UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Isolation and characterization of high chlorophyll fluorescence mutants of Arabidopsis thaliana Dinkins, Randy Ray 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-ubc_1992_spring_dinkins_randy_david.pdf [ 6.06MB ]
JSON: 831-1.0086680.json
JSON-LD: 831-1.0086680-ld.json
RDF/XML (Pretty): 831-1.0086680-rdf.xml
RDF/JSON: 831-1.0086680-rdf.json
Turtle: 831-1.0086680-turtle.txt
N-Triples: 831-1.0086680-rdf-ntriples.txt
Original Record: 831-1.0086680-source.json
Full Text

Full Text

ISOLATION ANT) CHARACTERIZATION OF HIGH CHLOROPHYLLFLUORESCENCE MUTANTS OF Arabidopsis thalianabyRANDY DAVID DINKINSBA. (Philosophy) Saint Andrews Presbyterian CollegeM. Sc. (Agronomy) Oklahoma State UniversityA THESIS SUBMIPED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDoctor of PhilosophyinTHE FACULTY OF GRADUATE STUDIESDEPARTMENT OF BOTANYWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAFebruary 1992© Randy David Dinkins, 1992Signature(s) removed to protect privacyIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of BotanyThe University of British ColumbiaVancouver, CanadaDate February 28, 1992DE-6 (2188)Signature(s) removed to protect privacyABSTRACTPhotosynthetic electron transport mutants ofArabidopsis thaflana were isolatedusing the high chlorophyll fluorescence (hcf) phenotype as a screen. Seeds from small bulkpopulations (7 plants each) of chemically mutagenized (ethyl methane sulfonate) plants werecollected and the progeny screened for the presence of hcf phenotypes. Of 570 bulkpopulations screened, 251 contained hcf seedlings in the M2 generation. This suggests thatthere are a high number of nuclear loci that are required for functional photosyntheticelectron transport. Twenty-one mutant lines, each originating from a separate M1 bulksegregating for the hcf phenotype as a single nuclear recessive gene were isolated. Eightlines that appeared to be blocked early in the photosynthetic electron transport chain wereselected for further study. They were characterized with respect to fluorescence inductionkinetics, chloroplast pigment composition, photosystem I and H electron transport activityrates, chloroplast proteins, chlorophyll-protein complexes, and RNAs. Chloroplast pigmentswere found in different proportions in the mutants compared to the wild-type siblings,however, all the major pigments were present. Thus pigment biosynthesis does not appearto be the cause of the hcf phenotype, but loss of one or more of the thylakoid membranecomplexes involved in electron transport is responsible. Four of the mutants (hcf2, hct3, hcf5and hcf6) are reduced in all components of the electron transport chain (photosystems I andII, and the cytochrome complex). The block in electron transport in these mutants can becorrelated with abnormal processing and/or stability of chioroplast RNA transcripts or steadystate levels of chloroplast RNAS encoding protein products components of photosystem II andcytochrome complex. In mutant hcf2, the chloroplastpetA transcript (encoding thecytochrome fapoprotein) is at significantly higher steady state levels in the mutant. The11reason for this is unknown at the present, as the cytochrome fpolypeptide is in the thylakoidmembrane at reduced levels. Mutant hcf3 appears to have a lower level of chloroplastmRNAs, but the differences were not quantified. Mutants hcf5 and hcf6 both have alteredlevels in some, but not all, of the transcripts derived from polycistronic chioroplast operons.The results on these mutants suggests that a nuclear gene product involved in the stabilityor processing of the chioroplast transcripts is missing in these mutants. The hcf phenotypein the remainder of the mutants (hcfl, hcf4, hc17 and hct) is due to a block in electrontransport at the photosystem I complex. Photosystem II and the cytochrome complex arenormal, except for mutant hcf4 which appears to have a slightly lower level of PhotosystemII electron transport activity rate. There was a correlation between photosystem I ChIprotein complex (CPI), photosystem I thylakoid membrane polypeptides and photosyntheticelectron transport activity rate for mutants hcfl and hcf8. Mutant hcf4 on the other hand, inspite of the lower level of photosystem I ChI-protein complex (CPI) and photosystem Ipolypeptides, had a normal level of photosystem electron transport activity rate. The reasonfor this is unknown, although it is possible that a soluble component in the electrontransport chain is lost in the mutant. Chloroplast and nuclear steady state mRNA levels areat wild-type levels in all of the mutants reduced in photosystem I. This suggests that anuclear gene product is required post-transcriptionally in assembly and/or stability of thephotosystem I complex in these mutants.111TABLE OF CONTENTSABSTRACTTABLE OF CONTENTS ivLIST OF FIGURES viiLIST OF TABLES ixLIST OF ABBREVIATIONS xACKNOWLEDGEMENTS .xiCHAFFER 1INTRODUCTION 11.1. Arabidopsis as a Model System 21.2. Photosynthesis and Components of Electron Transport 51.2.1. Photosystem II 71.2.2. Cytochromeb6If 111.2.3. Photosystem I 131.2.4. Coupling Factor 171.3. Chlorophyll Fluorescence 171.4. High Chlorophyll Fluorescence Mutants 221.5. Chioroplast Biogenesis 251.5.1. Light Receptors .271.5.2. Chloroplast Signal 301.5.3. Nuclear Genes Involved in Chioroplast Biogenesis 31CHAPTER 2MATERIALS AND METHODS 352.1. Plant Material and Growth Conditions 352.2. Fluorescence Measurements 372.3. Isolation of Thylakoid Membranes and Pigment Analysis 372.4. Electrophoresis and Immunoblotting 382.5. Photosynthetic Electron Transport Measurements 392.6. RNA Isolation and RNA Blot Analysis 402.7. Electron Microscopy 422.8. Genetic Analysis 422.8.1. Marker Lines 422.8.2. Linkage Analysis 432.9. Arabidopsis Tissue Culture 452.9.1. Plant Material and Growth Conditions 452.9.2. Tissue Culture 452.9.3. Genetic Analysis of the Columbia X Point Grey Explants 462.9.4. Transformation 46ivCHAP’I’ER 3RESULTS .473.1. Isolation of hcf Mutants 473.2. Characterization ofWild-type Arabidopsis Grown Under Sterile Conditions. .. .543.2.1. Physical Characteristics and Pigment Composition of Wild-Type Plants. .543.2.2. Fluorescence and Electron Transport 553.2.3. Chlorophyll-Protein Complexes, and Thylakoid Membrane Polypeptides. 593.2.4. Electron Microscopy 633.2.5. RNA Levels in the Wild-Type 633.3. Characterization of Eight hcf Mutant Lines 663.3.1.hcf 673.3.1.1. Mutant Isolation and Genetic Analysis 673.3.1.2. Physical Characteristics and Electron Microscopy 683.3.1.3. Pigment Composition 693.3.1.4. Fluorescence and Electron Transport 693.3.1.5. Chlorophyll-Protein Complexes and Thylakoid Membrane Proteins703.3.1.6. Steady State RNA Levels 743.3.2. hcffi 783.3.2.1 Mutant Isolation and Genetic Analysis 783.3.2.2. Pigment Composition and Electron Microscopy 793.3.2.3. Fluorescence and Electron Transport 803.3.2.4. Chlorophyll-Protein Complexes 823.3.2.5. RNA Gel Blots 823.3.3. hcf2 873.3.3.1. Mutant Isolation and Genetic Analysis 883.3.3.2. Physical Characteristics and Pigment Analysis 883.3.3.3. Fluorescence and Electron Transport 883.3.3.4. Chlorophyll-Protein Complexes 903.3.3.5. Thylakoid Membrane Polypeptides 923.3.3.6. RNA Gel Blots 953.3.4.hcf3 983.3.4.1. Mutant Isolation and Pigment Composition 983.3.4.2. Fluorescence Kinetics and Electron Transport 993 3 4 3 Chlorophyll-Protein Complexes 1013.3.4.4. RNA Gel Blots 1013.3.5. hcfl 1053.3.5.1. Mutant Isolation and Genetic Analysis 1053.3.5.2. Physical Characteristics of hcfl mutant line 1063 3 5 3 Chlorophyll-Protein Complexes and Thylakoid Proteins 1073.3.5.4. Electron Transport and Fluorescence Induction Kinetics. . . . 1103.3.5.5. RNA Gel Blots of Mutant Line hcft 1123.3.6. hcfB 1153 3 6 1 Mutant Isolation and Genetic Analysis 1153 3 6 2 Physical Characteristics of the hcf Mutant Line 1153.3.6.3. Fluorescence Induction Kinetics and Electron Transport. . . . 1163.3.6.4. Chlorophyll-Protein Complexes and Thylakoid Proteins 1183.3.7.hcf4 119V3.3.7.1. Mutant Isolation .1193.3.7.2. Physical Characteristics and Electron Microscopy 1193.3.7.2. Pigment Composition 1203.3.7.3. Fluorescence and Electron Transport -. 1203.3.7.4. Chlorophyll-Protein Complexes and Thylakoid MembranePolypeptides 1223.3.7.5. RNA Gel Blots 1253.3.8. hcfl 1253.4 Additional Lines Segregating for the hcf Phenotype 1293.5. Arabidopsis Tissue Culture Studies 1343.5.1. Response of Wild-type and Point Grey Cotyledon Explants 1343.5.2. Genetic Analysis 1373.5.3. Transformation 140CHAFPER 4DISCUSSION 1434.1. hef Mutant Isolation and Genetic Analysis 1434.2. Physical Characteristics and Pigment Content of hcf Mutant Lines 1454.3. Fluorescence, Electron transport, Chi-proteins and Polypeptides 1474.3.1. PSI Mutants 1474.3.2. PSII Mutants 1524.4. Chioroplast RNA 1564.6. Genetic Analysis of Regeneration and Transformation of Wild-Type Arabidopsis .161REFERENCES 165viLIST OF FIGURESFigure 1. Macromolecular complexes involved in photosythetic electron transport in higherplant thylakoid membranes 6Figure 2. Idealized in vivo fluorescence kinetics from leaves 20Figure 3. Flow diagram of hcf mutant isolation procedures 36Figure 4. Arabidopsis thaliana hcfl mutant and wild-type plants 48Figure 5. Fluorescence induction kinetics of wild-type and hef mutant plants in the M2generation 51Figure 6. HPLC chromatograms of pigments extracted from leaves of wild-type and hcfl. . 57Figure 7. Comparison of photosystem I polypeptides from Arabidopsis and spinach 62Figure 8. Thin section electron microscopy of hcf mutant and wild-type chloroplasts 64Figure 9. Chlorophyll-protein complexes of hcf5 and wild-type siblings 71Figure 10. Immunoblot of hcf5 and wild-type siblings 73Figure 11. RNA gel blot of hcf5 and wild-type siblings 75Figure 12. Fluorescence kinetics of hcf6 and wild-type siblings 81Figure 13. Chlorophyll-protein complexes of hcf6 and wild-type siblings 83Figure 14. Immunoblot of hcf6 and wild-type siblings 84Figure 15. RNA gel blot of hcf6 and wild-type siblings 85Figure 16. Fluorescence kinetics of hcf2 and wild-type siblings 89Figure 17. Chlorophyll-protein complexes of hct2 and wild-type siblings 91Figure 18. Thylakoid membrane polypeptide profile and heme specific staining of hct2 andwild-type siblings 93Figure 19. Immunoblot of hc12 and wild-type siblings 94Figure 20. RNA gel blot of hcf2 and wild-type siblings 96Figure 21. Fluorescence kinetics of hcfl3 and wild-type siblings 100vuFigure 22. Chlorophyll-protein complexes of hcf3 and wild-type siblings.102Figure 23. RNA gel blot of hcf3 and wild-type siblings 103Figure 24. Chlorophyll-protein complexes of hefi and hcf8 and their respective wild-typesiblings 108Figure 25. Immunoblot of hcfl and hcf8 and their respective wild-type siblings 109Figure 26. Fluorescence kinetics of hcfl and wild-type siblings 111Figure 27. RNA gel blot of hcfl and wild-type siblings 113Figure 28. Fluorescence kinetics of hcf9 and wild-type siblings 117Figure 29. Fluorescence kinetics of hcf4 and wild-type siblings 121Figure 30. Chlorophyll-protein complexes of hcf4 and wild-type siblings 123Figure 31. Immunoblot of hef4 and wild-type siblings 124Figure 32. RNA gel blot of hcf4 and wild-type siblings 126Figure 33. Fluorescence kinetics of hçf7 and wild-type siblings 128Figure 34. Chlorophyll-protein complexes of hcf7 and wild-type siblings 130Figure 35. Fluorescence induction kinetics of wild-type and hcf mutant plants from lines notselected for characterization 130Figure 36. Fluorescence induction kinetics of wild-type and hcf mutant plants from lines notselected for characterization due to additional morphological phenotypes in the line . 133Figure 37. Response of Columbia and Point Grey cotyledon explants on shoot inducingmedium 136yinLIST OF TABLESTable 1. Genes encoding photosystem II polypeptides 8Table 2. Genes encoding cytochrome complex polypeptides 12Table 3. Genes encoding photosystem I polypeptides 14Table 4. Genes encoding coupling factor polypeptides 18Table 5. Genotype ofArabidopsis thaliana visible markers used in crosses 44Table 6. Arabidopsis high chlorophyll fluorescence M1 bulk screen .53Table 7. Physical characteristics ofArabidopsis thatiana hcfmutant.s and wild-type siblings.56Table 8. HPLC determined of the pigment composition of wild-type and hcf mutnat ofArabidopsis thaliana 58Table 9. Electron transport activity rates for hcf mutants and wild-type siblings 60Table 10. Response of Columbia, Point Grey and Columbia x Point Grey F1 explants inculture 135Table 11. Root and shoot regeneration of Columbia and Point Grey F2 and F3 explants onSM2 139Table 12. Explants infected with Agrobacterium forming shoots on selective SJM2 142Table 13. Summary of data for the hcf mutants and wild-type siblings 148ixLIST OF ABBREVIATIONSCAFVIS = computer-aided fluorescence video immaging systemChi = chlorophyllCIM = callus inducing mediumCP = chlorophyll proteinF0 = initial level of fluorescence= dip in the fluorescence during inductionF1 = inflection in the fluorescence during inductionFM = maximum level of fluorescenceFS = steady state level of fluorescencehcf = high chlorophyll fluorescencekb = kilobasekDa = kilo daltonLHC = light harvesting complexNAA = -Naphthaleneacetic acid= micro EinsteinsqN = non-photochemical quenchingqP = photochemical quenchingPSI = photosystem IP811= photosystem IISIM = shoot inducing mediumxACKNOWLEDGEMENTSI would like to thank the following researchers for gifts of antibodies and DNA clones: AliceBarkan; Nam-Hai Chua, William Cramer, Wilhelm Gruissem, Reinhold Herrmann, NathanNelson, Richard Malkin, Peter Westhoff. Special thanks for assistance to the followingpeople who helped me in this research in their areas of expertise: Mary Ellard, Mario Sa andGopal Subramanian with the RNA work; Michael Weis and Susan Weilesko with the electronmicroscopy; Jim Fenton and Tony Crofts with the fluorescence and use of the CAFVIS;Dingren Shen and Ted Peng for maintaining stocks, antibodies and other laboratoryequipment in a ready to use form; Roxana Moslehi for the help in the tissue culture; IngoDamm for help with the HPLC, electron transport activity measurements, (no) P700; (no)77K fluorescence and especially the poor German humor - maybe it was good German humor,just bad english humor. Special thanks to Edith Camm and Carl Douglas for numeroushelpfull discussions and allowing me free access to their laboratories and equipment. ToTony Giffiths and Beverley Green, my supervisors, many thanks for continued challengesand support. Last, and especially, I would like to thank my wife Judy, for continued supportthroughout this projectxiCHAPTER 1INTRODUCTIONThe objectives of this research were to isolate and characterize nuclear mutations ofArabidopsis thaliana that affect photosynthetic electron transport. Nuclear mutations thatdisrupt photosynthetic electron transport probably would be located in genes that encodegene products that are essential for chioroplast biogenesis and assembly or stability of thephotosynthetic complexes. The screening procedure used took advantage of the fact thatlight energy that cannot be utilized in photosynthesis is re-emitted as chlorophyllfluorescence. Thus mutations that block photosynthetic electron transport display a highlevel of fluorescence and are termed high chlorophyll fluorescent phenotype or hcfCharacterization of selected hcf lines that appeared to segregateas single nuclear recessivemutations entailed analysis of the fluorescence kinetics, Chi-protein complexes,immunoblotting with monospecific polyclonal antibodies, electron transport activity rates,and analysis of steady state RNA levels of nuclear and chioroplast genes encodingpolypeptides known to be an integral part of the electron transport system.The rationale of the approach taken in this research is that the chioroplast, althoughbiochemically complex, is genetically simple. Sequencing the chloroplast genome fromseveral species has revealed that there are approximately 130 known genes and open readingframes (Hiratsuka et al., 1989; Ohyama et al., 1986; Shinozaki et al., 1986).The majority ofthe genes encode products involved in chioroplast transcriptionand translation, and encodeprotein components that are part of complexes involved in photosynthetic electron transport(Sugiura, 1988). It is well established that the nucleus plays a major role in chioroplastbiogenesis. Nuclear genes not only encode components thatform an integral part of thephotosynthetic electron transport system, but also encode components that are required in1the assembly and stability of the complexes. Little is known about these genes. Mutantanalysis provides the opportunity to study the action of nuclear genes required forchioroplast biogenesis that would not otherwise be accessible through biochemical andphysiological means.1.1. ARABIDOPSIS AS A MODEL SYSTEMThe advantage ofArabidopsis as a model system for studies in plant biology andgenetics is due to the short generation time (6-8 weeks), small genome size (Leutwiller et al.,1984), little repetitive DNA (Pruit and Meyerowitz, 1985), availability of mutants (Estelleand Somerville, 1986, Meyerowitz, 1989), availability of genetic and molecular markers formapping (Koornneff et al., 1983; Chang et al., 1987; Nam et al., 1989), and availability oflarge DNA segments in yeast artificial chromosomes to facilitate chromosome walking (Grilland Somerville, 1991). Arabidopsis is a dicotyledonous plant, and can be transformed viaAgrobacterium tumefaciens using leaf (Lloyd et al., 1986), cotyledon (Schmidt andWillmitzer, 1988) and root (Valvekens et al., 1988) explants, and seeds (Feldmann andMarks, 1987). The latter system has also led to the isolation of mutants by the insertion ofthe T-DNA ofA.grobacterium (Feldmann et al., 1989; Feldmann, 1991), facilitating isolationof genes of interest (Yanofsky et al., 1990).Each model system also has disadvantages, and Arabidopsis is no exception. Thesmall size of the plants complicates harvest of sufficient material for analyses (especially forseedling lethal mutants such as is the case for the present study), and the small size of theflowers requires expertise for crossing. There are no known functional transposableelements in Arabidopsis. However, the maize transposable element Ac has been transformedinto Arabidopsis, and can ‘hop’ at low frequencies (Baker et al., 1987; Schmidt and2Willmitzer, 1989). The transformation frequency ofArabidopsis explants is not on par withother systems such as tobacco (Nicotiana tabacum L.) and petunia (Petunia hybrickL L.). Inaddition the favored ecotypes for molecular and genetic studies (Columbia and Landsburg)have consistently been the hardest ecotypes from which to obtain fertile transformed plants(Chaudhury and Signer, 1989; Schmidt and Willmitzer, 1988; Valvekens et aL, 1988).The success ofArabidopsis as a model system in plant genetics and molecular biologyis due to a large increase in the number of researchers using Arabidopsis in the last 5-10years. This is evidenced by the increase in numbers ofjournal articles that appear monthlyin which Arabidopsis is the subject organism. The uses ofArabidopsis can be grouped intothe following (arbitrary) topics: isolation of novel mutant phenotypes, molecularcharacterization of the Arabidopsis genome, and characterization of known proteins andgenes already isolated in other species. The combination of the first two above is the mostpromising aspect ofArabidopsis as a model system. It is virtually impossible to isolate agene for which the protein product is not known, unless a molecular tagging’ technique isavailable. This is especially true in cases where the gene of interest is regulatory, where theprotein product will probably be at extremely low levels. Genes that give rise to mutantphenotypes which are probably regulatory are on the verge of being isolated and analyzed inArabidopsis through the combined use of mutant analysis and gene isolation by chromosomewalking. One of the more advanced studies of this nature derived from mutant analysis, isin flower development. To date, two genes involved in flower development have been cloned(Yanofsky et al., 1990). Granted, both genes were isolated due to an insertion of T-DNA inthe gene of interest. However, isolation of genes with known mutant phenotypes bychromosome walking should begin to appear in the literature shortly.3As the possibility of chromosome walking becomes reality in Arabidopsis, manyresearchers have concentrated on isolating mutant phenotypes of interest. Arabidopsis hasyielded phenotypes that are auxotrophic (Last et al., 1988; Schnieder et al., 1989), hormoneinsensitive (Klee and Estelle, 1991); phytochrome deficient (Chory et al., 1989); have alteredresponses to blue light (Karuna and Poff, 1989; Liscum and Hangarter, 1991) and reactsimilarly to that of light grown plants in the dark (Chory et al., 1989; 1991; Beng et al.,1991). The mutants with altered responses to light are of particular interest with respect tothe research presented in this thesis and will be further described below.Studies of the Arabidopsis photosynthetic apparatus have been meager compared toChiamydornonas reinhardtii, Synechocystis spp., barley (Hordeum vulgare L.), maize (Zeamays L.) and spinach (Spiniacea oleracea L.). Most of the research has been on lipidbiosynthesis in the chioroplast (reviewed in Browse and Somerville 1991), mutants withaltered CO2 requirements (Somerville and Ogren 1982; Artus and Somerville, 1988), andmutants blocked in pigment biosynthesis (Duckham et al., 1991; Murray and Kohorn, 1991;Ruhie et al., 1983). Work has been done on the characterization and expression of the cabgenes (Leutwiller et al., 1986; Karlin-Neuman and Tobin, 1988). A gene required forphotosynthetic competence in Arabidopsis has been isolated from a mutant caused by T-DNAinsertion (Koncz et al., 1990). The T-DNA mutant did not complement a known Arabidopsismutant (ch-42) with a similar phenotype. The wild-type CH-42 gene has been isolated andthe gene product is targeted to the chloroplast, although the function and location of thisgene product in the chloroplast is as yet unresolved.Using the knowledge ofhigher plant chioroplast thylalcoid membrane complexes(Green, 1988), chloroplast development (Deng and Gruissem, 1987), chloroplast geneorganization (Westhoff and Herrmann, 1988) and gene expression (Deng and Gruissem41988), Arab idopsis photosynthetic complexes and gene expression can be reasonablyspeculated on until further characterization of the Arabidopsis system is done.1.2. PHOTOSYNTHESIS AND COMPONENTS OF ELECTRON TRAI”ISPORTOxygenic photosynthesis is the process by which radiant energy, in the form ofsunlight is converted into chemical energy in the form of biomass. The light reactions ofphotosynthesis are involved in light capture, stabilization and transfer of energy into thechemical currency in the formation of ATP and generation of reductant in the form ofNADPH and ferredoxin. Functionally, electron transport occurs from intermediate tointermediate on thylakoid membrane complexes and soluble electron carriers of chioroplastsin a sequential manner outlined by Hill and Bendall (1960). In the transfer of electrons fromPSII to ferredoxin and NADPH an electrochemical potential gradient is generated and ATPformation ensues. The energy and reducing power convert carbon, nitrogen and sulfur whicharrive at the plastid in the form of carbon dioxide, nitrite, and sulfate ions, intocarbohydrates, amino acids, nucleotides, and many other compounds.The thylakoid membrane macrocomplexes involved in photosynthetic electrontransport are: photosystem II (PSII), cytochrome b6/f, and photosystem I (PSI) in addition tothe mobile electron carriers (Figure 1). Crude fractionation studies indicate that the activePSII complexes are primarily confined to the appressed regions of the grana, whereas PSI isfound in the stroma and at the margins of grana (Anderson, 1989). The cytochrome b6Ifcomplex is found in both the grana and stroma lamellae (Anderson, 1989). The level of eachof the complexes in the thylakoid membrane is continually in a dynamic equilibrium inresponse to environmental conditions and energy requirements. Plants grown under low5StromaFigure 1. Macromolecular complexes involved in photosynthetic electrontransport in higher plant thylakoid membranes. The complexes arephotosystem I, cytochrome b6/f, photosystem II. Gene sequences arepresented on Tables 1-3. This figure is adapted with modifications fromGreen et aL (1991).psaGl’hotosystem 11 ComplexPhotosystem I ComplexCytoehrome b6-fComplex6light levels, for example, were found to have a cytochrome/PSI ratio of 0.7 (Chow andAnderson 1987; Chow and Hope, 1987). When the plants were transferred to high light, thecytochrome content in the thylakoid membranes increased to a cytochromelPSl ratio of 1.0-1.25. Thus when the rate of light absorption is limited, the thylakoid membranes containlow levels of the cytochrome complex. Conversely, under high irradiances, where light is notlimited, the increased levels of the cytochrome complex in the membrane ensure increasedlevels of intermediate electron transport carriers. In addition, heterogeneity within PSII andPSI complexes has been documented (Govindjee, 1990; Svensson et aL, 1991). This is in partdue to subpopulations of each of the complexes that are probably being repaired or assembled(Melis, 1991). Further details of the macromolecular organization, function and geneticorigin of each of the complexes follows.12.1. Photosystem IIPSII is responsible for the oxidation of water and reduction of a stable electronacceptor, plastoquinone. The oxidation-reduction potential of this reaction is +0.9 V. Lightenergy is absorbed by the ChI antenna which is transferred to the reaction center ChI wherecharge separation occurs. The reaction center Chi P-680 through a series of four oxidizingphotochemical reactions involving a cluster of four manganese results in the oxidation ofwater leading to oxygen evolution, and the release of protons into the thylakoid lumen.By analogy to the Rhodopseudomonas viridis reaction center for which the 3-1)structure has been elucidated (fleisenhofer et al., 1985), it is thought that the core proteins,Dl and 02, bind a special pair chlorophyll (P680), two pheophytin molecules and each have aquinone binding site (Table 1). Both Dl and 02 are predicted to contain 5 membrane7Table 1. Genes encoding photosystem II polypeptidesGene Origin Gene ProductpsbA Chioroplast 34 kDa reaction center polypeptide (Dl)psbB Chloroplast 45-54 kDa Chi binding PSII core polypeptide (CP47)psbC Chloroplast 40-45 kDa Chl binding PSII core polypeptide (CP43)psbD Chloroplast 34 kDa reaction center polypeptide (D2)psbE Chloroplast 9 kfla Cytochrome b559 binding polypeptidepsbF Chloroplast 4 kfla Cytochrome b559 binding polypeptidepsbG Chloroplast PSII “G”protein (?); NADH dehydrogenase (?)psbH Chloroplast 9 kDa phospho proteinpsbl Chioroplast PSII ‘T proteinpsbJ Chioroplast PSII “J” proteinpsbK Chioroplast PSII “K” proteinpsbL Chioroplast PSII “L” proteinpsbM Chioroplast PSII “M’ proteinpsbN Chioroplast PSII “N” proteinpsbO Nuclear 33 kDa polypeptide of OECpsbP Nuclear 23 kDa polypeptide of OECpsbQ Nuclear 16 kDa polypeptide of OECpsbR Nuclear 10 kDa polypeptide of OEC8spanning helixes, and have been shown to contain one electron transport intermediatetyrosine residue (denoted Z and D, respectively) near the manganese atoms in the oxygenevolving enhancer complex. The Dl contains the QB binding site for the mobile electroncarrier, plastoquinone.In addition to the Dl and D2 polypeptides, the smallest photochemicafly activereaction center PSU preparation isolated also contains the polypeptides that bindcytochromeb559; (psbE and psbF gene products), and the psbl gene product (Namba andSatoh, 1987). The role of cytochrome b559 in PSII is still unknown, although mostresearchers do not believe that it is involved in non-cyclic electron transport (for a dissentingview see Ortega et aL, 1989). Possible hypotheses for the role of cytochrome b559 in PSIIare: 1) a protectant against photooxidation of PSII core chlorophylls (Herber et al., 1979), 2)participation in cyclic electron flow around PSII (Arnon and Tang, 1988), and 3) a redoxfunction in water oxidation (Cramer et al., 1986). All mutants that lackb559 are devoid ofPSII activity (Maroc and Gamier, 1981; Pakrasi et aL, 1989). It is interesting to