Molecular Vision 2005; 11:88-96 <>
Received 17 November 2004 | Accepted 23 January 2005 | Published 26 January 2005

Zebrafish α-crystallins: protein structure and chaperone-like activity compared to their mammalian orthologs

Jason M. Dahlman,1 Kelli L. Margot,1 Linlin Ding,2 Joseph Horwitz,2 Mason Posner1

1Department of Biology, Ashland University, Ashland, OH; 2Jules Stein Eye Institute, School of Medicine, University of California-Los Angeles, Los Angeles, CA

Correspondence to: Mason Posner, Department of Biology, Ashland University, 401 College Avenue, Ashland, OH, 44805; Phone: (419) 289-5691; FAX: (419) 289-5283; email:


Purpose: The vertebrate small heat shock proteins αA- and αB-crystallin contribute to the transparency and refractive power of the lens and may also prevent the aggregation of non-native proteins that would otherwise lead to cataracts. We previously showed that zebrafish (Danio rerio) and human αB-crystallin have diverged far more in primary structure and expression pattern than the orthologous αA-crystallins. In this current study we further compare the structure and function of zebrafish and mammalian α-crystallins.

Methods: Near UV CD spectroscopy was used to analyze the tertiary structure and thermal stability of recombinant zebrafish α-crystallins. The chaperone-like activities of zebrafish and human α-crystallins were compared by assaying their ability to prevent the chemically induced aggregation of several target proteins at temperatures between 25 °C and 40 °C.

Results: Zebrafish and human αA-crystallin showed very similar tertiary structures, while the αB-crystallin orthologs showed differences related to the presence of additional aromatic amino acids in the zebrafish protein. The denaturation temperatures of zebrafish crystallins were lower than those of mammals. The chaperone-like activities of the two zebrafish α-crystallins were highly divergent, with αA-crystallin showing much greater activity than αB-crystallin.

Conclusions: αA-crystallin serves a similar physiological function in both zebrafish and mammals as a lens specific chaperone-like molecule. The reduced chaperone-like function of zebrafish αB-crystallin and its lack of extralenticular expression indicates that it plays a different physiological role from its mammalian ortholog. Future comparative studies of α-crystallin from closely related vertebrate species can help identify specific structural changes that lead to alterations in chaperone-like activity.


The crystallins are a diverse group of proteins that make up over 90% of the total soluble protein of the vertebrate eye lens [1,2]. α-Crystallin, which accounts for up to 40% of the protein content of the mammalian lens, is found in the lens as a heteroaggregate of two proteins, αA-crystallin and αB-crystallin. Both proteins are members of the small heat shock protein family, which are typically stress inducible and act as molecular chaperone-like proteins by binding to and preventing the aggregation of denaturing proteins. Unlike true molecular chaperones, α-crystallins are not able to refold non-native proteins. Both mammalian α-crystallins are induced by stress [3,4] and both have chaperone-like activity [5,6]. The function of α-crystallins in the lens appears to be two fold. While they both contribute structurally to the transparency and refractive index of the lens necessary for vision, their chaperone-like activity also prevents the aggregation of aging and stressed proteins that would otherwise lead to lens opacity. αB-crystallin is also expressed in multiple non-lenticular tissues and may play numerous physiological roles that are not well understood [7].

Most studies on α-crystallins have focused on humans and other mammalian species. However, a smaller number of comparative studies on non-mammalian species has contributed greatly to the understanding of α-crystallin structure and function [8,9]. Amino acid sequence comparisons from diverse vertebrate taxa helped identify the structural and functional domains of the α-crystallins [10]. Kokke et al. [11] used a related small heat shock protein from the roundworm Caenorhabditis elegans to investigate the function of the α-crystallin domain. Other non-vertebrate small heat shock proteins have been useful for understanding the structure of α-crystallins. Because αA- and αB-crystallin form large polydisperse heteroaggregates with molecular weights ranging from 300,000 to over 1 million it has been difficult to produce crystal structures for x-ray crystallographic studies [12]. However, small heat shock proteins from non-vertebrate species have been crystallized. Crystal structures of Hsp 16.5 from the thermophilic archea Methanococcus janaschii and Hsp 16.9 from wheat provide some of the best insights to small heat shock protein tertiary and quaternary structure [13,14]. Lastly, comparative studies provide a powerful tool for examining α-crystallin physiological roles and correlating changes in protein structure with changes in function. The discovery of naturally occurring mutations in human αA-crystallin and αB-crystallin helped identify single amino acid substitutions that drastically alter protein structure and eliminate chaperone-like activity [15-17]. However, few comparative functional studies have included non-mammalian species. Studies on the blind mole rat [18,19] and the blind cave fish [20] have examined the impact of eye regression on the evolution of αA-crystallin structure and function. Both studies found, surprisingly, that the structure and function of αA-crystallin has not changed dramatically, suggesting that this protein plays some non-refractive physiological role.

