Measuring plant cell shapes with PaCeQuant
Scientists from IPB and MLU team up to develop OpenSource software for life sciences
A. IPB researcher Katharina Bürstenbinder faced the challenge to quantify the cell shape features of mutants (oxIQD16, right) in a high-throughput manner and compare them with wild type Arabidopsis plants (Col-0, left). B. Birgit Möller and colleagues from the Halle University collaborated with Bürstenbinder to develop PaCeQuant – an OpenSource software that automatically recognizes leaf epidermal cells in microscopic images (left) and provides 27 cell shape parameters using unambiguously defined measuring points (right). This method allows a robust comparison of wildtype and mutant cell shapes. (Figure A: Möller et al. 2017, Plant Physiol, DOI: 10.1104/pp.17.00961; Figure B: http://mitobo.informatik.uni-halle.de/index.php/Applications/PaCeQuant).
Plant leaves are covered by a layer of epidermal cells that resemble irregular puzzle pieces which seamlessly interdigitate. Therefore, leaf epidermal cells are also called pavement cells, as they provide a dense, protective tissue at the leaf surface. Katharina Bürstenbinder, group leader at the Leibniz Institute of Plant Biochemistry Halle (IPB), encountered aberrant epidermis cell shapes when investigating proteins that regulate the plant cytoskeleton. Normally, every pavement cell has a number of protrusions, called lobes, that reach into the the neighbouring cell’s gaps between its lobes. Those gaps are called necks. Lobes and necks form when cell growth is stimulated or inhibited, respectively. For that, the cytoskeleton, which represents the cell’s inner support and transport scaffolding, is extended or degraded. When Bürstenbinder and colleagues manipulated specific cytoskeletal regulators, they observed changes in the overall shape and in the lobe shape of leaf epidermal cells (see Figure A). However, precisely comparing mutant and wild-type cell shapes required methods to quantify all the diverse shape features.
Therefore, Bürstenbinder teamed up with Birgit Möller of Martin-Luther-University Halle-Wittenberg and together, they developed a software called PaCeQuant – an acronym for Pavement Cell Quantification (Birgit Möller, Yvonne Poeschl, Romina Plötner, Katharina Bürstenbinder. PaCeQuant: A Tool for High-Throughput Quantification of Pavement Cell Shape Characteristics, Plant Physiology, Sep 2017, pp.00961.2017; DOI: 10.1104/pp.17.00961).
PaCeQuant allows the reliable automatic recognition of pavement cells in microscopy images, and additionally determines 27 exactly defined cell shape descriptors (Figure B). This proves to be an enormous reduction of workload, since otherwise the cell contours and distances to measure would have to be marked manually for every image. By automatizing the procedure, many more images can be processed in shorter time, and with that, the analyses become more reproducible and statistically robust. Moreover, every experimenter involuntarily introduces subjective errors, counting e.g. more or less lobes, as the researchers show.The automatic procedure eliminates such fluctuations introduced by manual image analysis.
PaCeQuant’s 27 cell shape descriptors comprise circumference and area of the pavement cell, the number of lobes per cell or the similarity with a perfect circle, for example. In addition, for the first time, the Halle scientists came up with unambiguous criteria to describe pavement cell-specific features like lobe size and shape. With these features, PaCeQuant offers the currently most comprehensive way to quantify pavement cell shape in comparison to all other methods published so far. Mutants with altered epidermal cell shapes can be compared in greater detail and more reliably now.
PaCeQuant is freely available as extension for the widely used OpenSource image analysis software ImageJ/Fiji. Thus, many labs can use PaCeQuant and produce comparable results or even adapt the software to fit their needs. As of now, interested parties can find a link to the PaCeQuant download site on our website "Databases". The Halle University’s website provides documentation for the software.