Identification of Age- and Cataract-Related Changes in High-Density Lens Protein Aggregates

Abstract

Purpose: Cataract is believed to be caused by protein-protein and protein-membrane aggregation in the eye lens. After middle age, there is extensive binding of crystallins to the lens cell membranes as evidenced by sedimentation at high densities. Multiple protein modifications have been linked with cataract, whereas others have been associated with aging. The purpose of this study was to characterize protein constituents within high density protein-membrane fractions from normal aged or cataractous lenses and to compare these proteins and their modifications.

Methods: The inner nuclear regions of cataract or age-matched normal lenses were homogenized and proteins were separated using sucrose density gradient centrifugation. The low-density fractions (LDFs) and high-density fractions (HDFs) were analyzed by mass spectrometry using both top-down matrix-assisted desorption/ionization-mass spectrometry (MALDI-MS) and bottom-up liquid chromatography-tandem mass spectrometry (LC-MS/MS)-based proteomic methods. Quantification of low molecular weight crystallin peptides, deamidation, and isomerization were performed.

Results: Compared with normal aged-lenses, membrane-associated protein aggregates in high density fractions of cataract lenses exhibited significantly higher levels of γ-crystallins, as well as γS- and γD-crystallin C-terminal peptides. Deamidation of γ-crystallin, but not of β-crystallin, was increased in cataract lens membrane-bound aggregates. A very high level of Asp isomerization was detected in bound α-crystallins from both aged and cataract lenses.

Conclusions: Binding of crystallin aggregates to human lens cell membranes is associated with protein truncation, deamidation, and isomerization, and was observed in normal aged and cataract lenses. However, the protein aggregates bound to membranes in cataract lenses exhibit distinct modifications to γ-crystallins that may arise as a consequence of additional protein degradation.

Figure 1.

Representative MALDI-MS analysis of intact proteins and peptides in LDF (A, B) and HDF (C, D). HDFs and LDFs from normal lenses (B, D) and cataract lenses (A, C) were mixed with DHA matrix and analyzed by MALDI-MS using a Bruker Rapiflex Tissuetyper in linear positive mode (m/z range 5000–24,000). Representative spectra are shown from a 73-year-old healthy lens and 60-year-old cataract lens. The major signals in LDF corresponded to γ-crystallins (+1 or +2 charged) and β crystallins and the major signals in HDFs corresponded to truncated ɑA-crystallin fragments. * Indicates signals that have not been identified.

Figure 2.

MALDI-MS analysis of HDF in low m/z range. HDFs from a healthy lens (73-year-old, bottom) and a cataract lens (60-year-old, top) were mixed with CHCA matrix and analyzed in a Bruker SolariX 15T FT-ICR mass spectrometer (m/z range 1000–5000). Strong signals from γS-crystallin 153-177 and γS-crystallin 144-177 and a series of other γS-crystallin peptides were detected in the HDF from cataract lenses whereas these signals were barely detected in HDF from healthy lenses. A series of βA3- and αA-crystallin peptides was detected in both normal and cataract lenses. * Indicates signals that have not been identified.

Figure 3.

Quantification of γS- and βA3-crystallin C-terminal peptides in HDFs and LDFs. LMW peptides were extracted from HDFs and LDFs from both normal and cataract lenses. The γS-crystallin peptides (A) and βA3-crystallin peptides (B) were measured by LC-MS/MS. CLDF, cataract low density fraction; NLDF, normal low-density fraction; CHDF, cataract high density fraction; NHDF, normal high density fraction. The same abbreviations were used for other figures. Peaks corresponding to the peptides were extracted and peak intensities were used for quantification. # Indicates significantly different in HDFs than corresponding LDFs; * indicates significantly different in cataract lenses compared to corresponding fraction of normal lenses (P < 0.05). Data are presented as mean ± SD (n = 4).

Figure 4.

Quantification of Asn deamidation. The levels of deamidation were calculated as ratios of the intensities of deamidated peptides versus the intensities of their corresponding non-deamidated homologues. * Indicates significant difference between HDFs in cataract lenses compared with HDFs in healthy lenses; # Indicates significantly higher in HDFs compared with corresponding LDFs (P < 0.05). Data are presented as mean ± SD (n = 4).

Figure 5.

Percentage of Asp isomerization/racemization in alpha and beta crystallins from aged lenses. The total isomer peak intensity (I_isomer) was calculated by combing peak intensities of all isomers except L-Asp form and the total peak intensity was the summed peak areas of all peaks (I_total). The percent of isomerization was calculated by 100 × (I_isomer/I_total). # Indicates significantly higher in HDFs compared with corresponding LDFs; * indicates significantly lower HDFs in cataract lenses compared with the same fraction from healthy lenses (P < 0.05). Data are presented as mean ± SD (n = 4).