Overexpression of the Phosphatidylcholine:DiacylglycerolCholinephosphotransferase (PDCT) gene increases carbon flux toward triacylglycerol (TAG) synthesis in Camelinasativa seeds.

Camelinasativa has considerable promise as a dedicated industrial oilseed crop. Its oil-based blends have been tested and approved as liquid transportation fuels. Previously, we utilized metabolomic and transcriptomic profiling approaches and identified metabolic bottlenecks that control oil production and accumulation in seeds. Accordingly, we selected candidate genes for the metabolic engineering of Camelina. Here we targeted the overexpression of Camelina PDCT gene, which encodes the phosphatidylcholine: diacylglycerol cholinephosphotransferase enzyme. PDCT is proposed as a gatekeeper responsible for the interconversions of diacylglycerol (DAG) and phosphatidylcholine (PC) pools and has the potential to increase the levels of TAG in seeds. To confirm whether increased CsPDCT activity in developing Camelina seeds would enhance carbon flux toward increased levels of TAG and alter oil composition, we overexpressed the CsPDCT gene under the control of the seed-specific phaseolin promoter. Camelina transgenics exhibited significant increases in seed yield (19-56%), seed oil content (9-13%), oil yields per plant (32-76%), and altered polyunsaturated fatty acid (PUFA) content compared to their parental wild-type (WT) plants. Results from [14C] acetate labeling of Camelina developing embryos expressing CsPDCT in culture indicated increased rates of radiolabeled fatty acid incorporation into glycerolipids (up to 64%, 59%, and 43% higher in TAG, DAG, and PC, respectively), relative to WT embryos. We conclude that overexpression of PDCT appears to be a positive strategy to achieve a synergistic effect on the flux through the TAG synthesis pathway, thereby further increasing oil yields in Camelina.

Biophysical carbon concentrating mechanisms in land plants: insights from reaction-diffusion modeling.

Carbon Concentrating Mechanisms (CCMs) have evolved numerous times in photosynthetic organisms. They elevate the concentration of CO2 around the carbon-fixing enzyme rubisco, thereby increasing CO2 assimilatory flux and reducing photorespiration. Biophysical CCMs, like the pyrenoid-based CCM of Chlamydomonas reinhardtii or carboxysome systems of cyanobacteria, are common in aquatic photosynthetic microbes, but in land plants appear only among the hornworts. To predict the likely efficiency of biophysical CCMs in C3 plants, we used spatially resolved reaction-diffusion models to predict rubisco saturation and light use efficiency. We find that the energy efficiency of adding individual CCM components to a C3 land plant is highly dependent on the permeability of lipid membranes to CO2, with values in the range reported in the literature that are higher than used in previous modeling studies resulting in low light use efficiency. Adding a complete pyrenoid-based CCM into the leaf cells of a C3 land plant is predicted to boost net CO2 fixation, but at higher energetic costs than those incurred by photorespiratory losses without a CCM. Two notable exceptions are when substomatal CO2 levels are as low as those found in land plants that already employ biochemical CCMs and when gas exchange is limited such as with hornworts, making the use of a biophysical CCM necessary to achieve net positive CO2 fixation under atmospheric CO2 levels. This provides an explanation for the uniqueness of hornworts' CCM among land plants and evolution of pyrenoids multiple times.

Model validation and selection in metabolic flux analysis and flux balance analysis.

13C-Metabolic Flux Analysis (13C-MFA) and Flux Balance Analysis (FBA) are widely used to investigate the operation of biochemical networks in both biological and biotechnological research. Both methods use metabolic reaction network models of metabolism operating at steady state so that reaction rates (fluxes) and the levels of metabolic intermediates are constrained to be invariant. They provide estimated (MFA) or predicted (FBA) values of the fluxes through the network in vivo, which cannot be measured directly. These fluxes can shed light on basic biology and have been successfully used to inform metabolic engineering strategies. Several approaches have been taken to test the reliability of estimates and predictions from constraint-based methods and to compare alternative model architectures. Despite advances in other areas of the statistical evaluation of metabolic models, such as the quantification of flux estimate uncertainty, validation and model selection methods have been underappreciated and underexplored. We review the history and state-of-the-art in constraint-based metabolic model validation and model selection. Applications and limitations of the χ2 -test of goodness-of-fit, the most widely used quantitative validation and selection approach in 13C-MFA, are discussed, and complementary and alternative forms of validation and selection are proposed. A combined model validation and selection framework for 13C-MFA incorporating metabolite pool size information that leverages new developments in the field is presented and advocated for. Finally, we discuss how adopting robust validation and selection procedures can enhance confidence in constraint-based modeling as a whole and ultimately facilitate more widespread use of FBA in biotechnology.

