Membrane potential controls adipogenic and osteogenic differentiation of mesenchymal stem cells

Abstract

Background: Control of stem cell behavior is a crucial aspect of developmental biology and regenerative medicine. While the functional role of electrophysiology in stem cell biology is poorly understood, it has become clear that endogenous ion flows represent a powerful set of signals by means of which cell proliferation, differentiation, and migration can be controlled in regeneration and embryonic morphogenesis.

Methodology/principal findings: We examined the membrane potential (V(mem)) changes exhibited by human mesenchymal stem cells (hMSCs) undergoing adipogenic (AD) and osteogenic (OS) differentiation, and uncovered a characteristic hyperpolarization of differentiated cells versus undifferentiated cells. Reversal of the progressive polarization via pharmacological modulation of transmembrane potential revealed that depolarization of hMSCs prevents differentiation. In contrast, treatment with hyperpolarizing reagents upregulated osteogenic markers.

Conclusions/significance: Taken together, these data suggest that the endogenous hyperpolarization is a functional determinant of hMSC differentiation and is a tractable control point for modulating stem cell function.

Figure 1. V_mem hyperpolarization exhibited by OS- and AD-differentiated cells.

(A) Cell culture timeline for V_mem studies. Cells were seeded in control medium, then switched to OS or AD differentiation medium (OS or AD) at various time points over the course of 4 weeks. After 4 weeks, cells that had differentiated for a total of 0, 1, 2, 3, or 4 weeks (samples 0wk-diff, 1wk-diff, 2wk-diff, 3wk-diff, and 4wk-diff, respectively) were imaged on the same day. (B) Fluorescence measurements from cells cultured according to the timeline in OS or AD media. Cells were stained with the voltage-sensitive dye DiSBAC, which exhibits higher intensity with membrane depolarization. Data points are mean pixel intensity±standard deviation (N = 5–15 cell fields). Marked samples are statistically different, * relative to 0wk-diff OS sample (p<0.0005), § relative to 4wk-diff OS sample (p<0.0005), # relative to 0wk-diff AD sample (p<0.0005), † relative to 3wk-diff AD sample (p<0.0002), ‡ relative to 4wk-diff AD sample (p<0.005).

Figure 2. Measurement of resting and depolarized membrane potentials during OS and AD differentiation.

(A) Intracellular recordings of resting and depolarized membrane potentials (V_mem) in hMSCs during OS and AD differentiation. Cells were impaled individually and the V_mem recorded until a stable baseline was reached (pre-treatment), then 10 nM ouabain (OS-ouab, AD-ouab samples) or 80 mM K⁺ (OS-K, AD-K) was added and the V_mem recorded until a new equilibrium was reached (post-treatment). Data points are mean potentials±standard deviation (N = 6–7 cells). Marked samples are statistically different, * relative to respective pre-treatment samples (p<0.03), # relative to AD-ouab post-treatment sample (p<0.04). (For clarity, statistical significances marked by # are reported among post-treatment samples only.) (B) Intensities of DiSBAC₂(3)-loaded cells at resting and depolarized potentials during OS and AD differentiation. Pre-treatment values are the fluorescence intensities of OS and AD cells at rest, while post-treatment values are the fluorescence intensities after depolarization with 10 nM ouabain (OS-ouab, AD-ouab) or 80 mM K⁺ (OS-K, AD-K). Data points are mean pixel intensity±standard deviation (N = 15–20 cell fields). Marked samples are statistically different, * relative to respective pre-treatment samples (p≪0.0001), # relative to OS-K post-treatment sample (p≪0.0001), † relative to AD-K post-treatment sample (p≪0.0001). (For clarity, statistical significances marked by # and † are reported among post-treatment samples only).

Figure 3. Depolarization suppresses AD gene expression.

