Molecular Vision 2007; 13:1758-1768 <http://www.molvis.org/molvis/v13/a196/> Received 3 July 2007 | Accepted 18 September 2007 | Published 19 September 2007

Download Reprint

The role of the conserved COOH-terminal triad in αA-crystallin aggregation and functionality

Ying Li,¹ Karl R. Schmitz,² John C. Salerno,¹ Jane F. Koretz¹
 
 

¹Biochemistry and Biophysics Program, Rensselaer Polytechnic Institute, Troy, New York; ²Biochemistry and Biophysics Department, University of Pennsylvania, School of Medicine, Philadelphia, PA

Correspondence to: Jane F. Koretz, Ph.D., Biochemistry and Biophysics Program, Rensselaer Polytechnic Institute, Science Center, 110 8th Street, Troy, NY, 12180-3590; Phone: (518) 276-6492; FAX: (518) 276-2344; email: koretj@rpi.edu

Abstract

Purpose: The sequentially variable COOH-terminal region of small heat shock protein superfamily members usually contains a conserved IXI/V feature where X is typically a proline. When present in solved sHsp crystal structures (e.g. MjHsp16.5 and wheat Hsp16.9), this short sequence forms an isolated β strand apparently involved in the alignment of dimers into larger oligomers. Because it is a common feature of many sHsp family members, it is possible that this triad has a similar role in αA-crystallin. This study was undertaken to determine the contribution of this conserved triad to the quaternary structure and function of αA-crystallin.

Methods: A series of site-directed mutants was generated in both wild type αA and in an αA deletion mutant lacking the NH₂-terminal residues 1-50. After overexpression and purification, each protein's oligomer size was characterized by size-exclusion fast protein liquid chromatography (FPLC), thermal transition temperature by non-denaturing composite gel electrophoresis, and chaperone activity by the inhibition of DL-dithiothreitol (DTT)-induced insulin aggregation.

Results: Using the αA-crystallin NH₂-deletion mutant, the hydrophobic triad was changed from IPV to TPT, GPG, IGV, ITV, or GGG. All six D51 mutants associated into tetramers with small amounts of dimer and monomer also present. Chaperone-like activity was reduced but not eliminated in some of these triad mutants with GGG and ITV the most strongly affected. Similar modifications to wild type αA-crystallin (IPV to ITV, IGV, or GGG) restored oligomer sizes similar, but not identical to, native αA-crystallin, with additional small amounts of tetramer and dimer. Interestingly, equivalent mutants of wild type αA-crystallin did not have reduced chaperone-like activity but differed considerably in their thermal transition temperatures.

Conclusions: The conserved COOH-terminal triad does not appear to have a strong effect on the steady-state aggregation of wild type αA-crystallin or its 50-residue deletion mutant at 25 °C. However, it can exert a considerable effect on chaperone-like activity in the absence of the NH₂-terminal 50-residue sequence extension and can influence the thermal transition temperature in its presence. These results suggest that the conserved triad in αA-crystallin contributes to the stability of higher order oligomers but is not essential for the formation of tetramers.

Introduction

Crystallins are the predominant proteins in the mammalian eye lens, where their high concentration contributes to lens refractive power [1,2]. The largest of these, α-crystallin, belongs to the family of small heat shock proteins (sHsps) [3]. In the lens, it is a dynamic multimeric complex composed of two types of homologous subunits, αA-crystallin and αB-crystallin, each of which has molecular mass of about 20 kDa [4,5]. This complex not only contributes to lens refractive power but also assists in maintaining transparency [6-9]. The α-crystallins, particularly αB-crystallin, are also found individually in nonlenticular tissues. Whether present as hetero- or homo-oligomers, however, they have the capacity to respond to several different environmental stress situations such as heat shock by chaperone-like or other activities to provide protection at the cellular level [10-14]. Additionally, αB-crystallin has been shown to be associated with neurological disorders such as Alzheimer disease, Alexander disease, and amyotrophic lateral sclerosis (ALS) and to protect against apoptosis [15-27].

Small Hsp subunits can range in size from 11 kDa to over 40 kDa, but all share a conformationally conserved α-crystallin domain (core region), which consists of two β-sheets held together by extensive hydrophobic interactions [28,29]. This core region contributes to sHsp resistance to thermal denaturation and has been highly conserved throughout the superfamily's evolution [29-32]. In contrast, the NH₂- and COOH-terminal extensions outside the core region vary in length, sequence, and hydropathicity, and it is likely that these differences provide the basis for sHsp structural and functional diversity [29]. Earlier studies have established a critical role for the NH₂-terminal region in the assembly of sHsps into high molecular masses while the COOH-terminal sequence seems concerned primarily with activity [33-36].

