Báo cáo khoa học: Structural and functional roles for b-strand 7 in the a-crystallin domain of p26, a polydisperse small heat shock protein from Artemia franciscana pdf

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,

termed diapause. Encystment is marked by the synthesis of p26, a polydis-
perse small heat shock protein thought to protect embryos from stress. In
order to elucidate structural ⁄ functional relationships within p26 and other
polydisperse small heat shock proteins, and to better define the protein’s
role during diapause, amino acid substitutions R110G, F112R, R114A and
Y116D were generated within the p26 a-crystallin domain by site-directed
mutagenesis. These residues were chosen because they are highly conserved
across species boundaries, and molecular modelling indicates that they are
part of a key structural interface between dimers. The F112R mutation,
which had the greatest impact on oligomerization, placed two charged resi-
dues at the p26 dimer–dimer interface, demonstrating the importance of
b-strand 7 in tetramer formation. All mutated versions of p26 were less
able than wild-type p26 to confer thermotolerance on transformed bacteria
and they exhibited diminished chaperone action in three in vitro assays;
however, all variants retained protective activity. This apparent stability of
p26 may, by prolonging effective chaperone life in vivo, enhance embryo
stress resistance. All substitutions modified p26 intrinsic fluorescence, sur-
face hydrophobicity and secondary structure, and the pronounced changes
in variant R114A, as indicated by these physical measurements, correlated
with the greatest loss of function. Although mutation R114A had the
greatest effect on p26 chaperoning, it had the least on oligomerization.
These results demonstrate that in contrast to many other small heat shock
proteins, p26 effectiveness as a chaperone is independent of oligomeriza-
tion. The results also reinforce the idea, occasioned by modelling, that
R114 is removed slightly from dimer–dimer interfaces. Moreover, b-strand
7 is shown to have an important role in oligomerization of p26, a function
first proposed for this structural element upon crystallization of wheat
Hsp16.9, a small heat shock protein with different quaternary structure.
Abbreviations
ANS, 1-anilino-8-naphthalene-sulphonate; sHSP, small heat shock protein.

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]

and development [7,41], thereby playing important
roles in cell activities.
Artemia females release offspring as swimming lar-
vae (ovoviviparous development) or encysted gastrulae
(oviparous development), termed cysts. The cysts enter
diapause, a resting stage where metabolic activity is
extremely low, even under favourable conditions [41].
Activation of encysted embryos by desiccation pre-
cedes reinitiation of development in the presence of
appropriate hydration, temperature and aeration. Arte-
mia cysts are exceptionally resistant to harsh condi-
tions, and when fully hydrated, either during diapause
or in a postdiapause state of metabolic arrest, termed
quiescence, they survive for several years without oxy-
gen. This is arguably the ultimate indifference to
anoxia of any metazoan [42], and qualifies the organ-
ism, as do other of its characteristics, as an extremo-
phile. Because activated Artemia embryos resume
development immediately upon return to favourable
circumstances, macromolecular components must be
preserved in the presence of limiting ATP, an activity
within sHSP functional capability. Just such a protein,
named p26, is synthesized in large quantities by ovi-
parous, but not ovoviviparous, embryos [17,41,43–46].
The p26 a-crystallin domain is similar in sequence to
this region in other sHSPs, including wheat Hsp16.9;
the protein confers thermotolerance on transformed
Escherichia coli and it has chaperone activity in vitro.
The objectives of the work described here are to
reveal structural and functional characteristics of poly-

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.

p26 synthesized in bacteria possessed an N-terminal
His-tag and an additional short N-terminal peptide
(PRAAGIRHELVLK) encoded by the clone used for
site-directed mutagenesis, but comparisons throughout
the study to p26 from Artemia and transfected mam-
malian cells lacking these residues indicated that they
had almost no effect on structure and function. Cell-
free extracts prepared from transformed bacteria
induced with anhydrotetracycline (aTc) exhibited
lightly stained bands, corresponding in size to p26
when electrophoresed in SDS polyacrylamide gels and
stained with Coomassie blue, and these polypeptides
reacted with anti-p26 immunoglobulin (Fig. 2A,B).
Expression levels in bacteria were very similar for all
Y. Sun et al. a-crystallin domain of p26
FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS 1021
variants and there was no indication of protein degra-
dation. After purification on TALON
tm
affinity col-
umns, a major polypeptide of the expected size
recognized by anti-p26 immunoglobulin was obtained
for each variant (Fig. 2C,D).
p26 in COS-1 cells
COS-1 cells were transiently transfected with p26
cDNAs cloned in the eukaryotic expression vector,
pcDNA4 ⁄ TO ⁄ myc-His.A, and p26 synthesis was veri-
fied by immunofluorescent staining and confocal laser-
scanning microscopy (Fig. 3). Wild-type (WT) p26 was
localized predominantly, if not exclusively, in the cyto-

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.

