CD146/MCAM defines functionality of human bone marrow stromal stem cell populations

Abstract

Background: Identification of surface markers for prospective isolation of functionally homogenous populations of human skeletal (stromal, mesenchymal) stem cells (hMSCs) is highly relevant for cell therapy protocols. Thus, we examined the possible use of CD146 to subtype a heterogeneous hMSC population.

Methods: Using flow cytometry and cell sorting, we isolated two distinct hMSC-CD146(+) and hMSC-CD146(-) cell populations from the telomerized human bone marrow-derived stromal cell line (hMSC-TERT). Cells were examined for differences in their size, shape and texture by using high-content analysis and additionally for their ability to differentiate toward osteogenesis in vitro and form bone in vivo, and their migrational ability in vivo and in vitro was investigated.

Results: In vitro, the two cell populations exhibited similar growth rate and differentiation capacity to osteoblasts and adipocytes on the basis of gene expression and protein production of lineage-specific markers. In vivo, hMSC-CD146(+) and hMSC-CD146(-) cells formed bone and bone marrow organ when implanted subcutaneously in immune-deficient mice. Bone was enriched in hMSC-CD146(-) cells (12.6 % versus 8.1 %) and bone marrow elements enriched in implants containing hMSC-CD146(+) cells (0.5 % versus 0.05 %). hMSC-CD146(+) cells exhibited greater chemotactic attraction in a transwell migration assay and, when injected intravenously into immune-deficient mice following closed femoral fracture, exhibited wider tissue distribution and significantly increased migration ability as demonstrated by bioluminescence imaging.

Conclusion: Our studies demonstrate that CD146 defines a subpopulation of hMSCs capable of bone formation and in vivo trans-endothelial migration and thus represents a population of hMSCs suitable for use in clinical protocols of bone tissue regeneration.

Fig. 1

Flow cytometric cell sorting and cell characterization. a Details from FacsDIVA sorting demonstrating the overall distribution of CD146 in hMSC-TERT and the gated CD146⁺ (P4) and CD146⁻ (P5) populations. b Flow cytometric validation of the hMSC-CD146⁺ and hMSC-CD146⁻ populations and expression of canonical mesenchymal stem cell (MSC) markers: CD44, CD63, CD73, CD105, CD14, CD34, and CD146. c Analysis of population doublings demonstrating no significant difference between the different cell populations. FSC-A forward scatter-A, hMSC human mesenchymal stem cell

Fig. 2

Morphological and texture studies using high-content single-cell imaging. Fluorescent staining of hMSC-TERT, hMSC-CD146⁺, and hMSC-CD146⁻ (n = 3) demonstrates morphological changes assessed by tubulin and F-actin staining. Significant changes are demonstrated in cell roundness, area, and width-to-length ratio (*P < 0.05). Changes in F-actin are demonstrated by analysis of depth and height of ridges of the actin fibres. *P < 0.05. Scale bar = 100 μm. hMSC human mesenchymal stem cell, SER saddle, edge, ridge

Fig. 3

Characterization of sorted CD146 populations undergoing osteoblast (OB) differentiation. a Reverse transcription-polymerase chain reaction data (mean ± standard error of the mean, n = 3 independent experiments) analyzed from sorted population at day (d) 0, 5, 10, and 15. *P < 0.05. sh = short form and lg = long form of CD146. b Alkaline phosphatase (ALP) activity/cell viability at d6 of osteoblastic differentiation. c ALP staining at d6 and d10 of OB differentiation. Alizarin red (AZR) staining at d10 and d15 of OB differentiation. n = 3 independent experiments. white box, hMSC-CD146⁺; black box, hMSC-CD146⁻; gray box, lgCD146⁺

Fig. 4

In vivo heterotopic bone formation. hMSC-LUC2 (a, d, g, j, m), hMSC-CD146⁺ (b, e, h, k), and hMSC-CD146⁻ (c, f, i, l) were implanted subcutaneously (n = 4) mixed with hydroxyapatite tricalcium phosphate/Triosite (HA/TCP) into immune-compromised mice for 8 weeks. Analysis was performed on three serial sections at three depths with 100 μm between each depth. Hematoxylin and eosin (H&E) staining of bone formation demonstrating distribution of bone within the implants (a-i, m). d-f Blood vessel formation within the implants. g-i Establishment of bone marrow within the implants. j-l Human-specific vimentin staining of developing bone demonstrating the human origin. o, p Quantification of total bone or bone marrow volume. Scale bar on all H&E and vimentin staining = 100 μm

Fig. 5

Functional assessment of migrational ability of CD146 sorted human mesenchymal stem cell (hMSC) populations. a Analysis of in vitro transwell (Boyden chamber) migration assay. Data demonstrate statistically significant migration of hMSC-CD146⁺ in 10 % fetal bovine serum (FBS) as compared with hMSC-CD146⁻ under the same conditions. gray box, 0.2 % FBS hMSC-CD146^+/−; white box, 10 % FBS hMSC-CD146⁺; black box, 10 % FBS hMSC-CD146, n = 2 independent experiments. *P < 0.05. b IVIS^® in vivo imaging of hMSC-CD146⁺ and hMSC-CD146⁻ cells at days 0, 3, and 6 after intravenous tail vein injection. As can be observed in the images and from the radiance plots, hMSC-CD146⁺ cells (n = 4 mice) demonstrate migration away from the lung area at days 3 and 6, and hMSC-CD146⁻ cells (n = 5 mice) display less migration in comparison with hMSC-CD146⁺