Proteomic Analysis of Interstitial Fluid in Bone Marrow Identified That Peroxiredoxin 2 Regulates H2O2 Level of Bone Marrow during Aging Wei Wang, Lantu Gou, Gang Xie, Aiping Tong, Fei He, Zejun Lu, Yuqin Yao, Kang Liu, Jie Li, Minghai Tang, Lijuan Chen, Jinliang Yang,* Huozhen Hu,* and Yu-Quan Wei State Key Laboratory of Biotherapy, West China Hospital and College of Life Sciences, Sichuan University, Chengdu, China Received December 21, 2009
Hematopoiesis in bone marrow declines during aging owing to alteration of the hematopoietic niche. However, due to difficult accessibility and other complexities, senescence-related alteration of the hematopoietic niche is largely unknown. The interstitial fluid of bone marrow (IFBM), a pivotal component of the hematopoietic niche, includes soluble secretory factors that are present between bone marrow cells. To characterize the proteomic profile changes of IFBM during aging, we analyzed the IFBMs of young, adult, and senescent rats using 2-DE combined with ESI/MALDI-Q-TOF MS. Finally, 31 differentially expressed proteins involved in multiple biological functions were identified. Peroxiredoxin 2 (Prx2), down-regulated during aging, was further analyzed and demonstrated that it is produced by bone marrow stromal cells. Interestingly, higher levels of hydrogen peroxide (H2O2) were detected in the bone marrow with lower Prx2 expression. Moreover, exogenous Prx2 reduced the intracellular H2O2 level in bone marrow stromal cells in vitro. Therefore, Prx2 is implied in the regulation of H2O2 production in the bone marrow during aging. Our data characterized the dynamic protein profiles of the bone marrow microenvironment during aging and we provided clues to elucidate the mechanism of creating a low ROS level in the hematopoietic niche. Keywords: Aging • Proteomics • Hematopoiesis • Bone marrow • Peroxiredoxin 2 • H2O2 • Niche
1. Introduction Hematopoiesis produces highly differentiated blood cells developed from primitive, multipotential progenitors in a manner related to physiological requirements. This complex process requires the involvement of hematopoietic stem cells (HSCs) and the microenvironment, which precisely regulates the balance between HSC differentiation and quiescence. Disrupting the interaction between HSCs and the microenvironment, namely within the hematopoietic niche, affects hematopoiesis. During aging, reduced hematopoiesis occurs, such as decreased responses to hemorrhage.1 The balance of HSC differentiation also is upset,2 which results in a decline in immune function and increases the incidence of myelogenous disease.3 Previous studies have ascribed these functional alterations to the senescence of HSCs. The HSCs extracted from elderly patients have reduced repopulating abilities, homing, and mobilization.2,4 Many intrinsic factors, including chromosomal instability,5 DNA mutations,6 shortening of telomere and/or telomerase activities,7,8 and reactive oxygen species (ROS)induced protein damage,9 cause HSC senescence. Recently, * To whom correspondence should be addressed. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu 6l0041, China. Phone: +86-028-85503304. Fax: +86028-85164060: E-mail:
[email protected] (J.Y.);
[email protected].
