OPTOMETRY

I INVITED REVIEW I
a-crystallin: a review of its structure and function

Clin Ex# Optom 2004; 87: 6: 356-366

Robert C Augusteyn BSc PhD DipEd
Vision Cooperative Research Centre,
University of NSW, Sydney, Australia.
Neuroimmunology Laboratory,
Biochemistry Department, LaTrobe
University, Bundoora, Australia
Submitted: 29 March 2004
Revised: 23 June 2004
Accepted for publication: 18 July 2004
a-crystallin, the major protein of the mammalian lens in most species, is an aggregate
assembled from two polypeptides, each with a molecular weight around 20,000 Da. It is
polydisperse and can be isolated in a variety of forms, including spherical particles with
molecular weights ranging upwards from about 200 kDa.
Sequence comparisons reveal that it is a member of the small heat shock protein
(shsp) family. These proteins are aggregates assembled from polypeptides of 10 to 25
kDa that share a common central domain of about 90 residues (the ‘a-crystallin domain’)
with variable N- and Gterminal extensions.
a-crystallin has been intensively studied for more than 50 years but its three-dimensional
structure remains unknown because it has not been possible to obtain crystals for X-ray
studies and it is too large for NMR measurements. Structural information has been
derived from a variety of solution studies. Because of the protein’s polydispersity, inter-
pretation of data has been difficult. This led to different viewpoints and vigorous debate
on its structure and properties. Recently, the crystal structures of two closely-related
small heat shock proteins have been determined. These have provided some insight
into the structure of a-crystallin and explanations of previous observations.
Like many other heat shock proteins, a-crystallin exhibits chaperone-like properties,
including the ability to prevent the precipitation of denatured proteins and to increase
cellular tolerance to stress. It has been suggested that these functions are important for
the maintenance of lens transparency and the prevention of cataract.

Key words: a-crystallin, chaperone, crystalline lens, heat shock proteins, structure

The name crystallin was first coined by
Berzelius in 1830 to describe the gelati-
nous substances that could be isolated
from the crystalline lens.’ Fractionation of
this material led to the recognition in 1896
that there were several different constitu-
ents a n d these were proteins, t h e
crystallins, but studies on their structures
did not begin in earnest until the 1 9 6 0 ~ . ~ not understood. It appears that there is no
a-crystallin is the major protein of the
mammalian lens, accounting for about half
of the total protein in most species. It is usu-
ally isolated as a polydisperse mixture of
large aggregates assembled from two types
of polypeptides, the CrA and a B chains, with
molecular weights around 20 kDa. The p r e
portions of the two chains vary with cellular
differentiation, with age and with species
but the significance of this variability is
specific requirement for either polypeptide
in the formation of a stable protein.
a. My first exposure to a-crystallin came in 1965
when, as a young postgraduate student, I read a
paper on its dissociation. I thought it had a
strange name and assumed this indicated i t had
been crystallised, not yet having heard of the
crystalline lens. Three years later, Abraham
Spector, one of the paper’s authors, invited me
to join his laboratory and study a-crystallin. So
began a journey of almost 40 years, frustrating
at times hut always interesting and stimulating.
The journey is not yet complete. It has become
more difficult recently but my goal remains
clear-understanding the structure and func-
tions of a-crystallin.

Clinical and Experimental Optometry 87.6 November 2004

356

Traces of the a-crystallin polypeptides
are found in some other tissues but the
distribution of a B is ubiquitous.' It is
thought that production of a B is part of
the cell's response to stress and involves
its chaperone-like activity.
This review focuses on current knowl-
edge of the structure of a-crystallin and
its possible functions in the lens. For other
discussions on the properties and func-
tions of a-crystallin, as well as the other
crystallins, the reader is referred to reviews
by Horwitz,3 van Montford, Slingsby and
Vierlir~g,~Sax and Piatigorsky,' Augusteyn
and Stevens.ti
SIZE AND SHAPE
When isolated in t h e laboratory,
a-crystallins are usually polydisperse
populations of macromolecules large
enough to be seen with the electron mi-
croscope. Apparently spherical particles
with diameters of 12 to 15 nm are most
commonly A highly magnified
example is shown in Figure 1 (top).
Although there appear to be surface
features, areas where stain molecules con-
centrate, it has not been possible to deter-
mine the subunit arrangement. Smaller
particles (approximately 9.5 n m ) of
similar appearance are obtained with
proteins assembled from the purified
polypeptides.','" Sheet-like forms of the
protein are also observed."
The spherical particles assemble into
chain-like structures (Figure 1, bot-
tom) .8.'".12 These are fragile and fragment
easily but large numbers are seen if they
are stabilised by cross-linking with glutar-
I t is probable that the ob-
served polydispersity of the a-crystallin
populations is due to variations in the
length of these chains.I2 How these relate
to the structure of the protein in the
intact lens remains to be determined,
however, it is likely they are generated by
rearrangement of the in vivo form, as the
protein can be affected by conditions
used for disrupting lens cells. For exam-
ple, isolation of a-crystallin at five degrees
Centigrade yields macromolecules with
an average molecular weight of about 800
kDa. If a lens is disrupted at 37 degrees
Clinical and Experimental Optometry 117.6 November 2004
a-crystall in A ugustqrn
Centigrade, close to the temperature at
which it exists in the eye, the average is
320 kDa." Similar sizes are obtained with
proteins assembled from purified sub-
u n i t ~ , ' " .or
' ~by renaturation of denatured
protein.15 This suggests that the 800 kDa
species of a-crystallin is a n artefact
generated when the lens is cooled.
Determining the true molecular weight
of a-crystallin has not yet been achieved,
not only because of its polydispersity but
also because of the large variations in the
size of the isolated proteins. Average mo-
lecular weights for different preparations
range from 280 kDaI6 to in excess of
10 mDa." The reasons for the variability
are not yet fully understood, although, as
pointed out earlier, isolation conditions
can have an effect.
Some of the polydispersity is due to an
increase in the molecular weight of the
protein, as the lens grows larger. Figure 2
shows that the size of a-crystallin increases
towards the centre of the lens, in both
adult and foetal lenses. Because of the
unique growth pattern, in which n o cells
or their contents are lost during an ani-
mal's lifetime, a-crystallins isolated from
whole lenses will comprise proteins of
different sizes.
T h e increase in molecular weight is
usually attributed to age-related post-
translational modifications and conforma-
tional changes that make the protein more
hydrophobic.14 It is thought to be part of
pathological processes that culminate in
Figure 1. Electron micrographs of an
the insolubilisation of proteins and the
a-crystallin particle (top) and a chain of
formation of cataract in the old lens.
particles (below) magnified 10 to 20 x106
Both post-translational modifications
fold. The proteins were negatively stained
and molecular weight increases start dur-
with uranyl acetate. Dark areas correspond
ing prenatal lens growth. In the centre of
to stain deposited on and around protein.
the foetal lens, the molecular weight is al-
The bar on each micrograph is 10 nm.
ready more than double that in the out-
side (Figure 2). This could hardly be con-
sidered as pathological or age-related in a
tissue no more than six months old. More
likely, it is a normal event associated with
the compression of cells into the lens cen-
tre and plays a role in generation of the
refractive index gradient.
An increase in size reduces the surface
area/volume ratio of a protein and de-
creases the size of its hydration shell rela-
tive to its mass. Consequently, higher pro-
357
a-crystallin Augustqrn