In spite of considerable variation in the sequence of the COOH-terminal region of sHsps, it usually contains a conserved IXI/V triad where X is typically a proline [28]. The IXI/V motif undergoes intermolecular hydrophobic interactions with a β-strand in the α-crystallin domain during the oligomeric assembly of wheat Hsp16.9 and Mj Hsp16.5 [37,38] and may have a similar function in mammalian sHsps. Mutational studies of the conserved isoleucine in the IXI/V motif and of truncations spanning this region in some bacterial and plant sHSPs show a dramatic reduction in complex size and solubility and a loss of chaperone activity [39,40] while COOH-terminal truncations in mammalian α-sHSPs have been shown to be associated with myofibrilllar myopathy and cataract [41-43]. Pasta et al. [44] constructed two triad mutants of α-crystallin, αA-gxg and αB-gxg, mutating the isoleucine/valine residues in the triad to glycine to investigate the role of this conserved triad. They found that the modified proteins retained their capacity to oligomerize and showed enhanced chaperone-like activity. In this study, a series of triad mutants with differing degrees of conformational freedom and hydrophobicity were generated in both the wild type and NH₂-terminal truncated versions of αA-crystallin to determine the possible contribution of this conserved triad to both quaternary structure and functionality.

Methods

Plasmid construction

A linker-NH₂-terminal deletion mutation was introduced by polymerase chain reaction (PCR) using bovine α-crystallin as the template. Special primers were designed to incorporate the NdeI and XhoI restriction enzyme recognition sites, start codon, stop codon, and a hydrophilic linker sequence (MetHVDGGSGSGGSGGSA) into the PCR product. The primer sequences used were the following: upstream: 5'CAT ATG CAT GTC GAC GGT GGT TCC GGT TCC GGT GGT TCC GGT GGT AGC GCT 3'; downstream: 5'GCT TTG TTA GCA GCT CGA GCC TTA GGA CGA G 3'.

Each mutated gene was ligated into the pET20b vector (Novagen, Madison, WI). Site-directed mutations leading to single amino acid exchanges in the COOH-terminus of αA-crystallin were introduced by primer-based mutagenesis (Quick-Change Site-Directed Mutatagensis Kit; Stratagene, Los Angeles, CA). Mutagenesis was performed based on the manufacturer's instructions and some modifications. The specific NH₂-terminal deletion and COOH-terminal changes are presented in Figure 1. Coding sequences for the αA-wild type, its linker-NH₂-terminal deletion mutants, and COOH-terminal mutants were confirmed by automated DNA sequencing. (CFG, SUNY Albany, NY)

Protein expression and purification

E. coli BL21 (DE3) was freshly transformed with expression plasmids, and inoculated cultures were grown at 37 °C until the A₆₀₀ rose to around 0.5. Expression was induced by the addition of isopropyl thio-β-D-galactoside (0.5 mM). The cultures were grown for an additional three hours before cells were harvested and resuspended in ice-cold lysis buffer (50 mM Tris, 1 mM EDTA, 100 mM NaCl, pH 8.0). Cell lysis and supernatant preparation were performed according to Horwitz et al. [14]. Proteins were purified under native conditions first by ion exchange chromatography using Hiprep 16/10 Q XL column (Amersham/Pharmacia, Piscataway, NJ) that had been equilibrated with Tris buffer (20 mM Tris, 100 mM NaCl, pH 8.0) and then used a NaCl gradient (100 mM to 1000 mM) at a speed of 5 ml per min. The fractions that contained the most pure proteins were pooled together and concentrated with a Centricon spin column to about 1-2 ml, then further purified by gel filtration using a HiPrep Sephacryl S-400 column. The column was equilibrated with Tris buffer (20 mM Tris, 100 mM NaCl, pH8.0), and elution performed at roughly approximate to 1.0 ml/min. Each mutant was overexpressed and treated to a similar purification scheme with only the fractions containing homogenous protein being removed for subsequent analysis. The purity of the mutants was checked using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The samples were diluted 1:1 with Laemmli sample buffer (62.5 mM Tris-HCl, pH 6.8, 2% SDS, 25% glycerol, 0.01% bromophenol blue) and boiled for about 10 min before applied to the SDS-PAGE gels. The gels were run in electrophoresis buffer (25 mM Tris, 192 mM glycine, 0.1% SDS, pH 8.3) at 200 Volts for approximately 60 min. After the run, the gels were removed from the apparatus and stained in coomassie blue R-250 solution for 20 min. The stain was then poured off and the gels were washed with dH₂O and immersed in destain solution (10% methanol, 10% glacial acetic acid, 80% dH₂O).

Oligomer size

The sizes of the COOH-terminal mutants of the αA-crystallin wild-type and COOH-terminal mutants of linker-d51 D51 were determined using a Superose 6 HR 10/30 and Superose 12 HR 10/30 gel exclusion column (Amersham-Pharmacia), respectively, both running at 0.3 ml/min in 20 mM Tris, pH 8.0, and 100 mM NaCl. Oligomer size determination of unknown proteins are made by comparing the ratio of Ve/Vo for the protein in question to the Ve/Vo of protein standards of known molecular weight (Ve is the elution volume and Vo is the void volume). The Vo is based on the volume of effluent required for the elution of a large molecule, blue dextran 2,000,000. Standards used were thyroglobulin 669,000; apoferritin 443000; β-amylase 200,000; alcohol dehydrogenase 150,000; bovine serum albumin (BSA) 66,000; carbonic anhydrase 29,000; and cytochrome C 12,400 (Sigma, St. Louis, MO). Y-Axis intensities were normalized by Origin 6.1 software (Northampton, MA) so that the total area beneath the curves for each protein variation was equivalent.

