Proteomic Characterization of Mouse Cytosolic ... - ACS Publications

In addition, free SUMO-1 levels are relatively abundant in the prostate. ... Our data suggest that the identified free SUMO peptides play an important...
0 downloads 0 Views 1MB Size
Proteomic Characterization of Mouse Cytosolic and Membrane Prostate Fractions: High Levels of Free SUMO Peptides Are Androgen-Regulated ´ ric Winstall,‡ Yutaka Inaguma,§,| Se´bastien Michaud,†,| Francine Lettre,| Danielle Caron,† E ‡ Sylvie Bourassa, Isabelle Kelly,‡ Guy G. Poirier,‡ Robert L. Faure,*,† and Robert M. Tanguay| Department of Pediatrics, Proteomic platform, CHUL Research Center, Que´bec G1V 4G2, Canada, Institute for Developmental Research, Aichi Human Service Center, Kasugai, Aichi 480-0392, Japan, and Laboratory of Cellular and Developmental Genetics, Department of Medicine and CREFSIP, Universite´ Laval, Que´bec G1K 7P4, Canada Received April 2, 2008

The prostate is a relatively homogeneous tissue that is highly specialized in synthetic and secretory functions. The frequency of malignant growth explains its great clinical significance. We used here a combination of subcellular fractionation, 1-DE (one-dimensional gel electrophoresis) protein separation and mass spectrometry, to establish a prostate protein expression profile in mice. Analysis of proteins present in cytosolic (C) and membrane (P) prostate fractions led to the identification of 619 distinct proteins. A majority of abundant proteins were found to compose the metabolism and protein synthesis machinery. Those identified also correspond to known endoplasmic reticulum and Golgi residents, chaperones and anterograde cargos. They included a series of proteins involved in exocytic/endocytic trafficking. Among the signaling proteins, we identified the ubiquitin-like peptides smt3. We showed that both free small ubiquitin-related modifier SUMO-2/3 and SUMO-1 levels are subject to tight control by the androgen 5R-dihydrotestosterone (DHT). By contrast with SUMO-2/3, free SUMO-1 peptides are particularly abundant in the prostate when compared with other tissues. Therefore, we report prostate protein expression profiles of cytosolic and membrane fractions in mice. Our data suggest that the identified free SUMO peptides play an important role in this secretory tissue. Keywords: Prostate • Proteome • Subcellular fractionation • SUMO

Introduction The prostate is a highly specialized gland whose sole known function is secretion of prostatic fluid.1 The frequency with which the gland becomes infected, hyperplasic, or the site of malignant growth explains its great clinical significance.2,3 One characteristic is its strict dependency upon androgenic hormones for its functional and structural integrity. The rapid rates of involution following castration, in combination with the equally rapid rate of reactivation of tissue growth upon androgen replacement, make this a useful and well-established experimental model.1 Rodent and human prostate tissues are relatively homogeneous. Both have a tubular alveolar topology consisting of epithelium-lined acini surrounded by a stromal matrix. Each acinus consists of a single layer4 of columnar epithelial cells resting on a basal lamina surrounded by smooth muscle cells. Highly specialized secretory prostatic cells are characterized * Address correspondence to: Robert L. Faure, Centre de Recherche du CHUL, Pediatrics Research Unit, 2705 Blvd. Laurier, RC-9800, Que´bec QC, G1V 4G2, Canada. Phone: (418) 654-2152. Fax: (418) 654-2753. E-mail: [email protected]. † Department of Pediatrics, CHUL Research Center. ‡ Proteomic platform, CHUL Research Center. § Institute for Developmental Research, Aichi Human Service Center. | Department of Medicine and CREFSIP, Universite´ Laval.

4492 Journal of Proteome Research 2008, 7, 4492–4499 Published on Web 08/27/2008

at the ultrastructural level by an extensive rough endoplasmic reticulum (RER) along with a developed Golgi apparatus. Most of the remaining intracellular space is packed with secretory granules. The membranes of these secretory granules fuse with the luminal plasma membrane for cargo secretion into the lumen.5,6 The secretory material in vacuoles consists of flocculent substances containing abundant specific proteins such as prostate-binding protein, prostate acid phosphatase and spermine binding protein as well as many enzyme systems that lead to the formation of citrate.1 Early observations have revealed that, although castration results in cessation of synthesis of specific proteins, the secretory organelles in the regressing prostate remain almost unaltered.4 Here, we use crude subcellular fractions to survey a proteome of the mouse prostate. Among the characteristics observed, our attention focuses on the small ubiquitin-related modifier (SUMO). Covalent attachment of SUMO to numerous target proteins is perceived as an important mechanism involved in cell cycle progression, nuclear import, protein transport, modulation of protein-protein interactions, degradation and the regulation of transcriptional activity.7-11 We demonstrate an androgen dependency of free SUMO levels, suggesting an important role of SUMOylation in prostate homeostasis. 10.1021/pr8002497 CCC: $40.75

