Variations of Protein Levels in Human Amniotic Fluid Stem Cells

Sep 30, 2009 - Determination of specific stability markers, karyotyping, and ... karyotype, cell cycle, apoptosis rate, and 10 markers for characteriz...
0 downloads 0 Views 2MB Size
Variations of Protein Levels in Human Amniotic Fluid Stem Cells CD117/2 Over Passages 5-25 Wei-Qiang Chen,†,‡ Nicol Siegel,†,§ Lin Li,†,‡ Arnold Pollak,‡ Markus Hengstschla¨ger,§ and Gert Lubec*,‡ Department of Pediatrics, Medical University of Vienna, Wa¨hringer Gu ¨ rtel 18, 1090 Vienna, Austria, and Department of Medical Genetics, Medical University of Vienna, Wa¨hringer Gu ¨ rtel 18, 1090 Vienna, Austria Received July 17, 2009

Stability of cell lines is the prerequisite for all in vitro research, but literature on the stability of protein expression over passages is limited. Determination of specific stability markers, karyotyping, and morphology may not provide full information on this subject. It was the aim of the study to test protein level fluctuations in a human amniotic fluid stem cell line from passages 5, 7, 11, and 25. While karyotype, cell cycle, apoptosis rate, and 10 markers for characterization of the cell line remained unchanged (carried out at passages 5 and 25), cell volume was increased at passage 25. Significant protein fluctuations were observed for signaling, antioxidant, guidance cue, proteasomal, connective tissue, cytoskeleton proteins, chaperones, a chloride channel, and prothymosin at passages 5, 7, 11, and 25. Herein, the use of this gel-based proteomic screen, checking protein stability for the characterization of cell lines in addition to corresponding published markers, is proposed, in particular when experiments are run over several passages. Keywords: adult stem cells • passages of cell lines • protein expression • 2D gel electrophoresis

Introduction Stability of cell lines is a prerequisite for in vitro studies, and stability of protein expression is a major determinant. While morphological, functional criteria, karyotyping, and specific markers for the characterization of cell lines are widely used, information on protein stability is anecdotal and limited to determination of some individual proteins.1,2 Moreover, most studies addressing stability of individual protein levels over cell passages were carried out in tumor cell lines and systematic study on protein stability in nontumor cells is limited.3,4 A screen for protein stability to characterize a cell line as well as screens for protein stability over several passages in addition to the specific markers used, would be a major step forward for the fair proteomic characterization of cell lines forming the basis for comparability of results and representing quality control. Zhang and co-workers used a gel-based proteomic approach to evaluate stability of several lung adenocarcinoma cell lines over several cell passages to select a stable cell line for further in vitro studies,5,6 and it was the aim of the current study to introduce a robust comparable method,7-10 two-dimensional gel electrophoresis with subsequent quantification of proteins followed by mass spectrometrical unambiguous protein identification, for evaluation of protein stability in human amniotic * To whom correspondence should be addressed. Univ.Prof. Dr. Gert Lubec, Medical University of Vienna, Department of Pediatrics, Wa¨hringer Gu ¨ rtel 18-20, A-1090 Vienna, Austria. Tel.: +43-1-40400-3215. Fax: +43-140400-6065. E-mail: [email protected]. † These authors contributed equally. ‡ Department of Pediatrics. § Department of Medical Genetics. 10.1021/pr900630s CCC: $40.75

 2009 American Chemical Society

fluid stem cells at passages 5, 7, 11, and 25. This technique was already suitable for identification of proteins involved in cell differentiation11 and for identification of mitosis-related proteins.12 The human amniotic fluid stem cell line CD117/2 was selected because of its potential importance for clinical use in stem cell therapy,13 and it was intended not to examine any tumor cell line to avoid introduction of factors generated by malignancy per se. The method proposed herein comprehensively determines protein expression in cell lines for the characterization and follow up of a given cell line. Results of this method reveal information about possible derangements of a large series of protein pathways and cascades that may well be respected when in vitro studies are designed. In addition, potential information about posttranslational modifications of proteins may be obtained, a subject currently studied in our laboratory. And indeed, significant protein fluctuations along passaging were observed for proteins involved in signaling, antioxidant response, protein synthesis, chaperoning and degradation, connective tissue, and cytoskeleton architecture.