Nondenaturing composite gel

Nondenaturing composite gels were prepared using a modification of the method of Moulin et al. [45]. The resolving gel was prepared to a final concentration of 3% acrylamide and 0.7% agarose (Fisher, Pittsburgh, PA) in 375 mM Tris-HCl buffer, pH 8.8 in a Mini Protean II Gel Electrophoresis Cell (Bio-Rad, Hercules, CA) with frosted inner glass plates. The stacking gel was prepared to a final concentration of 2% acrylamide and 0.7% agarose in a 125 mM Tris-HCl buffer, pH 6.8. Samples of native αA-crystallin and all the chimera were each incubated for 1 h at RT, 35 °C, 40 °C, 45 °C, 50 °C, 55 °C, 60 °C, 65 °C, and 70 °C, then allowed to cool back to RT for 1 h as previously described [46]. Before incubation, the samples were each diluted to a concentration of 50-75 μM with imidazole buffer (50 mM imidazole, 100 mM NaCl, pH 8.0), which is more stable than Tris buffer during temperature changes. Each sample (4 μg) was added to 5X sample buffer (312.5 mM Tris-HCl, 50% glycerol, 0.05% bromophenol blue, pH 6.8) and run at 200 V for 30 min at 4 °C with 2 μg of apoferritin (MW: 443,000) run as a standard. Following the run, the gels were stained and destained using the same method as was used for SDS-PAGE gels.

Chaperone-like activity assays

The chaperone-like activity of the mutant proteins was analyzed with the DL-dithiothreitol (DTT)-induced insulin denaturation method at room temperature (approximately 22 °C). Bovine insulin (Sigma) was denatured with DTT, and the chaperone-like activity of the sample was represented by the difference in turbidity of insulin with DTT only and insulin with DTT and the sample. Lower turbidity suggests that aggregation of denatured insulin was reduced by the sample protein, thus, a lower turbidity is correlated with a higher chaperone-like activity. Both mutant protein and insulin were dialyzed in the buffer (50 mM imidazole, 100 mM NaCl, pH 7.9), and concentrations were determined by spectrophotometry. The desired volumes of the testing sample protein and insulin were calculated according to their concentration. The final concentrations of insulin and DTT were 60 μM and 20 μM, respectively, in the 200 μl reaction. The denaturing agent DTT was added just before the starting point of the measurement. A 96 well plate was used to perform the assay. Light scattering (OD) was read at 360 nm at room temperature with the SPECTRA MAX 190 every 30 s for 60 min.

Results

In this study, two sets of bovine αA-crystallin mutants were constructed to test the role of the IPV triad [28,47]. The first series of mutants were termed D51 mutants [48] because the first 50 amino acid residues of the NH₂-terminal region were deleted up to the beginning of the sHsp core region consensus sequence, and a mixed serine/glycine leader sequence was added [49] to improve stability. This set includes D51-αA, which has the normal IPV triad in the COOH-terminus and five triad variations to test the importance of triad flexibility and hydrophobicity: D51-TPT, D51-GPG, D51-ITV, D51-IGV, and D51-GGG. The second series of mutants were constructed based on results obtained from the first set as described the following section. The triad at the COOH-terminal of wild type αA-crystallin was changed either to ITV, IGV, or GGG and termed wt-ITV, wt-IGV, or wt-GGG, respectively, altering the COOH-terminal extension flexibility and polarity of the native sequence.

Triad mutants of D51 αA-crystallin

All six mutants were successfully overexpressed and recovered predominantly in the soluble fraction when host cells were extracted by lysozyme treatment. Sequential chromatography over ion exchange and gel exclusion columns was sufficient to purify D51-αA-crystallin and its mutants to apparent homogeneity. Purity of the mutant αA-crystallins was examined by SDS-PAGE. (Figure 2).

The oligomeric sizes of D51-αA-crystallin and its various COOH-terminal triad mutants were obtained by molecular sieve fast protein liquid chromatography (FPLC) of purified proteins (Figure 3). Plotting the logarithms of the known molecular weights of protein standards versus their respective Ve/Vo values produces a linear calibration curve, which is used to calculate the molecular weights of the unknown proteins. All six D51 mutants exhibited primary peaks corresponding either to tetramers (D51-αA, D51-IGV, D51-GPG, and D51-TPT) or trimers (D51-ITV and D51-GGG) with broad downstream shoulders. Additionally, a small peak corresponding to approximately 15 kDa was observed in D51-αA-crystallin, D51-TPT, D51-ITV, and D51-GGG mutants.

