Derivation of functional smooth muscle cells from multipotent human hair follicle mesenchymal stem cells

Abstract

We investigated the potential of human hair follicle cells for multilineage differentiation and as a source of functional smooth muscle cells (SMCs). We report that human hair follicle stem cells (HFCs) isolated from individual follicles expressed surface markers that are characteristic of mesenchymal stem cells such as CD44, CD49b, CD73, CD90, and CD105 but lacked hematopoietic markers CD45 and CD34. In addition, HFCs differentiated toward adipocytes, chondrocytes, osteoblasts, or SMCs in the appropriate induction medium. Treatment with basic fibroblast growth factor increased proliferation and prevented myogenic differentiation, suggesting that basic fibroblast growth factor can be used to expand the population of undifferentiated HFCs to the large numbers needed for therapeutic applications. SMCs were isolated from HFCs using tissue-specific promoters and flow cytometry sorting. Cylindrical vascular constructs engineered with HF-SMCs showed remarkable contractility in response to receptor and nonreceptor agonists such KCl, endothelin-1, and the thromboxane mimetic, U46619, as well as superior mechanical properties compared to their counterparts with human vascular SMCs. Our results suggest that HF is a rich source of mesenchymal stem cells with great potential for myogenic differentiation providing functional SMCs for tissue regeneration and cell therapies.

FIG. 1.

HFC morphology and proliferation potential. HFCs were plated in six-well plates at 10⁴ cells/well and cultured in DMEM with 10% FBS alone (Ctrl) or with 2 ng/mL bFGF. (A) Photomicrograph of HFCs in the presence or absence of bFGF. (B) Cells were cultured in the presence or absence of bFGF and at the indicated times trypsinized, counted, and re-plated at the same density. The cumulative cell number was plotted as mean ± SD of three independent experiments (n = 3). HFCs, hair follicle stem cells; bFGF, basic fibroblast growth factor; DMEM, Dulbecco's modified Eagle's medium; FBS, fetal bovine serum; SD, standard deviation.

FIG. 1.

HFC morphology and proliferation potential. HFCs were plated in six-well plates at 10⁴ cells/well and cultured in DMEM with 10% FBS alone (Ctrl) or with 2 ng/mL bFGF. (A) Photomicrograph of HFCs in the presence or absence of bFGF. (B) Cells were cultured in the presence or absence of bFGF and at the indicated times trypsinized, counted, and re-plated at the same density. The cumulative cell number was plotted as mean ± SD of three independent experiments (n = 3). HFCs, hair follicle stem cells; bFGF, basic fibroblast growth factor; DMEM, Dulbecco's modified Eagle's medium; FBS, fetal bovine serum; SD, standard deviation.

FIG. 2.

HFC surface marker profile. Flow cytometry with antibodies for mesenchymal stem-cell-specific markers as indicated. Representative flow cytometry plots from three independent experiments are shown. Color images available online at www.liebertonline.com/ten.

FIG. 3.

HFCs demonstrated adipogenic and chondrogenic differentiation potential. HFCs were cultured in the adipogenic (A–D) or chondrogenic (E–G) differentiation medium for 2 weeks. (A) reverse-transcription (RT)-PCR for adipogenic markers aP2 and PPARγ. Oil Red O staining of HFCs that were treated with the adipogenic (B) or control (C) medium. (D) Higher magnification of (B). (E) RT-PCR for chondrogenic markers Sox9 and Col II. Type II collagen immunostaining of HFC pellets cultured with chondrogenic (F) or control (G) medium. Cells stained with secondary antibody only served as negative control. All samples were counterstained with 4′,6-diamidino-2-phenylindole (DAPI): to observe cell nuclei. Representative images are shown from one out of three independent experiments. PCR, polymerase chain reaction; GAPDH, glyceraldehyde 3-phosphate dehydrogenase. Color images available online at www.liebertonline.com/ten.

FIG. 4.