Experimental Section Cell Line Characteristics. Establishment of an Amniotic Fluid Stem Cell Line. The human amniotic fluid cell sample was obtained from amniocentesis performed for routine prenatal diagnosis. This project has been reviewed and accepted by the ethics committee of the University of Vienna, Austria (project number: 036/2002) warranting written informed patient’s consent. Isolation of human amniotic fluid stem cells expressing CD117 antigen was performed via magnetic cell sorting using Journal of Proteome Research 2009, 8, 5285–5295 5285 Published on Web 09/30/2009

research articles

Chen et al.

Table 1. Markers Analysed in This Study

Marker for pluripotent stem cells Mesenchymal marker Epithelial marker Hematopoietic stem cell marker Marker for podocytes Smooth muscle cell marker a

abbreviation

full name

accession numbera

references

pos control

Oct4 CD117 CD29 CD44 CD51 ZO-1 CD133 CD34 CD2AP Cdh5

octamer-binding transcription factor 4 c-kit integrin beta-1 hermes antigen integrin alpha-V zona occludens 1 prominin 1 hematopoietic progenitor cell antigen CD34 CD2-associated protein vascular endothelial cadherin

NM_203289 NM_000222.2 NM_002211.2 NM_000610.3 NM_002210 NM_003257.3 NM_006017.2 NM_001773.2 NM_012120.2 NM_001795.2

48 49 50 51 52 53 53 54 53

Jurkat CaoV3 CaoV3 K562 kidney extract MCF7 Hela Jurkat MCF7 CaoV3

Taken from NCBI (http://www.ncbi.nlm.nih.gov/sites/entrez?db)Nucleotide).

the CD117 MicroBead Kit (Miltenyi Biotec, Bergisch Gladbach, Germany). The amniotic fluid stem cell line used, named CD117/2, was established by cell cultivation of a positively selected and eluted cell fraction.14 The various passages of the human amniotic fluid stem cell line CD117/2 were cytogenetically analyzed according to standard protocols. Chromosome banding was produced by means of a conventional 550-band trypsin-Giemsa analysis and a significant number of metaphases were analyzed for numerical and structural chromosome aberrations.15 Cell Cultivation. Human amniotic fluid stem cells were grown in alpha-MEM minimal essential medium (Gibco, Invitrogen, Carlsbad, CA), supplemented with 15% ESFBS (PAA, Pasching, Austria), 2.5 mM L-glutamine, antibiotics (30 mg/L penicillin, 50 mg/L streptomycin sulfate, Sigma, St. Louis, MI), 18% Chang B and 2% Chang C (Irvine Scientific, Santa Ana, CA, USA), at 37 °C and 5% CO2. Cell lines used as positive controls were grown in either DMEM-LG (Gibco, Invitrogen, Carlsbad, CA) (Jurkat, MCF7, Hela, CaoV3) or in RPMI 1640 medium (Gibco, Invitrogen, Carlsbad, CA) (K562), both supplemented with 10% fetal bovine serum (Gibco, Invitrogen, Carlsbad, CA) and antibiotics (60 mg/L penicillin, 100 mg/L streptomycin sulfate; Sigma, St. Louis, MI). All cultures were incubated at 5% CO2 and 37 °C. Photographs were taken using a magnification of 40× on an inverse Olympus IMT-2 microscope using the ColorView III soft imaging system (Olympus, Tokyo, Japan). Cells Used as Positive Controls. For each gene wellcharacterized cell types were used as positive controls due to specific mRNA expression levels, detected via RT-PCR (see below). Jurkat is a human T cell leukemia cell line. CaoV3 and MCF7 are derived from ovary and breast adenocarcinomas. The K562 cell line is from human chronic myelogenous leukemia cells and the HeLa cell line is from human cervix carcinoma. All cell lines were obtained from the American type Culture Collection. As another positive control for the expression of various genes the commercially available human kidney cDNA (AM3331, Applied Biosystems, Darmstadt, Germany) was used (positive controls see Table 1). CASY and Flow Cytometry. Cells were harvested by gentle trypsinisation. A subset of the cell suspension was used to analyze the cell volume with the CASY 1 cell counter according to the manufacturers’ instructions (Scha¨rfe System, Reutlingen, Germany). To determine cell cycle distribution, DNA of remaining cells was stained with 0.25 mg/mL propidium iodide, 0.05 mg/mL RNase and 0.1% Triton X-100 in 4 mM citrate buffer, pH 7.8. Cells were collected in the linear amplification 5286

