Influence of Glucose Concentration on Colony-Forming Efficiency and Biological Performance of Primary Human Tissue-Derived Progenitor Cells

Abstract

Objective: Glucose concentrations used in current cell culture methods are a significant departure from physiological glucose levels. The study focuses on comparing the effects of glucose concentrations on primary human progenitors (connective tissue progenitors [CTPs]) used for cartilage repair.

Design: Cartilage- (Outerbridge grade 1, 2, 3; superficial and deep zone cartilage), infrapatellar fatpad-, synovium-, and periosteum-derived cells were obtained from 63 patients undergoing total knee arthroplasty and cultured simultaneously in fresh chondrogenic media containing 25 mM glucose (HGL) or 5 mM glucose (NGL) for pairwise comparison. Automated ASTM-based quantitative image analysis was used to determine colony-forming efficiency (CFE), effective proliferation rates (EPR), and sulfated-proteoglycan (GAG-ECM) staining of the CTPs across tissue sources.

Results: HGL resulted in increased cell cultures with CFE = 0 compared with NGL in all tissue sources (P = 0.049). The CFE in NGL was higher than HGL for superficial cartilage (P < 0.001), and contrary for synovium-derived CTPs (P = 0.046) when CFE > 0. EPR of the CTPs did not differ between the media in the 6-day assay time period (P = 0.082). The GAG-ECM area of the CTPs and their progeny was increased in presence of HGL (P = 0.027).

Conclusion: Glucose concentration is critical to progenitor's physiology and should be taken into account in the setting of protocols for clinical or in vitro cell expansion strategies.

Keywords: cartilage; cell therapy; colony-forming efficiency; glucose; stem and progenitors.

Figure 1.

Effect of glucose concentration on colony-forming efficiency (CFE = 0). (A) Percentage of total samples (n = 623 observations) where the CFE was zero for cells derived from tissue source studied (cartilage grade 1, 2, G1-2; cartilage grade 3, G3; superficial cartilage, C_sp; deep zone cartilage, C_dp; infrapatellar fat pad, IPFP; synovium, SYN; and periosteum, PERI) when treated with chondrogenic media with normal glucose levels (NGL, 5 mM) and high glucose levels (HGL, 25 mM). HGL group had significantly higher percent of samples where CFE = 0 than NGL group (P = 0.049). (B) Percentage of total sample where the CFE = 0 when treated with NGL- and HGL-containing chondrogenic media, respectively, for cells derived from each of the tissue source studied—G1-2 (n = 32), G3 (n = 32), C_sp (n = 32), C_dp (n = 28), IPFP (n = 64), SYN (n = 69), and PERI (n = 56). IPFP and SYN-derived cells and progenitors were the least affected by glucose concentrations in comparison with other tissue sources.

Figure 2.

Effect of glucose concentration on colony-forming efficiency (CFE ≠ 0). (A) Violin plot showing distribution of CFE (%) of all the tissue-derived connective tissue progenitors (CTPs; n = 555) when treated with chondrogenic media with normal glucose levels (NGL, 5mM) and high glucose levels (HGL, 25mM). The bottom, middle, and the top dotted lines represent the 25% quartile, median, and 75% quartile for the data, respectively. Wider sections of the violin plot denote a higher probability that members of the population will take the given value; the skinnier sections denote a lower probability. (B) Violin plot showing the distribution of CFE (%) for each of the tissue source–derived CTPs—G1-2 (n = 26), G3 (n = 30), C_sp (n = 29), C_dp (n = 24), IPFP (n = 59), SYN (n = 68), and PERI (n = 50). CFE was significantly lower in HGL than in NGL for C_sp-derived cells, whereas for SYN, the CFE in NGL was significantly lower than in HGL.

Figure 3.

Effect of glucose concentration on effective proliferation rate (EPR) of tissue-derived progenitors. (A) Violin plot showing distribution of EPR (divisions per day) of individual tissue-derived connective tissue progenitors (CTPs) when treated with chondrogenic media with normal glucose levels (NGL, 5mM) and high glucose levels (HGL, 25 mM). The bottom, middle, and the top dotted lines represent the 25% quartile, median, and 75% quartile for the data, respectively. Wider sections of the violin plot denote a higher probability that members of the population will take the given value; the skinnier sections denote a lower probability. It was interesting to note the 2 peaks in both treatment groups, suggesting that there are 2 populations of progenitors, a slower but larger population of progenitors and a faster but smaller population of progenitors. (B) Violin plot showing the distribution of EPR for each of the tissue source–derived CTPs. No significant differences were noted in the EPR between tissue sources or treatment groups. It was interesting to note that each tissue source has at least 2 distinct population of progenitors—faster and slower proliferating progenitors (denoted by the 2 peaks), with IPFP and SYN tissue sources containing the larger population of faster proliferating progenitors.

Figure 4.

Effect of glucose concentration on sulfated proteoglycans (GAG-ECM) production as measured by acridine orange (AO) stained area per cell. (A) Violin plot showing distribution of AO stained area per cell (µm²) of the colonies derived from the tissue-resident connective tissue progenitors (CTPs) when treated with chondrogenic media with normal glucose levels (NGL, 5 mM) and high glucose levels (HGL, 25 mM). The bottom, middle and the top dotted lines represent the 25% quartile, median, and 75% quartile for the data, respectively. Wider sections of the violin plot denote a higher probability that members of the population will take the given value; the skinnier sections denote a lower probability. Significantly higher magnitude of AO area/cell were seen in HGL- than in NGL-treated groups. (B) Violin plot showing the distribution of AO stained area per cell for each of the tissue source–derived CTPs. AO area per cell was significantly higher for cartilage-derived CTPs (G1-2, G3, C_dp, and C_sp) than from the IPFP, PERI, and SYN tissue sources.