Daylength variation affects growth, photosynthesis, leaf metabolism, partitioning, and metabolic fluxes.

Daylength, a seasonal and latitudinal variable, exerts a substantial impact on plant growth. However, the relationship between daylength and growth is nonproportional, suggesting the existence of adaptive mechanisms. Thus, our study aimed to comprehensively investigate the adaptive strategies employed by plants in response to daylength variation. We grew false flax (Camelina sativa) plants, a model oilseed crop, under long-day (LD) and short-day (SD) conditions and used growth measurements, gas exchange measurements, and isotopic labeling techniques, including 13C, 14C, and 2H2O, to determine responses to different daylengths. Our findings revealed that daylength influences various growth parameters, photosynthetic physiology, carbon partitioning, metabolic fluxes, and metabolite levels. SD plants employed diverse mechanisms to compensate for reduced CO2 fixation in the shorter photoperiod. These mechanisms included enhanced photosynthetic rates and reduced respiration in the light (RL), leading to increased shoot investment. Additionally, SD plants exhibited reduced rates of the glucose 6-phosphate (G6P) shunt and greater partitioning of sugars into starch, thereby sustaining carbon availability during the longer night. Isotopic labeling results further demonstrated substantial alterations in the partitioning of amino acids and TCA cycle intermediates between rapidly and slowly turning over pools. Overall, the results point to multiple developmental, physiological, and metabolic ways in which plants adapt to different daylengths to maintain growth.

Integrative teaching of metabolic modeling and flux analysis with interactive python modules.

The modeling of rates of biochemical reactions-fluxes-in metabolic networks is widely used for both basic biological research and biotechnological applications. A number of different modeling methods have been developed to estimate and predict fluxes, including kinetic and constraint-based (Metabolic Flux Analysis and flux balance analysis) approaches. Although different resources exist for teaching these methods individually, to-date no resources have been developed to teach these approaches in an integrative way that equips learners with an understanding of each modeling paradigm, how they relate to one another, and the information that can be gleaned from each. We have developed a series of modeling simulations in Python to teach kinetic modeling, metabolic control analysis, 13C-metabolic flux analysis, and flux balance analysis. These simulations are presented in a series of interactive notebooks with guided lesson plans and associated lecture notes. Learners assimilate key principles using models of simple metabolic networks by running simulations, generating and using data, and making and validating predictions about the effects of modifying model parameters. We used these simulations as the hands-on computer laboratory component of a four-day metabolic modeling workshop and participant survey results showed improvements in learners' self-assessed competence and confidence in understanding and applying metabolic modeling techniques after having attended the workshop. The resources provided can be incorporated in their entirety or individually into courses and workshops on bioengineering and metabolic modeling at the undergraduate, graduate, or postgraduate level.

Accurate flux predictions using tissue-specific gene expression in plant metabolic modeling.

MOTIVATION: The accurate prediction of complex phenotypes such as metabolic fluxes in living systems is a grand challenge for systems biology and central to efficiently identifying biotechnological interventions that can address pressing industrial needs. The application of gene expression data to improve the accuracy of metabolic flux predictions using mechanistic modeling methods such as flux balance analysis (FBA) has not been previously demonstrated in multi-tissue systems, despite their biotechnological importance. We hypothesized that a method for generating metabolic flux predictions informed by relative expression levels between tissues would improve prediction accuracy. RESULTS: Relative gene expression levels derived from multiple transcriptomic and proteomic datasets were integrated into FBA predictions of a multi-tissue, diel model of Arabidopsis thaliana's central metabolism. This integration dramatically improved the agreement of flux predictions with experimentally based flux maps from 13C metabolic flux analysis compared with a standard parsimonious FBA approach. Disagreement between FBA predictions and MFA flux maps was measured using weighted averaged percent error values, and for parsimonious FBA this was169%-180% for high light conditions and 94%-103% for low light conditions, depending on the gene expression dataset used. This fell to 10%-13% and 9%-11% upon incorporating expression data into the modeling process, which also substantially altered the predicted carbon and energy economy of the plant. AVAILABILITY AND IMPLEMENTATION: Code and data generated as part of this study are available from https://github.com/Gibberella/ArabidopsisGeneExpressionWeights.

