Oligomerization with wt αA- and αB-Crystallins Reduces Proteasome-Mediated Degradation of C-Terminally Truncated αA-Crystallin

Abstract

Purpose.: We previously demonstrated that the ubiquitin-proteasome pathway (UPP) is a general protein quality control system that selectively degrades damaged or abnormal lens proteins, including C-terminally truncated αA-crystallin. The objective of this work was to determine the effects of wt αA- and αB-crystallins on the degradation of C-terminally truncated αA-crystallin (αA_1–162) and vice versa.

Methods.: Recombinant wt αA, αB, and αA_1–162 were expressed in Escherichia coli and purified to homogeneity by chromatography. Subunit exchange and oligomerization were detected by fluorescence resonance energy transfer (FRET), multiangle-light scattering and coprecipitation assays. Protein substrates were labeled with ¹²⁵I and lens epithelial cell lysates were used as the source of the UPP for degradation assays.

Results.: FRET, multiangle light scattering, and coprecipitation assays showed that αA_1–162 exchanged subunits with wt αA- or wt αB- crystallin to form hetero-oligomers. αA_1–162 was more susceptible than wt αA-crystallin to degradation by the UPP. When mixed with wt αA-crystallin at 1:1 or 1:4 (αA_1–162 : wt) ratios to form hetero-oligomers, the degradation of αA_1–162 was significantly decreased. Conversely, formation of hetero-oligomers with αA_1–162 enhanced the degradation of wt αA-crystallin. The presence of αA_1–162, but not wt αA-crystallin, decreased the degradation of wt αB-crystallin.

Conclusions.: αA_1–162 forms hetero-oligomers with wt αA- and αB-crystallins. Oligomerization with wt αA- or αB-crystallins reduces the susceptibility of αA_1–162 to degradation by the UPP. In addition, the presence of αA_1–162 in the hetero-oligomers also affects the degradation of wt αA- and αB-crystallins.

Introduction

Lens fiber cells contain high concentrations of crystallins in the cytoplasm. These crystallins provide the lens with a high refractive index and minimize light scattering at the membrane-cytoplasm interface. The proper packing of these proteins in cells is essential for maintaining lens transparency. ^1,2 Disruption of the proper arrangement of these crystallins by adverse modifications is causally related to cataractogenesis. ^3–6 With aging and environmental insults, lens proteins undergo various modifications, such as oxidation, glycation, deamidation, and truncations. ^7–16 These modified or damaged proteins are usually unstable in aqueous solution and tend to aggregate and precipitate. The aggregation and precipitation of damaged/modified proteins accounts for a large portion of the age- and cataract-related increases in water-insoluble proteins in the lens. ^17–24 Thus, efficient removal or repair of the damaged or adversely modified proteins before their aggregation and precipitation appears to be essential for lens transparency. 

The ubiquitin-proteasome pathway (UPP) is an important protein quality control mechanism that selectively recognizes and degrades proteins with abnormal structures. ^25–31 We and others have demonstrated that the lens has a fully functional UPP and that this proteolytic pathway selectively degrades various forms of damaged proteins, including oxidized, glutathiolated, and thermally denatured proteins, ^25–43 but not glycated proteins. ⁴⁴ To establish and maintain lens protein homeostasis, the proper function of the UPP is essential for timely degradation of damaged, adversely modified proteins and many regulatory proteins. Disruption of the normal function of the UPP in the lens results in abnormal lens development and cataracts. ⁴⁵  

Truncation is one of the dominant forms of posttranslational modifications of lens proteins on aging and cataractogenesis. ^46–53 Although other proteases, such as βA3-crystallin or βA3-crystallin–associated proteinases may be involved in cleavage of lens protein, ^54,55 calpains play a major role in the generation of various truncated forms of crystallins. ^60–63 Calpains are a superfamily of structurally related, calcium-activated cysteine proteases. ^56–58 Calpain-mediated cleavage of lens proteins plays an important role in the aggregation and insolubilization of lens proteins, including crystallins and cytoskeleton proteins.^24,59–64 Calpain-mediated cleavage also causes the loss of chaperone activity of α-crystallins. ⁶⁵  

In addition to calpain- or other protease-mediated degradation, lens proteins in long-lived species such as humans can also be truncated nonenzymatically. For example, a recent publication postulated that a major cleavage site in lens AQP0 was caused nonenzymatically. ⁶⁶  