note that ina mutant of Synechocystis in which thepsbE and psbF genes were deleted, the CP47 andCP43 proteins were observed to accumulate in the membrane (Pakrasi et al., 1989). Inhigher plants and Chlamydomonas any disruption of a PSII component appears todestabilize the P511 complex leading to loss of all P511 polypeptides (Metz and Miles, 1982;Rochaix and Erickson, 1988).The PSII core complex also includes two chloroplast encoded ChIa binding proteins,CP47 (psbB gene product) and CP43 (psbC gene product). Although not components of thereaction center, they function in the transfer of light energy to the reaction center. Both areessential for P511 stability in vivo, as inactivation results in PSII deficiency (Vermas et al.,1986).9The psbH gene encodes a small (10 kDa) phosphoprotein that is co-transcribed withthe 47 kDa ChIa binding protein in the plastid. The function of this polypeptide in PSII3 aswith some of the other smaller proteins associated with PSII, is not known. In some casestheir association with PSII has yet to be confirmed. For example, the psbG gene has beenproposed to be a component of NADH dehydrogenase and not a component of PSU (Nixon etaL, 1989). Insertional inactivation of the psbG gene in Synechocystis had no phenotypiceffect on P511 activity (Steinmuller et aL, 1991). However, due to a second plasmid-bornecryptic psbG gene in Synechocystis, no conclusions as to the role of the psbG can be made.Associated peripherally with PSII are nuclear encoded 33, 23 and 16 kflapolypeptides (psbO, psbP and psbQ, respectively) which form the oxygen enhancer complexand serve to stabilize the manganese atoms in P511. Originally this was called the oxygenevolving complex as the integrity of the complex is associated with oxygen evolution.However, oxygen evolution can be partially restored by calcium and chloride ions in theabsence of these polypeptides. It is generally accepted that the 33 kDa protein serves tomaintain the functional conformation of the manganese cluster. Deletion of the psbO gene(33 kDa) does not 1ead to loss of oxygen evolution in Synechocystis sp. PCC 6803, but doesincrease the susceptibility to photoinhibition (Mayes et al., 1991). The extrinsic 23 and 16kfla proteins do not play an obligatory role in binding manganese. However, they probablyhave regulatory functions serving to stabilize the manganese, protect the cluster from otherreducing agents and modulate the binding of the calcium and chloride ions (Ghanotakis andYocum, 1990).Associated with PSII, and comprising part of the PSII macro complex are the Chi a-i-blight harvesting antenna complexes, LHCII (reviewed in Green, 1988; Thornber et al., 1990).These are encoded by the nuclear cab genes. Several forms of the light harvesting antenna10have been found in the thylakoid membranes. Some of the protein complexes are in closeassociation with PSII, eg. CP29 and CP24 and do not appear to disassociate from thecomplex in vivo. The mobile 111011 complexes constitute a second form. These are foundassociated with PSII under favorable PSIT light conditions, whereas under high light orunder PSI light conditions they are phosphorylated, migrating to the stromal lamelae,thereby decreasing the PSII antenna size. Evidence has accumulated that the mobile formcan become associated with PSI in the phosphorylated state (Kyle et al., 1983). It ishowever, unresolved if the mobile LHCII complexes are actually involved in transfer ofenergy to PSI.1.2.2 Cytochrome hG/fThe cytochrome b6/fcomplex oxidizes plastoquinone and reduces plastocyanin. Inthis process protons are translocated across the membrane generating a protonmotive force.The cytochrome complex is composed functionally of the 34 kDa cytochrome f associatedpolypeptide (encoded by the petA gene), a 23 kDa polypeptide (encoded by the petB gene)which binds two cytochrome b563 (b6)heme molecules, a nuclear encoded 20 kDa iron-sulfurRieske protein (encoded by the petC gene), and a 17 kDa polypeptide termed subunit 4encoded by the petD gene (Westhoff et al., 1986) (Table 2). Recently Haley and Bogorad(1989) have shown that a small 4 kDa chloroplast-encoded polypeptide is also associated withthe complex. The function of this polypeptide in the complex is unknown.The petB and petD genes are co-transcribed in the same polycistronic message as thepsbB and psbH genes of photosystem II. The operon has been extensively studied as thesteady state levels of RNA messages display a complex pattern on northern blots (Deng andGruissem, 1987; Rock et aL, 1987; Westhoffet al., 1986). This is due to complex11Table 2. Genes encoding cytochrome complex and other chloroplast polypeptidesGene Origin Gene ProductpetA Chioroplast cytochrome fpetB Chloroplast cytochrome b6petC Nuclear Rieske Fe-S polypeptide, subunit IIIpetD Chloroplast 9 kDa, subunit TVpetE Nuclear plastocyaninpetF Nuclear ferredoxinpetG Chioroplast subunit V, function unknownetH Nuclear ferredoxin-NADPH reductase (FNR)petl Nuclear flavodoxin12processing events whose functional roles are not yet completely understood. In addition, thegene products from this operon do not accumulate simultaneously, as the photosystem IIgene products only accumulate during illumination, whereas the petB and petD geneproducts are present both in the dark and light. Co-transcription ofpetB and petD is notsurprising in light of the fact that the petB and petD gene products together are homologousto the N terminal and C terminal regions, respectively, of the mitochondrial cytochrome bgene (Widger et al., 1984).Also included in the pet (photosynthetic electron transport) designation is the mobileelectron carrier plastocyanin (petE), ferredoxin (petF), ferredoxin-NADPH reductase (FNR)(petiT) and flavodoxin (petl) (Hallick, 1989).1.2.3 Photosystem IPSI oxidizes reduced plastocyanin, and transfers electrons to ferredoxin which reducesNADP+. Charge separation occurs at the reaction center Chi, P700, and the electron istransferred to the primary electron acceptor Aj. The electron is transferred through a seriesof membrane-bound receptors to ferredoxin.Presently there are twelve gene products known to be part of the PSI complex (psaAthrough psaL; see table 3). Five are encoded in the chloroplast genome (psaA-C, psal, psaJ),and the remainder are of nuclear origin. With the exception ofpsaA, psaB and psaC, thefunctions of the other polypeptides in the complex have not been established unequivocally,although the functions of the psaD and psaF polypeptides are fairly certain.The psaA and psaB gene products form the PSI core complex. They are homologouspolypeptides and form a heterodimer. Dimerization of the psaA and psaB polypeptides ispostulated to be aided by presence of a leucine zipper motif (Weber and Malkin, 1990).13Table 3. Genes encoding photosystem I polypeptidesGene Origin Gene ProductpsaA Chloroplast P700 Chla binding core polypeptide, subunit IpsaB Chloroplast P700 Chia binding core polypeptide, subunit IIpsaC Chioroplast 9 kDa Fe-S polypeptide, subunit VIIpsaD Nuclear ferredoxin docking polypeptide, subunit IIpsaE Nuclear 18-20 kDa accesory polypeptide, subunit IVpsaF Nuclear plastocyanin binding polypeptide, subunit IIIpsaG Nuclear 14-16 kDa accessory polypeptide, subunit V(?)psaH Nuclear 10-12 kDa polypeptide, subunit VIpsal Chloroplast PSI T polypeptide, unknown functionpsaJ Chloroplast PSI “J” polypeptide, unknown functionpsaK Nuclear PSI ‘K’ polypeptide, unknown functionpsaL Nuclear PSI “L” polypeptide, subunit V(?)14Both genes encode polypeptides of 82-83 kDa, although the polypeptides are observed tomigrate on gels with an apparent molecular weight of 58-70 kDa. The anomalouselectrophoretic migration has been ascribed to the hydrophobic nature of these proteins. ThepsaA and psaB polypeptides bind the P700 reaction center chlorophyll, the primary electronacceptor A3, two phyloquinone (vitamin Ki) molecules which are postulated to be theacceptor A1, and the iron-sulfur (4Fe-4S) cluster F (Golbeck and Bryant, 1991). There areapproximately 60- 130 ChIa and 12-16 13-Carotene molecules in the core complex. ThepsaA/psaB core can function at room temperature to produce stable, light dependent chargeseparation (Golbeck and Bryant, 1991).The psaC gene encodes a small polypeptide (9 kfla) which binds the two iron sulfurcenters (4Fe-4S) FA and FB (Hoj et al., 1987). The FAIFB clusters are the terminal electronacceptors of the PSI complex and reduce the soluble electron carrier ferredoxin. It ispresently under debate whether the electron flow from FX to the FA and FB centers is linearor functions in parallel and which of the two centers donate electrons to ferredoxin (Golbeckand Bryant, 1991).The psaD gene encodes a precursor polypeptide of 23 kfla containing a 50 amino acidtransit peptide directing the protein to the chloroplast stroma (Hoffman et al., 1988). Themature protein migrates on polyacrylamide gels with an apparent molecular weight of 18-22kDa and was designated subunit II by Bengis and Nelson (1975). The psaD gene wasoriginally isolated from tomato (Hoffman et al., 1988) and subsequently from spinach(Lagoutte, 1988; Munch et al., 1988), Synechocystis sp PCC 6803 (Riley et al., 1988) andSynechocystis sp PCC 6301 (Wynn et al., 1989). Partial amino acid sequences have also beenobtained from the higher plants barley (Schefler et al., 1988) and pea (Dunn et al., 1988).There is a high degree of conservation in the PSI subunit II protein, and antibodies to the15spinach and swiss chard proteins cross react with bean, oats, Chiamydomonas andMastigocladus laminosum (Nechushtai and Nelson, 1985; Nechushtai et aL, 1983). In spiteof this the functional role has not been established although studies suggest that the psaDprotein is a docking site for ferredoxin (Zanetti and Morelli, 1987; Zilber and Malkin, 1988).A mutant of Synechocystis sp PCC 6803 that lacks the psaD gene product due to aninsertional mutation, grew slowly under photoautotrophic conditions, lacked some of thesmaller PSI polypeptides, and had a P700 activity rate that was reduced by 50%, but had aPSI electron transport rate similar to the wild-type (Chitnis et al., 1989b).The psaE gene encodes a 10 kDa polypeptide although it is frequently observed tomigrate with an apparent molecular weight mass of 14-16 kDa on polyacrylamide gels(Golbeck and Bryant, 1991). There is a high degree of conservation between the protein fromcyanobacteria and higher plants. The functional role ofthepsaE gene product is not known.Insertional inactivation of this gene in Synechocystis resulted in only minor differences inphotoautrophic growth and PSI activity rates compared to the wild-type (Chitnis et al.,1989a).The role of the psaF gene product [subunit III in the terminology of Bengis andNelson (1975)] is fairly well established. It is encoded in the nucleus and contains an N-terminal sequence similar to other lumenally directed proteins. Cross linking studies haveconfirmed that it is lumenally located, and that it cross links to plastocyanin (Wynn andMalkin, 1988; Hippler et al., 1989). The psaF gene product appears to be dispensable, asinsertional inactivation of the gene in Synechocystis does not impair autotrophic growth(Chitnis et al., 1991).The function of the remainder of the PSI gene products (psaG-psaL) has not beenestablished. The psaG and psaH gene products probably interact with the light harvesting16complex LHCI, as these protein products have only been observed in eukaryotes. (Li et al.,1991). Additional small polypeptides that appear to be part of PSI have been observed, butthey have not been characterized.12.4 Coupling FactorAlthough the coupling factor is not directly involved in electron transport per Se, thecomplex is an integral component of the chloroplast thylakoid membrane. It is composed oftwo structures, the catalytic section (CF1), and a hydrophobic structure in the membrane(CF0). The function of the complex is to harness the energy of the electro-chemical gradientto generate ATP from ADP and Pi. The coupling factor is found primarily on the stromallamellae and grana margins, probably due to the physical contraints from the CF1 knobprotruding from the thylakoid. The coupling factor is encoded by both nuclear andchloroplast genes (Table 4). In the chioroplast, two polycistronic operons encode the couplingfactor gene products (atpl-atpF-atpH-atpA and atpB-atpE).1.3. CHLOROPHYLL FLUORESCENCEChlorophyll a fluorescence has been used as a non-invasive method for monitoringphotosynthetic events and assessing the physiological state of the plant. Many fluorescencestudies have used isolated protoplasts or isolated chloroplast membranes and particlepreparations. Fluorescence emanating from intact leaves under normal physiologicalconditions is still enigmatic (Krause and Weis, 1991). Compilation of results obtained fromthese sources and contributions from studies on mutants allows for some generalizations tobe made about fluorescence observed under normal conditions. A typical idealizedfluorescence induction observed from dark adapted leaves exhibits a fast rise from F0 to Fj17Table 4. Genes encoding coupling factor polypeptidesGene Origin Gene ProductCF1:atpA Chioroplast a subunit; nucleotide binding and regulationatpB Chioroplast 3 subunit; Active siteatpC Nuclear y subunit; binding of CF1 to membraneatpD Nuclear 6 subunit; induction of proper bindingatpE Chioroplast subunit; necessary for phosphorylationCFO:atpF Chioroplast subunit I; 27 kfla polypeptide, function unknownatpG Nuclear subunit II; assembly of CFO (?)atpH Chioroplast subunit ifi; proton conductionatpl Chloroplast subunit IV: binding to CF1 (?)18(Iinflection) followed by a plateau or ‘dip’ (FD) and a slower rise to the peak (Fp) ormaximal fluorescence (FM) (Figure 2). Theoretically Fp equals FM, however Fp is lowerthan FM when complete reduction of QB is not achieved (ie. when light levels are notsaturating).The first fast rise (F0 to Fj) is thought to be due to the reduction of QA in thepopulation of inactive PSII centers lacking QB. This is suggested by results obtained usingthe P511 electron acceptors dimethylquinone and ferricyanide which quench the D-Pfluorescence but not the 0-I peak (Cao and Govindjee, 1989; Melis, 1985). Guenther andMelis (1990) suggest that this population is a subset of PSIIf3 centers in the process of beingassembled or repaired. The I-D phase is reported to be due to re-oxidation of QA by PSIelectron transport (Hansen et al., 1991). The D-P rise represents the reduction of theplastoquinone pool and has been postulated to be biphasic representing the PSIICL and lunits (Melis, 1991).In vivo fluorescence kinetics actually observed can vary with the source of plantmaterial and is influenced by: 1) rate of electron donation toP680,2) PSII cooperativity, 3)PSII heterogeneity, 4) size of the plastoquinone pool and the rate of its reoxidation, and 5)rate of electron transport beyond PSI including carbon metabolism (Krause and Weiss,1991). The F0 fluorescence observed at 680 nm under normal conditions is primarily emittedfrom PSII and thus is an indication of PSII integrity. Although PSI can contribute to overallvariable fluorescence this is normally only observed at wavelengths above 700 nm.Holzwarth et al. (1990) suggest that fluorescence from PSI only contributes 1-2% of thefluorescence observed at 685 nm.The origin of the variable fluorescence observed from intact leaves at roomtemperature is still uncertain. It was originally proposed by Klimov and Krasnovskii (1981)1940C)35(I)C)0302520Figure 2. Idealized in vivo fluorescence kinetics from dark adapted leaves.Features that can be observed are: O=the initial (F0) level prior to actiniclight; 1= inflection ; D= dip; P= the peak or maximal fluorescence. S= steadystate fluorescence which is observed following fluorescence quenching byphotochemical and non-photochemical mechanisms. Minor oscillations canalso be observed (see peak between 50 and 60 seconds) until CO2 fixationreaches a steady state level. Time scale indicates time from actinic light on.0 2 4 20 40 60 80Time (s)20and Breton (1982) that a charge recombination in closed reaction centers ofP680 Pheowas the source of the variable fluorescence. This is supported by data from Shatz et al.(1988) using picosecond fluorescence decay kinetics, and Schlodder and Brettel (1988) usingflash-induced absorbance changes related to the primary radical pair of P511, where chargeseparation formingP68 Pheo is inhibited by the electrostatic effect of QA. Thishypothesis has recently been questioned and van Dorssen et al. (1987) have proposed thatthe variable fluorescence originates from Chia molecules in the core antenna complex, CP47.These authors have shown that at 77 K, the fluorescence of CP47 (695 nm florescence)follows the same kinetics observed at room temperature. -Fluorescence measurements done on isolated chloroplasts or thylakoid membranepreparations and plant material at 77 K have been useful in the analysis of the fluorescenceinduction kinetics and fluorescence decay kinetics. However, for measurements involvingphotochemistry, such as reoxidation of QA, P511 cooperativity and electron transport eventsbeyond PSI, in vivo measurements at room temperature are more informative. Typicalfluorescence in vivo under continuous illumination results in a decline in the fluorescenceyield following FM, and this is termed the Kautsky effect (Krause and Weis, 1991).Fluorescence quenching is known to be caused by a variety of factors. The majorfactors contributing to fluorescence quenching are photochemical energy conversion used in002 reduction (qp), and non-photochemical quenching (q). Non-photochemical quenchingis the result of energization across the thylakoid membrane due to the formation of a protongradient (q), changes in the PSH cooperativity or connected centers due to phosphorylationof LHCII (q’p), and photoinhibition (q1). The use of pulse-modulated fluorometers hasallowed analysis and resolution of the quenching components (Schreiber et al., 1986).21However, the relative contribution of each in the decline from FM to F5 is presentlyunresolved.1.4 HIGH CHLOROPHYLL FLUORESCENCE MUTANTSFluorescence induction kinetics can be used to isolate mutants with alteredphotosynthetic capacity due to the consistency of the fluorescence induction of dark-adaptedplants or chioroplasts. This was the basis for the mutant isolation procedures used in thisstudy. The rationale for selecting mutants that are blocked in photosynthetic electrontransport by screening for high chloroplyll fluorescence was first shown by Bennoun andLevine (1967) in Chiamydomonas and subsequently in maize (Miles and Daniel, 1973).Screening for plants with pigmentation differences has also yielded mutants with alteredfluorescence kinetics in barley, Oenothera, and tobacco (Chia et al., 1986; Simpson and vonWettstein, 1980; Johnson and Sears, 1990).Hcf mutants tend to exhibit limited pleiotropic effects, but usually the mutationaffects a related set of polypeptides (Barkan et al., 1986; Metz and Miles, 1982; Metz et al.,1983; Taylor et al., 1989). The hcf mutation does not appear to disrupt pigment biosynthesisper Se, but disrupts the assembly of one or more of the complexes involved in electrontransport (Miles, 1980). Hcf mutants were instrumental in convincing researchers thatchlorophyll was associated with proteins and not lipids (Thornber 1986), helped to show therole of the cytochromes in non-cyclic electron flow (Levine, 1968) and have aided incorrelating loss of function with loss of particular sets of polypeptides (Barkan et al., 1986;Chua and Bennoun, 1975; Chua et al., 1975; Jensen et aL, 1986; Lemaire and Wollman,1989a; Leto and Miles, 1980; Metz and Miles, 1982; Moller et al., 1980; Rochaix andErickson, 1988; Taylor et al., 1987; Wollman and Lemaire, 1986). Studies on mutants of22barley, Chiamydomonas, and maize have yielded hcf mutations which specifically affect thePSII complex (Chua et al., 1975); Leto and Miles, 1980; Metz et al., 1984; Rochaix andErickson, 1988; Simpson and von Wettstein, 1980), cytochromeb6If(Lemaire et al., 1986;Metz et al., 1983), PSI complex (Bennoun and Jupin, 1976; Hiller et al.,1980; Cook andMiles, 1990), as well as photophosphorylation and carbon fixation (Edwards et al., 1988;Lemaire and Woliman, 1989b; Miles, 1980).Photosynthetic mutation affecting electron transport may either be due to defects inthe chioroplast or nuclear genome. It is assumed that any mutation in the chioroplastgenome which disrupts a component of a complex involved in electron transport will result inthe hcf phenotype. This is the case for Oenothera and maize mutants induced through theaction of a nuclear chloroplast mutator locus (Johnson and Sears, 1990; Mourad et al., 1989).Also, through the use of insertional inactivation via transformation, chloroplast genes havebeen deleted or altered in Chlamydomonas and Synechocystis resulting in the hcf phenotype(Blowers et al., 1990; Takahashi et al., 1991).The aim of this research was to isolate nuclear mutants defective in photosyntheticelectron transport. Possible nuclear hcf gene products are: 1) nuclearly-derived proteinsessential for a given complex; 2) proteins ivolved in assembly or stability of a complex; 3)proteins involved in specific translation, processing, or stability of particular chloroplasttranscripts; 4) proteins involved in transport of cytoplasmic derived polypeptides across themembrane, or insertion into the thylakoid membrane.The nuclear gene product responsible for the hcf phenotype has not been determinedfor any of the hcf mutants reported to date. The wild-type gene of the maize hcflO6 mutant,that was isolated due to the insertion of the transposable element mu has been cloned23(Barkan and Martienssen, 1991; Martienssen et al., 1989). However, the gene sequence andpossible function of the wild-type allele have not been reported.The majority of hcf phenotypes due to nuclear mutations have normal chioroplasttranscripts suggesting that the wild-type allele is required post-transcriptionally (Barkan etal 1986; Jensen et a]., 1986). Three exceptions to this have been reported in Chiamydornoriasand one in maize. Jensen et al. (1986) reported on a Chiamydomonas mutant GE2.1O,lacking chloroplast transcripts, originating from the psbB operon. No other chloroplasttranscripts were affected in this mutant. In vitro transcription of other chioroplast genesrevealed that translatable RNA was present, and pulse labeled experiments showed thatother components of PSII were translated, but did not accumulate. Seiburth et al. (1991)have shown that the psbB gene is transcribed in this mutant, but the message does notaccumulate. The simplest explanation for the results is that a nuclear factor is requiredspecifically for the stability of the psbB transcript.Other nuclear mutations in Chlamydomonas that appear to involve stability ofspecific chioroplast transcripts are the mutants 6.2z5, in which the psbC transcript isdepleted (Sieburth et al., 1991); and nczc2 , involving the psbD transcript (Kuchka et al.,1989). In both instances, the message appears to be transcribed at wild-type levels, but nosteady state mRNAs accumulate. Other chloroplast transcripts appear to be translated inthese mutants, but no PSII polypeptides accumulate in the thylakoid membranes. As in thecase of the GE2.1O mutant above, the simplest explanation for the results is that a nuclearprotein product involved in stability of the chloroplast transcript is missing. These examplesserve to illustrate that mutations that give rise to the hcf phenotype are not simply due tothe lack of one of the polypeptides of a complex.24Many hcf mutations have also been shown to affect two or more of the complexesinvolved in electron transport (Barkan et al., 1886; Simpson and von Wettstein, 1980; Tayloret al., 1989). The cause of the multiple pleiotropic effect in these mutants is not understoodwith one exception, hcf38 in maize (Barkan et al., 1986). Analysis of chloroplast mRNAtranscripts of this mutant revealed an altered transcript originating from the psbBpolycistronic operon. Also, the petA transcript was not present. Thus it can be inferred thatboth P511 and cytochrome b6i’fcomplexes would be affected, which was indeed observed. Noother differences at the mRNA. level with loss of multiple complexes have been reported.Thus, factors which may be involved in assembly andlor stability of more than one complexremain to be determined. The isolation of the hcflO6 wild-type gene, which results in theloss of several chloroplast complexes, should aid in the analysis when it is reported (Barkanand Martienssen, 1991).1.5. CHLOROPLAST BIOGENESISThe chloroplast’s progenitors are small undifferentiated proplastids that, in mostcases, are inherited maternally. The differentiation of the proplastids into chioroplasts inmesophyll cells, chromoplasts in petals and fruit, and amyloplasts in tubers, is dependent onpositional information. Nuclear derived positional information can further establishspecialized chloroplasts within a single leaf. For example, mature chloroplasts are foundonly in mesophyll cells. Furthermore, chioroplasts in mesophyll cells of C4 plants aredeficient in proteins involved in CO2 fixation, whereas the bundle sheath cell chloroplastsare deficient in PSII proteins. Regulation of these phenomena can be observed at the level ofmRNA accumulation for the proteins involved (Langdale et al., 1988).25Although development of photosynthetic plastids from proplastids is regulated by avariety of environmental factors, such as nutrient availability (carbon, nitrogen), the primarycontrol in vascular plants appears to be light. Some plants, for example barley, are able toproceed with the early phases of chioroplast and leaf development in the absence of light but,most dicots will follow a developmental strategy termed skotomorphogenesis in the absenceof light. The proplastids develop into small undifferentiated etioplasts under theseconditions. Leaf development is arrested, and cellular differentiation is retarded. Control ofthe skotomorphogenic strategy in these dicots appears to be due to negative nuclearregulatory genes. This conclusion is based on mutants that fail to follow theskotomorphogenic strategy under dark grown conditions (Chory et al., 1989a). Uponillumination the proplastids differentiate into photosynthesizing plastids. Chloroplastdifferentiation leads to activation of many nuclear and chloroplast genes (Deng andGruissem, 1986; Rodomel and Bogorad, 1985). Although it has not been possible to uncoupleleaf development and chioroplast differentiation (Iba et al., 1991), light regulation ofchloroplast development appears to manifest itself primarily at the post-transcriptional andtranslational levels (Krupinska and Apel, 1990; Reinbothe and Parthier, 1990). In spinachand maize, chioroplast transcription is constitutive and only increases 2-3 fold uponillumination, (Deng and Gruissem, 1987; Deng et al., 1987; Rock and Barkan, 1987).Essentially no change in transcription was observed in dark grown barley illuminated for16h (Krupinska and Apel, 1989) except for the psbA gene, where an increase was found(Klein and Mullet, 1987). The difference observed in relative chloroplast mRNA abundancemay be due to stronger promoters in the light. Differences in transcriptional start sites inlight vs dark have been reported for several chloroplast operons (Eisermann et al., 1990;Gamble et al., 1988; Haley and Bogorad, 1990).26Quantitative and qualitative increases in mRNA transcripts were observed uponillumination in Sorghum bicolor, a C4 plant (Schrubar et al., 1990). Thus the actual role oftranscriptional regulation may be different in some C4 plants. Further studies are necessaryto determine if the transcriptional regulation reported for sorghum is representative of C4regulation.Although some light and developmental transcriptional differences can be observed,the primary control of accumulation of chloroplast-gene products appears to be at thetranslational level. Accumulation of chlorophyll-associated proteins follows a sequential andorderly pattern (Croxdale and Omasa, 1990) and is dependent on Chi formation. In theabsence of Chi, Chl-binding proteins are rapidly degraded (Mullet et al., 1990).1.5.1 Light ReceptorsSeveral photoreceptors, namely phytochrome, a blueftJV-A light receptor (alsotermed cryptochrome), an UV-B light