‘ l I, ‘36.“; . I P t ”ii?" I 9:353:91. J", P1234“: mgr. . "JV-“"1“"! I 7 u'f'}! r, DI :"AI ‘I-..,”HI'. ”5% . .. .v.oa..,” 3.71:, -,r .. r;--'-" . “f. p: .... 99' I r ‘ '1 SAC”, ,,\-' “ ""’:':' , virgin r «bx?! 35.! -.1 I“ .-I “_ . J I raz-a- ”fig? '23:" u, ? ; ‘ 7 ., ""1. 2: {LT}; -' H“ ~- N , cu ' ' . HIVI‘: ‘ 4993-5, 1 V ,‘ _ '1‘}. ,-; ”.3ng ”35?. 'HZHU 511, , . I n - n? ' , . V Ii,” £21,? 3:}- I“: ‘ E" . .I ' 4.; . ‘1‘, fribk {at t: .’ - y)- ' . I. I MMMMMMM MM 1 .1 9L I m o v v . 3 1293 005387 LIBRARY Michigan State University This is to certify that the dissertation entitled MUTANTS OF ARABIDOPSIS THALIANA THAT EITHER REQUIRE OR ARE SENSITIVE TO HIGH ATMOSPHERIC C02 CONCENTRATIONS presented by NANCY NICOLE ARTUS has been accepted towards fulfillment of the requirements for Ph . D . degree in Botany :ESM ‘ Major professor April 30, 1988 [)ate MSU i: an Affirmative Anion/Equal Opportunity Imrirurion 0-12771 MSU RETURNING MATERIALS: Place in book drop to LJBRARJES remove this checkout from All-InllilL. your record. FINES will be charged if book is returned after the date stamped below. MUTANTS OF ARABIDOPSIS THALIANA THAT EITHER REQUIRE 0R ARE SENSITIVE TO HIGH ATMOSPHERIC CO2 CONCENTRATIONS BY Nancy Nicole Artus A DISSERTATION - Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Botany and Plant Pathology 1988 (TD ABSTRACT MUTANTS OF ARABIDOPSIS THALIANA THAT EITHER REQUIRE 0R ARE SENSITIVE TO HIGH ATMOSPHERIC CO2 CONCENTRATIONS BY Nancy Nicole Artus A major objective of this study was to isolate mutants of the cruciferous plant, Arabigopsis thaljana (Heynh.), with defects in photorespiration. Previously, mutants were isolated with a screen utilizing an atmosphere enriched with 1% CO2 as a permissive condition for growth, and air as a nonpermissive condition. Many potential classes of photorespiratory mutants were not recovered. In the first part of this study, two new screens were employed for the detection of mutants that have either a more subtle or a more severe phenotype in air than the previously isolated mutants. One mutant was recovered from each screen with a unique phenotype however, neither one of these was a novel photorespiratory mutant. In addition, 20 photorespiratory mutants were isolated which have phenotypes similar to the previously isolated mutants. Four of these appear to be novel photorespiratory mutants. The possibility that certain mutants do not exist because there is more than one gene, or more than one function for particular enzymes, was examined. A unique high COZ-requiring mutant of A; thaliana was isolated which requires 2% CO2 for normal growth. The mutant develops chlorosis and necrosis within a few days in air. This line has no apparent defect in photosynthetic COZ-fixation, photorespiration, photosynthetic electron transport, or in the detoxification of reactive oxygen species. The mutant displayed slightly altered kinetics of room temperature induction fluorescence, and the concentrations of a-tocopherol and glutathione were elevated 2- and 1.5-fold, respectively, in mutant as compared to wild-type plants grown in 2% C02. It was not possible to determine the biochemical basis of the mutation. Two mutants of A, thaliana were isolated which grow well with a normal atmospheric CO2 concentration but become chlorotic in 2% C02. The development of the chlorosis follows distinctive patterns in each mutant. In one case, the veins turn yellow before the interveinal regions do so. In the other case, the chlorosis begins at the base of the leaf and spreads towards the tip. In both cases, the CO2 sensitivity is not caused by a requirement for photorespiration or by a mineral deficiency. Some possible explanations for the phenotypes are discussed. ACKNOWLEDGMENTS There are many people to thank for guidance, technical assistance, and moral support. I thank Dr. Chris Somerville for his guidance and the many insights he provided. I also thank the other members of my committee, Dr. Andrew Hanson, Dr. Ed Tolbert and Dr. Jan Zeevaart for their guidance, their reviews of this dissertation (special thanks to AH for his thoroughness in this respect), and the encouragement they gave me, particularly in the last few weeks of preparing the dissertation. All of the past and present members of Chris Somerville’s lab deserve acknowledgment for much helpful advice, and for being very nice people to work with. I particularly want to thank Dr. Peter McCourt for his assistance in the early stages of this work, soon-to-be Dr. Tim Caspar for all of the useful advice and discussions, and Dr. Micha Volokita for teaching me some molecular biology. Last but not least, I want to thank a special chemist (who happens to be my husband), Dr. Curtis Hoganson, for many consultations regarding this work, and for a lot of moral support over the past several years. iv TABLE OF CONTENTS PAGE List of Tables ......................... viii List of Figures ......................... x List of Abbreviations ...................... xi Introduction ........................ 1‘ Chapter 1. A Search for New Photorespiration Mutants in Arabidopsis thaliana. ..... 2 Introduction ........................ 3 Materials and Methods ................... 11 Special Chemicals ................... 11 Growth of Plants ................... 11 M3 Lines ....................... 11 Screen for Photorespiratory Mutants with Subtle Phenotypes in Air .11 Screen for Photorespiratory Mutants with Severe Phenotypes in Air .12 CO2 Exchange and 14C02 Labeling ............ 12 Complementation Analysis ............... 13 Enzyme Assays ..................... 13 Inhibition of Glycolate Oxidase ............ 14 Southern Blot ..................... 14 Results .......................... 15 Screens for Photorespiratory Mutants ......... 15 Phenocopy of a Glycolate Oxidase Mutant ........ 25 PAGE Southern Blot ..................... 28 Discussion ......................... 29 Literature Cited ...................... 35 Chapter 2. A Mutant of AtihiQQESiS thaliana that Exhibits Chlorosis in Air but Not Atmospheres Enriched in C02 ...... 38 Introduction ....................... 39 Materials and Methods ................... 40 Plant Materials and Growth Conditions ......... 4O Tissue Culture Methods ................ 40 CO2 Exchange and 14CO2 Labeling ............ 41 Leaf Fluorescence and Thylakoid Electron Transport. . .42 Carotenoid Measurements ................ 42 Enzyme Assays ..................... 42 SOD Isozymes ..................... 44 Antioxidant Assays ................... 44 Other Measurements ................... 44 Results ........................... 45 Phenotype of the Mutant ................ 45 Genetic Analysis ................... 49 Effects of Oxygen Concentration ............ 50 Effects on Photosynthetic C02 Assimilation ...... 51 Labeling Studies ................... 53 Effects on Electron Transport ............. 56 Ability to Detoxify Reactive Oxygen and H202 Levels in Leaf Extracts .56 Discussion ......................... 64 Literature Cited ...................... 69 vi PAGE Chapter 3. Mutants of Arabjggpsis thaliana that Exhibit Chlorosis in an Atmosphere Enriched with CO2 .......... 72 Introduction ........................ 73 Materials and Methods ................... 74 Plant Materials and Growth Conditions ......... 74 Tissue Culture Methods ................ 74 Mineral Analysis ................... 75 Chlorophyll ...................... 75 Epidermal Impressions ................. 75 Results .......................... 75 Genetic Analysis ................... 75 Phenotypes of the Mutants ............... 76 Effects of Oxygen Concentration ............ 83 Mineral Analyses ................... 86 Discussion ......................... 86 Comparison of the Phenotypes of the Two Mutants . . . .86 Possible Causes of High C02 Sensitivity . . . . . . . .87 Literature Cited ...................... 93 LIST OF TABLES TABLE PAGE 1-1. Possible reasons why there are no mutants for certain 8 enzymes and transport steps of photorespiration. 1-2. Distribution of 14C Label in Hater Soluble Products of 19 Photoassimilation by Photorespiratory Mutants of Arabldgpsi1s. 1-3. Genetic crosses of photorespiratory mutants with charac- 22 terized mutants. 1-4. Activities of some C cycle enzymes in photorespiratory . 23 mutants of Arabidopsis. 1-5. Lines of photorespiratory mutants of Arabidopsis thaliana 24 isolated between 1984 and 1986. 1-6. Germination of seeds of various species on 2-Hydroxy-3- 28 butynoic acid. 1-7. Frequencies of photorespiratory mutants in Arabidopsis, 34 thaliana and Horggum yglgarg. 2-1. Chlorophyll concentrations in wild-type and mutant 51 Arabidopsis grown in air containing 2% CO2 then trans- ferred for 2.5 days to air or 1% 02. 2-2. Photosynthetic gas exchange in air of wild- -type and 54 mutant Arabidopsis 2-3. Distgibution of 14C label in water soluble products 55 CO photoassimilation by wild- -type and mutant Arabidgfisis. 2-4. Electron transport rates of thylakoids from wild-type 58 and mutant Arabidopsis. 2-5. Activities of enzymes involved in detoxification of 61 activated oxygen species. 2-6. Levels of antioxidants in leaves of wild-type and 62 mutant grown in 2% CO and after four days (gluta- thione) or five days Iascorbic acid) in air. viii TABLE PAGE 2-7. Relative concentrations of carotenoids in wild-type 63 and mutant Arabidopsis. 3-1. Growth rates of calli derived from mutants sensitive 81 to elevated atmospheric C02. 3-2. Mineral content of leaves from wild-type and two high 85 COZ-sensitive mutant lines of Arabidopsis. ix LIST OF FIGURES FIGURE PAGE 1-1. The pathway of photorespiratory metabolism, also called 4 the C2 cycle. 1-2. Tracings of CO gas exchange for wild-type and 17 lines 16 of photorespiratory mutants. 1-3. Phenocopy of a glycolate oxidase mutant. 26 2-1. Mutant line C3208 and wild-type Arabidopsis grown in air 46 containing 2% C02 and transferred to air for six days. 2-2. Chlorophyll and carotenoid content in wild— —type and 47 mutant Arabidopsis following transfer of plants from 2% C02 to air. 2-3. Effect of light intensity on chlorophyll content of wild- 48 type and mutant Arabidopsis 4 days after transfer from 2% C02 to air (0.03% C02). 2-4. Effect of 3 atmospheric conditions on wild-type, C5208, 52 and a glycine decarboxylase deficient line, C5116. 2-5. Typical fluorescence induction kinetics of leaves from 57 wild-type and mutant Arabidopsis. 3-1. Effect of air enriched with 2% CO2 on two mutant lines 77 and wild-type Arabidopsis. 3-2. Chlorophyll concentrations in air enriched with 2% C02 80 as a function of light intensity for two high C02- sensitive mutants of Arabiggspsis. 3-3. Rates of primary root elongation for seedlings of two 82 mutant lines and wild-type Arabidopsis in air or air enriched with 2% C02. 3-4. Effect of a low 0 atmosphere on chlorophyll content of 84 two mutant lines and wild- -type Arabidopsi1s. 3-5. Stomatal impressions from wild-type Arabidopsis grown 89 in air and after one day in air enriched with 2% C02. C2 cycle C3 cycle CAT chl DCT dwt EMS Fd fwt GGAT GK GLYD GOGAT GOX GS HBA HPR M2 LIST OF ABBREVIATIONS C2 oxidative carbon cycle C3 reductive carbon cycle catalase chlorophyll dicarboxylate transport protein symbol for a genetic locus regulating dicarboxylate transport activity dry weight ethyl methane sulphonate ferredoxin fresh weight glutamate,glyoxylate aminotransferase glycerate kinase symbol for a locus regulating glutamate synthase activity glycine decarboxylase symbol for a locus regulating glycine decarboxylase activity glutamate synthase glycolate oxidase glutamine synthetase 2-hydroxy-3-butynoic acid 2-hydroxypyruvate reductase liter second generation of a mutagenized population xi PGP PSI PSII 0A 0B Rubisco SAL SGAT SHMT SIM THF Abbreviations cont’d third generation of a mutagenized population singlet oxygen superoxide radical hydroxyl radical symbol for a locus regulating phosphoglycolate phosphatase activity phosphoglycolate phosphatase photosystem I photosystem 11 primary electron accepting plastoquinone of PSII second electron accepting plastoquinone of PSII ribulosebisphosphate carboxylase/oxygenase symbol for a locus regulating serine,glyoxylate amino-transferase activity serine,glyoxylate aminotransferase serine hydroxymethyltransferase symbol for a locus regulating serine hydroxymethyltransferase activity tetrahydrofolic acid xii INTRODUCTION Carbon dioxide has a variety of functions and effects in plant metabolism, the major one being its function as a substrate for carbon fixation in photosynthesis. The aim of this study was to isolate mutants for use as tools to study the physiological processes that are influenced by C02‘ Mutants are valuable experimental tools superior to inhibitors because one need not worry about the inhibitor reaching the site of action or its nonspecific effects. Arabidopsis thaliana was chosen as the experimental organism because it offers many advantages for use in genetic studies, including its small size, rapid life cycle, high seed yield, and small genome. The first chapter describes new approaches for isolating high COZ—requiring mutants with defects in the C2 carbon oxidation cycle (a pathway intimately connected with photosynthetic carbon fixation). Because the enzymatic reactions of the C2 cycle are well understood, the primary value of new mutants in this pathway is their contribution to our understanding the feasibility of saturating a metabolic pathway with mutants. The second chapter describes a detailed attempt to characterize a high COZ-requiring mutant that is not a C2 cycle mutant. The properties of this mutant are suggestive of an unidentified role for CO2 in plant metabolism. The third chapter offers a description of two mutants that do not tolerate an elevated CO2 concentration. This is the first time that CO2 sensitive mutants have been described in any organism. Several possible explanations for this phenotype are discussed. CHAPTER 1 A SEARCH FOR NEW PHOTORESPIRATION MUTANTS WWW INTRODUCTION Photorespiration is the light dependent uptake of 02 and release of C02 in green tissues of plants. The process is intimately connected to photosynthetic carbon reduction because of the dual functions of the enzyme Rubisco. This enzyme will use either C02 or 02 as the gaseous substrate, producing either two molecules of 3-phosphoglyceric acid as the result of carboxylation, or one molecule of 3-phosphoglyceric acid and one molecule of 2-phosphoglycolic acid in the case of oxygenation. The phosphoglycolic acid is oxidized in the C2 oxidative photosynthetic carbon cycle (or photorespiratory pathway) which serves to recycle carbon back to the C3 reductive photosynthetic carbon cycle (Fig. 1-1). The C2 cycle is amenable to genetic analysis because the flow of carbon into the cycle can be regulated by altering the ratio of the atmospheric concentrations of C02 and 02. A high C02 to O2 ratio (e.g. 0.05) suppresses photorespiration and serves as a permissive condition for growth of mutants which are defective in photorespiration, while a low COz to 02 ratio (e.g. 0.002) promotes photorepiration and serves as a nonpermissive condition for photorespiratory mutants. It is presumed that the reason why photorespiratory mutants die in a photorespiratory atmosphere is because the accumulation of C2 cycle intermediates before the block is toxic. Somerville and Ogren (1982) employed this selection scheme to isolate mutants of Arabidgp§j§_thaliana (L.) with lesions in the photorespiratory pathway. They grew second generation plants from EMS mutagenized seed in 1% C02, balance air, to the rosette stage, then transferred them to air. Plants which were healthy in 1% CO2 but became chlorotic in air were the putative photorespiratory mutants and were returned to high C02 to regain their green color and Figure 1-1. The pathway of photorespiratory metabolism, also called the C cycle. Because of the two complementary routes of glycine metabdfiism in the mitochondrion, two