
eral scheme of protein homeostasis in eukaryotic cells is
summarized in Fig. (4). When an unfolded protein exits the
ribosome or an import pore, it may spontaneously fold into a
native protein or misfold into a potentially toxic species.
Stress, such as heat-shock, may transiently unfold native
proteins, which after the stress, will either natively refold, be
degraded by gated proteases, or, unless prevented by hol-
dases, misfold and aggregate. Unfoldases will convert such
already formed aggregates into individual unfolded, natively
refoldable or degradable non-toxic species. Unfoldase-
resistant aggregates will be compacted by the aggresome into
less toxic amyloids [31, 146]. Noticeably, at this terminally
committed stage of the clearance mechanism, the
reactivation of unfoldases might transiently increase the con-
centration of toxic species. Misfolded protein conformers
that failed unfolding by chaperones or compaction by the
aggresome, may still be degraded into non-toxic products by
lysosomal autophagy  or may be secreted. In animal
tissues, failure of this last line of cellular defence against
proteotoxic species will generally induce a strong inflamma-
tory response, apoptosis and tissue loss .
As illustrated in Fig. (4), the broad range of molecular
tools that is used by cells to counter disturbances in protein
homeostasis is astounding, both in qualitative (Table 1) and
quantitative terms. The stability of proteins in thermophilic
archaea convincingly demonstrates that design of remarkably
thermostable proteins is possible and that the particular sen-
sitivity of proteins to variations in the environment is not an
Fig. (4). The role of holdase and unfoldase chaperones in eu-
karyotic protein homeostasis. The formation of toxic protein con-
formers is prevented by holdase chaperones. Failing holding, toxic
aggregated conformers are actively converted by unfoldase chaper-
ones into degradable or natively refoldable non-toxic species. Fail-
ing unfolding, toxic aggregates may be compacted and deposited by
aggresomes into less toxic inclusions. Failing that, aggregates may
be degraded by lysosomal autophagy into non-toxic degradation
products or may be secreted.
unavoidable property of all proteins. Rather, some proteins
are unavoidably thermosensitive likely because of the par-
ticular mechanistic dictates of their catalytic function or al-
losteric regulation: a compromise on stability for the sake of
conformational freedom. Thus, in addition to striving for
enhanced efficiency and new protein functions, evolution
must have taken into account the need for optimal (least ag-
gregation-prone) folding pathways. Sometimes, it had to
sacrifice stability for more functional flexibility or increased
complexity, at the cost of increased propensity to misfold
and aggregate. At this stage, the co-evolution of molecular
chaperones may have given proteins additional freedom to
increase their functional flexibility and complexity, by fenc-
ing more delicate folding pathways of more sophisticated
proteins with efficient holdases chaperones and establishing
effective rescuing unfoldase chaperones for astray off-
pathway species .
 Anfinsen, C.B. Principles that govern the folding of protein chains.
Science, 1973, 181, 223-230.
Bagriantsev, S.; Kushnirov, V.V.; Liebman, S.W. Analysis of amy-
loid aggregates using agarose gel electrophoresis. Methods Enzy-
mol., 2006, 412, 33-48.
Uversky, V.N. Amyloidogenesis of natively unfolded proteins.
Curr. Alzheimer Res., 2008, 5, 260-287.
Khare, S.D.; Dokholyan, N.V. Molecular mechanisms of polypep-
tide aggregation in human diseases. Curr. Protein Pept. Sci., 2007,
Diamant, S.; Goloubinoff, P. Temperature-controlled activity of
DnaK-DnaJ-GrpE chaperones: protein-folding arrest and recovery
Disaggregating Chaperones: An Unfolding StoryCurrent Protein and Peptide Science, 2009, Vol. 10, No. 5 443
during and after heat-shock depends on substrate protein and GrpE
concentration. Biochemistry, 1998, 37, 9688-9694.
Ellis, R.J. Macromolecular crowding: obvious but underappre-
ciated. Trends Biochem. Sci., 2001, 26, 597-604.
Zimmerman, S.B.; Trach, S.O. Estimation of macromolecule con-
centrations and excluded volume effects for the cytoplasm of
Escherichia coli. J. Mol. Biol., 1991, 222, 599-620.
Mogk, A.; Tomoyasu, T.; Goloubinoff, P.; Rüdiger, S.; Röder, D.;
Langen, H.; Bukau, B. Identification of thermolabile E. coli pro-
teins: prevention and reversion of aggregation by DnaK and ClpB.
EMBO J., 1999, 18, 6934-6949.
Uversky, V.N. ?-Synuclein misfolding and neurodegenerative
diseases. Curr. Protein Pept. Sci., 2008, 9, 507-540.
