Combining gene network, metabolic, and leaf-level models show means to future-proof soybean photosynthesis under rising CO₂

Abstract

2Global population increase coupled with rising urbanization underlies the predicted need for 2 3 60% more food by 2050, but produced on the same amount of land as today. Improving 2 4 photosynthetic efficiency is a largely untapped approach to addressing this problem. Here, we 2 5 scale modeling processes from gene expression through photosynthetic metabolism to predict 2 6 leaf physiology in evaluating acclimation of photosynthesis to rising [CO 2 ]. Model integration 2 7with the yggdrasil interface enabled asynchronous message passing between models. The 2 8 multiscale model of soybean photosynthesis calibrated to physiological measures at ambient 2 9[CO 2 ] successfully predicted the acclimatory changes in the photosynthetic apparatus that 3 0 were observed at 550 ppm [CO 2 ] in the field. We hypothesized that genetic alteration is 3 1 1 necessary to achieve optimal photosynthetic efficiency under global change. Flux control 3 2 analysis in the metabolic system under elevated [CO 2 ] identified enzymes requiring the 3 3 greatest change to adapt optimally to the new conditions. This predicted that Rubisco was less 3 4 limiting under elevated [CO 2 ] and should be down-regulated allowing re-allocation of 3 5 resource to enzymes controlling the rate of regeneration of ribulose-1:5 bisphosphate (RubP). 3 6 By linking the GRN through protein concentration to the metabolic model it was possible to 3 7 identify transcription factors (TF) that matched the up-and down-regulation of genes needed 3 8 to improve photosynthesis. Most striking was TF GmGATA2, which down-regulated genes 3 9 for Rubisco synthesis while up-regulating key genes controlling RubP regeneration and starch 4 0 synthesis. The changes predicted for this TF most closely matched the physiological ideotype 4 1that the modeling predicted as optimal for the future elevated [CO 2 ] world. 4 2 4 3 KEYWORDS: Gene network model, metabolic model, photosynthesis, global change, Soybean, 4 4 transcription factors, multiscale modeling, model integration 4 5 4 6As the world's most important seed legume and most widely grown dicotyledonous crop, 4 7 the future-proofing of photosynthesis in soybean (Glycine max (L.) Merr.) under rising 4 8 atmospheric concentrations of CO 2 ([CO 2 ]) is of importance. Down-regulation of light-saturated 4 9net leaf CO 2 uptake (A sat ) at elevated [CO 2 ] has been reported for many C 3 crops, yet the 5 0 mechanism underlying this response is poorly understood. Under current [CO 2 ], A sat in C 3 crops 5 1 is most commonly limited by the in vivo Rubsico activity (V c,max ) (Long et al., 2004). However, 5 2 as [CO 2 ] continues to rise, it follows from the steady-state biochemical model of photosynthesis 5 3 of (Farquhar et al., 1980) and its subsequent modifications (Von Caemmerer, 2000) that control 5 4 will shift from Rubisco to RubP regeneration (Long et al., 2004), which is represented by the 5 5maximum in vivo rate of whole chain electron transport (J max ). While described by electron 5 6 transport, most evidence now points to t...

Can Crop Models Identify Critical Gaps in Genetics, Environment, and Management Interactions?

“…Once again, these macro approaches can be robust enough (if well used) to simulate growth and can help define stress environments, and with expert use can suffice to explore the biological boundaries to growth. Beyond this relatively simple step, the task of evaluating the potential to genetically manipulating the expression of these traits belongs to more detailed photosynthesis models (e.g., Kannan et al, 2019;Wu et al, 2019). In our opinion, the expert user of a detailed simulation models must have a profound understanding of simplified approaches that retain core explanatory power and shed peripheral processes.…”

Plant Networks as Traits and Hypotheses: Moving Beyond Description

“…Example relationships include known or predicted proteinprotein interactions, gene coexpression, and regulatory binding information [35,36]. Multiscale modeling can be performed using a correlation approach [17,20], by designing a model that uses the output of one biological process as input for the next [37][38][39], or by using information across scales to constrain model parameters [40,41].…”

An open Internet of Things (IoT)-based framework for feedback controlof photosynthetic activities

“…Marshall-Colon et al (2017) used multi-scale models to facilitate the simulation of the entire plant. Feng et al (2018) developed a sevenstate chlorophyll fluorescence model for photosystem II activities, and Kannan et al (2019) evaluated the modeling process from gene expression to photosynthetic metabolism. Further, Fu et al (2020) evaluated the state of photosynthesis by a model, which included the extended Kalman filtering method, and used the estimated chlorophyll fluorescence as feedback to establish a light intensity feedback regulation model.…”