The comparative approach seems to be an underutilized strategy for examining α-crystallin structure, function, and physiological role. Comparisons of α-crystallins from taxa facing diverse physiological challenges can provide important data on how α-crystallin structures have been evolutionarily modified to function in different conditions. Ultimately, these data can provide insights into the function of human α-crystallins not available by studying mammalian α-crystallins alone. Specifically, comparisons of α-crystallin chaperone-like activity from organisms with varying physiological temperatures can be used to examine relationships between structure and function. Multiple studies have indicated that rising temperature increases the ability of α-crystallin to bind non-native protein [21-23]. Thermally induced changes in chaperone-like activity have been attributed to decreased structural stability [21], exposure of additional hydrophobic residues [24], increase in available binding sites [22], and increase in subunit exchange [25]. Because humans and other mammals are restricted to a relatively narrow range of similar body temperatures, these species do not provide a good model group for studying the relationship between physiological temperature and α-crystallin structure and function. Ectothermic vertebrates, however, provide a large group of species covering a broad range of physiological temperatures. The diverse bony fishes, with over 25,000 species, range in physiological temperature from -2 °C to over 35 °C. The zebrafish (Danio rerio) is adapted to a narrow temperature range centered near 27 °C, 10 °C lower than that of humans. At temperatures below 25 °C and above 31 °C, its development becomes impaired [26]. Because of its widespread use in developmental and genetics studies we chose the zebrafish for an analysis of chaperone-like activity of non-mammalian vertebrate α-crystallins.

We have previously cloned αA- and αB-crystallin from the zebrafish and examined their tissue specific expression [27,28]. We found that zebrafish and human αA-crystallin are similar in primary structure and tissue specific expression. The two orthologs show 73% identity in amino acid sequence and the expression of both is limited primarily to the lens. Zebrafish αB-crystallin, on the other hand, has diverged far more from its human ortholog in both structure and expression. These orthologs show only 58% identity in amino acid sequence. There are also several changes to amino acid residues in zebrafish αB-crystallin thought to be important to chaperone-like activity, such as a C-terminal extension that is four amino acids shorter than the human ortholog. Perhaps most interestingly, unlike human αB-crystallin, zebrafish αB-crystallin is not expressed in nervous or muscular tissue. Overall these studies showed that zebrafish and mammalian αA-crystallins are more conserved in both primary structure and expression pattern than αB-crystallins. Preliminary assays of zebrafish α-crystallin chaperone-like activity also demonstrated differences in the abilities of orthologous zebrafish and human α-crystallins to prevent the chemically induced aggregation of α-lactalbumin [8].

In this current study we examined the tertiary structures, thermal stability, and further assayed the chaperone-like activity of zebrafish αA- and αB-crystallin compared to their mammalian orthologs. We hypothesized that the conservation of αA-crystallin primary structure would lead to similar tertiary structure and chaperone-like activity. We also hypothesized that the divergence between zebrafish and human αB-crystallin primary structure and the reduced expression outside of the lens would be reflected in a reduction in chaperone-like activity. Indeed we found that zebrafish αB-crystallin was a worse molecular chaperone than human αB-crystallin. However, the chaperone-like activities of zebrafish and human αA-crystallin were also dissimilar in that zebrafish αA-crystallin showed greater protection than its human ortholog, suggesting a possible adaptation to the lower body temperatures of this ectothermic organism.


