Structural and functional roles for b-strand 7 in the
a-crystallin domain of p26, a polydisperse small heat shock
protein from Artemia franciscana
Yu Sun, Svetla Bojikova-Fournier and Thomas H. MacRae
Department of Biology, Dalhousie University, Halifax, NS, Canada
Protein folding and maintenance of an appropriate 3D
structure occur with the assistance of molecular chap-
erones, including Hsp60 (chaperonins), Hsp70, Hsp90,
Hsp104 ⁄ ClpB, Hsp110 and the small heat shock pro-
teins (sHSPs) [1–6]. Several chaperones are actively
involved in protein folding, whereas others, and in par-
ticular the sHSPs, protect proteins during stresses such
as heat shock, oxidation and hypoxia ⁄ anoxia. Mole-
cular chaperones also remove damaged proteins
through the action of CHIP, a ubiquitin ligase [6].
Keywords
a-crystallin domain; Artemia franciscana;
molecular chaperone; p26 structure ⁄
function; small heat shock protein
Correspondence
T. H. MacRae, Department of Biology,
Dalhousie University, Halifax, NS,
Canada B3H 4J1
Fax: +1 902 4943736
Tel: +1 902 4946525
E-mail:
(Received 5 July 2005, revised 26 December
2005, accepted 5 January 2006)
doi:10.1111/j.1742-4658.2006.05129.x
Oviparous development in the extremophile crustacean, Artemia franciscana,
generates encysted embryos which enter a profound state of dormancy,

1020 FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS
sHSPs usually occur as oligomers composed of sub-
units ranging in molecular mass from 12 to 43 kDa,
and they protect proteins from irreversible denatura-
tion independent of ATP [3,7–13]. The conserved
a-crystallin domain of  90 amino acid residues,
located towards the C terminus, is important for oligo-
mer formation and chaperoning [14–16]. The a-crystal-
lin domain is preceded by a poorly conserved
N-terminal region proposed to function in oligomer
assembly, subunit exchange and substrate binding [17–
22], and is followed by a flexible, polar, C-terminal
extension of variable sequence that influences solubility
and oligomerization [17,21,23–25]. sHSP secondary
structure is dominated by b-pleated sheet, but the
quaternary structure is variable [26,27]. Hsp16.5 from
the archaeon, Methanococcus jannaschii, and Hsp16.9
from wheat, Triticum aestivum, assemble monodisperse
oligomers and they have been crystallized, revealing
important sHSP structural attributes [14,16]. sHSPs,
most of which form polydisperse oligomers, interact
with several substrates and a reservoir of intermediates
accrues, a progression involving oligomer dissociation
and subunit exchange [19,28–30], but which may also
occur upon rearrangement of oligomer structure in the
absence of dissociation [31]. When stress is relieved,
substrates are released and renatured, processes occur-
ring spontaneously or with assistance from other
molecular chaperones [32,33]. The sHSPs influence
cytoskeleton organization [34–36], apoptosis [37–40]

disperse sHSPs by introducing single-site mutations in
p26, and to better define the relationship between p26
and stress resistance in A. franciscana. The amino acids
selected for study are highly conserved from species
to species (Fig. 1); at least one causes disease when
mutated [47–52] and, as indicated by molecular model-
ling, they reside in a key structural interface, suggest-
ing that their modification will affect oligomerization
and chaperoning. The role of b-strand 7 in oligomeri-
zation was demonstrated in this work. Additionally, as
for other sHSPs, changing the conserved p26 a-crystal-
lin domain arginine (R114) reduced chaperone activity,
but in this case with only a minor effect on oligomeri-
zation. This showed, in concert with analysis of the
F112R mutation, that oligomerization and chapero-
ning are not linked in p26. The resistance of p26 chap-
eroning activity to single-site mutations suggests a
stable protein and this, in concert with the large
amount of p26 present during oviparous development,
undoubtedly contributes to the remarkable stress
resistance of encysted Artemia embryos.
Results
Site-directed mutagenesis and purification
of bacterially produced p26
cDNAs encoding the amino acid substitutions R110G,
F112R, R114A and Y116D in the a-crystallin domain
of p26 were cloned in the prokaryotic expression
vector, pPROTet.E233, and used to transform E. coli
BL21PRO. Sequencing demonstrated that each p26
cDNA contained only the introduced substitution.