Fig. 2. Purification of p26 synthesized in Escherichia coli BL21PRO.
Cell-free extracts from transformed E. coli BL21PRO induced with
anhydrotetracycline (aTc) were electrophoresed through SDS poly-
acrylamide gels and either stained with Coomassie blue (A) or blot-
ted to nitrocellulose and reacted with antibody to p26 (B). Proteins
purified by affinity chromatography were electrophoresed through
SDS polyacrylamide gels and either stained with Coomassie blue
(C), or blotted to nitrocellulose and reacted with antibody to p26
(D). Lane 1, R110G; lane 2, F112R; lane 3, R114A; lane 4, Y116D;
lane 5, wild-type (WT) p26; lane 6, vector lacking p26 cDNA. Lanes
in panels A and B received 4.5–5.5 lg of protein and lanes in pan-
els C and D received 1 lg of protein. Arrow, p26. M, molecular
mass markers of 97, 66, 45, 31, 21 and 14 kDa.
a-crystallin domain of p26 Y. Sun et al.
1022 FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS
p26 oligomerization
WT p26 produced the largest oligomers, while, for
modified proteins, oligomer size was greatest for
R114A and smallest for F112R (Fig. 4; Table 1). The
molecular mass of p26 variants synthesized in bacteria
was unaffected by purification (Fig. 4A,B), indicating
that the methods employed had little effect on protein
structure, an important observation in relation to ana-
lysis of chaperone function. Except for WT p26, the
maximum monomer number per oligomer was higher
for p26 synthesized in COS-1 cells than in bacteria,
but the variation, although observed consistently, was
minor (Fig. 4, Table 1), indicating little difference
between the proteins from either source. Of equal sig-
nificance, the F112R substitution greatly reduced p26

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

progressively less effective in protecting citrate
Fig. 3. p26 synthesis in COS-1 cells. COS-1 cells transiently trans-
fected with the p26 cDNA-containing vector pcDNA4/TO/myc-His.A
were incubated with antibody to p26 followed by fluorescein iso-
thiocyanate-conjugated goat anti-rabbit IgG (green). Nuclei were
stained with propidium iodide (red). The scale bar represents
100 lm and both figures are the same magnification.
A
B
C
Fig. 4. p26 oligomer formation. Bacterially produced p26 either
before (A) or after (B) purification, and p26 synthesized in transfect-
ed COS-1 cells (C), were centrifuged at 200 000 g for 12 h at 4 °C
in 10–50% (w ⁄ v) continuous sucrose gradients. The gradients were
fractionated and 15-lL samples from each fraction were electro-
phoresed in SDS polyacrylamide gels, blotted to nitrocellulose and
reacted with antibody to p26 followed by horseradish peroxidase
(HRP)-conjugated goat anti-rabbit IgG. The top of each gradient is
to the right and fractions are numbered across the top. The posi-
tions of the 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, are indi-
cated by labeled arrows.
Y. Sun et al. a-crystallin domain of p26
FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS 1023
synthase, with the latter two not significantly differ-
ent from one another. However, the modified p26
variants exhibited appreciable chaperone activity, and
at 1200 nm the variants were almost as good as WT
p26 (Supplementary Fig. 1). R114A provided the

