Modeling the Evolution of Stelar Architecture in Vascular Plants

Abstract

The control of primary vascular tissue differentiation by auxin produced by the shoot apex ranks among the better known hormone signaling systems in vascular plants. Surprisingly little is known, however, about the relationship between the pattern-forming properties of this system and stelar architectures in different major groups of vascular plants, including those represented in an extensive and systematically important fossil record. Expanding upon suggestions by Wight, I present a simple computer model relating known properties of the auxin signaling system with explicit predictions about the configuration of stelar tissues suitable for comparison with morphometric data derived from both living and fossil plants. The model divides the continuous acropetal process of vascular tissue determination into a series of vignettes in which target meristematic tissue receives quantified concentrations of hormone, i.e., auxin, from "organizing centers" (OCs), branch or leaf primordia, serving as hormone sources. Specific parameters of the model include longitudinal spacing, phyllotaxis, variable hormone production rates of the OCs, size of the shoot apex, and reaction of the target tissue to specific signals received. To test the applicability of this approach, model predictions are compared with detailed morphometric data collected from Psilophyton and members of the Aneurophytales. Close correspondence is observed between predicted hormone concentrations from the model and the configuration of stelar tissues, including relative positions of primary xylem, primary phloem, secondary xylem, and the presence of morphological gradients in the primary xylem related to the protoxylem/metaxylem continuum. In addition to the morphometric comparisons, a wider exploration of the model involving both orthostichous and Fibonacci phyllotaxis of OCs is conducted. The results are compared with members of the Iridopteridales, Cladoxylopsida, Calamopityaceae, and Medullosales. Striking correspondence is observed allowing for reinterpretation of the evolution of vascular plant stelar architecture not just in terms of historical patterns of important mature structures but also as a system of evolving developmental dynamics underlying these structures. All comparisons made so far serve to underscore the fundamental relationship in early vascular plants between the evolution of increased complexity in stelar architectures and the evolution of complex lateral branches and leaves. Among evolutionary events occurring in the fossil record, the model of hormone determination offers important insights into (1) origin of ribbed protostelic systems from primitive columnar architectures as a consequence of increased compactness of the shoot apex and orthostichous phyllotaxis; (2) conspicuous differences in the three-dimensional configuration of protoxylem strands supporting a developmental distinction between Devonian "radiate protoxylem" and "permanent protoxylem" groups; (3) quantifiable differences in protoxylem/metaxylem tissue fabric probably related to differences in hormonal activity of lateral appendage primordia during early development; (4) origin of pith at the center of the stele related to changes in the geometry of the shoot apex under several possible models of hormone determination; and (5) dissection of the stele into discrete primary vascular bundles possibly related to changing receptivity to the hormone signal and the geometric consequences of flow rates in three dimensions over developmentally significant intervals of time. In each instance, the computer model provides new ways to interpret evolutionary change within and between major groups of vascular plants, with specific stelar architectures of these plants offering reciprocal illumination for further improvement of the model. Because of the unique nature of their development and excellent fossil record, I suggest that vascular plants in many ways seem to be the ideal study group for integrating developmental process directly into analysis of homology and estimation of phylogenetic relationship.