Coen Lab - Research

From Flower Development, E Coen , Cell & Developmental Biology Department - JIC UK

Revision as of 15:57, 6 December 2010 by JeromeAvondo (Talk | contribs)
Jump to: navigation, search

Overview of Research Programme

Our research programme involves collaborations with several labs, including Computer scientists Andrew Bangham at the University of East Anglia and Przemyslaw Prusinkiewicz at the University of Calgary and ecological population geneticists Nick Barton at Vienna and Edinburgh and Christophe Thebaud at the University of Toulouse. We study developmental and evolutionary problems at a range of spatial and temporal scales, from subcellular events at minute intervals, to events at geographic scale over millions of years. Our long term goal is to link these different scales of analysis to give an integrated understanding of evolution and development. A critical component in our approach is storage and analysis of key images and data, such as confocal microscopy and OPT images1,2 and field data14.

A major focus of our group is on leaf development in Arabidopsis, which has the advantage of good genetics and convenience for imaging. Starting at the subcellular scale, we are analysing the role of microtubules and nuclear movements in the control of leaf cell division and growth. This involves extensive confocal imaging, image processing and computational modelling1,2,3. This work is closely related to studies at the cellular scale on division, growth and differentiation in developing leaves, involving a combination of time-lapse imaging and modelling5,6. At the tissue scale we are studying leaf development through clonal analysis, time-lapse imaging7 and Optical Projection Tomography1,2,9. Leaf growth is also being modelled at this scale by treating the leaf as a growing sheet of material8,10. In addition to these studies on leaf development, we are analysing the growth of flowers using similar approaches in both Arabidopsis12 and Antirrhinum10,11,13.

The other major focus of our work is on the evolution of leaf and flower form, using the Antirrhinum species group as a model system. One aspect is to study evolutionary events over a few generations, at a naturally occuring hybrid zone in the Pyrenees. This uses a combination of DNA pedigree, phenotypic and pollinator studies13,14,15. We also aim to identify the key genes controlling flower colour that are under selection. On a longer evolutionary timescale we are studying variation in the function and structure of genes controlling flower colour between different Antirrhinum species using a combination of molecular and genetic approaches13,15. In addition to the work on flower colour, we are also studying genes underlying flower and leaf shape variation between species. These were identified through QTL analysis and are being studied at the molecular and phenotypic levels16,17.

1 - Visualization and Data Management of Microscopy Images

Volume render of an Arabidopsis leaf with ATH8::GUS expression in the veins (red).

Jerome Avondo - email

The field of biological imaging has seen incredible advances due to new imaging techniques and molecular markers. These advances in data aquisition result in the creation of multidimensional images. How these images are to be visualized and data managed remains an open challenge.

To visualize multidimensional images I research the field of real time volume rendering and the use of the graphics processing unit (GPU).This allows the creation of rendering algorithms for the interactive visualization of microscopy images through space and time. Results from my work is available to the community as an open source cross platform application here.

Due to the large amounts of data being produced by bio-image data acquisition systems, such as a confocal microscope, it is possible for a single lab to generate many terabytes of imaging data. To address the data management issues involved with such large data sets I am customizing and deploying the OMERO system for our lab.

2 - Image Processing for Biological Development

3D segmentation of cells (green) from a volumetric image (Arabdidopsis leaf).

Paul Southam - email

Extracting quantitative measurements from image data is a fundamental step towards understanding how biological forms develop and evolve. I work on a number of projects including:

Modelling microtubule dynamics during growth and division - Jordi Chan & Scott Grandison.

Modelling Dynamic Growth Maps of Leaf Development - Samantha Fox.

Capturing plant development in 3D with Optical Projection Tomography - Karen Lee.

I am also interested in cell segmentation, cell registration, leaf classification and texture analysis.

3 - Microtubule dynamics during growth and division

Pre-prophase band formation in a leaf epidermis cell.

Jordi Chan - email

Microtubules are highly dynamic subcellular filaments that adopt different patterns of alignment during growth, division and differentiation. During growth, microtubules are found in close association with the plasma-membrane where they are needed to guide the movement of cellulose synthase to generate appropriate wall architectures. Like animal cells, microtubules are required for chromosome segregation during cell division via the spindle apparatus. However, unlike animal cells, plant microtubules bunch up at the onset of mitosis to form a cortical ring, known as the pre-prophase band (PPB), which is involved in determining the future plane of cell division. Since plant cells within tissues are immobilised by their own walls and can only respond to cellular and environmental cues by altering their directions of growth and division planes, microtubules are indispensible elements of plant morphogenesis.

