Final Report Summary - PLANTORIGINS (PLANT developmental biology: discovering the ORIGINS of form)
WORKPACKAGE1
The aim of this work package was to use the uniquely high quality preservation of the plants preserved in the Rhynie chert (rock) to describe for the first time elements of the developmental biology of these enigmatic plants. All research involved the analysis of thin sections of the chert and painstaking reconstructions. This discovered new organs in Rhynia gwynn-vaughnii and demonstrated that genome size likely varied considerably in Asteroxylon mackei.
During the course of this research we completed the life cycle of Rhynia and discovered that it produced asexual propagules. Propagules of this kind have never been found in any of the plants in the Rhynie chert. The development of the propagules was first described using thin sections and transmitted light microscopy. Then synchrotron X-ray tomography was used to define the three dimensional structure of propagules. This gave unprecedented resolution and detail. Not only did it facilitate the imaging of the development of propagules but it also demonstrated that the apical cells of Rhynia are tetrahedral and differentiation takes place very soon after cells are cut off from this initial. Furthermore it allowed the description of stomatal development (mesogenous). By way of comparison the apex of Aglaophyton was described and shown to be much larger and lacking a single, tetrahedral apical cell.
The main challenge in understanding the life history of Asteroxylon mackei is the identification of the haploid phased (gametophyte) of the life cycle. To do this a protocol was developed to quantify the size of stomatal guard cells (since their size is indicative of their ploidy). Stomatal dimensions on thousands of specimens were characterized. While a large variation in cell size was discovered these did not constitute a biphasic distribution which would be expected if haploid and diploid plants were among the specimens measured. However the variation in cell size in Asteroxylon is larger than that observed in any other land plant. This suggests either that Asteroxylon in the chert represents a set of different species or subspecies with different ploides or that there is natural variation for genome size in Asteroxylon. The latter is more likely because the distribution of cell sizes clusteraing around a single mean and is normally distributed.. To determine if the stomatal distribution observed in Asteroxylon was unusual, the stomatal sizes of huperzia (the nearest extant species) was measured. The distribution of stomatal sizes in Huperzia was considerably smaller than in Asteroxylon. This highlights the uniqueness of stomatal size distribution in Asteroxylon.
The first detailed description of the developmental biology of the alga, Paleonitella in the chert has been produced. The morphology of these organisms is remarkably similar to those of extant relatives, indicating extreme evolutionary stasis over a more than 300 million year period.
WORKPACKAGE 2
A major cooperative activity between partners 1, 2 and 6 was the generation of a transcriptome of the different phases of the Physcomitrella patens life cycle. This included, rhizoids, caulonema, chloronema, a variety of stages of sporophyte development and spermatozoids. This work is currently been written up as a manuscript. These detailed data provided a platform for achieving each of the 15 specific goals outlined in work package 2 “Genetic mechanisms underpinning plant life cycles”.
Role of transcription factors in controlling the development of Physcomitrella development (includes all goals in Work Package 2)
Class 2 RSL genes are transcription factors that were first identified as controlling the development of root hairs in Arabidopsis. It was hypothesized that they would control some aspect of gametophyte development in Physcomitrella (either caulonema and or rhizoid development). Four Class2 RSL genes were identified in Physcomitrella and showed that they control caulonema development (and found no evidence that they control rhizoid development). Furthermore we showed that they are part of a regulatory network that includes class 1 RSL genes and auxin.
LRL basic helix loop helix transcription factors are involved in the development of the root hairs in Arabidopsis and it was hypothesized that they were involved in the development of rhizoids in the moss Physcomitrella. Two LRL genes were identified as expressed in rhizoids and loss of LRL activity (Pplrl1 Pplrl2 double mutants) gametophytes were rhizoidless. This indicates that RSL genes are required for the development of rhizoids in the gametophyte of Physcomitrella.
The sporophyte transcriptome of Physcomitrella showed that a TCP transcription factor was expressed at high level. TCP transcription factors regulate the development sporophyte branching in vascular plants. To determine if this activity was ancestral among land plants loss of function mutants were generated that lacked the activity of PpTCP1. Sporophyte development was defective in these mutants indicating that the activity of PpTCP1 was required for the development of the Physcomitrella sporophyte. This suggests that the function of TCP gene was involved in sporophyte development in early land plants and the role in the control of branching evolved later, perhaps through gene duplication and neofunctionalization.
The growth of plant cells required the activity of an energized plasma membrane through the activity of the proton-pumping AHA proton ATPases. This causes polarization of membrane facilitating the action of glutamine receptors which are calcium channels that are required for the transport of calcium across the plasma membrane. The transcriptome analysis identified two AHAs (PpAHA1 and PpAHA2) that were expressed at high levels in protonema and two glutamine receptors (PpGLR1 and PpGLR2) which were expressed during the tip growth of caulonema. Mutants that lacked the function of these GLRs (Ppglr1 Pglr2 double mutants) developed defective protonema and few sporophytes. This is consistent with the hypothesis that GLRs are active as calcium channels in ancestral groups of land plants such as Physcomitrella. Mutants that lack the function of the PpAHA1 and PpAHA2 have been generated and their phenotypes are being characterized.
We predicted that ancestors of RSL genes would be present in the genomes of Streptophyte algae (Coleochaete nitelarum). We identified a basic helix loop helix transcription factor that is the closest known relative to the land plant RSL genes. This suggests that the RSL program that controls the development of rhizoids and root hairs in land plants is derived from a program that existed in the algal ancestors of the land plants.
CONCLUSION
Genes functions have been defined during the life cycle of Physcomitrella and demonstrate that many of the key regulatory processes that exist in derived groups of land plants were inherited from their Palaeozoic ancestors and likely diversified in function as land plants diversified in mornphology.
WORKPACKAGE 3
The data from WORKPACKAGE 1 and 2 have been formulated into testable hypotheses that are being tested and will form the core of activity in this discipline for the next 5 to 10 years.
DISSEMINATION
These results have been, and continue to be, disseminated though the scientific literature. They are also disseminated through the general public through our public engagement activities at museums, media (radio and TV) websites (http://www.plants.ox.ac.uk/plantorigins/) and other public engagement activities.