biochimica et biophysica acta -...

Biochimica et Biophysica Acta 1860 (2016) 333–343

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r .com/ locate /bbagen

Gamma crystallins of the human eye lens☆

Venkata Pulla Rao Vendra a, Ismail Khan b, Sushil Chandani c,Anbukkarasi Muniyandi d, Dorairajan Balasubramanian b,⁎a Ophthalmic Molecular Genetics Section, National Eye Institute, Building 5635FL, Room 1S24, 5625 Fishers Lane, Rockville, MD 20852, United Statesb Prof. Brien Holden Eye Research Centre, Hyderabad Eye Research Foundation, L. V. Prasad Eye Institute, Hyderabad 500034 Telangana, Indiac Plot 32, LIC Colony, W Marredpally, Secunderabad 500026, Telangana, Indiad Department of Animal Science, Bharathidasan University, Tiruchirappalli 620 024, Tamil Nadu, India

☆ This article is part of a Special Issue entitled Crysand Disease.⁎ Corresponding author.

E-mail addresses: [email protected] (V.P.R. Vendr(I. Khan), [email protected] (S. Chandani), anbu(A. Muniyandi), [email protected] (D. Balasubramanian).

http://dx.doi.org/10.1016/j.bbagen.2015.06.0070304-4165/© 2016 Elsevier B.V. All rights reserved.

a b s t r a c t
a r t i c l e i n f o
Article history:

Received 17 April 2015Received in revised form 8 June 2015Accepted 19 June 2015Available online 25 June 2015

Keywords:Human gamma crystallinsGreek key motifStructure–function correlationCataractogenic mutationsCongenital cataract

Background: Protein crystallins co me in three types (α, β and γ) and are found predominantly in the eye, andparticularly in the lens, where they are packed into a compact, plastic, elastic, and transparent globule of properrefractive power range that aids in focusing incoming light on to the retina. Of these, the γ-crystallins are foundlargely in the nuclear region of the lens at very high concentrations (N400mg/ml). The connection between theirstructure and inter-molecular interactions and lens transparency is an issue of particular interest.Scope of review:We review the origin and phylogeny of the gamma crystallins, their special structure involvingthe use of Greek key supersecondary structural motif, and how they aid in offering the appropriate refractiveindex gradient, intermolecular short range attractive interactions (aiding in packing them into a transparentball), the role that several of the constituent amino acid residues play in this process, the thermodynamic andkinetic stability and how even single point mutations can upset this delicate balance and lead to intermolecularaggregation, forming light-scattering particleswhich compromise transparency.We cite several examples of this,

and illustrate this by cloning, expressing, isolating and comparing the properties of the mutant protein S39C ofhuman γS-crystallin (associated with congenital cataract-microcornea), with those of the wild type molecule.In addition, we note that human γ-crystallins are also present in other parts of the eye (e.g., retina), wheretheir functions are yet to be understood.Major conclusions: There are several ‘crucial’ residues in and around the Greek key motifs which are essential tomaintain the compact architecture of the crystallin molecules. We find that a mutation that replaces even one ofthese residues can lead to reduction in solubility, formation of light-scattering particles and loss of transparencyin the molecular assembly.General significance: Such a molecular understanding of the process helps us construct the continuum ofgenotype–molecular structural phenotype–clinical (pathological) phenotype. This article is part of a SpecialIssue entitled Crystallin Biochemistry in Health and Disease.

© 2016 Elsevier B.V. All rights reserved.

1. Gamma crystallins: phylogeny, presence in the lens, functions

Lens crystallins constitute about 90% of the total soluble protein con-tent of the lens, and 35% of the total lens mass. They are composed oftwo broad classes: α and βγ. The distribution of the three crystallinsin the lens is asymmetric and biphasic [1]. The central portion of thelens is rich in β- and γ-crystallins, and the embryonic nuclear region,from where the lens grows, is particularly rich in γ-crystallins. Theseproteins are the earliest to be expressed as the lens is formed andgrows, and present in very high concentrations (N400 mg/ml in

tallin Biochemistry in Health

a), [email protected]@gmail.com

mammalian eye lenses and N1000 mg/ml in some fish lenses). Theyare packed compactly in short range spatial order as in dense liquidsor glasses such that they aid transparency (no scattering) and the ap-propriate refractive index gradient, high in the nuclear region andlower in the cortical region of the lens. These two features are achievedby the unique and stable structure and packing that they have adopted,a classic example of structure dictating function.

Of the three crystallins found in the vertebrate lens (α, β and γ),gamma crystallins are the smallest and simplest members. Each ofthem has about 175 amino acids in its sequence, with a molecularweight of about 21 kDa. Each of them is folded in a compact globularmanner and exists as a monomer. At least six different types ofγ-crystallins have been reported so far, namely γA–γF. In addition,another one, which used to be called βS-crystallin, is now reclassified asγS-crystallin, based on its similarity with the other γ-crystallins. Thehuman eye lens is known to contain γC, γD and γS crystallins (γE and

http://crossmark.crossref.org/dialog/?doi=10.1016/j.bbagen.2015.06.007&domain=pdfhttp://dx.doi.org/10.1016/j.bbagen.2015.06.007mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]://dx.doi.org/10.1016/j.bbagen.2015.06.007http://www.sciencedirect.com/science/journal/03044165www.elsevier.com/locate/bbagen

Fig. 1. The Greek key motif. A: an artistic rendering of the motif; B: a schematic of the ar-rangement of beta strands in one Greek key domain of theγ-crystallins; C: ribbon diagramof γS crystallin with the motifs in blue, green, yellow and red, and D: Corey–Pauling–Koltun space-filling model rendering the same perspective as in Fig. 1C and E rotated90° for a view from the top.

Table 1Primary structures of human γC-, γD- and γS-crystallins.

An alignment of the amino acid sequences of human γC, γD and γS crystallins. Positions with invariant amino acids are shown inbold. Serine 39 in γS crystallin is underscored. Colored stretches correspond to the four Greek key motifs.

334 V.P.R. Vendra et al. / Biochimica et Biophysica Acta 1860 (2016) 333–343

γF- are pseudogenes in humans). While γC and γD are synthesizedduring the embryonic stage, γS is synthesized post-natally andwith de-velopment their relative proportion decreases with concomitantincrease of α- and β-crystallins.

Gamma crystallin is part of the βγ-family and known to have evolvedfrom primordial sources, namely archaeabacteria, and a description ofsuch microbial βγ-crystallins has been provided recently by Mishraet al. [2]. The solution state structure of such a primordial βγ-crystallinfrom Archaea has been published by Barnwal et al. [3]. The native statestability and related properties of the βγ-crystallins have just beenreviewed by Serebryany and King (4). How these proteins have evolvedover time in order to play a role in the vertebrate eye lens, and whattheir functions are in andout of the lenshavebeen comprehensively sum-marized recently by Slingsby, Wistow and Clark [5], and Wistow andSlingsby [6]. Given that these comprehensive reviews [2,4–6] have justappeared, and that the present article is the lone one on γ-crystallins inthis special issue, we first present here a ‘review of reviews’ on humanγ-crystallins, and follow it upwith howmutations in them are associatedwith congenital cataracts in humans, an area of special interest to eye carecenters such as ours; our center alone treats as many as 1200 such pa-tients each year. Congenital cataracts are largely genetic in origin (muta-tions in over 45 genetic loci and over 38 genes in humans [7]), unlikeother forms of cataract where metabolic, environmental and post-translational modifications can alter the structural and functional proper-ties of the constituents of the lens; also a large fraction of them are Men-delian in origin (one gene–one phenotype) so that a genotype–phenotype correlation is easier. Further, since the detailed molecularstructures of several lens crystallins, particularly the βγ-crystallins are in-creasingly being unraveled, we can attempt a genotype–molecular struc-tural phenotype–clinical phenotype correlation.

We are interested here in congenital cataracts associated withmutations in human γ-crystallins (as many as 30 are reported to datein γC, γD and γS-crystallins). Since the detailed molecular structuresof thesemolecules have been solved, making such a genotype–structuralphenotype–clinical phenotype correlation appears to be a possibility.

We have divided this paper into three broad sections. In the firstsection, we review the results of a significant number of papers inthe literature which have described the unique structural features ofγ-crystallins, namely, the Greek key motif, and summarize how foldingof the molecule using this motif offers these proteins the right set ofproperties needed in the human lens, such as compactness, rightrefractive properties, stability and intermolecular packing to generatea transparent assembly.

In the second section,we showhow single pointmutations in their se-quence can lead to perturbations in the structure leading to intermolecu-lar aggregation and the consequent compromise in lens transparency. Asan example, we present our own ongoing research on themutant S39C ofhuman γS-crystallin, associated with cataract andmicrocornea in a child.

In the third section, we briefly discuss the presence of crystallins intissues outside the lens and some ongoing research on their role, if

any, there. The putative role of lens crystallins, particularly γ-crystallins, is an area that is drawing considerable current attention.

2. Section I: the Greek key motif: role in structure and function

The primary structural sequences of human γ-crystallins are shownin Table 1.

A common protein structural folding pattern that has been pre-served through the phylogeny is the so called Greek key motif, asuper-secondary structural fold that was originally defined and namedby Richardson [8]. The Greek key motif involves a run of four β-sheetstructures folded in an antiparallel manner to generate a neatly foldedcompact domain, as shown in Fig. 1. Several archaeal proteins such asNitrollin or M-crystallin are folded using one such Greek key domainwhile the eye lens βγ-crystallins are two-domain structures (with twoGreek keymotifs per domain), and the protein referred to asmammali-anAbsent inMelanoma 1 (AIM-1) uses asmany as six such domains [2].

