PHYSIOLOGICAL CHARAC TERISTICS LEADING TO DIFFERENCES IN DROUG HT TOLERANCE IN PHASEOLUS VULGARIS AND P. ACUTIFOLIUS By Jesse Riaz Traub A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Plant Breeding, Genetics and Biotechnology - Horticulture - Doctor of Philosophy 2015 ABSTRACT PHYSIOLOGICAL CHARAC TERISTICS LEADING TO DIFFERENCES IN DROUG HT TOLERANCE IN PHASEOLUS VULGARIS AND P. ACUTIFOLIUS By Jesse Riaz Traub Common beans ( Phaseolus vulgaris L.) are a nutritious food that provides quality protein, dietary fiber, iron, and zinc, and it is an i mportant crop in many parts of the world, especially Central America, East Africa, and South America. Drough t stress is one of the greatest limits to common bean production , for not only is drought common in areas that rely the most on beans, but many common bean cultivars in use are also sensitive to drought stress. Furthermore, under field conditions, heat st ress often coincides with and exacerbates the effects of drought stress in beans. As a result, one of the major goals of bean breeding efforts is to improve drought and heat tolerance within available germplasm. To support these efforts, the research descr ibed in this dissertation examined the physiology of drought and heat stress in a selection of bean genotypes with varying degrees of stress tolerance. T hese genotypes included tepary bean ( Phaseolus acutifolius A. Gray), a particularly stress tolerant spe cies closely related to common bean. The response of different metabolites to drought stress was a major focus. Beans exposed to drought stress had no differences in free proline concentration in their leaves, either between treatments or among genotype s. For soluble carbohydrates, no differences among genotypes were found under control conditions, but the concentration of malic acid, glucose, fructose, inositol, and raffinose all increased in the leaf tissues of plants exposed to drought stress. Glucose, fructose, and inositol were all found in higher concentrations in more tolerant genotypes, so it is likely that their accumulation is correlated with drought tolerance. These compounds accumulated in sufficient quantities to osmotically adjust bean leaf ti s sues, and those genotypes that accumulated more soluble carbohydrates under drought stress also had lower leaf water potentials while no differences among genotypes existed for leaf water potentials under control conditions. Abscisic acid was responsive t o d rought stress in beans, but differences in its concentration among genotypes did not seem directly related to drought tolerance. Grafting experiments revealed that it is shoot identity that controls the concentration of A BA in root tissues under drought stress. Drought stress also affects a number of photosyn thesis related traits in beans. Photosynthesis vs. intercellular CO 2 concentration curves revealed that none of the photosynthetic parameters derived were related to drought tolerance, but the maximu m carboxylation rate of rubisco and the rate of electron transport could be related to general productivity. Based on measureme nt s of gas exchange on control and drought stressed beans, lower stomatal conductances are associated with drought tolerant genot ypes regardless of water treatment. Lower stomatal conductances would allow a plant to conserve more water during periods of drought stress. Grafting experiments showed that stomatal conductance is controlled mainly by factors located in the shoot tissue a nd not the root tissue. However, these factors are unrelated to leaf density or the density of stomata on leaf surfaces. Bean plants exposed to temperatures of 45 °C for two days showed measurable signs of heat stress. Measures of gas exchange, chlorophyll fluorescence , and oxidative stress were for the most part only affected by this high temperature and not by any tempera tures below 45 °C . These measures also correlated well with visual signs of damage on leaf tissue caused by heat stress. The method was useful for screening a large group of germplasm for heat tolerance, but this heat tolerance only partially related to dr ought tolerance observed in the field. Plant breeders can utilize some of the methods described in this dissertation to supplement field data and further characterize the stress tolerance of later generation bean lines. iv ACKNOWLEDGEMENTS As no scientific work can happen in isolation, I would like to acknowledge several people who made the research of this dissertation possible. First, my advisor Dr. Wayne Loescher helped m e to consider new experiments that complimented what was already done and was always available to review any materials that I wanted another pair of eyes on. My guidance committee of Drs. Rebecca Grumet, Jim Kelly, and Tom Sharkey were generous with their time and expertise when I had questions for them. Jim Klug and Cody Keilen of the Growth Chamber Facilities kept the growth chambers and greenhouses where this research took place in perfect condition and gave advice on how best to control environmental co nditions. Drs. Lijun Chen and Dan Jones of the Mass Spec Facility were invaluable for designing protocols and helping me execute them for the preliminary metabolite and abscisic acid work. Dr. Sean Weise helped me analyze A - C i curves and was always availab le to answer any photosynthesis questions I had. Drs. David Kramer and Jeffrey Cruz, and Greg Austic and Linda Savage provided essential help in using phenometric systems and the PhotosynQ platform. Visiting scholars to our lab Muhammad Naeem, Chen Ji, and Dr. Kawa Ali helped take chlorophyll fluorescence and soluble carbohydrate measurements. The MSU Bean Breeding Lab members always helped to fill the gaps in my knowledge about humble Phaseolus , and the Joint Lab members aided me with critiques of and idea s for my research . The Department of Horticulture's administrative staff were so good at their jobs that I only ever had to worry about my research and not whether I was in good standing with the university. Lastly, I gratefully acknowledge MSU's Universit y Distinguished Fellowship program, USAID, and USDA - NIFA for the funding they provided both me and my research. v TABLE OF CONTENTS LIST OF TABLES ................................................... ................................ ............................. . ..... v ii i LIST OF FIGURES ........................................................................................................... . ........... ix Chapter 1: Background information .......................................... .....................................................1 Introduction ................................................................................................................ .........2 Origins and genetics of common beans and tepary beans ..................................................3 Common beans ........................................................................................................3 Tepary beans ................................................................... ........................................6 Plant response strategies to abiotic stress ............................................ ...............................7 Gas exchange in leaves .................................................................... ...................................8 Stomata ...................................................................................................................8 Gas exchange and photosynthesis .......................................................... .................9 Photosynthesis, stress, and the problem of excess energy ................ ................................ 10 Excess energy and reactive oxygen species...........................................................10 Photorespiration .... ................................................................................................11 Metabolites and stress ...................................................................................................... .13 Compatible solutes ...... ..........................................................................................13 Sugars ....................................................................................................................15 Abscisic acid .................... .................................................................................................16 Studies of common beans and abiotic stress .....................................................................19 Different factors of drought ......... .........................................................................19 Metrics related to yield .........................................................................................20 Physiological factors in stress response of beans ......... ........................................21 Other factors ..........................................................................................................23 Comparisons of common bean and tepary bean ............................................ ...................24 The research of this dissertation .......................................................................................26 Chapter 2: Drought stress and metabolites .................................................................. .................27 Introduction ................................................................................................................ .......28 Materials and methods ............................................................................. .........................30 Preliminary metabolite screen ...............................................................................30 Proline content in stressed leaves ........................................... ..............................32 Soluble carbohydrates in bean leaves ........................................................... ........34 Starch determination in bean leaves ....................................... ..............................35 Application of abscisic acid ....... ........................... .................. ..............................36 Determination of endogenous abscisic acid ............................ ..............................37 Results ................................................................ ............................................................... 40 Preliminary metabolite screen ............................................................................... 40 Proline content in stressed leaves ....................................... ................................ ..40 Soluble carbohydrates and starch in stressed bean leaves ...... .............................. 40 Exogenous abscisic acid ......................................................... ..............................47 vi End ogenous abscisic acid of leaf tissue and root tissue under drought ................47 Discussion .................................................................................................................. ....... 51 Proline and carbohydrates ...... ................................................. ..............................51 Abscisic acid ........................................................................... ..............................55 Conclusions ......................................................................................... ..............................58 Chapter 3: Drought stress and photosynthesis ................................................ ..............................60 Introd uction ......................................................................................... ..............................61 Materials and methods ........................................................................ ..............................63 Rates of pod abscission under drought stress ......................... ..............................63 Chlorophyll fluorescence of young stressed bean leaves ....... ..............................65 Leaf water potential of stressed plants ........... ......................... ..............................66 A - C i curves in well - watered conditions . ..... ............................. ..............................67 Gas exchange measurements over increasing drought stress ............... ................ 68 Survey of tepary varieties ....................................................... ..............................69 Gas exchange of grafted beans ............................................... ..............................72 Leaf density and stoma tal density ........................................... ..............................72 Results ................................................................................................. ..............................73 Rates of pod abscission under drought stress ......................... ..............................73 Chlorophyll fluorescence of young stressed bean leaves ....... ..............................73 Leaf water potential of stressed plants ...................................... ............................ 77 A - C i curves ........................................................................................................... 7 7 Gas exchange measurements over increasing drought stress ............................... 81 Survey of tepary varieties ..................................................................................... 86 Gas exchange of grafted beans ............................................................................. 88 Leaf density and stomatal density ......................................................................... 88 Discussion ...................................................................... ..................... ..............................91 Effects of drought on pod abscission, water potential, and chlorophyll fluorescence ............................................................................. ..............................91 Photosynthesis and drought ............................... ............................ ................... ....94 Variation and grafting of tepary .............................................. ..............................98 Leaf traits ........................................................ .................. ............ .............. ........101 Conclusions ......................................................................................... ............................102 Chapter 4: Heat stress screens ................................................... .......................... ............. ...........103 Introduction ......................................................................................... ............................104 Materials and methods ............................................ ........................... ............ .................106 Fifteen genotype screen across three temperatures ................ .............................106 Select genotype screen over increasing temperature ............. ............................. 109 Results ................................................................................................ .............................111 Fifteen genotype screen across three temperatures ................ .............................111 Select geno type screen over increasing temperature ............. .............................119 Discussion .......................................................................................... .............................124 Significance of different measure s of stress .......................... . ............................124 Insights from the focused experiment .................................... .............................127 Genotype performance in all measures of stress .................... ........ .....................128 vii Conclusions ........................................................................................ .............................130 Chapter 5: Conclusions and recommendations ........................................................................... 131 BIBLIOGRAPHY .......................................................................................... .............................135 viii LIST OF TABLES Table 1 - Bean genotypes used in experiments this chapter ........................... ..............................64 Table 2 - Photosynthetic parameters ............................................................... ..............................80 Table 3 - Tepary genotype survey .................................................................. ..............................87 Table 4 - Bean genotypes used in screen ................................................. ...... .............................107 Table 5 - ANOVA of heat screen experiment ............................................... .............................112 Table 6 - Correlations of measured parameters in heat screen experiment ..................... ........... 113 ix LIST OF FIGURES Figure 1 - A b s c i s i c a c i d perception pathway . ... ......... ................. . .................................................17 Figu re 2 - Proline content in bean leaves ...................................................................................... 41 Figure 3 - Soluble organic acids and carbohydrates in bean leaves ............... ..............................42 Figure 4 - Exogen ous abscisic acid ................................................................. ....................... .......48 Figure 5 - Endogenous abscisic acid ............................................................... ..............................49 Figure 6 - Abscisic acid in roots and shoots ................................................... ................ ..............50 Figure 7 - Bean pod abscission ....................................................................... .............................. 74 Figure 8 - Drought and chlorophyll fluorescence ........................................... ..............................75 Figure 9 - Leaf water potential ........................................................................ ..............................78 Figure 10 - A - C i curves ................................................................................... ..............................79 Figure 11 - Gas exchange over days . ............................................ ................. ..............................82 Figure 12 - Gas exchange with pot weights .................................................... ..............................84 Figure 13 - Gas exchange in grafted plants .................................. .................. ..............................89 Figure 14 - Heat screen with fifteen genotypes ............................................. .............................114 Figure 15 - Heat screen with five genotypes ................................................. .............................120 1 Chapter 1: Background information 2 Introduction Common beans ( Phaseolus vulgaris L.) are one of the world's vital crops. Nearly 400 million people, the majority of whom live in South and Central America and East Africa, depend upon common beans as their primary source of dietary protein (Broughton et al., 2003) . Additionally, common b eans contain high quantities of iron and zinc, two micronutrients essential to the human diet (Graham et al., 2007) . Although cereal crops such as wheat, rice, and maize provide a majority of the calories consumed by humans throughout the world, legume cro ps such as common beans are arguably as important in the human diet for providing the protein content, essential amino acids, and nutrients that are deficient in most cereals and other starchy foods (Beebe, 2012) . Furthermore, common beans are the legume c rop grown in the greatest quantity for direct human consumption, its acreage grown being greater than the next two legume crops in this category combined (FAOSTAT, 2014) . Improvements in the common bean germplasm available to growers could thus benefit mil lions of people throughout the world, and not just consumers of beans but also growers, many of whom are smallholder farmers who depend upon common beans for their livelihood (Beebe et al., 2013) . Among the many constraints to common bean production, the abiotic stressors of heat and drought are especially yield - limiting (Beebe et al., 2012; Araújo et al., 2014) . Farmers are increasingly expanding bean production into marginal areas that have greater incidences of drought and soil nutrient deficiency (Beeb e et al., 2013) . It is estimated that more than 60% of the world's common bean production is yield - limited by drought (Cavalieri et al., 2011) . Furthermore, climatologists studying the effects of climate change on agricultural production have predicted tha t the incidence and severity of heat stress and drought stress will increase over 3 the coming century (Battisti and Naylor, 2009; Lobell and Gourdji, 2012) . It will be essential for plant breeders to develop new common bean varieties adapted to withstand th ese stresses and meet the challenges imposed by climate change. The research contained within this dissertation investigates the physiology of drought and heat stress responses within common beans and a related species with the goal that plant breeders wil l find this information useful when designing crosses between different varieties and evaluating the resulting lines in the field and greenhouse. Such a strategy of focusing on phenotype and stress physiology is, in conjunction with developing genetic reso urces and plant breeding, a key part of a strategy for common bean improvement (McClean et al., 2011) . As a basis for this research, this chapter reviews the origins and characteristics of Phaseolus spp., its characterized responses to stress, and more general stress physiology. Origins and genetics of common beans and tepary beans Common beans The nucleotide diversity at different loci sampled from several geographically distinct populations provides strong evidence for a Mesoamerican origin of common beans, arising in what is today central Mexico (Bitocchi et al., 2012) . From there, common bean s spread south through Central America, northern South America, and into the Andes Mountains, diverging into five distinct wild populations: Mesoamerican, Guatemalan, Colombian, Ecuadorian - northern Peruvian, and Andean (Blair, et al., 2012) . Within this ra nge of the wild species, two separate domestication events occurred, giving rise to the Andean and Mesoamerican gene pools, which are in turn subdivided into different races (Gepts and Bliss, 1985; Singh et al., 1991) . While 4 some accessions of wild common beans are drought tolerant and well - adapted to arid environments, in the bottleneck event of domestication, much of this tolerance was lost, resulting in domesticated common beans generally being susceptible to drought stress (Cortés, et al., 2013) . These wild populations could be valuable sources of abiotic stress tolerance that could be introgressed into domesticated cultivars, but as with any crossing of wild varieties into domesticated ones, plant breeders would have to avoid also introgressing any trai ts that negatively impact yield or quality. Within domesticated common bean germplasm today, the majority of drought resistance has come from the Mesoamerica and Durango races of the Mesoamerican genepool (Singh, 1995; Beebe et al., 2013) . As with any crop , because of environmental fluctuations and the challenge of reproducing treatments, breeding for drought in common beans is a difficult process (Terán and Singh, 2002 a ; Blum, 2011) . The length of time necessary for selection for drought resistance compoun ds this difficulty , as little benefit exists in selecting for drought resistance in early generations of common beans. The low heritability of seed yield under drought and the large genotype by environment interactions argues that drought selection gains a re too small and unreliable to justify the time and cost (Terán and Singh, 2002 b ) . The genetics of common beans are relatively unexplored, especially when compared to a legume crop like soybean. However, p revious research to develop genetic markers and de ploy them in quantitative trait loci (QTL) for marker - assisted selection has been successful , especially for traits like disease resistance, which are often linked to a smaller number of major - effect genes (Liu et al., 2009; Navarro et al., 2009; Singh and Schwartz, 2010) . T he resources of common bean genetics are otherwise scarce, and they are especially so for resources related to drought. An early study found marker - assisted selection ineffective for improving yield under drought in common beans (Schneider et al., 1997 a ; Kelly and Miklas, 1998) . However, as the tools 5 available for computation and molecular genetics advanced and allowed for better genomic coverage , recent QTL studies have found SNP markers with a small but significant association w ith seed yield under drought stress despite the more quantitative nature of drought tolerance (Asfaw and Blair, 2012; Blair et al., 2012; Mukeshimana et al., 2014) . Many more drought - related QTLs are likely to be discovered , but of those that have been dis covered, few have been linked with their underlying gene (s) (Mukeshimana et al., 2014) . The future is promising: c omplementing the development of genetic markers in common beans are the increasing genomic resources for the crop. The release of the sequenc ed common bean genome (Schmutz et al., 2014) now provides opportunities for investigation using comparative genomics, transcriptomics, and further marker development, all of which researchers plan to use in bean improvement (McClean et al., 2011) . Still, r esearchers have not yet exploited these new genomic resources. Nonetheless, a survey of drought - related expressed sequence tags (ESTs) in common beans (Blair et al., 2011) helped to support a later study of the drought - activated transcriptome of two variet ies contrasting in their tolerance to this stress (Müller et al., 2014) . Furthermore, a proteomic analysis of two common bean varieties served to highlight the systemic response to drought stress and its intervarietal and interenvironmental variation (Zadr . Further research is necessary on the genetics and genomics of common beans, and researchers have not yet fully exploited the published genome. Part of the reason is that common beans are recalcitrant to tissue regeneration, so it is difficult to obtain viable transformants for the study of bean genetics. However, the efficiency of sequencing is now so great that the genomes of multiple lines can be compared with each other, and a careful application of this method can be a powerful pr obe of genetics. An increase in the understanding of common bean stress 6 physiology, as this dissertation provides, enables forward genetics approaches to investigating the underlying genetic basis for stress response in beans. Tepary beans Tepary beans ( Phaseolus acutifolius A. Gray) are a close relative of common beans native to the Sonoran Desert located in northwestern Mexico and southwestern United States. The appearance of tepary beans closely resembles that of common beans although the former tends to have smaller leaves, more leaves, and a highly - branched, bushy architecture. Like common beans, the size and coat color of tepary bean seeds show variation based on landrace or variety. In the process of adapting to its sub - tropical desert environment, tepary beans developed a high degree of tolerance to heat and drought stresses. The domestication of tepary beans occurred approximately 5,000 years ago, and the crop served as a staple for a number of native cultures in the region, among them the Hopi Pu eblos, Papago, Seri, and Mohave (Nabhan and Felger, 1978) . Within the genus Phaseolus , species are classified within several distinct genepools, with P. vulgaris placed in the primary genepool and P. acutifolius , based on its evolutionary and genetic distance, placed in the tertiary genepool (Smartt, 1981) . Common beans and tepary beans are partially reproductively compatible with each other, requiring embryo rescue after fertilization to obtain viable hybrids (B eebe, 2012) . Despite this difficulty, breeders have created a number of common bean lines into which they introgressed tepary bean genetics (Mejía - Jiménez et al., 1994; Singh and Muñoz, 1999) . Indeed, some of these introgressed lines gained prominence rece ntly when researchers at CIAT discovered their great capacity for heat tolerance as measured by pollen viability (CGIAR, 2015) . 