3812 Journal of Proteome Research 2010, 9, 3812–3819 Published on Web 06/22/2010
numerous studies have investigated senescence-related alterations of the HSC extrinsic microenvironment. Previous data indicate that alterations of the hematopoietic niche during aging may be attributed to the modulation of HSC properties, such as quiescence, homing, repopulating, aging, and efficacy oftransplantation,which,inturn,directlyaffecthematopoiesis.2-4,10,11 The hematopoietic niche is composed of cells that support HSC function. Osteoblasts and endothelial cells support the two defined hematopoietic niche, endosteal niche and vascular niche, respectively.12 Other cells, including fibroblasts, adipocytes, and myocytes, also are involved in the creation of the hematopoietic niche architecture.12,13 In fact, these niche cells are not independent with each other, with other stromal cells and even HSCs.14 They may reciprocally affect their physiological function via secretory molecules, such as the extracellular matrix (ECM) and soluble cytokines. For example, the basement membrane-like extracellular matrix produced by endothelial cells maintains the proliferative and multilineage potential of mesenchymal stem cells, which reconstitute the hematopoietic microenvironment by differentiating into niche cells.13,15 Osteoblasts, which produce vascular endothelial growth factor (VEGF), may also promote endothelial cell proliferation. Therefore, analysis of cytokine expression in bone marrow cells may increase our understanding of the interactions between HSCs and the hematopoietic niche and the mechanism by which hematopoiesis is regulated in the bone marrow. 10.1021/pr901180w
2010 American Chemical Society
Proteomics Profiling of Aging-Induced IFBM Alterations The interstitial fluid of bone marrow (IFBM) includes all the extracellular cytokines between the cells of bone marrow and constitutes the external microenvironment of the bone marrow stromal cells. Previous studies have developed several methods to investigate the properties of these cytokines. Interestingly, the co-culture system has identified many factors that are essential for stromal cell and hematopoietic cell communication, such as IL-3 and stem cell factor (SCF).14 However, some inherent pitfalls of the co-culture system limit its application. This system simulates the intercourse that occurs within the hematopoietic niche, and only two or more cell types are involved in a single co-culture system. It is difficult to precisely reconstruct the actual microenvironment, which includes numerous cell types and secretory cytokines. During senescence, the composition of the IFBM may be altered. These changes may lead to a functional alteration of stromal and hematopoietic cells, which are responsible for a decline in hematopoiesis. To characterize changes in the expression of soluble factors in the IFBM, we performed a comparative analysis of the IFBMs proteome from young, adult, and senescent rats by 2-DE combined with ESI/MALDI-Q-TOF MS. We identified 31 proteins that had an altered expression. Further investigation of peroxiredoxin 2 (Prx2) indicated that Prx2 was produced by stromal cells and regulated the hydrogen peroxide (H2O2) level in the bone marrow during aging. Our results identified numerous proteins that deserve further analysis, and our strategy is an alternative method to investigate the complex niche network in the hematopoietic system.
2. Materials and Methods 2.1. Experimental Animals. The experimental procedures were performed in accordance with the National Drug Research Administration and with the permission of a local ethics committee. Young (2 weeks old, n ) 30), adult (12 weeks old, n ) 20), and senescent (90 weeks old, n ) 15) Sprague-Dawley (SD) rats (either sex) were purchased from the Experimental Animal Center at Sichuan University (Chengdu, China). 2.2. Sample Preparation. 2.2.1. IFBM Extraction. Total IFBM proteins from the bone marrow were carefully extracted as previously described16 with minor modification. After animals were anesthetized by an intraperitoneal injection of pentobarbital sodium (30 mg/kg), the femurs were dislocated and tissue attached to the surface of the bone was carefully cleared. One end of the femur was drilled and the opposite end was severed. Femurs (n ) 1-4) were transferred into a tube provided with nylon mesh (pore size 15-25 µm; generously bestowed by Helge Wiig of Bergen University) and then centrifuged (239g, 20 min) using an Allegra 64R centrifuge (Beckman Coulter, CA). Each centrifugation was performed using this centrifuge unless otherwise specified. To reduce the egress of cells