For Peer Review Time-lapse microscopy and classification of 2D 1 human mesenchymal stem cells based on cell shape 2 picks up myogenic from osteogenic and adipogenic 3 differentiation 4 5 Christof Seiler 1 , Amiq Gazdhar 2 , Mauricio Reyes 1 , Lorin M Benneker 3 , Thomas 6 Geiser 2 , Klaus A Siebenrock 3 & Benjamin Gantenbein-Ritter 1,4 7 8 1 Institute for Surgical Technology and Biomechanics, University of Bern, Bern, 9 Switzerland 10 2 Department of Pulmonary Medicine, Insel University Hospital, Bern, Switzerland 11 3 Orthopeadic Department, Insel University Hospital, Bern, Switzerland 12 4 ARTORG Center for Biomedical Engineering Research, University of Bern, Bern, 13 Switzerland 14 15 16 17 To: Journal of Tissue Engineering and Regenerative Medicine 18 19 Corresponding Author: 20 Prof. Dr. Benjamin Gantenbein 21 University of Bern 22 Medical Faculty 23 ARTORG Center for Biomedical Engineering Research 24 Institute for Surgical Technology and Biomechanics 25 Stauffacherstrasse 78 26 CH-3014 Bern 27 Tel +4131 631 5926 28 Fax +4131 631 5960 29 e-mail: [email protected]30 31 Page 1 of 25 http://mc.manuscriptcentral.com/term Journal of Tissue Engineering and Regenerative Medicine 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
25
Embed
Time-lapse microscopy and classification of 2D human ... · For Peer Review 1 Time-lapse microscopy and classification of 2D 2 human mesenchymal stem cells based on cell shape 3 picks
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
For Peer Review
Time-lapse microscopy and classification of 2D 1
human mesenchymal stem cells based on cell shape 2
picks up myogenic from osteogenic and adipogenic 3
differentiation 4 5 Christof Seiler1, Amiq Gazdhar2, Mauricio Reyes1, Lorin M Benneker3, Thomas 6 Geiser2, Klaus A Siebenrock3 & Benjamin Gantenbein-Ritter1,4 7 8 1 Institute for Surgical Technology and Biomechanics, University of Bern, Bern, 9 Switzerland 10 2Department of Pulmonary Medicine, Insel University Hospital, Bern, Switzerland 11 3Orthopeadic Department, Insel University Hospital, Bern, Switzerland 12 4ARTORG Center for Biomedical Engineering Research, University of Bern, Bern, 13 Switzerland 14 15
16
17
To: Journal of Tissue Engineering and Regenerative Medicine 18
19
Corresponding Author: 20
Prof. Dr. Benjamin Gantenbein 21
University of Bern 22
Medical Faculty 23
ARTORG Center for Biomedical Engineering Research 24
Institute for Surgical Technology and Biomechanics 25
This is the peer reviewed version of the following article: Seiler, C., Gazdhar, A., Reyes, M., Benneker, L. M., Geiser, T., Siebenrock, K. A., & Gantenbein‐Ritter, B. (2014). Time‐lapse microscopy and classification of 2D human mesenchymal stem cells based on cell shape picks up myogenic from osteogenic and adipogenic differentiation. Journal of Tissue Engineering and Regenerative Medicine, 8(9), 737-746, which has been published in final form at http://dx.doi.org/10.1002/term.1575. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions.