residues, respectively. The amino acid

4000 1 sequences of bovine a.4-and dogfkh a&

-!
In
C
crystallin, separated according to the exon

/
3000
locations, are shown in Figure 4.For ease
B
Y of reference, the numbering includes
gaps. Thus, the end of exon 1 in OA is at
position 77, corresponding to 63 amino
acids and 14 gaps.

AMINO ACID SEQUENCE

OJ
Amino acid sequence comparisons of a
0 200 400 600 800
large number of OA and aB polypeptides
Protein accumulatedfrom the outside of the lens (mg)
from different species have indicated that
the two share a common ancestral gene
Figure 2. The molecular weight of u-crystallinsin concentric layers
and have evolved slowly, averaging only
removed from single prenatal (eight-month gestation, -.- and
.-.-)
three amino acid changes per 100 residues
adult (two years, -.- bovine
. - lenses.
. - Protein accumulated is a
)
(three per cent) every 100 million years.2n
measure of the distance into the lens, with 0 representing the outside
By contrast, haemoglobin has changed by
and 800 the centre in the adult. The relationship is not linear.
20 per cent in the same time. Compari-
son of the sequences of the modern bovine
a.4-and the ancient dogfish aB-crystallins
in Figure 4 shows the close relationship.
It would appear that there are structural or
functional constraints on the a-crystallin
Figure 3. The u-crystaUingene, showing the arrangement of exons molecule, which prohibit extensive amino
and introns ( ) acid changes. This is not surprising for a
protein that must interact with other lens
crystallins to form a transparent protein
matrix and with denatured proteins to
prevent their insolubilisation.
Comparison of the sequences with
tein concentrations and refractive index dissociation of subunit homopolyrners.l8 those of other proteins has revealed that
can be attained without an attendant It seems that the smallest a-crystallin mol- a-crystallins are members of the small heat
increase in osmotic pressure. This is pre- ecule may be under 200 kDa.I4sL8 A three- shock protein family.21This family com-
cisely what is required in the central areas dimensional structure would provide the prises 10 to 27 kDa polypeptides that share
of the lens. required information but, because of the a common central sequence of about 90
The variability in particles that can be polydispersity, it has not been possible to residues, called the a-crystallin domain.
isolated and in the molecular weights of crystallise the protein and the aggregate They differ in the variable N- and Ctermi-
different preparations, as well as the is too large for complete structural nal extensions and form aggregates of vari-
polydispersity,suggest that a-crystallins are determination using NMR. ous sizes. Examples are shown in Figure 4.
aggregates of a smaller molecule.
Several approaches have been used in
THE &CRYSTALLIN GENE DOMAIN STRUCTURE
attempts to isolate this molecule. Re-
peated fractionation of a-crystallin a-crystallin comprises two primary gene Exons often code for independent struc-
populations, to remove large aggregates, products, the OlA and aB polypeptides, tural units or domains in a protein.22Thus,
yielded species with average molecular each 170 to 180 amino acids long.4 a-crystallins may contain three domains.
weights of 280 kDa.I6 These were still They are not related to any of the other Hydropathy plots indicate that the amino
polydisperse, indicating that the mini- lens proteins (p-, 'y- or taxon specific- acid sequences corresponding to the exon
mum was lower, perhaps a dodecamer crystallins). boundaries are strongly hyd ro p h ili~This
.~~
(240 kDa) . l 3 < I 4 Similar sizes are obtained The OA and aB genes (Figure 3) con- is a characteristic of domain interfaces,
by partial dissociation with weal5 and by sist of three exons (coding regions) sepa- which are often found on the surface of a
reassembly of purified s ~ b u n i t s . ' ~Even
-'~ rated by two introns (n~n-coding).'~ The protein.
smaller species can be obtained by acid exons code for 63 to 67,41 and 67 to 69 There is little doubt that domain 1 is an

Clinical and Experimental Optometry 87.6 November 2004

358
 a-crystallin Augustqrn

a - C r y s t a l l i n exon 1

20 40 60 71
bVaA MDIAI-QHP-WFKRTLG--PFY-----PSRLFDQFFGEGLFEYDLLPFLSSTISPYY---RQSLFR--T~DSGIS~
dfaB MDIAI-QHP-WLRRPLF--PSSIF---PSRIFDQNFGEH-FDPDLFPSFSSMLSPF~GAP~PSWAQ~LSE
hsp2O M E I P V P V Q P S W L R R A S A P L P - - - G L S A P O R L F D Q W D E G L L Q
hsp25 MT-ERRV-PFSLLRSPSWEPFRDWYPAHSRLFDQAFGVPRL-PDEWSQWFSAAGW (28 aa)LNR---QLSSGVSE
hspl6.5 MFGR----DPFD-------SL~~~FF----ATPMTGTTMIQSST-----GIQISGKGFMPISI
hsp16.9 MSIVRRSNVFDPFA-------DLWADPFDTFR---SIVPA-----

a-Crystallin exon 2

80 100 120 126

bvaA VRSDRDKFVIFLDVKHFSPEDLTVKVQ-EDFVEIHG-------KHNERQ
dfa B LRLDKDKFAIHLDVKHFTPEELRVKIL-GDFIEVQA-------QHEERQ
hsp2 0 VPTDPGHFSVLLDVKHFSPEEIAVKW-GEMA--------RHEERP

- - --
hsp2 5 IRQTADRWRVSLDVNHFAPEELTVKTKEG-WEITG-------KHEERQ
hsp16.5 IEGD-QHIKVIAWLPGVNKEDIILNAV-GDTLEIRAKRSPLMITESERI
hsp16.9 WXETPEAHVFKADLPGVKKEEVKVEVEDGNVLWSGERTKEKEDKNDKW
I
P2 P3 P4 I35 P6

a-Crystallin exon 3

127 140 160 180 197

bvaA DDHGY-ISREFHRRYRLPSNQSALSCSLSADOMLTFSGPKIPSGVDAGHSE~IPVSREEKPS-SAPSS-
df a B DEHGY-VSREFHRKYKVPAG~PLVITCSLSTITGPR-~ADV---PERS~ISRDEKPAVAGPQQK
hsp20 DEHGF-VAREFHRRYRLPPGMPAAVTSALSPEGnSIQAP-