Indeed, it is this folding using the Greek key motif that rendersthe γ-crystallins with a variety of favorable features and properties

335V.P.R. Vendra et al. / Biochimica et Biophysica Acta 1860 (2016) 333–343

suited for their role in the lens. The intramolecular packing involvingthe coming together of the N-terminal and C-terminal domainsgenerates a compact monomeric globule of about 5 nm in size,i.e., Stokes radius of 2.13 nm [9], and a low frictional ratio of about1.21 (only slightly above the value of 1.12 for a perfect compact smoothprotein sphere), suggesting that they have low hydration and a lowpropensity for sticky interactions with solvent and perhaps, withother proteins [10]. These small globules, packed at such high concen-trations in the lens (400–1000 mg/ml), are seen to exhibit short rangeorder which is necessary for transparency, particularly in a crowdedmacromolecular environment as in the lens [11]. Interestingly, it hasbeen reported that while interactions between γ-crystallins are weaklyattractive, those betweenβ-crystallins appear to bemildly repulsive [12].

Table 1 lists the sequences of the three γ-crystallins found inhumans. Human γC-crystallin, a protein of 174 residues, is folded infour Greek key motifs, residues 2–40 forming the first motif, residues41–83 forming the second (both in the N-terminal domain of themolecule) and residues 88–128 making up the third and 129–171making up the fourth and last Greek key fold. Residues 84–87 makeup the ‘linker region’ which, being a short one, brings the N-terminal

Fig. 2. Tyrosine corners and domain interface. Fig. 2A: The tyrosine corner domain interface in hforms a H-bond with the backbone oxygen of R59 (residue N-4). In the C-terminal domain, it iintervening residues are in gray. Fig. 2B: Residues involved in stabilizing the domain interface iCPK structural format; the two polar residues of each domain are shown in the ball-and-stick forN-terminal domain are colored green, while those in the C-terminal domain are in pink.

and C-terminal domains fold upon each other, thus making the mole-cule a compact globule. Human γD-crystallin is folded in exactly thesame way as human γC-crystallin. In human γS-crystallin, shown inFig. 1, the sequence 6–44 folds into one Greek key motif, the residues45–87 fold as the second, while the sequence 94–134 forms the thirdand the stretch 135–177 adopts the last motif. Residues 88–93 formthe connecting peptide that allows the intramolecular coming togetherof the N-terminal and C-terminal domains to produce a compactglobular molecule. (In contrast, the linker region plus N-terminal andC-terminal extensions in the β-crystallins together lead to inter-molecular, rather than intra-molecular, interactions generating dimersand multimers [13,14] rather than monomers like in γ-crystallins.)Also, while the βγ-crystallins fold using the Greek key motifs,α-crystallin, which belongs to the small heat shock protein family,does not. It is a multimeric protein with molecular weights as highas 800 kDa. The structures and properties of some members of the β-crystallin family, and those of α-crystallins are dealt with by othercontributors to this special issue.

The crystal structure of calf lens gamma-II crystallin was first pub-lished by Blundell et al. [15] and Wistow et al. [16], and that of human

uman γD-crystallin. In the N-terminal domain (left), the side chain hydroxyl group of Y63s Y151 and M147. The inset shows only the N-terminal domain tyrosine corner; the threen human γD-crystallin. Three hydrophobic residues from each domain are depicted in themat. The hydrogens have been stripped for clarity, and the carbon atoms in residues in the


γD-crystallin by Basak et al. [17]. (Also, the crystal structure of the C-terminal domain of human γS-crystallin has been determined [18].)The molecule is seen to pack in a highly symmetric manner, with eachof the two domains as tightly packed β-sheet sandwiches made ofGreek key motifs. As Chen et al. [12] point out, this template structureis highly conserved, thanks to several key residues in the sequencewhich play essential roles in the folding pattern, yet tolerates some var-iations in the non-essential residues in the sequence. The relation be-tween the role of the Greek key structure and the values and gradientsin the refractive index of lens proteins has been drawn attention to insome recent papers [19,20].

The function of the lens is to focus the light coming through the cor-nea into the retina. This would need not only the appropriate refractiveindex values, but also a refractive index gradient which minimizesspherical aberration. The lens should also be plastic (changeable inshape) and elastic (return to original shape after pressure is released),so as to accommodate its size and shape in a manner such that near ob-jects aswell as far objects should be focused on the retina.While the av-erage refractive index of the nuclear region of the porcine lens is about1.45, it decreases to about 1.36 as we move from the central to the cor-tical region on the equatorial plane [21]. The macromolecular concen-tration gradient (very high levels to gradually lower ones) that is seenas we go from the nuclear to the cortical region of the live lens is inwhat has been called opto-biological synchrony [22]. The refractiveindex of a protein can be estimated based on the refractive index incre-ment (dn/dc) values from the amino acid composition, using known

Table 2Mutations reported in human γD-, γC- and γS-crystallins (taken with permission from Vendra

S. no. Mutation Type of cataract seen Reference Comments

A: Mutations reported in human γD-crystallin1 R14C Nuclear and perinuclear [45,46] Only surface ch2 R14S Coralliform [48] Slight change in3 P24T Coralliform, cerulean, fasciculiform [49–55] Extensive confo

surface change4 A36P Nuclear [61] Greek key com

(see ref. 62)5 R37P Nuclear opacity [63] No data yet6 R37S Birefringent crystals [64] Crystal structur7 W43R Dominant congenital cataract [44] Minor tert stru8 M44V Blue dot opacity [66] No structural d9 Y56X Nuclear [67] Large scale trun10 R58H Aculeiform [68] H bonding lost,11 G61C Coralliform [69,70] Mutation main

to unfolded for12 R77S Polar coronary cataract [71,72] No change seen13 E107A Nuclear [73,74] Same as above.

(see ref. 62)14 Y134X Not reported [75] No data, but Gr15 R140X Nuclear [76] Greek key 4 los16 W157X Nuclear [77] Greek key 4 los17 G165fs Nuclear [79] High hydropho

B: Mutations reported in human γC-crystallin1 T5P Coppock-like cataract [68] No data2 G62fs Zonular pulverulant [80] No data3 C109X Nuclear cataract [81] Truncation and4 S119S Nuclear cataract [67] Polymorphism5 W157X Nuclear cataract [82] Truncation mu6 R168W Nuclear cataract [77,83] pI change, as in7 R48H Nuclear cataract [84] Similar to the γ8 G129C Nuclear cataract [85] Tertiary structu

C: Mutations reported in human γS-crystallin1 G18V Progressive cortical cataract [86] Extensive data2 D26G Coppock cataract [61] Very little chan3 S39C Microcornea-cataract [76] Surface hydrop4 V42M Autosomal dominant cataract [90] Compact packin5 G57W Autosomal dominant pulverulent cataract [93] G highly conser

Residue numbers in column 2 are as reported with cited references, regardless of whether metout of the center of the lens; cerulean: small bluish dots; fasciculiform: fibrous radiative strandcentral lens, slowly progressive cortical opacity; Coppock-like: bilateral progressive opacity of ttain layers between nucleus and cortex.

amino acid refractivity values of each constituent residue [23], and notnecessarily its actual sequence. It is noted that the dn/dc values ofdouble Greek key domains are consistently higher (0.2009 ml/g forthe Greek key motifs of crystallins, cf. 0.1899 for other human proteins[19]). Likewise, Zhao et al. [20] had earlier argued that “one could spec-ulate that this flexibility of the Greek key motif to accommodate differ-ent residues lends itself well as a stable scaffold for lens proteins,providing the opportunity of replacing residues with low dn/dc withhigh ones”. It should also be noted that γ-crystallins have the highestdn/dc values, namely 0.203 ml/g while β-crystallins have the lowest(0.187 ml/g, and α-crystallins in between [24]). This is in keepingwith the fact that γ-crystallins are the most abundant proteins in thecore of the lens where the refractive index is the highest.

Table 1 above lists the amino acid sequences of the three human γ-crystallins, namely γC, γD and γS. Note the remarkable sequence ho-mology between the three, suggesting aswell that theywill all be foldedin much the same manner, which has proved to be the case.

The crystal structural analysis of γ-crystallins shows all nonpolarresidues in the molecule to be packed in the interior and polar oneson the surface. Solution state studies also bear this out. This alsomakes the molecule both thermodynamically and kinetically stable.The intra- and inter-molecular packing features of themolecule (partic-ularly human γC, γD and γS), and the roles played by various constitu-ent residues have been dissected and reported in a series of about 30papers by Jonathan King's group [25] and in about 12 papers by thePande group [26] using optical spectroscopic and thermodynamic

et al. [62]).

ange; extensive disulfide intermolecular bridges (see ref. 47)hydrophobic content, opening up a phosphorylation site (no conformational work)rmational analysis done by many shows no change in sec/tert structure, minors, drop in solubility (see refs. 56–60)pactness distorted, higher hydrophobic exposure upon mutation; solubility drops

e reveals no changes (see ref. 65)ct. change; pI must be different, solubility not affected (see ref. 44)ata yetcation of chain; three Greek key motifs lostbut no other change. pI altered (see ref. 65)ly affected the transition from native to intermediate state but not from intermediatem.in structure upon mutation. pI change? Solubility high (see ref. 62)Solubility high. pI change is seen to lead to heteroaggregation with α-crystallin

eek key 4 lost (see ref. 62)t; solubility lost; surface exposure extensive (see ref. 62)t; solubility lost; surface exposure extensive (see refs. 62,78)bic exposure; solubility lost (see ref. 62)

loss of Greek key

tant; data shows solubility loss, exposure of residues to surface (see refs. 62,78)E107A? (see ref. 62)D- mutantre appears impaired

shows no change in 2° or 3°, but compaction of GK 1 likely to alter (see refs. 87,88)ge in 2° or 3°, solubility high (see ref. 89)hobicity increases, causing the tendency to self aggregate (current study)g of Greek key is distorted (see refs. 91,92)ved. Substitution by W partially disrupts the second Greek key motif.

hionine1 was counted or not. Note-2: Coralliform: round or elongated processes radiatings emanating from the center; aculeiform: frosted lens; polar coronary: ring around a clearhe embryonic nucleus, pulverulent (fine powderish); zonular/lamellar: affecting only cer-

Fig. 3.Mutation leads to increased surface exposure. Panel A: An overlay of the wild type(green; PDB code 2M3T) and S39C mutant (the red arrow points to the S residue). PanelB: Y11, F16 and S39 residues in the wild type. Panel C: Y11, F16 and C39 in the mutantmodel. Panel D: an overlay of B and C.


methods, and by NMR methods, notably by the groups of Gronenborn[27], Martin [28] and others.

