November 2000
Volume 41, Issue 12
Free
Lens  |   November 2000
In Vitro Filament-like Formation upon Interaction between Lens α-Crystallin and βL-Crystallin Promoted by Stress
Author Affiliations
  • Orly Weinreb
    From the Department of Biochemistry, University of Nijmegen, The Netherlands; and
  • Anke F. van Rijk
    From the Department of Biochemistry, University of Nijmegen, The Netherlands; and
  • Ahuva Dovrat
    B. Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel.
  • Hans Bloemendal
    From the Department of Biochemistry, University of Nijmegen, The Netherlands; and
Investigative Ophthalmology & Visual Science November 2000, Vol.41, 3893-3897. doi:
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Orly Weinreb, Anke F. van Rijk, Ahuva Dovrat, Hans Bloemendal; In Vitro Filament-like Formation upon Interaction between Lens α-Crystallin and βL-Crystallin Promoted by Stress. Invest. Ophthalmol. Vis. Sci. 2000;41(12):3893-3897.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

purpose. To determine whether α-crystallin is capable of forming filament-like structures with other members of the crystallin family.

methods. Water-soluble crystallins were isolated from calf lenses and fractionated into α-, βH-, βL-, and γ-crystallins according to standard procedures. Chaperone-like activity of α-crystallin was determined in control and UV-A–irradiated lenses by the heat-induced aggregation assay of βL-crystallin. Protein samples from this assay were analyzed by electron microscopy. In vitro filament formation was examined by transmission immunoelectron microscopy using specific antibodies directed against the crystallins. Involvement of intermediate filament constituents was excluded by the results of Western blot analysis, which were all negative. Moreover, the in vitro amyloid fibril interaction test using thioflavin T (ThT) was also performed.

results. At the supramolecular level heating at 60°C has no effect on the morphologic appearance of α-crystallin as observed by transmission electron microscopy. Moreover α-crystallin obtained from UV-A–irradiated lenses shows a virtually identical shape. However, heating in the presence of βL-crystallin results in the formation of filament-like αβ-hybrids as demonstrated by immunoelectron microscopy using specific antibodies directed either against α- orβ L-crystallin. Parallel experiments with α-crystallin derived from UV-A–irradiated lenses showed even more pronounced filamentous structures, compared with the controls. Nonetheless, we were able to show that the UV-light treatment affected the chaperone-like capacity of α-crystallin, as revealed by a diminished ability to inhibit in vitro denaturation of βL-crystallin. To exclude the presence of cytoskeletal contamination in the crystallin preparations, vimentin antibodies were also tested. These latter experiments were negative. The filamentous nature of the hybrids was further confirmed by the results obtained with the ThT assay earlier applied for the detection of amyloid fibrils.

conclusions. Crystallin hybrids have previously been detected in the water-soluble lens crystallin fraction. Our findings indicate that such endogenous hybrids, formerly called “rods,” may result from stress-induced interaction between α-crystallin and other lens constituents such as βL-crystallin. Because the hybrid formation is enhanced when α-crystallin from UV-A–irradiated lenses is used as one of the two components of the hybrid, one can only speculate that this formation may be one of the factors leading to UV-A cataract.