-
hsp25 DEHGY-ISRCFTRKYTLPPGVDPTLVSSSSLSPEGTL-VEAP-LPKAVTQS-AEITIP~FE~QIGGPEAGK
hsp16.5
hsp16.9
m
P6
- --
IYSEIPEEEEIYRTIKLPA~EENASAKFEN-OVLSVILPESSI-----KKGINIE---
HRVERSSGKFVRR-FRLLEDAKVEEVKAGLEN-GVLTVTVPKAEV--KKPEVK-AIQISG

P7 P8 P9 P I0
Figure 4. Alignment of bovine UA- (bvaA) and dogfish UB (dfuB) crystallin amino acid sequences with those of human (hsptO), mouse
(hsp25, truncated near the Cterminus), methanoeoccusj a m c h i i (hsp16.5) and wheat (hsp16.9) heat shock proteins. The sequences are
arranged in three segments corresponding to the exons in the UA-crystallingene. Identical or highly conserved residues (D/E; S/T; I/L/v,
K / R Y/F/W) in at least four of the six sequences are shown in bold. The c cry stall in domain is shaded. The numbering at the top is for the
ucrystallin sequences and includes the gaps. The location of secondary structure elements in the hsp 16.9 u-crystallindomain is indicated
with black bars below the sequences.

independent structural unit. As can be
seen in Figure 4, there are large variations
in the size and sequence of this domain in
the shsp family. It is hard to accept that
these variations could be accommodated
in a larger domain without serious disrup
tion of the protein's structure. Further-
more, a mutant polypeptide lacking the first
63 amino acids forms a stable dimeric struc-
ture,24t25indicating that the domain is re-
quired only for assembly of the aggregates.
The homology with invertebrate shsps
(hsp 16.5 and 16.9) appears to reside only
in the sequences corresponding to the sec-
ond exon and a large part of the third.
This suggests that the a-crystallin gene may
have arisen through a process of exon
shuffling, with the second and third exons
derived from the same ancestral gene as
the shsps and the first exon from some
other unknown source. In some vertebrate
shsps even the first exons may have a com-
mon origin. As can be seen in Figure 4,
the hsp20, hsp25 and both a-crystallins
also show substantial homology in their
N-terminal sequences.
In the vertebrate hsps, domain 1 con-
tains an internal duplication with residues
1 to 34 homologous to residues 35 to 63.26
This suggests the domain has evolved
through duplication and fusion of an an-
cestral gene coding for 30 to 40 amino
acids. Removal of the first of these

Clinical and Experimental Optometry 87.6 November 2004

359
a-crystallin Augusteyn

Figure 5. Secondary structure of the hsp 16.9 subunit with an
ordered N-terminus.ss The N-terminal domain (green) contains
three helical segments, as shown, in half of the subunits. The
remaining subunits in hsp 16.9 and all subunits in hsp 16.5 have
unstructured N-termini. The u-crystallindomain (red) consists
of a seven-stranded B-sandwich, an interdomain loop containing
one pstrand (B6 at the top) and a Gterminal extension (at the
bottom), which is largely unstructured except for the short BlO-
strand. (Reproducedwith permission from Nature Structural and
MokcuZur Biology; http://www.nature.com/nsmb/)
Figure 6.A possible micelldike structure for c cry stall in.'^*^ The
subunits contain two domains and assemble into large aggregates
through interactions between their hydrophobic N-terminal
domains (dark grey), which are located in the centre of the
aggregate. The hydrophilic Gterminal (u-crystallin) domains
(light grey) are on the surface of the assembly.

repeated segments by mutagenesis has
no discernible effect on the size of the
a-crystallin molecule or on its stability.25
In view of the conserved nature of the
a<rystallin domain, itwould be expected that
exons 2 and 3 would be interdependent.
The similarities between the hsps and
a-crystallin also extend to their aggre-
gates. Antibodies against a-crystallin re-
act with the Drosophila s h s p ~ . ~Some
’ 16.5 and 16.9.s1*s2
shsps form 200 to 800 kDa aggregates,
which appear very similar to those seen
in electron micrographs of a-crystallin.
Furthermore, hsp25 and uA-crystallin can
coaggregateZ8and complexes of shsps and
aB-crystallin have been observed in other
tiss~es.~~~~~
SECONDARY STRUCTURE
Circular dichroism and infrared measure-
ments indicate that 60 to 70 per cent of
the a-crystallin polypeptide is arranged in
P-strands and there is very little or no
a - h e l i ~ .How
’ ~ these strands are arranged
can be deduced from a consideration of
the structures of the closely related hsp
A ribbon diagram, showing the second-
ary structure of one subunit in hsp 16.9,32
is shown in Figure 5, with the N-terminal
domain, green, and the a-crystallin do-
main, red. The amino acids comprising
the P-strands are identified under the
sequence in Figure 4. The hsp 16.5subunit
is very similar except for its N-terminal
domain, which is unstruct~red.~’ Because
of the large differences between the N-
terminal domains, both in sequence and
in conformation, it is not possible to pre-
dict its structure in a-crystallin.
The a-crystallin domain consists of the
compact Psheet sandwich in the centre,
the interdomain loop at the top and the
Cterminal extension below. The bsheet
sandwich contains seven P-strands and
closely resembles the sandwich found in
the immunoglobulins.ss There is little
doubt that the secondary structure of
a-crystallin would be similar in this highly
conserved domain. .
There will be some differences. For