Both α- and β-crystallins, but not γ-crystallins, are susceptible to calpain-mediated cleavage. ⁶⁷ Whereas α-crystallins are cleaved by calpains at their C-terminus. ⁶⁸ β-crystallins are cleaved closer to their N-terminus. ⁶⁹ C-terminal cleavage of αA-crystallin by calpains can occur at several sites. The major cleaved products of αA-crystallin in the rat lenses include αA_1–151, αA_1–156, αA_1–163, and αA_1–168. ⁶⁸ αA_1–162 can be generated by incubation of αA-crystallin with calpain-2 in vitro. ⁶⁸ but it is barely detectable in normal rat lenses. ^68,70 However, αA_1–162 is readily detected in diabetic cataract rat lenses and is also found in small quantities in the older fibers of normal human lenses. ^70,71 Although there is considerable evidence of calpain-mediated crystallin truncation in mouse or rat lenses, there is far less evidence for calpain-mediated cleavage in human lenses, because there are no active calpain-3 isoforms in human lenses. ⁷² Calpain-2, the enzyme that may produce αA_1–162, has much lower activity in the human lens. Thus, the levels of αA_1–162 in human lens are very low. ⁷¹  

Accumulation of truncated αA-crystallins in cataractous lenses can result from both an increase in production and/or a decrease in degradation of the truncated products by other proteases, such as the UPP. In a previous study we compared the susceptibility of wt and the C-terminally truncated αA-crystallins to UPP-mediated degradation. ⁷³ We found that the sites of truncation at the C-terminus of αA-crystallin affect its conformation and susceptibility to UPP-mediated degradation. Whereas αA_1–168 is degraded at a reduced rate as compared with wild-type (wt) αA-crystallin, αA_1–162 is degraded at a much faster rate than wt αA-crystallin. Because the αA_1–162 is less thermally stable and prone to aggregation, it would appear that the timely degradation of the αA_1–162 is critical to prevent its accumulation and aggregation in the lens. The rapid degradation of αA_1–162 by the UPP may explain why αA_1–162 is barely detectable in normal lenses. 

α-Crystallin normally exists as large heterogeneous oligomers composed of two types of subunits, αA-crystallin and αB-crystallin, in a molar ratio of 3 to 1 in most mammalian lenses. ⁷⁴ It has been reported that C-terminally truncated αA-crystallins also form oligomers with wt αA- and αB-crystallins. ⁷⁵ Our previous work demonstrated that isolated αA_1–162 is rapidly degraded by the UPP. ⁷³ but it remained unknown whether the αA_1–162 in complexes with wt αA- or αB-crystallins was also selectively degraded. To fill this gap of knowledge, we investigated the effects of oligomerization with wt αA- or αB-crystallins on degradation of αA_1–162. Conversely, we determined the effects of incorporation of the αA_1–162 into hetero-complexes on degradation of wt αA- and αB-crystallins. The data indicate that whereas oligomerization with wt αA- or αB-crystallin reduced the degradation of αA_1–162, incorporation of αA_1–162 into wt αA-crystallin complex enhanced the degradation of wt αA-crystallin. In contrast, incorporation of αA_1–162 into wt αB-crystallin complex reduced the degradation of wt αB-crystallin. 

Materials and Methods

Construction of C-Terminally Truncated αA-Crystallin

To mimic C-terminal cleavage by calpains, αA_1–162-crystallin was generated by PCR-based cloning. ⁷³ Human αA-crystallin cDNA in pAED4 vecto. ⁷⁶ was used as the template, 5′-ACTCCATGGACGTGACCATCCAG-3′ and 5′-CATATGTTACGACACGGGGATGG-3′ were used as the pair of primers for the PCR reaction. The recombinant wt αA- and C-terminally truncated αA-crystallins were expressed in Escherichia coli and purified to homogeneity as described previously. ⁷³ Recombinant wt αB-crystallin was expressed and purified as previously described. ⁷⁷  

Labeling of WT α-Crystallins and αA_1–162 with Fluorescence Probes

To monitor the formation of hetero-oligomers between wt α-crystallins and αA_1–162, recombinant α-crystallins were first labeled with fluorescence probes (lucifer yellow iodoacetamide [LYI] or 4-acetamido-4′-((iodoacetyl)amino)-stilbene-2,2′-disulfonic acid [AIAS]) as described previously. ⁷⁸ Recombinant wt αA- and αB-crystallins, as well as αA_1–162 solutions, were prepared in 20 mM MOPS buffer containing 100 mM NaCl, pH 7.9, at a concentration of 1 mg/mL. Solid AIAS or LYI was added to a final concentration of 4 and 10 mM, respectively, and the reaction was allowed to proceed for 12 hours at room temperature (23°C) in the dark. For labeling with LYI, the reaction was extended for another 6 hours at 37°C. Unreacted AIAS and LYI were separated from the fluorescently labeled αA-crystallins on a G-25 Sephadex desalting column equilibrated with 50 mM sodium phosphate buffer containing 100 mM NaCl, 2 mM dithiothreitol (DTT), pH 7.5. 