 2008 American Chemical Society

Proteomic Characterization of Mouse Prostate Fractions

Experimental Section Animals. Male C57BL6 mice (12-15 week old, weight ) 25-30 g) supplied by Charles River Canada, Inc. (St. Constant, Canada) were maintained under standard laboratory conditions with food and water available ad libidum except that food was removed 18 h prior to organ collection. Work was conducted with the approval of Laval University Animal Care committee. For immunoblot analysis, mice were assigned to 4 groups of 12 animals each: (1) control, (2) GDX, (3) DHT 24 h, and (4) DHT 96 h groups. With the exception of the control group, mice were gonadectomized (GDX) via the scrotal route under isoflurane-induced anesthesia 7 days prior to organ collection. Mice assigned to the GDX group received a single subcutaneous injection of 0.2 mL of 0.4% methyl cellulose/5% ethanol (vehicle) 24 h prior to organ collection, while DHT mice received a single subcutaneous injection of DHT (5R-dihydrotestosterone at 0.1 mg/mouse, from Steraloids, Newport, RI) 24 and 96 h prior to organ collection. The prostates (ventral and dorsal lobes) were visually examined, collected, weighted, and processed immediately. Parallel protocols of histological studies confirmed that prostates were normal and exempt of infection. Subcellular Fractionation. For each of the four groups, 12 prostates were pooled and homogenized in ice-cold homogenization buffer (250 mM sucrose, 50 mM HEPES, pH 7.4, 3 mM benzamidine, 30 mM NaF, 2 mM EGTA, 10 mM pyrophosphate, 2 mM Na3VO4, 1 mM PMSF) using a Polytron homogenizer (Kinematica Switzerland; Probe PTA 7 K1, speed 5/10, 4 × 5 s). The homogenate was centrifuged 10 min at 3 300gav (rotor 50ti, Beckman Coulter, Inc., Fullerton, CA) and the sediment (N) discarded. The supernatant (T) was further centrifuged 10 min at 25 000gav to obtain the mitochondrial fraction (sediment M) and the resultant supernatant (SN) was centrifuged 35 min at 180 000gav to obtain the crude membrane fraction (sediment P) and the cytosolic fraction (supernatant C) (see flowchart in Figure 1A). The yield of each fraction was as follows: (C) control 41.1 ( 3.9, GDX 38.8 ( 5.5, DHT 24 h 40.2 ( 5.3, and DHT 96 h 33.1 ( 5.5 mg protein/g prostate (n ) 3, ( SD); (P) control 5.1 ( 1.1, GDX 6.1 ( 2.2, DHT 24 h 6.6 ( 1.5, and DHT 96 h 6.7 ( 1.2 mg protein/g prostate (n ) 3, ( SD). Protein content was measured using the Bradford microtechnique (#500-0006, Bio-Rad, Hercules, CA) in comparison with a BSA standard. Fractions were stored at -80 °C. Protein In-Gel Digestion. Proteins from control groups (1) were resolved by 12.5% SDS-PAGE (Bio-Rad, Hercules, CA). Protein visualization was made using SYPRO Ruby protein stain (Bio-Rad, Hercules, CA) and gel scans were obtained with ProExpress (PerkinElmer Life Science, Boston, MA). Bands of interest were extracted from gels using a “lanepicker” (The Gel Company, San Francisco, CA) and placed in 96-well plates and then washed with water. Tryptic digestion was performed on a MassPrep liquid handling robot (Micromass, Manchester, U.K.) according to specifications of the manufacturer and according to the protocol of Shevchenko et al.12 incorporating modifications suggested by Havlis et al.13 Briefly, proteins were reduced with 10 mM DTT and alkylated with 55 mM iodoacetamide. Trypsin digestion was performed using 105 mM of modified porcine trypsin (sequencing grade, Promega, Madison, WI) at 58 °C for 1 h. Digestion products were extracted using 1% formic acid/2% acetonitrile followed by 1% formic acid/50% acetonitrile. Recovered extracts were pooled, dried