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

spectra, whereas R114A possessed the lowest fluores-
cence and the emission was red-shifted. The results
indicate altered microenvironments for aromatic amino
acid residues, such as tryptophan, of which two reside
in the p26 N-terminal region at positions 6 and 17,
with the greatest change caused by the R114A modi-
fication. All mutated versions of p26 exhibited less
1-anilino-8-naphthalene-sulphonate (ANS)-binding than
WT p26, an indication of reduced surface hydropho-
bicity (Fig. 6B), with R114A at the lowest level. The
enhancement of surface hydrophobicity by increasing
the temperature from 25 °Cto43°C was reduced for
the p26 variants in comparison to WT p26 (Fig. 6B).
Additionally, as shown by far-UV CD, the spectra for
p26 a-crystallin domain mutants possessed wider, more
negative, shoulders ranging from 208 to 230 nm, and
with one exception, a related positive shoulder peaking
near 194 nm (Fig. 7A,B). R114A exhibited the greatest
variation from WT in CD spectra, this reflecting
decreased b-structure content and an increase in a-heli-
cal constituents (Table 2).
WT p26 from E. coli and Artemia gave comparable
fluorescence intensities in ANS-binding experiments at
25 °C and 43 °C (Fig. 6B). The far-UV CD spectra of
p26 from both sources were characteristic of b-sheet
enrichment, with a negative shoulder near 214 nm and
a positive shoulder near 194 nm (Fig. 7A). The only
indication of a difference was the slight red-shifted
intrinsic fluorescence of bacterial WT p26; however,
the fluorescence intensities for p26 from bacteria and

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

mass (kDa)
Oligomer mass
range (kDa)
Maximum
monomer
number
E. coli R110G 25.4 29–443 17
F112R 25.5 29–150 6
R114A 25.4 29–500 20
Y116D 25.5 29–443 17
WT 25.5 29–669 26
COS-1 R110G 20.7 14.2–443 21
cells F112R 20.8 14.2–150 7
R114A 20.7 14.2–500 24
Y116D 20.8 14.2–443 21
WT 20.8 14.2–500 24
a-crystallin domain of p26 Y. Sun et al.
1024 FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS
C-terminal extensions which extend to neighboring
monomers, and the remaining b-strands, with the
exception of strand 6, are arranged in two antiparallel
beta sheets within the a-crystallin domain. The inter-
face between monomers of a dimer involves interaction
between strands b2 and b6 of neighboring monomers.
The p26 modifications examined in this study are not
located in either of these strands and they are not con-
sidered further.
As the basic p26 oligomer building block, dimers
interact to form tetramers, the next level of struc-
ture. Modeling indicates that tetramer formation

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

360 nm. The A
360
values were multiplied by 1000 for construction of the curves. The standard error ranged from 4.2 to 7.0%. (D) Citrate
synthase at 150 n
M was heated at 43 °C for 1 h in either the absence or presence of p26, and then enzyme activity was determined. p26
concentrations are indicated and the p26 variants are in the same order in each histogram group. The standard error ranged from 3.3 to
10.0%.
Y. Sun et al. a-crystallin domain of p26
FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS 1025
residues had little effect on oligomerization, although
the R114A substitution reduced chaperoning to the
greatest extent.
Discussion
p26, an abundantly expressed, polydisperse sHSP
thought to protect encysted Artemia embryos against
physiological stress, was investigated by site-directed
mutagenesis of a-crystallin domain residues and
molecular modeling of protein structure. The p26
a-crystallin domain contains nine b-strands arranged
predominantly as paired b-sheets and possesses resi-
dues conserved in many other sHSPs, including those
in the sequence 110REFRRRY116, where substitu-
tions were generated. Examination of mutations within
the selected sequence indicated that b-strand 7 is
involved in dimer–dimer interactions, leading to
higher-order oligomer structure. In addition, it was
concluded that p26 structural characteristics would
promote Artemia survival during encystment, diapause
and stress exposure.
Fig. 7. Secondary structure of p26. Far-UV CD spectra were

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

4
, pH 7.1, was determined by oversaturation with 1-anilino-8-naphthalene-sulphonate (ANS). Fluorescence was measured at an exci-
tation wavelength of 388 nm and band pass of 8 nm, with emission wavelength at 473 nm and band pass of 8 nm. Measurements were
made at either 25 °C (grey) or 43 °C (black). Fluorescence generated by buffer containing ANS, but no p26, was subtracted. The standard
error ranged from 6.7 to 10%.
Table 2. Secondary structure elements of p26. The secondary ele-
ment percentages were calculated using the
CDNN v2.1 deconvolu-
tion program for each p26 variant generated by site-directed
mutagenesis and for purified wild-type (WT) p26 from transformed
Escherichia coli and Artemia embryos.
Structural
elements
R110G
(%)
F112R
(%)
R114A
(%)
Y116D
(%)
WT
(E. coli)
(%)
WT
(Artemia)
(%)
a-helix 20.3 20.2 25.7 19.3 17.7 17.7
b-antiparallel 18.0 18.2 12.9 20.5 21.6 23.0
b-parallel 10.2 10.2 9.7 10.1 10.3 10.0