plasm of transfected cells. In contrast, all COS-1 cells
transfected with cDNA containing the R114A muta-
tion had p26 in nuclei as well as in the cytoplasm
(Fig. 3). p26 R110G, F112R and Y116D occurred in
the cytoplasm and nuclei of transfected cells, although
some nuclei lacked the protein (not shown). p26 was
subsequently prepared from transfected COS-1 cells
for determination of oligomer size.
Fig. 1. Multiple sequence alignment of rep-
resentative small heat shock proteins
(sHSPs). The amino acid sequences of
selected sHSPs were analyzed by
CLUSTAL W
(1.82). HHSP27, Homo sapiens Hsp27,
P04792; MHSP25, Mus musculus Hsp25,
P14602; HCRYAA, H. sapiens aA-crystallin,
P02489; HCRYAB, H. sapiens aB-crystallin,
P02511; Ap26, Artemia franciscana p26,
AF031367; and WHSP16.9, wheat Hsp16.9,
1GME. sHSP domains based on the
sequence of p26 are indicated above the
alignment, secondary structure elements
based on the sequence of wheat Hsp16.9
are below the alignment, and the conserved
a-crystallin domain amino acid residues
selected for mutational analysis are shaded.
Residue number is indicated on the right.
–, no residue; *, identical residues; :, con-
served substitution; ., semiconserved sub-
stitution.

oligomer size upon synthesis in COS-1 cells, demon-
strating that results obtained upon synthesis in bacteria
were not specific to the organism or the recombinant
construct.
Amino acid substitutions in the p26 a-crystallin
domain reduced chaperone activity
Although all p26 variants conferred thermotolerance
on bacteria, WT p26 was the most effective (Fig. 5A).
Thermotolerance levels induced by R110G, F112R
and Y116D were similar to (P > 0.05) and significantly
higher than those conferred by R114A (P<0.05),
which provided the least protection. Bacteria lacking
p26 failed to survive the 60 min heat shock.
WT p26 at 1.6 lm, representing a chaperone to sub-
strate molar ratio of 0.4 : 1 if monomers are compared,
almost completely prevented dithiothreitol-induced
insulin aggregation at 25 °C, and even at 0.1 lm
p26 aggregation was inhibited by more than 40%
(Fig. 5B). At all concentrations, WT p26 prevented
insulin aggregation the most and R114A the least, fol-
lowed by F112R, Y116D and R110G, with the latter
two not significantly different from one another. Chap-
eroning of insulin by p26 purified from Artemia [45]
and E. coli was very similar, whereas BSA and IgG at
1.6 lm had no effect on dithiothreitol-induced insulin
aggregation (not shown).
At 600 nm, a chaperone to target (monomer to
dimer) molar ratio of 4 : 1, WT p26 inhibited citrate
synthase aggregation almost completely after 1 h at
43 °C (Fig. 5C). R110G, Y116D and F112R were

least protection, but was still about 60% as potent
as WT p26 at 600 nm. WT p26 also shielded citrate
synthase enzyme activity against heat-induced inacti-
vation better than mutated p26, and at 1200 nm the
activity remaining was essentially the same as in
unheated preparations (Fig. 5D). R114A was the
least effective of all p26 variants in protecting
enzyme activity, although differences disappeared at
lower concentrations. Of the remaining p26 mutants,
chaperone activity decreased from R110G to Y116D,
which were similar, and then to F112R. Chaperone
activities of p26 from Artemia [45] and E. coli with
citrate synthase were similar, whereas BSA and IgG
at 1200 nm neither prevented citrate synthase aggre-
gation nor preserved enzyme activity (data not
shown).
To summarize, as determined by thermotolerance
induction in E. coli, dithiothreitol induced insulin
aggregation at 25 °C, heat-induced citrate synthase
aggregation at 43 °C, and maintenance of citrate syn-
thase enzyme activity at 43 °C, WT p26 possessed the
greatest chaperone activity and R114A the least. It is
noteworthy, however, that all p26 variants protected
bacteria and substrate proteins in vitro.
Modification of p26 structure by amino acid
substitutions
Measurement of intrinsic fluorescence demonstrated
that the maximum emission peak for each p26 mutant
was less than for WT p26 (Fig. 6A). Three variants
(R110G, F112R and Y116D) had very similar emission

