Human bone marrow-derived MSC aggregates produced in the 96-well plates are larger in comparison to those produced using previous systems such as that described by Kato et al. (4
) (). Chondrogenic assessment by means of immunohistochemistry indicated no type I collagen in aggregates of 15-mL tubes or 96-well plates (). Conversely, type II collagen immunoreactivity was present throughout the extracellular matrix in both the tube and plate aggregates, except for a small centralized region in the tube aggregates (). Type X collagen immunoreactivity was also present throughout the extracellular matrix in both aggregate systems, with the exception of the outermost cell layers (). All negative controls remained unstained (not shown), indicating that the type I, II, and X collagen immunoreactivity was specific to the antigens of interest. There was no central necrosis detected in the plate or tube aggregates. However, this can become an issue, due to mass transport limitations, if the size of the aggregates is increased significantly beyond what is shown here. In this regard, however, we also did not notice any differences between plate and tube-based aggregates.
Figure 1. Microscopic characteristics of aggregates. (A and B) Toluidine blue-stained human bone marrow-derived mesenchymal stem cell (hMSC) aggregates prepared in 15-mL polypropylene tubes or 96-well plates cultured in chondrogenic medium for 3 weeks. (C-H) hMSC (more ...)
DNA analysis demonstrated higher DNA content in the plate aggregates compared to the tubes (). Similarly, glycosaminoglycan (GAG) analysis of the plate aggregates indicates higher GAG content per aggregate () and normalized to DNA content (). We have also successfully extracted RNA from 96-well plate aggregates. The typical RNA yields from aggregates grown in 96-well plates were in the range of 3-7 μg RNA per aggregate.
Figure 2. DNA and GAG content for hMSC aggregates made in 15-mL polypropylene tubes (black) or 96-well plates (gray). (A) DNA content for five replicate aggregates. (B) GAG content for five replicate aggregates. (C) GAG content for five replicate aggregates normalized (more ...)
The typical time course for chondrogenesis in aggregates has been described previously (6
). Briefly, however, metachromatic toluidine blue staining for GAG usually first appears around day 5, with type I collagen present from day 1, and decreasing through day 14. Type II collagen typically appears at day 5, while type X collagen is generally detected in the matrix by day 7, with both being detected throughout the aggregate by day 14. When cultures in tubes and plates were compared, there were no detectable differences between these markers at 1, 2, or 3 weeks, suggesting that chondrogenesis proceeds in parallel under both culture methods.
Biochemical analyses indicate that the larger size of the plate aggregates are a result of an increase in the number of cells as well as higher matrix production, which suggests that the 96-well plates are a seemingly better choice for the preparation of aggregates. Although our group has successfully used the Kato et al. (4
) protocol to promote chondrogenesis, the method is cumbersome, time-consuming, and expensive for experiments requiring large numbers of replicate cultures or study variables. We present here a protocol that will result in an overall cost and time reduction. A typical aggregate culture experiment for our group frequently consists of 400 aggregates or more, which would require a substantial amount of incubator space to contain the aggregates if using 15-mL tubes. At typical plastic-ware prices, replacing the tubes necessary for the production of 400 aggregates with 96-well plates results in a 90% cost savings. Another nontrivial improvement with our system is a decrease in the amount of time required for the preparation and maintenance of the aggregates. Using our method, 400 aggregates can be prepared and maintained in about one-fourth of the time needed when using the tubes. This time savings is due to the time required to uncap/recap and rack the tubes.
To adapt this assay for use with microplates, we tested a variety of 96-well polypropylene microplates (Fisherbrand®; Fisher Scientific) with different-shaped bottoms to select those which produce consistent aggregate formation and chondrogenic differentiation of bone marrow-derived MSCs. Microplates that had wells with wide-angle bottoms did not allow the cell-cell interactions that are necessary for the coalescence of the cells, while those with narrow angles limited the circumferential growth of the aggregates as new extracellular matrix was fabricated by the differentiating chondrocytes. The 96-well, V Bottom, 300-μL polypropylene plates from Phenix (cat. no. MPU-8355) were found to be the most effective for the production of our aggregates. Although polypropylene plates are normally intended for use in PCR applications and are supplied nonsterile, there are no apparent adverse effects from autoclave sterilization.
To summarize, we have successfully produced consistent and reproducible chondrogenic aggregates using this high-throughput culture system. This protocol can be a practical approach to examine multiple variables to study the chondrogenic potential of hMSCs.