receptor, and protochlorophyllide reductase, are usedto mediate informational signals from incident radiation in the control of plant development.Proto-chlorophylide reductase requires light to convert proto-chlorophyllide to Chl, and thusfalls into the category of light receptor. However, the major light receptors that are involvedin plant development are phytochrome and the bluefLIV light receptors.Phytochrome is the best characterized plant photoreceptor (Colber, 1988). It absorbseither red or far red light, depending on the conformation of the molecule, and triggers acascade response involved in photomorphogenesis. The best studied effect of phytochromehas involved dark-grown etiolated seedlings which are then subjected to a particular lighttreatment. Upon illumination with red light, phytochrome is converted to the active form(Pfr), and illumination with far red light converts it back to the inactive Pr form. Although27the changes in the structure of the phytochrome molecule have been documented, how thesechanges occur especially in relation to the protein portion of phytochrome and what thesechanges ellicit with respect to the signal transduction mechanism is still unresolved.In spite of the wealth of information on the effects of red and far red light on seedgermination, greening, nuclear transcription, stern elongation, and flower induction, verylittle is known of the role of phytochrome in chloroplast development and its role in greentissue under normal physiological conditions. Mutants lacking detectable phytochrome arefully autotrophic and survive to produce seeds (Chory et al., 1989b). The mutant plants aresmaller, have smaller leaves, fewer chioroplasts per mesophyll cell, reduced grana densityand higher Chia to Chib ratios. Thus while phytochrome does not appear to play a majorrole in triggering chioroplast development nor in regulation of chloroplast genes, it maymodulate the amount of chioroplast development.In higher plants, response to different light treatments in the blue and ultra violetregion include phototropism, growth inhibition, and stomatal opening. To date, no pigmenthas been demonstrated convincingly to be a blue light photoreceptor (and for this reason theblue light receptor has been designated cryptochrome), although carotenoids and/orflavanoids are the most likely candidates. Photomorphogenesis, per Se, does not appear torequire blue light to proceed, as red light alone is sufficient. Thus blue light, acting inconcert with phytochrome, exerts additional stimulatory effects on nuclear gene expression.In addition to the bluef[JV-A responses there is also a response in the UV-B region of thespectrum, which is not directly involved in photomorphogenesis, but has been shown toinduce nuclear genes primarily in the phenyipropanoid pathway (Halbrock and Scheel,1989).28The physiological responses above are at least several minutes removedfrom thetime of light reception, and thus probably represent events later in the transductionpathway. As expected, several biochemical, biophysical and enzymatic changeshave beenreported that are due to different fluence rates in the blue and UV light regions (Short andBriggs, 1990). Gamble and Mullet (1989) reported that the chioroplast dicistronic messageencoding the PSH D2 (psbD) and CP47 (psbC) polypeptides is induced by blue light.Although a pulse of far red light can attenuate partially the transcript accumulation, itappears that phytochrome is not involved specifically in induction as redlight by itselfelicited no response. The induction by blue light was not observed in the presence ofcyclohexamide (an inhibitor of cytosolic protein synthesis), thus suggesting that a blue lightresponsive nuclear gene product is required.One method for analysis of the blue light responses is through the use of mutants.Extensive use of mutants in fungi has revealed numerous mutants with altered response toblue light (Senger, 1987). Similar use of mutants in higher plants has been meager, butmutants with altered curvature response to blue light, presently under studyby Poff and coworkers, should begin to bear fruit in the near future (Khurana and Poff 1989; Kurana et al1989). In one such mutant no phototropic curvature was observed. The lack of responseinthis mutant appears to be solely due to lack of phototropic response, as normal gravitropicresponse was noted.In another study, Liscum and Hangarter (1991) isolated three mutants ofArabidopsis that did not respond normally to blue light. The screening technique employedwas to select for long hypocotyl elongating seedlings exposed to blue light. This is a similartechnique as was used to isolate hy mutants described above. However, the mutants that areinsensitive to blue light are not allelic to the hy mutants. These mutants, calledblu (for hlue29light linresponsive), are only unresponsive to blue light, and responded normally to whitelight and far red light. No phenotypic differences were observed in the mutant plants grownunder white light, and the wild-type product appears to be required only during earlyseedling development.1.5.2 Chioroplast SignalThe state of the chloroplast has been shown to be important in determining whethernuclear genes are transcribed (Taylor, 1989). This has led to the speculation that a plastidsignal regulates nuclear transcription. The requirement for a plastid signal was postulatedby Bradbeer et al. (1979) in an analysis of carotenoid-deflcient mutants of barley. Themutants are deficient in plastid ribosomes, thus fail to accumulate the chioroplast encodedproteins, but also do not accumulate transcripts of nuclear encoded chioroplast directedproteins. Similar results have been shown in carotenoid deficient mutants of maize(Harpster et al., 1984, Mayfleld and Taylor., 1986).Further evidence for the requirement of a plastidic signal has been from the work ofOelmuller el al. (1986; 1988). Using the herbicide Norfiurazon, which blocks carotenoidbiosynthesis and leads to photooxidation of Chi, the authors found that the level of nucleartranscripts of protein products destined for the chioroplast were reduced. Treated seedlingsgrown under far red or dim light accumulate normal levels of Chi, and the levels of nuclearcab transcripts is normal. Thus, it appears that it is the photooxidation which is responsiblefor the low levels of the transcripts observed, and not the carotenoid deficiency per Se. Thisconclusion was further supported by the use of a double mutant in maize blocked in bothchlorophyll and carotenoid biosynthesis (Burgess and Taylor, 1988). Under low lightintensities, cab mRNA was observed to be at normal levels in the double mutant compared to30a Norfiurozan treated wild-type. However, once transferred to high light, photobleachingoccured in the wild-type, and lower levels of the cab transcript were found, whereas therewas no effect on the mRNA level in the double mutant.At present the nature of the plastidic signal is unknown. In each of the proceduresabove, multiple pleiotropic effects are also observed which make it difficult to draw specificconclusions regarding the nature of the signal. It is well established that organelle signalscan regulate nuclear genes in yeast mitochondrial biogenesis and development. In yeast,heme, which is synthesized in the mitochondria, has been shown to be a componentregulating nuclear gene transcription (reviewed in Forsburg and Guarente, 1989). However,the genes involved in heme biosynthesis are nuclear. The fact that mutants of yeastimpaired in electron transport show very little difference in nuclear steady state RNA levels,including those part of the respiratory electron transport, has led to speculation that there isno mitochondrial-specific signal directed to the nucleus (Tzagoloff and Myers, 1986).However, Parikh et al. (1987) have identified nuclear transcripts that are increased in yeastcells with deletions or complete loss of mitochondrial genomes. The identity and function ofthe protein products of these transcripts are not known. The authors speculated that somemetabolic compound synthesized in the mitochondria might function as a regulatorymessenger, although identification of such a compound has eluded researchers.1.5.3 Nuclear Genes Involved in Chioroplast BiogenesisOne of the best examples to illustrate the importance of nuclear genes in chloroplastdevelopment are the deticop mutants ofArabi.dopsis (Chory et al., 1989a; Chory et al., 1991;Deng et al., 1991). The mutants were isolated by screening for normal light-grownmorphology in plants germinated in the dark. Presently, three mutant phenotypes have31been reported in the literature: deti (Chory et al., 1989a), det2 (Chory et al., 1991) and copi(Deng et al., 1991). These mutants display many characteristics of light grown plants, suchas expanded cotyledons and primary leaves, short hypocotyls, accumulation of anthocyaninsand some differentiation of chloroplasts in the dark (deti only). The results from thesemutants suggest that the etiolated morphology in the dark is due to a nuclear negativecontrol factor. The deti mutation also de-regulates the tissue specificity of the light-inducible genes, as chioroplast development is observed in the roots. Some phenotypicdifferences are also observed in the det mutants grown under normal light conditions,suggesting that the mutation also affects a component necessary for normal development(det2 and copi).The det/cop mutations apppear to act very early in chioroplast develoment.Mutations that affect the assembly of components involved in electron transport, on theother hand, would act later in the signal transduction pathway. With the exception ofnuclear gene products that code for strurtural components of photosynthetic complexes, andenzymes of carbon and nitrogen metabolism in the chioroplast, knowledge of the role ofnuclear derived genes in the chioroplast is meager. One example is nuclear gene productsthat bind to the 3’ end of chioroplast transcripts (Stern et al., 1989). Additional nuclearencoded chioroplast-localized RNA-binding proteins have been isolated (Li and Sigiura 1989),however, the function of these is unknown at the present. In the case of the former,chloroplast 3’ mRNA binding proteins were observed to be general in nature, that is, thebinding proteins were found associated with all chloroplast transcripts analyzed, whileothers appear to be transcript-specific. One such general 3’ binding protein has beenpurified, and the gene has been isolated (Kiaff and Gruissem, 1991). The relevance of the 3’binding proteins has not been unequivocally demonstrated, but they have been postulated to32be stabilizing proteins, and may also be involved in directing splicing. Precedence for theinvolvement of nuclear-derived organelle mRNA binding proteins in stability and processinghas been shown in mitochodrial transcripts of yeast and Neurospora. For example, in yeast,the MRS1 gene appears to be specifically required for excision of two introns of the genescob-box and coxl (Bousquet et al. 1990). Yeast strains that have the two introns deleted inthese genes are phenotypically wild-type and deletions in other introns had no affect. Thusthe authors conclude that the sole raison d’etre of the MRS1 gene is in splicing the twointrons.Several nuclear encoded genes which are involved in mitochondrial transcriptionalprocessing also have a second function. While most are also involved in translation (Fox,1986; Grivell, 1989), two interesting examples were found in yeast and Neurospora where theprotein involved in mitochodrial RNA processing is also an aminoacyl tRNA synthetase. TheNAM2 gene in yeast encodes the mitochondrial leucyl tRNA synthetase (Herbert et al., 1988)and in Neurospora the cyt-18 mutant gene encodes the mitochondrial tyrosyl tRNAsynthetase (Askins and Lambowitz, 1987). The mutant phenotype was associated in bothcases with the splicing defect in which the target intron was not properly excised For thecyt 18 gene product, the splicing was found to be specifically associated with a small portionof the amino terminus of the protein (Cherniack et al, 1990) The amino acid sequencefound on this portion of the protein was not homologous with other tyrosyl tRNA synthetasesfrom any other species studied.While no such factors have been isolated which function similarly in the chioroplast,maize and Chiamydornonas hcf mutants with altered chioroplast RNA transcripts (Barkan etal., 1986, Jensen et al., 1986; Sieburth et al., 1991) suggest that similar nuclear factorsinvolved in processing are present. The present study describes three hcf mutants that also33have altered pattern of chioroplast transcripts. Although the molecular basis for the mutantphenotype has not been determined, the abnormal transcription pattern observed wascorrelated with mutant phenotype. Using procedures that are presently being developed inthe Arabidopsis genome mapping program, it should be possible to clone these genes in thenear future.34CHAPTER 2MATERIALS AND METHODS2.1. PLANT MATERIAL AND GROWTH CONDITIONSSeeds from Arabidopsis thaliana (L.) Heynh. wild-type Columbia were mutagenizedin 0.3% (vlv) ethyl methane sulfonate for 12.5 hr. M2 seeds from 570 small M1 bulkpopulations (average 7 plants) were surface sterilized and germinated on 112 MSO (112concentration MS salts [Murashige and Skoog, 1962], 5 g L1 sucrose) petri plates andscreened for the high chlorophyll fluorescence (hcf) phenotype visually under ‘UVillumination as described by Miles (1980). Approximately 100 M2 plants from each M1 bulkfound to contain the hcf phenotype were grown in soil, seed collected individually from eachplant, and a small sample screened for the hcf phenotype. Plants which segregated at a 3wild-type: 1 hcf ratio were rescreened the following generation to confirm the hcf phenotype.Each line is derived from a single M2 plant from each M1 bulk. Aflow chart of the mutantisolation procedure is presented in Figure 3. Since the hcf phenotype is seedling lethal, theexpected segregation frequency of self-pollinated heterozygous plants grown in soil is: onethird homozygous wild-type and two thirds heterozygous for the hcfphenoytype. Each linewas maintained by screening individual plants segregating 3 wild-type: 1 hcf on a sucrosesupplemented medium (112 MSO) each generation.Plants were grown in a controlled environment chamber under fluorescent lights(100-150 pE m2s1, 23±3 C, 16 h light’ 8 h dark photoperiod) on a vermiculite:peat (3:1)mixture and bottom watered with the addition of a 20-20-20 all purpose fertilizer (PlantProducts, Co., Ontario, Canada) at a rate of 0.8 g as needed. Harvest of material foranalyses was done on plants which were germinated on 112 MSO petri plates under35MutagenizeArabidopsis wild-type Columbia seeds in 0.3% EMS for 12.5 hrsPlant 7 mutagenized seeds per pot‘I,Harvest each pot separately. Each constitutes one M1 bulk populationScreen a small sample of progeny of each M1 bulk population (M2)on 1/2 MSO petri platesfor high fluorescencePlant M2 seeds (approximately 100) from promising M1 bulks in soil..1Harvest seeds from each self-pollinated M2 plant separately.Screen the progeny from each M2 plant (M3 generation) for the presence of the hcfphenotype segregating at a ratio indicative of a single nuclear recessive (3 wild-type: 1 hcO.“Plant seeds from selected M2 plants (if more than one), self pollinate and screen the progenyfor verification of the hcf phenotype and for additional morphological phenotypes.“The progeny of M3 plants confirmed to segregate for the hcf phenotype in the M4 generationat a ratio indicative of a single nuclear recessive are kept, and now constitute an hcf line.One additional generation of self-pollination and screening was done prior to experiments.Each line can be traced to a single M2 plant from each M1 bulk population.Figure 3. Flow diagram of hcf mutant isolation procedures. The proceduredepicted was used to isolate hcf lines derived from 21 M1 bulk populations.36sterile conditions and seedlings transferred to fresh 250X15 mm 1/2 MSO plates to allow forleaf expansion following germination and identification of hcf plants. Each plate typicallycontained 15-20 hcf plants and 10-15 wild-type plants. Plates were placed in environmentalgrowth chambers under full spectrum fluorescent lights (Vita-light, GE) (25-40 1iE m21),16h lightJ 8h dark, 23±3 C. After approximately four weeks (at the onset of bolting), theaerial portion of the plants was harvested. Each experiment consisted of harvesting the hcfand wild-type siblings (homozygous wild-type and heterozyous plants) from 3-5 plates(typically 40-80 plants of each) grown under the same conditions.2.2. FLUORESCENCE MEASUREMENTSIn vivo fluorescence measurements were made at room temperature on plants grownon 112 MSO medium using a computer-aided video fluorescence imaging system (CAVFIS)(Fenton and Crofts, 1990) in the M2 generation, and in later generations using a pulsemodulated fluorescence apparatus (PAM 101, Walz, FGR) (Schreiber et al., 1986). Plantswere dark-adapted for 5 mm. prior to induction, and when using the PAM 101 a modulatingbeam (1.6 kH) determined the F0. Actinic light (70 iLE m2i for the CAVFIS and 10 iEm2s or 100 pE m2 when using the PAM fluorometer) was used to measure thefluorescence kinetics. When using the PAM fluorometer, saturing light pulses (2000 1iE m2s’- ) were applied in some instances in order to determine photochemical (qp) and nonphotochemical (q) quenching (van Kooten and Snel, 1990)2.3. ISOLATION OF THYLAKOID MEMBRANES AND PIGMENT ANALYSISThe thylakoid membranes were isolated by grinding the leaves in 20 mM Tricine (pH8.0), 10 mM NaCI, and 0.4 M sucrose with a mortar and pestle. The homogenate was filtered37through 125 tm bolting silk, and centrifuged at 4340 g for 12 mm. The pellet was washedonce in the same buffer and suspended in 20 mM Tricine (pH 8), 150 mM NaCI, 5 mM MgCl2.Total Chi per plant was determined by assaying the thylakoid membrane fraction andfraction retained on the bolting silk by the method of Arnon (1949).For pigment analysis, 3-4 week old hcf and wild-type plants were dark adapted for 1-2 h, collected and ground to a fine powder in liquid nitrogen. The powder was dispersed inbuffered (10 mM Hepes pH 7.5) acetone (80% v/v). An equal volume of diethyl ether wasadded to the acetone extract and the acetone was removed by extraction with 3-4 vol. of 10%KCI. After extensive washing with water the organic phase was evaporated under nitrogenand the pigments were redissolved in ethanol and used immediately for HPLC analysis.Pigments were seperated by reversed phase HPLC using a LiChrospher 100 RP-18 column (5aim, 4 mm I.D. X 125 mm length) (Merck, Darmstadt, FRG). Separation of the pigments wasdone in a single run with a linear gradient of solvent A (75/15/10 acetonitrile! methanol!tetrahydrofuran, v/v/v) with 12% water (vlv) to 100% solvent A after 12 mm. at a flow rate 2ml mm ‘1. The pigments were detected at 445 nm using a Waters 994 photodiode arraydetector. IdentifIcation of the pigments was done by comparison of their absorption spectraand retention behavior using purified pigments. Calibration was done by using Chia (Sigma)as a standard.2.4. ELECTROPHORESIS AND IMMUNOBLOTTINGFor separation of ChI-protein complexes, thylakoid membrane samples correspondingto 25 j.tg ChI were pelleted, washed with 65 mM Tris-maleate (pH 7.0), and solubilized in 88mM octyl-13-D-glucopyranoside at a detergent:Chl ratio of 30:1. Electrophoresis was run at 25mA, for 4-5h, at 4C in the dark on a 1.5 mm thick, 10% polyacrylamide gel.38Denaturing gels were run on 10% polyacrylamide gel containing 0.1 M Tricine in thecathode buffer as described by Schagger and von Jagow (1987) modified so that the final.concentration of Tris (pH 8.25) in the gel was 1.0 M. Thylakoid membrane samples werepelleted and resuspended in 65 mrvi Tris-HC1 (pH 6.8), 20 mM dithiothreitol, 10% ethyleneglycol and 2% SDS and heated at 65C for 15-20 mm prior to loading onto the gel. Samplesof 25 jig Chi (for Coomassie Blue staining) or 5 jig Chi (for western blotting) were loaded fromboth hcf and wild-type plants. Electrophoresis was run at 35 mA for 5-6 h, at roomtemperature. Proteins were transferred onto nitrocellulose and visualized as described byWhite and Green (1987).Anti-sera prepared against CPI and CPu from barley (White and Green, 1987); PSIsubunits H, V and VI from spinach (Bengis and Nelson, 1975); cytochrome [(a gift from R.Malkin); PSII chlorophyll-protein core complexes CP47 CP43 (a gift from N.-H. Chua), PSIIcore complex Dl (a gift from L Mcintosh), cytochromeb559 (a gift from W. Cramer); andcoupling factor CF1 (Moase and Green, 1981) were used to determine the presence ofpolypeptides. Following staining with one antibody, the nitrocellulose filter was strippedovernight in a solution containing 0.1 M glycine (pH 2.2), 20 mM Mg acetate, and 50 mM KCIand reblotted (Legocki and Verma, 1981).Visualization of cytochromes on non-denaturing gels was done using 3,3’,5,5’-tetramethylbenzidine (TMBZ) and hydrogen peroxide following the procedures of HoyerHansen (1980).2.5. PHOTOSYNTHETIC ELECTRON TRANSPORT MEASUREMENTSPSI-dependent electron transport was assayed at 25C in the presence of 0.1% NaN3,0.5 mM N,N,N’,N’-tetramethyl-p-phenylenediamine (reduced with 2.5 mi ascorbate), and 139jiM 3-[3,4-1)ichlorophenyl]1,1-dimethyl-urea (DCMIJ), monitoring 02 consumption in thepresence of 0.1 m methyl viologen in a YSI electrode (Allen and Holmes, 1986). Thylakoidmembranes (25 pg ChI) were added to the reaction mixture which contained 100 mM sorbitol,20 mivi Hepes (pH 7.8), 10 mM NaCI, and 2.5 mM MgC12. Saturating white light wasprovided through Fibre-light 180 (Dolan-Jenner) light pipe. PSII-mediated 2,6-dichiorophenolindophenol (DCPIP) reduction was measured at 595 nm using a Cary 210(Varian, Palo Alto CA) spectrophotometer (Allen and Holmes, 1986). Thylakoid membranes(5 pg ChI) were added to the reaction mixture (20mM MES pH 6.0, 25 jiM DCPIP). Electrondonors used were water or 0.5 mM diphenylcarbazide (DPC). For each experiment in whichboth water and DPC were used, a measurement was first obtained with water, DPC added tothe same sample, and a second measurement taken.2.6. RNA ISOLATION AND RNA BLOT ANALYSiSTotal RNA from hcf mutants and their wild-type siblings was isolated from 3-4 weekold plants grown on 1/2 MSO plates (described above). All plant material was collectedbetween 12:30 and 14:00 hours, weighed, and immediately frozen in liquid nitrogen untiluse, except for hcf5 which was harvested at 10:00 and used immediately.RNA was isolated following essentially the procedures of Parsons et al. (1989). Theleaf material was ground to a fine powder in liquid nitrogen. The leaf powder was thensuspended at 6WC in 4/5 extraction buffer: 115 buffer equilibrated phenol (v/v) at 1 ml pergram of tissue. The extraction buffer consisted of 100mM Tris, pH 8, 20mM EDTA, 0.5 MNaCI, 0.5% SDS; 0.5% 2-mercaptoethanol. The slurry was agitated at 65C for 5 mm, and anequal volume of chloroform added. The aqueous layer was saved, and an equal volume of 1:1phenol: chloroform (v/v) at 6WC added. The aqueous layer was extracted twice more with 140volume of chloroform. LiCI (115 volume of 10 M) was added to the final aqueous layer. Thesample was placed at 4°C overnight, and centrifuged (17000 g for 10 mm) the followingmorning. The RNA pellet suspended in water and concentration determinedspectrophotometrically (Sambrook et al., 1989).RNA (approximately 10 rig) was separated in formaldehyde-containing agarose (1.2%w/v) gels, and blotted onto Hybond.NTM (Amersham) membrane (Sambrook et al., 1989).The filters were prehybridized for 2 to 4 hr at 65°C in a solution containing 0.9 M NaCJ, 0.09M sodium citrate, 0.5% SDS, 5X Denhart’s reagent (0.1% bovine serum albumin, 0.1% Ficoll,and 0.1% polyvinylpyrrolidone), 100 ig!mL denatured salmon sperm DNA. Specifichybidization probes were prepared by labeling DNA with 32P using the random primermethod following manufacturer’s recomendation (Boehringer Mannheim), and hybridized for12-16 hr at 65°C. The filters were washed several times at 65°C in 2X SSC (0.3 M NaCI, 0.03M sodium citrate), 0.1% SDS (w/v) for 2 to 4 hr and exposed to Kodak XAR-5 film at -90°Cwith an intensifying screen.DNA probes for nuclear genes used in this study were cabil, a 0.7 kb PstlXbafragment of the tomato gene for the light-harvesting chlorophyll protein associated with PSI(Kirsch et al., 1986) ; psaD, a 0.3 kb EcoRI fragment containing the tomato PSI subunit IIgene (Hoffman et a!., 1988). DNA probes for chioroplast genes used were atpA, a 1.1 kbBamHhJSalI fragment of the maize coupling factor (CF1)alpha subunit (Barkan et al., 1986);atpB/E, a 2.2 kb Xbal fragment from the spinach CF1 beta and epsilon subunits (Henningand Herrmann, 1984); petA, a 478 bp EcoRliXhol fragment containing the spinachcytochrome f gene (Alt and Herrmann, 1984); petB, a 309 bp Xhol fragment containing thespinach cytochrome b6 gene (Heinemeyer et al., 1984); psaB, a 1.7 kb BamHl fragment ofthe spinach 82 kd PSI reaction center polypeptide (Kirsch et a!., 1986); psaC, a 0.3 kb RsaI41fragment of the spinach PSI subunit VII polypeptide gene (Steppuhn et al., 1989); psbB, a 1.1kb BamHlLXbal fragment containing the PSU 47 kd polypeptide gene from spinach (Morrisand Herrmann, 1984); psbD, a 989 bpEcoRIJPvuII fragment containg the PSIJ D2 coreprotein gene from spinach (Alt et al., 1984).2.7. ELECTRON MICROSCOPYElectron microscopy was carried out on leaves of the mutants and wild type siblingsgrown on 1/2 MSO for two weeks. The leaf tissue was fixed in 2.5% (v/v) glutaraldehyde in0.1 M sodium cacodylate buffer (pH 7.2) for 1 h, and stained in 1% (wlv) osmium tetroxide inthe same buffer for 1 h. The tissues were then dehydrated in a graded ethanol series,followed by propylene oxide, and embedded in Spurs resin. Ultra thin sections were stainedwith uranyl acetate and lead citrate and examined m a Zeiss (EM1O) electron microscope.2.8. GENETIC ANALYSIS2.8.1 Marker LinesThe following tester lines containing visible phenotypes were used in crosses.Chromosome 1 (W2:an, disi; W4: chi, api, g12), chromosome 2 (W6: as, cer8), chromosome 3(W7: hy2, gil, tt5; W8:gii, cer7; RDM3: hy2,gil, cer?), chromosome 4 (W10: cer2, ap2;RDM4: bp, ap2, cer2), chromosome 5 (W13: ttg, yi). All lines are in the Landsburgbackground. The W lines were obtained from Dr. David Meinke (Oklahoma StateUniversity, Stillwater, OK), and were originally constructed by Dr. Maarten Koornneef(Agricultural University, Generaal Foulkesweg, Wageningen, Netherlands). The two RDM42lines were constructed by the author (sandy inkins Marker). The phenotype of each visiblemarker and relative location is presented in Table Linkage analysis- Immature flowers from plants homozygous for the marker lines were opened and thestigma was dusted with pollen from heterozygous hcf plants. The visible marker lines wereused in all cases as the female parent to facilitate the identification of accidental selfs amongthe F1 progeny. Since the hcf mutations are lethal, it was not possible to evaluate the hcfphenotype in relation to the visible markers. Instead, deviations greater than the expected3:1 segregation for the visible markers would be indicative of linkage of the hcf phenotypewith a visible marker. Complete linkage, for example, should give a 2:1 segregation ratio forthe visible marker in the F2 generation.Since it is not possible to distinguish the heterozygous hcfY+ plants from homozygouswild-type (+1+) several crosses using different plants from each hcf line were made to eachmarker line. Putative hcf7+ plants used in crosses were self-pollinated and heterozygousplants identified. The F1 were grown from crosses that were verified to have been doneusing the heterozygous hcf plants. The F1 was self-pollinated and individual plantsharvested and the F2 analysed for the presence of the hcf phenotype. Followingidentification of lines which segregated for the hcf phenotype, 100-400 F2 plants were grownand screened for the presence of the visible markers. Since the hcf phenotype is lethal insoil, segregation frequency greater than 25% for a given visible marker should be indicativeof linkage of a lethal phenotype with the visible marker.