Artemia were similar (Fig. 6A). This structural resem-
blance, and the functional analysis mentioned previ-
ously, indicate data obtained by examining bacterially
produced p26 are indicative of the protein synthesized
in Artemia.
Localization of amino acid substitutions within
p26
The p26 tetramer, modeled on the 3D crystal structure
of wheat Hsp16.9, consists of two dimers, with mono-
mers A and B in dimer 1 and C and D in dimer 2
(Fig. 8). The a-crystallin domain of each monomer is
composed of nine b-strands (labeled b2–b10), with the
b6 strand situated in a large loop, L5 ⁄ 7, located
between b-strands 5 and 7. b-strand 10 inhabits
Table 1. Characteristics of p26 oligomers. The molecular mass of
p26 oligomers produced in transformed Escherichia coli BL21PRO
and transfected COS-1 cells was determined by sucrose density
gradient centrifugation. Monomer mass refers to the mass of p26
monomers and was calculated using a p26 molecular mass of
20.8 kDa, as determined by
GENERUNNER (version 3.05, Hastings
Software, Inc.) with corrections for protein modifications. Oligomer
mass range represents the smallest to largest oligomers observed,
and maximum monomer number refers to the number of subunits
in oligomers of maximum mass.
Expression
system
p26
mutant
Monomer

depends upon contact of b-strand 10 from the C-ter-
minal extensions of monomers A and D with
b-strands 4 and 8 in the a-crystallin domain of
monomers C and B, respectively (Fig. 8). A more pro-
minent dimer–dimer interface occurs with b-strand 7
of monomer B interacting with loop L5 ⁄ 7 of mono-
mer C, and b-strand 7 of monomer C reacting with
L5 ⁄ 7 of monomer B, regions of high similarity
between p26 and Hsp16.9 (Fig. 8). The p26 residues
examined in this study are situated in b-strand 7,
with mutations R110G and F112R directly in the
dimer–dimer interface. As a result of the spatial dis-
position of monomers and their b-strand elements in
the a-crystallin domain, the amino acid substitution
F112R introduces two changes at the dimer–dimer
interface. Although Y116 and R114 are in b-strand
7, neither is shown by the model to reside directly
in the dimer–dimer interface. Modification of these
AB
CD
Fig. 5. Chaperone activity of p26. (A) Transformed Escherichia coli BL21PRO cells were incubated at 54 °C for 1 h with samples removed
periodically, diluted, and plated in duplicate on Luria–Bertani (LB) agar followed by incubation at 37 °C for 16 h. The log
10
values of colony-
forming units (CFU) per ml were plotted against heat shock in min. Bacteria containing the pPROTet.E233 vector lacking p26 cDNA did not
survive the entire 60 min. Standard errors ranged from 3.3 to 7.1%. (B) Bacterially produced p26 purified to apparent homogeneity was incu-
bated with insulin for 30 min in the presence of dithiothreitol, and solution turbidity was measured at 400 nm. The p26 variants tested are
indicated in the figure and they appear in the same order in each histogram group. The standard error ranged from 4.2 to 5.8%. (C) Purified,
bacterially produced p26 at 600 n
M was heated at 43 °C for 1 h with 150 nM citrate synthase and the solution turbidity was measured at