Journal of Proteome Research • Vol. 8, No. 11, 2009

mode and analyzed on a Beckton Dickinson FACScan using CellQuest and ModFit software. The cell size distribution according to the different cell cycle phases was determined using a 2-D contour plot of FSC versus DNA-content. Contour plots, representing a density-related presentation of two parameters in relation to cell number, were generated in a probability density mode at 20%. The outermost contour line represents 10% of the total number of events, the next one 30%, then 50, 70, and 90%.16 The sub G1 fraction of the cells, representing apoptotic cells, was cytofluorometrically analyzed via one-dimensional DNA-content-histograms.17 RT-PCR Analyses. RNA of the individual CD117/2 passages (5, 7, 11, 25) was prepared using the Cell-to-cDNA-II kit (Ambion, Applied Biosystems, Darmstadt, Germany) according to the manufacturers’ instructions. The RNA was checked for DNA contamination using GAPDH-primers specific for DNA recognition. Exclusively non-DNA-contaminated samples were transcribed into cDNA following the Cell-to-cDNA-II kit (Ambion, Applied Biosystems, Darmstadt, Germany) manufacturers’ instructions. The PCR reactions were prepared to contain 0.2 mM dNTPs and 1.25 U GoTaq-Flexi-DNA polymerase in the according reaction buffer (Promega, Madison, WI). Two micromolar of each primer was added to the reaction. The sequences and annealing temperature for each primer pair are presented in Supplementary Table 1, Supporting Information. The PCR reaction was composed of an initial denaturation step (2 min/ 95 °C), followed by 40 PCR cycles, consisting of denaturation (30 s/95 °C), annealing (30 s at temperatures given in Supplementary Table 1, Supporting Information) and elongation (90 s/72 °C). Finally, a last elongation step (7 min/ 72 °C) was performed. Statistical Analysis. The groups were normally distributed, therefore independent Student’s t test was used for statistical analysis. In all instances, a probability level of P < 0.05 was considered as statistically significant. All calculations were performed using SPSS version 17.0 (SPSS Inc., Chicago, IL). Protein Studies. Sample Preparation. Cell pellets were homogenized and suspended in 1.8 mL sample buffer (20 mM Tris, 7 M urea, 2 M thiourea, 4% w/v CHAPS, 10 mM 1,4dithioerythritol, 1 mM EDTA, 1 mM PMSF, 1 tablet CompleteTM from Roche Diagnostics, Risch, Switzerland, and 0.2% v/v phosphatase inhibitor cocktail from Calbiochem, Merck, Darmstadt, Germany). The suspension was sonicated on ice for approximately 30 s and centrifuged at 14 000× g for 60 min at 12 °C. Desalting was carried out with an Ultrafree-4 centrifugal filter unit at a cutoff molecular weight of 10 000 Da (Millipore, Bedford, MA) at 4400× g at 12 °C until the eluted volume was about 4 mL and the remaining volume reached 100-200 µL.18