Two members of the small heat shock protein family, 1 2 αA- and αB-crystallin, possess molecular chaperone properties. 3 A decade ago it became apparent that these two polypeptides, which form 800-kDa aggregates in the lens, also exist in a variety of other tissues. 4 5 6 Current studies indicate that small heat shock proteins like αB-crystallin are able to interact with intermediate filaments in response to stress and to function as molecular chaperones. 7 8 9 10 Earlier ultrastructural observations showed that crude fractions from chicken lens consisted of 5- to 6-nm-thick core filaments and irregularly sized globular particles 15 to 20 nm in diameter called “beaded filaments.” 11 It was noted that the “beads” had dimensions that were similar to native α-crystallin. 12 Moreover, two proteins with molecular masses of 115 and 49 kDa, respectively (named filensin and phakinin), have been localized in the beaded filament fraction of the lens with the aid of immunoelectron microscopy. 13 14 However, the question still remains whether or not other lens proteins may be involved in the formation of filamentous structures. In this report we demonstrate that water-soluble α-crystallin has the ability to form, in response to heat stress, in vitro filament-like structures with one other crystallin, namely βL-crystallin. This filament formation is enhanced by UV-A irradiation. 
Methods
UV-A Irradiation
Lenses, excised from 2- to 4-year-old bovine eyes, were irradiated in special organ culture glass vessels described previously by Dovrat and Weinreb. 15 Briefly, a 400 W UV lamp (Vilber, Lourmat Cedex, France) contained a filter that provided radiation of 33 J/cm2 for 75 minutes at 365 nm. 
Fractionation of Crystallins
Lenses were dissected under a binocular stereomicroscope. The lens cortex was homogenized in 100 mM Tris buffer at pH 7.5 and spun at 4°C at 13,000g for 30 minutes. The supernatant comprises the water-soluble lens fraction. Separation of this fraction into α-,β H-, βL-, and γ-crystallin was carried out by gel filtration on a Sephacryl S-300 (Pharmacia-LKB, Uppsala, Sweden) HR column. 16  
Chaperone-like Activity
The chaperone-like activity of α-crystallin from control and UV-irradiated lenses was determined by the heat-induced aggregation assay of βL-crystallin at 60°C. 3 The proteins were dissolved in a solution of 20 mM sodium phosphate, 100 mM Na2SO4, 10 mM EDTA, at pH 6.9. The assay was performed at a concentration of 0.25 mg/ml substrate protein and 0.05 mg/ml α-crystallin. 
Electron Microscopy
Protein samples from the heating assay were also analyzed by electron transmission microscopy. Samples were negatively stained with uranyl acetate (1% v/v). The in vitro filament formation ofα -crystallin from control and UV-treated lenses with βL-crystallin was followed by immunoelectron microscopy using antibodies against vimentin, αA-, αB-, and βL-crystallin. The grids were examined with a transmission electron microscope (Jeol TEM1210, Tokyo, Japan) using 70 to 80 kV. 
Thioflavin T Interaction Assay
Fluorometric determinations were carried out using the thioflavin T (ThT) interaction assay at excitation and emission of 450 and 482 nm, respectively. 17  
Results
Electron Microscopy
Micrographs of mixtures of α- and βL-crystallin, obtained from control lenses and α-crystallin from UV-A–treated lenses, which were separately heated at 60°C, are shown in Figure 1 . Apparently heating has no effect on α-crystallin obtained from the control lenses (Fig. 1A) , because comparison with previously published electron micrographs of nonheated α-crystallin revealed no detectable morphologic differences. 18 Normal α-crystallin consists of molecules having an apparent spherical structure, with a diameter of approximately 17 nm. Likewise α-crystallin obtained from irradiated lenses shows a similar shape, albeit the size is somewhat smaller (Fig. 1B) . After incubation at 60°C, βL-crystallin (Fig. 1C) lost its irregular spherical shape when compared with nonheated β-crystallin as described earlier. 18  
The in vitro filament formation was also examined by immunoelectron microscopy (Fig. 2) . Anti–αB- and anti–βL-crystallin labeling yielded identical results (not shown). The formation of filament-like structures can be observed after the heating assay at 60°C using 0.05 mg α-crystallin obtained from control lenses with 0.25 mg βL-crystallin. The identical experiment with α-crystallin from irradiated lenses (Figs. 3A 3B) showed more pronounced filament-like structures compared with the control. The results show that UV-A irradiation promotes the filament-like formation. Experiments with anti-vimentin were negative showing that no intermediate filament component was involved in the crystallin hybrid formation. 
