GENETIC AND MECHANISTIC LINKAGES THAT GOVERN PHOTOSYNTHETIC RESPONSES TO FLUCTUATING ENVIRONMENTS

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 1021393
• Project Start Date: Dec 9, 2019
• Project End Date: Nov 30, 2024
• Program Code: [(N/A)]- (N/A)

Recipient Organization
MICHIGAN STATE UNIV
(N/A)
EAST LANSING,MI 48824

Performing Department
Plant Research Laboratory

Non Technical Summary
Our world is facing a confluence of unprecedented challenges that directly involve the plant sciences. The U.N. estimates that food production will have to increase by 50-70% over the next 30-50 years (http://www.un.org/waterforlifedecade/food_security.shtml), whereas the increases in plant productivity seen in the early years of the "green revolution" have flattened out (1), at least in part caused by rapid climate change, depletion of nonrenewable resources, eutrophication of waters etc. that may severely diminish plant productivity (2). At the same time, plants and algae are increasingly viewed as sources of fuel to supplant fossil fuels, leading to potential competition for arable land (e.g. 3). There is thus an urgent need to develop highly productive, environmentally robust and sustainable energy and food production under a rapidly changing environment. Traditional breeding of crops has focused on maximizing many of the easily modifiable plant parameters (e.g., crop architecture, plant growth cycle), leaving the (more challenging) energy storing reactions of photosynthesis and plant responses to environmental changes as the remaining opportunities for improvement (4). The long-term aim of this proposal is to 1) Understand when and why photosynthesis is limiting for plant productivity; 2) determine when photosynthetic processes can be used as an indicator of plant physiological status or to predict diseases and yield; and 3) to use these tools to identify genetic and mechanistic linkages that underlie improved performance; 4) make this information available to accelerate the development and deployment of crop lines with improved photosynthetic efficiency and robustness; and 5) assess the utility of photosynthetic phenotype measurements to predict crop yield and diseases. This project, and its separately-funded components, address the AgBioResearch mission of Food, Energy and Natural Resources because it directly connects basic understanding of photosynthesis to the breeding and monitoring of crops for improved performance.

Animal Health Component

20%

Research Effort Categories

Basic

60%

Applied

20%

Developmental

20%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
203	1411	1000	100%

Knowledge Area
203 - Plant Biological Efficiency and Abiotic Stresses Affecting Plants;

Subject Of Investigation
1411 - Beans (fresh, fresh-processed);

Field Of Science
1000 - Biochemistry and biophysics;

Keywords

photosynthesis

phenotyping

genomics

bioenergy

plant breeding

cowpea

genome wide association

quantitative trait loci

physiology

biochemistry

instrumentation

Goals / Objectives
Our world is facing a confluence of unprecedented challenges that directly involve the plant sciences. The U.N. estimates that food production will have to increase by 50-70% over the next 30-50 years (http://www.un.org/waterforlifedecade/food_security.shtml), whereas the increases in plant productivity seen in the early years of the "green revolution" have flattened out (1), at least in part caused by rapid climate change, depletion of nonrenewable resources, eutrophication of waters etc. that may severely diminish plant productivity (2). At the same time, plants and algae are increasingly viewed as sources of fuel to supplant fossil fuels, leading to potential competition for arable land (e.g. 3). There is thus an urgent need to develop highly productive, environmentally robust and sustainable energy and food production under a rapidly changing environment. Traditional breeding of crops has focused on maximizing many of the easily modifiable plant parameters (e.g., crop architecture, plant growth cycle), leaving the (more challenging) energy storing reactions of photosynthesis and plant responses to environmental changes as the remaining opportunities for improvement (4).The long-term aim of this proposal is to 1) Understand when and why photosynthesis is limiting for plant productivity; 2) determine when photosynthetic processes can be used as an indicator of plant physiological status or to predict diseases and yield; and 3) to use these tools to identify genetic and mechanistic linkages that underlie improved performance; 4) make this information available to accelerate the development and deployment of crop lines with improved photosynthetic efficiency and robustness; and 5) assess the utility of photosynthetic phenotype measurements to predict crop yield and diseases. This project, and its separately-funded components, address the AgBioResearch mission of Food, Energy and Natural Resources because it directly connects basic understanding of photosynthesis to the breeding and monitoring of crops for improved performance.

Project Methods
The general methods are to combine genomics with detailed, high throughput phenotyping to identify the genetic and mechanistic bases for plant responses to fluctuating light and other environmental conditions. To probe and understand such variation, we developed a concept we term "Light Potential" (LP), defined as the ability of a phototroph to productively capture and store rapidly increased PAR. In the preliminary experiments shown below, we measured increases in LEF over a 10-s period following an abrupt increase from ambient to full sunlight. (Measuring LEF is very convenient for field studies, but should be confirmed by measuring assimilation to distinguish between productive LEF and photorespiration or dissipative O2 reduction, see below).As a proof of concept, we used these tools to bio-prospect for plants and environments showing various extents and limitations of LP, taken from a range of plants on MSU campus gardens and research plots and around the world. The results showed large variations in LP, that are dependent on species (or genotype), light intensity, temperature and time of day. Most important, though, the data shows a range of mechanisms that appear to limit LP, that fall into several distinct categories of phenotypes. This basic proof of concept shows that detailed phenotyping can reveal hidden connections, and begin to define the genome X environmentàphenome (G X E à P) relationships that govern the fitness and productivity of a crop. The next level of question that we address in the renewal involve linking these relationships to specific mechanisms and genetic components.The Aims in the proposed umbrella project focus on identifying the mechanistic and genetic bases of these variations, i.e. we aim to identify "new" photosynthetic control mechanisms by exploring natural variations. More specifically, the approach involves collection of multiple photosynthetic and related phenotypes using our high throughput DEPI (1) and MultispeQ platforms, in combination with the rapid phenotyping methods developed in our current and past work, to assess variations in key limiting factors in response to chilling and light intensity changes. The aggregated results are then analyzed using software developed in the lab, to identify time- and condition-dependent QTLs for each process (or phenotype) of interest. By comparing the QTL profiles for the different processes or phenotypes we can ask if, to a reasonable statistical level, the genetic diversity in one process is linked to that of another. By "linked" we mean that it is either controlled by the same genetic loci, or is mechanistically related so that one process influences the other. This "comparative QTL" approach may allow us to assess the mechanistic bases of natural variations in LP, other properties and ultimate photosynthetic yield, and determine whether the genetic variation in one property is likely to be linked to the other. Observation that QTLs for two phenotypes do not overlap, would strongly indicate that genetic diversity controlling these processes are not genetically or mechanistically linked, at least in this particular population, and at the experimental conditions and timeframe, i.e. a linkage could exist in another population or under different conditions. For instance, if we observe QTLs for LP without any overlapping QTLs for stomatal conductance, it is highly unlikely that the genetic variations in these properties are linked. On the other hand, while observing QTL overlaps can be considered as strong evidence for linkage, either genetically or mechanistically, it does not prove it, as there are likely to be multiple genes under each QTL. However, such overlaps are good clues to possible linkages, that can indicate either cause-effect relationships, which may be identified by their temporal dependence, or co-regulation by genetic control. In this case, we are interested in whether LP is linked to several hypothetical mechanisms, including photodamage to PSII and PSI, delayed fluorescence, ATP synthase activity, the protein levels and activities of PSI, b6f, PSII, maximal Assimilation capacity, xanthophyll cycle content, storage of the thylakoid proton motive force (pmf) into Dψ and DpH. For instance, the over-reduction of PSI leads to oxidative damage that in turn shuts off the CBC, we should see linkages between natural variations in our PSI assays at early times and triose-phosphate utilization (TPU)-related phenomena at later times. Note that this approach does not require the identification of specific genes under the QTLs, which is difficult. However, when possible, we will attempt to identify specific genes, using knockout lines and genetic replacements, in Arabidopsis.