These studies have revealed several interesting properties of themolecule. For example, γD-crystallin is highly stable, not denatured bythe conventional denaturing agent urea, and needs over 2.5 M of thestronger denaturant guanidinum chloride [29]. In contrast, βB1, βA1,βA3 and βB2 crystallins denature in urea [30,31]. The free energy ofdenaturation of human γD-crystallin has been estimated to be about8.9 kcal/mol [32], and of γS to be 10.5 ± 0.9 kcal/mol [33]; in compari-son, that of αA-crystallin is 6.38 kcal/mol and αB 5.04 kcal/mol [34].Similarly, the thermal denaturation temperature of γS is about 74.1 °Cand of γD is 83.8 °C [33], although it is pH dependent [35], whilethose of β-crystallins are lower; e.g., that of βB2 is around 58 °C [36]and βB1 about 67 °C [37]. It is also interesting to note that the stabilityof the N-terminal domains of γ-D and γ-S crystallins are inherentlylower than those of their C-terminal domains [33]. Also, γ-crystallinsare kinetically stable, making them long-lived. The half-life of the initialunfolding step of γD-crystallin is estimated to be around 19 years [38]!

3. Roles of some constituent amino acid residues in the structureand stability; relation between structural changes andlens malfunction

While it is true that the Greek key motif is adopted by a variety ofproteins with differing primary structural sequences [2], there aresome constituent residues that play an important role in maintainingthe architecture of human γ-crystallins. These have been studied ingreat detail by a number of workers cited above.

The role played by the aromatic residues needs special mention.Human γC-crystallin has a total of 22 aromatic residues (14Y, 4W and4F), γD has 24 of them (14Y, 4W and 6F), while γS-has 27 of them(14Y, 4W and 9F). Of these, the Y residues have a particularly importantrole in the stability of the Greek key fold, by placing themselves in whathas been termed as the tyrosine corner [39], wherein a tyrosine residuenear the beginning or the end of an antiparallel β-strand hydrogenbonds with a residue (often the Y-4 residue, or Δ4 Tyr corner) thus sta-bilizing the β-barrel structure of the Greek key motif. For example, inγC-crystallin we find two Δ4 Tyr type of homologous tyrosine corners,one in N-terminal domain around Y63 (Y56, Y63 andW69) and anotherone in the C-terminal domain around Y151 (Y144, Y151, W157), andthe same is true with γD, while the tyrosine corners in γS are (Y60,Y67 and W73) and (Y150, Y157, W163).

Fig. 2A highlights two such tyrosine corners in human γD-crystallin,those of Y63 and Y151. Yang et al. [40] have recently dissected the con-tributions of the β-hairpin tyrosine corners to the stability and long lifeof the γ-crystallins.

(Note: The amino acid residues in several publications are numberedcounting the N-terminal starting residue methionine as number 1(e.g., P24T, R77S), while other publications discount met 1 and countresidues as (P23, R76). In order to avoid any confusion and maintainuniformity, we number residues here counting the starting methionineas met 1).

Besides tyr, the clustering of aromatic side chains within the interiorof the molecule is a feature that allows the γ-crystallins to take on acompact shapewith all nonpolar residues buriedwithin. These aromaticresidues cluster into several structural elements. Besides the tyrosinecorners, six aromatic pairs (four tyr/tyr, one tyr/phe and one phe/phe)are present in the β-hairpin sequences of the Greek key. The role ofthese aromatic pairing to the stability of the molecule has beenhighlighted by Kong and King [41]. The role that trp residues (W) playis also worth commenting upon. In human γC and γD crystallins, allthe four trp residues are well buried at positions 43, 69, 131, 157. Ofthe four trp residues, W69 and W157 are surrounded by aromaticamino acids: Y56 and Y63 in the case of W69, and Y144 and Y151 sur-round W157. Similarly, in γS, the trps are in positions 47, 73, 137 and163. How these trp and tyr/cys clusters go to stabilize and protect γ-

crystallin has recently been studied [42,43]. Interestingly replacementof even one of these, namely W43 (by R) is seen to lead to weakeningof the stability, loss of solubility and protein aggregation in humanγD-crystallin, as in the case of a human patient with cataract [44].

The two domains are covalently linked by a four/six residue-longconnecting peptide which allows them to interact non-covalentlythrough inter-domain contacts. This inter-domain interface is com-posed of a cluster of six hydrophobic residues and two pairs of polarperipheral interactions surrounding the hydrophobic cluster. These in-terfacial residues of human γD-crystallin are illustrated in Fig. 2B. Thehydrophobic cluster consists of M44, F57, and I82 from the N-terminaldomain and V132, L145 and V170 from C-terminal domain. Peripheralpair-wise interactions are between Q55/Q143 and R80/M147. Howsuch a hydrophobic inter-domain interface contributes to the foldingand stability of the molecule has been described [32].

Quite apart from the aromatics, several other residues (e.g., arginine,cysteine, proline, serine) also contribute to the structure and stability ofhuman γ-crystallins. Replacement of even one such ‘critical’ residue bymutation can lead to drastic changes in the protein's properties, whichtranslate into compromise in transparency of the lens and cataract.

4. Section II: mutations in human γ-crystallins and pathologyTable 2 lists as many as 30 such cataract-associated mutations in

human γC, γD and γS crystallins reported to date in human patients.(It is interesting to note that most of these are single point mutations,while some others are truncation mutations, one of them is a deletionvariant, and one a 5-base pair insertion.) They are all inherited (congen-ital) cataracts with no post-translational or age-relatedmodifications inphenotype. These make it easy to clone the variant by site-directedmu-tagenesis, isolate and purify the molecule and compare its structuralproperties with those of the wild type protein, so as to understandwhat the mutation has done to the properties of the protein.

The role played by the arginine residues (20 of them in human γC,21 in γD, and 13 in γS) in maintaining the overall structural folding ofthe native protein is exemplified in the Table, where we see that replac-ing any of these by other amino acid residues (e.g., R15, R37, R59, R77Sand R140X in γD, R48 and R168 in γC) is associated with congenital


cataracts. The γDmutants R37S and R59H are seen to be more prone tocrystallization than the wild type protein [65].

And the role played by cysteine residues (6 in γD, 8 in γC and 7 inγS) has been studied in detail by the Pandes, who have, for example,shown how when R15 is replaced by C in γD-, the resultant mutantmolecule undergoes intermolecular disulfide crosslinking, whichinitiates protein aggregation [47]. Giblin et al. [94] have suggested thatS-thiolation within the molecule may act to delay the insolubilizationof the lens proteins.

Though the imino acid proline, both through its ring structure andthe lack of hydrogen-bonding ability, causes a break in the α-helicaland β-sheet runs, it seems to play a role in the folding of theGreek key motif. For example, in γD-, replacing the residue P24 with T(or other residues such as S or V), as reported in detail by the Pandes[56–60], decreases the solubility of the protein by increasing hydropho-bicity and restricting backbone flexibility, altering the temperature-dependent solubility and the phase diagram. The roles of other aminoacids in the sequence, whose replacement is associated with lens mal-function, are yet to be understood. We present some results obtainedwith one such example involving the serine residue below.

5. Study of a serine mutant: S39C of human γS-crystallinThe role of serine residues in the molecule is yet to be clarified.

There are as many as 13 S residues in human γC, 17 in γD and 11 inγS-crystallin. Yet, interestingly, Table 2 reports mutation in just one ofthese so far, namely the single point mutation S39C in human γS-crystallin, which is associated with micro-cornea and cataract [76].S39 is conserved across all γ-crystallins (it is S35 in γC and γD), and isthus thought to be important for the structure. To get an idea of the

Fig. 4. S to C change at 39 does not affect the back bone conformation but slightly alters its tertiaIf: fluorescence emission intensity in arbitrary units. λexc: 295 nm; the protein concentrations uand mutant human γS-crystallin S39C. θ: elipticity inmillidegrees. Protein concentration in eacwas 2 mm and all spectrawere recorded at 37 °C, corrected for background buffer signal and eaandmutant human γS-crystallin S39C. θ: elipticity inmillidegrees. Protein concentration in eachother conditions ofmeasurementwere the same as above. PanelD: Intrinsicfluorescence ofwildunits. λexc: 295 nm; The protein concentrations usedwere 10 μM (0.2mg/mL) in 100mM sodiu2.5 nm and spectra were recorded at 37 °C.

change in local environment in the vicinity of residue 39, we built amo-lecular model of the C39 mutant using as template the γS-crystallinstructure provided by Kingsley et al. [95]. Fig. 3 shows the overlay ofthe structures of wild type (WT) human γS and the S39C mutant. Ascan be seen in Fig. 3C, the Y11 and, particularly, the F16 side chainshave moved outward and are somewhat more exposed in the mutant.In order to get a molecular understanding of this cataract, namely,how this replacement of S by C affects the structure of the mole-cule, we have used site-directed mutagenesis of the mutant S39CγS-crystallin molecule, cloned, expressed, purified and studied itssolution state properties, and compared them with those of native WTmolecule, and present some preliminary results here. The materialsandmethods used for model building, cloning, over expression, isolationand purification of the recombinant wild type and S39C mutant ofhuman γS-crystallin and the analysis of their structural and stability fea-tures were done as described in our earlier publications [62,78,89,91].We present the results here.