HFCs demonstrated osteogenic differentiation potential. HFCs were cultured in osteogenic differentiation medium for 2 weeks. (A) RT-PCR for osteogenic markers alkaline phosphatase (ALP), Runx2, and osteocalcin (OC). Alkaline phosphates activity of HFCs that were treated with the osteogenic (B) or control (C) medium. von Kossa staining of HFCs that were cultured in the osteogenic (D) or control (E) medium for 4 weeks. Representative images are shown from one out of three independent experiments. Color images available online at www.liebertonline.com/ten.

FIG. 5.

HFCs demonstrated myogenic differentiation potential and force generation ability. HFCs were cultured in the myogenic differentiation medium for 1 week. (A) RT-PCR for αSMA, and smoothelin. (B) Immunostaining for αSMA and calponin. Cells stained with Alexa 488 secondary antibody only served as negative control. All samples were counterstained with DAPI to observe the nuclei. Representative images are shown from one out of three independent experiments. (C) HFCs were cultured in the presence of bFGF (2 ng/mL) for 5 days until they reached 85%–90% confluence. At that time the cells were trypsinized and embedded in fibrin that was allowed to polymerize in 24-well plates to form disks. One hour after polymerization, the gels were detached from the walls and allowed to compact in the presence of FBS alone or supplemented with bFGF (2 ng/mL) or TGF-β1 (2 ng/mL). At the indicated times, gels were photographed and their area was measured using ImageJ software. The ratio of gel area at the indicated times over the initial area was plotted as percentage of initial hydrogel area over time. (D) HFCs were cultured in DMEM with 10% FBS alone or supplemented with bFGF (2 ng/mL) or TGF-β1 (2 ng/mL) for 5 days. At that time, the cells were embedded in fibrin hydrogels that were incubated with the medium of the same composition. The percentage of initial hydrogel area was plotted over time. All values represent the mean ± SD of triplicate samples in a representative experiment (n = 3). Asterisks denote p < 0.05 between the TGF-β1- and FBS-treated hydrogels at the same time point. bFGF-treated samples were significantly different (p < 0.05) than either the TGF-β1- or FBS-treated samples at all times. P-αSMA, smooth muscle alpha-actin promoter; TGF-β1, transforming growth factor-beta 1. Color images available online at www.liebertonline.com/ten.

FIG. 5.

HFCs demonstrated myogenic differentiation potential and force generation ability. HFCs were cultured in the myogenic differentiation medium for 1 week. (A) RT-PCR for αSMA, and smoothelin. (B) Immunostaining for αSMA and calponin. Cells stained with Alexa 488 secondary antibody only served as negative control. All samples were counterstained with DAPI to observe the nuclei. Representative images are shown from one out of three independent experiments. (C) HFCs were cultured in the presence of bFGF (2 ng/mL) for 5 days until they reached 85%–90% confluence. At that time the cells were trypsinized and embedded in fibrin that was allowed to polymerize in 24-well plates to form disks. One hour after polymerization, the gels were detached from the walls and allowed to compact in the presence of FBS alone or supplemented with bFGF (2 ng/mL) or TGF-β1 (2 ng/mL). At the indicated times, gels were photographed and their area was measured using ImageJ software. The ratio of gel area at the indicated times over the initial area was plotted as percentage of initial hydrogel area over time. (D) HFCs were cultured in DMEM with 10% FBS alone or supplemented with bFGF (2 ng/mL) or TGF-β1 (2 ng/mL) for 5 days. At that time, the cells were embedded in fibrin hydrogels that were incubated with the medium of the same composition. The percentage of initial hydrogel area was plotted over time. All values represent the mean ± SD of triplicate samples in a representative experiment (n = 3). Asterisks denote p < 0.05 between the TGF-β1- and FBS-treated hydrogels at the same time point. bFGF-treated samples were significantly different (p < 0.05) than either the TGF-β1- or FBS-treated samples at all times. P-αSMA, smooth muscle alpha-actin promoter; TGF-β1, transforming growth factor-beta 1. Color images available online at www.liebertonline.com/ten.

FIG. 6.