Assays of chaperone-like activity were performed by monitoring the ability of purified αA-crystallin and the six D51 triad variations to prevent the denaturation and aggregation of insulin induced by adding DTT at room temperature. The ratio of monomers of αA-crystallin or D51 mutants to target protein was 1:1. The D51-αA crystallin construct and its triad mutants all suppressed insulin aggregation to some degree, falling into two groups (Table 1). D51-IGV, D51-GPG, and D51-TPT suppressed aggregation equal to or better than D51-αA-crystallin; the other two mutants, D51-ITV and D51-GGG, showed activity poorer than D51-αA-crystallin. However, none of the samples exhibited chaperone-like activity as good as that of wild-type αA-crystallin.

Triad mutants of wt-αA-crystallin

Three triad variations were selected for testing with the native αA-crystallin sequence based on their chaperone-like activity relative to D51-αA-crystallin. Comparing the initial functionality of each mutant, D51-IGV and D51-GPG showed better chaperone-like activty, and D51-ITV and D51-GGG showed poorer chaperone-like activity relative to D51-αA-crystallin. Therefore, ITV and IGV were selected because they represent a progression of increasing mobility relative to the native IPV while GGG eliminates both conformational and steric restraints and destroys the ability of the two outer residues to form hydrophobic interactions with the interiors of neighboring monomers.

The mutant proteins with one exception were overexpressed and recovered predominantly in the soluble fraction when host cells were extracted by lysozyme treatment. However, the wt-GGG mutant partitioned with the insoluble phase following lysozyme treatment. Even when recovered by solubilizing in cracking buffer containing 6 M urea, it was still insoluble following slow removal of urea by dialysis. Sequential chromatography over ion exchange and gel exclusion columns was sufficient to purify wt-αA-crystallin and the remaining mutants to apparent homogeneity. Purity of the wild type and mutant αA-crystallins was examined by SDS-PAGE. (Figure 4).

When allowed to aggregate, Wt-αA-crystallin produced a peak centered around 540 kDa (Figure 5) as assessed by size exclusion FPLC. Two other smaller peaks possibly corresponding to tetramers and dimers were also observed. The primary peaks of wt-ITV and wt-IGV mutants were each broader than that of αA-crystallin (Figure 5) with the former exhibiting a smaller oligomeric size (410 kDa) and the latter a larger one (610 kDa). Peaks representing smaller oligomers were again apparent but represented a much larger proportion of the overall population.

Assays of chaperone-like activity were performed by monitoring the ability of αA-crystallin and the triad mutants to prevent the denaturation and aggregation of insulin induced by adding DTT at room temperature. The equivalent mutants of wild type αA-crystallin, wt-ITV, and wt-IGV exhibited chaperone-like activity similar to wt-αA-crystallin in the presence of 3:1, 2:1, and 1:1 molar ratios of insulin:mutants (Figure 6, Table 1).

Changes in oligomeric size after heat incubation of the wt-IGV/ITV mutants and wt-αA-crystallin were observed using non-denaturing composite gel electrophoresis (Figure 7). The nondenaturing composite gel of wt-αA-crystallin (Figure 7A) shows an increase in oligomeric size after incubation at temperatures of 50 °C or higher. This increase in oligomeric size is progressive with increasing incubation temperature to 70 °C. Wt-ITV showed a similar trend in oligomeric size with increasing incubation temperatures and a similar transition temperature (Figure 7B). The wt-IGV mutant also increases in oligomeric size with temperature (Figure 7C), but this transition appears to be initiated at about 40 °C, 10 °C lower than the transition temperature of wt-αA-crystallin and wt-ITV.

Discussion

At present, there are four available crystal structures of the conformationally conserved α-crystallin core domain, Hsp16.5 from Methanococcus Janaschii, Hsp16.9 from wheat, p23 from human, and Tsp36 from Taenia saginata [37,38,50,51]. Of these, the first two will assemble into ordered quaternary structures while p23 is found in vivo as a monomeric cofactor for the Hsp 90 chaperone system and Tsp36 as a dimer in reducing conditions and a tetramer in oxidizing conditions [51]. The role of the conserved triad present in the COOH-terminal sequences of Hsp16.5 and Hsp16.9 appears to be to guide the organization of dimer subunits into higher order symmetric assemblies. Interestingly, neither the monomeric p23 nor the Tsp36 sequences contains the IXI/V motif in the COOH-terminal extension.

The α-crystallins, along with related animal sHsps, contain the COOH-terminal arm triad even though they do not appear to aggregate into ordered symmetric quaternary structures [52,53]. This conservation of the short β-strand motif in sHsp assemblies across kingdoms but not in monomeric or dimeric sHsp relatives suggests that the triad may serve an essential function in the subunit assembly process. To test this, a series of variations on the native IPV sequence was generated in αA-crystallin to assess its contribution to tetramer formation in the absence of the influence of the NH₂-terminal on aggregation. Flanking the central proline with either threonine (TPT) or glycine (GPG) rather than isoleucine and valine tests the importance of hydrophobicity and side chain size, respectively, in the presence of an amino acid backbone conformationally constrained by proline as graphically shown by Ramachandran plots of these residues' allowed dihedral angles. Changing the central proline to either theonine (ITV) or glycine (IGV) gradually reduces backbone conformational constraints in the orientation of the flanking residues by first replacing the proline's imine with an amide bond and then reducing steric hindrance caused by side chain size. GGG eliminates all of these steric, hydrophobic, and conformational constraints.