Turning to the results, we note from Fig. 4A that the gel elution profileof the WT and the mutant S39C are essentially identical; the two mole-cules elute at about the same volume, suggesting that the mutant isquite soluble and has about the same size as the wild type. (There is ahint of a higher molecular species eluting a little ahead (around17.5 ml) of the main band in the mutant, but not in the wild type. Weare currently following this up in order to checkwhether it suggests mo-lecular aggregation, e.g., dimerization, in the mutant protein.) Fig. 4B,which describes the far-UV circular dichroism (CD) profiles of the twoproteins in the far-UV region (250–190 nm), shows that the secondarystructure of the twomolecules is about the same. The characteristic neg-ative band at 218 nm, positive band around 195 nm and a shoulderaround 206 nm are indicative of a high degree of β-sheet conformation

ry structure. Panel A: Gel elution profile of wild type andmutant humanγS-crystallin S39C.sed were 0.825mg/mL in 50mM Tris-Cl (pH 7.3). Panel B: Far-UV CD spectra of wild-typeh case was 0.150 mg/ml in 10mM sodium phosphate buffer (pH 7.3). The cell path lengthch spectrum is an average of 3 independent runs. Panel C: Near UV CD spectra of wild typecasewas 0.825mg/ml in 50mMTris-Cl (pH 7.3). The cell path lengthwas 10mmand thetype andmutant humanγS-crystallin S39C. If: fluorescence emission intensity in arbitrarym phosphate buffer (pH 7.0). The cell path length was 3 mm, excitation and emission slits

Fig. 5. Panel A: Surface exposure of residues in thewild type andmutant humanγS-crystallins S39C,monitored using bis-ANS as the extrinsic probe.λexc: 390 nm, If at 490 nmof the probewasmeasured as a function of its increasing concentration. The cell path lengthwas 3mm, excitation and emission slits 2.5 nm, and spectrawere recorded at 37 °C. Each point is an averageof 3 independent runs. Panel B: Aggregation tendencies of thewild type andmutant humanγS-crystallin S39C, estimated usingNile Red as the extrinsic probe.λexc: 540nm, If at 605nmofthe probewasmeasured as a function of its increasing concentration. The cell path lengthwas 3mm, excitation and emission slits 10 nm, and spectrawere recorded at 37 °C. Each point isan average of 3 independent runs.


in the secondary structure of the protein chain, expected of theGreek keyfold. CD spectra in the near-UV region (340–250 nm), shown in Fig. 4C,focusing on the micro-environmental region of the aromatic residues,shows a slight difference between the two molecules, essentially intheir intensities. But such comparisons are better made and more infor-mative, rather than near UV-CD, when one looks at the intrinsic fluores-cence features of the aromatic residues, particularly trp, which has thehighest quantum yield of emission (with the wavelength of excitationat 295 nm).When themicroenvironment of the aromatic residue is non-polar, the emissionmaximumoccurs in the 320nmregion, and the emis-sion quantum yield is rather low. But when the residue experiences amore polar environment, its emission band is red-shifted and the inten-sity of emission increases. Fig. 4D compares the emission profiles of thetwo proteins; we find that while the band maximum is only slightlyred-shifted (by 1 nm) in the mutant compared to WT, and the band in-tensity here has increased modestly, by about 10% compared to that ofthe wild type, suggesting that the local environment around the onceburied aromatics has become somewhat more exposed.

Such changes in the surface polarity are bettermonitored using dyessuch as bis-ANS (4,4′-dianilino-1,1′-binaphthyl-5,5′-disulfonate) [96],and 9-diethylamino-5H-benzo α-phenoxazine-5-(one) called Nile Red[97] as extrinsic probes which, upon binding to exposed hydrophobicsurfaces display significant changes in their emission maxima and in-tensity. Fig. 5A shows that when 140 μM of bis-ANS is titrated withthe proteins, its emission intensity is significantly higher (19 arbitrary

Fig. 6.Mutant has less thermal stability and is prone to aggregate faster. Panel A: Thermal denat350 nmbandwas compared to that of the 327 nmband andmonitored as a function of time. Prthe range of 30–80 °C. The cell path lengthwas 10mm, excitation and emission slits 10 nmand sPanel B: Thermal aggregation ofwild type andmutant at different temperatures. Light scatteringA protein concentration of 0.260 mg/ml in 50 mM Tris buffer was used. The cell path length w

units) when bound to S39C than when bound to the WT (10 units),suggesting that the mutant molecule has opened up a little more, thusoffering a greater surface for dye binding.

Does amore exposed surface allow themolecule to display a greatertendency to make intermolecular interactions and thus aggregate? Thispoint can be checked by using the aggregation monitor dye Nile Red.Fig. 5B compares the binding of this neutral dye Nile Red. Upon bindingto the wild type, it displays a broad band maximum centered around651 nmwith an intensity of about 0.65 arbitrary units, while upon bind-ing to S39C, the band blue-shifted by 15nmto636 nm, and the intensitydoubled to over 1.3 units; this result too suggests that as the S residue isreplaced by the slightly bigger C, the tendency of the protein to offer asomewhat more polar surface and promote intermolecular aggregationbecomes noticeable.

The temperature-dependent unfolding of the two proteins was nextstudied by following the relative intensity ratios of fluorescence at350 nm (representative of the red-shifted band maximum obtainedupon total unfolding of the molecule) and at 327 nm (representingthe native form). Fig. 6A shows thatwhile theWTmolecule starts to un-fold after 60 °C, with amidpoint at around 74 °C, themutant unfolds at amuch earlier temperature, well below 70 °C, with the midpoint around59 °C, suggesting that the mutant has a lower thermal stability. Indeed,themutant starts scattering light already by about 50 °C during the runs,hampering the accuracy of the experiment. Fig. 6B describes such time-dependent increase in light scattering. While the wild type molecule

uration of wild type andmutant S39C; λexc: 295 nm; The relative emission intensity of theotein concentration in each casewas 0.065mg/ml in 50mM Tris-Cl (pH 7.3) and heated inpectrawere recorded at 37 °C. Samplewas allowed for 1min prior to recording the spectra.wasmonitored at 600nm light for 625 s at different temperatures (43.5, 53.5 and 54.0 °C).as 10 mm, excitation and emission slits 5 nm and spectra were recorded at 37 °C.

Fig. 7. Legend: Mutant is less stable towards the chemical denaturant GuHCl and capable of forming an intermediate state during unfolding. Guanidine hydrochloride-induced denatur-ation ofwild type (curveA) and S39C (curve B)γS-crystallin.λexc: 295 nm; the relative emission intensity of the 350nmband (of the denatured form)was compared to that of the 327 nmband (of the native protein) and monitored as a function of denaturant concentration. The solid line indicates the fitted data and solid blocks stand for raw data. Protein concentration ineach samplewasfixed at 0.2mg/ml in 50mMTris buffer, 1mMEDTA and5mMDTT. The cell path lengthwas 3mm, excitation and emission slits 5 nm, and spectrawere recorded at 37 °C.


remains clear in solution at 54 °C for as long as 600 s, as monitored bylight scattering at 600 nm, the mutant aggregates to produce scatteringparticles even when allowed to stand at 53.5 °C at 400 s; this is furtherhastened to 300 s, when it is incubated at 54 °C.

We next studied the chemical denaturant-induced unfolding of themutant. The denaturant guanidine hydrochloride (GuHCl) was used tounfold the protein and the relative fluorescence ratio (350 nm/327 nm, representative of the ratio of unfolded/native forms of theprotein) was used as the assay. As Fig. 7A shows, thewild typemoleculedisplays a clear, sharp two-state transition from the native to the un-folded form, with themidpoint of unfolding occurring at the concentra-tion (Cm) of 2.8 M GuHCl. Data from this equilibrium native to theunfolded two-state (N–U) curve were fitted to the two-state transitionmodel of Greene and Pace [98], yielding a free energy (ΔG0N–U) value of9.145 ± 0.30 kcal/mol. In contrast, the denaturation curve (Fig. 7B) forthe mutant displays a multiple step process; the first transition alreadyoccurs well before 1.5 M GuHCl, and the second at 2.8 M GuHCl, just aswith theWTmolecule.When these datawere analyzed using the three-state model, proposed by Clark et al. [99], we obtained a Cm value of0.73 M GuHCl and a (ΔG0N–I) value of 4.32 ± 0.26 kcal/mol, and a Cmvalue of 2.8 M GuHCl for the second transition (I–U) with a ΔG0I–Uvalue of 4.88 ± 0.20 kcal/mol (see Table 3). The behavior of S39C isthus reminiscent of some other mutants that we had described earlier,e.g., V42M human γS- and A36P γD-crystallins.

We next examined the changes that the mutation brings about inour molecular model which was built following protocols described inour earlier reports [62,78,89,91] using the Swiss-Model workspace[100] and refined by molecular dynamics procedures in Gromacs 4.5.5[101]. Fig. 3, shown above, suggests that in the wild type molecule, res-idues Y11 and F16 are in close contact, and S39 iswithin interacting dis-tance. When it is replaced by C in the mutant, both Y11 and F16 moveoutward in order to accommodate the bulkier sulfur atom in C39. Fur-ther analysis shows that there is a change in the aromatic–aromatic

Table 3Equilibrium unfolding parameters for wild type human γS-crystallin and its S39Cmutant.