Fluorescence Energy Transfer Assays

Fluorescence energy transfer was used to monitor subunit exchange or formation of hetero-oligomers. The reaction was initiated by mixing an equal volume of 0.5 mg/mL AIAS-labeled αA- or αB-crystallin and 0.5 mg/mL LYI-labeled αA_1–162 at 37°C in 50 mM sodium phosphate buffer containing 100 mM NaCl and 2 mM DTT, pH 7.5. At time 0, 15, and 30 minutes, 20 μL of the reaction mixture was taken and diluted 100 times with the same buffer. The emission spectrum of the sample excited at 335 nm was recorded using a Perkin-Elmer LS-5 spectrofluorometer (Perkin-Elmer, Waltham, MA). The decrease in fluorescence intensity at 415 nm of AIAS-labeled α-crystallin and the concomitant increase in fluorescence intensity at 525 nm of LYI-labeled α-crystallin were indicative of energy transfer owing to subunit exchange between the two labeled populations. Pilot experiments indicated that the subunit exchange reached equilibrium by 60 minutes of incubation. 

Size Exclusion-Dynamic Light-Scattering Assay

To verify the formation of hetero-oligomers between wt αA-crystallin and αA_1–162, solutions of wt αA-crystallin and αA_1–162, as well as the 1:1 mixtures of wt αA-crystallin and αA_1–162, were analyzed by size exclusion chromatography with inline light-scattering, absorbance, and refractive index detectors. ⁷⁴ A Superose 6HR 10/300 GL column (GE Healthcare, Piscataway, NJ) was connected in-line to an AKTA basic system (GE Healthcare), a Dawn-EOS multiangle laser light scattering detector (Wyatt Technology Corp., Santa Barbara, CA), and an Optilab-DSP refractive index detector (Wyatt Technology Corp.).Samples were loaded onto the column at a concentration of 1 mg/mL and eluted with 50 mM sodium phosphate buffer containing 100 mM NaCl, pH 7.0. 

Labeling of α-Crystallins with Biotin and Avidin Pulldown Assays

The wt α-crystallin and αA_1–162 solutions were prepared in PBS at a concentration of 2 mg/mL at pH 7.4. Sulfo-NHS-biotin stock solution at a concentration of 100 mM was made in dimethyl sulfoxide. To label wt αA-crystallin and αA_1–162 with sulfo-NHS biotin, 5 μL sulfo-NHS-biotin stock solution was added to 1 mL crystallin solutions and the reaction was carried out on ice for 2 hours with occasional shaking. The reaction was then terminated by adding lysine to a final concentration of 20 mM. Unreacted biotin was removed from the biotin-labeled crystallins using a G-25 Sephadex desalting column. To determine the formation of protein complexes, 100 μg of biotin-labeled wt αA-crystallin and αA_1–162 were mixed with an equal amount of unlabeled α-crystallin and incubated at 37°C for 30 minutes. The complexes were isolated by affinity pulldown using 50 μL NeutrAvidin beads. After 5 washes with PBS, 1 mL/wash, the specifically bound proteins were retrieved from the beads using SDS-gel loading buffer, resolved by SDS-PAGE (15% gel), and stained with Coomassie blue. 

Preparation of Lens Epithelial Cell Lysates

Human lens epithelial cells (SRA 01/04) were cultured using standard cell culture conditions (37°C, 5% CO₂, 10% fetal bovine serum) and collected by scraping at subconfluence. The cell pellets were homogenized with 50 mM Tris-HCl buffer containing 1 mM DTT, pH 7.6. After centrifugation at 100,000g for 10 minutes at 4°C, the supernatant was used as the source of the UPP for degradation assays. 

Proteolytic Degradation Assay and Statistical Analysis

The wt αA-crystallins and αA_1–162 were labeled with ¹²⁵I by the chloramine T method. ²⁶ Free ¹²⁵I was removed by Sephadex G25 desalting columns. The specific activity of ¹²⁵I-labeled proteins ranged from 0.2 to 0.5 μCi/μg. To determine the effects of hetero-oligomerization with wt αA- or αB-crystallin on degradation of αA_1–162, ¹²⁵I-labeled αA_1–162 was mixed with unlabeled wt αA- or αB-crystallins at the indicated ratios and incubated at 37°C for 30 minutes to form hetero-oligomers, and the mixtures were used for degradation assays. Unlabeled αA_1–162 was used as control to ensure each mixture contained the same amount of crystallins. To determine the effects of αA_1–162 on degradation of wt αA- or αB-crystallins, ¹²⁵I-labeled αA- or αB-crystallins were mixed with unlabeled αA_1–162 at the indicated ratios and incubated at 37°C for 30 minutes to allow the formation of hetero-oligomers. Unlabeled wt αA- or αB-crystallins were used as respective controls. Equal amounts of labeled substrates and unlabeled crystallins in each assay were used for the degradation assay (100 ng/assay). Degradation of the ¹²⁵I-labeled crystallins was determined as described by Huang et al.. ³¹ using human lymphatic endothelial cell (HLEC) lysates as sources of ubiquitinating and proteolytic enzymes. Briefly, the proteolysis reaction mixture, in a final volume of 25 μL, contained 50 mM Tris-HCl (pH 7.6), 5 mM MgCl₂, 1 mM DTT, 2 mM ATP, 10 mM creatine phosphate, 6 μg of creatine phosphokinase, 2 μg ubiquitin, 0.4 μg recombinant Ubc4, and 15 μL HLEC lysate (10 mg/mL protein). Ubc4 was expressed and purified essentially as described by Wing and Jain. ⁷⁹ Degradation was initiated by addition of ¹²⁵I-labeled αA-crystallins (4 to 10 × 1. ⁴ cpm/assay) and the reaction mixtures were incubated at 37°C for 2 hours. Reactions were terminated by addition of 200 μL of ice-cold 10 mg/mL bovine serum albumin, immediately followed by 50 μL of 100% trichloroacetic acid (TCA) (yielding a final concentration of 18.2% TCA), after which the samples were left on ice for 10 minutes. The extent of degradation was determined as the amount of TCA-soluble ¹²⁵I-labeled fragments of αA-crystallin. The total TCA-insoluble count at time 0 was defined as 100%. The percentage of degradation was proportional to the incubation time for the first 2 hours, thus we chose the 2-hour time point to compare the susceptibilities of substrates for degradation in this study. Because the substrate was not saturating and the degradation was expressed as percentage of the labeled substrate that was degraded during the 2-hour period, the susceptibility to degradation was not affected by variation concentrations of the labeled substrates under these experimental conditions. The portion of degradation that was inhibited in the presence of 20 μM MG132, a potent proteasome inhibitor, was designated as UPP-mediated degradation. For both wt α-crystallin and αA_1–162, more than 85% of the degradation in the lysates was proteasome-dependent. 