Since previous studies of others and our own have shown that the deletion of the first 50 amino acids of αA-crystallin mainly results in tetramers [48,54], finding that all five variants of D51-αA-crystallin will also primarily aggregate into tetramers provides clear evidence that the IPV triad is not critical to the interactions for tetramer formation [55,56]. However, there are subtle variations in both the structural and functional characteristics of the mutants that may be correlated with specific changes. The trailing shoulders of the mutants D51-ITV, D51-GGG, and D51-TPT were probably contributed by dimers and monomers. When comparing D51-ITV with D51-IGV and D51-TPT with D51-GPG, the fact that the percentage of the tetramers in the samples is 88.85% versus 100% and 84.09% versus 100%, respectively, suggests that introduction of one or two more hydrophilic amino acids into the hydrophobic triad will decrease the stability of the protein as tetramers. Furthermore, D51-GPG and D51-IGV, which exhibit a higher percentage of tetramers than D51-αA-crystallin, also show the best chaperone-like activities compared to D51-αA-crystallin. For D51-GPG, the replacement of the isoleucine and valine with glycines while maintaining the presence of the central proline results in larger molecular weight and better chaperone-like activity than the D51-αA-crystallin. This result is consistent with the study of Pasta et al. [44] on intact α-crystallin and extends it to an NH₂-terminal deletion variant. In contrast, D51-GGG, which has the greatest possible conformational freedom of all the variants, exhibits the poorest structural and functional properties. Thus, retaining the constraint of either the central proline or the flanking isoleucine and valine of IPV appears to be sufficient in the deletion mutants for equivalent or better structural and functional behavior.

In the wt-experiments, the central proline was replaced by either a threonine or a glycine. Threonine exhibits a range of allowed dihedral angles that is similar to most other amino acids and, therefore, less constrained than proline while the range of glycine's allowed dihedral angles is much greater than any other amino acid. The major peaks in the elution profiles of wild type αA-crystallin and its mutants are not as sharp as those of the D51 mutants. In this series of mutants, there appears to be small peaks at the leading edge and the trailing shoulders overlapping the major peak and partially eluting with it. This is probably due to the dynamic equilibration between different oligomeric states on the time scale of the experiment. Wt-ITV oligomeric size was significantly smaller than wt-αA-crystallin, and wt-IGV oligomeric size was significantly larger (Figure 5, Table 1). Both mutants showed a greater range of oligomeric sizes relative to wild type as indicated by elution peak width at the half peak height, and both showed a higher percentage amount of total protein in smaller oligomers (Table 1). These differences in average oligomeric size suggest differences in the location, the strength, and/or specificity of the triad interaction with surface features of neighboring subunits. If the larger oligomeric size of wt-IGV is due to weaker constraints on its quaternary structure, then the 10 °C reduction in its transition temperature relative to the wild type and ITV (Figure 7) may also be attributable to this factor. Unlike the D51 series, however, chaperone-like activity was essentially identical for all wt mutants over all of the molar ratios tested (Figure 6, Table 1). The likely explanation is that chaperone-like activity is not totally dependent on but may be strongly associated with larger oligomeric states.

Although the GGG mutant of D51 could be expressed and characterized as well it could show properties similar to the other D51 mutants, its equivalent variant of the wild type αA-crystallin differs substantially from wt-IGV and wt-ITV. Wt-GGG can be overexpressed successfully but is found in the insoluble phase rather than being soluble. If solubilized in the presence of 6 M urea, it drops out of solution as soon as the urea is removed. These properties of wt-GGG, combined with the results from wt-ITV and wt-IGV, suggest that the triad contributes substantially and specifically to the proper progression of the quaternary aggregation process in the intact sequence. Since it appears to have little or no effect on tetramer formation in the absence of first 50 amino acids, it may be either that the triad's interactions do not specifically facilitate a tetrameric intermediate or that the deletion mutant's tetramer subunit orientation differs substantially from that of the intact subunits in native-like oligomers. Additionally, just as the structural differences observed for the native-like assemblies are not exhibited by the tetramers, the functional differences observed for the tetrameric triad variants are not seen in the intact subunit assemblies. These functional differences in chaperone-like activity may have arisen as a result of differences in the stability of the different tetramers or in the mobility of the COOH-terminal arm relative to the tetrameric core. However, whatever factors contributed differentially to the chaperone-like activity, results shown for the D51 mutants are substantially reduced in significance in the intact subunits (Table 1).

