| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
D crystallin, a protein involved in cataract formation
Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA
Reprint requests to: Jonathan King, Department of Biology, Massachusetts Institute of Technology, Building 68, Room 330, 31 Ames Street, Cambridge, MA 02139, USA; e-mail: jaking{at}mit.edu; fax: (617) 252-1843.
(RECEIVED July 26, 2002; FINAL REVISION December 13, 2002; ACCEPTED December 12, 2002)
Article and publication are at http://www.proteinscience.org/cgi/doi/10.1110/ps.0225503.
 |
Abstract |
|---|
 TOP Abstract IntroductionResultsDiscussionMaterials and methodsReferences |
|---|
D crystallin (HD-Crys), a major protein of the human eye lens, is a primary component of cataracts. This 174-residue primarily ß-sheet protein is made up of four Greek keys separated into two domains. Mutations in the human gene sequence encoding HD-Crys are implicated in early-onset cataracts in children, and the mutant protein expressed in Escherichia coli exhibits properties that reflect the in vivo pathology. We have characterized the unfolding, refolding, and competing aggregation of human wild-type HD-Crys as a function of guanidinium hydrochloride (GuHCl) concentration at neutral pH and 37°C, using intrinsic tryptophan fluorescence to monitor in vitro folding. Wild-type HD-Crys exhibited reversible refolding above 1.0 M GuHCl. The GuHCl unfolded protein was more fluorescent than its native counterpart despite the absence of metal or ion-tryptophan interactions. Aggregation of refolding intermediates of HD-Crys was observed in both equilibrium and kinetic refolding processes. The aggregation pathway competed with productive refolding at denaturant concentrations below 1.0 M GuHCl, beyond the major conformational transition region. Atomic force microscopy of samples under aggregating conditions revealed the sequential appearance of small nuclei, thin protofibrils, and fiber bundles. The HD-Crys fibrous aggregate species bound bisANS appreciably, indicating the presence of exposed hydrophobic pockets. The mechanism of HD-Crys aggregation may provide clues to understanding age-onset cataract formation in vivo.
Keywords: Human D crystallin; protein folding; aggregation; cataracts; atomic force microscopy; hysteresis
|
Introduction |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Pathological studies of cataractous lenses have revealed that cataracts are composed of protein aggregates that precipitate in lens cells of the eye. The insoluble protein species obstructs the passage of light through the lens, thereby blocking light from reaching the photoreceptors in the retina (Benedek 1997).
The human eye lens, as a tissue, is composed of layers of fibrous cells that continuously grow with age. Crystallins comprise 90% of the total protein content of the lens (Oyster 1999). The ubiquitous crystallins are expressed primarily early in life and, therefore, must remain transparent throughout a person’s lifetime despite the high protein concentration in the lens and the continued presence of oxidative stress from atmospheric oxygen and UV/Vis light.
In addition to the unique cellular structure of the lens, the overall protein concentration within these cells is extremely high (approaching 70% g/g wet volume). A short-range order exists between the tightly packed crystallin proteins in the lens that provides minimal solution turbidity and a high degree of light transparency (Delaye and Tardieu 1983).
Cataracts removed from the human eye lens are composed of different species of aggregated crystallins. The cataractous crystallin proteins may be divided into two categories: 
-crystallins and ß-crystallins. -Crystallin is a member of a class of small heat-shock proteins thought to bind to unfolded polypeptide chains during times of stress and is thus crucial to preventing protein aggregation (Clark and Muchowski 2000; Horwitz 2000; MacRae 2000).
ß-Crystallins are small 20–30-kD proteins primarily composed of antiparallel ß-sheets. ß-Crystallins and -crystallins are structurally similar. They are both comprised of four Greek-key motifs separated into two domains (Wistow et al. 1983). The domains are very similar and appear to be the result of gene duplication during evolution. The ß-crystallins form domain-swapped dimers in solution owing to their flexible linker sequence, whereas the -crystallins are monomeric in solution (Jaenicke 1999). In addition, the -crystallins are the only known crystallins having attractive forces between molecules (Tardieu et al. 1992; Clark 1994). Extensive biophysical studies have been performed on members of the ß-crystallin family proteins in vitro (Rudolph et al. 1990; Pande et al. 1991; Mayr et al. 1997; Norledge et al. 1997; Slingsby et al. 1997; Wenk et al. 2000).
Human D crystallin (HD-Crys) is a 173-amino-acid protein. HD-Crys has a high sequence similarity to its bovine homologs, bovine D crystallin (
86%) and bovine B crystallin (76%; Slingsby et al. 1997). The human sequence homology modeled using the bovine -crystallin 3D structures is shown in Figure 1
(Peitsch 1995, 1996; Guex and Peitsch 1997). The structure of the human protein has been determined from crystals grown from the human recombinant protein and is fully homologous with the bovine structure (C. Slingsby, pers. comm.). HD-Crys has four intrinsic tryptophans that may be used to probe the unfolding/refolding progression of the molecule with fluorescence spectroscopy.