molecules of phosphoglycolate must enter the pathway for each molecule of serine, C0 , and NH produced. The encircled numbers correspond to the followizng enzym s: [1] D-Ribulose-l,5-bisphosphate carboxylase/oxygenase, [2] phosphoglycolate phosphatase, [3] glycolate oxidase, [4] catalase, [5] glutamate:glyoxylate aminotransferase, [6] serine:glyoxylate aminotransferase, [7] glycine decarboxylase, [8] serine hydroxymethyltransferase, [9] 2-hydroxypyruvate reductase, [10] glycerate kinase, [11] glutamine synthetase, [12] glutamate synthase. Figure l-l Ca Chloroplaot Cycle 91120904 ATP ADP a 9'0 _ H-c-Ol-I NH °° H-c-ou 3 ® ”‘ ‘°" cuppa-2 cuppa? co2 3 a-Phoephogl ycorato Rlbuloae- 1 .s-blaphosphate E J ! Glutamate Glutamino 02 W253. coo- ® ATP 2-Phoephoalycoleto Q) \91 Ed” Fdrod oo- cflzo... Foroxlooulo 4| 153°" ¢°°’ 20" Glycolate Glycerate ”20 NM)" ® 2 (9 KO: 0 NADH g H "2 2 <6 900- ' _ 6'0 GOO CHZOH GIYOXYIIIO a-Hydroxypyruvate ‘ -WCC? : _ 2-Oxoglutarato NH; HO CH-é-H OO‘ L-Serino NADH NAD" Iltoobondrlon 6 eventually produce seed. Six classes of mutants with defects in photorespiration plus one mutant unable to activate Rubisco in yjyg were isolated in this manner (Somerville and Ogren, 1982a). In addition, two mutants of Arabidopsis have been isolated recently which require high CO2 but have normal photorespiratory metabolism. The biochemical lesions in these mutants were not determined (Artus and Somerville, 1988; Chastain, 1985). The photorespiratory mutants lack activity of either PGP, GLYD, SHMT, SGAT, GOGAT, or the chloroplast envelope DCT. Physiological studies with the mutants confirmed the .roles of each of these enzymes and the transport protein in the C2 cycle as depicted in Figure 1. More recently, photorespiratory mutants in barley and pea have been described with deficiencies in PGP, CAT, GOGAT, GS, the glycine to serine conversion, SGAT and DCT (Blackwell et al, 1988). Theoretically, it should have been possible to isolate several other classes of mutants in the C2 cycle. There are several enzymes in the pathway for which there are no corresponding mutants in Arabidgpsis. They include GOX, CAT, HPR, GK, GGAT, and GS. The interorganellar transport of metabolites may be susceptible to genetic alteration. Mutants of this sort would be useful in determining the mechanism of metabolite transport and the nature of any transport proteins which are present. Another type of mutant that would be informative would be one unable to provide reducing equivalents to the peroxisome for hydroxypyruvate reduction. Attempts to recover additional mutants of Arabidopsisonly resulted in more lines of all of the existing classes of mutants except PGP. 7 The possible explanations why mutants with defects in many enzymatic reactions associated with photorespiration have not been recovered can be divided into two categories (Table 1-1). Either the mutants are present in the mutagenized population but they are continuously missed in the screen due to a bias in the screening procedure, or the mutants do not exist. The previously isolated Arabidgnsis mutants all begin to turn yellow after two to three days of exposure to continuous illumination in air. If some mutants do not begin to turn yellow until four or more days in air, or are already too severely injured to be rescued after three days in air, then they would have been missed in the original screen performed by Somerville and Ogren. Variable phenotypes may be expected for photorespiratory mutants if the cause for yellowing in a photorespiratory atmosphere is the accumulation of a C2 cycle intermediate. The severity of the mutation would then depend on the particular metabolite that accumulates. For example, pools of glycine and serine are normally present in leaves and are not toxic, whereas phosphoglycolate is toxic because it inhibits the triose phosphate isomerase reaction (Anderson, 1971). Phosphoglycolate accumulates in the PGP mutant (Somerville and Ogren, 1979) and this mutant is more sensitive to a photorespiratory atmosphere than either the GLYD or the SHTM mutants, both of which accumulate glycine (Somerville and Ogren, 1982c; Somerville and Ogren, 1981), or the SGAT mutant which accumulates serine (Somerville and Ogren, l980a). Accumulation of glycolic acid, hydroxypyruvic acid or glyceric acid may be more toxic than accumulation of an amino acid since acidification of the cell may result. Thus, mutants defective in GOX, HPR or GK may have a more severe phenotype than the mutants 8 Table 1-1. Possible reasons why there are no mutants for certain enzymes and transport steps of photorespiration. A. The Phenotypes are Unique Relative to the Existing Classes of Mutants. 1. Symptoms take longer than 3 to 4 days in air to appear. . The mutants are already dead after 3 to 4 days in air. . The mutants are indistinguishable from wild-type because another protein is able to substitute for the mutant protein. Mutants Do Not Exist or Are Very Rare. . The mutants are lethal in high C0g because there is more than one function for the pr tein. . There is more than one gene. . A bias in the mutagenesis discriminates against certain genes so that the frequency of recovering these mutants is low. missing GLYD, SHMT, SGAT or GS, the latter of which accumulate amino acids. On the other hand, the chlorosis observed upon transferring the mutants to air may be at least in part due to nitrogen deficiency resulting from an accumulation of nitrogen—containing metabolites. If this hypothesis is true, mutants deficient in GLO, HPR or GK may turn yellow more slowly than mutants of GOGAT, GLYO, SHMT or SGAT because the latter are defective in the portion of the C2 pathway where nitrogen cycles, and the former are not. A third possible phenotype for a photorespiratory mutant that Somerville and Ogren would have missed would be a plant that displays no deleterious effects in air because another isozyme or a nonspecific ennzyme is able to substitute for the missing protein. There are three possible reasons why a particular mutant may not exist regardless of the screen employed or the number of plants screened. One reason may be that there is more than one active gene which codes for the enzyme. The multiple copies of the enzymes may be either the same or different isozymes. If there are two active genes, then in order to obtain a mutant, both must be altered by mutagenesis in a single seed in such a way that they both code for inactive proteins. The chance of this occurring in Arabidopsis mutagenized with EMS according to the standard procedure is approximately 1 in 4 x 106 (assuming recessive, loss of function mutations occur at a frequency of 1 in 2000, Haughn and Somerville, 1987). Another potential reason why a mutant would not exist may be that there is more than one essential function for a particular protein so that the mutant would be lethal in a nonphotorespiratory atmosphere. l0 This may be why there are no GS mutants in Arabidopsis. Glutamine synthetase is required in roots as well as in leaves for NH3 assimilation (Miflin and Lea, 1982). Most species have two isozymes of G5, one located in the chloroplast which functions in photorespiratory NH3 assimilation, and one in the cytosol which presumably satisfies the nonphotosynthetic needs for ammonia assimilation. Some species have only one or the other isozyme (McNally et al, 1983). Photorespiratory mutants of barley and pea exist which lack the chloroplast isozyme but have an active cytosolic isozyme (Nallsgrove et al, 1987). If Arabidopsis has only one of the isozymes, then a GS mutant might be lethal even in a nonphotorespiratory atmosphere. Another explanation why particular photorespiratory mutants were not found may be that certain genes are not susceptible to mutation by EMS. For reasons that are not understood, mutation frequency varies considerably between loci (Koornneef et al., 1982). It may be that a very large number of mutants must be screened in order to find particular mutants. The purpose of this study was to determine why so many potential classes of mutants are not represented in Somerville and Ogren’s collection of Arabidopsis photorespiratory mutants. First, mutagenized populations were screened for photorespiratory mutants with phenotypes more subtle or more severe than the existing classes of mutants. Some new mutants were discovered, but they do not have unique phenotypes. The potential reasons why certain mutants may not exist were investigated using glycolate oxidase as a model. ll MATERIALS AND METHODS Special Chemicals. 2-Hydroxy-3-butynoic acid (HBA) was synthesized as described (Jewess et al., 1975) by Chris Somerville. Growth of Plants. Arabidopsis thaliana (L.) Heynh. (Columbia) plants were grown at 23C in continuous fluorescent illumination (80-100 uE 111'2 s'1 PAR) on a perlite/vermiculite/sphagnum (1:1:1) mixture irrigated with mineral nutrients (Somerville and Ogren, 1982a). Carbon dioxide enrichment was achieved with a CO2 regulator (Forma Scientific). The mutagenesis procedure for the production of M2 seed has been described (Haughn and Somerville, 1986). M3 Lines. M3 lines were generated by collecting seed from individual M2 plants grown in 1% C02 and storing them in a vial. Each of the 2,500 lines was designated an upper case letter, a lower case letter, and a number according to the vial’s position in a box. Screen for Photorespiratory Mutants with Subtle Phenotypes in Air. To detect mutants that stop growing or grow slowly but do not necessarily become chlorotic in air, second generation mutagenized plants (M2) were photographed before they were transferred to air. M2 plants were grown at a density of ca. 40 plants per five inch diameter pot in 1% CO2 until two weeks old. The pots were individually photographed with color slide film, then transferred to a chamber with a standard atmosphere. After five to six days, each plant was compared to its picture. Plants that had grown well in high C02 but grew slowly or not 12 at all in air, and plants that became chlorotic in air, were returned to high CO2 and allowed to set seed. These putative mutants were assigned numbers beginning with NA. The M3 generation was rescreened by growing two pots of each line in 1% CO2 for two weeks, then one pot was moved to air for comparison. One third of the M2 plants screened by this method were generated from a glabrous, phosphoglucomutase deficient line designated TC7 (Caspar et al., 1985). Screen for Photorespiratory Mutants with Severe Phenotypes in Air. M3 lines were screened for mutants that die rapidly in air. Approximately 10 seeds from each M3 line were planted in a sector of a tray and grown in 2% C02. After 14 to 16 days, notes were taken on the appearance of each line, and the trays were moved to air. The mutants of interest died within a few days, but any plant that became chlorotic was noted. Putative photorespiratory mutants were rescreened in the M4 generation if the M3 survived, otherwise by growing more M3 plants. 602 Exchange and 14002 Labeling. Photosynthetic gas exchange was measured on whole plants at 23 C, 200 pE 111'2 s'1 and 330,ul CO2 l'l, 50% 02’ balance N2 with an infrared gas analyzer (Analytical Development Co., Hoddesdon, England) (Somerville and Ogen, 1982). Photoassimilation of 1400 2 2 was performed at 23C, and 200 to 300,uE m' s‘l. Nhole plants were equilibrated in darkness with either air (21% 02) or air enriched to 50% 02, then label was added at the start of illumination or after 7 anutes, as indicated. The incorporation period lasted 10 minutes. Labeled plants were plunged into liquid N2, and the water soluble products were separated and identified by ion 13 exchange chromatography and thin layer chromatography (Somerville and Ogren, 1982b). ”Basics" refer to the fraction that elutes from a Dowex-SO (H+) column with 8 N NH4OH but not H20; ”Neutrals" elute from both Dowex-SO (H+) and Dowex-l (formate) with H20; 'Acidl", "AcidZ", and 'Acid3” refer to fractions eluted from Dowex-l (formate) with 0.5 N formic acid, 8 N formic acid or 4 N HCl, respectively. Complementation Analysis. Genetic crosses were performed to determine if two mutations were at the same locus. The sepals, petals, and stamens were dissected away from unopen flowers on the mutant selected to serve as the female parent. Then anthers from open flowers on the mutant selected to serve as the male parent were brushed against the stigmas. The F1 plants resulting from the seed obtained from the crosses were grown in high CO2 for 2 weeks and then transferred to air. If the F1 plants turned yellow in air, it was concluded that the mutations of the two parents are at the same locus. If the F1 plants remained green and continued to grow in air, then the mutations were considered to be located in separate genes. Enzyme Assays. The extraction and assay of GOGAT (Somerville and Ogren, 1980b), SGAT and GGAT (Somerville and Ogren, 1980a), PGP (Somerville and Ogren, 1979), and HPR (Kleczkowski and Randall, 1988) have been described. GS was extracted by homogenizing 0.2 9 leaf material on ice in 2 ml 50 mM Tricine, pH 7.6, and 50 mM M9504, then clarifying the brei by centrifugation for 15 min at 30,0009. The assay was modified after Nallsgrove et al. (1979). One ml contained 50 mM MOPS, pH 7.2, 45 11M MgSO4, 100 M Na-glutamate, 8 mM ATP, 6 mM l4 neutralized hydroxylamine, and 200 pl extract. The reaction was terminated after 20 min by adding 1 ml of 0.67 N HCl, 0.20 N TCA, 10% FeCl The Fe3+-glutamylhydroxamate product was detected at 500 nm and 30 compared to standards. Inhibition of Glycolate Oxidase in 11,19. Seventeen day-old whole plants and detached leaves from 25 day-old plants were transferred to aqueous solutions containing 4 mM HBA and placed in air or air enriched with 1% COZ. Controls were put in water acidified to the same pH as the HBA. For seed germination experiments, surface sterilized seed of Arabidopsis. Herdeum 191931;: (Himalaya). Estunia mm; (Mitchell variety) and Brassica glgragga (turnip) were plated on sterile filter paper saturated with Murashige and Skoog nutrients (Gibco Laboratories) plus or minus 3 mM HBA and placed in darkness. Southern Blot. Total Arabiggpsis leaf DNA (Leutmiler et al., 1984) was digested in 7.5 pg aliquots, phenol extracted, then run on a 1% agarose gel and transferred to nitrocellulose by blotting (Maniatis et al, 1982). The blot was prehybridized in a solution containing 50% formamide, 5X Denhardt’s solution, 5X SSC (1X is 0.15 M NaCl, 0.015 M sodium citrate) and 100 119 ml'1 denatured salmon sperm DNA, and hybridized in an equivalent solution plus 10% dextran sulfate and approximately 80 ng ml.1 GOX cDNA insert labeled by nick translation with [a32P]dCTP. The GOX cDNA was provided by M. Volokita and C. Somerville (1987). Hybridization was carried out at 4 C overnight and the blot was washed several times in 2X SSC and 0.1% SOS at room temperature, then in 1x SSC and 0.1% SOS at 65 C, and finally in 0.2x l5 SSC and 0.1% SDS at 65 C. The blot was exposed a few hours to X-ray film at -70 C with an intensifying screen. RESULTS Screens for Photorespiratory Mutants. Ten thousand M2 plants were screened by the photographic method for photorespiratory mutants with phenotypes more subtle than the mutants isolated by Somerville and Ogren. Eighteen mutants were isolated, one (designated NA94) of which became chlorotic at a slower rate and to a lesser extent than the previously isolated mutants. As a diagnostic tool, photosynthetic gas exchange in air enriched with 02 was measured on most of the putative photorespiratory mutants. Photosynthesis was inhibited within 30 minutes for 14 of the 15 mutants examined (Fig. 1-2). This response is typical of all of the photorespiratory mutants described to date (Somerville and Ogren, 1982a). The mutant line NA94, with the uniquely subtle phenotype, was the one which did not display an inhibition of photosynthesis in a photorespiratory atmosphere. Most of the mutants were fed 14002 in air, or air enriched with 02, during the induction of photosynthesis (before photosynthesis is strongly inhibited). The 14C labeling patterns gave clues as to the site of the lesion since label generally accumulated in the intermediate immediately preceding the block (Table 1-2). The exceptions to this rule were mutants of GOGAT and OCT, since carbon originating from fixed CO2 does not get diverted to a-ketoglutarate, glutamate, and glutamine in the C2 cycle (Fig 1-1). These mutants had no labeling abnormality or showed label accumulation in malate. 