Effects of auricular chondrocyte expansion on neocartilage formation in photocrosslinked hyaluronic acid networks

Abstract

The overall objective of this study was to examine the effects of in vitro expansion on neocartilage formation by auricular chondrocytes photoencapsulated in a hyaluronic acid (HA) hydrogel as a next step toward the clinical application of tissue engineering therapies for treatment of damaged cartilage. Swine auricular chondrocytes were encapsulated either directly after isolation (p = 0), or after further in vitro expansion ( p = 1 and p = 2) in a 2 wt%, 50-kDa HA hydrogel and implanted subcutaneously in the dorsum of nude mice. After 12 weeks, constructs were explanted for mechanical testing and biochemical and immunohistochemical analysis and compared to controls of HA gels alone and native cartilage. The compressive equilibrium moduli of the p = 0 and p = 1 constructs (51.2 +/- 8.0 and 72.5 +/- 35.2 kPa, respectively) were greater than the p = 2 constructs (26.8 +/- 14.9 kPa) and the control HA gel alone (12.3 +/- 1.3 kPa) and comparable to auricular cartilage (35.1 +/- 12.2 kPa). Biochemical analysis showed a general decrease in glycosaminoglycan (GAG), collagen, and elastin content with chondrocyte passage, though no significant differences were found between the p = 0 and p = 1 constructs for any of the analyses. Histological staining showed intense and uniform staining for aggrecan, as well as greater type II collagen versus type I collagen staining in all constructs. Overall, this study illustrates that constructs with the p = 0 and p = 1 auricular chondrocytes produced neocartilage tissue that resembled native auricular cartilage after 12 weeks in vivo. However, these results indicate that further expansion of the chondrocytes (p = 2) can lead to compromised tissue properties.

Figure 1

General schematic of chondrocyte expansion, photoencapsulation, and subsequent analysis.

Figure 2

Explanted HA constructs 12 weeks after subcutaneous implantation in nude mice. Scale bar = 1cm.

Figure 3

Compressive equilibrium modulus of constructs after 12 weeks of subcutaneous implantation in nude mice compared to controls of the HA gel alone and native auricular and articular cartilage. The moduli of the p = 0 and p = 1 constructs are significantly greater than the HA gel and the average modulus of the p = 1 constructs is significantly greater than both the p = 2 constructs and auricular cartilage. Additionally, the average modulus of the articular cartilage is significantly greater than all other groups.

Figure 4

Water content of constructs after 12 weeks of subcutaneous implantation in nude mice compared to controls of the HA gel alone and auricular and articular cartilage. The HA gel exhibits a significantly greater water content than all other groups. The water content also showed slight increases with auricular chondrocyte passage number, although no significant differences were measured. The water content of auricular and articular cartilage is significantly lower than both the HA gel and the tissue engineered constructs.

Figure 5

DNA content normalized to wet weight for constructs after 12 weeks of subcutaneous implantation in nude mice compared to controls of the HA gel alone and native auricular and articular cartilage. No significant differences were detected among groups with the exception of p = 1 constructs versus articular cartilage. The HA gel showed insignificant DNA measurements and thus, exhibited no interference with the fluorescent assay.

Figure 6

Glycosaminoglycan content of samples normalized to construct wet weight after 12 weeks of subcutaneous implantation in nude mice compared to controls of the HA gel alone and native auricular and articular cartilage. Articular cartilage has significantly greater GAG content than all other groups while no significant difference was detected between auricular cartilage and the p = 0 constructs. In the engineered constructs, the GAG content generally decreased with chondrocyte passage, with the GAG content of the p = 0 and p = 1 constructs significantly greater than that of the p = 2 constructs. The GAG content detected in the HA gels was minimal and statistically lower than all constructs and native cartilage.

Figure 7

Collagen content of constructs normalized to construct wet weight after 12 weeks of subcutaneous implantation in nude mice compared to controls of the HA gel alone and native auricular and articular cartilage. Collagen content in articular cartilage is significantly greater than the HA gel and all engineered constructs. No significant difference was found between auricular cartilage and the p = 0 constructs, but there is significantly more collagen in auricular cartilage than the HA gel and the p = 1 and p = 2 constructs. In general, the collagen content decreased with chondrocyte passage.

Figure 8

Elastin content of constructs normalized to construct wet weight after 12 weeks of subcutaneous implantation in nude mice compared to controls of the HA gel alone and native auricular and articular cartilage. There was no significant difference between auricular cartilage when compared to either the p = 0 or p = 1 constructs. Minimal elastin was found in articular cartilage and the elastin content in the p = 2 constructs was significantly lower than that in the p = 0 and p = 1 constructs and auricular cartilage. No elastin was detected in the HA gels.

Figure 9

Histological sections of constructs stained for chondroitin sulfate and type I and type II collagen compared to non-immune controls (NIC), with no primary antibody, after 12 weeks of subcutaneous implantation in nude mice. Chondroitin sulfate is evenly distributed throughout all constructs with similar intensity, regardless of passage number. All constructs exhibited greater type II collagen staining versus type I collagen, where the p = 0 constructs exhibit the greatest intensity and distribution of type II collagen. Scale bar =100 μm.