The small effect of the triad mutants on tetramer formation and the much more significant effect on the formation of higher order oligomeric states can potentially be understood by careful examination of the available structures of superfamily members MjHsp 16.5 and TaHsp 16.9. In these structures, each monomer is part of a well-defined dimer, characterized by a two-fold symmetry axis, a common conserved interface, and by swapped strand stabilization. Dimers are connected in part by the COOH-terminal segment that contains the IPV type triad, but the interdimer connections do not define tetramers because the dimer serving as the "acceptor" for the triad "donated" by a monomer of an adjacent dimer donates its triad to a third dimer [38,39]. Additionally, these relationships between dimers connected by triad interactions can be defined in terms of symmetry operations that are different for each oligomerization. These operations do not form small closed groups corresponding to tetramers but instead form extended closed groups that correspond to the entire oligomerization process. It is likely that part of the variability of α-crystallin oligomers is the loss of this symmetry-related pattern of interactions due to differences in the connections between the triad and the conserved core and/or due to differences in subunit orientation imposed by other regions such as the NH₂-terminal or the extended loop that forms the swapped β-strand in plant and bacterial sequences.

Furthermore, the results from other studies of sequential truncated mutants of small heat shock proteins also offer some clues to the function of this triad. A R157STOP mutant αA-crystallin, which deletes the COOH-terminal 16 residues including the IXI/V triad, appears to have reduced solubility following the removal of urea by dialysis and elutes as an asymmetric peak with a trailing descending shoulder [57]. This is consistent with our observation that the wt-GGG mutant is not soluble. Abraham et al. [36] generated a set of COOH- terminal sequential truncated mutants at positions before and after the IXI/V triad. It is interesting to see that αA1-157, a mutant truncated two amino acids before the triad, is similar in size to αA1-162, a mutant truncated just one amino acid after the triad. The molecular masses of both mutants were around 150 kDa, probably indicating octamers, and were generally found in the insoluble pellet. In contrast, αA1-163, a mutant with one more residue at the COOH-terminal than αA1-162, aggregates into large oligomers around 500 kDa. Combined with the results reported here, this suggests the triad is not critical to the oligomerization of the αA-crystallin but affects the process to some extent.

In summary, the conserved COOH-terminal triad does not appear to have a strong effect on the steady-state aggregation of wild type αA-crystallin or its NH₂-terminal deletion mutant at 25 °C. It can exert a considerable effect on chaperone-like activity in the absence of the NH₂-terminal 50 residues and can influence the thermal transition temperature in its presence. These results suggest that the conserved triad in αA is not essential for the formation of tetramers but contributes to the stability and solubility of higher order oligomers.

Acknowledgements

The authors are grateful for the support of NIH grant EY010011 and the Vincentine Cundari Memorial Research Fund for portions of this work, and for the technical assistance of Ms. Katie Mahony.

References

1. Brown NP, Koretz JF, Bron AJ. The development and maintenance of emmetropia. Eye 1999; 13:83-92.

2. Koretz JF, Cook CA. Aging of the optics of the human eye: lens refraction models and principal plane locations. Optom Vis Sci 2001; 78:396-404.

3. Ingolia TD, Craig EA. Four small Drosophila heat shock proteins are related to each other and to mammalian alpha-crystallin. Proc Natl Acad Sci U S A 1982; 79:2360-4.

4. van der Ouderaa FJ, de Jong WW, Bloemendal H. The amino-acid sequence of the alphaA2 chain of bovine alpha-crystallin. Eur J Biochem 1973; 39:207-22.

5. Van Der Ouderaa FJ, De Jong WW, Hilderink A, Bloemendal H. The amino-acids sequence of the alphaB2 chain of bovine alpha-crystallin. Eur J Biochem 1974; 49:157-68.

6. Veretout F, Delaye M, Tardieu A. Molecular basis of eye lens transparency. Osmotic pressure and X-ray analysis of alpha-crystallin solutions. J Mol Biol 1989; 205:713-28.

7. Rao PV, Huang QL, Horwitz J, Zigler JS Jr. Evidence that alpha-crystallin prevents non-specific protein aggregation in the intact eye lens. Biochim Biophys Acta 1995; 1245:439-47.

8. Tardieu A. alpha-Crystallin quaternary structure and interactive properties control eye lens transparency. Int J Biol Macromol 1998; 22:211-7.

9. Horwitz J. The function of alpha-crystallin in vision. Semin Cell Dev Biol 2000; 11:53-60.

10. Horwitz J. Alpha-crystallin can function as a molecular chaperone. Proc Natl Acad Sci U S A 1992; 89:10449-53.

11. Nicholl ID, Quinlan RA. Chaperone activity of alpha-crystallins modulates intermediate filament assembly. EMBO J 1994; 13:945-53.

12. Boelens WC, de Jong WW. alpha-Crystallins, versatile stress-proteins. Mol Biol Rep 1995; 21:75-80.

13. van den IJssel PR, Smulders RH, de Jong WW, Bloemendal H. alpha-Crystallin: molecular chaperone and heat shock protein. Ophthalmic Res 1996; 28:39-43.

14. Horwitz J, Huang QL, Ding L, Bova MP. Lens alpha-crystallin: chaperone-like properties. Methods Enzymol 1998; 290:365-83.