Tolleter, D.; Jaquinod, M.; Mangavel, C.; Passirani, C.; Saulnier,
P.; Manon, S.; Teyssier, E.; Payet, N.; Avelange-Macherel, M.H.;
Macherel, D. Structure and function of a mitochondrial late em-
bryogenesis abundant protein are revealed by desiccation. Plant
Cell, 2007, 19, 1580-1589.
Esposito, A.; Dohm, C.P.; Kermer, P.; Bähr, M.; Wouters, F.S.
alpha-Synuclein and its disease-related mutants interact differen-
tially with the microtubule protein tau and associate with the actin
cytoskeleton. Neurobiol. Dis., 2007, 26, 521-531.
Martinez, J.; Moeller, I.; Erdjument-Bromage, H.; Tempst, P.;
Lauring, B. Parkinson's disease-associated alpha-synuclein is a
calmodulin substrate. J. Biol. Chem., 2003, 278, 17379-17387.
Morimoto, R.I. Proteotoxic stress and inducible chaperone net-
works in neurodegenerative disease and aging. Genes Dev., 2008,
Hinault, M.P.; Ben-Zvi, A.; Goloubinoff, P. Chaperones and prote-
ases: cellular fold-controlling factors of proteins in neurodegenera-
tive diseases and aging. J. Mol. Neurosci., 2006, 30, 249-265.
Ben-Zvi, A.P.; Goloubinoff, P. Proteinaceous infectious behavior
in proteins is controlled by molecular chaperones. J. Biol. Chem.,
2002, 277, 49422-49427.
Kawahara, K.; Hashimoto, M.; Bar-On, P.; Ho, G.J.; Crews, L.;
Mizuno, H.; Rockenstein, E.; Imam, S.Z.; Masliah E. alpha-
Synuclein aggregates interfere with parkin solubility and distribu-
tion: role in the pathogenesis of Parkinson disease. J. Biol. Chem.,
2008, 283, 6979-6987.
Masliah, E. Neuropathology: Alzheimer's in real time. Nature,
2008, 451, 638-639.
Gidalevitz, T.; Ben-Zvi, A.; Ho, K.H.; Brignull, H.R.; Morimoto,
R.I. Progressive disruption of cellular protein folding in models of
polyglutamine diseases. Science, 2006, 311, 1471-1474.
Solary, E.; Droin, N.; Bettaieb, A.; Corcos, L.; Dimanche-Boitrel,
M.T.; Garrido, C. Positive and negative regulation of apoptotic
pathways by cytotoxic agents in hematological malignancies.
Leukemia, 2000, 14, 1833-1849.
Cohen, E.; Bieschke, J.; Perciavalle, R.M.; Kelly, J.W.; Dillin, A.
Opposing activities protect against age-onset proteotoxicity.
Science, 2000, 313, 1604-1610.
Lashuel, H.A.; Lansbury, P.T.; Jr. Are amyloid diseases caused by
protein aggregates that mimic bacterial pore-forming toxins? Q.
Rev. Biophys., 2006, 39, 167-201.
Sharma, S.K.; Goloubinoff, P.; Christen, P. Heavy metal ions are
potent inhibitors of protein folding. Biochem. Biophys. Res. Com-
mun., 2008, 372, 341-345.
Huang, X.; Atwood, C.S.; Moir, R.D.; Hartshorn, M.A.; Tanzi,
R.E.; Bush, A.I. Trace metal contamination initiates the apparent
auto-aggregation, amyloidosis, and oligomerization of Alzheimer's
Abeta peptides. J. Biol. Inorg. Chem., 2004, 9, 954-960.
Zawia, N.H.; Basha, M.R. Environmental risk factors and the de-
velopmental-basis for Alzheimer's disease. Rev. Neurosci., 2005,
Hirakura, Y.; Azimov, R.; Azimova, R.; Kagan, B.L. Polyglu-
tamine-induced ion channels: a possible mechanism for the neuro-
toxicity of Huntington and other CAG repeat diseases. J. Neurosci.
Res., 2000, 60, 490-494.
Lashuel, H.A. Membrane permeabilization: a common mechanism
in protein misfolding diseases: If it look like a pore and acts like a
pore, is it a pathogenic pore? Sci. Aging Knowledge Environ., 2005,
Tsigelny, I.F.; Sharikov, Y.; Miller, M.A.; Masliah, E. Mechanism
of alpha-synuclein oligomerization and membrane interaction:
theoretical approach to unstructured proteins studies. Nanomedi-
cine, 2008, 4, 350-357.
 Schrödel, A.; de Marco, A. Characterization of the aggregates
formed during recombinant protein expression in bacteria. BMC
Biochem., 2005, 6, 10.
LeVine, H, III, Quantification of beta-sheet amyloid fibril struc-
tures with thioflavin T. Methods Enzymol., 1999, 309, 274-284.