PPARG and LPL expression were suppressed on Days 2, 7, 14, and 22 by addition of 80 mM K⁺ (AD-80K) during AD differentiation. Similarly, PPARG and LPL expression were suppressed on Days 7, 14, and 22 by addition of 10 nM ouabain (AD-ouab) during AD differentiation. Data points are mean relative expression±standard deviation (N = 6). Marked samples are statistically different, * relative to PPARG expression of untreated AD samples (p<0.05), † relative to LPL expression of untreated AD samples (p<0.003), ‡ relative to LPL expression of AD-80K samples (p<0.0005). (For clarity, statistical significances are reported among samples taken within the same day.) Undiff, hMSCs cultured in control medium; AD, hMSCs cultured in AD medium; AD-80K, hMSCs cultured in AD medium supplemented with 80 mM K⁺; AD-ouab, hMSCs cultured in AD medium supplemented with 10 nM ouabain.

Figure 4. Depolarization reduces fat droplet accumulation during AD differentiation.

Oil Red O staining revealed less accumulation of fat droplets in K⁺- and ouabain-treated AD cells. Depolarized cells were cultured in AD medium supplemented with 80 mM K⁺ (AD-K; images A, B) or 10 nM ouabain (AD-ouab; images C, D). Control cells were cultured in AD medium (AD; images E, F) or control medium (Undiff; images G, H). After 7 days, cells were stained with Oil Red O and imaged at magnifications of 100× (images A, C, E, G; scale bar = 100 µm) and 320× (images B, D, F, H; scale bar = 40 µm). Oil Red O was then extracted from each sample and measured spectrophotometrically (I). Data points are mean absorbance±standard deviation (N = 6). Marked samples are statistically different, * relative to untreated AD samples (p<0.009).

Figure 5. Shorter, earlier depolarization times are sufficient to suppress AD differentiation.

Cells were exposed to 80 mM K⁺ (AD-K) or 10 nM ouabain (AD-ouab) during Days 1–2 (A) or Days 1–4 (B), then washed and continued in culture in AD medium. Gene expression was evaluated on Day 7. Two days of exposure to 80 mM K⁺ or four days of exposure to 10 nM ouabain was sufficient to effect a change in AD marker expression. Data points are mean relative expression±standard deviation (N = 6). Marked samples are statistically different, * relative to PPARG expression of untreated AD samples (p<0.002), # relative to LPL expression of untreated AD samples (p<0.002). Undiff, hMSCs cultured in control medium; AD, hMSCs cultured in AD medium; AD-80K, hMSCs cultured in AD medium supplemented with 80 mM K⁺; AD-ouab, hMSCs cultured in AD medium supplemented with 10 nM ouabain.

Figure 6. Membrane potential of AD and OS cells recovers after washout of early depolarization treatments.

hMSCs in AD or OS differentiation media were depolarized with 80 mM K⁺ (AD-K, OS-K) or 10 nM ouabain (AD-ouab, OS-ouab) during Days 1–4. Control cells were cultured in normal AD or OS media (AD or OS). Depolarization treatment was washed out after Day 4 and replaced with normal AD or OS media. Intracellular recordings were performed after washout on Days 5 or 6. Data points are mean potentials±standard deviation (N = 7–10 cells). Neither treated AD cells nor treated OS cells were statistically different from their respective untreated controls (p<0.05).

Figure 7. Depolarization suppresses OS gene expression.

Expression of ALP (A) and BSP (B) decreased by Day 7 of OS differentiation in response to addition of 40–80 mM K⁺ (OS-40K, OS-60K, OS-80K) or 10 nM ouabain (OS-ouab). Data points are mean relative expression±standard deviation (N = 6). Marked samples are statistically different, * relative to ALP expression of untreated OS samples (p<0.04), # relative to BSP expression of untreated OS samples (p<0.002). Undiff, hMSCs cultured in control medium; OS, hMSCs cultured in OS medium; OS-20K, OS-40K, OS-60K, OS-80K, hMSCs cultured in OS medium supplemented with 20, 40, 60, 80 mM K⁺, respectively; OS-ouab, hMSCs cultured in OS medium supplemented with 10 nM ouabain.