—43an 1:0ap-1 1:102ap-2 4:51as 2:33bp 4:11cer-2 4:42cer-7 3:85cer-8 2:51c/i-I 1:60cp-2 2:7dis-1 1:18er 2:16gl-1 3:40gl-2 1:122hy-2 2:0tt-5 2:78ttg 5:32yi 5:87 Yellow inflorescence, yellowish flower budsTable 5. Genotype ofArabidopsis thaliana visible markers used in crossesLocus Chr.a Name and Description -Angustifolia: Narrow leavesApetela: no peta1sApetela: no petalsAsymetric leavesBrevipedicellus: very short pedicels, siliques bend downwardsEceriferum: no epidermal waxEceriferum: like cer-2Eceriferum: reduced epidermal waxChlorina: light green color due to loss of ChibCompacta: semi-dwarfDistorted trichomesErecta: compact inflorescence, fruits bluntGlabra: no trichomes on leaves or stemsGlabra: reduced trichomes on lower leaves, no trichomes on upper leavesLong hypocotyl, reduced level of phytochrometransparent testa, yellow seed due to transparent testaTransparent testa, glabra: (see tt-5 and gl-1)a Chromosome: Chromosomal location. From Koornneef et al. (1983)442.9. ARABIDOPSIS TISSUE CULTURE2.9.1. Plant Material and Growth ConditionsArabidopsis thaliana seeds were kindly provided by 11W. Meinke (Oklahoma StateUniversity: Columbia), M. Jacobs (Vrije Universiteit Brussels: C24), and A.R. Kranz (J.W.Goethe-Unversität Frankfurt: Landsburg erecta). The Point Grey ecotype was collected bythe author and A.J.F. Griffiths on the Point Grey peninsula in British Columbia, Canada.All plant material was grown in a controlled enviromental chamber under fluorescent lights(100-150 jiE m2 s ) 16 h lightl8 h dark, at 23 ± 3’ C.2.9.2 Tissue Culture.Seeds were surface sterilized in 95% EtOH (30 sec.), and 50% Clorox (6 mm.), rinsedin sterile distilled water 3-4 times and allowed to dry on a sterile filter paper. Seeds wereeither immediately sown on a germination medium (GM: 112 X MS salts; 5% sucrose; pH 5.8and solidified with 0.8% agar) by gently displacing the seeds from the filter paper with asterile forceps, or left in a sterile petri sealed with paraflim until use. Explant sources werecotyledons cut at the petiole from 7-10 d. old seedlings (Schmidt and Willmitzer, 1988,Chaudbury and Signer, 1989) or roots from 2-3 week old plants (Valvekens et al., 1988).Explants were placed on a thoot inducing medium reported by Lloyd et al. (1986) (SIM1: MSsalts, B5 vitamins, 3% sucrose, 1 mg L 6-benzylaminopurine, 0.1 mg L1 anaphthaleneacetic acid, pH 5.7, solified with 0.8% agar) or on a slightly modified SIMmedium (SIM2: same as SIM1 except a-naphthaleneacetic acid was increased to 0.5 mg L1)452.9.3 Genetic analysis of the Columbia X Point Grey explants.Seeds were harvested from the Columbia parent in which immature flowers wereopened, the anthers removed and pollen from the Point Grey ecotype applied (F1). SeveralF1 plants were also self-fertilized, the F2 seeds surface sterilized and germinated on 1/2MSO. Both cotyledons from Columbia, Point Grey, F1 and F2 seedlings were placed on SIM2side by side. The progeny from 20 random F2 plants were also evaluated on SJM2 in asimilar manner.2.9.4. Transformation.The Agrobacterium strains used for transformation experiments were A208 carryingthe avirulent nopaline-type plasmid pTiT37-SE and pMON41O binary plasmid (Rogers et al.,1987) (cotyledons) and C58C1 rif containing the PGV3850 Ti plasmid recombined with E. colipKU4 plasmid (Baker et al., 1987). Agrobacterium was grown overnight in Luria brothcontaining the appropriate antibiotics at 2W C on a rotary table (150 rpm). Explants wereplaced in 18 ml 1/2 X MS salts liquid medium and two ml ofAgrobacterium for 1-2 mm,rinsed with fresh liquid medium, blotted dry on filter paper and placed on SIM. In someexperiments root explants were placed on a callus inducing medium (Valvekens et al., 1988).After 4-5 days, the explants were rinsed in 1/2 MS liquid medium and transfered to SIM2containing 25 mg L1hygromycin and 500 mg L1 carbinicillin or 250 mg L1 cefotaxime(cotyledon explants), or 500 mg L1 vancomycin (root explants).46CHAPTER 3RESULTS3.1. ISOLATION OF HCF MUTANTSSeeds of small bulk populations of ethyl methane sulfonate (EMS) seed-mutagenizedArabidopsis thaliana M1 plants were individually collected and screened visually in the M2generation for the high chlorophyll fluorescence (hcf) phenotype (Figure 4). The majority ofthe bulk populations contained seven plants. A few pots contained less than seven plantsdue to mortality, and a few pots contained eight plants that were not thinned properly fromthe initial 10-15 seeds planted in each pot. The rationale for collecting small M1 bulkpopulations was that a relatively small number ofM2 plants from each M1 bulk would haveto be screened in the M3 in order to isolate an M2 plant heterozygous for a nuclear mutation,a necessary prerequisite to isolate mutations that are lethal in the homozygous condition. Asecond reason was that mutant lines derived from the different M1 bulks would be due todifferent mutagenic events, and not simply a re-isolation of the same mutation. Of 570 M1bulks screened visually for the hcf phenotype, approximately half (251) contained at leastone hcf seedling. This result, while surprising, was not completely unexpected due to thehigh number of known hcfloci in maize (Cook and Miles, 1988) which is a third reason smallM1 bulks were collected.The screening was done in two phases. Bulk populations 1-240 were screenedvisually only, while the remainder were also screened using a computer aided fluorescencevideo imaging system (CAFVIS) (Fenton and Crofts, 1990). The CAFVIS software allowedscreening the fluorescence kinetics on up to sixteen individual M2 plants simultaneously,quickly quantifying the fluorescence kinetics of putative hcf mutants and the wild-type47Figure 4. Arabidopsis thaliana hcfl mutant and wild-type plants.Photograph depicts plants under normal white light (A), under UV light, nofilter (B); and under UV light red filter (C). Mutant plants fluoresce redunder UV light and appear white to the eye when a red filter is used.4861siblings on the same petri plate, and it also provides a measure of the variability that will beobserved among the plants.A sample of fluorescence kinetic curves obtained using this method is presented inFigure 5. The fluorescence kinetics from wild-type Columbia plants grown on 112 MSO(Figure 5A) is essentially the same as the fluorescence kinetics observed in plants grown insoil (Artus and Somerville, 1988). This suggests that photosynthetic competence of wild-typeArabidopsis, at least with respect to the fluorescence kinetics, is normal when grown on 112MSO plates.• Many different patterns in the fluorescence kinetics were observed during the initialscreening (Figure 5). Miles (1980) outlined some of the more common fluorescence kineticsobtained from maize hcf mutants. Mutant plants from M1 bulk 60 appear to have a normalinduction to the maximal level of fluorescence (FM), but there is no decline to a steady statefluorescence (Fs). This indicates that charge separation and electron flow in PSII to QB isoccurring. Since there is essentially no decrease from the maximal fluorescence level (FM) inthe mutant plants, it suggests that QA is not being reoxidized. This would result if electronflow is blocked at some point past QB such as cytochrome b6If, or PSI (Miles 1980). A linederived from this M1 bulk that appears to segregate as a single nuclear recessive and isdefective in the PSI complex will be further described below (hcfl).M2 plants from M1 bulks 239, 302 and 530 have a high initial fluorescence (F0), andlittle variable fluorescence (Fv). This suggests that there is very little, if any, electron flowthrough PSII. Similar fluorescence kinetics has been observed in PSII hcf mutants ofChiamydomonas (Bennoun and Chua, 1976) and maize (Leto and Miles, 1980). A linederived from M1 bulk 302 has been isolated and will be further described below (hcf).50550400 B500______________________450_________-zz400300200 250350300100______ ______ ___ __ ___ ______ ___________ __________) 4812162024 2100 4812162024 28700 V 900 D800600700C.)500 600400 500—400 V300300200V 200100048 1216202428 1) 4 8 1216202428900 700F600700flz500400300200 200100 4812162024 28 1004Time (s)Figure 5. Fluorescence induction kinetics of wild-type and hcf mutant plantsin the M2 generation. Fluorescence kinetics of a representative sample fromwild-type Columbia (A) and mutant and wild-type siblings of M2 plants frombulk populations 60 (B), 239 (C), 302 (D), 530 (E) and 559 (F) taken on acomputer-aided video fluorescence apparatus (CAFVIS). Measurementstaken from 6-10 individual plants simultaneously. Fluorescencemeasurement are in arbitrary units.51Although attempts were made to isolate lines segregating for the hcf phenotype from M1bulk 239, none have been recovered. Possible reasons for this will be discussed below.The fluorescence kinetics ofM2 plants presented thus far are examples ofM1 bulksthat were selected as candidates for further study. Figure 5F provides an example of an M1bulk which was not selected for further investigation. The reasons are twofold. First, thereare two distinct patterns of fluorescence originating from different plants. This could meanthat more than one mutation, which gives rise to the hcf phenotype, originated in this M1bulk, or that a second mutation has some synergistic or pleiotropic effect on the hcfphenotype. A second feature of this M1 bulk, is that while the initial fluorescence fromseveral plants is higher than the wild-type, the variable fluorescence (F) and quenchingappears to be relatively normal. Similar fluorescence kinetic patterns have been described inmaize, and are correlated with mutations that affect CO2 fixation (Edwards et al., 1988;Miles, 1980). Since the the goal of this project was to select mutants that might be blockedearly in electron transport, only M1 bulks containing hcf plants that displayed little or nodecrease in the maximal fluorescence, and with all the hcf plants having similar inductionkinetics, were selected as candidates for further study. Some of the lines isolated from M1bulks prior to analysis with the CAFVIS that appear to be blocked subsequent to PSI, wereisolated, but have not been analyzed (see below).In order to isolate heterozygous plants segregating for the hcf phenotype as a singlenuclear recessive, approximately 100 individual M2 plants from each selected bulk weregrown in soil, and the progeny were screened on 1/2 MSO petri plates. Individual M2 plantsfrom 27 M1 bulks have been screened, and 21 have yielded at least one M2 plant segregatingat a ratio indicative of a single nuclear recessive in the M3 generation (Table 6). Screeningadditional M2 plants from the six M1 bulk populations in which no hcf lines have been52Table 6. Arabidopsis high chlorophyll fluorescence M1 bulk screen.M1 Bulk Present No. M2 plants M2 plantsnumber designationa screenedb seg. for hcf’16 N 72 14B,31C,34020 hcf2 71 1A,12A,.141 N 72 8A,29C,34B44 N 64 23C,28A45 N 57 4054 N 72 31B,32B57 N 52 14B60 hcfl 68 4A,80j70 hc/3 59 2.7,29C72 NV 86 12A,22CVV N 56 7B146 hcf4 102 7A,155 N 99 7Aj.j207 N 48 hA215 - 48 None239 - 169 None302 hcf5 92 7C,,20B310 hc.f7 95 ,9C313 hcf6 69322 N 99 4A,1OC343 - 116 None379 - 138 None386 - 93 None392 N 92 4A393 hcf8 76 1B,530 N 108 250550 - 181VNonea Designation used in the thesis hefi through hcf8; N= no present designation. A dash (-)signifies that no hcf lines have been identified originating from these bulk populations.b Number of individual plants from which the progeny were screened on 1J2 MSO petriplates.C Hcf lines have been propagated by single seed descent, originating from the M2 plantunderlined. V53found, may yield an M2 plant which segregates as a single nuclear recessive. However,another possibility is that the hcf phenotype observed dining screening was due to acytoplasmic mutation. Cytoplasmically inherited photosynthetic mutants would probablynot have survived in soil in the M2 generation. Over 160 M2 plants have been screened fromM1 bulks 239 and 550, for example, without yielding a line which segregates as a nuclearrecessive. Also, the number of M2 hcf plants observed from the original M1 bulk 239appeared to be higher than expected for a single nuclear recessive (14 versus the expected 4-7 of 500-750 total M2 plants screened). These results provide circumstantial evidence thatthe phenotype observed in M1 bulks 239 and 550 may have been due to a cytoplasmicmutation. However, genetic analysis was not done to verify that this was indeed the case.3.2. CHARACTERIZATION OF WILD-TYPE ARABIDOPSIS GROWN UNDER STERILECONDITIONSWhenever possible, each hcf line was compared with wild-type siblings. The resultsof the wild-type siblings were essentially the same as the Columbia wild-type parent for allcharacteristics analyzed. This section details initial experiments to determine the results forthe wild-type, and comparisons among wild-type siblings from the different mutant lines.3.2.1. Physical Characteristics and Pigment Composition of Wild-Type PlantsPhysical characteristics of wild-type plants on 112 MSO petri plates were dependenton the number of plants growing on the same petri plate, placement in the growth chamberand age of the plants at harvest. Attempts to minimize this by transferring an equal numberof plants to fresh petri plates, and harvesting of plant material at roughly the same age (at54four weeks) still resulted in large variations in plant size and total Chi per fresh weight(Table 7).Analysis of carotenoids and chlorophylls by HPLC revealed that the relative amountof the thylakoid pigments was similar to published results for Arabidopsis (Figure 6, Table 8)(Duckham et al 1991, Rock and Zeevart, 1991). No significant differences between thedifferent wild-type siblings in relative amounts of pigments were found, so the results werecombined (Table 8). All the pigments found in the wild-type siblings were found in themutants. This suggests that disruption of pigment biosynthesis was not the cause of the hcfphenotype in the mutants, but that the differences in pigment composition were the result ofloss of specific pigement complexes (see below). One pigment, antheraxanthin, was found atsignificantly higher levels relative to Chia in all the mutants compared to the wild-type(Table 8). Since antheraxanthin is an intermediate in the pathway from B-carotene toviolaxanthin and neoxanthin (Sandmann, 1991), both of which are present in the mutants,the significance of this finding is not presently understood. A second intermediate,zeaxanthin, was present only at trace levels in the wild-type and mutants. Zeaxanthin andlutein elute at roughly the same time on the HPLC column used. Since the plants were darkadapted prior to pigment isolation, the level of zeaxanthin was expected to be low (DemmigAdams et al., 1990). A few samples were run on a column that would resolve the twopigments, and zeaxanthin was found only at trace levels in the mutants (data not shown),thus no further attempts to resolve the two pigments were undertaken.3.2.2. Fluorescence and Electron TransportThe fluorescence characteristics of the wild-type plants in culture have beenpresented above, and the fluorescence kinetics observed in the plants grown under sterile55Table 7. Physical characteristics ofArabidopsis thaliana hcf mutants and wild-typesiblingsaLine Plant Fresh Weigh Chi (pg) per Chi a/bWeight (g)b Fresh Weight (mgY Ratiohcfl 18.4±9.5 0.217±0.012 2.68±0.12hcfl Wt 40.7±14.1 0.763±0.012 2.70±0.11hcf2 20.0±6.4 0.35±0.11 1.95±0.02hc/2 Wt 51.0±10.4 0.75±0.05 2.70±0.06hcf3 8.3 0.30 2.12hcf3 WT 45.3 0.70 2.71hcf4 21.3 0.22 2.34hcf4Wt 51.4 1.14 2.63hcf5 17.6±7.7 0.22±0.02 1.60±0.08hcf5Wt 54.3±4.1 1.04±0.07 2.67±0.17hcf6 18.3 0.10 1.59hcfli Wt 54.5 0.85 2.47hc/7 16.4 0.43 2.59hcf7Wt 108.2 1.44 2.78hcf8 20.0±6.5 0.445 2:52±0.09hcf8Wt 47.9±7.1 1.29 2.68±0.05a Each experiment consisted of harvesting 40-80 mutant and wild-type siblings from 3-5 112MSO petri plates at four weeks following germination. Values are means±SD, where threeor more experiments were done.b Fresh weight of aerial portion per plant.C Total Chi was calculated from the combined thylakoid membrane fraction and filter residuefraction (see Materials and Methods).56A Bo 10 20lhCflLuChIhCbV.a, j1.4 .i No0Nco•AI (I 21) III 2))Retention Time (mm)Figure 6. IIPLC chromatogram of pigments extracted from leaves ofArabidopsis wild-type (A) and mutant hcfl (B). Peaks: Neoxanthin (Neo),Violaxanthin (Vio), Lutein (Lu), Chlorophyll b (Chlb), Chlorophyll a (Chia)and 13-carotene (BCar). The peak between violaxanthin and lutein that isprominent in the mutant but only found at trace levels in the wild-type isantheraxanthin.57Table 8. HPLC analysis of the relative pigment composition of wild-type and hcf mutants ofArabibickipsis thaliana.Pigmenta wtb hcflC hcf2 hcf3 hcf4 hcf5 hcf6 hcf7 hct8Chlorophyll a 100 100 100 100 100 100 100 100 100Chlorophyll b 33.9±1.8 35.5 43.3 38.5 38.4 56.0 43.3 35.3 40.3Lutein 18.2±1.34 28.3 24.8 23.8 24.1 36.7 37.0 24.1 27.3Violaxanthin 4.1±0.6 5.8 5.8 6.7 4.9 9.5 14.3 5.6 6.7Neoxanthin 4.8±0.51 7.0 6.5 7.7 6.0 9.4 7.4 5.4 7.0Antheraxanthin 0.09±0.03 2.9 2.1 0.5 3.2 1.7 1.9 1.6 2.613-Carotene 10.1±1.19 6.4 4.4 6.5 6.0 3.4 4.4 5.9 5.5a Expressed as mol pigment/100 mol chlorophyll ab Average of 5-9 preparations of wild-type chioroplasts from the wild-type siblings of each hcfmutant, and 6-12 injections. No significant differences were found between the wild-typesiblings, so the results were combined.C No standard errors were calculated for the mutants since an insufficient number ofinjections were done. hcfl: Average of 3 extractions of plant material, each injected once;hcf2: 2 extraction, each injected once; hcf3: 2 extractions, 3 injections; hcf4: 2 extractions, 3injections; hcf5: 2 extractions, each injected twice; hcfi3: 2 extractions, each injected twice;hcfl: 1 extraction, injected 3 times; hcf: 2 extractions, 3 injections.58conditions on the 1/2 MSO petri plates were similar to wild-type plants grown in soil (Artusand Somerville 1988). As previously stated, the conditions on the petri plates did not appearto have any detrimental effect on electron transport as observed from the fluorescencekinetics.PSI and PSH electron activity rate in the wild-type siblings varied from experimentto experiment. This probably reflects the manner in which thylakoid membranepreparations were done, the age of the plants, conditions in the petri plates or time at whichthe plants were harvested, rather than real differences among the wild-type siblings. Thisconclusion is based on a strong negative correlation (-0.89) observed between the PSI andPSII electron transport activity rates in the wild-type siblings (Table 9) suggesting thatplants were subjected to some condition favoring either PSII or PSI electron transport priorto isolation of the thylakoid membranes. Attempts were made to minimize this by onlyharvesting plants early in the morning following a 1-2 dark adaptation period, however; thiswas not always possible. For this reason, comparison of the hcf mutant lines was only donein relation to the wild-type siblings harvested from the same petri plates, at the same time.3.2.3. Chlorophyll-Protein Complexes, and Thylakoid Membrane PolypeptidesThylakoid membrane samples from wild-type Arabidopsis plants grown on 1/2 MSOpetri plates and electrophoresed under non-denaturing conditions have the normalcomplement of ChI-protein complexes observed from a number of species (Green 1988) aswell as Arabidopsis plants grown in soil (Hugly et al., 1989). There were no differences thatcould be observed between any of the wild-type siblings from the different mutant lines.The presence or absence of specific polypeptides was detected using monospecificpolyclonal antibodies. Since none of the antibodies were raised from proteins isolated59Table 9. Electron transport activity rates for hcf mutants and wild-type siblings. Values aremeans ± SD; n=3-6.Line p5ja P511bj.tmol 02 reduced pmol DCPIP reduced.(mg Chl)1hr4 (mg ChIY1hrH20 DPChcfl 280±39 99±8 119±9hcflWt 1121±100 115±10 123±13hcf2 577±60 NDC 36±5hcf2Wt 1320±175 ND 115±11hcfl3 768±37 ND 65±13hcfl) Wt 954±83 NI) 164±21hcf4 1231±48 113±19 112±18hcf4Wt 1270±82 136±11 141±141vf5 759±51 <5 <5hc/5Wt 1043±80 147±7 145±7hcf6 1044±38 ND <5hef6Wt 822±78 ND 172±5hc/7 827±22 ND 116±10hcj7Wt 1184±37 ND 115±11hc18 607±30 ND 111±5hctWt 1493±159 ND 107±4a Measured as 02 uptake in the presence of 0.1% NaN3,2.5 mM ascorbate, 0.5mMtetramethyl-p-phenylenediamine, 0.1 mM methyl viologen, 1 j.iM DCMU, pH 7.8.b Measured as light-driven reduction of DCPJP at 595 nm, pH 6.0. Electron donors werewater and diphenylcarbazide (see Materials and Methods for experimental protocol).C ND = no data60from Arcthidopsis, detection of the corresponding polypeptide in Arabidopsis confirmedimmunological relatedness. Only in the case of PSI subunit III and IV did antibodies raisedto spinach proteins fail to react to a homologous polypeptide in the Arabidopsis thylakoidmembrane preparations (Figure 7). Subunit ifi is the psaF gene product and has beenshown to become easily disassociated from the PSI core complex (Bengins and Nelson, 1977).The antibody weakly recognized several bands corresponding to CPI core polypeptides inArabidopsis. Thus either the subunit III polypeptide did not dissassociate completely fromthe core complex under the conditions used, there is some immunological relatednessbetween the A.rabidopsis core polypeptides and the spinach subunit HI, or the reactionobserved is simply non-specific. The subunit 111 antisera prepared from spinach cross-reactsto a homologous protein in oats, bean, (Nechushtai and Nelson, 1985) and Synechocystis(Wynn et al., 1989), it suggests that the bulk of the subunit ifi polypeptide was lost duringisolation. It is interesting to note however, that the thylakoid membrane fraction isolationprocedure used, was essentially the same for both spinach and Arabidopsis. Thus while forthe spinach a significant amount of subunit Ill was associated with the PSI fraction, it waslost in Arabidopsis. For PSI subunit IV, there was no cross reaction observed at all.The function of subunit IV, the psaE gene product, is not presently known. There isa high degree of homology observed among different species, leading Golbeck and Bryant(1991) to speculate that the protein is either ftinctionally or structurally constrained. In viewof this, the lack of observed immunological relatedness between Arabidopsis and spinach isintriguing.The antibody to the spinach subunit VI polypeptide recognized three polypeptides inthe Arabidopsis thylakoid membrane preparations (Figure 7), and only the larger two whena PSI-200 preparation was isolated by the procedures of Mullet et al. (1980) (data not61AS ASSub. IIISub. VISub. WFigure 7. Comparison of Arabidopsis (A) and spinach (S) photosystem IpoLypeptides. Thylakoid membranes were separated by denaturing SDSPAGE and were immunoblotted with antibodies prepared against spinachphotosystem I subunits Ill (Sub. III), IV (Sub. IV) and VI (Sub. VI).62shown). Only two polypeptides are also immunostained when spinach thylakoid membranesare probed with the antibody (Figure 7). In this case, the antibody may be recognizing bothsubunit VI (psaH gene product) and subunit VII (psaC gene product) or another PSIpolypeptide which may have been co-purified. The identity of the lower molecular weightpolypeptide observed in the Arabidopsis thylakoid membrane preparations which reacts toanti-PSI subunit VI is unknown. Since it is not observed when PSI preparations wereimmunoblotted, it is either not a PSI polypeptide or it is lost during PSI isolation. As thislower band is at normal levels in the mutants that are reduced in the PSI complex (see hcfland hcf8 below) the former possibility is more likely.3.2.4. Electron MicroscopyChloroplast ultrastructure was compared among the mutants hcfl, hcf4, hcf5, hcf6and hcf8 and the wild-type siblings. No Columbia wild-type plants were analysed, however,there were no ultrastructuraldifferences observed between the wild-type siblings from thedifferent mutant lines. A representative example of the wild-type and mutants hcf4, hct5and hcf6 is presented in Figure 8. There were no obvious ultrastructural differencesobserved between the wild-type and hcfl and hcf8 at the stage at which plants were analysed(early seedling; not shown).3.2.5. RNA Levels in the Wild-TypeThere were no differences in the bands observed between the different wild-typesiblings on northern blots. The banding pattern profiles from the different chioroplasttranscriptional units observed on northern blots, are also similar to those published forspinach (see below).63Figure 8. Thin section electron microscopy of hcf mutants and wild-typechioroplasts. Arabidopsis wild-type (A), mutant hcf4 (B), hcf5 (C) and hcf6(D).64S9(.)3.3. CHARACTERIZATION OF EIGHT HCF MUTANT LINESWithout exception, all the mutant lines when first isolated contained additionalmutant traits segregating with the hcf phenotype. The most common co-segregating mutanttrait was embryo lethality. Each line was self-pollinated at least one additional generationprior to commencing the experiments described. All measurements were done on mutantand wild-type siblings grown on the same petri plates. This ensured that all comparisons ofthe hcf phenotype were performed in a common genetic background.In order to harvest sufficient plant material for analysis, seeds from heterozygousplants were sown under sterile conditions on agar plates containing 5% sucrose andnutrients (see materials and methods). Following identification of the hcf and wild-typeplants under UV light, small numbers of hcf (typically 15-20) and wild-type (typically 10-15)plants were transferred to fresh agar plates to allow for leaf expansion. Each experimentusually involved harvesting plants from 3-5 petri plates, and combining similar phenotypes.Thus, the results of each experiment are based on an average of 45-75 plants.The results on the individual mutants given below are based on fluorescence kinetics,photosynthetic electron transport activity rates, presence and amount of Chi-proteincomplexes and thylakoid membrane polypeptides. The mutants are grouped according tophenotype. In the first class (represented by hcf5 and hcf6) electron transport is blocked atthe PSII complex, although, as will be shown, all the complexes in the thylakoid membraneare also reduced in these mutants. In the second class PSII electron transport is impaired,yet some activity is observed (hct2 and hcf3). In the third class PSII electron transport isnormal, but the PSI complex is reduced (hefi and hcf8). The last two mutant lines presented66(hcf4 and hcf7) probably also have a damaged PSI complex, however, this conclusion cannotbe unequivically substantiated by the results obtained to date.3.3.1.hcf53.3.1.1. Mutant Isolation and Genetic AnalysisMutant line hcf5 was derived from M1 bulk population 302 and was found tosegregate for the hcf phenotype in the M3 generation. The mutant line has been self-pollinated 5 generations and backcrossed to the wild-type Columbia parent. The phenotypeis inherited through the pollen. Homozygous wild-type and heterozygous plants areindistinguishable in morphological characteristics (except for the embryo-lethal phenotypedescribed below) and the progeny must be screened each generation in order to identifyheterozygous plants.Heterozygous plants segregated for the hcf phenotype at a frequency of 13%-18%(jt=14.9; 194 hcf: 1111 wild-type). This is below the expected 25% level for a single nucleargene, perhaps because of selective loss of the hcf allele in the gametophytic stage. Thisphenomenon, termed certation, has been previously observed in mutant lines where thephotosynthetic apparatus is affected (Deleart, 1980; Simpson and von Wettstein, 1980). Asecond explanation could be a linkage to an embryo-lethal allele. The hcf5 mutant line wasscreened for the presence of embryo-lethal phenotypes in response to anomolous segregationratios observed in F2 populations containing a glabrous marker on chromosome 3 (gl-1; Table5) in routine mapping of the hcf locus (see Materials and Methods). Of 695 F2 plantsscreened, 268 were glabrous and 427 were wild-type. The frequency of glabruos plants (39%)is significantly higher than the expected 2 wild-type: 1 glabrous ratio, assuming a single67linked lethal locus in trans (X2=8.54; P<O.01). In F2 populations containing visible markerson chromosome 1 (an, dis-l), chromosome 2 (er) and chromosome 4 (cer-2, ap-2) thesegregation frequencies for the visible markers were not found to be significantly differentfrom the expected 25%. F2 populations with visible markers on chromoéome 5 have not beenanalysed. Screening revealed that an embryo-lethal mutation was present in the hcf5 plantswhich were used in crosses. In fact, two distinct embryo-lethals have been found cosegregating with hcf5. One is a late globular to torpedo, and in the second, the embryodevelops outside the seed coat (Meinke, 1991). The second embryo-lethal phenotype has beenmapped to chromosome 1 (DW Meinke, personal communication). The fact that nosegregation differences were observed using either marker on this chromosome, suggeststhat the markers analysed (an and dis-1) are greater than 50 map units from the embryo-lethal locus, or it calls into question the reliability of the method employed. Similarproblems using this method have also been encountered in the mapping of embryonic lethals(Patton et aL, 1991). The results suggest that the hcf5 locus could be on chromosome 3;however, due to the presence of the embryo-lethal mutations, further analysis will berequired. Physical Characteristics and Electron MicroscopyThe hcf seedlings can be differentiated from the wild-type siblings by their lightercolor even without screening for the high fluorescent phenotype. This is due to a loweramount of Chi per plant in the mutants (Table 7). After 4 weeks on 1/2 MSO petri plates, themutant plants were significantly smaller than the wild-type siblings. The plants bolt andflower following an extended time on 112 MSO plates (> 6 weeks), but no seed set has been68observed. It is not known if this is due to the high humidity in the 1/2 MSO magenta boxes,or if this is part of the hcf phenotype.Striking differences in chioroplast ultrastructure were observed between the hcfmutant line hcf5 and wild-type (Figure 7). The thylakoid membranes were almostexclusively stacked grana, and appeared devoid of stroma lamellae. Since the PSII complexis severely depleted (see below), the stacking should be due solely to interactions between theChla+b light harvesting complex proteins (Anderson, 1982; Thomber, 1990). Pigment CompositionAll the pigments that are normally present in the wild-type siblings are also presentin the mutants, albeit reduced and in different proportions (Table 8). The Chia lb ratio wassignificantly altered, resulting in a reduction of 82% and 70% for Chla and Chlb, respectively(Table 7 and 8). The least affected were the xanthophylls; neoxanthin, violaxanthin, andlutein which were decreased 34%, 41% and 45%, respectively, on a fresh weight basis. 