research articles by vacuum centrifugation, resuspended into 10 µL of 0.1% formic acid, and analyzed by mass spectrometry. Mass Spectrometry (MS). Peptide samples were separated by online reversed-phase (RP) nanoscale capillary liquid chromatography (nanoLC) and analyzed by electrospray mass spectrometry (ES MS/MS). Analyses were performed with a Thermo Surveyor MS pump connected to an LCQ Deca XP mass spectrometer (Thermo Electron, San Jose, CA) supplied with a nanoelectrospray ion source (Thermo Electron, San Jose, CA). Peptides were bound on a cap trap (Michrom Bioresources, Auburn, CA) at 10 µL/min after which chromatographic separation took place using a PicoFrit column BioBasic C18, 10 cm × 0.075 mm internal diameter (New Objective, Woburn, MA) set with a linear gradient from 2 to 50% of solvent B (acetonitrile, 0.1% formic acid) for 30 min, at 200 nL/min (obtained by flow-splitting). Mass spectra were acquired using a data-dependent acquisition mode using Xcalibur software version 1.2. Each full scan mass spectra (400-2000 m/z) was followed by collision-induced dissociation of the three most intense ions. The dynamic exclusion (30 s exclusion duration) function was enabled, and the relative collisional fragmentation energy was set to 35%. Database Searching. All MS/MS samples were analyzed using Mascot (Matrix Science, London, U.K.; version 2.2.0) and X! Tandem (www.thegpm.org; version 2006.04.01.2). All data were converted automatically from raw instrument output to the Mascot Generic Format using the script LCQ_DTA.exe and Mascot Daemon (version 2.2.0) using default parameters. Mascot and X! Tandem were set up to search the Mus musculus Uniref100 database (version 8.0, 87 442 entries) assuming that the digestion enzyme was trypsin. Mascot and X! Tandem were searched with a fragment ion mass tolerance of 0.50 Da and a parent ion tolerance of 2.0 Da. An iodoacetamide derivative of cysteine was specified as fixed modification in both Mascot and X! Tandem. Deamidation of asparagine and glutamine, oxidation of methionine, and acetylation of the N-terminus were specified as variable modifications in X! Tandem. Oxidation of methionine was specified in Mascot as a variable modification. Two missed cleavages were allowed. Criteria for Protein Identification. Scaffold (version Scaffold-01-06-19, Proteome Software, Inc., Portland, OR) was used to validate MS/MS based peptide and protein identifications. Peptide identifications were accepted if they could be established as being greater than 95.0% probability as specified by the Peptide Prophet algorithm.14 Protein identifications were accepted if they could be established at greater than 95.0% probability and contained at least 1 identified peptides. Protein probabilities were assigned by the Protein Prophet algorithm.15 Proteins that contained similar peptides and that could not be differentiated by MS/MS analysis alone were grouped in order to satisfy the principles of parsimony. Plasmid Constructs and Antibodies. Human SUMO-1 cDNA was amplified by RT-PCR using primers, 5′-GTTCTGCTTACCCGAGGCCGCTGCTGTGC-3′ and 5′-AAAGAATATCTAAACTGTTGAATGACCCCC-3′, using total RNA from U373MG glioma cells as a template. The amplified fragment was ligated to pGEM-T easy vector (Promega, Madison, WI). The NcoI-SalI fragment, including a coding sequence, was transferred into pET-30a (+) (Novagen, Inc., Madison, WI). Histidine-tagged SUMO-1 was induced in Escherichia coli BL21 (DE3) and purified with HisTrap (GE Healthcare Bio-Science Corp., Piscataway, NJ) according to the manufacturer’s protocol. A polyclonal antibody against SUMO peptides was raised using Journal of Proteome Research • Vol. 7, No. 10, 2008 4493

research articles histidine-tagged SUMO-1 in rabbits. The antibody was affinitypurified using an antigen-coupled Sepharose column. This antibody recognizes SUMO-1 peptides but not ubiquitin.16 HeLa cells were transfected using standard calcium-phosphate precipitation methods with the pCEP4-SUMO vector, carrying the murine SUMO-1 open reading frame. After 24-48 h of expression, cellular lysates were solubilized in sample buffer. For Western blot analysis, samples were prepared in Laemmli sample buffer (Bio-Rad, Hercules, CA) and boiled for 5 min. Proteins were resolved by SDS-PAGE and transferred onto PVDF membranes overnight at 4 °C (150 mA, transfer buffer: 20% methanol, 50 mM Tris base, 192 mM glycine, 0.1% SDS). Membranes were verified for equal proteins loading. They were rinsed using TBS 0.05% Tween solution and blocked with a solution containing TBS 0.05% Tween and 5% nonfat dry-milk. Membranes were then incubated with SUMO-1 (dilution 1:1000) or SUMO-2/3 (#4974 Cell Signaling Technology, Danvers, MA, dilution 1:2000) in TBS 0.05% Tween 1% nonfat drymilk for 2 h, washed in TBS 0.05% Tween and further incubated using the appropriate horseradish peroxidase-conjugated secondary antibody (Jackson ImmunoResearch Laboratories, Westgrove, PA, dilution 1:10 000 in TBS 0.05% Tween 1% nonfat drymilk). Membranes were then washed in TBS 0.05% Tween solution, revealed using chemiluminescence (ECL, Chemiluminescence Reagent, PerkinElmer Life Science, Boston, MA), and exposed on Kodak film (X-Omat Blue XB-1, Kodak). Semiquantitative whole band analysis of each band utilized a Bio Image system (Bio Image System, Inc., Jackson, MI).