Figure 8. Depolarization suppresses ALP activity and reduces calcium content during OS differentiation.

(A) ALP activity decreased during OS differentiation in cells treated with 20–80 mM K⁺ (OS-20K, OS-40K, OS-60K, OS-80K) or 10 nM ouabain (OS-ouab). Data points are mean ALP activity units normalized to relative cell viability±standard deviation (N = 6). Marked samples are statistically different * relative to untreated OS samples (p<0.008). Undiff, hMSCs cultured in control medium; OS, hMSCs cultured in OS medium; OS-20K, OS-40K, OS-60K, OS-80K, hMSCs cultured in OS medium supplemented with 20, 40, 60, 80 mM K⁺, respectively; OS-ouab, hMSCs cultured in OS medium supplemented with 10 nM ouabain. (B) Total calcium content of cells undergoing OS differentiation was lowered by addition of 20–80 mM K⁺ (OS-20K, OS-40K, OS-60K, OS-80K) or 10 nM ouabain (OS-ouab). Data points are mean calcium content normalized to relative cell viability±standard deviation (N = 6). Marked samples are statistically different * relative to untreated OS samples (p<0.002). Undiff, hMSCs cultured in control medium; OS, hMSCs cultured in OS medium; OS-20K, OS-40K, OS-60K, OS-80K, hMSCs cultured in OS medium supplemented with 20, 40, 60, 80 mM K⁺, respectively; OS-ouab, hMSCs cultured in OS medium supplemented with 10 nM ouabain.

Figure 9. Shorter, earlier depolarization times are sufficient to suppress OS differentiation.

Cells were exposed to 80 mM K⁺ or 10 nM ouabain during Days 1–2 (A) or Days 1–4 (B), then washed and continued in culture in OS medium. Gene expression was evaluated on Day 7. Two or four days of exposure to depolarization treatment was sufficient to effect a change in OS marker expression. Data points are mean relative expression±standard deviation (N = 6). Marked samples are statistically different * relative to ALP expression of untreated OS samples (p<0.009), # relative to BSP expression of untreated OS samples (p<0.04). Undiff, hMSCs cultured in control medium; OS, hMSCs cultured in OS medium; OS-80K, hMSCs cultured in OS medium supplemented with 80 mM K⁺; OS-ouab, hMSCs cultured in OS medium supplemented with 10 nM ouabain.

Figure 10. Hyperpolarization upregulates OS gene expression.

(A) K-_ATP-channel openers pinacidil and diazoxide hyperpolarized hMSCs undergoing OS differentiation. Cells were impaled individually and the V_mem recorded until a stable baseline was reached (pre-treatment), then 10 µM pinacidil or diazoxide was added and the V_mem recorded until a new equilibrium was reached (post-treatment). Data points are mean potentials±standard deviation (N = 5 cells). Marked samples are statistically different * relative to respective pre-treatment samples (p<0.04). (B, C) Exposure to K-_ATP-channel openers pinacidil (B) and diazoxide (C) resulted in slight upregulation of OS markers compared to untreated cells. When treated with 1 and 10 µM pinacidil (OS-1pin and OS-10pin, respectively), cells showed upregulated BSP expression compared to untreated OS cells (p<0.04). When treated with 10 and 100 µM diazoxide, cells upregulated ALP and BSP expression compared to untreated OS cells. Data points are mean relative expression±standard deviation (N = 6). Marked samples are statistically different * relative to ALP expression of untreated OS samples (p<0.05), # relative to BSP expression of untreated OS samples (p<0.05). Undiff, hMSCs cultured in control medium; OS, hMSCs cultured in OS medium; OS-80K, hMSCs cultured in OS medium supplemented with 80 mM K⁺; OS-ouab, hMSCs cultured in OS medium supplemented with 10 nM ouabain.