16 Figure 1-2. Tracings of CO gas exchange for wild-type and 17 lines of photorespiratory mutants. 2The curves were normalized so that the difference between zero net CO exchange and CO evolution in the dark were equal for all lines. Thg arrows indicatg when the light was turned on. Net CO2 Exchange Figure 1-2 17 NA60 NA77 _l 1 1 1 NA94 NA1 OO NA1 09 O 20 4O Time(min) Not 002 Exchange 18 Figure 1-2 (cont'd) NA115 . NA1 26 NA133 -fxe L..f.$‘.-- 1 1 NA138 NA1 57 NA164 Time(min) 19 Table 1-2. Distribution of 14C label in water soluble products of 14CO photoassimilation by photorespiratory mutants of Arabidopsis, Plants were preequilibrated in darkness in humidified air(*) or air enriched with O to 50% (#) as indicated. Label was added at the start of illumiEation and the incorporation period lasted ten minutes unless noted: 0, 20 min incorporation; S, label added after 7 min of illum- ination and incorporation lasted 10-12 minutes. The fractions are defined in Materials and Methods. 2 Line Basics Neutrals Acids Total 1 2 3 (Percent) *HT 29 30 6 23 13 101 *NA60 17 31 10 27 14 99 *ONT 31 36 6 21 7 101 *0NA60 14 36 18 22 6 96 *NT 44 19 9 16 5 93 *NA66 61 11 7 16 5 99 *NT 32 27 5 27 18 109 *NA77 39 20 5 29 16 110 *NT 35 24 3 29 13 104 *NA78 47 14 6 23 12 102 #SHT 42 30 5 16 8 101 #SNA78 69 12 3 12 6 102 #NT 35 28 5 22 11 101 #NA79 56 9 3 20 15 103 #NA85 26 24 12 25 13 100 #NT 55 14 4 17 -- --- #NA94 59 10 4 22 -- --- #Eh5 38 13 10 20 13 94 *SHT 39 27 8 21 11 106 *SNAIOO 15 36 17 24 12 104 *SNA109 . 48 15 4 24 10 101 *SNAIIS 62 I8 3 12 9 104 Table 1-2 cont’d 20 Line Basics Neutrals Acids Total 1 2 tfis #TC7 42 17 11 19 6 96 #NA126 41 14 7 25 7 94 #NA138 76 5 1 9 7 98 #NA157 70 5 4 l4 4 99 #NA164 33 7 13 26 15 94 #STC7 28 28 9 20 6 91 #SNA126 20 9 22 34 3 88 #NT 35 27 5 26 9 102 #0113 20 27 3 18 28 96 #SNT 44 19 9 16 5 93 #SNA91 33 18 10 19 9 89 #SSe16 56 15 4 13 4 92 21 Based on these results, genetic complementation analysis or enzyme assays were performed to confirm or identify the site of the mutation for each mutant (Tables 1-3 and 1—4). Fourteen of the 18 mutants were found to belong to five of the six classes of photorespiratory mutants described previously. The final analyses of the 14 mutants are summarized in Table 1-5. There was no abnormality in the 14C labeling pattern for NA94 as compared to the wild-type (Table 1-2). This observation together with the normal photosynthesis (Fig 1-2) suggest that the mutation in line NA94 is not related to photorespiration. 14C One mutant, NA66, accumulated approximately 50% of the soluble in glycine. It was found by genetic crossing to the GLYD and SHMT mutants to belong to a new complementation group (Table 1-3). Glycine decarboxylase is a complex composed of four protein components: a glycine decarboxylase; a carrier of the resulting aminomethyl moiety; a protein that transfers the aminomethyl moiety to THF and releases NH3; and a dehydrogenase (see Husic et al., 1987). Because the enzyme complex is insoluble, a satisfactory system has not been developed to assay the individual components. Therefore, the particular components affected in the GLYD mutants have not been determined. All of the previous GLYD mutants that were analyzed by complementation lacked the same component. NA66 probably lacks one of the other components. 1 have tentatively designated the locus glyDZ. Two mutants, NA91 and NA126, displayed an inhibition of photosynthesis (Fig. 1-2), either no or minor labeling abnormalities, and complemented both the GOGAT and DOT mutants (Table 1-3). Both mutants had normal activities of GS (Table 1-4). Additionally, NA91 22 Table 1-3. Genetic crosses of photorespiratory mutants with characterized mutants. Tester Unknown M/F No. Crosses Color of F1 Plants _Lin9 Line C551 NA66 F 2 GREEN (sat) NA126 F 1 GREEN NA138 M 1 GREEN Se16 F 2 GREEN C5116 NA66 M 2 GREEN (glyD) NA77 M 2 YELLOW NA91 F 1 GREEN NA164 F 1 GREEN Se16 M 1 GREEN F 2 GREEN C5119 NA164 F 1 GREEN (pcoA) C5156 NA91 M 1 GREEN (dct) NA126 M 1 GREEN NA157 F 1 GREEN NA164 M 3 GREEN Se16 M 3 GREEN C5253 NA66 M 1 GREEN (stm) NA77 M 1 GREEN NA109 M l YELLOW NA115 M 4 YELLOW NA126 F 2 GREEN NA138 M 1 YELLOW 5e16 M 1 GREEN 05254 NA85 M 1 YELLOW (glus) NA91 M 3 GREEN NA94 M 3 GREEN NAIOO M 1 YELLOW NA126 F 2 GREEN NA133 F 1 YELLOW NA164 F 3 . YELLOW Eh5 M 1 YELLOW Kle M 1 YELLOW Oi6 M 1 YELLOW 05299 NA131 F 1 YELLOW _Lglu51 * M - unknown line used as male parent; F- unknown l1ne used as female 23 Table 1-4. Activities of some C2 cycle enzymes in photorespiratory mutants of Arabidopsls. GOGAT umol l4C-glu h'1 mg chl'1 WT 78 NAQO 5 SGAT GGAT pmol h‘l mg of1 WT 30 54 NA78 1 so NA79 2 51 NA91 21 48 NA126 25 49 NA157 o 59 WT + NA78: 16 ND w1 + NA79 19 ND PGP pmol h"1 g fwt'1 HT 582 3113 14 HT + 3113* 360 as pmol h'1 mg chl'1 WT 22.1 NA91 19.3 NA126 23.9 §§l§ I§§I§ HPR -1 -1 pmol h mg prot NT 41 NA91 43 NA94 41 §e16 37 * Enzymes were mixed in equal proportions to distinguish between a lack of enzyme activity or the presence of inhibitor activity. 24 had normal activites of HPR and GGAT (Table 1-4). Further analyses of these mutants are in progress. Table 1-5. Lines of photorespiratory mutants of thaljana isolated between 1984 and 1986. A, identified in screen for subtle mutants. B, identified in screen for severe mutants. The following mutants were isolated from a phosphoglucomutase deficient population: NA126, NA131, NA133, NA138, NA157 and NA164. pcoA gluS glyDl glyDZ‘ stm sat ? A. NA60 NA77 NA66 NA109 NA78 NA91 NA85 NA115 NA79 .NA126 NAIOO NA138 NA157 NA131 NA133 NA164 B. Jil3 Eh5 Se16 Kle Qifii 1”Tentative designation. Two thousand five hundred M3 lines were screened for photorespiratory' mutants with more severe phenotypes than the previously isolated mutants. The advantages of using M3 lines are i) it does not matter if the mutant dies during the screen because there is more seed of the same line; ii) The M3 progeny from a M2 plant which is heterozygous for a mutation of interest will segregate, thereby increasing the frequency of finding the mutant. Five photorespiratory mutants were isolated, of which one, designated 0113, died within two days of exposure to air. The photosynthetic gas exchange on 0113 in 50% 02 was unique among the photorespiratory mutants in that the rate was always below zero net CO2 exchange (Fig. 1-2). l4CO2 feeding resulted in 28% of the water soluble label in the strongly acidic fraction compared to 9% in the wild-type. The weak acids and the bases 25 had proportionately less label in the mutant (Table 1-2). A mutation in PGP was the most likely explanation for this result. Indeed, PGP activity was 2% of the wild-type activity (Table 1-4). Three of the other four mutants isolated in this screen were found by complementation analyses to be GOGAT mutants (Table 1-3). The fourth mutant, Se16, accumulated 56% of the 14C in the basic fraction compared to 44% for the wild-type (Table 1-2), of which 61% of the 56% was in serine, compared to 40% of the 44% for the wild-type. Se16 complemented the other mutants which accumulate glycine or serine as well as the DCT mutant (Table 1-3). Thus, Se16 appears to belong to a unique class of photorespiratory mutants. The activities of G5 and HPR are normal (Table 1-4). Further analysis of this mutant is in progress. Phenocopy of a Glycolate Oxidase Mutant In order to determine what the phenotype would be for a glycolate oxidase mutant, the specific suicide inhibitor, 2-hydroxy-3-butynoic acid, was fed to seeds, seedlings, rosettes, and leaves of wild-type Arabidopsis in air or air enriched with 2% CO2 for 6 days. Rosettes and leaves in the nonphotorespiratory atmosphere were unaffected by the inhibitor for the duration of the experiment, whereas those in air shriveled and died within 12 hours (Fig. 1-3). This result in itself would suggest that a glycolate oxidase mutant would be viable when grown in an atmosphere that suppresses photorespiration but would quickly die in a photorespiratory atmosphere. However, seeds did not germinate and seedlings died on HBA regardless of the atmosphere. This latter result indicates that either glycolate oxidase is required 26 Figure 1- 3. Phenocopy of a Glycolate Oxidase Mutant. Whole plants or detached leaves of were treated with (right) or without (left) 4 mM HBA and were placed in air enriched with 1% CO2 (upper) or air (lower). The photograph was taken after two days. 27 during seed germination and seedling growth in Arabidopsis, or the inhibitor has a nonspecific effect on seeds and seedlings of Arabjdgp§15.1f the former hypothesis is true, then glycolate oxidase mutants would be lethal. Glycolate oxidase is present at low levels in glyoxysomes of seeds that perform p-oxidation of fatty acids, but it is not known to have a function (Huang et al, 1983). Tolbert (personal communication) has speculated that glycolate oxidase may function in w~oxidation of lipids in peroxisomes, as it does in mammalian kidneys. To determine if the inhibitory effect of HBA on germination is a general phenomenon, other species were examined. Seeds of four species were imbibed with 3 mM HBA in darkness. [Two of these species, Arabidopsis and turnip, store oil in their seeds and therefore rely on B-oxidation for germination and seedling growth. The other two species, petunia and barley, presumably store starch in their seeds and are not dependent on p-oxidation. There was 100% inhibition of germination for Arabiggpsis and petunia, whereas germination of turnip and barley was unaffected (Table 1-6). seedling growth on 3 mM HBA in darkness was inhibited for Arabidopsis, petunia and turnip, but not barley. This result suggests that the effect of HBA on germination and seedling growth is not specific to species which perform B-oxidation. Greening of seedlings in light with HBA was inhibited in Arabidopsis, petunia and turnip, but not barley. Barley probably did not take up the HBA, since greening should be inhibited by HBA in all 03 species. In summary, these results suggest that glycolate oxidase mutants may not be viable in a nonphotorespiratory atmosphere. However, the cause of the inhibition of germination and seedling growth by HBA should be investigated in 28 more detail before drawing firm conclusions. It was not possible to continue this study because the supply of HBA was limited and efforts to synthesize more by a local organic synthesis laboratory failed. Table 1-6. Germination of seeds of various species on 2-hydroxy- 3-butynoic acid. Germination was scored four days after imbibition. Controls + 3 mM HBA £033.12; 11.139121111011131 11 mm Arabidopsis 41 98 46 O Turnip 27 96 27 '100 Barley 16 50 16 69 Petunia 47 96 48 0 Southern Blot. In an attempt to determine if Arabidopsis has more than one gene encoding GOX, digests of total DNA were probed with 3zP-labeled GOX cDNA from spinach. Digestion with the restriction enzymes HindIII, XbaI and XhoI yielded a single band on autoradiograms of approximately 5.0 to 5.2 KB. A triple digestion with these three enzymes resulted in a single 5.0 KB band. Digestion with EcoRI resulted in two bands between 1 and 2 KB. The gene for GOX must be at least 1.6 KB without introns based on the size of the complete cDNA. If introns are present, it may be much larger. Two reasonable interpretations of these results are that either there is only one gene for glycolate oxidase which contains an EcoRI site, or there are two genes in tandem. 29 DISCUSSION The purpose of this study was to determine why certain loci are not represented in the collection of Ambjfimj; photorespiratory mutants. Mutants for six loci have been described (Somerville and Ogren, 1982a), and theoretically, it should have been possible to find mutants for at least five other loci. The rationale behind this study was that either the missing mutants belong to one or more phenotypic classes so that they were missed in the screen employed by Somerville and Ogren, or the mutants are extremely rare or do not exist (Table 1-1). Phenocopies of a GOX mutant died within one day in air (Fig. 1-3), suggesting that a GOX mutant would have a much more severe phenotype than the existing photorespiratory mutants. However, it is unlikely that novel photorespiratory mutants with unique phenotypes are present in the mutagenized populations since the screening of 10,000 M2 plants for mutants that are slow at displaying deleterious symptoms in a photorespiratory atmosphere, and of 2,500 M3 lines for mutants that die rapidly in a photorespiratory atmosphere, did not result in a new photorespiratory mutant with a unique phenotype. However, four new mutants were discovered that have phenotypes indistinguishable from the previously isolated mutants. The complementation between the four mutants has not been determined, but the finding of these mutants indicates that mutation frequency varies considerably between loci. The recovery of one mutant from each screen with a phenotype that the screen was designed to detect is evidence that the screens were reliable. The new mutant line, NA94, has a more subtle phenotype than the other mutants, but photorespiratory metabolism is not affected by the mutation. Complementation analysis with two other high 30 COz-requiring mutants which have normal photorespiratory metabolism (chapter 2) is in progress. The new mutant line Ji13 has a more severe phenotype than the other mutants. This mutant has 2% of the wild-type activity of PGP. A mutant line deficient in PGP that was described previously (Somerville and Ogren, 1979) has 4% of the wild-type activity. Though the phenotype of the previously identified PGP mutant is more severe than that of the other photorespiratory mutants, it is not as severe as that of the new mutant, probably because it has twice the PGP activity. Clearly, any mutants that are indistinguishable from wild-type would not have been detected in this study. It is apparent from the HBA feeding that a GOX mutant would not have this phenotype, but it is possible that an HPR mutant would. Recently, a second isozyme of HPR, which is imunologically distinct from the peroxisomal HPR, was identified in spinach, pea and wheat. It is located outside of the peroxisome and preferentially uses NADPH as cofactor (Kleczkowski and Randall, 1988). Considering that the peroxisomal membrane is not an absolute barrier for hydroxypyruvate (Liang and Huang, 1983), it is reasonable to expect that this isozyme may contribute to the C2 pathway by reducing hydroxypyruvate that leaks from the peroxisome. Thus, a mutant deficient in one HPR isozyme may not be completely blocked in hydroxypyruvate utilization and therefore would not display the phenotype of a photorespiratory mutant. Most plant species have two isozymes for GS (McNally et al, 1983). In this case, the cytosolic isozyme cannot substitute for the chloroplastic isozyme in photorespiratory NH3 assimilation. This was demonstrated with a mutant of barley lacking chloroplast GS (Kendall et 31 al 1986). A possible explanation for the lack of a GS mutant in Arabidopsis is that Arabidgpsis may not have a cytoplasmic isozyme of GS. The mutant would have no means of assimilating NH3 in nonphotosynthetic tissues and would therefore be lethal even in a nonphotorespiratory atmosphere. The GS isozyme composition in Arabidopsis has not been examined. Because most of the C2 cycle enzymes and transport