In vitro growth factor-induced bio engineering of mature articular cartilage

Abstract

Articular cartilage maturation is the postnatal development process that adapts joint surfaces to their site-specific biomechanical demands. Maturation involves gross morphological changes that occur through a process of synchronised growth and resorption of cartilage and generally ends at sexual maturity. The inability to induce maturation in biomaterial constructs designed for cartilage repair has been cited as a major cause for their failure in producing persistent cell-based repair of joint lesions. The combination of growth factors FGF2 and TGFβ1 induces accelerated articular cartilage maturation in vitro such that many molecular and morphological characteristics of tissue maturation are observable. We hypothesised that experimental growth factor-induced maturation of immature cartilage would result in a biophysical and biochemical composition consistent with a mature phenotype. Using native immature and mature cartilage as reference, we observed that growth factor-treated immature cartilages displayed increased nano-compressive stiffness, decreased surface adhesion, decreased water content, increased collagen content and smoother surfaces, correlating with a convergence to the mature cartilage phenotype. Furthermore, increased gene expression of surface structural protein collagen type I in growth factor-treated explants compared to reference cartilages demonstrates that they are still in the dynamic phase of the postnatal developmental transition. These data provide a basis for understanding the regulation of postnatal maturation of articular cartilage and the application of growth factor-induced maturation in vitro and in vivo in order to repair and regenerate cartilage defects.

Fig. 1

FGF2 and TGFβ1 induce potent morphological changes in immature articular cartilage during in vitro culture. Explants were taken from adjacent sites from the same joint and cultured either in the absence (A) or continual presence (B) of 100 ng ml⁻¹ FGF2 and 10 ng ml⁻¹ TGFβ1 for 21 days in ITS containing serum-free medium. Growth factor-treated explants undergo significant resorption resulting in the disappearance of hypertrophic chondrocytes that reside in the deep zone of immature articular cartilage (bracketed in (A)]. Bar equals 500 μm. Polarising light microscopy of picro-sirius red stained sections of untreated (C) and growth factor-treated cartilage explants (D). The surface cartilage of growth factor treated cartilage displays extensive changes in collagen orientation many fibrils are anti-parallel to the surface axis (D). Also, a thin fluorescent line parallel to the surface (white arrows) delineates the lamina splendens, a collagen and lipid rich structure that is approximately 3 microns deep. This structure is absent in growth factor-treated cartilage. Bar equals 50 μm. Electron microscopy of surface chondrocytes (×7500) in untreated (E) and growth factor-treated (F) cartilage explants. Note the appearance of a thickened pericellular coat surrounding individual surface chondrocytes in growth factor-treated cartilage (black arrow in F) compared to untreated cartilage where this micro-anatomical unit of mature chondrocytes is absent (arrow in E). Also noteworthy is the increased collagen fibril density in growth factor-treated cartilage explants (F). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Changes in nanoscale elasticity and adhesive properties at the apical surface of articular cartilage. Representative load–displacement curves for nanoindendation analysis of cartilage samples (A). Despite having heterogeneous surfaces , freshly isolated immature cartilage explants exhibited significantly different ranges of both elasticity (B) and adhesion (D) when compared to their mature tissue sample counterparts. These differences are visualised in the boxplot representation and were shown to be significant through Mann–Whitney statistical analysis; P < 0.05. Similar results were observed for elasticity and adhesion measurements for immature growth factor-treated cartilage explants and their untreated controls (C, E). A significant (P < 0.05) increase in sample Young's modulus, and therefore, stiffening of the surface was observed following growth factor treatment when compared to the untreated control samples (C). A decrease in the adhesive status of the surface of growth factor-treated cartilage was observed compared to the control samples, a significant lowering of the 50% interquartile range is depicted in the boxplot analysis, indicating a reduction in the maximum force needed to withdraw the AFM stylus from the sample surface (E). This behavioural characteristic was reversed when analysing the approach curves and sample elasticity of cartilages where we observed an increase in the 50% interquartile range (B, C).

Fig. 3

AFM surface scan of freshly isolated immature and mature, and, serum-free (ITS) and growth factor-treated (ITS-FGF2-TGFβ1) cultured articular cartilages. Roughness analysis (seeTable 2) revealed that mature cartilages and samples treated with growth factors were smoother than immature cartilages or control samples. In addition, we also note that in surface scans, mature and growth factor-treated (ITS-FGF2-TGFβ1) cartilage appeared more fibrous and exhibit a less dimpled appearance than immature or cultured immature (ITS) cartilages.

Fig. 4

Biotribological analysis of articular cartilage explants. The coefficient of friction (CoF) was measured for freshly isolated immature (7-day-old) and mature (>18 month old) cartilages, and, in vitro cultured growth factor-treated and untreated immature cartilages (see Materials and Methods). The CoF of freshly isolated mature cartilages was significantly higher than their immature counterparts (P < 0.01). The CoF of growth factor-treated explants (ITS-FGF2-TGFβ1) was also significantly higher than untreated (ITS) explants (P < 0.01). The CoF of growth factor-treated cartilage explants increased approximately 3-fold (P < 0.05) following in vitro culture for 21 days compared to freshly isolated immature tissue.

Fig. 5

Collagen type I expression during in vivo and in vitro articular cartilage developmental maturation. In freshly isolated tissue, labelling for anti-collagen type I antibodies was observed as a broad and diffuse surface layer, and, localised as a thinner layer at the surface in mature cartilage (A). Labelling in cartilage explants cultured in serum-free (ITS) or growth factor medium (ITS-FGF2-TGFβ1) was more intense and was consolidated at the surface of these cartilages. Bar equals 50 μm. The ratio of collagen type IαI (in nanograms) normalised to 18S rRNA (in nanograms) is shown (B). Transcript levels of collagen type IαI decrease approximately 3-fold (P < 0.05) as cartilage matures, but following in vitro experimental maturation transcript levels increase approximately 6-fold compared to serum-free cultured (ITS) cartilage explants (P < 0.02).