Clinical and Experimental Optometry 87.6 November 2004

360

example, as can be seen in the sequence
comparisons, there are gaps in the region
corresponding to the interdomain loop
(residues 100 to 120) in the a-crystallins.
Thus, the loop will be smaller. As the loop
is important in subunit interactions, it is
likely that subunit orientations are differ-
ent in a-crystallin.
These observations are consistent with
studies of the location and arrangement
of the pstrands, using an elegant approach
called site-directed-spin-labelling
(SDSL)..943fi
They generated sets of mutant a -
crystallins, in which consecutive amino
acids were replaced with cysteine. A
nitroxide spin label group was then
attached to the side chain of the cysteine
to produce sets of proteins containing
nitroxide spin labels in consecutive posi-
tions. By examining the characteristics of
the label in each protein, the secondary
structure of the polypeptide and its inter-
actions with other parts of the protein
could be deduced. For example, when
sequences are in a Pstrand configuration,
accessibilityand mobility of the label vary
with a periodicity of two (that is, high, low,
high, low et cetera).
Mchaourab’s laboratorys43fiwas able to
identify seven P-strands in residues 60 to
155 (the a-crystallin domain) of aA-crys-
tallin, determine their orientations and
identify their packing in pairs. They com-
pared the arrangement of P-strands in
aA-crystallin, hsp 27 and hsp 16.3 and
found that, as expected, the secondary
structure of a-crystallin was very similar
to that of the larger hsp. Also, they were
able to conclude that the basic building
block of the hsp aggregates was a dimer
formed through a two-fold symmetric
interface between two P-strands near the
N-termini.
The a-crystallin and hsp 25 sequences
extend well beyond the P l O strand (Fig-
ure 4). NMR measurements have shown
that these residues are unstructured and
highly m~bile.~’,~*They contain a high pro-
portion of hydrophilic amino acids and are
thought to contribute to the solubility of
the protein.
Clinical and Experimental Optometry 87.6 November 2004
QUATERNARY STRUCTURE
To determine the subunit arrangement or
quaternary structure of a protein, an ac-
curate molecular weight is required. So far,
this has been unattainable for a-crystallin.
Nevertheless, models for the quaternary
structure abound. They include the three-
layered models, the dodecamer, the
micelles and those involving a tetrameric
arrangement of subunits. Some of these
are based on observations now known to
be incorrect and have been largely aban-
doned. For a detailed discussion on these
models, the reader is directed to the re-
view by Augusteyn and Stevens.6
A key feature of a-crystallin molecules
is that there is no specific requirement for
either CLA or a B s u b ~ n i t s .Any
‘ ~ combina-
tion of subunits can be accommodated
and, regardless of aggregate size, all of the
subunits are equally accessible, in quasi-
equivalent locations on the surface of the
aggregate^.'^,^' It is probable that the
subunits are in two different orientations
as only half can interact with monoclonal
antibodies directed to the a-crystallin
d~main.~’
Therefore, any model proposed for the
quaternary structure of a-crystallin must
have the subunits in two orientations and
in equivalent locations. It must also be able
to accommodate the huge variations
observed in the molecular weight.
Models that best meet these criteria in-
clude those in which subunits are in a
micelle-like arrangement (Figure 6). The
idea of a micellar quaternary structure was
first floated by Thomson in 1985,42and
later developed by Augusteyn and
Koret~T . ~h e~ idea is supported by
previous observations that subunits are
mobile and readily exchange between ag-
gregates under mild conditions;@that the
a-crystallin polypeptide is amphiphilic
with distinct hydrophobic N-terminal and
hydrophilic Gterminai domains;2sand that
the subunits assemble through interac-
tions between the N-terminal domains.45
The micelle model has been adopted
enthusiastically in many laboratories. How-
ever, our further research has shown that
a-crystallin does not exhibit some of the
characteristic properties of lipid micelles,
361
a-crystallin Augustqrn
such as a critical micelle c ~ n c e n t r a t i o n . ~ ~
Furthermore, the model is not consistent
with the uniformity of particles observed
with the electron microscope. Large vari-
ations in diameter would be expected for
micelles with different numbers of
subunits.
Therefore, we are considering an alter-
native model, a tetramer of tetramers, de-
rived in part from our studies on the acid
dissociation of subunit homopolymers.6t18
This is also consistent with many of the
observations on the protein. A similar
model has been suggested for the closely
related murine h ~ p 2 5 . ~Further
’ work is
required to explore this new model.
More information on the possible
subunit arrangement can be gleaned from
consideration of the hsp quaternary struc-
tures that are shown in Figure 7. Hsp 16.5
contains 24 subunits, arranged on the sur-
face of a hollow sphere while hsp 16.9 is a
dodecamer, consisting of two stacked
hexameric rings. In both cases, all of the
subunits are in quasi-equivalent locations
on the surface of the aggregate, as is the
case for a-crystallin.Although the two shsp
molecules contain different numbers of
subunits, they share several common struc-
tural features that would also be expected
in a-crystallin. These can be seen by refer-
ence to the ribbon diagrams in Figure 7b.
In both hsps, the subunits are arranged
head to head in dimeric building blocks,
stabilised by interactions between the
a-crystallin domains (through the top
surface of the domain shown in Figure 5).
The orientations of the subunits in the
dimers are slightly different for the two
molecules, but the similarities are evident
in Figure 7. A key feature of the dimer
structure is strand sharing, where the P 6
strand from each subunit becomes part of
the sandwich in the other subunit. This is
possible because of the large loop in which
this strand is located (Figure 5). In
a-crystallins, the loop is much shorter be-
cause of amino acid deletions (Figure 4).
This changes the nature of the interactions
at the interface but, as shown with SDSL
and X-ray solution scattering, the subunit
dimer is still f ~ r m e d . ~ ~ - ~ ~
A dimeric arrangement is also consist-
ent with the limited binding of antibodies
a-crystallin Augustqn

Figure 7a. Spacefilling models of the 24mer hsp 16.5 (LHS)and the 12-merhsp 16.9 (RHS)molecules. Subunits are
shown in different colours. Different shades of similar colours are used for the two subunits in a dimeric building
block. The hsp 16.5 molecule is a hollow sphere with triangular and square windows on the surface. Hsp 16.9 consists
of two hexameric rings with a triangular central cavity. (Reproducedwith permission from Natum Strueeuml and Moleculat.
Biology; http://www.nature.com/nsmb/)

Figure 7b. Ribbon diagrams showing the secondary structure of a portion of hsp 16.5 around one of the triangular
windows (LHS)and the top ring of hsp 16.9 (RHS). Subunits in the dimeric building block are the same colour. In the
hsp 16.9 molecule, submits in the bottom hexameric ringare shown in lighter shadesof the top ring colours. (Reproduced
with permission from Natutv Structural and Molecular Biology; http://www.nature.com/nsmb/)