obtained for purified p26 dissolved in 10 m
M NaH
2
PO
4
, pH 7.1, to
0.2 mgÆmL
)1
. The absorption data are expressed as molar ellipticity
in degrees cm
2
Ædmol
)1
(m deg), with each spectrum the average of
three scans.
AB
Fig. 6. Tertiary structure perturbation of p26. (A) The intrinsic fluorescence of purified p26 diluted in 10 mM NaH
2
PO
4
, pH 7.1, to
0.06 mgÆmL
)1
was determined. The excitation wavelength was 280 nm, with a 2-nm band pass, and fluorescence emission was detected
from 310 to 400 nm. The standard error ranged from 3.5 to 7.5%. (B) Surface hydrophobicity of purified p26 at 0.06 mgÆmL
)1
in 10 mM
NaH
2
PO

b-turn 16.7 16.6 16.9 16.9 16.5 16.8
Random coil 34.8 34.8 34.9 33.3 34.0 32.5
a-crystallin domain of p26 Y. Sun et al.
1026 FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS
Oligomers for each exogenously produced p26 vari-
ant are composed of similar numbers of monomers
when synthesized in mammalian and bacterial cells,
and oligomerization is unaffected by protein purifica-
tion, observations important for subsequent analysis of
the protein in in vitro assays. Single-site mutations to
the p26 a-crystallin domain generally decreased oligo-
mer size in comparison to WT p26, with mutation
F112R reducing oligomerization most dramatically. A
tetramer model of p26 was constructed on the basis of
the crystallin structure of wheat Hsp16.9 [14], a mono-
disperse sHSP used for modeling of human aA- and
aB-crystallins [47], in order to position residues within
the a-crystallin domain, better understand the conse-
quences of amino acid substitutions, and identify pro-
tein regions involved in oligomer assembly. The four
modified a-crystallin domain residues are spatially
close to one another in the p26 model, with R110 and
F112 occupying central positions in the dimer–dimer
interface. The R110G mutation had relatively limited
effect on oligomer size, indicating that p26, and by
extrapolation, other polydisperse sHSPs tolerate charge
reduction at the dimer–dimer interface. The F112R
modification, on the other hand, placed two positively
charged residues in the dimer–dimer interface and the
maximum oligomer mass dropped, as indicated by

side the dimer–dimer interface. Interestingly, in Chi-
nese hamster Hsp27, mutation R148G had a limited
effect on chaperone activity and reduced oligomers to
dimers [54], contrasting the results obtained with p26
R114A. Whether this indicates fundamental differences
between the two proteins awaits further study.
WT p26, purified from transformed bacteria, almost
completely prevented heat-induced citrate synthase
aggregation and loss of enzyme activity at a molar
ratio of 4 : 1 (monomer to dimer), a result obtained
previously [17] and which was similar to the activity of
p26 from Artemia embryos (data not shown). Chemic-
ally induced insulin aggregation at 25 °C was inhibited
at a monomer to monomer ratio of 0.4 : 1, the first
measurement of p26 chaperone activity in vitro at a
temperature near the optimum for Artemia growth.
Although it is difficult to compare chaperone activity
across species owing to variation in experimental tech-
niques, effective chaperone to substrate molar ratios
determined by heating citrate synthase in the presence
of other representative sHSPs are 2 : 1 for Bradyrhizo-
bium japonicum sHSPs [55],  3 : 1 for Caenorhabditis
elegans Hsp16–2 [56], 15 : 1 for Mycobacterium tuber-
culosis Hsp16.3, and 5 : 1 for human aB-crystallin [57].
The bovine a-crystallin to substrate ratio for protec-
tion against dithiothreitol induced denaturation ranges
from 2 : 1 for insulin and a-lactalbumin, 8 : 1 for BSA
and 10 : 1 for ovotransferrin, with the ratio rising as
the molecular mass of the substrate increases [58]. For
human aB-crystallin, the ratio is 1 : 1 [48]. The p26