Results Proteome elucidation is an important step toward a better understanding of prostate function. Establishing protein profiles will also facilitate diagnosis of prostate diseases as well as highlight potential new therapeutic targets. Here, total homogenate (T), cytosolic (C), and crude membrane (P) subcellular fractions were subjected to 1-DE separation. This approach permits analysis of cytosolic and membrane proteins by removing nuclear and mitochondrial compartments. Fractions were characterized first in terms of gel protein staining (Figure 1). Marked differences were observed. Most noticeably, abundant proteins present in the 7-14 kDa region of the T and C fractions were not present in the P fraction. Also, several protein bands were distinctly enriched in the 10, 23, 35, 50, and 70 kDa gel regions of the P fraction. Still, other bands in the 29 kDa gel region were not present (Figure 1B). In order to evaluate both the sensitivity and reproducibility of the identification procedure, we analyzed the first 16 gel bands excised from the 10-30 kDa regions of the T fraction. A total of 43 proteins, with appropriate apparent molecular weight, were unambiguously identified (mean identification: 2.7 ( proteins per band). These proteins are part of the cell cytoskeleton (28%), protein translation process (12%), or metabolic processes (9%). Five percent of the proteins appeared in the signal transduction category including proteins involved in vesicular transport events, like the ADP ribosylation factors and proteins involved in ubiquitin-related mechanisms such as Ube2n (Table 1 and Supplementary Data Prostate Total Reports 95_95 in Supporting Information). These results support the argument that our methodology is efficient in systematically identifying the same proteins belonging to the categories of “translation and RNA processing” and “signal transduction”. Further fractionation was performed in order to increase the identifications efficiency by removing especially nuclear and cytosolic proteins. 4494

Journal of Proteome Research • Vol. 7, No. 10, 2008

Caron et al.

Figure 1. Cellular fractionation. (A) Diagram of the subcellular fractionation procedure. See Experimental Section for specific details. (B) A total of 5 µg of the control mouse total prostate and subcellular fractions was prepared with Laemmli buffer and separated by 12.5% SDS-PAGE. Proteins were visualized with SYPRO Ruby protein stain. T (total), C (cytosol), P (membrane).

Table 1. Summary Table of Protein Families Comparing the Distribution of Proteins Isolated from Mouse Subcellular Fractions by 1-DE Gela total extract* family

NB

Intracellular trafficking Protein folding and processing Translation and RNA processing Metabolism Signal transduction Cytoskeleton Stress response Others Unknown Total

2 3 5 4 2 12 6 8 1 43

%

cytosol NB

%

membrane NB

%

5 27 8 54 13 7 46 14 60 15 12 40 12 91 22 9 111 33 73 18 5 26 8 37 9 28 43 13 46 11 14 4 1 5 1 19 24 7 33 8 2 13 4 11 3 100 334 100 410 100

a Number of proteins and their distribution by family as categorized by their reported or predicted function. A total of 787 proteins were identified. Complete listings of all proteins appear in the Supporting Information (Prostate Total, Cytosol and Membrane Reports 95_95.xls). NB ) number of proteins. Asterisk (*): 10-30 kDa region.

A systematic analysis of 44 gel bands representing whole C and P lanes was then pursued. Analysis of fraction C yielded identification of 334 proteins (mean identification: 7.6 proteins per band). Of these, a majority are involved in metabolism (33%), protein folding and processing (14%), translation and RNA processing (12%), cytoskeleton (13%), intracellular trafficking (8%), and signal transduction (8%) (Table 1 and Supporting Information Prostate Cytosol Reports 95_95). The same analysis yielded 410 proteins from the P fraction, with a preponderance of proteins from the translation and RNA processing family (Table 1 and Supporting Information Prostate Membrane Reports 95_95; mean identification: 9.3 proteins per band). A total of 125 proteins were identified in both the C and the P fractions (Table 2 and Supporting Information Common Proteins C vs P fractions 95_95). Of these, 18% fell into the cytoskeleton category, 21% in the metabolism category, and 20% in the protein folding and processing category. Several proteins present in the metabolic category come from mitochondrial contamination (