Variations of Protein Levels in Human Amniotic Fluid Stem Cells The protein content of the supernatant was determined by the Bradford assay.19 Two-Dimensional Gel Electrophoresis (2-DE). 2-DE was performed essentially as reported previously.20 Samples of 700 µg protein (6 gels per group, total 24 gels) were subjected to immobilized pH 3-10 nonlinear gradient strips (18 cm length). Focusing started at 200 V and the voltage was gradually increased to 8000 at 4 V/min and kept constant for further 3 h (approximately 150 000 Vh totally). Prior to the second dimensional run strips were equilibrated twice for 15 min with gentle shaking in 10 mL of SDS equilibration buffer (50 mM pH 8.8 Tris-HCl, 6 M urea, 30% v/v glycerol, 2% w/v SDS, trace of bromophenol blue). DTT (1%) w/v was added at the first incubation for 15 min and 4% iodoacetamide w/v instead of DTT at the second incubation step for 15 min. The seconddimensional separation was performed on 10-16% gradient SDS-PAGE. After protein fixation for 12 h in 50% methanol and 10% acetic acid, the gels were stained with colloidal Coomassie blue (Novex, San Diego, CA) for 8 h and excess of dye was washed out from the gels with distilled water. Molecular masses were determined by running precision protein standard markers (Bio-Rad Laboratories, Hercules, CA), covering the range of 10-250 kDa. Isoelectric point values were determined as given by the supplier (GE Healthcare, Buckinghamshire, UK) of the immobilized pH gradient strips. Quantification of Protein Levels. Protein spots from each gel from passages 5, 7, 11, and 25 (6 per group; total n ) 24) were outlined (first automatically and then manually) and quantified using the Proteomweaver software (Definiens, Munich, Germany). The percentage of the volume of the spots representing a certain protein was determined in comparison with the total proteins present in the 2-DE gel.21 Analysis of Peptides by Nano-LC-ESI-(CID/ETD)-MS/MS (High capacity ion trap, HCT). Twenty-three spots which showed different levels between passage groups were manually excised and placed into 0.5 mL lobind Eppendorf tubes. In-gel digestion and sample preparation for HCT analysis was performed as described before.22 Gel plugs were washed with 10 mM ammonium bicarbonate and subsequently with 50% acetonitrile in 10 mM ammonium bicarbonate, repeatedly. Addition of 100% acetonitrile resulted in gel shrinking and the shrunk gel plugs were then speedVac dried in a Speedvac Concentrator 5301 (Eppendorf, Hamburg, Germany). The dried gel pieces were reswollen, in-gel digested with 40 ng/µL trypsin (Promega, Madison, WI) in digestion buffer (consisting of 5 mM Octyl β-D-glucopyranoside (OGP) and 10 mM ammonium bicarbonate) and incubated overnight at 37 °C. Peptide extraction was performed with 20 µL of 1% TFA in 5 mM OGP for 30 min, and subsequently 0.1% TFA in 4% acetonitrile for 30 min. The extracted peptides were pooled for HCT analysis. Forty microliters of extracted peptides were analyzed by HCT. The HPLC used was a biocompatible Ultimate 3000 system (Dionex Corporation, Sunnyvale, CA, USA) equipped with a PepMap100 C-18 trap column (300 µm × 5 mm) and PepMap100 C-18 analytic column (75 µm × 150 mm). Peptides were first loaded to the trap column and after washing were run on the separation column. The gradient in the separation column was (A ) 0.1% formic acid in water, B ) 0.08% formic acid in acetonitrile) 4-30% B from 0 to 105 min, 80% B from 105 to 110 min, 4% B from 110 to 125 min. The flow rate was 300 nL/min from 0 to 12 min, 75 nL/min from 12 to 105 min, 300 nL/min from 105 to 125 min. A HCT ultra ETDII PTM discovery system (Bruker Daltonics, Bremen, Germany) was