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

sucrose density gradient centrifugation, from 669 kDa,
for WT p26, to150 kDa for the F112R variant.
Replacement of Y116 with negatively charged aspar-
tic acid had limited effects on oligomerization, prob-
ably as a result of the residue’s location at the edge of
the dimer–dimer interface. The maximum oligomer size
obtained with p26 R114A was  500 kDa, closer to
the mass of the WT oligomer than the other variants.
This compares to oligomers of 2–4 MDa for mutation
R116C of aA-crystallin and 0.7–2 MDa and larger for
R120G aB-crystallin [48–53], both significant increases
in mass when compared with oligomers of WT a-crys-
tallins. Modification of R114 in p26 obviously has less
effect on oligomerization than equivalent substitutions
in aA and aB-crystallin. In agreement with the limited
effect on oligomer mass and the proposed importance
Fig. 8. Structural model of a p26 tetramer. (A) Sequence alignment of amino acid residues 59–158 of Artemia p26 (AAB87967) and residues
45–151 of wheat, Triticum aestivum, Hsp16.9 (1GME) used to generate the 3D structural model of the p26 tetramer. The proteins share
25.9% sequence identity and overall similarity of 69.4% in the regions compared. The boxed residues labeled b2–b10 indicate the Hsp16.9
b-strand positions [14] and the corresponding residues in p26. Residues highlighted in yellow were modified in p26 by site-directed mutagen-
esis. Residue numbers are given on the right. (B) A structural model of the p26 tetramer generated by comparison to wheat Hsp16.9 is rep-
resented in a ribbon diagram. Mutated residues Arg110, Phe112, Arg114 and Tyr116 are shown in gray in ball-and-stick and are labeled
along the dimer–dimer interface by using the three-letter amino acid code in the color of the parent monomer. Monomers A (green) and B
(yellow) form dimer 1, while monomers C (red) and D (blue) form dimer 2. L5 ⁄ 7, the loop between b-strands 5 and 7 which contains
b-strand 6; N term, amino terminus of a p26 monomer; C term, carboxy terminus of a p26 monomer; the b-strands 2–10 are labeled in
monomer A.
Y. Sun et al. a-crystallin domain of p26
FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS 1027
of tetramer formation in higher-order structure, the
p26 model predicts R114 to be positioned slightly out-

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

chaperone activity therefore approximates that of
other sHSPs and this, in concert with its abundance,
provides a large capacity for storage of partially dena-
tured proteins in oviparous Artemia embryos during
diapause and quiescence. Upon return of embryos to
permissive conditions, proteins would be released from
p26 and renatured, permitting rapid resumption of
metabolism, cell growth and development, an advant-
age to the organism under most circumstances.
In contrast to a marginal impact on oligomerization,
substitution R114A had the greatest detrimental effect
on p26 chaperone activity in all assays. R114 is prob-
ably buried within the a-crystallin domain, stabilized
by a salt bridge with another charged residue(s) [59].
The R114A substitution would destroy ionic linkages
and expose negatively charged residues within mono-
mer interiors, with ensuing structural changes and
reduced chaperone activity. In agreement with this
idea, modified intrinsic fluorescence spectra and sur-
face hydrophobicity – the latter an effector of sHSP
chaperone activity [60] – indicate that p26 structural
changes are greater for R114A than for other muta-
tions. Additionally, far-UV CD measurements showing
decreased b-structure were most prominent for R114A,
with similar observations reported for R120G in
aB-crystallin [48–50] and R116C aA-crystallin at 37 °C
but not 25 °C [51]. Mutation R114A had the least
effect on p26 oligomerization but the greatest conse-
quence for function, demonstrating that chaperoning is
independent of oligomerization.