For Peer Review
Seiler et al. Classification of 2D hMSCs based on shape
2
Abstract 1
Current methods to characterize mesenchymal stem cells (MSCs) are limited to CD 2
marker expression, plastic adherence and their ability to differentiate into adipo-, 3
osteo- and chondrogenic precursors. It seems evident that stem cells undergoing 4
differentiation should differ in many aspects such as morphology and possibly also in 5
behavior, however such correlation has not yet being exploited for fate prediction of 6
MSCs. 7
Primary human MSCs from bone marrow were expanded and pelleted to form high-8
density cultures and were then split/ (randomly divided) into four groups to 9
differentiate into adipo-, osteo-, chondro-, and myogenic progenitor cells. Cells were 10
expanded as heterogeneous and clonal populations and tracked with time-lapse 11
microscopy to record cell shape using phase-contrast microscopy. Cells were 12
segmented using a custom-made image processing pipeline. Seven morphological 13
features were extracted for each of the segmented cells. Statistical analysis was 14
performed on the 7-dimensional feature vectors using a tree-like classification 15
method. Differentiation of cells was monitored with key marker genes and histology. 16
Cells in differentiation media were expressing the key genes for each of the three 17
pathways after 21 days, i.e. adipo-, osteo-, chondrogenesis, which was also confirmed 18
by histological stains. Time-lapse microscopy data was obtained and contained new 19
evidence that two cell shape features, eccentricity and filopodia (= “fingers”) are 20
highly informative to classify myogenic differentiation from all others. However, no 21
robust classifiers could be identified for the other cell differentiation paths. 22
Results suggest that non-invasive automated time-lapse microscopy could be 23
potentially used to predict stem cell fate of hMSCs for clinical application based on 24
morphology for earlier time-points. Classification is challenged by cell density, 25
proliferation and possible unknown donor-specific factors, which affect the 26
in orthopedics. Tissue Eng 11(7-8): 1198-1211. 9 Chamberlain, JS. 2006; Stem-cell biology: a move in the right direction. Nature 10
444(7119): 552-553. 11 Cohen, AR, Gomes, FL, Roysam, B, et al. 2010; Computational prediction of neural 12
progenitor cell fates. Nat Methods 7(3): 213-218. 13 Dainiak, MB, Kumar, A, Galaev, IY, et al. 2007; Methods in cell separations. Adv 14
Biochem Eng Biotechnol 106 1-18. 15 da Silva Meirelles, L, Caplan, AI, Nardi, NB. 2008; In search of the in vivo identity of 16
mesenchymal stem cells. Stem Cells 26(9): 2287-2299. 17 Dominici, M, Blanc, KL, Mueller, I, et al. 2006; Minimal criteria for defining 18
multipotent mesenchymal stromal cells. The International Society for Cellular 19 Therapy position statement. Cytotherapy 8(4): 315-315. 20
Flanagan, LA, Lu, J, Wang, L, et al. 2008; Unique dielectric properties distinguish 21 stem cells and their differentiated progeny. Stem Cells 26(3): 656-665. 22
Gantenbein-Ritter, B, Benneker, LM, Alini, M, et al. 2011; Differential response of 23 human bone marrow stromal cells to either TGF-β(1) or rhGDF-5. Eur Spine J 24 20 962-971. 25
Giordano, A, Galderisi, U, Marino, IR. 2007; From the laboratory bench to the 29 patient's bedside: an update on clinical trials with mesenchymal stem cells. J 30 Cell Physiol 211(1): 27-35. 31
Glauche, I, Lorenz, R, Hasenclever, D, et al. 2009; A novel view on stem cell 32 development: analysing the shape of cellular genealogies. Cell Prolif 42(2): 33 248-263. 34
Gorelick, L, Galun, M, Sharon, E, et al. 2006; Shape representation and classification 35 using the poisson equation. IEEE Trans Pattern Anal Mach Intell 28(12): 1991-36 2005. 37
Klauschen, F, Qi, H, Egen, JG, et al. 2009; Computational reconstruction of cell and 38 tissue surfaces for modeling and data analysis. Nat Protoc 4(7): 1006-1012. 39
