Review

# The next green movement: Plant biology for the environment and sustainability

See allHide authors and affiliations

Science  16 Sep 2016:
Vol. 353, Issue 6305, pp. 1241-1244
DOI: 10.1126/science.aag1698

## Abstract

From domestication and breeding to the genetic engineering of crops, plants provide food, fuel, fibers, and feedstocks for our civilization. New research and discoveries aim to reduce the inputs needed to grow crops and to develop plants for environmental and sustainability applications. Faced with population growth and changing climate, the next wave of innovation in plant biology integrates technologies and approaches that span from molecular to ecosystem scales. Recent efforts to engineer plants for better nitrogen and phosphorus use, enhanced carbon fixation, and environmental remediation and to understand plant-microbiome interactions showcase exciting future directions for translational plant biology. These advances promise new strategies for the reduction of inputs to limit environmental impacts and improve agricultural sustainability.

Our reliance on plants as resources is ancient and only promises to expand in the future. Faced with a growing population, estimates suggest the need for a 70% increase in agricultural production by 2050 and highlight the parallel challenges of maintaining food security, minimizing environmental damage, and managing water and resource use under shifting climate conditions (1). Along with efforts to reduce post-harvest losses and waste in the food chain, the intensification of agriculture will be essential to increase productivity (2). Precision agriculture, the breeding and genetic engineering of new crop varieties, and better management of energy, fertilizer, field systems, and irrigation inputs will also help to meet this challenge (2, 3). This growth in global production will come in regions where land and water assets are either overused or constrained and in developing countries where smallholder farmers dominate (4). Any intensification of agriculture also risks increased fertilizer use, degradation of soil quality, water availability issues, salinization, potential contamination from chemicals, and loss of genetic diversity.

The path to increased production requires a variety of approaches that minimize inputs, maximize efficiencies, and limit ecological impact. These goals drive efforts to translate advances in fundamental plant biology toward the application of plants for sustainability. Current agricultural technologies require large energy inputs for soil preparation, irrigation, and the synthesis and application of fertilizers and pesticides; these processes also produce greenhouse gases. One route to reducing those inputs is the development of plants that more efficiently use fertilizers, remove more carbon from the environment, and maintain soil and water quality. Are these challenges new? Not really. The friction between efficient production and use of inputs that drive agriculture to feed the world has always been there; that friction drove the Green Revolution of the 1960s and continues to motivate innovation and the integration of new technologies to tackle the problem (5, 6). What has changed is the increased rate of population growth and climate change that will push agriculture to the limits in the coming decades, but the tools available to meet these problems have also expanded. Building from a deeper understanding of plant genetics, biochemistry, microbiology, chemistry, and systems biology, plant scientists are working toward the next generation of crops that more efficiently use inputs for growth and mitigate environmental damages.

## The modern plant engineering toolkit

Since completion of sequencing the first plant genome in 2000, the increasing amount of data and the development of new technologies have accelerated the discovery and evaluation cycle for plant science (Fig. 1). Cheaper next-generation sequencing and improved computational power for data analysis are key for marker-assisted selection and quantitative trait locus (QTL)–guided breeding (7). The selection of plant traits by using these tools accelerates market entry of new varieties. These same tools are also essential for studies of plant-associated microbiomes that are unraveling the interplay between plants, bacteria, and fungi (8). In addition to genomic data, imaging technologies now allow large-scale phenotyping of plant growth and development. Noninvasive sensors and high-resolution spectroscopy enable image reconstruction of root and leaf architectures and real-time capture of mineral and nutrient flow (9). Similarly, monitoring technologies, including drones and satellites, and experimental systems, such as free-air concentration enrichment facilities that simulate altered climate under field conditions, expand experimental trials beyond growth chambers and greenhouses. Precision agriculture is also bringing soil, water, and atmospheric chemists together with plant scientists to simulate and anticipate changes across geographic regions.

Along with “big data” that connect genomes, phenotypes, and the environment, molecular technologies have evolved since the first generation of transgenic crops. A variety of new gene silencing and genome editing methods offers accuracy not possible with earlier methods (10, 11). Comparative genomics and computational modeling of gene and protein networks are invaluable for mapping control points and the fine-tuning of biochemical, signaling, and transcriptional pathways (12, 13). At the molecular level, protein engineering can improve activity and generate new functionality that extends metabolism. Better techniques for the introduction of multiple genes and for the targeting of gene expression to specific tissues, cells, and organelles are also bringing metabolic engineering to new levels of sophistication (14). The pallet of tools with which to understand how plants work and interact with their environments now spans from genomes to proteins and metabolism to agro-ecosystems (Fig. 1). Recent progress highlights how new technologies allow research to move from the molecular to ecosystem scales to tackle environmental and sustainability problems.