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-

tion disrupts aB-crystallin speckle localization, with lit-
tle of the modified protein entering nuclei [67], and
there is a tendency for R120G aB-crystallin to form
inclusion bodies in the cytosol [68], but this was not
observed with R114A p26. Human R116C aA-crystal-
lin occurs mainly in the cytoplasm of epithelial cells
[53]. How p26 enters nuclei is unknown, as is true for
most, but not all, sHSPs [69]. Oligomers of p26 R114A
enter all COS-1 nuclei in which the protein is expressed
and they are equivalent in mass to WT oligomers,
which, in contrast to Hsp27 and a-crystallin [67],
reside only in the cytoplasm of unstressed cells. How-
ever, the much smaller F112R oligomers exhibit
reduced translocation efficiency and they are not found
a-crystallin domain of p26 Y. Sun et al.
1028 FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS
in the nuclei of all transfected COS-1 cells synthesizing
this p26 variant; these results are in agreement with
earlier work showing that p26 oligomers, reduced in
size by C-terminal truncation, remain in the cytoplasm
[17]. p26 nuclear migration is apparently not accom-
plished by simple diffusion across the membrane upon
oligomer size reduction, and why a-crystallin domain
modifications promote translocation remains uncer-
tain.
To summarize, analysis of individual amino acid
substitutions, coupled with molecular modeling of
protein structure, indicate that b-strand 7 of the
a-crystallin domain is an integral component of the
p26 dimer–dimer interface in polydisperse sHSPs. p26

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 ⁄

myc-His.A (Invitrogen, San Diego, CA, USA) and transfor-
mation of E. coli DH5a (Invitrogen, Carlsbad, CA, USA).
The p26 cDNAs were also cloned into pPROTet.E233
(Clontech Laboratories, Inc., Palo Alto, CA, USA), a His-
tag-containing prokaryotic expression vector, using the
BamHI and XbaI restriction sites. Polypeptides encoded by
pPROTet.E233 were longer than those encoded by
pcDNA4 ⁄ TO ⁄ myc-His.A because the former employed a
start codon upstream of the His-tag, while the latter initi-
ated translation from the p26 start codon. All p26 cDNA
inserts were sequenced (DNA Sequencing Facility, Center
for Applied Genomics, Hospital for Sick Children,
Toronto, ON, Canada).
Bacterial synthesis and purification of p26
p26 was synthesized in transformed E. coli BL21PRO
(Clontech Laboratories, Inc., Mississauga, ON, Canada)
induced with 100 ngÆmL
)1
anhydrotetracycline (aTc) (Clon-
tech Laboratories). p26 was recovered from bacterial
extracts using BD TALON resin (BD Biosciences Clontech,
Mississauga, ON, Canada) and concentrated in Centrip-
repYM-10 centrifugal filter devices (Amicon Bioseparations,
Billerica, MA, USA) [17]. Protein samples were electro-
phoresed in 12.5% SDS polyacrylamide gels and either
stained with Coomassie Brilliant Blue R-250 (Sigma) or
blotted onto nitrocellulose (Bio-Rad, Hercules, CA, USA)
for reaction with anti-p26 immunoglobulin [46] and Omni-
probe (Santa Cruz Biotechnology, Inc., Santa Cruz, CA,
USA), a monoclonal antibody recognizing the (His)

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

in lysis buffer consisting of 50 mm Tris ⁄ HCl, pH 7.8,
Table 3. Primers for site-directed mutagenesis of p26. Single
amino acid substitutions were generated within the p26 a-crystallin
domain by site-directed mutagenesis using primers presented as
sense and antisense, respectively, for each mutation.
p26
mutation Primer
R110G 5¢-GGACACGTACAAGGAGAATTTCGACGACG-3¢
5¢-CGTCGTCGAAATTCTCCTTGTACGTGTCC-3¢
F112R 5¢-CACGTACAAAGAGAACGTCGACGACG-3¢
5¢-CGTCGTCGACGTTCTCTTTGTACGTG-3¢
R114A 5¢-GAGAATTTCGAGCACGATACAGACTCCC-3¢
5¢-GGGAGTCTGTATCGTGCTCGAAATTCTC-3¢
Y116D 5¢-CGACGACGAGACAGACTCCCAGAACATGTC-3¢
5¢-GACATGTTCTGGGAGTCTGTCTCGTCGTCG-3¢
Y. Sun et al. a-crystallin domain of p26
FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS 1029
150 mm NaCl, 0.5% Nonidet P-40, 1 mm phenyl-
methanesulfonyl fluoride (Sigma, Oakville, ON, Canada),
1 lgÆmL
)1
pepstatin A (Sigma) and 1 lgÆmL
)1
leupeptin
(Sigma). Lysates were centrifuged at 10 000 g for 10 min,
supernatants were transferred to fresh tubes and protein
concentrations were determined using the Bradford assay
(Bio-Rad). p26 was detected on western blots, as
described above, and localized in transfected COS-1 cells
by staining with anti-p26 immunoglobulin and propidium