Protein Equilibrium transition 1 Equilibrium transition 2

Cm Apparent ΔG0 P value Cm Apparent ΔG0 P value

Wild type 2.8 M 9.15 ± 0.3[ΔG0N–U]

0.001

S39C 0.73 M 4.32 ± 0.26[ΔG0N–I]

0.002 2.8 M 4.88 ± 0.20[ΔG0I–U]

0.002

[Cm] = transition midpoint in the units of M GuHCl; [ΔG0] = free energy of unfolding inthe absence of GuHCl in units of kcal/mol−1; N–U refers to native to unfolded transition,N–I native to intermediate and I–U intermediate to unfolded state.

interactions between the two residues, with the centroid–centroid dis-tance between Y11 and F16 increasing to 6.2 Å in the mutant whencompared to 5.5 Å in the wild type γS-crystallin. Both S39 and C39form main chain–main chain H-bonds with the N of Y11. The sidechain oxygen of S39 forms H-bonds with the main chain N atom ofF16 and the O of D13. The larger S atom in the mutant forms thesetwo H-bonds and, in addition, forms H-bonds with the N atom of N15and the O of P68. The S atom is also involved in the aromatic–sulfur in-teraction with the F16, the centroid-sulfur atom distance being 4.5 Å.However, the larger bulk of the S atom pushes out F16, which alsoleads to a disruption, in themutant. Cation-pi interaction seen betweenF16 and K41 (distance 3.7Å) in thewild type protein. Fig. 8 gives a com-parison in our model of the inter-atom interactions between Y11, F16and S39 (or C39 in the mutant), while Table 4 estimates the distances.

Note the small but noticeable increase in the distance between thebackbone C atoms of residues 16 (F16) and 39 (S39 or C39) (from6.60 to 7.34 Å) and also of the N atoms (5.42 to 5.991 Å), while thereis some shrinkage between those of Y11 and S/C39.

It is also worth pointing out that the mutation involves the replace-ment of serine (with a relative polarity ranking of 14 (based on the rel-ative hydrophobicity scale), an average percentage of about 8% burial inproteins, a van derWaals volume of 73 Å3, and accessible surface area of

Fig. 8.Mutation alters the inter-atomic distances: H-bond interactions around residue 39in the wild type and S39C mutant of γ-S crystallin. (A) Wild type protein, with the sidechain oxygen of S39 shown as a Corey–Pauling–Koltun rendering and (B) the S39Cmutant, with the sulfur atom rendered in CPK. Arrows point to hydrogen bonds betweenneighboring residues and residue 39. Note that the H-bond pattern between theseresidues is not significantly altered in the mutant.

Table 4Alterations in the inter-atom distances upon mutation.

Distance,in Å

2M3T, wildType C–C,distance, Å

S39C mutantC–Cdistance, Å

2M3T, wildtype N–Ndistance, Å

S39C mutantN–Ndistance, Å

Y11–F16 6.12 6.14 6.57 6.98Y11–S/C39 5.50 4.82 5.24 4.74F16–S/C39 6.60 7.34 5.42 5.99

C–C: backbone carbon to carbon distance, and N–N backbone nitrogen atoms distance,both in Å units. 2M3T refers to the PDB code for human γS-crystallin.


80 Å2 by Cys (relative polarity ranking 7, average percentage burialvalue of 3%, van der Waals volume of 86 Å3 and accessible surface areaof 104 Å2) [102]; in other words, even this rather minor alterationweakens the stability of the molecule and makes it a little more hydro-phobic on the surface, causing the tendency to self-aggregate andgenerate light scattering particleswhich, in the lens, can lead to the clin-ical phenotype of lens opacification and cataract. We had earlier showna similar genotype–molecular phenotype correlation with the muta-tions V42M and D26G in human γS-crystallin [89,91], and Bharat et al.[92] have confirmed this in greater detail in V42M, usingNMR spectros-copy; others had shown similar effects with the mutant G18V of thesame molecule [87,88], and DiMauro et al. [103] have shown thateven a subtle alteration such as acetylation of gly1 and lys 2 in humanγD-crystallin promotes aggregation of the otherwise highly solublemolecule. We had earlier argued that the structural integrity of theGreek key fold seen in lens crystallins appears essential for lens trans-parency, and a minor distortion in even one of the Greek key folds inlens crystallins leads to compromise in lens transparency [62]. The pre-liminary results we have described above with S39C appear to be in ac-cordwith this argument.We still need to investigate the ultimate statusof the C residues in the self-aggregation.

6. Section III: gamma crystallins outside the lens

Having looked at the presence of, and the role played by γ-crystallins in the lens, we now move to their presence outside thelens, in other tissues of the eye. While there are many reports for thenon-lenticular presence and functions of α-crystallins, and also somemembers of the β-crystallin family, particularly in the retina, the rolesof γ-crystallins are just getting to be understood [104,105]. Based onthe idea that γS-crystallin is an outlier of the γ family, with some char-acteristics that appear more likely to retain non-lens function, Sinhaet al. [106] looked for the Crygs gene in the mouse eye and found theγS mRNA to be present in the retina as well as the cornea, though notin the iris. They wondered whether γS may have a role as a stress-modulator in the tissues of the eye. Most of these studies so far haveconcentrated on the murine eye, where they are suggested to play therole of stress-modulators during the development of the eye. Sincethen, several papers have appeared which have tried to follow up onthis idea [107,108]. However, all of these studies, while mentioningβγ-crystallins, focus on the roles of specific β-crystallins such as βA3/A1 [108], or βB2 [105], and these too on rodent eyes, rather thanhumans. Andwhen talking aboutγ-crystallins, reference ismade invari-ably to the mixture rather than a single member such as γS, γD or γC.The role of individual members of human γ-crystallins is thus an issueworth researching on. This is particularly worthwhile since we find,while working with clinical colleagues in the pediatric ophthalmic sec-tion at our institute, occasional cases of children diagnosed with bothcongenital cataract and persistent fetal vasculature (PFV, wherein thehyaloid artery involved in nourishing the growth of the eye throughvasculo- and angiogenesis does not regress from the anterior portionof the eye at the appropriate juncture, thus leaving the vasculature topersist and block vision). Zhang et al. [109] had seen the expression ofγ-crystallins in the hyaloid tissue of an infant with microphthalmiaand PFV. Ongoing genetic analysis in one such patient in our center,

who has cataract, PFV andmicrocornea, has revealed a single point mu-tation in human γC crystallin (R48H), but it is not clear what role themutation plays in each one of these conditions. Work of this kind is ex-pected to give us an indication of the role that γ-crystallins play in non-lenticular parts of the eye.

Conflict of interest

None of the authors has any financial disclosures to make.

Acknowledgments

This work was funded by the Hyderabad Eye Research Foundation.AM thanks the Science Academies of India for a summer researchfellowship. We thank Drs. Yogendra Sharma and Rajeev Raman of theCentre for Cellular and Molecular Biology, Hyderabad for the use ofthe circular dichroigraph, and for helpful comments. Also, owing tothe page and reference number limitations, we could not cite all rele-vant original research articles and thereby apologize to the authorswhose publications are not cited.

VPRV and DB conceived the idea, VPRV and AM did much of the ex-perimental work, VPRV and DB analyzed and interpreted the data andwrote themanuscript, IK participated in the analysis,modified theman-uscript, and checked all the references, SC did the protein modelingwork, helped analyze the data, andmodified themanuscript. All the au-thors read and cleared the final manuscript.

References

[1] R.C. Augusteyn, On the growth and internal structure of human lens, Exp. Eye Res.90 (2010) 643–654.

[2] A. Mishra, B. Krishnan, S.S. Swaroop, Y. Sharma, Microbial βγ-crystallins, Prog.Biophys. Mol. Biol. 115 (2014) 42–51.

[3] R.P. Barnwal, M.K. Jobby, K.M. Devi, Y. Sharma, K.V.R. Chary, Solution structure andcalcium binding properties of M-crystallin, a primordial βγ-crystallin fromArchaea, J. Mol. Biol. 386 (2009) 675–689.

[4] E. Serebryany, J.A. King, The βγ-crystallins: native state stability and pathways toaggregation, Prog. Biophys. Mol. Biol. 115 (2014) 32–41.

[5] C. Slingsby, G.J. Wistow, A.R. Clark, Evolution of crystallins for a role in thevertebrate eye lens, Protein Sci. 22 (2013) 367–380.

[6] G.J. Wistow, C. Slingsby, Editorial for special issue: crystallins of the eye, Prog.Biophys. Mol. Biol. 115 (2014) 1–2.

[7] A. Shiels, J.F. Hejtmancik, Genetics of human cataract, Clin. Genet. 84 (2013)120–127.

[8] J.S. Richardson, Beta-sheet topology and the relatedness of proteins, Nature 268(1977) 495–500.

[9] H. Zhao, Y. Chen, L. Rezabkova, Z. Wu, G. Wistow, P. Schuck, Solution properties ofγ-crystallins: hydration of fish and mammal γ-crystallins, Protein Sci. 23 (2014)88–99.

[10] Y. Chen, H. Zhao, P. Schuck, G. Wistow, Solution properties of γ-crystallins:compact structure and low frictional ratio are conserved properties of diverseγ-crystallins, Protein Sci. 23 (2014) 76–87.

[11] M. Delaye, A. Tardieu, Short-range order of crystallin proteins accounts for eye lenstransparency, Nature 302 (1983) 415–417.

[12] A. Tardieu, F. Veretout, B. Krop, C. Slingsby, Protein interactions in the calf eye lens:interactions between beta-crystallins are repulsive whereas in gamma-crystallinsthey are attractive, Eur. Biophys. J. 21 (1992) 1–12.

[13] B. Bax, R. Lapatto, V. Nalini, H. Driessen, P.F. Lindley, D. Mahadevan, T.L. Blundell, C.Slingsby, X-ray analysis of beta B2-crystallin and evolution of oligomeric lensproteins, Nature 347 (1990) 776–780.

[14] R. Jaenicke, C. Slingsby, Lens crystallins and their microbial homologs: structure,stability, and function, Crit. Rev. Biochem. Mol. Biol. 36 (2001) 435–499.