from the bone marrow during centrifugation, the drilled end was placed downward and the cut end was upward. After centrifugation, the centrifugate was resuspended in 500 µL of phosphate buffered sodium (PBS) and centrifuged (239g, 5 min). Three volumes of acetone were added to the supernatant, and protein was precipitated overnight. Finally, the precipitated protein samples were collected together and centrifuged (20 000g, 4 °C) for 2 h. After acetone removal, protein was resuspended in rehydration lysis buffer or stored at -80 °C. 2.2.2. Protein Preparation for Western Blot Analysis. Both ends of the femurs (n ) 4) were severed, and the bone marrow was flushed out with 5 mL of PBS. Then, the bone marrow was
research articles vortexed gently and centrifuged (239g, 5 min). Erythrocytes were lysed by adding 2 mL of erythrocyte lysis buffer (139.6 mM NH4Cl, 16.96 mM Tris, pH 7.2). After removal of the supernatant, the remaining cells were lysed in 2 mL of RIPA buffer, and the supernatant was collected and indicated as bone marrow cells protein sample. To prepare the samples of red blood cells (RBCs) and serum, peripheral blood (2 mL) was collected by cardiac puncture. After coagulation, the serum was collected. Five milliliters of erythrocyte lysis buffer was added to the remaining cells and RBCs sample was prepared by collecting the supernatant after centrifugation (239g, 10 min) of the lysate. 2.3. Two-DimensionalElectrophoresis.IFBMproteinsamples from young, adult, and senescent rats were suspended in the rehydration lysis buffer (8 M urea, 2 M thiourea, 4% CHAPS, 50 mM dithiothreitol (DTT), 0.2% 3-10 ampholyte, and 0.001% bromophenol blue; Bio-Rad, Hercules, CA). The protein concentration was determined using the Bradford Protein Assay kit (Bio-Rad); bovine serum album was used as a standard. At first, the samples (2-5 mg) were diluted with small amount of rehydration lysis buffer. The protein concentration was adjusted by addition of rehydration lysis buffer to equilibrate the loading amount (1 or 2 mg) for each sample. Following a 16 h rehydration step using the pH 3-10 nonlinear IPG strip (BioRad), IEF was performed in a Protean IEF Cell (Bio-Rad) at 80 000 V/h. The strips were incubated sequentially in equilibration buffer I (50 mM Tris, 6 M urea, 30% glycerol, 2% SDS, and 20% DTT) and equilibration buffer II (50 mM Tris, 6 M urea, 30% glycerol, 2% SDS, and 25% iodoacetamide (IAA)) (Bio-Rad) for 15 min each. This was followed by SDS-PAGE on a 12% acrylamide gel. The gels were stained with Coomassie Brilliant Blue G-250 (Bio-Rad) and imaged using a GS-800 Scanner (BioRad). Image analysis was performed using PDQuest v. 7.1 software (Bio-Rad). The quantity of each spot in a gel was normalized as a percentage of the total quantity of all spots in that gel and evaluated in terms of OD. Only proteins that had consistent and significant differences (over (1.5-fold either in comparison of young vs adult or senescent vs adult) were selected for MS analysis. 2.4. In-Gel Trypsin Digestion. Proteins that were differentially expressed were excised and digested with Trypsin Gold (Promega, Madison, WI). The protein spots were destained in buffer I, which contains 50% acetonitrile (ACN) and 50 mM ammonium bicarbonate, for 2 h and then dehydrated in ACN for 1 h. Afterward, the protein spots were incubated in a trypsin solution (10 ng/µL) and digested for 16 h. One hundred microliters of the peptide extraction solution, which contains 5% trifluoroacetic acid (TFA) and 50% ACN, was added, and the samples were vibrated for 15 min. Peptides were collected and dried in a SPD131DDA SpeedVac (Thermo, Waltham, MA). 2.5. Protein Identification by ESI/MALDI-Q-TOF Mass Spectrometry. Proteins were identified by either ESI- or MALDI-Q-TOF MS. For ESI-Q-TOF MS identification, mass spectra were acquired using a Q-TOF mass spectrometer (Micromass, Manchester, U.K.) coupled with ESI ions (Micromass). The automatic scan rate was 1.0 s with an interscan delay of 0.02 s. Spectra were accumulated until a satisfactory signal/noise ratio was obtained. Only peaks that had two or more charges in a mass range from 400 to 1600 m/z were considered for MS/MS. Ions that exhibited a detection intensity exceeding 10 counts/s were selected for CID. A switch to the MS survey was made when either the duration of 10 s had elapsed or the ion intensity had fallen below 2 counts/s. MS/ Journal of Proteome Research • Vol. 9, No. 8, 2010 3813
