Original video: https://www.youtube.com/watch?v=Mu19qFRSfPU
Sequential voxel removal beneath the sepal inner epidermis in longitudinal sections using the clipping plane to fully expose the inner epidermis. The video shows the 3D dimensional view of a typical Arabidopsis stage 7 flower, with cells on both epidermal layers of the outer sepal segmented. To expose the inner epidermis, voxels within a given longitudinal section defined by the clipping plane are removed. The steps for voxel removal involve tilting and moving the clipping plane to the adjacent regions, resulting in a complete view of the inner epidermis. The 3D image of the inner epidermis can then be used to generate a mesh surface onto which the membrane signal is projected for segmenting the cells.
To delete the remaining floral meristem voxels underneath the inner epidermis and expose it for viewing, we undertake a two-step approach. Firstly, we move the clipping plane to the adjacent section while having some overlap with the previously exposed region. Having an overlap reveals a residual border from voxel removal in the preceding section (Figure 3H). This border serves as a guide for subsequent voxel removal in the adjacent section. This strategy of using the border as a guide ensures uniform voxel removal by eliminating the chances of under- or over- removal of voxels across discrete longitudinal sections. Maintaining this uniformity is critical for obtaining a smooth mesh surface and accurately projecting membrane signal in subsequent steps. Secondly, we tilt the clipping plane such that it is roughly perpendicular to the curvature of the sepal in the region from which we intend to remove voxels. ‘Voxel Edit’ deletes voxels all the way through the thickness of the section. Since the sepal is a curved tissue, the clipping plane must be perpendicular so that voxels associated with the inner epidermal layer on the opposite side are not accidentally deleted.
The two steps (tilting and moving the clipping plane for voxel removal) are repeatedly performed throughout the flower in a longitudinal section-by-section manner, until we achieve a complete view of the inner epidermis. Note that all the steps described above are error-prone; whether voxel removal was optimally performed or not can only be assessed later, based on how good the membrane signal projection is on the mesh surface (described in the next section). Therefore, saving our progress regularly is a smart practice that allows us to revert to any previous versions if any anomalies arise, without having to start all over. To save the current image version, we simply copy the image in the ‘Work’ stack back to the ‘Main’ stack using the process ‘Stack/Multistack/Copy Work to Main Stack’ and then save the stack.
Overall, the steps described above allow one to view both the inner and outer epidermis of growing sepals without the need for physically dissecting the sepal. Although we illustrate the ‘voxel removal’ steps using the 48-hour time point, the same can be applied to visualize the inner epidermis for the other time-points.
Keywords: growth, 2.5D segmentation, live imaging, image processing, deep tissue imaging, Arabidopsis, MorphoGraphX, sepals
Citation: Singh Yadav A and Roeder AHK (2024) An optimized live imaging and multiple cell layer growth analysis approach using Arabidopsis sepals. Front. Plant Sci. 15:1449195. doi: https://doi.org/10.3389/fpls.2024.1449195
Received: 14 June 2024; Accepted: 29 July 2024;
Published: 03 September 2024.
Attribution 4.0 International — CC BY 4.0 - Creative Commons
Comments