Production and purification of recombinant proteins

PCR constructs containing the coding regions for zebrafish αA- and αB-crystallin were ligated into the pET20b(+) expression vector (Novagen, Madison, WI) and used to transform BL21(DE3) bacterial cells (Novagen) or BL21(DE3) gold cells (Stratagene, La Jolla, CA). Expression constructs for the production of human α-crystallins were obtained from other laboratories. PCR was also used to produce a modified zebrafish αB-crystallin expression construct (Hybrid αB) containing a C-terminal extension similar to human αB-crystallin (Table 1). Protein induction, cell lysis and purification were performed essentially as described by Horwitz et al. [29] except for the following changes: Cell lysates were loaded onto a Mono-Q Hi Trap column (Amersham, Piscataway, NJ) and eluted with 20 mM Tris, pH 8.5 with stepwise concentrations of 0.1 M, 0.2 M, 0.3 M, and 0.4 M NaCl. Zebrafish αA-crystallin eluted from the column with the 0.3 M NaCl buffer, and zebrafish αB-crystallin and the hybrid αB-crystallin eluted from the column with the 0.4 M NaCl buffer. Fractions from the Mono-Q Hi Trap column containing the recombinant crystallin were concentrated in Amicon centrifugal filters (30,000 MW cutoff; Millipore, Billerica, MA) and passed through a 90 cm x 2.5 cm size exclusion column containing Sephacryl S-200 High Resolution bedding material (Amersham) at a flow rate of 0.4 ml/min and a temperature of 8 °C. Fractions containing purified α-crystallin were concentrated to approximately 5 mg/ml in Centricon YM-30 centrifugal concentrators (Millipore) and used in chaperone assays. Bradford assays were used to determine the relationship between absorbance and protein concentration of the recombinant zebrafish proteins and total zebrafish lens soluble protein. The relevant human α-crystallin or total bovine lens soluble protein was used to produce standard curves from which these relationships were determined.

Production of a polyclonal antibody to zebrafish αB-crystallin

A polyclonal antibody to purified zebrafish αB-crystallin was made in rabbits (Alpha Diagnostic International, San Antonio, TX) and used to probe the total soluble lens protein from cow and zebrafish. A dilution of 1:5,000 was optimal for probing total zebrafish lens soluble protein. A polyclonal antibody to human αB-crystallin was used at a dilution of 1:1,000 to probe both cow and zebrafish total lens soluble protein. Colorimetric detection of the western blots was performed with the Opti-4CN kit (Bio-Rad, Hercules, CA).

Circular dichroism (CD) spectroscopy of zebrafish and mammalian α-crystallins

Near UV spectra were measured using a J-810 spectropolarimeter (Jasco, Easton, MD) with 1.0 cm path length cells. Each spectrum is an average of 32 scans. The concentration of protein used was 1.5 mg/ml. Near UV spectra were also measured at various temperatures to determine the thermal stability of the zebrafish α-crystallins.

Assays of chaperone-like activity

Chaperone-like activities of purified zebrafish and human α-crystallins were assayed by measuring their ability to prevent the chemically induced aggregation of several target proteins [29]. Insulin (I5500; Sigma, St. Louis, MO) was denatured with 20 mM dithiothreitol (DTT) in a buffer containing 50 mM sodium phosphate, 0.1 M NaCl, pH 7.0. α-Lactalbumin (L6010; Sigma) was denatured with 20 mM DTT in a buffer containing 50 mM sodium phosphate, 0.1 M NaCl, pH 6.75. Lysozyme (L6876; Sigma) was denatured with 1 mM Tris (2-carboxyethyl) phosphine hydrochloride (TCEP) in a buffer containing 50 mM sodium phosphate, 0.1 M NaCl, pH 7.0. Both DTT and TCEP denature these target proteins by reducing their disulfide bonds. α-Crystallins are not affected as they lack disulfide bonds. Absorption due to light scattering produced in the reactions with or without different α-crystallins was measured at 360 nm for 60 to 120 min at temperatures between 25 °C and 40 °C in 5 °C increments. All reactions were in a total of 500 μl using a 5 mm path length cuvette. The percent protection of each α-crystallin was calculated by dividing the difference between the increases in light scattering of the target protein alone and the α-crystallin solution by the increase in light scattering of the target protein alone. The target protein concentrations used in the assays were chosen to produce sufficient aggregation without allowing precipitation of aggregated protein. For each target protein, the concentration used was kept constant for all temperatures tested, as indicated in the Figures. However, concentrations varied between target proteins.


SDS-PAGE and western blot analysis of native and recombinant lens proteins

A wide variety of fish species contain lower concentrations of lens α-crystallin than the typical 30% found in many mammals [30,31]. This reduced α-crystallin concentration was also evident in the zebrafish lens when total lens soluble protein was analyzed by SDS-PAGE. The two bands from bovine lens above the 21.5 kDa standard represent the two α-crystallins (Figure 1A). The zebrafish α-crystallins are not as highly resolved on the gel, as they are obscured by the more abundant γ-crystallins. Western blot analysis using an antibody to human αB-crystallin recognized bovine αB-crystallin and produced a very weak band from zebrafish lens soluble protein between the 20.7 and 28.8 kDa standards, which is the correct size to be zebrafish αB-crystallin (Figure 1B, arrow 1). A polyclonal antibody to zebrafish αB-crystallin recognized two tightly clustered bands in the zebrafish total soluble lens protein between the 20.7 and 28.8 kDa standards that could be α-crystallins (Figure 1C, arrow 2) and recognized both recombinant zebrafish αA- and αB-crystallin (data not shown). The zebrafish anti-αB-crystallin antibody consistently recognized a third band from zebrafish lens of unknown identity that was smaller than the 20.7 kDa standard (Figure 1C, arrow 3). The zebrafish anti-αB-crystallin antibody did not recognize any soluble bovine lens proteins (Figure 1C) nor either of the recombinant human α-crystallins (data not shown).

