Differential Protein Expression in Human Articular Chondrocytes Expanded in Serum-Free Media of Different Medium Osmolalities by DIGE Jane Koo,† Kang-Il Kim,‡ Byoung-Hyun Min,§ and Gyun Min Lee*,† Department of Biological Sciences and Graduate School of Nanoscience & Technology (WCU), KAIST, 335 Gwahangno, Yuseong-gu, Daejeon 305-701, Republic of Korea, Center for Joint Diseases and Rheumatism, Department of Orthopaedic Surgery, Kyung Hee University East-West Neo Medical Center, Sangil-dong 149, Gangdong-gu, Seoul, 134-727 Republic of Korea, Department of Molecular Science and Technology, College of Engineering, Department of Orthopaedic Surgery, Ajou University, School of Medicine, San 5, Woncheon-dong, Yeongtong-gu, Suwon 443-721, Republic of Korea Received December 16, 2009
Expansion of human articular chondrocytes (HACs) in serum-free medium (SFM) has been shown to be enhanced by the control of medium osmolality. HACs attained better growth when cultured at osmolalities lower than the average osmolality found in articular cartilage in vivo, although the specific causes for this improved expansion were not speculated. The aim of this study was to perform monolayer cultures of HACs in SFM at two different medium osmolalities, 320 and 400 mOsm/kg, and observe changes in protein content. A proteomics approach using differential in gel electrophoresis (DIGE) revealed differences in 20 spots. MALDI-TOF/TOF mass spectrometry identified 18 of the 20 spots. Proteins affected by higher osmolality in SFM were identified and found to function, traditionally, in the cytoskeleton, protection against reactive oxygen species, mRNA biogenesis, and the heat shock response. All of these proteins were correlated with the cell cycle and proliferation, emphasizing the importance of medium osmolality control in the expansion of HACs in SFM. In addition, no protein expression changes were observed regarding chondrogenic markers, which supports our previous result that adjusting medium osmolality in SFM, while enhancing growth, does not deter the tissue-forming capability of HACs. Keywords: human articular chondrocytes • serum-free medium • protein expression • DIGE
Introduction Human articular cartilage is a tissue composed of a single cell type, the chondrocyte, and an extracellular matrix (ECM) secreted by the cells themselves consisting of water, proteoglycans, and collagens. This tissue has a poor capacity for selfrepair due to its lack of access to the blood, leading to the development of cell therapy procedures such as autologous chondrocyte implantation (ACI). ACI involves the in vitro expansion of a patient’s own chondrocytes obtained by biopsy and subsequent injection back into the patient under a periosteal flap.1 For such procedures, a large number of cells must be available for injection, which is obtained by the use of serum during expansion cultures. However, the use of a serum-free medium (SFM) alleviates problems associated with serum such as limited availability and nonphysiological responses resulting in the cells. The SFM must be able to obtain * Corresponding Author: Gyun Min Lee. Department of Biological Sciences, KAIST, Gwahangno, Yuseong-gu, Daejeon 305-701, Republic of Korea. Phone: 82-42-350-2618. Fax: 82-42-350-2610. E-mail:
[email protected]. † Department of Biological Sciences and Graduate School of Nanoscience & Technology (WCU), KAIST. ‡ Kyung Hee University East-West Neo Medical Center. § Ajou University.