All experiments were performed in triplicate and repeated two to four times. For statistical analyses, data from several experiments were pooled and analyzed using the Student's t-test. 

Results

αA_1–162 Forms Hetero-oligomers with wt αA- and αB-Crystallins via Subunit Exchange

In the lens, αA-crystallin normally forms hetero-oligomers with αB-crystallin. To determine if αA_1–162 forms hetero-oligomers with wt αA- and αB-crystallins, we performed FRET assays using AIAS- and LYI-labeled wt α-crystallin or αA_1–162. As shown in Figure 1, AIAS-labeled α-crystallin exchanged subunits with LYI-labeled α-crystallin. The time-dependent decrease in fluorescence intensity at 415 nm of AIAS-labeled α-crystallin and the concomitant increase in fluorescence intensity at 525 nm of LYI-labeled α-crystallin indicate energy transfer between the two fluorescently labeled populations owing to subunit exchange. Comparison of Figure 1B with Figure 1A showed that αA_1–162 exchanged subunits similarly to that of wt αA-crystallin, but the efficiency of FRET of αA_1–162 was lower in comparison with wt αA-crystallin, indicating the rate of subunit exchange of the αA_1–162 was slower. As shown in Figure 1C, αA_1–162 also exchanged subunits with wt αA-crystallin. Again, the efficiency of subunit exchange was lower as compared with the subunits exchange between the two differently labeled wt αA-crystallin populations (compare Fig. 1C with Fig. 1A). Consistent with the hetero-oligomers of αA- and αB-crystallins in the lens, wt αA-crystallin exchanged subunits readily with purified αB-crystallin as indicated by FRET (Fig. 1D). αA_1–162 also exchanged subunits with αB-crystallin, but at a reduced efficiency compared with wt αA-crystallin (Fig. 1, compare panel E with D). 

Figure 1.

 

C-terminal truncation of αA-crystallin reduces subunit exchange between αA- and αB-crystallins. Recombinant wt α-crystallins and αA_1–162 were labeled with LYI or AIAS. FRET was used to determine the interaction between subunits of recombinant α-crystallins. A decrease in fluorescence intensity at 415 nm of AIAS-labeled α-crystallin with a concomitant increase in fluorescence intensity at 525 nm of LYI-labeled α-crystallin indicates subunits exchange. The data shown are representative of three reproducible independent experiments. (A) Subunit exchange between AIAS-labeled wt αA-crystallin and LYI-labeled wt αA-crystallin; (B) subunit exchange between AIAS-labeled αA_1–162 and LYI-labeled αA_1–162; (C) subunit exchange between AIAS-labeled wt αA-crystallin and LYI-labeled αA_1–162; (D) subunit exchange between AIAS-labeled wt αB-crystallin and LYI-labeled wt αA-crystallin; (E) subunit exchange between AIAS-labeled wt αB-crystallin and LYI-labeled αA_1–162.

