1 Graduate School of Pharmaceutical Sciences, Nagoya University, Nagoya, Aichi, Japan.
2 Mitsubishi Tanabe Pharma Corporation, Yokohama, Japan.
SLAS Discov. 2019 Jan;24(1):47-56. doi: 10.1177/2472555218793374. Epub 2018 Aug 13.
The development of new drugs depends on the efficiency of drug screening. Phenotype-based screening has attracted interest due to its considerable potency for the discovery of first-in-class drugs. In general, fluorescently labeled imagery is the leading technique for phenotype-based screening; however, there are growing requirements to understand total culture profiles, which are unclear after end-point assays. In this study, we demonstrate that morphology-based cellular evaluation of unlabeled cells is an efficient approach to evaluate myotube formation assays. One of our aims was to study the myogenic differentiation process in C2C12 cells to discern the differences between cellular responses to different medium conditions (serum concentrations and insulin dosages). Our results show that predictive morphological profiles that strongly correlate with myogenic differentiation can be generated from myotube images, even in the confluent stage. The differentiation rate after 14 days can be quantitatively predicted with the highest accuracy by means of images taken on days 0-11.5. In addition, for the application of our morphology-based cellular evaluation, the effect of cyclic guanosine monophosphate (cGMP) on myogenic differentiation was analyzed. Our results show that the quantitated morphological profile from these images can be an effective descriptor for analysis of the myotube-recovering effect of cGMP.
新药的开发取决于药物筛选的效率。基于表型的筛选因其对发现首创类药物的巨大潜力而引起了关注。一般来说，荧光标记的图像是基于表型筛选的主要技术；然而，人们越来越需要了解总培养物的概况，而终点测定后这些情况尚不清楚。在本研究中，我们证明了对未标记细胞进行基于形态的细胞评估是评估肌管形成测定的有效方法。我们的目标之一是研究 C2C12 细胞中的成肌分化过程，以区分细胞对不同培养基条件（血清浓度和胰岛素剂量）的反应差异。我们的结果表明，即使在细胞汇合阶段，也可以从肌管图像中生成与成肌分化强烈相关的预测性形态学特征。通过对 0-11.5 天的图像进行定量分析，可在第 14 天以最高的准确度预测分化速度。此外，为了应用我们基于形态的细胞评估，分析了环鸟苷单磷酸（cGMP）对成肌分化的影响。我们的结果表明，从这些图像中定量的形态学特征可以作为分析 cGMP 对肌管恢复效果的有效描述符。