research articles

used to record peptide spectra over the mass range of m/z 350-1500, and MS/MS spectra in information- dependent data acquisition over the mass range of m/z 100-2800. Repeatedly, MS spectra were recorded followed by three data-dependent CID MS/MS spectra and three ETD MS/MS spectra generated from three highest intensity precursor ions. An active exclusion of 0.4 min after two spectra was used to detect low abundant peptides. The voltage between ion spray tip and spray shield was set to 1100 V. Drying nitrogen gas was heated to 170 °C and the flow rate was 10 L/min. The collision energy was set automatically according to the mass and charge state of the peptides chosen for fragmentation. Multiple charged peptides were chosen for MS/MS experiments due to their good fragmentation characteristics. MS/MS spectra were interpreted and peak lists were generated by DataAnalysis 4.0 (Bruker Daltonics, Bremen, Germany). Searches were done by using the MASCOT 2.2.04 (Matrix Science, London, UK) against latest UniProtKB database for protein identification. Searching parameters were set as follows: enzyme selected as trypsin with two maximum missing cleavage sites, species limited to human, a mass tolerance of 0.2 Da for peptide tolerance, 0.2 Da for MS/MS tolerance, fixed modification of carbamidomethyl(C) and variable modification of methionine oxidation and phosphorylation (Tyr, Thr, and Ser). Positive protein identifications were based on a significant MOWSE score. After protein identification, an error-tolerant search was done to detect unspecific cleavage and unassigned modifications. Protein identification and modification information returned from MASCOT were manually inspected and filteredtoobtainconfirmedproteinidentificationandmodifications. The Modiro software computed enzymes “selected as used”, with three maximum missing cleavage sites, species limited to human. Peptide mass tolerance was 0.2 Da for peptide tolerance, 0.2 Da for fragment mass tolerance, selecting modification 1 as carbamidomethyl and modification 2 of methionine oxidation. Protein identification was first of all listed by inspection of spectra and subsequently significant peptide identification was based upon the ion-charge status of the peptide, b- and y-ion fragmentation quality, ion score >200 and a significance score >80 as suggested by the manufacturer’s manual (http://www.protagen.com/customers_downloads/ MAN_Modiro_v1.1.zip). Searches for unknown mass shifts, for amino acid substitution and calculation of significance were selected on advanced PTM explorer search strategies (http://www.protagen.com/ customers_downloads/MAN_Modiro-Advanced-Search-Strategies.pdf). Western Blotting. Sample aliquots from 2-DE containing 10 µg of protein were loaded onto 12.5% ExcelGel SDS) homogeneous gels (GE Healthcare, Buckinghamshire, UK). Electrophoresis was performed with Multiphor II Electrophoresis System (Amersham Pharmacia Biotech, Uppsala, Sweden) and transfer to membranes was performed as described recently.23 Primary antibodies used were as follows: rabbit antibody against cofilin-1 (Cell Signaling, Danvers, MA, #3312, in a dilution of 1:1000); rabbit antibody against stathmin (Cell Signaling, Danvers, MA, #3352, in a dilution of 1:1000); rabbit polyclonal antibody against collagen type III (Abcam, Cambridge, UK, ab24137, in a dilution of 1:2000); rabbit polyclonal antibody against hnRNP H (Abcam, Cambridge, UK, ab10374, in a dilution of 1:5000); rabbit polyclonal antibody against CLIC1 (Abcam, Cambridge, UK, ab28722, in a dilution of 1:2000); rabbit polyclonal antibody against TARCBP (Abcam, Journal of Proteome Research • Vol. 8, No. 11, 2009 5287

research articles

Chen et al.

Figure 1. (A and B) Characterization of CD117/2 positive cells, Characterization of CD117/2 cells of passages (A) 5 and (B) 25. Cells were photographed with a magnification of 40× on an inverse Olympus IMT-2 microscope. The Chromosome banding was produced by means of a conventional 550-band trypsin-Giemsa analysis. The doubling time (26 h) was estimated by growth curve analysis. The karyotype was 46, XY at passages 5 and 25. (C and D) Cell volume of logarithmically growing CD117/2 cells. (C) Cell cycle distribution of CD117/2 cells of passage 5 and 25 was determined by FACS analysis. Logarithmically growing cells were analyzed for their cell cycle distribution using the program ModFit. (*P < 0.05, **P < 0.005, n.s. P < 0.5). (D) Logarithmically growing cell samples of passage 5 and 25 were analyzed for their cell volume [fl] using the CASY 1 cell counter (***P < 0.0005). (E, F and G) Relative cell size of CD117/2 cells Passage 5 and 25 of CD117/2 cells of a comparable DNA distribution were analyzed for their relative cell size and DNA content using FACS analysis. (E) Cell size according to different cell cycle phases was examined via a 2-D contour blot analysis of FSC versus DNA content. (F) DNA profiles of the passages 5 and 25 of CD117/2 cells are presented. (G) For the same samples the overall cell size was investigated via FSC analysis. (H) Sub G1 cell fraction of CD117/2 cells H Passage 5 and Passage 25 of CD117/2 cells were analyzed for their sub G1 fraction, representing apoptotic cells, by comparing the amount of their respective sub G1 cell fraction via FACS analysis (n.s. P < 0.5).