Selkoe, D.J. Cell biology of protein misfolding: the examples of
Alzheimer's and Parkinson's diseases. Nat. Cell Biol., 2004, 6,
Kopito, R.R. Aggresomes, inclusion bodies and protein aggrega-
tion. Trends Cell Biol., 2000, 10, 524-530.
Bancher, C.; Brunner, C.; Lassmann, H.; Budka, H.; Jellinger, K.;
Wiche, G.; Seitelberger, F.; Grundke-Iqbal, I.; Iqbal, K.;
Wisniewski, H.M. Accumulation of abnormaly phosphorylated tau
precedes the formation of neurofibrillary tangles in Alzheimer’s
disease. Brain Res., 1989, 477, 90-99.
Togo, T.; Akiyama, H.; Iseki, E.; Uchikado, H.; Kondo, H.; Ikeda,
K.; Tsuchiya, K.; de Silva, R.; Lees, A.; Kosaka, K. Immunohisto-
chemical study of tau accumulation in early stages of Alzheimer-
type neurofibrillary lesions. Acta Neuropathol., 2004, 107, 504-
Berson, E.L. Retinitis pigmentosa: unfolding its mystery. Proc.
Natl. Acad. Sci. USA, 1996, 93, 4526-4528.
Kisselev, A.F.; Goldberg, A.L. Proteasome inhibitors: from re-
search tools to drug candidates. Chem. Biol., 2001, 8, 739-758.
Anukanth, A.; Khorana, H.G. Structure and function in rhodopsin:
requirements of a specific structure for the intradiscal domain. J.
Biol. Chem., 1994, 269, 19738-19744.
Chaudhuri, T.K.; Paul, S. Protein misfolding diseases and chaper-
one based therapeutic approaches. FEBS J., 2006, 273, 1331-1349.
Diamant, S.; Eliahu, N.; Rosenthal, D.; Goloubinoff, P. Chemical
chaperones regulate molecular chaperones in vitro and in cells un-
der combined salt and heat stresses. J. Biol. Chem., 2001, 276,
Diamant, S.; Rosenthal, D.; Azem, A.; Eliahu, N.; Ben-Zvi, A.P.;
Goloubinoff, P. Dicarboxylic amino acids and glycine-betaine
regulate chaperone-mediated protein-disaggregation under stress.
Mol. Microbiol., 2003, 49, 401-410.
Mittler, R. Abiotic stress, the field environment and stress combi-
nation. Trends Plant Sci. 2006, 11, 15-19.
Nollen, E.A.; Brunsting, J.F.; Roelofsen, H.; Weber, L.A.;
Kampinga, H.H. In vivo chaperone activity of heat shock protein 70
and thermotolerance. Mol. Cell. Biol., 11991, 9, 2069-2079.
Sousa, R.; Lafer, E.M. Keep the traffic moving: mechanism of the
Hsp70 motor. Traffic, 2006, 7, 1596-1603.
Weiss, Y.G.; Bromberg, Z.; Raj, N.; Raphael, J.; Goloubinoff, P.;
Ben-Neriah, Y.; Deutschman, C.S. Enhanced Hsp70 expression
alters IB kinase proteasomal degradation in experimental ARDS.
Crit. Care Med., 2007, 35, 2128-2138.
Joglekar, A.P.; Hay, J.C. Evidence for regulation of ER/Golgi.
SNARE complex formation by HSC 70 chaperones. Eur. J. Cell
Biol., 2005, 84, 529-542.
Picard, D. Chaperoning steroid hormone action. Trends Endocri-
nol. Metab., 2006, 17, 229-235.
Voellmy, R.; Boellmann, F. Chaperone regulation of the heat shock
protein response. Adv. Exp. Med. Biol., 2007, 594, 89-99.
Vojta, L.; Soll, J.; Bölter, B. Requirements for a conservative pro-
tein translocation pathway in chloroplasts. FEBS Lett., 2007, 581,
Rapoport, T.A. Protein translocation across the eukaryotic ER and
bacterial plasma membranes. Nature, 2007, 450, 663-669.
Matouschek, A.; Pfanner, N.; Voos, W. Protein unfolding by mito-
chondria: the Hsp70. import motor. EMBO Rep., 2000, 1, 404-410.
Sur, R.; Lyte, P.A.; Southall, M.D. Hsp27 regulates pro-
inflammatory mediator release in keratinocytes by modulating NF-
kappaB signaling. J. Invest. Dermatol., 2008, 128, 1116-1122.
Bromberg, Z.; Raj, N.; Goloubinoff, P.; Deutschman, C.S.; Weiss,
Y.G. Enhanced expression of 70-kilodalton heat shock protein lim-
its cell division in a sepsis-induced model of acute respiratory dis-
tress syndrome. Crit. Care Med., 2008, 36, 246-255.