To confirm that αA_1–162 forms complexes with wt αA- or αB-crystallin, we labeled wt αA- or αA_1–162 crystallins with biotin and then mixed them with unlabeled αA- or αB-crystallins. After incubation at 37°C for 30 minutes, the complexes were pulled down with NeutrAvidin beads and analyzed by SDS-PAGE. αA_1–162 and wt αA-crystallin migrated as distinctive bands on SDS- PAGE (Fig. 2, compare lane 2 with lane 1). Biotin labeling slowed migration slightly, resulting in a smearing of the bands on SDS-PAGE, probably owing to heterogeneity of labeling. After incubation with biotin-labeled αA_1–162, unlabeled wt αA- or αB-crystallins were pulled down together with biotin-labeled αA_1–162 (Fig. 2, lanes 3 and 6). Incubation with biotin-labeled wt αA-crystallin also resulted in pulling down of both unlabeled αA_1–162 (Fig. 2, lane 4) and αB-crystallin (Fig. 2, lane 5). However, biotin-labeled wt αA-crystallin migrated together with αB-crystallin on SDS-PAGE (Fig. 2, lane 5). To ensure that the pulldown of unlabeled crystallins by NeutrAvidin beads was because of formation of hetero-oligomers with biotin-labeled αA-crystallins, we did a mock pulldown assay. In the absence of biotin-labeled αA-crystallins, only marginal levels of unlabeled αA- or αB-crystallins were pulled down (Fig. 2, lanes 7 and 8), indicating that the observed pulldown of unlabeled α-crystallins was because of formation of hetero-complexes between biotin-labeled and -unlabeled α-crystallins (Fig. 2, lanes 3–6). 

Figure 2.

 

C-terminally truncated αA-crystallin forms complexes with wt αA- and αB-crystallins. The wt αA-crystallin and αA_1–162 were labeled with biotin, mixed with unlabeled α-crystallins and incubated at 37°C for 30 minutes. The complexes were isolated by affinity pulldown. The specifically bound proteins were retrieved from the beads, resolved by SDS-PAGE and stained by Coomassie blue. Lanes 1 and 2, biotin-labeled wt αA-crystallin and αA_1–162, respectively; Lane 3, pulldown of wt αA-crystalllin with biotin-labeled αA_1–162; Lane 4, pulldown of αA_1–162 with biotin-labeled wt αA-crystallin; Lane 5, pulldown of wt αB-crystallin with biotin-labeled wt αA-crystallin; Lane 6, pulldown of wt αB-crystallin with biotin-labeled αA_1–162; Lane 7, mock pulldown of unlabeled αA_1–162 and wt αB-crystallin; lane 8, mock pulldown of unlabeled of αA- and αB-crystallins.

Oligomers formed by purified C-terminally truncated αA-crystallins were smaller than those formed by purified wt αA-crystallin. ⁷³ To verify that αA_1–162 forms hetero-oligomers with wt αA-crystallin, we determined the sizes of oligomers of pure wt αA-crystallin, pure αA_1–162, and the 1:1 mixture of αA-crystallin and αA_1–162. Consistent with our previous findings, oligomers formed by pure αA_1–162 were heterogeneous and in general smaller than those of pure wt αA-crystallin (Fig. 3). In addition, a small fraction of αA_1–162, but not wt αA-crystallin, existed as high-mass aggregates. This is consistent with our previous finding that the αA_1–162 was less thermally stable and prone to aggregation. The oligomer sizes of the 1:1 mixture of wt αA-crystallin and αA_1–162 were slightly smaller than those of pure wt αA-crystallin, but were larger than those formed with pure αA_1–162 (Fig. 3). The shape of the elution curve of the mixture was similar to that of pure wt αA-crystallin, except for the presence of high-mass aggregates of αA_1–162 (Fig. 3). Similarly, the oligomer sizes of the 1:1 mixture of αA_1–162 and wt αB were slightly smaller than those of pure wt αB, but were larger than those in the main peak of pure αA_1–162 (Supplemental Fig. 1). Together, these data indicate that αA_1–162 forms hetero-oligomers with wt αA- or αB crystallins. However, it appears that the aggregated form of αA_1–162 was not incorporated into αA- or αB crystallins, as the proportions and sizes of the aggregates remained the same after incubation with wt αA-crystallin (Fig. 3 and Supplemental Fig. 1, available a). 

Figure 3.

 

C-terminally truncated αA-crystallin forms oligomers with wt αA-crystallin. Oligomer states of wt (αA-crystallin [red]), αA_1–162 (blue) and the 1:1 mixture of wt (αA-crystallin and αA_1–162 [green]) were determined by size-exclusion chromatography coupled with multiangle light-scattering detectors. The sizes of oligomers formed by αA_1–162 were smaller than those formed by wt αA-crystallin. High-mass aggregates were also detected in the preparation of αA_1–162, but not in the sample of wt αA-crystallin. The elution curve and sizes of oligomers of the 1:1 mixtures of wt αA-crystallin and αA_1–162 were similar to those of wt αA-crystallin, except for the existence of high-mass aggregates of αA_1–162.