13-carotene was the pigment most severely affected in the mutants and only 6% of wild-typelevel was observed. 13-Carotene has been primarily associated with the reaction centres ofPSI and PSII, whereas the xanthophylls are primarily associated with peripheral lightharvesting complexes (Siefermann-Harms, 1985). Fluorescence and Electron TransportThe room temperature fluorescence induction curve of the mutant line hcf5 in latergenerations was similar to the fluorescence curves found in the original M1 bulk 302 (Figure5). The wild-type siblings on 1/2 MSO plates displayed normal fluorescence kinetics asdescribed above. The hcf mutants, on the other hand, have a high initial fluorescence (F0),69and essentially no variable fluorescence (Fv) (Figure 5). These results suggest thatreduction of QA is not occuring and thus PSII is damaged or absent. Similar fluorescencekinetics have been observed in PSII mutants of Chiamydomonas and maize (Chua andBennoun, 1975; Leto and Miles, 1980; Miles, 1980).In order to estimate the degree of damage to PSII, and to determine if PSI was alsoaffected, electron transport activity rates of isolated thylakoids were assayed. As shown inTable 9, PSII electron transport activity was undetectable above the background level,confirming the fluorescence data which suggests that electron flow through PSH is impaired.PSI electron transport rates were also somewhat reduced at 73% of the wild-type levels. Theresults suggest that both P511 and PSI are damaged in the mutant, although the highfluorescence phenotype is due to a disruption in electron transport in PSII. Chlorophyll-Protein Complexes and Thylakoid Membrane ProteinsOn non-denaturing (green) gels, the CP47 and CP43 bands, which are the core ChIabinding proteins of PSU, are not present (Figure 9). The CPI band associated with PSI arealso severely depleted. The Chla+b complexes, CPII*, CPu, and CP29, appear to be normal.CPI antenna and PSII (CP43 and CP47) Chla binding proteins are chioroplast encoded,whereas the Chla+b binding proteins are the nuclear encoded products of the cab genefamily.These results confirm the pigment and electron transport data, which suggested thatPSI and PSII are severely affected. The low ChlaIb ratio suggests that the majority of theChI present in the mutants is associated with the peripheral light harvesting complexes,particularly LHCII which has a Chla/b ratio of 1.2.70hcf5hcf wtCPICPU*CP47CP43CP29CPUFigure 9. Chlorophyll-protein complexes of hcf5 and wild-type siblings.Unstained thylakoid membrane complexes run on a non-denaturing 10%SDS-PAGE. CM-protein complexes: CPI: Photosystem I Chi a binding corecomplex; CP47 and CP43: Photosystem II 47 and 43 kDa Chl a binding corecomplex polypeptides; CPIP and CPu: the oligomeric and monomeric forms,respectively, of the Chi a+b light harvesting complex polypeptides.FP71Thylakoid polypeptides were completely denatured, separated by gel electrophoresisand successively immunoblotted with several antibodies to determine the presence orabsence of specific polypeptides. The PSII Chla binding polypeptide of CP47 was almostundetectable and the PSII core polypeptide cytochome b559 was not detected (Figure 10).The levels of PSI polypeptides were also severely reduced (Figure 10). Thepolypeptides of CPI, the PSI reaction center, were almost undetectable. There was a lowamount of the PSI subunit II polypeptide, encoded by the nuclear psaD gene. All of thepolypeptides which react to the spinach PSI subunit VI antiserum in Arabidopsis were alsoreduced. The levels of PSI polypeptides present in the mutant appear to be less than wouldbe expected based on the electron transport activity rates. The reason for this is unknown(but see Discussion).Immunoblotting against other thylakoid membrane proteins revealed that both thecytochrome complex and coupling factor are also affected in the mutant. Cytochrome f levelsappear to be reduced to less than 1% of the wild-type. Some of the observed reaction appearsto be due to a polypeptide which ran slightly lower on the gel in the mutant lane. Theidentity of this polypeptide is unknown. The CF1 antisera used (Moase and Green, 1981)reacts primarily against the alpha and beta subunits in the wild-type, although the smallergamma and delta subunits could be resolved following extended staining (data not shown).In the mutant, only one band was recognized by the antiserum. It appeared to be the upperband which is normally the alpha subunit in higher plants (Westhoff et al., 1985), but in viewof the data that suggests that the alpha transcript is affected in the mutant (see below) itmust be pointed out that the relative migration of the different subunits has not beendetermined on the gel system used.72hcf5W’T hcf hcf VTCF1 CPICP47f CytfCpu iJSub. IIJ Sub.VICytb559Figure 10. Immunoblot of hcf5 and wild-type siblings. Denaturing gels wereloaded with equal amounts of Chi from the mutant and wild-type. The leftnitrocellulose filter was successvely immunoblotted with antibodies tocoupling factor (CF1), PSII 47 kfla ChI a binding protein (CP47), lightharveting complex (CPu) and 9 kfla polypeptide of photosytem II cytochromeb559 (Cyt b559). The right nitrocellulose filter was successivelyimmunoblotted with antibodies to PSI 80-82 kfla core complex (CPu),cytochrome flCyt t), PSI subunit II (Sub. II) and PSI subunit VI (Sub. VI).733.3.1.6. Steady State RIsIA LevelsTo determine if any of the observed differences in the mutant hcf5 line could be dueto transcriptional abnormalities, total leaf RNA was isolated and probed with specificnuclear and chloroplast DNA sequences. RNA gel blots loaded with approximately equalamounts of total RNA were probed with the psbB gene encoding the 51 kDa ChIa apoproteinof CP47 and the petB gene encoding the cytochrome b6 apoprotein (Figure 11). Both of thesegenes are part of the psbB-psbH-petB-petD transcriptional unit, which also encodes the 10kDa phosphoprotein of P811 (psbH) and subunit 4 of the cytochrome complex (petD)(Westhoff et al., 1986; Heinemeyer et a]., 1984). This cistron has been the subject ofconsiderable interest as it displays a complex pattern due to post-transcriptional processingwhich is conserved over a wide range of plant species (Barkan, 1988; Gruissem, 1989;Westhoff and Herrmann, 1988). The level of the primary transcript, as well as the 4.7 and3.7kb transcripts derived from the petB and petD intron splicing were reduced in the mutantas detected by both the psbB and petB probes. Some of the smaller processed products werealso at lower levels, specifically the 2.2 and 1.4 kb transcripts using the petB as a probe. Themutant had normal levels of the 2.4, 1.9 and 1.1kb transcripts, all which include the psbHcoding sequence, whereas the 2.2 and 1.4 kb transcripts do not. Since it is believed that thesmaller transcripts are derived solely from the processing of the primary transcript, and noalternate promoters have been detected, transcription per se does not appear to be impairedin the mutant.The second PSII chioroplast transcript analysed was psbD, encoding the PSII core D2protein. In other higher plants, psbD is part of a dicistronic transcript which includes thepsbC gene encoding the 43 kD Chla-binding apoprotein (Alt et al., 1984). In the mutant, the74Figure 11. RNA gel blot of hcf5 and wild-type siblings. Gels were loadedwith equal amounts of total RNA. DNA probes used (labeled above each blot)are described in the Material and Methods. Molecular weights of theprominent bands were estimated using RNA standards (BoehringerMannhejm)75rbcLhcfWTpsbBhcfWTkbpsaBhcfVTkbpetBhcfVTpsbDhcfWTa.)atpB/EhcfVTcabllhcfWTkb1.81.6kb1.0level of the primary transcript appears at the wild-type level, however, a 3.1 kb transcriptappears to be reduced (Figure 11).The atpA gene is part of an Atpase gene cluster also encoding atpF, atpH and atpl(Henning and Herrmann, 1986). On RNA gel blots, overlapping transcripts accumulatesimilar to those of the psbB region. The overall steady state levels were reduced in themutant and a 3.3 kb transcript observed in the wild-type is lacking. The level of the largerexpected transcript (4.5 kb) (Westhoff and Herrmann, 1985) was low in the wild-type, thuscomparison with the mutant was not possible.The rbcL gene encoding the large subunit of ribulose- 1,5-bisphosphatecarboxylase/oxygenase was also used as a probe, although the protein product was notassayed. No differences at the transcriptional level were expected since the steady statelevels of the rbcL gene have been shown to remain constant over a wide range ofdevelopmental and environmental conditions (Deng and Gruissem, 1988; Piechulla et al.,1986). Surprisingly, however, less than 5% of wild-type levels were observed in the mutant,which suggests that there is a problem in transcription or in stability.Not all chloroplast transcripts were affected. Steady state levels of the two atpBtranscripts that encode the coupling factor beta subunit were normal and of the expectedsizes. In higher plants the atpB gene is part of a discistronic unit also encoding the atpEgene (Henning and Herrmann, 1986). Immunoblots revealed that, in the mutant, apolypeptide with relative mobility corresponding to the CF1 alpha subunit was present butno band corresponding to the beta subunit was present. These results suggest that thepolypeptide band observed is probably not the alpha subunit, thus the identity of the band isunknown. No significant differences were found with the psaB transcripts (PSI reaction77center polypeptide B), psaC (PSI iron-sulfur polypeptide) the atpB /E transcript describedabove (Figure 11 and data not shown).Nuclear genes cab3C (not shown) and cabil (Figure 11) encoding the peripheral lightharvesting complexes of PSI! and PSI, respectively, appear at wild-type levels. In spite ofthe decreased accumulation of the nuclear encoded psaD PSI subunit II polypeptide, themRNA was at wild-type levels. Gels were loaded as closely as possible with equal amounts oftotal RNA, and no differences were observed in ribosomal RNA levels when the gels werestained with ethidium bromide. Each of the filters was stripped and hybridized with achloroplast rDNA probe to verify that the differences observed were not due to a generaldecrease in chloroplast RNA.3.3.2. hcf63.3.2.1 Mutant Isolation and Genetic AnalysisMutant line hcf6 was originally isolated from M1 bulk 313. The original mutant linesegregated at a frequency higher than would be expected for a single nuclear recessive(j.t=32.5; 96 hcf: 196 wild-type). This result suggested that the mutant phenotype wasinherited maternally and chioroplast sorting was still occuring in the line, or that more thanone nuclear hcf mutation was responsible for the hcf phenotype, or that an embryo lethallinked in trans was also present in the line. Crosses to marker lines in the Landsburgbackground confirmed that the hcf phenotype is inherited through the pollen, it is thereforeof nuclear origin. Visual screening for embryo-lethal phenotypes confirmed the presence ofat least two distinct embryo-lethal phenotypes segregating in the line. Recently an M6 planthas been isolated that segregates for the hcf phenotype at a ratio suggesting a single nuclear78recessive locus (11=21; 38 hcf: 144 wild-type). These combined results suggest that theoriginal segregation ratios were due to the presence of co-segregating embryo lethalphenotypes and that the hcf6 mutant phenotype is due to a single nuclear recessivemutation. Pigment Composition and Electron MicroscopyThe hcf seedlings can be differentiated from the wild-type siblings by their yellowcolor even without screening for the hcf phenotype under UV light. Mutant plants aresmaller than the wild-type siblings and the yellow phenotype is due to the very low level ofChl compared to the wild-type (Tables 7 and 8). When normalized to Chia, all the pigmentspresent in the wild-type were present in the mutant siblings and were found at increasedlevels, with the exception of 13-carotene which was lower (Table 8). When compared to Chia,the ratios observed are similar to other Arabidapsis hcf mutants analysed, with theexception of violaxanthin. Violaxanthin was found to be twice the level normallyencountered when compared to the wild-type, or to the other mutants analysed. It isunknown if the high value of vio1axanthin is significant or simply due to randon deviations inthe levels of pigments in the mutant. Violaxanthin is a carotenoid in the xanthophyll cycleand undergoes de-epoxidation and is converted to zeaxanthin in high light. The latter hasbeen found to be associated with dissipation of excess excitation energy in PSII and ispostulated to be involved in photoprotection (Demmig-Adams et al., 1989). Since the plantswere dark-adapted prior to isolation of material for pigment analysis, zeaxanthin was onlyfound in trace amounts in the mutant, and was not quantified.79Striking differences were observed in chloroplast ultrastructure between hcf6 plantsand wild-type siblings (Figure 8). Chloroplasts appear to be mostly small and immature andcontain very few grana and stromal lamellae. Fluorescence and Electron TransportThe hcf phenotype is due to a disrupted PSII as can be seen by the high F0 and lowF (Figure 12A). However, there is a small amount of plastoquinone reduced based onfluorescence kinetics. There is a slow and well pronounced 0-I-fl-P kinetics using the lightlevel tested (Figure 12B). The I-I) kinetics is thought to be due to re-oxidation by PSI(Hansen et al., 1991). This is at variance with the results which suggest that the PSIcomplex is at a lower levelin the mutant. On the other hand, PSI electron transport activityrates were greater than the wild-type when normalized to Chi (Table 9). The hcf6 PSIresults were based on only two replications due to the small amount of mutant thylakoidmembrane preparation, but the two experiments were similar. There is no obviousexplanation for the contradiction between protein and PSI assays. It may be that due to theimmature state of the chloroplast, photooxidation of some compound is occuring in themutant, which results in the high PSI electron transport activity rates. Higher thanexpected PSI electron transport activity rates have been observed in cases where thechloroplast were immature and chlorophyll levels low (Bar-Nun et al., 1977; flaniell et al.,1985; Ohashi et al., 1989).PSII electron transport activity in the mutant was difficult to ascertain due to lightscattering. A large amount of thylakoid membrane preparation from the mutant wasrequired in the assay in order to reach the same Chi concentration as the wild-type.However, no activity was observed above the background noise. Electron transport data80::-heftS700600500Q 200 wiC.)10-0 i I I I I I I I I I I I I0 4 8 12 16 20 24 2806961 0.59.57 .0 2 4 6Time (s)Figure 12. Fluorescence induction kinetics of hcf6 and wild-type siblings.(A) Fluorescence kinetics from a single mutant and wild-type plant takenusing the CAFVIS. (B) A representative example of the early induction ofthe mutant displaying a prominent dip during induction obtained using thePAM fluorometer.81were complicated by the fact that the wild-type PSII was somewhat higher than normallyobserved, while the PSI is somewhat lower than normal. Both measurements were done onthe same day from the same plant material. As pointed out above, it may be that the plantswere subjected to conditions that favored PSII activity prior to harvesting. Chlorophyll-Protein ComplexesA non-denaturing (green) gel shows that there are no Chla binding proteins CPI,CP43 and CP47 (Figure 13). The Chla+b binding proteins appear at nomal levels. Thisresult suggests that both PSII and PSI complexes are diminished in the mutant.Thylakoid polypeptides were completely denatured, separated by gel electrophoresisand successively immunoblotted with several antibodies to determine the presence orabsence of specific polypeptides. The PSI Chia binding apoprotein CPI was almostundetectable and the PSII core polypeptide cytochome b559 was not detected (Figure 14).The cytochrome f polypeptide was also found at a very low level in the mutant. The resultssuggest that all the thylakoid membrane complexes are probably at lower levels in themutant. RNA Gel BlotsAbnormal steady state levels of some chioroplast transcripts are similar to thatobserved for hcf5 (Figure 15). In the case of the rbcL transcript, it is even at lower levels.When probed with chioroplast ribosomal RNA, the overall steady state level of transcriptswere reduced in the mutant compared to the wild-type. This suggests that some of theprimary chioroplast RNA transcripts could be due in part to a lower transcription rate. Alower transcription rate would allow for faster turnover of unprocessed primary transcripts82hcf6hcf WtCPIa- cPIcPII*CP47CP43___CpuFPFigure 13. Chlorophyll-protein complexes of hcf6 and wild-type siblings.Unstained thylakoid membrane complexes run on a non-denaturing 10%SDS-PAGE. Chi-protein complexes: CPI: Photosystem I ChI a binding corecomplex; CP47 and CP43: Photosystem II 47 and 43 kDa ChI a binding corecomplex polypeptides; CPII and CPU: the oligomeric and monomeric forms,respectively, of the Chi a-i-b light harvesting complex polypeptides.83cpICytfSub. IICyt b559 r—• Vhcf6Wt hcf1’,•—..Figure 14. Immunoblot of hcf6 and wild-type siblings. Denaturing gels wereloaded with equal amounts of Chi from the mutant and wild-type. Thenitrocellulose filter was successvely immunoblotted with antibodies to to PSI80-82 kDa core complex (CPI), cytochrome fiCyt f), PSI subunit II (Sub. II)and the 9 kfla polypeptide of photosytem II cytochrome b559 (Cyt b559).84Figure 15. RN, gel blot of hcf6 and wild-type siblings. Gels were loaded withequal amounts of total RNA. DNA probes used (labeled above each blot) aredescribed in the Material and Methods. Molecular weight of the prominentbands were estimated using RNA standards (Boehringer Mannheim).85kb5. Wt2. WtatpAhcf •Wtkb4. WtrbcLhcf Wtkb4.44.123S rRNAhcf WtpsaDhcf Wt kb86in favor of processed transcripts observed in the steady state. However, such a conclusionwould require in vitro transcription assays to determine if this is the case.In some of the gels, the hcf6 lane was significantly overloaded, thereby makingcomparisons with the wild-type difficult. In one of the gels where fairly equal loading wasachieved, the rbcL transcript was assayed, and was found to be lower than for hcf5. ThepetB and psbB transcripts also appear to lack the large un-processed transcripts.Unfortunately, these gel were overloaded.An interesting anomaly was noted when the filters were probed with the psaB gene.There appear to be two high molecular weight transcripts present in the mutant (ie. largerthan the ‘primary’ transcript). It is not possible to rule out that this is DNA contamination.However, the bands did not co-migrate with other DNA bands observed in some of the otherwild-type and mutant lanes. In addition these bands were not observed in the hcf6 lanewhen probed with other chloroplast genes. If the bands observed are in fact RNA, it suggeststhat there is a readthrough into the psaB gene from another 5’ chloroplast gene, and/or the 3’end is not being processed properly.3.3.3. hcf23.3.3.1. Mutant Isolation and Genetic AnalysisMutant hcf2 was originaly isolated from M1 bulk 20. The line was self-pollinated for5 generations and crossed to marker lines in the Landsburg background. Most of theseedlings derived from self-pollinated plants segregating for the hcf phenotype are lightgreen to yellow in soil and die shortly after gennination; no cotyledon expansion is observed.In a few instances however, some seedlings do survive in soil to form small rosettes.87Following a prolonged period (8 weeks) these small plants begin to flower, but no seed sethas ever been observed. It has not been possible to confirm if this phenotype is truly hcf or isdue to a second mutation. If the phenotype is due to the hcf mutation, it suggests that themutation is ‘leaky’ in some instances. The hcf phenotype is inherited through the pollen andis therefore of nuclear origin. Only crosses to the marker line RIJM4 (chromosome 4) havebeen evaluated and no linkage was found. Physical Characteristics and Pigment AnalysisOn 1/2 MSO plates hcf2 plants vary considerably in size. Overall the hcf plants wereapproximately half the fresh weight of the wild-type siblings (Table 7). All the pigmentsnormally found in the wild-type were present in the mutant, although, in differentproportions (Table 8), as with other hcf mutant lines. On a fresh weight basis Chia and Chibwere reduced to 43% and 55% of the wild-type levels respectively. Since ChIa was morereduced than ChIb, it resulted in a decrease in the ChlaIb ratio. The major xanthophyllsdecreased proportionally. Relative to ChIa, neoxanthin, violaxanthin and lutein, decreased59%, 61% and 59%, respectively. The pigment most severely affected was 13-carotene whichwas only at 19% of the wild-type levels. Fluorescence and Electron TransportRoom temperature fluorescence induction kinetics from dark adapted plants (Figure16) suggest that while electron transport is occuring in the mutant, the PSII complex isdisturbed. This conclusion is based on the increase in the F0 level, and the observation ofsome variable fluorescence. The induction of those PSII centers that are present, appears tobe normal, as the induction kinetics is normal. However, at least 50% of the variable8880•0hcf250200 20 40 60 807068666462600 20 40 60 80(s)Figure 16. Fluorescence induction kinetics of hcf2 and wild-type siblings. (A)Fluorescence kinetics average from 4-7 mutant and wild-type plant takenusing the PAM fluorometer. (B) Same as panel (A) except that the y axis wasexpanded to better show the kinetics of the mutant.89fluorescence observed is below F0 as the steady state fluorescence (F5)is below the initialfluorescence (Figure 16B).Using the PAM 101 pulse modulated fluorometer (Schreiber et al., 1986), the relativecontribution of qP and qNP were calculated (van Kooten and Snel 1990). After 3112 minutesat low actimc light levels the contribution of qP was less than half of the value observed forthe wild-type (0.22 versus 0.46). However, the contribution of non-photochemical quenchingwas essentially the same for the mutant and wild-type. Although the relative contributionsof qE, qT and qI in qNP are still under debate, it appears that the factors responsible fornon-photochemical quenching are normal in the mutant, in spite of the apparent abnormalPSH kinetics.Electron transport activity measurements support the conclusion that PSII isreduced, yet functional in the mutant, as 31% of wild-type rate was observed (Table 9). ThePSI activity rate is also reduced to 50% of the wild-type level in the mutant. Chlorophyll-Protein ComplexesThylakoid membranes proteins separated on non-denaturing (green) gels confirmthat both PSU and PSI are reduced (Figure 17). The PSI band, CPI, is reduced and the PSIJChl-prot,ein complexes CP47 and CP43 were not present. Since some PSH electron transportactivity is observed, the reason why no PSII are seen bands is unknown. It may be that thelow Chi levels are not detectable, or the PSII complex is unstable and falls apart duringelectrophoresis. The Chla+b complexes, CPu and CPII*, appear at normal levels.90hcfhcf2wtcpi—*w-rCPII*CP47CP43CP29CpuFPFigure 17. Chlorophyll-protein complexes of hcf2 and wild-type siblings.Unstained thylakoid membrane complexes run on a non-denaturing 10%SDS-PAGE. Chi-protein complexes: CPI: Photosystem I Chl a binding corecomplex; CP47 and CP43: Photosystem II 47 and 43 kDa Chi a binding corecomplex polypeptides; CPII* and CPu: the oligomeric and monomeric forms,respectively, of the Chi a+b light harvesting complex polypeptides.913.3.3.5. Thylakoid Membrane PolypeptidesNon-denaturing gel electrophoresis was performed and the gel was stained withCoomassie Brilliant Blue (Figure 18). The higher level of protein observed in the mutantlane is due to loading based on equal levels of Chi. The polypeptide profile from the mutantand wild-type is identical with two exceptions. Polypeptides of approximately 34 and 15 kDaappear to be missing in the mutant lane. The identity of these polypeptides is not known,but one likely candidate for the 34 kDa polypeptide is cytochrome f. In order to determinethe presence of cytochrome fin the membrane two approaches were undertaken. First aheme specific stain, TMBZ H20(see Materials and Methods) was done on a non-denaturedgel (Figure 18B). In the mutant lane it can be observed that no band corresponding to thecytochrome f in the wild-type is found in the mutant lane. However, a slightly largermolecular weight diffuse band is stained for heme and suggests that the cytochrome fproteinis not migrating normally in the gel and is at a reduced level in the mutant. The stainedhigher molecular weight bands (43-47 kDa) appeared prior to the addition ofH20and thusare probably due to some non-specific staining by TMBZ. The second approach used was todetermine the presence of specific polypeptides using polyclonal polypeptide specificantibodies on denatured gels transferred to nitrocellulose. When the antibody raised againstcytochrome fwas used, a band corresponding to the correct molecular weight is observed,although the amount of stain is at lower levels (Figure 19). The reasons for the discrepancyobserved in the results above are not known. Attempts to do the heme stain on denaturedgels have been unsuccessful, for unknown reasons. Thus while the results suggest that themissing band in the denatured gels may not be the cytochrome f polypeptide, as the antibody92MWhcfWt46.0 -30.0-14.3-ABhcf wtCpI*—CpuFigure 18. Thylakoid membrane polypeptide profile (A) and heme-specificstaining of non-denaturing gel (B) of hcf2 and