Chaperone-like Activity
The chaperone-like activity determined with the aid of the protein scattering at 360 nm is depicted in Figure 4 . It can be seen that this property of the water-soluble α-crystallin was affected by UV-A light. Compared with controls (curve II),α -crystallin derived from UV-A–irradiated lenses revealed a decreased ability to inhibit βL-crystallin denaturation in vitro (curve III). Curve IV represents βL-crystallin in the absence ofα -crystallin. Furthermore, α-crystallin obtained from control and UV- irradiated lenses did not denature during 30 minutes of incubation at 60°C (compare the coinciding curves Ia and Ib). These results are consistent with previous reports that described decreased chaperone-like activity of α-crystallin on UV-B irradiation. 19 20 21  
ThT Interaction Assay
The results of the ThT test are depicted in Figure 5 This assay, in which fibrils convert to a β-sheet configuration in vitro, has previously been successfully applied for detection of amyloid fibrils. 17 It can be seen that heatedβ L-crystallin or heated βL-crystallin plus α-crystallin from control lenses produced a 10 times higher fluorescence value than heated α-crystallin obtained from control and UV-irradiated lenses alone. The amount of fluorescence increased when heated βL-crystallin is assembled with α-crystallin from UV-treated lenses. 
Conclusion
Previously Slingsby et al. 22 suggested a new model for crystallin assembly in lens fiber cells. In the highly hydrated solution-like region of the lens, it is envisaged that weak interaction between subunits such as those of β-crystallin will occur, forming elements of a network with dynamic branching. An open gel structure would maintain protein–protein interactions at a high concentration, covering the more prominent hydrophobic regions and preventing random aggregation of subunits. This may possibly explain the present observation that (heated) βL-crystallin assembles withα -crystallin, resulting in filament-like structures. It cannot be excluded that one or more of UV-A–provoked alterations 23 are related to the ability of water-soluble α-crystallin to form filaments in vitro more efficiently than with α-crystallin derived from control lenses. The in vitro filament-like chains identified by electron microscopy after irradiation have a high degree of morphologic similarity to the αβ-hybrids that have been described previously after reconstitution of the dissociated total mixture of the water-soluble crystallins. 18 Dhir et al. 24 have recently shown by in vitro UV-A irradiation of recombinantα A-crystallin that sensitized photooxidation can occur in amino acids other than Trp in the presence of kynurenine or 3-hydroxykynurenine with effects similar to, albeit smaller than, direct UV-B photooxidation. In the old lens, other types of sensitizers may be operative, such as advanced glycation end products (AGE). Finley et al., 25 studying the photooxidation sites in bovineα A-crystallin, found that in addition to Trp, Met and His were photooxidized. Their conclusion is that the N-terminal region ofα A-crystallin is exposed to an aqueous environment and is in the vicinity of Trp from neighboring subunits. Albeit we did not try to identify the exact site of photooxidation being beyond the aim of our study, it might well be that particularly AGE could play a role as sensitizer because we used adult bovine lenses. Besides, the relatively large amount of NAD(P)H in bovine lens could also initiate photochemical processes as it does in human and rabbit lens cells. 26 Furthermore, the ThT interaction assay, which is used as a method for the demonstration of β-sheet conformation and which appeared previously to be a useful tool for detection of amyloid fibrils in vitro, 17 provided additional evidence for possible αβ-crystallin filament formation (Fig. 5) . According to Levine, 27 it is very likely that both the β-sheet conformation and the aggregation state provide the environment to stabilize the long wavelength ThT fluorescent complex, regardless of the identity of the participating peptides. Therefore, at least some of the endogenous filament-like structures that have been demonstrated in the lens may result from interaction of α-crystallin with other proteins such as βL-crystallin under stress conditions. This might provide a clue regarding the processes leading to the development of UV cataract. 
 
Figure 1.
 
Electron micrographs of (A) α-crystallin obtained from control lenses (bar, 200 nm); (B) α-crystallin from UV-A–treated lenses (bar, 200 nm); (C) βL-crystallin obtained from control lenses separately heated at 60°C (bar, 500 nm). Complexes were visualized by negative staining with uranyl acetate.
Figure 1.
 
Electron micrographs of (A) α-crystallin obtained from control lenses (bar, 200 nm); (B) α-crystallin from UV-A–treated lenses (bar, 200 nm); (C) βL-crystallin obtained from control lenses separately heated at 60°C (bar, 500 nm). Complexes were visualized by negative staining with uranyl acetate.
Figure 2.
 
Immunogold labeling with anti–αA-crystallin of samples from the heating assay. 0.05 mg of α-crystallin from control lenses incubated with 0.25 mg βL-crystallin at 60°C. F, filament-like chains; αA, labeling (bar, 500 nm).
Figure 2.
 
Immunogold labeling with anti–αA-crystallin of samples from the heating assay. 0.05 mg of α-crystallin from control lenses incubated with 0.25 mg βL-crystallin at 60°C. F, filament-like chains; αA, labeling (bar, 500 nm).
Figure 3.
 
(A) Filament samples from the heating assay. α-Crystallin, 0.05 mg, from UV-A–treated lenses incubated with 0.25 mgβ L-crystallin at 60°C (bar, 500 nm); (B) filaments at higher magnification (bar, 100 nm). F, filament-like structures; long arrows in (B), labeling with anti–αA-crystallin.
Figure 3.
 
(A) Filament samples from the heating assay. α-Crystallin, 0.05 mg, from UV-A–treated lenses incubated with 0.25 mgβ L-crystallin at 60°C (bar, 500 nm); (B) filaments at higher magnification (bar, 100 nm). F, filament-like structures; long arrows in (B), labeling with anti–αA-crystallin.
Figure 4.
 
Chaperone-like activity of water-soluble α-crystallin fraction obtained from the cortex of control and UV-A–irradiated bovine lenses determined by heating assay at 60°C with 0.25 mg ofβ L-crystallin and 0.05 mg of α-crystallin.
Figure 4.
 
Chaperone-like activity of water-soluble α-crystallin fraction obtained from the cortex of control and UV-A–irradiated bovine lenses determined by heating assay at 60°C with 0.25 mg ofβ L-crystallin and 0.05 mg of α-crystallin.
Figure 5.
 
Fluorescence determination using Thioflavin T (ThT) interaction with samples obtained from the heating assay, measured at excitation of 450 nm and emission of 482 nm. ThT concentration was 250 nM. The reaction buffer contained 50 mM glycine-NaOH at pH 6.0. Bars, SD (in four experiments).
Figure 5.
 
Fluorescence determination using Thioflavin T (ThT) interaction with samples obtained from the heating assay, measured at excitation of 450 nm and emission of 482 nm. ThT concentration was 250 nM. The reaction buffer contained 50 mM glycine-NaOH at pH 6.0. Bars, SD (in four experiments).
We thank Wilfried W. de Jong for fruitful discussions and Lucio Benedetti for advice concerning electron microscopy. 
van den Heuvel R, Hendriks W, Quax W, Bloemendal H. Complete structure of the hamster αA-crystallin gene: reflection of an evolutionary history by means of exon shuffling. J Mol Biol. 1985;185:273–284. [CrossRef] [PubMed]
Bloemendal H, de Jong WW. Lens proteins and their genes. Progr Nucleic Acids Res Mol Biol. 1991;41:259–281.
Horwitz J. α-Crystallin can function as a molecular chaperone. Proc Natl Acad Sci USA. 1992;89:10449–10453. [CrossRef] [PubMed]
Bhat SD, Nagineni CN. αB Subunit of lens-specific protein α-crystallin is present in other ocular and non-ocular tissues. Biochem Biophys Res Commun. 1989;158:319–325. [CrossRef] [PubMed]
Dubin RA, Wawrousek EF, Piatigorsky J. Expression of the murine αB-crystallin gene is not restricted to the lens. Mol Cell Biol. 1989;9:103–191.
Kato K, Shinohara H, Kurobe N, Goto S, Inaguma Y, Ohshima K. Immunoreactive αA-crystallin in rat non-lenticular tissues detected with a sensitive immunoassay method. Biophys Biochim Acta. 1991;1080:173–180. [CrossRef]
Nicholl ID, Quinlan RA. Chaperone activity of α-crystallin modulates intermediate filaments assembly. EMBO J. 1994;13:945–953. [PubMed]
Liang P, MacRae TH. Molecular chaperones and the cytoskeleton. J Cell Sci. 1997;110:1431–1440. [PubMed]
Djabali K, de Nechaud B, Landon F, Portier MM. αB-Crystallin interacts with intermediate filaments in response to stress. J Cell Sci. 1997;110:2759–2769. [PubMed]
Muchowski PJ, Valdez MM, Clark JI. αB-Crystallin selectively targets intermediate filament proteins during thermal stress. Invest Ophthalmol Vis Sci. 1999;40:951–958. [PubMed]
Maisel H, Perry MM. Electron microscope observation on some structural proteins of the chick lens. Exp Eye Res. 1972;14:7–12. [CrossRef] [PubMed]
Benedetti EL, Dunia I, Bentzel CJ, Vermorken AJM, Kibbelaar M, Bloemendal H. A portrait of plasma membrane specialization in eye lens epithelium and fibers. Biochim Biophys Acta. 1976;457:353–384. [CrossRef] [PubMed]
Ireland M, Maisel H. A family of lens fiber cell specific proteins. Lens Eye Toxic Res. 1989;6:623–638. [PubMed]
FitzGerald PG, Gottlieb W. The Mr 115 KD fiber cell-specific protein of a component of the lens cytoskeleton. Curr Eye Res. 1989;8:801–811. [CrossRef] [PubMed]
Dovrat A, Weinreb O. Recovery of lens optics and epithelial enzymes after ultraviolet A radiation. Invest Ophthalmol Vis Sci. 1995;36:2417–2424. [PubMed]
Slingsby C, Bateman OA. Rapid separation of bovine β-crystallin subunits βB1, βB2, βB3, βA3 and βA4. Exp Eye Res. 1991;51:21–26.
Naiki H, Higuchi K, Hossokawa M, Takeda T. Fluorometric determination of amyloid fibrils in vitro using fluorescence dye, Thioflavine T. Anal Biochem. 1989;177:244–249. [CrossRef] [PubMed]
Bloemendal H, Zweers A, Benedetti EL, Walter H. Selective reassociation of the crystallins. Exp Eye Res. 1975;20:463–478. [CrossRef] [PubMed]
Borkman RF, McLaughlin J. The molecular chaperone function of α-crystallin is impaired by UV photolysis. Photochem Photobiol. 1995;62:1046–1051. [PubMed]
Schauerte JA, Gafni A. Photodegradation of tryptophan residues and attenuation of molecular chaperone activity in α-crystallin are correlated. Biochem Biophys Res Commun. 1995;212:900–905. [CrossRef] [PubMed]
Ellozy AR, Ceger P, Wang RH, Dillon J. Effect of the UV modification of α-crystallin on its ability to suppress nonspecific aggregation. Photochem Photobiol. 1996;64:344–348. [CrossRef] [PubMed]
Slingsby C, Norledge B, Simpson A, et al. X-ray diffraction and structure of crystallins. Prog Retinal Eye Res. 1997;16:3–29. [CrossRef]
Weinreb O, van Rijk AF, Steely HT, Dovrat A, Bloemendal H. Analysis of UV-A-related alterations upon aging of eye lens proteins by mini-two-dimensional polyacrylamide gel electrophoresis. Ophthal Res. 2000;32(5)196–204.
Dhir P, Nila JA, Sun TX, Liang JJN. Photooxidized products of recombinant αA-crystallin and W9F mutant. Photochem Photobiol. 1999;69:329–335. [CrossRef] [PubMed]
Finley EL, Busman M, Dillon J, Crouch RK, Schey KL. Identification of: photooxidation sites in bovine α-crystallin. Photochem Photobiol. 1997;66:635–641. [CrossRef] [PubMed]
Atherton SJ, Lambert C, Schultz J, Williams N, Zigman S. Fluorescence studies of lens epithelial cells and their constituents. Photochem Photobiol. 1999;70:823–828. [CrossRef] [PubMed]
Levine H. Thioflavin T interaction with synthetic Alzheimer’s disease β-amyloid peptides: detection of amyloid aggregation in solution. Protein Sci. 1993;2:404–410. [PubMed]
Figure 1.
 
Electron micrographs of (A) α-crystallin obtained from control lenses (bar, 200 nm); (B) α-crystallin from UV-A–treated lenses (bar, 200 nm); (C) βL-crystallin obtained from control lenses separately heated at 60°C (bar, 500 nm). Complexes were visualized by negative staining with uranyl acetate.
Figure 1.
 
Electron micrographs of (A) α-crystallin obtained from control lenses (bar, 200 nm); (B) α-crystallin from UV-A–treated lenses (bar, 200 nm); (C) βL-crystallin obtained from control lenses separately heated at 60°C (bar, 500 nm). Complexes were visualized by negative staining with uranyl acetate.
Figure 2.
 
Immunogold labeling with anti–αA-crystallin of samples from the heating assay. 0.05 mg of α-crystallin from control lenses incubated with 0.25 mg βL-crystallin at 60°C. F, filament-like chains; αA, labeling (bar, 500 nm).
Figure 2.
 
Immunogold labeling with anti–αA-crystallin of samples from the heating assay. 0.05 mg of α-crystallin from control lenses incubated with 0.25 mg βL-crystallin at 60°C. F, filament-like chains; αA, labeling (bar, 500 nm).
Figure 3.
 
(A) Filament samples from the heating assay. α-Crystallin, 0.05 mg, from UV-A–treated lenses incubated with 0.25 mgβ L-crystallin at 60°C (bar, 500 nm); (B) filaments at higher magnification (bar, 100 nm). F, filament-like structures; long arrows in (B), labeling with anti–αA-crystallin.
Figure 3.
 
(A) Filament samples from the heating assay. α-Crystallin, 0.05 mg, from UV-A–treated lenses incubated with 0.25 mgβ L-crystallin at 60°C (bar, 500 nm); (B) filaments at higher magnification (bar, 100 nm). F, filament-like structures; long arrows in (B), labeling with anti–αA-crystallin.
Figure 4.
 
Chaperone-like activity of water-soluble α-crystallin fraction obtained from the cortex of control and UV-A–irradiated bovine lenses determined by heating assay at 60°C with 0.25 mg ofβ L-crystallin and 0.05 mg of α-crystallin.
Figure 4.
 
Chaperone-like activity of water-soluble α-crystallin fraction obtained from the cortex of control and UV-A–irradiated bovine lenses determined by heating assay at 60°C with 0.25 mg ofβ L-crystallin and 0.05 mg of α-crystallin.
Figure 5.
 
Fluorescence determination using Thioflavin T (ThT) interaction with samples obtained from the heating assay, measured at excitation of 450 nm and emission of 482 nm. ThT concentration was 250 nM. The reaction buffer contained 50 mM glycine-NaOH at pH 6.0. Bars, SD (in four experiments).
Figure 5.
 
Fluorescence determination using Thioflavin T (ThT) interaction with samples obtained from the heating assay, measured at excitation of 450 nm and emission of 482 nm. ThT concentration was 250 nM. The reaction buffer contained 50 mM glycine-NaOH at pH 6.0. Bars, SD (in four experiments).
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×