From cytogenomic to epigenomic profiles: monitoring the biologic behavior of in vitro cultured human bone marrow mesenchymal stem cells

Abstract

Introduction: Bone marrow mesenchymal stem cells (BM-MSCs) are multipotent cells that can differentiate into different cell lineages and have emerged as a promising tool for cell-targeted therapies and tissue engineering. Their use in a therapeutic context requires large-scale in vitro expansion, increasing the probability of genetic and epigenetic instabilities. Some evidence shows that an organized program of replicative senescence is triggered in human BM-MSCs (hBM-MSCs) on prolonged in vitro expansion that includes alterations in phenotype, differentiation potential, telomere length, proliferation rates, global gene-expression patterns, and DNA methylation profiles.

Methods: In this study, we monitored the chromosomal status, the biologic behavior, and the senescence state of hBM-MSCs derived from eight healthy donors at different passages during in vitro propagation. For a more complete picture, the telomere length was also monitored in five of eight donors, whereas the genomic profile was evaluated in three of eight donors by array-comparative genomic hybridization (array-CGH). Finally, an epigenomic profile was delineated and compared between early and late passages, by pooling DNA of hBM-MSCs from four donors.

Results: Our data indicate that long-term culture severely affects the characteristics of hBM-MSCs. All the observed changes (that is, enlarged morphology, decreased number of cell divisions, random loss of genomic regions, telomere shortening) might be regulated by epigenetic modifications. Gene Ontology analysis revealed that specific biologic processes of hBM-MSCs are affected by variations in DNA methylation from early to late passages.

Conclusions: Because we revealed a significant decrease in DNA methylation levels in hBM-MSCs during long-term culture, it is very important to unravel how these modifications can influence the biologic features of hBM-MSCs to keep track of this organized program and also to clarify the conflicting observations on hBM-MSC malignant transformation in the literature.

Figure 1

Human bone marrow stem cell (hBM-MSC) cultures at P3, P6, P9, and P12. At all culture passages examined, hBM-MSCs displayed a fibroblast-like morphology. At P9, some extra- and intracellular debris (arrows) appeared and became more evident at P12. Images from donor 6 are shown by way of example. Bars, 100 μm.

Figure 2

Human bone marrow stem cells (hBM-MSCs) characterization. hBM-MSC expression of CD105 (A) and CD34 (B) by flow-cytometric analysis. hBM-MSC mesengenic differentiation capability (C-E): Alizarin red staining of osteogenic differentiated hMSCs (C); oil red O staining of adipogenic differentiated hBM-MSCs (D); safranin O staining of chondrogenic differentiated hBM-MSCs (E). Bars, 50 μm (D) and 25 μm (C).

Figure 3

Human bone marrow stem cell (hBM-MSC) senescence. β-Galactosidase staining (blue) of hBM-MSCs from donor 2 at P10 (A), donor 3 at P4 (B), donor 5 at P13 (C), donor 6 at P16 (D), and donor 8 at P14 (E). SHSY-5Y cell line (F) was used as negative control of the β-galactosidase staining. Bars, 50 μm.

Figure 4

Telomere length in human bone marrow stem cells (hBM-MSCs) at different culture passages. (A) Telomere length evaluated by Southern blotting in hBM-MSCs (donors 1, 2, 5, 7, and 8) at P0, P3, P6, P9, and P12. The size marker is indicated on the left, and the positive control DNA used is a purified genomic DNA from immortal cell lines. Differences in telomere length were observed between P0, P3, and the following passages (P6, P9, P12) examined. (B) ANOVA statistical analysis of telomere length (donors 1, 2, 5, 7, and 8). A medium spot in the range of telomere length (smear, A) was calculated (see Material and Methods), and a mean value between donors (x axis) was calculated for each passage. Data are expressed as mean ± SEM. No significant differences in the mean values of telomere lengths were observed between P0 and P3, whereas significant differences were observed between P3 and P6 (P < 0.05), and between P0/P3/P6 and P9/P12 (P < 0.001).

Figure 5

Example of array-CGH profiles of human bone marrow stem cells (hBM-MSCs) at three different passages of culture. Example of chromosome 3 from donor 8 in three overlapping experiments (P0, green line; P4, blue line; P9, red line): the profiles were almost overlapping (black arrow, a common copy-number variation (CNV), but P9 showed some exclusive CNVs (dotted rectangle).

Figure 6

CpG methylation profile of human bone marrow stem cells (hBM-MSCs) at early and late passages of culture. Histograms with percentages of methylation of each chromosome at early passages (top) and late passages (bottom) of hBM-MSCs. Red, methylated; green, unmethylated.

Figure 7

Gene ontology (GO) analysis by GOstat software. Histogram with percentages of gene promoters associated with a change in the methylation profile in late passages, classified by category of biologic process. Black, percentages of demethylated gene promoters; gray, percentages of methylated gene promoters.