Aghdassi, A.; Phillips, P.; Dudeja, V.; Dhaulakhandi, D.; Sharif,
R.; Dawra, R.; Lerch, M.M.; Saluja, A. Heat shock protein 70 in-
creases tumorgenicity and inhibits apoptosis in pancreatic adeno-
carcinoma. Cancer Res., 2007, 67, 616-625.
So, A.; Hadaschik, B.; Sowery, R.; Gleave, M. The role of stress
proteins in prostate cancer. Curr. Genomics, 2007, 8, 252-261.
444 Current Protein and Peptide Science, 2009, Vol. 10, No. 5 Sharma et al.
Brown, I.R. Heat shock proteins and protection of the nervous
system. Ann. N. Y. Acad. Sci., 2007, 1113, 147-158.
Horwitz, J. Alpha-crystallin can function as a molecular chaperone.
Proc. Natl. Acad. Sci. USA, 1992, 89, 10449-10453.
Buchner, J. Hsp90 & Co. - a holding for folding. Trends Biochem.
Sci., 1999, 24, 136-141.
Zahn, R.; Buckle, A.M.; Perrett, S.; Johnson, C.M.; Corrales, F.J.;
Golbik R.; Fersht, A.R. Chaperone activity and structure of mono-
meric polypeptide binding domains of GroEL. Proc. Natl. Acad.
Sci. USA, 1996, 93, 15024-15029.
Ehrnsperger, M.; Graber, S.; Gaestel, M.; Buchner, J. Binding of
non-native protein to Hsp25 during heat shock creates a reservoir
of folding intermediates for reactivation. EMBO J., 1997, 16, 221-
Veinger, L.; Diamant, S.; Buchner, J.; Goloubinoff, P. The small
heat-shock protein IbpB from Escherichia coli stabilizes stress-
denatured proteins for subsequent refolding by a multichaperone
network. J. Biol. Chem., 1998, 273, 11032-11037.
Suzuki, C.K.; Rep, M.; van Dijl, J.M.; Suda, K.; Grivell L.A.;
Schatz, G. ATP-dependent proteases that also chaperone protein
biogenesis. Trends Biochem. Sci., 1997, 22, 118-123.
Licht, S.; Lee, I. Resolving individual steps in the operation of
ATP-dependent proteolytic molecular machines: from conforma-
tional changes to substrate translocation and processivity. Biochem-
istry, 2008, 47, 3595-3605.
Weber-Ban, E.U.; Reid, B.G.; Miranker, A.D.; Horwich, A.L.
Global unfolding of a substrate protein by the Hsp100 chaperone
ClpA. Nature, 1999, 401, 90-93.
Braun, B.C.; Glickman, M.; Kraft, R.; Dahlmann, B.; Kloetzel,
P.M.; Finley, D.; Schmidt, M. The base of the proteasome regula-
tory particle exhibits chaperone-like activity. Nat. Cell Biol., 1999,
Ito, K.; Akiyama, Y. Cellular functions, mechanism of action, and
regulation of FtsH protease. Annu. Rev. Microbiol., 2005, 59, 211-
Van Dyk, T.K.; Gatenby, A.A.; LaRossa, R.A. Demonstration by
genetic suppression of interaction of GroE products with many pro-
teins. Nature, 1989, 342, 451-453.
Genevaux, P.; Georgopoulos, C.; Kelley, W.L. The HSP 70 chap-
erone machines of Escherichia coli: a paradigm for the repartition
of chaperone functions. Mol. Microbiol., 2007, 66, 840-857.
Muchowski, P.J.; Wacker, J.L. Modulation of neurodegeneration
by molecular chaperones. Nat. Rev. Neurosci., 2005, 6, 11-22.
Jiang, J.; Lafer, E.M.; Sousa, R. Crystallization of a functionally
intact Hsc70 chaperone. Acta Crystallogr. Sect. F. Struct. Biol.
Cryst. Commun., 2006, 62, 39-43.
Palleros, D.R.; Reid, K.L.; Shi, L.; Welch, W.J.; Fink A.L. ATP-
induced protein-Hsp70 complex dissociation requires K+ and does
not involve ATP hydrolysis. Nature, 1993, 365, 664-666.
Schmid, D.; Baici, A.; Gehring H.; Christen, P. Kinetics of molecu-
lar chaperone action. Science, 1994, 263, 971-973.
Mayer, M.P.; Laufen, T.; Paal, K.; McCarty, J.S.; Bukau, B. Inves-
tigation of the interaction between DnaK and DnaJ by surface
plasmon resonance. J. Mol. Biol., 1999, 289, 1131-1144.
Siegenthaler, R.K.; Christen, P. Tuning of DnaK chaperone action
by non-native-protein sensor DnaJ and thermosensor GrpE. J. Biol.
Chem., 2006, 281, 34448-34456.
