Accurate and versatile 3D segmentation of plant tissues at cellular resolution
- EMBL, Germany
- Heidelberg University, Germany
- Technical University of Munich, Germany
- Max Planck Institute for Plant Breeding Research, Germany
- University of Warwick, United Kingdom
Abstract
Quantitative analysis of plant and animal morphogenesis requires accurate segmentation of individual cells in volumetric images of growing organs. In the last years, deep learning has provided robust automated algorithms that approach human performance, with applications to bio-image analysis now starting to emerge. Here, we present PlantSeg, a pipeline for volumetric segmentation of plant tissues into cells. PlantSeg employs a convolutional neural network to predict cell boundaries and graph partitioning to segment cells based on the neural network predictions. PlantSeg was trained on 1xed and live plant organs imaged with confocal and light sheet microscopes. PlantSeg delivers accurate results and generalizes well across different tissues, scales, acquisition settings even on non plant samples. We present results of PlantSeg applications in diverse developmental contexts. PlantSeg is free and open-source, with both a command line and a user-friendly graphical interface (https://github.com/hci-unihd/plant-seg).
Data availability
All data used in this study have been deposited in Open Science Framework: https://osf.io/uzq3w/Additionally Arabidopsis 3D Digital Tissue Atlas is available under https://osf.io/fzr56/
Article and author information
Author details
Funding
Deutsche Forschungsgemeinschaft (FOR2581)
- Jan U Lohmann
- Miltos Tsiantis
- Fred Hamprecht
- Kay Schneitz
- Alexis Maizel
- Anna Kreshuk
Leverhulme Trust (RPG-2016-049)
- George Bassel
The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
Reviewing Editor
- Dominique C Bergmann, Stanford University, United States
Version history
- Received: April 6, 2020
- Accepted: July 28, 2020
- Accepted Manuscript published: July 29, 2020 (version 1)
- Version of Record published: August 25, 2020 (version 2)
Copyright
© 2020, Wolny et al.
This article is distributed under the terms of the Creative Commons Attribution License permitting unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 11,147
- views
-
- 1,326
- downloads
-
- 151
- citations
Views, downloads and citations are aggregated across all versions of this paper published by eLife.
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Accurate and versatile 3D segmentation of plant tissues at cellular resolution
eLife 9:e57613.
https://doi.org/10.7554/eLife.57613
Further reading
-
- Plant Biology
Root gravitropic bending represents a fundamental aspect of terrestrial plant physiology. Gravity is perceived by sedimentation of starch-rich plastids (statoliths) to the bottom of the central root cap cells. Following gravity perception, intercellular auxin transport is redirected downwards leading to an asymmetric auxin accumulation at the lower root side causing inhibition of cell expansion, ultimately resulting in downwards bending. How gravity-induced statoliths repositioning is translated into asymmetric auxin distribution remains unclear despite PIN auxin efflux carriers and the Negative Gravitropic Response of roots (NGR) proteins polarize along statolith sedimentation, thus providing a plausible mechanism for auxin flow redirection. In this study, using a functional NGR1-GFP construct, we visualized the NGR1 localization on the statolith surface and plasma membrane (PM) domains in close proximity to the statoliths, correlating with their movements. We determined that NGR1 binding to these PM domains is indispensable for NGR1 functionality and relies on cysteine acylation and adjacent polybasic regions as well as on lipid and sterol PM composition. Detailed timing of the early events following graviperception suggested that both NGR1 repolarization and initial auxin asymmetry precede the visible PIN3 polarization. This discrepancy motivated us to unveil a rapid, NGR-dependent translocation of PIN-activating AGCVIII kinase D6PK towards lower PMs of gravity-perceiving cells, thus providing an attractive model for rapid redirection of auxin fluxes following gravistimulation.
-
- Plant Biology
Plant pathogens secrete proteins, known as effectors, that function in the apoplast or inside plant cells to promote virulence. Effector recognition by cell-surface or cytosolic receptors results in the activation of defence pathways and plant immunity. Despite their importance, our general understanding of fungal effector function and recognition by immunity receptors remains poor. One complication often associated with effectors is their high sequence diversity and lack of identifiable sequence motifs precluding prediction of structure or function. In recent years, several studies have demonstrated that fungal effectors can be grouped into structural classes, despite significant sequence variation and existence across taxonomic groups. Using protein X-ray crystallography, we identify a new structural class of effectors hidden within the secreted in xylem (SIX) effectors from Fusarium oxysporum f. sp. lycopersici (Fol). The recognised effectors Avr1 (SIX4) and Avr3 (SIX1) represent the founding members of the Fol dual-domain (FOLD) effector class, with members containing two distinct domains. Using AlphaFold2, we predicted the full SIX effector repertoire of Fol and show that SIX6 and SIX13 are also FOLD effectors, which we validated experimentally for SIX6. Based on structural prediction and comparisons, we show that FOLD effectors are present within three divisions of fungi and are expanded in pathogens and symbionts. Further structural comparisons demonstrate that Fol secretes effectors that adopt a limited number of structural folds during infection of tomato. This analysis also revealed a structural relationship between transcriptionally co-regulated effector pairs. We make use of the Avr1 structure to understand its recognition by the I receptor, which leads to disease resistance in tomato. This study represents an important advance in our understanding of Fol-tomato, and by extension plant–fungal interactions, which will assist in the development of novel control and engineering strategies to combat plant pathogens.