Clinical and Experimental Optometry 87.6 November 2004

362

to a-crystallin."2The monoclonal antibod-
ies used were directed at residues within
the sequence numbered 111 to 130 in Fig-
ure 4 (HG ... GY in bovine aA). These in-
clude the p6 strand and the interdomain
loop before it. In the hsp molecules, these
are on the surface of the molecule close
to the subunit-subunit dimer interface.
Binding of antibody to one of the subunits
would make inaccessible the binding site
in the second.
In both hsps, the dimers are arranged
around a central cavity. T h e actual
arrangement is determined by the
Gterminal extensions, which fit into hy-
drophobic grooves on the surface of
subunits in adjoining dimers.
In hsp 16.5, the extension adopts a sin-
gle conformation, which results in the
dimers being located in identical positions
on a spherical surface. This can be seen
in Figure 7b. The C-terminal extension
from the blue subunit on the left lies over
the red subunit beside it. The C-terminal
extension, from the blue subunit on the
right, heads in the opposite direction,
around the surface of the sphere, to in-
teract with an adjacent subunit. Although
the colours are different in the space-
filling model, the extensions are clearly
visible.
The C-terminal extension in hsp 16.9
contains additional residues between the
P9 and PlO strands. It is obviously longer
in Figures 7a and 7b, allowing greater sepa-
ration between the subunit dimers. The
extra length generates a hinge, so that the
angle between the extension and the
a-crystallin domain can vary by 30 degrees.
As a result, there are two different
orientations of the dimers, giving rise to a
quaternary structure different from that
in hsp16.5." One extension is similar to
that in hsp 16.5, lying over the top surface
of the adjoining red subunit, and can be
seen clearly in Figures 7a and 7b. The sec-
ond, which cannot be seen in the views
presented, lies down the side of the mol-
ecule (at right angles to the plane of the
paper) and joins together subunits in the
two rings.
It is probable that the Gterminal exten-
sion functions in the same way in the a-
crystallins.The aA chain contains a cysteine
Clinical and Experimental Optometry 87.6 November 2004
a-crystallin Augustqrn
residue in position 153. This corresponds of the aggregate^.'^.^^ This may be neces-
to a P8 strand residue, which lies in the sary because the Gterminal extensions are
groove where the hsp extension binds. The too flexible to maintain a compact, or-
side chain of the cysteine is buried in dered aggregate. Under conditions where
a-crystallin but becomes exposed with age- the Gterminal extensions have dissociated
related Gterrninal truncation and during from their grooves and the a-crystallin
partial denaturation of the molecule, indi- domain starts to unfold (exposure to 4 to
cating it is covered by a flexible region of 5M urea), the aggregates remain intact.15
polypeptide near the Gterrninus.39 By contrast, the shsp molecules would be
In aA-crystallin, there are even more expected to dissociate. Mutant a-crystal-
residues between the p9 and P l O strands lin polypeptides, lacking the N-terminal
(Figure 4). This suggests that the exten- domain, do not assemble into aggregates
sion may be even more flexible, that is, but form dimers and small amounts of
the hinge may be larger, allowing more tetramer~.'~.'~ Presumably, as in the com-
subunit orientations. The a B hinge is plete molecule, the dimers consist of two
smaller and only one residue longer than a-crystallin domains in the head to head
that in hsp 16.9. These observations sug- configuration, while the tetramers also
gest that the aA and a B homopolymer involve the Gterminal extensions.
molecules may have different quaternary As mentioned earlier, the N-terminal
structures. It is interesting to note that domain of a-crystallin consists of two re-
a-crystallin molecules rich in aB polypep peated sequences. The first of these is
tides appear to be more compact, have partly exposed on the surface49 but the
lower average molecular weights and are second is buried inside the aggregate.
less polydisperse than those rich in cA.'".'~ Mutants lacking the first repeat still form
This may reflect a lower flexibility in a B a-crystallin-sized aggregates. However,
because of the smaller hinge. removal of any part of the second repeat
It seems likely that the a B homopoly- generates the dimers, indicating that
mer is a dodecamer, similar to hsp 16.9. residues 36 to 63 are critical for aggregate
aA may be larger, perhaps the tetramer of f ~ r m a t i o n . * ~ .far,
~~S it ohas not been possi-
tetramers.'" It is difficult to imagine the ble to determine the secondary structure
structure of the mixed a-crystallin aggre- of the repeats.
gates. Indeed, there may be several qua- Although great advances have been
ternary structures, all with the subunits in made in recent years, much remains to be
quasiequivalent locations, and the specific learned before we can fully understand the
structure adopted could depend o n the structure and properties of a-crystallin.
proportions of aA and a B subunits as well
as the molecule's environment. This could
FUNCTIONS
explain some of the variability observed
in the structure of a-crystallin. Until about 10 years ago, it was thought
As can be seen in Figure 7b, the hsp16.9 that the roles of a-crystallin in the lens
molecule is also stabilised, by the N-termi- were passive, that is, to contribute to the
nal domains, which are located within the refractive properties of the lens and to
interior of the aggregate. Half of these remain soluble for the lifetime of the or-
domains have an ordered structure con- ganism. The discovery that akrystallin is
taining three a-helical segments."' By expressed in other tissues2and is elevated
contrast, all of the N-terminal domains in in a number of human diseases"' led to
hsp 16.5 appear to be unstructured,3' sug- speculations that the protein may have
gesting the domain plays little or no role specific functions in the lens and else-
in stabilisation of the aggregate. where.
In this regard, the a-crystallins differ Apart from the structural similarities
substantially from the two small hsps. The discussed earlier, a-crystallin resembles