15. Bhat SP, Horwitz J, Srinivasan A, Ding L. Alpha B-crystallin exists as an independent protein in the heart and in the lens. Eur J Biochem 1991; 202:775-81.

16. Lutsch G, Vetter R, Offhauss U, Wieske M, Grone HJ, Klemenz R, Schimke I, Stahl J, Benndorf R. Abundance and location of the small heat shock proteins HSP25 and alphaB-crystallin in rat and human heart. Circulation 1997; 96:3466-76.

17. Vicart P, Caron A, Guicheney P, Li Z, Prevost MC, Faure A, Chateau D, Chapon F, Tome F, Dupret JM, Paulin D, Fardeau M. A missense mutation in the alphaB-crystallin chaperone gene causes a desmin-related myopathy. Nat Genet 1998; 20:92-5.

18. Banwell BL, Engel AG. AlphaB-crystallin immunolocalization yields new insights into inclusion body myositis. Neurology 2000; 54:1033-41.

19. Zabel C, Chamrad DC, Priller J, Woodman B, Meyer HE, Bates GP, Klose J. Alterations in the mouse and human proteome caused by Huntington's disease. Mol Cell Proteomics 2002; 1:366-75.

20. Liu B, Bhat M, Nagaraj RH. AlphaB-crystallin inhibits glucose-induced apoptosis in vascular endothelial cells. Biochem Biophys Res Commun 2004; 321:254-8.

21. Link CD, Taft A, Kapulkin V, Duke K, Kim S, Fei Q, Wood DE, Sahagan BG. Gene expression analysis in a transgenic Caenorhabditis elegans Alzheimer's disease model. Neurobiol Aging 2003; 24:397-413.

22. Stege GJ, Renkawek K, Overkamp PS, Verschuure P, van Rijk AF, Reijnen-Aalbers A, Boelens WC, Bosman GJ, de Jong WW. The molecular chaperone alphaB-crystallin enhances amyloid beta neurotoxicity. Biochem Biophys Res Commun 1999; 262:152-6.

23. Castellani RJ, Perry G, Harris PL, Monnier VM, Cohen ML, Smith MA. Advanced glycation modification of Rosenthal fibers in patients with Alexander disease. Neurosci Lett 1997; 231:79-82.

24. Soti C, Csermely P. Chaperones and aging: role in neurodegeneration and in other civilizational diseases. Neurochem Int 2002; 41:383-9.

25. Vleminckx V, Van Damme P, Goffin K, Delye H, Van Den Bosch L, Robberecht W. Upregulation of HSP27 in a transgenic model of ALS. J Neuropathol Exp Neurol 2002; 61:968-74.

26. Calabrese V, Scapagnini G, Colombrita C, Ravagna A, Pennisi G, Giuffrida Stella AM, Galli F, Butterfield DA. Redox regulation of heat shock protein expression in aging and neurodegenerative disorders associated with oxidative stress: a nutritional approach. Amino Acids 2003; 25:437-44.

27. Wyttenbach A. Role of heat shock proteins during polyglutamine neurodegeneration: mechanisms and hypothesis. J Mol Neurosci 2004; 23:69-96.

28. Caspers GJ, Leunissen JA, de Jong WW. The expanding small heat-shock protein family, and structure predictions of the conserved "alpha-crystallin domain". J Mol Evol 1995; 40:238-48.

29. de Jong WW, Caspers GJ, Leunissen JA. Genealogy of the alpha-crystallin--small heat-shock protein superfamily. Int J Biol Macromol 1998; 22:151-62.

30. de Jong WW, Leunissen JA, Voorter CE. Evolution of the alpha-crystallin/small heat-shock protein family. Mol Biol Evol 1993; 10:103-26.

31. Berengian AR, Bova MP, Mchaourab HS. Structure and function of the conserved domain in alphaA-crystallin. Site-directed spin labeling identifies a beta-strand located near a subunit interface. Biochemistry 1997; 36:9951-7.

32. Ganea E. Chaperone-like activity of alpha-crystallin and other small heat shock proteins. Curr Protein Pept Sci 2001; 2:205-25.

33. Eifert C, Burgio MR, Bennett PM, Salerno JC, Koretz JF. N-terminal control of small heat shock protein oligomerization: changes in aggregate size and chaperone-like function. Biochim Biophys Acta 2005; 1748:146-56.

34. Kumar LV, Rao CM. Domain swapping in human alpha A and alpha B crystallins affects oligomerization and enhances chaperone-like activity. J Biol Chem 2000; 275:22009-13.

35. Pasta SY, Raman B, Ramakrishna T, Rao ChM. Role of the C-terminal extensions of alpha-crystallins. Swapping the C-terminal extension of alpha-crystallin to alphaB-crystallin results in enhanced chaperone activity. J Biol Chem 2002; 277:45821-8.

36. Thampi P, Abraham EC. Influence of the C-terminal residues on oligomerization of alpha A-crystallin. Biochemistry 2003; 42:11857-63.