research articles MS data was acquired with Masslynx v 4.1 (Micromass) using the default settings. Keratin-derived precursor ions and trypsin autolysis products were automatically excluded. Proteinlynx 2.2.5 software (Micomass) was used to convert the raw data to the peak list (pkl) files. MASCOT 2.2 software (Matrix Science) was used to search against the Swiss-Prot database (Swiss-Prot release of 57.0, 428 650 sequence entries), and the search parameters were set as follows: taxonomy, Rattus; enzyme, trypsin; mass tolerance, (1.2 Da; MS/MS tolerance, (0.6 Da; and an allowance of one missed cleavage. Fixed cysteine carbamidomethylation modifications, variable modifications of methionine oxidation (Met) and phosphorylation (ST) were allowed. The data format was set to Micromass (pkl), and the instrument was selected as ESI-QUAD-TOF. Proteins with probability-based MOWSE scores that exceeded their threshold (p < 0.05) were considered to be positively identified. Coupled with the MALDI ion source (Micromass), all settings of MALDIQ-TOF MS identification were the same as ESI-Q-TOF MS. The exceptions were that the tryptic digests were diluted in 10 µL of 50% ACN and then mixed with 10 µL of the MALDI-matrix solution (Micromass). In addition, a standard calibration peptide (Glu-fibrinopeptide) was used, and the peak lists were generated with Masslynx v. 4.1 software (Micromass) under default settings. The 10 most abundant peaks (mass > 800 m/z) were selected for MS/MS analysis, and the MALDI-QUAD-TOF was selected when performing the MASCOT database search. To eliminate the redundancy of proteins appearing in the database under different names or accession numbers, the one protein member with the highest MASCOT score and belonging to the Rattus species was further selected from the relevant multiple member protein family. 2.6. Western Blot Analysis. Prior to SDS-PAGE, protein concentration was measured by a Bradford assay using the Protein Assay kit (Bio-Rad). Bovine serum album was used as a standard. Proteins were separated on a 12% polyacrylamide gel and transferred to a PVDF membrane (Millipore, MA). Membranes were blocked in TBST (20 mM Tris, 0.14 M NaCl, 0.1% Tween 20, pH 7.6) containing 5% nonfat milk before overnight incubation with rabbit anti-Prx2 primary antibodies (Abcam, Cambridge, U.K.), which were diluted in 5% nonfat milk TBST (1:1000). Membranes were then incubated with goat anti-rabbit horseradish peroxidase (HRP)-conjugated secondary antibodies (Abcam) diluted in TBST at a concentration of 1:10 000 for 2 h at room temperature. The protein bands were visualized using the Chemiluminescent Kit (Promega). Images were scanned using a DS 800 Scanner (Bio-Rad), and band intensity was normalized and quantified using Quantity One software (Bio-Rad). 2.7. Detection of the Intracellular H2O2 Concentration in Bone Marrow Stromal Cells. The stroma of bone marrow was flushed gently from the femur with PBS, and the cells were resuspended in PBS. Erythrocytes were lysed in lysis buffer. After centrifugation (1500 rpm, 5 min), the supernatant was removed, and the cells were resuspended in PBS. Cells were counted using a hemocytometer. To evaluate the effects of the extracellular Prx2 expression on the intracellular H2O2 level, bone marrow stromal cells were incubated in low glucose Dulbecco’s Modified Eagle Media (Gibco, Langley, OK) containing L-glutamine and 10% fetal bovine serum (HyClone, Logan, UT). The media was renewed every 24 h in order to remove nonadhered cells. Three days later, the media was incubated with or without 50 µM Prx2 for 2 days (Prx2+ and Prx2-, respectively; USBiological, MA), and the intracellular 3814
Journal of Proteome Research • Vol. 9, No. 8, 2010
Wang et al.
Figure 1. (A) 2-DE analysis of the IFBMs from young, adult and senescent rat bone marrow. The vertical arrows indicate the molecular weight gradient and the horizontal arrows indicate the pH gradient. Thirty-one differentially expressed proteins were identified (arrow heads). The figure was one representative of three parallel experiments. (B) Three representative differentially expressed (DE) spots were enlarged to indicate the consistent expression variation in three independent experiments. Y, A, and S indicate young, adult, and senescent samples, respectively. The arrow heads indicate the corresponding DE spots on the gels.
H2O2 concentration was determined. To detect the H2O2 concentration in bone marrow stromal cells, 1 × 107 cells were lysed by repeated freeze-thaw. The H2O2 level in the lysate supernatant was determined using the Hydrogen Peroxide Detection Kit (Jiancheng Biotech, Nanjing, China) following the manufacturer’s instructions. 2.8. Statistical Analysis. Data were presented as the mean ( SD. To assess statistical significance of differences, Student’s t-test was performed. P-values