Derivation of SMCs from HFCs using tissue-specific promoters. (A) Schematics of lentiviral vectors encoding for ZsGreen under the control of αSMA promoter (P-αSMA) or DsRed under the P-MHC. (B) HFCs were transduced with either lentivirus and P-αSMA (ZsGreen+) or P-MHC (DsRed+) cells were sorted by flow cytometry. Fluorescence images of P-αSMA (ZsGreen+) or P-MHC (DsRed+) that were cultured in the presence or absence of bFGF (2 ng/mL). (C) Sorted cells were plated in six-well plates (10⁴ cells/well) and cultured in the presence or absence of bFGF (2 ng/mL). On day 7, the cells were trypsinized and counted, and the cell number was plotted as mean ± SD of triplicate samples in a representative experiment (n = 3). SMCs, smooth muscle cells; P-MHC, myosin heavy chain promoter. Color images available online at www.liebertonline.com/ten.

FIG. 6.

Derivation of SMCs from HFCs using tissue-specific promoters. (A) Schematics of lentiviral vectors encoding for ZsGreen under the control of αSMA promoter (P-αSMA) or DsRed under the P-MHC. (B) HFCs were transduced with either lentivirus and P-αSMA (ZsGreen+) or P-MHC (DsRed+) cells were sorted by flow cytometry. Fluorescence images of P-αSMA (ZsGreen+) or P-MHC (DsRed+) that were cultured in the presence or absence of bFGF (2 ng/mL). (C) Sorted cells were plated in six-well plates (10⁴ cells/well) and cultured in the presence or absence of bFGF (2 ng/mL). On day 7, the cells were trypsinized and counted, and the cell number was plotted as mean ± SD of triplicate samples in a representative experiment (n = 3). SMCs, smooth muscle cells; P-MHC, myosin heavy chain promoter. Color images available online at www.liebertonline.com/ten.

FIG. 7.

HFC-derived P-αSMA cells decrease αSMA expression in response to bFGF. P-αSMA cells were seeded at 10³ cells/cm² and cultured in the presence or absence of bFGF (2 ng/mL). On day 7, the percentage of ZsGreen+ cells (A) and mean green fluorescence intensity (B) of ZsGreen+ cells were measured by flow cytometry. All values are the mean ± SD of triplicate samples in a representative experiment (n = 3). (C, D) P-αSMA cells were fixed, permeabilized, and stained mouse anti-αSMA followed by incubation with Alex Fluor 647-R-phycoerythin goat anti-mouse secondary antibody. The percentage of cells and the level of αSMA expression were measured by flow cytometry. (C) Percentage of αSMA+ cells and (D) mean green fluorescence intensity of αSMA+ cells. All values are the mean ± SD of triplicate samples in a representative experiment (n = 3). (E) P-αSMA cells were cultured in the presence of bFGF (2 ng/mL) for 5 days before embedding in fibrin gels that were allowed to compact in DMEM with 10% FBS alone or supplemented with bFGF (2 ng/mL) or TGF-β1 (2 ng/mL). The percentage of initial hydrogel area was plotted over time. All values represent the mean ± SD of triplicate samples in a representative experiment (n = 3). Symbols (*) and (^#) denote significant difference (p < 0.05) between bFGF- and FBS- or bFGF- and TGF-β1-treated hydrogels, respectively, at the indicated time point.

FIG. 8.

Vascular constructs with P-αSMA cells expressed human vascular SMC markers. P-αSMA cells were embedded in fibrin hydrogels and cultured around 6-mm mandrels for 2 weeks to form rings. (A) Hematoxylin and eosin (H&E) staining showed uniform distribution of P-αSMA cells within fibrin hydrogels (scale bar = 100 μm). Immunostaining for (B) αSMA and (C) calponin (scale bar = 20 μm). Color images available online at www.liebertonline.com/ten.

FIG. 9.

Vascular constructs with P-αSMA cells demonstrated remarkable vasoreactivity. P-αSMA cells were embedded in fibrin hydrogels and cultured around 6-mm mandrels for 2 weeks to form rings. Vascular reactivity was measured using an isolated tissue bath system. (A) Representative graphs of isometric contraction over time in response to the indicated agonists. (B) Vascular reactivity (N/g dry tissue weight) in response to KCl (118 mM), endothelin (ET)-1 (20 nM), or U46619 (10⁻⁶ M). All values represent the mean ± SD of triplicate samples in a representative experiment (n = 3).