N-terminal domain is much larger in the many of the other small heat shock pro-
a-crystallins (Figure 4) and several obser- teins in its ability to inhibit the precipita-
vations indicate it is critical for assembly tion of denatured proteins, including the
363
a-crystallin A u p t e y n
CtA
C(B
- --
VRSDRDKFVIFLDVKHFSPEDLTWVQEDFVEIHGKHNERQ
LRLDKDRFSVNLDVKHFSPEELKVKVLGDVIFVHGKHEERQ
P3
Figure 8. The putative chaperone binding sites in bovine aA-and
akystallins
j3- and y - ~ r y s t a l l i n sSimilar
. ~ ~ ~ ~ ~protection
is offered from other types of stress caus-
ing protein denaturation and precipita-
tion. In addition, it appears that the pro-
tein can impart thermotolerance to
cultured cells.54These protective functions
are referred to as the molecular chaper-
one activity (see review by Hendrick and
HartP).
The protective activity involves binding
of denatured polypeptides to the hydro-
phobic a-crystallin to form larger soluble
complexes. The stoichiometry of the bind-
ing is one denatured polypeptide per
subunit.5fiUnlike the large heat shock pro-
teins, ATP is not required and the dena-
tured protein is not refolded and released,
although it may be presented to the large
chaperones for r e f ~ l d i n g . ~ ~ . ~ ~
The mechanism involved is not under-
stood but binding has been attributed to
hydrophobic sequences in the a-crystallin
domain and in the N-terminal domain.
I t has been suggested that these become
exposed only following dissociation of the
m o l e c ~ l e .It~ would
~ ~ ~ ~appear that
a-crystallins and many of the small heat
shock proteins have mobile structures,
being able to exchange subunits between
different aggregates under mild condi-
t i o n ~ . ~This
~ *must
~ ~ .involve
~ ~ dissociation
of the molecule at least into the subunit
dimers. Whether such dissociation is
involved in the chaperone activity is
questionable.
The complex formed between a-crystal-
lin and denatured proteins is larger than
the original a-~rystallin,5~ indicating that
the subunits must reassociate. Given that
the N-terminal domain is required for the
formation of aggregates, it is hard to im-
agine how it could bind a denatured pro-
tein and still participate in the interac-
tions, which generate the a-crystallin
Clinical and Experimental Optometry 87.6 November 2004
P4 P5
aggregate. A covalently cross-linked
a-crystallin, which is incapable of dissocia-
tion, shows undiminished chaperone ac-
tivity.60In addition, aA and a B mutants,
lacking the N-terminal domain, exhibit
substantial chaperone activit~.~’ Thus, it is
unlikely that residues in the N-terminal
domain or dissociation of the molecule are
required for the activity.
Numerous studies have addressed the
relationship between structure and func-
tion by examining the effects of tempera-
ture, pressure, dissociation/association,
chemical modification and denaturation
on the hydrophobicity and chaperone ac-
tivity. As with earlier studies on the p r o p
erties of a-crystallin,fi there is a large
amount of information but little agree-
ment between investigators. This may be
due, in part, to differences in the meth-
ods used for isolating and handling the
proteins. In addition, one must question
the value of studies conducted with pro-
tein isolated in the cold when lens tem-
perature is around 36 degrees Centigrade
and it has been demonstrated clearly that
the protein has different structures at five
degrees and 37 degrees Centigrade.’3,44f’’
Site directed mutagenesis has been used
in attempts to identify residues involved
in the chaperone activity. Unfortunately,
the results are confusing. For example, a
mutation at position 27 in a B was initially
reported to be inactive@but subsequently
shown to be active.fiSIn 1999, it was re-
ported that mutations at positions 19 and
45 in the N-terminal domain of aB were
without effect, while a mutation at 59 pro-
duced a modest decrease.fi4In 2001, the
same laboratory reported that mutations
in positions, 45 and 59 decreased chaper-
one activity.- Swapping of the aA and aB
N-terminal domains has been reported to
abolish activity in one hybrid but enhance
364
it in the other,fi5yet removal of the whole
domain does not appear to have any
effect.4sSeveral mutations in the a-crystal-
lin domain, which affect activity, have also
been described. However, in view of the
confusion in data on the N-terminus, it is
difficult to evaluate these at present.
A major difficulty in mutational studies
is the risk that the mutation affects the
overall structure of the protein and this is
responsible for the change, rather than
any specific effect on the chaperone. It
may not always be possible to detect such
structural changes.
Sharma and co-workers%allowed hydro-
phobic fluorescent probes to interact with
a-crystallin and then covalently cross-
linked the probes and protein. On the
assumption that the probes would have
bound at or near the hydrophobic c h a p
erone binding site, they identified two se-
quences in the a B chain and one in a4
that may be involved in the chaperone-like
activity. The sequences are shown in Fig-
ure 8. The putative binding sites contain
the P3, P4 and P5-strands and, in the hsps,
line the groove on the surface of the
subunit, in which the Gterminal extension
from an adjoining subunit binds. This
would suggest that denatured proteins
bind in the groove and displace at least
part of the Cterminal extension, consist-
ent with the report that immobilisation of
the extension reduces chaperone activity.*
Sharma and his colleagues@ also synthe-
sised a peptide corresponding to the aA
site and, surprisingly, found that this
exhibited chaperone activity.
Much has been made of the chaperone
property and the relationship of a-crystal-
lin with stress proteins (see reviews by
Derham and Harding,fiyGanea,7”Groenen
a n d associate^,^' Horwi tz3 a n d van
Montford, Slingsby and Vierling4). I t is
now generally thought that the chaperone-
like activity plays an important role, espe-
cially in the ageing lens by preventing the
accumulation of insoluble proteins, which
would scatter light and cause a loss of trans-
This is an
parency, that is, cataract.3.5’,m.6y
interesting idea as a-crystallin represents
almost 50 per cent of the proteins in the
mammalian lens72and theoretically (based
on the 1:l stoichiometry), would be capa-