7 Plant response strategies to abiotic stress When studying abiotic stress, it is often helpful to place a plant's response to stress in one of three categories: escape, resistance, or tolerance (Levitt, 1972) . An escape strategy involves a plant structuring its life cycle so that its quiescent phase corresponds with a period of stress and its growth and reproduction occur when e nvironmental conditions are favorable. For example, desert ephemerals exemplify an escape strategy by rapidly growing, flowering, and setting seed when a rare precipitation event happens; by the time moisture is once again scarce and drought stress would b e high, the ephemerals exist as dormant seeds in the soil. However, an escape strategy can also simply be earlier flowering and seed set in a crop exposed to moderate stress. A resistance strategy involves a mechanism that prevents a stress from penetratin g a plant's tissues; a high rate of transpiration that lowers a leaf's temperature through latent heat loss would be an example of a resistance strategy to heat stress. Finally, there is the tolerance strategy, in which the plant mitigates adverse effects from stress that does penetrate the tissues; again using heat stress as an example, a plant may produce more heat shock proteins within its leaves when exposed to high temperatures, so while the tissues of the plant do rise in temperature, the heat shock p roteins prevent some of the damage that the rise would cause. These strategies are not always clearly separate from each other, and in many cases, a mechanism for responding to stress could fall into more than one category. However, this classification of stress responses does help when comparing responses, considering how they would interact with each other, and understanding what responses would best aid survival or yield in a given environment. It should be noted that the term "tolerance" is used in thi s dissertation to describe an organism's general ability to withstand damage from adverse 8 conditions and, unless noted otherwise, is not used to describe the specific strategy detailed above. As will be seen below, the majority of stress studies in common beans have been conducted in field setting s where yield and maturity characteristics are the main parameters being measured. As a result, these studies have investigated the relation between stress escape mechanisms and agronomic success. Yet, stress resis tance and stress tolerance mechanisms remain mostly unexplored in common beans. One of the aims of this research is to more fully investigate the resistance and tolerance strategies of common beans as a compliment to the already well - characterized escape m echanisms. Gas exchange in leaves Stomata The means by which plants take up atmospheric carbon dioxide and transport it to rubisco for assimilation have ramifications for the response of plants to water stress. Models of gas diffusion in a leaf include parameters for carbon dioxide concentration in the atmosphere (C a ), in the intercellular leaf spaces (C i ), and within cells at the site of carboxylation (C c ); the diffusion between atmosphere and leaf interior is controlled by stomatal conductance (g s ), an d the diffusion between leaf interior and the site of carboxylation is controlled by mesophyll conductance (g m ) (Flexas et al., 2002) . Stomata, the pores on leaf surfaces that flanking guard cells can open or close, not only facilitate the diffusion of car bon dioxide into intercellular leaf spaces but also regulate the transpirational loss of water from the leaf. Exposure to water stress upregulates the production of abscisic acid (ABA), and ABA upregulates the production of nitric oxide, which is necessary to conduct the drought signal to guard cells and induce stomatal 9 closure (Desikan et al., 2002; Neill et al., 2008) . The nitric oxide triggers an efflux of K + from the guard cells as well as an increase in the cytosolic pH and concentration of Ca 2+ (McAin sh et al., 1992; Blatt and Armstrong, 1993; MacRobbie, 1998) . O ther phytohormones involved in the regulation of the stomatal aperture rely wholly on or have crosstalk with ABA (Hossain et al., 2011; Daszkowska - Golec and Szarejko, 2013) . However, the pathwa y responsible for ABA perception and induction of stomatal closure is not the same pathway responsible for inhibiting the opening of stomata (Yin et al., 2013) . The basic mechanisms of stomatal control are thus elucidated at the level of general plant phys iology. However, few studies have examined these mechanisms in common beans specifically. Examining the variation that exists for these traits and how well they correlate with agronomic success is essential to understanding the physiology of stress toleran ce in common beans. While the loss of water through stomata can be a problem under water stress, it also performs valuable functions. Transpiration creates a gradient in water potential that helps drive the transport of water and nutrients in the vascular tissue and the uptake of soil water by the roots. Additio nally, the latent heat loss provided by transpiration increases the yield of plant s under high temperatures (Fischer et al., 1998; Araus et al., 2002) . This last mechanism creates an antagonistic relationship between how a plant responds to heat and drought stress: it must balance its need to conserve water under drought stress with its need to avoid high temperatures under heat stress. Gas exchange an d photosynthesis The review by Flexas et al. (2004) summarizes the large body of evidence for stomatal factors and but not metabolic factors being the greater limitation to photosynthesis in plants 10 experiencing water stress. Stomata are not merely a limit ing factor of photosynthesis, but photosynthesis can regulate stomatal aperture; under conditions of limited photosynthesis, such as low irradiance, guard cells will narrow the stomatal openings in response (Wong et al., 1979) . Nor are stomata the only bar riers to gas diffusion. Under drought stress, P. vulgaris leaves maintained high C i and an unperturbed photosynthetic apparatus at 30% leaf water deficit, yet rates of assimilation were still negligible (Cornic et al., 1989) . This phenomenon gives strong e vidence for g m as a second diffusional constraint to photosynthesis that water stress can transiently increase (Delfine et al., 1999; Centritto et al., 2003) . How a plant changes its g m in response to stress helps to determine its water use efficiency (Hom mel et al., 2014) , and this effect has implications for the study of drought within and between species. While metabolic factors certainly should not be ignored, the above research highlights the importance of diffusive limitations when studying the e ffect s of stress on plants. The research reported in this dissertation probes further the diffusive resistances in beans in order to better understand their physiology under stress. Photosynthesis, stress, and the problem of excess energy Excess energy and re active oxygen species Stress reduces a plant's ability to fix carbon, but the plant is often still under high solar radiation at various points during the day, so it must safely dissipate this intercepted energy lest reactive oxygen species (ROS) damage t he plant (Allen and Ort, 2001; Foyer and Noctor, 2009) . Via morphological mechanisms, paraheliotropism allows a plant to reduce the solar radiation it is intercepting, and this response can effectively mitigate stress, especially in beans (Pastenes et al., 11 2004; Lizana et al., 2006) . However, paraheliotropism can only reduce incident solar radiation to a certain extent, and plants use other mechanisms to ameliorate high light stress. Among these mechanisms is energy dependent quenching (q E ) : under excess ex citation energy, the pH decreases in the lumen of the chloroplasts, and this acidification drives the xanthophyll cycle conversion of violaxanthin into zeaxanthin, which is necessary for q E which quenches the energy of excited chlorophyll and dissipate s it as heat (Demmig - Adams and Adams, 1996) . If this energy is not dissipated, then it can reduce oxygen to form ROS like singlet oxygen, which can in turn be converted to a less reactive state via pathways involving glutathione and ascorbate (Asada, 2006; Foy er and Noctor, 2011) . Although ROS can be damaging when unregulated, they are an integral part of redox signaling within the plant, so plant breeding or genetic engineering strategies should focus more on ROS regulation than their complete prevention (Foy er and Noctor, 2009) . For example, cotton plants engineered to have twenty times the level of glutathione reductase activity than wild type still had no photosynthetic or yield advantage in a field setting (Kornyeyev et al., 2005) despite having an advanta ge under controlled conditions (Kornyeyev et al., 2001, 2003) . Before energy dissipation and ROS scavenging traits can be breed into common beans, the strategies of tolerant varieties must first be understood. This dissertation increases the characterizati on of disparate responses in bean varieties to excess energy and the resulting damage that it causes. Photorespiration Photorespiration is another mechanism that is intimately linked to stress and excess energy, and it is especially relevant to C 3 c rops like common bean. Because r ubisco has affinities for both carbon dioxide and oxygen, it will often perform an oxygenase reaction on ribulose - 1,5 - 12 bisphosphate (RuBP) to produce one molecule of 3 - phosphoglycerate and one molecule of 2 - phosphoglycolate (Bowes et al., 1971) . The 2 - phosphoglycolate then goes through a series of reactions in multiple organelles that ultimately convert it to 3 - phosphoglycerate at the expense of reducing energy and the release of carbon dioxide and ammonia (Ogren, 1984) . Some view photorespiration's use of energy, carbon, and nitrogen as inefficient and propose alterations to the photoresp iratory pathway, engineering a r ubisco with greater affinity to carbon dioxide, or the insertion of carbon concentration mechanisms as ways to in crease productivity in C 3 plants (Long et al., 2006; Moroney et al., 2013; Singh et al., 2014) . The photorespiratory pathway is necessary for survival in photosynthetic organisms, and plants deficient in its operation cannot survive under normal atmosphe ric conditions (Somerville and Ogren, 1982; Timm et al., 2012) . Even C 4 plants with compromised photorespiratory pathways will accumulate 2 - phosphoglycolate and eventually die when they are exposed to normal atmospheric conditions although these effects ca n be prevented in low oxygen environments (Zelitch et al., 2009) . Thus, photorespiration cannot be wholly removed from plants although careful modifications are still possible. Different groups have used endogenous genes or genes inserted from the bacteriu m Synechocystis to engineer photorespiratory pathways that release carbon dioxide in the chloroplast or avoid the release of ammonia, and researchers are still evaluating these engineered lines (Kebeish et al., 2007; Peterhansel et al., 2013 ). Instead of v iewing it as a waste, it may be equally valid to view photorespiration as a mechanism of dissipating excess energy under conditions of high light (Kozaki and Takeba, 1996) . This dissipation would be especially helpful during hot and dry periods when the st omates are closed and the concentration of carbon dioxide at rubisco is low. From this viewpoint, the plant expends carbon to protect itself from photooxidative damage. 13 As for carbon concentrating mechanisms, when comparing C 3 and C 4 plants, C 4 plants theoretically will only have a photosynthetic advantage at temperatures above 30 ° C, depending on the species (Ehleringer and Pearcy, 1983) . C 4 crops perform better in warmer temperatures in part because they avoid the rise in photorespiration caus ed by rubisco 's increasing ratio of affinity for oxygen versus CO 2 as temperature increases (Jordan and Ogren, 1984) . If not from C 4 plants, a carbon concentrating mechanism from algae or cyanobacteria could be engineered into C 3 plants although different barriers to their functioning in C 3 plants exist (Moroney et al., 2013) . So even if carbon concentrating mechanisms would not be as much of a boon to C 3 crops in temperate regions, they could perhaps still increase the productivity of such crops in tropica l and subtropical areas. However, C 4 metabolism independently arose at least 45 times across a wide range of higher plants (Sage, 2004) , but C 4 plants are not ubiquitous in tropical areas, and this lack of ubiquity supports the viability of alternative met abolic strategies in hot environments. Metabolites and stress Compatible solutes Small compatible solutes, namely sugars, sugar alcohols, free amino acids, and quaternary ammonium compounds, are known to play a significant role in many plants' responses to drought stress. Compatible solutes purportedly improve stress tolerance by lowering tissue water potential, stabilizing the structure of proteins, and scavenging reactive oxygen species (Ingram and Bartels, 1996) . However, some researchers debate where compatible solutes exert their greatest effect and point out that some of these putative roles are based more on theory than in vivo evidence (Hare et al., 1998) . Indeed, based on total concentrations, 14 inorganic ion s play a greater role in adjusting water potential than organic compatible solutes in most plants (Hare and Cress, 1997) . Hyperaccumulation of compatible solutes is a physiological response most often seen in desiccation tolerant species (Hoekstra et al., 2001) , but the mechanisms of desiccation tolerance are not necessarily the same as those for drought tolerance (Serraj and Sinclair, 2002) . Indeed, Serraj and Sinclair (2002) were especially critical of previous studies looking at native and genetically engineered effects of compatible solutes, declaring that the importance of these studies were inflated because the experimenters failed to consider the results within the context of agricultural systems, which would never be productive at the low moisture levels used. Although not so bluntly critical, other authors have presented a tempered view of the efficacy of compatible solutes by considering each compound's place within its biochemical pathway and how that placement would affect the wider metabolism o f the plant (Hare et al., 1998; Zhang et al., 1999) . As will be discussed further below, the research on compatible solutes in common beans is limited, and further research is needed to fully determine the role these may play in stress response s in this sp ecies. Glycinebetaine is an amino - acid derived compatible solute that accumulates in large quantities in certain species (Chen and Murata, 2008) and helps to stabilize protein structure under abiotic stress (Murata et al., 1992) . A pplication of glycinebet aine to common beans does confer greater tolerance to drought than control treatments (Xing and Rajashekar, 1999) , but beans produce very little glycinebetaine themselves, especially when compared with a number of stress tolerant species (Storey et al., 19 77) or even when compared to Arabidopsis ( Arabidopsis thaliana) (Xing and Rajashekar, 2001) . The amino acid proline also acts as a compatible solute in certain species (Verbruggen and Hermans, 2008) , but common beans only have a modest increase in proline concentration under drought stress, and in one study comparing two varieties 15 differing in stress tolerance, the drought susceptible variety accumulated more proline under stress (Rosales et al., 2012) . Few other studies address the role of free proline in drought response in common beans, so one of the aims of this dissertation is to extend the study of proline to a greater number of varieties and conditions to better ascertain proline's effect. Sugars Sugar and carbohydrate metabolism also tend to have predictable shifts under drought tolerance. Like with proline and glycinebetaine, the concentration of soluble sugars increases with the incidence of drought stress in a number of species (Souza et al., 2004; Rizhsky et al., 2004; Xiao et al., 2009) . Conve rsely, levels of starch decrease in plants exposed to drought because of a reduction in synthesis caused by the inhibition of photosynthesis (Vassey and Sharkey, 1989; Souza et al., 2004; Muller et al., 2011) . Sugars not only reflect the metabolic state of a stressed plant but also act as signaling molecules that control a number of stress - related transcriptional pathways (Gupta and Kaur, 2005; Rolland et al., 2006) . For example, the KIN10 and KIN11 transcription factors transfer signals between stress perc eption and sugar metabolism and regulate starch production under stress (Baena - Gonzalez et al., 2007) . Drought stress initially causes growth rates and carbon metabolism to fall out of sync with each other (Muller et al., 2011) . Because photosynthesis wil l be affected more by diffusive limitations than metabolic ones under drought stress (Flexas et al., 2004) , the rate of tissue growth, which depends on cell turgor (Frensch and Hsiao, 1994; Geitmann and Ortega, 2009) , may decrease before the rates of photo synthesis do, and this decrease would result in an accumulation in the cell of carbon compounds that are no longer being utilized in the synthesis of structural components. Only one other study (Ramalho et al., 2014) has looked at the 16 accumulation of sugar s in common beans under stress; the work of this dissertation extends the study and examines the concentration of sugars and organic acids that were not included in previous studies. Abscisic acid Abscisic acid (ABA) is one of the most important stress signaling hormones in plants. ABA is mainly produced through an offshoot of the carotenoid pathway in plant chloroplasts when nine - cis epoxycarotenoid dioxygenase cleaves violaxanthin to form xanthoxin, the first committed precursor in ABA biosynthesis (Sc hwartz et al., 1997; Qin and Zeevaart, 1999) . The recently discovered interactions between ABA, 2C protein phosphatases (PP2Cs), pyrabactin resistance and pyrabactin resistance - like genes (PYR/PYLs), and a family of SNF1 related kinases (SnRK2s) allowed fo r the elucidation of the ABA perception pathway (Ma et al., 2009; Park et al., 2009) . Normally, PP2C binds to SnRK2, preventing the latter's activity. ABA facilitates the binding of PYR/PYL to PP2C, and this binding releases SnRK2 and allows it to phosphor ylate transcription fac tors to induce their activity (F igure 1). While ABA, SnRK2, and PP2C are all ancient components found in many kingdoms of life, the ABA receptor PYR/PYL arose only in higher plants after they had colonized land (Hauser et al., 2011) . After formation, ABA can be reversibly inactivated through conjugation with glucose, or it can be permanently degraded by ABA 8' hydroxylase to form phaseic acid and, further downstream, diphaseic acid (Cutler and Krochko, 1999; Yang and Zeevaart, 2006) . 17 Figure 1 - A bscisic acid perception pathway Taken from Nakashima and Yamaguchi - Shinozaki (2013) , this figure illustrates the major actors of the ABA perception pathway. ABA binds to PYR/PYL, which facilitates the latter's binding to PP2C. This binding releases SnRK2, which then induces the activity of certain transcription factors via phosphorylation . 18 Historically, some debate existed over whether plants produce ABA primarily in the roots or the shoots in response to stress (Sreenivasulu et al., 2012) . A drought signal affecting growth and stomatal response when a part of the root system was placed in dry soil was used as evidence for ABA originating in the root (Hartung et al., 2002; Wilkinson and Davies, 2002) . However, the studies using grafting bet ween wildtype and ABA deficient varieties showed that ABA - producing shoot tissue is both necessary and sufficient for ABA accumulation (Holbrook et al., 2002; Christmann et al., 2007) . Nonetheless, ABA is transported to the roots and can act as a signaling molecule there as well, so the study of root ABA is justified from a stress physiology perspective. As little research exists on ABA in common beans in general, let alone on tissue specific concentrations of ABA, this dissertation examine s these subjects in more detail within the crop of common beans. Under stress, the flux of material through the ABA synthesis and degradation pathways significantly increases, with a large increase in the breakdown products of phaseic acid and di hydro phaseic acid (Seiler et al., 2011) . Some studies have shown that in comparing stress tolerant and stress susceptible varieties within a species, it is the tolerant variety that produces less ABA and ABA breakdown products when exposed to stress (Asch et al., 1995; Seiler et al ., 2014) . A likely hypothesis is that stress tolerant varieties are able to regulate their response to stress and produce a moderated ABA signal while stress susceptible varieties produce a large, unregulated response via ABA and other signaling molecules that derange the plant's metabolism (Sreenivasulu et al., 2012) . The accumulation of ABA in reproductive tissues could be of greater relevance than leaf accumulation when looking at plants from an agricultural perspective. Under drought stress, most genes in the ABA pathway are upregulated in floral tissue except for those related to the catabolism of ABA (Kakumanu et al., 2012) . Additionally, the ABA content of 19 seed heads better correlates with reproductive success than leaf ABA content, and mutant lines d eficient in ABA catabolism show reduced yield (Ji et al., 2011) . Again, the majority of the research on the agronomic implications of ABA have been done in major cereal crops. Orphan crops like common bean have received little attention in this area. Consi dering the position of ABA as a master regulator of stress perception and response, the study of ABA in common beans is necessary to a better understandin g of its stress physiology. The research in this dissertation was designed to establish a basic unders tanding of ABA in common beans and how it reacts to drought stress that futu re studies can use. Studies of common beans and abiotic stress Different factors of drought When breeding common beans for drought tolerance, whether the beans will be in areas with terminal end of season drought or intermittent drought throughout the season influences which traits will be important for their agricultural success (Muñoz - Perea et al., 2006) . Higher plasticity of reproductive phase timing provides common beans an a dvantage under conditions of terminal drought stress (Acosta - Gallegos and White, 1995) , but this same plasticity could reduce yield in an intermittent drought setting. Perhaps just as important as the type of drought is the physiological stage of the bean plant when drought occurs. Drought that occurs during the reproductive phase has a much more negative impact on yield than drought that occurs during vegetative or grain - filling stages (Nielsen and Nelson, 1998) . Carbon isotope discrimination assays on a collection of diverse common bean lines revealed significant differences in water use efficiency based on the region the lines bred for but not on ancestral genepool (Ehleringer et al., 20 1991) . Nonetheless, materi als adapted to one type of drought may still be of value for breeding in the other type of drought environment (Frahm et al., 2004) . Metrics related to yield A field study of common beans under nonstress and drought conditions showed no clear correlatio ns between yield and water use efficiency, total biomass, or harvest index under either treatment (Muñoz - Perea et al., 2007) ; rather, high - yielding varieties were able to compensate with a high score in one metric for a deficiency in another. This observat ion makes sense given the mathematical interrelatedness of the terms: Y = T x WUE x HI (1) where Y is yield, T is transpiration, WUE is water use efficiency in terms of weight aboveground biomass per amount of water transpired, and HI is harvest index (Passioura, 1996) . Blum (2009) argues that water use efficiency is an unhelpful target for plant breeders and that the concept of effective use of water better represents the agronomic yield of crops in water - limiting conditions, but as long as its mathematical relationship to yield is known , water use efficiency can still serve as a helpful summary number. In a study examining the yield of common beans under terminal drought and intermittent drought conditions, harvest index decreased slightly in response to drought while yield strongly correlated with aboveground biomass and strongly negatively correlated with days until flowering for both drought types (Rosales - Serna et al., 2004) ; these results support overall plant vigor under drought and esca pe from drought through earliness as the greatest contributors to 21 yield under stress. Early maturity correlating positively with yield under drought stress has been observed in other studies (Singh, 1995; Terán and Singh, 2002 a ) . When breeders select for d rought tolerance, they want to also make sure that they are not inadvertently selecting against yield under well - watered conditions. For two quantities, their geometric mean is the square root of their product. The geometric mean of a cultivar's yield unde r drought stress and yield under well - watered conditions in the same environment has proven a simple way to help identify which lines show superior performance under both conditions (Schneider et al., 1997 b ; Terán and Singh, 2002 a ; Frahm et al., 2004) . The geometric mean can be a helpful metric in heat stress experiments as well (Porch, 2006) . Physiological factors in stress response of beans Although some varieties yield well in all conditions (Terán and Singh, 2002 a ; Rao et al., 2013) , field yields unde r well - watered conditions were poorly correlated with yields under drought conditions as varieties yielding well under drought stress are surpassed by other varieties under well - watered condtions (Terán and Singh, 2002 b ; Lizana et al., 2006) . In one study (Lizana et al., 2006) , the tolerant genotype maintained its growth rate and abscised fewer of its reproductive organs under drought stress while the susceptible genotype had large rates of floral and pod abscission. Additionally, soil and leaf water statu s more strongly controlled stomatal aperture than light levels in the tolerant genotype while the reverse was true for the susceptible genotype (Lizana et al., 2006) . In another study using the same two genotype s exposed to different light intensities, the tolerant genotype was able to utilize higher light intensities for photosynthesis, and this utilization correlated more with plasticity of stomatal development than with accumulation of carotenoids (Wentworth et al., 2006) . 22 In a proteomic study of commo n beans under drought stress, when compared to the susceptible genotype , the tolerant genotype had more abundant changes in proteins related to photosynthesis and fewer in proteins related to stress response and ROS scavenging et al., 2013) . Int erestingly, the strongest individual protein group contrast was that the tolerant genotype down regulated its abundance of oxygen evolving enhancer proteins (OEEs) while the susceptible genotype upregulated all of its OEEs under drought stress. Hypothetica lly, the difference in abundance of OEEs could represent the broader stress response strategies of these genotype s: while the susceptible variety enhances its photosynthetic capacity in response to stress and dangerously uncouples its available reducing energy from its carbon supply, the tolerant variety reduces its photosynthetic capacity and thus dissipates the incident energy in other ways. In co mparisons of three bean genotype s under drought stress, the genotype that showed the greatest increase in lipid peroxidation and membrane disruption also had higher amounts of ROS and lower activity for ROS - scavenging enzymes (Zlatev et al., 2006) . A simil ar experiment using the same genotype s studied chlorophyll fluorescence, gas exchange, metabolite, and water potential parameters (Ramalho et al., 2014) . The significant difference among genotype s was that the high ROS line from Zlatev et al.'s (2014) expe riment also had lower rates of photosynthesis, stomatal conductance, and photosystem II efficiency under drought stress, but it also had higher levels of carotenoids (Ramalho et al., 2014) . Water deficits also reduce common beans' capacity to fix nitrogen although resistance to this reduction exists within the germplasm (Devi et al., 2013) . Different cultivars have different rates of pollen derangement under heat stress with commensurate rates of pollen viability and seed yield per plant (Porch and Jahn, 2001) . Although increases in temperature only moderately affected rates of photosynthesis and conductance, it did 23 severely reduce pollen viability and seed set, and increasing the ambient carbon dioxide concentration could not compensate for stress induced by heat (Prasad et al., 2002) . These findings suggest that the warmer global temperatures of the future will restrict the growth of common beans in production zones as the higher carbon dioxide concentration will do little to improve pollen viability. The recent release by the International Center for Tropical Agriculture (CIAT) of a number of heat tolerant common bean lines (CGIAR, 2015) may allow farmers to maintain production in the areas where they grow common beans over the coming decades. The relati ve contributions of shoot tissue and root tissue to drought response is an issue of much discussion. In a grafting experiment involving several different lines of common beans, White and Castillo (1992) found that root identity had the largest impact on s mall plot yield under water - limiting conditions. Shoot identity affected days to maturity, but the authors considered the shoot's effect on seed yield to be insignificant. Sponchiado et al. (1989) also concluded that common bean's yield under drought stres s is primarily associated with rooting depth, but this association was not true across all sites that they tested, and the drought susceptible varieties accumulated near identical weights of above ground biomass as the tolerant varieties. Other factors While a crop in a field setting is sometimes affected by just one stress, in many cases, a crop will undergo multiple coincident stresses. These stresses can exacerbate each other; for example, high temperatures will damage a drought stressed plant more be cause it lacks the water to cool itself through evapotranspiration. Most studies focus on a single stress because it is easier to draw conclusions about the effects of that stress when it is the only variable being 24 manipulated. However, a plant's biochemic al and transcriptional response to two coincident stresses can be substantially different from its responses to those stresses individually (Rizhsky et al., 2004; Rasmussen et al., 2013) . One inadvertent form of stress in greenhouse and growth chamber exp eriments comes from an inadequate consideration of root conditions (Poorter et al., 2012 b ) . Small pot sizes limit root growth and can negatively impact photosynthesis while using improper soil media and watering regimes can place the roots in a near consta nt state of hypoxia (Passioura, 2006; Poorter et al., 2012 a ) . In designing experiments, Poorter (2012a) recommends not exceeding 2 g plant biomass per L of soil volume. Comparisons of common bean and tepary bean Common bean and tepary bean have signific antly diverged for some of their responses to stress to adapt to their respective native environments. Nonetheless, the two species are closely related, and their shared genetics is evidenced by their similarities in morphology and metabolism. Common bean and tepary bean can both follow a drought escape strategy of early maturity during a late season terminal drought although this escape strategy is not ubiquitous in former as it is in latter (Nabhan and Felger, 1978; White and Singh, 1991) . Exposure to hig h temperatures for prolonged periods completely prevents seed set in many common bean lines, but many tepary beans are able to yield seed, albeit a reduced yield, at the same extreme temperatures (Rainey and Griffiths, 2005 b ) . In a field experiment tepary bean lines yielded more under water limited conditions and fair to excellent under irrigated conditions in comparison with all the common bean lines tested (Rao et al., 2013) . However, the more drought tolerant 25 common beans were closer in performance to tepary beans than to other common bean lines under drought stress (Rao et al., 2013) . Interspecific inbred backcross lines yielded poorly under both conditions, and this poor performance raises concerns about the ease with which drought tolerance from tepary bean can be bred into common bean. The total aboveground biomass of the tepary lines was not substantially different from biomass of the common bean lines, so the high yield of the tepary bean lines results in part from their superior harvest index under drought conditions (Rao et al., 2013) . When comparing the two species, tepary bean's root systems grows to greater depth, and tepary bean reduces its stomatal conductance more severely at higher leaf water pot entials (Markhart, 1985) . Conversely, tepary bean lost its root length advantage when exposed to soil aluminum stress, and even under combined aluminum and drought stress, tepary bean had lower root lengths than common bean (Butare et al., 2011) . It is lik ely that tepary bean is more sensitive to metal toxicity in general. In drought experiments featuring grafts between common bean and tepary bean, leaf water potential associated most closely with root identity, and plants with a tepary rootstock tended to have higher leaf water potentials (Sanders and Markhart, 1992) . These results support the hypothesis that tepary bean has higher root hydraulic conductivity than common bean. However, when increased osmoticum in the watering solution was used as the source of water stress and when cuticular barriers to diffusion were removed, both species had similar leaf water potentials and photosynthetic responses (Castonguay and Markhart, 1991) . A similar experiment measuring photosynthesis in intact leaves revealed tha t tepary bean had a higher carboxylation efficiency and a higher water use efficiency than common bean under water stress (Castonguay and Markhart, 1992) . While both common bean and tepary bean exhibit paraheliotropic movement of their leaves, tepary bean increases its leaf angle to a greater degree 26 under conditions of high light and temperature (Bielenberg et al., 2003) . Studies on ABA levels within the genus Phaseolus have also been carried out. During extended exposure to salt stress, common beans have i ncreasing levels of leaf ABA with increasing levels of salinity, but tepary beans show no increase in leaf ABA content, whatever the salt level (Yurekli et al., 2004) . Tepary bean accumulates more ABA in its leaves under heat stress than either a heat tole rant common bean line or a heat susceptible one (Udomprasert et al., 1995) . Overall, tepary bean uses a mix of increased resistance and tolerance strategies that allow it to perform better under stress than common beans; tepary bean's increased rooting de pth and reduced stomatal conductance during stress serve to place more water in its tissues, and its increased carboxylation efficiency allows it to better tolerate a decreased CO 2 supply caused by closed stomates. The research of this dissertation As e xamined in this review, common beans are an important crop for many regions, but drought stress often limits yields and heat stress limits areas of production. Building off previous research in beans and other species, this dissertation examines the physio logical response to drought or heat stress of a small number of bean genotypes that differ in their tolerance to abiotic stress. The effect of drought on concentrations of proline, carbohydrates, and abscisic acid is examined in C hapter 2. The relation of drought tolerance to photosynthetic and leaf charac teristics is examined in Chapter 3. Finally, C hapter 4 examines the effects of high temperatures on photosynthesis and indicators of stress on a larger group of bean germplasm. 