All a-crystallin domain mutants, including R114A,
retain significant amounts of chaperone activity. In
comparison, loss of chaperone activity reported upon
introduction of substitution R116C into aA-crystallin
ranges from 40% to almost 100% [49,51,52]. The aB-
crystallin mutation R120G promotes protein aggrega-
tion in in vitro turbidimetric assays, reduces in vitro
chaperone activity [48–50], and decreases thermotoler-
ance induction by 70% while promoting inclusion
body formation [61], the latter not being observed for
p26 R114A. p26 chaperone activity appears to be more
resistant to modification of this conserved a-crystallin
domain arginine, suggesting that the residue is less crit-
ical than in mammalian a-crystallins where modifica-
tion leads to disease [62–64]. The ramifications of these
observations for p26 are worthy of note. For example,
a-crystallins function in the mammalian lens for a life-
time, indicating, by comparison, that p26 is sufficiently
stable to protect Artemia for long periods of time, as
required in encysted embryos.
p26 oligomers synthesized in mammalian and bacter-
ial cells are similar in size to one another and to Arte-
mia p26, indicating that characteristics derived by
studying bacterially produced p26 are reflective of the
protein from Artemia. Moreover, p26 localization in
transfected cells is interesting because the protein
migrates into Artemia nuclei during diapause and
stress [65]. Other sHSPs, such as Hsp20, aB-crystallin
and Hsp27, access nuclei where they may be associated
with speckles and nucleoli [66,67]. The R120G muta-

chaperoning is not dependent upon oligomerization,
and chaperone activity effectively tolerates structural
perturbation, this potentially contributing to stress
resistance in Artemia embryos. The ability of p26 to
prevent aggregation and loss of enzyme activity, in
concert with its abundance, indicate a large protective
capacity during oviparous development. Proteins shiel-
ded by p26 would be readily available upon termin-
ation of diapause to initiate development, conferring a
marked advantage on encysted Artemia embryos.
Experimental procedures
Construction of p26 cDNAs
p26 amino acid substitutions were generated by site-direc-
ted mutagenesis by using the QuikChange
tm
Site-directed
Mutagenesis kit (Stratagene, La Jolla, CA, USA), using
pRSET.C-p26–3-6-3 as template [46] and designated prim-
ers (Table 3). PCR mixtures were incubated for 30 s at
95 °C prior to 12 cycles of 30 s at 95 °C, 1 min at 55 °C
and 8 min at 68 °C. DNA products were digested with
DpnIat37°C for 1 h and used to transform E. coli XL1-
blue supercompetent cells (Stratagene). p26 cDNA inserts
were recovered from pRSET.C plasmids by digestion with
BamHI and XhoI, electrophoresis in agarose and purifica-
tion with the GFX
tm
PCR DNA and Gel Band purification
kit (Amersham Biosciences, Piscataway, NJ, USA) before
cloning in the eukaryotic expression vector, pcDNA4 ⁄ TO ⁄

6
tag.
Blots were then incubated with either horseradish peroxi-
dase (HRP)-conjugated goat anti-rabbit IgG or HRP-conju-
gated goat anti-mouse IgG (Jackson ImmunoResearch,
Mississauga, ON, Canada) and immunoconjugates were
detected with Western Lightning Enhanced Chemilumines-
cence (ECL) Reagent Plus (PerkinElmer Life Sciences,
Boston, MA, USA).
p26 synthesis and localization in transiently
transfected COS-1 cells
Cloned p26 cDNA in SuperFect
tm
(Qiagen, Mississauga,
ON, Canada) was employed to transiently transfect COS-1
cells [17]. The cells were trypsinized 24 h after transfection
for preparation of protein extract, centrifuged at 1500 g for
5 min, washed with 1 mL of phosphate-buffered saline
(NaCl ⁄ P
i
) (140 mm NaCl, 2.7 mm KCl, 8.0 mm Na
2
HPO
4
,
1.5 mm KH
2
PO
4
, pH 7.4), and incubated on ice for 20 min