proteins are not known to have nonphotosynthetic functions, this would be an unsatisfactory explanation for the absence of several photorespiratory mutants. Catalase is the only enzyme besides GS that is known to be required in nonphotosynthetic tissue. Peroxisomal CAT detoxifies H202 produced by the fatty acleoA oxidase reaction in the glyoxylate cycle of oil seeds as well as that produced by the GOX reaction in the C2 cycle. If the same isozyme performs both of these functions, a CAT photorespiratory mutant in Arabidopsis would die during germination. Barley, on the other hand, does not rely on the glyoxylate cycle for germination and seven mutant lines have been identified to date with deficiencies in peroxisomal CAT (Blackwell, 1988). Glycolate oxidase was shown to be present in glyoxysomes of seeds and in unspecialized peroxisomes of roots, stems, buds and etiolated leaves of some plants (Huang et al, 1983). This may be due simply to a low constitutive expression of the gene since it has no known function other than in the C2 cycle. However, the observation in this study that HBA inhibits seed germination in Arabidopsis may imply that GOX has a vital function in this process. HBA works by mimicking glycolate as a substrate so that it is enzymatically converted to a compound which irreversibly binds and inactivates the flavin cofactor (Jewess et 32 ‘ al., 1975). If HBA inhibits another enzyme, that enzyme would have to have both a similar active site as GOX and a flavin cofactor. Further investigations are necessary to determine the mode of inhibition of HBA in seeds. Arabidopsis has a small genome with very little repetitive DNA compared to other plant species (Pruitt and Meyerowitz, 1986). Most genes can be expected to occur as single copies except for those which encode regulatory enzymes or very abundant proteins. Enzymes such as most of those in the C2 cycle would not be expected to have multiple genes. It is possible, however, that some do, and if so, this would account for the lack of certain mutants. There are two reasonable interpretations of the results of the Southern blot of Arabidopsis genomic DNA probed with GOX cDNA from spinach. Either Arabidopsis has one gene for GOX which contains one EcoRI site, or there are two genes in tandem with less than 5 KB between the start of the first gene and the end of the second gene. The HindIII, XbaI and XhoI fragments would span the two genes, and there would be an EcoRI site between the two genes or within one gene. In either case, there must be two clusters of HindIII, XbaI and XhoI sites since a triple digestion with these three enzymes did not yield a band much smaller than the individual digestions. The isolation and characterization of a genomic clone for GOX from the Arabidopsis lambda sep6 library is in progress. It should be possible to determine conclusively the number of GOX genes in Arabidgpsis with this clone. If GOX is only required for photorespiration and if it has only one gene, then the question why there is no GOX mutant still remains. The answer may be in the susceptibility of different genes to 33 mutagenesis. Mutation frequency within a phenotypic class varies considerably. For example, there are mutations at three loci that result in long hypocotyls when seedlings are grown in light. Of the EMS generated mutations, one was at the h1;1 locus, three at the by;z locus and nine at the hy;3 locus (Koornneef et al, 1982). Differences were also observed in Arabidopsis with the gibberellin sensitive and glabra mutants (Koornneef et al., 1982). No explanation has been offered for these differences, but Koornneef et al (1982) rejected the notion that they may be due to differences in DNA content per gene. The photorespiratory mutants also have variable mutation frequencies (Table 1-7). Fifty-six percent of the photorespiratory mutations in Arabidopsis are at the glufi locus, while 2% are at the dct locus. The finding in this study of NA66, NA91, NA126, and Se16, novel mutants with similar phenotypes to the other photorespiratory mutants, supports the hypothesis that there is a bias in the mutagenesis. It is possible that a different mutagen would yield different mutation frequencies than EMS at given loci. Mutagen specificity has been known to occur in certain cases. For example, 93 site-specific mutations at the ad-8 locus of Negrgspgra were mapped to determine the precise base pair that was altered (see Fincham and Day, 1965). Of the 15 which were generated by S-bromodeoxyuridine mutagenesis, all mapped to the same site on the gene. All of the other mutagens used in this study, including EMS, resulted in a random distribution of mutagenized sites. However, some of these same mutagens have exhibited site specificity in other systems (Fincham and Day, 1965). In a study on mutation frequencies in Arabidopsis, similar frequencies were obtained for a given locus irrespective of the mutagen used (Koornneef et al, 1982). 34 Azide was used to generate the photorespiratory mutants in barley, where the GOGAT mutant is also most common (Table 1-7). Thus, trying a different mutagen may not solve the problem. Table 1-7. Frequencies of photorespiratory mutants in 81391101515 thaljana and Hordggm yglgarg. The Arabidopsis mutants were generated with EMS, and the barley mutants were generated with azide. Deficient A. thaliana LL. xulsare Ewan: PGP 2 3 CAT 0 7 SGAT * 11 1 Glycine to Serine 27 2 G5 0 13 GOGAT 58 13 OCT 2 3 . Unidentified 3 TOTAL 103 42 * Of the glycine accumulating mutants of Arabidopsis, genetic crosses have been performed for 15 out of 27. Of the 15, Glare deficient in GLYD, 8 are deficient in SHTM, and 1 belongs to a third complementation group. The precise biochemical lesions of the two barley mutants and their complementation is not known. In summary, two reliable screens were developed to identify photorespiratory mutants whith phenotypes in air more subtle or more severe than the previously isolated mutants. A few conclusions can be drawn from the results obtained with these screens. First, certain mutants are probably not missing because they have a uniquely subtle phenotype. Second, the low frequency of PGP mutants may be due to the severe phenotypes of these mutants, rather than a biased mutagenesis. It is evident from HBA feeding experiments (Fig. 1-3) that if GOX mutants exist, they too would have a severe phenotype. It was not possible to rule out the possibility that GOX has a second vital 35 function or that it has two genes. If either is true, it would explain the absence of a GOX mutant. The finding of novel photorespiratory mutants with typical phenotypes indicates that some mutants were not identified previously because they occur at a very low frequency relative to the mutants described by Somerville and Ogren. Other factors besides the intrinsic nature of a base pair may play a role in determining the site preference of a mutagen, thereby resulting in variable mutation frequencies between loci. LITERATURE CITED Anderson LE 1971 Chloroplast and cytOplasmic enzymes. 11. Pea leaf triosephosphate isomerases. Biochem BiOphys Acta 235: 237-244 Artus NN, CR Somerville 1988 A mutant of Arabidopsis thaliana that displays chlorosis in atmospheres enriched with C02. Plant Physiol (in press Blackwell RD, AJS Murray, PJ Lea, AC Kendall, NP Hall, JC Turner, RM Nallsgrove 1988 The value of“ mutants unable to carry out photorespiration. Photosynthesis Research (in press) Caspar T, SC Huber, C Somerville 1985 Alterations in growth, photosynthesis, and respiration in a starchless mutant of Arabidopsis (L. ) deficient in chloroplast phosphoglucomutase activity. Plant Physiol 79:11-17 Chastain CJ 1985 Photosynthesis inhibition in the Arabidopsis thaliana photorespiration mutants. Ph.D. thesis. University of Illinois, Urbana Fincham JRS, PR Day 1965 Fungal Genetics. second edition, Botanical Monographs, WO James, ed. vol4 FA Davis company, Philadelphia, PA Haughn GW, CR Somerville 1986 Sulfonylurea-resistant mutants of Arabidopsis thaliana. Mol Gen Genet 20: 430-434 Haughn GW and CR Somerville 1987 Selection for herbicide resistance at the whole-plant level. In: Biolechnology in agricultural chemistry, HM Le Baron, RO Mumma, RC Honeycutt, JH Deusing, eds., American Chemical Society, Washington D.C., pp.98-198 Huang AHC, RN Trelease, TS Moore 1983 Plant Pgrngsgmes, Academic Press, New York 36 Music OH, HO Husic, NE Tolbert 1987 The oxidative photosynthetic carbon or C2 cycle. CRC Crit Rev Plant Sci 5: 45-100 Jewess PJ, MW Kerr, DP Whitaker 1975 Inhibition of glycollate oxidase from pea leaves. FEBS Letters 53: 292-296 Kleczkowski LA, DD Randall 1988 Purification and characterization of a novel NADPH(NADH)-dependent hydroxypyruvate reductase from spinach leaves. Biochem J (in press) Koornneef‘ M, LWM Dellaert, JH van der Veen 1982 EMS- and radiation-induced mutation frequencies at individual loci in Agaaigapaia thaljana Heynh. Mutation Research 93: 109-123 Leutwiler LS, BR Hough- Evans, EM Meyerowitz 1984 The DNA of Arahigaaaia thaljana. Mol Gen Genet 194: 15- 23 Liang Z, AHC Huang 1983 Metabolism of glycolate and glyoxylate in intact spinach leaf peroxisomes. Plant Physiol 73: 147-152 Maniatis T, EF Fritsch and J Sambrook 1982 Mglaggla: glgnjng, a laboratory manual, Cold Springs Harbor Laboratory, Cold Springs Harbor, New York McNally SF, B Hirel, P Gadal, AF Mann, GR Stewart 1983 Glutamine synthetase of higher plants. Evidence for a specific isoform content related to their possible physiological role and their compartmentation within the leaf. Plant Physiol 72: 22-25 Miflin BJ, PJ Lea 1982 Ammonia assimilation and amino acid metabolism. In Ngglajg Agiga and Ergtejn jg Elanta 1,, D Boulter and B Partheir, eds. , Encyc Plant Physiol vol 14A Springer- Verlag, New York pp 3- 64 Pruitt RE, EM Meyerowitz 1986 Characterization of the genome of ,Anapidanaia thaliana. J Mol Biol 187: 169-184 Somerville CR, WL Ogren 1979 A phosphoglycolate phosphatase deficient mutant of Arabidaagla. Nature 280: 833-836 Somerville CR, WL Ogren 1980 Photorespiration mutants of ALBDIQQDSIS thaliana deficient in serine-glyoxylate aminotransferase activity. Proc Natl Acad Sci 77: 2684-2687 Somerville CR, WL Ogren 1980b Inhibition of photosynthesis in A_aaidgp§1§ mutants lacking leaf glutamate synthase activity. Nature 286: 257- 259 Somerville CR, WL Ogren 1981 Photorespiration-deficient mutants of Arabidgpsia thaliana lacking mitochondrial serine transhydroxymethylase activity. Plant Physiol 67: 666-671 Somerville CR, WL Ogren 1982a Genetic modification of photorespiration. Trends Biochem Sci 7: 171-174 37 Somerville CR, WL Ogren 1982b Isolation of photorespiration mutants in Aratiganaia thaliaaa. In Methods in Chloroplast Biology. M Edelman, R Hallick, NH Chua, eds., Elsevier, New York, pp. 129-138 Somerville CR, WL Ogren 1982c Mutants of the cruciferous plant Arabidapaia lacking glycine decarboxylase activity. Biochem J 202: 373-380 Volokita M, CRS Somerville 1987 The primary structure of spinach glycolate oxidase deduced from the DNA sequence of a cDNA clone. J Biol Chem 262: 15825-15828 Wallsgrove RM, JC Turner, NP Hall, AC Kendall, SWJ Bright 1987 Barley mutants lacking chloroplast glutamine synthetase- biochemical and genetic analysis. Plant Physiol 83: 155-158 Wallsgrove RM, PJ Lea, BJ Miflin 1979 Distribution of the enzymes of nitrogen assimilation within the pea leaf cell. Plant Physiol 63: 232-236 CHAPTER 2 A Mutant of Araaidgnaia thaliana that Exhibits Chlorosis in Air but not in Atmospheres Enriched in CO2 38 39 INTRODUCTION A relatively large number of mutants of higher plants have been described which will only grow in atmospheres enriched with about 1% C02. All but one of these are mutants of Arabidopsis thaliana and flagdaam yalgaga with defects in photorespiratory metabolism (Somerville, 1986; Blackwell et al., 1988; chapter 1). These mutants are nonviable in air but they are able to grow normally in an atmosphere enriched with CO2 because ribulose bisphosphate oxygenase activity is suppressed and carbon is not diverted into the photorespiratory pathway. In addition to the mutants with defects in photorespiration, a high COz-requiring mutant was isolated that was unaltered in photorespiratory metabolism, but that had severely reduced levels of activated Rubisco j_n_ yjya (Somerville et al., 1982). Characterization of the biochemical lesion in this mutant led to the discovery of Rubisco activase, a protein required for the Ln yjyg activation of Rubisco (Salvucci et al., 1985). In this report, another mutant of At thaliana is described that requires elevated levels of atmospheric C02 for growth but has normal photorespiratory metabolism. Although the precise biochemical lesion in this mutant is not known, the properties of the mutant suggest the existence of a previously unknown role for C02. A preliminary report on this mutant has appeared (Artus and Somerville, 1987). ' 40 MATERIALS AND METHODS Plant Materials and Growth Conditions. The lines of mu; thaliana (L.) Heynh. described here were descended from the Columbia wild-type. The mutant line C5208 was isolated from an M2 population of plants which had been mutagenized with ethyl methane sulfonate as described previously (Haughn and Somerville, 1986). To reduce the number of background mutations, the mutant line was advanced for several generations by self-fertilization and then backcrossed to the wild-type and reselected in the F2 generation two times. The glycine decarboxylase deficient line C5116 has been described (Somerville and Ogren, 1982b). The mutant line CC126 was generously provided by C. J. Chastain and M. L. Ogren. Except as noted otherwise, the plants were grown on an artificial medium irrigated with mineral nutrients (Haughn and Somerville, 1986) 2 s'l) in air enriched at 23 C in continuous illumination (80-100 uE m' with 2% (v/v) C02. The C02 concentration in the chamber was maintained with a C02 regulator (Forma Scientific). For the growth of plants in low 02 and high C02, a gas mixture (1500‘pl 1'1 C02, 1% 02, balance N2) flowed continuously at a rate of 1 l min'1 through a 2 l Plexiglas chamber which contained a 1 l pot with 8-10 plants of each genotype. The Plexiglas chamber was contained within a larger plant growth chamber which provided the standard conditions for temperature and light. Tissue Culture Methods. The media for induction and maintenance of callus have been described (Haughn and Somerville, 1986). Seeds were 4l surface sterilized and germinated on solid callus induction medium. After three weeks, the resulting callus cultures were fragmented and transferred at a density of 1 g per plate to callus maintenance media overlaid with Whatman #1 filters. To measure callus growth rate, the filter papers with callus were weighed under sterile conditions at periodic intervals. CO2 Exchange and 14(:02 Labeling. Photosynthetic C02 exchange was measured on whole plants at 23 C and 200 pE 111'2 s'1 with an infrared gas analyzer (Analytical Development Co., Hoddesdon, England). Photoassimilation of 14C02 (1 mCi mmol’l) was performed in 330 pl 1'1 C02, 50% 02, balance N2, at 24 C. For labeling plants during the induction phase of photosynthesis, plants were equilibrated with the atmosphere in darkness, then label was added at the start of illumination. The incorporation period lasted 23 minutes. To label during steady state photosynthesis, plants were illuminated for 30 minutes prior to the addition of label. The incorporation period lasted 20 minutes. Methods for gas exchange measurements, 14CO2 feeding in the light, and analysis of the water soluble products of 14C labeling have been described in detail (Somerville and Ogren, 1982a). For dark 14C02 assimilation, a 12.5 cm pot with approximately 20 mutant and 20 wild-type plants was enclosed in a covered Plexiglas chamber that was in line with a pump and a flask containing 8% phosphoric acid. After 20 minutes of preequilibration, the