How microtubules change alignment and construct different arrays during growth and cell division is unknown but is likely to reside within fundamental properties of the microtubules themselves, such as the dynamic behaviour of their filamentous ends, regulation of their sites and angles of nucleation (birth) and outcomes of encounters between themselves and the geometry of the cell. It is interesting to think that these simple subcellular behaviours or rules concerning microtubule dynamics may not only shape the cell but also be amplified during morphogenesis from cell-to-cell to shape the entire plant.

The aim of my work is to reconstitute microtubule dynamics and cortical array organisation in-silico to understand and test mechanisms behind microtubule reorientation and division plane alignment (PPB formation) in-planta. This will involve time-lapse microscopy of growing and dividing leaf cells expressing appropriate fluorescent reporter genes, the development of software to quantify microtubule dynamics at both subcellular and tissue levels, and computer modelling. Models will be tested using cytoskeletal-specific drugs and mutants harbouring defective cytoskeletal genes.

4 - Modelling microtubule dynamics

A frame from a microtubule simulation.

Scott Grandison - email

Microtubules are small, filamentous polymers that perform a variety of important developmental functions that have an impact, not only within the cell, but on a macroscopic scale. They help to provide structural integrity to the cell, by guiding the mechanisms which lay down cellulose microfibrils. If these microfibrils are laid down with a preferential direction then the cell can be encouraged to grow anisotropically which is important when forming complex shapes during development. Another important role for microtubules is their role in determining where exactly a cell will divide during mitosis.

I'm interesting in studying these functions by making accurate mathematical and computer simulations of microtubule dynamics. I work closely with Dr. Jordi Chan and Dr. Paul Southam who are interesting in the experimental and image processing part of the project.

5 - Cell division in the Arabidopsis leaf epidermis

Tracked divisions of a stomata lineage

Sarah Robinson - email

Collaborators: Przemyslaw Prusinkiewicz, Dominique Bergmann, Pierre Barbier de Reuille, Jill Harrison

The Arabidopsis leaf epidermis is a tissue consisting of a single layer of cells of various sizes, shapes and functions, arranged in an intricate two-dimensional pattern. We aim at understanding this pattern with a combination of time-lapse imaging and computational modeling techniques.

We have developed the ability to image Arabidopsis seedlings continuously for up to seven days using time-lapse confocal microscopy. The resulting movies capture the dynamic nature of the leaf development. From these movies we extract the information about cell growth, timing of cell divisions and placement of division walls, which is used to produce an initial descriptive model. We then gradually replace direct data with hypothetical deterministic rules of cell division within a growing leaf epidermis, and verify the results by comparing the output to the data.

The model has already made it possible to test existing theories of how and when cells divide, and resulted in the falsification, in the case of Arabidopsis leaves, of several rules previously reported in the literature. We have also verified rules that predict the position of dividing walls in non-differentiated cells. . The model is currently being extended to include rules for the timing of cell division and the differentiation of stomata. The model thus provides a framework for understanding how complex patterns of epidermis cells develop and accommodate differentiated cells, and suggests that the observed complexity may by an emergent property of a small set of simple, possibly deterministic rules.

6 - Modelling and analysis of growing plant tissues. Application to flower development

No caption

Pierre Barbier-De-Reuille - email

My main research project is about methods and tools for the modelling and the analysis of growing plant tissues.

For the modelling part, I am developing VVe, a modelling environment for the modelling of 2D structures in 3D space. VVe includes specialized tools for dealing with the development of plant tissues at the cellular scale, but can also be used to model tissue considered as a continuum. It is already distributed, as it is used by Sarah Robinson in our lab, in the lab of the professor Prusinkiewicz in Calgary and in the lab of the professor Kuhlemeier in Bern.

For the data analysis, I am developing digitizing and computation tools. I am in particular working on tracking cell growth and division, and extracting useful information from this (i.e. growth tensor, division rate, ...).

Currently, my developments aim at helping the understanding of the growth patterns in the mature petal of Arabidopsis.

7 - Dynamic Growth Maps of Leaf Development

Arabidopsis leaf one expressing GFP in the cell membranes

Samantha Fox - email

To understand how genes control leaf shape and size we are characterising the dynamics of leaf growth in Arabidopsis. We first established a robust staging sytem that can be used to accurately determine the developmental phase of an Arabidopsis leaf. Time-lapse imaging was then used to capture leaf growth during each phase. Growth parameters were extracted from the resulting images using computational techniques. This approach was complemented by sector analysis in which clones expressing GFP were induced and visualised at a range of stages. The results have informed mechanistic models of leaf growth and allowed the role of genes in leaf growth to be explored.

8 - Modelling Leaf Development and Growth Arrest in Arabidopsis


Erika Kuchen - email

The aim of my research project is to use Arabidopsis leaf phenotype descriptions together with expression patterns of genes involved in leaf development to construct a leaf growth models. These models will help to answer the following questions. How is leaf growth maintained and coordinated to produce such similarities in leaf size and shape across plants of the same species? How does a leaf know that it has reached its final/desired size?