iodide [17].
p26 oligomerization
p26, either purified from transfected bacteria or in extracts
from transfected COS-1 cells and transformed bacteria, was
centrifuged at 200 000 g for 12 h at 4 °C in 10 mL of con-
tinuous 10–50% (w ⁄ v) sucrose gradients prepared in 0.l m
Tris ⁄ glycine buffer, pH 7. p26 was detected in gradient frac-
tions by immunoprobing of western blots [17]. Molar ratios
were calculated using a p26 molecular mass of 20.8 kDa, as
determined by generunner (version 3.05, Hastings Soft-
ware, Inc., Hastings on Hudson, NY, USA) with corrections
for protein modifications. Molecular mass markers alpha-
lactalbumin (14.2 kDa), carbonic anhydrase (29 kDa), BSA
(66 kDa), alcohol dehydrogenase (150 kDa), apoferritin
(443 kDa) and thyroglobulin (669 kDa) (Sigma) were
centrifuged separately and localized in sucrose gradients by
measuring the A
280
of fractions.
p26 induced thermotolerance in E. coli
Transformed E. coli, incubated overnight with shaking at
37 °C in 2 mL of LB medium, containing spectinomycin,
chloramphenicol and aTc, were diluted 1 : 10 in fresh LB
medium and incubated at 54 °C. Samples were removed at
timed intervals during heating, plated on LB agar and col-
onies were counted after incubation for 24 h at 37 °C. All
experiments were performed in triplicate. Between data
groups, two-sample t-tests were performed at a confidence
level of 95% with the statistical software minitab 14.12.0
(Minitab Inc., State College, PA, USA) to evaluate the sig-

PO
4
, pH 7.1, and fluorescence spectra were measured
at 25 °C with a SPECTRAmax GEMINIXS fluorescence
spectrophotometer (Molecular Devices). The emission
wavelength was set initially at 340 nm with a 2-nm band
pass, and fluorescence excitation was detected from 250 to
310 nm. The excitation wavelength was then set to 280 nm
with a 2-nm band pass, and fluorescence emission was
detected from 310 to 400 nm. To measure surface hydro-
phobicity, mixtures containing 80 lm ANS (Molecular
Probes, Eugene, OR, USA) and 0.06 mgÆmL
)1
p26 in
10 mm NaH
2
PO
4
, pH 7.1, were incubated for 5 min at
either 25 °Cor43°C. The excitation wavelength was set to
388 nm with a band pass of 8 nm, and emission wavelength
was 473 nm with a band pass of 8 nm. Measurements were
made with an AMINCO Bowman series z luminescence
spectrometer (AMINCO, Rochester, NY, USA) equipped
with a thermostated circulating water bath. All spectra were
recorded in duplicate using two independently prepared
samples. Far-UV CD spectra were recorded at 25 °C over
180–260 nm in a JASCO J-810 spectropolarimeter (Japan
Spectroscopic, Tokyo, Japan). A 0.1-cm path length quartz
cuvette containing 0.2 mgÆmL

zation. p26 a-crystallin domain tetramer models were based
on the corresponding Hsp16.9 tetramer. One hundred mod-
els were generated, and the structure displaying the lowest
objective function value was used to represent p26. Model
evaluation was made using verify3d without further
energy minimization to preserve the conserved residue side
chain conformation [73,74]. Using the Hsp16.9 monomer as
a template, the root mean square deviation (RMSD) [75]
was 2.3 A
˚
, an acceptable value that decreased to 0.6 A
˚
when flexible protein regions were excluded. Application of
procheck [76] revealed that the stereochemical quality of
the model was reliable, with 84% of the residues in the
most favored regions of the tetramer model and none in
disallowed regions. Graphical representations were made
using vmd [77].
Acknowledgements
We thank Dr Stephen Bearne and Dr Neil Ross for
experimental support with biophysical studies and Mr
Carey Isenor for expert assistance in image processing.
This work was supported by a Natural Sciences and
Engineering Research Council of Canada Discovery
Grant, a Nova Scotia Health Research Founda-
tion ⁄ Canadian Institutes of Health Research Regional
Partnership Plan Grant, and a Heart and Stroke Foun-
dation of Nova Scotia Grant to T.H.M. and a
NSHRF Student Fellowship to Y.S.
References

