Modulation of sirtuins during monolayer chondrocyte culture influences cartilage regeneration upon transfer to a 3D culture environment

Abstract

This study examined the role of sirtuins in the regenerative potential of articular chondrocytes. Sirtuins (SIRT1-7) play a key role in regulating cartilage homeostasis. By inhibiting pro-inflammatory pathways responsible for cartilage degradation and promoting the expression of key matrix components, sirtuins have the potential to drive a favourable balance between anabolic and catabolic processes critical to regenerative medicine. When subjected to osmolarity and glucose concentrations representative of the in vivo niche, freshly isolated bovine chondrocytes exhibited increases in SIRT1 but not SIRT3 gene expression. Replicating methods adopted for the in vitro monolayer expansion of chondrocytes for cartilage regenerative therapies, we found that SIRT1 gene expression declined during expansion. Manipulation of sirtuin activity during in vitro expansion by supplementation with the SIRT1-specific activator SRT1720, nicotinamide mononucleotide, or the pan-sirtuin inhibitor nicotinamide, significantly influenced cartilage regeneration in subsequent 3D culture. Tissue mass, cellularity and extracellular matrix content were reduced in response to sirtuin inhibition during expansion, whilst sirtuin activation enhanced these measures of cartilage tissue regeneration. Modulation of sirtuin activity during monolayer expansion influenced H3K27me3, a heterochromatin mark with an important role in development and differentiation. Unexpectedly, treatment of primary chondrocytes with sirtuin activators in 3D culture reduced their matrix synthesis. Thus, modulating sirtuin activity during the in vitro monolayer expansion phase may represent a distinct opportunity to enhance the outcome of cartilage regenerative medicine techniques.

Keywords: cartilage tissue engineering; chondrocyte; glucose restriction; nicotinamide adenine dinucleotide; pellet culture; sirtuin.

FIGURE 1

Osmolarity and glucose concentration regulate SIRT1 but not SIRT3 gene expression. (A,B) Gene expression of (A) sirtuin-1 (SIRT1) and (B) sirtuin-3 (SIRT3) normalised to β-2-microglobulin (B2M) endogenous control and presented relative to 250 mOsm. Mean ± s.e.m., n = 4 donors represented by unique symbols, General linear model with Dunnet multiple comparisons against 250 mOsm: *p = 0.011, **p = 0.005, ****p < 0.0001. (C,D) Gene expression of (C) SIRT1 and (D) SIRT3 normalised to β-actin (ACTB) endogenous control and presented relative to 10 mM glucose. Mean ± s.e.m., n = 4 donors represented by unique symbols, paired t-test: *p = 0.030, n. s. P > 0.05. Donors in A,B are distinct from donors in C,D.

FIGURE 2

Cartilage matrix and sirtuin 1 gene expression declines during in vitro monolayer expansion of primary chondrocytes. (A–C) Gene expression of (A) aggrecan (ACAN), (B) collagen type II α-1 chain (COL2A1) and (C) sirtuin 1 (SIRT1) is normalised to β-2-microglobulin (B2M) and β-actin (ACTB) endogenous control genes and presented relative to 0 population doublings (PDs). Mean ± s.e.m., n = 4 from two experiments with individual data points represented by open and closed circles respectively, General linear model with Tukey pairwise comparisons: *p < 0.05, **p < 0.01 and ***p < 0.001 vs. 0 PDs, ^# p < 0.05, ^## p < 0.01 and ^### p < 0.001 vs. 2 PDs, ^††† p < 0.001 vs. 4 PDs. (D) Representative nuclear SIRT1 immunofluorescent staining. Scale bar 5 µm. (E) Nuclear SIRT1 protein levels relative to 2 PDs. Boxes represent median and interquartile range, with whiskers extending to 1.5 × interquartile range or the max/min data points, n _2PD = 140, n _4PD = 113, n _25PD = 354 cells, General linear model with Tukey pairwise comparisons: ^### p < 0.001 vs. 2 PDs, ^††† p < 0.001 vs. 4 PDs. (F,G) Scatter plots of (F) ACAN, and (G) COL2A1 against SIRT1 expression. n = 20 from two experiments with individual data points represented by open and closed circles respectively, Pearson correlation coefficient (r) and p-value < 0.001.

FIGURE 3

Chondrocytes have similar morphology and early proliferation kinetics in the presence or absence of the sirtuin inhibitor, NAM. (A) Chondrocytes expanded in monolayer exhibited similar morphology regardless of glucose concentration or sirtuin inhibition with NAM, illustrated by representative images at day 6. Scale bar 100 µm. (B,C) Proliferation kinetics data for chondrocytes expanded for 4PD in (B) 10 mM glucose and (C) 1 mM glucose with and without NAM. Data from two experiments indicated by circles and squares. The population doubling times calculated from the linear region of the population doubling/days in monolayer plots (B,C) were 1.49 ± 0.11 and 1.47 ± 0.12 days (mean ± s.e.m.) for control and NAM treated cells respectively. The final population doubling number at which cells were harvested to prepare cell pellets, indicated by the dashed line in (B) and (C) was 4.06 ± 0.08 and 4.05 ± 0.06 (mean ± s.e.m.) for control and NAM treated cells respectively. (D,E) Long-term proliferation kinetics in which a linear fit indicates the trajectory of the initial linear growth phase. During long-term culture, premature proliferative arrest was observed in the presence of NAM treatment for both 10 mM and 1 mM glucose media (D,E), illustrated by earlier deviation from linear growth compared to cells without NAM.