Progress 12/09/19 to 09/30/20

Outputs
Target Audience: Plant scientists, plant breeders, local and global communities of farmers, researchers, extension agents, global modeling and big data efforts, entrepreneurs and international development program. Changes/Problems:Our overall aims have not changed, and we continue to develop both the tools and computational methods to assess natural variations in the responses of photosynthesis to rapidly fluctuating light. However, we have adapted our approaches to work within the constraints of the COVID-19 situation, as described in previous sections. What opportunities for training and professional development has the project provided?This section reports on our efforts to maintain education and outreach during COVD-19. Over the past few years, I have organized and taught a hands-on graduate level course on Photosynthesis. This year, partly because of the COVD-19 situations, I taught in an innovative way that allowed students to learn about photosynthesis in a way that let then continue to have hands on research experience as well to learn new skills in coding, data science and hypothesis testing. Each student or student team received a PhotosynQ MultsipeQ device, and devised group experiments that involved making measurements of plants available to them on lock-down, while asking real scientific hypotheses. The PhotosynQ platform captured all data and metadata from these experiments allowing the class to analyze them together. The main goal was to explore the data "live" and have the students try to interpret it. Sometimes these discussions revealed gaps the students' understanding, which we took advantage of to cover basic/foundational information and theory. The fact that the device measures so many different parameters simultaneously and metadata, so that the students could explore "post hoc" hypotheses generation and testing. We also used the platform to tap into on-going research project ongoing around the world. For example, one of the students has been studying sun and shade plants. Based on his results, we connected to researchers measuring plants in the Amazon rain forest to test the hypothesis that mid-canopy plants have more "dynamic" photosynthesis responses than the upper or lower canopies. The results are highly relevant to the research in the current project and partly inspired some of our new approaches. The final part of the class involved live, real-time programming using Zoom sessions. The students took on a project to update and re-code in Python, a very useful (but somewhat dated) program to fit gas exchange data. I asked my full-time lab programmer to help the students when needed. There were several cool things about this approach. First, the learned a lot about coding and the fact that the coding was often done live, they were able to see the process of building, debugging and testing that programmers use to generate robust code. Essentially, it demystified the process. Second, in order to generate the code, they had to fully understand the assumptions, logic and physics of gas exchange processes. Finally, the resulting program was applied to known data sets and came up with some very interesting results. In particular, the analysis of previously published data clearly showed that the interpretation of these experiments depends strongly on the certain assumptions and that the conclusions of many of these papers is less certain than implied in the publications. This outcome, I think, was most satisfying because it taught the students not to simply believe an outcome but to go to the basics and understand the bases of such interpretations. I am excited to report that the students have continued working on this model and are preparing a publication. Student feedback from this approach has been astonishingly positive. I finding that they are much more engaged than I've experienced in the past and they are the ones guiding the discussion and lectures, not the other way around. In addition to the classroom-based projects, we initiated an open science project, open to all researchers at PRL (and beyond) exploring the "light potentials" of photosynthesis, titled: Productive and Photoprotective Light Potential