64–93.
10 Sun W, Van Montagu M & Verbruggen N (2002) Small
heat shock proteins and stress tolerance in plants. Bio-
chim Biophys Acta 1577, 1–9.
11 Horwitz J (2003) Alpha-crystallin. Exp Eye Res 76,
145–153.
12 Laksanalamai P & Robb FT (2004) Small heat shock
proteins from extremophiles: a review. Extremophiles 8,
1–11.
13 Taylor RP & Benjamin IJ (2005) Small heat shock pro-
teins: a new classification scheme in mammals. J Mol
Cell Cardiol 38, 433–444.
14 van Montfort RLM, Basha E, Friedrich KL, Slingsby C
& Vierling E (2001) Crystal structure and assembly of a
eukaryotic small heat shock protein. Nat Struct Biol 8,
1025–1030.
15 Koteiche HA & Mchaourab HS (2002) The determi-
nants of the oligomeric structure in Hsp16.5 are
encoded in the a-crystallin domain. FEBS Lett 519,
16–22.
16 Kim KK, Kim R & Kim S-H (1998) Crystal structure
of a small heat-shock protein. Nature 394, 595–599.
17 Sun Y, Mansour M, Crack JA, Gass GL & MacRae
TH (2004) Oligomerization, chaperone activity, and
nuclear localization of p26, a small heat shock protein
from Artemia franciscana. J Biol Chem 279, 39999–
40006.
18 Stromer T, Fischer E, Richter K, Haslbeck M & Buch-
ner J (2004) Analysis of the regulation of the molecular
chaperone Hsp26 by temperature-induced dissociation:

Esposito G, Behlke J, Lutsch G, Kotlyarov A & Gaestel
M (2000) Mouse Hsp25, a small shock protein. The role
of its C-terminal extension in oligomerization and cha-
perone action. Eur J Biochem 267, 1923–1932.
26 Haslbeck M, Braun N, Stromer T, Richter B, Model N,
Weinkauf S & Buchner J (2004) Hsp42 is the general
small heat shock protein in the cytosol of Saccharo-
myces cerevisiae. EMBO J 23, 638–649.
27 Haley DA, Bova MP, Huang QL, Mchaourab HS &
Stewart PL (2000) Small heat-shock protein structures
reveal a continuum from symmetric to variable assem-
blies. J Mol Biol 298, 261–272.
28 Regini JW, Grossmann JG, Burgio MR, Malik NS, Ko-
retz JF, Hodson SA & Elliott GF (2004) Structural
changes in a-crystallin and whole eye lens during heat-
ing, observed by low-angle X-ray diffraction. J Mol Biol
336, 1185–1194.
29 Benesch JL, Sobott F & Robinson CV (2003) Thermal
dissociation of multimeric protein complexes by using
nanoelectrospray mass spectrometry. Anal Chem 75,
2208–2214.
30 Lentze N, Studer S & Narberhaus F (2003) Structural
and functional defects caused by point mutations in the
a-crystallin domain of a bacterial a-heat shock protein.
J Mol Biol 328, 927–937.
31 Franzmann TM, Wu
¨
hr M, Richter K, Walter S &
Buchner J (2005) The activation mechanism of Hsp26
does not require dissociation of the oligomer. J Mol

is a novel inhibitor of TRAIL-induced apoptosis that
suppresses the activation of caspase-3. J Biol Chem 280,
11059–11066.
39 Mao Y-W, Liu J-P, Xiang H & Li DW-C (2004)
Human aA- and aB-crystallins bind to Bax and Bcl-XS
to sequester their translocation during staurosporine-
induced apoptosis. Cell Death Differ 11, 512–526.
40 Concannon CG, Gorman AM & Samali A (2003) On
the role of Hsp27 in regulating apoptosis. Apoptosis 8,
61–70.
41 MacRae TH (2003) Molecular chaperones, stress resist-
ance and development in Artemia franciscana. Semin
Cell Dev Biol 14, 251–258.
42 Clegg JS, Willsie JK & Jackson SA (1999) Adaptive sig-
nificance of a small heat shock ⁄ a-crystallin protein
(p26) in encysted embryos of the brine shrimp, Artemia
franciscana. Am Zool 39, 836–847.
43 Crack JA, Mansour M, Sun Y & MacRae TH
(2002) Functional analysis of a small heat shock ⁄
a-crystallin protein from Artemia franciscana.
Oligomerization and thermotolerance Eur J Biochem
269, 933–942.
44 Liang P & MacRae TH (1999) The synthesis of a small
heat shock ⁄ a-crystallin protein in Artemia and its rela-
tionship to stress tolerance during development. Dev
Biol 207, 445–456.
45 Liang P, Amons R, MacRae TH & Clegg JS (1997)
Purification, structure and in vitro molecular-chaperone
activity of Artemia p26, a small heat-shock ⁄ a-crystallin
protein. Eur J Biochem 243, 225–232.

