3D multicellular spheroids: regularization time for obtaining a homogeneous model

Text (Regular style): up to 450 words OBJECTIVE: 3D multicellular spheroids with homogeneous morpho-biological features are fundamental to obtain reliable data in biological experiments. This represents a compelling requirement especially for large-scale analysis, like High Content Screening (HCS). In this regard, the spheroidization time has been recently defined as the time required by freshly generated spheroids to lose their variability in sphericity. However, a wider definition of spheroids homogeneity can be attained considering other morphological parameters, such as the volume. Here, we widen the spheroidization time definition introducing the regularization time as the time required by spheroids to reach an asymptotic morpho-biological equilibrium point, ending drifts in sphericity and volume. In this work, we tested three generation systems to analyse initial morphological features and regularization time of the obtained spheroids. MATERIALS AND METHODS: Spheroids of Mesenchymal Stromal Cells have been generated using (a) the pellet culture method (PCM); spheroids of A549 cells using (b) a rotatory bioreactor (Synthecon Inc.) and (c) hanging drop plates (GravityPLUS, InSphero). For each generation system we chose an appropriate cell seeding density; additional densities have been tested with the hanging drop plates. The generated spheroids have been maintained in vitro for 30 days, during which we acquired brightfield images and computed the morphological features using AnaSP (http://sourceforge.net/p/anasp/) to determine the associated regularization times. RESULTS: We defined for the spheroidization time a threshold of 0.9 for sphericity and a Coefficient of Variation among volumes (CVv) of 0.20 for the regularization time. Since the bioreactor-generated spheroids strongly differed in the initial dimension, without pre-selecting volumes the 70% of the generated spheroids showed a 0.9 sphericity after 7 days, but never reached a 0.20 CVv. However, selecting at the beginning the spheroids with a volume of mean±std, a 0.20 CVv was achieved after approximately one week. 70% of spheroids generated with hanging drop plates showed a 0.9 sphericity after 7 days and a 0.20 CVv already one day after generation (i.e. no regularization time needed). More than 70% of the spheroids generated with PCM showed a 0.9 sphericity and a 0.20 CVv already one day after generation (i.e. no spheroidization and regularization time needed). Regarding the different cell seeding densities tested with the hanging drop plates, it is worth noting that all the generated spheroids evolved toward an equilibrium point of 400-micrometer diameter, which is the theoretical diameter for the origin of the necrotic core. CONCLUSION: These preliminary experiments show that each generation system is characterized by a different spheroidization and regularization time. Awareness of these times allows the operators to predict when the generated spheroids can be considered as a homogeneous model, and it is possible using them to obtain reliable and reproducible data in HCS experiments, accordingly.