Designing Plants: Modeling Ideal Shapes.
Yuling Jiao
DOI: https://doi.org/10.1016/j.molp.2018.12.004
IF: 27.5
2018-01-01
Molecular Plant
Abstract:In an industrialized world, virtually everything we are using is first designed and then manufactured. Let us use the automobile manufacturing industry as an example: engineers design vehicles based on laws of physics; select proper materials for machine parts; and assemble parts following blueprints into cars, buses, and trucks. It is evident that we need not only parts but also technical drawings and instructions for the assembly of parts into a functional pump or clutch. In biology, we have long been doing “reverse engineering”. Using genetics, biochemistry, cell biology, and other approaches, we deconstruct an organism, an organ, a type of tissue, or even a cell to reveal the underlying designs and architecture to expand our knowledge base. Thanks to the accumulating advances over the past decades, we have learned a great deal about biological “parts,” including genes, proteins, and organelles. How are these “parts” assembled to yield a living organism? This is clearly an outstanding question in biology. As plant biologists, we work hard to understand plants, and we dream to advance plants—to design plants for the well-being of humans. For example, we need crops with high yield, crops requiring minimal fertilizer and pesticide use, and crops adapted to harsh environments. Our ancestors selected natural variants for crop domestication and improvement, and the practice continues today. Such natural variations include spontaneous mutations, favorable offspring from hybridizations, and rare polyploidy events. With our expanding knowledge of the biological “parts” in plants, we are coming closer to designing crops in the way we design automobiles and airplanes. One missing piece preventing us from assembling “parts” into a living plant is a full understanding of their regulatory systems: their structures, constraints, and possibilities. This is a challenging task insofar as there are simply too many “parts” in a plant, and disabling each of them, as mutants have taught us, do not always explain how the entire “machine” functions. In terms of the automobile analogy, it is almost impossible to understand the physical laws and designing principles by only disassembling a car. Systems biology is well suited to deal with this particular challenge because it uses computational and mathematical modeling to explain complex biological systems. By testing a range of possible regulatory principles, modeling results indicate which regulatory circuits may explain a biological process. Because model predictions can in turn be subjected to reciprocal experimental testing and experimental data can be used to revise models, systems biology has the power to decipher the “designing principles” of biological “parts” (Roeder et al., 2011Roeder A.H. Tarr P.T. Tobin C. Zhang X. Chickarmane V. Cunha A. Meyerowitz E.M. Computational morphodynamics of plants: integrating development over space and time.Nat. Rev. Mol. Cell Biol. 2011; 12: 265-273Crossref PubMed Scopus (65) Google Scholar, Prusinkiewicz and Runions, 2012Prusinkiewicz P. Runions A. Computational models of plant development and form.New Phytol. 2012; 193: 549-569Crossref PubMed Scopus (130) Google Scholar). For example, the power to dissect cybernetics in plants, i.e., control and communicate in plants, has been nicely illustrated by the modeling of plant organogenesis. The collective results of organogenesis determine plant architecture, which broadly affects crop traits (Wang et al., 2018Wang B. Smith S.M. Li J. Genetic regulation of shoot architecture.Annu. Rev. Plant Biol. 2018; 69: 437-468Crossref PubMed Scopus (295) Google Scholar). Inflorescences, the branching reproductive shoot systems, are major determinants of crop yield. In crops, inflorescence complexity is often a selection target during domestication and improvement, especially in cereal crops. After flowering transition, the shoot apical meristems (SAMs) at the growing tips switch to reproductive growth to produce flowers. An iterative pattern of decisions at the SAM determines the inflorescence architecture (Kyozuka et al., 2014Kyozuka J. Tokunaga H. Yoshida A. Control of grass inflorescence form by the fine-tuning of meristem phase change.Curr. Opin. Plant Biol. 2014; 17: 110-115Crossref PubMed Scopus (44) Google Scholar). The SAM may either transition and adopt the floral identity or produce branching meristems. Each branching meristem reiterates this developmental decision-making process using the same program or a revised one. In the model plant Arabidopsis thaliana, two genes, LEAFY (LFY) and TERMINAL FLOWER1 (TFL1), regulate inflorescence architecture (Weigel et al., 1992Weigel D. Alvarez J. Smyth D.R. Yanofsky M.F. Meyerowitz E.M. LEAFY controls floral meristem identity in Arabidopsis.Cell. 1992; 69: 843-859Abstract Full Text PDF PubMed Scopus (1191) Google Scholar, Bradley et al., 1997Bradley D. Ratcliffe O. Vincent C. Carpenter R. Coen E. Inflorescence commitment and architecture in Arabidopsis.Science. 1997; 275: 80-83Crossref PubMed Scopus (660) Google Scholar). lfy mutants have delayed floral transition and produce highly branched inflorescences, whereas tfl1 mutants switch from an indeterminate inflorescence to a determinate structure with terminal flower formation. Inflorescence structures are highly diversified in nature. In fact, many inflorescence types observed in nature have never been observed in Arabidopsis. Therefore, how much can studies in a model plant reveal about the wide range of inflorescence architectures found in nature? Computational modeling has provided a plausible answer (Prusinkiewicz et al., 2007Prusinkiewicz P. Erasmus Y. Lane B. Harder L.D. Coen E. Evolution and development of inflorescence architectures.Science. 