Cambridge, UK, ab41881, in a dilution of 1:1000). Secondary antibodies used were antirabbit IgG, HRP linked (Cell Signaling, 5288

Journal of Proteome Research • Vol. 8, No. 11, 2009

Danvers, MA, #7074) and antimouse IgG, HRP-linked (Cell Signaling, Danvers, MA, #7076). Both antibodies were used in

Variations of Protein Levels in Human Amniotic Fluid Stem Cells

research articles

the same dilution as primary antibodies. Controls were run without primary antibodies. Membranes were developed with the Amersham ECL Plus Western Blotting Detection System (GE Healthcare, Buckinghamshire, USA). Densities of immunoreactive bands were measured by Image J software program (http://rsb.info.nih.gov/ ij/). Statistical Analysis. Statistical analysis to reveal betweengroup differences was performed by ANOVA followed by unpaired Student’s t test. Bonferroni-Holm correction was applied for correction of multiple testing. If data violated a principal assumption of a parametric distribution nonparametric Mann-Whitney-U-test was carried out. In all proteomic studies, a probability level of P < 0.001 was considered as statistically significant. All calculations were performed using SPSS version 14.0 (SPSS Inc., Chicago, IL).

Results As shown in Figure 1 no morphological differences between passages 5 (Figure 1A) and 25 (Figure 1B) were observed and doubling time was 26 h at both passages. Chromosome banding revealed identical karyotypes. Cell volume of logarithmically growing CD117/2 cells was determined by Casy measurement analysis. The cell cycle distribution was determined by FACS analysis. The results of the analyses are presented in Figure 1C and D. There was no significant difference in the percentage of G0/ G1 cells between the passages. However, at passage 25 there was a significantly higher percentage of cells in the S phase and a significantly lower percentage in the G2/M phase as compared to passage 5. Cell volume, expressed as femtoliters, at passage 25 was significantly higher than in passage 5. Because a difference in the cell cycle distribution as well as in the cell volume between the passages was observed, evidence was needed on whether the difference in the cell volume was a real physical trait of cells, or simply due to shifted cell cycle distribution. Two-dimensional contour blot analysis (Figure 1E) of relative DNA content (Figure 1F) versus Forward Scatter (FSC, Figure 1G) of the passages 5 and 25 revealed that the observed difference in cell size and volume is not a consequence of the different cell cycle distributions between the passages. Figure 1E clearly shows that the G0/G1 cells of passage 25 are larger than the cells in G0/G1 of passage 5. There was no difference between the amount of apoptotic cells in passage 5 and 25 as shown by the sub G1 cell fraction in Figure 1H. As given in Table 1 and Figure 2, the expression of a series of markers for pluripotent stem cells, mesenchymal cells, epithelial cells, hematopoietic stem cells, podocytes and smooth muscle cell markers, was compared between passage 5 and 25. mRNA levels of Oct4, CD117, CD29, CD44, CD51, ZO-1, CD2AP, and CD34 were expressed in both passages. There was no expression of CD133 and Cdh5 observed in the human amniotic fluid stem cells. All observed genes were either expressed in both or in none of the passages. Therefore we can conclude that there is no difference in the expression status of the analyzed marker genes in the passages analyzed. Protein Studies. A representative two-dimensional gel identifying proteins with differential levels at passages 5 and 25 along with their corresponding UniProtKB/Swissprot numbers is shown in Figure 3.

Figure 2. Gene expression analyses in CD117/2 cells. Passage 5 and Passage 25 of CD117/2 were analyzed for the expression of the indicated genes via RT-PCR analysis. Representative examples are shown for each gene. For the negative controls water was used instead of cDNA. For each gene we have characterized cells, which can be used as positive controls due to a significantly detectable specific mRNA expression. The used positive control cells are indicated in Table 1. For a more detailed description of these cells, see materials and methods. The * indicates primer dimers. The arrow indicates the specific band.