Oligomerization with WT αA- or αB-Crystallin Reduced the Degradation of αA_1–162

As we demonstrated previously, αA_1–162 was more susceptible to degradation than wt αA-crystallin by the UPP. When incubated with lens epithelial cell lysates for 2 hours, approximately 4% of αA_1–162 was degraded by the UPP (Fig. 4A), whereas only approximately 2% of the wt αA-crystallin was degraded under the same condition (Fig. 5A). To determine the effects of formation of the hetero-oligomers with wt αA- or αB-crystallins on degradation of αA_1–162, ¹²⁵I-labeled αA_1–162 was mixed with unlabeled wt αA- or αB- crystallins at 1:1 or 1:4 ratios and incubated at 37°C for 30 minutes to form hetero-oligomers. To rule out the possibility of substrate competition, unlabeled αA_1–162 was used as a control for wt αA- or αB- crystallins. Incubation with wt αA-crystallin at both 1:1 and 1:4 ratios resulted in an approximately 40% decrease in degradation of αA_1–162 (Fig. 4A), whereas incubation with wt αB-crystallin at the 1:4 ratio, but not at the 1:1 ratio, decreased the degradation of αA_1–162 (Fig. 4B). These data indicate that formation of hetero-oligomers with wt αA- and αB-crystallins significantly reduces the degradation of αA_1–162 (Fig. 4). The effects wt αA-crystallin on degradation of αA_1–162 appeared to be stronger than those of wt αB-crystallin (Fig. 4, compare A with B). 

Figure 4.

 

Oligomerization with wt αA- or αB-crystallins reduces the degradation of C-terminally truncated αA-crystallin. ¹²⁵I-labeled αA_1–162 was mixed with indicated molar ratios of wt αA-crystallin (A) or wt αB-crystallin (B) to form complexes and then the mixtures were subjected to UPP-mediated degradation in lens epithelial cell lysates. The degradation of ¹²⁵I-labeled αA_1–162 was monitored. The degradation assays were performed in triplicate and repeated four times. Data presented are means + standard deviations. *Indicates P < 0.05 when compared with the degradation in the absence of wt αA- or αB-crystallins.

Figure 5.

 

Incorporation of C-terminally truncated αA-crystallin promotes the degradation of wt αA-crystallin but inhibits the degradation of wt αB-crystallin. ¹²⁵I-labeled wt αA- or ¹²⁵I-labeled αB-crystallin were mixed with αA_1–162 or wt αA-crystallin at indicated molar ratios and incubated at 37°C for 30 minutes to form hetero-complexes. The mixtures were then subjected to degradation in lens epithelial cell lysates. The degradation of ¹²⁵I-labeled wt αA- or αB-crystallins was monitored. (A) Effects of αA_1–162 on degradation ¹²⁵I-labeled wt αA-crystallin; (B) effects of αA_1–162 on degradation ¹²⁵I-labeled wt αB-crystallin; (C) effects of wt αA-crystalllin on degradation ¹²⁵I-labeled wt αB-crystallin. The degradation assays were performed in triplicate and repeated three times (A, B) or two times (C). Data presented are means + standard deviations. *Indicates P < 0.05 when compared with the degradation in the absence of αA_1–162.

Oligomerization with αA_1–162 Altered the Degradation of WT αA- and αB-Crystallins

The above data indicate that oligomerization with wt αA- and αB-crystallins reduced the degradation of αA_1–162. To determine the reciprocal effects of αA_1–162 on degradation of wt αA- or αB-crystallins, we incubated ¹²⁵I-labeled wt αA- or αB-crystallins with unlabeled αA_1–162 at different ratios and then determined the degradation of ¹²⁵I-labeled wt αA- or αB-crystallins. The data in Figure 5A suggest that the presence of αA_1–162 increased the degradation of wt αA-crystallin in a dose-dependent manner, although the increase was not statistically significant. In contrast, incubation with αA_1–162 reduced the degradation of wt αB-crystallin in a dose-dependent manner (Fig. 5B). To determine whether the inhibitory effects on degradation of wt αB-crystallin were specific to αA_1–162, we incubated ¹²⁵I-labeled wt αB-crystallin with the same amount of unlabeled wt αA-crystallin. As shown in Figure 5C, incubation with wt αA-crystallin had no detectable effects on degradation wt αB-crystallin. Together, these data indicate that oligomerization with αA_1–162 differentially affects the degradation of wt αA- and αB-crystallins. 

Discussion

We and others have demonstrated that the UPP is an important protein quality control system and that various forms of damaged or abnormal protein are selectively degraded by the UPP.^{25,27,29,31,73,80–83} However, it remains largely unknown whether formation of hetero-complexes with normal proteins alters the degradation behaviors of damaged proteins. Because most proteins in the lens exist as complexes, investigating the impact of formation of hetero-complexes on degradation of damaged proteins by the UPP would provide more physiologically relevant information regarding the role of the UPP in lens protein quality control. In this study, we examined the effects of formation of hetero-oligomers with wt α-crystallins on degradation of C-terminally truncated αA-crystallins using αA_1–162 as a model substrate. The data show that αA_1–162 readily interacts with wt αA- or αB-crystallins and forms hetero-oligomers with wt αA- or αB-crystallins. Consistent with our previous work. ⁷³ αA_1–162 was more susceptible than wt αA-crystallin to proteasome-mediated degradation. Oligomerization with wt αA- or αB- crystallins reduced the degradation of αA_1–162. Conversely, the presence of αA_1–162 in the hetero-complexes enhanced the degradation of wt αA-crystallin, but reduced the degradation of wt αB-crystallin. These results suggest that formation of hetero-complexes not only alters the degradation behavior of αA_1–162, but also the degradation behaviors of wt α-crystallins. 