FIGURE 4

Inhibition of sirtuin activity during monolayer expansion abrogates chondrocyte regenerative capacity. (A) Representative cell pellets at day 28 of 3D culture created from chondrocytes which were either freshly isolated, 0 population doublings (0 PD), or monolayer expanded for 4 PDs in media containing either 10 mM or 1 mM glucose in the presence or absence of the sirtuin inhibitor nicotinamide (NAM). Scale bar 1 mm. (B) Pellet wet weight, (C) collagen content, (D) sulphated glycosaminoglycan (sGAG) content, and (E) DNA content at day 28. Mean ± s.e.m, n _0PD = 8 (B–E), n _{4PD 10mM Ctrl} = 9 (B) 7 (C–E), n _{4PD 10mM NAM} = 10 (B) 8 (C–E), n _{4PD 1mM Ctrl} = 10 (B) 6(C–E), n _{4PD 1mM NAM} = 12 (B) 6 (C–D) 7 (E) from two experiments indicated by symbol shape, General linear model with Tukey pairwise comparisons: *p < 0.05, ***p < 0.001; ⁺⁺⁺ p < 0.001 vs. 0 PD.

FIGURE 5

Treatment with the sirtuin inhibitor, NAM, decreased expression of SIRT1 and abundance of the epigenetic mark H3K27me3. (A) Treatment of primary (0 PD) chondrocytes for 24 h with 10 mM NAM significantly inhibited SIRT1 gene expression. Mean ± s.e.m., n = 4 from two experiments with individual data points represented by open and closed circles respectively, Mann-Whitney test: *p < 0.05. (B) Representative SIRT1 and H3K27me3 immunofluorescent nuclear staining in chondrocytes expanded for four population doublings (PDs) in media containing either 10 mM or 1 mM glucose in the presence or absence of the sirtuin inhibitor nicotinamide (NAM). Scale bar 5 µm. (C) Nuclear SIRT1 and (D) H3K27me3 levels in chondrocytes at 4 PDs relative to 10 mM glucose control cells. Boxes represent median and interquartile range, with whiskers extending to 1.5 × interquartile range or the max/min data points. (E,F) Scatter plots of nuclear H3K27me3 against nuclear SIRT1 intensity in 4 PD chondrocytes cultured in (E) 10 mM glucose or (F) 1 mM glucose in the presence or absence of NAM. n _{10mM Ctrl} = 270 cells, n _{10mM NAM} = 246 cells, n _{1mM Ctrl} = 416 cells, n _{1mM NAM} = 468 cells from two experiments. (C,D) Mann-Whitney test: ***p < 0.001. (E,F) Pearson correlation coefficient (r) and p-value presented.

FIGURE 6

Regeneration of cartilage tissue following transfer of monolayer cells to 3D pellet culture. (A) Representative tissue masses formed after 21 days culture of cell pellets created from (i) freshly isolated cells (0 PD) or (ii) monolayer expanded for four population doublings (4 PDs) in media containing 10 mM glucose supplemented with the pan-sirtuin activator, 100 µM NMN, or the SIRT1 specific activator, 100 nM SRT1720, or untreated controls. Scale units cm. (B) Pellet wet weight, (C) collagen content, (D) sulphated glycosaminoglycan (sGAG) content, and (E) DNA content at day 21 relative to the 4 PD control. Mean ± s.e.m, n _0PD = 12, n _4PD = 18 from three donors indicated by symbol shape. 2-factor ANOVA with replication and bonferroni correction: *p < 0.05, **p < 0.01 ***p < 0.001, ^ns p > 0.05. The cell number seeded to pellets for each treatment group is shown in Supplementary Figure S3.

FIGURE 7

Regeneration of cartilage tissue by freshly isolated (0 PD) chondrocytes is impaired by the presence of sirtuin activators during 3D pellet culture. (A) Cell pellet wet weights, (B) collagen, (C) sulphated glycosaminoglycan and (D) DNA content after 21 days pellet culture in the presence or absence of sirtuin activators 100 µM NMN or 100 nM SRT1720. DNA content is normalised to pre-culture pellet values at day 0. n = 6 for all groups except SRT1720 where n = 3. The freshly isolated (0 PD) cells were pooled from four donors before pellet formation. Two-way t-test with Bonferroni correction: ***p < 0.001 ****p < 0.0001, ^NS p > 0.05.