wild-type siblings. (A)Denaturing gels were loaded based on an equal amount of ChI. Arrow pointsout to major band missing in the mutant. (B) TMBZ H20 heme-specificstain band is marked by the arrow. The higher molecular weight bands(asterisk) were visible prior to the addition of H20. Green Chi proteincomplexes: CPI: Photosystem I Chi a binding core complex; and CPu: themonomeric form of the Chi a+b light harvesting complex polypeptides.93hcf Wthcf2hcf Wt‘p*Cytf*Sub. ii•—Cytf— Cytb6Cytb559Figure 19. Immunoblot of hcf2 and wild-type siblings. Denaturing gels wereloaded on equal amounts of chlorophyll from the mutant and wild-type. Theleft nitrocellulose filter was successvely immunobloted with antibodies to PSI80-82 kDa core complex (CPI), eytochrome f (Cyt t) and PSI subunit II (Sub.II). Some non-specific staining of LHCII was observed when the PSI subunitII antibody was used (asterisk). The right nitrocellulose filter wassuccessively immunoblotted with antibodies to coupling factor (CF1),cytochrome f (Cyt /), cytochrome b6 (Cyt b6) and 9 kfla polypeptide ofphotosytem II cytochromeb559 (Cytb559).94recognized a similar migrating polypeptide, there appears to be some abnormality associatedwith the native cytochrome fprotein.Immunoblotting with cytochrome b6 antibody revealed that the polypeptide waspresent, although at lower levels. This suggests that the entire cytochrome complex isreduced in the mutant. Immunoblotting with the antiserum raised against the cytochromeb559 associated polypeptide of PSI! shows the 9 kDa polypeptide is at lower levels butpresent in the thylakoid membranes of the mutant.Immunoblotting with PSI specific polypeptides shows that subunit II, V and VI arereduced (Figure 19 and data not shown). There does not appear to be any difference in thelevel of CPI in the mutant. It is not clear why this should be the case. Mutants that have areduced PSI activity, normally display a pleiotropic effect whereby all the polypeptides of thecomplex are similarly reduced. This is the case for PSI mutants of barley (Hiller et al., 1980,Simpson and von Wettstein, 1980), Chiamydomonas (Bennoun and Jupin, 1976) and maize(Cook and Miles 1989), and the results of other Arabidopsis mutants reported in this thesis(see hefi, hcffl). Since the antiserum used for hc/2 is the same used for the other mutantsanalysed in this study, the results suggest that the CPJ polypeptides are at normal levels.The reasons for this descrepancy is, as of yet, unexplained. RNA Gel BlotsSteady state levels of mRNAs from nuclear and chioroplast encoded genes were allnormal except the chioroplast derived petA transcript which encodes the cytochrome fapoprotein (Figure 20). Overall the steady state level of chloroplast RNA appears to beslightly lower in the mutant based on results of chioroplast ribosomal RNA. It is not knownif this was due to slightly underloading hcf2 RNA, or due to a lower level of steady state95Figure 20. RNA gel blot of hc/2 and wild-type siblings. Gels were loaded withequal amounts of total RNA. DNA probes used (labeled above each blot) aredescribed in the Material and Methods. Molecular weights of the prominentbands were estimated using RNA standards (Boehringer Mannheim). ThepetA blot depicted also includes hcf6 and the wild-type siblings to show thatover loading was not the cause of the high level of the transcripts in hcf2.The hcf6 lane on the other hand was overloaded, but the major bands arebetter seen than the wild-type siblings.96atpAbcf •WtpetAbcf6 Wt bcf2 WtetBbcfWt23S rUNAlid Wtkbc.44.73.7241.9i.41.10.770.52psaBhcf 1bpsbDbcf WtrbcLlid Wtkb97chioroplast transcription. The petA transcript, on the other hand, is found to be at asignificantly higher steady state level in the mutant, although it is difficult to discern thewild-type level due to the contrast in the mutant and wild-type. The transcript sizescorresponding the expected wild-type are shown by the hcf6 mutant lane, which wasoverloaded (see above). Experiments are still in progress to calculate the level ofoverexpression of this transcript in the mutant. The transcript encoding the cytochrome fapoprotein is the 3’ gene of a polycistronic message also containing the 4 kDa PSI gene psal,and two other open reading frames whose protein products remain to be determined. One ofthe open reading frames encodes a putative zinc finger protein, designated zfpA (Sazaki etal., 1989) and the second open reading frame encodes a putative heme-binding polypeptide(Wiley and Gray, 1990). The primary transcript is processed, leading to a complex pattern oftranscripts similar to that described for petB-petD transcript (Sazaki, 1989; Wiley and Gray,1990).3.3.4. hcf33.3.4.1. Mutant Isolation and Pigment CompositionMutant line hcf3 was originally isolated from M1 bulk 70. The line was self-pollinated for four generations and crossed to marker lines in the Landsburg background.The hcf phenotype is inherited through the pollen, therefore the mutation is nuclear.Segregation ratios have been examined for markers on chromosomes 2, 3 and 4. The resultssuggest that a lethal phenotype may be responsible for a higher than expected number ofglabrous (gl-2; Table 5) plants in the F2 generation. This would be expected if the hcf locuswere linked in trans to the glabrous gene on chromosome 1. However, it must be pointed out98that the hcf3 line has not been screened for the presence of embryo lethals. Thus it ispossible that an embryo-lethal phenotype is co-segregating in the original line used in thecrosses, which could give the same result.The phenotype of plants grown on 112 MSO petn plates is light green to yellow. Theanalysis of the mutant is complicated by the appearance of a small yellow-green phenotypewhich survives in soil. At present it is unknown if the small yellow-green plants are hcf.The hcf plants are lighter green than the wild-type siblings and, based on two experiments,the hcf plants are slightly smaller (Table 7). All the pigments present in the wild-type arealso observed in the mutant, but, in different proportions (Table 8). No standard errors werecalculated to compare the fresh weight pigment content as only two replicates were done.When related to Chia, the level of all the pigments increased fairly proportionally,except for 13-carotene which was reduced. ChIb increased slightly thus resulting in a lowerChla/b ratio. Fluorescence Kinetics and Electron TransportThe level of the initial fluorescence (F0)is significantly increased in the mutantsuggesting that PSII is abnormal (Figure 21). Upon illumination, there is a reduced, butnormal fluorescence induction and decline to steady state fluorescence. Steady statefluorescence is slightly higher than the initial fluorescence, under low light (10 iE sm1).However under higher light intensities (100 jiE s1 m1)the F was below F0.Photosynthetic electron transport activity rates of the hcf siblings suggest electrontransport is primarily blocked at PSII (Table 9). PSII and PSI activity rates were 40% and80% of the wild-type level, respectively. This result agrees with the fluorescence kinetics andsuggests that energy transfer in P511 is somewhat impaired. However, for the PSII centers9965hcf3Time (s)Figure 21. Fluorescence induction kinetics of hcfl3 and wild-type siblings.Fluorescence kinetics are the average from 4-7 mutant and wild-tyie plantsmeasured using the PAM fluorometer with 1ow light levels (10 ilE m &1)100that are present, electron transport appears to be relatively normal based on the variablefluorescence kinetics. Chlorophyll-Protein ComplexesOn non-denaturing (green) gels all the Chi-protem complexes were observed, but theChia binding proteins of PSI (CPI), and PSII (CP43 and CP47) were all at lower levels(Figure 22). The Chi a+b antenna band CPII* appears to be slightly decreased, and the CPuband was increased (Figure 22). Variation in the ratio of the oligomeric and monomericforms of CPu caused by the use of different detergents during resuspension has previouslybeen observed (Green 1988). Since both the mutant and wild-type were treated in the samemanner, the differences observed could either be due to the different nature of the CPUcomplexes in the mutant, or to a real increase in the monomeric forms of the light harvestingcomplexes on a per ChI basis. A significantly higher level of the free pigment (FP) was alsoconsistently observed in the mutant.The results suggest that both PSI and PSII are reduced, but appear to be present inthe thylakoid membranes. No experiments were carried out to determine the presence ofspecific polypeptides in the mutant. As both PSI and PSII electron transport activity wasobserved, and Chl-protein complexes associated with each complex were also present in themutant, the corresponding polypeptides are also expected to be in the thylakoid membranesat reduced levels. RNA Gel Blots ofMutant Line hcf3All chloroplast and nuclear derived transcripts appear to be processed normally(Figure 23). When some the blots were probed with chloroplast ribosomal DNA, the steady101CP29CPUPPhcfWt hcfFigure 22. Chlorophyll-protein complexes of mutant hcf3 and wild-typesiblings. Unstained thylakoid membrane complexes run on a non-denaturing10% SDS-PAGE. Chi-protein complexes: CPJ: Photosystem I Chi a bindingcore complex; CP47 and CP43: Photosystem II 47 and 43 kfla Chi a bindingcore complex polypeptides; CPII* and CPu: the oligomeric and monomericforms, respectively, of the Chi a+b light harvesting complex polypeptides.cPICPII*CP47CP43102Figure 23. RNA gel blot of hcflE? and wild-type siblings. Gels were loaded withequal amounts of total RNA. DNA probes used (labeled above each blot) aredescribed in the Material and Methods. Molecular weights of the prominentbands were estimated using RNA standards (Boehringer Mannheim).103kb5. •Wtkb4. Wt kb0.52psbDhcf WtrbcLhcf Wt23S rRNAhcf Wtcabilhcf Wtkb2.81.41..10.4kb1.0104state level of chioroplast transcripts appears to be slightly lower in the mutant compared tothe wild-type siblings. It is unknown if this difference can be accounted for by unequalamounts of RNA loading from the mutant. However, the differences in loading do not appearto be significant, and this suggests that the steady state level of RNA is lower in the mutant.3.3.5.hcfl3.3.5.1. Mutant Isolation and Genetic AnalysisThe hcfl mutant line originated from M1 bulk 60. No morphological differences havebeen observed between plants that segregate for the hcfl phenotype and those that do not.Segregation ratios for the hcf phenotype have approximated a single nuclear recessive ratiofor six generations. For example, ten self-pollinated m5 plants were analysed for hcf progenyin the M6. Four of the ten did not segregate for the hcf phenotype. This is expected becausewhen hcf phenotype is maintained in the heterozygous condition, one-third will behomozygous wild-type, and two-thirds will be heterozygous. For the remaining six M5plants, 507 M6 progeny were scored and 107 were hcf. This frequency (21%) while slightlylower than the expected for a single nuclear recessive mutation (25%), may reflectdifferential survival in the gametophytic or early sporophytic stage (Delaert 1980; Simpsonand von Wettstein 1980). Similar results have been found in all the Arabidopsis hcf mutantsexamined.The genetic basis for the mutant phenotype was examined by crossing heterozygousmutant plants to several lines containing visible markers (as described in Materials andMethods). The resulting F1 plants grew normally in soil. Analysis of the F2 progeny of hcflcrosses suggest that the hcf phenotype is not inherited through the pollen. No hcf plants105were ever observed among F2 plants from several crosses. The hcfl plants used as thepaternal parent in the crosses were confirmed to segregate for the hcf phenotype when sell-pollinated. This result suggests that: 1) the hcf phenotype is not due to a nuclear mutation,but chioroplast mutation, and sectoring has led to segregation ratios which for sixgenerations approximated a nuclear recessive ratio; 2) the mutation disrupts photosyntheticcompetence in the Columbia background, and not in the Landsburg background; 3) twomutations occurred during mutagenesis, one nuclear and one cytoplasmic, neither of which islethal by itself. Although the mutant line has been designated hcfl, further genetic analysiswill be necessary prior to assignment of a genotype. However, having stated that, themutant line will be called hcfl for the remainder of the thesis for clarity. Physical characteristics of hcfl mutant lineMutant plants germinate in soil, but the cotyledons do not expand, and no primaryleaves are observed to form. On 112 MSO petri plates leaf expansion is normal, although thehcf plants are smaller (Table 7). The hcf plants flower after 2-3 months on the sucrosesupplemented medium, but no seed set was observed. They are lighter green and arevisually distinguishable shortly after germination even without UV light. At four weeks allthe pigments which are normally present in the wild-type siblings are also present in themutant (Table 8). Total Chi in the mutant line is only 28.4% on a fresh weight basis;however, the Chl a/b ratio was essentially unchanged. This implies that approximately 2.6Chia molecules were lost for each Chib molecule lost. The xanthophylls were lower by 58.5%,59.8% and 55.8%, for neoxanthin, violaxanthin, and lutein, respectively, on a fresh weightbasis. 13-Carotene was the pigment most severely affected, and was only 17% of the wild-typeon a fresh weight basis in the mutant.1063.3.5.3. Chlorophyll-Protein Complexes and Thylakoid ProteinsAs shown in Figure 24, CPI is almost completely missing. Occasionally, a very faintband can be observed. CP43 and CP47, the Chla antenna complexes of PSII, appearunchanged. The Chi a+b antenna complex CPII* is slightly lower, and the CPH band ishigher in the mutant. Variation in the ratio of the oligomeric and monomeric forms of thelight harvesting complex was noted above in the mutant hct3. Since both the mutant andwild-type were treated in the same manner, the differences observed could either be due tothe nature of the CPu complexes in the mutant being different, changes in the lipidcomposition surrounding the light harvesting complex, or to a real increase in the monomericforms of the light harvesting complexes on a per Chi basis. A significant increase in the levelof the free pigment (FP) is also consistently observed in this mutant.To determine if the PSI proteins were present in the thylakoids, but they were notassociated with chlorophyll, denaturing gels were run and immunoblotted. Results indicatethat for all PSI polypeptides tested: the core PSI Chla binding polypeptides (CPI), PSIsubunit II and subunit V are similarly decreased in the mutant (Figure 25). Whenimmunoblotted with antiserum prepared against PSI subunit VI, the two upper bands are atlower levels in the mutant, while the lower band was at wild-type levels. The identity of thelower molecular weight polypeptide is unknown but the results suggest that it is probablynot a PSI polypeptide.When ixnmunoblotted using antibodies against PSII (CP47, CP43, D2 and cytochromeb559), cytochrome b fr (cytochrome f) and coupling factor (CF1)polypeptides, no differenceswere observed between the mutant and wild-type (Figure 25 and data not shown). Thus themutation appears to be specific for PSI. Previous analysis of mutants reduced in PSI haveshown that the loss of one component has a pleiotropic effect and leads to the loss of the107Figure 24: Chlorophyll-protein complexes of hcfl and hcf8 and theirrespective wild-type siblings. Unstained thylakoid membrane complexes runon a non-denaturing 10% SDS-PAGE. Chl-protein complexes: CPI:Photosystem I Chi a binding core complex; CP47 and CP43: Photosystem II47 and 43 kDa Chi a binding core complex polypeptides; CPII* and CPH: theoligomeric and monomeric forms, respectively, of the Chl a+b light harvestingcomplex polypeptides.hcf1hcf Wthcf8hcf WtcPIcpJI*CP47CP43CP29CpuFPI108hcfl hcfSWt hcf Wt hcfcpi__Cytf *L__ ____Sub. II__Sub V -rSub.VI {‘— mmriFigure 25. Immunoblot of hcfl and hcf8 and their respective wild-typesiblings. Denaturing gels were loaded with equal amounts of Chi from themutant and wild-type. Both nitrocellulose filters were successvelyimmunoblotted with antibodies PSI 80-82 kDa core complex (CPI),cytochrome flCyt /), PSI subunit II (Sub. II), PSI subunit V (Sub. V) and PSIsubunit VI (Sub. VT).109entire complex (Chitnis et al 1989; Cook and Miles 1990). The fact that all polypeptidestested were present in the mutants precludes the possibility of a null mutation in one of thegenes encoding these polypeptides. Electron Transport and Fluorescence induction KineticsPSI electron transport activity rate, measured by TMPD/ascorbate to methylviologen, was found to be 25% of the wild-type siblings (Table 9). The activity measurementscorrespond to the fact that hcfl has a low level of PSI Chi-protein complex, although the rateis higher than would be expected on the basis of the level of PSI ChI-protein complex and PSIpolypeptides. PSII electron transport, as determined by DCPIP reduction using both waterand DPC as a donor, was found to be identical in the wild-type and mutant.Fluorescence induction of the mutant lines in later generations was assayed underlow actinic light (10 iE m1)using a PAM fluorometer (Schrieber et al 1986). This wasdone to confirm the fluorescence kinetics observed with the CAFVIS (Figure 5). Low lightlevels were used to resolve the early events in electron transport. The mutant appears tohave a normal induction to the maximal level of fluorescence (FM), but there is no decline toa steady state fluorescence (F8). This result agrees with the fluorescence kinetics observedin the M2 plants from the original M1 bulk 60 (Figure 5 and 26A). It indicates that chargeseparation and electron flow in PSII to QB is occurring, but since there is essentially nodecrease from the maximal fluorescence level (FM) in the mutant plants, it suggests that QAis not being reoxidized. However, it is interesting to note the lack of the O-I-D-P kinetics inthe mutant (Figure 26B). Using the low light levels the early events in the fluorescenceinduction were very clearly resolved in the wild-type. Thus while electron flow in P811 isoccuring, the fluorescence kinetics is not completely normal. The kinetics responsible for the11080A70 hcflQ6OCfiC. 5020;II4ba6bIsb1. Do•1.0 p I I I I0 2 4 6Time (sec)Figure 26. Fluorescence induction kinetics of hcfl and wild-type siblings. (A)Fluorescence kinetics is an average from 4-7 mutant and wild-type plantstaken using the PAM fluorometer using low light levels (10 tE m2 s’). (B)A representative example of the early induction of the mutant. Note the lackthe dip during induction. The flourescence induction was normalized (FV/F0)so both mutant and wild-type could be shown on the same panel.111fast rise (0-I) is suspected to be due to a population of non-reducing PSII centers, ie. PSIIcomplexes lacking or defective at the QB site (Govindjee, 1990; Melis, 1985). However, sincethe electron transport activity rates in the mutant were similar to the wild-type, it does notappear that P511 electron transport is impaired or reduced. It, therefore, appears that nore-oxidation of QA is occuring due to the block in the electron transport at PSI. RNA Gel Blots ofMutant Line hcflChioroplast transcripts encoding proteins which are part of the PSI complex wereassayed to determine if any gross abnormalities could be detected, since genetic analysissuggested that the phenotype of mutant line hcfl may be inherited, at least in part,maternally. As shown in Figure 27, the transcripts of the psaB gene encoding one of the PSI82 Kd core Chia binding apoproteins appeared at normal levels. The gel used to probe forpsaC RNA encoding the 8-9 kDa iron-sulfur protein was of poor quality, but no obviousdifferences were observed (not shown). The fourth PSI chioroplast encoded gene, psal, whichencodes a 4 kDa polypeptide of unknown function was not assayed, but the petA geneencoding the cytochrome fapoprotein, which is co-transcribed with the psal gene was used asa probe (Wiley and Gray, 1990), and no differences were observed (not shown). RNAsencoding other chioroplast proteins (atpA, atpB IE, petB, psbD, and rbcL) were normal.Nuclear PSI derived psaD encoding subunit II mRNA was also normal. The results suggestthat the hcf phenotype of the hcfl line is not due to abnormalities at the level oftranscription, RNA processing or stability.112Figure 27. ENA gel blot of hcf3 and wild-type siblings. Gels were loadedwith equal amounts of total RNA. Molecular weights of the prominent bandswere estimated using RNA standards (Boehringer Mannheim).113psaBhcf Wt kbatpAhcf •WtkbI4. WtkbrbcLhcf Wtkb114psbDhcf WtpetBhcf Wtkb5. Wtkb3.3.6. hcf83.3.6.1. Mutant Isolation and Genetic AnalysisThe mutant line hqc8 originated from M1 bulk 393. The mutant phenotype isinherited through the pollen, therefore it is a nuclear mutation. Analysis of the F2 progenyfrom crosses to lines with markers on chromosomes one through four did not reveal anysignificant deviations from the expected frequency of the markers. Therefore the hc/8 locusis either on chromosome 5, or not close enough to any of the markers tested to give sufficientresolution by the method used. Physical characteristics of the hcf8 mutant lineMutant plants germinate in soil but the cotyledons do not expand, no primary leavesare observed to form and the seedlings eventually die in 7-14 days. On 1J2 MSO petri plates,hc/8 seedlings are visually indistinguishable from their wild type siblings both with respectto pigmentation and fluorescence. Only after 10-14 days can the hcf and wild-type plants bedifferentiated by their fluorescence, at which time the hcf plants become noticeably lightergreen compared to the wild-type siblings. After four weeks on 1/2 MSO plates, hcf8 plantsare significantly smaller and lighter green than the wild-type siblings (Table 7). However,the newer growth remained green. The fluorescence was also normal in these. In an effortto determine if light or temperature would affect the expression of the hef phenotype, hcf8seeds were germinated under different light (25-40 iiE m2 16/8 h; and 100-150 jiE m2s, 16/8 h and 24/0 h lightJdark ), and temperature (23 and 27 C under 100- 150115jiE m2s1, 16/8 h lightldark photoperiod) regimes. No differences were observed in thetiming of expression of the hcf phenotype.At four weeks all the pigments which are normally present in the wild-type siblingswere also present in the mutant, albeit reduced and in different proportions (Table 8). TotalChi on a per fresh weight basis was reduced by 65.5% in the mutant; however, the Chl a/bratio decreased only slightly (8%) in the mutant. This implies that approximately 2.4 Chl amolecules were lost for each ChI b molecule lost. The xanthophylls were also reduced, butwere less affected.Electron microscopy revealed that no ultrastructural differences were evidentbetween the mutant and wild-type siblings (not shown). Fluorescence Induction Kinetics and Electron TransportThe fluorescence kinetics of M2 plants recorded from the original 393 M1 bulk ispresented in Figure 28k However, it is surprising that the hcf phenotype could not beobserved during the early stages of growth on 1/2 MSO plates (see above), yet the kineticsobtained from the original M2 plants was done prior to leaf expansion. The discrepancy wasresolved after analysing the fluorescence kinetics in the later generations which shows thatthe kinetics obtained in the subsequent generations did not resemble those obtained in theM2 generation (Figure 28B). The fluorescence induction of mutant hcf8 line is similar to thatof hcfl after the hcf phenotype manifests itself. Thus the fluorescence kinetics of the latergenerations is in agreement with electron transport activity rates, Chl-protein complexes andimmunoblot data (see below) which suggest that the mutant phenotype is due to a disruptedPSI complex. The result above suggests that there were two distinct hcf mutations whichoccurred in the original 393 M1 bulk population.116300C)200C)100•L4 ‘1000C)8006004000 20 40 60Time (s)Figure 28. Fluorescence induction kinetics of hcf8 and wild-type siblings. (A)Fluorescence kinetics is an average from 4-7 mutant and wild-type plantstaken using the CAFVIS in the M2 generation. (B) Fluorescence kinetics isan average from 4-7 mutant and wild-type plants taken using the PAMfluorometer under low light levels (10 pE m2s4) in the M4 generation.117800700600500400hcfI I I I I I I I I I I I I I0 4 8 12 16 20 24 282000Electron transport activity rates were assayed to veriI’ the lack of PSI electrontransport. PSI activity, measured by TMPD/ascorbate to methyl viologen, was found to be40% of the wild-type siblings in the mutant (Table 9). The activity measurements correspondto the reduction in PSI proteins, however, the rate does appear to be higher than the level ofPSI polypeptides (see below). PSII electron transport activity rate, as determined by DCPIPreduction using DPC as an electron donor, was identical to the wild-type. Chlorophyll-Protein Complexes and Thylakoid ProteinsAs shown in Figure 24, the PSI Chi-protein bands are severely reduced in themutant. CP43 and CP47, the ChIa antenna complexes of PSII, appear unchanged. The Chla+b antenna complexes CPII* , CPII and CP29 appear to be normal.To determine if the PSI proteins were present in the thylakoids but were notassociated with chlorophyll, denaturing gels were run and immuno-blotted. Results indicatethat all PSI polypeptides are similarly decreased in the mutant and that there is a slightlyhigher level of CPI apoprotein than observed in hcfl (Figure 25). When immunoblotted withantiserum prepared against PSI subunit VI, the lower band was at wild-type levels, whereasthe upper two bands are at similar low levels as the other PSI polypeptides.When immunoblotted using antibodies against PSII (CP47, Dl and cytochromeb559), and cytochrome b6/fcomplex (cytochrome /) polypeptides, no differences wereobserved between the mutants and wild-type (Figure 25 and data not shown). Thus theblock in electron transport appears to be the result of a reduced PSI complex in the mutant.Since the phenotype is only observed in the older tissues, this suggests that the mutationmay be involved in stability of the PSI complex in the membrane and not in assembly per se.For this reason, electron transport activity rates and the level of PSI ChI-protein and118polypepides could be biased upwards, as the entire plant was harvested to isolate thylakoidmembranes.3.3.7. hcf43.3.7.1. Mztant IsolationMutant line hcf4 was originally isolated from M1 bulk 146. The line was self-pollinated for four generations. Analysis of single plants suggests that the hcf phenotypesegregates as a single nuclear recessive. Most of the plants segregate at a frequency of 21-26% for the hcf phenotype, although a few plants were observed to segregate 31-33% for thehcf phenotype in the M4 generation. The frequency observed in the later plants suggeststhat an embryo-lethal mutation co-segregated in the original M2 plant from which the hcf4line was derived. The line has not been backcrossed to the Columbia wild-type nor crossed tomarker lines, therefore, assignment of the hcf4 designation should be viewed as tentativeuntil nuclear inheritance is confirmed. Physical Characteristics and Electron MicroscopyThe hcf seedlings can be differentiated from their wild-type siblings in soil and on 112MSO plates by their lighter color even without screening for the hcf phenotype. In soil thehcf seedlings die within 10-14 days following germination. On 1/2 MSO plates, the mutantplants are significantly smaller than the wild-type siblings after four weeks growth (Table 7).There were no observed differences in chloroplast ultrastructure between the mutantand wild-type siblings grown on 1/2 MSO plates for four weeks (Figure 8).1193.3.7.2. Pigment CompositionAll the pigments present in the wild-type were also present in the mutant. However,the level of Chl on a fresh weight basis was significantly lower in the mutant (Table 8).Relative to ChIa all the pigments, except 13-carotene, were only slightly increased comparedto the wild-type. On a fresh weight basis, Chia was reduced by 83% and Chib by 78%reflecting the lower Chla/b ratio. The xanthophylls were all similarly reduced in themutant. Lutein, violaxanthin, and neoxanthin were reduced by 75-78% in the mutant on afresh weight basis. 