Leaves are being imaged in 3-dimensions using optical projection tomography by Karen Lee. Clonal analysis has been used to follow cell division and growth. Additionally, tracking techniques are used to calculate localised growth rates across leaf development. Leaf expansion consists of two main phases, cell division and cell expansion. The spatio-temporal pattern of cell division has been analysed using cyclinB1:GFP. Cell expnasion arrest will be adressed using life tracking of the growing leaf.

This data will be used to model leaf growth in 3-dimensions until leaf maturation and growth arrest. As a first approach the leaf model has been created using hypothetical morphogens that help to describe the observed shape and sectors. The relatively well fitting model is now used to find candidate genes for these theoretical morphogens.

The model will be confirmed by modelling a growth arrest mutant, in which several TCP (TEOSINTE BRANCHED 1, CYCLOIDEA ,PFC) family memebers have been down regulated by overexpression of miR-JAW.

9 - Capturing plant development in 3D with Optical Projection Tomography

Optical Projection Tomography of an Arabidopsis seedling.
Volume Rendered Plant Specimens

Karen Lee - email

We have explored the use of Optical Projection Tomography (OPT) as a method for capturing 3D morphology and gene activity at a variety of developmental stages and scales from plant specimens, in collaboration with the Medical Research Council, James Sharpe and Bioptonics .

OPT can be conveniently applied to a wide variety of plant material including seedlings, leaves, flowers, roots, seeds, embryos and meristems. At the highest resolution large individual cells can be seen in the context of the surrounding plant structure. 3D domains of gene expression can be visualized using either marker genes such as β-glucuronidase, or more directly by whole-mount in situ hybridization. For naturally semi-transparent structures, such as roots, live 3D imaging using OPT is possible. 3D gene expression patterns in living transgenic plants expressing fluorescent GFP markers can also be visualised.

To interactively analyse and quantify 3D OPT data we are developing software to visualise 3D volumes (Jerome Avondo), use haptics to accurately place points on volumes in 3D space (Johann Strasser), and extract growth measurements from 3D point co-ordinates (Paul Southam and Pierre Barbier-De-Reuille).

The aim of my project is to use these tools to capture 3D leaf shape and leaf growth in Arabidopsis, contributing to leaf growth models being developed by Erika Kuchen, and to the atlas of leaf development with Samantha Fox. The combination of 3D shape analysis, 3D growth tracking and modelling will allow us to understand mechanisms controlling growth and shape from earliest stages of leaf growth to maturity.

Please see OPT Gallery.

10 - Modelling Growth at the Tissue Scale

Model for Snapdragon cyc dich mutant

Richard Kennaway - email

I am developing finite element methods for modelling the growth and development of curved two-dimensional tissues such as leaves and petals. The interactive software tool I am developing, called GFtbox, is available for download.

11 - Testing models for polarity and asymmetry in Antirrhinum

An Antirrhinum rad mutant flower

Xana Rebocho - email

Bilateral symmetry of flowers has evolved several times independently from an ancient radially symmetrical condition. Antirrhinum’s flowers are comprised of 2 ventral, 2 lateral and 1 dorsal petal. One of the key genes involved in the control of dorsal identity is RADIALIS. However, the way in which this gene acts to control the asymmetry, shape and form of the flower remains to be unravelled. I am addressing this issue using a combination of genetic, developmental and computational approaches. The goal of my research will be to understand how key genes act within and between cells to modify growth patterns that contributed to the final morphology of the flower and the evolution of a new trait.

12 - Modelling Arabidopsis floral organs and testing organisers of growth

Arabidopsis petal sectors.

Susana.Sauret-Gueto - email

How regulatory genes co-ordiante cell behaviour to produce the macroscopic shape of organs is key to understand how plants develop.

A fascinating example of this process is flower development, as in the flower bud spatially close undifferentiated primorida will differentiate into the four types of mature floral organs: sepals, petals, stamens and carpels. Regulatory gens responsible to trigger the developmental programme that will generate the different types of floral organs are well known and plants with homeotic transformations have been characterised.

Nevertheless, how growth is sequentially organised to produce different final shapes is not well-known. In order to gain better understanding of how shape is patterned, we aim to characterise and manipulate, both in the flower and leaf of the model plant Arabidopsis, candidate genes to function as growth organisers.