Mayer, M.P.; Bukau, B. Hsp70 Chaperone systems: diversity of
cellular functions and mechanism of action. Biol. Chem., 1998,
Laufen, T.; Mayer, M.P.; Beisel, C.; Klostermeier, D.; Mogk, A.;
Reinstein J. and Bukau, B. Mechanism of regulation of Hsp70
chaperones by DnaJ co-chaperones. Proc. Natl. Acad. Sci. USA,
1999, 96, 5452-5457.
Han, W.; Christen, P. Mechanism of the targeting action of DnaJ in
the DnaK molecular chaperone system. J. Biol. Chem., 2003, 278,
Goloubinoff, P.; De Los Rios, P. The mechanism of Hsp70 Chap-
erones: (entropic) pulling the models together. Trends Biochem.
Sci., 2007, 32, 372-380.
Glover, J.R.; Lindquist, S. Hsp104, Hsp70 and Hsp40: a novel
chaperone system that rescues previously aggregated proteins. Cell,
1998, 94, 73-82.
Goloubinoff, P.; Mogk, A.; Ben-Zvi, A.P.; Tomoyasu, T.; Bukau,
B. Sequential mechanism of solubilization and refolding of stable
protein aggregates by a bi-chaperone network. Proc. Natl. Acad.
Sci. USA, 1999, 96, 13732-13737.
Rüdiger, S.; Germeroth, L.; Schneider-Mergener, J.; Bukau, B.
Substrate specificity of the DnaK chaperone determined by screen-
ing cellulose-bound peptide libraries. EMBO J., 1997, 16, 1501-
Han, W.; Christen, P. cis-Effect of DnaJ on DnaK in ternary com-
plexes with chimeric DnaK/DnaJ-binding peptides. FEBS Lett.,
2004, 563, 146-150.
Pierpaoli, E.V.; Gisler, S.M.; Christen, P. Sequence-specific rates
of interaction of target peptides with the molecular chaperones
DnaK and DnaJ. Biochemistry 1998, 37, 16741- 16748.
De Los Rios, P.; Ben-Zvi, A.; Slutsky, O.; Azem, A.; Goloubinoff,
P. Hsp70 chaperones accelerate protein translocation and the un-
folding of stable protein aggregates by entropic pulling. Proc. Natl.
Acad. Sci. USA, 2006, 103, 6166-6171.
Rothman, J.E. Polypeptide chain binding proteins - catalysts of
protein folding and related processes in cells. Cell, 1989, 59, 591-
Hubbard, T.J.P.; Sander, C. The role of heat-shock and chaperone
proteins in protein folding: Possible molecular mechanisms. Pro-
tein Eng., 1991, 4, 711-717.
Rüdiger, S.; Schneider-Mergener, J.; Bukau, B. Its substrate speci-
ficity characterizes the DnaJ co-chaperone as a scanning factor for
the DnaK chaperone. EMBO J., 2001, 20, 1042-1050.
Zhu, X.; Zhao, X.; Burkholder, W.F.; Gragerov, A.; Ogata, C.M.;
Gottesman, M.E.; Hendrickson, W.A. Structural analysis of sub-
strate binding by the molecular chaperone DnaK. Science, 1996,
Wawrzynów, A.; Zylicz, M. Divergent effects of ATP on the bind-
ing of the DnaK and DnaJ chaperones to each other, or to their
various native and denatured protein substrates. J. Biol. Chem.,
1995, 270, 19300-19306.
Feifel, B.; Schönfeld, H.J.; Christen, P. D-Peptide ligands for the
co-chaperone DnaJ. J. Biol. Chem., 1998, 273, 11999-12002.
Ben-Zvi, A.; De Los Rios, P.; Dietler, G.; Goloubinoff, P. Active
solubilization and refolding of stable protein aggregates by coop-
erative unfolding action of individual HSP70 chaperones. J. Biol.
Chem., 2004, 279, 37298-37303.
Schiene-Fischer, C.; Habazettl, J.; Schmid, F.X.; Fischer, G. The
hsp70 chaperone DnaK is a secondary amide peptide bond cis-trans
isomerase. Nat. Struct. Biol., 2002, 9, 419-424.
Gisler, S.M.; Pierpaoli, E.V.; Christen, P. Catapult mechanism
renders the chaperone action of Hsp70 unidirectional. J. Mol. Biol.,
1998, 279, 833-840.
Rüdiger, S.; Buchberger, A.; Bukau, B. Interaction of Hsp70 chap-
erones with substrates. Nat. Struct. Biol., 1997, 4, 342-349.
Goloubinoff, P.; Gatenby, A.A.; Lorimer, G.H. GroE heat-shock
proteins promote assembly of foreign prokaryotic ribulose bisphos-
phate carboxylase oligomers in Escherichia coli. Nature, 1989,
Jewett, A.I.; Shea, J-E. Folding on the chaperone: Yield enhance-
ment through loose binding. J. Mol. Biol., 2006, 363, 945-957.