[15] T. Blundell, P. Lindley, L. Miller, D. Moss, C. Slingsby, I. Tickle, B. Turnell, G. Wistow,The molecular structure and stability of the eye lens: X-ray analysis of gamma-crystallin II, Nature 289 (1981) 771–777.

[16] G. Wistow, B. Turnell, L. Summers, C. Slingsby, D. Moss, L. Miller, P. Lindley, T.Blundell, X-ray analysis of the eye lens protein gamma-II crystallin at 1.9 Å resolu-tion, J. Mol. Biol. 170 (1983) 175–202.

[17] A. Basak, O. Bateman, C. Slingsby, A. Pande, N. Asherie, O. Ogun, G.B. Benedek, J.Pande, High-resolution X-ray crystal structures of human gamma D crystallin(1.25 Å) and the R58H mutant (1.15 Å) associated with aculeiform cataract, Mol.Biol. 328 (2003) 1137–1147.

[18] A.G. Purkiss, O.A. Bateman, J.M. Goodfellow, N.H. Lubsen, C. Slingsby, The X-raycrystal structure of human γS-crystallin C-terminal domain, J. Biol. Chem. 277(2002) 4199–4208.

[19] K. Mahendiran, C. Elie, J.C. Nebel, A. Ryan, B.K. Pierscionek, Primary sequence con-tribution to the optical function of the eye lens, Sci. Rep. 4 (2014) 5195.

http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0005http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0005http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0010http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0010http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0015http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0015http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0015http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0020http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0020http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0025http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0025http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0030http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0030http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0035http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0035http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0040http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0040http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0045http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0045http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0045http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0050http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0050http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0050http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0055http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0055http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0060http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0060http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0060http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0065http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0065http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0065http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0070http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0070http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0075http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0075http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0075http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0080http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0080http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0080http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0085http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0085http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0085http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0085http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0090http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0090http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0090http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0095http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0095


[20] H. Zhao, P.H. Brown, M.T. Magone, P. Schuck, The molecular refractive function oflens γ-crystallins, J. Mol. Biol. 411 (2011) 680–699.

[21] M. Hoshino, K. Uesugi, N. Yagi, S. Mohri, J. Regini, B. Pierscionek, Optical propertiesof in situ eye lenses measured with X-ray Talbot interferometry: a novel measureof growth processes, PLoS One 6 (2011) e25140.

[22] B.K. Pierscionek, J.W. Regini, The gradient index lens of the eye: an opto-biologicalsynchrony, Prog. Retin. Eye Res. 31 (2012) 332–349.

[23] T.L. McMeekin, M.L. Groves, N.J. Hipp, Refractive indices of amino acids, proteinsand related substances, in: J. Stekol (Ed.), Amino Acids and Serum Proteins,American Chemical Society, Washington DC, 1964.

[24] B.K. Pierscionek, G. Smith, R.C. Augusteyn, The refractive increments of bovinealpha, beta, and gamma-crystallins, Vis. Res. 27 (1987) 1539–1541.

[25] See the King web page at http://web.mit.edu/king-ab/www/publications/pubs_cat.htm#Lens (for all relevant papers on their structural studies on γ-crystallin).

[26] See the Pande web page at http://www.albany.edu/chemistry/jpande.shtml#recentpub (for relevant papers on their structural studies on γ-crystallin).

[27] See the Gronenborn web page at http://www.structbio.pitt.edu/webusers/amg/?page_id=21 (for their structural studies on γ-crystallin).

[28] See the Martin web page at http://www.faculty.uci.edu/profile.cfm?faculty_id=5277 (for relevant papers on their structural studies on γ-crystallin).

[29] M.S. Kosinsky-Collins, J. King, In vitro unfolding, refolding, and polymerization ofhuman gammaD crystallin, a protein involved in cataract formation, Protein Sci.12 (2003) 480–490.

[30] O.A. Bateman, A. Sarra, S.T. Van Genesan, G. Kappe, N.H. Lubsen, C. Slingsby, Thestability of human acidic beta-crystallin oligomers and hetero-oligomers, Exp.Eye Res. 77 (2003) 409–422.

[31] K.J. Lampi, P.A. Wilmarth, M.R. Murray, L.L. David, Lens β-crystallins: the role ofdeamidation and related modifications in aging and cataract, Prog. Biophys. Mol.Biol. 115 (2014) 21–31.

[32] S.L. Flaugh, M.S. Kosinski-Collins, J.A. King, Contributions of hydrophobic domaininterface interactions to the folding and stability of human gammaD-crystallin,Protein Sci. 14 (2005) 569–581.

[33] I.A. Mills, S.L. Flaugh,M.S. Kosinski-Collins, J.A. King, Folding and stability of the iso-lated Greek key domains of the long-lived human lens proteins gammaD-crystallinand gammaS-crystallin, Protein Sci. 16 (2007) 2427–2444.

[34] T.X. Sun, N.J. Akhtar, J.J. Liang, Thermodynamic stability of human lens recombi-nant alphaA- and alphaB-crystallins, J. Biol. Chem. 274 (1999) 34067–34071.

[35] J.W.Wu,M.E. Chen,W.-S.Wen,W.-A. Chen, C.-T. Li, C.K. Chang, C.H. Lo, H.S. Liu, S.S.Wang, Comparative analysis of human γD-crystallin aggregation under physiolog-ical and low pH conditions, PLoS One 9 (11) (2014) e112309, http://dx.doi.org/10.1371/journal.pone.0112309.

[36] M. Maiti, M. Kono, B. Chakrabarti, Heat induced changes in the conformation ofalpha- and beta-crystallins: unique thermal stability of alpha-crystallin, FEBSLett. 236 (1988) 109–114.

[37] K.J. Lampi, Y.H. Kim, H.P. Bachinger, B.A. Boswell, R.A. Linder, J.A. Carver, T.R.Shearer, L.L. David, D.M. Kapfer, Decreased heat stability and increased chaperonerequirement of modified human betaB1-crystallins, Mol. Vis. 8 (2002) 359–366.

[38] I.A.R. Mills-Henry, Stability, unfolding, and aggregation of the gamma D andgamma S human eye lens crystallins(Ph. D. thesis) Department of Biology,Massachusetts Institute of Technology, 2007.

[39] J.M.Hemmingsen, K.M. Gernert, J.S. Richardson,D.C. Richardson, The tyrosine corner:a feature of most Greek key beta-barrel proteins, Protein Sci. 3 (1994) 1927–1937.

[40] Z. Yang, Z. Xia, T. Huynh, J.A. King, R. Zhou, Dissecting the contributions ofβ-hairpin tyrosine pairs to the folding and stability of long lived γD-crystallins,Nanoscale 6 (2014) 1797–1807.

[41] F. Kong, J. King, Contributions of aromatic pairs to the folding and stability of long-lived human γD-crystallin, Protein Sci. 20 (2011) 513–528.

[42] N. Schafheimer, J. King, Tryptophan cluster protects human γD-crystallin from ul-traviolet radiation-induced photoaggregation in vitro, Photochem. Photobiol. 89(2013) 1106–1115.

[43] N. Schafheimer, Z. Wang, K. Schey, J. King, Tyrosine/cysteine cluster sensitizinghuman γD-crystallin to ultraviolet radiation-induced photoaggregation in vitro,Biochemistry 53 (2014) 979–990.

[44] B. Wang, C. Yu, Y.B. Xi, H.C. Cai, J. Wang, S. Zhou, S. Zhou, Y. Wu, Y.B. Yan, X. Ma, L.Xie, A novel CRYGD mutation (p.Trp43Arg) causing autosomal dominantcongenital cataract in a Chinese family, Hum. Mutat. 32 (2010) E1939–E1947.

[45] D.A. Stephan, E. Gillanders, D. Vanderveen, D. Freas-Lutz, G. Wistow, A.D.Baxevanis, C.M. Robbins, A.V. Auken, M.I. Quesenberry, J.B. Wilson, S.H.H. Juo,J.M. Trent, L. Smith, M.J. Brownstein, Progressive juvenile-onset punctate cataractscaused bymutation of the gamma D-crystallin gene, Proc. Natl. Acad. Sci. U. S. A. 96(1999) 1008–1012.

[46] F. Gu, R. Li, X.X. Ma, L.S. Shi, S.Z. Huang, X. Ma, A missense mutation in the gammaD-crystallin gene CRYGD associated with autosomal dominant congenital cataractin a Chinese family, Mol. Vis. 12 (2006) 26–31.

[47] A. Pande, D. Gillot, J. Pande, The cataract associated R14Cmutant of human gammaD-crystallin shows a variety of intermolecular disulfide cross-links: a Raman spec-troscopic study, Biochemistry 48 (2009) 4937–4945.

[48] L.Y. Zhang, B. Gong, J.P. Tong, D.S. Fan, S.W. Chiang, D. Lou, D.S. Lam, G.H. Yam, C.P.Pang, A novel gamma D-crystallin mutation causes mild changes in protein prop-erties but leads to congenital coralliform cataract, Mol. Vis. 15 (2009) 1521–1529.

[49] L. Hilal, E. Nandrot, M. Belmekki, M. Chefchaouni, S. El Bacha, B. Benazzouz, Y.Hajaji, O. Gribouval, J. Dufier, M. Abitbol, A. Berraho, Evidence of clinical and genet-ic heterogeneity in autosomal dominant congenital cerulean cataracts, OphthalmicGenet. 23 (2002) 199–208.

[50] E. Nandrot, C. Slingsby, A. Basak, M. Cherif-Chefchaouni, B. Benazzouz, Y. Hajaji, S.Boutayeb, O. Gribouval, L. Arbogast, A. Berraho, M. Abitbol, L. Hilal, Gamma-D

crystallin gene (CRYGD) mutation causes autosomal dominant congenital ceruleancataracts, J. Med. Genet. 40 (2003) 262–267.

