Wide-Range Segmentation of Cotyledon Epidermal Cells for Morphometrical Analysis and Mechanical Simulation

Smooth Elongation of Pavement Cells Induced by RIC1 Overexpression Leads to Marginal Protrusions of the Cotyledon in Arabidopsis thaliana

The interdigitated pavement cell shape is suggested to be mechanically rational at both the cellular and tissue levels, but the biological significance of the cell shape is not fully understood. In this study, we explored the potential importance of the jigsaw puzzle-like cell shape for cotyledon morphogenesis in Arabidopsis. We used a transgenic line overexpressing a Rho-like GTPase-interacting protein, ROP-INTERACTIVE CRIB MOTIF-CONTAINING PROTEIN 1 (RIC1), which causes simple elongation of pavement cells. Computer-assisted microscopic analyses, including virtual reality observation, revealed that RIC1 overexpression resulted in abnormal cotyledon shapes with marginal protrusions, suggesting that the abnormal organ shape might be explained by changes in the pavement cell shape. Microscopic, biochemical and mechanical observations indicated that the pavement cell deformation might be due to reduction in the cell wall cellulose content with alteration of cortical microtubule organization. To examine our hypothesis that simple elongation of pavement cells leads to an abnormal shape with marginal protrusion of the cotyledon, we developed a mathematical model that examines the impact of planar cell growth geometry on the morphogenesis of the organ that is an assemblage of the cells. Computer simulations supported experimental observations that elongated pavement cells resulted in an irregular cotyledon shape, suggesting that marginal protrusions were due to local growth variation possibly caused by stochastic bias in the direction of cell elongation cannot be explained only by polarity-based cell elongation, but that an organ-level regulatory mechanism is required.

Machine learning and feature analysis of the cortical microtubule organization of Arabidopsis cotyledon pavement cells

The measurement of cytoskeletal features can provide valuable insights into cell biology. In recent years, digital image analysis of cytoskeletal features has become an important research tool for quantitative evaluation of cytoskeleton organization. In this study, we examined the utility of a supervised machine learning approach with digital image analysis to distinguish different cellular organizational patterns. We focused on the jigsaw puzzle-shaped pavement cells of Arabidopsis thaliana. Measurements of three features of cortical microtubules in these cells (parallelness, density, and the coefficient of variation of the intensity distribution of fluorescently labeled cytoskeletons [as an indicator of microtubule bundling]) were obtained from microscopic images. A random forest machine learning model was then used with these images to differentiate mutant and wild type, and Taxol-treated and control cells. Using these three metrics, we were able to distinguish wild type from bpp125 triple mutant cells, with approximately 80% accuracy; classification accuracy was 88% for control and Taxol-treated cells. Different features contributed most to the classification, namely, coefficient of variation for the wild-type/mutant cells and parallelness for the Taxol-treated/control cells. The random forest method used enabled quantitative evaluation of the contribution of features to the classification, and partial dependence plots showed the relationships between metric values and classification accuracy. While further improvements to the method are needed, our small-scale analysis shows the potential for this approach in large-scale screening analyses.

Metal-Nano-Ink Coating for Monitoring and Quantification of Cotyledon Epidermal Cell Morphogenesis

During cotyledon growth, the pavement cells, which make up most of the epidermal layer, undergo dynamic morphological changes from simple to jigsaw puzzle-like shapes in most dicotyledonous plants. Morphological analysis of cell shapes generally involves the segmentation of cells from input images followed by the extraction of shape descriptors that can be used to assess cell shape. Traditionally, replica and fluorescent labeling methods have been used for time-lapse observation of cotyledon epidermal cell morphogenesis, but these methods require expensive microscopes and can be technically demanding. Here, we propose a silver-nano-ink coating method for time-lapse imaging and quantification of morphological changes in the epidermal cells of growing Arabidopsis thaliana cotyledons. To obtain high-resolution and wide-area cotyledon surface images, we placed the seedlings on a biaxial goniometer and adjusted the cotyledons, which were coated by dropping silver ink onto them, to be as horizontal to the focal plane as possible. The omnifocal images that had the most epidermal cell shapes in the observation area were taken at multiple points to cover the whole surface area of the cotyledon. The multi-point omnifocal images were automatically stitched, and the epidermal cells were automatically and accurately segmented by machine learning. Quantification of cell morphological features based on the segmented images demonstrated that the proposed method could quantitatively evaluate jigsaw puzzle-shaped cell growth and morphogenesis. The method was successfully applied to phenotyping of the bpp125 triple mutant, which has defective pavement cell morphogenesis. The proposed method will be useful for time-lapse non-destructive phenotyping of plant surface structures and is easier to use than the conversional methods that require fluorescent dye labeling or transformation with marker gene constructs and expensive microscopes such as the confocal laser microscope.