Jakob, U.; Gaestel, M.; Engel, K.; Buchner, J. Small heat shock
proteins are molecular chaperones. J. Biol. Chem., 1993, 268,
Buchner, J.; Schmidt, M.; Fuchs, M.; Jaenicke, R.; Rudolph, R.;
Schmid, F.; Kiefhaber, T. GroE facilitates refolding of citrate syn-
thase by suppressing aggregation. Biochemistry, 1991, 30, 1586-
Jakob, U.; Lilie, H.; Meyer, I.; Buchner, J. Transient interaction of
Hsp90 with early unfolding intermediates of citrate synthase-
implications for heat shock in vivo. J. Biol. Chem., 1995, 270,
Cyr, D.M. Cooperation of the molecular chaperone Ydj1 with
specific Hsp70 homologs to suppress protein aggregation. FEBS
Lett., 1995, 359, 129-132.
Arimon, M.; Grimminger, V.; Sanz, F.; Lashuel, H.A. Hsp104
targets multiple intermediates on the amyloid pathway and sup-
presses the seeding capacity of Abeta fibrils and protofibrils. J.
Mol. Biol., 2008, 384, 1157-1173.
Diamant, S.; Ben-Zvi, A.P.; Bukau, B.; Goloubinoff, P. Size-
dependent disaggregation of aggregated protein particles by the
DnaK chaperone machinery of Escherichia coli. J. Biol. Chem.,
2000, 275, 21107-21113.
Disaggregating Chaperones: An Unfolding StoryCurrent Protein and Peptide Science, 2009, Vol. 10, No. 5 445
 Török, Z.; Goloubinoff, P.; Horváth, I.; Tsvetkova, N.M.; Glatz,
A.; Balogh, G.; Varvasovszki, V.; Los, D.A.; Vierling, E.; Crowe,
J.H.; Vígh, L. Synechocystis hsp17 is an amphitropic protein that
stabilizes heat-stressed membranes and binds denatured proteins
for subsequent chaperone-mediated refolding. Proc. Natl. Acad.
Sci. USA, 2001, 98, 3098-3103.
Mogk, A.; Deuerling, E.; Vorderwühlbecke, S.; Vierling, E.;
Bukau, B. Small heat shock proteins, ClpB and the DnaK system
form a functional triade in reversing protein aggregation. Mol. Mi-
crobiol., 2003, 50, 585-595.
Goloubinoff, P.; Christeller, J.T.; Gatenby, A.A.; Lorimer, G.H.
Reconstitution of active dimeric ribulose bisphosphate carboxylase
from an unfolded state depends on two chaperonin proteins and
Mg-ATP. Nature, 1989, 342, 884-889.
Saidi, Y.; Finka, A.; Zryd, J.P.; Schaefer, D.G.; Goloubinoff, P.
Controlled expression of recombinant proteins in Physcomitrella
Patens by a conditional heat-shock promoter: a tool for plant re-
search and biotechnology. Plant Mol. Biol., 2005, 59, 695-709.
Saidi, Y.; Domini, M.; Choy, F.; Zryd, J.P.; Schwitzguebel, J.P.;
Goloubinoff, P. Activation of the heat shock response in plants by
chlorophenols: transgenic Physcomitrella patens as a sensitive bio-
sensor for organic pollutants. Plant Cell Environ., 2007, 130, 753-
Van Montfort, R.; Slingsby, C.; Vierling, E. Structure and function
of the small heat shock protein ?-crystallin family of molecular
chaperones. Adv. Protein Chem., 2001, 59, 105-156.
Caspers, G.-J.; Leunissen, J.A.M.; De Jong, W.W. The expanding
small heat-shock protein family, and structure predictions of the
conserved "alpha-crystallin domain". J. Mol. Evol., 1995, 40, 238-
Garrido, C. Size matters: of the small Hsp27 and its large oli-
gomers. Cell Death Differ., 2002, 9, 483-485.
Litt, M.; Kramer, P.; LaMorticella, D.M.; Murphey, W.; Lovrien,
E.W.; Weleber, R.G. Autosomal dominant congenital cataract as-
sociated with a missense mutation in the human alpha crystallin
gene CRYAA. Hum. Mol. Genet., 1998, 7, 471-474.
Benndorf, R.; Welsh, M.J. Shocking degeneration. Nat. Genet.,
2004, 36, 547-548.