27 Chapter 2: Drought stress and metabolites 28 Introduction Common beans ( Phaseolus vulgaris ) are important food crop s in the United States, Central America, South America, and East Africa (Akibode and Maredia, 2011) and are the most widely grown legume for direct human consumption (FAOSTAT, 2014) . Beans are a nutritious foodstuff, and in addition to being high in quality protein, they provide the nutrients iron and zinc in appreciable quantities as well. Deficiencies in these two nutrients are common throughout the world (Fletcher et al., 2004) , so greater access to and consumption of beans are relevant to the public health of both developed and developing countries. Drought stress is among the greatest limiting facto rs to the production of common beans (Beebe et al., 2013) , so breeding drought tolerant cultivars would help create a more stable supply of beans. A greater understanding of bean's physiological response to drought stress helps inform plant breeders of th ose traits to focus on during the selection process. A number of metabolites are known to be involved in the abiotic stress response in plants. The amino acid proline was shown to accumulate in Arabidopsis in response to cold and drought stress (Wanner and Junttila, 1999; Urano et al., 2009) , but the extent of variation among tolerant and susceptible genotypes in the model organisms for proline accumulation remains understudied (Verslues and Juenger, 2011) . Proline also accumulated in crop plants such as wh eat and rice experiencing osmotic stress (Garcia et al., 1997; Nayyar, 2003) , and exogenous proline alleviated salt stress imposed on suspension - cultured tobacco cells (Hoque et al., 2007) . Sugar metabolism also changes in response to drought stress. As d rought stress increases and photosynthesis decreases , a smaller supply of photosynthate is available to developing reproductive tissues, which can lead to their abortion (Pinheiro and Chaves, 2011) ; supplying 29 sucrose to drought stressed maize ears rescued their seed yield (Zinselmeier et al., 1995) . However, at moderate levels of stress, the supply of soluble carbohydrates often increases in leaves because of the cessation of growth and export (Chaves and Oliveira, 2004) . In maize, drought stress affected t he transcriptional and metabolic profiles of many sugars (Witt et al., 2012; Kakumanu et al., 2012) . The accumulation of sugars under drought stress is thought to result in osmotic adjustment of plant tissues (Levitt, 1972; Morgan, 1984) , but the degree to which sugars contribute to osmotic adjustment is questioned, especially when compared to contribution by inorganic solutes (Hare and Cress, 1997; Hare et al., 1998; Serraj and Sinclair, 2002; Gagneul et al., 2007) . Perhaps just as important to drought str ess, the concentrations of certain soluble carbohydrates act as signals of stress and metabolic status (Carrari et al., 2004; Koch, 2004; Smith and Stitt, 2007; Usadel et al., 2008; Pinheiro and Chaves, 2011) . Although sugar responses in common beans are n ot well studied, Ramalho et al. (2014) measured an increase in soluble sugars in leaf tissues when bean plants were exposed to drought stress. Abscisic acid (ABA) is the major plant hormone for signaling abiotic stresses, especially drought stress (Wilkin son and Davies, 2002) . Upon exposure to drought stress, a plant's concentration of ABA increases, which in turn causes the stomata to close by triggering an efflux of ions from the guard cells (Mori and Murata, 2011) . While the closure of stomata conserves water, it also limits photosynthesis by decreasing the influx of carbon dioxide to sites of carbon fixation (Chaves, 1991) . ABA is thus critical to determining the balance between productivity and water loss. However, drought tolerance is not simply a mat ter of accumulating ABA as no relation exists between ABA concentration under drought stress and a plant's degree of drought tolerance (Chaves and Oliveira, 2004) . Rather, a plant's survival and productivity under drought stress depends on the maintenance of ABA homeostasis (Sreenivasulu et al., 2007) . While ABA 30 has been well studied in Arabidopsis and cereal crops (Tian et al., 2004; Harb et al., 2010; Kanno et al., 2010; Ji et al., 2011; Kakumanu et al., 2012; Yin et al., 2013; Seiler et al., 2014; Dalal and Inupakutika, 2014) , a dearth of information on ABA in beans exists, despite the crucial step in ABA biosynthesis first being discovered in common bean (Qin and Zeevaart, 1999) . The current study used a diverse group of bean genotypes contrasting in th eir degree of drought tolerance, including the drou ght tolerant related species tepary bean ( Phaseolus acutifolius ), to investigate the role of several metabolites in drought response in beans. The concentration of the amino acid proline, the concentration of several carbohydrates, and the concentration of , and sensitivity to , ABA were all measured in a range of bean genotypes. Materials and m ethods Preliminary metabolite screen To determine which compounds were associated with drought resp onse in beans, a preliminary metabolite screen was performed. Plants were grown in a growth chamber at a temp erature of 25 ° C during the light period and 20 ° C during the dark period , and the chamber ran according to a 14 h light and 10 h dark cycle. Light intensity was 400 µ mol photons m - 2 s - 1 PAR during the day cycle. Plants were planted in 10 L black plastic pots filled with Suremix perlite potting media (Michigan Grower Products, Galesburg, Michigan). Common bean genotypes Jaguar, Zorro, and SER - 16 were used in the experiment as well as the improved tepary bean line TB1 provided by Dr. Timothy Porch from USDA - ARS in Mayaguez, Puerto Rico (Porch et al., 2013) . Jaguar and Zorro are two elite black bean cultivars bred in Michigan by Michigan State University ; Jaguar is an older cultivar that Zorro completely replaced 31 commercially after the latter's release because of its higher and more stable yield (Kelly et al., 2009 a ) . SER - 16 is a small red bean line bred in Colo mbia by CIAT as part of their development of abiotic stress tolerant lines, and it has shown tolerance to stress in field settings (Beebe et al., 2008; Rao et al., 2013) . In terms of drought tolerance, Jaguar is a susceptible genotype, Zorro is moderately tolerant, SER - 16 is tolerant, and tepary is very tolerant. Several seeds of each of the four varieties were sown in the same pot, and after germination, the seedlings we re thinned so that each pot contained four plants representing one plant of each variety. Two water treatments were imposed on the plants: well - watered and drought stress, and both treatments had six replicates each. The pots were arranged in a randomized complete block design with pots as the blocking factor. After sowing, all plants were regularly watered with half - strength Hoagland solution mixed with a soluble fertilizer to bring the final nitrogen concentration of the solution to 15 mM. All plants deve loped mature trifoliate leaves within four days of each other, and after the last plant had matured , water was withheld from the drought stress treatment while the well - watered treatment continued to be regularly supplied with deionized water. After five d ays of withholding water, the plants in the drought stress treatment were wilting, so one leaflet was harvested from all plants by using a razor to cut the leaflet from the petioule , and the harvested leaflets were flash - frozen in liquid nitrogen, stored a t - 80 ° C, and finally lyophilized. After lyophilization, the samples were finely ground using a Wiley Mill (Thomas Scientific, Swedesboro, New Jersey, USA) and weighed, and approximately 100 mg of each sample was then extracted with 2 mL of 80% ethanol fo r 30 minutes at 65 ° C. Then, the samples were centrifuged, and the supernatant decanted. The remaining pellets were extracted twice more as above, and the supernatants were pooled and 3 mL of water added to each one. Then, 3 mL of 32 chloroform was added to th e supernatant, they were vortexed together, and the phases were allowed to separate overnight in a 4 °C refrigerator . The next day, the aqueous upper portion was pipetted into a separate tube and completely dried down on a Speedvac with minimal heat. Dry s amples were resuspended in 0.5 mL of pyridine containing 30 mg/mL hydroxylamine hydrochloride and 1 mg/mL xylitol (as an internal standard) and placed on a heating block set to 75 ° C for one hour. After the samples cooled, 1 mL of hexamethyldisilazane was added to each sample followed by 0.1 mL of trifluoracetic acid. The samples were vortexed and allowed to sit for one hour. Twelve leaf extracts , taken evenly from the genotypes Jaguar and tepary and the drought stressed and control treatments, were then ta ken to Michigan State University's Mass Spectrometry and Metabolomics Core , and the compounds contained in them were identified using gas chromatography - mass spectrometry. Compounds accumulating in large quantities in any of the samples were marked for fur ther study. Proline content in stressed leaves Bean plants were grown in a growth chamber on a 14 hour light and 10 hour dark cycle, with 300 µ mol photons m - 2 s - 1 of PAR supplied by a combination of fluorescent and incandescent lamps during the light period . Temperature was maintained at 21 ° C. Plants were grown in 1 L black plastic pots filled with Suremix Pearlite potting media (Michigan Grower Products, Galesburg, Michigan). As leaf samples were taken from young plants, these pots had sufficient cap acity for the plants' small sizes. Four different common bean genotype s were used in this experiment: Jaguar, Zorro, SER - 16 , and RAB - 651 . With the first three genotypes described above in 'Preliminary metabolite screen', RAB - 651 is a red bean line bred in Colo mbia by CIAT as part of their 33 development of abiotic stress tolerant lines, and it has also shown tolerance to stress in field settings (Beebe et al., 2008; Rao et al., 2013) . Each variety was sown into two different water treatments: well - watered and drought stressed. Each variety by treatment comb ination had six pots assigned to it, with one plant per pot , and the plants were grown in a randomized complete block design. All pots were irrigated with half - strength Hoagland solution and kept well - wate red until the first trifoliate leaf was fully mature, and then were subjected to differential water treatments. Three days elapsed between the first plants to have mature trifoliates and the last plants to have mature trifoliates, and the watering treatmen ts did not begin until these last plants were fully mature. T he well - watered treatment continued to be watered with deionized water and maintained between 60 - 100% of pot water - holding capacity (pot capacity) as determined by weighing the pots. W ater was wi thheld from the drought stressed t reatment for three days, at which point, their pot capacity was in the range of 20 - 30%. Then, one leaflet from the first tr ifoliate leaf of each plant was cut with a razor at the connection of petioule to leaf blade , immed iately frozen in liquid nitrogen, and stored at - 80 ° C . All leaf samples were lyophilized and proline content was determined according to (Bates et al., 1973 ) with the following modifications. Briefly, each leaf sample was ground using a Wiley Mill , weighed , and then extracted with 0.5 mL of 3% sulfosalicylic acid four times, and the homogenate was pooled and centrifuged. Acid ninhydrin was made by combining 1.25 g of ninhydrin with 30 mL of glacial acetic acid and 20 mL of 6 M phosphoric acid until d issolved. The homogenate of each sample was then reacted with 2 mL of glacial acetic acid and 2 mL of acid ninhydrin for 1 hour at 100 ° C. The samples were then cooled in an ice bath, extracted with 4 mL of toluene, and the absorbance of the proline contai ning toluene was measured at 520 nm 34 using a spectrophotometer. Pure toluene was used as a blank, and the proline concentration of each sample was determined from a standard curve. All statistical analysis was performed using the program SAS (SAS Institute , Cary, North Carolina). The data was analyzed as a two - factor randomized complete block design using the code PROC MIXED. For factors that were significant, the individual treatment means were compared using Least Significant Difference. An alpha value of 0.05 was set as the threshold for which results would be considered statistically significant. Soluble carbohydrates in bean leaves Plants were grown in a growth chamber using the same environmental conditions as described above in 'Preliminary metaboli te screen'. Several seeds of each of the genotypes Jaguar, Zorro, SER - 16, and tepary were sown in the same pot, and after germination, the seedlings were thinned so that each pot contained four plants representing one plant of each variety. Two water treat ments were imposed on the plants: well - watered and drought stress, and both treatments had six replicates each. The pots were arranged in a randomized complete block design with pots as the blocking factor. After sowing, all plants were regularly watered w ith half - strength Hoagland solution mixed with a soluble fertilizer to bring the final nitrogen concentration of the solution to 15 mM. All plants developed mature trifoliate leaves within three days of each other, and after the last plant had matured, wat er was withheld from the drought stress treatment while the well - watered treatment continued to be regularly supplied with deionized water. After five days of withholding water, the plants in the drought stress treatment were wilting, so one leaflet was ha rvested from all plants by using a razor to cut the leaflet from 35 the petioule, and the harvested leaflets were flash - frozen in liquid nitrogen, stored at - 80 ° C, and finally lyophilized. After lyophilization, the samples were finely ground using a Wiley Mill (Thomas Scientific, Swedesboro, New Jersey, USA) and weighed, and approximately 100 mg of each sample was then extracted with 2 mL of 80% ethanol for 30 minutes at 65 ° C. Then, the samples were centrifuged, and the supernatant decanted. The remaining p ellets were extracted twice more as above, and the supernatants were pooled and 3 mL of water added to each one. Then, 3 mL of chloroform was added to the supernatant, they were vortexed together, and the phases were allowed to separate overnight in a 4 °C refrigerator. The next day, the aqueous upper portion was pipetted into a separate tube and completely dried down on a Speedvac with minimal heat. Dry samples were resuspended in 0.5 mL of pyridine containing 30 mg/mL hydroxylamine hydrochloride and 1 mg/ mL xylitol (as an internal standard) and placed on a heating block set to 75 ° C for one hour. After the samples cooled, 1 mL of hexamethyldisilazane was added to each sample followed by 0.1 mL of trifluoracetic acid. The samples were vortexed and allowed t o sit for one hour. T he concentration of compounds identified previously in the metabolite screen was analyzed for each sample using gas chromatography. Starch determination in bean leaves Starch levels were determined from leaf samples taken from an ide ntical replication of the 'Soluble carbohydrates in bean leaves' experiment . The concentration of soluble sugars in leaf samples from this experiment was measured as described in 'Soluble carbohydrates in bean leaves', and both experiments had similar conc entrations of soluble carbohydrates . The procedure to measure starch was carried out according to the protocol found in Ebell (1969). Briefly, the 36 pellets remaining from extraction of soluble carbohydrates with ethanol were dried on a Speedvac. 2 mL of ac etate buffer was added to each sample and reacted for an hour at 100 ° C. After cooling, 0.1 mL of an amyloglucosidase solution was added to each sample, and the samples were incubated for 16 hours at 55 ° C. An aliquot of 80 µ L from each sample was taken an d deionized water added to it to make a final volume of 1 mL. Each sample was split into three replications of 0.25 mL each, and 2 mL of a glucose oxidase, peroxidase, and O - dianisidine d ihydrochloride color reagent was added to each replication (Keller an d Loescher, 1989) . After sitting for 40 minutes, the reaction was stopped by adding 2 mL of 6 M sulfuric acid. Absorbance of the samples was then read at 540 nm, and concentrations were determined from a standard curve of glucose ranging from 0 to 80 µg/mL , and the technical replicates' readings were averaged to represent the sample. Application of abscisic acid The three common bean genotype s, Jaguar, SER - 16 , and Zorro, and TB1 tepary were grown in a growth chamber set to 25 ° C, 400 µ mol photons m - 2 s - 1 PAR light intensity during the day, and 14 hour light / 10 hour dark periods. Seeds were sown into 10 L black plastic pots filled with Suremix pearlite mix potting media (Michigan Grower Products, Galesburg, Michigan, USA). After germination, seedlings we re thinned to one per pot. Six replicates of each variety were planted. Pots were fertilized with 5 g of Osmocote 15 - 9 - 12 slow - release fertilizer (Bloomington Brands, LLC, Bloomington, Indiana, USA) and supplemented once a week by watering with half - streng th Hoagland solution until the pots were saturated and dripping. When the first trifoliate leaf was fully expanded, the treatments with ABA began. Treatment solutions consisted of a 0.01% (volume/volume) solution of the surfactant Tween 20 37 mixed with enou gh of the commercial ABA product ConTego Pro SL (donated by Valent Biosciences, Libertyville, Illinois, USA) to create a solution with a known ABA concentration. On the first day of treatments, in the afternoon, the plants were sprayed to drip with the Twe en solution with no ABA. The following day, in the mid - morning, the stomatal conductance of the youngest fully mature leaf was measured using the LI - COR 6400XT portable gas exchange analyzer (LI - COR Biosciences, Lincoln, Nebraska). The conditions inside th e LI - COR 6400XT's measuring chamber were 1000 µ mol photons m - 2 s - 1 PAR, 400 µ mol mol - 1 CO 2 , a relative humidity ranging from 52 - 58%, and block temperature ranging from 26 - 30 °C . This pattern of spraying in the afternoon and measuring the following mid - morn ing continued using a 0.1 mM solution of ABA, then a 0.5 mM solution of ABA, and finally a 1 mM solution of ABA. The same group of plants were being sprayed each time, so the plants were exposed to progressively increasing concentrations of ABA. The experi ment was carried out over four days to minimize the effects of leaf age on stomatal conductance. Statistical analysis was done in SAS (SAS Institute, Cary, North Carolina) using the PROC MIXED procedure. The experiment was analyzed as a completely randomi zed design with ABA as a repeated measures variable. Determination of endogenous abscisic acid The bean lines Jaguar, SER - 16 , tepary, and Zorro were planted and grown in growth chamber conditions similar to th ose described above in the 'Application of ab scisic acid' section but with the following modifications. The bean lines were grown in a common pot setup: each 10 L black plastic pot contained one plant of each of the four bean lines. The well - watered treatment and the drought stressed treatment each had six common pot replications. When the third trifoliate leaf had fully expanded in the common bean lines, watering was stopped for the 38 drought stress ed treatment while continuing for the well - watered treatment. One day after the start of differential wa tering, in the morning, a 17 mm diameter leaf punch was taken from the latest fully expanded leaf of each plant . . Two days after the start of differential watering, the drought stressed treatment was wilting, and leaf punches were taken from all plants as before. Three days after the start of differential watering, watering of the drought stressed treatment resumed, and leaf punches were taken four and five days after differential watering, with these two days representing the recovery period for the droug h t stress treatment plants. To analyze the leaf punches, a solution consisting of 70% methanol, 30% water, 0.1% formic acid (volume / volume), and 0.1 µ M concentration of labeled ABA d6 (an internal standard containing six deuturated hydrogens) was used for extracting ABA from plant tissue. Each leaf punch taken above was placed directly into 1 mL of extraction solution and left to extract overnight at 4 ° C. The following morning, the samples were centrifuged for 5 minutes at 5000 g, the supernatant was coll ected and evaporated to dryness on a Speedvac, and the resulting pellet redissolved in 100 µ L of 50% methanol. The redissolved samples were quantitatively analyzed for ABA concentration using a protocol developed by Dr. Dan Jones using tandem ultra high pe rformance liquid chromatography and mass spectrometry with negative ion electrospray at the Mass Spectrometry and Metabolomics Core of the Research Technology Support Facility at Michigan State University. For determining endogenous levels of ABA in root s and shoots of grafted bean plants, plants were grown and treated as above but with the following differences. T he varieties Jaguar and TB1 t epary were grown as they represented the two extremes of susceptibility and tolerance to drought . They were initia lly sown in two flats of eighteen 10 cm diameter pots, with two seeds per pot. Three days after germination, plants were grafted in four combinations: J ag uar shoot 39 grafted onto a Jaguar root, J aguar shoot on tepary root, tepary shoot on Jaguar root, and te pary shoot on tepary root. Plants were grafted used a sharp razor blade to cut the stems midway between the soil and the unifoliate leaves. The shoot's stem end was pared on two sides to form a pointed wedge, and then a vertical cut was made 1 cm deep into the top of the rootstock's stem. The shoot wedge was inserted into the rootstock stem's cut so that the cut sides of both were in full contact with each other, and the entire graft junction was then wrapped in a small piece of Parafilm . A bag was placed o ver each plant to maintain a humid environment, and the plants were kept well - watered in the growth chamber for two weeks before the bags were removed and the plants transferred to 10 L common pots with one of each graft type in the same pot, six of these common pots total. After a month of normal growth, water was withheld from all plants for six days to impose drought stress, at which point the photosynthesis and stomatal conductance of all plants were near zero. It was assumed that when gas exchange was near zero, the plants were experiencing severe stress, but leaf ABA concentrations actually were closer to a period of moderate stress (Figure s 5 and 6) . Leaf samples for ABA determination were taken as above. Plants were then uprooted, the roots quickly w ashed twice in a mild detergent solution (1% w/v solution of Alconox powdered detergent (Alconox Inc., White Plains, New York, USA)) to remove soil, and the crown and tap roots frozen in liquid nitrogen and later lyophilized before determination of ABA as above. 40 Results Preliminary metabolite screen Gas chromatography - mass spectrometry identified a multitude of compounds in the leaf extracts (Supplemental Data 1). Compounds that accumulated in appreciable concentrations, as determined from their percentage of total peak area, were considered for further analysis. The compounds so identified were malic acid, glucose, fructose, inositol, sucrose, and raffinose. Proline content in stressed leaves No significant differences were found in leaf proline content between the well - watered control and the drought - stressed treatment, nor were there any significant differences in leaf proline content among the four common bean varieties tested (Figure 2 ). Pr oline levels for all variety by treatment combinations were statistically indistinguishable from each other. Soluble carbohydrates and starch in stressed bean leaves For most all soluble carbohydrates, drought stress caused a significant increase in thei r concentration in bean leaves. Malic acid increased roughly four - fold to 4 mg per g of dry tissue in plants exposed to drought stress, but no differences were found among genotypes (Figure 3A). For fructose (Figure 3B), tepary and Zorro had significantly higher concentrations than Jaguar or SER - 16 under drought stress although all genotypes had similarly low levels of fructose under well - watered cond i tions. Glucose followed a similar pattern (Figure 3C) with indistinguishable and low levels under well - wat ered conditions and a significant increase under drought stress, but compared to fructose, glucose was generally pr esent in greater concentrations 41 Figure 2 - Proline content in bean leaves The amount of proline in leaves of different bean genotypes in exp osed to drought stress . Blue bars represent the amount of proline in leaves of the well - watered control, and red bars represent proline levels in the leaves of the drought stress treated plants. Within the well - watered treatment, bars that do not share any lowercase letters are significantly different from each other, and within the drought stressed treatment, bars that do not share any uppercase letters are significantly different from each other (alpha=0.05). Error bars represent standard error. 0 100 200 300 400 500 600 Jaguar Zorro RAB 651 SER 16 µg proline / g dry leaf weight Genotype Watered Drought a a b ab ab ab a ab 42 Figure 3 - Soluble organic acids and carbohydrates in bean leaves The amount of (A) malic acid, (B) fructose, (C) glucose, (D) inositol, (E) sucrose, (F) raffinose, and (G) starch present in the leaves of four different bean lines exposed to drought str ess (red bars) or maintained in a well - watered condition (blue bars). Letters located above bars are used to indicate significant differences among lines within a stress treatment; bars that share no letters are significantly different from each other (alp ha= 0.05). Error bars represent standard error. 0 1 2 3 4 5 6 Jaguar Zorro SER Tepary mg malic acid / g dry leaf weight Malic Acid watered drought A b b b b a a a a 43 Figure 3 (cont'd) 0 5 10 15 20 25 30 Jaguar Zorro SER Tepary mg fructose / g dry leaf weight Fructose watered drought B b b c d a a a a 0 10 20 30 40 50 Jaguar Zorro SER Tepary mg glucose / g dry leaf weight Glucose watered drought C bc c b d a a a a 44 Figure 3 (cont'd) 0 4 8 12 16 Jaguar Zorro SER Tepary mg inositol/ g leaf dry weight Inositol watered drought D b b c b a a a a 0 4 8 12 16 Jaguar Zorro SER Tepary mg sucrose / g leaf dry weight Sucrose watered drought E ab bc ab ab c ab ab a 45 Figure 3 (cont'd) 0 2 4 6 Jaguar Zorro SER Tepary mg raffionose / g leaf dry weight Raffinose watered drought F b b b b a a a a 0 20 40 60 80 Jaguar Zorro SER Tepary mg starch/ g dry leaf weight Starch watered drought G a a a a b b b b 46 for all genotypes and treatments. Under drought stress, Zorro had the highest levels of glucose among the common bean lines, but tepary bean had an even higher concentration. Inositol overall had concentrations similar to fructose, and under drought stress, tepary had a higher concentration than the other genotypes (Figure 3D). Levels of sucrose were similar between the well - watered and drought stress treatments; only genotype Zorro had a significant difference between the two treatments (Figure 3E). Conversely compared to other sugars, although the drought treatment had no differences among the four genotypes, the well - w atered treatment had a narrow spectrum of differences among the genotypes, with tepary having the lowest concentration and Zorro having the greatest concentration. No raffinose was detected in any of the well - watered samples, but the drought treated genoty pes all had similar amounts of raffinose in their leaves (Figure 3F). The concentration of raffinose was similar to that of malic acid and much lower than that of glucose or fructose. Well - watered samples contained significantly greater concentrations of starch than drought treated samples. However, within a treatment, no differences in starch content could be detected among genotype s. When comparing common bean genotype s tested in this experiment, all genotype s had the same level of carbohydrates under well - watered conditions, and they also had similar increases when exposed to drought stress. The tepary bean genotype , although similar to the common bean genotype s under well - watered conditions, had a significantly larger increase in a number of carbohydrates when exposed to drought stress. 47 Exogenous abscisic acid As expected, increasing concentrations of exogenous ABA decreased the stomatal conductance of al l bean lines that were tested ( Figure 4 ). N ot all lines showed similar decreases in conductance. Overall, tepary was the least responsive to ABA, with the smallest decrease in conductance relative to the no ABA treatment. However, tepary also had the lowest stomatal conductance under the control tr eatment. SER - 16 was the most sensitive to ABA exposure, having the largest relative drop in stomatal conductance at low and medium concentrations of exogenous ABA. Bean lines Jaguar and Zorro were similarly sensitive to ABA and did not appreciably drop the ir stomatal conductance until moderate concentrations of ABA (5 mM) were applied. The greatest differences among lines were found at low to moderate concentrations of ABA (0.1 mM and 0.5 mM), where SER - 16 is especially affected by the ABA exposure while Ja guar and Zorro were not affected as much. At 1 mM ABA, all lines had similarly low stomatal conductances. Endogenous abscisic acid of leaf tissue and root tissue under drought Under well - watered conditions, no significant differences in endogenous ABA le vels were found for any of the bean lines tested, nor did mean ABA content differ on any of the days (Figure 5A). ABA concentrations were very low in all bean varieties under well - watered conditions. Under drought stress, the concentration of ABA increased by nearly two orders of magnitude (Figure 5B). Jaguar is the only line that was statistical ly different from the other lines in its concentration of ABA under drought stress: it accumulated more ABA in its leaves in response to drought stress than the oth er lines. After rewatering, ABA concentrations were reduced for all varieties. The ABA concentrations on days 4 and 5 after initiation of drought 48 Figure 4 - Exogenous abscisic acid The stomatal conductance of four bean lines after exposure to several conc entrations of exogenous abscisic acid . Bars represent standard error. Treatment means that share no letters are significantly different from each other (alpha = 0.05). 0 0.1 0.2 0.3 0.4 0.5 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 Conductance (mol H 2 O m - 2 s - 1 ) ABA Concentration (mM) Jaguar SER Tepary Zorro .0 a a a ab a ab ab c c c c c 49 Figure 5 - Endogenous abscisic acid The endogenous levels of abscisic acid in the leaf tissue of four different bean lines exposed to either (A) well - watered conditions or (B) drought stress and recovery. In (B), the drought treatment was started on day 0, and plants were rewatered and allowed to recover on days 3 - 5. Note the di fference in scale in the y - axis between (A) and (B). Bars represent standard error. Treatment means that share no letters are significantly different from each other (alpha = 0.05). 0 0.0005 0.001 0.0015 0.002 0.0025 0 1 2 3 4 5 6 ABA (nmol /mg dry weight) Days after start of water treatment ABA in well - watered plants Jaguar SER - 16 Tepary Zorro A a a a a a a a a a a 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0 1 2 3 4 5 6 ABA (nmol/mg dry weight) Days after start of water treatment ABA in drought treated plants Jaguar SER - 16 Tepary Zorro B Drought Period Recovery Period a a b b c c c d 50 Figure 6 - Abscisic acid in roots and shoots The concentration of abscisic acid in root tissues (blue bars) and shoot tissues (red bars) of different graft combinations between common bean genotype Jaguar and an improved tepary bean line. Tissues samples were taken during a period of moderate drought stress imposed by withholding water . Graft types are identified as shoot/root. Error bars represent standard error. Root concentrations not sharing a capital letter above them are significantly different from each other while shoot concentrations not shar ing a lowercase letter are significantly different from each other (alpha=0.05). 0 0.01 0.02 0.03 0.04 0.05 0.06 jag/jag jag/tep tep/jag tep/tep ABA (nmol/mg dry weight) Graft type roots shoots a a b b c c c c 51 treatment (1 and 2 days after rewatering, respectively) were statistically indistinguishable from the ABA concentrations of the well - watered controls. Bean plants h ave lower concentrations of ABA in their roots compared to the leaf tissue concentrations (Figure 6). No significant differences were found among the different graft types for the concentration of ABA in leaf tissues. However, graft types tepary/Jaguar and tepary/tepary had significantly higher root concentrations of ABA than graft types Jaguar/Jaguar and Jaguar/tepary. The leaf ABA concentration in this experiment was similar to that found one day after the imposition of drought stress in the previous expe riment (Figure 5B). This level indicates a moderate level of drought stress, and for both experiments, no varietal differences in leaf ABA concentration were detected at this level of drought stress. Discussion Proline and c arbohydrates When considering levels of free proline in bean leaves, no significant differences could be detected between well - watered bean plants and those experiencing drought stress. Proline content thus apparently plays no role in common beans during moderate droug ht stress events; because its levels do not change, proline is unlikely to mitigate the stress that beans experience nor are proline levels perturbed as a downstream effect of drought stress. These results are similar to previously reported effects of mode rate drought stress on proline levels in common beans (Rosales et al., 2012) . While some xerophytes use free proline accumulation in shoot tissues to mitigate severe drought stress (Hoekstra et al., 2001) , this same mechanism seems unlikely to play a role in common beans. 52 Unlike free proline, many of the carbohydrates studied in this experiment had a significant and consistent response to drought stress. Although proline and many carbohydrates were present in comparable quantities under well - watered condi tions, several soluble carbohydrates had dramatic increases in concentration under drought stress. Although too imprecise and expensive to use as a main analysis, the preliminary metabolite screen was especially useful for i dentifying compounds that could be analyzed in more detail in the soluble carbohydrates study. When comparing only common bean genotype s, very few differences were found amongst them for the accumulation of carbohydrates under drought stress. Zorro accumulated significantly more of the hexoses , glucose and fructose , under drought stress than at least one of the other common bean genotype s. Zorro is more stress tolerant than Jaguar (Kelly et al., 2009 a ) , so its greater concentration may be an adaptive mechanism to stress. However, this in crease in concentration is not seen in SER - 16 , which is a genotype with greater drought tolerance than Zorro (Beebe et al., 2008; Kelly et al., 2009a) . This lack of the mechanism may be explained by SER - 16 having other mechanisms of drought tolerance that have a greater impact on yield under stress, so it can lack high concentrations of hexoses under drought stress but still perform better than Zorro. Zorro and Jaguar belong to the same market class and are genetically close to each other, so it is possibl e that Zorro's higher concentration of these compounds gives it the advantage in terms of performance under drought stress. When tepary bean is compared to the common bean genotype s, it shows a much greater accumulation of the hexoses glucose and fructose as well as the cyclic sugar alcohol inositol, a direct derivative of glucose 6 phosphate. Tepary's higher concentration of these compounds in its 53 leaf tissues under drought stress could contribute to its greater drought tolerance when compared to common b ean genotype s. Using the form of the van 't Hoff equation: = - M i RT (E quation 1) where is the osmotic potential, M is molar concentration of the solute, i is the van 't Hoff dissociation factor, R is the ideal gas constant, and T is the absolute temperature, the approximate contribution of the soluble carbohydrates to osmotic adjustment in tepary under stress is - 0.1 MPa compared to approximately - 0.05 MPa for the common bean genotype s. This difference in osmotic adjustment is due to tepary's higher concentrations of glucose, fructose, and inositol. The accumulation of these compounds and the ensuing decrease in leaf water potential could allow tepary to del ay wilting and continuing growing under water - limited conditions. Relevant to all genotype s in this experiment, i nositol conjugates with UDP - D - galactose to form a precursor of raffinose biosynthesis (Loewus and Murthy, 2000) , so the increase seen in inosi tol under drought stress in all genotype s may be related to their higher concentrations of raffinose under drought stress. Raffinose was undetectable in bean leaf tissues under well - watered conditions, but drought stress greatly increased the raffinose con tent in all genotype s. Raffinose accumulation in response to abiotic stress is seen in a number of species (Barchet et al., 2014; Richter et al., 2015; Wenzel et al., 2015) , and it possibly acts as an osmoprotectant to protect against both osmotic and oxid ative stress (Nishizawa et al., 2008) . Moderate drought stress in bean plants led to a decrease in the concentration of starch in leaf tissue. A decrease in starch in response to drought stress is seen in beans and in other plant 54 species and most likely r esults from an inhibition of synthesis (Vassey and Sharkey, 1989; Geigenberger et al., 1997; Escobar - Gutierrez et al., 1998) . Well - watered Zorro leaves had higher concentration s of s ucrose than the leaves of other genotype and treatment combinations , but n o other physiologically significant differences in sucrose concentration were observed among treatment means . Sucrose's stability is remarkable not only because a decrease in starch synthesis could shunt glyceraldehyde - 3 - phosphate towards sucrose synthesis but also because the increase in fructose and glucose concentrations could result from the breakdown of sucrose by invertase or sucrose synthase. Under drought stress, the sucrose pool could be a sensitive regulator of reduced photosynthate from stress, r educed starch synthesis, and increased fructose and glucose pools. Indeed, sucrose regulates the expression of a number of genes independent of fructose and glucose (Rolland et al., 2006) , so any increase or decrease in the size of the sucrose pool could t rigger negative - feedback loops that return sucrose to normal values. Malic acid was seen in higher concentrations in the leaves of beans exposed to drought stress. M alic acid pools increased in response to hypoxia in moss and heavy metal exposure in Silen e cucubalus (Bailey et al., 2003; Rut et al., 2010) and decreased in response to chemically induced oxidative stress and a combination of heat and drought stress in Arabidopsis (Koussevitzky et al., 2008; Obata et al., 2011) . Interestingly, NADP - malic enzyme increases in activity under drought or osmotic stress in many C 3 plants (Liu et al., 2007; Doubnerová and . NADP - malic enzyme converts malate and NADP + to pyruvate, CO 2 , and NADPH. The increase in malic acid, especially if it took pl ace in the chloroplasts of leaf cells, could play a small role in pH related stress signaling (Edwards et al., 1998) . Indeed, one paralog of NADP - malic enzyme in Arabidopsis functions in the chloroplast (Wheeler et al., 2005) . NADP - malic enzyme increases i n content and activity in response to ultraviolet B stress (Casati 55 et al., 1999) , and in a study of three bean genotype s, NADP - malic enzyme content was positively correlated with tolerance to ultraviolet B stress (Pinto et al., 1999) . In tests of several b arley genotype s, only the drought tolerant genotype s upregulated the expression of NADP - malic enzyme in response to drought stress (Guo et al., 2009) . However, NADP - malic enzyme breaks malic acid down. What about its synthesis? In Arabidopsis, drought stre ss significantly increases the expression of fumarase, the enzyme responsible for the production of malic acid from fumarate, malic acid's direct precursor, but has little to no effect on the expression of other genes in the tricarboxylic acid cycle (Kilia n et al., 2007; Winter et al., 2007) . Additionally, in maize roots exposed to stress, the flux of carbon to malic acid via PEP carboxylase is an order of magnitude greater than the flow of carbon out of malic acid via NADP malic enzyme (Edwards et al., 199 8) . The increase in malic acid concentration seen in beans under drought stress could allow them to regulate cellular pH and produce reducing power and CO 2 via the NADP - malic enzyme pathway, with the possibility that the CO 2 released could be refixed in th e chloroplast. Further, malic acid could act as a counter - ion to potassium and thus aid in the regulation of stomatal closure. Abscisic acid Different bean genotype s have different stomatal sensitivities to ABA ( Figure 4 ). This sensitivity is not correla ted with a genotype 's drought tolerance. While the drought tolerant tepary bean is comparatively insensitive to ABA, the drought tolerant common bean genotype SER - 16 is more sensitive than any other genotype tested. These differential sensitivities could i ndicate a divergence in the drought stress strategies of these two bean genotypes; despite both being drought tolerant, tepary bean and SER - 16 belong to different species, so the genetic 56 distance between them could have allowed different stress adaptation strategies to evolve in each one. SER - 16 has a highly responsive strategy that regulates metabolism at early or smaller stress signals while tepary has a more conservative strategy that keeps primary productivity low under all conditions with less reliance on sensitivity to stress signals for survival. Differences in sensitivity to ABA among Phaseolus genotypes is a new finding, and future studies should include additional genotypes not used in this study to test if the patterns of sensitivity found in this experiment hold true for the wider bean germplasm. Leaf ABA levels rose sharply in bean plants as drought intensifie d , but they remain ed low and stable under well - watered conditions ( Figure 5 ). Varietal differences in ABA concentration only appeared under severe stress; drought susceptible common bean Jaguar accumulated more ABA in its leaves than any other genotype . While ABA is necessary to activate many stress protective responses, the hype raccumulation of ABA in Jaguar could indicate a disruption of signaling that is coincident with a harmful, unregulated stress response (Seiler et al., 2011; Sreenivasulu et al., 2012) . Tepary bean, the most drought tolerant of the lines tested, had the low est concentrations of ABA in its leaves during the most severe stress period, and its low concentration could indicate a well regulated ABA signal that is an advantage to survival and reproduction under stress conditions. Thus, as the bean plants transitio n from moderate to severe drought stress, the ABA levels of tolerant and susceptible genotype s diverge as the former maintain ABA homeostasis while the latter has its drought signaling disrupted (Sreenivasulu et al., 2012) . When comparing the bea n lines' sensitivities to ABA (F igure 4) and endogenous levels of ABA under stress ( Figure 5 B ), we gain a greater insight into their stress response than from either of those facets alone. The extremely drought tolerant tepary is neither as sensitive to ABA 57 nor doe s it produce as much under se vere stress, suggesting a strong constitutive mechanism of protection less reliant on stress signaling. The drought tolerant SER - 16 has both high levels of ABA under stress and is very sensitive to ABA exposure, suggesting a mo re reactive stress response strategy. Jaguar, a susceptible genotype , and Zorro, a moderate ly tolerant genotype , both have the same sensitivities to ABA, but Jaguar produces more ABA under severe drought stress. Zorro's ability to maintain ABA homeostasis under severe stress could be why it has an advantage over Jaguar under stress conditions despite sharing its sensitivity to ABA. As a whole, on a dry plant tissue basis, leaves produce more ABA than roots in drought stressed bean plants ( Figure 6 ). These results correlate with previous studies that find shoot tissue the site of ABA production (Holbrook et al., 2002; Christmann et al., 2007) . For shoot ABA concentrations, no differences were found among any of the graft types, and these results matched prev ious results about shoot ABA levels under moderate stress ( Figure 5 ). However, root ABA concentration significantly differed among graft types: tepary/Jaguar and tepary/tepary had higher ABA concentrations in their roots than Jaguar/Jaguar and Jaguar/tepar y. Given that the graft types with higher root ABA shared their shoot identity but had different root identities, shoot identity appears to determine ABA concentrations in root tissues. Interestingly, the drought tolerant tepary bean had a higher concentr ation of root ABA than the drought susceptible common bean genotype Jaguar, but further bean genotype s' root ABA levels will need testing to determine if high root ABA correlates with drought tolerance in field settings. Mechanically, this correlation is p lausible because ABA can influence root hydraulic conductivity (Thompson et al., 2007; Kudoyarova et al., 2011) and change the characteristics of water supply from the roots. In Figure 5 B , two days after the imposition of s tress and when drought stress w as severe, ABA concentrations in leaves reach ed their highest point as well as the point at which 58 appreciable differences in ABA among cultivars appear ed . The ABA levels in the shoot tissue in Figure 6 were very close to the ABA l evels of leaves one day af ter the imposition of stress, and drought stress was only moderate at that point. Using these ABA concentrations as a referenc e, as drought stress intensified , genotypic differences in root ABA appear ed sooner than genotypic differences in shoot ABA. Tepar y sent a greater initial stress sign al to its roots than Jaguar did , and the metab olic changes this signal induced could have better prepare d tepary for upcoming stress and prevent ed a larger stress sign al in the shoot when stress did arrive. Conclusions Free proline plays little to no role in stress protection or response in common and tepary beans, but the concentration s of certain carbohydrates and organic acids do. Tepary's higher concentration of some of these carbohydrates likely contributes to its greater drought tolerance. Future studies should look at the subcellular localization of some of the compounds studied, like malic acid, glucose, fructose, and raffinose, as well as the activity of the NADP - malic enzyme, in tepary and common bean to inves tigate any differences and further define the function of these compounds in stress response. Analyzing the concentrations of glucose and fructose in drought stressed leaf tissue could aid in screening for drought tolerant genotypes, but tolerant genotypes lacking this mechanism would be undetected, and the time and specialized equipment required would limit its application. Both within and among species in Phaseolus , variation exists for sensitivity to ABA and concentrations of ABA produced under stress, b ut neither measurement alone correlates with a variety's drought tolerance. However, combining information about sensitivity to ABA and endogenous levels of ABA allows for a more informative picture of a 59 genotype's stress response. Shoots are responsible f or ABA production and its concentration in the roots of bean plants. Additional studies on root ABA concentration in stressed bean plants in a wider range of germplasm would help determine if correlations exist between root ABA and drought tolerance. 60 Chapter 3: Drought stress and photosynthesis 61 Introduction Common beans ( Phaseolus vulgaris L.) are a staple source of protein and nutrients in many parts of the world, especially Central America and East Africa (Cavalieri et al., 2011) . Drought stress limits the yield of common beans in roughly 60% of the regions in which it is produced in any given year (McClean et al., 2011) . Breeding bean cultivars with improved drought tolerance is thus crucial to the stability of the global bean crop (Beebe et al., 2012) , which in turn supports the food security and economic well - being of the farmers producing beans. To assist breeders in developing more tolerant cultivars, the physiology of drought tolerance in common beans should be investigated further (Beebe et al., 2013) . While research in the drought tolerance mechanisms of other crops (Cattivelli et al., 2008; Passioura, 2012) helps inform efforts in beans, ultimately the presence and range of these mechanisms in common bean germ plasm must be investigated before they can be used in a breeding program. Furthermore, there is a need for drought studies that integrate several different types of metabolic and physiological measurements (Pinheiro and Chaves, 2011) . The present study inv estigates several physiological traits and their response to drought stress in a small group of bean genotypes contrasting for drought tolerance. Water stress impacts a plant's productivity by decreasing its rate of photosynthesis. Considerable debate sti ll exists about the relation between photosynthesis and agronomic yield (Long et al., 2006) , but measurements of photosynthesis still offer rough insight into the integrated metabolism of a plant and how water stress affects that (Chaves, 1991) . The sensit ivity of reproductive tissues to drought stress creates a disconnect between photosynthesis and agronomic yield. Drought stress causes the abortion of developing reproductive tissue. Tolerant 62 genotypes in an agronomic setting have lower fruit abortion rate s, often by remobilizing photosynthate reserves from other plant parts (Araus et al., 2002; Yang and Zhang, 2006; Tolk et al., 2013) . While much of the previous research was done on cereals, tolerant common bean genotypes also have lower pod abortion rates (Boutraa and Sanders, 2001; Lizana et al., 2006) . Additionally, comparisons of photosynthesis within a species avoid the variance attendant with broader studies and allow meaningful correlations to be revealed (Fischer et al., 1998) . Closely linked to pho tosynthesis is stomatal conductance, how open a leaf's stomata are to the influx of carbon dioxide and the efflux of water vapor. Plants must balance their carbon and water resources. Under drought stress, plants close their stomata to conserve water, but this also reduces the supply of carbon dioxide available for photosynthesis. Closure of stomata is the main limiting mechanism of photosynthesis under drought stress (Flexas et al., 2004) . The balance a plant strikes between carbon gain and water loss dete rmines the degree of its drought tolerance (Pinheiro et al., 2005) . Tepary beans ( Phaseolus acutifolius A. Gray) are a drought tolerant species native to the Sonoran desert and used as an arid land crop within Northwest Mexico and the American Southwest f or hundreds of years (Nabhan and Felger, 1978) . Tepary is closely related to common beans, and with embryo rescue, tepary and common bean can be hybridized with each other (Mejía - Jiménez et al., 1994; Araújo et al., 2014) . A handful of studies compared rel ative responses of common beans and tepary beans subjected to heat and drought stress (Markhart, 1985; Castonguay and Markhart, 1991, 1992; Sanders and Markhart, 1992; Udomprasert et al., 1995) , but research into tepary waned for some years until tepary's potential as a genetic resource for improving common beans was more fully utilized (Butare et al., 2011; CGIAR, 2015) . In 63 addition to its useful genetics, tepary's physiological traits are a model for which traits to alter in common beans to improve their abiotic stress tolerance (Rao et al., 2013) . To investigate drought response, the current research studied physiological traits within common bean and tepary beans and how these two species compared with each other. Drought's effects on photosynthesis in beans were a particular focus and were measured using gas exchange, chlorophyll fluorescence, and grafting. Stress and its relation to morphology was also investigated by looking at leaf traits and the abscission of reproductive tissues. Materials and m e thods Rates of pod abscission under drought stress The common bean varieties Fuji and Zorro were sown in 1 L square, black plastic pots and grown in a growth chamber set to a constant 22 ° C, 12 h light and 12 h dark cycle, and a light intensity of 200 µ mol photons m - 2 s - 1 . Plants were grown in Suremix Pearlite potting media (Michigan Grower Products, Galesburg, MI, USA) with 32 replications per variety. Fuji is a cultivar within the specialty Otebo bean market class that has substantially lower yields un der drought stress (Kelly et al., 2009 b ) . Zorro is an elite cultivar within the black bean market class that is moderately tolerant to drought stress and generally yields well (Kelly et al., 2009 a ) . After sowing, the plants were regularly watered with a ha lf - strength Hoagland nutrient solution. Once the unifoliate leaves were fully expanded, all plants were placed on a limited watering regime: every two days, each pot was only given enough water to bring it up to 30% of pot's total water capacity, as determ ined by weighing on a scale. A month after sowing, the plants began to flower. Two weeks after flowering, the number of pods and trifoliate leaves on each plant was 64 Table 1 - Bean genotypes used in experiments this chapter The table below contains brief d escriptions of the bean genotypes used by the experiments that are discussed in this chapter. Name Description Level of drought tolerance Big Fields Brown Pa, landrace Very tolerant Black Tepary Pa, landrace Very tolerant Fuji Pv, otebo bean Susceptible Jaguar Pv, black bean Susceptible RAB - 651 Pv, small red bean bred by CIAT Tolerant Sacaton White Pa, landrace Very tolerant San Ignacio Pa, landrace Very tolerant SER - 16 Pv, small red bean bred by CIAT Tolerant SER - 95 Pv, small red bean bred by CIAT Tolerant TB1 tepary Pa, improved tepary line bred by Dr. Tim Porch Very tolerant Tohono O'dham Pa, landrace Very tolerant Tucson Brown Pa, landrace Very tolerant Wild tepary - Chihuahua Pa, wild accession Very tolerant Wild tepary - Tibur ó n Island Pa, wild accession Very tolerant Zorro Pv, elite black bean Moderately tolerant Legend: Pv - Phaseolus vulgaris , Pa - Phaseolus acutifolius 65 counted, and two weeks after that first counting, the number of pods and leaves on each plant was counted again. Pods of any size or developmental stage were included in the count, but only fully expanded trifoliate leaves were included in the count. Varietal and time means were compared using PROC MIXED in SAS version 9.4 (SAS Institute, Cary, NC , USA). Chlorophyll fluorescence of young stressed bean leaves The plants were grown as described in the ' Rates of pod abscission under drought stress ' section above with the following modifications. Four common bean genotype s were used: Jaguar, RAB - 651, SER - 16, and Zorro. Jaguar is an older black bean variety whose susceptibility to biotic and abiotic stresses led to its replacement by Zorro. SER - 16 and RAB - 651 are two small red bean line s bred in Colo mbia by CIAT as part of their development of abiotic stress tolerant lines, the latter bred for tolerance to low phosphorous, and both have shown tolerance to stress in field settings (Beebe et al., 2008, 2013) . Plants were split into well watered and drought stressed treatments with six replications of each genotype per treatment. The experiment was arranged as a randomized complete block design. After the first trifoliate leaves fully matured, the watered treatment continued to receive daily watering with deionized water while the drought treatment received no additional water. Three days after the start of differential watering, measurements of maximal photosynthetic efficiency (Fv/Fm) were made on the leaves that were dark - adapted over night . Measurements were made at the end of the dark period in the da rk (a flashlight was used to provide illumination for the work) to avoid disrupting dark adaptation. The measurements were made with the fluorescence head of the LI - COR 6400XT (LI - COR Biosciences, Lincoln, NE, USA). After the Fv/Fm measurements, the chambe r lights turned on and the plants adapted to the 66 light for two hours. Then, the LI - COR's chlorophyll fluorescence head was again used to measure photosystem II efficiency ( PSII ). All plants were then rewatered, and after one week, the aboveground tissues of the plants were harvested, dried in a 60 ° C oven, and the final dry weight of each plant measured. Data were statistically analyzed and treatment means compared using PROC MIXED in SAS 9.4. An alpha of 0.05 was used as the threshold for significance. Wh en a treatment or interaction was not significant according to the analysis of variance (ANOVA), the individual means within that treatment or interaction were compared using the much more conservative Tukey's adjustment. L eaf water potential of stressed plants This experiment was grown under the same conditions as described above in the 'Rates of pod abscission under drought stress' section but with the following modifications. Four genotype s were planted: Jaguar, SER - 16, Zorro, and a drought tolerant te pary bean line TB1 provided by Dr. Timothy Porch from USDA - ARS in Mayaguez, Puerto Rico. Plants were grown in a growth chamber with the temperature set to 25 ° C during the light period and 20 ° C during the dark period and a 14 h light and 10 h dark cycle. Light intensity (PAR) was 400 µ mol photons m - 2 s - 1 . Plants were grown in 10 L black plastic pots in a common pot setup: one plant of each genotype per pot. Twelve pots arranged in a randomized complete block design were grown in total. Four weeks af ter germination, the third trifoliate leaves of the common beans had fully expanded, and the pots were watered to pot capacity. Then, water was withheld for three days, and the plants appeared drought stressed by the end of this period. In the midmorning o f the third day after water ing stopped, the latest fully expanded trifoliate of each plant was cut off at the base of the petiole, the leaves sealed in plastic bags, and the bags kept in a cool, dark container 67 until their analysis. A pressure bomb (PMS Ins truments, Albany, OR, USA) was used to measure the leaf water potential of each leaf. The leaves were sealed in the pressure bomb with the cut petiole sticking out, and nitrogen from a compressed gas tank was used to increase pressure in the chamber until water returned to the cut surface of the petiole, at which point the pressure reading was recorded as the leaf's water potential. A - C i curves in well - watered conditions This experiment was performed as the leaf water potential experiment described above but with the following differences. The four genotype s planted were Jaguar, Tepary, Zorro, and SER - 95, a small red variety bred for drought tolerance by CIAT. The plants were grown individually in 5 L plastic pots with four to six replications per variety. The temperature regime was 28 ° C during the light period and 20 ° C during the dark period . Four weeks after sowing, the third trifoliate leaves were fully expanded, and measurements of the photosynthetic rate at different intercellular CO 2 concentrations ( A - C i ) commenced. The plants were kept well - watered for the duration of the experiment. For three consecutive days, A - C i measurements were made from mid morning to mid afternoon, measuring two replications per day using the autoprogram feature of the LI - CO R 6400XT. Six replications of each genotype were measured in total. In addition to the limited time period during which measurements took place, time of day effects were also controlled by randomizing within a replication the order in which genotypes were measured. An entire block of plants, consisting of one of each genotype, was measured before moving on to the next block. The reference CO 2 concentration started at 1500 µ mol mol - 1 and decreased to 1250, 1000, 800, 600, 500, 400, 300, 200, 150, 100, 50, and finally 0. Measurements were automatically taken 68 when the stability criteria involving the rate of change of photosynthesis and stomatal conductance were met; to measure the A - C i curve of one plant took approximately 25 minutes. The entire measurement period took place under a saturating light intensity of 1000 µ mol photons m - 2 s - 1 . Afterwards, the measured data were analyzed using a spreadsheet to determine the maximum carboxylat ion efficiency of rubisco (V cmax ), electron transport rate (J), triose phosphate use (TPU), and mesophyll conductance (g m ) (Sharkey et al., 2007) . The day respiration rate (R d ) was fixed at 0.66 µ mol m - 2 s - 1 (at 25 ° C), and calculations were done with a * of 3.97 Pa (at 25 ° C); the R d and * values come from Sean Weise's experimental determination of these parameters in soybean (unpublished data). The values of V cmax , J, TPU, and g m were analyzed in SAS 9.4 using the PROC MIXED statement as a randomized co mplete block design experiment with replication as the blocking factor. For ANOVAs that were significant, means were separated using the least significant difference and an alpha of 0.05. Gas exchange measurements over increasing drought stress These exp eriments used the same bean genotypes, common pot setup, and environmental conditions described in the 'Leaf water potential of stressed plants' section above but with the following modifications. The pots in which plants were growing were split into two t reatments: a well - watered control and a drought stress treatment, each with six replications. All treatments were kept well - watered until the third trifoliate leaf was fully expanded. Then, the LI - COR 6400XT was used to take gas exchange measurements on th e latest fully expanded trifoliate leaf while the plants of both treatments were still well watered. Gas exchange measurements were taken under a light intensity of 1000 µ mol photons m - 2 s - 1 , a CO 2 concentration of 400 µ mol mol - 69 1 , uncontrolled re lative humidity ranging from 54 - 59 %, and a block temperature that ranged from 27 - 31 ° C. Watering then stopped on the plants in the drought stress treatment while the well - watered treatment continued to receive water daily. Gas exchange measurements were taken dai ly for the next five days as stress in the drought treatment increased. On the sixth day, all treatments were rewatered to pot capacity, and the drought treatment was again allowed to dry down from days seven through ten. At midmorning on day ten, the late st fully expanded leaf was cut from each plant, and the leaf water potential of these leaves were measured according to the procedure described above in the leaf water potential experiment. This experiment was replicated, and after watering was stopped on the drought treatment, the weight of each pot was measured using a high capacity electronic scale before gas exchange measurements were taken. Separately, the weight of individual empty pots, pots filled with dry soil mix, and pots filled with fully satur ated soil mix were weighed. From these measurements, the maximum water holding capacity of the 10 L pots was determined as well as the percentage of maximum capacity each pot was at during the stress period. Survey of tepary varieties Eight different tepary landrace and wild varieties, kindly donated by Native Seeds (Native Seeds/SEARCH, Tucson, AZ, USA) for experimental purposes: Black Tepary, San Ignacio, Tucson Brown, Tohono O'dham, Big Fields Brown, Sacaton White, wild tepary collected from the Chi huahuan Desert, and wild tepary collected from Tibur ó n Island, Mexico. The common bean Zorro was included as a check. The improved TB1 tepary line was not included in this experiment because of space limitations, but it was grown with Zorro in the experime nt 'Gas exchange measurements over increasing drought stress' detailed above, so rough 70 comparisons via comparative performance to Zorro should be possible. Previous work revealed that germination was a problem, so all seeds were scarified with a file, plac ed on moist filter paper in petri dishes, and kept in a 30 ° C environment . After three days, the majority of seeds had germinated, and they were transplanted to 10 L pots. The seeds were grown according to a randomized incomplete block design; four differe nt genotypes were planted in each pot, and each of the nine genotypes used were replicated three times for a total of seven pots. The plants were then grown under the same conditions as described in the above section 'Leaf water potential of stressed plant s' . When the third trifoliate leaf was fully expanded, water was withheld from all pots. Gas exchange measurements were taken one day and three days after the cessation of watering, representing mild and severe drought stress, respectively. Five days later , watering was reintroduced, and three days after watering was reintroduced, gas exchange measurements were taken again. Gas exchange measurements were taken according to the procedure described above in the gas exchange measurements over increasing drough t stress experiment. The experiment was analyzed in SAS as a randomized incomplete block design using the PROC MIXED procedure. As the ANOVA from the PROC MIXED procedure was significant for both genotype and stress level, Fischer's Least Significant Dif ference (LSD) was used to compare means. Gas exchange of grafted beans In order to examine the relative contributions of root tissue and shoot tissue to drought response, grafting experiments were performed between two contrasting genotypes: drought susc eptible Jaguar and drought tolerant tepary. A flat each of Jaguar seeds and tepary seeds were planted and germinated in a growth chamber. One day after full germination, the plants 71 underwent grafting. The grafting procedure involved using a razor blade to cut the seedling in two at the stem 2 cm below the shriveled but still attached cotyledons. This was done for a second plant as well, yielding two shoots and two rootstocks. One shoot's stem end was pared on two sides to form a pointed wedge, and then a ve rtical cut was placed 1 cm deep into the top of the second rootstock's stem. The shoot wedge was inserted into the rootstock stem's cut so that the cut sides of both were in full contact with each other, and the entire graft junction was then wrapped in a small piece of Parafilm . The same was done to the remaining shoot and rootstock. A bag was placed over each plant to maintain a humid environment, and the plants were kept well - watered in the growth chamber. Grafts were performed in four combinations: Jagu ar shoot grafted onto Jaguar rootstock, Jaguar shoot grafted onto tepary root, tepary shoot on Jaguar root, and tepary shoot on tepary root. Ten days after grafting, the bags were removed, and plants were transferred to 10 L common pots with one of each gr aft type in the same pot, eight of these common pots total. The plants were then grown normally under the environmental conditions described in the leaf water potential experiment described above. After one month of recovery and growth, all plants were w atered to pot capacity, and then all watering ceased. Gas exchange measurements were taken daily in midmorning for the five days after the cessation of watering as the pots gradually dried down. Gas exchange measurements were taken according to the procedu re described above in the 'G as exchange measurements over increasing drought stress ' section . Intrinsic water use efficiencies were calculated by dividing each measurement's photosynthetic rate by its stomatal conductance. Gas exchange data was analyzed a s a repeated measures experiment in the statistical program SAS using the PROC MIXED procedure and the "repeated" statement. The resulting 72 ANOVAs found both graft type and genotype significantly affected the dependent variables tested, so individual means were compared using LSD. Leaf density and stomatal density Jaguar and tepary were examined for density of leaf tissue and the density of stomata on the abaxial surface of the leaves. These two genotypes were chosen because they are genetically distinct a nd differ in drought tolerance. Fifteen replications of each genotype were planted, and plants were grown individually in 5 L plastic pots. Otherwise, the plants were grown in the same environmental condit ions as described above in the 'L eaf water potentia l of stressed plants' section . After the third trifoliate leaf was fully expanded, the plants were all exposed to moderate drought stress by withholding watering for four days and letting the pots dry down before continuing watering. The plants were grown normally for two more weeks to allow leaves that were expanding during the drought period to mature, and then leaflets from these leaves were excised to obtain a whole leaf blade with no petiole or petiolule. Immediately after a leaflet was detached, it w as weighed on an electronic balance and was photographed. The resulting photograph was analyzed using the software Easy Leaf Area (Easlon and Bloom, 2014) to obtain the surface area of the leaflet, and density was calculated from weight and area. Other de tached leaflets were used to determine stomatal density with four replication s per genotype. A thin layer of clear nail polish was applied to the abaxial surface of a leaf. After the nail polish dried, forceps were used to peel off and remove a small patch of the dried polish, and this thin film of dried polish was examined under a microscope, revealing a relief of the abaxial 73 leaf surface with visible stomatal pores. A photograph of a portion of this nail polish film was taken at 200x magnification, and th e number of stomata in each image was counted. As comparisons were only being made between two means, control of entire experiment error was not needed, and statistical comparisons were made between means using Student's t - test with an alpha of 0.05. Results Rates of pod abscission under drought stress For the first count of reproductive pods, which took place two weeks after flowering, Fuji had approximately three times the number of pods per plant than Zorro ( Figure 7 ). In the second count of pods, which took place four weeks after flowering, the number of pods per plant for both genotype s was statistically the same. Fuji's number of pods decreased sharply between the first and second counts as the majority of pods abscised from their plants. Zorro also had a significant decrease in the number of pods per plant from the first count to the second count, but the magnitude of Zorro's decrease was smaller than Fuji's decrease. The number of trifoliate leaves was unvarying between varieties and between th e two count periods. Chlorophyll fluorescence of young stressed bean leaves The parameter Fv/Fm did not vary by either genotype or water treatment ( Figure 8 A). The moderate drought stress to which these plants were exposed (the leaves of the drought stre ss treatment wilted the day after measurements) had no impact on Fv/Fm. PSII was more responsive to drought stress and had more variation among genotype s. Drought stress significantly lowered the PSII of all genotype s ( Figure 8 B). For the well - watered tr eatment, no 74 Figure 7 - Bean pod abscission A graph of the average number of bean pods per plant for two different bean genotype s: the drought susceptible Fuji and the moderately drought tolerant Zorro. The plants were grown under consistent drought stress . Pod counts were taken two weeks after flowering and four weeks after flowering. Error bars represent standard error. Means that do not share letters are significantly different from each other (alpha= 0.05) . 0 2 4 6 8 10 12 Fuji Zorro Number of pods Genotype 2 weeks after flowering 4 weeks after flowering a a b c 75 Figure 8 - Drought and chlorophyll fluorescence (A) The average maximum photosystem II efficiency of dark adapted leaves (Fv/Fm) and (B) the average photosystem II efficiency of light exposed leaves ( PSII ) of four different common bean genotype s exposed to either well - watered or drought stress conditions. (C) The final dry weight of the aboveground biomass of the four bean genotype s after one week of exposure to either well - watered or drought stress conditions. Error bars represent standard erro r. Means that do not share letters are significantly different from each other (alpha= 0.05). 0 0.2 0.4 0.6 0.8 Jaguar Zorro RAB 651 SER 16 Fv/Fm Genotype Fv/Fm watered drought A a a a a a a a a 76 Figure 8 (cont'd) 0 0.2 0.4 0.6 0.8 Jaguar Zorro RAB 651 SER 16 Photosystem II efficiency Genotype PSII watered drought B a ab b b c c c c 0 1 2 3 4 Jaguar Zorro RAB 651 SER 16 Dry Weight aboveground biomass (g) Genotype Final Dry Weight watered drought C a a a a b b b b 77 differences were found among any of the genotypes. Under drought stress, RAB - 651 had a significantly lower PSII than Jaguar and Zorro. Drought stress also significantly reduced the final biomass of all genotypes tested (Figure 8C). Within either water treatment, no significant differences were found among any of the genotypes. Leaf water potential of stressed plan ts Leaf water potential of drought stressed bean plants varied significantly by genotype ( Figure 9 ). Because the plants were grown in common pots, all the genotype s in a replication experienced the same soil moisture conditions throughout the experiment a nd when the leaf water potential measurements were taken. At similar levels of water availability, the drought tolerant tepary had an average leaf water potential of - 1.34 MPa, the lowest of all the genotype s tested. While higher than tepary, the drought t olerant SER - 16 still had a lower leaf water potential than either Jaguar or Zorro. Drought susceptible Jaguar and moderately tolerant Zorro had leaf water potentials that were essentially identical and were the highest of all genotype s tested. A - C i curv es Representative A - C i curves for each of the four genotypes measured are shown in Figure 10 . From these curves, the averages of the photosynthetic parameters V cmax , J, TPU, and g m were determined for each genotype ( Table 2 ). No significant differences in either TPU or g m were found among the genotypes tested. However, Zorro had a significantly higher V cmax than the other genotypes tested. Furthermore, both Tepary and Zorro had a higher J than Jaguar and SER - 95. 78 Figure 9 - Leaf water potential The leaf wat er potential of four bean genotype s under moderate to severe drought stress. Error bars represent standard error. Columns with different letters are significantly different from each other (alpha=0.05). - 1.6 - 1.4 - 1.2 - 1 - 0.8 - 0.6 - 0.4 - 0.2 0 Jaguar Zorro SER Tepary Leaf Water Potential (MPa) Genotype a a b c 79 Figure 10 - A - C i curves Representative photosynthesis versus intercellular CO 2 concentration (A - C i ) curves of four different bean lines. 0 5 10 15 20 25 30 35 40 0 200 400 600 800 1000 1200 1400 Photosynthesis (umol m - 2 s - 1 ) Intercelluar [CO 2 ] Zorro Jaguar SER 95 Tepary 80 Table 2 - Photosynthetic parameters Averages of the photosynthetic parameters of maximum carboxylation rate of rubisco (V cmax ), photosynthetic electron transport rate (J), triose phosphate use (TPU), and mesophyll conductance (g m ) as derived from the A - C i curves of four different bean lines. Means within a row that share no letters are significantly different from each other (alpha = 0.05). Bean line Jaguar SER - 95 Tepary Zorro Photosynthetic parameters V cmax 98 a 82 a 101 a 124 b J 129 a 129 a 139 b 145 b TPU 8.9 a 9.1 a 9.6 a 9.7 a g m 1.89 a 2.12 a 2.21 a 1.64 a Legend: V cmax - maximum carboxylation rate of rubisco ( µ mol m - 2 s - 1 ), J - photosynthetic electron transport rate ( µ mol m - 2 s - 1 ), TPU - triose phosphate utilization ( µ mol m - 2 s - 1 ), g m - mesophyll conductance ( µ mol m - 2 s - 1 Pa - 1 ) 81 Gas exchange measurements over increasing drought stress Both genotype and water treatment affected rates of photosynthesis ( Figure s 11 A and 12A ). The well - watered treatment had stab le photosynthetic rates over the course of the measurement period with no appreciable day to day differences. Within the well - watered treatment, Jaguar and Zorro had the highest rates of photosynthesis, SER - 16 was intermediate, and tepary had the lowest ph otosynthesis under these conditions. Under drought stress conditions, the photosynthesis of all genotypes decreased as the pots dried from evaporation and transpiration. Generally, Jaguar and Zorro had the highest photosynthetic rates under drought stress, SER - 16 was intermediate ( Figure 11 A) or equal to tepary ( Figure 12A ), and tepary had the lowest photosynthesis; the relative performance of genotypes was similar to that under well - watered conditions. However, under the most extreme stress, all genotypes had similarly low photosynthetic rates. Jaguar and Zorro also recovered their photosynthesis to a greater extent than SER - 16 or tepary after the pots were rewatered ( Figure 11 A). Photosynthesis was unaffected at pot water contents of 70% of maximum capacit y ( Figure 12A ), but by a pot capacity of 30%, photosynthesis was moderately impacted, and by 13% it was severely impacted. During the course of the experiment, the well - watered treatment's average pot capacity stayed within the range of 75 - 85%. Stomatal c onductance had the same general trends as photosynthesis , and as stress increased in the drought treatment, conductance decreased as well ( Figure s 11 B and 12B ). Again, Jaguar and Zorro had the highest stomatal conductance, followed by SER - 16, and finally t epary with the lowest conductance. 82 Figure 11 - Gas exchange over days The measurement of gas exchange parameters on four bean genotypes subjected to two treatments: a well - watered control (dotted line with closed circles) and a progressively increasing drought stress treatment (solid line with closed triangles). A) The photosynthetic rate of these genotypes over the course of nine days. The drought stress treatment was rewatered once on day six to examine rates of recovery. B) The stomatal conductance o f the same plants from (A ) C ) The leaf water potential of the latest fully expanded trifoliate from the well - watered and drought stress treatments at day 9 from the plants measured in (A - B). Means that share no assigned letters are significantly different from each other (alpha=0.05). All error bars represent standard error. 0 5 10 15 20 25 30 0 1 2 3 4 5 6 7 8 9 10 Photosynthesis ( mol CO 2 m - 2 s - 1 ) Day Photosynthesis Jaguar SER 16 Tepary Zorro Jaguar SER16 Tepary Zorro A Rewatering a a a b b b bc bc bc bc bcd cd cd cd bcd de de de e e e e e e de 83 Figure 11 (cont'd) 0 0.4 0.8 1.2 1.6 0 1 2 3 4 5 6 7 8 9 10 Stomatal Conductance (mol H 2 O m - 2 s - 1 ) Day Conductance Jaguar SER 16 Tepary Zorro Jaguar SER16 Tepary Zorro B Rewatering a a ab ab ab bc bc bc bc bc bc c cd cd bc cd d de e e de cd cd d - 1.4 - 1.2 - 1 - 0.8 - 0.6 - 0.4 - 0.2 0 Jaguar Zorro SER Tepary LWP (MPa) Leaf Water Potential well - watered drought C a b c c a a a b 84 Figure 12 - Gas exchange with pot weights The measurement of gas exchange parameters on four bean genotypes subjected to two treatments: a well - watered control (dotted line with closed circles) and a progressively increasing drought stress treatment (solid line with closed triangles). A) The photosynthetic rates of these genotypes over the course of five days. The average pot water capacity o f the drought stress treatment is listed below each day of measurements. B) The stomatal conductance of the same plants from (A). Means that share no assigned letters are significantly different from each other (alpha=0.05). All error bars represent standa rd error. 0 5 10 15 20 25 0 1 2 3 4 5 6 Photosynthesis (µmol CO2 m - 1 s - 1 ) Day Photosynthesis Jaguar SER16 Tepary Zorro Jaguar SER 16 Tepary Zorro A 71% 28% 21% 13% Pot capacity: a ab b b bc c c cd cde cde de de de de e e e 85 Figure 12 (cont'd) 0 0.4 0.8 1.2 0 1 2 3 4 5 6 Conductance (mol H 2 O m - 2 s - 1 ) Day Conductance Jaguar SER16 Tepary Zorro Jaguar SER 16 Tepary B 71% 28% 21% 13% Pot capacity: a ab a ab bc bc bc bc c cd cd cde de de e e ef ef f f f f 86 At day nine of the measurement period ( Figure 11 A), leaf samples from all treatments and genotypes were taken, and the leaf water potential of these samples was determined ( Figure 11 C ). At the point of sampling, the drought treatment was experiencing severe stress. Under well - watered conditions, all genotypes had similar leaf water potentials. The drought stress treatment had lower leaf water potentials than the well - watered control, and signifi cant differences arose among genotypes in this treatment. Jaguar and SER - 16 had higher leaf water potentials than Zorro and tepary under drought stress. Survey of tepary varieties The ANOVA found that genotype and stress level significantly affect photos ynthesis and stomatal conductance, and within certain treatments, significant differences exist within the tepary germplasm and between the tepary germplasm and Zorro, the common bean check. Under mild drought stress, differences were found among the tepar y varieties for both photosynthesis and stomatal conductance ( Table 3 ). Those varieties with higher photosynthetic rates also had commensurately higher stomatal conductance: varieties such as Tohono O'dham and San Ignacio. Additionally, the two wild tepary varieties also had higher photosynthesis and conductance than some of the domesticated varieties. However, under mild drought stress, Zorro had much higher photosynthesis and conductance than any tepary variety tested. For severe drought stress, few diff erences existed for the genotypes tested, and all gas exchange rates were low. Wild tepary - Chihuahua, Tohono O'dham, and Zorro all had higher photosynthetic rates at this level of stress, but only Tohono O'dham had a higher stomatal conductance. Under re covery conditions, Zorro again had much higher photosynthesis and conductance than any tepary variety although it did not recover to levels it attained under mild 87 Table 3 - Tepary genotype survey The average rate of photosynthesis (A) and stomatal conductance (g s ) for eight tepary bean genotypes and the elite common bean Zorro after exposure to mild drought stress, severe drought stress, and recovery from drought stress. Within a column, means that share no letters are significantly di fferent from each other (alpha = 0.05). Bean lines Gas e xchange t raits A g s Mild Severe Recovery Mild Severe Recovery Zorro 19.5 a 4.3 a 12.5 a 0.585 a 0.03 ab 0.223 a W ild tepary - Chihuah 13.6 b 2.8 bc 5.4 b 0.175 bc 0.019 b 0.058 b San Ignacio 13 b 1.6 c 4.7 b 0.239 b 0.009 b 0.044 b W ild tepary - Tiburón 12.2 bc 3.5 ab 2.9 b 0.186 bc 0.024 ab 0.025 b Tohono O'dham 12 bc 4.8 a 4.1 b 0.171 bc 0.048 a 0.042 b Big Fields Brown 11.2 bcd 2.4 bc 4.2 b 0.141 cd 0.014 b 0.049 b Black Tepary 9.8 cd 1.8 c 4.8 b 0.137 cd 0.011 b 0.05 b Sacaton White 8.4 de 1.5 c 5.9 b 0.099 cd 0.01 b 0.058 b Tucson Brown 6.3 e 1.8 c 3.1 b 0.057 d 0.011 b 0.024 b LSD 3.1 1.3 3.1 0.095 0.026 0.041 Mean 11.8 2.7 5.3 0.199 0.02 0.064 Legend - A - photosynthetic rate ( µ mol CO 2 m - 2 s - 1 ); g s - stomatal conductance (mol H 2 O m - 2 s - 1 ); LSD - Least Significant Difference, using an alpha of 0.05 8 8 stress. All the tepary varieties had slight increases in their gas exchange under recovery compared to severe stress, but no tepary variety had higher rates than any other during recovery, and all their rates were approximately a third to a half of what they were under mild stress. Gas exchange of grafted beans Photosynthesis decreased in response to increasing drought stress for all graft types ( Figure 13 A). Overall, a Jaguar shoot on a Jaguar root had the highest photosynthetic rates during the drought period while Jaguar/tepary had the second highest rates. Tepary/Jaguar had the lowest photosynthetic rates, and tepary/tepary had rates intermediate between Jaguar/tepary and tepary/Jaguar. By the most severe drought stress five days after watering, most graft types had similarly low photosynthetic rates. Stomat al conductance had more pronounced differences. Jaguar/Jaguar and Jaguar/tepary had similarly high conductances, especially under mild to moderate drought, while tepary/Jaguar and tepary/tepary had substantially lower conductances in comparison ( Figure 13 B ). Tepary/Jaguar and tepary/tepary also maintained higher intrinsic water use efficiencies than the other two graft types throughout the experiment ( Figure 13 C). Five days after watering, gas exchange in all leaves was too low and variable to derive any me aning from water use efficiency. Leaf density and stomatal density Average leaf density was not significantly different between Jaguar and tepary. Both genotypes had fresh leaf densities that were approximately 0.018 g cm - 2 . Average stomatal density was not significantly different between Jaguar and tepary. Their stomatal density was approximately 230 - 238 stomata per mm 2 . 89 Figure 13 - Gas exchange in grafted plants Reciprocal and self - grafts of drought susceptible Jaguar and drought tolerant tepary expose d to progressively increasing drought stress. Water was withheld after day 0, and the pots dried down over the subsequent days. (A) Average photosynthesis measurements for each graft type during each day of the drought stress period. (B) Average stomatal c onductance during this period. (C) Average water use efficiency during this period. Outside marker color indicates shoot genotype, inside marker color represents root genotype; blue represents Jaguar and green represents tepary. Solid lines indicate self - g rafts, and dashed lines indicate interspecific grafts. Error bars represent standard error. Means that do not share letters are significantly different from each other (alpha = 0.05). 0 5 10 15 20 25 0 1 2 3 4 5 Photosynthesis (µmol CO 2 m - 2 s - 1 ) Days after withholding water Photosynthesis jag/jag jag/tep tep/jag tep/tep A a b c cd d d de de de de e e 90 Figure 13 (cont'd) 0 0.2 0.4 0.6 0.8 1 0 1 2 3 4 5 Stomatal Conductance (mol H 2 O m - 2 s - 1 ) Days after withholding water Conductance jag/jag jag/tep tep/jag tep/tep B a ab bc c cd d d d d d e e 0 75 150 225 300 375 0 1 2 3 4 5 Water Use Efficiency (µmol CO 2 mol - 1 H 2 O) Days after withholding water Water Use Efficiency jag/jag jag/tep tep/jag tep/tep C a a a b c d c e de de 91 Discussion Effects of drought on pod abscission, water potential, and chlorophyll fluorescence The common bean genotype Fuji had a high rate of pod abscission under drought stress, especially when compared to the Zorro. Fuji's high abscission rate explains in part its poor agr onomic performance under drought stress conditions. The effects of drought on plants in their seed - filling stage is well characterized although most of the research has been done on cereal crops (Boyer and Westgate, 2004; Bahieldin et al., 2005; Tolk et al., 2013) . The seed - filling stage is especially crucial: in maize, plants exposed to only three days of drought stress had the same decrease in kernel number as plants with a prolonged exposure to drought (Westgate and Boyer, 1986) . The abortion of develo ping seed embryos under drought stress is not caused by the lack of water to embryo but by the lack of photosynthate available to the embryo (Gengenbach, 1977; Boyle et al., 1991) . However, low water availability reduces rates of photosynthesis, and low ph otosynthetic rates reduce the amount of photosynthate available to the embryos unless reserves from other tissues, such as the stem, are remobilized (Yang and Zhang, 2006) . Supply of photosynthate is thus connected closely to reproductive development, and the higher concentrations of soluble sugars that we see in some drought tolerant genotype s exposed to drought stress (Chapter 2) likely contribute to their reproductive success and yield. One of the few studies in common beans found that a drought tolerant variety and a drought susceptible variety had a proportionally similar reduction in pod and seed number between control and drought treatments, but the drought tolerant variety had higher seed yield in all cases (Boutraa and Sanders, 2001) . However, the a uthors did not state whether the reduced yield was because of reduced pod formation or increased pod abortion. A similar experiment found no difference in 92 pod number under control conditions but that the drought susceptible bean genotype had a much higher rate of pod abscission than the drought tolerant genotype (Lizana et al., 2006) . In the current study, although Zorro and Fuji ended having the same number of pods per plant, the moderately drought tolerant Zorro formed fewer pods initially and aborted fe wer pods after two weeks while the drought susceptible Fuji formed a large number of pods initially and aborted many of them over the course of the two weeks. Differences in plant architecture likely played a large role in differences seen between Fuji and Zorro. Fuji has a Type I determinate bush architecture while Zorro has a Type II indeterminate bush architecture. Under well - watered conditions, bean plants with Type I determinate bush architecture abscise a greater number of pods during the pod - filling stage of development than bean plants with other architectures (Izquierdo and Hosfield, 1983) . Ultimately, Fuji expended more photosynthate on pods that were ultimately unproductive. Zorro's greater ability to regulate development allowed it allowed it to waste less photosynthate under the photosynthesis - limited condition of drought . Even moderate to severe levels of drought stress do not perturb Fv/Fm (Figure 8A ). The stresses that photosystem II faced during the day, as seen in a decrease in PSII (Figur e 8B ) , were not sufficient to permanently damage it, as Fv/Fm was able to recover after the dark period. Granted, the plants had an entire ten hours to dark adapt and recover from any stresses during the day. A shorter period of dark adaptation might have had a Fv/Fm still affected by transient or slow - relaxing photoinhibition, but this information is already more accurately conveyed by changes in PSII (Baker, 2008) . In both common bean, tepary bean, and cowpea, Fv/Fm did not vary between control and even severe drought stress (Castonguay and Markhart, 1991; Souza et al., 2004) while drought stress significantly impacted PSII and related parameters like photochemical quenching and electron transport rate (Souza et al., 2004; Wentworth et al., 93 2006) . Both h eat and drought stress affected PSII at mild to moderate levels, but Fv/Fm did not shift until long exposure to extreme conditions of stress (Havaux, 1992) . A wheat screening study did find small variations in Fv/Fm under stress, but it used detached leav es, its period of dark adaption was only 30 minutes, and even then no differences were found until the leaves were exposed to a combination of high light and high temperature directly preceding the dark adaption (Sharma et al., 2012) . The decrease in PS II under drought stress mirrors the decrease in accumulated biomass under drought stress (Figure 8C ). However, while RAB - 651 had a significantly lower PSII than Jaguar and Zorro under drought stress, no significant differences in biomass were found among genotype s in the drought stressed treatment. Averaging both the well - watered and drought stress treatments , Jaguar's biomass is significantly lower than that of RAB - 651 or SER - 16. PSII is much more responsive to drought stress and is a better parameter fo r screening for drought response than Fv/Fm. The drought tolerant tepary and SER - 16 had lower leaf water potentials under drought stress than the more drought susceptible Jaguar and Zorro (Figure 9 ). The physical flow and retention of water in plant tissu es has been well studied with regards to stress. A water stressed drought tolerant soybean genotype had lower leaf hydraulic conductivity than other genotype s (Sinclair et al., 2008) , which could lead to lower leaf water potentials. Tying leaf water potent ial to the abscisic acid work of Chapter 2, abscisic acid decreased xylem conductance (Chaves et al., 2003) but increased root hydraulic conductivity (Thompson et al., 2007) . Thus, stress and abscisic acid causes more water to flow into the plant through t he roots but decreases how quickly it moves through the plant and how quickly it escapes through the leaves. More drought tolerant genotype s like tepary and SER - 16 thus conserved water by creating and tolerating lower 94 leaf water potentials. The more suscep tible genotype s Jaguar and Zorro had higher leaf water potentials, so the water in their leaves was more likely to transpire out of the plant from the leaf and its stomates. The increase in soluble sugars could also play a role in decreasing leaf water pot ential (Chapter 2). However, while tepary had the highest concentration of soluble sugars (Chapter 2) and the lowes t leaf water potential (Figure 9 ), SER - 16, despite having a lower concentration of soluble sugars, still had a lower leaf water potential tha n Zorro. Thus, other factors beyond the measured solute accumulation s play a role in controlling leaf water potential; however, the ability to create and maintain lower leaf water potentials aids plants in surviving and yielding under drought stress. Pho tosynthesis and drought The photosynthetic parameters V cmax and J vary significantly among bean genotypes while TPU and g m do not ( Table 2 ). Although no differences were found in the g m for these genotypes as calculated from A - C i curves, using direct measurement methods instead of estimation would likely yield more accurate and precise values (Sharkey, 2015) ; methods such as carbon isotope discrimination (Evans et al., 1986) or combined A - C i and chlorophyll fluorescence (Flexas et al., 2002) could reveal unseen differences. That the A - C i curves were performed on non - stressed plants in this experiment could also contribute to a lack of differences. The g m of grapevines decreases in response to drought stress (Flexas et al., 2002) , and both interspecific and intraspecific differences in the response of g m to increased temperature have been found (Pimentel et al., 2013; von Caemmerer and Evans, 2014) . It seems likely that drought - induced differences in g m would also exist among bean g ermplasm as well. That no differences in TPU were found among bean genotypes is unsurprising : TPU is rarely the limiting 95 rate under natural conditions (Sharkey et al., 2007) , so no adaptive advantage would be gained from a higher TPU. However, V cmax and J are often limiting factors under conditions that plants face in the field, and it is for these parameters that genotypic differences were found. V cmax is the limiting factor under low C i while J is the limiting factor under high C i (Long and Bernacchi, 2 003) . Because drought stress induces stomatal closure and lowers C i , Zorro's higher V cmax could contribute to its productivity under moderate drought stress; however, this would be a drought tolerance mechanism unique to Zorro and not shared by either SER - 95 or Tepary. Additionally, Zorro's higher V cmax would give it a greater amount of photosynthate , especially under drought stress because the effect would be greatest at low C i . This improved supply of photosynthate could in turn be transported to developi ng reproductive tissues to prevent their abortion and abscission, thus accounting for Zorro's lower rate of pod abscission ( Figure 7 ). The higher J of both Tepary and Zorro could contribute to their higher general productivity, but it is unlikely to contri bute to their productivity under drought stress as it would not be a limiting factor at the low C i of drought stress. However, Zorro yields more than other genotype s within its market class, including Jaguar (Kelly et al., 2009 a ) , and this productivity co uld be due to its higher J that operates under favorable conditions. Likewise, Tepary behaves in some regards like a desert ephemeral (Nabhan and Felger, 1978) , so its high J would make it more productive after the seasonal rains of the Sonoran Desert when water, heat, and light are all plentiful for a brief period. Bean genotype and water treatment predictably affected photosynthetic rates and stomatal conductance ( Figure s 11 and 12 ). The four genotypes had different performances under well - watered condit ions, and although drought stress caused both photosynthesis and stomatal 96 conductance to decrease in all genotypes, their relative ordinal performance was unchanged; i.e., Jaguar and Zorro had the highest rates of photosynthesis and stomatal conductance, S ER - 16 intermediate, and tepary the lowest. In all treatments and genotypes, photosynthetic rates followed the same trends as stomatal conductance. This was an expected result, as drought stress induces stomatal closure, which in turn limits the flow of CO 2 to the sites of carbon fixation, thus lowering photosynthetic rates (Chaves and Oliveira, 2004) . Interestingly, while drought susceptible Jaguar and moderately tolerant Zorro had higher stomatal conductances, it was the more drought tolerant SER - 16 and te pary that had lower stomatal conductances under both drought and well - watered conditions ( Figure 11 B). Because stomata are the primary site of water loss from plant tissues, the regulation of their opening and closing also impacts a plant's water use and w ater content (Medrano et al., 2002; Flexas et al., 2004; Daszkowska - Golec and Szarejko, 2013) . While causing a slight reduction in photosynthesis under well - watered conditions, SER - 16 and tepary's lower stomatal conductances may allow them to survive and y ield under water limited conditions. In soybean, leaf hydraulic conductance was unchanged by exposure to drought stress while stomatal conductance and leaf water potential decreased (Locke and Ort, 2014) , again emphasizing the importance of stomatal contr ol to drought response. Other studies on drought stress and gas exchange in common beans also found a strong c onnection between photosynthetic rates and stomatal conductance, but they either used few genotype s (Rosales et al., 2012) or uncharacterized geno type s (Ramalho et al., 2014) , preventing any extrapolation of a genotype's level of drought tolerance and its rates of photosynthesis and conductance. The current study also adds a greater temporal resolution to onset and progression of drought stress in P haseolus . In Eucalyptus spp. originating from different regions, those species adapted to hot, 97 dry environments had higher water use efficiencies than those from cooler, humid environments, even when all species were grown in the same common garden with mi ld to moderate drought stress (Héroult et al., 2013) . The drought tolerant SER - 16 and tepary also had higher water use efficiencies, achieved by their lower stomatal conductances ( Figure 11 B). Getting closer agronomically and genetically, comparisons betwe en a cultivated soybean variety ( Glycine max (L.) Merr.) and a stress tolerant wild relative ( Glycine soja Sieb. and Zucc.) showed that G. soja reduced its transpiration to a greater extent than cultivated soybean as their soil gradually dried (Seversike e t al., 2014) . G. max and G. soja. are analogous to common bean and tepary bean, and in both cases, it was the stress tolerant related species ( G. soja and tepary) that control to a greater extent the water flowing out of their stomata. While tepary's lowe r stomatal conductance under drought stress had been examined previously (Markhart, 1985; Castonguay and Markhart, 1992) , here the results also showed that drought tolerant common bean genotype s also have a similar response to drought stress. Previous stud ies reported no differences in the leaf water potential of common bean and tepary bean under drought stress (Castonguay and Markhart, 1991, 1992) , but those studies imposed stress via quick acting methods of applying osmotica and dry air streams to detache d leaves. In the present experiments, even when drought stress was imposed by withholding water and allowing potting soil to dry gradually, different results were obtained: the prolonged and more severe stress caused Zorro to have a significantly lower lea f water potential ( Figure 11 C ) while a shorter and more moderate stress caused SER - 16 to have a lowe r leaf water potential (Figure 9 ). Given the variation that arose when drought stress was imposed by withholding water, it is unsurprising that using differ ent methods of imposing drought stress would produce different results. 98 The approximate pot water capacities at which stomatal conductance and photosynthesis were affected were also examined ( Figure 12 ). Special care should be taken in translating this in formation to field conditions; the height of the water column in the pots was much shorter than a typical water column in the field, so the suction force is much weaker for the pots, causing them to have a higher absolute water content for a given relative capacity (Passioura, 2006) . For that reason, the term "pot capacity" is used instead of the more familiar "field capacity". Variation and grafting of tepary While variation was found within the tepary germplasm tested, the variation between the tepary germplasm and a model common bean genotype , represented by Zorro, was still greater ( Table 3 ). Given the variation within tepary, future breeders may want to car efully consider which accessions of the tepary genepool to include in future introgressions into common bean germplasm so that they continue to build on the advances of previous introgressions (Mejía - Jiménez et al., 1994; Muñoz et al., 2004, page 200; Rao et al., 2013; CGIAR, 2015) . Such considerations would also be important in efforts to improve domesticated tepary germplasm, as even this basic survey indicated that sufficient variation on which improvements could be built exists within the tepary genepoo l. When considering tepary bean as an ideotype for breeding drought tolerant common bean lines (Blum, 2011; McClean et al., 2011; Beebe, 2012) , not only did all the tepary lines have lower stomatal conductances compared to Zorro at the onset of stress, but they also had lower conductances well into the recovery phase ( Table 3 ). This slow recovery of stomatal conductance would be especially advantageous in production areas that experience terminal drought stress. In these areas, after the onset of water stre ss, any subsequent precipitation is likely to be low in volume and transient. By not immediately recovering stomatal 99 conductance, increasingly scarce water resources are conserved further into the end of the season. However, it is important to note that Zo rro's quicker recovery of stomatal conductance would be beneficial in production areas that experience intermittent drought (Beebe et al., 2012) . These stomatal behaviors also match the stronger drought escape mechanisms of tepary (Nabhan and Felger, 1978) in comparison to common bean. The drought escape of tepary would give it a yield advantage under terminal drought conditions but a yield disadvantage under intermittent drought conditions compared to a common bean with weaker drought escape mechanisms. I n previous studies involving tepary bean, usually only one tepary accession was used for comparisons to common bean (Markhart, 1985; Castonguay and Markhart, 1991, 1992; Udomprasert et al., 1995; Butare et al., 2011) although exceptions exist (Rainey and Griffiths, 2005 b ; Rao et al., 2013) . Although considerations of space and time will restrict the use of multiple tepary accessions in experiments, even as they have in the present research, this survey of tepary landraces and wild accessions shows that eno ugh variation exists that caution must be used when extrapolating results from one genotype to the entire species. Further physiological characterization of tepary germplasm should be done to complement the ecological survey of the species (Nabhan and Felg er, 1978) . Grafting two disparate genotypes onto each other revealed the broad effects of root and shoot tissue on a plant's response to drought. Among other factors, tissue specific signaling contributed to the interactions between root and shoot that we re seen for photosynthesis. Shoot tissues produce abscisic acid and transport it to the roots, and root ABA content is dependent on shoot identity (Chapter 2). Likewise, root tissues send drought stress signals, some of which are still uncharacterized, to shoots and leaves (Chaves et al., 2003) . Additionally, the plant hormone cytokinin is produced in both the shoot and root tissue (Frébort et al., 2011) and induces cell 100 division and stomatal opening, acting as an antagonist to abscisic acid (Zwack and Rash otte, 2015) . The production and reception of some of these signals likely varies between Jaguar and tepary. Because of these differences in signaling, even though the graft type Jaguar shoot on tepary root was intermediate between Jaguar/Jaguar and tepary/ tepary in terms of photosynthesis, the tepary/Jaguar was not intermediate between the two self - grafts but instead had a lower photosynthesis than tepary/tepary. The effects of graft type on stomatal conductance were much more apparent. Shoot genotype comp letely determined stomatal conductance independent of root genotype ( Figure 13 B). Any graft type that had tepary for its shoot tissue had a lower stomatal conductance under both well - watered and drought conditions. Thus, because of the lower stomatal condu ctances, water use efficiency was also primarily determined by shoot genotype ( Figure 13 C). Under conditions of drought, a plant avoids water stress by increasing water uptake through the roots and decreasing water loss through the leaves by closing stomat a (Chaves et al., 2003) . But even in the root limited environment these plants were grown in, which would prevent increased water uptake by root exploration of the soil, stomatal control still operated as an important drought avoidance mechanism for tepary . A previous study involving reciprocal grafts of common bean and tepary bean and their water relations under drought stress found that leaf water potential was determined by root genotype while the root:shoot ratio was determined by the additive effects o f root and shoot genotypes (Sanders and Markhart, 1992) . That study also examined stomatal conductance, but its results were inconclusive. In the current study, it was found that because of tepary shoot's lower stomatal conductance, it conserved water and avoided drought stress while still having a higher photosynthesis per unit of water lost than Jaguar shoots, all regardless of root identity. 101 Leaf traits Neither density of leaf tissue nor stomatal density varied between Jaguar and tepary (Figures 7 and 8). Not only are tepary and Jaguar distinct in their level of drought tolerance, with tepary being very tolerant and Jaguar being susceptible, but they are also genetically distinct, the two being different species belonging to the same genus. Thus, if any drought tolerance - connected differences in leaf or stomatal density were to be found, they would be found between Jaguar and tepary. As only two genotypes were used, these results do not prove a lack of variation for leaf and stomatal density within Phase olus , but they do indicate that these traits have no effect on the drought tolerance of genotypes within this genus. Other plant families do exhibit variation in stomatal density (Hetherington and Woodward, 2003) . However, within the genus Banksia , no clear relationship exists between a species' stomatal density and the availability of water within its ecological niche (Drake et al., 2012) . As tepary and Jaguar have the same stomatal density, the differences observed in their stomatal conductance ( Figure 11 B) are caused by differences in the regulation of stomatal opening. Leaf density varies not only by vegetation type but also by ecological conditions; species adapted to dry environments tend to have a higher leaf density (Reich et al., 1999; Wrig ht et al., 2005) . As tepary has small leaves, another adaptation often seen in desert - adapted plants, it was hypothesized that it might also have denser leaves as well, but tepary's leaves were no denser than it s drought susceptible relative. Leaf density has no connection to drought tolerance within Phaseolus . 102 Conclusions The response of beans to drought stress varied in predictable ways according a genotype's drought tolerance. Often, differences among genotypes were not apparent under well - watered c onditions and only manifested when the plants experienced drought stress. Under drought stress, the more tolerant genotype shed fewer developing pods. While no differences existed under well - watered conditions, under drought stress, drought tolerant genoty pes maintained lower leaf water potentials. Other drought - related traits were constitutive and observable even under well - watered conditions. Tolerant genotypes had lower photosynthetic rates and stomatal conductances under both control and stress treatmen ts. While photosynthetic parameters such as V cmax and J did not relate directly to drought tolerance, they might contribute to a plant's primary productivity, which in turn can contribute to a plant's tolerance to stress. Traits such as leaf tissue density and stomatal density did not vary and play no part in drought tolerance in Phaseolus . While shoot and root tissues both affected photosynthesis, shoot tissue solely affected stomatal conductance, and thus, it is the most important determinant of water los s and the water use efficiency. Significant variation exists within P. acutifolius germplasm in their response to drought, but the variation between P. acutifolius and P. vulgaris is likely greater. From these results, an ideotype of a drought tolerant bea n can be built and used to aid breeding decisions. This drought tolerant ideotype would have low stomatal conductance under both well - watered and drought stress conditions, have a low rate of pod abscission, small leaves, and, for areas that frequently exp erience terminal drought, a flowering stage that could be induced by drought conditions. 103 Chapter 4: Heat stress screens 104 Introduction Common beans ( Phaseolus vulgaris L.) are subject to a number of stresses in their production areas. Chief among these stresses is drought, which affects a large portion of the world bean crop every year (Cavalieri et al., 2011; Beebe, 2012) . Heat stress also impacts bean production, not by reducing yields to the same extent as drought bu t by limiting the areas where beans can be grown (Beebe et al., 2011) ; climate change will most likely increase the area of land unsuitable for bean production because of high temperatures. Additionally, h eat stress is often coincident with drought stress in the field and can increase its severity (Beebe et al., 2013) . As with many abiotic stresses, heat and drought stress induce some of the same responses in plants. Plants responses to both involve the sam e redox pathways (Koussevitzky et al., 2008) , signaling plant hormones (Wahid et al., 2007; Cutler et al., 2010) , heat - shock proteins and molecular chaperones (Wang et al., 2004) , and transcription factors (Schramm et al., 2008) . However, heat and drought stress also activate different transcriptional pathways involved in independent metabolic and physiological responses (Rizhsky et al., 2004; Mittler, 2006; Shulaev et al., 2008; Rasmussen et al., 2013) . The extent of overlap between heat and drought tolera nce in common bean has been studied little, and it remains a question whether it is the differences or the commonalities in heat and drought pathways that have the greatest practical effect in plant performance. Much of the work done on heat stress in bea ns focused on its detrimental effects on reproduction. High temperatures render pollen unviable and thus prevent seed set and limit yield (Gross and Kigel, 1994; Prasad et al., 2002) . Greenhouse studies successfully identified genotypes with tolerant repro ductive systems that could produce seed at high temperatures 105 (Rainey and Griffiths, 2005 a , b ) . Using the same principles, a screen of multiple bean lines at a field site with temperatures prohibitive to the cultivation of most beans separated heat tolerant from susceptible genotypes based on their ability to set seed (CGIAR, 2015) . In both greenhouse and field studies, tepary bean ( Phaseolus acutifolius A. Gray) genotypes or common bean genotypes with tepary in their pedigree were well - represented among the heat tolerant group (Rainey and Griffiths, 2005 b ; CGIAR, 2015) . Tepary originated in the Sonoran Desert, located in present - day northwest Mexico and southwest United States, and was used as a reliable staple crop in this region for hundreds of years (Nabh an and Felger, 1978) . Compared to common bean, tepary has a great tolerance for both heat and drought stress (Castonguay and Markhart, 1992; Udomprasert et al., 1995; Rao et al., 2013) . Tepary thus serves as a useful check and ideotype for abiotic stress t olerance in beans (Beebe et al., 2013) . While these previous studies focused on the effects of heat on bean reproduction, the present study examined the effects of heat stress on vegetative tissues. Field tests are potentially the best measure of crop per formance, but unpredictable weather patterns, such as too much rain or low temperatures, can prevent the evaluation of a desired trait. Thus, in addition to studying the heat stress physiology of beans, the present study also tested the feasibility of usin g high temperatures in a controlled environment to screen germplasm for heat and drought tolerance. These screens could then be used to supplement information from field evaluations and help inform breeding decisions. A selection of genotypes, some evaluat ed for drought tolerance in the field and others tested in previous experiments, were subjected to gradually increasing temperatures over the course of a week in a growth chamber. The genotypes were measured periodically for photosynthetic traits and stres s - related damage. 106 Materials and m ethods Fifteen genotype screen across three temperatures Fifteen bean genotypes were used in this experiment; ten genotypes came from a drought diversity panel and were recommended based on their diversity of drought response by Dr. Carlos Urrea of University of Nebraska - Lincoln at the Scottsbluff research station and Dr. Timothy Porch of the USDA Tropical Agriculture Research Station in Mayaguez, Puerto Rico. Five other genotypes were included from previous stress e xperiments (the full list of genotypes is listed in Table 4 ). The fifteen genotypes were planted sequentially in four blocks with one week of spacing between each block and two replications of each genotype per block. Plants were first grown in 5 L black p lastic pots filled with Suremix Pearlite potting media (Michigan Grower Products, Galesburg, MI, USA) in a greenhouse set to 25 ° C in East Lansing, MI from December 2014 to February 2015. Pots were fertilized with 12 g of Osmocote - 15 - 9 - 12 slow - release fert ilizer (Bloomington Brands, LLC, Bloomington, Indiana, USA) and supplemented twice a week by watering with half - strength Hoagland solution until the pots were saturated and dripping. One month after its planting, when the third trifoliate was fully mature , a block would be moved into a growth chamber set to 35 ° C during the light period and 30 ° C during the dark period; the chamber had a 14 h light and 10 h dark cycle with a light intensity of 400 µ mol photons m - 2 s - 1 . For each block, no more than five day s elapsed between when the earliest plants had mature third trifoliates and the latest plants had them, and movement was delayed until every plant had mature third trifoliates. After two days in this environment, gas exchange and 107 Table 4 - Bean genotypes used in screen A list of bean genotypes used in the progressively increasing heat stress experiment along with their parentage or origin and their level of drought tolerance. Characterization of drought tolerance in the numbered lines comes from yield dat a taken field plots experiencing moderate drought stress (Tim Porch, personal communication). Genotype Origin Drought Response I14538 Tacana x VAX6 Tolerant I14539 (Morales x XAN176) x (BAT 477 x B98311) Susceptible I14541 Black Rhino x SEN 10 Moderately tolerant I14544 (BelMiDak RMR 10 x B01741) x ( BAT 477 x L88 - 63) Tolerant I14545 Matterhorn x SER 21 Moderately tolerant I14546 USPT - ANT x ( Matterhorn x 98078 - 5 - 1 - 5 - 1) Moderately tolerant I14548 Merlot x (Merlot x SER - 16 ) Susceptible I14549 Merlot x (98020 - 3 - 1 - 6 - 2 x Tacana) Moderately tolerant I14550 Merlot x (98020 - 3 - 1 - 6 - 2 x Tacana) Moderately tolerant I14553 Merlot x (05F - 5055 - 1 x 98020 - 3 - 1 - 6 - 2) Tolerant Fuji Hime tebo x Matterhorn (Kelly et al., 2009 b ) Susceptible Jaguar B90211 x N90616 (Kelly et al., 2001) Susceptible SER - 95 CIAT breeding program Tolerant Tepary (TB1) Breeding program, USDA - ARS Mayaguez, Puerto Rico Very tolerant Zorro B00103*2 x X00822 (Kelly et al., 2009 a ) Moderately tolerant 108 chlorophyll fluorescence measurements were taken on the second most recently mature trifoliate leaf. Then the temperature was raised to 40/35 ° C, and the acclimation time and measurements repeated, and the temperature again raised to 45/40 ° C, and the acclimation time and measurements repeated. Finally, canopy temperature and visual ratings were taken on all plants. The plants were then cleared out from the growth chamber, and the entire cycle was repeated using the next block of plants. Gas exchange m easurements of photosynthesis and stomatal conductance were made with a LI - COR 6400XT (LI - COR Biosciences, Lincoln, NE, USA). Gas exchange measurements were taken under a light intensity of 1000 µ mol photons m - 2 s - 1 , a CO 2 concentration of 400 µ mol mol - 1 , ambient relative humidity of approximately 50 - 70%, and ambient temperature. Chlorophyll fluorescence measurements of photosystem II efficiency ( PSII ) and non - photochemical quenching (NPQ) were taken using the MultispeQ as part of the PhotosynQ platform ( photosynq.org). An E30bx infrared camera (FLIR Systems, Inc., Wilsonville, OR, USA) was used to measure the leaf canopy temperature at an ambient air temperature of 45 ° C. The camera was positioned 1 m above each plant and pointed perpendicular to the gro und for each infrared picture taken. The temperature of the entire visible canopy was averaged and used as that plant's leaf canopy temperature. One hour after being removed from the growth chamber and placed back in the greenhouse, each plant was rated o n its qualitative appearance on a 1 - 5 scale ; examples of this scale are found in Figure 14 G . A rating of ' 1 ' represented a plant that was nearly dead , '2' a plant that showed severe damage , '3' a plant that had moderate damage, '4' a plant that had little visible damage, and ' 5 ' represented a plant that was undamaged . 109 The data were analyzed using the statistical software program SAS version 9.4 (SAS Institute, Cary, NC, USA). The procedure PROC MIXED was used to determine the analysis of variance (ANOVA) f or various factors and if they significantly impacted the dependent variable. The data for photosynthesis, stomatal conductance, PSII , and NPQ were treated as repeated measures with the factors of genotype and temperature; the "repeated" statement was use d in the analysis. The canopy temperature and visual rating data were treated as single factor experiments with genotype being the only contributing independent variable. The predetermined alpha cutoff value was 0.05. Select genotype screen over increasin g temperature Having tested the effects of increasing temperature on fifteen genotypes, five genotypes were selected from that group for a more in - depth study to determine on a finer temporal scale when and at what temperatures differences in genotypes ap pear. The five genotypes were I14538, I14541, I14553, Jaguar, and Tepary, and they were chosen to cover a range of responses to heat stress based on the previous fifteen genotype study. Six replications of each genotype were planted in a gr owth chamber set to 25/20 °C light/dark temperature; other environmental conditions were the same as described above in the 'F ifteen genotype screen across three temperatures' section . After one month of growing in these conditions, the third trifoliate leaf of each plant was mature. Then, the light and dark period temperatures were raised by 5 and 4 °C, respectively, every two days until the chamber reached a maximum of 45/36 °C. Measurements of gas exchange and chlorophyll fluorescence were taken as described above on th e day before the first temperature increase and every day thereafter. 110 Destructive measurements of electrolyte leakage and thiobarbituric acid - reactive substances (TBARS) were taken from leaf samples harvested every two days to measure stress - induced damag e to cellular membranes and oxidative damage, respectively. For electrolyte leakage, 17 mm diameter leaf punches were taken from the latest fully expanded leaf of each plant. The leaf punches were immediately placed in 20 mL of deionized water. The leakage of ions from the leaf punches into the surrounding water was measured as an increase in electrical conductivity using a conductivity probe (OAKTON Instruments, Vernon Hills, IL, USA). Conductivity was measured every 30 - 45 minutes until the readings platea ued. Samples were then frozen in a conventional - 20 ° C freezer to fully disrupt membranes, thawed, and conductivity was measured again. Electrolyte leakage was determined by dividing a sample's plateaued conductivity by its post thaw conductivity. TBARS measurements were made according to the protocol in Wang et al. (2009) with slight modifications. Briefly, 0.5 g of fresh leaf tissue was ground up in 5 mL of water containing 0.1% (weight/volume) trichloroacetic acid. The solution was centrifuged, and 0.3 mL of the supernatant was combined with 1.2 mL solution of 20% (w/v) trichloroacetic acid and 0.5% (w/v) thiobarbituric acid. The resulting solution was placed in a heating block set to 100 °C for 30 minutes, after which the reaction was terminated by rem oving the solution and placing in an ice water bath. After cooling, the solution was centrifuged, and then its absorbance at 535 nm was read using a spectrophotometer. The TBARS concentration was calculated using an extinction coefficient of 155 mM - 1 cm - 1 . The data was statistically analyzed as described above in the "F ifteen genotype screen across three temperatures' section. 111 Results Fifteen genotype screen across three temperatures Among the 15 bean genotypes, variation was found for all of the respon ses measured. The ANOVA found that both genotype and temperature have p - values below the alpha value (0.05) and thus significantly contribute to the outcome of photosynthesis, stomatal conductance, and NPQ ( Table 5 ) . Genotypic effects also significantly af fected leaf canopy temperature and qualitative visual rating ( Table 5 ) . However, PSII 's p - value for genotype was 0.08, above the alpha value of 0.05, and thus, genotype does not meet the criteria for being considered a significant contributor to PSII . Te mperature was , however, a significant contributor to PSII ( Table 5 ) . Photosynthesis remained fairly constant at 35 °C and 40 °C ( Figure 14 A). Little variation existed among genotypes within a temperature or between these two temperatures within a genotyp e. However, at 45 °C, differences among genotypes became apparent, and photosynthesis ranged from 0 - 12 µ mol CO 2 m - 2 s - 1 . Rates of photosynthesis at 45 °C were significantly lower than those at 35 or 40 °C. Stomatal conductance was significantly higher at 40 °C than it was at 35 °C. At 45 °C, stomatal conductances fell below those at 35 °C, and genotypes that had comparatively high conductances at previous temperatures did not necessarily have them at 45 °C. At this highest temperature, a wide range of stomatal conductances was again found among genotypes, and they correlate d well with photosynthetic rates at this same temperature (Table 6) . PSII did not vary between 35 and 40 °C, and the variation among genotypes was also small at these two temperatur es ( Figure 14 C). When the temperature was raised to 45 °C, PSII 112 Table 5 - ANOVA of heat screen experiment Displayed below are the p - values of the ANOVA test of the effect of the independent factors of genotype, temperature, and the interaction of genotype and temperature on the parameters of photosynthesis (A), stomatal conductance (g s ), photosystem II efficiency ( PSII ), non - photochemical quenching (NPQ), leaf temperature (L temp) , and visual rating (vis). An alpha value of 0.05 was used as the cut off for determining significance. The analysis was made with the PROC MIXED statement in the statistical analysis software SAS 9.4. Indep. factor P - values A g s PSII NPQ L temp vis Genotype <.0001 <.0001 0.0821 <.0001 0.0380 <.0001 Temperature <.0001 < .0001 <.0001 <.0001 - - Gen x Temp 0.0272 0.0251 0.0784 0.0422 - - 113 Table 6 - Correlations of measured parameters in heat screen experiment This table contains the Pearson correlation coefficients for the parameters measured in the 'Fifteen genotype screen across three temperatures' experiment. Correlated with each other are photosynthesis, conductance, photosystem II efficiency, non - photochemical quenching, plant canopy temperature, and visual rating. The analysis used data from all fifteen genotypes and three temperatures, with the exception of plant canopy temperature and visual rating: because they were only measured at the end of the experiment, these two parameters were correlated with the other parameters using only the 45 °C data. The p - values for each Pearson correlation coefficient are in parenthesis. A g s PSII NPQ T canopy Visual A 1 0.76997 (<.0001) 0.4961 (<.0001) - 0.33288 (<.0001) - 0.63448 (<.0001) 0.54955 (<.0001) g s 0.76997 (<.0001) 1 0.34305 (<.0001) - 0.26313 (<.0001) - 0.57997 (<.0001) 0.3371 (0.0002) PSII 0.4961 (<.0001) 0.34305 (<.0001) 1 - 0.63364 (<.0001) - 0.40963 (<.0001) 0.65241 (<.0001) NPQ - 0.33288 (<.0001) - 0.26313 (<.0001) - 0.63364 (<.0001) 1 0.15189 (0.1873) - 0.31965 (0.0046) T canopy - 0.63448 (<.0001) - 0.57997 (<.0001) - 0.40963 (<.0001) 0.15189 (0.1873) 1 - 0.36485 (<.0001) Visual 0.54955 (<.0001) 0.3371 (0.0002) 0.65241 (<.0001) - 0.31965 (0.0046) - 0.36485 (<.0001) 1 Legend: A - photosynthesis, g s - stomatal conductance, PSII - photosystem II efficiency, NPQ - non - photochemical quenching, T canopy - plant canopy temperature, visual - qualitative visual rating 114 Figure 14 - Heat screen with fifteen genotypes 15 bean genotypes subjected to first two days at 35 °C, then two days at 40 °C, and finally two days at 45 °C. (A) Average rates of photosynthesis for each genotype at the three different temperatures. Measurements were taken at the end of the two day period of each temperature regim e, in the mid - morning. (B) Stomatal conductance of the 15 genotypes, taken simultaneously with photosynthesis. (C) Photosystem II efficiency of the 15 genotypes, taken just preceding photosynthesis measurements. (D) Non - photochemical quenching of the 15 ge notypes, taken simultaneously with photosystem II efficiency. (E) The leaf canopy temperature of each genotype, taken on the last day of their exposure to 45 °C. (F) The qualitative visual rating of each genotype one hour after the end of the 45 °C period. Plants were rated on a scale of 1 to 5, with 1 being a plant that appeared mostly dead, and 5 being a plant that appeared undamaged. (G) Representative plants from each of the five visual rating scores. All error bars represent standard error. (A - C) Genot ypes are organized from greatest to least based on values at 45 °C. (D) Genotypes are organized from least to greatest based on values at 45 °C. (E) Genotypes organized from least to greatest. (F) Genotypes organized from greatest to least. 0 5 10 15 20 25 tep fuj zor 538 549 539 548 546 jag 553 ser 544 541 550 545 Photosynthesis (µmol CO 2 m - 2 s - 1 ) Genotype Photosynthesis 35 C 40 C 45 C A 115 Figure 1 4 (cont'd) 0 0.2 0.4 0.6 0.8 1 1.2 tep fuj 544 zor 549 553 538 539 548 546 545 jag ser 550 541 Stomatal Conductance (mol H 2 O m - 2 s - 1 ) Genotype Stomatal Conductance 35 C 40 C 45 C B 0.4 0.5 0.6 0.7 0.8 tep ser 549 zor 538 539 jag fuj 550 544 541 548 546 545 553 Photosystem II efficency Genotype Photosystem II efficiency 35 C 40 C 45 C C 116 Figure 14 (cont'd) 0 0.2 0.4 0.6 0.8 1 tep 544 jag zor ser 538 545 553 fuj 539 549 546 550 548 541 Non - photochemical quenching Genotype Non - photochemical quenching 35 C 40 C 45 C D 32 34 36 38 40 42 44 fuj 539 tep 545 zor ser 546 jag 538 550 548 553 549 544 541 Leaf temperature ( ° C) Genotype Leaf temperature at 45 ° C E 117 Figure 14 (cont'd) 1 2 3 4 5 tep ser 549 538 539 546 zor fuj 541 544 545 548 550 jag 553 Visual rating (1 - 5 scale) Genotype Visual rating post stress F 118 Figure 14 (cont'd) Legend : (1) top left is a picture of genotype I14538 after exposure to 45 °C for two days, (2) top center is I14538, (3) top right is I14538, (4) bottom left is I14541, and (5) bottom center in TB1 tepary G 119 decreased for almost every genotype. However, genotype to genotype variation also increased, and no significant differences among genotypes were found, even when the 45 °C data was analyzed on its own, separate from the other two temperatures . Like PSII , NPQ had few differences between temperature or genotype at 35 and 40 °C, but at 45 °C, many genotypes saw an increase in NPQ ( Figure 14 D). At this highest temperature, genotypes with the highest PSII also tended to have the lowest NPQ. Leaf canopy temperatures at 45 °C ranged from 38 to 41 °C, but there were few significant differences among genotypes ( Figure 14 E). Qualitative visual ratings taken after the stress treatments showed significant differences between genotypes, and significa nt variation was found among the genotypes' average ratings ( Figure 14 F). Select genotype screen over increasing temperature For most measured parameters and most temperatures, the five genotypes performed similarly. Photosynthesis remained stable for al l genotypes until the temperature reached 40 °C, causing a small drop in photosynthetic rates ( Figure 15 A). Upon reaching 45 °C, photosynthetic rates were very low , falling somewhere between 1 - 5 µ mol CO 2 m - 1 s - 1 , for all genotypes except tepary . Stomatal c onductance increased with increasing temperatures until temperature reached 40 - 45 °C; at that point, conductance decreased to low rates for all genotypes except tepary ( Figure 15 B). Most genotypes had an increase in PSII from 25 to 30 °C, and then PSII was fairly stable until temperatures reached 45 °C, at which point average PSII values decreased greatly for all genotypes except tepary ( Figure 15 C). Conversely, NPQ was uniformly low for all genotypes 120 Figure 15 - Heat screen with five genotypes Five bean genotypes were exposed to increasing temperatures: starting at 25 °C on day one, the temperature increased by 5 °C every two days until it reached a maximum of 45 °C at days eight and nine. (A) Photosynthetic rates of the five bean genotypes taken dai ly during this period. The dotted black line indicates the temperature at which each day's measurements were taken both in this graph and the following graphs in this figure. (B) The stomatal conductance of these genotypes over this period. (C) The photosy stem II efficiency of these genotypes during this period. (D) The non - photochemical quenching of these genotypes during this period. (E) The TBARS content of leaf samples taken from each genotype every two days. (F) The electrolyte leakage of leaf samples taken from each genotype every two days. (G) The qualitative visual rating of each genotype at the end of the nine - day heat stress period. Plants were rated on a scale of 1 - 5, with 1 assigned to a plant that seemed dead and 5 being assigned to a plant that seemed undamaged. Bars that share no letters are significantly different from each other (alpha = 0.05). See Figure 14 G above for example pictures of each rating. All error bars represent standard error. 20 25 30 35 40 45 50 0 5 10 15 20 25 0 1 2 3 4 5 6 7 8 9 10 Temperature ° C Photosynthesis (µmol CO 2 m - 2 s - 1 ) Day Photosynthesis 538 541 553 jag tep temp A 121 Figure 15 (cont'd) 0 0.2 0.4 0.6 0.8 1 1.2 0 1 2 3 4 5 6 7 8 9 10 Conductance (mol H 2 O m - 2 s - 1 ) Day Stomatal conductance 538 541 553 jag tep B 0.3 0.4 0.5 0.6 0.7 0.8 0 1 2 3 4 5 6 7 8 9 10 Photosystem II efficiency Day PSII 538 541 553 jag tep C 122 Figure 15 (cont'd) 0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 8 9 10 Non - Photochemical Quenching Day NPQ 538 541 553 jag tep D 0 50 100 150 200 250 20 25 30 35 40 45 50 TBARS (nmol TBARS / g fresh leaf weight Temp ( ° C) TBARS 538 541 553 jag tep E 123 Figure 15 (cont'd) 0 10 20 30 40 50 60 20 25 30 35 40 45 50 Electrolyte leakage, % total Temperature ( ° C) Electrolyte Leakage 538 541 553 jag tep F 1 2 3 4 5 6 538 541 553 jag tep Visual rating, 1 - 5 Variety Visual rating after extreme stress G a a a b c 124 until the 45 °C period (Figure 15D). At that temperature, all except tepary saw a large increase in NPQ. Tepary was the exception at 45 °C for gas exchange and chlorophyll fluorescence measurements. It had higher values than the other genotypes for photosynthesis, stomatal conductance, and PSII and a lower NPQ at 45 °C. TBARS concentrations were unchanging for all genotypes from 25 - 40 °C, but at 45 °C, TBARS increased for all genotypes ( Figure 15 E). Likewise, electrolyte leakage was stable for most genotypes until 45 °C when values for this parameter more than doubled for most genotypes ( Figure 15 F). I14541 had lo wer TBARS and electrolyte leakage values at 45 °C than the other genotypes. For the final qualitative visual ratings, tepary scored significantly higher than the other genotypes, appearing undamaged on average, and I14541 scored higher than the other thr ee common bean genotypes because it showed fewer signs of d amage from heat stress ( Figure 15 G). I14538, I14553, and Jaguar appeared greatly damaged on average and had similarly low visual scores. Discussion Significance of different measures of stress Heat stress affected most of the measured parameters, but heat stress did not become apparent until the most extreme temperatures tested. For example, photosynthesis was not appreciably reduced by he at stress until 45 °C (Figures 14 A and 15 A). This stands in contrast with a previous study that showed bean plants experiencing significant reductions in photosynthesis at 40 °C (Pimentel et al., 2013) , but different genotypes and environmental 125 conditions were used in that study; most significant ly, it lacked a gradual increase in the ambient air temperature. The decline in photosynthesis at these high temperatures was likely the result of a decrease in the specificity of rubisco for CO 2 compared to O 2 , the deactivation of rubisco, and a decline i n electron transport capacity (Haldimann and Feller, 2004; June et al., 2004; Sharkey, 2005; Sage and Kubien, 2007) . Conversely, stomatal conductance increased with increasing temperature until it shar ply declined at 45 °C (Figures 14 B and 15 B). From 25 - 40 °C, the increase in stomatal conductance could represent a stress resistance response that allows plants to maintain lower leaf temperatures via the transfer of latent heat. It is not likely that CO 2 concentration was limiting photosynthesis under these c onditions, so stomatal conductance increased without a concomitant increase in photosynthesis; usually, the two processes correlate because of their regulation of each other (Wong et al., 1979) . The chlorophyll fluorescence data combined with the gas exch ange data give a more complete picture of the response of the bean plants to increasing temperatures. From 25 to 40 °C, photosynthesis, stomatal conductance, and PSII were all stable and relatively high, indicating that the photosynthetic system was relat ively unstressed. Stomatal conductance was high and so allow ed a greater influx of CO 2 to maintain the high photosynthetic rates, and the high PSII indicated that most of the light energy absorbed was being partitioned towards fixing carbon (Baker, 2008) ; NPQ, a measure of excess energy being dissipated by various mechanisms (Kaiser et al., 2015) , was low. At 45 °C, the photosynthetic systems of most genotypes experienced stress. The stress damaged components of the photosynthetic system and lowered photos ynthetic rates; additionally, less absorbed light energy was being used by the damaged photosynthetic apparatus, so more energy was dissipated or quenched by other mechanisms, causing a decrease in PSII and an increase in NPQ. Finally, because photosynthe sis was lower, 126 the stomata closed more to decrease conductance and match the influx of CO 2 with rate of CO 2 consumption. Leaf canopy temperature showed little variation among genotypes especially when considering the large variation within a genotype ( Fig ure 14 E). Still, at an air temperature of 45 °C and with sufficient water, all genotypes lowered their leaf canopy temperatures to 38 to 41 °C . While plants rarely have canopy temperatures less than the air temperature in field settings, in this experiment , the low radiant energy (400 µ mol photons m - 2 s - 1 ) in the chamber and the well - watered condition of the plants allowed them to attain such low temperatures . In high temperature, high light conditions, many plants require the latent heat loss provided by high stomatal conductances to prevent damage by heat stress (Beerling et al., 2001; Schymanski et al., 2013) . Additionally, plants with smaller leaves have smaller boundary layers and thus greater sensible heat exchange with the surrounding air (Okajima e t al., 2012) . Of the fifteen b ean genotypes tested, tepary had noticeably smaller leaves, and the greater sensible heat loss enabled by such leaves would confer an advantage in hot desert environment s, environments similar to those where tepary originated (Nabhan and Felger, 1978) . Tepary's advantages in photosynthesis and stomatal conductance translate to productive benefits as well; a previous heat stress study showed that common bean's plant dry weight was decreased at elevated temperatures while tepary' s weight was unchanged (Lin and Markhart, 1996) . Considering all genotypes, stomatal conductance and canopy temperature correlated strongly with each other; if a genotype had a higher than average stomatal conductance, it also tended to have a lower than a verage canopy temperature (Table 6) . 127 Insights from the focused experiment The fifteen - genotype - screen indicated that the plants were experiencing heat stress only at 45 °C, so the smaller screen using five genotypes was performed to confirm and more nar rowly identify at what point heat stress occurs. In this more focused experiment ranging from 25 - 45 °C, no measure showed consistent signs of stress until the air temperature reached 45 °C, and this observation was true for measures of gas exchange ( Figure 15 A - B), chlorophyll fluorescence ( Figure 15 C - D), and biochemical damage ( Figure 15 E - F). Duration of exposure is also a factor in heat stress; Udomprasert et al. (1995) found few differences between tepary and common bean when they were exposed to 40 or 45 °C for 6 hours. The current study suggests 24 - 48 hours are necessary to see the effects of heat stress. Some slight differences in a genotype's visual ratin g were seen between the fifteen - genotype - screen and the small er screen experiments (Figures 14 F an d 15 G). These differences likely resulted from growing conditions of each experiment: because of space restrictions, the fifteen genotype experiment was grown to maturity in a greenhouse setting and only spent one week in a growth chamber, but the smaller experiment was grown to maturity in the growth chambers themselves. However, this change only made I14538 have a slightly lower visual rating and I14541 have a higher one in the smaller experiment. Also of interest in the smaller experiment was that tepa ry's photosynthesis, stomatal conductance, PSII , and NPQ were least affected by 45 °C temperatures ( Figure 15 A - D) while I14541's TBARS and electrolyte leakage were least affected at this temperature ( Figure 15 E - F), and both genotypes looked the least damaged and had the highest visual ratings at the end of the experiment ( Figure 15 G). These results suggest two different heat tolerance strategies. Tepary's heat - capable photosynthetic machinery allowed it withstand high temperatures and shunt most of 128 its absorbed energy towards photosynthesis while I14541, despite having more a heat - compromised photosynthesis, was able to regulate reactive oxygen signaling and prevent damage to its cellular membranes, whose fluidity, permeability, and interactions with pr oteins are all affected by heat stress (Havaux, 1996; Pastenes and Horton, 1996; Bukhov et al., 1999) . Genotype performance in all measures of stress For the fifteen - genotype - screen , some parameters correlated with each other in mechanistically expected ways : s tomatal conductance and canopy temperature were negatively correlated with each other, PSII and NPQ were negatively correlated with each other, and PSII and visual rating were positively correlated with each othe r (Table 6). When examined individu ally, some genotypes had interesting trends across all parameters. Using genotype I14550 as an example, if the genotypes were ranked, it always placed in the bottom half, often the bottom fifth, for all measurements at 45 °C : it had low photosynthesis, sto matal conductance, and PSII , and it had high NPQ and canopy temperature. Consequently, it showed signs of damage to its leaves and had a low visual rating as well. Conversely, tepary ranked high for all of these measurements and had a high visual rating. Some exceptions do exist: SER - 95 had low photosynthesis and conductance but high PSII and low NPQ, and it ended up having a high score for visual rating. The qualitative visual rating was an excellent indicator of leaf damage caused by heat stress and cor related well with other measures (Table 6) . However, these parameters did not have a strong connection to a genotype's level of drought tolerance in the field. Drought tolerant genotypes such as tepary, SER - 95, I14538, and I14549 did have the highest vis ual ratings of the fifteen genotypes, but then drought tolerant genotypes such as I14553 and I14544 ranked relatively low in terms of their visual rating ( Figure 129 14 F). This weaker connection could have been the result of a variety of factors. For all the g enotypes labeled as a number, their characterization comes from a single drought field test performed in Mayaguez, Puerto Rico (Tim Porch, personal communication). Under both drought and nonstress conditions, common bean yield can vary greatly in different years and locations (Terán and Singh, 2002 b ; Muñoz - Perea et al., 2006; Beebe et al., 2008; Builes et al., 2011) , so further field tests in different locations and years could create a more accurate characterization of each genotype's drought tolerance tha t escapes the current study that relies on a single field trial. The other factor contributing to the discrepancy is that heat and drought stress are two distinct abiotic stresses with different response pathways (Mittler, 2006) , so it is possible for a genotype to be tolerant of one and not the other. While heat and drought stress share a number of transcriptional response and redox - signaling pathways (Swindell et al., 2007; Foyer and Noctor, 2009) , much like any stress, heat an d drought cause distinctly different changes in the transcriptome and metabolome (Shulaev et al., 2008; Rasmussen et al., 2013) , and the combination of heat and drought stress produce a third, unique response as well (Rizhsky et al., 2002, 2004) . Heat, dro ught, and the combination of the two share relatively few genes in their response pathways compared to the number they do not share (Rizhsky et al., 2004) . Thus, depending on the alleles found in each pathway, it is possible for a genotype to be resistant to only drought stress, only heat stress, or both. Some of the genotypes that performed well under the drought - stressed field trial may have performed poorly in this heat stress screen simply because they are drought tolerant but not heat tolerant. The hea t - screening method was useful in characterizing germplasm, especially in identifying some of the most tolerant genotypes, but it should not be relied upon exclusively and its results should be considered alongside the results of several other methods. Alth ough based on the high coincidence of heat and drought stress in 130 several bean production areas, breeders may not want a drought tolerant cultivar if it cannot also withstand heat (Araújo et al., 2014) , so a consideration of combined stresses may make the h eat screening method a more attractive tool. Conclusions Under these controlled conditions, the bean genotypes did not experience significant heat stress until air temperatures were increased to 45 °C. This temperature was sufficient to affect photosynthesis, stomatal conductance, PSII , NPQ, electrolyte leakage, and TBARS and to cause visible damage to plant leaves. For a single genotype, the degree of change in one of these measures was ofte n well - correlated with the degree of change in another. Because of this correlation, any or all of these measurements would be appropriate in other screens of germplasm using heat stress. Even an experimenter limited by time or available equipment could us e one or two measures and still perform a meaningful screen. The stress tolerant, related species tepary bean outperformed all common bean genotypes for most measures of heat tolerance, and known drought - tolerant common bean genotype SER - 95 performed bette r than most other genotypes as well. As would be expected, the correlation between drought tolerance in the field and heat tolerance under controlled conditions was not perfect because great differences exist between the two environments and drought and he at tolerance share only some of their pathways; however, the screen still identified the more tolerant genotypes with few false positives. Used in conjunction wi th field data and other tests, heat screening is a simple and quick method of characterizing la rge groups of germplasm. 131 Chapter 5: Conclusions and recommendations 132 The research of this dissertation adds to an understanding of stress - related physiology in common bean and t e pary bean. While proline had little response to drought st ress in beans and likely plays no role in acclimation to drought, certain organic acids and sugars such as malic acid, glucose, fructose, inositol, and raffinose significantly increase d in response to drought stress. These soluble carbohydrates accumulate d in sufficient quantities to contribute to the osmotic adjust ment of plant tissues, and drought tolerant bean genotypes accumulate d a higher concentration of these compounds than susceptible genotypes. Supporting these results, the same genotypes with higher concentrations of soluble carbohydrates also had lower leaf water potentials under drought stress, but no significant differences in leaf water potential were found among genotypes in the control tre atment. The differences among genotypes were the result of drought stress and were not cons titutive, for all genotypes had similar concentrations of soluble carbohydrates under control conditions. Whether as osmolytes, osmoprotectants, or as signaling fact ors, these soluble carbohydrates are an important part of stress response in beans. While measuring individual sugar concentrations would be onerous for a plant breeder who has to evaluate large quantities of germplasm , it could be effectively used on a sm aller scale to evaluate a smaller group of genotypes being considered as parents in future crosses. Furthermore, in field trials that experience a period of drought stress, leaf water potential measurements could be taken on a moderate number of genotypes, perhaps those of special interest, to test for the degree of osmotic adjustment and thus indirectly for accumulation of sugars. Despite being an integral stress - signaling hormone, abscisic acid (ABA) concentrations under drought stress had little correla tion with drought tolerance . Under the most severe drought stress, the most susceptible genotype had the highest concentration of ABA in leaf tissue. Such high ABA concentrations probably do not result in drought susceptibility per se , but rather they 133 refl ect the level of stress a plant is experiencing, so high ABA concentrations in bean plants result from the failure of other mechanisms to mitigate the effects of drought stress. ABA is one of the prime regulators of stomatal aperture, and the gas exchange data from stressed plants revealed characteristics of drought tolerance in beans. Drought tolerant genotypes had lower stomatal conductances under both control and drought stressed conditions and consequently tended to have lower photosynthetic rates as we ll. Factors downstream of ABA signaling could be more sensitive to ABA and thus result in lower stomatal conductances in tolerant genotypes, even when their concentrations of ABA are similar to susceptible genotypes. The exogenous ABA experiment showed tha t only one of the two tolerant genotypes displayed increased sensitivity, so other factors determining stomatal conductance must exist in beans. That tolerant genotypes also had lower stomatal conductances under control conditions, when ABA is not a contri buting factor, supports t he existence of other mechanisms, mechanisms which are perhaps constitutive. Based on grafting experiments, whatever mechanisms that do control stomatal conductance are located in shoot tissues and not root tissues. A following exp eriment found stomatal density is not the factor behind differences in stomatal conductance as two genotypes with large contrasts in phylogeny and drought tolerance had the same density of stomata. Future research should investigate further the nature of t he differences in stomatal conductance between susceptible and tolerant genotypes, for it seems to be key to drought tolerance in beans. Breeding for reduced stomatal conductances should be an easier goal than some because it is a constitutive trait that d oes not require drought stress to be measureable. Of course, the tradeoff between lower conductances and lower photosynthetic rates must also be considered , and extreme selection for this trait could result in reduced productivity. Based on this tradeoff a nd the time required to take measurements, selecting for stomatal conductance should be done at more 134 advanced stages of the breeding cycle when productivity is already established and there are fewer lines to evaluate. When doing heat stress experiments, it is essential that sufficiently high temperatures are used and that duration of exposure is long enough to induce stress . For the bean genotypes used in the heat screening experiment, 45 °C for two days served to induce measureable stress effects. The di fferent measures of physiology and stress correlated well with each other , so any one of them could be used in similar screening experiments. Overall, the genotypes' heat tolerance and drought tolerance were not perfectly correlated with each other, but it is interesting to note that the genotypes that displayed the greatest heat tolerance in the screen were among the most drought tolerant in the field. Heat and drought stress are often coincident in field settings, so having some degree of heat tolerance c ontributes to better tolerance of most drought events as well. Additionally, now that the basic heat screening method described in Chapter 4 was shown to have some utility in beans, researchers can conduct experiments to tweak the method to screen for the more technically challenging stress of drought or even combined drought and heat stress. Breeders can then use these screening methods in the off - season to add an extra layer of characterization to lines also being evalu ated in the field. 135 BIBLIOGRAPHY 136 BIBLIOGRAPHY Acosta - Gallegos J, White JW . 1995. Phenological plasticity as an adaptation by common bean to rainfed e nvironments. Crop Science 35 , 199 204. Akibode S, Maredia M . 2011. Global and Regional Trends in Produ ction, Trade and Consumption of Food Legume Crops. CGIAR. Allen DJ, Ort DR . 2001. Impacts of chilling temperatures on photosynthesis in warm - climate plants. Trends in Plant Science 6 , 36 42. Araújo SS, Beebe S, Crespi M, et al. 2014. Abiotic stress responses in legumes: strategies used to cope with e nvironmental c hallenges. Critical Reviews in Plant Sciences 34 , 237 280. Araus JL, Slafer GA, Reynolds MP, Royo C . 2002. Plant breeding and drough t in C3 cereals: what should we breed f or? Annals of Botany 89 , 925 940. Asada K . 2006. Production and scavenging of reactive oxygen species in chloroplasts and their functions. Plant Physiology 141 , 391 396. Asch F, Dörffling K, Dingkuhn M . 1995. Response of rice varieties to soil salinity and air humidity: A possible involvement of root - borne ABA. Plant and Soil 177 , 11 19. Asfaw A, Blair M . 2012. Quantitative trait loci for rooting pattern traits of common beans grown under drought stress versus non - stre ss conditions. Molecular Breeding 30 , 681 695. Baena - Gonzalez E, Rolland F, Thevelein JM, Sheen J . 2007. A central integrator of transcription networks in plant stress and energy signalling. Nature 448 , 938 942. Bahieldin A, Mahfouz HT, Eissa HF, Saleh OM, Ramadan AM, Ahmed IA, Dyer WE, El - Itriby HA, Madkour MA . 2005. Field evaluation of transgenic wheat plants stably expressing the HVA1 gene for drought tolerance. Physiologia Plantarum 123 , 421 427. Bailey NJC, Oven M, Holmes E, Nicholson JK, Zenk MH . 2003 . Metabolomic analysis of the consequences of cadmium exposure in Silene cucubalus cell cultures via H - 1 NMR spectroscopy and chemometrics. Phytochemistry 62 , 851 858. Baker NR . 2008. Chlorophyll fluorescence: A probe of photosynthesis in vivo. Annual Revi ew of Plant Biology 59 , 89 113. Barchet GLH, Dauwe R, Guy RD, Schroeder WR, Soolanayakanahally RY, Campbell MM, Mansfield SD . 2014. Investigating the drought - stress response of hybrid poplar genotypes by metabolite profiling. Tree Physiology 34 , 1203 1219. Bates L, Waldren R, Teare I . 1973. Rapid determination of free proline for water stress studies. Plant and Soil 39 , 205 207. 137 Battisti DS, Naylor RL . 2009. Historical warnings of future f ood insecurity with unprecedented seasonal h eat. Science 323 , 240 244. Beebe S . 2012. Common Bean Breeding in the Tropics. Plant Breeding Reviews. John Wiley & Sons, Inc., 357 426. Beebe S, Ramirez J, Jarvis A, Rao IM, Mosquera G, Bueno JM, Blair MW . 2011. Genetic Improvement of Common Beans and the C hallenges of Cli mate Change. Yadav SS, Redden RJ, Hatfield JL, Lotze - Campen H, Hall AE, eds. Crop Adaptation to Climate Change. Wiley - Blackwell, 356 369. Beebe S, Rao I, Blair M, Acosta J . 2013. Phenotyping common beans for adaptation to drought. Frontiers in Physiology 4 , 347 - 362 . Beebe SE, Rao IM, Cajiao C, Grajales M . 2008. Selection for drought resistance in common bean also improves yield in phosphorus limited and favorable environments. Crop Science 48 , 582 592. Beebe S, Rao I, Mukankusi C, Buruchara R . 2012. Improvi ng resource use efficiency and reducing risk of common bean production in A frica, Latin America, and the Carribean. Eco - Efficienc y: From Vision to Reality. CIAT . Beerling DJ, Osborne CP, Chaloner WG . 2001. Evolution of leaf - form in land plants linked to at mospheric CO2 decline in the Late Palaeozoic era. Nature 410 , 352 354. Bielenberg D, Miller J, Berg V . 2003. Paraheliotropism in two Phaseolus species: combined effects of photon flux density and pulvinus temperature, and consequences for leaf gas exchange . Environmental and Experimental Botany 49 , 95 105. Bitocchi E, Nanni L, Bellucci E, et al. 2012. Mesoamerican origin of the common bean (Phaseolus vulgaris L.) is revealed by sequence data. Proceedings of the National Academy of Sciences 109 , E788 E796. B lair M, Fernandez A, Ishitani M, Moreta D, Seki M, Ayling S, Shinozaki K . 2011. Construction and EST sequencing of full - length, drought stress cDNA libraries for common beans (Phaseolus vulgaris L.). BMC Plant Biology 11 , 171. Blair MW, Galeano CH, Tovar E , Munoz Torres MC, Velasco Castrillon A, Beebe SE, Rao IM . 2012. Development of a Mesoamerican intra - genepool genetic map for quantitative trait loci detection in a drought tolerant x susceptible common bean (Phaseolus vulgaris L.) cross. Molecular Breedin g 29 , 71 88. Blair, MW, Soler, A, Cortés, AJ . 2012. Diversification and population structure in common b eans (Phaseolus vulgaris L.). PLoS ONE 7 , e49488. Blatt M, Armstrong F . 1993. K+ channels of stomatal guard cells: abscisic acid evoked control of the o utward rectifier mediated by cytoplasmic pH. Planta 191 , 330 341. 138 Blum A . 2011. Breeding Considerations and Strategies . Boutraa T, Sanders FE . 2001. Influence of water stress on grain yield and vegetative growth of two cultivars of b ean (Phaseolus vulgaris L.). Journal of Agronomy & Crop Science 187 , 251. Bowes G, Ogren WL, Hageman RH . 1971. Phosphoglycolate production catalyzed by ribulose diphosphate carboxylase. Biochemical and Biophysical Research Communications 45 , 716 722. Boyer JS, Westgate ME . 2004. Grain yields with limited water. Journal of Experimental Botany 55 , 2385 2394. Boyle MG, Boyer JS, Morgan PW . 1991. Stem infusion of liquid culture medium prevents reproductive failure of maize at low water p otential. Crop Science 31 , 1246 1252. Broughton WJ, Hernández G, Blair M, Beebe S, Gepts P, Vanderleyden J . 2003. Beans (Phaseolus spp.) model food legumes. Plant and Soil 252 , 55 128. Buckley TN, Mott KA . 2013. Modelling stomatal conductance in response to environmental facto rs. Plant Cell and Environment 36 , 1691 1699. Builes VHR, Porch TG, Harmsen EW . 2011. Genotypic differences in water use efficiency of common bean under drought s tress. Agronomy Journal 103 , 1206 1215. Bukhov N, Wiese C, Neimanis S, Heber U . 1999. Heat sen sitivi ty of chloroplasts and leaves: l eakage of protons from thylakoids and reversible activation of cyclic electron transport. Photosynthesis Research 59 , 81 93. Butare L, Rao I, Lepoivre P, Polania J, Cajiao C, Cuasquer J, Beebe S . 2011. New genetic sour ces of resistance in the genus Phaseolus to individual and combined aluminium toxicity and progressive soil drying stresses. Euphytica 181 , 385 404. von Caemmerer S, Evans JR . 2014. Temperature responses of mesophyll conductance differ greatly between spec ies. Plant, Cell & Environment, n/a n/a. Carrari F, Fernie AR, Iusem ND . 2004. Heard it through the grapevine? ABA and sugar cross - talk: the ASR story. Trends in Plant Science 9 , 57 59. Casati P, Drincovich M, Edwards G, Andreo C . 1999. Malate metabolism b y NADP - malic enzyme in plant defense. Photosynthesis Research 61 , 99 105. Castonguay Y, Markhart A . 1991. Saturated rates of photosynthesis in water - stressed leaves of common bean and tepary bean. Crop Science 31 , 1605 1611. Castonguay Y, Markhart A . 1992. Leaf gas - exchange in water - stressed common bean and tepary bean. Crop Science 32 , 980 986. 139 Cattivelli L, Rizza F, Badeck F - W, Mazzucotelli E, Mastrangelo AM, Francia E, Mare C, Tondelli A, Stanca AM . 2008. Drought tolerance improvement in crop plants: An integrated view from breeding to genomics. Field Crops Research 105 , 1 14. Cavalieri A, Merchant A, van Volkenburgh E . 2011. Why not beans? Foreword. Functional Plant Biology 38 , III VI. Centritto M, Loreto F, Chartzoulakis K . 2003. The use of low [CO2] to estimate diffusional and non - diffusional limitations of photosynthetic capacity of salt - stressed olive saplings. Plant, Cell & Environment 26 , 585 594. CGIAR . 2015. Developing Beans that Can Beat the Heat. Chaves M . 1991. Effects of water deficits on carbon a ssimilation. Journal of Experimental Botany 42 , 1 16. Chaves M, Maroco J, Pereira J . 2003. Understanding plant responses to drought - from genes to the whole plant. Functional Plant Biology 30 , 239 264. Chaves MM, Oliveira MM . 2004. Mechanisms unde rlying plant resilience to water deficits: prospects for water - saving agriculture. Journal of Experimental Botany 55 , 2365 2384. Chen THH, Murata N . 2008. Glycinebetaine: an effective protectant against abiotic stress in plants. Trends in Plant Science 13 , 499 505. Christmann A, Weiler EW, Steudle E, Grill E . 2007. A hydraulic signal in root - to - shoot signalling of water shortage. The Plant Journal 52 , 167 174. Cornic G, Le Gouallec J - L, Briantais JM, Hodges M . 1989. Effect of dehydration and high light on p hotosynthesis of two C3 plants (Phaseolus vulgaris L. and Elatostema repens (Lour.) Hall f.). Planta 177 , 84 90. Cortés, AJ, Monserrate, FA, Ramírez - Villegas, J, Madriñán, S, Blair, MW . 2013. Drought tolerance in wild plant populations: the case of common b eans (Phaseolus vulgaris L.). PLoS ONE 8 , e62898. Cutler A, Krochko J . 1999. Formation and breakdown of ABA. Trends in Plant Science 4 , 472 478. Cutler SR, Rodriguez PL, Finkelstein RR, Abrams SR . 2010. Abscisic acid: emergence of a core signaling n etwork . Merchant, S and Briggs, WR and Ort, D ed s . Annual Review of Plant Biology. Annual Review of Plant Biology 61 , 651 679. Dalal M, Inupakutika M . 2014. Transcriptional regulation of ABA core signaling component genes in sorghum (Sorghum bicolor L. Moench). Molecular Breeding 34 , 1517 1525. Damour G, Simonneau T, Cochard H, Urban L . 2010. An overview of models of stomatal conductance at the leaf level. Plant Cell and Environment 33 , 1419 1438. 140 Daszkowska - Golec A, Szarejko I . 2013. Open or close the gate sto mata action under the control of phytohormones in drought stress conditions. Frontiers in Plant Science 4 . Delfine S, Alvino A, Villani MC, Loreto F . 1999. Restrictions to carbon dioxide conductance and photosynthesis in spinach leaves r ecovering from salt s tress. Plant Physiology 119 , 1101 1106. Demmig - Adams B, Adams W . 1996. The role of xanthopyll cycle carotenoids in the protection of photosynthesis. Trends in Plant Science 1 , 21 26. Desikan R, Griffiths R, Hancock J, Neill S . 2002. A new role for an old enzyme: n itrate reductase - mediated nitric oxide generation is required for abscisic acid - induced stomatal closure in Arabidopsis thaliana. Proceedings of the National Academy of Sciences 99 , 16314 16318. Devi MJ, Sinclair TR, Beebe SE, Rao IM . 2013. C omparison of common bean (Phaseolus vulgaris L.) genotypes for nitrogen fixation tolerance to soil drying. Plant and Soil 364 , 29 37. . 2011. What can enzymes of C4 photosynthesis do for C3 plants under stress? Plant Science 180 , 575 583. Drake PL, Froend RH, Franks PJ . 2012. Smaller, faster stomata: scaling of stomatal size, rate of response, and stomatal conductance. Journal of Experimental Botany 64 , 549 - 505. Easlon HM, Bloom AJ . 2014. Easy Leaf Area: Automated Digital Image Analys is for Rapid and Accurate Measurement of Leaf Area. Applications in Plant Sciences 2 , 421 - 428 . Ebell LF . 1969. Specific total starch determinations in conifer tissues with glucose oxidase. Phytochemistry 8 , 25 36. Edwards S, Nguyen B - T, Do B, Roberts JKM . 1998. Contribution of malic enzyme, p yruvate k inase, p hosphoenolpyruvate c arboxylase, and the k rebs c ycle to r espiration and b iosynthesis and to i ntracellular pH r egulation during h ypoxia in m aize r oot t ips o bserved by n uclear m agnetic r esonance i maging and g as c hromatography - m ass s pectrometry. Plant Physiology 116 , 1073 1081. Ehleringer J, Klassen S, Clayton C, Sherril D, Fullerholbrook M, Fu Q, Cooper T . 1991. Carbon isotope discrimination and transpiration efficiency in common bean. Crop Science 31 , 1611 1615. Ehleringer J, Pearcy RW . 1983. Variation in q uantum y ield for CO u ptake among C and C p lants. Plant Physiology 73 , 555 559. Escobar - Gutierrez A, Zipperlin B, Carbonne F, Moing A, Gaudillere J . 1998. Photosynthesis, carbon partitioning an d metabolite content during drought stress in peach seedlings. Australian Journal of Plant Physiology 25 , 197 205. 141 Evans J, Sharkey T, Berry J, Farquhar G . 1986. Carbon isotope discrimination measured concurrently with gas exchange to investigate CO2 diffu sion in leaves of higher plants. Australian Journal of Plant Physiology 13 , 281 292. FAOSTAT . 2014. FAOSTAT. Fischer RA, Rees D, Sayre KD, Lu Z - M, Condon AG, Saavedra AL . 1998. Wheat y ield p rogress a ssociated with h igher s tomatal c onductance and p hotosynth etic r ate, and c ooler c anopies. Crop Science 38 , 1467 1475. Fletcher RJ, Bell IP, Lambert JP . 2004. Public health aspects of food fortification: a question of balance. Proceedings of the Nutrition Society 63 , 605 614. Flexas J, Bota J, Escalona JM, Sampol B, Medrano H . 2002. Effects of drought on photosynthesis in grapevines under field conditions: an evaluation of stomatal and mesophyll limitations. Functional Plant Biology 29 , 461 471. Flexas J, Bota J, Loreto F, Cornic G, Sharkey T . 2004. Diffusive and m etabolic limitations to photosynthesis un der drought and salinity in C3 plants. Plant Biology 6 , 269 279. Foyer CH, Noctor G . 2009. Redox r egulation in p hotosynthetic o rganisms: s ignaling, a cclimation, and p ractical i mplications. Antiooxidants & Redox Sign aling 11 , 861 905. Foyer CH, Noctor G . 2011. Ascorbate and g lutathione: t he h eart of the r edox h ub. Plant Physiology 155 , 2 18. Frahm M, Rosas J, Mayek - Perez N, Lopez - Salinas E, Acosta - Gallegos J, Kelly J . 2004. Breeding beans for resistance to terminal dr ought in the lowland tropics. Euphytica 136 , 223 232. Frébort I, Kowalska M, Hluska T, Frébortová J, Galuszka P . 2011. Evolution of cytokinin biosynthesis and degradation. Journal of Experimental Botany 62 , 2431 2452. Frensch J, Hsiao TC . 1994. Transient r esponses of c ell t urgor and g rowth of m aize Roots as a ffected by c hanges in w ater p otential. Plant Physiology 104 , 247 254. Gagneul D, Aienouche A, Duhaze C, Lugan R, Larher FR, Bouchereau A . 2007. A reassessment of the function of the so - called compatible solutes in the halophytic Plumbaginaceae Limonium latifolium. Plant Physiology 144 , 1598 1611. Garcia A, Engler J, Iyer S, Gerats T, VanMontagu M, Caplan A . 1997. Effects of osmoprotectants upon NaCl stress in rice. Plant Physiology 115 , 159 169. Geigenberger P, Reimholz R, Geiger M, Merlo L, Canale V, Stitt M . 1997. Regulation of sucrose and starch metabolism in potato tubers in response to short - term water deficit. Planta 201 , 502 518. 142 Geitmann A, Ortega JKE . 2009. Mechanics and modeling of plant cell growth. Trends in Plant Science 14 , 467 478. Gengenbach BG . 1977. Development of maize caryopses resulting from in - vitro pollination. Planta 134 , 91 93. Gepts P, Bliss FA . 1985. F1 hybrid weakness in the common bean: Differential geographic origin su ggets two gene pools in cultivated bean germplasm. Journal of Heredity 76 , 447 450. Graham RD, Welch RM, Saunders DA, et al. 2007. Nutritious s ubsistence f ood s ystems. Donald L. Sparks ed. Advances in Agronomy. Academic Press, 1 74. Gross Y, Kigel J . 1994. Differential sensitivity to high temperature of stages in the reproductive development of common bean (Phaseolus vulgaris L.). Field Crops Research 36 , 201 212. Guo P, Baum M, Grando S, Ceccarelli S, Bai G, Li R, von Korff M, Varshney RK, Graner A, Valkoun J . 2009. Differentially expressed genes between drought - tolerant and drought - sensitive barley genotypes in response to drought stress during the reproductive stage. Journal of Experimental Botany 60 , 3531 3544. Gupta A, Kaur N . 2005. Sugar signall ing and gene expression in relation to carbohydrate metabolism under abiotic stresses in plants. Journal of Biosciences 30 , 761 776. Haldimann P, Feller U . 2004. Inhibition of photosynthesis by high temperature in oak (Quercus pubescens L.) leaves grown un der natural conditions closely correlates with a reversible heat - dependent reduction of the activation state of ribulose - 1,5 - bisphosphate carboxylase/oxygenase. Plant, Cell & Environment 27 , 1169 1183. Harb A, Krishnan A, Ambavaram MMR, Pereira A . 2010. Mo lecular and p hysiological a nalysis of d rought s tress in A rabidopsis r eveals e arly r esponses l eading to a cclimation in p lant g rowth. Plant Physiology 154 , 1254 1271. Hare PD, Cress WA . 1997. Metabolic implications of stress - induced proline accumulation in plants. Plant Growth Regulation 21 , 79 102. Hare PD, Cress WA, Van Staden J . 1998. Dissecting the roles of osmolyte accumulation during stress. Plant, Cell & Environment 21 , 535 553. Hartung W, Sauter A, Hose E . 2002. Abscisic acid in the xylem: where does it come from, where does it go to? Journal of Experimental Botany 53 , 27 32. Hauser F, Waadt R, Schroeder JI . 2011. Evolution of a bscisic a cid s ynthesis and s ignaling m echanisms. Cur rent Biology 21 , R346 R355. Havaux M . 1992. Stress t olerance of p hotosystem II in v ivo: a ntagonistic effects of water, heat, and photoinhibition s tresses. Plant Physiology 100 , 424 432. 143 Havaux M . 1996. Short - term responses of p hotosystem I to heat stress. Photosynthesis Research 47 , 85 97. Héroult A, Lin Y - S, Bourne A, Medlyn B, Ellsworth D . 2013. Optimal stomatal conductance in relation to photosynthesis in climatically contrasting Eucalyptus species under drought. Hetherington AM, Woodward FI . 2003. The r ole of stomata in sensing and driving environmental change. Nature 424 , 901 908. Hoekstra FA, Golovina EA, Buitink J . 2001. Mechanisms of plant desiccation tolerance. Trends in Plant Science 6 , 431 438. Holbrook N, Shashidhar V, James R, Munns R . 2002. Sto matal control in tomato with ABA - deficient roots: response of grafted plants to soil drying. Journal of Experimental Botany 53 , 1503 1514. Hommel R, Siegwolf R, Saurer M, Farquhar GD, Kayler Z, Ferrio JP, Gessler A . 2014. Drought response of mesophyll cond uctance in forest understory species impacts on water - use efficiency and interactions with leaf water movement. Physiologia Plantarum 152 , 98 114. Hoque MA, Banu MNA, Okuma E, Murata Y . 2007. Exogenous proline and glycinebetaine increase NaCl - induced asc orbate - glutathione cycle enzyme activities, and proline improves salt tolerance more than glycinebetaine in tobacco Bright Yellow - 2 suspension - cultured cells. Journal of Plant Physiology 164 , 1457 1468. Hossain MA, Munemasa S, Uraji M, Nakamura Y, Mori I C, Murata Y . 2011. Involvement of endogenous abscisic acid in methyl jasmonate - induced stomatal c losure in Arabidopsis. Plant Physiology 156 , 430 438. Ingram J, Bartels D . 1996. The molecular basis of dehydration tolerance in plants. Annual Review of Plant Physiology and Plant Molecular Biology 47 , 377 403. Izquierdo J, Hosfield G . 1983. The relationship of seed filling to yield among dry beans with differing architectural forms. Journal of the American Society for Horticultural Science 108 , 106 111. Ji X, Dong B, Shiran B, Talbot MJ, Edlington JE, Hughes T, White RG, Gubler F, Dolferus R . 2011. Control of abscisic acid catabolism and abscisic acid homeostasis is i mportant for reproductive stage stress tolerance in c ereals. Plant Physiology 156 , 647 662. Jordan D, Ogren W . 1984. The CO2/O2 specificity of ribulose 1,5 - bisphosphate carboxylase/oxygenase. Planta 161 , 308 313. June T, Evans JR, Farquhar GD . 2004. A simple new eq uation for the reversible temperature dependence of photosynthetic electron transport: a study on soybean leaf. Functional Plant Biology 31 , 275 283. 144 Kaiser E, Morales A, Harbinson J, Kromdijk J, Heuvelink E, Marcelis LFM . 2015. Dynamic photosynthesis in d ifferent environmental conditions. Journal of Experimental Botany 66 , 2415 2426. Kakumanu A, Ambavaram MMR, Klumas C, Krishnan A, Batlang U, Myers E, Grene R, Pereira A . 2012. Effects of drought on gene expression in maize reproductive and leaf meristem ti ssue revealed by RNA - s eq. Plant Physiology 160 , 846 867. Kanno Y, Jikumaru Y, Hanada A, Nambara E, Abrams SR, Kamiya Y, Seo M . 2010. Comprehensive hormone profiling in developing Arabidopsis seeds: e xamination of the site of ABA biosynthesis, ABA transport and hormone i nteractions. Plant and Cell Physiology 51 , 1988 2001. Kebeish R, Niessen M, Thiruveedhi K, Bari R, Hirsch H - J, Rosenkranz R, Stabler N, Schonfeld B, Kreuzaler F, Peterhansel C . 2007. Chloroplastic photorespiratory bypass increases photosynthe sis and biomass production in Arabidopsis thaliana. Nat Biotech 25 , 593 599. Keller J, Loescher W . 1989. Nonstructural carbohydrate partitioning in perennial parts of sweet cherry. Journal of the American Society for Horticultural Science 114 , 969 975. Kelly JD, Hosfield GL, Varner GV, Uebersax MA, Taylor J . black b ean. Crop Science 41 , 1647 1648. Kelly J, Miklas P . 1998. The role of RAPD markers in breeding for disease resistance in common bean. Molecular Breeding 4 , 1 11. . 2009 a ean. Journal of Plant Registrations 3 , 226 230. Kelly JD, Varner GV, Roman B, Long B . 2009 b ean. Journal of Plant Registrations 3 , 223 225. Kilian - Bauer E, Kudla J, Harter K . 2007. The AtGenExpress global stress expression data set: protocols, evaluation and model data analysis of UV - B light, drought and cold stress response s. The Plant Journal 50 , 347 363. Koch K . 2004. Sucrose metabolism: regulatory mechanisms and pivotal roles in sugar sensing and plant development. Current Opinion in Plant Biology 7 , 235 246. Kornyeyev D, Logan BA, Allen RD, Holaday AS . 2005. Field - grown cotton plants with elevated activity of chloroplastic glutathione reductase exhibit no significant alteration of diurnal or seasonal patterns of excitation energy partitioning and CO2 fixation. Field Crops Research 94 , 165 175. Kornyeyev D, Logan BA, Payto n P, Allen RD, Holaday AS . 2001. Enhanced photochemical light utilization and decreased chilling - induced photoinhibition of photosystem II in cotton 145 overexpressing genes encoding chloroplast - targeted antioxidant enzymes. Physiologia Plantarum 113 , 323 331. Kornyeyev D, Logan BA, Payton PR, Allen RD, Holaday AS . 2003. Elevated chloroplastic glutathione reductase activities decrease chilling - induced photoinhibition by increasing rates of photochemistry, but not thermal energy dissipation, in transgenic cotton . Functional Plant Biology 30 , 101 110. Koussevitzky S, Suzuki N, Huntington S, Armijo L, Sha W, Cortes D, Shulaev V, Mittler R . 2008. Ascorbate p eroxidase 1 p lays a k ey r ole in the r esponse of Arabidopsis thaliana to s tress c ombination. Journal of Biologi cal Chemistry 283 , 34197 34203. Kozaki A, Takeba G . 1996. Photorespiration protects C3 plants from photooxidation. Nature 384 , 557 560. Kudoyarova G, Veselova S, Hartung W, Farhutdinov R, Veselov D, Sharipova G . 2011. Involvement of root ABA and hydraulic conductivity in the control of water relations in wheat plants exposed to increased evaporative demand. Planta 233 , 87 94. Levitt J . 1972. Responses of plants to environmental stresses . New York: Academic Press. Lin T, Markhart A . 1996. Phaseolus acutifoli us A. Gray is more heat tolerant than P - vulgaris L in the absence of water stress. Crop Science 36 , 110 114. Liu S, Cheng Y, Zhang X, Guan Q, Nishiuchi S, Hase K, Takano T . 2007. Expression of an NADP - malic enzyme gene in rice (Oryza sativa. L) is induced by environmental stresses; over - expression of the gene in Arabidopsis confers salt and osmotic stress tolerance. Plant Molecular Biology 64 , 49 58. Liu S, Yu K, Park SJ . 2009. Marker - assisted breeding for resistance to common ba cterial blight in common bean. Huttunen N , Sinisalo T, eds. Plant Breeding. 211 226. Lizana C, Wentworth M, Martinez J, et al. 2006. Differential adaptation of two varieties of common bean to abiotic stress - I. Effects of drought on yield and photosynthes is. Journal of Experimental Botany 57 , 685 697. Lobell DB, Gourdji SM . 2012. The i nfluence of c limate c hange on g lobal c rop p roductivity. Plant Physiology 160 , 1686 1697. Locke AM, Ort DR . 2014. Leaf hydraulic conductance declines in coordination with phot osynthesis, transpiration and leaf water status as soybean leaves age regardless of soil moisture. Journal of Experimental Botany 65 , 6617 6627. Loewus FA, Murthy PPN . 2000. myo - Inositol metabolism in plants. Plant Science 150 , 1 19. Long S, Bernacchi C . 2 003. Gas exchange measurements, what can they tell us about the underlying limitations to photosynthesis? Procedures and sources of error. Journal of Experimental Botany 54 , 2393 2401. 146 Long S, Zhu X, Naidu S, Ort D . 2006. Can improvement in photosynthesis increase crop yields? Plant Cell and Environment 29 , 315 330. MacRobbie EAC . 1998. Signal transduction and ion channels in guard cells. Philosophical Transactions of the Royal Society of London B: Biological Sciences 353 , 1475 1488. Markhart AH III . 1985. Comparative w ater r elations of Phaseolus vulgaris L. a nd Phaseolus acutifolius Gray. Plant Physiology 77 , 113 117. Ma Y, Szostkiewicz I, Korte A, Moes D, Yang Y, Christmann A, Grill E . 2009. Regulators of PP2C p hosphatase a ctivity f unction as a bscisic a cid s ensors. Science 324 , 1064 1068. McAinsh MR, Brownlee C, Hetherington AM . 1992. Visualizing c hanges in c ytosolic - f ree Ca2+ during the r esponse of s tomatal g uard c ells to a bscisic a cid. The Plant Cell 4 , 1113 1122. McClean PE, Burridge J, Beebe S, Rao I M, Porch TG . 2011. Crop improvement in the era of climate change: an integrated, multi - disciplinary approach for common bean (Phaseolus vulgaris). Functional Plant Biology 38 , 927 933. Medrano H, Escalona J, Bota J, Gulias J, Flexas J . 2002. Regulation of photosynthesis of C - 3 plants in response to progressive drought: Stomatal conductance as a reference parameter. Annals of Botany 89 , 895 905. Mejía - Jiménez A, Muñoz C, Jacobsen HJ, Roca WM, Singh SP . 1994. Interspecific hybridization between common and tep ary beans: increased hybrid embryo growth, fertility, and efficiency of hybridization through recurrent and congruity backcrossing. Theoretical and Applied Genetics 88 , 324 331. Mencuccini M, Comstock J . 1999. Variability in hydraulic architecture and gas exchange of common bean (Phaseolus vulgaris) cultivars under well - watered conditions: interactions with leaf size. Australian Journal of Plant Physiology 26 , 115 124. Mittler R . 2006. Abiotic stress, th e field environment and stress combination. Trends in Plant Science 11 , 15 19. Morgan JM . 1984. Osmoregulation and w ater s tress in h igher p lants. Annual Review of Plant Physiology 35 , 299 319. Mori IC, Murata Y . 2011. ABA signaling in stomatal guard cells: lessons from Commelina and Vicia. Journal of Plant Research 124 , 477 487. Moroney J, Jungnick N, DiMario R, Longstreth D . 2013. Photorespiration and carbon concentrating mechanisms: two adaptations to high O2, low CO2 conditions. Photosynthesis Research 1 17 , 121 131. Mukeshimana G, Butare L, Cregan PB, Blair MW, Kelly JD . 2014. Quantitative t rait l oci a ssociated with d rought t olerance in c ommon b ean. Crop Sci ence . 54 , 923 938. 147 Muller B, Pantin F, Génard M, Turc O, Freixes S, Piques M, Gibon Y . 2011. Water deficits uncouple growth from photosynthesis, increase C content, and modify the relationships between C and growth in sink organs. Journal of Experimental Botany 62 , 1715 1729. Müller B deFaria, Sakamoto T, Silveira R, et al. 2014. Different ially e xpressed g enes during f lowering and g rain f illing in c ommon b ean (Phaseolus vulgaris) g rown under d rought s tress c onditions. Plant Molecular Biology Reporter 32 , 438 451. Muñoz LC, Blair MW, Duque MC, Tohme J, Roca W . 2004. Introgression in c ommon b ean × t epary bean interspecific congruity - backcross lines as measured by AFLP m arkers. Crop Sci ence . 44 , 637 645. Muñoz - Perea C, Allen R, Westermann D, Wright J, Singh S . 2007. Water use efficiency among dry bean landraces and cultivars in drought - stressed and non - stressed environments. Euphytica 155 , 393 402. Muñoz - Perea CG, Terán H, Allen RG, Wright JL, Westermann DT, Singh SP . 2006. Selection for d rought r esistance in d ry b ean l andraces and c ultivars. Crop Sci ence . 46 , 2111 2120. Murata N, Mohanty PS, Hayashi H, Papageorgiou GC . 1992. Glycinebetaine stabilizes the association of extrinsic proteins with the photosynthetic oxygen - evolving complex. FEBS Letters 296 , 187 189. Nabhan G, Felger R . 1978. Teparies in Southwestern North America - biogeographical and ethnohistorical study of Phaseolus acutifolius. Economic Botany 32 , 2 19. Navarro FM, Sass ME, Nienhuis J . 2009. Marker - f acilitated s election for a m ajor QTL a ssociated with r oot r ot r esistance in s nap b ean (Phaseolus vulgaris L.). Crop Science 49 , 85 0 856. Nayyar H . 2003. Accumulation of osmolytes and osmotic adjustment in water - stressed wheat (Triticum aestivum) and maize (Zea mays) as affected by calcium and its antagonists. Environmental and Experimental Botany 50 , 253 264. Neill S, Barros R, Brigh t J, Desikan R, Hancock J, Harrison J, Morris P, Ribeiro D, Wilson I . 2008. Nitric oxide, stomatal closure, and abiotic stress. Journal of Experimental Botany 59 , 165 176. Nielsen DC, Nelson NO . 1998. Black b ean s ensitivity to wa ter s tress at v arious g rowt h s tages. Crop Science 38 , 422 427. Nishizawa A, Yabuta Y, Shigeoka S . 2008. Galactinol and raffinose constitute a novel function to protect plants from oxidative damage. Plant Physiology 147 , 1251 1263. Obata T, Matthes A, Koszior S, Lehmann M, Araujo WL, Bock R, Sweetlove LJ, Fernie AR . 2011. Alteration of mitochondrial protein complexes in relation to metabolic regulation under short - term oxidative stress in Arabidopsis seedlings. Phytochemistry 72 , 1081 1091. 148 Okajima Y, Taneda H, Noguchi K, Terashima I . 2012. Optimum leaf size predicted by a novel leaf energy balance model incorporating dependencies of photosynthesis on light and temperature. Ecological Research 27 , 333 346. Ogren WL . 1984. Photorespiration: p athways, r egulation, and m odification. Annua l Review of Plant Physiology 35 , 415 442. Park S - Y, Fung P, Nishimura N, et al. 2009. Abscisic a cid i nhibits t ype 2C p rotein phosphatases via the PYR/PYL f amily of START p roteins. Science 324 , 1068 1071. Passioura JB . 1996. Drought and drought tolerance. P lant Growth Regulation 20 , 79 83. Passioura JB . 2006. The perils of pot experiments. Functional Plant Biology 33 , 1075 1079. Passioura JB . 2012. Phenotyping for drought tolerance in grain crops: when is it useful to breeders? Functional Plant BIology 39 , 8 51 859. Pastenes C, Horton P . 1996. Effect of h igh t emperature on p hotosynthesis in b eans II. CO a ssimilation and m etabolite c ontents. Plant Physiology 112 , 1253 1260. Pastenes C, Porter V, Baginsky C, Horton P, Gonzalez J . 2004. Paraheliotropism can protect water - stressed bean (Phaseolus vulgaris L.) plants against photoinhibition. Journal of Plant Physiology 161 , 1315 1323. Peterhansel C, Krause K, Braun H - P, Espie GS, Fernie AR, Hanson DT, Keech O, Maurino VG, Mielewczik M, Sage RF . 2013. Engineering photorespiration: current state and future possibilities. Plant Biology 15 , 754 758. Pimentel C, Ribeiro RV, Machado EC, dos Santos MG, de Oliveira RF . 2013. In vivo temperature limitations of photosynthesis in Phaseolus vulgaris L. Environmental and Experimental Botany 91 , 84 89. Pinheiro C, Chaves MM . 2011. Photosynthesis and drought: can we make metabolic connections from available data? Journal of E xperimental Botany 62 , 869 882. Pinheiro H, DaMatta F, Chaves A, Loureiro M, Ducatti C . 2005. Drought tolerance is associated with rooting depth and stomatal control of water use in clones of Coffea canephora. Annals of Botany 96 , 101 108. Pinto ME, Casati P, Hsu T - P, Ku MSB, Edwards GE . 1999. Effects of UV - B radiation on growth, photosynthesis, UV - B - absorbing compounds and NADP - malic enzyme in bean (Phaseolus vulgaris L.) grown under different nitrogen conditions. Journal of Photochemistry and Photobiology B: Biology 48 , 200 209. Poorter H, Buehler J, van Dusschoten D, Climent J, Postma JA . 2012 a . Pot size matters: a meta - analysis of the effects of rooting volume on plant growth. Functional Plant Biology 39 , 839 850. 149 Poorter H, Fiorani F, Stitt M, et al. 20 12 b . The art of growing plants for experimental purposes: a practical guide for the plant biologist Review. Functional Plant Biology 39 , 821 838. Porch TG . 2006. Application of s tress i ndices for h eat t olerance s creening of c ommon b ean. Journal of Agronomy & Crop Science 192 , 390 394. Porch TG, Beaver JS, Brick MA . 2013. Registration of Tepary Germplasm with Multiple - Stress Tolerance, TARS - Tep 22 and TARS - Tep 32. Journal of Plant Registrations 7 , 358 364. Porch TG, Jahn M . 2001. Effects of high - temperature stress on microsporogenesis in heat - sensitive and heat - tolerant genotypes of Phaseolus vulgaris. Plant, Cell & Environment 24 , 723 731. Prasad PVV, Boote KJ, Allen LH, Thomas JMG . 2002. Effects of elevated temperature an d carbon dioxide on seed - set and yield of kidney bean (Phaseolus vulgaris L.). Global Change Biology 8 , 710 721. Qin X, Zeevaart JAD . 1999. The 9 - cis - epoxycarotenoid cleavage reaction is the key regulatory step of abscisic acid biosynthesis in water - stress ed bean. Proceedings of the National Academy of Sciences 96 , 15354 15361. Rainey K, Griffiths P . 2005 a . Differential response of common bean genotypes to high temperature. Journal of the American Society for Horticultural Science 130 , 18 23. Rainey K, Grif fiths P . 2005 b . Evaluation of Phaseolus acutifolius A. Gray plant introductions under high temperatures in a controlled environment. Genetic Resources and Crop Evolution 52 , 117 120. Ramalho JC, Zlatev ZS, Leitão AE, Pais IP, Fortunato AS, Lidon FC . 2014. Moderate water stress causes different stomatal and non - stomatal changes in the photosynthetic functioning of Phaseolus vulgaris L. genotypes. Plant Biology 16 , 133 146. Rao I, Beebe S, Polania J, Ricaurte J, Cajiao C, Garcia R, Rivera M . 2013. Can tepary bean be a model for improvement of drought resistance in common bean? African Crop Science Journal 21 , 265 281. Rasmussen S, Barah P, Suarez - Rodriguez MC, Bressendorff S, Friis P, Costantino P, Bones AM, Nielsen HB, Mundy J . 2013. Transcriptome r esponses t o c ombinations of s tresses in Arabidopsis. Plant Physiology 161 , 1783 1794. Reich PB, Ellsworth DS, Walters MB, Vose JM, Gresham C, Volin JC, Bowman WD . 1999. Generality of l eaf t rait r elationships: a t est across s ix b iomes. Ecology 80 , 1955 1969. Richter JA, Erban A, Kopka J, Zörb C . 2015. Metabolic contribution to salt stress in two maize hybrids with contrasting resistance. Plant Science 233 , 107 115. Rizhsky L, Liang H, Mittler R . 2002. The c ombined e ffect of d rought s tress and h eat s hock on g ene e xpres sion in t obacco. Plant Physiology 130 , 1143 1151. 150 Rizhsky L, Liang H, Shuman J, Shulaev V, Davletova S, Mittler R . 2004. When d efense p athways c ollide. The r esponse of Arabidopsis to a c ombination of d rought and h eat s tress. Plant Physiology 134 , 1683 1696 . Rolland F, Baena - Gonzalez E, Sheen J . 2006. Sugar s ensing and s ignaling in p lants: c onserved and n ovel m echanisms. Annual Review of Plant Biology 57 , 675 709. Rosales MA, Ocampo E, Rodríguez - Valentín R, Olvera - Carrillo Y, Acosta - Gallegos J, Covarrubias A A . 2012. Physiological analysis of common bean (Phaseolus vulgaris L.) cultivars uncovers characteristics related to terminal drought resistance. Plant Physiology and Biochemistry 56 , 24 34. Rosales - Serna R, Kohashi - Shibata J, Acosta - Gallegos JA, Trejo - L ópez C, Ortiz - Cereceres J, Kelly JD . 2004. Biomass distribution, maturity acceleration and yield in drought - stressed common bean cultivars. Field Crops Research 85 , 203 211. Rut G, Rzepka A, Krupa J . 2010. Effect of hypoxia and post - hypoxia on the fluctuat ions in contents of malate and citrate, the activity of malic enzyme, and on the intensity of gas exchange in moss gametophores. Photosynthetica 48 , 79 86. Sage RF . 2004. The evolution of C4 photosynthesis. New Phytologist 161 , 341 370. Sage R, Kubien D . 2007. The temperature response of C - 3 and C - 4 photosynthesis. Plant Cell and Environment 30 , 1086 1106. Sanders P, Markhart A . 1992. Interspecific grafts demonstrate root - system control of leaf water status in water - stressed phaseolus. Journal of Experim ental Botany 43 , 1563 1567. Schneider KA, Brothers ME, Kelly JD . 1997 a . Marker - a ssisted s election to i mprove d rought r esistance in c ommon b ean. Crop Science 37 , 51 60. Schneider KA, Rosales - Serna R, Ibarra - Perez F, Cazares - Enriquez B, Acosta - Gallegos JA, Ramirez - Vallejo P, Wassimi N, Kelly JD . 1997 b . Improving c ommon b ean p erformance under d rought s tress. Crop Science 37 , 43 50. Schramm F, Larkindale J, Kiehlmann E, Ganguli A, Englich G, Vierling E, Von Koskull - Döring P . 2008. A cascade of transcription fa ctor DREB2A and heat stress transcription factor HsfA3 regulates the heat stress response of Arabidopsis. The Plant Journal 53 , 264 274. Schmutz J, McClean PE, Mamidi S, et al. 2014. A reference genome for common bean and genome - wide analysis of dual domestications. Nat Genet 46 , 707 713. Schwartz SH, Tan BC, Gage DA, Zeevaart JAD, McCarty DR . 1997. Specific o xidative c leavage of c arotenoids by VP14 of m aize. Science 276 , 1872 1 874. Schymanski SJ, Or D, Zwieniecki M . 2013. Stomatal c ontrol and l eaf t hermal and h ydraulic c apacitances under r apid e nvironmental f luctuations. PLoS ONE 8 , e54231. 151 Seiler C, Harshavardhan VT, Rajesh K, Reddy PS, Strickert M, Rolletschek H, Scholz U, Wob us U, Sreenivasulu N . 2011. ABA biosynthesis and degradation contributing to ABA homeostasis during barley seed development under control and terminal drought - stress conditions. Journal of Experimental Botany 62 , 2615 2632. Seiler C, Harshavardhan VT, Redd y PS, et al. 2014. Abscisic a cid f lux a lterations r esult in d ifferential a bscisic a cid s ignaling r esponses and i mpact a ssimilation e fficiency in b arley under t erminal d rought s tress. Plant Physiology 164 , 1677 1696. Serraj R, Sinclair TR . 2002. Osmolyte accumulation: can it really help increase crop yield under drought conditions? Plant, Cell & Environment 25 , 333 341. Seversike T, Sermons S, Sinclair T, Carter T Jr, Rufty T . 2014. Physiological properties of a drought - resistant wild soyb ean genotype: Transpiration control with soil drying and expression of root morphology. Plant and Soil 374 , 359 370. Sharkey TD . 2005. Effects of moderate heat stress on photosynthesis: importance of thylakoid reactions, rubisco deactivation, reactive oxyg en species, and thermotolerance provided by isoprene. Plant, Cell & Environment 28 , 269 277. Sharkey TD . 2015. Commentary: what gas exchange data can tell us about photosynthesis. Plant, Cell & Environment. Sharkey TD, Bernacchi CJ, Farquhar GD, Singsaas E L . 2007. Fitting photosynthetic carbon dioxide response curves for C - 3 leaves. Plant Cell and Environment 30 , 1035 1040. Sharma DK, Andersen SB, Ottosen C - O, Rosenqvist E . 2012. Phenotyping of wheat cultivars for heat tolerance using chlorophyll a fluoresc ence. Functional Plant Biology 39 , 936 947. Shulaev V, Cortes D, Miller G, Mittler R . 2008. Metabolomics for plant stress response. Physiologia Plantarum 132 , 199 208. Sinclair TR, Zwieniecki MA, Holbrook NM . 2008. Low leaf hydraulic conductance associated with drought tolerance in soybean. Physiologia Plantarum 132 , 446 451. Singh SP . 1995. Selection for w ater - s tress to lerance in i nterracial p opulations of c ommon b ean. Crop Science 35 , 118 124. Singh SP , Gepts P, Debouck DG . 1991. Races of Common Bean (Phaseolus vulgaris, Fabaceae). Economic Botany 45 , 379 396. Singh SP, Muñoz CG . 1999. Resistance to c ommon b acterial b light among Phaseolus s pecies and c ommon b ean i mprovement. Crop Sci ence . 39 , 80 89. Singh J, Pandey P, James D, Chandrasekhar K, Achary VMM, Kaul T, Tripathy BC, Reddy MK . 2014. Enhancing C3 photosynthesis: an outlook on feasible interventions for crop improvement. Plant Biotechnology Journal 12 , 1217 1230. 152 Singh SP, Schwartz HF . 2010. Breeding c ommon b ean for r esistance to d iseases: a r eview. Crop Science 50 , 2199 2223. Smartt J . 1981. Gene pools in Phaseolus and Vigna cultigens. Euphytica 30 , 445 449. Smith AM, Stitt M . 2007. Coordination of carbon supply and plant growth. Plant Cell and Environment 30 , 1126 11 49. Somerville CR, Ogren WL . 1982. Genetic modification of photorespiration. Trends in Biochemical Sciences 7 , 171 174. Souza RP, Machado EC, Silva JAB, Lagôa AMMA, Silveira JAG . 2004. Photosynthetic gas exchange, chlorophyll fluorescence and some associated metabolic changes in cowpea (Vigna unguiculata) during water stress and recovery. Environmental and Experimental Botany 51 , 45 56. Sreenivasulu N, Harshavardhan VT, Govind G, Seiler C, Kohli A . 2012. Contrapuntal role of ABA: Does it mediate stress tolerance or plant growth retardation under long - term drought stress? Gene 506 , 265 273. Sreenivasulu N, Sopory SK, Kavi Kishor PB . 2007. Deciphering the regulatory mechanisms of ab iotic stress tolerance in plants by genomic approaches. Gene 388 , 1 13. Storey R, Ahmad N, Jones R . 1977. Taxonomic and ecological aspects of distribution of glycinebetaine and related compounds in plants. Oecologia 27 , 319 332. Swindell W, Huebner M, Webe r A . 2007. Transcriptional profiling of Arabidopsis heat shock proteins and transcription factors reveals extensive overlap between heat and non - heat stress response pathways. BMC Genomics 8 , 125. Terán H, Singh SP . 2002 a . Comparison of s ources and l ines s elected for d rought r esistance in c ommon b ean. Crop Science 42 , 64 70. Terán H, Singh SP . 2002 b . Selection for drought resistance in early generations of common bean populations. Canadian Journal of Plant Science 82 , 491 497. Thompson AJ, Andrews J, Mulhol land BJ, et al. 2007. Overproduction of abscisic acid in tomato increases transpiration efficiency and root hydraulic conductivity and influences leaf expansion. Plant Physiology 143 , 1905 1917. Tian L, DellaPenna D, Zeevaart J . 2004. Effect of hydroxylate d carotenoid deficiency on ABA accumulation in Arabidopsis. Physiologia Plantarum 122 , 314 320. Timm S, Mielewczik M, Florian A, Frankenbach S, Dreissen A, Hocken N, Fernie AR, Walter A, Bauwe H . 2012. High - to - Low CO2 Acclimation r eveals p lasticity of the p hotorespiratory p athway and i ndicates r egulatory l inks to c ellular m etabolism of Arabidopsis. PLoS ONE 7 , e42809. 153 Tolk JA, Howell TA, Miller FR . 2013. Yield component analysis of grain sorghum grown under water stress. Field Crops Research 145 , 44 51. Udomprasert N, Li PH, Davis DW, Markhart AH . 1995. Effects of r oot t emperatures on l eaf g as e xchange and g rowth at h igh a ir t emperature in Phaseolus acutifolius and Phaseolus vulgaris. Crop Sci ence . 35 , 490 495. Urano K, Maruyama K, Ogata Y, et al. 2009. C haracterization of the ABA - regulated global responses to dehydration in Arabidopsis by metabolomics. Plant Journal 57 , 1065 1078. Usadel B, Blaesing OE, Gibon Y, Retzlaff K, Hoehne M, Guenther M, Stitt M . 2008. Global transcript levels respond to small cha nges of the carbon status during progressive exhaustion of carbohydrates in Arabidopsis rosettes. Plant Physiology 146 , 1834 1861. Vadez V, Kholova J, Medina S, Kakkera A, Anderberg H . 2014. Transpiration efficiency: new insights into an old story. Journal of Experimental Botany 65 , 6141 6153. Vassey T, Sharkey T . 1989. Mild water stress of Phaseolus vulgaris plants leads to reduced starch synthesis and extractable sucrose phosphate synthase activity. Plant Physiology 89 , 1066 1070. Verbruggen N, Hermans C . 2008. Proline accumulation in plants: a review. Amino Acids 35 , 753 759. Verslues PE, Juenger TE . 2011. Drought, metabolites, and Arabidopsis natural variation: a promising combination for understanding adaptation to water - limited environments. Current Op inion in Plant Biology 14 , 240 245. Wahid A, Gelani S, Ashraf M, Foolad MR . 2007. Heat tolerance in plants: An overview. Environmental and Experimental Botany 61 , 199 223. Wang W, Vinocur B, Shoseyov O, Altman A . 2004. Role of plant heat - shock proteins and molecular chaperones in the abiotic stress response. Trends in Plant Science 9 , 244 252. Wanner LA, Junttila O . 1999. Cold - induced freezing tolerance in Arabidopsis. Plant Physiology 120 , 391 399. Wentworth M, Murchie EH, Gray JE, Villegas D, Pastenes C, Pinto M, Horton P . 2006. Differential adaptation of two varieties of common bean to abiotic stress: II. Acclimation of photosynthesis. Journal of Experimental Botany 57 , 699 709. Wenzel A, Frank T, Reichenberger G, Herz M, Engel K - H . 2015. Impact of induce d drought stress on the metabolite profiles of barley grain. Metabolomics 11 , 454 467. Westgate ME, Boyer JS . 1986. Reproduction at l ow and p ollen w ater p otentials in m aize. Crop Science 26 , 951 956. 154 Wheeler MCG, Tronconi MA, Drincovich MF, Andreo CS, Flüg ge U - I, Maurino VG . 2005. A c omprehensive a nalysis of the NADP - m alic e nzyme g ene f amily of Arabidopsis. Plant Physiology 139 , 39 51. White JW, Singh SP . 1991. Sources and inheritance of earliness in tropically adapted indeterminate common bean. Euphytica 5 5 , 15 19. Wilkinson S, Davies W . 2002. ABA - based chemical signalling: the co - ordination of responses to stress in plants. Plant Cell and Environment 25 , 195 210. Winter D, Vinegar B, Nahal H, Ammar R, Wilson GV, Provart NJ e lectronic f luorescent p b rowser for e xploring and a nalyzing l arge - s cale b iological d ata s ets. PLoS ONE 2 , e718. Witt S, Galicia L, Lisec J, Cairns J, Tiessen A, Luis Araus J, Palacios - Rojas N, Fernie AR . 2012. Metabolic and p henotypic r esponses of g reenhouse - g rown m aize h ybrids to e xperimentally c ontrolled d rought s tress. Molecular Plant 5 , 401 417. Wong SC, Cowan IR, Farquhar GD . 1979. Stomatal conductance correlates with photosynthetic capacity. Nature 282 , 424 426. Wright IJ, Reich PB, Cornelissen JHC, et al. 2005 . Assessing the generality of global leaf trait relationships. New Phytologist 166 , 485 496. Xiao X, Yang F, Zhang S, Korpelainen H, Li C . 2009. Physiological and proteomic responses of two contrasting Populus cathayana populations to drought stress. Physi ologia Plantarum 136 , 150 168. Xing W, Rajashekar C . 1999. Alleviation of water stress in beans by exogenous glycine betaine. Plant Science 148 , 185 192. Xing W, Rajashekar C . 2001. Glycine betaine involvement in freezing tolerance and water stress in Arab idopsis thaliana. Environmental and Experimental Botany 46 , 21 28. Yang SH, Zeevaart JAD . 2006. Expression of ABA 8 ` - hydroxylases in relation to leaf water relations and seed development in bean. Plant Journal 47 , 675 686. Yang J, Zhang J . 2006. Grain filling of cereals under soil drying. New Phytologist 169 , 223 236. Yin Y, Adachi Y, Ye W, Hayashi M, Nakamura Y, Kinoshita T, Mori IC, Murata Y . 2013. Difference in a bscisic a cid p erception m echanisms between c losure i nduction and o pening i n hibition of s tomata. Plant Physiology 163 , 600 610. Yurekli F, Porgali Z, Turkan I . 2004. Variations in abscisic acid, indole - 3 - acetic acid, gibberellic acid and zeatin concentrations in two bean species subjected to salt stress. Acta Biologica Cracoviensi a Series Botanica 46 , 201 212. 155 - - . 2013. Differential proteomic analysis of drought stress response in leaves of common bean (Phaseolus vulgaris L.). Journal of Proteomics 78 , 254 272. Zelitch I, Schultes NP, Peterson RB, Brown P, Brutnell TP . 2009. High g lycolate o xidase a ctivity is r equired for s urvival of m aize in n ormal a ir. Plant Physiology 149 , 195 204. Zhang J, Nguyen HT, Blum A . 1999. Genetic analysis of osmotic adjustment in crop plant s. Journal of Experimental Botany 50 , 291 302. Zinselmeier SA, Lauer MJ, Boyer JS . 1995. Reversing d rought - i nduced l osses in g rain y ield: s ucrose m aintains e mbryo g rowth in m aize. Crop Science 35 , 1390 1400. Zlatev ZS, Lidon FC, Ramalho JC, Yordanov IT . 20 06. Comparison of resistance to drought of three bean cultivars. Biologia Plantarum 50 , 389 394. Zwack PJ, Rashotte AM . 2015. Interactions between cytokinin signalling and abiotic stress responses. Journal of Experimental Botany 66 , 4863 4871.