Under Diverse Fluctuating Light Environments. The entire project data can be viewed at: https://photosynq.org/projects/productive-and-photoprotective-light-potentials-1) and thus far about 2500 data sets have been collected. The ability of plant photosynthesis to respond to rapid fluctuations in light are governed by severe tradeoffs among efficiency of light capture, the need to avoid photodamage, the need to maintain downstream metabolic processes, and other costs, e.g. optimal partitioning of nitrogen and other resources. Thus, different genotypes under different environmental constraints will exhibit different "light potentials" for LEF and NPQ. The potential for plants to respond to rapidly increased light intensities is dependent on genotype (species), developmental factors and local environmental conditions, especially previous exposure to rapid fluctuations in light. The kinetics of productive light potential (P-LP) and photoprotective (NPQ-LP) of different leaves will be related to the frequency of light fluctuations previously experienced. Leaves that are completely exposed to light or in deep inner canopies will have lower LPs than leaves in intermediate canopies, which experience rapid light flecks. Plants with stiff petioles will experience "damped" wind-induced leaf motions, while those with loose or springy petioles will show rapid motions. These leaf motions will result in damped (low frequency) or fluttering/quaking motions that will impose slower or more rapid light fluctuations. The experiment termed NPQ Light Potential, was designed to detect the change in NPQ under different light intensities. The resulting data should reveal differences in NPQ that indicates the efficiency of down regulation. How have the results been disseminated to communities of interest?Publications, presentations at meetings including talks and posters (both in person and remote), as well as depositing of data, protocols and analysesthrough the hotosynQ.org openscience platform. What do you plan to do during the next reporting period to accomplish the goals?Our overall aims have not changed, and we continue to develop both the tools and computational methods to assess natural variations in the responses of photosynthesis to rapidly fluctuating light. However, we have adapted our approaches to work within the constraints of the COVID-19 situation. For example, although we continue to perform high throughput DEPI experiments, distancing requirements allow us to employ only one technician to operate the system. This decreased the number of this type of experiment that we can perform. However, we used the time to develop new methods for measuring rapid photosynthetic responses that can be performed using more mobile spectroscopic tools. For example, we have demonstrated a new method to deconvolute absorbance changes measured in leaves to yield estimates of changes in energy-dependent non-photochemical quenching (qE)in vivo. This allows us to measure changes in qE at much higher time resolutions and without the need for saturation pulses or dark acclimation that disturb the system. Using this approach, we can screen for effects of mutations or natural variations on the responses of qEto fluctuating light and test whether it is controlled by partitioning of the thylakoid proton motive foce into electric field and osmotic components. We alsodeveloped the "microPAR" sensors (described above)that can be attached directly to leaves through extremely fine and flexible wires to a data logger that captures PAR values with ~1 ms time resolution. We have already manufactured enough prototype instruments to send to colleagues at MSU to collect data from a range of species under different weather conditions etc., as well as on the KEA and related mutants developed by our collaborators. These should allow us to achieve our original Aims in a more COVID-19 suitable way.Overall, we will adapt the downstream analyses and modeling approaches to understanding these new experimental approaches, and the phenomena the reveal. In particular, we are currently updating our simulations to include the microPAR-measured responses of photosynthetic processes measured by fluorescence, electrochromic shift etc.

Impacts
What was accomplished under these goals? The most immediate goals of the past year were to write proposals to continue and expand our work. Our aim was to use the opportunity of the new and renewal proposals to develop new collaborative research directions based on our past work and technologies. This effort led to very positive interactions with colleagues, exciting new research proposals. Three externally-funded proposals were successfully funded. Specific research outcomes include the publication of 13 Journal Articles, one book chapter (peer reviewed), one patent application. Two additional papers are under review. Many of these publications reflect the broad impact of our work on collaborative projects both at MSU and internationally. We are currently finalizing a publication on the QTL-mapping of lipids and cold sensitivity of photosynthesis in cowpea, which resulted from a collaboration between the Kramer and Benning laboratories at MSU. Despite being disrupted by the COVID19 situation, we have been able to maintain, and in some cases extend our research plan by expanding thedistribution and exchange of instruments, techniques and data through the PhotosynQ.org platform, as well as by expanding our computational simulations and analyses of global PhotosynQ data sets. We have adapted our approaches to work within the constraints of the COVID-19 situation. For example, although we continue to perform high throughput experiments using the Dynamic Environmental Phenotype Imager (DEPI), distancing requirements allow us to employ only one technician to operate the system. This decreased the number of this type of experiment that we can perform. However, we used the time to develop new methods for measuring rapid photosynthetic responses that can be performed using more mobile spectroscopic tools. We have demonstrated a new method to deconvolute absorbance changes measured in leaves to yield estimates of changes in energy-dependent non-photochemical quenching (qE)in vivo. This allows us to measure changes in qE at much higher time resolutions and without the need for saturation pulses or dark acclimation that disturb the system. Using this approach, we can screen for effects of mutations or natural variations on the responses of qEto fluctuating light and test whether it is controlled by partitioning of Dψ/DpH. A major goal for this cycle is to realize the power of the tools and methods we have developed and the massive flow of data that these enabled. We are focusing on the understanding the big data itself, and the basic mechanisms that underlie the observed behaviors. This approach has been revealing unexpected phenomena in basic photosynthetic mechanisms as well as in applications to improvement of agriculture. Most recently, we have begun to concentrate on understanding the large, complex data sets we (and others) have generated, connecting our work to data science, machine learning, computational biology and related statistical methods and theory. The realization that we are all becoming data scientists and that tapping into new methods will be essential for understanding complex biological interactions, led me to propose a DOE Council Workshop entitled "At the Tipping Point: A Future of Fused Chemical and Data Science (Hypothesis driven data science)", which is scheduled for September, 2020. One unexpected advance came directly from the COVID-19 locked down. The time spent just looking out of our windows, started us thinking about the true nature of the "light fluctuations" that we have been exploring. The inspiration for this set of experiments was in watching Aspen leaves flutter in the wind. They seem to be selected to have rapid leaf motions, that should result in equally rapid fluctuations in light intensity. The question is: are the photosynthetic regulatory systems also more rapid in these species? A search through the literature suggests that, thus far, the kinetics of light fluctuations have been estimated using stationary light sensors either in the open (exposed to the sky) or under canopies. However, the light experienced by a chloroplast in a real leaf will be strongly affected not only by the canopy, but hypothetically even more so by the motions of the leaves themselves. If this is true, then the frequencies of light fluctuations will be: 1) should be strongly affected by leaf and petiole structure and physical properties; 2) be show strong variations across species and genotypes; and 3) be much high frequency than previously considered. All of these possibilities should directly impact how we address our proposed questions of ion movements and regulation of photosynthesis. We thus developed a (unique to our knowledge) Photosynthetically Active Radiation (PAR) sensor that can measure PAR experienced by leaves in real time. These "microPAR" sensors are very small (~2 X 2 mm) and weigh less than 1 mg. The microPAR sensors can be attached directly to leaves through extremely fine and flexible wires to a data logger that captures PAR values with ~1 ms time resolution. We have already manufactured enough prototype instruments to send to colleagues at MSU to collect data from a range of spies under different weather conditions etc., as well as on the KEA and related mutants developed by our collaborators. Our initial data is quite exciting: we found: 1) the extent and frequency components of these fluctuations are much stronger than those seen using conventional light sensors; 2) there is strong evidence for the involvement of KEA3 and other processes that impact the thylakoidpmf, in the responses of photosynthesis to specific patterns of light fluctuations. We plan to make a version of the microPAR sensor that can be used by students around the world. We also have focused on our computation and modeling components, incorporating modeling of our new results from "flicking" and "fluttering" light obtained using the microPAR sensors. Combining these tools with computational simulations and genetics (mutation analyses) suggests that KEA3 is regulated by both ATP and NADPH levels. Consistent with such regulation, a truncated form of the putative K+/H+antiporter,KEA3 is likely to be continuously active, i.e. is no longer regulated, leading to altered photosynthetic responses to fluctuating lightin vivo. We have used our methods to conduct a QTL mapping experiments both on an Arabidopsis MAGIC population using our Dynamic Environmental Phenotype Imager (DEPI) at MSU and on a tomato introgression population at the Max Planck Institute for Molecular Plant Physiology (Germany). The latter experiment was, unfortunately, truncated by extreme weather and the need to limit the number of researchers in the field. Nevertheless, we were able to obtain sufficient results to indicate that changes in qE kinetics were statistically significant and showed high heritability. As described below, we have also developed methods for measuring real-time fluctuations of light at leaf surfaces from a range of plant species and genotypes and "playing these back" (reproducing them) onto any leaf while measuring photosynthetic responses. To understand the complex interactions between genetic components and specific regulatory mechanisms, we adapted Exploratory Factor Analysis methods to generate networks of interactions defined by "latent variables", each of which represents a distinct mode of action of photosynthesis that links the measured parameters. We show that the response to different stresses, e.g. combinations of heat and light fluctuations activate distinct responses that are impacted by distinct sets of polymorphisms, likely indicating that multiple genetic components tune photosynthesis in different environments.

Publications

• Type: Journal Articles Status: Published Year Published: 2020 Citation: Walker, B. J., Kramer, D. M., Fisher, N., and Fu, X. (2020) Flexibility in the energy balancing network of photosynthesis enables safe operation under changing environmental conditions, Plants 9, 301.
• Type: Journal Articles Status: Under Review Year Published: 2020 Citation: Osei-Bonsu, I., Hoh, D., Chattopadhyay, A., Tessmer, O., Roberts, P. A., Huynh, B. L., Kuhlgert, S., Chen, J., Kanazawa, A., and Kramer, D. M. (2020) On the genetic and mechanistic bases of natural variations in photosynthesis responses to abiotic stresses, Plant Cell andEnvironment in press.
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Major, I. T., Guo, Q., Zhai, J., Kapali, G., Kramer, D. M., and Howe, G. A. (2020) A Phytochrome B-independent Pathway Restricts Growth at High Levels of Jasmonate Defense, Plant physiology 183, 734-749.
• Type: Book Chapters Status: Published Year Published: 2020 Citation: Kanazawa, A., Davis, G. A., N., F., Neofotis, P., and Kramer, D. M. (2020) Diversity in photoprotection and energy balancing in terrestrial and aquatic phototrophs, in Photosynthesis in Algae (Larkum, A. W. D., Grossman, A., and Raven, J. A., Eds.).
• Type: Journal Articles Status: Published Year Published: 2020 Citation: J Peng, J., Lu, J., Hoh, D., Dina, A. S., Shang, X., Kramer, D. M., and Chen, J. (2020) Identifying emerging phenomenon in long temporal phenotyping experiments, Bioinformatics 36, 568-557.
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Huang, J., Zarzycki, J., Gunner, M. R., Parson, W. W., Kern, J. F., Yano, J., Ducat, D. C., and Kramer, D. M. (2020) Mesoscopic to Macroscopic Electron Transfer by Hopping in a Crystal Network of Cytochromes, J Am Chem Soc 142, 10459-10467.
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Huang, J., Ferlez, B. H., Young, E. J., Kerfeld, C. A., Kramer, D. M., and Ducat, D. C. (2020) Functionalization of Bacterial Microcompartment Shell Proteins With Covalently Attached Heme, Frontiers in Bioengineering and Biotechnology 7, 43.
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Flood, P. J., Theeuwen, T., Schneeberger, K., Keizer, P., Kruijer, W., Severing, E., Kouklas, E., Hageman, J. A., Wijfjes, R., Calvo-Baltanas, V., Becker, F. F. M., Schnabel, S. K., Willems, L. A. J., Ligterink, W., van Arkel, J., Mumm, R., Gualberto, J. M., Savage, L., Kramer, D. M., Keurentjes, J. J. B., van Eeuwijk, F., Koornneef, M., Harbinson, J., Aarts, M. G. M., and Wijnker, E. (2020) Reciprocal cybrids reveal how organellar genomes affect plant phenotypes, Nat Plants 6, 13-21.
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Davis, G. A., and Kramer, D. M. (2020) Optimization of atp synthase crings for oxygenic photosynthesis, Frontiers in Plant Science 10, 10, 1778.
• Type: Journal Articles Status: Published Year Published: 2019 Citation: Li, J. Y., Tietz, S., Cruz, J. A., Strand, D. D., Xu, Y., Chen, J., Kramer, D. M., and Hu, J. P. (2019) Photometric screens identified Arabidopsis peroxisome proteins that impact photosynthesis under dynamic light conditions, Plant Journal 97, 460-474.
• Type: Journal Articles Status: Published Year Published: 2019 Citation: Li, J., Weraduwage, S. M., Preiser, A. L., Tietz, S., Weise, S. E., Strand, D. D., Froehlich, J. E., Kramer, D. M., Hu, J., and Sharkey, T. D. (2019) A Cytosolic Bypass and G6P Shunt in Plants Lacking Peroxisomal Hydroxypyruvate Reductase, Plant physiology 180, 783-792.