Lee, WC, Bhagat, AA, Huang, S, et al. 2011; High-throughput cell cycle 40 synchronization using inertial forces in spiral microchannels. Lab Chip 11(7): 41 1359-1367. 42
Levin, M. 2007; Large-scale biophysics: ion flows and regeneration. Trends Cell Biol 43 17(6): 261-270. 44
Livak, KJ, Schmittgen, TD. 2001; Analysis of relative gene expression data using 45 real-time quantitative PCR and the 2^(-Delta Delta C(T)) Method. Methods 46 25(4): 402-408. 47
Page 13 of 25
http://mc.manuscriptcentral.com/term
Journal of Tissue Engineering and Regenerative Medicine
Seiler et al. Classification of 2D hMSCs based on shape
14
Lutolf, MP, Gilbert, PM, Blau, HM. 2009; Designing materials to direct stem-cell 1 fate. Nature 462(7272): 433-441. 2
Majd, H, Wipff, PJ, Buscemi, L, et al. 2008; A Novel Method of Dynamic Culture 3 Surface Expansion Improves Mesenchymal Stem Cell Proliferation and 4 Phenotype. Stem Cells 27(1): 200-209. 5
Mareddy, S, Broadbent, J, Crawford, R, et al. 2009; Proteomic profiling of distinct 6 clonal populations of bone marrow mesenchymal stem cells. J Cell Biochem 7 106(5): 776-786. 8
Mareddy, S, Crawford, R, Brooke, G, et al. 2007; Clonal isolation and 9 characterization of bone marrow stromal cells from patients with osteoarthritis. 10 Tissue Eng 13(4): 819-829. 11
Mehlhorn, AT, Schmal, H, Kaiser, S, et al. 2006; Mesenchymal stem cells maintain 12 TGF-beta-mediated chondrogenic phenotype in alginate bead culture. Tissue 13 Eng 12(6): 1393-1403. 14
Meinshausen, N. 2009; Node harvest: simple and interpretable regression and 15 classification. Electronic Journal of Statistics arXiv 0910.2145v1. 16
Morris JH, Apeltsin L, Newman AM, et al. 2011; clusterMaker: a multi-algorithm 17 clustering plugin for Cytoscape. BMC Bioinformatics 12: 436. 18
Pierantozzi, E, Gava, B, Manini, I, et al. 2010; Pluripotency Regulators in Human 19 Mesenchymal Stem Cells: Expression of NANOG But Not of OCT-4 and SOX-20 2. Stem Cells Dev 20(5): 915-923. 21
Pittenger, MF. 2008; Mesenchymal stem cells from adult bone marrow. Methods Mol 22 Biol 449 27-44. 23
Pittenger, MF, Mackay, AM, Beck, SC, et al. 1999; Multilineage Potential of Adult 24 Human Mesenchymal Stem Cells. Science 284(5411): 143. 25
Prockop, DJ. 1997; Marrow stromal cells as stem cells for nonhematopoietic tissues. 26 Science 276(5309): 71-74. 27
Prockop, DJ. 2001; Stem cell research has only just begun. Science 293(5528): 211-28 212. 29
Raimondo, S, Penna, C, Pagliaro, P, et al. 2006; Morphological characterization of 30 GFP stably transfected adult mesenchymal bone marrow stem cells. J Anat 31 208(1): 3. 32
Ravin, R, Hoeppner, DJ, Munno, DM, et al. 2008; Potency and fate specification in 33 CNS stem cell populations in vitro. Cell Stem Cell 3(6): 670-680. 34
Roeder, I, Radtke, F. 2009; Stem cell biology meets systems biology. Development 35 136(21): 3525-3530. 36
Sarugaser, R, Hanoun, L, Keating, A, et al. 2009; Human mesenchymal stem cells 37 self-renew and differentiate according to a deterministic hierarchy. PLoS ONE 38 4(8): e6498. 39
Schmittgen, TD, Zakrajsek, BA. 2000; Effect of experimental treatment on 40 housekeeping gene expression: validation by real-time, quantitative RT-PCR. J 41 Biochem Biophys Methods 46(1-2): 69-81. 42
Sengers, BG, Dawson, JI, Oreffo, RO. 2009; Characterisation of Human Bone 43 Marrow Stromal Cell heterogeneity for skeletal regeneration strategies using a 44 two-stage colony assay and computational modeling. Bone 46(2): 496-503. 45
Solchaga, LA, Johnstone, B, Yoo, JU, et al. 1999; High variability in rabbit bone 46 marrow-derived mesenchymal cell preparations. Cell Transplant 8(5): 511-519. 47
Sundelacruz, S, Levin, M, Kaplan, DL. 2008; Membrane potential controls 48 adipogenic and osteogenic differentiation of mesenchymal stem cells. PLoS 49 ONE 3(11): e3737. 50
Page 14 of 25
http://mc.manuscriptcentral.com/term
Journal of Tissue Engineering and Regenerative Medicine
