Adipose stem cells can secrete angiogenic factors that inhibit hyaline cartilage regeneration

Abstract

Introduction: Adipose stem cells (ASCs) secrete many trophic factors that can stimulate tissue repair, including angiogenic factors, but little is known about how ASCs and their secreted factors influence cartilage regeneration. Therefore, the aim of this study was to determine the effects ASC-secreted factors have in repairing chondral defects.

Methods: ASCs isolated from male Sprague Dawley rats were cultured in monolayer or alginate microbeads supplemented with growth (GM) or chondrogenic medium (CM). Subsequent co-culture, conditioned media, and in vivo cartilage defect studies were performed.

Results: ASC monolayers and microbeads cultured in CM had decreased FGF-2 gene expression and VEGF-A secretion compared to ASCs cultured in GM. Chondrocytes co-cultured with GM-cultured ASCs for 7 days had decreased mRNAs for col2, comp, and runx2. Chondrocytes treated for 12 or 24 hours with conditioned medium from GM-cultured ASCs had reduced sox9, acan, and col2 mRNAs; reduced proliferation and proteoglycan synthesis; and increased apoptosis. ASC-conditioned medium also increased endothelial cell tube lengthening whereas conditioned medium from CM-cultured ASCs had no effect. Treating ASCs with CM reduced or abolished these deleterious effects while adding a neutralizing antibody for VEGF-A eliminated ASC-conditioned medium induced chondrocyte apoptosis and restored proteoglycan synthesis. FGF-2 also mitigated the deleterious effects VEGF-A had on chondrocyte apoptosis and phenotype. When GM-grown ASC pellets were implanted in 1 mm non-critical hyaline cartilage defects in vivo, cartilage regeneration was inhibited as evaluated by radiographic and equilibrium partitioning of an ionic contrast agent via microCT imaging. Histology revealed that defects with GM-cultured ASCs had no tissue ingrowth from the edges of the defect whereas empty defects and defects with CM-grown ASCs had similar amounts of neocartilage formation.

Conclusions: ASCs must be treated to reduce the secretion of VEGF-A and other factors that inhibit cartilage regeneration, which can significantly influence how ASCs are used for repairing hyaline cartilage.

Figure 1

Effects of microencapsulation and chondrogenic medium (CM) on angiogenic factor expression and secretion. (A) Diagram of adipose stem cell (ASC) monolayer (ML) and microbead (μB) treatments with growth medium (GM) and CM prior to mRNA and media collections. (B) mRNA levels isolated after 5 days of culture and (C) growth factor production over the last 24 hours from ASCs, chondrocytes (chond), and liver cells (liv) (n = 6 ± SE). ^*P < 0.05 vs. chond; ^#P < 0.05 vs. ASCs; ^{^}P < 0.05 vs. ASC microbeads (ASC + μB).

Figure 2

Effects of adipose stem cell (ASC) co-culture and ASC-conditioned media on chondrocyte mRNA levels. (A) Diagram of ASC co-culture and (B) mRNA levels of chondrocytes after 7 days. (C) Diagram of ASC-conditioned media treatment and (D) mRNA levels of chondrocytes after 12 hours. X-axis labels refer to treatments and culture type for ASCs and liver cells (Liv) prior to co-culture and conditioning, which include growth medium (GM), chondrogenic medium (CM), monolayer (ML), and microbead (μB) (n = 6 ± SE). ^*P < 0.05 vs. control; ^#P < 0.05 vs. ASCs, ^{^}P < 0.05 vs. ASC microbeads (ASCs + μB).

Figure 3

Effects of adipose stem cell (ASC)-conditioned medium on chondrocyte phenotype, proliferation, apoptosis, and angiogensis. (A) Diagram of ASC-conditioned media experiments, (B) [³⁵S]-sulfate incorporation, (C) alkaline phosphatase activity, (D) [³H]-thymidine incorporation, (E) caspase-3 activity, (F) bax/bcl2 mRNA levels, (G) DNA fragmentation, and (H) endothelial tube length. All experiments were assayed after 24 hours of conditioned media treatment except for bax/bcl2 mRNA levels and endothelial tube length, which were measured after 12 hours of treatment. X-axis labels refer to treatments and culture type for chondrocytes (Chond), ASCs, and liver cells (Liv) prior to conditioning, which include growth medium (GM), chondrogenic medium (CM), monolayer (ML), and microbead (μB) (n = 6 ± SE). ^*P < 0.05 vs. control; ^#P < 0.05 vs. ASCs; ^{^}P < 0.05 vs. ASC microbeads (ASCs + μB).

Figure 4

Effects of exogenous vascular endothelial growth factor (VEGF)-A and fibroblast growth factor (FGF)-2 on chondrocytes. (A) [³⁵S]-sulfate incorporation of chondrocytes treated with recombinant human VEGF-A and FGF-2, (B) caspase-3 activity of chondrocytes treated with recombinant human VEGF-A and FGF-2, and (C) [³H]-thymidine incorporation of chondrocytes treated with recombinant human VEGF-A and FGF-2 after 24 hours of treatment (n = 6 ± SE). ^*P < 0.05 vs. control; ^#P < 0.05 vs. adipose stem cells (ASCs).

Figure 5

Effects of adipose stem cell (ASC)-secreted vascular epithelial growth factor (VEGF)-A and fibroblast growth factor (FGF)-2 on chondrocytes. (A) Schematic outlining chondrocytes treated with ASC-conditioned medium with VEGF-A and FGF-2 neutralizing antibodies and assayed for (B) [³⁵S]-sulfate incorporation, (C) caspase-3 activity, and (D) [³H]-thymidine incorporation after 24 hours of treatment (n = 6 ± SE). ^*P < 0.05 vs. control; ^#P < 0.05 vs. ASCs.

Figure 6

Effects of adipose stem cells (ASCs) on cartilage regeneration. (A) Radiographic scoring with arrows highlighting the defect (n = 4 blinded scoring averages ± SE). ^*P < 0.05 vs. empty defect; ^#P < 0.05 vs. ASCs; ^{^}P < 0.05 vs. ASC + chondrogenic medium, CM). (B) Three-dimensional EPIC-micro computer tomography (μCT) images of xiphoids with ^*marking the defect and calculated cartilage volume within defects (n = 7 ± SE). ^*P < 0.05 vs. empty defect; ^#P < 0.05 vs. ASCs. (C) Representative H&E staining. Bar represents 100 μm at 20 × magnification (D = defect, × = xiphoid, AG = autograft).