[51] D.S. Mackay, U.P. Andley, A. Shiels, A missense mutation in the gamma D crystallingene (CRYGD) associated with autosomal dominant “coral-like” cataract linked tochromosome 2q, Mol. Vis. 10 (2004) 155–162.

[52] W.Z. Xu, S. Zheng, S.J. Xu, W. Huang, K. Yao, S.Z. Zhang, Autosomal dominantcoralliform cataract related to a missense mutation of the gamma D-crystallingene, Chin Med J (Engl.) 117 (2004) 727–732.

[53] X. Shentu, K. Yao, W. Xu, S. Zheng, S. Hu, X. Gong, Special fasciculiform cataractcaused by a mutation in the gamma D-crystallin gene, Mol. Vis. 10 (2004)233–239.

[54] A.O. Khan, M.A. Aldahmesh, F.E. Ghadhfan, S. Al-Mesfer, F.S. Alkuraya, Founderheterozygous P23T CRYGD mutation associated with cerulean (and coralliform)cataract in 2 Saudi families, Mol. Vis. 15 (2009) 1407–1411.

[55] G. Yang, C. Xiong, S. Li, Y. Wang, J.A. Zhao, A recurrent mutation in CRYGD is asso-ciated with autosomal dominant congenital coralliform cataract in two unrelatedChinese families, Mol. Vis. 17 (2011) 1085–1089.

[56] A. Pande, O. Annunziata, N. Asherie, O. Ogun, G.B. Benedek, J. Pande, Decrease inprotein solubility and cataract formation caused by the Pro23 to Thr mutation inhuman gamma D-crystallin, Biochemistry 44 (2005) 2491–2500.

[57] J.J. McManus, A. Lomakin, O. Ogun, A. Pande, M. Basan, J. Pande, G.B. Benedek,Altered phase diagram due to a single point mutation in human gammaD-crystallin, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 16856–16861.

[58] A. Pande, J. Zhang, P.R. Banerjee, S.S. Puttamadappa, A. Shekhtman, J. Pande, NMRstudy of the cataract-linked P23T mutant of human gammaD-crystallin showsminor changes in hydrophobic patches that reflect its retrograde solubility,Biochem. Biophys. Res. Commun. 382 (2009) 196–199.

[59] A. Pande, K.S. Ghosh, P.R. Banerjee, J. Pande, Increse in surface hydrophobicity ofthe cataract-associated P23T mutant of human gammaD-crystallin is responsiblefor its dramatically lower, retrograde solubility, Biochemistry 49 (2010)6122–6129.

[60] P.R. Banerjee, S.S. Puttamadappa, A. Pande, A. Shekhtman, J. Pande, In-creased hydrophobicity and decreased backbone flexibility explain thelower solubility of a cataract-linked mutant of γD-crystallin, J. Mol. Biol.412 (2011) 647–659.

[61] W. Sun, X. Xiao, S. Li, X. Guo, Q. Zhang, Mutation analysis of 12 genes in Chinesefamilies with congenital cataracts, Mol. Vis. 17 (2011) 2197–2206.

[62] V.P.R. Vendra, G. Agarwal, S. Chandani, V. Talla, N. Srinivasan, D. Balasubramanian,Structural integrity of the Greek key motif in βγ-crystallins is vital for central eyelens transparency, PLoS One 8 (2013) e70336.

[63] L. Wang, X. Chen, Y. Lu, J. Wu, B. Yang, X. Sun, A novel mutation in gamma D-crystallin associated with autosomal dominant congenital cataract in a Chinesefamily, Mol. Vis. 17 (2011) 804–809.

[64] S. Kmoch, J. Brynda, B. Asfaw, K. Bezouska, P. Novak, P. Rezácova, L. Ondrova, M.Filipec, J. Sedlacek, M. Elleder, Link between a novel human gamma D-crystallinallele and a unique cataract phenotype explained by protein crystallography,Hum. Mol. Genet. 9 (2000) 1779–1786.

[65] A. Pande, J. Pande, N. Asherie, A. Lomakin, O. Ogun, J. King, Crystal cataracts: humangenetic cataract caused by protein crystallization, Proc. Natl. Acad. Sci. U. S. A. 98(2001) 6116–6120.

[66] S.T. Santhiya, G.S. Kumar, P. Sudhakar, N. Gupta, N. Klopp, T. Illig, T. Soker, M. Groth,M. Platzer, P.M. Gopinath, J. Graw, Molecular analysis of cataract families in India:new mutations in the CRYBB2 and GJA3 genes and rare polymorphisms, Mol. Vis.16 (2010) 1837–1847.

[67] A. Santana, M.Waiswol, E.S. Arcieri, Cabral de Vasconcellos JP, M. Barbosa de Melo,Mutation analysis of CRYAA, CRYGC, and CRYGD associated with autosomal dom-inant congenital cataract in Brazilian families, Mol. Vis. 15 (2009) 793–800.

[68] E. Heon, M. Priston, D.F. Schorderet, G.D. Billingsley, P.O. Girard, N. Lubsen, F.L.Munier, The gamma-crystallins and human cataracts: a puzzle made clearer, Am.J. Hum. Genet. 65 (1999) 1261–1267.

[69] F. Li, S. Wang, C. Gao, S. Liu, B. Zhao, M. Zhang, S. Huang, S. Zhu, X. Ma, MutationG61C in the CRYGD gene causing autosomal dominant congenital coralliform cat-aracts, Mol. Vis. 14 (2008) 378–386.

[70] W. Zhang, H.C. Cai, F.F. Li, Y.B. Xi, X. Ma, Y.B. Yan, The congenital cataract-linkedG61C mutation destabilizes gamma D-crystallin and promotes non-native aggre-gation, PLoS One 6 (2011) e20564.

[71] M. Roshan, P.H. Vijaya, G.R. Lavanya, P.K. Shama, S.T. Santhiya, J. Graw, P.M.Gopinath, K.A. Satyamoorthy, A novel human CRYGD mutation in a juvenile auto-somal dominant cataract, Mol. Vis. 16 (2010) 887–896.

[72] F. Ji, J. Jung, A.M. Gronenborn, Structural and biochemical characterization of thechildhood cataract associated R76S mutant of human gamma D-crystallin, Bio-chemistry 51 (2012) 2588–2596.

[73] O.M. Messina-Baas, L.M. Gonzalez-Huerta, S.A. Cuevas-Covarrubias, Two affectedsiblings with nuclear cataract associated with a novel missense mutation in theCRYGD gene, Mol. Vis. 12 (2006) 995–1000.

[74] P.R. Banerjee, A. Pande, J. Patrosz, G.M. Thurston, J. Pande, Cataract associated mu-tant E107A of human gamma D-crystallin shows increased attraction to alpha-crystallin, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 574–579.

[75] L. Hansen, W. Yao, H. Eiberg, K.W. Kjaer, K. Baggesen, J.F. Hejtmancik, T. Rosenberg,Genetic heterogeneity in microcornea-cataract: five novel mutations in CRYAA,CRYGD, and GJA8, Invest. Ophthalmol. Vis. Sci. 48 (2007) 3937–3944.

[76] R.R. Devi, W. Yao, P. Vijayalakshmi, Y.V. Sergeev, P. Sundaresan, J.F. Hejtmancik,Crystallin gene mutations in Indian families with inherited pediatric cataract,Mol. Vis. 14 (2008) 1157–1170.

[77] S.T. Santhiya, M. Shyam Manohar, D. Rawlley, P. Vijayalakshmi, P.Namperumalsamy, P.M. Gopinath, J. Loster, J. Graw, Novel mutations in the

http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0100http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0100http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0105http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0105http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0105http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0110http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0110http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0115http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0115http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0115http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0120http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0120http://web.mit.edu/king-ab/www/publications/pubs_cat.htm#Lenshttp://web.mit.edu/king-ab/www/publications/pubs_cat.htm#Lenshttp://www.albany.edu/chemistry/jpande.shtml#recentpubhttp://www.albany.edu/chemistry/jpande.shtml#recentpubhttp://www.structbio.pitt.edu/webusers/amg/?page_id=21http://www.structbio.pitt.edu/webusers/amg/?page_id=21http://www.faculty.uci.edu/profile.cfm?faculty_id=5277http://www.faculty.uci.edu/profile.cfm?faculty_id=5277http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0145http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0145http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0145http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0150http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0150http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0150http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0155http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0155http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0155http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0160http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0160http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0160http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0165http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0165http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0165http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0170http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0170http://dx.doi.org/10.1371/journal.pone.0112309http://dx.doi.org/10.1371/journal.pone.0112309http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0180http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0180http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0180http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0185http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0185http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0185http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0190http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0190http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0190http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0195http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0195http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0200http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0200http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0200http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0205http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0205http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0210http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0210http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0210http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0215http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0215http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0215http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0220http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0220http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0220http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0225http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0225http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0225http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0225http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0225http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0230http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0230http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0230http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0235http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0235http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0235http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0240http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0240http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0240http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0245http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0245http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0245http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0245http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0250http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0250http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0250http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0250http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0255http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0255http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0255http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0260http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0260http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0260http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0265http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0265http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0265http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0270http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0270http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0270http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0275http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0275http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0275http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0280http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0280http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0280http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0285http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0285http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0285http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0290http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0290http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0290http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0290http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0295http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0295http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0295http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0295http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0300http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0300http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0300http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0300http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0305http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0305http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0310http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0310http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0310http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0315http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0315http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0315http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0320http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0320http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0320http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0320http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0325http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0325http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0325http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0330http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0330http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0330http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0330http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0335http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0335http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0335http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0340http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0340http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0340http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0345http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0345http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0345http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0350http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0350http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0350http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0355http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0355http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0355http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0360http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0360http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0360http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0365http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0365http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0365http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0370http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0370http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0370http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0375http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0375http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0375http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0380http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0380http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0380http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0385http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0385


gamma-crystallin genes cause autosomal dominant congenital cataracts, J. Med.Genet. 39 (2002) 352–358.