|
D-Crys, including R114C, R38H, and R36S (Heon et al. 1999; Kmoch et al. 2000). Site-specific mutagenesis of recombinant HD-Crys in collaboration with the laboratory of George Benedek (MIT Physics Department), confirmed that these substitutions influence the protein’s phase-transition characteristics in vitro (Pande et al. 2000, 2001). The mutation R114C resulted in the presence of an extra solvent-exposed cysteine that formed a disulfide bond with the endogenous solvent-exposed cysteine, Cys 110 (Pande et al. 2000). The disulfide linkage caused the formation of high-molecular-weight oligomers that precipitated out of solution in vitro. The mutations R38H and R36S are associated with aculeiform and congenital juvenile-onset cataracts, respectively. These juvenile cataracts have distinctive morphologies that indicate protein crystallization occurs within the lens. The purified mutant proteins exhibited decreased solubility in vitro (Pande et al. 2001). This alteration in the phase-separation characteristics was caused by an increased crystal nucleation rate of HD-Crys, consistent with the pathology of the inherited amino acid substitutions. Based on the importance of HD-Crys in juvenile-onset cataracts, the protein is likely to be an important substrate in the formation of age-onset cataracts as well.
Numerous studies using polypeptides, including many conferring disease phenotypes, such as -synuclein, transthyretin, and the Aß-42 peptide, have demonstrated that protein aggregation is often an ordered polymerization process that proceeds via a distinct mechanistic pathway involving a series of specific nonnative interactions (Speed et al. 1995; Harper et al. 1997; Lashuel et al. 1998; Betts et al. 1999; Li et al. 2001). The aggregation pathway of the ß-sheet P22 tailspike protein is known to proceed from a destabilized folding intermediate (Haase-Pettingell and King 1988; Speed et al. 1995). We were interested in the possibility that the mechanism of aggregation of HD-Crys in the aging lens differed from juvenile-onset genetic cataracts and proceeded from a partially unfolded conformer of the wild-type protein (Mitraki and King 1989; Wetzel 1994).
The unfolding and refolding of bovine B crystallin has been carefully studied by Jaenicke and his colleagues (Rudolph et al. 1990; Mayr et al. 1997; Jaenicke 1999). They described a complex unfolding transition indicating the presence of a partially folded intermediate with one of the domains ordered and the other disordered.
To explore the relationship of protein folding and cataract formation, we have performed a series of unfolding/refolding studies on HD-Crys at or near physiological pH and temperature (pH 7.0, 37°C). The experiments identify an aggregation-prone state of HD-Crys populated in low concentration of guanidine hydrochloride (GuHCl). We present direct evidence from atomic force microscopy that the aggregation pathway of human D crystallin is ordered and that the protein forms fibrils in vitro.
|
Results |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
D-crystallinD crystallin was expressed from a pET.16 plasmid in Escherichia coli strain BL21(DE3). The recombinant HD-Crys expressed in E. coli folded into soluble protein subunits that were stable in the bacterial cytoplasm. Methods developed for the purification of D crystallin from the bovine lens were used to purify the crystallin subunits from the bacterial lysate (Pande et al. 2000). Recombinant HD-Crys contained an N-terminal methionine that is posttranslationally cleaved in the native protein. The crystallin purified from E. coli had far-UV CD (Fig. 2) and absorption spectra (data not shown) similar to that observed for soluble bovine D crystallin extracted from calf lenses (Hay et al. 1993). X-Ray diffraction studies of crystals grown from the recombinant protein indicated the same fold as that found for the bovine protein (C. Slingsby, pers. comm.).
|
D crystallin did not denature in urea at concentrations in excess of 8 M, but did exhibit a denaturation transition in GuHCl. Unfolding and refolding transitions were examined using GuHCl as the denaturant in equilibrium analyses. Buffers were maintained at pH 7 and temperatures of 25°C or 37°C. Under these conditions an unfolding transition occurred in the range of 2.0–3.3 M GuHCl.
With respect to the overall tryptophan fluorescence, the native state of the protein was quenched to a greater extent than the denatured state (Fig. 3A). Based on the hypothesized structure of HD-Crys, native-state fluorescence quenching may occur because each of the four tryptophan side chains present in the protein are within close proximity to polar residues. The maximum fluorescence intensity of native HD-Crys occurred at 325 nm, whereas the maximum fluorescence intensity of denatured protein occurred at 348 nm.
|
D-Crys at 350 nm showed that a single major unfolding transition occurred (Fig. 3B
). Control titrations with sodium chloride demonstrated that fluorescence changes observed in high and low denaturant concentrations reflected structural alterations in the protein as opposed to high salt effects from GuHCl (data not shown). UV/Vis solution turbidity experiments did not reveal the presence of any high-molecular-weight structures in the unfolding samples.
At 37°C, the refolding transition of HD-Crys appeared to be reversible in GuHCl concentrations above 1.0 M GuHCl, although with the suggestion of a hysteresis in the transition region (Fig. 3B). Dilution of denatured protein to GuHCl concentrations lower than 1.0 M gave rise to a protein species with nonnative tryptophan environments as demonstrated by increased fluorescence intensity and wavelength maximum shifts to 330 nm (Fig. 3A). These samples contained high-molecular-weight protein aggregates as evidenced by solution turbidity measurements.
Equilibrium refolding and unfolding of HD-Crys as a function of GuHCl concentration at 25°C was similar to 37°C, but the refolding and unfolding transition curves did not overlay and exhibited hysteresis. The midpoint of the unfolding transition at 25°C was 3.7 M GuHCl, whereas the midpoint of the refolding transition was 2.7 M. At 37°C the midpoints of both the unfolding and refolding transitions were 2.7 M GuHCl. The refolding aggregation reaction at low GuHCl concentration observed at 37°C was also visible at 25°C.