Vicart, P.; Caron, A.; Guicheney, P.; Li, Z.; Prevost, M.C.; Faure,
A.; Chateau, D.; Chapon, F.; Tome, F.; Dupret, J-M.; Paulin, D.;
Fardeau, M. A missense mutation in the alphaB-crystallin chaper-
one gene causes a desmin-related myopathy. Nat. Genet. 1998, 20,
Evgrafov, O.V.; Mersiyanova, I.; Irobi, J.; Van Den Bosch, L.;
Dierick, I.; Leung, C.L.; Schagina, O.; Verpoorten, N.; Van Impe,
K.; Fedotov, V.; Dadali, E.; Auer-Grumbach, M.; Windpassinger,
C.; Wagner, K.; Mitrovic, Z.; Hilton-Jones, D.; Talbot, K.; Martin,
J.J.; Vasserman, N.; Tverskaya, S.; Polyakov, A.; Liem, R.K.;
Gettemans, J.; Robberecht, W.; De Jonghe, P.; Timmerman, V.
Mutant small heat-shock protein 27 causes axonal Charcot-Marie-
Tooth disease and distal hereditary motor neuropathy. Nat. Genet.,
2004, 36, 602-606.
Irobi, J.; Van Impe, K.; Seeman, P.; Jordanova, A.; Dierick, I.;
Verpoorten, N.; Michalik, A.; De Vriendt, E.; Jacobs, A.; Gerwen,
V.V. Hot-spot residue in small heat-shock protein 22 causes distal
motor neuropathy. Nat. Genet., 2004, 36, 597-601.
Arrigo, A.P. The cellular "networking" of mammalian Hsp27 and
its functions in the control of protein folding, redox state and apop-
tosis. Adv. Exp. Med. Biol., 2007, 594, 14-26.
Arrigo, A.P.; Virot, S.; Chaufour, S.; Firdaus, W.; Kretz-Remy, C.;
Diaz-Latoud, C. Hsp27 consolidates intracellular redox homeosta-
sis by upholding glutathione in its reduced form and by decreasing
iron intracellular levels. Antioxid. Redox Signal, 2005, 7, 414-422.
Horwitz, J.; Huang, Q.L.; Ding, L.; Bova, M.P. Lens alpha-
crystallin: Chaperone-like properties. Methods Enzymol., 1998,
Lee, G.J.; Roseman, A.M.; Saibil, H.R.; Vierling, E. A small heat
shock protein stably binds heat-denatured model substrates and can
maintain a substrate in a folding-competent state. EMBO J., 1997,
Haslberger, T.; Zdanowicz, A.; Brand, I.; Kirstein, J.; Turgay, K.;
Mogk, A.; Bukau, B. Protein disaggregation by the AAA+ chaper-
one ClpB involves partial threading of looped polypeptide seg-
ments. Nat. Struct. Mol. Biol., 2008, 15, 641-650.
Li, J.; Sha, B. Crystal structure of the E. coli Hsp100 ClpB
N-terminal domain. Structure (Camb), 2003, 11, 323-328.
 Weibezahn, J.; Tessarz, P.; Schlieker, C.; Zahn, R.; Maglica, Z.;
Lee, S.; Zentgraf, H.; Weber-Ban, E.U.; Dougan, D.A.; Tsai, F.T.;
Mogk, A.; Bukau, B. Thermotolerance requires refolding of aggre-
gated proteins by substrate translocation through the central pore of
ClpB. Cell, 2004, 119, 653-665.
Ben-Zvi, A.; Goloubinoff, P. Mechanisms of chaperone-mediated
protein disaggregation. J. Struct. Biol., 2001, 135, 84-93.
Schlieker, C.; Tews, I.; Bukau, B.; Mogk, A. Solubilization of
aggregated proteins by ClpB/DnaK relies on the continuous extrac-
tion of unfolded polypeptides. FEBS Lett., 2004, 578, 351-356.
Liberek, K.; Lewandowska, A.; Zietkiewicz, S. Chaperones in
control of protein disaggregation. EMBO J., 2008, 27, 328-335.
Doyle, S.M.; Shorter, J.; Zolkiewski, M.; Hoskins, J.R.; Lindquist,
S.; Wickner, S. Asymmetric deceleration of ClpB and Hsp104
ATPase activity unleashes protein-remodeling activity. Nat. Struct.
Mol. Biol., 2007, 14, 114-122.
Shorter, J.; Lindquist, S. Hsp104, Hsp70 and Hsp40 interplay regu-
lates formation, growth and elimination of Sup35 prions. EMBO J.,
2008, 27, 2712-2724.
Weibezahn, J.; Schlieker, C.; Tessarz, P.; Mogk, A.; Bukau B.
Novel insights into the mechanism of chaperone-assisted protein
disaggregation. Biol. Chem., 2005, 386, 739-744.
Rutherford, S.L.; Lindquist, S. Hsp90 as a capacitor for morpho-
logical evolution. Nature, 1998, 396, 336-342.
Horwich, A.L.; Fenton, W.A.; Chapman, E.; Farr, G.W. Two fami-
lies of chaperonin: Physiology and mechanism. Annu. Rev. Cell.