The success of modern agriculture is largely due to the availability of fertilizers that provide essential nutrients, such as nitrogen and phosphorus, to plants (Fig. 2). Without nitrogen fertilization, agriculture would support less than half our current population (15). Annually, chemical synthesis by means of the Haber-Bosch process provides ~118 million metric tons of nitrogen valued at $100 billion, with nearly half of this tonnage applied to wheat, rice, and maize worldwide (15). The Haber-Bosch process, which burns 3% of the world’s natural gas and contributes the same percentage to global carbon emissions, mixes nitrogen and hydrogen gas under high pressure (270 atm) and temperature (450°C) with an iron catalyst to generate ammonia, which is then used in fertilizers. Overuse of nitrogen fertilizers can also disrupt terrestrial and aquatic ecosystems. Compared with chemical synthesis, the symbiosis between nitrogen-fixing rhizobial bacteria and legumes, such as soybean, accounts for <25% of global nitrogen fixation (16). The development of plants with improved nitrogen use would help reduce fertilizer applications, lower energy costs and greenhouse gas emissions associated with its synthesis, and help mitigate the consequences of nitrogen loss into soil and water sources. Plant nitrogen uptake, assimilation, and storage pathways offer targets for breeding and genetic engineering of enhanced efficiency (16). For example, transgenic canola overexpressing alanine aminotransferase required 40% less fertilizer compared with wild-type plants under low-nitrogen conditions (16). Substantial efforts also aim to improve rhizobial colonization of legume roots (17). This symbiosis provides an oxygen-free environment for nitrogenase, a multisubunit bacterial metalloenzyme that converts nitrogen into ammonia for the plant (Fig. 2); however, engineering nitrogen fixation into nonlegume plants remains a challenge (18). Introducing nitrogenase into plants requires the transfer of the nif gene cluster, which encodes multiple nitrogenase subunits and accessory proteins required for metal-cofactor and subunit assembly and is under highly coordinated transcriptional control (18). Moreover, the oxygen-free environment needed by nitrogenase requires targeted expression in plants. These hurdles have led to investigations of other plant-microbial interactions that could enhance nitrogen use, such as nonlegume recognition by Rhizobium, altering plant-mycorrhizal interactions, and engineering nitrogen-fixing bacteria that associate with crops. The signaling pathway required for rhizobial nodulation in legumes is also involved in mycorrhizal fungi associations (19). Because mycorrhizal interaction is common across plants, dissecting and rewiring the system for nonlegume-Rhizobium symbiosis may be possible (19). Likewise, ectomycorrhizal and arbuscular mycorrhizal associations—in which the extensive fungal hyphae allow for efficient acquisition and delivery of soil nutrients, including nitrogen, in exchange for carbon from the plant—are also the focus of intensive research. Ectomycorrhizal fungi associated with plant roots are critical for overcoming nitrogen limitation and maintaining plant growth (20). Alternatively, nitrogen-fixing endophytic bacteria that form nodule-independent associations with crops and free-living nitrogen-fixing bacteria, such as Azotobacter and Azospirillum, are routinely used as biofertilizers and could be engineered for enhanced nitrogen fixation (19). Nitrogen tracer studies that have used the model grass Setaria viridis and bacterial inoculants suggest that this plant uptakes bacterial nitrogen, but the efficiency and universality of the process is unclear (21). If widespread, engineering of the nif cluster into bacteria that associate with nonlegume crops is possible. Experiments demonstrating the transfer of nif clusters between bacterial species and synthetic biology optimization for nitrogenase expression have been successful (22). Given the complexity of plant-rhizosphere interactions, the application of plant-microbe niche engineering needs further investigation but could provide ways for improving nitrogen use efficiency. There is also the exciting possibility of hybrid biological-chemical approaches for nitrogen fixation. Synthetic nanoparticles as generators of fixed nitrogen may be feasible; a cadmium sulfide nanocrystal/nitrogenase molybdenum-iron protein hybrid driven by light, instead of adenosine triphosphate hydrolysis, was recently described (23). Whether such nanoparticles can be stable enough to compete with the Haber-Bosch process or work under field conditions with plants remains to be explored, but synthesis of ammonia by using a light-driven catalyst is an intriguing future direction that showcases the value of multidisciplinary approaches to sustainability. ## Improving phosphorus use Phosphorus, like nitrogen, is an essential nutrient and limits crop growth in nearly 40% of agricultural land, but estimates suggest this could be as much as 70% (24, 25). Orthophosphate is the only bioavailable form for plants and is a nonrenewable resource obtained from mining. Because of its reactivity with soils rich in iron and aluminum oxides and conversion by soil microbes to forms inaccessible to plants, crops capture <30% of phosphorus applied as fertilizer. Moreover, phosphate excesses are a primary contributor to eutrophication. Recent progress toward the development of plants with improved phosphorus use will help maintain supplies of this resource and could limit overuse (Fig. 2). Traditional breeding and QTL analysis identified a rice variety with a protein kinase [phosphorus-starvation tolerance 1 (PSTOL1)] that acts as an enhancer of early root growth and provides tolerance to phosphorus deficiency (26). PSTOL1 overexpression enhanced grain yield in phosphorus-deficient soil through altered root architecture that allows for better nutrient uptake. Variations in root structure that enhance phosphorus mobilization could also be combined with improved knowledge of key plant-fungal interactions. For example, the endophytic fungus Colletotrichum tofieldiae, which associates with Arabidopsis thaliana in the wild, transfers phosphorus to shoots to promote plant growth and increases plant fertility under phosphorus-limited conditions (27). Complementing these efforts, our molecular understanding of how phosphorus is sensed was advanced by the identification of the SPX inositol polyphosphate-binding domain shared by phosphate transporters, signal transduction proteins, and inorganic polyphosphate polymerase (28). This work suggests that association of SPX domain proteins with transcription factors modulates phosphorus starvation responses. There are also efforts to alter the form of phosphorus fertilizer used by plants, with the added benefit of weed control (24). Transgenic plants expressing a bacterial protein that oxidizes phosphite to orthophosphate can grow with 30 to 50% less phosphorus input. Moreover, because orthophosphate does not discriminate between crops and weeds, use of phosphite fertilizer allowed transgenic plants to grow with up to 10-fold greater biomass in competition with weeds than when provided with orthophosphate. Deciphering the phosphorus response network in plants promises to reveal new targets for improving nutrient use. ## Photosynthesis and carbon fixation As part of terrestrial ecosystems, which sequester ~25% of anthropogenic CO2, plants use clean and renewable processes for removal of this greenhouse gas: photosynthesis and carbon fixation. The application of modern plant biology approaches toward improving photosynthesis and carbon fixation may provide solutions for carbon capture and sequestration. Because photosynthesis links sunlight to carbon fixation for the production of seeds and grains, this will also affect agriculture. Each step—from efficient light capture by leaves to the conversion of sunlight into high-energy compounds to the partitioning of carbon into plant growth and seed production—is a target for improvements through breeding and genetic engineering (Fig. 3) (29). Advances in genome data and phenotyping contribute to understanding how plants adapt to light environment and control leaf anatomy and morphology in sunlight versus shade (30). Such insights can guide the breeding of plants with improved photosynthesis. Light-capture research ranges from the macroscopic to the molecular scales. Altering developmental programs that affect leaf area and geometry would allow light to penetrate to lower-canopy leaves for maximizing light-harvesting efficiency (31). Similarly, the discovery of new chlorophylls and strategies for altering the structure of light-harvesting antenna complexes to match photosynthetic efficiency may extend the spectrum of energy available for photosynthesis and carbon fixation (32). Ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO), the entry point for CO2 into the Calvin cycle, is an obstacle for improving carbon fixation. Plant RuBisCO has a slow activation response and catalyzes an oxygenation reaction that competes with CO2 fixation. Recent work describes the first success in replacing a plant RuBisCO with a more rapid cyanobacterial homolog (33). Although the cyanobacterial RuBisCO is faster at CO2 fixation, it is also more reactive with oxygen. Expression of a subunit from the cyanobacterial carboxysome, a microcompartment that excludes oxygen and concentrates CO2, was required for higher fixation rates in transgenic plants (33). Future improvements of photosynthetic efficiency by using protein engineering to reduce RuBisCO oxygenation and to express carboxysomes and other parts of the cyanobacterial CO2-concentration machinery are logical next steps (33). A more technically demanding approach is the conversion of plants that use C3 photosynthesis to C4 photosynthesis (34). Most plants use C3 photosynthesis with RuBisCO in leaf mesophyll cells to directly fix CO2. In contrast, C4 plants separate CO2 fixation from the Calvin cycle. Making up only 5% of plant species, C4 plants fix ~30% of global CO2. Fixation of CO2 into the C4 acid oxaloacetate occurs in mesophyll cells, followed by diffusion of oxaloacetate to bundle sheath cells. There, decarboxylation of oxaloacetate provides CO2 to RuBisCO and the Calvin cycle. The oxygen-free environment enhances the efficiency of RuBisCO. Evolutionary history suggests that such radical engineering is possible; C4 plants independently evolved from C3 plants in 60 different taxa (34). All of the genes and proteins for C4 photosynthesis are known; however, generation of transgenic plants would require the introduction of more than 20 genes—a technical problem similar to that of engineering nitrogenase into plants. The more formidable challenge is how to alter C3 plants for the compartmentalization of carbon fixation and the Calvin cycle in distinct cell types. As yet, the control of leaf cell development and identification of promoters that target expression to mesophyll versus bundle sheath cells are not well understood. ## Water and salinity Irrigation of agricultural crops accounts for 70% of global fresh water consumption (35). Sustainable agriculture requires improved plant survival during short-term periods without water and more efficient water use. Although breeding has altered flowering time, height, and other traits that affect water use, the multigenic nature of plant drought responses and a lack of natural variability for crop drought tolerance have limited progress (35). Engineering plants for drought tolerance emphasizes the need for systems-level approaches that center on molecular control points with broad effects across multiple downstream targets (Fig. 3). Development of the first commercialized drought-tolerant plants used a bacterial RNA/cold stress chaperone protein to minimize drought stress effects on photosynthesis, carbon fixation, and water loss (stomata closure) in early development of maize (36). Since then, genes and proteins linked to osmoprotectants, the detoxification of reactive oxygen species produced under drought stress, and plant hormone systems that regulate stomata closure and protein degradation have been engineering targets. For example, expression of trehalose-6-phosphate phosphatase—an enzyme that alters levels of the signaling molecule trehalose—in maize improved crop yields by 49 and 123% under mild and severe drought conditions, respectively (37). Although efforts to improve crop drought tolerance are promising, the variability of crop cultivars, field conditions, and regional weather necessitate continued field studies. In some regions of the world, such as the tropics, increasingly wetter climates with prolonged rainfall and severe floods affect farming with losses comparable with those from drought (38). In Asia, flooding affects rice with annual losses of$1 billion in areas dominated by smallholder farms (38). Waterlogged roots and submergence of aerial parts typically kill plants within a week; however, adaptations in certain rice cultivars allow for survival up to twice as long. QTL analysis identified submergence 1 (Sub1) as a mediator of stress-induced expression of ethylene-responsive transcriptional regulators in rice. As with drought, the search for submergence tolerance traits is identifying key control points in low-oxygen-sensing and -signaling pathways and regulators of pathways linked to flooding survival (38); however, the extension of such knowledge from rice to other crops is just beginning.

Directly related to crop irrigation and coastal flooding is the problem of soil salinity or sodium toxicity. Crop irrigation increases salinity because of trace salts in water that accumulate over time. Saline soil conditions affect 7% of land, and nearly 30% of irrigated crops suffer from sodium contamination (39). The identification of the high-affinity potassium transporters (HKTs), which are selective for sodium and prevent accumulation of this toxic ion in leaves, provides a molecular mechanism for salt tolerance. Natural variation in HKTs is a resource for introducing salt tolerance into crops. Breeding of a wheat relative with a HKT gene into durum wheat produces a variety with reduced sodium in leaves and 25% higher grain yield on saline soils (39). Other pathways involved in salinity tolerance—including osmoprotectant synthesis, degradation of reactive oxygen species, and altered signaling systems—are also being explored for the breeding and engineering of plants that tolerate growth on saline soils (40). In addition to excluding contaminants such as sodium, plants can serve as environmental cleanup tools.

## Environmental engineering for remediation

All human activity has the potential to create environmental contaminants, including organic molecules from fuel and chemical spills, military activities, agriculture, industry, and forestry and naturally occurring inorganic contaminants that are mobilized and concentrated by mining, irrigation, and geochemical cycles. The use of microbes and plants for remediation dates back to the Romans and offers noninvasive, cost-effective, and environmentally friendly options to clean up contamination sites (41). Plants can stabilize, extract, degrade, and/or volatilize organic and inorganic contaminants, and extensive metabolic engineering efforts can improve the remediation efficiency of various pathways (41).

Better understanding of plant metabolism is leading to new approaches for using plants as environmental remediation tools. Although many explosives, including 2,4,6-trinitrotoluene (TNT), are recalcitrant to degradation, plants can detoxify TNT to a limited extent (42). Metabolism of TNT in plants generates a nitro-radical that reacts with oxygen to form toxic superoxide. Plants with a knockout of the gene encoding monodehydroascorbate reductase, which generates the nitro-radical, display enhanced TNT tolerance and may be useful for degradation of explosives on military sites (42). In addition to knockout engineering, fine-tuning enzymatic activity can also tailor plant responses to toxic metals. Directed evolution of phytochelatin synthase (PCS), an enzyme that synthesizes heavy metal–binding peptides, revealed an unexpected way of improving heavy metal tolerance through the attenuation of enzymatic activity (43). A catalytically inferior PCS resulted in phenotypic superiority because it maintained cellular redox homeostasis while synthesizing peptides for detoxification of cadmium (43). In addition to remediation applications, plants can also serve as biosensors for monitoring contamination sites. Introduction of a fluorescent zinc biosensor into Arabidopsis and poplar trees demonstrates a useful strategy for the development of sentinel plants (44). Given the physical dimensions of many waste sites and the expense of soil sampling to monitor and define contamination, improvements in plant-based remediation and the development of biosensors can provide cost-effective tools for environmental cleanup.

## Conclusions

During the past century, agriculture moved beyond the knowledge of local farmers to become a global endeavor. Investments in fundamental plant biology and agriculture revolutionized food production, but larger problems loom ahead. To meet the challenges of this century, plant biologists are working on many fronts, some highlighted here, to make sustainability a reality. In this next green movement, new technologies and multidisciplinary approaches are enabling the translation of fundamental plant biology knowledge toward the reduction of inputs and ecological footprints.

Aside from technical hurdles, agro-economics, the evolutionary constraints of plants, and the public perception of plant biotechnology remain issues. Basic science opens many routes toward a sustainable future; however, crops are commodities with a narrow profit margin, which can temper the implementation of new varieties and practices. Breakthroughs must be useful, make economic sense, and be adaptable to multiple regions and agro-ecosystems worldwide. More frequently, the combination of classic breeding and genetic engineering supported by new technologies provides greater options. Natural variation remains essential for plant improvement, but biotechnology can target complicated traits not amenable to traditional breeding. This is especially noticeable as efforts shift away from a one-gene-one-trait model toward broader systems-level targets, in which key control points that regulate downstream steps are exploited or combined to regulate multiple pathways. This also highlights the potential limits of plant genetics. Modern crops evolved under “normal” conditions, and the diversity needed for breeding new traits to meet changing climate is not always available. Moreover, efforts to improve particular traits often ignore combinations of stresses, diseases, and altered inputs. Additional work needs to consider multiple simultaneous challenges to plant growth and can be aided by the integration of large data sets and modeling studies. There is also substantial progress toward a mechanistic understanding of the plant microbiome and how bacteria and fungi work with plants for nutrient mobilization. In particular, recent discoveries on the relationships between plants and fungi in nitrogen (20) and phosphorus (27) may lead to enhancing plant-microbe interactions to minimize inputs and environmental impacts. Last, the public perception of plant biotechnology necessitates the continued education of people about the potential, and the limitations, of both breeding and genetic engineering.

The coming decades will be an exciting time for plant biologists with an eye for how to use plants for environmental and sustainability applications. A single look at teosinte and maize and one sees how adaptable plants can be for meeting our needs. What remains to be seen is whether we can harness our understanding of plants to innovate fast enough to meet the challenges poised by 9 billion people in 2050.

Correction (6 October 2016): Review: "The next green movement: Plant biology for the environment and sustainability" by J. M. Jez et al. (16 September 2016, p. 1241). On p. 1242, the first sentence of the section “Improving phosphorous use” has been updated to read, “Phosphorus, like nitrogen, is an essential nutrient and limits crop growth in nearly 40% of agricultural land, but estimates suggest this could be as much as 70% (24, 25).” A new reference (25) was added, and subsequent references were renumbered.

## References and Notes

Acknowledgments: The authors acknowledge support from the NSF (MCB-1157771 to J.M.J. and DGE-1143954 to A.M.S.). We thank our colleagues for helpful comments.
View Abstract