2480 Journal of Proteome Research 2010, 9, 2480–2487 Published on Web 03/23/2010
at least similar growth parameters as serum-containing media (SCM) to be useful. Medium osmolality is a key physical property of culture media that is defined as the concentration of particles in solution expressed as osmoles of solute per kilogram of solvent. The osmolality of media varies depending on various factors such as the concentrations of nutrients, the basal medium and the accumulation of byproducts during the culture itself. Changes in medium osmolality can lead to hypo or hyperosmolality, which are known to have detrimental effects on mammalian cells. For this reason, control of medium osmolality, which can be accomplished via the adjustment of salts, is essential. Studies regarding medium osmolality in the culture of human articular chondrocytes (HACs) are limited although there have been several performed with articular chondrocytes from other sources dealing with viability, apoptosis and effects on proteoglycans.2-6 In a previous study, we focused on the effects of medium osmolality in the culture of HACs with SFM, which had not been investigated before.7 Because of the advantages of SFM, it is important to explore medium osmolality in SFM for HACs, which we found to enhance expansion by control with sodium chloride (NaCl) adjustment. 10.1021/pr100136q
2010 American Chemical Society
SFM Osmolality and Chondrocyte Protein Expression Proteomics is a tool used in the study of cell proteomes and their differential expression. One of the most common techniques used in proteomics is two-dimensional polyacrylamide gel electrophoresis (2-D PAGE) in which proteins are first separated according to their isoelectric points and then their molecular weights.8 Protein spots of interest, which are visualized by staining, can be excised from gels, digested and then analyzed by mass spectrometry and database searching for identification. Differential in gel electrophoresis (DIGE) allows the analysis of two samples in a single gel, by labeling with different fluorescent dyes, and direct quantitative analysis of changes via use of an internal standard.9 Thus far, proteomics in human articular cartilage has been used primarily in the study of arthritic disease, although other investigations via proteomic analysis have been performed in immortalized human chondrocytes and human chondrosarcoma cells.10-18 In the present study, we utilized 2-D DIGE technology with mass spectrometry (MS) and tandem MS (MS/MS) in order to investigate proteome changes caused by different medium osmolalities of SFM in the expansion of HACs.
Experimental Section HAC Isolation. Human articular cartilage samples were received, after informed consent, from the knee joints of 12 patients suffering from osteoarthritis and requiring arthroplasty between the ages of 55 and 81 (mean age of 68.3 years). Cartilage was sliced from regions with no visual degeneration. Cartilage pieces were minced into small fragments, washed in PBS and digested overnight in a humidified 5% CO2 incubator at 37 °C with Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen, Carlsbad, CA) containing 0.1% collagenase type II (Worthington, Lakewood, NJ). Following incubation, undigested fragments were removed by passing the solution through a 70 µm cell strainer (BD Biosciences, San Jose, CA). The isolated HACs were then pooled, counted, and cryopreserved for subsequent seeding. Preparation of Expansion SFM of Various Osmolalities. Proprietary SFM at different osmolalities were made in our laboratory by the use of a customized DMEM (SAFC Biosciences, Lenexa, KS), which was modified to be absent of the NaCl normally formulated into the basal medium. SFM made in the customized DMEM without NaCl resulted in SFM with an osmolality of approximately 191 mOsm/kg. Amounts of 5 M NaCl were then added to this SFM to construct two SFM with osmolalities of 320 and 400 mOsm/kg, respectively. Osmolalities were measured using a Fiske 210 freezing-point osmometer (Fiske Associates, Norwood, MA). HAC Expansion. Cryopreserved, primary HACs were thawed and then washed 3 times with PBS to remove all traces of serum. The cells were counted, resuspended in the two SFM of different osmolalities and seeded at 5 × 103 cells/cm2 into BD BioCoat Fibronectin 75 cm2 Flasks (BD Biosciences, San Jose, CA) and cultured in a humidified 5% CO2 incubator at 37 °C. Media were exchanged two times per week. HACs were harvested at semiconfluency, while cells were undergoing proliferation, during the growth phase. Cells were also seeded into BD BioCoat Fibronectin 24-well Multiwell Plates (BD Biosciences) and cultured under the same conditions as above in order to confirm dissimilar HAC growth in the SFM of different osmolalities. The proliferative activity was evaluated for 28 days by sacrificing wells via trypsinization and cell counting with a hemacytometer every 2 days. Viable cells were distinguished from dead cells using the trypan blue