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-

nificance of difference, which was accepted at a P-value
of < 0.05. The presence of p26 in bacteria that were
stressed was confirmed by immunoprobing of western blots
containing protein extracts obtained from aTc-induced cells
prior to heating at 54 °C.
p26 chaperone activity in vitro
Citrate synthase protection against heat-induced aggrega-
tion was determined [17], as was the ability of p26 to shield
citrate synthase enzyme activity at 43 °C. Reaction mixtures
for measuring citrate synthase activity contained 940 lL
of TE (50 mm Tris ⁄ HCl, pH 7.5, 2 mm EDTA), 10 lLof
10 mm oxaloacetic acid (Sigma), 10 lLof10mm 5,5¢-di-
thiobis (2-nitrobenzoic acid) (Sigma) and 30 lLof5mm
acetyl-CoA (Sigma) [70]. Reactions initiated by adding
10 lL of 150 nm citrate synthase were monitored at 25 °C
as an increase in absorption at 412 nm determined with
a SPECTRAmax PLUS spectrophotometer (Molecular
Devices, Sunnyvale, CA, USA). Insulin (Sigma) at a final
concentration of 4.0 lm in 10 mm phosphate buffer,
pH 7.4, was mixed with p26, dithiothreitol (Sigma) was
added to 20 mm and solution turbidity was measured at
400 nm in a SPECTRAmax PLUS spectrophotometer at
25 °C. Assays were performed in triplicate.
p26 intrinsic fluorescence, ANS binding
and secondary structure
Purified p26 was diluted to 0.06 mgÆmL
)1
in 10 mm
NaH
2

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

)1
p26 in 10 mm NaH
2
PO
4
,
pH 7.1, was employed, and three scans were averaged for
each spectrum. Bandwidth was 2 nm, with all scans correc-
ted for buffer and smoothed to eliminate background noise.
Secondary structure parameters were calculated using the
cdnn v2.1 deconvolution program (Martin-Luther-Univer-
sita
¨
t, Halle-Wittenberg, Germany).
Modeling of p26 structure
The a-crystallin domains of p26, aA ⁄ aB-crystallin, Hsp27
and Hsp16.9 were aligned by clustal w followed by man-
ual adjustment, according to secondary structure elements
predicted with psipred [71]. The Hsp16.9 tetramer, com-
posed of subunits 1gme_1:A, 1gme_1:B, 1gme_1:G and
1gme_1:H, was obtained from the protein quaternary struc-
a-crystallin domain of p26 Y. Sun et al.
1030 FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS
ture file server PQS at EMBL-EBI where the crystal struc-
ture coordinates of Hsp16.9 (accession number: 1GME)
were used to reconstruct the Hsp16.9 oligomer by crystallo-
graphic symmetry transformations. The p26 a-crystallin
domain models were constructed using modeler [72] and
the structure was obtained by optimizing the probability
objective functions (pdfs) and simulated annealing minimi-

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

1 Frydman J (2001) Folding of newly translated proteins
in vivo: the role of molecular chaperones. Annu Rev
Biochem 70, 603–647.
2 Hartl FU & Hayer-Hartl M (2002) Molecular chaper-
ones in the cytosol: from nascent chain to folded pro-
tein. Science 295, 1852–1858.
3 Haslbeck M (2002) sHsps and their role in the chaper-
one network. Cell Mol Life Sci 59, 1649–1657.
4 Craig EA (2003) Eukaryotic chaperonins: Lubricating
the folding of WD-repeat proteins. Curr Biol 13, R904–
R905.
5 Mogk A & Bukau B (2004) Molecular chaperones:
Structure of a protein disaggregase. Curr Biol 14, R78–
R80.
6 Petrucelli L, Dickson D, Kehoe K, Taylor J, Snyder H,
Grover A, De Lucia M, McGowan E, Lewis J, Prihar G
et al. (2004) CHIP and Hsp70 regulate tau ubiquiti-
nation, degradation and aggregation. Hum Mol Genet
13, 703–714.
7 MacRae TH (2000) Structure and function of small heat
shock ⁄ a-crystallin proteins: established concepts and
emerging ideas. Cell Mol Life Sci 57, 899–913.
8 Scharf K-D, Siddique M & Vierling E (2001) The
expanding family of Arabidopsis thaliana small heat
stress proteins and a new family of proteins containing
a-crystallin domains (Acd proteins). Cell Stress Chaper-
ones 6, 225–237.
9 Narberhaus F (2002) a-crystallin-type heat shock pro-
teins: socializing minichaperones in the context of a
multichaperone network. Microbiol Mol Biol Rev 66,