Model Validation and Selection in Metabolic Flux Analysis and Flux Balance Analysis.

13C-Metabolic Flux Analysis (13C-MFA) and Flux Balance Analysis (FBA) are widely used to investigate the operation of biochemical networks in both biological and biotechnological research. Both of these methods use metabolic reaction network models of metabolism operating at steady state, so that reaction rates (fluxes) and the levels of metabolic intermediates are constrained to be invariant. They provide estimated (MFA) or predicted (FBA) values of the fluxes through the network in vivo, which cannot be measured directly. A number of approaches have been taken to test the reliability of estimates and predictions from constraint-based methods and to decide on and/or discriminate between alternative model architectures. Despite advances in other areas of the statistical evaluation of metabolic models, validation and model selection methods have been underappreciated and underexplored. We review the history and state-of-the-art in constraint-based metabolic model validation and model selection. Applications and limitations of the χ2-test of goodness-of-fit, the most widely used quantitative validation and selection approach in 13C-MFA, are discussed, and complementary and alternative forms of validation and selection are proposed. A combined model validation and selection framework for 13C-MFA incorporating metabolite pool size information that leverages new developments in the field is presented and advocated for. Finally, we discuss how the adoption of robust validation and selection procedures can enhance confidence in constraint-based modeling as a whole and ultimately facilitate more widespread use of FBA in biotechnology in particular.

Do betaine lipids replace phosphatidylcholine as fatty acid editing hubs in microalgae?

Acyl editing refers to a deacylation and reacylation cycle on a lipid, which allows for fatty acid desaturation and modification prior to being removed and incorporated into other pools. Acyl editing is an important determinant of glycerolipid synthesis and has been well-characterized in land plants, thus this review begins with an overview of acyl editing in plants. Much less is known about acyl editing in algae, including the extent to which acyl editing impacts lipid synthesis and on which lipid substrate(s) it occurs. This review compares what is known about acyl editing on its major hub phosphatidylcholine (PC) in land plants with the evidence for acyl editing of betaine lipids such as diacylglyceryltrimethylhomoserine (DGTS), the structural analog that replaces PC in several species of microalgae. In land plants, PC is also known to be a major source of fatty acids and diacylglycerol (DAG) for synthesis of the neutral lipid triacylglycerol (TAG). We review the evidence that DGTS contributes substantially to TAG accumulation in algae as a source of fatty acids, but not as a precursor to DAG. We conclude with evidence of acyl editing on other membrane lipid substrates in plants and algae apart from PC or DGTS, and discuss future analyses to elucidate the role of DGTS and other betaine lipids in acyl editing in microalgae.

The carbon-concentrating mechanism of the extremophilic red microalga Cyanidioschyzon merolae.

Cyanidioschyzon merolae is an extremophilic red microalga which grows in low-pH, high-temperature environments. The basis of C. merolae's environmental resilience is not fully characterized, including whether this alga uses a carbon-concentrating mechanism (CCM). To determine if C. merolae uses a CCM, we measured CO2 uptake parameters using an open-path infra-red gas analyzer and compared them to values expected in the absence of a CCM. These measurements and analysis indicated that C. merolae had the gas-exchange characteristics of a CCM-operating organism: low CO2 compensation point, high affinity for external CO2, and minimized rubisco oxygenation. The biomass δ13C of C. merolae was also consistent with a CCM. The apparent presence of a CCM in C. merolae suggests the use of an unusual mechanism for carbon concentration, as C. merolae is thought to lack a pyrenoid and gas-exchange measurements indicated that C. merolae primarily takes up inorganic carbon as carbon dioxide, rather than bicarbonate. We use homology to known CCM components to propose a model of a pH-gradient-based CCM, and we discuss how this CCM can be further investigated.

Flux-Balance Analysis and Mobile CRISPRi-Guided Deletion of a Conditionally Essential Gene in Shewanella oneidensis MR-1.