ble of binding all o t h e r lens proteins. It
still remains to b e convincingly established
that t h e c h a p e r o n e activity operates in uivo
a n d that its loss would lead t o increased
light scatter.
There are m a n y r e p o r t s t h a t a g e n t s
linked with cataract formation, f o r exam-
p l e , UV e x p o s u r e , o x i d i s i n g a g e n t s ,
glycation (diabetes), e t cetera inhibit t h e
c h a p e r o n e a ~ t i v i t y . ~ "I' n~ a d d i t i o n , a
n u m b e r o f m u t a t i o n s a s s o c i a t e d with
h u m a n cataract has b e e n r e p o r t e d f o r
a-crystallin. T h e s e have been reviewed by
H o m k 3 Most are i n the a-crystallin do-
main a n d could affect c h a p e r o n e activity.
T h e s e a n d o t h e r observations s u p p o r t
t h e concept of a universal mechanism for
cataract formation, that is, chaperone fail-
ure, where many different triggers lead to
t h e same outcome. However, it is difficult
to accept t h a t t h e brunescent cataracts,
generally attributed t o oxidative damage,
arise through t h e same mechanism as the
swollen hydrated lens observed i n osmotic
ca tat-dcts.
Many questions remain to be answered
concerning the possible role of a-crystai-
lin's chaperone activity in maintaining lens
transparency.
REFERENCES
1. FrutonJ. Mokcules of Life. NewYork: Wiley
Interscience; 1972.
2. Bhat SP, Nagineni CN. Alpha B subunit of
lens specific protein alpha-crystallin is
present in other ocular and non-ocular tis-
sues. Biochem BiophysRes Commun 1989; 158:
319-325.
3. HonvitzJ. Alpha-crystallin. Exp EyeRes2003;
7 6 145-153.
4, Van Montford R, Slingsby C, Vierling E.
Structure and function of the small heat-
shock protein/alpha-crystallin family of
molecular chaperones. Adv Protein Chem
2002; 59: 105-156.
5. Sax CM, Piatigorsky J. Expression of the
alpha-crystdlin/small heat-shock protein/
molecular chaperone. Adv Enzymol Related
Areas Mol B i d 1994; 69: 155-201.
6. Augusteyn RC, Stevens A. Macromolecular
structure of the eye lens. Prog Polymer Sri
1998; 23: 375-413.
7. Koretz JF, Augusteyn RC. Electron
microscopy of native and recoristituted
alpha-crystallin aggregates. Curr Eye Res
1988; 7: 25-30.
8. Siezen RJ, Bindels JG, Hoenders HJ. The
interrelationship between monomeric, oli-
Clinical and Experimenral Optometry 87.6 November PO04
gomeric and polymeric alpha-crystallin in
the calf lens nucleus. Exp Eye Res 1979; 28:
551-567.
9. Haley DA, HorwitzJ, Stewart PL. The small
heat shock protein, aB-crystallin, has avari-
able quaternary structure. J Mol Biol 1998;
277: 27-35.
10 Augusteyn, RC, Koretz JF, Schurtenberger
P. The effect of phosphorylation on the
structure of a-crystallin. Biochirn Bqbhys Ada
1989; 999: 293'299.
11. Stevens A, Walsh R, Augusteyn RC. A sheet-
like form of a-crystallin. CurrEye Res 1996;
15: 215-218.
12. Schurtenberger P, Augusteyn RC. Struc-
tural properties of polydisperse biopolymer
solutions-a light scattering study of bovine
alpha-crystallin. Biopolyms 1991; 31: 1229-
1240.
13. Thomson JA, Augusteyn RC. am-crystallin:
the native form of the protein? Exp Eye Res
1983; 37: 367-377.
14. Thomson JA, Augusteyn RC. On the struc-
ture of a-crystallin: construction of hybrid
molecules and homopolymers. Biochim
Biophys Acta 1989; 994: 24&252.
15. Thomson JA, Augusteyn RC. On the struc-
ture of am-crystallin: The reversibility of
urea-dissociation. J Biol Chem 1984; 259:
4339-4345.
16. Thomson JA, Augusteyn RC. On the struc-
ture of alpha-crystallin: The minimum mo-
lecular weight. CurrEye Res 1988 7: 563-569.
17. Spector A, Li L-K, Augusteyn RC, Schnei-
der A, Freund T. a-crystallin: the isolation
and characterization of distinct macro-
molecular fractions. BiochemJ 1971; 124:
337-343.
18. Stevens A, Augusteyn RC. Acid dissociation
of a-crystallin homopolymers. Biophys J
1993; 65: 1648-1655.
19. van den Heuvel R, Hendricks W, Quax W,
Bloemendal H, de Jong WW, Hendricks W.
Complete structure of the hamster aA-
crystallin gene. Reflection of an evolution-
ary history by means ofexon shufHing.JMo1
Ev01 1986; 24: 121-129.
20. de Jong WW, Hendricks W. The eye lens
crystallins: ambiguity as an evolutionary
strategy. JMoZEvol1986; 24: 121-129.
21. d e Jong WW, Caspers GJ, Leunissen JA.
Geneology of the alpha-crystallin-small
heat-shock protein superfamily. Int J Biol
Mammoll998; 22: 151-162.
22. Gilbert W. Why genes in pieces? Nature
1978; 271: 501.
23. Puri N, Augusteyn RC, Owen E, Siezen RJ.
Immunochemical properties of vertebrate
alphacrystallins. Eur J Biochem 1983; 134:
321-326.
24. Merck KB, de Haard-Boekman WA, Oude-
Essink BB, Bloemendal H, d e Jong WW.
Expression and aggregation of recombi-
nant aA-crystallin and its two domains.
365
acrystallin Auguslqrn
Biochim Biophys Acta 1992; 1130: 267-276.
25. Augusteyn RC. a-crystallin polymers and
polymerization: the view from down under.
In1J Macromoll998; 22: 253-262.
26. Barker WC, Ketcham LK, Dayhoff MO. A
comprehensive examination of protein
sequences for evidence of internal gene
duplication. J Mol Euol1978; 10, 265-281.
27. Augusteyn RC, Boyd A, Kelley L. Immuno-
logical comparison of heat-shock proteins
and alpha-crystallin. Curr Eye Res 1986; 5:
759-762.
28. Merck KB, G r o e n e n PJ, Voorter E,
d e Haard-Hoekman WA, Horwitz J,
Bloemendal H, d e Jong WW. Structural
a n d functional similarities of bovine
a-crystallin and mouse small heat shock
protein. J B w l Chem 1993; 268: 10461052.
29. Zantema A, Verlaan-De Vries M, Maasdam
D, Bol S, van der Eb A. Heat shock protein
27 and alphakrystallin can form a com-
plex, which dissociates by heat shock. JBiol
Chem 1992; 267: 1293612941.
30. Kato K, Shinohara H,Goto S, Inaguma,Y,
Morishita R, Asano T. Copurification of
small heat shock protein with aB-crystallin
from human skeletal muscle. J Bid Chem
1992; 267: 7718-7725.
31. Kim KK, Kim R, Kim SH. Crystal structure
of a small heat shock protein, Nature 1998;
3 9 4 595-599.
32. van Montford R, Basha E, Friedrich KL,
Slingsby C, Vierling E. Crystal structure
and assembly of a eukaryotic small heat-
shock protein. Nature Struct Biol2002; 8:
1025-1030.
33. Mornon JP, Halaby D, Malfois, M, Durand
P, Callebaut I, Tardieu A. a-crystallin C-
terminal domain: on the track of an Ig fold.
Inl J Biol Mammol1998; 22: 219-227.
34. Berengian AR, Bova MP, Mchaourab HS.
Structure and function of the conserved
domain in aA-crystallin. Sitedirected spin
labelling identifies a @strand located near
a subunit interface. Biochemisfry 1997; 36
995 1-9957.
35. Koteiche HA, Berengian AR, Mchaourab
HS. Identification of protein folding pat-
terns using site-directed- spin labelling.
Structural characterization of a P-sheet and
putative substrate binding regions in the
conserved domain of a-crystallin. Biochem-
isfry 1998; 37: 12681-12688.