3-Carotene was only 11.2% of wild-type levels on a fresh weight basis.The xanthophyll, antheraxanthin, present in all the other Arabidopsis hcf mutants to date,was found to be at its highest level in the hcf4 mutant line. The significance of this in so faras the mutant phenotype is concerned is not known. Fluorescence and Electron TransportRoom temperature induction curves for the mutant and wild-type are shown inFigure 29. The mutant has slightly higher F0, yet it appears that the induction is normal.No decline in the maximal fluorescence level (FM) was observed. These results suggest thatwhile PSII is somewhat affected in the mutant, electron transport appears to be blocked atsome point after QB. Since there was no decline from FM, it suggests that electron transportis blocked either at the cytochrome complex or PSI complex.PSII activity rates suggest that while the level of PSII centers may be slightlyreduced accounting for the observed increase in F0, the difference observed between the wildtype and mutant siblings was not statistically significant (Table 9). PSI activity rates, on theother hand, were essentially no different than the wild-type sibling rates. Thus whilefluorescence kinetics data suggest that electron transport is blocked at some point after QB.120807060hcf450040• 3020100 I I I I I I—0 10 20 30 40Time (s)Figure 29. Fluorescence induction kinetics of hcf4 and wild-type siblings.Fluorescence kinetics is an average from 4-7 mutant and wild-type planttaken using the PAM fluorometer using low light levels (10 jiE m2s1-).121PSI electron transport is not disrupted between the site ofTMPD/ascorbate electrondonation (P700) and reduction of methyl viologen (the iron sulfur cluster Fx). Chlorophyll-Protein Complexes and Thylakoid Membrane PolypeptidesOn non-denaturing (green) gels all the bands present in the wild-type were alsoobserved in the mutant (Figure 30). CPI, CP47 and CP43 appear to be at slightly lowerlevels, when gels were loaded with an equal amount of Chi, but otherwise, they appearnormal. Since the electron transport activity rates suggested that neither PSI and PSII weresignificantly affected, these results were not unexpected.To determine if any differences could be detected at the polypeptide level, thethylakoid proteins were completely denatured, separated by gel electrophoresis andsuccessively immunoblotted with polypeptide specific polyclonal antibodies (Figure 31). Allpolypeptides present in the wild-type were also found in the mutant. Differences wereobserved only when the filter was blotted with antibodies to PSI polypeptides. Subunit VI,subunit II and CPI were all at slightly lower levels on a per Chi basis. At the polypeptidelevel, no other differences were apparent. The PSII polypeptide CP47 appears to be atnormal levels (not shown), suggesting that PSII is not significantly reduced. The cytochromef and coupling factor polypeptides appear at wild-type levels (Figure 31 and not shown).At the polypeptide level, except for a slightly lower level of PSI polypeptides, no grossabnormalities could be detected that would account for the block in electron transport in themutant. In addition, while the fluorescence kinetics suggest that the hcf phenotype may bedue to a block in PSI, they contrast with the PSI electron transport rates which showed nodifferences in the mutant and wild-type. It is possible that a polypeptide in the PSI complexwhich was not screened for, has been lost. Another possibility is that a soluble component in122cPIcPII*CP47CP43CpuFigure 30. Chlorophyll-protein complexes of hcf4 and wild-type siblings.Unstained thylakoid membrane complexes run on a non-denaturing 10%SDS-PAGE. Chi-protein complexes: CPI: Photosystem I Chi a binding corecomplex; CP47 and CP43: Photosystem II 47 and 43 kDa Chi a binding corecomplex polypeptides; CPII* and CPu: the oligomeric and monomeric forms,respectively, of the Chi a+b light harvesting complex polypeptides.hcf4Wt hcfFP123hcf4Wt hcfCytfSub. IISub.V.L {Figure 31. Immunoblot of hcf4 and wild-type siblings. Denaturing gels wereloaded on equal amounts of chl from the mutant and wild-type. Thenitrocellulose filter was successvely immunoblotted with antibodies to to PSI80-82 kDa core complex (CPI), cytochrome flCyt f), PSI subunit II (Sub. H)and PSI subunit VI (Sub. VI)cPI124the electron transport chain is missing or non-functional. Since only the thylakoidmembrane fraction was assayed in this study, the presence or absence of the solublechloroplast proteins can only be speculated upon. RNA Gel BlotsNo differences were observed in the chloroplast and nuclear encoded transcriptsbetween the mutant and wild-type siblings (Figure 31). Some of the higher RNA levelsobserved in the mutant lanes were due to overloading. This was verified using a chioroplastDNA probe containing the genes for the chloroplast ribosomal RNA. The results suggest thatthe hcf phenotype of the hcf4 line is not due to abnormalities at the level of transcription,RNA processing or stability.3.3.8. hcf7The hcf7 line was originally isolated from Ml bulk 310. In soil and on 112 MSO petriplates the hcf siblings were yellow. As genetic analysis of this line progressed, it becameevident that the yellow phenotype observed was not part of the hef phenotype, but due toanother locus, probably a linked gene. Isolation of a plant which was not yellow segregatingfor the hcf phenotype in the M4 generation confirmed that the yellow phenotype was not partof the hcf phenotype. Thus, previously collected data on the mutant were discarded. Presentanalysis of this mutant line has been restricted to electron transport activity rates andpigment data, in addition to the fluorescence kinetics (Figure 33).Pigment analysis shows that although the Chlalb ratio was slightly reduced, all thepigments normally found in Arabidopsis are present in the mutant and the level does not125Figure 32. RNA gel blot of hcf4 and wild-type siblings. Gels were loaded withequal amounts of total RNA. DNA probes used (labeled on above each blot)are described in the Material and Methods. Molecular weights of theprominent bands were estimated using RNA standards (BoehringerMannheim).1260.52kb2.6- 2.2cablihcf Wtkbkb2. WtatpAhcf Wtkb5. Wt kb0aatpBIEhcf WtpsbDhcf Wtkb23S rRNAhcf Wt-I12780hcf760 hcf2010 I I I I I I I I I p0 20 40 60 80 100 120Time (s)Figure 33. Fluorescence induction kinetics of hcf7 and wild-type siblings. (A)Fluorescence kinetics average from 4-7 mutant and wild-type plant takenusing the PAM fluorometer in the M4 generation.128appear to be significantly different. Chl-protein complexes associated with PSII are also atnormal levels, whereas CPI is reduced (Figure 34).Both the fluorescence kinetics and electron transport activity rates point to the causeof the mutant phenotype being a disruption at some point past QB. Fluorescence inductionis normal, and PSU electron transport appears to be normal (Table 9). Unlike other PSImutants described in this thesis, there appears to be a slight, but detectable decrease in theM suggesting that some electron transport is occurring. This may be the reason for thequite high PSI electron transport activity rate observed seen, although the fluorescencekinetics suggest that electron transport is blocked. A lower level of CPI is observed on non-denaturing gels, however, confirmation of the level of PSI polypeptides has not been done. Inspite of the small amount of data presently available on this line, I suggest that theArabidopsis hcf7 designation be reserved and published as the data becomes available.3.4 ADDITIONAL LINES SEGREGATING FOR THE HCF PHENOTYPEOf the twenty-one hcf lines isolated, only eight have been further characterized anddescribed above (Table 6). An additional thirteen lines that appear to segregate as nuclearrecessives, have also been isolated but have not been further characterized. The reasons fornot characterizing the addtional lines are fourfold. First, the fluorescence induction kineticsof some of the lines (from M1 bulks 41, 44, 54, 57, 72 and 80) display a significant decreasefrom the peak fluorescence (Figure 35). All of these lines were isolated prior to screening theplants from M1 bulks on the CAFVIS. As previously mentioned, this fluorescence pattern isexpected to yield mutations which affect electron transport subsequent to PSI (Edwards etal., 1988; Miles, 1980). Since the goal of this project was to isolate mutants that wereblocked early in electron transport, only the most promising lines were analysed.129CP47CP43wthcf7CpI pJpcPfl*CpuPphcfFigure 34. Chlorophyll-protein complexes of hcf7 and wild-type siblings.Unstained thylakoid membrane complexes run on a non-denaturing 10%SDS-PAGE. Chi-protein complexes: CPI: Photosystem I ChIa binding corecomplex; CP47 and CP43: Photosystem II 47 and 43 kDa ChIa binding corecomplex polypeptides; CPIP and CPu: the oligomeric and monomeric forms,respectively, of the Chl a-i-b light harvesting complex polypeptides.IJI_•••130A:400450I_350300 500250150200100 10010..20 30 1ö2030U0800 800V—700700600500 500400400300100 V00 10 2) V3o 401000 0 - 3 40Time (s)Figure 35. Fluorescence induction kinetics of wild-type and hcf mutantplants from lines in the M3 generation not selected for characterization. (A)Line from M1 bulk 44; (B) Line from M1 bulk 54; (C) Line from M1 bulk 57;(D) Line from M1 bulk 72.131Second, some of the lines (eg. from M1 bulks 45, 155, 207 and 322) contain additionalmorphological phenotypes which co-segregate with the hcf phenotype (fluorescence inductionkinetics of these lines is presented in Figure 36). Although the extent of additionalmutations in each line is unknown, only lines which appeared morphologically normal werecharacterized. Further characterization of these lines will require that they be backcrossedto the wild-type parent.Third, two of the lines (from M1 bulks 16 and 392) appear to segregate for more thanone hcf mutation. Thus, either more than one hcf mutation occurred in these lines, or asecond mutation exerts a pleiotropic effect on the hcf mutation. Regardless, furthercharacterization will require that the lines be backcrossed to the wild-type, in order to get a‘clean’ mutant phenotype.Fourth, one of the lines (originating from M1 530; see Figure 5 for fluorescenceinduction kinetics) has an extremely low level of Chi, making analysis of the thylakoidmembrane proteins difficult. The phenotype is inherited through the pollen, therefore it is ofnuclear origin. Segregation ratios (ji=24.4%; 177 hcf: 723 wild-type; from combined F2 datafrom crosses derived from two different marker lines) suggest that the phenotype is due to asingle nuclear mutation.Germinating hcf seedlings from the Ml: 530 line are distinguishable on 1/2 MSOpetri plates due the their pale green color even without screening for the hcf phenotype. Asthe tissue ages, bleaching occurs, leading to loss of all photosynthetic activity and pigments.The new growth is pale green and exhibit,s the hcf phenotype. Bleaching does not appear tobe due to a disruption in pigment biosynthesis per Se, as all the major pigments present inthe wild-type are also observed to be present in the mutant. Chloroplast ultrastucture in thepale green tissues show that only rudimentary thylakoid membranes are present (not132900800 A 700700 600600400 400500300 300rjJ2000 4812162024 28 200-1’4 800700 1000600 900800500 700600400 500300200100 10000 4812 162024 2 4812162024 28Time (s)Figure 36. Fluorescence induction kinetics of wild-type and hcf mutantplants in the M3 generation from lines not selected for characterization dueto additional morphological phenotypes. (A) Line from M1 bulk 16; (B) linefrom M1 bulk 45; (C) line from M1 bulk 207; (B) line from M1 bulk 322.133shown). Analysis of the chioroplast ultrastructure in the older, bleached tissues, has notbeen done.3.5. ARABIDOPSIS TISSUE CULTURE STUDIES3.5.1. Response of Wild-type and Point Grey Cotyledon ExplantsThe results using cotyledon explants agree with previous reports that the Columbiaecotype responds poorly in culture (Table 10). In addition, many of the regenerated shootsdid not survive to produce seeds. Typical losses of 50-90% at this stage are not uncommon(Chaudhury and Signer, 1989).The Point Grey ecotype was evaluated as it appeared to regenerate shoots and rootssimultaneously in a preliminary screen of wild-type ecotypes using the leaf segment protocolof Lloyd et al., (1986). These results were confirmed using cotyledon explants placed onSIM1 or SIM2 (Table 10). In experiments where Point Grey and Columbia explants wereplaced on the same petri plate, the response of the Point Grey ecotype, especially on theSIM2 media was quite rapid (Figure 37). Within 7-10 d. many root hairs and roots formedover the callus, and shoots differentiated within 14-2 1 d. Sectioning was not done to confirmthe vascular continuity; however, most shoots which were normal in.appearance grew to seedset following transfer from SIM medium to soil. A similar response has been observed usingthe Columbia ecotype when the explant source was embryonic cotyledons (Patton andMeinke, 1988). These results were confirmed in this study on SIM2 medium, usingColumbia and Point Grey ecotypes (data not shown). Regeneration of shoots and roots isvery rapid when embryonic sources are used, and shoots can be transferred from SIMdirectly to soil. However, due to the small size it is technically difficult and time consuming134Table 10. Response of Columbia, Point Grey and Columbia X Point Grey Fl explants inculture. Number and percent of explants forming roots or shoots on shoot inducing medium.SEMiExplant SIM2Roots Shoots Roots ShootsColumbia 0/200 12/200 5/212 7/212% 0 0 2 3:Point Grey 411112 58/112 75/76 49/76% 37 52 99 64F1 11/80 9/80 100/100 45/100% 14 11 100 45135Figure 37. Response of Columbia and Point Grey cotyledon explants on shootinducing medium. A comparison of both ecotypes (A) and a close up ofColumbia (B) and Point Grey (C)136to obtain explants from Arabidopsis embryonic sources for routine purposes. It appears thatthe Point Grey ecotype retains its regenerative potential following germination, whereas thepotential is lost in the Columbia ecotype during seed desication and/or germination forunknown reasons. One possible explanation is the difference in .the level of endogeneoushormone in the embryonic tissues compared to the germinating seedlings.3.5.2. Genetic AnalysisThe Point Grey and Columbia ecotypes were crossed and the progeny were evaluatedto search for a line which would germinate with the ease and uniformity of the Columbiaecotype and respond to the shoot inducing medium as the Point Grey ecotype. The responseof the F1 was different on the two media tested. On the SIM2 root and shoot regenerationappears to be under the control of dominant inheritance. This agrees with the results ofChaudhury and Signer (1989) where the F1 in crosses between high and low regenerativeecotypes, Nossen and Landsburg respectively, responded as the high regenerative ecotype.However, on SJM1 the rooting response was slightly below the midparent value. Shooting,on the other hand, was only slightly above the response observed for the Columbia ecotype,and well below the midparent value. Thus on SIM1 the response of the F1 appears to beintermediate to recessive. Since the two shoot inducing media were compared at the sametime, the difference in response appears to be due to the change in the level of NAA in themedium. The most likely explanation for the results is that a threshold level in the auxinconcentration, or the cytokinin to auxin ratio, is responsible for the different response of theF1 on the two media. Furthermore, the threshold level required for the F1 explants isdifferent from either the Columbia or Point Grey ecotypes. Even though only yes/noresponses were recorded insofar as regeneration of shoots and roots is concerned, it was137observed that in the F1 the roots were not as numerous as the Point Grey ecotype. Thissuggests that there is some intermediate response, thus inheritance may not be under simplegenetic control. Further comparisons with slight variations in the auxin and cytokinin ratiosare required to define the threshold level is for both parents and the F1.Since the F1 responded similarly to the Point Grey parent on SIM2, giving theimpression of dominant inheritance, only SIM2 was used to evaluate the F2 and F3generations. Both cotyledon explants from each seedling were placed side by side on eachpetri plate, thus it was possible to evaluate the response of each seedling. The highproportion ofF2 seedlings in which only one of the two explants responded by regeneratingroots and/or shoots suggests that there is a high degree of environmental influence on thesetraits (Table 11). This is not surprising due to variability observed when essentiallyhomozygous lines are evaluated for regeneration. Although root and shoot regeneration areprobably not coupled, explants that regenerate roots appear to have a higher probability ofalso regenerating shoots. The low number ofF2 explants that shooted was surprising. It ispossible that the different F2 genotypes have different auxin to cytokinin requirements. Asecond possibility is that the light intensity was too high in the growth chamber. Shortlyprior to experiments in which the F2 explants were evaluated, new lights were placed in thegrowth chamber. The light levels used (100- 150 iE m2 1), unknown to us at the time,have been shown to be suboptimal for shoot regeneration in Arabidopsis (Chaudhury andSigner 1989). A third possibility, is that there is a correlation between cold treatmentrequirement and propensity for regeneration. The F2 seedlings used were germinatedshortly after harvest, and no cold treatment was applied. Since only F2 seedlings whichgerminated were used, a high number of the progeny tested would be similar to theColumbia ecotype.138Table 11. Root and shoot regeneration of Columbia X Point Grey F2 and F3 seedlings onSIM2.Explant Total Rooting ShootingBoth One Neither Shoots On Shoots OnExplants Explant Explant Rooting Non-rootingRooted Rooted Rooted Explants ExplantsF2 177 91 45 41 37 4% 53 26 24 16 3F3 Individual Plants:1 59 26 23 10 21 02 44 19 14 11 4 03 27 11 14 2 18 04 28 0 0 28 0 15 43 20 9 14 7 16 44 8 15 21 1 07 43 16 11 16 4 38 28 1 9 18 1 09 14 4 3 7 3 010 27 8 10 9 1 111 14 13 1 0 1 012 59 38 12 9 12 013 28 11 8 9 12 014 30 22 1 7 10 215 29 12 7 10 12 616 44 19 9 10 5 117 28 7 5 16 7 118 13 2 3 7 4 019 11 1 4 6 4 020 15 13 2 0 4 0F3 Combined:627 251 160 216 131 16% 40 25 34 20. 3139Since over 50% of the F2 were observed to have both explants rooted, seeds fromindividual F2 plants were harvested and the F3 generation was evaluated to determine ifthis trait is simply inherited on SIM2 (Table 11). As was observed in the F2, the high level ofvariability makes interpretation difficult. However, it is interesting to note that the overallvariability observed in the F3 is on the same order as observed in the F2. Thus, while rootregeneration may be under the influence of a few genes under these conditions, additional F2lines will be required to get a better estimate of the number. No attempt was made tointerpret the shooting results except to note that, as was observed in the F2, explants thatformed roots also appeared to be more apt to form shoots (Table 11).3.5.3. TransformationThe high morphogenic potential of the Point Grey ecotype suggests that it would be agood ecotype to use for transformation. Cotyledons infected with Agrobacterium resulted ina high number of transformed calli, but regeneration of seed bearing plants was no moresuccessful than for the Columbia ecotype. The reason for this was that roots formed shortlyafter being placed on selective medium, but they died once in contact with the medium. Ifthe dying roots were not excised, the entire callus eventually died. When the roots wereexcised, the callus remained green, and proliferated, however, shoot regeneration was slow(only after 4-6 weeks) and none of the calli formed subsequent roots. Thus it appears thatroot morphogenesis is very rapid, and the roots were not derived from cells which weretransformed. In this respect the morphogenic potential appears to be a detriment fortransformation of cotyledon explants.When the root transformation protocol of Valvekens et al. (1988) was used,uninfected Point Grey root explants were found to be highly morphogenic, and resulted in140numerous roots and shoots from each explant. It was interesting to note that the Point Greyroot explants did not require a Lallus inducing medium (CIM) pretreatment prior to placingon SIM to result in a non-polar response (data not shown). Since previous reports indicatedthat a pretreatment on CIM leads to better shoot regeneration, all explants exposed toAgrobacterium infection were placed on CIM prior to transfering to selective SIM2. Theresults comparing the transformation frequencies of the Point Grey ecotype with the C24,Columbia and Landsburg ecotypes are presented in Table 12. Although the lowtransformation efficiencies of the Columbia and Landsburg ecotypes were not unexpected,the meager regeneration of the C24 ecotype was surprising. This ecotype was selected forcomparison as it has been shown to have high regenerative potential from root explants(Valvekens et al., 1988). The observed result is probably due to the use of different levelsand type of auxins and cytokinins in the shoot inducing medium. One observation of interestwas noted when the explants were transferred to a selective SIM2 medium containingCefotaxime instead of Vancomycin (which was not available at the time) to contain bacterialgrowth. Cefotaxime is an inhibitor of prokaryotic cell wall formation, but has also beenshown to inhibit regeneration from root explants for unknown reasons (Valvekens et al.,1988). Regeneration from root explants from the Columbia and Landsburg ecotypes on theseplates was completly inhibited and almost no callus growth was observed. Root explantsfrom the Point Grey ecotype however, were only slighly inhibited and resulted intransformed shoots. The reason for this difference is unknown.141Table 12. Explants infected with Agrobacterium forming shoots on selective SIM2.Genotype Total Explants TransformationExplants Shooting EfficiencyPoint Grey 104 42 40.4%Columbia 110 13 11.8%Landsburg 120 9 8.3C24 40 9 22.5%Columbia XPoint Grey F3 171 31 18.1%Columbia X Point Grey F3 line used was line 1 from Table 11.142CHAPTER 4DISCUSSION4.1. HCF MUTANT ISOLATION AND GENETIC ANALYSISThis manuscript describes the isolation of photosynthetic mutants ofArabidopsisthaliana using a high chlorophyll fluorescence screening procedure. The high frequency ofhcf plants observed in the original M1 bulks suggests that there are many nuclear loci inArabidopsis which are required for correct assembly and function of the thylakoidmembranes. Since Miles and collaborators had established that there are a high number ofnuclear hcf loci that have been genetically established in maize (Cook and Miles, 1988), smallM1 bulks were screened in the present study, and only one M2 line from each wasmaintained. The use of the computer-aided fluorescence video imaging system (CAFVIS)(Fenton and Crofts, 1990) allowed for rapid quantification of the fluorescence kinetics of themutant and wild-type siblings on the same petri plates. In this respect, the small size ofArabidopsis allows for manipulations similar to those used for microorganisms such as,Chlamydomonas where the hcf phenotype was first described (Bennoun and Levine, 1967).As the goal of this project was to isolate mutants blocked early in photosynthetic electrontransport, the CAFVIS allowed concentration on the most promising lines. Visual screeningmethods were sufficient to find putative hef plants, but some of the lines isolated by visualmeans were eliminated after screening with the CAFVIS because they displayed variableresults.Of the 27 M1 bulks originally selected that displayed the hcf phenotype, 6 failed toyield any M2 plants segregating as nuclear recessives in the M3. This suggests that, at leastin some cases, the hef phenotype observed in the original bulks may have been due to a143cytoplasmic mutation. This is especially relevant in light of the inheritance of the hefimutant line. The hcfl line appears to segregate as a nuclear recessive, but the phenotype isnot inherited through the pollen. One possible explanation is that the phenotype is inheritedmaternally and the ratios observed are coincidental. This seems unlikely, for if theinheritance is strictly maternal, the hcfl wild-type gene product is not cell autonomous. Thisconclusion is based on the fact that no sectoring has been observed between plantssegregating for the hcf phenotype, and those that do not. A second possibility is that twomutations occurred in the original M1 plant, one nuclear and one cytoplasmic. If this is thecase, neither of the two mutations appear to have any detrimental effects on Arabidopsisgrowth and development in the absence of the other. Reciprocal crosses would be required toveri1’ this hypothesis. However, the hcfl mutation may offer a unique opportunity to studythe interaction of the nuclear and plastid genomes.All of the other hcf lines analysed genetically segregated for the hcf phenotype as asingle nuclear recessive gene. At present, the location of the mutations have not beenmapped in relation to known genetic markers. The presence of additional mutations, namelyembryo-lethals, co-segregating with the hcf phenotype interfered with mapping. The geneticmethod used in this study was dependent on deviations from the expected 3 wild-type:1visible marker ratio. However, any co-segregating lethal mutation would change the ratios.With the exception of hcf7, where the yellow phenotype appeared to be linked to the hcflocus, no other visible morphological differences were observed between plants whichsegregated for the hcf phenotype and those that did not in the lines that were characterized.Arabidopsis hcf lines that do contain additional morphologial phenotypes co-segregating withthe hcf phenotype have been isolated, and will have to be backcrossed the the wild-typeparent prior to analysis. Complementation tests have not been done to verify that all effects144are non-allelic. Since the the sample ofArabidopsis hcf mutants analyzed in this study wassmall, and the phenotype of each is quite different, it is probable that each is affected atdifferent locus. However, this cannot be stated with certainty, and future analysis of theseand additional mutants should verify that this is indeed the case. Genetic mapping of thehcf loci to particular chromosomes should eliminate the need for some crosses.4.2. PHYSICAL CHARACTERISTICS AND PIGMENT CONTENT OF HCF MUTANTLINESHomozygous hcf plants on i12 MSO plates could be distinguished visually from thewild-type due to pigmentation differences. The only exception was hcf8 (see below), whichdeveloped the hcf phenotype after leaf expansion. The difference in pigmentation could becorrelated with a general decrease in all the pigments in the mutants. Pigment analysisestablished that all the major pigments in wild-type Arabidopsis are also found in themutants. B-carotene was the carotenoid most severely affected in all the mutants. Since B-carotene is the biosynthetic precursor of all the xanthophylls detected in this study; [ie.antheraxanthin, zeaxanthin, violaxanthin, and neoxanthin (Sandmann, 1991)], all of whichwere found in the mutants, pigment biosynthesis per se does not appear to be impaired. B-carotene itself has been primarily associated with the reaction centres of PSI and P811 and isonly a small component of the peripheral light-harvesting antenna, whereas thexanthophylls are primarily associated with peripheral light harvesting complexes(Siefermann-Harms, 1985). Since the peripheral light harvesting complexes appeared to beat normal levels, the low level of B-carotene is presumed to be due to loss of the reactioncenter complexes.145The xanthophylls, violaxanthin and neoxanthin are normally at a 1:1 ratio in thewildtype (Rock and Zeevart, 1991; this work, see table 8). The same ratio, with minor -deviations was observed in all the mutants except hcf6 where violaxanthin was twice thelevel of