Figure 3. Master gel showing the map of significantly changed 23 spot volumes from cell passage samples. UniProtKB/Swissprot accession numbers are provided.

Quantification results of 23 protein spots with significantly different levels between passages 5 and 25 are listed in Table 2. Identification of these 23 protein spots along with their matched peptide numbers, sequence coverage and identification by Mascot and Modiro software including ion scores/mass errors and MS/MS peptides determined are listed in Supplementary Table 2, Supporting Information. Posttranslational modifications as well as protein modifications probably repJournal of Proteome Research • Vol. 8, No. 11, 2009 5289

5290

Journal of Proteome Research • Vol. 8, No. 11, 2009

Heat shock 70 kDa protein 4 Ankycorbin

P34932 Q9P0K7

0.0001

1.169 ( 0.158 0.390 ( 0.098

0.783 ( 0.287 0.335 ( 0.145

0.0009 0.0003 0.0001

0.752 ( 0.186

0.377 ( 0.211

2 proteins (2 spots) varied during passages 0.268 ( 0.095 0.299 ( 0.047 0.514 ( 0.105 0.337 ( 0.068 0.047 ( 0.007 0.022 ( 0.008 0.030 ( 0.007 0.043 ( 0.011

1.216 ( 0.242

1.518 ( 0.101

0.0008 0.0000 0.0000 0.0001 0.0000

7 proteins (7 spots) increased during passages 1.723 ( 1.332 1.531 ( 1.063 2.448 ( 2.085 6.044 ( 2.326 0.199 ( 0.030 0.164 ( 0.057 0.155 ( 0.022 0.476 ( 0.048 0.202 ( 0.042 0.395 ( 0.052 0.189 ( 0.044 0.519 ( 0.106 0.509 ( 0.274 0.759 ( 0.116 0.729 ( 0.269 1.418 ( 0.354 0.157 ( 0.040 0.134 ( 0.038 0.178 ( 0.043 0.321 ( 0.082

0.540 ( 0.054 0.166 ( 0.040 0.075 ( 0.032 0.314 ( 0.054 0.071 ( 0.023 0.253 ( 0.109 2.217 ( 0.837 2.618 ( 0.177 0.378 ( 0.159 0.0002 0.0001 0.0005 0.0008 0.0001 0.0004 0.0009 0.0003 0.0000

0.758 ( 0.093 0.156 ( 0.037 0.088 ( 0.039 0.316 ( 0.076 0.081 ( 0.053 0.536 ( 0.106 2.259 ( 0.519 2.673 ( 0.400 0.869 ( 0.232

0.0003 0.0000 0.0000 0.0003 0.0001

ANOVA p-value

0.546 ( 0.148 0.124 ( 0.039 0.082 ( 0.018 0.260 ( 0.082 0.084 ( 0.009 0.216 ( 0.108 0.961 ( 0.575 1.913 ( 0.374 0.149 ( 0.051

P25

0.776 ( 0.055 0.247 ( 0.031 0.164 ( 0.041 0.453 ( 0.061 0.251 ( 0.045 0.397 ( 0.140 2.614 ( 0.456 2.883 ( 0.317 1.059 ( 0.463

P11

passages 0.171 ( 0.059 0.204 ( 0.046 0.168 ( 0.036 0.172 ( 0.068 0.016 ( 0.013

P7

11 proteins (14 spots) decreased during 0.419 ( 0.047 0.372 ( 0.098 0.292 ( 0.111 0.524 ( 0.075 0.529 ( 0.167 0.317 ( 0.069 0.506 ( 0.114 0.556 ( 0.169 0.326 ( 0.089 0.504 ( 0.124 0.564 ( 0.208 0.345 ( 0.091 0.047 ( 0.006 0.029 ( 0.008 0.024 ( 0.006