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:

the N-terminal domain is important for oligomer assem-
bly and the binding of unfolding proteins. J Biol Chem
279, 11222–11228.
19 Wintrode PL, Friedrich KL, Vierling E, Smith JB &
Smith DL (2003) Solution structure and dynamics of a
heat shock protein assembly probed by hydrogen
exchange and mass spectrometry. Biochemistry 42,
10667–10673.
20 Salerno JC, Eifert CL, Salerno KM & Koretz JF (2003)
Structural diversity in the small heat shock protein
Y. Sun et al. a-crystallin domain of p26
FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS 1031
superfamily: control of aggregation by the N-terminal
region. Prot Eng 16, 847–851.
21 Studer S, Obrist M, Lentze N & Narberhaus F (2002)
A critical motif for oligomerization and chaperone
activity of bacterial a-heat shock proteins. Eur J Bio-
chem 269, 3578–3586.
22 Fu X, Zhang H, Zhang X, Cao Y, Jiao W, Liu C, Song
Y, Abulimiti A & Chang Z (2005) A dual role for the
N-terminal region of Mycobacterium tuberculosis
Hsp16.3 in self-oligomerization and binding denaturing
substrate proteins. J Biol Chem 280, 6337–6348.
23 Hasan A, Yu J, Smith DL & Smith JB (2004) Thermal
stability of human a-crystallins sensed by amide hydro-
gen exchange. Prot Sci 13, 332–341.
24 Thampi P & Abraham EC (2003) Influence of the
C-terminal residues on oligomerization of aA-crystallin.
Biochemistry 42, 11857–11863.
25 Lindner RA, Carver JA, Ehrnsperger M, Buchner J,

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

Biol 350, 1083–1093.
32 Basha E, Lee GJ, Breci LA, Hausrath AC, Buan NR,
Giese KC & Vierling E (2004) The identity of proteins
associated with a small heat shock protein during heat
stress in vivo indicates that these chaperones protect a
wide range of cellular functions. J Biol Chem 279, 7566–
7575.
33 Friedrich KL, Giese KC, Buan NR & Vierling E (2004)
Interactions between small heat shock protein subunits
and substrate in small heat shock protein-substrate com-
plexes. J Biol Chem 279, 1080–1089.
34 Duverger O, Paslaru L & Morange M (2004) HSP25 is
involved in two steps of the differentiation of PAM212
keratinocytes. J Biol Chem 279, 10252–10260.
35 Day RM, Gupta JS & MacRae TH (2003) A small heat
shock ⁄ acrystallin protein from encysted Artemia
embryos suppresses tubulin denaturation. Cell Stress
Chaperones 8, 183–193.
36 Panasenko OO, Kim MV, Marston SB & Gusev NB
(2003) Interaction of the small heat shock protein with
molecular mass 25 kDa (hsp25) with actin. Eur J Bio-
chem 270, 892–901.
37 Arrigo A-P, Firdaus WJ, Mellier G, Moulin M, Paul C,
Diaz-Iatoud C & Kretz-remy C (2005) Cytotoxic effects
induced by oxidative stress in cultured mammalian cells
and protection provided by Hsp27 expression. Methods
35, 126–138.
38 Kamradt MC, Lu M, Werner ME, Kwan T, Chen F,
Strohecker A, Oshita S, Wilkinson JC, Yu C, Oliver PG
et al. (2005) The small heat shock protein aB-crystallin

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.