α-Crystallins are the major components of the ocular lens and they play essential roles in maintaining lens transparency, both as structural proteins and as molecular chaperones. ^74,84 Mutation and adverse modifications of α-crystallins are associated with cataractogenesis.^{3,5,77,85–90} C-terminal truncation of αA-crystallin is one type of posttranslational modification that is associated with cataract.^46,51,68,70 Modification of α-crystallins not only alters their chaperone activity,^{3,5,77,86–90} but also their stability and susceptibility to proteolytic degradation. For example, αA_1–162 is thermally unstable and prone to aggregations. ⁷³ If not removed in a timely manner from the lens, accumulation and precipitation of this modified crystallin would cause light scattering and lens opacity. We showed that isolated αA_1–162 was more susceptible than wt αA-crystallin to proteasome-mediated degradation. ⁷³ The preferential degradation of αA_1–162 by the UPP may play a role in preventing its accumulation in the lens. Accumulation of αA_1–162 and other forms of modified proteins in cataractous lenses may be related to impairment of UPP or attributable to altered susceptibility of the modified proteins to proteasome-mediated degradation. Consistent with this hypothesis, experimental impairment of the UPP by targeted expression of dominant negative ubiquitin in the lens results in accumulation of truncated crystallins and cataract. ⁴⁵ We also found that glycation of αA_1–162 reduced its susceptibility to proteasome-mediated degradation. ⁴⁴ The reduced susceptibility of glycated proteins to proteasome-mediated degradation may explain why αA_1–162 was detected in diabetic lenses, but not detected in normal lenses. ^68,70  

In the lens, α-crystallins normally exist as hetero-oligomers. ⁷⁴ C-terminally truncated crystallins also form hetero-oligomers with wt α-crystallins. ⁹¹ Damaged proteins in the lens may form complexes with wt crystallins via subunit exchange or through chaperone functions of α-crystallins. Data from this work indicate that that αA_1–162 readily exchanges subunits with wt αA- and αB-crystallins, but the efficiency of subunit exchanges was lower compared with wt αA-crystallin (Fig. 1). This was consistent with previous findings. ⁷⁵ αA_1–162 is thermally unstable and prone to form high-mass aggregates. Dynamic light-scattering analysis shows that a significant portion of αA_1–162 exists as high-mass aggregates (Fig. 3). It is plausible that the aggregated form of αA_1–162 is not competent to subunit exchange, thus reducing the efficiency of subunit exchange, as we observed that the proportion and sizes of high-mass aggregates of αA_1–162 remained the same after incubation with wt αA-crystallin (Fig. 3). 

Our data show that formation of hetero-oligomers with wt αA- or αB-crystallin significantly reduced the degradation of αA_1–162 (Fig. 4); however, the mechanisms by which oligomerization with wt αA- or αB-crystallin reduces the degradation of αA_1–162 remain to be elucidated. We hypothesize that the enhanced degradation of αA_1–162 is because of subtle conformation changes that result in the exposure of intrinsic degradation signals, such as hydrophobic patches. ⁷³ The presence of wt αA- or αB-crystallin in the hetero-oligomers may partially mask the degradation signal on αA_1–162 and reduce its degradation. 

Interestingly, we found that incorporation of αA_1–162 into oligomers of wt αA- or αB-crystallins also altered the degradation behaviors of wt αA- and αB-crystallins (Fig. 5), but in different manners. Whereas incorporation of αA_1–162 increased the degradation of wt αA-crystallin, the presence of αA_1–162 decreased the degradation of wt αB-crystallin. The different effects of αA_1–162 on degradation of wt αA- and αB-crystallins suggest different models of recognition and degradation. We propose that the hetero-oligomers formed between αA_1–162 and wt αA-crystallin, rather than their individual subunits, are recognized and degraded by the UPP. Formation of hetero-oligomers with wt αA-crystallin may partially mask the degradation signals of αA_1–162 and reduce its degradation. On the other hand, if degradation signals on αA_1–162 in the hetero-oligomers were recognized by the UPP, wt αA-crystallin subunits in the hetero-oligomers would be degraded together with αA_1–162 (Fig. 6A). Thus, incorporation of αA_1–162 into oligomers of wt αA-crystallin promotes the degradation of wt αA-crystallin. Consistent with this hypothesis, the degradations of αA_1–162 and wt αA-crystallin in the hetero-complexes were comparable (compare Fig. 4A with Fig. 5A). 

Figure 6.