Carbon-neutral production of valuable bioproducts is critical to sustainable development but remains limited by the slow engineering of photosynthetic organisms. Improving existing synthetic biology tools to engineer model organisms to fix carbon dioxide is one route to overcoming the limitations of photosynthetic organisms. In this work, we describe a pipeline that enabled the deletion of a conditionally essential gene from the Shewanella oneidensis MR-1 genome. S. oneidensis is a simple bacterial host that could be used for electricity-driven conversion of carbon dioxide in the future with further genetic engineering. We used Flux Balance Analysis (FBA) to model carbon and energy flows in central metabolism and assess the effects of single and double gene deletions. We modeled the growth of deletion strains under several alternative conditions to identify substrates that restore viability to an otherwise lethal gene knockout. These predictions were tested in vivo using a Mobile-CRISPRi gene knockdown system. The information learned from FBA and knockdown experiments informed our strategy for gene deletion, allowing us to successfully delete an "expected essential" gene, gpmA. FBA predicted, knockdown experiments supported, and deletion confirmed that the "essential" gene gpmA is not needed for survival, dependent on the medium used. Removal of gpmA is a first step toward driving electrode-powered CO2 fixation via RuBisCO. This work demonstrates the potential for broadening the scope of genetic engineering in S. oneidensis as a synthetic biology chassis. By combining computational analysis with a CRISPRi knockdown system in this way, one can systematically assess the impact of conditionally essential genes and use this knowledge to generate mutations previously thought unachievable.

Re-Programing Glucose Catabolism in the Microalga Chlorella sorokiniana under Light Condition.

The microalga Chlorella sorokiniana has attracted much attention for lipid production and wastewater treatment. It can perform photosynthesis and organic carbon utilization concurrently. To understand its phototrophic metabolism, a biomass compositional analysis, a 13C metabolic flux analysis, and metabolite pool size analyses were performed. Under dark condition, the oxidative pentose phosphate pathway (OPP) was the major route for glucose catabolism (88% carbon flux) and a cyclic OPP-glycolytic route for glucose catabolism was formed. Under light condition, fluxes in the glucose catabolism, tricarboxylic acid (TCA) cycle, and anaplerotic reaction (CO2 fixation via phosphoenolpyruvate carboxylase) were all suppressed. Meanwhile, the RuBisCO reaction became active and the ratio of its carbon fixation to glucose carbon utilization was determined as 7:100. Moreover, light condition significantly reduced the pool sizes of sugar phosphate metabolites (such as E4P, F6P, and S7P) and promoted biomass synthesis (which reached 0.155 h-1). In addition, light condition increased glucose consumption rates, leading to higher ATP and NADPH production and a higher protein content (43% vs. 30%) in the biomass during the exponential growth phase.

13C-labeling reveals how membrane lipid components contribute to triacylglycerol accumulation in Chlamydomonas.

Lipid metabolism in microalgae has attracted much interest due to potential utilization of lipids as feedstocks for biofuels, nutraceuticals, and other high-value compounds. Chlamydomonas reinhardtii is a model organism for characterizing the synthesis of the neutral lipid triacylglycerol (TAG), from which biodiesel is made. While much of TAG accumulation under N-deprivation is the result of de novo fatty acid (FA) synthesis, recent work has revealed that approximately one-third of FAs, especially polyunsaturated FAs (PUFAs), come from preexisting membrane lipids. Here, we used 13C-isotopic labeling and mass spectrometry to analyze the turnover of glycerol backbones, headgroups, FAs, whole molecules, and molecular fragments of individual lipids. About one-third of the glyceryl backbones in TAG are derived from preexisting membrane lipids, as are approximately one-third of FAs. The different moieties of the major galactolipids turn over synchronously, while the FAs of diacylglyceryltrimethylhomoserine (DGTS), the most abundant extraplastidial lipid, turn over independently of the rest of the molecule. The major plastidic lipid monogalactosyldiacylglycerol (MGDG), whose predominant species is 18:3α/16:4, was previously shown to be a major source of PUFAs for TAG synthesis. This study reveals that MGDG turns over as whole molecules, the 18:3α/16:4 species is present in both DAG and TAG, and the positional distribution of these PUFAs is identical in MGDG, DAG, and TAG. We conclude that headgroup removal with subsequent acylation is the mechanism by which the major MGDG species is converted to TAG during N-deprivation. This has noteworthy implications for engineering the composition of microalgal TAG for food, fuel, and other applications.

Reimport of carbon from cytosolic and vacuolar sugar pools into the Calvin-Benson cycle explains photosynthesis labeling anomalies.

SignificancePhotosynthesis metabolites are quickly labeled when 13CO2 is fed to leaves, but the time course of labeling reveals additional contributing processes involved in the metabolic dynamics of photosynthesis. The existence of three such processes is demonstrated, and a metabolic flux model is developed to explore and characterize them. The model is consistent with a slow return of carbon from cytosolic and vacuolar sugars into the Calvin-Benson cycle through the oxidative pentose phosphate pathway. Our results provide insight into how carbon assimilation is integrated into the metabolic network of photosynthetic cells with implications for global carbon fluxes.

Kinetic complexities of triacylglycerol accumulation in developing embryos from Camelina sativa provide evidence for multiple biosynthetic systems.

Quantitative flux maps describing glycerolipid synthesis can be important tools for rational engineering of lipid content and composition in oilseeds. Lipid accumulation in cultured embryos of Camelina sativa is known to mimic that of seeds in terms of rate of lipid synthesis and composition. To assess the kinetic complexity of the glycerolipid flux network, cultured embryos were incubated with [14C/13C]glycerol, and initial and steady state rates of [14C/13Cglyceryl] lipid accumulation were measured. At steady state, the linear accumulations of labeled lipid classes matched those expected from mass compositions. The system showed an apparently simple kinetic precursor-product relationship between the intermediate pool, dominated by diacylglycerol (DAG) and phosphatidylcholine (PC), and the triacylglycerol (TAG) product. We also conducted isotopomer analyses on hydrogenated lipid class species. [13C3glyceryl] labeling of DAG and PC, together with estimates of endogenous [12C3glyceryl] dilution, showed that each biosynthetically active lipid pool is ∼30% of the total by moles. This validates the concept that lipid sub-pools can describe lipid biosynthetic networks. By tracking the kinetics of [13C3glyceryl] and [13C2acyl] labeling, we observed two distinct TAG synthesis components. The major TAG synthesis flux (∼75%) was associated with >95% of the DAG/PC intermediate pool, with little glycerol being metabolized to fatty acids, and with little dilution from endogenous glycerol; a smaller flux exhibited converse characteristics. This kinetic heterogeneity was further explored using postlabeling embryo dissection and differential lipid extractions. The minor flux was tentatively localized to surface cells across the whole embryo. Such heterogeneity must be recognized in order to construct accurate gene expression patterns and metabolic networks describing lipid biosynthesis in developing embryos.

Large fluxes of fatty acids from membranes to triacylglycerol and back during N-deprivation and recovery in Chlamydomonas.

Microalgae accumulate triacylglycerol (TAG) during nutrient deprivation and break it down after nutrient resupply, and these processes involve dramatic shifts in cellular carbon allocation. Due to the importance of algae in the global carbon cycle, and the potential of algal lipids as feedstock for chemical and fuel production, these processes are of both ecophysiological and biotechnological importance. However, the metabolism of TAG is not well understood, particularly the contributions of fatty acids (FAs) from different membrane lipids to TAG accumulation and the fate of TAG FAs during degradation. Here, we used isotopic labeling time course experiments on Chlamydomonas reinhardtii to track FA synthesis and transfer between lipid pools during nitrogen (N)-deprivation and resupply. When cells were labeled before N-deprivation, total levels of label in cellular FAs were unchanged during subsequent N-deprivation and later resupply, despite large fluxes into and out of TAG and membrane lipid pools. Detailed analyses of FA levels and labeling revealed that about one-third of acyl chains accumulating in TAG during N-deprivation derive from preexisting membrane lipids, and in total, at least 45% of TAG FAs passed through membrane lipids at one point. Notably, most acyl chains in membrane lipids during recovery after N-resupply come from TAG. Fluxes of polyunsaturated FAs from plastidic membranes into TAG during N-deprivation were particularly noteworthy. These findings demonstrate a high degree of integration of TAG and membrane lipid metabolism and highlight a role for TAG in storage and supply of membrane lipid components.

The metabolic origins of non-photorespiratory CO2 release during photosynthesis: a metabolic flux analysis.

Respiration in the light (RL) releases CO2 in photosynthesizing leaves and is a phenomenon that occurs independently from photorespiration. Since RL lowers net carbon fixation, understanding RL could help improve plant carbon-use efficiency and models of crop photosynthesis. Although RL was identified more than 75 years ago, its biochemical mechanisms remain unclear. To identify reactions contributing to RL, we mapped metabolic fluxes in photosynthesizing source leaves of the oilseed crop and model plant camelina (Camelina sativa). We performed a flux analysis using isotopic labeling patterns of central metabolites during 13CO2 labeling time course, gas exchange, and carbohydrate production rate experiments. To quantify the contributions of multiple potential CO2 sources with statistical and biological confidence, we increased the number of metabolites measured and reduced biological and technical heterogeneity by using single mature source leaves and quickly quenching metabolism by directly injecting liquid N2; we then compared the goodness-of-fit between these data and data from models with alternative metabolic network structures and constraints. Our analysis predicted that RL releases 5.2 µmol CO2 g-1 FW h-1 of CO2, which is relatively consistent with a value of 9.3 µmol CO2 g-1 FW h-1 measured by CO2 gas exchange. The results indicated that ≤10% of RL results from TCA cycle reactions, which are widely considered to dominate RL. Further analysis of the results indicated that oxidation of glucose-6-phosphate to pentose phosphate via 6-phosphogluconate (the G6P/OPP shunt) can account for >93% of CO2 released by RL.