Dev. Biol., 2007, 23, 115-145.
Saibil, H.R.; Horwich, A.L.; Fenton, W.A. Allostery and protein
substrate conformational change during GroEL/GroES-mediated
protein folding. Adv. Protein. Chem., 2001, 59, 45-72.
Azem, A.; Diamant, S.; Kessel, M.; Weiss, C.; Goloubinoff, P. The
protein-folding activity of chaperonins correlates with the symmet-
ric GroEL-14(GroES-7)-2 heterooligomer. Proc. Natl. Acad. Sci.
USA, 1995, 92, 12021-12025.
Török, Z.; Vigh, L.; Goloubinoff, P. Fluorescence detection of
symmetric GroEL14(GroES7)2 hetero-oligomers involved in pro-
tein-release during the chaperonin cycle. J. Biol. Chem., 1996, 271,
Koike-Takeshita, A.; Yoshida, M.; Taguchi, H. Revisiting the
GroEL-GroES reaction cycle via the symmetrical intermediate im-
plied by novel aspects of the GroEL (D398A) mutant. J. Biol.
Chem., 2008, 283, 23774-23781.
Sameshima, T.; Ueno, T.; Iizuka, R.; Ishii, N.; Terada, N.; Okabe,
K.; Funatsu, T. Football- and bullet-shaped GroEL-GroES com-
plexes coexist during the reaction cycle. J. Biol. Chem., 2008, 283,
Ellis, R.J.; Minton, A.P. Protein aggregation in crowded environ-
ments. Biol. Chem., 2006, 387, 485-497.
Shtilerman, M.; Lorimer, G.H.; Englander, S.W. Chaperonin func-
tion: folding by forced unfolding. Science, 1999, 284, 822-825.
Apetri, A.C.; Horwich, A.L. Chaperonin chamber accelerates pro-
tein folding through passive action of preventing aggregation. Proc.
Natl. Acad. Sci. USA, 2008, 105, 17351-17355.
Tang, Y.C.; Chang, H.C.; Roeben, A.; Wischnewski, D.; Wisch-
newski, N.; Kerner, M.J.; Hartl, F.U.; Hayer-Hartl, M. Structural
features of the GroEL-GroES nano-cage required for rapid folding
of encapsulated protein. Cell, 2006, 125, 903-914.
Sharma, S.; Chakraborty, K.; Müller, B.K.; Astola, N.; Tang, Y.C.;
Lamb, D.C.; Hayer-Hartl, M.; Hartl, F.U. Monitoring protein con-
formation along the pathway of chaperonin-assisted folding. Cell,
2008, 133, 142-153.
Lin, Z.; Madan, D.; Rye, H.S. GroEL stimulates protein folding
through forced unfolding. Nat. Struct. Mol. Biol., 2008, 15, 303-
Donnelly, A.; Blagg, B.S. Novobiocin and additional inhibitors of
the Hsp90 C-terminal nucleotide-binding pocket. Curr. Med.
Chem., 2008, 15, 2702-2717.
Gadelle, D.; Graille, M.; Forterre, P. The HSP90 and DNA topoi-
somerase VI inhibitor radicicol also inhibits human type II DNA
topoisomerase. Biochem. Pharmacol., 2006, 72, 1207-1216.
Wiech, H.; Buchner, J.; Zimmermann, R.; Jakob, U. Hsp90 chaper-
ones-protein folding in vitro. Nature, 1992, 358, 169-170.
Wandinger, S.K.; Richter, K.; Buchner, J. The Hsp90 chaperone
machinery. J. Biol. Chem., 2008, 283, 18473-18477.
446 Current Protein and Peptide Science, 2009, Vol. 10, No. 5 Sharma et al.
Xu, Y.; Lindquist, S. Heat shock protein hsp90 governs the activity
of pp60v-src kinase. Proc. Natl. Acad. Sci. USA, 1993, 90, 7074-
Shiau, A.K.; Harris, S.F.; Southworth, D.R.; Agard, D.A.; Struc-
tural analysis of E. coli hsp90 reveals dramatic nucleotide-
dependent conformational rearrangements. Cell, 2006, 127, 329-
 Kaganovich, D.; Kopito, R.; Frydman, J. Misfolded proteins parti-
tion between two distinct quality control compartments. Nature,
2008, 454, 1088-1095.
Winslow, A.R.; Rubinsztein, D.C. Autophagy in neurodegeneration
and development. Biochim. Biophys. Acta., 2008, 1782, 723-729.
Touriki, N.; Tawfik, D.S. Chaperonin overexpression promotes
genetic variation and enzyme evolution. Nature, 2009, 459, 668-
Received: January 12, 2009 Revised: March 09, 2009 Accepted: March 11, 2009
Sandeep Kumar Sharma