CD106/VCAM-1 distinguishes a fibroblast subpopulation with high colony-forming capacity and distinct protein expression from the uterosacral ligament

Abstract

Background: Pelvic organ prolapse (POP) is a common degenerative disease in women which may diminish quality of life. Investigating the pathological changes of the uterosacral ligament, including the functional changes of fibroblasts, is critical to understanding the pathophysiology of POP. This study was designed to isolate CD106-positive (CD106⁺) fibroblasts from the human uterosacral ligament and assess the function and expression of this subpopulation.

Methods: We separated CD106⁺ fibroblasts and CD106 negative (CD106^-) fibroblasts by fluorescence-activated cell sorting (FACS) and cultured them for subsequent experiments. Flow cytometric analysis was used to test the sorting efficiency, CD106 expression, and typical mesenchymal stem cell (MSC) phenotype marker expression. A colony-forming unit (CFU) assay was applied to evaluate the colony-forming ability of the fibroblasts. Trilineage differentiation capacities were assessed after in vitro induction. The protein levels of vimentin, fibroblast specific protein-1 (FSP-1), collagen I (COL 1), matrix metallopeptidase-1 (MMP-1), and α-smooth muscle actin (α-SMA) were detected by western blot analysis. The expression of CD106 was verified by flow cytometric analysis and immunohistochemistry (IHC) in the POP and non-POP groups.

Results: The CD106⁺ fibroblasts were isolated with a purity of (93.50±3.91)%. The CD106⁺ fibroblasts exhibited higher colony-forming capacity than that of CD106^- fibroblasts, but neither of them showed adipogenic or osteogenic differentiation similar to that of MSCs. The protein levels of MMP-1 and α-SMA were lower, and the level of COL 1 was higher in the CD106⁺ fibroblasts than in the CD106- fibroblasts. In addition, we observed a decreased expression of CD106 in the POP group compared with the non-POP group.

Conclusions: Our results suggest that CD106⁺ fibroblasts possess a high colony-forming capacity and distinct protein expression, and this subpopulation is reduced in POP.

Keywords: CD106; Fibroblasts; pelvic organ prolapse (POP); uterosacral ligament.

Figure 1

Isolation of CD106⁺ and CD106⁻ human uterosacral ligament fibroblasts. (A) CD106⁺ fibroblasts and CD106⁻ fibroblasts were sorted by FACS. Viable cells were selected by their FSC versus their SSC profile and single cells were selected by their height versus SSC-A. (B) Morphology of CD106⁺ fibroblasts and CD106⁻ fibroblasts, scale bar, 100 μm. Cultured cells were observed by an inverted microscope at low magnification (×4) and medium magnification (×10). (C) Flow cytometry analysis of CD106⁺ fibroblasts, CD106⁻ fibroblasts, and the unsorted fibroblasts. The blue areas represent unstained cells serving as controls; the red areas represent stained cells; the percentages of positively labeled cells are listed. (D) The percentage of CD106 positive cells in CD106⁺ fibroblasts during passaging (P3, P4, P5), P: passage. The data shown are the means ± SD of 3 independent experiments. ***, P<0.001 compared with CD106⁺ fibroblasts at P3. FACS, fluorescence-activated cell sorting; FSC, forward scatter; SSC, side scatter; SSC-A, side scatter-area; SSC-H, side scatter-height; FSC-H, forward side scatter-height; SD, standard deviation.

Figure 2

Phenotyping of MSCs and human uterosacral ligament fibroblasts for typical MSC phenotypic markers. (A) Human uterosacral ligament fibroblasts and MSCs were positive for the expression of markers CD90, CD73, and CD105 but negative for CD34, CD45, and HLA-DR in a similar pattern. (B) The positive rate of CD106 in MSCs and in uterosacral ligament fibroblasts. The blue areas represent unstained cells serving as controls; the red areas represent stained cells; the percentages of positively labeled cells are shown. The data shown are the means ± SD of 3 independent experiments. ***, P<0.001. MSC, mesenchymal stem cell; HLA-DR, human leukocyte antigen-DR; SD, standard deviation.

Figure 3

Colony-forming assay of the sorted cell populations and unsorted cells. The colony-forming number of CD106⁺ fibroblasts was significantly higher than that of CD106⁻ fibroblasts. There was no difference between CD106⁻ fibroblasts and unselected cells. Cells were stained with crystal violet after being fixed by 4% paraformaldehyde. The data shown are the means ± SD of 3 independent experiments. **, P<0.01; ***, P<0.001. ns, not significant; SD, standard deviation.

Figure 4

Exploring the trilineage differentiation capacity of MSCs and CD106⁺ and CD106⁻ fibroblasts. (A) Adipogenesis was detected by Oil Red O staining. The lipid droplets were quantified by calculating the positive area in each group relative to the control. (B) Osteogenesis was detected by Alizarin Red staining. The mineralized nodules were quantified by calculating the positive area in each group relative to the control. (C) Chondrogenesis was detected by Alcian Blue staining. Fibroblasts incubated in expansion medium served as the controls. Scale bar, 100 μm. The data shown are the means ± SD of 3 independent experiments. *, P<0.05 versus the MSC group. MSC, mesenchymal stem cell; NS, not significant; SD, standard deviation.

Figure 5

Comparison of the typical fibroblast proteins between CD106⁺ and CD106⁻ fibroblasts. (A) A series of typical fibroblast protein levels were determined by western blot analysis in MSCs, CD106⁺ and CD106⁻ fibroblasts, and unsorted fibroblasts. (B) Protein densitometric quantification of COL 1, vimentin, MMP-1, FSP-1, and α-SMA. The data shown are the means ± SD of 3 independent experiments. **, P<0.01; ***, P<0.001 versus the CD106⁺ groups. COL 1, collagen I; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MMP-1, matrix metallopeptidase-1; FSP-1, fibroblast specific protein-1; α-SMA, α-smooth muscle actin; MSC, mesenchymal stem cell; SD, standard deviation.

Figure 6

Comparison of CD106 expression between the POP and non-POP groups. (A) IHC analysis of CD106 expression, scale bar, 100 μm. (B) Scatterplot showing the positive area of CD106 in the POP and non-POP groups (n=10). Error bars represent mean ± SD. ****, P<0.0001. (C) Flow cytometry analysis of the CD106⁺ cells. The blue areas represent unstained cells serving as controls; the red areas represent stained cells; the percentages of positively labeled cells are shown. (D) Scatterplot showing the percentages of CD106⁺ cells in fibroblasts from the POP and non-POP groups (n=5). Error bars represent mean ± SD. *, P<0.05. POP, pelvic organ prolapse; SSC, side scatter; IHC, immunohistochemistry; SD, standard deviation.

Figure 7

Schematic diagram of the isolation and characteristics of CD106⁺ fibroblasts. The primary fibroblasts isolated from uterosacral ligaments were expanded to passage 3 and then sorted by the surface marker CD106 using FACS. CD106⁺ fibroblasts exhibited a high colony-forming capacity and different protein expressions compared with CD106⁻ fibroblasts. FACS, fluorescence-activated cell sorting; MMP-1, matrix metallopeptidase-1; α-SMA, α-smooth muscle actin.