gms | German Medical Science

German Congress of Orthopedic and Trauma Surgery (DKOU 2018)

23.10. - 26.10.2018, Berlin

Image processing workflow to visualize and quantify MSCs in 3D

Meeting Abstract

  • presenting/speaker Carolina Gamez - Department for experimental Orthopaedics and Trauma Surgery, Orthopaedics and Trauma Surgery Centre (OUZ), Medical Faculty Mannheim of the University of Heidelberg, Mannheim, Germany
  • Arif ul Maula Khan - Medical Research Center (ZMF), Medical Faculty Mannheim of the University of Heidelberg, Mannheim, Germany
  • Angelo Torelli - Medical Research Center (ZMF), Medical Faculty Mannheim of the University of Heidelberg, Mannheim University of Applied Sciences, Mannheim, Germany
  • Barbara Schneider-Wald - Department for experimental Orthopaedics and Trauma Surgery, Orthopaedics and Trauma Surgery Centre (OUZ), Medical Faculty Mannheim of the University of Heidelberg, Mannheim, Germany
  • Norbert Gretz - Medical Research Center (ZMF), Medical Faculty Mannheim of the University of Heidelberg, Mannheim, Germany
  • Ivo Wolf - Department of Computer Science, Mannheim University of Applied Sciences, Mannheim, Germany
  • Karen Bieback - Institute of Transfusion Medicine and Immunology, German Red Cross Blood Service of BW-Hessen, Medical Faculty Mannheim of the University of Heidelberg, Mannheim, Germany
  • Markus Schwarz - Sektion experimentelle Orthopädie und Unfallchirurgie, OUZ, Medizin. Fakultät Mannheim, Universität Heidelberg, Mannheim, Germany

Deutscher Kongress für Orthopädie und Unfallchirurgie (DKOU 2018). Berlin, 23.-26.10.2018. Düsseldorf: German Medical Science GMS Publishing House; 2018. DocGF16-955

doi: 10.3205/18dkou491, urn:nbn:de:0183-18dkou4913

Published: November 6, 2018

© 2018 Gamez et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at http://creativecommons.org/licenses/by/4.0/.


Outline

Text

Objectives: When aiming at cartilage tissue engineering with mesenchymal stromal cells (MSCs), 3D culture appears more promising and physiologic than the typical monolayer 2D culture. But, visualization and quantification of cells in a 3D scaffold is hampered. Confocal microscopy is a tool to visualize cells in a volume. However, whether it also allows discrimination of viable cells and accurate quantification of primary cells like MSCs is the question. The aim of the study was to compare given numbers of MSCs seeded in scaffolds with the cell numbers after 3D culturing quantified by an image processing tool.

Methods: Bone marrow-MSCs from 3 months old pigs were isolated and expanded (DMEM, 10% FBS, 1% antibiotics, 1% L-Gln). MSCs immunophenotyping was done by flow cytometry.

For the 3D culture, MSCs were embedded in 1.5% alginate scaffolds at defined cell numbers: 32, 45, and 64 cells/mm3 (theoretical values). Alginate polymerization was induced by CaCl2. Scaffolds were kept in culture for 24 hours; then, stained with Live/Dead kit for 30 minutes, and visualized by confocal microscopy immediately. Alive cells were identified at the wavelength of 494/517 nm (Ex/Em). The imaged Region of Interest (ROI) was in the middle of the scaffold with volumes of 0.6-0.7 mm3. The expected cell numbers were compared against the counted numbers obtained by Leica Application Suite X (LAS X) from Leica Microsystems.

3D images were processed for cell segmentation, counting and statistical analysis using the following workflow: 3D median filter (to reduce background noise), manual threshold adjustment, 3D binary hole filling, 3D image opening, binary image editing, features calculation, features histogram, and report/results. The binary data was used to count and calculate 29 different features based on shape and intensity. Cells were identified as structures with diameters between 8-25 μm. The expected cell numbers were compared against the counted numbers obtained by LAS X software.

Results and conclusion: A method was performed to culture, visualize and count cells in a 3D-cell culture system in vitro. We visualized cells in the ROI through a depth of 500-600 μm. The quantification was performed semi-automatically. We counted 33, 49, and 56 cells/mm3, which correlated with the theoretical numbers of 23, 45, and 64 cells/mm3 (Pearson correlation coefficient = 0.94). Thus, this quantification method seems to be valid.

In addition, several features like diameter, volume, surface area, and sphericity of cells were defined using the LAS X software. In further studies, these morphometric parameters could be useful as following-up criteria in case of changes of cells' shape.

The cells in the scaffold can be visualized by confocal microscopy, and quantified by our image processing workflow. The results suggest that with this protocol, the number of cells can be counted correctly in a 3D scaffold. However, we cannot exclude that some cells could have proliferated and others died over the time period of the experiment.