system was closed and sodium [14C]bicarbonate was injected into the acid to yield a specific activity of 3.5 mCi mmol'1 and a C02 concentration of 390 pl l'l. Sixty minutes later, additional sodium [14C]bicarbonate was injected to 42 bring the specific activity up to approximately 5 mCi mmol'l. After another 30 min, the plants were harvested and processed as for plants labeled in light. Leaf Fluorescence and Thylakoid Electron Transport. Fluorescence was measured on detached leaves after a one hour dark incubation using a portable fluorometer (Richard Brancker Research Ltd). The voltage output of the photodiode was monitored on a Nicolet model 206 digital oscilloscope. Electron transport was measured on thylakoids that were prepared as described (McCourt et al., 1987) except that 0.1% (w/v) BSA was included in the grinding buffer. The assays (whole chain, PSI and P511) were performed as described (McCourt et al., 1987) except that the standard assay medium contained 0.02 M Hepes-KOH (pH 7.9), 0.3 M sorbitol, 0.01 M NaCl, 2 mM M9012, 2.5 mM EDTA and 0.1% BSA. Carotenoid Measurements. Total carotenoids were quantified in 80% acetone extracts by monitoring the absorbance at 480 nm and correcting for the absorbance by chlorophyll (Kirk and Allen, 1965). Carotenoids in acetone extracts were separated by reverse phase HPLC on a 4.6 X 20 cm C18 column (Varian, MCH-S-n-cap) using a linear gradient of 90% 1 aqueous methanol to 100% ethylacetate over 20 min at 1 ml min' (Casadero et al., 1983). Enzyme Assays. Carbonic anhydrase was assayed according to Wilbur and Anderson (1948) by monitoring the change in pH during the course of the reaction. Leaf extracts were prepared by homogenizing 50 mg ml'1 leaf material in 20 mM sodium phosphate (pH 7.0), 1 mM EDTA and 1 mM 43 ascorbic acid. A unit of activity is defined as follows: U - 10[(E/N)-1] where E and N are the times required for the enzymatic and nonenzymatic reactions, respectively, to change the pH of the medium from 8.0 to 7.0. Catalase was assayed by monitoring 02 evolution from H202 with an oxygen electrode. Leaf material was homogenized on ice in 0.05 M potassium phosphate (pH 7.2) at a fwt to volume ratio of 50 mg ml'l. The extracts were centrifuged for 10 min at 10,0009. The assay medium contained 0.05 M potassium phosphate (pH 7.2), 0.2 % (v/v) H202 and extract. Superoxide dismutase was assayed by measuring the inhibition of the reduction of nitro blue tetrazolium by superoxide as described (Gianopolitis and Ries, 1977). Leaf material was homogenized on ice in 0.1 M potassium phosphate (pH 8.3), 0.5 mM EDTA at a fresh weight to volume ratio of 50 mg ml.1 and centrifuged 15 min at 13,0009 before assay. Dehydroascorbate reductase was assayed spectrophotometrically by measuring the change in absorbance at 265 nm due to the production of ascorbate (Nakano and Asada, 1981). Extracts were prepared by homogenizing leaf material on ice in 0.05 M potassium phosphate (pH 7.0), 0.1 mM EDTA, centrifuged for 10 min at 18,0009, and dialyzed 4 hours against the same buffer. Glutathione reductase was assayed spectrophotometrically by measuring the change in absorbance at 340 nm due to oxidation of NADPH. Leaf material was homogenized on ice in 0.1 M Hepes-KOH (pH 7.5), 3 mM EDTA and 2 mM GSSG. The GSSG was essential to prevent irreversible inactivation of the enzyme. The extract was filtered through miracloth and centrifuged 10 min at 11,0009. The assay medium contained 0.05 M Hepes-KOH (pH 8.0), 1 mM GSSG, 1 mM dithiothreitol, 0.24 mM NADPH and extract. 44 - SOD Isozymes. To resolve isozymes of superoxide dismutase in mutant and wild-type Anabidapaig, leaf extracts were prepared by homogenizing 500 mg leaf material in 3 ml 0.1 M potassium phosphate (pH 7.8), 0.1 mM EDTA and clarifying by centrifugation for 15 min at 13,0009. Samples containing 300 pg protein were electrophoresed on an (7.2%) acrylamide gel and the regions of SOD activity were localized (Beuchamp and Fridovich, 1971). Antioxidant Assays. Assays for glutathione and ascorbic acid were performed as described (Law et al., 1983) on acid leaf extracts prepared by homogenizing 100 mg leaf material in 1.5 ml of either 5% (w/v) sulfosalicylic acid and 6 mM EDTA for glutathione, or 3.5% (w/v) trichloroacetic acid for ascorbic acid. For the isolation of a-tocopherol, leaves were extracted in methanolzchloroform (2:1, v/v), filtered, and partioned against water. The chloroform layer was dried onto alumina powder. The latter steps of the isolation, beginning with the transfer of the dried powder to a neutral A1203 chromatography column, was performed as described by Wise and Naylor (1987). a-Tocopherol was then quantified by HPLC as for the carotenoids except detection was at 280 nm. Other Measurements. Chlorophyll was measured in 95%1 ethanol (Wintermans and DeMots, 1965) or in 80% acetone (MacKinney, 1941). Protein was measured according to Bradford (1976). Leaf area was measured with a Licor portable area meter (Lambda Instruments Corporation). Hydrogen peroxide was measured in leaf extracts by a colorimetric assay (Patterson et al., 1984). 45 RESULTS Phenotype of the Mutant. The mutant line 05208 was isolated by screening a mutagenized population for individuals which grow in 1% CO2 but become chlorotic in air. When grown in air enriched with 2% (v/v) C02, the mutant is slightly lighter in color than the wild-type but otherwise appears healthy. However, the leaves become chlorotic and develop necrotic regions when the mutant is transferred to ordinary air (Fig. 2-1, 2-2). Chlorophyll and carotenoid contents of mutant plants grown in 2% CO2 were 80% of the wild-type level. After 5 days of continuous illumination in air, the chlorophyll and carotenoid content of the mutant was reduced by 60% and 40%, respectively, but was unchanged in the wild-type (Fig. 2-2). Likewise, chlorophyll was reduced 60%, from 1.4 mg g fwt'l 1 to 0.6 mg g fwt' , after 5 days in air on a 16 hour photoperiod. In order to determine if the bleaching of the mutant in air was dependent on light intensity, the chlorophyll contents of mutant and wild-type were compared after four days of exposure to various light intensities in normal atmospheric conditions (Fig. 2-3). Chlorophyll was reduced by approximately 40% in both mutant and wild-type in '2 s'l) had no effect on darkness. The lowest light intensity (3O pE m the chlorophyll content of the wild-type but resulted in a 40% reduction of chlorophyll in the mutant. The mutant exhibited a 60% loss of chlorophyll in the highest light intensity (270 pE m '2 s'l). Since varying the light intensity from 0 to 270‘uE m'2 5'1 had little if any effect on the chlorophyll content of C5208, it cannot be concluded that light is required for the injurious effects of normal 46 Figure 2-1. Mutant line C5208 and wild-type Arabidgpsis grown in air containing 2% C0 and transferred to air for six days. (A) wild-type in 2% C02; (B) wild-type in air; (C) C5208 in 2% C02; (D) C5208 in air. Chlorophyll (mg/9 fwt) A u E U! \ O'l E v E O C d) a O h m 0 O- - 1 L L I I I O 1 2 3 4 5 Time in Air (days) Figure 2-2. Chlorophyll [A] and carotenoid [B] content in wild-type (DI) and mutant (0.) Arabidopsis following transfer of plants from 2% CO7 to air (solid- lines). Controls (closed symbols, dashed lines) remained in 2% C02. 48 0.03% co2 2.0-l Chlorophyll( mg g fwt") E; l Light lntensltyULE 01-2 5'1) Figure 2-3. Effect of light intensity on chlorophyll content of wild-type (shaded) and mutant (open) AW 4 days after transfer from 2% CO2 to air (0.03% C02). 49 atmospheric conditions. However, the bleaching caused by the mutation is not additive with the normal reduction of chlorophyll that occurs following transfer of plants to darkness. In addition, necrotic lesions are not produced in darkness as they are in light. Mutant plants kept in high C02 at light intensities ranging from 75 to 270‘pE '2 s'l maintained 1.5 mg Chl g fwt'1 over a 9 day period. m To determine whether the phenotype is expressed in nonphotosynthetic tissue, the effect of atmospheric CO2 on calli growth and primary root elongation was examined. In air, calli of mutant and wild-type had growth rates of 98 1 21 and 85 i 22 m9 100 mg'1 (1'1 (means 1 5.0., N-4 plates), respectively. Similarly, after 4 days in air the mean lengths (N-lO) of primary roots for mutant and wild-type seedlings were 5.7 1 0.9 and 5.3 i 0.6 (5.0.) mm, respectively. These results suggest that the mutant phenotype is expressed only in photosynthetic tissue. In other high COz-requiring mutants of Arabidopsis, loss of chlorophyll was readily reversed by returning plants that had become chlorotic in air to an atmosphere enriched with C02 (Chapter 1). By contrast, illumination of this mutant in air for as little as 1 day leads to irreversible injury to the vegetative tissue. Once the leaves become chlorotic they do not regreen when placed back in high C02. However, the plants survive by growing new leaves. In this respect the mutant line C5208 differs from the previously described mutants which require high levels of CO2 for growth with the exceptions of two lines (CC112 and CC126) identified by Chastain (1985). Genetic Analysis. The genetic basis for the mutant phenotype was 50 examined by crossing the line 05208 with wild-type. The resulting F1 plants were able to grow normally in air. The F2 progeny resulting from self-fertilization of the F1 heterozygotes were scored for the mutant phenotype by growing the F2 plants for 16 days in 2% CO2 then transferring them to air. Of 260 F2 individuals examined in this way, 198 grew normally in air and 62 turned chlorotic and died. This fits the 3:1 hypothesis (XZ- 0.13; P>0.70), indicating that the phenotype is due to a single, recessive nuclear mutation at a locus I have designated hgrl for ' high 902 requiring". Chastain (1985) has isolated a mutant line (CC126) that has a similar phenotype to C5208 except that CC126 becomes chlorotic in air at a significantly slower rate than C5208. The biochemical basis of the lesion in C0126 is not known. F1 hybrids obtained by crossing 05208 with CC126 had the mutant phenotype indicating that the two mutant lines carry defective alleles at the same locus. I have designated the alleles [15111;], and 335131;: for C5208 and CC126, respectively. Effects of Oxygen Concentration. The flux of carbon into the photorespiratory pathway can be suppressed by high CD2 or low 02. Thus, if the high C02 requirement of the mutant were related to a defect in the C2 cycle, it should be possible to grow the mutant in an atmosphere in which the oxygen concentration is reduced to the point that the ratio of CO2 to 02 is equal to or greater than that in air containing 2% C02. Therefore, the phenotype of plants that were grown in air containing 2% C02 then transferred for 2.5 days to an atmosphere of 0.15% C02, 1% 02, balance N2 was examined. This yields a COz/O2 ratio 51 of 0.15 compared to a ratio of 0.10 in 2% C02. The combination of 0.15% CO2 and 1% 02 did not prevent the development of chlorosis in the mutant (Fig. 2-4, Table 2-1). As a control, to ensure that the atmospheric conditions in the chamber were maintained under conditions which would effectively prevent the flux of carbon into the photorespiratory pathway, a line of 8119913192513. with a defect in glycine decarboxylase activity (Somerville and Ogren 1982b) was included. The observation that this mutant remained green and healthy in the low oxygen and high C02 chambers throughout the course of the experiment indicated that the atmospheric conditions were suitable (Fig. 2-4). Table 2-1. Chlorophyll concentration in wild-type and mutant Arabidopsis grown in air containing 2% CO then transferred for 2.5 days to air or 1% 0 Values re means i so (N=2). 2 Atmospheris_gonditlens thoreehxll_senteet -1 -1 CO2 (ul 1 ) o2 (%) mg g fwt WT C8208 20000 21 2.0 i 0.1 1.9 i 0 1 350 21 2.2 i 0 1 1.1 i 0 1 1500 1 1.3 i 0 1 1.2 i 0 1 Effects on Photosynthetic CO2 Assimilation. Since a major requirement for CO2 is in photosynthesis, the possibility that high CO2 is needed to enhance Rubisco activity in the mutant was examined. It is 52 Figure 2-4. Effect of 3 atmospheric conditions on wild-type (left triads), C5208 (right triads), and a glycine decarboxylase deficient line, C5116 (bottom triads). All plants were grown in 2% C0 for days, then transferred to 0.15% C0 , 1% 0 , balance N (top right); air (bottom); or maintained in air enriched with 2% C02 (top left). 53 conceivable that if there were a high resistance to CO2 diffusion in the mutant, or if there were a biochemical impairment in carbon fixation, high C02 might be required to maintain an adequate level of (202 fixation by Rubisco. These possibilities were examined by measuring the rate of photosynthetic gas exchange by wild-type and mutant plants grown in high C02. The rate of photosynthetic COZ-fixation in air was similar in wild-type and mutant when expressed on a protein, leaf area or fresh weight basis (Table 2-2). When expressed on a chlorophyll basis the photosynthetic COZ-fixation rate was higher in the mutant since the mutant was slightly chlorotic (Fig. 2-2). The rate of gas exchange for the mutant in 50% 02’ 330 ppm C02, balance N2 remained stable at 850 ug CO2 mg Chl"1 h"1 during a continuous thirteen hour measurement. Thus, it appears that whatever leads to the injurious effects of air in the mutant does not directly affect the activity of the C3 cycle or the diffusion of C02 t0 the chloroplast stroma. The interconversion of C02 and H2C03 is catalyzed by carbonic anhydrase. A defective carbonic anhydrase in a higher plant may, in principle, result in a requirement for elevated C02. However, carbonic anhydrase activities in leaf extracts were 210 U mg prot'1 for the mutant and 220 U mg prot'1 for wild-type. (See Methods section for definition of U.) Labeling Studies. Since C02 exerts an effect on the phenotype, it seemed possible that CO2 metabolism was altered in the mutant. Therefore, the products of 14(:02 fixation in both light and dark conditions were examined. Wild-type and mutant plants were labeled 54 see « cc.” as s as“ o. a ma can « ccs.. ccmmu can a ace“ s. « ~_~ a s as com « ccm.¢ h: :8 9.. NS Ex 7.. 733...... a... «8 .1 7.. «-5 «S a; 7.. 7:: 3 NS 9‘ 52; .Anuz. am M menus «as mm=—a> .m_m:c=c.sag< “caszz s=a «gau-=__= co macsgoxu mam o_.m=.=>mo.o=a. N-~ «pack 55 h 14C02 under photorespiratory conditions (330 pl l'1 C02, 50% 02, wit balance N2) either during induction of photosynthesis or during steady state photosynthesis. The distribution of label revealed no difference between mutant and wild-type (Table 2-3). Table 2-3. Distribution of 14C label in water soluble products of C0 photoassimilation by wild-type and mutant Arabidopsis. Labeling was performed either during the induction of photosynthesis (0 to 23 min after illumination) or during 20 min of steady state photosyn- thesis. Values are averages of two samples. ___Indusj_i.on__ 4mm Fraction in cszoa WT C8208 % of total incorporation Basic 48 45 44 48 Neutral 24 25 21 23 (sucrose) Acid 1 6 4 9 6 (organic acids) Acid 2 14 15 15 11 (acid phosphates) Acid 3 6 7 6 7 (sugar bisphosphates) Mutant and wild-type plants were labeled with 14 f 14 C02 in the dark for 1.5 hours. The average incorporation o 602 in mutant plants was 278 dpm pg chl'1 compared to 232 dpm pg chl.1 in wild-type plants. The 20% higher incorporation in the mutant is due to the 15% reduction in chlorophyll. The distribution of label between fractions was the same for mutant and wild-type, with 78% of the label in the basic fraction. Thus, there is no obvious quantitative defect in any of the major leaf carboxylases that operate in the absence of light. 