To further examine the nature of the aggregation reaction, we characterized the aggregation reaction as a function of sample incubation time after initiation of refolding at 37°C. Native protein exhibited no appreciable increase in solution turbidity after 40 h of incubation under these conditions. The location of the aggregation transition was not altered with increasing times of incubation after dilution from 3 to 40 h. This indicated that the reaction depended on the population of a distinct conformer during refolding in 1 M GuHCl and did not require a rate-limiting nucleation event (Fig. 4). Samples containing a final GuHCl concentration >1.0 M did not exhibit appreciable solution turbidity even after 40 h. We did not observe aggregation in the major conformational transition occurring from 2 to 3.3 M GuHCl, indicating that intermediates populated in this transition were not the precursors of the aggregated state.
|
D-Crys protein refolded in aggregation-prone conditions (0.55 M GuHCl) was centrifuged at 12,000 rev/min. Solution turbidity measurements indicated no high-molecular-weight aggregates remained in the supernatant after centrifugation (data not shown). The fluorescence spectrum of the resulting supernatant revealed that a significant portion of the protein was lost to precipitation, but the fraction remaining had a maximum intensity wavelength indistinguishable from native (Fig. 3A). This indicated that there was a native-like species present at the end of the reaction and that the off-pathway aggregation process competed with a productive refolding pathway under aggregation-prone conditions. Integration of the fluorescence wavelength spectra of the resulting supernatant in the experiment shown in Figure 3 revealed that 50% ± 2% of the protein was refolded productively and 50% ± 2% of the protein became incorporated into the aggregate.
Kinetic unfolding of HD-Crys through a partially unfolded intermediate
The GdnHCl-induced unfolding of HD-Crys was monitored over time with fluorescence at 37°C (Fig. 5; Table 1). A possible early intermediate (Iu1) may have been populated within the dead time of the experiment as indicated by the burst of fluorescence intensity at the onset of the experiment. The only observable intermediate (Iu2) formed with a t
of 55 sec. This intermediate was not as quenched as native protein, indicating the polar-tryptophan interaction had been disrupted. In the final unfolding transition (Iu2 
denatured), occurring with a t of 2700 sec, a slow requenching took place, indicative of solvent rearrangement or isolated local restructuring in the unfolded state. The unfolding process was completed in 2 h.
 | (1) |
D-Crys showed the same kinetic rates and intermediates during unfolding regardless of pH, temperature, starting concentration of protein, and presence or absence of DTT (data not shown).
|
|
D-Crys under aggregation prone conditions (Fig. 6; Table 1). The solution turbidity of the first observable species was significantly higher than the native control, indicating that a high-molecular-weight species (I1*) had already formed within the dead time of the experiment. The formation of the first visible high-molecular-weight species (I2*) occurred with an approximate t = 60 sec, whereas the second state change (I2* to aggregate) had an approximate t = 3000 sec. This second transition may have reflected a slow structural transition of the multimeric protein from a highly turbid species to a less turbid species, but also may have corresponded to the progressive precipitation of the aggregate from solution.
 | (2) |
|
D-Crys without interference from light scattering, fluorescence experiments were performed on HD-Crys refolding to a final denaturant concentration of 1.5 M GuHCl (Fig. 7; Table 1). Under these conditions, fluorescence equilibrium data revealed the refolding process favored a native-like conformation and no high-molecular-weight species were detected in solution turbidity measurements. Productive refolding had a possible early intermediate (I1) that may have formed within the dead time of the experiment as evidenced by the burst in fluorescence observed at 15 sec. Spectroscopically, the only visible intermediate (I2) formed with a t = 73 sec and then was converted to native protein with a t = 1030 sec. Under these conditions, productively refolded HD-Crys reached equilibrium in 3 h, whereas refolding HD-Crys under aggregation-prone conditions did not reach equilibrium until 4 h had passed.
 | (3) |
|
D-Crys aggregate has a fibril structureD-Crys. Native crystallin from an S buffer solution was visible upon nonselective binding to the mica surface. The molecules had an apparent height of 0.5 nm, a length of 10 nm, and a width of 10 nm (Fig. 8A). The discrepancy between this value and the predicted dimensions of the protein was presumably a result of molecular flattening upon binding, resolution limits of the technique, or dimerization of the protein under the sample preparation conditions. The molecules appeared to contain a region of low height in the very center.
|
D-Crys was denatured in 5.5 M GuHCl and subsequently refolded using the aggregation-prone conditions (dilution to 0.55 M GuHCl). Samples of 10 µL were removed and applied to a mica grid at various intervals. Within the first minute of refolding, small globular structures were visible (Fig. 8B). In the background of this image, native-like HD-Crys monomers are already visible, supporting the competing aggregation and productive pathway model for refolding under these conditions. After 5 min of refolding, long, thin, 1-nm-high, 15-nm-wide protofibril structures were observed (Fig. 8C). The protofibrils had an approximate width and height of the soluble crystallin. After 24 min of refolding, protofibrils were visible but were found only in associated masses (Fig. 8D). The protofibrils had begun to associate from the center, producing a species with one thick midregion and multiple fibril-width tail ends. By 1 h of refolding, virtually all of the protofibrils had disappeared, presumably being incorporated into the thick 5-nm-high, 50-nm-wide branched and unbranched fiber bundles (Fig. 8E). Although the exact height and width of the fiber bundles varied, their overall appearance and approximate size were similar. After several hours of refolding, aggregated masses were visible, presumably containing associated fiber bundles (Fig. 8F).
Characterization of the HD-Crys aggregate shows it contains exposed hydrophobic pockets
HD-Crys aggregates were characterized by binding assays with bisANS. All aggregated species, including nuclei presumably present as early as 30 sec after the start of refolding and fiber bundles present at 4 h, bound bisANS significantly better than either the unfolded or folded controls (Fig. 9A,B). This indicated that all HD-Crys aggregates under these conditions had nonnative patches of exposed hydrophobic residues.
|
|
Discussion |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
D crystallin was resistant to denaturation by concentrations of urea up to 8 M at neutral pH, but could be denatured upon incubation with high concentrations of GuHCl. The midpoint of the transition occurred at 2.7 M GuHCl, at 37°C.
The fluorescence emission intensity of native HD-Crys was representative of the emission spectrum of quenched tryptophans. Wistow et al. (1983) suggested that fluorescence quenching of the native state observed in bovine -crystallin proteins is caused by tryptophan–cysteine interactions. The native-state quenching of HD-Crys may be due to the interaction of cysteine thiols 19 and 79 with tryptophan 43, cysteine 33 with tryptophans 69, and/or histidine 88 with tryptophan 131. This phenomenon has been observed in other proteins such as cellular retinoic-acid-binding protein and human serum transferring N-lobe (Eyles and Gierasch 2000; He et al. 2001). The packing of cysteines against tryptophans may provide a mechanism of protection in the eye from free radical damage initiated by ultraviolet radiation absorption by tryptophans (Davies and Truscott 2001). However, the source of the quenching has not been experimentally determined and may be associated with other residues in the hydrophobic core.
The denatured chains of HD-Crys could be refolded by dilution from GuHCl, and the reaction was reversible with one distinct transition in the range of 1–5 M GuHCl. Bovine B crystallin, which is denatured by urea at pH 2.0, but not at pH 7.0, was reversible over all reported concentrations of GuHCl, but exhibited a three-stage transition in equilibrium unfolding studies, representing sequential denaturation of the C-terminal and N-terminal domains at pH 2.0 (Rudolph et al. 1990; Jaenicke 1999). On the other hand, the closely homologous protein S, a protein also containing two Greek key domains, unfolded and refolded without evidence of separate domain transitions (Wenk et al. 2000). HD-Crys may possess differential domain stability that is not detected in the apparent two-state unfolding transition observed during equilibrium unfolding. Future experiments will attempt to characterize the stability of each domain of the human protein.
The in vitro unfolding and refolding steps of human D crystallin were relatively slow compared to some small proteins such as RNase or barstar (Nolting et al. 1995; Hollien and Marqusee 2002), but similar to rates found for some other ß-sheet proteins like apo-pseudoazurin (Reader et al. 2001).
During refolding, dilution into concentrations of GuHCl below 1.0 M at 37°C resulted in the population of an intermediate that aggregated irreversibly. The aggregating species was populated at low GuHCl concentrations both at 37°C and at 25°C. At both temperatures, the major conformational transition measured by fluorescence spectroscopy was in the range of 2–3.3 M GuHCl. This was well separated from the GuHCl concentration in which the aggregation reaction was detected. This indicates that the aggregation-prone intermediate differs from the species populated in the transition or that perhaps similar intermediates are populated under both conditions but have increased solubility in higher GuHCl concentrations.
This behavior differs somewhat from other proteins in which aggregation competes with productive refolding, such as phosphoglycerate kinase, ß-galactosidase, interleukin 1-ß, and transthyretin (Ghelis and Yon 1982; Colon and Kelly 1992; Wetzel and Chrunyk 1994). For these proteins, dilution to intermediate concentrations of denaturant generated an aggregating intermediate, whereas dilution to lower concentrations led to the recovery of the native fold. Equilibrium analysis with GuHCl demonstrated that in HD-Crys, the aggregating species probably was not generated from a transition-state intermediate.
Although not well defined at 37°C, a hysteresis between the unfolding and refolding equilibrium curves was very clear at 25°C. The refolding curves had the same midpoint of transition at both temperatures; however, unfolding at 37°C required lower GuHCl concentrations, indicating a thermal destabilization of the native state.
The existence of hysteresis in protein refolding experiments is thought to reflect transformations that are kinetically controlled through high energy barriers between the conformational transitions needed for refolding. The folding pathways of a number of ß-sheet proteins are kinetically controlled. The parallel ß-helical tailspike adhesin contains an interdigitated triple ß-sheet that acts as a molecular clamp (Sturtevant et al. 1989; Chen and King 1991; Kreisberg et al. 2001). The hysteresis that occurs during refolding of the recombinant mouse prion protein, in the transformation from the - to the ß-isoform, reflects a very high energy barrier (Baskakov et al. 2001). Under refolding conditions, alkaline phosphatase exhibits a reorganization of the environment around tryptophan 109 to the native state that takes days (Subramaniam et al. 1995). For HD-Crys, the hysteresis was reduced at higher temperature, consistent with overcoming a kinetic barrier. Correct packing of the domain buried cores or the formation of the correct domain interface are candidates for the slow step. The molecular basis of crystallin hysteresis requires further investigation (Sinclair et al. 1994; Lai et al. 1997).
Formation of filamentous aggregates
AFM images reveal that the in vitro aggregate of HD-Crys was ordered. The aggregate had a filamentous appearance as would be expected from the polymerization of defined subunits. At early times during aggregation, globular molecules, presumably representing soluble crystallin species, were present. These were 2–5 times the size of the native crystallin, estimated from the AFM images. These species are candidates for aggregation nuclei. The exposed surfaces at the growing tip may serve as intermolecular interfaces for polymerization. The multimeric nuclei appeared to polymerize into elongated protofibrils. The thin protofibrils appeared to wind around each other, forming fiber bundles. All protofibrils were incorporated into fiber bundles by 4 h of refolding. Additional exposed hydrophobic surfaces may be the site of association, or intermolecular contacts may be made by an unknown mechanism.
The aggregation pathway competed with a productive refolding pathway under the described conditions, indicating that a partially folded intermediate might be the polymerizing species. Neither equilibrium nor kinetic analyses of HD-Crys unfolding revealed any evidence for aggregation from the native state of the protein. The native-like species forming during aggregation-prone refolding had a fluorescence spectrum and AFM structure similar to native HD-Crys. Owing to the low concentration of protein present during refolding experiments, no other structural evidence was obtainable to confirm this.
Figure 10 shows a model for HD-Crys refolding and aggregation in vitro. The initial step on the productive refolding pathway is envisaged to be a global hydrophobic collapse (Kuwajima 1992; Ptitsyn 1995) during which large bulky, hydrophobic residues are sequestered away from the polar solvent as evidenced by an increase in the overall fluorescence of the protein. A kinetic partitioning of HD-Crys chains occurs between the productive refolding pathway and the aggregated complexes. The early collapsed intermediate may be the same for the two pathways. In both aggregation-prone refolding experiments and control productive refolding experiments, a burst phase was visible at 15 sec that may correspond to the formation of a similar early folding intermediate that is prone to polymerization.
|
D-Crys, as evidenced by the presence of a species that has a high solution turbidity within the dead time (15 sec) of the spectroscopic kinetic experiments.
The presence of 1.5 M GuHCl appeared to shift the equilibrium of the branch point between productive and aggregation-prone folding to favor the productive process, as evidenced by equilibrium analysis as well as kinetic experiments in which HD-Crys was refolded by dilution into 1.5 M GuHCl.
The HD-Crys aggregate bound bisANS significantly even as early as the formation of nuclei species, further emphasizing that these aberrant exposed hydrophobic residues exist from the onset of refolding. The hydrophobic bisANS binding surfaces may correspond to the intermolecular interaction domains that facilitate protofibril and fiber bundle formation. Further work is needed to elucidate which area of the molecule is responsible for these aggregation-facilitating interactions and which area, if any, retains a native-like fold.
The aggregation and refolding pathways were dependent on the concentration of HD-Crys, whereas the unfolding pathway was concentration-independent. When the overall concentration of protein was increased by a factor of 10, the presence of 1.5 M GuHCl was not sufficient to quench the aggregation pathway. This high-concentration aggregation reaction may or may not be the same as the aggregation phenomenon previously described, but these results do indicate a significant divergence of the in vivo and in vitro pathways. The amount of crystallin found in the eye lens is 70% (g/g wet weight), whereas the amount of crystallin used for these experiments was as low as 10 µg/mL (Jaenicke 1999).
Cataract formation in the lens
Many human diseases such as glaucoma and Meretoja’s syndrome result in the expression of insoluble protein fibers in the eye (Kivela et al. 1994; Nelson et al. 1999). Experiments performed with -crystallin have demonstrated that a fibrous aggregate forms when this protein is complexed with ßL crystallin (Weinreb et al. 2000), and amyloid-like protein species have been identified in situ in the mammalian ocular lens using Congo red and thioflavin T binding assays (Frederikse 2000). The results reported here indicate that old age cataracts should be reexamined for the presence of fibrillar aggregates containing human D crystallin.
|
Materials and methods |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
D-CrysD crystallin was prepared from E. coli as described (Pande et al. 2000). Briefly, the protein was purified by fractionating cell lysate on a size exclusion column followed by cation-exchange chromatography as described (Broide et al. 1991).
Circular dichroism
Far-UV CD measurements of native HD-Crys were collected on an Aviv Associates model 202 circular dichroism spectrometer. All readings were performed on 0.1 mg/mL protein samples.
Equilibrium refolding and unfolding
For the unfolding equilibrium titration, purified HD-Crys was diluted to 10 µg/mL in increasing amounts of GuHCl in S buffer from 0 to 5.5 M. S buffer contained 10 mM NaPO4, 5 mM DTT, and 1 mM EDTA, pH 7.0. The samples were incubated at 37°C until equilibrium was reached (6 h). For the refolding titration, 100 µg/mL protein was denatured in 5.5 M GuHCl in S buffer at 37°C for 5 h. The protein was subsequently refolded by dilution to 10 µg/mL into decreasing concentrations of GuHCl from 5.5 to 0.55 M. The fluorescence spectra of the equilibrated samples were determined using a Hitachi 4500 fluorimeter equipped with a continuous temperature-control system with excitation at 295 nm and emission from 310 to 420 nm. The emission intensities at 350 nm were used for data analysis. The excitation and emission slits were both set to 10 nm.
Unfolding fluorescence kinetics
Tryptophan environment changes with unfolding were monitored using a Hitachi 4500 fluorimeter equipped with a continuous temperature-control system. Native protein (100 µg/mL in S buffer at 37°C) was unfolded by dilution into S buffer to final concentrations of 10 µg/mL HD-Crys and 5.5 M GuHCl. Loss of global structure was monitored with continuous excitation at 295 nm and emission at 350 nm at 37°C.
Refolding fluorescence kinetics
Tryptophan environment changes with unfolding were monitored using a Hitachi 4500 fluorimeter equipped with a continuous temperature-control system. Native protein was denatured at 100 µg/mL concentration in 5.5 M GuHCl in S buffer at 37°C for 5 h. The unfolded protein was refolded by dilution into S buffer to final concentrations of 10 µg/mL HD-Crys and 0.55 M GuHCl or 1.5 M GuHCl. The increase in global structure while refolding was monitored with continuous excitation at 295 nm and emission at 350 nm. Fluorescence wavelength spectra were obtained after 5 h of refolding both before and after a 20-min spin at 12,000 rpm. A background fluorescence correction was made by obtaining spectra of S buffer containing 0.55 M GuHCl and subtracting it from the refolded sample data.
Refolding solution turbidity kinetics
Solution turbidity changes caused by formation of high-molecular-weight species were monitored with refolding in a Cary 50 Bio UV/Vis spectrophotometer. Samples were prepared as discussed for the refolding fluorescence kinetic experiments. Absorbance spectra from 260 to 350 nm were monitored at various time intervals. Absorbance values at 280 nm were used for data analysis.
Atomic force microscopy
Atomic force microscopy (AFM) analysis was performed using the tapping method as described (Marini et al. 2002). Ten µL of sample was allowed to nonspecifically bind to a mica surface for a total drying time of 75 sec. The mica was then washed with 150 µL of milli-Q water and allowed to air-dry before imaging.
BisANS binding assay
The character of the HD-Crys aggregate was probed by monitoring bisANS fluorescence changes upon binding. HD-Crys aggregates were prepared by refolding denatured protein in 0.55 M GuHCl in S buffer at 10 µg/mL HD-Crys at 37°C. Samples were removed at refolding time intervals ranging from 30 sec to 4 h. BisANS was added to the resulting sample to reach a final small molecule concentration of 20 nM. The fluorescence spectrum of the sample was determined by excitation at 350 nm and emission from 400 to 600 nm on a Hitachi 4500 fluorimeter. Background fluorescence of bisANS in 0.55 M GuHCl was subtracted from the resulting spectrum for data analysis purposes.
|
Acknowledgments |
|---|
D-Crys. We would also like to acknowledge Stephen Raso and Shannon Flaugh for helpful discussions and technical assistance regarding experimental procedures and data interpretation, and Davide Marini for his vital contribution to the AFM portion of this work. This research was supported by NIH grant GM17980 (awarded to J.K.). The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.
|
References |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
-helical conformation is under kinetic control. J. Biol. Chem. 276: 19687–19690.
Benedek, G. 1997. Cataract as a protein condensation disease: The Proctor Lecture. Invest. Ophthalmol. Vis. Sci. 38: 1911–1921.
Betts, S.D., Speed, M., and King, J. 1999. Detection of early aggregation intermediates by native gel electrophoresis and native Western blotting. Methods Ezymol. 309: 333–350.
Broide, M.L., Berland, C.R., Pande, J., Ogun, O.O., and Benedek, G.B. 1991. Binary-liquid phase separation of lens protein solutions. Proc. Natl. Acad. Sci. 88: 5660–5664.
Chen, B. and King, J. 1991. Thermal unfolding pathway for the thermostable P22 tailspike endorhamnosidase. Biochemistry 30: 6260–6269.[CrossRef][Medline]
Clark, J.I. 1994. Principle and practice of ophthalmology. Saunders College Publishing, Philadelphia, PA.
Clark, J.I. and Muchowski, P.J. 2000. Small heat-shock proteins and their potential role in human disease. Curr. Opin. Struct. Biol. 10: 52–59.[CrossRef][Medline]
Colon, W. and Kelly, J. 1992. Partial denaturation of transthyretin is sufficient for amyloid fibril formation in vitro. Biochemistry 31: 8654–8660.[CrossRef][Medline]
Davies, M.J. and Truscott, R.J.W. 2001. Photo-oxidation of proteins and its role in cataractogenesis. J. Photochem. Photobiol. 63: 114–125.
Delaye, M.T. and Tardieu, A. 1983. Short-range order of crystallin proteins accounts for eye lens transparency. Nature 302: 415–417.[CrossRef][Medline]
Eyles, S.J. and Gierasch, L.M. 2000. Multiple roles of prolyl residues in structure and folding. J. Mol. Biol. 301: 737–747.[CrossRef][Medline]
Frederikse, P.H. 2000. Amyloid-like protein structure in mammalian ocular lens. Curr. Eye Res. 20: 462–468.[CrossRef][Medline]
Ghelis, C. and Yon., J. 1982. Overview of the reversibility of the unfolding–folding process. In Protein folding (eds. B. Horecker et al.), pp. 285–295. Academic Press, New York.
Guex, N.P. and Peitsch, M.C. 1997. SWISS-MODEL and the Swiss-PdbViewer: An environment for comparative protein modeling. Electrophoresis 18: 2714–2723.[CrossRef][Medline]
Haase-Pettingell, C.A. and King J. 1988. Formation of aggregates from a thermolabile folding intermediate in P22 tailspike maturation. A model for inclusion body formation. J. Biol. Chem. 263: 4977–4983.
Harper, J.D., Leber, C.M., and Lansbury, P.T. 1997. Atomic force microscopic imaging of seeded fibril formation and fibril branching by the Alzheimer’s disease amyloid-ß protein. Chem. Biol. 4: 951–959.[CrossRef][Medline]
Hay, R.E., Andley, U.P., and Petrash, J.M. 1993. Expression of recombinant bovine B-, C-, and D-crystallins and correlation with native protien. Exp. Eye Res. 58: 573–584.
He, Q., Mason, A.B., Lyons, B.A., Tam, B.M., Nguyen, V., MacGillivray, R.T.A., and Woodworth, R. C. 2001. Spectral and metal-binding properties of three single-point tryptophan mutants of the human transferrin N-lobe. Biochem. J. 354: 423–429.[CrossRef][Medline]
Heon, E., Priston, M., Schorderet, D.F., Billingsley, G.D., Girard, P.O., Lubsen, N., and Munier, F.L. 1999. The -crystallins and human cataracts: The puzzle made clearer. Am. J. Hum. Genet. 65: 1261–1267.[CrossRef][Medline]
Hollien, J. and Marqusee, S. 2002. Comparison of the folding processes of T. thermophilus and E. coli ribonucleases H. J. Mol. Biol. 316: 327–340.[CrossRef][Medline]
Horwitz, J. 2000. The function of -crystallin in vision. Sem. Cell Dev. Biol. 11: 53–60.[CrossRef][Medline]
Jaenicke, R. 1999. Stability and folding of domain proteins. Prog. Biophys. Mol. Biol. 71: 155–241.[CrossRef][Medline]
Kivela, T., Tarkkanen, A., Frangione, B., Ghiso, J., and Haltia, M. 1994. Ocular amyloid deposition in familial amyloidosis, an analysis of native and variant gelsolin in Meretoja’s syndrome. Invest. Ophthalmol. Vis. Sci. 10: 3759–3769.
Kmoch, S., Brynda, J., Asfaw, B., Bezouska, K., Novak, P., Rezacova, P., Ondrova, L., Filipec, M., Sedlacek, J., and Elleder, M. 2000. Link between a novel human D-crystallin allele and a unique cataract phenotype explained by protein crystallography. Hum. Mol. Genet. 9: 1779–1786.
Kreisberg, J.F., Betts, S.D., Haase-Pettingell, C., and King, J. 2001. The interdigitated ß-helix domain of the P22 tailspike protein acts as a molecular clamp in trimer stabilization. Protein Sci. 11: 820–830.
Kuwajima, K. 1992. Protein folding in vitro. Curr. Opin. Biotechnol. 3: 462–467.[CrossRef][Medline]
Lai, Z., McCulloch, J., Lashuel, H.A., and Kelly, J.W. 1997. Guanidine hydrochloride-induced denaturation and refolding of transthyretin exhibits a marked hysteresis: Equilibria with high kinetic barriers. Biochemistry 36: 10230–10239.[CrossRef][Medline]
Lashuel, H., Lai, Z., and Kelly, J.W. 1998. Characterization of the transthyretin acid denaturation pathways by analytical ultracentrifugation: Implications for wild-type, V30M, and L55P amyloid fibril formation. Biochemistry 37: 17851–17864.[CrossRef][Medline]
Li, J., Uversky, V.N., and Fink, A.L. 2001. Effect of familial Parkinson’s disease point mutations A30P and A53T on the structural properties, aggregation, and fibrillation of human -synuclein. Biochemistry 40: 11604–11613.[CrossRef][Medline]
MacRae, T. 2000. Structure and function of small heat shock/-crystallin proteins: Established concepts and emerging ideas. Cell Mol. Life Sci. 57: 899–913.[CrossRef][Medline]
Marini, D.M., Hwang, W., Laffenburger, D.A., Zhang, S., and Kamm, R.D. 2002. Left-handed helical ribbon intermediates in the self-assembly of a ß-sheet peptide. Nano Lett. 2: 295–299.[CrossRef]
Mayr, E.M., Jaenicke, R., and Glockshuber, R. 1997. The domains in B crystallin: Identical fold—different stabilities. J. Mol. Biol. 269: 260–269.[CrossRef][Medline]
Mitraki, A. and King, J. 1989. Protein folding intermediates and inclusion body formation. Biotechnology 7: 690–697.[CrossRef]
Nelson, G.A., Edward, D.P., and Wilensky, J.T. 1999. Ocular amyloidosis and secondary glaucoma. Ophthalmology 106: 1363–1366.[CrossRef][Medline]
Nolting, B., Golbik, R., and Fersht, A.R. 1995. Submillisecond events in protein folding. Proc. Natl. Acad. Sci. 92: 10668–10672.
Norledge, B.V., Har, R.E., Bateman, O.A., Slingsby, C., and Driessen, H.P.C. 1997. Towards a molecular understanding of phase separation in the lens: A comparison of the x-ray structure of two high Tc -crystallins, E and F, with two low Tc -crystallins, B and D. Exp. Eye Res. 65: 609–630.[CrossRef][Medline]
Oyster, C.W. 1999. The human eye structure and function. Sinauer, Sunderland, MA.
Pace, C.N., Shirley, B.A., and Thomson, J.A. 1989. Measuring the conformational stability of a protein. In Protein structure: A practical approach (ed. T.E. Creighton), pp. 311–330. Oxford University Press, UK.
Pande, A., Pande, J., Asherie, N., Lomakin, A., Ogun, O., King, J.A., Lubsen, N.H., Walton, D., and Benedek, G.B. 2000. Molecular basis of a progressive juvenile-onset hereditary cataract. Proc. Natl. Acad. Sci. 97: 1993–1998.
Pande, A., Pande, J., Asherie, N., Lomakin, A., Ogun, O., King, J., and Benedek, G.B. 2001. Crystal cataracts: Human genetic cataract caused by protein crystallization. Proc. Natl. Acad. Sci. 98: 6116–6120.
Pande, J., McDermott, M.J., Callender, R.H., and Spector, A. 1991. The calf -crystallins—A Raman spectroscopic study. Exp. Eye Res. 52: 192–197.
Peitsch, M.C. 1995. Protein modeling by E-mail. Biotechnology 13: 658–660.[CrossRef]
———. 1996. ProMod and Swiss-Model: Internet-based tools for automated comparative protein modeling. Biochem. Soc. Trans. 24: 274–279.[Medline]
Ptitsyn, O.B. 1995. Structures of folding intermediates. Curr. Opin. Struct. Biol. 5: 74–78.[CrossRef][Medline]
Reader, J.S., Van Nuland, N.A., Thompson, G.S., Ferguson, S.J., Dobson, C.M., and Radford, S.E. 2001. A partially folded intermediate species of the ß-sheet protein apo-pseudoazurin is trapped during proline limited folding. Protein Sci. 10: 1216–1224.
Rudolph, R., Sienbendritt, R., Nesslauer, G., Sharma, A.K., and Jaenicke, R. 1990. Folding of an all-ß protein: Independent domain folding in II-crystallin from calf eye lens. Proc. Natl. Acad. Sci. 87: 4625–4629.
Sinclair, J.F., Ziegler, M.M., and Baldwin, T.O. 1994. Kinetic partitioning during refolding yields multiple native states. Nat. Struct. Biol. 1: 320–326.[CrossRef][Medline]
Slingsby, C., Norledge, B., Simpson, A., Bateman, O.A., Wright, G., Driessen, H.P.C., Lindley, P.F., Moss, D.S., and Bax, B. 1997. X-Ray diffraction and structure of crystallins. Prog. Retin. Eye Res. 16: 3–29.
Speed, M.A., Wang, D.I.C., and King, J. 1995. Multimeric intermediates in the pathway to the aggregated inclusion body state for P22 tailspike polypeptide chains. Protein Sci. 4: 900–908.[Abstract]
Sturtevant, J.M., Yu., M.-H., Haase-Pettingell, C., and King, J. 1989. Thermostability of temperature sensitive folding mutants of the P22 tailspike protein. J. Biol. Chem. 264: 10693–10698.
Subramaniam, V., Bergenhem, N.C.H., Gafni, A., and Steel, D. 1995. Phosphorescence reveals a continued slow annealing of the protein core following reactivation of Escherichia coli alkaline phosphatase. Biochemistry 34: 1133–1136.[CrossRef][Medline]
Tardieu, A., Veretout, F., Krop, B., and Slingsby, C. 1992. Protein interactions in the calf eye lens: Interactions between ß-crystallins are repulsive whereas in -crystallins they are attractive. Eur. Biophys. J. 21: 1–21.[Medline]
Weinreb, O., Van Rijk, A.F., Dovrat, A., and Bloemendal, H. 2000. In vitro filament-like formation upon interaction between lens -crystallin and ßL-crystallin promoted by stress. Invest. Opthalmol. Vis. Sci. 41: 3893–3897.
Wenk, M., Herbst, R., Hoeger, D., Kretschmar, M., Lubsen, N. H., and Jaenicke, R. 2000. S-crystallin of bovine and human eye lens: Solution structure, stability and folding of the intact two-domain protein and its separate domains. Biophys. Chem. 86: 95–108.[CrossRef][Medline]
Wetzel, R. 1994. Mutations and off-pathway aggregation of proteins. Trends Biotechnol. 12: 193–198.[CrossRef][Medline]
Wetzel, R. and Chrynuk, B.A. 1994. Inclusion body formation by interleukin-1 ß depends on the thermal sensitivity of a folding intermediate. FEBS Lett. 350: 245–248.[CrossRef][Medline]
Wistow, G., Turnell, B., Summers, L., Slingsby, C., Moss, D., Miller, L., Lindley, P., and Blundell, T. 1983. X-ray analysis of the eye lens protein -II crystallin at 1.9 Å resolution. J. Mol. Biol. 170: 175–202.[CrossRef][Medline]
), denatured protein in 5.5 M GuHCl (
), H
), and after a 20-min centrifugation (X) at 37°C. (B) Raw fluorescence intensity of unfolding (•, black) and refolding (
), 17 h (
).