52 Cobb BA & Petrash JM (2000) Structural and func-
tional changes in the aA-crystallin R116C mutant in
hereditary cataracts. Biochemistry 39, 15791–15798.
53 Andley UP, Patel HC & Xi J-H (2002) The R116C
mutation in a A-crystallin diminishes its protective abil-
ity against stress-induced lens epithelial cell apoptosis.
J Biol Chem 277, 10178–10186.
54 Cha
´
vez Zobel AT, Lambert H, The
´
riault JR & Landry
J (2005) Structural instability caused by a mutation at a
conserved arginine in the a-crystallin domain of Chinese
hamster heat shock protein 27. Cell Stress Chaperones
10, 157–166.
55 Studer S & Narberhaus F (2000) Chaperone activity and
homo- and hetero-oligomer formation of bacterial small
heat shock proteins. J Biol Chem 275, 37212–37218.
56 Leroux MR, Melki R, Gordon B, Batelier G & Candido
EPM (1997) Structure–function studies on small heat
shock protein oligomeric assembly and interaction with
unfolded polypeptides. J Biol Chem 272, 24646–24656.
57 Valdez MM, Clark JI, Wu GJS & Muchowski PJ (2002)
Functional similarities between the small heat shock
proteins Mycobacterium tuberculosis HSP 16.3 and
human aB-crystallin. Eur J Biochem 269, 1806–1813.
58 Lindner RA, Kapur A, Mariani M, Titmuss SJ & Car-
ver JA (1998) Structural alterations of a-crystallin dur-
ing its chaperone action. Eur J Biochem 258, 170–183.

Proc Natl Acad Sci USA 101, 10132–10136.
65 Willsie JK & Clegg JS (2001) Nuclear p26, a small heat
shock ⁄ a-crystallin protein, and its relationship to stress
resistance in Artemia franciscana embryos. J Exp Biol
204, 2339–2350.
66 van de Klundert FAJM & de Jong WW (1999) The
small heat shock proteins Hsp20 and aB-crystallin in
cultured cardiac myocytes: differences in cellular local-
ization and solubilization after heat stress. Eur J Cell
Biol 78, 567–572.
67 van den IJssel P, Wheelock R, Prescott A, Russell P &
Quinlan RA (2003) Nuclear speckle localisation of the
small heat shock protein aB-crystallin and its inhibition
by the R120G cardiomyopathy-linked mutation. Exp
Cell Res 287, 249–261.
68 Ito H, Kamei K, Iwamoto I, Inaguma Y, Tsuzuki M,
Kishikawa M, Shimada A, Hosokawa M & Kato K
(2003) Hsp27 suppresses the formation of inclusion
bodies induced by expression of R120G aB-crystallin, a
cause of desmin-related myopathy. Cell Mol Life Sci 60,
1217–1223.
69 Siddique M, Port M, Tripp J, Weber C, Zielinski D,
Calligaris R, Winkelhaus S & Scharf K-D (2003)
Tomato heat stress protein Hsp16.1-CIII represents a
member of a new class of nucleocytoplasmic small heat
stress proteins in plants. Cell Stress Chaperones 8, 381–
394.
70 Rajaraman K, Raman B, Ramakrishna T & Rao CM
(2001) Interaction of human recombinant aA- and
aB-crystallins with early and late unfolding intermedi-

This material is available as part of the online article
from
a-crystallin domain of p26 Y. Sun et al.
1034 FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS