Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M603882200 on August 4, 2006

J. Biol. Chem., Vol. 281, Issue 41, 30782-30793, October 13, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/41/30782    most recent
M603882200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Flaugh, S. L.
Right arrow Articles by King, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Flaugh, S. L.
Right arrow Articles by King, J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Glutamine Deamidation Destabilizes Human {gamma}D-Crystallin and Lowers the Kinetic Barrier to Unfolding*

Shannon L. Flaugh1, Ishara A. Mills2, and Jonathan King3

From the Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139

Received for publication, April 24, 2006 , and in revised form, July 28, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Human eye lens transparency requires life long stability and solubility of the crystallin proteins. Aged crystallins have high levels of covalent damage, including glutamine deamidation. Human {gamma}D-crystallin (H{gamma}D-Crys) is a two-domain beta-sheet protein of the lens nucleus. The two domains interact through interdomain side chain contacts, including Gln-54 and Gln-143, which are critical for stability and folding of the N-terminal domain of H{gamma}D-Crys. To test the effects of interface deamidation on stability and folding, single and double glutamine to glutamate substitutions were constructed. Equilibrium unfolding/refolding experiments of the proteins were performed in guanidine hydrochloride at pH 7.0, 37 °C, or urea at pH 3.0, 20 °C. Compared with wild type, the deamidation mutants were destabilized at pH 7.0. The proteins populated a partially unfolded intermediate that likely had a structured C-terminal domain and unstructured N-terminal domain. However, at pH 3.0, equilibrium unfolding transitions of wild type and the deamidation mutants were indistinguishable. In contrast, the double alanine mutant Q54A/Q143A was destabilized at both pH 7.0 and 3.0. Thermal stabilities of the deamidation mutants were also reduced at pH 7.0. Similarly, the deamidation mutants lowered the kinetic barrier to unfolding of the N-terminal domain. These data indicate that interface deamidation decreases the thermodynamic stability of H{gamma}D-Crys and lowers the kinetic barrier to unfolding due to introduction of a negative charge into the domain interface. Such effects may be significant for cataract formation by inducing protein aggregation or insolubility.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Transparency of the human eye lens depends on high concentrations and short range order of the crystallin proteins (1, 2). Enucleated mature lens fiber cells do not actively synthesize protein, and consequently, the crystallin proteins of nuclear lens cells are as old as the lens itself. This unique phenomenon necessitates life long protein stability and solubility despite elevated concentrations and exposure to environmental stresses.

A striking feature of cataract is the sharp rise in prevalence with increasing age (3). The crystallin proteins accumulate high levels of covalent damage as they age, suggesting that covalent damage of the crystallins may cause or contribute to disease onset. Cataract is the leading cause of blindness worldwide and affects one in six people over the age of 40 in the United States (3). Cataract is associated with the presence of insoluble light-scattering crystallin inclusions. Formation of these inclusions is likely caused by loss of crystallin solubility and/or aggregation.

The {alpha}-, beta-, and {gamma}-crystallins are found in all vertebrate lenses. The {alpha}-crystallins associate to form large polydisperse multimers that possess in vitro molecular chaperone activity (4, 5). In contrast, it is believed that in the lens, the beta- and {gamma}-crystallins function solely as structural proteins. The beta- and {gamma}-crystallins adopt similar two domain, beta-sheet, Greek key motif folds. The wild-type {gamma}-crystallins are remarkably stable in vitro and generally have free energies of unfolding ({Delta}G0) that are higher than the wild-type {alpha}- and beta-crystallins (14). The two domains of the monomeric {gamma}-crystallins interact intramolecularly via side chain contacts across a domain interface (6-8). In contrast, the beta-crystallins form a range of multimeric states. For instance, the domains of betaB2-Crys4 interact intermolecularly to form a domain-swapped dimer (6). Whether they occur intra- or intermolecularly, domain interface interactions are important for folding, stability, and oligomerization of the beta- and {gamma}-crystallins (9-13).

Glutamine and asparagine deamidation has been observed in all of the major crystallin proteins recovered from cataractous lenses (15-17). At the atomic level, deamidation can cause backbone isomerization and introduces a negative charge at physiological pH. It has been suggested that deamidation of Asn-143 in human {gamma}S-crystallin is specifically associated with mature-onset cataract formation (18). Similarly, deamidated beta-crystallins isolated from human lenses have an increased tendency to associate into noncovalent aggregates (19). Deamidation may cause changes in protein structure or solubility that could instigate cataract formation.

Critical investigations by Lampi and co-workers (20-23) observed that the effects of deamidation on the in vitro properties of human beta-crystallins varied depending on the site of damage. Deamidation of a glutamine in the connecting peptide of betaB1-Crys caused the protein to form larger multimers and increased the tendency for thermal aggregation (20). In contrast, deamidation of glutamines in the domain interfaces of dimeric betaB1-Crys and betaB2-Crys decreased stabilities of the proteins and caused them to populate partially unfolded intermediates in equilibrium experiments (21, 22, 24). Studies by Gupta and Srivastava (25, 26) showed that {alpha}-crystallin also displays context-dependent effects of deamidation, including decreased in vitro chaperone activity, changes in secondary and tertiary structures, and altered oligomerization properties (25, 26). Specific effects of covalent damage on the properties of the {gamma}-crystallins have not been addressed previously.


Figure 1
View larger version (38K):
[in this window]
[in a new window]
  		FIGURE 1.
Structure of wild-type H{gamma}D-Crys (31) depicted in ribbon representation showing the location of interface residues Gln-54 and Gln-143 and four nearby crystallographic waters that form hydrogen bonds to the glutamine side chains (Protein Data Bank code 1HK0).

 
Human {gamma}D-crystallin (H{gamma}D-Crys) is one of the most abundant {gamma}-crystallins of the lens nucleus, and covalently damaged forms accumulate in aged lenses (27-30). H{gamma}D-Crys adopts the typical two domain fold of the {gamma}-crystallins where the two domains interact intramolecularly (31). In vitro stability and folding of H{gamma}D-Crys have been studied previously by using fluorescence as a probe of conformation (12, 13, 32-34). In equilibrium unfolding/refolding experiments at pH 7.0 and 37 °C, wild-type H{gamma}D-Crys populated a partially folded intermediate that likely had a folded C-terminal domain (C-td) and unfolded N-terminal domain (N-td) (12). Off-pathway aggregation competed with productive refolding when H{gamma}D-Crys was diluted out of high concentrations of guanidine hydrochloride (GdnHCl) into buffer (34).

As shown in Fig. 1, the domain interface of H{gamma}D-Crys includes a pair of buried glutamines that form hydrogen bonds to four water molecules present in the crystal structure (31). A cluster of six hydrophobic residues that are buried in the domain interface are also in close proximity to these apical glutamines. It was shown previously that although neither Gln-54 nor Gln-143 are critical determinants of structure, mutating these residues to alanine destabilized the N-td and slowed its rate of refolding in vitro (13). Glutamine is conserved in these positions among beta- and {gamma}-crystallins from diverse species at ~78% identity for Gln-54 and ~80% identity for Gln-143.

Given the importance of Gln-54 and Gln-143 in stability and folding, and that deamidation of analogously positioned glutamines in the beta-crystallins causes destabilization, we suspected that deamidation of these residues would destabilize H{gamma}D-Crys due to introduction of a negative charge into the domain interface. To test this, single and double glutamine to glutamate mutants were constructed, and in vitro properties of the mutant proteins were studied. The deamidation mutations destabilized the proteins at pH 7.0 but not pH 3.0. Additionally, the mutations reduced the rate of refolding and increased the rate of unfolding of the N-td resulting in a decreased kinetic barrier to unfolding.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Mutagenesis, Expression, and Purification of Recombinant H{gamma}D-Crys—Single and double glutamine to glutamate substitutions of residues Gln-54 and Gln-143 were constructed using site-directed mutagenesis. Mutant primers (IDT-DNA) were used to amplify the gene for H{gamma}D-Crys with an N-terminal His6 tag in a pQE.1 plasmid (33). Substitutions were confirmed by DNA sequencing of all amplified plasmids (Massachusetts General Hospital).

Wild-type and mutant H{gamma}D-Crys proteins were expressed and purified as described by Kosinski-Collins et al. (33). The proteins were expressed in Escherichia coli, purified by affinity chromatography using a nickel-nitrilotriacetic acid resin (Qiagen), and finally dialyzed into 10 mM ammonium acetate (pH 7.0). Concentrations of purified proteins were calculated from absorbance data at 280 nm using an extinction coefficient of 41,040 cm-1 M-1 for all proteins.

Calculating Solvent-accessible Surface Areas—Solvent-accessible surface areas of Gln-54 and Gln-143 were calculated from the crystal structure of H{gamma}D-Crys, Protein Data Bank code 1HK0 [PDB] (31), using the program GETAREA 1.1 (35).

Circular Dichroism Spectroscopy—CD spectra of the wild-type and mutant proteins were recorded with an AVIV model 202 CD spectrometer (Lakewood, NJ). Proteins were present at 100 µg/ml in 10 mM sodium phosphate (pH 7.0). The temperature was maintained at 37 °C using an internal Peltier thermoelectric temperature controller. Far-UV CD spectra were collected from 195 to 260 nm in a 1-mm cuvette. Buffer signal was subtracted from all spectra, after which mean residual ellipticity was calculated.

Fluorescence Emission Spectroscopy—Fluorescence spectra of the wild-type and mutant H{gamma}D-Crys proteins were measured with a Hitachi F-4500 fluorimeter. Samples at pH 7.0 and 37 °C contained 10 µg/ml purified protein in 100 mM sodium phosphate, 5 mM DTT, 1 mM EDTA (pH 7.0), and 5.5 M GdnHCl where appropriate. Samples at pH 3.0 and 20 °C contained 10 µg/ml protein in 100 mM sodium citrate (pH 3.0). Temperature was maintained at 37 or 20 °C using a circulating water bath. An excitation wavelength of 295 nm was used to selectively monitor tryptophan fluorescence. Emission spectra were recorded over a range of wavelengths from 310 to 400 nm using slit widths of 10 nm for both excitation and emission. Fluorescence emission spectra were corrected for the buffer signal.

Equilibrium Unfolding and Refolding—Equilibrium unfolding/refolding experiments at pH 7.0 were performed in GdnHCl at 37 °C. For equilibrium unfolding experiments, purified protein was diluted into solutions containing 0-5.5 M GdnHCl (purchased as an 8.0 M solution from Sigma) to give a final protein concentration of 10 µg/ml. All pH 7.0 unfolding samples contained 100 mM sodium phosphate, 5 mM DTT, and 1mM EDTA (pH 7.0). Unfolding samples were incubated at 37 °C for 24 h to ensure equilibrium had been reached. For equilibrium refolding experiments, purified proteins were first unfolded in 5.5 M GdnHCl at a protein concentration of 100 µg/ml. These unfolded stocks were incubated at 37 °C for 5 h to ensure complete unfolding. The unfolded stock solutions were then diluted into refolding samples, which contained 100 mM sodium phosphate, 5 mM DTT, 1 mM EDTA (pH 7.0), and GdnHCl from 0.55 to 5.5 M. Refolding samples were allowed to reach equilibrium by incubation at 37 °C for 24 h prior to recording fluorescence spectra.

Equilibrium unfolding/refolding experiments at pH 3.0 were performed in urea at 20 °C. Samples were set up in a manner identical to that described above for pH 7.0 experiments where the final protein concentration was 10 µg/ml. Unfolding and refolding samples were buffered with 100 mM sodium citrate (pH 3.0) and did not contain DTT or EDTA due to concerns of low solubility at pH 3.0. A 24-h incubation time was used for both unfolding and refolding samples.

Fluorescence emission spectra were recorded for each unfolding and refolding sample using a Hitachi F-4500 fluorimeter as described above. The concentration of GdnHCl or urea in the unfolding/refolding samples was determined by measuring the refractive index of each sample. Data were analyzed by plotting the concentration of GdnHCl for each sample versus the fluorescence intensity at 360 nm and the concentration of GdnHCl versus the ratio of fluorescence intensities at 360 and 320 nm (FI 360/320 nm). Equilibrium unfolding/refolding experiments of the wild-type and mutant proteins were performed three times each.

Equilibrium unfolding and refolding data were fit to a two-state model by the methods of Greene and Pace (36) or a three-state model by the methods of Clark et al. (37) using the curvefitting feature of Kaleidagraph (Synergy Software). The model that best fit the data was selected based on a random distribution of residuals. Transition midpoints, {Delta}G0, and m values were calculated for all transitions and averaged over the three trials.

Thermal Denaturation—Thermal denaturation experiments were performed using an AVIV model 202 CD spectrometer equipped with an internal Peltier thermo-electric temperature controller (Lakewood, NJ). All samples contained 100 µg/ml protein in 10 mM sodium phosphate (pH 7.0). The solution conditions differed from equilibrium experiments because of the optical interference of EDTA and DTT and higher concentrations of sodium phosphate. A 4-mm cuvette with an air-tight screw cap was used for the measurements to prevent loss of sample at high temperatures due to evaporation or boiling over. Changes in molar ellipticity at 218 nm were monitored every 1 °C from 25 to 90 °C. The samples were allowed to equilibrate at each temperature for 1 min before measuring ellipticity over a 3-s averaging time. The fraction of native (FN) protein at each temperature was calculated according to Equation 1,

Formula(Eq.1)

where y is the ellipticity at 218 nm; yU is the unfolded/aggregated base line, and yN is the native base line. Melting temperatures were calculated by determining the midpoints of the thermal transitions. The experiments were repeated three times for each protein, and melting temperatures of the three trials were averaged.

Productive Refolding Kinetics—Kinetic refolding experiments were performed by initially unfolding in 5.5 M GdnHCl at a protein concentration of 100 µg/ml. These unfolded stocks were incubated at 37 °C for 5 h to ensure complete unfolding. The unfolded stocks were injected into constantly stirred refolding buffer containing 10 mM sodium phosphate, 5 mM DTT, and 1 mM EDTA (pH 7.0) at 37 °C. A syringe-port injection system with a dead time of 1 s was used. Changes in fluorescence over time were monitored with a Hitachi F-4500 fluorimeter using an excitation wavelength of 295 nm and emission wavelength of 350 nm. The final protein concentration of refolding samples was 10 µg/ml in 1.0 M GdnHCl. Fluorescence emission spectra were measured at the end of each experiment to ensure that the proteins had completely refolded into native-like conformations. Kinetic refolding data were fit to two and three exponentials using the curve-fitting feature of Kaleidagraph and residuals of the fits were calculated (Synergy Software). The model with the most random distribution of low magnitude residuals was selected as the best fit. Kinetic refolding experiments of the wild-type and mutant proteins were performed three times each, and parameters of the fits were averaged.

Unfolding Kinetics—Kinetic unfolding experiments were performed by diluting purified proteins into 5.0 M GdnHCl and monitoring fluorescence emission at 350 nm over time using an excitation wavelength of 295 nm. The kinetic unfolding samples contained proteins at final concentrations of 10 µg/ml in 5.0 M GdnHCl, 10 mM sodium phosphate, 5 mM DTT, 1 mM EDTA (pH 7.0) at 37 °C. Fluorescence emission spectra were recorded at the end of each experiment to ensure that the proteins had fully unfolded. Kinetic unfolding data were fit to two and three exponentials, and residuals of the fits were calculated using the curve-fitting feature of Kaleidagraph (Synergy Software). The best fit was chosen by a random arrangement of residuals. Experiments were performed three times for each protein, and parameters were averaged.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Protein Purification and Structure Characterization—To probe the site-specific effects of domain interface glutamine deamidation in H{gamma}D-Crys, single and double glutamine to glutamate substitution mutants were constructed as follows: Q54E, Q143E, and Q54E/Q143E. The mutant proteins were purified out of E. coli, and the stabilities and folding properties of the proteins were determined in vitro.

The wild-type and mutant proteins all had exogenous N-terminal His6 tags that were utilized during affinity purification. Previous studies of wild-type H{gamma}D-Crys with and without the N-terminal His6 tag confirmed that the additional sequence did not discernibly affect the structure or thermodynamic and kinetic unfolding/refolding properties of the protein (33, 34). Therefore, the His6 tag was not removed for these investigations.


Figure 2
View larger version (14K):
[in this window]
[in a new window]
  		FIGURE 2.
A, far-UV CD spectra of native wild-type H{gamma}D-Crys (solid line), Q54E (short dotted gray line), Q143E (long dashed gray line), and Q54E/Q143E (short dashed gray line). All samples contained 100 µg/ml protein in 10 mM sodium phosphate (pH 7.0) at 37 °C. B, fluorescence spectra of native wild type and deamidation mutants at pH 7.0 (lines are the same as in A) and wild-type H{gamma}D-Crys denatured in 5.5 M GdnHCl (short dotted black line). Samples contained 10 µg/ml protein in 100 mM sodium phosphate, 5 mM DTT, 1 mM EDTA (pH 7.0) at 37 °C.

 
The single and double deamidation mutants of residues Gln-54 and Gln-143 expressed at levels similar to wild type and behaved like wild type during affinity purification. The mutant proteins were found primarily in the soluble fractions of cell lysates indicating that the mutations did not prevent in vivo folding into soluble native-like conformations.

Structures of wild-type H{gamma}D-Crys, Q54E, Q143E, and Q54E/Q143E were probed with CD and fluorescence spectroscopy. These experiments addressed gross changes in secondary and tertiary structure and were not aimed at distinguishing atomic resolution differences in conformation. Similar to previous results, the far-UV CD spectrum of wild-type H{gamma}D-Crys at pH 7.0 displayed a distinct minimum at 218 nm, indicative of high beta-sheet content, and a small shoulder at 208 nm (38, 39). The single and double deamidation mutants all displayed similar minima at 218 nm and a shoulder at 208 nm (Fig. 2). CD spectra of the single mutants Q54E and Q143E were indistinguishable from that of wild type in the region analyzed, whereas the spectra of Q54E/Q143E differed about 20% in intensity at lower wavelengths (Fig. 2). This difference may be due to actual changes in conformation or slight differences in solution conditions.

Domain tertiary structures of wild-type H{gamma}D-Crys and of the deamidation mutants were surveyed using tryptophan fluorescence emission. H{gamma}D-Crys has four intrinsic tryptophans, two per domain, that are buried in the hydrophobic cores of the domains. H{gamma}D-Crys also has 14 tyrosines located throughout the protein. Tryptophan fluorescence was selectively monitored by using an excitation wavelength of 295 nm and monitoring emission from 310 to 400 nm. In accord with previous results, native wild-type H{gamma}D-Crys displayed a fluorescence emission maximum at 325 nm (Fig. 2), and upon denaturing in 5.5 M GdnHCl, the fluorescence emission maximum shifted to 350 nm and increased in intensity (34). The native fluorescence emission spectrum of Q54E was indistinguishable from wild type (Fig. 2). In contrast, although the spectra of Q143E and Q54E/Q143E had identical emission maxima as wild type, the overall intensities were increased ~10% (Fig. 2).

Solvent-accessible surface areas of Gln-54 and Gln-143 in the crystal structure of H{gamma}D-Crys (31) were calculated with the program GETAREA 1.1 (35). The solvent-accessible surface area of the Gln-54 side chain was calculated to be 20.22 Å2, and the solvent-accessible surface area of Gln-143 was calculated to be 3.2 Å2. Relative to the total surface area of the glutamine side chain (143.7 Å2), 86% of the side chain surface area of Gln-54 is buried, and 97% of the side chain surface area of Gln-143 is buried. Despite being highly buried, the side chain amide nitrogen of Gln-54 and the side chain amide oxygen of Gln-143 form hydrogen bonds to two water molecules each as seen in the crystal structure and shown in Fig. 1 (31).

Equilibrium Unfolding/Refolding of Wild Type at pH 7.0—The thermodynamic stability of wild-type H{gamma}D-Crys was analyzed by equilibrium unfolding/refolding using the chemical denaturant GdnHCl, at pH 7.0 and 37 °C. GdnHCl was used in these experiments instead of urea because wild-type H{gamma}D-Crys has been shown previously to resist denaturation up to 8 M urea at pH 7.0 (34). The conformation of proteins in equilibrium experiments was probed with tryptophan fluorescence. To this end, the fluorescence intensity changes at 360 nm and changes in the ratio of intensities at 360 and 320 nm (FI 360/320 nm) were used for data analysis. The FI 360/320 nm ratio data are shown in Fig. 3 instead of changes in the intensity at 360 nm because the transition plateau was visually more perceptible. Nevertheless, parameters derived from the fits of both sets of data were indistinguishable.

Consistent with previous results, the unfolding/refolding transitions of wild-type H{gamma}D-Crys were best fit to a three-state model suggesting the presence of a partially folded intermediate in equilibrium with the native and unfolded states (12). This intermediate was apparent as an inflection in the transitions at ~2.3 M GdnHCl (Fig. 3). The first transition had a midpoint (Cm) of 2.2 M GdnHCl and an apparent {Delta}G0 of 7.7 kcal·mol-1 (Table 1). The second transition had a midpoint of 2.8 M GdnHCl and apparent {Delta}G0 of 8.9 kcal·mol-1. The first transition likely corresponded to unfolding/refolding of the N-td and the second transition to unfolding/refolding of the C-td (12). At pH 7.0 and 37 °C, the {Delta}G0 of wild-type H{gamma}D-Crys was ~16.6 kcal·mol-1.


View this table:
[in this window]
[in a new window]
  		TABLE 1
Equilibrium unfolding/refolding and thermal unfolding parameters for wild type and deamidation mutants at pH 7.0

 
Also consistent with previous results, wild-type H{gamma}D-Crys aggregated upon refolding into buffer (34). The aggregate was evident by the sharp increase in fluorescence at low concentrations of GdnHCl because of right angle light scattering by the aggregate (Fig. 3).

Equilibrium Unfolding/Refolding of Deamidation Mutants at pH 7.0—Thermodynamic stabilities of the single and double deamidation mutants were analyzed by the same method described above for wild-type H{gamma}D-Crys. Similar to wild type, data for all mutants were analyzed by changes in fluorescence intensity at 360 nm as well as changes in FI 360/320 nm. Parameters derived from fits of both analyses agreed within the error of the experiments. FI 360/320 nm data for the single and double deamidation mutants is shown in Fig. 3.


Figure 3
View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 3.
Equilibrium unfolding (solid symbols) and refolding (open symbols) of wild-type H{gamma}D-Crys ({diamondsuit}), Q54E ({blacktriangleup}), Q143E (bullet), and Q54E/Q143E ({blacksquare}) at pH 7. 0 and 37 °C in GdnHCl. Conformation was probed by fluorescence spectroscopy using an excitation wavelength of 295 nm. Changes in the ratio of fluorescence intensities at 360/320 nm are shown. All samples contained 10 µg/ml in 100 mM sodium phosphate, 5 mM DTT, 1mM EDTA (pH 7.0), and GdnHCl from 0 to 5.5 M at 37 °C. Solid lines represent three-state fits of the unfolding data.

 
As with wild type, the transitions of all mutants were best fit to a three-state model indicating the population of a partially folded intermediate. The presence of an intermediate is more apparent for the deamidation mutants than the wild-type protein as a prominent plateau is present in all transitions. {Delta}G0 values of the first transition were 5.8 kcal·mol-1·M-1 for Q54E, 6.1 kcal·mol-1·M-1 for Q143E, and 5.3 kcal·mol-1·M-1 for Q54E/Q143E (Table 1). These values are substantially less than the {Delta}G0 of the first transition for wild type (7.7 kcal·mol-1·M-1).

In contrast, the {Delta}G0 values of the second transition were very similar for wild type and all three deamidation mutants (Table 1). Similarly, Cm values for the first transition were decreased for all mutants compared with wild type, whereas Cm values for the second transition were identical (Table 1). Also similar to wild type, all deamidation mutants aggregated upon refolding into buffer (Fig. 3).

Thermal Denaturation at pH 7.0—Given that the ionic character of GdnHCl may mask or interfere with the effects of the deamidation charge change, stabilities of wild-type H{gamma}D-Crys and the deamidation mutants were further probed in a low ionic strength buffer by thermal denaturation. These experiments assessed stabilities by monitoring changes in ellipticity at 218 nm every 1 °C from 25 to 90 °C. Wild-type H{gamma}D-Crys and all deamidation mutants aggregated at high temperatures as seen by the presence of protein precipitate subsequent to heating to 90 °C. Therefore, loss of CD signal at high temperatures was probably caused by a combination of unfolding, and aggregation initiated by unfolding. Regardless of the origin of the signal change, an increase in ellipticity at 218 nm was interpreted as a loss of native structure and therefore a satisfactory measure of stability.

Under the conditions employed here the thermal denaturation transition of wild-type H{gamma}D-Crys appeared two state with no significant evidence of a stable intermediate (Fig. 4). The melting temperature (Tm) of wild type was 83.8 °C (Table 1). Similarly, the thermal denaturation transitions of all deamidation mutants also appeared two state (Fig. 4). However, all of the mutants denatured at lower temperatures than wild type with Tm values of 77.4 °C for Q54E, 75.7 °C for Q143E, and 71.0 °C for Q54E/Q143E (Table 1). Thermal unfolding of the deamidation mutants occurred over a narrower temperature range than wild type suggesting that the mutants may have unfolded more cooperatively.

Equilibrium Unfolding/Refolding Wild Type at pH 3.0—Equilibrium experiments of wild type, Q54E, Q143E, and Q54E/Q143E were performed at pH 3.0 to test the role of charge in destabilization at pH 7.0. The thermodynamic stability of wild-type H{gamma}D-Crys at pH 3.0 was determined by equilibrium unfolding/refolding experiments in urea at 20 °C. These experiments were performed at 20 °C rather than 37 °C because H{gamma}D-Crys has been shown to form amyloid fibers at pH 3.0 and 37 °C.5 Urea was used instead of GdnHCl because at pH 3.0 and 20 °C, low concentrations of GdnHCl also caused polymerization into an amyloid state.6 Finally, pH 3.0 was chosen because wild-type H{gamma}D-Crys was partially unfolded at lower pH even in the absence of denaturant (data not shown).


Figure 4
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 4.
Thermal denaturation transitions for wild-type H{gamma}D-Crys ({diamondsuit}), Q54E ({blacktriangleup}), Q143E (bullet), and Q54E/Q143E ({blacksquare}) as monitored by changes in ellipticity at 218 nm at increasing temperature. All samples contained 100 µg/ml protein in 10 mM sodium phosphate (pH 7.0).

 
There was no evidence of aggregation or amyloid fiber formation in unfolding samples under the conditions employed here (100 mM sodium citrate (pH 3.0) plus urea from 0 to 6 M at 20 °C). However, all proteins aggregated upon refolding into concentrations of urea below 2.5 M at pH 3.0 (data not shown). Fluorescence spectra of soluble protein present in the aggregation samples subsequent to a high speed spin did not match spectra of the native proteins at pH 3.0 (data not shown). The addition of 5 mM DTT and 1 mM EDTA did not result in recovery of native protein in refolding samples (data not shown). The morphology of the pH 3.0 refolding aggregates and the irreversibility of the reactions will be the subject of future investigations. Given the irreversibility of the reaction at pH 3.0, {Delta}G0 values were not calculated, and instead transition midpoints were used as an estimate of stability (Table 2). To reduce complexity of the plots, equilibrium refolding data are not shown in Fig. 5.


View this table:
[in this window]
[in a new window]
  		TABLE 2
Equilibrium unfolding/refolding parameters for wild-type and mutant H{gamma}D-Crys at pH 3.0

 


Figure 5
View larger version (23K):
[in this window]
[in a new window]
  		FIGURE 5.
Equilibrium unfolding of wild-type H{gamma}D-Crys ({diamondsuit}), Q54E ({blacktriangleup}), Q143E (bullet), Q54E/Q143E ({blacksquare}), and Q54A/Q143A ({blacktriangledown}) in urea at pH 3. 0 and 20 °C. Conformation was probed by fluorescence emission. Data are presented as the ratio of fluorescence intensities at 360/320 nm using an excitation wavelength of 295 nm. Protein was present at 10 µg/ml in 100 mM sodium citrate (pH 3.0) and urea from 0 to 5.0 M at 20 °C.

 
The equilibrium unfolding transition of wild-type H{gamma}D-Crys at pH 3.0 was best fit to a two-state model that assumes direct transition between the native and unfolded states (Fig. 5). The Cm of the pH 3.0 equilibrium unfolding transition was 2.4 M urea (Table 2). Discrepancy between the number of states observed for the pH 7.0 and pH 3.0 data is likely because of different effects of pH on the stabilities of the N- and C-tds. For instance, if the stabilities of each domain of wild-type H{gamma}D-Crys were more similar at pH 3.0 than pH 7.0, the partially folded intermediate would not be populated over as wide a range of denaturant concentrations if at all, and thus would not be observable at pH 3.0.

Equilibrium Unfolding/Refolding of Deamidation Mutants at pH 3.0—Stabilities of the mutants Q54E, Q143E, and Q54E/Q143E were also measured at pH 3.0, which is below the pKa of glutamate in aqueous solution. However, the introduced glutamate side chains are in proximity to the polar main chain and side chain atoms as well as water molecules that may alter pKa values and cause the glutamate side chains not to be fully protonated at pH 3.0 (40, 41). The pH 3.0 equilibrium unfolding transitions of the deamidation mutants were best fit to two-state models similar to wild type (Fig. 5). The transition midpoints of the deamidation mutants were identical to or comparable with that of wild-type H{gamma}D-Crys. The Cm value for Q54E was 2.4 M, the Cm value for Q143E was 2.3 M, and the Cm value for Q54E/Q143E was 2.1 M (Table 2).

Equilibrium unfolding of the double alanine mutant, Q54A/Q143A, was also measured at pH 3.0 as a control protein that is destabilized at pH 7.0 but has the same number of acidic and basic residues as wild type (13). At pH 3.0, the unfolding transition of Q54A/Q143A deviated significantly from that of wild type as it was best fit to a three-state model with Cm values of 1.3 and 2.5 M for the first and second transitions, respectively (Fig. 5 and Table 2).

Kinetic Refolding of Wild Type and Deamidation Mutants at pH 7.0—The effects of glutamine deamidation on the refolding kinetics of H{gamma}D-Crys were determined by performing productive kinetic refolding experiments at pH 7.0. Proteins were diluted from 5.0 to 1.0 M GdnHCl, and burial of the tryptophans was monitored as a decrease in the fluorescence intensity at 350 nm over time. Proteins were refolded to 1.0 M GdnHCl as this was the lowest concentration of GdnHCl where aggregation did not compete with productive refolding (34). A syringe-port injection system was used instead of a stopped-flow apparatus because the major refolding transitions of H{gamma}D-Crys occur on a second and not millisecond time scale (33). Potential sub-second refolding intermediates were not addressed in these experiments.


Figure 6
View larger version (23K):
[in this window]
[in a new window]
  		FIGURE 6.
Productive kinetic refolding data of wild-type H{gamma}D-Crys fit to two (A) and three exponentials (B). Protein was initially unfolded in 5.5 M GdnHCl and diluted into 10 mM sodium phosphate, 5 mM DTT, 1 mM EDTA (pH 7.0) at 37 °C to give a final GdnHCl concentration of 1.0 M. Changes in fluorescence intensity at 350 nm were monitored over time using an excitation wavelength of 295 nm. Fits of the data are shown as black lines and residuals of the fits are shown above.

 
Previous analyses of wild-type H{gamma}D-Crys refolding kinetics revealed the presence of a single major intermediate that likely has a folded C-td and unfolded N-td (33). It was further found that the domain interface of the folded C-td nucleates refolding of the N-td (12, 13). Refolding kinetics of the deamidation mutants did not fit well to two exponentials (three states) as described previously for wild type. This observation prompted a more in-depth analysis of wild-type H{gamma}D-Crys refolding kinetics. The refolding kinetics of wild type were fit to two and three exponentials, and residuals were examined to select the best fit (Fig. 6). Fitting to three exponentials (four states) resulted in a more random arrangement of residuals (Fig. 6). Compared with the two exponential fit, the residuals of the three exponential fit were improved for data in the range of 0-200 s, where the major structural transformations take place. Thus, as depicted in Reaction 1, wild-type H{gamma}D-Crys likely populated two major intermediates during refolding.Formula

First, a rapid decrease in fluorescence was observed with a half-time (t1/2) of 8 s, which corresponded to refolding from the unfolded (U) state into the first intermediate (IR1). Refolding into the second intermediate (IR2) then occurred with a t1/2 of 35 s, and finally refolding into the native (N) conformation occurred with a t1/2 of 130 s (Table 3).


View this table:
[in this window]
[in a new window]
  		TABLE 3
Productive kinetic refolding parameters for wild-type H{gamma}D-Crys and deamidation mutants

 
Refolding rates of the deamidation mutants were analyzed in an analogous manner as that described for wild type. All of the mutants exhibited a rapid decrease in fluorescence upon diluting into 1.0 M GdnHCl followed by a slower decrease in fluorescence (Fig. 7). All of the mutant proteins required significantly longer times than wild type to refold completely and were best fit by three exponentials similar to wild type (Fig. 7). For all mutants, the rates of refolding into IR1 and IR2 were similar to wild type, although the rate of the final transition from IR2 to N was decreased (Table 3). The t1/2 of the IR2 to N transition was increased more than 19 times for Q54E/Q143E (2500 s) as compared with wild type (130 s).

Kinetic Unfolding of Wild Type and Deamidation Mutants at pH 7.0—The effects of glutamine deamidation on the unfolding kinetics of H{gamma}D-Crys were analyzed by performing kinetic unfolding experiments. Native proteins were rapidly diluted into 5.0 M GdnHCl at pH 7.0, and the decrease in fluorescence intensity at 350 nm was monitored over time in order to follow the solvent exposure of buried tryptophans. As with the kinetic refolding experiments, a syringeport injection system was used because major transitions during unfolding occur on a second time scale (33). It was shown previously that the kinetic unfolding transitions of wild-type H{gamma}D-Crys were best fit to three exponentials, suggesting an unfolding pathway with two major intermediates as depicted in Reaction 2.Formula


Figure 7
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 7.
Normalized kinetic unfolding (gray line) and productive kinetic refolding (black line) transitions of wild-type H{gamma}D-Crys and deamidation mutants. For unfolding, native proteins were diluted into 5.0 M GdnHCl, and buffer at 37 °C as described in Fig. 6. For productive refolding, proteins were first unfolded in 5.5 M GdnHCl and subsequently diluted into buffer (as described in Fig. 6) to give a final GdnHCl concentration of 1.0 M at 37 °C. Insets shown for Q54E, Q143E, and Q54E/Q143E display the refolding transitions over the extended times that were required for complete refolding.

 
The kinetic unfolding transitions of wild type measured here agreed well with the previous analysis. A rapid increase in fluorescence at 350 nm corresponding to the native to the first intermediate (IU1) transition was followed by a slower increase corresponding to unfolding into the second intermediate (IU2) and finally unfolded (Fig. 7). The unfolding transitions were best fit to three exponentials with t1/2 values of 0.79, 33, and 200 s for the three phases, respectively (Table 4).


View this table:
[in this window]
[in a new window]
  		TABLE 4
Kinetic unfolding parameters for wild-type H{gamma}D-Crys and deamidation mutants

 
Kinetic unfolding rates of Q54E, Q143E, and Q54E/Q143E were measured in a manner analogous to that described for wild-type H{gamma}D-Crys. Kinetic unfolding transitions for all mutants exhibited a rapid increase in fluorescence followed by a slower increase, which were best fit to three exponentials (Fig. 7). The t1/2 values for unfolding into IU1 and IU2 were increased for all mutants compared with wild type. In contrast, the t1/2 values for the IU2 to unfolded transition were similar for the mutant and wild-type proteins (Table 4).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Glutamine and asparagine deamidation are very common forms of covalent damage found in the crystallins of aged and cataractous lenses. Chemically, deamidation results in replacement of an amide group with a carboxyl group and may cause side chain racemization and backbone isomerization, depending on steric hindrance induced by the protein backbone (42). In addition, the extent of deamidation depends on solvent accessibility of the asparagine or glutamine and the size and flexibility of the amino acid immediately following it (43, 44).

Many sites of deamidation have been identified in H{gamma}D-Crys from cataractous and noncataractous lenses of various ages (27, 28, 30). Gln-54 was identified as a potential site of deamidation in the water-soluble fraction of noncataractous lenses (28). In contrast, another study found no damage in H{gamma}D-Crys from the water-insoluble fraction of noncataractous lenses (27). A third study of the proteins from a 93-year-old cataractous lens found deamidation of H{gamma}D-Crys at a variety of sites, some of which agreed with previous results and others that were newly identified (30). Deamidation is a particularly difficult modification to detect because it results in a mass change of only 1 Da in the protonated state. Given the discrepancy between studies, and the difficulty in identifying deamidation, it is plausible that further sites of deamidation in H{gamma}D-Crys have yet to be identified. We focused our investigation on glutamines in the domain interface of H{gamma}D-Crys, a structurally significant location of the protein. Although Gln-54 and Gln-143 have low solvent accessibility in the native state, they are each within hydrogen bonding distances to two water molecules, critical reactants in nonenzymatic deamidation.

Effects of Interface Glutamine Deamidation on Structure—Single and double glutamine to glutamate substitutions of the interface residues, Gln-54 and Gln-143, did not prevent in vivo folding into soluble native-like conformations. Additionally, the CD and fluorescence spectra of the mutant proteins were similar to wild type suggesting that they had similar secondary and tertiary structures. Despite similar emission maxima, native fluorescence emission spectra of Q143E and Q54E/Q143E were increased about 10% in intensity compared with wild type (Fig. 2). This change may be due to relaxation of the anomalous native state quenching phenomenon that causes the fluorescence intensity of H{gamma}D-Crys to be lower in the native state than the unfolded state. Native state fluorescence quenching has been described previously for H{gamma}D-Crys and other betaand {gamma}-crystallins (21, 33, 34, 45). Increase in the fluorescence emission of Q143E and Q54E/Q143E may have been caused by relaxation of the native state quenching phenomenon, structural changes, or deviations in solution conditions.

Overall, the data observed here suggest that deamidation of Gln-54 and Gln-143 into L-glutamate does not cause considerable changes in the structure of H{gamma}D-Crys. High resolution crystal structures of the mutant proteins will be necessary to determine subtle changes in conformation not detectable by the spectroscopic methods employed here.

Effects of Interface Glutamine Deamidation on Stability—At pH 7.0 and at 37 °C, wild-type H{gamma}D-Crys exhibits three-state equilibrium unfolding/refolding transitions with an intermediate populated that likely has a structured C-td and unstructured N-td (12, 33). According to this model the first transition represents unfolding/refolding of the N-td and the second transition represents unfolding/refolding of the C-td (12).

At pH 7.0 the single and double interface deamidation mutations destabilized the N-td but not the C-td of H{gamma}D-Crys. Single and double alanine mutagenesis of Gln-54 and Gln-143 also caused selective destabilization of the N-td (13). Double mutant cycle analysis revealed that Gln-54 and Gln-143 have a very low interaction energy of ~0.7 kcal/mol (13). However, mutating the residues singly and doubly still resulted in notable destabilization. These results suggest that the residues are not positioned at the top of the domain interface to energetically clasp the domains together, but may instead function to shield the central domain interface hydrophobic cluster from solvent (13).

According to the model described above, mutating Gln-54 and Gln-143 to glutamate should not have a large effect on stability because the polar nature of the residues is maintained. However, significant destabilization was observed for the deamidation mutants, probably due to the introduction of negative charges. To test this hypothesis, we performed equilibrium unfolding/refolding experiments at acidic pH, below the pKa of the free glutamate side chain in aqueous solution. The lowest practical pH for these experiments was 3.0, because wild-type H{gamma}D-Crys partially unfolded at lower pH even in the absence of denaturant. In addition, urea was used instead of GdnHCl, and lower temperatures were employed for the pH 3.0 experiments to avoid amyloid fiber formation that was observed in the presence of GdnHCl.7

The pH 3.0 equilibrium unfolding transition of wild type appeared two state under the conditions employed here. Equilibrium unfolding transitions of Q54E, Q143E, and Q54E/Q143E at pH 3.0 were comparable with wild type with similar transition midpoints. Thus, the glutamine to glutamate mutations did not notably affect stability when the side chains were protonated. In contrast, the double alanine mutant Q54A/Q143A was destabilized at both pH 7.0 and 3.0, where it was best fit to a three-state transition. These results suggest that wild type and the deamidation mutants did populate an equilibrium intermediate at pH 3.0, but it was not readily observable because of similarities in the transition midpoints of the two domains at pH 3.0.

The results observed here indicate that subsets of the glutamates were protonated at pH 3.0. Additionally, the positively charged guanidinium ions present in denaturation samples did not completely mask the negatively charged side chains as has been described previously for acetylated ferricytochrome c and several coiled-coil proteins (46-49). The inability of guanidinium ions to completely or partially mask the glutamate side chains may be due to the low solvent accessibilities of these residues in the native state. In all probability, destabilization of H{gamma}D-Crys by deamidation of the interface glutamines occurred due to introduction of a negative charge into these buried locations. Introducing charged residues into the hydrophobic interior of staphylococcal nuclease has been shown previously to destabilize the protein in a pH-dependent manner (40, 50).

Thermal Stability of Wild-type H{gamma}D-Crys and Deamidation Mutants—To further avoid ionic effects of GdnHCl, we studied stability under low ionic conditions by using thermal denaturation. The {gamma}-crystallins exhibit high thermal stabilities (32, 51, 52). Sen et al. (52) reported a Tm of ~73 °C for bovine {gamma}D-crystallin, and Evans et al. (32) found that H{gamma}D-Crys resisted thermal aggregation after incubating at 70 °C for up to 10 min. In contrast, bovine {alpha}-crystallin has a Tm of 61 °C, and human betaB1-Crys has a Tm of 67 °C (14). We measured thermal stabilities of wild-type H{gamma}D-Crys and the deamidation mutants by observing changes in ellipticity at 218 nm at increasing temperatures using a CD spectrometer. By this method, wild-type H{gamma}D-Crys had a very high Tm of 83.8 °C. The deamidation mutants all had lower Tm values where Q54E was the least destabilized, Q143E was in-between, and Q54E/Q143E was the most destabilized.

The thermal denaturation data contradict the equilibrium data at pH 7.0 and 37 °C where, within error, Q54E and Q143E were similarly destabilized, and Q54E/Q143E was most destabilized. This discrepancy may be due to differences in the response of the N- and C-tds to thermal denaturation and GdnHCl-induced unfolding. The thermal denaturation transitions of the wild-type and mutant proteins all appeared two state, whereas the pH 7.0 equilibrium transitions were clearly three state, suggesting that the protein unfolds more cooperatively in response to heat than GdnHCl. This may be caused by the N-td and C-td having very similar Tm values or a cooperative unfolding mechanism not demonstrated in GdnHCl equilibrium experiments. Therefore, deamidation could have very different effects on stabilities of the domains under the two denaturation conditions. Alternatively, it is still possible that the differences are due to guanidinium ion interference in the equilibrium experiments. Similarly, unfolding and aggregation both occur in thermal experiments, which may obscure detection of intermediates.

Thermal stabilities of wild-type and the deamidation mutants were not measured at pH 3.0 because wild-type H{gamma}D-Crys has been observed to form amyloid fibers at pH 3.0 and temperatures higher than 20 °C.7

Kinetic Unfolding and Refolding—High thermodynamic stability is a critical property of the lens crystallins, which allows the proteins to remain folded despite advanced age. It has furthermore been suggested that the lens crystallins also utilize kinetic stability to prevent unfolding, where there exists an exceptionally high kinetic barrier to unfolding (53-55). The effects of deamidation on the free-energy barrier between the native and intermediate states of H{gamma}D-Crys were examined by performing productive kinetic refolding and unfolding experiments. As described above, the kinetic unfolding/refolding pathway of H{gamma}D-Crys has been investigated in detail by fluorescence (12, 13, 33).

Kinetic refolding transitions were best fit to three transitions suggesting two major intermediates. Rates of the third transitions were notably different for the mutants (Table 3). Given the previously described nucleation-dependent sequential domain refolding pathway, we hypothesize that the first and second transitions monitored refolding of the C-td and the third transition monitored refolding of the N-td. Kinetic unfolding transitions of wild type and the deamidation mutants were all best fit by three exponentials. The deamidation mutants had significant effects on rates of the second unfolding transition only (Table 4). These data suggest that the first and second transitions were N-td unfolding and the third transition was C-td unfolding (33).

Previous analyses of proteins adopting the beta{gamma}-crystallin domain fold have established that the beta-strands contributing to the Greek key motif closest to the domain interface, as well as a beta-hairpin between these strands, are highly stabilizing and may act as a nucleus for folding (55, 56). Given this, the partially folded intermediate (IR1) that had a partially folded C-td may have be structured in the region near the domain interface, and the intermediate (IU1) with a partially folded N-td may have been unstructured in the Greek key motif distant from the domain interface (Fig. 8).


Figure 8
View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 8.
Schematic reaction coordinate diagram describing qualitative relationships between the equilibrium and kinetic unfolding/refolding intermediates of H{gamma}D-Crys. Theorized effects of the interface deamidation mutations are shown as dashed lines. Line drawings of kinetic intermediates are schematics illustrating partially folded N- and C-terminal domains.

 
According to the hypotheses described above, and contrary to what might be expected for a reversible folding reaction, neither IU1 and IR2 nor IU2 and IR1 were equivalent. IU1 likely had a partially folded N-td where domain interface contacts were maintained, whereas IR2 likely had a folded C-td and unfolded N-td. The IU2 species was likely analogous to the IR2 species. Finally, IR1 likely had a partially folded C-td and fully unfolded N-td. The reason for these discrepancies may be due to differences in relative sizes of kinetic barriers on the forward and reverse pathways. For instance, during refolding, the U to IR1 transition was slow enough to detect, but during unfolding the IR1 to U transition may be too fast to detect. The reverse may be true of the N to IU1 transition explaining why it was only observed during unfolding. A schematic diagram describing potential qualitative relationships between the unfolding and refolding intermediates is shown in Fig. 8. It is also possible that the discrepancies are because of inherent differences in the unfolding and refolding pathways, such as the presence of off-pathway intermediates or parallel folding channels as has been described previously for the {alpha}-subunit of tryptophan synthase, an {alpha}/beta-barrel protein (57).

By comparing the kinetic unfolding and refolding data to thermodynamic stabilities determined by equilibrium experiments, it was possible to describe the effects of interface deamidation on the various stages of the unfolding and refolding pathways (Fig. 8). Consistent with what would be expected for mutations that disrupt the domain interface, the most significant effects were observed for transitions of the N-td. The native state was markedly destabilized by the deamidation mutants ({Delta}{Delta}GN-U = 2.0-3.0 kcal·mol-1·M-1). It was not possible to determine the stability of the kinetic unfolding intermediate IU1. However, given that the domains were likely still making contact, we expect that the mutants decreased the stability of IU1 similar to the native state. The height of the free energy barrier between N and IU1 was decreased slightly for the deamidation mutants ({Delta}{Delta}G{ddagger}1 = 0.32-0.47 kcal·mol-1·M-1). In contrast, the height of the free energy barrier between IU1 and IU2 was markedly decreased for the mutants ({Delta}{Delta}G{ddagger}2 = 0.94-1.65 kcal·mol-1·M-1), suggesting that this was the transition during which domain interface contacts were lost. On the refolding pathway, the free energy barrier between IR2 and N was increased considerably for the mutants (0.57-1.81 kcal·mol-1·M-1). Because it was not possible to unambiguously correlate intermediates on the unfolding and refolding pathways, we were not able to quantitatively compare changes in their respective free energy barriers. Regardless, the data here indicate that the deamidation mutants decreased the free energy barrier between N and IU2/IR2 enough to reduce the rate of refolding, but not as much as the native state was destabilized so that an increase in the rate of unfolding was also observed. In other words, in the model drawn in Fig. 8, {Delta}{Delta}Gu2 was less than {Delta}{Delta}GN-U for all of the mutants.

Covalent Damage and Lens Transparency—Cataract is associated with the presence of insoluble inclusions of covalently damaged crystallin proteins. Studying the effects of covalent damage on important physiological properties of the crystallins may give insight into in vivo alterations that could induce aggregation or insolubility. A general feature of many protein deposition diseases is aggregation from a partially folded or nonnative conformation (58-61). In order for aggregation to occur, it is first necessary to populate the problematic conformation. For the crystallins, destabilization of the native states and lowering the kinetic unfolding barrier by covalent damage may cause partial unfolding into aggregation-prone states. Interestingly, deamidation has also been shown to induce aggregation of amyloidogenic peptides (62, 63). Alternatively, covalent damage may alter surfaces of the crystallins in such a way that the short range order is disrupted, or insolubility is induced. Congenital cataracts caused by mutations in the gene encoding H{gamma}D-Crys instigate cataract formation through such mechanisms where native H{gamma}D-Crys associates by intermolecular disulfide bond formation, crystallization, or precipitation (32, 39, 64, 65). A congenital cataract mutant of human {gamma}C crystallin (T5P) had both altered conformation and was destabilized, suggesting that association of non-native conformations caused cataract (51). Although the exact means by which these mutations cause cataract almost certainly do not explain the mechanism of mature-onset cataract formation, these examples do provide evidence that alteration of surface properties or native-state destabilization can promote self-association or aggregation resulting in cataract.

Previous analyses have determined that the effects of deamidation on the structures, stabilities, and solubilities of the crystallins are dependent on the structural context of the damage (21, 22, 24-26). In the beta-crystallins, deamidation of domain interface glutamines has a large effect on stability where a monomeric intermediate is populated in equilibrium experiments (21, 22). Human betaB1 crystallin deamidated in the domain interface also had an increased tendency to aggregate at high temperatures and required more {alpha}-crystallin to suppress aggregation (24). Here we show that deamidation of the H{gamma}D-Crys domain interface glutamines destabilizes the protein and lowers the kinetic barrier to unfolding. These in vitro experiments generated deamidation mimics by site-directed substitution of L-glutamine or -asparagine with L-glutamate or -aspartate. The effects of backbone isomerization and side chain racemization that may also occur as a result of deamidation are expected to have even more severe effects on the stabilities and kinetic unfolding barriers of the crystallins. The effects of interface glutamine deamidation in H{gamma}D-Crys effectively increased the probability of populating partially unfolded conformations under conditions that favor the native state. These partially unfolded conformations may be prone to aggregation through mechanisms such as domain swapping or loop-sheet insertion (66, 67). For example, the single folded domain conformer may be susceptible to aggregation by domain swapping where the unfolded N-td of one monomer would use the C-td domain interface of another monomer during templated refolding. The intermediate with a folded C-td and partially unfolded N-td may be prone to aggregation through a reaction akin to loop-sheet insertion where an unstructured loop of one monomer inserts as a beta-strand into the beta-sheet of another monomer.

Rescue or repair of deamidated crystallins may be possible through the actions of {alpha}-crystallin and the enzyme isoaspartyl protein carboxyl methyltransferase. This enzyme, which functions to repair racemized aspartyl groups, has been identified in the lens where decreased expression correlates with cataract (68-70). It is unclear if there is a corresponding enzyme in the lens that may repair damaged glutamines. The onset of cataract may correspond to damage at regions of the crystallins particularly sensitive to chemical or structural alterations, such as the domain interfaces of the beta- and {gamma}-crystallins. Alternatively, onset may not depend on damage of these hot-spots but instead may occur when an overall threshold of damage has been exceeded at which time the {alpha}-crystallin chaperone complexes may be saturated or in an inactive aggregated state themselves.


   FOOTNOTES
 
* This work was supported in part by National Institutes of Health Grant GM17980 (to J. K.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 Supported by a Cleo and Paul Schimmel graduate fellowship. Back

2 Supported by a United Negro College Fund-Merck graduate fellowship. Back

3 To whom correspondence should be addressed: Dept. of Biology, Massachusetts Institute of Technology, 31 Ames St., Rm. 68-330, Cambridge, MA 02139. Tel.: 617-253-4700; Fax: 617-252-1843; E-mail: jaking{at}mit.edu.

4 The abbreviations used are: Crys, crystallin; H{gamma}D-Crys, human {gamma}D-crystallin; DTT, dithiothreitol; N-td, N-terminal domain; C-td, C-terminal domain; FI, fluorescence intensity. Back

5 K. Papanikiolopoulou, personal communication. Back

6 K. Papanikiolopoulou, I. A. Mills, S. L. Flaugh, and J. King, unpublished data. Back

7 K. Papanikiolopoulou, S. L. Flaugh, I. A. Mills, and J. King, unpublished data. Back


   ACKNOWLEDGMENTS
 
We thank Jiejin Chen and Melissa Kosinski-Collins for helpful discussions and Xiaonan Lin and Robin Nance for experimental assistance.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Delaye, M., and Tardieu, A. (1983) Nature 302, 415-417[CrossRef][Medline] [Order article via Infotrieve]
  2. Fernald, R. D., and Wright, S. E. (1983) Nature 301, 618-620[CrossRef][Medline] [Order article via Infotrieve]
  3. National Eye Institute (2002) Prevent Blindness in America, pp. 22-25, Bethesda, MD
  4. Boyle, D., and Takemoto, L. (1994) Exp. Eye Res. 58, 9-15[CrossRef][Medline] [Order article via Infotrieve]
  5. Horwitz, J. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 10449-10453[Abstract/Free Full Text]
  6. Bax, B., Lapatto, R., Nalini, V., Driessen, H., Lindley, P. F., Mahadevan, D., Blundell, T. L., and Slingsby, C. (1990) Nature 347, 776-780[CrossRef][Medline] [Order article via Infotrieve]
  7. Slingsby, C., and Bateman, O. A. (1990) Biochemistry 29, 6592-6599[CrossRef][Medline] [Order article via Infotrieve]
  8. Wistow, G., Turnell, B., Summers, L., Slingsby, C., Moss, D., Miller, L., Lindley, P., and Blundell, T. (1983) J. Mol. Biol. 170, 175-202[CrossRef][Medline] [Order article via Infotrieve]
  9. Hope, J. N., Chen, H. C., and Hejtmancik, J. F. (1994) J. Biol. Chem. 269, 21141-21145[Abstract/Free Full Text]
  10. Mayr, E. M., Jaenicke, R., and Glockshuber, R. (1994) J. Mol. Biol. 235, 84-88[Medline] [Order article via Infotrieve]
  11. Palme, S., Slingsby, C., and Jaenicke, R. (1997) Protein Sci. 6, 1529-1536[Medline] [Order article via Infotrieve]
  12. Flaugh, S. L., Kosinski-Collins, M. S., and King, J. (2005) Protein Sci. 14, 569-581[CrossRef][Medline] [Order article via Infotrieve]
  13. Flaugh, S. L., Kosinski-Collins, M. S., and King, J. (2005) Protein Sci. 14, 2030-2043[CrossRef][Medline] [Order article via Infotrieve]
  14. Bloemendal, H., de Jong, W., Jaenicke, R., Lubsen, N. H., Slingsby, C., and Tardieu, A. (2004) Prog. Biophys. Mol. Biol. 86, 407-485[CrossRef][Medline] [Order article via Infotrieve]
  15. Groenen, P. J., Merck, K. B., de Jong, W. W., and Bloemendal, H. (1994) Eur. J. Biochem. 225, 1-19[Medline] [Order article via Infotrieve]
  16. Lampi, K. J., Ma, Z., Hanson, S. R., Azuma, M., Shih, M., Shearer, T. R., Smith, D. L., Smith, J. B., and David, L. L. (1998) Exp. Eye Res. 67, 31-43[CrossRef][Medline] [Order article via Infotrieve]
  17. Lund, A. L., Smith, J. B., and Smith, D. L. (1996) Exp. Eye Res. 63, 661-672[CrossRef][Medline] [Order article via Infotrieve]
  18. Takemoto, L., and Boyle, D. (2000) Mol. Vis. 6, 164-168[Medline] [Order article via Infotrieve]
  19. Zhang, Z., Smith, D. L., and Smith, J. B. (2003) Exp. Eye Res. 77, 259-272[CrossRef][Medline] [Order article via Infotrieve]
  20. Harms, M. J., Wilmarth, P. A., Kapfer, D. M., Steel, E. A., David, L. L., Bachinger, H. P., and Lampi, K. J. (2004) Protein Sci. 13, 678-686[CrossRef][Medline] [Order article via Infotrieve]
  21. Kim, Y. H., Kapfer, D. M., Boekhorst, J., Lubsen, N. H., Bachinger, H. P., Shearer, T. R., David, L. L., Feix, J. B., and Lampi, K. J. (2002) Biochemistry 41, 14076-14084[CrossRef][Medline] [Order article via Infotrieve]
  22. Lampi, K. J., Amyx, K. K., Ahmann, P., and Steel, E. A. (2006) Biochemistry 14, 3146-3153
  23. Lampi, K. J., Oxford, J. T., Bachinger, H. P., Shearer, T. R., David, L. L., and Kapfer, D. M. (2001) Exp. Eye Res. 72, 279-288[CrossRef][Medline] [Order article via Infotrieve]
  24. Lampi, K. J., Kim, Y. H., Bachinger, H. P., Boswell, B. A., Lindner, R. A., Carver, J. A., Shearer, T. R., David, L. L., and Kapfer, D. M. (2002) Mol. Vis. 8, 359-366[Medline] [Order article via Infotrieve]
  25. Gupta, R., and Srivastava, O. P. (2004) Investig. Ophthalmol. Vis. Sci. 45, 206-214[Abstract/Free Full Text]
  26. Gupta, R., and Srivastava, O. P. (2004) J. Biol. Chem. 279, 44258-44269[Abstract/Free Full Text]
  27. Hanson, S. R., Hasan, A., Smith, D. L., and Smith, J. B. (2000) Exp. Eye Res. 71, 195-207[CrossRef][Medline] [Order article via Infotrieve]
  28. Hanson, S. R., Smith, D. L., and Smith, J. B. (1998) Exp. Eye Res. 67, 301-312[CrossRef][Medline] [Order article via Infotrieve]
  29. Lampi, K. J., Ma, Z., Shih, M., Shearer, T. R., Smith, J. B., Smith, D. L., and David, L. L. (1997) J. Biol. Chem. 272, 2268-2275[Abstract/Free Full Text]
  30. Searle, B. C., Dasari, S., Wilmarth, P. A., Turner, M., Reddy, A. P., David, L. L., and Nagalla, S. R. (2005) J. Proteome Res. 4, 546-554[CrossRef][Medline] [Order article via Infotrieve]
  31. Basak, A., Bateman, O., Slingsby, C., Pande, A., Asherie, N., Ogun, O., Benedek, G. B., and Pande, J. (2003) J. Mol. Biol. 328, 1137-1147[CrossRef][Medline] [Order article via Infotrieve]
  32. Evans, P., Wyatt, K., Wistow, G. J., Bateman, O. A., Wallace, B. A., and Slingsby, C. (2004) J. Mol. Biol. 343, 435-444[CrossRef][Medline] [Order article via Infotrieve]
  33. Kosinski-Collins, M. S., Flaugh, S. L., and King, J. (2004) Protein Sci. 13, 2223-2235[CrossRef][Medline] [Order article via Infotrieve]
  34. Kosinski-Collins, M. S., and King, J. (2003) Protein Sci. 12, 480-490[CrossRef][Medline] [Order article via Infotrieve]
  35. Fraczkiewicz, R., and Braun, W. (1998) J. Comput. Chem. 19, 319-333[CrossRef]
  36. Greene, R. F., Jr., and Pace, C. N. (1974) J. Biol. Chem. 249, 5388-5393[Abstract/Free Full Text]
  37. Clark, A. C., Sinclair, J. F., and Baldwin, T. O. (1993) J. Biol. Chem. 268, 10773-10779[Abstract/Free Full Text]
  38. Andley, U. P., Mathur, S., Griest, T. A., and Petrash, J. M. (1996) J. Biol. Chem. 271, 31973-31980[Abstract/Free Full Text]
  39. Pande, A., Pande, J., Asherie, N., Lomakin, A., Ogun, O., King, J., and Benedek, G. B. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 6116-6120[Abstract/Free Full Text]
  40. Fitch, C. A., Karp, D. A., Lee, K. K., Stites, W. E., Lattman, E. E., and Garcia-Moreno, E. B. (2002) Biophys. J. 82, 3289-3304[Medline] [Order article via Infotrieve]
  41. Forsyth, W. R., Antosiewicz, J. M., and Robertson, A. D. (2002) Proteins 48, 388-403[CrossRef][Medline] [Order article via Infotrieve]
  42. Takemoto, L., Fujii, N., and Boyle, D. (2001) Exp. Eye Res. 72, 559-563[CrossRef][Medline] [Order article via Infotrieve]
  43. Geiger, T., and Clarke, S. (1987) J. Biol. Chem. 262, 785-794[Abstract/Free Full Text]
  44. Lapko, V. N., Purkiss, A. G., Smith, D. L., and Smith, J. B. (2002) Biochemistry 41, 8638-8648[CrossRef][Medline] [Order article via Infotrieve]
  45. Bateman, O. A., Sarra, R., van Genesen, S. T., Kappe, G., Lubsen, N. H., and Slingsby, C. (2003) Exp. Eye Res. 77, 409-422[CrossRef][Medline] [Order article via Infotrieve]
  46. Hagihara, Y., Tan, Y., and Goto, Y. (1994) J. Mol. Biol. 237, 336-348[CrossRef][Medline] [Order article via Infotrieve]
  47. Kohn, W. D., Monera, O. D., Kay, C. M., and Hodges, R. S. (1995) J. Biol. Chem. 270, 25495-25506[Abstract/Free Full Text]
  48. Monera, O. D., Kay, C. M., and Hodges, R. S. (1994) Protein Sci. 3, 1984-1991[Medline] [Order article via Infotrieve]
  49. Monera, O. D., Kay, C. M., and Hodges, R. S. (1994) Biochemistry 33, 3862-3871[CrossRef][Medline] [Order article via Infotrieve]
  50. Nguyen, D. M., Leila Reynald, R., Gittis, A. G., and Lattman, E. E. (2004) J. Mol. Biol. 341, 565-574[CrossRef][Medline] [Order article via Infotrieve]
  51. Fu, L., and Liang, J. J. (2002) FEBS Lett. 513, 213-216[CrossRef][Medline] [Order article via Infotrieve]
  52. Sen, A. C., Walsh, M. T., and Chakrabarti, B. (1992) J. Biol. Chem. 267, 11898-11907[Abstract/Free Full Text]
  53. Das, B. K., and Liang, J. J. (1998) Int. J. Biol. Macromol. 23, 191-197[CrossRef][Medline] [Order article via Infotrieve]
  54. Jaenicke, R., and Slingsby, C. (2001) Crit. Rev. Biochem. Mol. Biol. 36, 435-499[CrossRef][Medline] [Order article via Infotrieve]
  55. MacDonald, J. T., Purkiss, A. G., Smith, M. A., Evans, P., Goodfellow, J. M., and Slingsby, C. (2005) Protein Sci. 14, 1282-1292[CrossRef][Medline] [Order article via Infotrieve]
  56. Bagby, S., Go, S., Inouye, S., Ikura, M., and Chakrabartty, A. (1998) J. Mol. Biol. 276, 669-681[CrossRef][Medline] [Order article via Infotrieve]
  57. Bilsel, O., Zitzewitz, J. A., Bowers, K. E., and Matthews, C. R. (1999) Biochemistry 38, 1018-1029[CrossRef][Medline] [Order article via Infotrieve]
  58. Booth, D. R., Sunde, M., Bellotti, V., Robinson, C. V., Hutchinson, W. L., Fraser, P. E., Hawkins, P. N., Dobson, C. M., Radford, S. E., Blake, C. C., and Pepys, M. B. (1997) Nature 385, 787-793[CrossRef][Medline] [Order article via Infotrieve]
  59. Jiang, X., Smith, C. S., Petrassi, H. M., Hammarstrom, P., White, J. T., Sacchettini, J. C., and Kelly, J. W. (2001) Biochemistry 40, 11442-11452[CrossRef][Medline] [Order article via Infotrieve]
  60. Mitraki, A., and King, J. (1989) Bio/Technology 7, 690-697[CrossRef]
  61. Wetzel, R. (1994) Trends Biotechnol. 12, 193-198[CrossRef][Medline] [Order article via Infotrieve]
  62. Nilsson, M. R., and Dobson, C. M. (2003) Protein Sci. 12, 2637-2641[CrossRef][Medline] [Order article via Infotrieve]
  63. Nilsson, M. R., Driscoll, M., and Raleigh, D. P. (2002) Protein Sci. 11, 342-349[CrossRef][Medline] [Order article via Infotrieve]
  64. Kmoch, S., Brynda, J., Asfaw, B., Bezouska, K., Novak, P., Rezacova, P., Ondrova, L., Filipec, M., Sedlacek, J., and Elleder, M. (2000) Hum. Mol. Genet. 9, 1779-1786[Abstract/Free Full Text]
  65. Pande, A., Pande, J., Asherie, N., Lomakin, A., Ogun, O., King, J. A., Lubsen, N. H., Walton, D., and Benedek, G. B. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 1993-1998[Abstract/Free Full Text]
  66. Carrell, R. W., Whisstock, J., and Lomas, D. A. (1994) Am. J. Respir. Crit. Care Med. 150, S171-175[Medline] [Order article via Infotrieve]
  67. Liu, Y., and Eisenberg, D. (2002) Protein Sci. 11, 1285-1299[CrossRef][Medline] [Order article via Infotrieve]
  68. Kodama, T., Mizobuchi, M., Takeda, R., Torikai, H., Shinomiya, H., and Ohashi, Y. (1995) Biochim. Biophys. Acta 1245, 269-272[Medline] [Order article via Infotrieve]
  69. McFadden, P. N., and Clarke, S. (1986) Mech. Ageing Dev. 34, 91-105[CrossRef][Medline] [Order article via Infotrieve]
  70. McFadden, P. N., Horwitz, J., and Clarke, S. (1983) Biochem. Biophys. Res. Commun. 113, 418-424[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Proc. Natl. Acad. Sci. USAHome page
K. Itoh and M. Sasai
Cooperativity, connectivity, and folding pathways of multidomain proteins
PNAS, September 16, 2008; 105(37): 13865 - 13870.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
J. R. Mazzulli, M. Armakola, M. Dumoulin, I. Parastatidis, and H. Ischiropoulos
Cellular Oligomerization of {alpha}-Synuclein Is Determined by the Interaction of Oxidized Catechols with a C-terminal Sequence
J. Biol. Chem., October 26, 2007; 282(43): 31621 - 31630.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/41/30782    most recent
M603882200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Flaugh, S. L.
Right arrow Articles by King, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Flaugh, S. L.
Right arrow Articles by King, J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2006 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement