Accumulation of Large Protein Fragments in Prematurely Senescent ARPE-19 Cells

Abstract

purpose. Senescence of retinal pigment epithelial (RPE) cells is a crucial event in the pathogenesis of age-related macular degeneration (AMD). This study was designed to improve the understanding of proteomic changes that underlie RPE senescence. Specifically, the levels of several protein fragments in prematurely senescent ARPE-19 cells were quantitatively compared with those in control cells.

methods. Premature senescence of human ARPE-19 cells was induced by repeated treatments with 6 mM tert-butylhydroperoxide (tert-BHP). Whole senescent cells were then treated with deuterated D₃-acrylamide, and control cells were treated with normal D₀-acrylamide. The D₃ and D₀ samples were mixed at a 1:1 ratio, and the proteins were separated by FPLC (fast protein liquid chromatography) and 2D-PAGE (two-dimensional polyacrylamide gel electrophoresis). After in-gel trypsinolysis, the relative quantification of selected proteins and fragments in the senescent cells versus control ARPE-19 cells was achieved by calculating the ratio of signal intensities for the deuterated and normal forms of cysteine-containing labeled peptides in MALDI-MS (matrix-assisted laser desorption/ionization mass spectrometry) spectra.

results. Several large fragments of typical cytosolic proteins, such as GAPDH, triosephosphate isomerase, and M2-type pyruvate kinase increased approximately two- to threefold in the prematurely senescent ARPE-19 cells.

conclusions. This study is the first demonstration that large fragments of cytosolic proteins can be accumulated in prematurely senescent ARPE-19 cells, the in vitro model of AMD. These data suggest that protein degradation processes are impaired in these cells and point to a new type of “waste” material in post-mitotic cells that may contribute to the senescent phenotype.

Age-related macular degeneration (AMD) is a retinal degenerative disease and retinal pigment epithelium (RPE) is the primary target tissue. The senescence and dysfunction of RPE have been proposed as a pathogenic pathway. One of the major changes in the senescent RPE is an alteration in lysosomal and proteasomal activities. ^{1 2 3 4 5 6 7} The lysosomal digestion of the continuously growing photoreceptor outer segments is the most important function of the healthy RPE. Proteasomal protein degradation is also a major intracellular function that is responsible for cellular homeostasis and survival. Alterations of these vital functions can cause accumulation of partially degraded proteins with unknown biological activity. For example, fibronectin fragments found in age-related degenerative diseases stimulate release of various proinflammatory factors from murine RPE cells ⁸ and may be linked to the development of AMD. Alternatively, partially degraded proteins can be a source of proteinaceous waste material that accumulates in the senescent cells and contributes to the senescent phenotype. 

In the current study, human ARPE-19 cells subjected to recurring oxidative stress ⁹ were used as an in vitro model of premature senescence. We focused on quantitative proteomic analysis of partially degraded proteins. Recent advantages in quantitative protein profiling have made possible the relative quantification of individual proteins within two different samples. ¹⁰ The approaches vary in detail but all are based on labeling the proteins in two separate samples with isotopically normal and heavy reagents. Our approach is based on the use of light (nondeuterated or D₀) and heavy (deuterated or D₃) acrylamide to alkylate cysteine residues in proteins. ¹¹ Heavy acrylamide has three deuterium atoms instead of three hydrogen atoms and is three mass units heavier than light acrylamide. From the chemical point of view, there is an insignificant difference between light and heavy forms of acrylamide. This similarity warrants equal efficiency of alkylation in separate samples and, after mixing, the co-migration of the same protein from separate samples on two-dimensional polyacrylamide gel electrophoresis (2D-PAGE). ^{12 13 14} The spots of interest are then in-gel digested with trypsin and identified using matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). A relative abundance of a protein in the original samples is determined as the ratio of signal intensities for the isotopically heavy and light forms of a cysteine-containing labeled peptide in MALDI-MS spectra. ^{11 12 13 14} In the present study, we found accumulation of large fragments of several typical cytosolic proteins in the prematurely senescent ARPE-19 cells in comparison to the control cells. 

Materials and Methods

Reagents

ARPE-19 cells were purchased from ATCC (Manassas, VA). Acrylamide, dithiothreitol (DTT), AG 501-X8 (D) resin, and supplies for 2D-PAGE were from Bio-Rad Laboratories (Hercules, CA). Acrylamide (2,3,3-D3, 98%) was purchased from Cambridge Isotope Laboratories (Andover, MA). A silver-stain protein detection kit (SilverSNAP kit) was from Pierce (Rockford, IL), and a senescence-associated β-galactosidase (SA β-gal) staining kit was from Cell Signaling (Danvers, MA). Sequencing grade modified trypsin was obtained from Promega (Madison, WI). All other chemicals were purchased from Sigma-Aldrich (St. Louis, MO). 

Cell Treatments and Analysis of Senescence

The human retinal pigment epithelium cells (ARPE-19) were maintained in growth medium: DMEM:F12, supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin at 37°C and 5% CO₂. Premature senescence of ARPE-19 cells was induced by oxidative stress similar to previous study by Glotin et al. ⁹ Ninety percent confluent ARPE-19 cells were treated with 6 mM of tert-butylhydroperoxide (tert-BHP) for 1 hour at 37°C. Cells were then washed twice with growth medium and allowed to recover for 24 hours. The procedure of tert-BHP treatment was repeated five times for five consecutive days. For control cells, the growth medium was replaced in parallel according the same time schedule. After the last tert-BHP treatment, the cells were allowed to recover for 3 days. Cells were then trypsinized, cell viability was assessed by trypan blue staining, and cells were replated at the concentration of 10,000 cells per square centimeter. Three days later, the replated cells reached 90% confluence. Cells were then trypsinized, washed once with PBS, and collected by centrifugation. Cell pellets were immediately frozen on dry ice/ethanol and stored at −80°C until further use. 

Cellular senescence was assayed with an SA β-gal staining kit according to the manufacturer’s protocol. Pictures were taken after 6 hours of incubation with the staining solution at 100× magnification (Axio Imager equipped with CCD camera; Carl Zeiss Meditec, Dublin, CA). 

Treatment with D₀- and D₃-Acrylamide

Whole control ARPE-19 cells (55 million) and whole prematurely senescent ARPE-19 cells (22 million) were directly resuspended in 3.6 mL of 20 mM Tris-HCl (pH 8.0)/8 M urea/20 mM DTT and sonicated with six 10-second pulses at 40% duty cycle (Sonifier 450; Branson Ultrasonics Corp., Danbury, CT). After 1 hour incubation at room temperature that allows protein sulfhydryl groups to be reduced, proteins in the control sample were alkylated with 400 mM light (nondeuterated, D₀) acrylamide, whereas the proteins in the senescent sample were alkylated with 400 mM heavy (deuterated, D₃) acrylamide for 5 hours at room temperature. The D₀- and D₃-labeled samples were then mixed in a 1:1 ratio: 10 mg of protein in the control sample (from 45 million cells) and 10 mg of protein in the senescent sample (from 18 million cells). Total cellular protein was measured in advance by protein assay (Quick Start Bradford assay; Bio-Rad). The mixed D₀/D₃-labeled sample was centrifuged at 105,000g for 1 hour, and the pellet was discarded while the supernatant was precipitated with chloroform/methanol and used for FPLC separation. 

Chloroform/Methanol Precipitation

Four sample volumes of methanol, 1 sample volume of chloroform, and 3 sample volumes of H₂O were added to the sample and briefly mixed. The following centrifugation at 14,000g for 10 minutes at room temperature caused separation into two phases with a precipitated protein layer on the interface. The upper phase was carefully discarded. The lower phase with protein layer was mixed with another four volumes of methanol. After centrifugation at 14,000g for 10 minutes at room temperature, the supernatant was discarded while the protein pellet was air dried for 5 minutes and kept at −20°C until further use. 

FPLC in the Presence of 4 M Urea

The precipitated D₀/D₃-labeled sample was dissolved in 2 mL of 20 mM citric acid (pH 3.0)/4 M urea and loaded onto a 1-mL column (HiTrap SP XL; GE Life Sciences, Piscataway, NJ). The proteins were separated by elution with a 0.0- to 0.7-M NaCl linear gradient in 20 mM citric acid (pH 3.0)/4 M urea on FPLC (AKTA; GE Life Sciences). Freshly deionized urea was made by passing 8 M urea through a resin (AG 501-X8 (D); Bio-Rad Laboratories) and was used to prepare all urea buffers. In addition, to prevent urea crystallization and excessive viscosity, all chromatographic operations were performed at room temperature. Collected fractions were precipitated with chloroform/methanol and used for 2D-PAGE separation. 

Two-Dimensional Polyacrylamide Gel Electrophoresis

The first dimension of separation was performed on 7-cm strips with a 3-10 immobilized pH gradient (Protean IEF cell; Bio-Rad Laboratories). The strips were rehydrated with 125 μL of a protein solution in 2 M thiourea, 7 M urea, 4% CHAPS, 0.2% (3-10) pH gradient (Biolytes; Bio-Rad), and bromophenol blue. IEF was conducted at 250 V for 15 minutes, linearly increased over 2 hours to a maximum of 4000 V, and then run to accumulate a total of 20,000 V/h. For the second dimension, the immobilized pH gradient strips were equilibrated for 15 minutes in 50 mM Tris-HCl (pH 8.8), 6 M urea, 30% glycerol, 2% SDS, and bromophenol blue. The strips were then embedded in 0.7% (wt/vol) agarose on the top of 10% polyacrylamide gels and proteins were separated by SDS-PAGE. 

In-gel Trypsinolysis

Staining of proteins in polyacrylamide gels was performed with a silver stain (SilverSNAP kit; Pierce). Silver-stained gel pieces were excised and destained in accordance with the manufacturer’s protocol. In-gel digestion was then performed with 1 to 4 μg/mL sequencing grade modified trypsin (Promega) in 25 mM ammonium bicarbonate (pH 7.9) for 15 hours at 37°C. After digestion, the peptides were extracted with 50% acetonitrile containing 0.1% TFA and dried. 

MS and Data Analysis

Dry peptide samples were dissolved in 5 mg/mL α-cyano-4-hydroxycinnamic acid in 40% acetonitrile containing 0.1% TFA and manually spotted onto a target plate (model 01-192-6-AB; Applied Biosystems, Inc. [ABI], Farmington, MA) and MS analysis was performed (AB4700 Proteomics Analyzer; ABI). MS-mode acquisitions consisted of 1000 laser shots averaged over 20 sample positions. For MS/MS-mode acquisitions, 3000 laser shots were averaged over 30 sample positions for postsource decay fragments. Automated combined acquisition of MS and MS/MS data was controlled with a system-associated software program (4000 Series Explorer software 3.0; ABI) and data analysis was performed (GPS Explorer software 3.5, with Mascot 2.0 search engine; MatrixScience, London, UK). During searching, the mass tolerance was 0.08 Da for the precursor ions and 0.2 Da for the fragment ions. A protein was listed as identified when the Mowse score ¹⁵ was higher than 63 or 64, the Mowse score at which statistical significance (P < 0.05) occurred for that particular search. 

Relative abundance of a protein in the senescent sample versus control sample was calculated as a D₃/D₀ ratio. In the MS spectrum, D₃-isotopic envelope for a peptide with one cysteine residues is shifted 3 Da relative to D₀-isotopic envelope. Depending on the mass of peptide, there might be an overlap between the D₀- and D₃-isotopic envelopes. In this case, the relative quantity was determined more accurately using peak area of the M+1 or M+2 isotopes. ¹³ As a general rule, the shorter peptides with no or minimal overlap of isotopic envelopes were selected for quantification and quantifications were made based on a single peptide per each particular protein. 

Results

Effect of Chronic Treatment with tert-BHP on Premature Senescence in ARPE-19 Cells

The prematurely senescent phenotype of tert-BHP-treated ARPE-19 cells has been recently proven by a variety of assays, ⁹ including increased cell surface area, SA β-gal staining, failure to respond to mitogens, and cell growth arrest in the G₁/G₂ phase of the cell cycle. We have reproduced the original protocol ⁹ with a minor change: 6 mM tert-BHP (instead of 8 mM tert-BHP) was used for treatment. To validate the senescent phenotype of tert-BHP-treated ARPE-19 cells, we have performed SA β-gal staining (Fig. 1A) . Cells were observed 3 days after replating and the population of SA β-gal-positive cells was determined by counting 200 cells per dish. The SA β-gal staining was approximately eight times greater in the tert-BHP-treated than in the control ARPE-19 cells (94% versus 12%). We did not directly measure the increase in cell surface, but tert-BHP-treated cells were visibly enlarged, flattened, and showed irregular morphology. We did measure total protein per 10,000 cells and the value for tert-BHP-treated cells was ≈2.5 times higher than the value for control ARPE-19 cells. Total protein measurements pointed to the enlarged senescent cells with increased cell surface. 

Using the D₃/D₀ ratio of acrylamide-labeled peptides, we also measured the relative abundance of many proteins in the tert-BHP-treated cells versus the control ARPE-19 cells. The full set of data will be published elsewhere. In the current report, we present only the data for 34 kDa active cathepsin D (Fig. 1B) . The Cys-containing peptide of active cathepsin D that was available for quantification was 18-amino-acids long. This causes partial overlap of the D₃ and D₀ isotopic envelopes used for quantification. In this case, the calculations were made based on the third peak in the isotopic envelopes. The mean ratio for three independent experiments with the same cell sample was D₃/D₀ = 1.82 ± 0.16. This implies that the protein level of active cathepsin D was almost twofold higher in the tert-BHP-treated cells versus the control ARPE-19 cells. This value concurs well with the published 1.7- to 1.8-fold increase of cathepsin D found in prematurely senescent ARPE-19 cells by human aging gene microarray (DualChip; Eppendorf Biochip System, Hamburg, Germany) and qRT-PCR. ⁹  

Taken together, the SA β-gal staining and increase in cathepsin D confirmed that chronic exposure to the tert-BHP produce a prematurely senescent phenotype in the ARPE-19 cells. ⁹  

Protein Fragment Accumulation in the Prematurely Senescent ARPE-19 Cells

To avoid the possibility of fragmentation of proteins during the cell processing leading to false positives, whole cells were directly dissolved in the buffer with 8 M urea. Control ARPE-19 cells were then treated with D₀-acrylamide, whereas prematurely senescent ARPE-19 cells were treated with D₃-acrylamide. Both samples were mixed at a 1:1 ratio based on total protein concentrations. Proteins of the combined sample were separated on the SP column by using a gradient of NaCl, and collected fractions were subjected to 2D-PAGE separation. The following MS-based identification of proteins revealed that some collected fractions were enriched with protein fragments more than others. These fractions were used for relative quantification. Figure 2shows representative 2D gels for the protein fractions eluted in the range of 0.30 to 0.35 M NaCl (Fig. 2A)and 0.65 to 0.70 M NaCl (Fig. 2B) . MS identification of proteins, and analysis of D₃/D₀ ratio of acrylamide-labeled peptides from individual spots, revealed an increased level of fragments for three proteins. These proteins are GAPDH (spots 1–10), triosephosphate isomerase (spots 11–13), and M2-type pyruvate kinase (spots 14–16). The quantitative data for all the spots are summarized in Table 1 . Table 1also includes parameters of protein identifications that validate spot assignments. 

The molecular mass of full-size proteins is 36 kDa for GAPDH (spot 1), 26.7 kDa for triosephosphate isomerase (spot 11), and 58 kDa for M2-type pyruvate kinase (spot 14). The molecular mass of fragments cannot be determined accurately from 2D-PAGE or MS analyses. This limits our ability to propose a potential mechanism of fragmentation and warrants a separate study focused on N- and C-terminal sequencing of observed fragments. The exact size of fragments would be valuable for assignment of protease(s) that can cause fragmentation. 

Figure 3shows representative quantifications for GAPDH based on D₀ and D₃ isotopic envelopes of ²³⁵VPTANVSVVDLTCR²⁴⁸ peptide. Spot 1 represents full-size GAPDH, which abundance in the prematurely senescent cells in comparison to control ARPE-19 cells was not changed (D₃/D₀ ratio = 0.97). Spot 7 (D₃/D₀ ratio = 1.57) and spot 3 (D₃/D₀ ratio = 2.85) represent fragments of GAPDH which abundance in the prematurely senescent cells in comparison to control ARPE-19 cells was increased 1.53- and 2.85-fold, respectively. It is important to emphasize that not all the GAPDH fragments shown in Figure 2were increased. Table 1shows that the abundance of a fragment in spot 6 was not changed (D₃/D₀ ratio = 0.99) and several others are on the border between increased and no changed (D₃/D₀ ratio ranging from 1.29 to 1.36). At the same time, the abundance of several other fragments was increased approximately two- to threefold (Table 1) . 

Figure 4shows representative quantifications for triosephosphate isomerase based on D₀ and D₃ isotopic envelopes of two different peptides from this protein, ⁵⁹IAVAAQNCYK⁶⁸ and ²⁰⁶IIYGGSVTGATCK.²¹⁸ Measurements for a specific protein based on more than one peptide demonstrate utility of the quantitative approach used. The D₃/D₀ ratios for two different peptides are similar and show a decrease for full-size protein (spot 11) and increase for a protein fragment (spot 13). 

Figure 5shows representative quantifications for M2-type pyruvate kinase. Based on D₀ and D₃ isotopic envelopes of ⁴⁴NTGIICTIGPASR⁵⁶ peptide from M2-type pyruvate kinase, the abundance of full-size protein (spot 14, D₃/D₀ ratio = 0.81) was slightly decreased in the prematurely senescent cells in comparison with control ARPE-19 cells. At the same time, the abundance of M2-type pyruvate kinase fragments (spots 15 and 16) was increased 2.00- and 2.90-fold, respectively. 

Discussion

RPE cells are long-lived postmitotic cells. Under normal conditions, these cells accumulate various biological waste materials at a rate that is inversely correlated to the normal lifespan of the organism. Under pathologic conditions, the rate of accumulation can be increased leading to functional decay and cell death. Typical examples of biological waste materials in the postmitotic cells are lipofuscin, irreversibly damaged mitochondria, and aberrant proteins. ¹⁶ In our study, prematurely senescent ARPE-19 cells tended also to accumulate large protein fragments. It is important to note that all these fragments are present in the control cells and presumably are a portion of the normal lifecycle of the ARPE-19 cells. However, the abundance of these fragments increased approximately two- to threefold in the prematurely senescent ARPE-19 cells. 

This observation raises two questions that merit investigation in future studies. The first question is whether these fragments may have a specific biological activity other than that of “waste” accumulation. GAPDH, triosephosphate isomerase, and M2-type pyruvate kinase are glycolytic enzymes that may possess different activities depending on their subcellular localization and conformational stability. For example, numerous studies have demonstrated the translocation of modified glycolytic enzymes in the nuclear compartments ^{17 18 19} and their potential role in regulation of cell growth and differentiation. Studies on GAPDH have demonstrated that GAPDH is conformationally and functionally altered under the conditions of oxidative stress ²⁰ and exerts various biological activities. ^{21 22} Nuclear localized M2-type pyruvate kinase has been shown to enhance cell proliferation. ¹⁹ Multiple electrophoretic and chromatographic forms have been reported for triosephosphate isomerase ²³ that presumably have different functions. Our observation is that many fragments found to be accumulated in the prematurely senescent ARPE-19 cells were large enough (Fig. 2)to be properly folded. These protein fragments may have been modified or may have shown new activity and compartmentalization in comparison to the full-size proteins. 

The second question is what was impaired in the normal process of fragmentation that caused fragments to accumulate in the prematurely senescent cells and whether the impairment correlates with senescent phenotype. GAPDH, triosephosphate isomerase, and M2-type pyruvate kinase are long-lived cytosolic proteins with KFERQ-like motifs and, like many others, long-lived cytosolic proteins are predisposed for chaperone-mediated autophagy in lysosomes. ^{24 25} Lysosomal function is sensitive to oxidative stress and aging, ^{1 2 3 4 26} and changes in lysosomes are likely to result in fragment accumulation in the prematurely senescent ARPE-19 cells. However, other cellular mechanisms responsible for protein degradation cannot be ruled out in assigning the mechanism of fragment accumulation observed in this study. 

In summary, we have demonstrated that prematurely senescent ARPE-19 cells can accumulate large fragments of typical cytosolic proteins, for which a mechanism of generation and a potential role in AMD remain to be established. 

 Certain commercial materials, instruments, and equipment are identified in the article, to specify the experimental procedure as completely as possible. In no case does such identification imply a recommendation or endorsement by the National Institute of Standards and Technology (NIST) nor does it imply that the materials, instruments, or equipment identified are necessarily the best available for the purpose.

 Submitted for publication March 5, 2009; revised May 6, 2009; accepted July 24, 2009.

 Disclosure: W.L. Liao, None; I.V. Turko, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Analysis of premature senescence of tert-BHP-treated ARPE-19 cells. (A) Ninety percent confluent ARPE-19 cells were treated with 6 mM tert-BHP for 1 hour. The cells were then washed twice with growth medium and allowed to recover for 24 hours. The tert-BHP treatment procedure was repeated five times for five consecutive days. For control cells, the growth medium was replaced in parallel according the same time schedule. After the last tert-BHP treatment, the cells were allowed to recover for 3 days and were then replated. Cellular senescence was assayed with SA β-gal staining kit according to the manufacturer’s protocol. (B) Relative quantification of cathepsin D. A representative MALDI mass spectrum shows that the level of cathepsin D was increased 1.82-fold in the tert-BHP-treated cells versus the control ARPE-19 cells.

Figure 2.

 

Silver-stained 2D-PAGE images of two eluted fractions. (A) A fraction eluted with 0.30 to 0.35 M NaCl. Blue numbers: GAPDH and its fragments; red numbers: triosephosphate isomerase and its fragments. (B) A fraction eluted with 0.65 to 0.70 M NaCl. Green numbers: M2-type pyruvate kinase and its fragments.

Table 1.

 

Identification and Quantification of Protein Fragments in the Numbered Silver-Stained Gel Spots in Figure 2

Spot and Protein	Peptide Matches	Peptides Validated by MS/MS	MOWSE Score*	D3/D0 Ratio, †
1. GAPDH	9	2	283	0.97 ± 0.07
2. GAPDH	6	2	192	2.14 ± 0.19
3. GAPDH	8	2	282	2.85 ± 0.27
4. GAPDH	12	2	410	2.11 ± 0.22
5. GAPDH	9	1	169	1.36 ± 0.11
6. GAPDH	4	2	154	0.99 ± 0.10
7. GAPDH	6	2	252	1.57 ± 0.12
8. GAPDH	4	1	112	1.30 ± 0.09
9. GAPDH	3	1	66	1.29 ± 0.11
10. GAPDH	7	3	294	1.87 ± 0.21
11. TPI	14	3	512	0.69 ± 0.11
12. TPI	14	3	463	3.03 ± 0.32
13. TPI	7	1	151	2.75 ± 0.30
14. Pyruvate kinase	22	2	392	0.81 ± 0.09
15. Pyruvate kinase	17	1	146	2.00 ± 0.25
16. Pyruvate kinase	18	1	171	2.90 ± 0.34

 TPI, triosephosphate isomerase.

*  The protein was listed as identified when the MOWSE score was higher than 63 or 64, the MOWSE score at which statistical significance (P < 0.05) occurred for that particular search.

†  Relative quantification was achieved based on the following peptides: ²³⁵V-R²⁴⁸ for GAPDH, ⁵⁹I-K⁶⁸ for TPI, and ⁴⁴N-R⁵⁶ for pyruvate kinase. Data are presented as the mean ± SD for three independent experiments with the same cell sample.

Figure 3.

 

Representative MALDI mass spectra for relative quantification of GAPDH and its fragments. Spot numbers correspond to those in Figure 2 .

Figure 4.

 

Representative MALDI mass spectra for relative quantification of triosephosphate isomerase and its fragments. Spot numbers correspond to those in Figure 2 .

Figure 5.

 

Representative MALDI mass spectra for relative quantification of M2-type pyruvate kinase and its fragments. Spot numbers correspond to those in Figure 2 .