CD spectroscopy of zebrafish and mammalian α-crystallins

Recombinant zebrafish and mammalian α-crystallins were examined by near UV CD spectroscopy to analyze differences in tertiary structure. The resulting spectra indicated no major differences between the tertiary structures of zebrafish and bovine αA-crystallin (Figure 2A). The similar peaks between 260 and 270 nm of their near UV CD spectra indicate that the position of phenylalanines are nearly identical in the two proteins. The peaks between 280 and 295 nm, due to the position of tyrosines and tryptophans, show a greater difference in intensity, but are still similar. The near UV spectra for zebrafish and human αB-crystallin indicated changes in the positioning of aromatic amino acid residues (Figure 1B). The two phenylalanine peaks between 260 and 270 nm are present in both proteins, although their intensities do vary. The negative shift in the zebrafish αB-crystallin spectrum between 270 and 290 nm is likely due to the addition of two tryptophans and six tyrosines compared to the human ortholog. Adding a human-like C-terminal extension to zebrafish αB-crystallin did not significantly affect the near UV spectrum (Figure 2B). Far UV CD spectroscopy indicated no major differences in secondary structure between the zebrafish and human orthologs (data not shown).

Thermal stability of zebrafish α-crystallins

Near UV CD spectroscopy was also used to determine the thermal denaturation temperatures for the two zebrafish α-crystallins. Zebrafish αA-crystallin remained stable between 25 °C and 50 °C, with its tertiary structure changing between 50 °C and 60 °C (Figure 3A). This is lower than the denaturation temperature for human αA-crystallin, which remains relatively unchanged up to 62 °C [32]. Aqueous solutions of zebrafish αA-crystallin became visibly opaque by 60 °C. Zebrafish αB-crystallin remained stable between 25 °C and 45 °C and became visibly opaque by 50 °C (Figure 3B). Human αB-crystallin only partially denatures at 62 °C, and remains in solution until 65 °C [32]. Both zebrafish proteins denature at temperatures lower than their human orthologs.

Comparisons of zebrafish and human α-crystallin chaperone-like activity

The chaperone-like activities of zebrafish α-crystallins were assayed by measuring their ability to prevent the aggregation of the target proteins α-lactalbumin, insulin, and lysozyme upon chemical denaturation. Zebrafish αA-crystallin had greater chaperone-like activity than human αA-crystallin when using insulin and lactalbumin as target proteins. At 37 °C greater than five times and three times more human αA crystallin was required to provide the same protection as zebrafish αA crystallin against the aggregation of α-lactalbumin (Figure 4A) and insulin, respectively (Figure 4B). The two αA-crystallins showed similar chaperone-like activity when lysozyme was used as a target protein (Figure 4C). Zebrafish αB-crystallin was a weaker chaperone than human αB-crystallin with all three target proteins at 37 °C (Figure 5). It required greater than four to ten times more zebrafish αB-crystallin to provide the same protection as human αB-crystallin against the aggregation of the various target proteins. The chaperone-like activity of all α-crystallins tested generally increased with temperature (Figure 6). The relative strength of chaperone-like activity among the α-crystallins depended on the target protein that was used. In general, zebrafish αA-crystallin and human αB-crystallin both provided significant protection while zebrafish αB-crystallin provided the least. Because the C-terminal region of human α-crystallin has been implicated in chaperone-like activity, we assayed an artificially constructed version of zebrafish αB-crystallin containing a human-like C-terminus. This modification did not significantly increase chaperone-like activity over that of wild type zebrafish αB-crystallin (Figure 6). Because the aggregation of denatured lysozyme changed greatly with temperature, it was not possible to directly compare the chaperone-like activity of each crystallin against lysozyme aggregation at all temperatures. In order to achieve adequate aggregation at 27 °C it was necessary to double the amount of lysozyme used in the assay. The relative strengths of zebrafish and human αA-crystallin in protecting against lysozyme aggregation did not change between 27 °C and 37 °C (data not shown). However, the relative strengths of the αB-crystallins were very different. Whereas it took more than four times more zebrafish αB-crystallin to provide the same protection as human αB-crystallin at 37 °C (Figure 5C), it only took two times more to produce the same protection at 27 °C (data not shown). A similar trend was seen with the α-lactalbumin assays in which human αB-crystallin was a worse chaperone than zebrafish αB-crystallin only at the lowest temperature measured (25 °C).


The data presented in this study show that zebrafish αB-crystallin has lower chaperone-like activity than its human ortholog, while zebrafish αA-crystallin has higher chaperone-like activity than its human ortholog. These findings support the hypothesis that zebrafish αB-crystallin plays a different physiological role than human αB-crystallin. Furthermore, both zebrafish α-crystallins exhibit features that reflect the lower physiological temperature of this ectothermic species. Specifically, both zebrafish proteins begin to denature at temperatures lower than their human orthologs, and the increased chaperone-like activity of zebrafish αA-crystallin relative to its human ortholog may be an adaptation to the lower physiological temperature of the zebrafish. Differences in structure, thermal stability, chaperone-like activity, and physiological function between zebrafish and mammalian α-crystallins indicate that ectothermic vertebrates can be a useful model group for investigating the impact of temperature on α-crystallin function and evolution.

The strong conservation of tertiary structure between zebrafish and human αA-crystallin suggests that these two proteins have maintained similar functions within the lens. It also shows that changes in tertiary structure, at least as identified by near UV CD, are not necessary to alter thermal stability or chaperone-like activity. The greater divergence of near UV spectra between the αB-crystallins is likely due to the presence of two additional tryptophans and six additional tyrosines in the zebrafish protein (Figure 2B). Because near UV CD spectra indicate the position of aromatic amino acids in the tertiary structure, the addition of tryptophans and tyrosines could alter the spectrum without necessarily indicating a dramatic change in protein conformation. Our results, therefore, do not conclusively indicate a major difference in tertiary structure between the zebrafish and human αB-crystallin.

Our data showing that the thermal denaturation temperatures of both zebrafish α-crystallins were reduced compared to their human orthologs concurs with a previous study on native vertebrate α-crystallin heteroaggregates. McFall-Ngai et al. [33] first showed that the susceptibility of vertebrate α-crystallin heteroaggreagates to thermal denaturation was dependent on their physiological temperature, with warm adapted species showing higher tolerances than cold adapted species. Interestingly, the difference in thermostability between the two zebrafish α-crystallins is similar to that found in mammals, in which αA-crystallin is also more thermostable than αB-crystallin [32]. It is unclear whether this similar difference in thermostability between orthologous pairs of zebrafish and mammalian α-crystallins is a coincidence, or whether it reflects a conserved functional difference. Our finding that zebrafish αB-crystallin solutions became opaque by 50 °C differs greatly from a recent report that recombinant catfish (Clarius batrachus) αB-crystallin stays in solution until near 70 °C [34].

Our data clearly indicate that zebrafish αA-crystallin provided better protection than human αA-crystallin against chemically induced protein aggregation. This result was seen with two of the three target proteins (α-lactalbumin and insulin) at all temperatures tested. The similar protection provided by zebrafish and human αA-crystallin against lysozyme aggregation reflects the variation in chaperone-like activity that can be seen when using different target proteins. It is possible that structural features in zebrafish αA-crystallin responsible for its increased ability to bind denaturing α-lactalbumin and insulin do not provide similar ability to bind denaturing lysozyme. From our data, it appears that zebrafish αA-crystallin is better than or at least equivalent to human αA-crystallin as a chaperone-like molecule. Zebrafish αB-crystallin was a consistently worse chaperone than human αB-crystallin in all three assays at most temperatures tested. Previous studies have shown that α-crystallin chaperone-like activity typically decreases with falling temperature [21]; a trend also found in our data (Figure 6). In our insulin assays the chaperone activity of human αB-crystallin changed faster with temperature than that of zebrafish αB-crystallin, resulting in lower protection at 25 °C (Figure 6B). But in every other situation zebrafish αB-crystallin was a far inferior chaperone than its human ortholog. The difference in protective ability is even greater when the two proteins are compared at their relevant physiological temperatures (27 °C and 37 °C; Figure 6). These results differ from the Yu et al. [34] study mentioned above, which indicated that recombinant catfish αB-crystallin provides greater protection than porcine αB-crystallin against the chemically induced denaturation of insulin at 37 °C. It is possible that the reduced protection of zebrafish αB-crystallin relative to the human ortholog demonstrated in this present study will not generally be found when comparing all bony fishes to mammals. However, the recent discovery of a possible second αB-crystallin transcript from zebrafish (G. Wistow, personal communication) [35] may provide another interpretation of the increased chaperone-like activity in catfish αB-crystallin. Ray-finned fishes, the group to which both zebrafish and catfishes belong, underwent a genome duplication event early in their evolution [36,37]. It is, therefore, likely that the gene duplication event that produced two αB-crystallins pre-dated the evolutionary separation between the zebrafish and catfish lineages. If this is the case, the catfishes might also contain genes for two αB-crystallins, and Yu et al. [34] may have cloned a different copy than the one we have assayed in this study. It is worth noting that our polyclonal antibody to zebrafish αB-crystallin recognized two bands (Figure 1C). The lighter of the two bands (arrow 3) was just below 20 kDa in size and could be a second zebrafish αB-crystallin. If there are indeed two copies of αB-crystallin in the zebrafish, they would both be orthologous to human αB-crystallin [38]. The two zebrafish αB-crystallins would be considered paralogous to each other.

Because the C-terminus of mammalian α-crystallins has been suggested to play a role in chaperone-like activity [39,40] we hypothesized that a truncation to this region of zebrafish αB-crystallin relative to its human ortholog could account for the reduced protective ability of the former. However, it does not appear that the modified C-terminus alone is responsible for the reduced chaperone-like activity of zebrafish αB-crystallin. The addition of eight C-terminal amino acids similar to those found in the human protein did not significantly change the chaperone-like activity of zebrafish αB-crystallin (Figure 6). Thampi and Abraham [41] found that the chaperone activity of mouse αA-crystallin was not affected until 11 C-terminal amino acids were removed. The modifications seen in zebrafish αB-crystallin might not be extensive enough to cause its reduced chaperone-like activity. Addition of a human-like C-terminus also had no apparent effect on zebrafish αB-crystallin tertiary structure (Figure 2B).

All four α-crystallins assayed in this study maintained their tertiary structure beyond the highest temperature (40 °C) used in the aggregation assays. This suggests that thermally induced changes in tertiary structure did not account for the differences in chaperone-like activity between zebrafish and human orthologues. The differing abilities of zebrafish αA- and αB-crystallin to prevent aggregation could be due to amino acid differences affecting protein binding sites or the dynamic nature of the α-crystallin multimere. These hypotheses can be directly addressed in future studies. It is tempting to conclude that zebrafish αA-crystallin has adapted to its physiological temperature by increasing its chaperone-like activity to produce adequate protection at lower body temperatures. Indeed, the percent protection of zebrafish and human αA-crystallin are similar at their respective physiological temperatures (Figure 6). Adaptation of protein function to physiological temperature is a widespread phenomenon [42]. However, because we have only examined two species we are not able to confirm a trend between body temperature and chaperone-like activity in this study. Additional αA-crystallins from species with varying physiological temperatures will need to be examined.

This study demonstrates the insights that can be gained from a comparative approach to vertebrate α-crystallin that includes non-mammalian species. Specifically, we have shown that the thermal stabilities of homomultimeric α-crystallin tertiary structures reflect physiological temperature. Furthermore, the increase in zebrafish αA-crystallin chaperone-like activity compared to human αA-crystallin does not require a change in tertiary structure, and the truncation of the zebrafish αB-crystallin C-terminus alone does not account for its reduced chaperone-like activity. Our data also provide insights into the evolution of α-crystallin physiological function. The conserved primary and tertiary structure of zebrafish and mammalian αA-crystallin, their similar tissue specific expression, and the maintenance of chaperone-like activity all suggest that these proteins have been under similar selection pressures for their protective role since the evolutionary divergence between ray-finned fishes and mammals. These findings suggest that zebrafish could serve as a model for studies of human αA-crystallin. The evolution of αB-crystallin is more complex. The zebrafish αB-crystallin described in this study has diverged greatly from its human ortholog in primary structure, expression pattern and chaperone-like ability. This αB-crystallin clearly plays a different and more limited role in the zebrafish than in mammals. That role is apparently lens specific and does not involve significant chaperone function. The ubiquitous protective function performed by mammalian αB-crystallin is either lacking from zebrafish, or is being performed by another protein. This role may have been assumed by the recently discovered second transcribed copy of zebrafish αB-crystallin, described above. It is common for ancestral protein functions to be divided between the two derived copies after a gene duplication event [36]. This possibility could explain the higher chaperone-like activity of the catfish αB-crystallin found by Yu et al. [34]. Examination of this second zebrafish αB-crystallin is needed to confirm that it has retained the higher chaperone-like activity and widespread expression found in the human ortholog.

What we are not able to identify in this study are the specific amino acid changes that alter thermal stability or chaperone-like activity in zebrafish α-crystallins. While the percent identity between zebrafish and human αA-crystallin is high (73%), there are still 43 amino acid differences that could contribute to these variations. There are even more amino acid differences between zebrafish and human αB-crystallin. Determining how the evolution of primary structure modifies α-crystallin function will require comparison of more closely related species with less divergent amino acid sequences. If physiological temperature influences the evolution of thermal stability and the ability to bind non-native protein, as suggested in this study, comparison of closely related species differing in physiological temperature could provide a powerful model for analyzing the relationships between α-crystallin structure and function.


This study was supported by a grant from the National Institutes of Health (EY 13535) to MP. We are grateful to Mark Petrash for supplying the expression construct for human αA-crystallin, Qingling Huang for technical advice on the chaperone assays and to Mike Danko, Amber Smith and Jeff Adams for help in the production of recombinant α-crystallins.


1. Wistow G, Piatigorsky J. Recruitment of enzymes as lens structural proteins. Science 1987; 236:1554-6.

2. Bloemendal H, de Jong W, Jaenicke R, Lubsen NH, Slingsby C, Tardieu A. Ageing and vision: structure, stability and function of lens crystallins. Prog Biophys Mol Biol 2004; 86:407-85.

3. Klemenz R, Frohli E, Steiger RH, Schafer R, Aoyama A. Alpha B-crystallin is a small heat shock protein. Proc Natl Acad Sci U S A 1991; 88:3652-6.

4. Hawse JR, Cumming JR, Oppermann B, Sheets NL, Reddy VN, Kantorow M. Activation of metallothioneins and alpha-crystallin/sHSPs in human lens epithelial cells by specific metals and the metal content of aging clear human lenses. Invest Ophthalmol Vis Sci 2003; 44:672-9.

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

6. Jakob U, Gaestel M, Engel K, Buchner J. Small heat shock proteins are molecular chaperones. J Biol Chem 1993; 268:1517-20.

7. Bhat SP, Nagineni CN. alpha B subunit of lens-specific protein alpha-crystallin is present in other ocular and non-ocular tissues. Biochem Biophys Res Commun 1989; 158:319-25.

8. Posner M. A comparative view of alpha crystallins: the contribution of comparative studies to understanding function. Integr Comp Biol 2003; 43:481-91.

9. 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.

10. de Jong WW, Zweers A, Versteeg M, Nuy-Terwindt EC. Primary structures of the alpha-crystallin A chains of twenty-eight mammalian species, chicken and frog. Eur J Biochem 1984; 141:131-40.

11. Kokke BP, Boelens WC, de Jong WW. The lack of chaperonelike activity of Caenorhabditis elegans Hsp12.2 cannot be restored by domain swapping with human alphaB-crystallin. Cell Stress Chaperones 2001; 6:360-7.

12. Horwitz J. Alpha-crystallin. Exp Eye Res 2003; 76:145-53.

13. Kim R, Kim KK, Yokota H, Kim SH. Small heat shock protein of Methanococcus jannaschii, a hyperthermophile. Proc Natl Acad Sci U S A 1998; 95:9129-33.

14. 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.

15. Bova MP, Yaron O, Huang Q, Ding L, Haley DA, Stewart PL, Horwitz J. Mutation R120G in alphaB-crystallin, which is linked to a desmin-related myopathy, results in an irregular structure and defective chaperone-like function. Proc Natl Acad Sci U S A 1999; 96:6137-42.

16. Pras E, Frydman M, Levy-Nissenbaum E, Bakhan T, Raz J, Assia EI, Goldman B, Pras E. A nonsense mutation (W9X) in CRYAA causes autosomal recessive cataract in an inbred Jewish Persian family. Invest Ophthalmol Vis Sci 2000; 41:3511-5.

17. Shroff NP, Cherian-Shaw M, Bera S, Abraham EC. Mutation of R116C results in highly oligomerized alpha A-crystallin with modified structure and defective chaperone-like function. Biochemistry 2000; 39:1420-6.

18. Smulders RH, van Dijk MA, Hoevenaars S, Lindner RA, Carver JA, de Jong WW. The eye lens protein alphaA-crystallin of the blind mole rat Spalax ehrenbergi: effects of altered functional constraints. Exp Eye Res 2002; 74:285-91.

19. Hendriks W, Leunissen J, Nevo E, Bloemendal H, de Jong WW. The lens protein alpha A-crystallin of the blind mole rat, Spalax ehrenbergi: evolutionary change and functional constraints. Proc Natl Acad Sci U S A 1987; 84:5320-4.

20. Behrens M, Wilkens H, Schmale H. Cloning of the alphaA-crystallin genes of a blind cave form and the epigean form of Astyanax fasciatus: a comparative analysis of structure, expression and evolutionary conservation. Gene 1998; 216:319-26.

21. van Boekel MA, de Lange F, de Grip WJ, de Jong WW. Eye lens alphaA- and alphaB-crystallin: complex stability versus chaperone-like activity. Biochim Biophys Acta 1999; 1434:114-23.

22. Mchaourab HS, Dodson EK, Koteiche HA. Mechanism of chaperone function in small heat shock proteins. Two-mode binding of the excited states of T4 lysozyme mutants by alphaA-crystallin. J Biol Chem 2002; 277:40557-66.

23. Koteiche HA, McHaourab HS. Mechanism of chaperone function in small heat-shock proteins. Phosphorylation-induced activation of two-mode binding in alphaB-crystallin. J Biol Chem 2003; 278:10361-7.

24. Datta SA, Rao CM. Differential temperature-dependent chaperone-like activity of alphaA- and alphaB-crystallin homoaggregates. J Biol Chem 1999; 274:34773-8.

25. Bova MP, McHaourab HS, Han Y, Fung BK. Subunit exchange of small heat shock proteins. Analysis of oligomer formation of alphaA-crystallin and Hsp27 by fluorescence resonance energy transfer and site-directed truncations. J Biol Chem 2000; 275:1035-42.

26. Westerfield M, The zebrafish book: A guide for the laboratory use of zebrafish (Danio rerio). 4 ed. 2000, Eugene, Oregon: University of Oregon Press.

27. Posner M, Kantorow M, Horwitz J. Cloning, sequencing and differential expression of alphaB-crystallin in the zebrafish, Danio rerio. Biochim Biophys Acta 1999; 1447:271-7.

28. Runkle S, Hill J, Kantorow M, Horwitz J, Posner M. Sequence and spatial expression of zebrafish (Danio rerio) alphaA-crystallin. Mol Vis 2002; 8:45-50 <>.

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

30. Kiss AJ, Mirarefi AY, Ramakrishnan S, Zukoski CF, Devries AL, Cheng CH. Cold-stable eye lens crystallins of the Antarctic nototheniid toothfish Dissostichus mawsoni Norman. J Exp Biol 2004; 207:4633-49.

31. Chiou SH, Chang WC, Pan FM, Chang T, Lo TB. Physicochemical characterization of lens crystallins from the carp and biochemical comparison with other vertebrate and invertebrate crystallins. J Biochem (Tokyo) 1987; 101:751-9.

32. Liang JJ, Sun TX, Akhtar NJ. Heat-induced conformational change of human lens recombinant alphaA- and alphaB-crystallins. Mol Vis 2000; 6:10-4 <>.

33. McFall-Ngai MJ, Horwitz J. A comparative study of the thermal stability of the vertebrate eye lens: Antarctic ice fish to the desert iguana. Exp Eye Res 1990; 50:703-9.

34. Yu CM, Chang GG, Chang HC, Chiou SH. Cloning and characterization of a thermostable catfish alphaB-crystallin with chaperone-like activity at high temperatures. Exp Eye Res 2004; 79:249-61.

35. Franck E, Madsen O, van Rheede T, Ricard G, Huynen MA, de Jong WW. Evolutionary diversity of vertebrate small heat shock proteins. J Mol Evol 2004; 59:792-805.

36. Postlethwait J, Amores A, Force A, Yan YL. The zebrafish genome. Methods Cell Biol 1999; 60:149-63.

37. Van de Peer Y, Taylor JS, Meyer A. Are all fishes ancient polyploids? J Struct Funct Genomics 2003; 3:65-73.

38. Fitch WM. Homology a personal view on some of the problems. Trends Genet 2000; 16:227-31.

39. Smulders RHPH, Carver JA, Lindner RA, van Boekel MA, Bloemendal H, de Jong WW. Immobilization of the C-terminal extension of bovine alphaA-crystallin reduces chaperone-like activity. J Biol Chem 1996; 271:29060-6.

40. Carver JA, Lindner RA. NMR spectroscopy of alpha-crystallin. Insights into the structure, interactions and chaperone action of small heat-shock proteins. Int J Biol Macromol 1998; 22:197-209.

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

42. Hochachka PW and Somero GN. Biochemical Adaptation: Mechanism and Process in Physiological Evolution. Oxford: Oxford University Press; 2002.

Dahlman, Mol Vis 2005; 11:88-96 <>
©2005 Molecular Vision <>
ISSN 1090-0535