Assay validation for the assessment of adipogenesis of multipotential stromal cells--a direct comparison of four different methods

Abstract

Background aims: Mesenchymal stromal cells (MSCs) are regenerative and immuno-privileged cells that are used for both tissue regeneration and treatment of severe inflammation-related disease. For quality control of manufactured MSC batches in regard to mature fat cell contamination, a quantitative method for measuring adipogenesis is needed.

Methods: Four previously proposed methods were validated with the use of bone marrow (BM) MSCs during a 21-day in vitro assay. Oil red staining was scored semiquantitatively; peroxisome proliferator activated receptor-γ and fatty acid binding protein (FABP)4 transcripts were measured by quantitative real-time polymerase chain reaction; FABP4 protein accumulation was evaluated by flow cytometry; and Nile red/4',6-diamidino-2-phenylindole (DAPI) ratios were measured in fluorescent microplate assay. Skin fibroblasts and MSCs from fat pad, cartilage and umbilical cord were used as controls.

Results: Oil red staining indicated considerable heterogeneity between BM donors and individual cells within the same culture. FABP4 transcript levels increased 100- to 5000-fold by day 21, with large donor variability observed. Flow cytometry revealed increasing intra-culture heterogeneity over time; more granular cells accumulated more FABP4 protein and Nile red fluorescence compared with less granular cells. Nile red increase in day-21 MSCs was ~5- and 4-fold, measured by flow cytometry or microplate assay, respectively. MSC proliferation/apoptosis was accounted through the use of Nile red/DAPI ratios; adipogenesis levels in day-21 BM MSCs increased ~13-fold, with significant correlations with oil red scoring observed for MSC from other sources.

Conclusions: Flow cytometry permits the study of MSC differentiation at the single-cell level and sorting more and less mature cells from mixed cell populations. The microplate assay with the use of the Nile red/DAPI ratio provides rapid quantitative measurements and could be used as a low-cost, high-throughput method to quality-control MSC batches from different tissue sources.

Figure 1

Semiquantitative scoring of adipogenesis with the use of oil red. (A) Visual grading scheme on the basis of lipid droplet accumulation: microphotographs of representative cells assigned to grades 1–4. Grades represent the proportion of cell cytoplasm occupied by lipid droplets. (B) Adipogenesis time course: fibroblasts (left panels) and BM MSCs (right panels) were differentiated over 21 days into fat lineage, and microphotographs were taken from a central area of a representative triplicate well. (C) Percentage of grade 1 cells. (D) Percentage of cells of grades 2 and above. BM1-3 represents the three different donors, and the percentage shown is the mean value from triplicate wells for each donor. X-axis numbers represent days of culture. Original magnifications: ×400 for A and ×100 for B.

Figure 2

Monitoring adipogenic progression of MSCs and fibroblasts with the use of q-PCR. MSCs and fibroblasts were cultured in adipogenic medium for 21 days. Expression of (A) PPAR-γ and (B) FABP4 was determined on days 0, 3, 7, 14 and 21. Relative levels of gene expression were normalized to reference gene GAPDH and displayed as fold increase over day 0. Each data point represents the mean of three replicates.

Figure 3

Monitoring adipogenic progression of MSCs with the use of flow cytometry for FABP4 and Nile red. (A) Representative histogram plots for FABP4 staining. (B) The increase in SSC characteristics for the whole population of cells as the time course progresses; error bars represent standard deviation of the mean for three donors tested. (C) Representative histogram plots for Nile red staining. (D) Nile red fluorescence of both populations of gated low- and high-SSC cells, demonstrating a parallel increase throughout the time course. Days 0 and 3 are omitted because of very low frequency of high-SSC early in differentiation. Numbers in the top right corners of histograms represent median fluorescence intensities. FI is fluorescent intensity.

Figure 4

Nile red and DAPI staining of MSCs. MSCs and fibroblasts were grown for 0, 3, 7, 14 and 21 days in adipogenic medium and stained with Nile red dye and DAPI. Representative images of (A) fibroblasts and (B) MSCs on day 21 are shown. Intracellular lipid droplets in cells stain yellow/gold, and the nucleus is stained blue. Images shown are ×200 magnification. (C) Levels of fluorescence for DAPI and Nile red were measured for fibroblasts cultured in adipogenic medium. Error bars shown are ± standard deviation (deviation between replicate wells). (D) Levels of fluorescence for DAPI and Nile red were calculated for BM MSCs 1–4 cultured in adipogenic medium, and the mean values normalized to day 0 shown. Error bars shown are ± standard deviation (deviation between donors). (E) Adipogenesis of MSCs isolated from umbilical cord (UC), knee cartilage (C), knee fat pad (FP) and bone marrow (BM) on day 21 of culture. Error bars shown are ± standard deviation (n = 3 donors each). (E) Correlation between Nile red/DAPI ratios and grades 2+3+4 semiquantitative scoring. These methods exhibited significant levels of correlation (r = 0.983, P < 0.0001).