2007; 316: 1452-1456Crossref PubMed Scopus (286) Google Scholar). In the model, maturation rates are hypothesized for the apical meristem and branching meristems. The time to reach a threshold for flower formation is denoted as T. Using different relative values for flower transition speed in the apical meristem (TA) and the branching meristems (TB), the model output encompasses a range of inflorescence architectures, including panicle, raceme, and cyme (Figure 1A). The modeling results highlight possible directions for manipulating inflorescence architecture to obtain higher yield. In fact, subsequent studies have provided evidence for a progressive maturation of meristems in tomato and demonstrated that manipulation of genes controlling meristem maturation by gene editing can indeed be used for such applications (Park et al., 2012Park S.J. Jiang K. Schatz M.C. Lippman Z.B. Rate of meristem maturation determines inflorescence architecture in tomato.Proc. Natl. Acad. Sci. U S A. 2012; 109: 639-644Crossref PubMed Scopus (124) Google Scholar). Researchers have used gene editing to generate a spectrum of mutants with varying meristem maturation speed (T in the model) and found that favorable inflorescence architectures could be obtained in tomato (Soyk et al., 2017Soyk S. Lemmon Z.H. Oved M. Fisher J. Liberatore K.L. Park S.J. Goren A. Jiang K. Ramos A. van der Knaap E. et al.Bypassing negative epistasis on yield in tomato imposed by a domestication gene.Cell. 2017; 169: 1142-1155.e12Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar). Leaf shape is another striking example of biological shape variation recently exploited by modelers. Leaves show tremendous morphological diversity, and these variations clearly affect the photosynthesis area as well as drought adaption. Leaf shapes range from simple to lobed to compound and are highly determined by margin morphology. Genetic analyses have shown that mutants in the auxin efflux carrier PIN FORMED1 (PIN1) and transcription factor CUP-SHAPED COTYLEDON2 (CUC2) both have smooth margins (Hay et al., 2006Hay A. Barkoulas M. Tsiantis M. ASYMMETRIC LEAVES1 and auxin activities converge to repress BREVIPEDICELLUS expression and promote leaf development in Arabidopsis.Development. 2006; 133: 3955-3961Crossref PubMed Scopus (252) Google Scholar, Nikovics et al., 2006Nikovics K. Blein T. Peaucelle A. Ishida T. Morin H. Aida M. Laufs P. The balance between the MIR164A and CUC2 genes controls leaf margin serration in Arabidopsis.Plant Cell. 2006; 18: 2929-2945Crossref PubMed Scopus (441) Google Scholar). It is puzzling how these two genes can generate iterative serrations, i.e., margin protrusions. A computational model has tested the significance of the reciprocal repression between CUC2 and the PIN1 module in the determination of leaf margin shape (Bilsborough et al., 2011Bilsborough G.D. Runions A. Barkoulas M. Jenkins H.W. Hasson A. Galinha C. Laufs P. Hay A. Prusinkiewicz P. Tsiantis M. Model for the regulation of Arabidopsis thaliana leaf margin development.Proc. Natl. Acad. Sci. U S A. 2011; 108: 3424-3429Crossref PubMed Scopus (308) Google Scholar). PIN1 transports auxin up the gradient to form auxin concentration maxima. High levels of auxin inhibit CUC2, restricting CUC2 expression to auxin minima. The model also assumes that CUC2 reorients PIN1 away. Based on these assumptions, the model recapitulates repeated serrations in the margin. Furthermore, by modulating the strength of feedback, a range of leaf marginal shapes can be derived. On top of that, veins, which serve as auxin sinks from leaf marginal auxin maxima, can model more complex serrations and compound leaves (Runions et al., 2017Runions A. Tsiantis M. Prusinkiewicz P. A common developmental program can produce diverse leaf shapes.New Phytol. 2017; 216: 401-418Crossref PubMed Scopus (78) Google Scholar) (Figure 1B). In addition to marginal serrations, leaf blade shapes are highly diversified among species, among ecotypes, and even within the same plant. A separate set of models introduces a set of spatiotemporal regulatory factors to describe the growth of each cell (Kuchen et al., 2012Kuchen E.E. Fox S. de Reuille P.B. Kennaway R. Bensmihen S. Avondo J. Calder G.M. Southam P. Robinson S. Bangham A. et al.Generation of leaf shape through early patterns of growth and tissue polarity.Science. 2012; 335: 1092-1096Crossref PubMed Scopus (165) Google Scholar). These factors define cell growth anisotropy and growth rate, as well as cell division direction and division rate. These regulatory factors are interconnected to form feedbacks from early growth patterns, which dynamically adjust factor values. These models can generate a wide range of blade shapes seen in nature. However, it is important to note that the regulatory factors used in the model remain hypothetical and that their biological meanings remain to be determined. The examples described above are only a small sampling of the available findings. There are many more examples of computational and mathematical modeling used to explain and predict morphogenesis patterns, such as phyllotaxis, apical dominance, and root patterning (Lavedrine et al., 2015Lavedrine C. Farcot E. Vernoux T. Modeling plant development: from signals to gene networks.Curr. Opin. Plant Biol. 2015; 27: 148-153Crossref PubMed Scopus (9) Google Scholar). Modeling has also provided critical insight into traits not directly connected with morphogenesis, such as circadian clock feedback loops. Although it is still insufficient, we now know a great deal about gene functions and have an expanding toolkit to manipulate genes. To precisely design and create new plant varieties, we need to understand the cybernetics of plants. How do genes reproducibly assemble to form regulatory circuits? How are biological shapes determined? How do organs respond to environmental cues? We have some answers and clues about each of these pathways. Simply overexpressing or knocking out one gene may lead to unexpected phenotypes and sometimes strong side effects. Similarly, in the automobile analogy, removing a part or adding a new part may not always boost the performance of a car. An understanding of design principles and access to technical drawings are necessary even for car tuning. Computational and mathematical modeling allows the prediction of the complex or even counterintuitive dynamics of the regulatory networks driving plant growth and crop traits. Such blueprints of regulatory networks will establish a foundation for the rational design of crops for the future. New advances in synthetic biology and gene editing speed up the move from principles to design.