46 Liang P, Amons R, Clegg JS & MacRae TH (1997)
Molecular characterization of a small heat shock ⁄
a-crystallin protein in encysted Artemia embryos. J Biol
Chem 272, 19051–19058.
a-crystallin domain of p26 Y. Sun et al.
1032 FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS
47 Guruprasad K & Kumari K (2003) Three-dimensional
models corresponding to the C-terminal domain of
human aA- and aB-crystallins based on the crystal
structure of the small heat-shock protein HSP16.9 from
wheat. Int J Biol Macromol 33, 107–112.
48 Bova MP, Yaron O, Huang Q, Ding L, Haley DA,
Stewart PL & Horwitz J (1999) Mutation R120G in aB-
crystallin, which is linked to a desmin-related myopathy,
results in an irregular structure and defective chaperone-
like function. Proc Natl Acad Sci USA 96, 6137–6142.
49 Kumar LVS, Ramakrishna T & Rao ChM (1999) Struc-
tural and functional consequences of the mutation of a
conserved arginine residue in aA and aB crystallins.
J Biol Chem 274, 24137–24141.
50 Perng MD, Muchowski PJ, van den IJssel P, Wu GJS,
Hutcheson AM, Clark JI & Quinlan RA (1999) The
cardiomyopathy and lens cataract mutation in aB-crys-
tallin alters its protein structure, chaperone activity, and
interaction with intermediate filaments in vitro. J Biol
Chem 274, 33235–33243.
51 Shroff NP, Cherian-Shaw M, Bera S & Abraham EC
(2000) Mutation of R116C results in highly oligomer-
ized aA-crystallin with modified structure and defective
chaperone-like function. Biochemistry 39, 1420–1426.

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.

59 Berengian AR, Bova MP & Mchaourab HS (1997) Struc-
ture and function of the conserved domain in aA-crystal-
lin. Site-directed spin labeling identifies a a-strand located
near a subunit interface. Biochemistry 36, 9951–9957.
60 Reddy GB, Das KP, Petrash JM & Surewicz WK
(2000) Temperature-dependent chaperone activity and
structural properties of human aA- and aB-crystallins.
J Biol Chem 275, 4565–4570.
61 Zobel ATC, Loranger A, Marceau N, The
´
riault JR,
Lambert H & Landry J (2003) Distinct chaperone
mechanisms can delay the formation of aggresomes by
the myopathy-causing R120G aB-crystallin mutant.
Hum Mol Genet 12, 1609–1620.
62 Litt M, Kramer P, LaMorticella DM, Murphey W,
Lovrien EW & Weleber RG (1998) Autosomal domi-
nant congenital cataract associated with a missense
mutation in the human alpha crystallin gene CRYAA.
Hum Mol Genet 7, 471–474.
63 Vicart P, Caron A, Guicheney P, Li Z, Prevost M-C,
Faure A, Chateau D, Chapon F, Tome
´
F, Dupret J-M
et al. (1998) A missense mutation in the aB-crystallin
chaperone gene causes a desmin-related myopathy. Nat
Genet 20, 92–95.
64 Sanbe A, Osinska H, Saffitz JE, Glabe CG, Kayed R,
Maloyan A & Robbins J (2004) Desmin-related cardio-
myopathy in transgenic mice: a cardiac amyloidosis.

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-

ates of citrate synthase on its thermal denaturation.
FEBS Lett 497, 118–123.
Y. Sun et al. a-crystallin domain of p26
FEBS Journal 273 (2006) 1020–1034 ª 2006 The Authors Journal compilation ª 2006 FEBS 1033
71 Jones DT (1999) Protein secondary structure prediction
based on position-specific scoring matrices. J Mol Biol
292, 195–202.
72 Sali A & Blundell TL (1993) Comparative modeling by
satisfaction of spatial restraints. J Mol Biol 234, 779–815.
73 Eisenberg D, Lu
¨
thy R & Bowie JU (1997) VERIFY3D:
assessment of protein models with three-dimensional
profiles. Methods Enzymol 277, 396–404.
74 Luthy R, Bowie JU & Eisenberg D (1992) Assessment
of protein models with three-dimensional profiles.
Nature 356, 83–85.
75 Zhu J & Weng Z (2005) FAST: a novel protein struc-
ture alignment algorithm. Proteins 58, 618–617.
76 Laskowski RA, MacArthur MW, Moss DS & Thornton
JM (1993) PROCHECK: a program to check the stereo-
chemical quality of protein structures. J Appl Cryst 26,
283–291.
77 Humphrey W, Dalke A & Schulten K (1996) VMD –
Visual Molecular Dynamics. J Mol Graph Model 14 ,
33–38.
Supplementary material
The following supplementary material is available
online:
Fig. S1. p26 protects citrate synthase against heat.

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