56 Effects on Electron Transport Bicarbonate is required for electron transport on the reducing side of PSII (Khanna et al., 1977). If thylakoids are depleted of bicarbonate, electron transport between QA and QB is reversibly inhibited. It has been proposed that bicarbonate binds to the QB protein or a nearby protein causing a conformational change in QB (Blubaugh and Govindjee, 1986). In bicarbonate-depleted thylakoids, the room temperature fluorescence induction transient rises rapidly, resembling a fluorescent transient in the presence of the 0A to QB electron transport inhibitor, DCMU (Vermaas and Govindjee, 1982). In order to examine the possibility that the mutant was altered in bicarbonate-stimulated electron transport, fluorescence induction of detached leaves was monitored at ambient C02 in the absence of DCMU. For 13 out of 15 mutant leaves, the rise in fluorescence was slower than that for 12 wild-type leaves examined (Fig. 2-5). This result suggests that the bicarbonate binding site is not altered in the mutant. This conclusion is supported by the observation that rates for photosystem II electron transport measured from H20 to 2,6-dimethylbenzoquinone were similar in the wild-type and mutant (Table 2-4). Similarly, there was no apparent difference in the rates of whole chain (H20 to methylviologen) or PSI (TMPD to methylviologen) uncoupled electron transport between wild-type and mutant (Table 2-4). Ability to Detoxify Reactive Oxygen And H202 Levels in Leaf Extracts. The tissue damage displayed by 65208 in air is suggestive of the production of reactive oxygen species. A mutant of barley that lacks 90% of the leaf catalase activity exhibits a similar phenotype, i.e., brown lesions, bleaching, and ultimate death of the leaves, (Kendall et 57 Fluorescence 1 O 2 4 6 8 Time (see) Figure 2- 5. Typical fluorescence induction kinetics of leaves from wild- -type (-—) and mutant (-....-) Anabjfiopjjj 58 Table 2-4. Electron transport rates of thylakoids from wild-type and mutant Arabidopsis. Values are means 1 SD (N - 3-6). HT CS208 peq mg chl'1 h'1 Whole Chain Hzo—anev 570 i 40 640 i 50 PSII H20 ——4 2,6-DMBQ 520 .t 40 510 i 50 PSI TMPD--9MeV 1250 i 150 1290 i 230 59 al., 1983). Reactive oxygen is produced either by reduction of molecular oxygen or by photodynamic excitation of oxygen to the singlet state, e.g., by intersystem energy transfer from triplet chlorophyll (Elstner, 1982). If the injurious effects of low CO2 on the mutant were caused by reactive oxygen, then either the mutant is not capable of detoxifying a particular species of reactive oxygen or it overproduces reactive oxygen. The possibility that the mutant is deficient in some aspect of the detoxification, as well as the level of H202 in leaf extracts, were examined. Plants grown in 2% CO2 were assayed for the scavenger enzymes catalase and superoxide dismutase, and the antioxidant regenerating enzymes dehydroascorbate reductase and glutathione reductase. The soluble antioxidants ascorbate and glutathione and the membrane located singlet oxygen scavengers a-tocopherol and the carotenoids were also examined. The results are presented in Tables 2-5 to 2-7. The activities of the four enzymes were similar in the mutant and wild-type when grown in 2% £02 (Table 25). Since there are several isozymes of SOD in the cell (Gianopolitis and Ries, 1977) and since the activities of the various isozymes may increase as a result of various environmental stresses or increased oxygen radical production (Elstner, 1982), the SOD activity in a crude extract may not reveal a deficiency in one particular isozyme. For this reason, leaf extracts were analyzed on a native acrylamide gel and stained for SOD activity. Four bands were discernable from both mutant and wild—type Arabidopsis. The bands from the mutant were indistinguishable from those obtained from wild-type with respect to both mobility and intensity (results not shown). The level of ascorbate in the mutant was similar to that in the 60 wild-type (Table 2-6). However, glutathione and tocopherol were elevated 1.5- and 2-fold, respectively, in the mutant compared to the wild-type. Glutathione continued to increase to over twice the wild-type level when transferred to air for 4 days, whereas the ascorbate level declined (Table 2-6). The four major carotenoids were present in similar proportions in mutant and wild-type (Table 2-7). The total concentration of carotenoids in the mutant was 80 to 90% of the wild-type level when grown in 2%.CO2 (Fig. 2-2). Because the amount of carotenoids declined when the mutant was in air, the reduction in the carotenoid concentration observed in 2% CO2 may be a secondary effect of the mutation rather than the primary defect. The elevated level of a-tocopherol is probably a response to a powerful oxidant in the thylakoid membranes, where a-tocopherol is located (Yerin et al., 1984). Together, the data in Tables 2-5, 6, and 7 indicate that 65208 has normal or greater than normal levels of the enzymes and metabolites thought to be involved in the detoxification of reactive oxygen species. The H202 concentration in leaf extracts from mutants grown in 2% C02 was 25 i 1 (N-Z) nmol g fwt'l. Mutants transferred to air for 2 days had 29 1 7 (N-4) nmol g fwt"1 -l , compared to 28 :2 (N-2) nmol g fwt in wild-type extracts. These results indicate that the damage that occurs in air is not due to an overproduction of H202. 61 Table 2-5. Activities of enzymes involved in detoxification of’ activated oxygen species. Values are means 1 SD (N - 2-5). UT 05208 Catalase _1 _1 1.7 i 0.2 1.8 i 0.1 mmol 02 min 9 fwt Superoxide .Diimutase 1 1.4 i 0.4 1.2 1 0.1 units min 9 fwt Dehdroascorbate Reductase 4.0 i 0.5 3.7 i 0.5 nmol mi n g fwt Glutathione Deductase 2.1 i 0.1 2.1 i 0.2 ‘pmol min 9 fwt *One unit is the amount of activity that causes 50% inhibition of NBT photoreduction. 62 oz _oo. o mmo.o oz coo. A a_o.o _ocm;aoooc-a Nos. +1m~o.o ewo.o soc. +1-o.o som.o ca~_c_xo .o. + cm.o .o. + om.c _o. + mw.o _o. + ON.o saozsma aco_;omsa_s as mamas 5 msts 8:23 m.~ o.~ + e.m N.¢ m._ + ¢.m caozuwm s_o< o_ncoom< N -uzc m —osa ~ 1_< co RM 1_< so am mo~m8 c3 .A¢-~ n zv om H mam a mom mm=_m> .c_a :_ Amumncoommv mamc m co Amco_;amu=_mv mzmc e Locum new ou gm :_ czocm “cause new ma»u-u__3 co mm>mm~ =_ mucmcwxomuca co.m~m>ma .m-~ mpnmp 63 Table 2-7. Relative concentrations of carotenoids in wild-type and mutant Arabidopsis. Values are averages (N-3) of percentages of total carotenoids in the sample. The mutant line 05208 contained 0.17 mg and the wild-type 0.22 mg of carotenoids per gram fresh weight. NT 05208 % of total carotenoids Lutein 49 46 B-Carotene 24 24 Violoxanthin 14 ' 15 Neoxanthin 13 15 64 DISCUSSION In attempting to determine the basis for the requirement for high 002 and the biochemical defect caused by the h_c_11_ mutation, the possibility of a lesion in photorespiration or photosynthetic C02 fixation was examined. While C5208 grows well in air containing 2% (:02, an atmosphere low in 02 offered no protection against the deleterious effects of low C02 (Table 2-1). Hence, the high COZ requirement of the mutant does not appear to be due to a lesion in photorespiratory metabolism. The lack of protection by low oxygen also suggests that the mutant does not require a rate of £02 assimilation higher than the normal rate in 21% 02 and 350 ppm 002, since suppression of photorespiration enhances the net 002 fixation rate. A disruption in either the C3 cycle or photorespiration would be revealed by a modified distribution of label relative to the wild-type after 14 C02 feeding in the light, as well as by a reduction in the rate of photosynthesis. However, the 14C labeling pattern revealed no anomalies in the mutant (Table 2-3), and the rate of £02 exchange was very similar in the mutant compared to that in the-wild-type (Table 2-2). The normal photosynthetic rate also reveals that there is no barrier in the diffusion of 002 to the stroma. The presence of normal levels of carbonic anhydrase indicated that the COZ-HCO3' interconversion is not limiting. Considered together, these observations indicate that the biochemical lesion imposed by the hgrl mutation is not in photosynthetic or photorespiratory carbon metabolism. 65 Other processes that involve C02 were investigated. PSII electron transport requires the binding of bicarbonate on or near the QB protein (Blubaugh and Govindjee, 1986). I have ruled out the possibility that the bicarbonate binding site is altered in the mutant since photosynthetic electron transport was not impaired (Table 2-4), and the fluorescence induction transient was consistently slower in the mutant than the wild-type (Fig. 2-5). This conclusion is also supported by the observation that the mutation is of nuclear origin, whereas all of the peptides known to occur on the reducing side of PSII are chloroplast encoded (Ort, 1986). It seems unlikely that the mutation affects a carboxylase which operates in the absence of light since equivalent amounts of 14002 were incorporated into mutant and wild-type plants in an experiment performed in the dark, and the distribution of label in the various fractions was identical. In addition, high CO2 was not required for normal growth of callus tissue or for primary root elongation. The possibility that the chlorosis which occurred in air may be attributed to senescence was considered. Senescence in plants is characterized, in ‘part, by enhanced sensitivity to ethylene and enhanced production of ethylene. Ethylene action is inhibited by 002 with a K1 of 0.49 mM or 1.55% in the gas phase (Burg and Burg, 1967). It is conceivable that the mutant is hypersensitive to ethylene so that it senesces unless high 602 is present. I consider this unlikely for the following reasons. First, wild-type Arabidopsis leaves and plants were insensitive to Slpl l'1 ethylene when illuminated at IOO‘pE m'2 2 5'1 lost all of their 2 5-1 s'1 for 8 days, whereas leaves kept at B‘pE m' chlorophyll within 3 days. Leaves without ethylene at 8 p6 m' were 66 only slightly affected (results not shown). Second, an inhibitor of ethylene action, 2,5-norbornadiene (Sisler and Yang, 1984), did not prevent the bleaching of mutant leaves in ambient C02 at a concentration of 10 pl 1'1 (results not shown). Third, ethylene action in pea, measured as growth inhibition, was markedly reduced by lowering the partial pressure of oxygen to 5% (Burg and Burg, 1967). However, an oxygen concentration of 1% offered no protection to the hgrl mutant in this study (Table 2-1, Fig. 2-4). 0 It also seems unlikely that the pigment loss in the mutant is a senescence phenomenon caused by a factor other than sensitivity to ethylene. It is well known that both light and cytokinins delay senescence of leaves. However, neither 0.5 mM benzyladenine applied to the foliage of mutant plants (results not shown), nor light (Fig. 2-2) offered protection against bleaching in air. Since the necrotic lesions displayed by the mutant in air are indicative of oxidative damage in the cells, the ability of the mutant to detoxify reactive oxygen species was examined. In the chloroplast, superoxide anions, which are produced on the reducing side of PSI, are detoxified by a series of reactions involving superoxide dismutase, ascorbate peroxidase, monodehydroascorbate reductase, dehydroascorbate reductase, and glutathione reductase (Badger, 1985). Based on the activities of some of these enzymes (Table 2-5) and the levels of glutathione and ascorbate (Table 2-6), this system appears to be functional in the hggl mutant. Singlet oxygen, produced when oxygen quenches triplet chlorophyll (Krinsky, 1977), is itself quenched by the carotenoids and a-tocopherol (Elstner, 1982). The carotenoids and a-tocopherol were also present at adequate levels in the mutant (Fig. 67 2-2, Tables 2-6 and 2-7). The mutant contained more glutathione and a-tocopherol than wild-type (Table 2-6). The glutathione level (Table 2-6) and the glutathione reductase activity (results not shown) continued to increase after mutant plants were transferred to air. These protective responses by the mutant indicate that the mutant may over-produce strong oxidants. A barley mutant deficient in catalase activity, and wild-type barley fed an inhibitor of catalase, produced a three-fold increase in the level of glutathione within 8 hours in air in response to an excess of H202 (Smith et al., 1984). However, the levels of H202 in leaf extracts from CS208 in air were normal compared to the wild-type. If these measurements accurately reflect the in yiyo concentration, it may be concluded that 02'1 and GM“ as well as H202 are not produced in excessive quantities in the mutant. This conclusion is based on the fact that SOD efficiently produces H202 from 02-1, and that the only known mechanism for the production of 0H° is from H202 and 02'1 (Elstner, 1982). Singlet oxygen may be produced in the membranes without significantly affecting H202 production. However, because 021 is produced primarily by photodynamic reactions (Knor and Dodge, 1985), one would expect the degree of chlorosis to be dependent on light intensity. This was not observed (Fig. 2-3). It is difficult to propose a model to explain the DES mutation that is consistent with all of the results discussed in this paper. The role of £02 in providing a permissive condition for growth is particularly perplexing. Perhaps 602 or bicarbonate has an as yet undiscovered function as an effector for some process in leaves. Bicarbonate may provide a needed counter charge to the positive charge of a basic amino acid residue. This has been proposed for the QB 68 apoprotein of PSII which has an arginyl residue buried within a hydrophobic portion so that its positive charge is uncompensated by a nearby counter charge (Vermaas and Govindjee, 1982). The binding of bicarbonate to the arginyl residue is believed to cause a configurational change in the QB protein. Alternatively, CO2 may be required to form a carbamate on a protein as it does with Rubisco and hemoglobin (Lorimer, 1983). A carbamate is formed as the result of a nucleophilic attack of an uncharged amine upon C02. In the case of Rubisco, carbamate formation occurs with the c-amino group of a lysyl residue on the large subunit and serves to activate the enzyme. Carbamates occur on the a-amino groups on the N-termini of the hemoglobin subunits and cause a decrease in the affinity for oxygen. The regulation of Rubisco and hemoglobin by COZ was no doubt discovered because these are extensively studied proteins. Lorimer (1983) has speculated that carbamate formation “may be considerably more widespread and of greater biological significance than was previously realized”. Thus, it is possible that the hcrl mutation results in an altered affinity for CD2 or HC03' by a protein so that the mutant must be grown in a high concentration of (202 in order to activate the protein. LITERATURE CITED Artus NN, CR Somerville 1987 A high CO -requiring mutant that displays photooxidation in air. lg J Biggins, 6d, Advances in Photosynthesis Research Vol IV Martinus Nijhoff Publishers, Dordrecht, pp 67-70 Badger MR, 1985 Photosynthetic oxygen exchange. Annu Rev Plant Physiol 36: 27-53 Beuchamp C, I Fridovich 1971 Superoxide dismutase: Improved assays and an assay applicable to acrylamide gel. Anal Biochem 44: 276-287 69 Blackwell RD, AJS Murray, PJ Lea, AC Kendall, NP Hall, 00 Turner, RM Wallsgrove 1988 The value of“ mutants unable to carry out ' photorespiration. Photosynth Research (in press) Blubaugh DJ, Govindjee 1986 Bicarbonate, not CO , is the species required for the stimulation of photosystem II glectron transport. Biochim Biophys Acta 848: 147-151 Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254 Burg SP, EA Burg 1967 Molecular requirements for the biological activity of ethylene. Plant Physiol 42: 144-152 Casadoro G, G Hoyer-Hanson, CG Kannangara, SP Cough 1983 An analysis of temperature and light sensitivity in tjgrjm mutants of barley. Carlsberg Res. Commun. 