[78] V. Talla, N. Srinivasan, D. Balasubramanian, Visualization of in situ intracellularaggregation of two cataract-associated human γ-crystallin mutants: lose a tail,lose transparency, Invest. Ophthalmol. Vis. Sci. 49 (2008) 3483–3490.

[79] L.Y. Zhang, G.H. Yam, D.S. Fan, P.O. Tam, D.S. Lam, C.P. Pang, A novel deletionvariant of gammaD-crystallin responsible for congenital nuclear cataract, Mol.Vis. 13 (2007) 2096–2104.

[80] Z. Ren, A. Li, B.S. Shastry, T. Padma, R. Ayyagari, M.H. Scott, M.M. Parks, M.I. Kaiser-Kupfer, J.F. Hejtmancik, A 5-base insertion in the gamma C-crystallin gene isassociated with autosomal dominant variable zonular pulverulent cataract, Hum.Genet. 106 (2000) 531–537.

[81] K. Yao, C. Jin, N. Zhu, W. Wang, R. Wu, J. Jiang, X. Shentu, A nonsense mutation inCRYGC associated with autosomal dominant congenital nuclear cataract in aChinese family, Mol. Vis. 14 (2008) 1272–1276.

[82] L. Zhang, S. Fu, Y. Ou, T. Zhao, Y. Su, P. Liu, A novel nonsense mutation in CRYGC isassociated with autosomal dominant congenital nuclear cataracts andmicrocornea, Mol. Vis. 15 (2009) 276–282.

[83] L.M. Gonzalez-Huerta, O.M. Messina-Baas, S.A. Cuevas-Covarrubias, A family withautosomal dominant primary congenital cataract associated with a CRYGCmutation: evidence of clinical heterogeneity, Mol. Vis. 13 (2007) 1333–1338.

[84] M. Kumar, T. Agarwal, S. Khokhar, P. Kaur, T.S. Roy, R. Dada, Mutation screeningand genotype phenotype correlation of alpha-crystallin, gamma-crystallin andGJA8 gene in congenital cataract, Mol. Vis. 17 (2011) 693–707.

[85] X.Q. Li, H.C. Cai, S.Y. Zhou, J.H. Yand, Y.B. Xi, X.B. Gao, W.J. Zhao, P. Li, G.Y. Zhao, Y.Tong, F.C. Bao, Y. Ma, S. Wang, Y.B. Yan, C.L. Lu, X. Ma, A novel mutation impairingthe tertiary structure and stability of human γC-crystallin (CRYGC) leads to cata-ract formation in humans and zebrafish lens, Hum. Mutat. 33 (2012) 391–401.

[86] H. Sun, Z. Ma, Y. Li, B. Liu, Z. Li, X. Ding, Y. Gao, W. Ma, X. Tang, X. Li, Y. Shen,Gamma-S crystallin gene (CRYGS) mutation causes dominant progressive corticalcataract in humans, J. Med. Genet. 42 (2005) 706–710.

[87] Z. Ma, G. Piszczek, P.T. Wingfield, Y.V. Sergeev, J.F. Hejtmancik, The G18V CRYGSmutation associatedwith human cataracts increases gamma S-crystallin sensitivityto thermal and chemical stress, Biochemistry 48 (2009) 7334–7341.

[88] W.D. Brubaker, R.W. Martin, 1H, 13C, and 15N assignments of wild-type humanγS-crystallin and its cataract-related variant γS-G18V, Biomol. NMR Assign. 6(2012) 63–67.

[89] S. Karri, R.B. Kasetti, V.P.R. Vendra, S. Chandani, D. Balasubramanian, Structuralanalysis of the mutant protein D26G of human γS-crystallin associated withCoppock cataract, Mol. Vis. 19 (2013) 1231–1237.

[90] V. Vanita, J.R. Singh, D. Singh, R. Varon, K. Sperling, Novel mutation in the gamma-Scrystallin gene causing autosomal dominant cataract, Mol. Vis. 15 (2009)476–481.

[91] V.P.R. Vendra, S. Chandani, D. Balasubramanian, The mutation V42M distorts thecompact packing of the human gamma-S-crystallin molecule, resulting in congen-ital cataract, PLoS One 7 (2012) e51401.

[92] S.V. Bharat, A. Shekhtman, J. Pande, The cataract-associated V41M mutant ofhuman γS-crystallin shows specific structural changes that directly enhance

local surface hydrophobicity, Biochem. Biophys. Res. Commun. 443 (2014)110–114.

[93] Z. Yang, Q. Li, S. Zhu, X. Ma, A G57Wmutation of CRYGS associated with autosomaldominant pulverulent cataracts in a Chinese family, Ophthalmic Genet. (2013)http://dx.doi.org/10.3109/13816810.2013.865761.

[94] F.J. Giblin, L.L. David, P.A. Wilmarth, V.R. Leverenz, M.F. Simpanya, Shotgun prote-omic analysis of S-thiolation sites of guinea pig lens nuclear crystallins followingoxidative stress in vivo, Mol. Vis. 19 (2013) 267–280.

[95] C.N. Kingsley, W.D. Brubaker, S. Markovic, A. Diehl, A.J. Brindley, H. Oschkinat, R.W.Martin, Preferential and specific binding of human αB-crystallin to a cataract-related variant of γS-crystallin, Structure 21 (2013) 2221–2227.

[96] C.G. Rosen, G. Weber, Dimer formation from 1-amino-8-naphthalenesulfonate cat-alyzed by bovine serum albumin, a new fluorescent molecule with exceptionalbinding properties, Biochemistry 8 (1969) 3915–3920.

[97] M. Sutter, S. Oliveira, N.N. Sanders, B. Lucas, A. Van Hoek, M.A. Hink, A.J. Visser, S.C.De Smedt, W.E. Hennink, W. Jiskoot, Sensitive spectroscopic detection of large anddenatured protein aggregates in solution by use of fluorescent dye Nile Red, J.Fluoresc. 17 (2007) 181–192.

[98] R.F. Greene Jr., C.N. Pace, Urea and guanidine hydrochloride denaturation of ribo-nuclease, lysozyme, α -chymotrypsin and β-lactoglobulin, J. Biol. Chem. 249(1974) 5388–5393.

[99] A.C. Clark, J.F. Sinclair, T.O. Baldwin, Folding of bacterial luciferase involves a non-native heterodimeric intermediate in equilibrium with the native enzyme andthe unfolded subunits, J. Biol. Chem. 268 (1993) 10773–10779.

[100] K. Arnold, L. Bordoli, J. Kopp, T. Schwede, The SWISS-MODEL Workspace: a web-based environment for protein structure homology modeling, Bioinformatics 22(2006) 195–201.

[101] D. Van Der Spoel, E. Lindahl, B. Hess, G. Groenhof, A.E. Mark, H.J.C. Berendsen,GROMACS: fast, flexible and free, J. Comput. Chem. 26 (2005) 1701–1718.

[102] http://www.proteinsandproteomics.org/content/free/tables_1/table08.pdf&gt.[103] M.A. DiMauro, S.K. Nandi, C.T. Raghavan, R.K. Kar, B. Wand, A. Bhunia, R.H. Nagaraj,

A. Biswas, Acetylation of Gly1 and Lys2 promotes aggregation of human γD-crystallin, Biochemistry 53 (2014) 7269–7282.

[104] C. Slingsby, G.J. Wistow, Functions of crystallins in and out of lens: roles in elongat-ed and post-mitotic cells, Prog. Biophys. Mol. Biol. 115 (2014) 52–67.

[105] S. Thanos, M.R.R. Bohm, M.M. zu Horste, V. Prokosch-Willing, M. Hennig, D. Bauer,A. Heiligenhaus, Role of crystallins in ocular neuroprotection and axonal regenera-tion, Prog. Retin. Eye Res. 42 (2014) 145–161.

[106] D. Sinha, N. Esumi, C. Jaworski, C.A. Kozak, E. Pierce, G. Wistow, Cloning and map-ping the mouse Crygs gene and non-lens expression of γS-crystallin, Mol. Vis. 4(1998) 8.

[107] S.E. Jones, C. Jomary, J. Grist, J. Makwana, M.J. Neal, Retinal expression of γS-crystallins in the mouse, Invest. Ophthalmol. Vis. Sci. 40 (1999) 3017–3020.

[108] J. Xi, R. Farjo, S. Yoshida, T.S. Kern, A. Swaroop, U.P. Andley, A comprehensive anal-ysis of the expression of crystallins in mouse retina, Mol. Vis. 9 (2003) 410–419.

[109] C. Zhang, P. Gehlbach, C. Gongora, M. Cano, R. Fariss, S. Hose, A. Nath, W.R. Green,M.F. Goldberg, J.S. Zigler Jr., D. Sinha, A potential role for β- and γ-crystallins in thevascular remodelling of the eye, Dev. Dyn. 234 (2005) 36–47.

http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0385http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0385http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0390http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0390http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0390http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0395http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0395http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0395http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0400http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0400http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0400http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0400http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0405http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0405http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0405http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0410http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0410http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0410http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0415http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0415http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0415http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0420http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0420http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0420http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0425http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0425http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0425http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0425http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0430http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0430http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0430http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0435http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0435http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0435http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0440http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0440http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0440http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0440http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0440http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0440http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0445http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0445http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0445http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0450http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0450http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0450http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0455http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0455http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0455http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0460http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0460http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0460http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0460http://dx.doi.org/10.3109/13816810.2013.865761http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0470http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0470http://refhub.elsevier.com/S0304-4165(15)00169-5/rf0470http://refh

biochimica et biophysica acta -...

Documents