Injectable tissue-engineered cartilage using a fibrin glue polymer.

Abstract

The purpose of this study was to demonstrate the feasibility of using a fibrin glue polymer to produce injectable tissue-engineered cartilage and to determine the optimal fibrinogen and chondrocyte concentrations required to produce solid, homogeneous cartilage. The most favorable fibrinogen concentration was determined by measuring the rate of degradation of fibrin glue using varying concentrations of purified porcine fibrinogen. The fibrinogen was mixed with thrombin (50 U/cc in 40 mM calcium chloride) to produce fibrin glue. Swine chondrocytes were then suspended in the fibrinogen before the addition of thrombin. The chondrocyte/polymer constructs were injected into the subcutaneous tissue of nude mice using chondrocyte concentrations of 10, 25, and 40 million chondrocytes/cc of polymer (0.4-cc injections). At 6 and 12 weeks, the neocartilage was harvested and analyzed by histology, mass, glycosaminoglycan content, DNA content, and collagen type II content. Control groups consisted of nude mice injected with fibrin glue alone (without chondrocytes) and a separate group injected with chondrocytes suspended in saline only (40 million cells/cc in saline; 0.4-cc injections). The fibrinogen concentration with the most favorable rate of degradation was 80 mg/cc. Histologic analysis of the neocartilage showed solid, homogeneous cartilage when using 40 million chondrocytes/cc, both at 6 and 12 weeks. The 10 and 25 million chondrocytes/cc samples showed areas of cartilage separated by areas of remnant fibrin glue. The mass of the samples ranged from 0.07 to 0.12 g at 6 weeks and decreased only slightly by week 12. The glycosaminoglycan content ranged from 2.3 to 9.4 percent for all samples; normal cartilage controls had a content of 7.0 percent. DNA content ranged from 0.63 to 1.4 percent for all samples, with normal pig cartilage having a mean DNA content of 0.285 percent. The samples of fibrin glue alone produced no cartilage, and the chondrocytes alone produced neocartilage samples with a significantly smaller mass (0.47 g at 6 weeks and 0.46 g at 12 weeks) when compared with all samples produced from chondrocytes suspended in fibrin glue (p < 0.03). Gel electrophoreses demonstrated the presence of type II collagen in all sample groups. This study demonstrates that fibrin glue is a suitable polymer for the formation of injectable tissue-engineered cartilage in the nude mouse model. Forty million chondrocytes per cc yielded the best quality cartilage at 6 and 12 weeks when analyzed by histology and content of DNA, glycosaminoglycan, and type II collagen.