36. Berengian AR, Parfenova M, Mchaourab
HS. Sitedirected spin labelling study of
subunit interactions in the acrystallin do-
main of small heat-shock proteins. J Biol
Chem 1999; 274: 6305-6314.
37. CarverJA, Probing the structure and inter-
actions of crystallin proteins by NMR
spectroscopy. Prog Retin Eye Res 1999; 18:
431462.
38. Carver JA, Esposito G, Schwedersky
G, Gaestel M. 1H NMRspectroscopy reveals
a-crystallin A ugusteyn
that mouse Hsp25 has a flexible Gterminal
extension of 18 amino acids. FEES Let 1995;
369: 305-310.
39. Augusteyn RC, H u m TP, Putilin T,
Thomson JA. The location of sulphydryl
groups in a-crystallin. Biochim Biophys Acta
1987; 915: 132-139.
40. Augusteyn RC, Ghiggino KP, Putilina TP.
Studies on the location of aromatic amino
acids in a-crystallin. Biochim Biophys Acta
1993; 1162: 61-71.
41. Winzor D, Stevens A, Augusteyn RC. Evalu-
ation of equilibrium constants from pre-
cipitin curves: interaction of a-crystallin
with an elicited monoclonal antibody. Arch
Biorhem Biophys 1989; 268: 221-226.
42. ThomsonJA. Alpha-crystallin: a model for
multi subunit protein assemblies. PhD
Thesis, University of Melbourne; 1985.
43. Augusteyn RC, KoretzJF. Hypothesis: apos-
sible structure for a-crystallin. FEES Lett
1987; 222: 1-5.
44. van den Oetelaars PJ, van Someren PF,
Thomson JA, Hoenders HJ. A dynamic qua-
ternary structure of bovine a-crystallin as
indicated from intermolecular exchange of
subunits. Biochemistry 1990; 29: 3488-3493.
45. Augusteyn RC. a-crystallin polymers and
polymerization: the view from down under.
Int J Macromoll998;22: 253-262.
46. Loutas J, Stevens A, Howlett GJ, Augusteyn
RC. The effect of concentration on the
structure of alpha-crystallin. Exp Eye Res
1996; 62: 613-620.
47. Ehrnsperger E, Lilie H, Gaestel M, Buch-
ner J. The dynamics of hsp25 quaternary
structure. Structure and function of differ-
ent oligomeric species. J Biol Chem 1999;
274: 14867-14874.
48. Feil IK, Malfois M, Hendle J, van der Zandt
H, Svergun DJ. A novel quaternary struc-
ture of the dimeric alpha-crystallin domain
with chaperone-like activity. J Biol Chem
2001; 276: 1202412029.
49. Takemoto LJ, Emmons T, Gransstrom D,
Jacobs S. Antisera to synthetic peptides
as probes of structural changes during
aging of alpha-crystallin from the bovine
lens. Biochim Biophys A c t a 1989; 995:
259-263.
50. Cannon JF. On the quaternary structure of
a-crystallin. PhD Thesis. University of Mel-
bourne 2000.
51. deJong WW, Leunissen JAM, Voorter CEM.
Evolution of the a-crystallin/small heat
shock protein family. MolBiolEvoll993; 10:
103126.
52. Klemenz R, Frhli E, Steiger RH, Schafer R,
Aovama A. Alpha B-crystallin is a small
heatshock protein. Proc Natl Acad Sci USA
1991; 88: 3652-3656.
53. Horwitz J. a-crystallin can function as a
molecular chaperone. Proc Natl Acad Sci
USA 1982; 89: 10449-10453.
Clinical and Experimental Optometry 87.6 November 2004
54. van den Ijsel PRL, Overkamp P, Knauf U, tide Identified as a functional element in
Gaestel M,d e Jong WW. aA-crystallin con- d-crystallin. J Biol C h a 2000; 275: 3767-
fers cellular thermoresistance. FEBS Lett 3771.
1994; 355: 5456. 69. Derham BK, HardingJ. Alpha-crystallin as
55. Hendrick J, Hart1 F. Molecular chaperone a molecular chaperone. Prog Ret Eye Res
functions of heat-shock proteins. Ann Rev 1999; 18: 463-509.
Biochpm 1994; 62: 349-384. 70. Ganea E. Chaperone-like activity of alpha-
56. Wang K, Spector A. The chaperone activity crystallin and other small heat shock pro-
of bovine a-crystallin.JBio1 Chem 1994; 269: teins. Curr Prot Pept Sci 2001; 2: 205-225.
13601-13608. 71. G r o e n e n P, Merck K, d e J o n g WW,
57. Narberhaus F. Alpha-crystallin-type heat Bloemendal H. Structure and modifica-
shock proteins: socializing minichaperones tions of the junior chaperone a-crystallin.
in the context of a multichaperone network. From lens transparency to molecular pa-
M i d i 0 1 Molec Biol Rev 2002; 66: 6493. thology. EurJBiochem 1994; 225: 1-19.
58. Raman B, Rao CM. Chaperone-like activity 72. Pierscionek BK, Augusteyn RC. Growth re-
and quaternary structure of alpha-crystal- lated changes in functional parameters in
IinJBiol Chem 1994; 269: 2726427268. the bovine lens. Biochim Biophys Acta 1992;
59. Das KP, SurewiczW K. Temperature-induced 11 16: 283-290.
exposure of hydrophobic surfaces and its 73. Cherian M, Abraham EC. Decreased mo-
effect on the chaperone activity of alpha- lecular chaperone property of a-crystallin
crystallin. EEBS Lett 1995; 369: 321-325. d u e to posttranslational modification.
60. Siezen RJ, Bindels J, Hoenders HJ. The Biochem Biophys Res Comm 1995; 208: 675-
quaternary structure of bovine a-crystallin. 679.
Effects of variations in alkaline pH, ionic 74. Kelley MJ, David LL, Iwasaki N, Wright J,
strength, temperature and calcium ion con- Shearer TR. a-crystallin chaperone activity
centration. EurJBiocha 1980; 11 1: 435444. is reduced by calpain I1 in uitro and in
61. Augusteyn RC. Dissociation is not required selenite cataract. J Biol Chem 1993; 268:
for a-crystallin’s chaperone function. Exp 1884418849.
Eye Res. In press. 75. SchaurteJA, Gafni A. Photodegradation of
62. Plater ML, Goode D, Crabbe MJ. Effects of tryptophan residues and attenuation of
sitedirected mutations on the chaperone- molecular chaperone activity in alpha-
like activity of alphaB-crystallimJBiol Chem crystallin are correlated. Biochem Biophys R e 5
1996; 271: 28558-28566. Comm 212; 95: 900-905.
63. Derham BK, van Boekel MA, Muchowski PJ,
ClarkJI, Horwitz J, Hepburne-Scott HW,d e Author’s address:
Jong WW, Crabbe MJ, HardingJ. Chaper- Professor RC Augusteyn
one function of mutant versions of alpha
30 Melcornbe Road
A- and alpha B-crystallin prepared to pin-
point chaperone binding sites. E u r J B i o c h Ivanhoe VIC 3079
2001; 268; 13-21. AUSTRALIA
64. Muchowski PJ, Wu GJ, LiangJ, Adman ET,
Clark JI. Sitedirected mutations within the
core ‘alpha-crystallin’ domain of the small
heat-shock protein, human alphaB-crystal-
lin, decrease molecular chaperone func-
tions. JMolBioll999; 289: 397-411.
65. Kumar LV, Rao CM. Domain swapping in
human alpha A and alpha B crystallins
affects oligomerization a n d enhances
chaperone-like activity. J Biol Chem 2000;
275: 22009-22013.
66. Sharma KK, Kumar GS, Murphy AS, Kester
K. Identification of l,11-bi(4-anilino)
naphthalene-5,5’-disulfonicacid binding
sequences in alpha-crystallin. J Biol Chem
1998; 273: 1547415478.
67. Smulders RHP,Carver JA, Lindner R4,van
Boekel MA, Bloemendal H, de Jong WW.
Immobilization of the Cterminal extension
of bovine d-crystallin reduces chaperone-
like activity. J Biol Chem 1996; 271: 29060-
29066.
68. Sharma KK, Kumar RS, Kumar GS, Quinn
PT. Synthesis and characterization ofa p e p
366