37. Kim KK, Kim R, Kim SH. Crystal structure of a small heat-shock protein. Nature 1998; 394:595-9.

38. van Montfort RL, Basha E, Friedrich KL, Slingsby C, Vierling E. Crystal structure and assembly of a eukaryotic small heat shock protein. Nat Struct Biol 2001; 8:1025-30.

39. Studer S, Obrist M, Lentze N, Narberhaus F. A critical motif for oligomerization and chaperone activity of bacterial alpha-heat shock proteins. Eur J Biochem 2002; 269:3578-86.

40. Kirschner M, Winkelhaus S, Thierfelder JM, Nover L. Transient expression and heat-stress-induced co-aggregation of endogenous and heterologous small heat-stress proteins in tobacco protoplasts. Plant J 2000; 24:397-411.

41. Selcen D, Ohno K, Engel AG. Myofibrillar myopathy: clinical, morphological and genetic studies in 63 patients. Brain 2004; 127:439-51.

42. Evgrafov OV, Mersiyanova I, Irobi J, Van Den Bosch L, Dierick I, Leung CL, Schagina O, Verpoorten N, Van Impe K, Fedotov V, Dadali E, Auer-Grumbach M, Windpassinger C, Wagner K, Mitrovic Z, Hilton-Jones D, Talbot K, Martin JJ, Vasserman N, Tverskaya S, Polyakov A, Liem RK, Gettemans J, Robberecht W, De Jonghe P, Timmerman V. Mutant small heat-shock protein 27 causes axonal Charcot-Marie-Tooth disease and distal hereditary motor neuropathy. Nat Genet 2004; 36:602-6.

43. Takeuchi N, Ouchida A, Kamei A. C-terminal truncation of alpha-crystallin in hereditary cataractous rat lens. Biol Pharm Bull 2004; 27:308-14.

44. Pasta SY, Raman B, Ramakrishna T, Rao ChM. The IXI/V motif in the C-terminal extension of alpha-crystallins: alternative interactions and oligomeric assemblies. Mol Vis 2004; 10:655-62 <http://www.molvis.org/molvis/v10/a78/>.

45. Moulin S, Fruchart JC, Dewailly P, Sezille G. [Lipoprotein electrophoresis on acrylamide-agarose plates, with discontinuous acrylamide gradient]. Clin Chim Acta 1979; 91:159-63.

46. Burgio MR, Bennett PM, Koretz JF. Heat-induced quaternary transitions in hetero- and homo-polymers of alpha-crystallin. Mol Vis 2001; 7:228-33 <http://www.molvis.org/molvis/v7/a32/>.

47. Salerno JC, Eifert CL, Salerno KM, Koretz JF. Structural diversity in the small heat shock protein superfamily: control of aggregation by the N-terminal region. Protein Eng 2003; 16:847-51.

48. Yang C, Salerno JC, Koretz JF. NH2-terminal stabilization of small heat shock protein structure: a comparison of two NH2-terminal deletion mutants of alphaA-crystallin. Mol Vis 2005; 11:641-7 <http://www.molvis.org/molvis/v11/a76/>.

49. Robinson CR, Sauer RT. Optimizing the stability of single-chain proteins by linker length and composition mutagenesis. Proc Natl Acad Sci U S A 1998; 95:5929-34.

50. Weaver AJ, Sullivan WP, Felts SJ, Owen BA, Toft DO. Crystal structure and activity of human p23, a heat shock protein 90 co-chaperone. J Biol Chem 2000; 275:23045-52.

51. Stamler R, Kappe G, Boelens W, Slingsby C. Wrapping the alpha-crystallin domain fold in a chaperone assembly. J Mol Biol 2005; 353:68-79.

52. Haley DA, Bova MP, Huang QL, Mchaourab HS, Stewart PL. Small heat-shock protein structures reveal a continuum from symmetric to variable assemblies. J Mol Biol 2000; 298:261-72.

53. Haley DA, Horwitz J, Stewart PL. The small heat-shock protein, alphaB-crystallin, has a variable quaternary structure. J Mol Biol 1998; 277:27-35.

54. Merck KB, De Haard-Hoekman WA, Oude Essink BB, Bloemendal H, De Jong WW. Expression and aggregation of recombinant alpha A-crystallin and its two domains. Biochim Biophys Acta 1992; 1130:267-76.

55. Schmitz KR, Koretz JF. Site-Directed Mutagenesis of a Conserved Region Along the C-Terminal Arm of Alpha-Crystallin. ARVO Annual Meeting; 2003 May 4-9; Fort Lauderdale (FL).

56. Koretz JF, Li Y, Salerno JC. The Role of the Conserved C-Terminal Triad in AlphaA-Crystallin Aggregation and Functionality. ARVO Annual Meeting; 2005 May 1-5; Fort Lauderdale (FL).

57. Andley UP, Mathur S, Griest TA, Petrash JM. Cloning, expression, and chaperone-like activity of human alphaA-crystallin. J Biol Chem 1996; 271:31973-80.