High Flux Through the Oxidative Pentose Phosphate Pathway Lowers Efficiency in Developing Camelina Seeds.

Many seeds are green during development, and light has been shown to play a role in the efficiency with which maternally supplied substrates are converted into storage compounds. However, the effects of light on the fluxes through central metabolism that determine this efficiency are poorly understood. Here, we used metabolic flux analysis to determine the effects of light on central metabolism in developing embryos of false flax (Camelina sativa). Metabolic efficiency in C. sativa is of interest because, despite its growing importance as a model oilseed and engineering target and its potential as a biofuel crop, its yields are lower than other major oilseed species. Culture conditions under which steady-state growth and composition of developing embryos match those in planta were used to quantify substrate uptake and respiration rates. The carbon conversion efficiency (CCE) was 21% ± 3% in the dark and 42% ± 4% under high light. Under physiological illumination, the CCE (32% ± 2%) was substantially lower than in green and nongreen oilseeds studied previously. 13C and 14C isotopic labeling experiments were used together with computer-aided modeling to map fluxes through central metabolism. Fluxes through the oxidative pentose phosphate pathway (OPPP) were the principal source of CO2 production and strongly negatively correlated with CCE across light levels. OPPP fluxes were greatly in excess of demand for NAD(P)H for biosynthesis and larger than those measured in other systems. Excess reductant appears to be dissipated via cyanide-insensitive respiration. OPPP enzymes therefore represent a potential target for increasing efficiency and yield in C. sativa.

Unfair trade underground revealed by integrating data with Nash bargaining models.

Mutually beneficial resource exchange is fundamental to global biogeochemical cycles and plant and animal nutrition. However, there is inherent potential conflict in mutualisms, as each organism benefits more when the exchange ratio ('price') minimizes its own costs and maximizes its benefits. Understanding the bargaining power that each partner has in these interactions is key to our ability to predict the exchange ratio and therefore the functionality of the cell, organism, community and ecosystem. We tested whether partners have symmetrical ('fair') or asymmetrical ('unfair') bargaining power in a legume-rhizobia nitrogen-fixing symbiosis using measurements of carbon and nitrogen dynamics in a mathematical modeling framework derived from economic theory. A model of symmetric bargaining power was not consistent with our data. Instead, our data indicate that the growth benefit to the plant (Medicago truncatula) has greater weight in determining trade dynamics than the benefit to the bacteria. Quantitative estimates of the relative power of the plant revealed that the plant's influence rises as soil nitrogen availability decreases and trade benefits to both partners increase. Our finding that M. truncatula legumes have more bargaining power than their rhizobial partner at lower nitrogen availabilities highlights the importance of context-dependence for the evolution of mutualism with increasing nutrient deposition.

Modelling nutritional mutualisms: challenges and opportunities for data integration.

Nutritional mutualisms are ancient, widespread, and profoundly influential in biological communities and ecosystems. Although much is known about these interactions, comprehensive answers to fundamental questions, such as how resource availability and structured interactions influence mutualism persistence, are still lacking. Mathematical modelling of nutritional mutualisms has great potential to facilitate the search for comprehensive answers to these and other fundamental questions by connecting the physiological and genomic underpinnings of mutualisms with ecological and evolutionary processes. In particular, when integrated with empirical data, models enable understanding of underlying mechanisms and generalisation of principles beyond the particulars of a given system. Here, we demonstrate how mathematical models can be integrated with data to address questions of mutualism persistence at four biological scales: cell, individual, population, and community. We highlight select studies where data has been or could be integrated with models to either inform model structure or test model predictions. We also point out opportunities to increase model rigour through tighter integration with data, and describe areas in which data is urgently needed. We focus on plant-microbe systems, for which a wealth of empirical data is available, but the principles and approaches can be generally applied to any nutritional mutualism.