To analyse the results thus obtained, a quantitative development framework for the flower and its floral organs it is being generated equivalent to the leaf staging system being developed in the lab. Analysis of whole and regional organ growth is gathered using OPT as well as confocal together with sector analysis, being the Arabidopsis petal the first organ of study. This data is the base to construct mechanistic models of leaf and floral organs growth that will guide hypothesis on gene co-ordiantion of cell behaviour at the tissue level to promote and organise growth. Hypothesis will be experimentally and computationally challenged in a crosstalk of approaches that will aim to model shape patterning.

13 - Evolution of organ size and shape between Antirrhinum species

Antirrhinum wild-type flower

Lucy Copsey - email

The "old world" Antirrhinum species, found growing naturally in southern Europe and North Africa, show an extensive range of diversity in growth habit, organ size, shape and flower colour. This variation is important as it highlights differences between individuals which may be due to either environmental effects or differences at gene level, these genetic differences underpins how diversity in form is generated through evolutionary time and is the basis of evolution. By exploiting evolutionary variation in size we hope to identify genes controlling organ size in Antirrhinum species (allometry project).

I am using a classical genetics approach involving the production of a number of plant resources developed by crossing Antirrhinum species to both our cultivated JI stock 7 line and interspecies crosses to study natural variation and domestication effects. These resources are available for use by group members and for the wider Antirrhinum community.

14 - Evolutionary Dynamics Underlying Species Diversification

Bee polinating A. pseudomajus

Matthew Couchman - email

We are part of a collaboration investigating factors affecting gene flow across populations. The project focuses on flower colour in two distinct subspecies of Antirrhinum. Within the Spanish Pyrenees there are a number of hybrid zones that provide ideal environments for analysing how genes influence flower colour. For each individual within our chosen hybrid zone we are annually recording their GPS location and colour scores as well as taking samples for genotyping and other molecular analysis.

My role within the project is to develop a relational database to capture these and other collaborator outputs as well as a website to act as a gateway to this database. The website will include visual tools such as a configurable map of recorded GPS positions and genetic and physical maps.

15 - Natural variation of flower colour

Antirrhinum flower colour variation.

Xana Rebocho - email, Desmond Bradley - email

Collaborators: Christophe Thebaud, Nick Barton, Ferran Palero, Christophe Andalo and Monique Burrus

Two subspecies of Antirrhinum (magenta and yellow), which generally occur in genetic isolation in the Pyrenees, have hybridized in nature to create a hybrid zone. Due to hybridization the Antirrhinum flowers display an array of parental as well as mixed flower colour phenotypes which are the result of genetic variation at three key loci ROSEA, ELUTA and SULFUREA. The two first genes are involved in the control of the magenta anthocyanin pigmentation while SULF is a repressor of aurone (yellow) pigmentation. As the HZ gives a great playground to study evolution in action, the goal of this research is to follow the genetic flow of these flower colour loci within the hybrid population as well as the fitness of each phenotype/genotype throughout several years.

16 - Evolution and genetic control of leaf shape

youngArabidopis mutant leaf diverging from WT morphology

Sascha Duttke - email

Despite their tremendous morphological diversity, all leaves start development as a small group of cells initiate the leaf primordium. Post initiation, cell division and differentiation results in morphological variation. Using a combination of stainings, 3D imaging and computational approaches, I am analysing the early stages of leaf development to determine when and how these variations arise.

17 - The genetic basis for natural variation of organ shape and size

A. majus flower (left) compared to A.charidemi flower (right)

Xianzhong Feng - email

Natural variation is one of the most striking features of the biological world; much of the variation between species and varieties involves correlated differences in shape and size (allometric variation). However, it is poorly understood the genetic, developmental and evolutionary basis of these variations. We addressed this problem by examining the genetic basis for organ size and shape differences between Antirrhinum species. Using statistical shape modeling, we quantified the differences of shape and size in both leaves and flowers . Allometry genes were identified by QTL analysis, and confirmed using recombinant inbred lines (RILs) and near isogenic lines (NILs). Key allometry loci is fine-mapped and transposon-mutagenesised with a view to isolating the corresponding genes. The allometry genes form different species are introgressed into a uniform background to compare their effects in evolution pathway.

18 - Evolutionary processes of trait diversity in the wild

Flowers of different Antirrhinum species

Hugo Tavares - email

Different species of Antirrhinum show conspicuous differences in flower color (see figure), a trait though to be under strong selective pressure in insect-pollinated species such as this one. To understand the molecular genetic basis of these differences I am focusing my work on two tightly-linked loci - ROSEA and ELUTA - which interact to regulate the intensity and pattern of red anthocyanin pigments. By combining genetic, molecular, and bioinformatic analysis I hope to understand how the different species' ROS and EL alleles (genotype) effectively contribute to their respective floral pigmentation patterns (phenotype), thus uncovering some of the evolutionary mechanisms that may lead to trait diversification in the wild.

modified on 6 December 2010 at 15:57 ••• 1,023,755 views