Seiler et al. Classification of 2D hMSCs based on shape
15
Sundelacruz, S, Levin, M, Kaplan, DL. 2009; Role of membrane potential in the 1 regulation of cell proliferation and differentiation. Stem Cell Rev Rep 5(3): 231-2 246. 3
Zuk, PA, Zhu, M, Mizuno, H, et al. 2001; Multilineage cells from human adipose 4 tissue: implications for cell-based therapies. Tissue Eng 7(2): 211-228. 5
6
Page 15 of 25
http://mc.manuscriptcentral.com/term
Journal of Tissue Engineering and Regenerative Medicine
Seiler et al. Classification of 2D hMSCs based on shape
16
Figure legends 1 2 Figure 1. A-H Illustration of the segmentation pipeline that was used to extract cells 3
from phase-contrast images. 10x magnification of the phase-contrast images. The 4
time-lapse microscope series (microscope, Incucyte, Essen Bioscience) was used. 5
6 Figure 2. A-H The seven shape features that were extracted from each cell for 7 morphological analysis. 8 9 Figure 3. Relative gene expression profiles for marker genes grouped for osteogenic, 10
adipogenic, chondrogenic and myogenic differentiation of human mesenchymal stem 11
cells (hMSCs). 12
13 Figure 4. Histological stainings for each of the four differentiation lines of primary 14
human mesenchymal stem cells after 12, 15 and 21 days. First, column: Osteogenic 15
differentiation: Von Kossa /fast red stain, second column: adipogenic differentiation, 16
red oils / Meyer’s hematoxylin, third and forth column: chondrogenic differentiation 17
Seiler et al. Classification of 2D hMSCs based on shape
19
Table 3 Real-time RT-PCR Primers used for Real-time RT-PCR. All primers were run at 61°C Ta 1 (annealing temperature) and a two-step protocol. 2 3 Gene Abbreviation Name Forward Reverse Hs18S
Reference Gene CGA TGC GGC GGC GTT ATT C
TCT GTC AAT CCT GTC CGT GTC C
Chondrogenic Differentiation ACAN Aggrecan core protein CAT CAC TGC AGC
Figure 1. A-H Segmentation pipeline that was used to extract shape from phase-contrast images using the 10x magnification of the time-lapse microscope images.
130x171mm (150 x 150 DPI)
Page 20 of 25
http://mc.manuscriptcentral.com/term
Journal of Tissue Engineering and Regenerative Medicine
Figure 4. Histological stainings for each of the four differentiation lines of primary human mesenchymal stem cells after 12, 15 and 21 days. First, column: Osteogenic differentiation: Von Kossa /fast red stain, second column: adipogenic differentiation, red oils / Meyer’s hematoxylin, third and forth column: chondrogenic
differentiation aclian blue and safranin O / fast green stain. 1411x705mm (72 x 72 DPI)
Page 23 of 25
http://mc.manuscriptcentral.com/term
Journal of Tissue Engineering and Regenerative Medicine
Figure 5. Confocal laser scanning microscope images of MyoD stain (ALEXA555) and for α-smooth muscle actin stain (FITC) for myogenic differentiation and negative controls. Nuclei were counter stained with DAPI.
A. and B myotube formation after 14 days of culture which express MyoD, C is negative control and D is positive stain for undifferentiated control, E is positive staining for α smooth muscle (sm) actin of myogenic
differentiation and F is negative control. 485x321mm (300 x 300 DPI)
Page 24 of 25
http://mc.manuscriptcentral.com/term
Journal of Tissue Engineering and Regenerative Medicine
Figure 6. Output of Node Harvest classification method. The response axis represents A: myogenic=0 and
control=1, B: myogenic=0 and adipogenic=1, C: myogenic=0 and osteogenic=1, D: myogenic=0 and adipgenic/osteogenic/control=1, where values between 0 and 1 represent a weighted combination of both types. The sample axis indicates the number of cells. Each node contains cells that fulfill one condition of one element of the 7-feature vector. The node size shows the relevance of that condition. Lines between
nodes symbolize subsets. 205x162mm (300 x 300 DPI)
Page 25 of 25
http://mc.manuscriptcentral.com/term
Journal of Tissue Engineering and Regenerative Medicine