 

Hypothetic models for the effects of wt αA- and αB- crystallins on degradation of C-terminally truncated αA_1–162-crystallin. (A) wt αA-crystallin is resistant to degradation because of the lack of degradation signals on the surface. C-terminal truncation results in conformational changes and exposure of degradation signal, such as hydrophobic patches (marked red in the diagram) and render αA_1–162 more susceptible to degradation by the UPP. Oligomerization with wt αA-crystallin masks some of the degradation signals on αA_1–162, thus reducing its degradation. However, any unmasked degradation signals on αA_1–162 subunits in the hetero-oligomers also bring the wt αA-crystallin subunits in the oligomers for degradation. Thus, oligomerization with αA_1–162 increases the degradation of wt αA-crystallin. (B) wt αB-crystallin is susceptible to degradation because of the availability of hydrophobic patches on the surface (marked dark red in the diagram). The hydrophobic patches serve both as degradation signals and as client-binding sites for its chaperone function. The degradation signal on αA_1–162 may overlap or share with chaperone-binding signal. In addition to subunit exchange, oligomerization between wt αB and αA_1–162 may involve chaperone-client binding, which masks the degradation signals on both wt αB-crystallin subunits and αA_1–162 subunits. Thus, the susceptibilities of both wt αB-crystallin and αA_1–162 in the hetero-oligomers are reduced.

However, the above model cannot explain why the presence of αA_1–162 reduced the degradation of wt αB-crystallin (Fig. 5B). We found that αB-crystallin is a better substrate than αA-crystallin for UPP-mediated degradation (compare Fig. 5B with Fig. 5A), indicating that wt αB-crystallin may have exposed signals for degradation by the UPP. Formation of hetero-oligomers with αA_1–162 may mask the degradation signals on αB-crystallin and reduce its degradation (Fig. 6B). If that is the case, the interaction between wt αB-crystallin and αA_1–162 must be different from the interaction between wt αB- and wt αA-crystallins, as the presence of wt αA-crystallin had no effects on degradation of wt αB-crystallin (Fig. 5C). 

αB-crystallin is a chaperone and it actively interacts with proteins with abnormal structures and protects them from aggregation and precipitation. The interaction between wt αA-crystallin and wt αB-crystallin is mainly through subunit exchange; however, the interactions between αA_1–162 and wt αB-crystallin may involve chaperone-client binding, in addition to subunit exchange (Fig. 6B). Hydrophobic interaction is often involved in binding of client proteins by chaperones, ^92,93 and hydrophobic patches on protein surface are also signals for UPP-mediated degradation.^{29,73,83,94–97} The chaperone-client interaction between wt αB-crystallin and αA_1–162 may mask the degradation signal on αB-crystallin and reduce its degradation by the UPP. 

Our data obtained from cell-free experiments show that recombinant αB-crystallin is degraded faster than recombinant αA-crystallin, even faster than C-terminally truncated αA-crystallin. The greater susceptibility of αB-crystallin to UPP-mediated degradation raises the question why the native αB-crystallin is not cleared in the lens, as there is an active UPP in lens fibers. This inconsistency could be because of potential difference in regulation and selectivity of the UPP between intact lens and lens epithelial cell lysate. This difference might explain why native αB-crystallin in intact lenses is not cleared, although recombinant wt αB-crystallin is rapidly degraded by the UPP in the cell-free system. Another possibility is that the conformation of the recombinant wt αB-crystallin might not be identical to that of native αB-crystallin in the intact lens. The subtle difference in conformation might contribute to the difference between recombinant wt αB-crystallin and native αB-crystallin in susceptibility to UPP-mediated degradation. Furthermore, native αB-crystallin in the lens is normally associated with αA-crystallin and other client proteins. The binding of client proteins (including truncated αA-crystallin) to αB-crystallin in the lens might mask the degradation signals in αB-crystallin and prevent it from degradation, as illustrated in Figure 6B. 

In summary, this study confirmed the previous finding that αA_1–162 is preferentially degraded by the UPP. Formation of hetero-oligomers with wt αA- or αB-crystallins reduces, but does not prevent, the selective degradation of αA_1–162. Furthermore, the presence of αA_1–162 also affects the degradation of wt αA- and αB-crystallins. Future structural characterization of the hetero-oligomers formed by αA_1–162 and wt αA- or αB-crystallins will help elucidate mechanism by which the presence of αA_1–162 alters the degradation of wt αA- and αB-crystallins. 

Supplementary Materials

PDF S1 

Footnotes

 This work is supported by NIH Grants EY011717, EY013250, and EY03897; USDA CRIS 1950-51000-060-01A, USDA AFRI Award 2009-35200-05014; Chinese NNSF 81070719; and Fundamental Research Funds for the Central Universities 3030901009017.

Footnotes

 Disclosure: M. Wu, None; X. Zhang, None; Q. Bian, None; A. Taylor, None; J.J. Liang, None; L. Ding, None; J. Horwitz, None; F. Shang, None