48: 95-129 Chastain CJ 1985 Photosynthesis inhibition in the photorespiration mutants. PhD thesis. University of Illinois, Urbana Elstner EF 1982 Oxygen activation and oxygen toxicity. Annu Rev Plant Physiol 37: 467-507 Gianopolitis CN, SK Ries 1977 Superoxide dismutases I. Occurrence in higher plants. Plant Physiol 59: 309-314 Haughn CH, CR Somerville 1986 Sulfonylurea- resistant mutants of Arabidopsis thaliana. Molec Gen Genet 20: 430- 434 Kendall AC, AJ Keys, JC Turner, PJ Lea, BJ Miflin 1983 The isolation and characterization of a catalase-deficient mutant of barley (flgrdgum vulgarg l.). Planta 159: 505-511 Khanna R, Govindjee, T Nydrizynski 1977 Site of bicarbonate effect in hill reaction: Evidence from the use of artificial electron acceptors and donors. Biochim Biophys Acta 462: 208-214 Kirk JTO, RL Allen 1965 Dependence of chloroplast pigment synthesis on protein synthesis: Effect of actidione. Biochem Biophys Res Commun 21: 523-530 Knox JP, AD Dodge 1985 Singlet oxygen and plants. Phytochemistry 24: 889-896 Krinsky NI 1977 Singlet oxygen in biological systems. Trends Biochem Sci 2: 35-38 Law MY, SA Charles,B Halliwell 1983 Glutathione and ascorbic acid in spinach (Spinacia oleracga) chloroplasts. The effect of hydrogen peroxide and of paraquat. Biochem J 210: 899-903 70 Lorimer GH 1983 Carbon dioxide and carbamate formation: the makings of a biological control system. Trends Biochem Sci 8: 65-68 MacKinney G 1941 Absorption of light by chlorophyll solutions. J Biol Chem 140: 315-322 McCourt P, l Kunst, J Browse, CR Somerville 1987 The effects of reduced amounts of lipid unsaturation on chloroplast ultrastructure and photosynthesis in a mutant of Arabidopsis. Plant Physiol 84: 353-360 Nakano Y, K Asada 1981 Hydrogen peroxide is scavanged by ascorbate-specific peroxidase in spinach chloroplasts. Plant Cell Physiol 22: 523-530 Ort DR 1986 Energy transduction in oxygenic photosynthesis: an overview of structure and mechanism. 111 LA Staehelin, CJ Arntzen, eds, Encyclopedia of Plant Physiology New Series vol 19 Springer-Verlag, Berlin, pp 143-196 Patterson 80, EA MacRae, IB Ferguson 1984 Estimation of hydrogen peroxide in plant extracts using titanium(IV). Anal Biochem 139: 487-492 Salvucci ME, AR Portis, HL Ogren 1985 A soluble chloroplast protein catalyzes activation of ribulosebisphosphate carboxylase/oxygenase 1n yiyg. Photosynth Res 7: 193-201 Sisler EC, SF Yang 1984 Antiethylene effects of cis-2-butene and cyclic olefins. Phytochem 23: 2765-2768 Smith, IK, AC Kendall, AJ Keys, JC Turner, PJ Lea 1984 Increased levels of glutathione in a catalase-deficient mutant of barley (Handgun vulgar: 1.). Plant Sci Lett 37: 29-33 Somerville CR 1986 Analysis of photosynthesis with mutants of higher plants and algae. Annu Rev Plant Physiol 37: 467-507 Somerville CR, HL Ogren 1982a Isolation of photorespiration mutants in Arabidopsis thaliana. In M Edelman, R Hallick, NH Chua, eds, Methods in Chloroplast Molecular BiolOQY, Elsevier, New York pp 129-138 Somerville CR, HL Ogren 1982b Mutants of the cruciferous plant Arabidopsis thaliana lacking glycine decarboxylase activity. Biochem J 202: 373-380 Somerville CR, AR Portis, HL Ogren 1982 A mutant of mm thaliana which lacks activation of RuBP carboxylase oxygenase in vivo. Plant Physiol. 70: 381-387 Vermaas HFJ, Govindjee 1982 Bicarbonate effects on chlorophyll a fluorescence transients in the presence and absence of diuron. Biochim Biophys Acta 68: 202-209 7l Nilbur KM, NC Anderson 1948 'Electrometric and colorimetric determination of carbonic anhydrase. J Biol Chem 176: 147-154 Hintermans JFGH, A DeMots 1965 Spectophotometric characteristics of chlorophyll and their pheophytins in ethanol. Biochem Biophys Acta 109: 448-433 Wise RR, AN Naylor 1987 Chilling-enhanced photooxidation. Evidence for the role of singlet oxygen and superoxide in the breakdown of pigments and endogenous antioxidants. Plant Physiol 83: 278-282 Yerin AN, AY Kormanovskii, II Ivanov 1984 Localization of a-tocopherol in chloroplasts. Biophys 29: 363-364 CHAPTER 3 Mutants of Arabiggpis thaliana that Exhibit Chlorosis in an Atmosphere Enriched with CO2 72 73 INTRODUCTION Carbon dioxide enrichment of the atmosphere has been shown to increase the rate of photosythesis in most fertilized C3 plants and consequently increase vegetative yields (Enoch and Kimball, 1986). Enrichment up to 1,000 pl 1'1 CO2 is commonly practiced to increase the productivity of glasshouse crops. Occasionally, undesirable effects of C02 enrichment have been observed. Leaf injury was reported in tomato and cucumber grown in glasshouses, particularly in the winter months or when overcast weather was followed by sunshine (Van Berkel, 1986). A severe response was reported with the ornamental composite, Cerbera, which displayed chlorosis followed by necrosis and finally leaf abscission at concentrations of C02 from 800 to 4,000 pl 1'1 (Van Berkel, 1986). There was no speculation on the cause of the sensitivity to high C02 in these studies. The small cruciferous plant, Arabidopsis thaliana, tolerates CO2 enrichment to at least 2% (20,000 pl l'l) without ill effects (see chapter 2). In this report, we describe two mutants of A, thaliana which become chlorotic in elevated levels of atmospheric C02. As far as we are aware, mutants sensitive to a high concentration of C02 have not been reported in any organism. The purpose of this study is to provide a description of the mutants and to discuss the possible causes of high C02 sensitivity. 74 MATERIALS AND METHODS Plant Materials and Growth Conditions. The lines of 613.111.199.51: thaliana (l.) Heynh. described here were descended from the Columbia wild-type. The mutant lines NA73 and CaZ were isolated from M2 and M3 Apopulations of plants, respectively, which had been mutagenized with ethyl methane sulfonate as described previously (Haughn and Somerville, 1986). To reduce the number of background mutations, the mutant lines were advanced for several generations by self-fertilization and then backcrossed to the wild-type and reselected in the F2 generation. The plants were grown on sphagnum/perlite/vermiculite (1:1:1) irrigated with mineral nutrients (Haughn and Somerville, 1986) at 23 C in continuous illumination (80-100‘pE m"2 s'1 PAR). when indicated, 2% (v/v) CO2 was provided with a C02 regulator (Forma Scientific). For sterile growth of plants in petri plates, surface-sterilized seeds were plated on 0.7% agar with the standard nutrients. For the growth of plants in a low 02 concentration, a gas mixture (1500,pl 1'1 C02, 1% 02, balance N2) flowed continuously at a rate of 1 l min'1 through a 2 liter Plexiglas chamber which contained a 1 l pot with 4 to 6 plants of each genotype. The Plexiglas chamber was contained within a larger plant growth chamber which provided the standard conditions for temperature and light. Tissue Culture Methods. The media for induction and maintenance of callus have been described (Haughn and Somerville, 1986). Seeds were surface sterilized and germinated on solid callus induction medium. After 19 days, the resulting callus cultures were fragmented and 75 transferred to callus maintenance media. To measure callus growth rate, calli were plated at a density of 500 mg per plate on callus maintenance media overlaid with Hhatman #1 filters. The filter papers with callus were weighed under sterile conditions at 5 day intervals for 20 days. Mineral Analysis. Approximately 17 9 leaf tissue from 4 week old plants that had been in 2% CO2 for 8 days was dried at 68 C for 4 days. Plasma emmission spectroscopy was performed on the dried leaves by the Michigan State University Soil Testing Laboratory, East Lansing, MI. Chlorophyll. Chlorophyll was measured in 80% acetone (MacKinney, 1941). Epidermal Impressions. To observe stomatal apertures, Duco cement (Devcon corporation, Danvers, MA) was applied to the abaxial surface of a leaf. The dried glue was pealed and observed using light microscopy. RESULTS Genetic Analysis. The mutant lines NA73 and Caz were isolated by screening a mutagenized population for individuals that grew in air but became chlorotic in air enriched with 2% (v/v) C02. The genetic bases for the mutant phenotypes were examined by crossing each line with wild-type. In each case, the resulting F1 plants were able to grow normally in 2% C02. The F2 progeny resulting from self-fertilization 76 of the F1 heterozygotes were scored for the mutant phenotype by growing the F2 plants for 16 days in air then transferring them to 2% C02. For NA73, 339 F2 individuals were examined in this way, of which 264 grew normally in 2% C02 and 75 turned chlorotic. For Ca2, 400 F2 plants were examined of which 291 grew normally in 2% CO2 and 109 turned chlorotic. In both cases, the Chi square values are in agreement with the 3:1 hypothesis (xz- 1.42; P>0.20 and x2- 0.96; P> 0.30, respectively), indicating that the phenotypes are due to single, recessive nuclear mutations. To determine if the two mutations are at the same locus, the mutants were crossed to each other. The F1 progeny grew normally in elevated C02, indicating that the mutations affect separate loci. I have designated the loci h9g1 and h9g2 (' high 902 sensitive“) for NA73 and Caz, respectively. Phenotypes of the Mutants. Hhen grown in air, NA73 is slightly lighter in color and grows more slowly than the wild-type. The other mutant, Ca2, grows slightly slower than the wild-type but otherwise appears healthy in air. The mutants become chlorotic when transferred to high C02 (Fig. 3-1). The pattern of the chlorosis is different for the two mutants. For NA73, there is a preferential yellowing of the veins on the younger leaves. The leaves which are mature before transferring to 2% CO2 are unaffected. For the mutant line Ca2, chlorosis develops on the younger leaves, and there is a tendency for the basal portion of the leaf blade to turn yellow before the tip. Eventually, all the leaves become chlorotic and necrotic regions appear (Fig. 3-1). For the mutant line Ca2, loss of chlorophyll was reversed by ' returning plants to air which had become chlorotic in high C02. Leaves 77 Figure 3-1. Effect of air enriched with 2% CO on two mutant lines and wild-type Arabidopsis. (A,C,E), grown in air 25 days; (B,D,F), grown in air 17 days then in 2% CO2 for 8 days. (A,B), HT; (C,D), NA73; (E,F), Ca2. 78 79 of the mutant line NA73 regreened only if they were not yet mature when transferred back to air. Mutants of both lines were able to germinate and complete their life cycle in 2% C02, although they were mostly yellow and they grew very slowly. The fresh weights of the rosettes at their maximum size (25 days old) were 30 i 8 mg for NA73 and 10 i 6 mg for Ca2 grown in 2% C02 compared to 90 i 24 mg and 95 i 13 mg for controls grown in air (means 1 SD, N-7). In order to determine if the bleaching of the mutants in high CO2 was affected by light intensity, the chlorophyll content of both mutants and wild-type were compared after 1 week of exposure to various light intensities in 2% C02 (Fig. 3-2). At the start of the experiment, the mutants had 10% (Ca2) or 20% (NA73) less chlorophyll than the wild-type. Chlorophyll in the wild-type was reduced by approximately 40% in the dark. This was expected because chlorosis is associated with dark-induced Senescence (Biswal and Biswal, 1984). The wild-type displayed a 10% reduction of chlorophyll at the low light 2 5'1), while the medium (90 pE 111'2 5'1) and high (200 pE m'z s'l) light intensities had no effect. Chlorophyll in the intensity (20 pE m- mutant line NA73 was reduced 60% in darkness but was not affected by the low light intensisty. Wild-type and NA73 had similar chlorophyll levels after 1 week in low light. The chlorophyll content of NA73 was reduced by approximately 35% in both the medium and high light intensities. The preferential yellowing of the veins characteristic of NA73 in high CO2 (Fig. 3-1) occurred at the medium and high light intensities but not in darkness or at the low intensity. 80 air air + 2% 002 ChlorOphyll (m9 9 “Vt—1) Light Intensity (pE m" Figure 3- 2. Arabidopsis. ....... ......... ..... ...... ..... ..... ..... ..... ..... ' . . . ..... .......... ..... ........ ..... ....... ..... . . . . ...... ....... '.“‘ .... ..... ..... ..... ..... ..... . ..... .‘. . ..... ..... .... 1... ...... ... ‘‘‘‘‘‘‘‘‘‘ ..... ........ ........ ...... ........ .-.. --------- ..... ......... ..... transferring 17 day old plants from air to 2% C02. (I) Caz Values are means + SD (N- 2- -5). ......... .... ......... ...... ......... ....... ......... .... ........ .......... .3. ...... .... ........ ......... ...... ..... ..... ...... ..... ..... ....... ...... ...... .... ...... Chlorophyll concentrations in air enriched with 2% C0 a function of light intensity for two high CO Chlorophyll was measured befor sensitive mutants 6 and 7 days after (ll) NT; ([3) NA73; 81 Chlorophyll in the mutant line Ca2 was reduced by 30 to 35% at all four of the light intensities. However, necrotic lesions were only observed on plants maintained at the medium and high light intensities, and there was more necrosis on the plants at the high intensity than at the medium intensity. To determine whether nonphotosynthetic tissues are affected by high C02, the effects of air and 2% CO2 on growth of callus (Table 3-1) and on primary root elongation (Fig. 3-3) were compared. The growth rates of calli were similar in the two atmospheric conditions for both of the mutants and for wild-type. The rates of primary root elongation for NA73 were significantly lower than for wild-type in both atmospheric conditions. The growth rates for the mutant line Ca2 were similar to wild-type, except for a two day lag in the mutant. The mean primary root lengths tended to be longer in high CO2 than in air for all three lines. Thus, undesirable effects of high C02 on the mutants are not observed in calli or in roots. Table 3-1. Growth rates of calli derived from mutants sensitive to elevated atmospheric C02. Values are means 1 SD for 4 plates. Line Air , Air + 2% C0., 9 fwt d ‘ ‘ NT 0.083 1 0.022 0.085 1 0.012 NA73 0.076 t 0.015 0.084 i 0.016 Ca2 0.076 3 0.012 0.094 + 0.014 82 Length (mm) Age (days) Figure 3- 3. Rates of primary root elongation for seedlings of two mutant lines and wild- -type Arabidopsjs in air (open symbols) or air enriched with 2% CO (closed symbols). ([3) HT; ((3) NA73; (21) Ca2. Values are means + SD (N-20). _ 33 Effects of Oxygen Concentration. Since a high concentration of atmospheric C02 suppresses the flux of carbon into the photorespiratory pathway, the possibility that either or both of the mutants may require the C2 cycle for normal growth was examined. If so, it should be possible to mimick the phenotype observed in high CO2 by growing the mutants in an atmosphere in which the oxygen concentration is reduced to the point that the ratio of CO2 to 02 is equal to or greater than that in air containing 2% C02. The experiment was performed in a different manner for each mutant because of differences in their ability to grow in high C02. For NA73, mutant and wild-type plants were grown in air containing 2% CO2 for 18 days and then transferred for 7 days to either air or an atmosphere of 0.15% C02, 1% 02, balance N2. This yields a COZ/O2 ratio of 0.15 compared to a ratio of 0.10 in 2%.C02. The chlorophyll content of NA73 increased both in low 02 and in air but not in 2% CO2 (Fig. 3-4A). To ensure that the atmospheric conditions in the chamber would effectively prevent the flux of carbon into the photorespiratory pathway, a line of Arabidopsis with a defect in glycine decarboxylase activity (Somerville and Ogren 1982b) was included. At the end of the experiment, the glycine decarboxylase mutant had 1.9 mg chl g fwt‘l in low 02 compared to 1.0 mg chl g th'l in air. For Ca2, plants were grown in air and transferred to 2% CD2 or 1% 02 for 5 days. The low 02 treatment had no effect on the chlorophyll content of the mutant whereas high CO2 resulted in a 30% loss of chlorophyll (Fig. 3-4B). These results indicate that the chlorosis observed in 2% C02 is not caused by a suppression of photorespiration for either mutant. 