P5

1 0.0003

1

0.0277

1 1 0.0004 0.7248 1

1 0.0022 0.0055 0.0163 0.0002 0.3133 1 1 1

1 1 1 1 0.0083

P5 vs P7

0.0005 0.0167

1

0.0114

1 0.6005 1 1 1

0.0022 0.0066 0.0010 0.0149 0.0002 0.2648 1 1 0.0019

0.0924 0.0115 0.0713 0.3130 0.0009

P5 vs P11

1 1

0.0018

0.0000

0.0026 0.0000 0.0000 0.0001 0.0002

0.0028 0.0001 0.0026 0.0007 0.0008 0.0854 0.0009 0.0003 0.0001

0.0003 0.0001 0.0003 0.0021 0.0000

P5 vs P25

0.0012 0.5842

1

1

1 1 0.0002 1 0.9991

0.0047 1 1 1 1 0.0026 1 1 0.0321

0.6646 0.0093 0.0125 0.0605 1

P7 vs P11

1 0.0016

0.0069

0.0559

0.0017 0.0000 0.0253 0.0023 0.0000

0.0061 0.8872 1 1 1 0.0007 0.0094 0.0041 0.0011

0.0027 0.0001 0.0000 0.0003 0.1271

P7 vs P25

0.0076 0.0867

0.0050

0.1296

0.0135 0.0000 0.0000 0.0015 0.0011

1 0.3689 1 1 1 1 0.0123 0.0082 0.9604

0.1220 0.3943 0.1504 0.2238 0.7613

p11 vs P25

a Proteins were analyzed by 2-DE and identified by nano-LC-ESI-MS/MS. Relative protein expression resulting from software-assisted quantification is given (mean ( SD). Statistical analysis to reveal between-group differences was performed using SPSS.

O00299

P09936

Vimentin Dihydropyrimidinase-related protein 2 Heat shock protein beta-6 Transgelin Thioredoxin-dependent peroxide reductase, mitochondrial Ubiquitin carboxyl-terminal hydrolase isozyme L1 Chloride intracellular channel protein 1

ELKS/RAB6-interacting/CAST family member 1 Vinculin Heterogeneous nuclear ribonucleoprotein H 26S protease regulatory subunit 7 TAR DNA-binding protein 43 Protein kinase C delta-binding protein Cellular retinoic acid-binding protein 2 Prothymosin alpha Stathmin Cofilin-1

Collagen alpha-1(III) chain

protein name

P08670 Q16555 O14558 Q01995 P30048

P18206 P31493 P35988 Q13148 Q969G5 P29373 P06454 P16949 P23528

P02461(1) P02461(2) P02461(3) P02461(4) Q8IUD2

accession number (spot)

Table 2. Quantification Result of 23 Significantly Changed Spots during Cell Passages (n ) 6 per Group)a

research articles Chen et al.

Variations of Protein Levels in Human Amniotic Fluid Stem Cells

research articles 28

Figure 4. Western blot analysis and densitometry of passages 5, 7, 11, and 25. Immunoblotting confirmed results from gel-based proteomic studies. The expressional patterns are revealed and the statistical outcome is provided in Table 3.

resenting artifacts resulting from chemical modification by the analytical procedure, are listed in the Supplementary Table 2 (Supporting Information) as well. These 23 protein spots represented 20 individual proteins. Eleven proteins along with their different expression forms probably representing splice variants or posttranslational modifications showed decreased levels at passage 25, levels of 7 proteins increased over passaging and 2 proteins showed variable fluctuations following passaging. Western blot results were partially confirming results from the gel-based proteomics approach. Images for the proteins collagen type III, heterogeneous nuclear ribonucleoprotein H, TAR DNA-binding protein 43, stathmin, cofilin 1 and chloride intracellular channel protein 1 (presenting with 2 individual immunoreactive bands) are given in Figure 4. Results for significantly different protein levels as evaluated by immunoblotting and expressed as arbitrary units of optical density are shown in Tables 3a and 3b.

Discussion The major outcome of this investigation is that protein levels from a series of pathways and cascades fluctuate over the period of passaging in CD117/2 cells. This may be hampering comparability of results obtained from these cells even if the corresponding markers for the characterization of this cell line, caryotyping and biological criteria remain unchanged. Decreased stability of the cell line per se could have been detected by morphological criteria, that is, changed cell volume, but this finding may be observed only when cell size is compared to other passages. Protein level fluctuations over the number of passages were detected using stringent statistical criteria using a P value of