research articles dye exclusion method. Growth curves were drawn for HACs cultured for each osmolality. Each experimental point was assessed in triplicate and expressed as total viable cells/well. Protein Sample Preparation, 2-D DIGE, and Image Analysis. Expanded HACs (approximately 11 × 106 cells) were detached from the 75 cm2 culture flasks by trypsinization, counted, washed in PBS, collected by centrifugation, and stored at -70 °C. Protein sample preparation, 2-D DIGE, and image analysis were performed, subsequently, as described previously.19 Briefly, HACs expanded at 320 mOsm/kg were labeled with Cy3, those cultured at 400 mOms/kg with Cy5, and internal standards were labeled by with Cy2 by mixing 400 pM of CyDye (GE Healthcare, Waukesha, WI) with 50 µg of protein samples. Labeled samples were then quenched and total proteins (150 µg) were mixed and denatured in 2D sample buffer. The protein solutions were rehydrated on IPG strips (GE Healthcare) with isoelectric focusing (IEF) at 95 000 Vh. 9-17% linear gradient polyacrylamide gels were utilized for separation in the second dimension at 40 mA per gel until the BPB reached the bottom of the gel. Gels were stained with Coomassie Brilliant Blue G250. The CyDye labeled gels were visualized using a Typhoon 9400 imager (GE Healthcare) with excitation and emission wavelengths chosen specifically for each of the dyes according to the manufacturer’s recommendations. Images were preprocessed to remove areas extraneous to those of interest using ImageQuant V 2005 (GE Healthcare). DeCyder difference in-gel analysis (DIA) v6.5.11 was used for intragel analysis (GE Healthcare). Intergel matching and statistical analysis were performed with DeCyder biological variance analysis (BVA) v6.5.11 (GE Healthcare). Mass Spectrometry and Database Searching. Proteins of interest were excised from the gels and subjected to digestion with trypsin (Promega, Madison, WI) as previously described.20 A POROS R2, Oligo R3 column (Applied Biosystems, Fostercity, CA) was washed with 70% acetonitrile, 100% acetonitrile and then 50 mM ammonium bicarbonate, before applying the peptides resulting from tryptic digestion and eluting with cyano-4-hydroxycinamic acid (CHCA; Sigma, St. Louis, MO) dissolved in 70% acetonitrile and 0.1% TFA.21 Samples were analyzed by the use of a 4800 Proteomics Analyzer (Applied Biosystems) and GPS Explorer software (Applied Biosystems). For MALDI-TOF/TOF-MS analysis, MS spectra were acquired in reflectron mode using 1000 shots per spectrum. Instrument calibrations were performed externally using 4700 calibration mixture (Applied Biosystems) and internally by the use of trypsin autolysis peaks. The mass range used was 800-4000 m/z. Data Explorer 4.4 (Applied Biosystems) was used for data acquisition and extraction of the monoisotopic masses. MS/ MS analysis was performed by selecting 15 precursors with a minimum S/N ratio of 50. Fragmentation experiments were performed with the collision energy set at 1 kV and atmosphere as the collision gas. A total of 2000 acquisitions were accumulated for each MS/MS spectrum. MS and MS/MS spectra were used to search for protein candidates using the SwissProt and NCBInr human protein database with the Matrix Science search engine (http://www.matrixscience.com). Restriction criteria were iodoacetamide on cysteine and methionine oxidation. Mass tolerances of 0.5 and 0.2 Da were used for precursor and fragment ions, respectively. Journal of Proteome Research • Vol. 9, No. 5, 2010 2481
research articles
Koo et al. tially expressed, 3 were up-regulated and the remaining 17 were down-regulated. Spot Identification. Peptides resulting from the differentially expressed spots were analyzed by MALDI-TOF-MS and MS/ MS. Eighteen out of the 20 spots were successfully identified and are listed in Table 1. However, two of the 18 spots were identified as an unnamed protein product and a hypothetical protein, respectively.
Discussion
Figure 1. Growth curves of HAC expansion cultures in SFM at various osmolalities. HACs were expanded in SFM of 320 mOsm/ kg (O) and 400 mOsm/kg (b). Growth curves were constructed as described in the Experimental Section. Experimental points were accessed in triplicate and are expressed as means ( standard deviations. Differences between experimental groups were assessed via the Mann-Whitney test and were shown to be statistically significant (p-value