84 IO Chlorophyll (mg 9 Mt") WT NA73 WT 082 Figure 3-4. Effect of a low 02 atmosphere on chlorophyll content of two mutant lines and wild-type Arabidopsis. A. Plants were grown in air enriched with 2% CO until 18 days old, then transferred to air (vertical slashes), maifitained in 2% C0 (open), or transferred to 1000 ul 1 CO , 1% 0 , balance N (horizodtal slashes) for 7 days. B. Plants we e grow? in air until 16 days old, then maintained in air (vertical_§lashes), transferred to 2% C0 (open), or transferred to 1000 ul l CO , 1% 0 , balance N2 (horiéontal slashes) for 5 days. Values are mead; 1 SD TN- 2-4). 85 Table 3-2. Mineral content of leaves from wild-type and two high CO sensitive mutant lines of AM. The plants were grown in air2 for 3 weeks, then transferred to air enriched with 2% CO2 for 8 days. The units are for dry weight. NT NA73. Ca2. Phosphorous (%) 0.80 1.04 1.08 Magnesium (%) 0.51 0.67 0.71 Calcium (%) 1.37 1.75 2.06 Potassium (%) 2.91 3.95 3.98 Sodium (ppm) 649 820 824 Boron (ppm) 57 55 106 Zinc (ppm) 39 45 47 Iron (ppm) 76 105 94 Manganese (ppm) 78 91 102 Copper (ppm) 3 5 34 Aluminum (ppm) 15 29 22 86 Mineral Analyses. To determine if the symptoms produced in high C02 are caused by a.nfineral deficiency or imbalance, plasma emmission spectroscopy was performed on leaf tissue from wild-type and mutant plants that were in high CO2 for 8 days. The results are presented in Table 3-2. The most striking difference was the ll-fold higher concentration of copper in Ca2 than in the wild-type. Chlorosis of the young leaves is a symptom of Cu toxicity (Lingle et al., 1963). Thus, the high concentration of Cu observed is consistent with the chlorosis displayed by Ca2 in high 002. Further experiments are required to demonstrate a correlation between the degree of chlorosis and the Cu level in the plants. Besides the Cu, both mutants appear to have approximately 1.2 to 1.4 times higher levels of all minerals than wild-type. This is probably a result of an error in the weighing of the samples rather than a true difference in the mineral content. DISCUSSION Comparison of the Phenotypes of the Two Mutants. Two nonallelic mutants of A, thaliana were described that become chlorotic in atmospheres enriched with C02. In both cases, the growing leaves are affected more than fully expanded leaves. However, there are several differences in the phenotypes of the two mutants. For NA73, the chlorosis begins on the veins and gradually spreads to the interveinal regions. For Ca2, it begins at the basal regions of the leaves and spreads to the tips (Fig. 3-1). The mature leaves on NA73 will not regreen if returned to air, whereas those on Ca2 will. NA73 87 Z s'1 for requires a minimum light intensity of between 20 and 90 pE m' the characteristic phenotype to develop in high C02, whereas the response by Ca2 was independent of the light intensity (Fig. 3-2). These differences suggest that the mutations affect separate processes in the plant. Possible Causes of High C02 Sensitivity. The high C02 sensitive phenotype may be caused by a direct action of CO2 on the leaves or by an indirect effect on the roots or the root environment. In leaves, CO2 is a substrate for photosynthesis, a suppressor of photorespiration, a regulator of stomatal aperture, and a promoter of ethylene synthesis. However, none of these functions are likely to be related to the causes of high C02 sensitivity in the mutants for the following reasons. Various proposals have been made regarding functions of the C2 cycle. It may aid in the dissipation of excess light energy when C02 is in short supply by utilizing reducing equivalents (Osmond, 1981), or it may serve as a source of glycine, serine and Cl units if the demand becomes high (Husic et al., 1987; Cossins, 1980). Because most plants are able to grow well in a high C02 concentration, it is clear that photorespiration is not required at least in high C02. However, it is conceivable that a mutation in an alternative mechanism f0r either energy dissipation or for the synthesis of glycine, serine or C1 units would impose a requirement for the C2 cycle. However, suppressing the C2 cycle by reducing the 02 concentration did not produce the high CO2 sensitive phenotype in either mutant (Fig. 3-4). 88 The stomata of some species are highly sensitive to changes in intracellular CO2 concentrations, while for other species the stomata are relatively insensitive, particularly in the absence of water stress and when the abscisic acid concentration is low (Dubbe et al., 1978; Raschke, 1975; Zelitch, 1969). In general, stomata close when the internal C02 concentration is saturating for photosynthesis, and open when it is subsaturating. The greatest changes in the response to C02 is generally observed below 300 pl 1'1 (Morrison, 1987). Concentrations of CO2 of 2% and higher have been reported to cause opening of stomata in some species (Coudret et al., 1985; Freudenberger, 1941; Longuet, 1965). I have found that this is true for Arabidopsis as well (Fig. 3-5). If one of the mutations results in stomatal closure at 2% C02, the reduced rate of transpiration may cause physiological effects on the mutant such as reduced mineral uptake. However, the effect on mineral uptake would probably not be severe enough to result in mineral deficiencies since the small amount of transpiration driven by diffusion from the leaf surface and root pressure are usually adequate to supply nutrients to a plant (Raschke, personal conununication; Bradfield and Guttridge, 1979). Though measurements were not made, stomatal apertures for the mutants in 2% CO2 did not appear to be significantly different than for the wild-type (results not shown). Carbon dioxide increases the rate of the production of ethylene from l-aminocyclopropane-l-carboxylic acid in the light (Ka - 0.06%) (Kao and Yang, 1982), yet C02 inhibits ethylene action (Ki - 1.55%) (Burg and Burg, 1967). A mutant insensitive to CO2 with respect to inhibition of ethylene action may produce symptoms caused by ethylene 89 Figure 3-5. Stomatal impressions from wild-type mm; grown in air (A) and after one day in air enriched with 2% C02 (8). 90 when grown in high C02, such as inhibition of shoot and root elongation and accelerated leaf senescence (Lieberman, 1979). These symptoms were not observed in the mutants described in this study. Root elongation was not inhibited in 2% CO2 (Fig. 3-3), and the chlorosis is not likely to be a senescence phenomenon since the old leaves are not affected (Fig. 3-1). Another manner by which CO2 may directly affect the leaves is by acidifying the cytoplasm of a mutant defective in a mechanism for pH regulation. The influence of 2% C02 on intracellular pH in the absence of active pH regulation can be calculated as follows (Bown, 1985). The molar ratio of HC03' to H2C03 plus C0z can be obtained from the Henderson-Hasselbach equation: pH - pKa + Log[HC03']/[CO2 + H2C03] The pKa for the dissociation of carbonic acid is 6.35, and the sum of the concentrations of C02 and H2C03 in water equilibrated with air enriched to 2% C02 is 634 pH at any pH. Thus, at pH 7.5, 9 mM HCO3’ and 9 mM H+ would be generated. Assuming that the passive buffering capacity in the cytoplasm is 20 mM H+ per pH unit between pH values of 6 and 8 (Leguay, 1977), the potential decrease in pH would be 9/20 or 0.45 units. This change may be expected to have a profound influence on the metabolism in the cell. As an indirect test of the mutants’ abilities to readjust cellular pH, I examined their response to hypoxia. Hypoxia causes cytoplasmic acidosis in roots (Roberts et al., 1984). One may expect that a mutant affected in pH regulation would have a lower tolerance to hypoxia. Plants were flushed with N2 for 4 hours in darkness, then returned to the standard growth environment. Seven out of nine wild-type plants survived the treatment, compared to 91 three out of three for NA73 and two out of two for Ca2. All of the surviving plants developed necrosis. These results suggest that mutants respond to hypoxia in a manner similar to the wild-type. Further investigations are needed to provide evidence to support or reject the hypothesis that one or both mutations affect a mechanism involved in pH regulation. There are two ways that high atmospheric CO2 may affect roots to cause a high CO2 sensitive phenotype. First, if the buffering capacity of the soil or soil substitute is low, hydration of C02 to H2C03 may lower the pH. The roots of the mutant may be sensitive either to the low pH or to another change in the root environment resulting from the lower pH. The media pH of 5.5 was found not to vary between the low and high C02 conditions used in this study. (The accuracy of the measurements was i 0.1 unit.) This may be attributed to the high ion-exchange capacity of vermiculite, and the buffering capacity of the 2.5 mM potassium phosphate in the nutrient solution. Second, CO2 may have an inhibitory effect on mineral uptake independent of its effect on the pH. Chang and Loomis (1945) showed that when C02 was bubbled into hydroponic media, water absorption and mineral uptake by roots of maize, wheat, and rice were inhibited. This effect was independent of the effect of C02 on the pH and on the 02 concentration of the media. Thus, if a mutant was already partly defective in the uptake of a mineral, CO2 may exacerbate the problem and induce a deficiency. However, the mineral analyses of the mutants are not consistent with the possibility of a mutation in mineral uptake (Table 3-2). Furthermore, mutant and wild-type plants were grown on agar solidified media deficient in either sulfur, iron, calcium, or a combination of 92 boron, manganese, copper, zinc, molybdenum, and cobalt. None of these treatments resulted in symptoms similar to those induced by high C02 in either the mutants or the wild-type (results not shown). The mineral analysis did reveal an abnormally high level of copper in the mutant line, Ca2. Arabidopsis wild-type had 3 ppm dwt, whereas the mutant line Ca2 had 34 ppm dwt (Table 3-2). The copper content of plants is normally between 2 and 10 ppm of the dry weight (Mengel and Kirby, 1978).~ The concentration of Cu that is toxic depends on the species. As little as 14 ppm dwt produced symptoms in rose (Rey and Tsujita, 1987). The hgsz mutation may affect the regulation of Cu uptake such that the mutant overaccumulates Cu in high atmospheric COZ° Carbon dioxide may act as an effector of the regulatory mechanism, or more likely, it may increase the availability of Cu to the plant. In the latter case, bicarbonate may facilitate the uptake of Cu2+ been reported for Ca2+, Mg2+ , as has and K+ (Wallace and Hsieh, 1966). Alternatively, CO2 may increase the availability of Cu2+ by providing protons to enhance the HT-Cu2+ exchange on the soil particles. Copper toxicity occurs, in part, because Cu competes with Fe uptake (Lingle et al., 1963). Thus, the symptoms resemble an Fe deficiency, i.e., interveinal chlorosis on the young leaves. Further studies are necessary to determine if the high CO2 sensitive phenotype of Ca2 is caused by the high concentration of Cu in the leaves. If it is, it should be possible to alleviate the symptoms by reducing the Cu and increasing the Fe in the nutrient solution. It is necesssary to demonstrate coincidence of the high C02 sensitive phenotype with the high Cu level, to rule out the possibility that they may be attributed 93 to independent mutations. The simplest way to do this would be to repeat the mineral analysis on healthy mutant plants grown in air. In summary, the high COZ-sensitive phenotypes of NA73 and Ca2 are not caused by a requirement for C2 oxidative photosynthetic carbon metabolism, or by an effect of C02 on the stomata, ethylene synthesis, or mineral uptake. Futher investigations are required to determine if C02 may be acidifying the cytoplasm of either mutant, or if the regulation of Cu uptake is altered in Ca2. LITERATURE CITED Biswal DC, 8 Biswal 1984 Photocontrol of leaf senescence. Photochem Photobiol 39: 875-879 Bown AH 1985 C02 and intracellular pH. Plant Cell Envir 8: 459-465 Bradfield EG, CG Guttridge 1979 The dependence of calcium transport and leaf tipburn in strawberry on relative humidity and nutrient solution concentration. Ann Bot 43: 363-372 Burg SP, EA Burg 1967 Molecular requirements for the biological activity of ethylene. Plant Physiol 42: 144-152 Chang HT, HE Loomis 1945 Effect of carbon dioxide on absorption of water and nutrients by roots. Plant Physiol 20: 221-232 Cossins EA 1980 One carbon metabolism. In Vol 2, Davies DD, ed, Academic Press, New York pp 365-418 Coudret A, F Ferron, D Laffray 1985 High C0 partial pressure effects on dark and light C0 fixation and metabolem in Vicia faba leaves. Photosynth Res 7: 1152-126 Dubbe OR, GO Farquhar, K Raschke 1978 Effect of abscisic acid on the gain of the feedback loop involving carbon dioxide and stomata. Plant Physiol 62: 413-417 Elstner EF 1982 Oxygen activation and oxygen toxicity. Annu Rev Plant Physiol 37: 467-507 Enoch HZ, BA Kimball 1986 WWW crops, CRC Press, Boca Raton, FL Two volumes. 94 Freudenberger H 1941 Protoplasma 35: 15-54 Haughn Gil, CR Somerville 1986 Sulfonylurea-resistant mutants of Arabidopsis thaliana. Molec Gen Genet 20: 430-434 Kao CH, SF Yang 1982 Light inhibition of the conversion of 1-aminocyclopropane-l-carboxylic acid to ethylene in leaves is mediated through carbon dioxide. Planta 155: 261-266 Leguay JJ 1977 The 5,5-dimethyloxazolidine-Z (14C), 4 dione distribution technique and the measurement of intracellular pH in Age; psggdgplatangs cells. Biochim et Biophys Acta 497: 329-333 Lieberman M 1979 Biosynthesis and action of ethylene. Annu Rev Plant Physiol 30: 533-591 Lingle JC, L0 Tifflin, JJ Brown 1963. Iron uptake-transport of soybeans as influenced by other cations. Plant Physiol 38: 71-76 MacKinney G 1941 Absorption of light by chlor0phyll solutions. J Biol Chem 140: 315-322 Mengel M, EA Kirby 1978 Er1nginlgs_gf_£13nt_flgtritign. International Potash Institute, Bern, Switzerland pp. 463-474 Morison JIL 1987 Intercellular C02 concentration and stomatal response to C0 . 13, E Zeiger GD Farquhar, IR Cowan, Stomatal Function. Stanford Univegsity Press, Stanford, CA pp.229-252 Osmond CB 1981 Photorespiration and photoinhibition. Biochim Biophys Acta 639: 77-98 Raschke K 1975 Stomatal Action. Annu Rev Plant Physiol 26:309-340 Raschke K 1986 The influence of the C0 content of the ambient air on stomatal conductance and the C0 concedtration in leaves. In, Enoch HZ, BA Kimball i ' i m e CRC Press, Boca Raton, FL. Vol 2, pp. 87-104 Rey FA, MJ Tsujita 1987 Copper nutrition of greenhouse roses relative to supplementary irradiation and growing medium. J Plant Nutr 10: 47-65 Roberts JKM, J Callis, 0 Jardetzky, V Halbot, M Freeling 1984 Cytoplasmic acidosis as a determinant of flooding intolerance in plants. Proc Natl Acad Sci 81: 6029-6033 Somerville CR 1986 Analysis of photosynthesis with mutants of higher plants and algae. Annu Rev Plant Physiol 37: 467-507 Somerville CR, NL Ogren 1982a Isolation of photorespiration mutants in Arabidopsis thaliana. In M Edelman, R Hallick, NH Chua, eds, Methods in mm Elsevier, New York pp 129-138 95 Somerville CR, HL Ogren 1982b Mutants of the cruciferous plant Arabjgnneis theliene lacking glycine decarboxylase activity. Biochem J 202: 373-380 van Berkel N 1986 C02 enrichment in the Netherlands. In, Enoch HZ, BA Kimball Cenbnn dinxide enrichment nf QEQEE ense ernpe. CRC Press, Boca Raton, FL. Vol 1, pp. 17-34 Hallace A, H Hsieh 1966 Effects of bicarbonate on cation absorption by plants. 1n A Wallace, Current Inning jn Plent untritjen. UCLA, Los Angeles, p. 57 Wilbur KM, NG Anderson 1948 Electrometric and colorimetric determination of carbonic anhydrase. J Biol Chem 176: 147-154 Zelitch I 1969 Stomatal Control. Annu Rev Plant Physiol 20: 329-350 "11111111111111l“