neoxanthin. It is unknown if the highvalue of violaxanthin is significant for hc/6, orsimply due to random deviations. Violaxanthin is a carotenoid in the so called xanthophyllcycle where it undergoes de-epoxidation and is converted to zeaxanthin in high light. Thelatter carotenoid has been found to be associated with dissipation of excess excitation energyin PSII and is postulated to be involved in photoprotection (Demmig-Adams et al., 1989).The roles of violaxanthin and neoxanthin in the thylakoid membranes are not as clear.Mutants ofArabidopsis have been isolated that are blocked in one or both of the epoxidationsteps from zeaxanthin to violaxanthin and almost completely lacking violaxanthin andneoxanthin (Duckham et a1 1991; Rock and Zeevart, 1991). Although color differences can beobserved in these mutants (Koornneef et al 1982), they do not appear to bephotosynthetically impaired as they are autotrophic. However, studies specifically dealingwith photosynthetic capacity have not been done.Unlike the other mutant lines in this study, the hcf8 mutant and wild-type seedlingsare indistinguishable in the seedling stage. The hcf phenotype appears gradually and can beobserved only in the older tissues. The light and temperature treatments tested did notchange the timing of the expression of the hcf phenotype. The fact that seedlings in soilexpress the phenotype shortly after germination suggests that conditions on 112 MSO platesretard the development of the hcf phenotype. A possible explanation is that a nutrient in themedium, possibly sucrose, is able to retard the expression of the hcf phenotype. Arabidopsismutants requiring specific nutrients have previously been identified (Last et al., 1989;Schnieder et al., 1989). Sucrose, for example,, is known to repress the expression of nuclear146genes involved in photosynthesis (Sheen, 1990). However, any model involving therepression of the HCF8 gene product by the addition of a component to the media, must alsotake into account the fact that the new growth appears normal with respect to thefluorescence and color. A second possibility is that the wild-type product is involvedspecifically in PSI stability or repair. Heterotrophic growth may retard the requirement forthe HCF8 gene product in the developing tissues, whereas plants grown in soil do not containsufficient reserves. Growing Arabidopsis lines in both soil and sucrose supplementedmedium, has allowed for the isolation of this novel phenotype.4.3. FLUORESCENCE, ELECTRON TRANSPORT, CHL-PROTEINS ANDPOLYPEPTIDESThe fluorescence kinetics observed in the mutants can be correlated to a block inelectron transport using the data from previous mutants analysed in Chiamydomonas andmaize (Chua and Bennoun, 1976; Miles 1980). In this study the mutants were divided intoPSI mutants (hcfl, hcf4, hcf7 and hcf8) and P511 mutants (hct2, hcf3, hcf5 and hcf6) (seesummary in Table 13) based on the fluorescence kinetics data, photosynthetic electrontransport, the presence or abscence of specific Chi-proteins and polypeptides. Each will bediscussed separately below.4.3.1. PSI Mutants VMutants blocked in electron transport at PSI have a normal induction to themaximal level of fluorescence (FM), but there is no decline to a steady state fluorescence(Fe). This indicates that charge separation and electron flow in P511 to plastoquinone isoccurring, but since there is no decrease from the maximal fluorescence level (FM) in theV147Table 13. Summary of data for the hcf mutants and wild-type siblings.V hcfl hcf2 hc/3 hcf4 hcf5 hcf hct7 Vhq8Electron transporta:PSII 100 97 31 40 79 <5 <5 100 100PSI 100 25 44 80 97 73 127 70 41ChI-protein complexes1:cPI + + ++ + + + ++ +CP47 +CP43 +++ +++ + ++ - -cPII* + Cpu +++ +1-+ +++ +++ •4-++ +++Thylakoid Membrane Polypeptidesc:CPI +++ +1- ++++ nd + +1- + nd +PSI sub. II +++ +1- +++ nd ++ +1- + nd +cytf +++ +++ +P1 nd +++ + + nd +++PSII CP47 +++ +++ + nd +++ + nd nd +++a Percent of wild-type electron transport activity rates.DPC to DCPIP only.From Table 9. PSII rates are fromb Chi-protein complexes estimated visually on non-denaturing gels. CPI: PSI Chi a bindingcore complex; CP47 and CP43: P511 ChI a binding core complex polypeptides; CPII andCPII: the oligomeric and monomeric forms, respectively, of the Chi z÷b light harvestingcomplex polypeptides.c Representative sample of polypeptides associated with PSI, cytochrome b6/fand PSIIcomplexes. CPI: 80-82 kDa PSI Chl a binding polypeptides; Sub. II, PSI nuclear psaD geneproduct; Cytf, cytochrome fpolypeptide; CP47, PSII 47 kDa Chl a binding polypeptide.Relative levels estimated visually from immunoblots. nd = not determined.d A heme staining protein that was observed at lower levels and migrated abnonnally onnon-denaturing gels in the hcf2 lane is presumed to be the cytochrome f apoprotein.148mutant plants, it suggests that QA is not being reoxidized. This would result if electron flowis blocked at some point past QB such as cytochrome b6/f, or PSI. Hcf mutants of barley,Chiamydomonas and maize which are defective in PSI have fluorescence kinetics similar tothose observed for hcfl, hcf4, hc/7 and hcf8 described herein (Chua and Bennoun 1975; Cookand Miles 1989; Miles 1980; Simpson and von Wettstein 1980).It is interesting to note the lack of O-I-D-P kinetics in the mutants disrupted in PSIphotosynthetic activity (eg. Figure 25). Using the low light levels on the PAM fluorometerthese early events in the fluorescence induction in the wild-type were very clearly resolved.Thus, while electron flow through PSII is occurring, the induction kinetics of dark adaptedplants is not completely normal. Several possibilities may account for the lack of the dipduring the induction. One is that there is a high proportion of non-reducing PSII centers (ie.P511 complexes lacking or defective at the QB site). PSH activity rates, and the level of P511Chi-protein complexes and polypeptides in the mutants were similar to the wild-type.Therefore it does not appear that PSII per Se, is affected. However, since there was asignificant decrease of Chl associated with PSI in the mutants (hcfl, hcf8, hcf4 and hcfl inorder of severity) an increase in PS11 activity rates would be expected when the rates arebased on equal amounts of Clii. This was not observed. PSII activity rates were identicalbetween each of the mutants and their wild-type siblings. It is possible that due to adisrupted PSI complex, many of the increased PSII centers are inactive. Heterogeneity ofthe P511 complexes is well documented (Govindjee 1990; Melis, 1985). The origin of inactivePSII centers is still uncertain, although Melis et al. (1988) suggest that these consist ofsmaller PSIII3 complexes which are in the process of being assembled or repaired. In themutant, impaired electron flow through PSI over an extended period (four weeks on 1/2 MSOplates in this case), could result in damage to the PSII complexes.149A second possibility is that the induction kinetics observed is due to plastoquinonenot being fully re-oxidized in the mutants prior to fluorescence measurements. Chow et al.(1991) argue that differences in the dark adaptation period prior to taking measurements isone of the primary reasons differences in postulated PSII populations are observed bydifferent authors. However, the present analysis does not provide for an accurate estimate offunctional versus non-functional PSII centers (Govindgee, 1990), and further experimentsare necessary.Another possibility is that the dip is due to oxidation of the plastoquinone pool by PSI(Hansen et al., 1991). Since the mutants all have impaired PSI activity no dip would beexpected during induction according to this view. PSI mutants such as these and of barley,maize and Chiamydomonas should confirm these results. Based on the results of this study,it seems logical that PSI oxidation of the plastoquinone pool is responsible for the dipobserved during induction of dark adapted plants.PSI activity rates confirmed the fluorescence kinetics of the mutants and is inaccordance with the observation that hcfl, followed by hcf8, hcf4 and hcfl have the lowestlevel of PSI associated Chl-protein complexes and PSI polypeptides. However, in all cases,the electron transport rate appeared to be higher than could be accounted for based on thelevel of observed PSI polypeptides in the membrane. This was especially true in the case ofhcf4, where no differences were observed between the mutant and wild-type electrontransport activity rate, and hcf6 where the rate was higher than for the wild-type (seebelow).PSI electron transport activity rates have been found to be correlated with CPI andP700 (Cook and Miles, 1990; Miles, 1980). Although this has been documented in mostcharacterized PSI deficient mutants, the correlation has been lacking in several instances.150Pea chiorina mutant (#5) was categorized as a PSI mutant due to lack of detectable P700, yetit was found to have 50% of PSI electron transport activity (Roshchina et al., 1983). Maizehcf mutants, hcf-2 and hcf-38 have PSI proteins which are reduced 2 to 3-fold, but no adverseeffects on electron transport were observed (Barkan et al., 1986). Mutant hcf- 101 wasreported to have only 15-20% of wild-type CPI apo-protein and P700 levels, yet it retained70% of electron transport activity rates (Cook and Miles, 1990). One possibility for thehigher electron transport rates in the mutants, postulated by Cook and Miles (1990), is thatthe lower relative amounts of PSI centers allows for more efficient use of the availableelectron donors and acceptors in the assay. However, the electron donors (ascorbateiTMEZ)and acceptor (methyl viologen) are not limiting factors in the PSI assay, and the wild-typeactivity rate should reflect the photosynthetic capacity or maximum under the assayconditions.Another possibility is that photooxidation of some unstable component is occurring inthe mutants or there are alternate pathways. PSI hyperactivity has been observed indeveloping immature chloroplasts of barley (Ohashi et al., 1989), Chlamydomonas (Bar-Nunet al., 1977) and cucumber (Daniell et al., 1985). In each instance, this phenomenon wasobserved in plant material in which the electron transport system was still developing, andthe Chi concentration was very low. This correlates with the fact that the Chi concentrationis significantly reduced in all the mutants studied. In addition, the electron transportsystem could be thought of as still in the process of ‘developing’ in the mutants, since themutations disrupt the assembly and/or stability of the PSI complex.In the case of hcf4, where wild-type electron transport rates were observed in spite ofthe apparent lower level of PSI polypeptides, it is also possible that the mutant has lost asoluble component of the electron transport chain. One such possible component is151ferrodoxin. If such were the case, it would lead to disruption of electron transport in vivogiving a similar fluorescence kinetics observed of PSI mutants, yet electron transport activityrates as measured by TMPD/ascorbate to methyl viologen would be unaffected. Furtheranalysis of this mutant will be required to determine where the block in electron transport isoccurring.4.3.2. PSII MutantsAlthough this section details mutants that are disrupted in PSII based on thefluorescence kinetics, both PSII and PSI were affected in these mutants. In fact, one of themore intriguing results from this study was that the PSI band on non-denaturing gels (CPI)was decreased in all of the mutants. There was no mutant found in this initial screen whereonly P511 was affected. This result is surprising because the PSI core Chl, P700, is one of themost stable components in the photosynthetic electron transport chain (Golbeck and Bryant,1991). The observed decrease in PSI in the mutants primarily affected at the PSII complexis not easily understood. However, the majority of maize hcf mutants appear to have apleiotropic effect on both PSI and PSII complexes (Miles, 1980; Barkan et al., 1986), and suchpleiotropic effects on both PSI and PSII may also be common in Arabidopsis.The fluorescence kinetics of mutations which affect the reducing side of PSII isexemplified by a high initial fluorescence (F’0), and little variable fluorescence (F). A highF0 level and no induction suggests that there is no electron flow through P511 due to theinability to use absorbed light energy in electron transport. Similar fluorescence kinetics hasbeen observed in P511 hcf mutants of Chiamydomonas and maize (Chua et al., 1975; Letoand Miles,1980; Rochaix and Erickson, 1988). Three mutants (hcf2, hefi) and hcI) thatappeared to have a PSII type mutant fluorescence kinetics were found upon closer inspection152to have some degree of variable fluorescence. The fluorescence correlated with electrontransport rates, which suggested that while the PSH complex was decreased, it was alwayspresent at some level (Table 9).For mutants hcf2 and hcf3 the fluorescence induction kinetics suggest that while thePSH complex is disturbed, electron transport is occurring. The induction kinetics for thePSII centers present appears to be normal as observed in the small rise from F0 to FM.However, at least 50% of the variable fluorescence observed in the mutant hcf2 is below theinitial F0. Fluorescence quenching is known to be due to two factors, namely, photochemicalenergy conversion used in CO2 reduction (qp) and non-photochemical quenching (q). Nonphot,ochemincal quenching is due to: a) energization across the thylakoid membrane due tothe formation of a proton gradient (q), b) changes in the PSII LHCII connected centers (qT),and c) quenching due to photoinhibition (qj) (Krause and Wiess, 1991). Although the relativecontribution of each of the factors involved in fluorescence quenching in normal tissues isstill under debate, in the mutant hcf2, it appears that the quenching mechanisms are normalas q is the same as in the wild-type.For hcf3, in spite of the increase in the F0 level, and a lower level of P511photosynthetic electron transport activity rate, the variable fluorescence appears normal.This type of fluorescence kinetics has been correlated with mutations that affect CO2fixation in maize (Edwards et al., 1988). Further investigations into the cause of the mutantphenotype of hc/3 should include CO2 fixation.The fluorescence induction kinetics of mutant hcffl suggests that the hcf phenotype isdue to a disrupted PSII complex. Although very little variable fluorescence was observed,there was some induction from F0 to FM. There was also a significant dip between F1 andFM. According to the theory of Hansen et aL, (1991), the dip is caused by oxidation of the153plastoquinone pool (see above). Although the levels of PSI ChI-protein complexes and PSIpolypeptides appeared to be reduced in the mutant, PSI photosynthetic electron transportrate was greater than for the wild-type. While the cause of the high electron transport rateis unknown, and the differences are difficult to reconcile, the dip observed during inductiontends to support the possibility that the results observed for the electron transport rates arereal. Possible reasons for the high electron transport activity have been presented above.Mutant hcf6 offers the opportunity to address questions on development of the chloroplastphotosynthetic electron transport, in particular PSI electron transport.An unresoved discrepancy found on western blots was that the CPI band of mutanthc/2 immunostained at the same level as the wild-type. In mutations primarily affecting thePSI complex (hcf7 and hcf), there was a correlation between the level of PSI Chi-proteinfound when assayed under non-denaturing gel electrophoresis and the level of PSIpolypeptides observed by immunoblotting (discussed above). The CPI antibody used was thesame for all experiments. For hcf2, in spite of the apparent lower level of CPI found undernon-denaturing conditions, and the lower level of the other PSI polypeptides (Figures 16 and18) the CPI band was found at wild-type levels. This suggests that CPI is present at normallevels in the mutant, but binding chlorophyll is not.Another result that merits further study was the observation that only one of thecoupling factor polypeptides was detected in the mutant hcf5. In higher plants, the missingband would normally correspond to the beta subunit (Westhoff et al., 1985; Lemaire andWoliman, 1989a). This result is intriguing in light of the fact that the mRNA of the mutanthcf5 beta subunit mRNA (atpB) appeared to be normal, whereas, the alpha subunit mRNA(atpA) did not appear to be processed normally (Figure 10). Mutations which affect onecomponent of the coupling factor in Chiamydomonas have shown to result in the loss of all154coupling factor polypeptides (Lemaire and Wofiman, 1990a; 1990b Robertson et al., 1989). Inaddition, mutations which affect the stability of the beta subunit tend to disrupt theassembly of CF1 in Chiamydomonas and suggesting that the beta subunit is required priorto assembly and stable integration of the coupling factor into the thylakoid membranes(Robertson et aL, 1989). It is possible that the mobilities of the beta and the alpha subunitsare switched on the Schagger and von Jagow (1987) gel system. It is also possible that theimmunostained band in Figure 9 is due to another protein which crossreacts with the CF1antiserum. A 70 kI)a thylakoid membrane-associated nucleotide-binding protein wasobserved to crossreact with CF1 beta subunit antiserum in Chiamydornonas (Lemaire andWoliman, 1989a; Robertson et al., 1989). A corresponding protein has not been reported inhigher plants, but such a candidate is the 64 kfla thylakoid membrane bound protein kinasefound to be associated with the cytochromeb6/fcomplex (Coughian and Hind, 1986). InChiamydomonas the crossreacting protein migrates slightly higher than CF1 polypeptides ongels and, if in higher plants a similar protein co-migrates with the CF1 alpha subunit, it mayhave remained undetected.One of the most striking phenotypic trait was the drastic alteration of thylakoidmembrane ultrastructure observed in the hcf5 and hcf6 mutant lines (Figure 8). For hc1almost all the thylakoids were appressed, although a few single thylakoids were observedaround the periphery of the thylakoid stacks. This morphology is consistent with theobservation that LHCII accounts for most of the ChI in the cell, since LHCII is believed to bethe major factor involved in thylakoid appression (Anderson, 1982; 1989). For hcft3, the lowlevels of thylakoid membranes is consistent with the fact the there is a very low level ofoverall Chi in the mutant. The mutant chioroplasts, in this instance, do not appear to befully developed. It is possible that for these mutants that the altered electron transport is a155consequence of a more general mutation, for example, lipid biosynthesis. This is unlikely,however, for several reasons. One, numerous lipid biosynthetic mutants have beencharacterized, and none appear to result in photosynthetic electron transport abnormalities(Browse and Somerville, 1991). A second reason that a lipid biosynthesis mutation isunlikely to be the major cause of the hcf phenotype is that both of the mutants appear tohave altered chioroplast mRNA levels (see below) which would be expected to result inchioroplast protein deficiencies. A third reason, is that while many of the thylakoidmembrane proteins are missing or at lower levels, the nuclear encoded LHCII Chia Ibcomplexes are present in the membrane.4.4. CHLOROPLASTRNAThree of the six mutants analysed at the transcriptional level were found to containaberrant chioroplast mRNAs. Since the majority of the Chiamydomonas and maize hefmutants that have been analysed have normal mRNA (Barkan et al., 1986; Jensen et al.,1986), these results were unexpected.Both hcf5 and hcf6 had similar phenotypes with regard to defects in accumulationand processing abnormalities of chloroplast transcripts. For hcf5, overall levels ofchloroplast transcription were apparently not affected, suggesting that the differences in thelevels of specific transcripts are due to real differences in the steady state levels of thechioroplast mRNA populations. In most respects, the loss of thykakoid complexes correlatedwith aberrant accumulation of corresponding mRNA’s. The deficiency in PSII andcytochrome f could be the result of abnormal processing of the psbB-psbH-petB-petD operonas well as the psbD transcript. The atpA trancript is also abnormal and may explain thepresence of an aberrant coupling factor in the thylakoid membrane. Although a band156corfesponding to the alpha polypeptide of the coupling factor was observed in the thylakoidmembranes when the CF1 antiserum was used, it may be that the immunostained bandresu1ted from a cross reacting polypeptide (discussed above). It is not as clear why PSIproteins are depleted in mutant hc. Both the psaA and psaC transcripts, encoding the twopolypeptides of CPI which carry the reaction center chromophores of PSI, and the PSI iron-sulfur binding polypeptide transcripts were normal. It is possible that the processing ofchloroplast polycistronic messages that include the PSI-I or PSI-J polypeptides is not normal,although it must be pointed out that the functional role of these polypeptides in PSI has notbeen established.The pleiotropic effect of the hcf5 mutation on several steady state levels ofchioroplast mRNAs suggests that the wild-type gene product may be a general proteinrequired for chloroplast mRNA stability and/or translation. Stern et al. (1989) have reportedon stabilizing proteins associated with the 3’ ends of chioroplast transcripts. In proteinbinding competition studies they observed both common and gene-specific proteins. It is alsopossible that stabilizing proteins would bind to the 5’ end. Chioroplast transcripts differingat the 5’ end have been documented (Gruissem, 1989). While direct evidence for 5’chloroplast mRNA stabilizing proteins has not been documented, stabilizing proteins thatbind to the 5’ end of mitochondrial transcripts that are required for stability in yeast havebeen identified (Dieckmann et al., 1984). Since the levels of unprocessed and processedintermediates differ depending on the gene assayed (compare psbB and petB with atpA andpsbD in Figure 10), the affinity (stabilizing effect) of the putative HCF5 gene product forparticular transcripts would probably be further determined by transcript specific factors.Another possible role of the wild-type HCF5 gene product may be in translation.Although no chioroplast specific translation factors have been isolated, nuclear mutations157that disrupt the translation of mitochodrial transcripts have been identified in yeast (Grivell,1989). All characterized nuclear mutations affecting translation characterized, have beenrecessive. This suggests that the missing product is a positive acting element. In theseyeast mutants, stable but untranslatable mRNAs accumulated (Fox, 1986). In the mutanthcf5 there were no ‘new’ transcripts observed, only an increase in the steady state levels ofsome processed intermediates and a decrease in others compared to the primary transcripts.Barkan (1988) reported that processing does not appear to be a prerequisite for translationin the chloroplast. AU transcripts derived from the psbB-psbH-petB-petD transcriptionalunit, for example, were found to be polysome bound and translated.. The rate of RNAprocessing tends to be slow relative to the rate of transcription, allowing for theaccumulation of unprocessed transcripts observed on northern blots (Rock et aL, 1987;Westhoff and Herrmann, 1988). Thus, if a translational block is occurring in the mutantline, the transcript population would tend to accumulate in the processed form, therebyleading to an under representation of the unprocessed transcripts.A third possibility is that the HCF5 gene product may function in both RNA stabilityand translation. The yeast nuclear CPB1 gene product was originally described as amitochodrial 5’ RNA stabilizing protein (Diekman and Mittelmeier, 1987) and hassubsequently been found to resemble prokaryotic and eukaryotic translation initiationfactors (Grivell 1989). Other yeast mitochondrial directed nuclear gene products have alsobeen found to be involved in both RNA processing and translation (Grivell (1989).The hc/ mutant phenotype, while similar to hcf5, is more extreme as the overalllevel of chloroplast transcripts is reduced. However, some of the differences observed, maybe due to a lower transcription rate in the chloroplast of the mutant. One difference wasnoted between hcf6 and hcf5 when the psaB gene (encoding PSI subunit II) was used as a158probe. There appear to be two high molecular weight transcripts present in the mutant thatare not observed in the wild-type, nor in any other mutant. While the possibility that thebands are due to DNA contamination cannot be ruled out, no corresponding band wasobserved in any of the other RNA gel blots either in the mutant hc/6 lane nor in any other ofthe mutant and wild-type lanes. If the bands are in fact RNA, a readthrough transcriptioninto the psaB gene from another 5’chloroplast gene is possible, and/or the 3’ end of thetranscript is not processed properly.Another transcriptional abnormality was observed in the expression of the petAchioroplast transcript in the mutant hct2. Although the majority of chioroplast transcripts inthis mutant were at a lower level, due to an apparently slightly lower level of steady statechloroplast transcripts, the petA transcript was observed to be significantly higher than thewild-type. This is in agreement with the observation that the cytochrome fprotein migratedabnormally when run on non-denaturing gels. The significance or role of the observed highsteady state level of the petA gene is not presently known. The petA gene is part of apolycistronic message also encoding a PSI 4 kDa polypeptide (psal gene) and two additionalopen reading frames from which the gene products have not been determined. One of theopen reading frames designated zfpA (Sazaki et al., 1989), potentially encodes a zinc fingerprotein. Smith et al. (1991) propose that this may be a protein involved in Ci metabolism,namely propionyl CoA carboxylase. This conclusion is based on comparisons of the sameenzyme from mitochondrial sources, leading the authors to postulate that the conservedcysteine amino acids might form a metal ion binding site involved in catalysis. The secondconserved open reading frame part of the petA transcript encodes a 229-23 1 amino acidpolypeptide. Sequence analysis of this open reading frame suggests that it may be a heme159binding polypeptide (Wiley and Gray, 1990). The role and location of this putative hemebinding polypeptide in the chioroplast remains to be determined.One possible explanation for the mutant phenotype is that the mutation leads to asingular overexpression of the petA region at the expense of other transcriptional units in thechioroplast. The steady state levels of other chloroplast transcripts including the genes forribosomal RNA appeared to be reduced in the mutant. However, all the processedintermediates associated with each transcript were present, and the protein productsassociated with various transcripts were present in the thylakoid membrane. If the hcf2mutation leads to an increase in transcription of the petA region, it would be due to a loss offunction of a nuclear factor that is responsible for negative regulation of this operon.Another possibility for the increased levels observed for the petA transcript may be ablock in processing or translation. The increased levels would be due to the lack ofdegradation of the transcript, or an increased transcription in an attempt to compensate inthe mutant plants. However, as pointed out above, mutants in yeast blocked at atranslational step are normal with respect to mRNA levels. Since the petA protein product(cytochrome I) is detected in the thylakoid membrane fraction at similar levels as the petBprotein product (cytochrome b6), this suggests that translation per se is not blocked. The roleof the additional open reading frames from the petA transcriptional unit have not beendetermined, therefore it is not possible to evaluate what effect a block in translation of thesegenes may have. Future work should include DNA sequences 5’ to the petA gene to verifythat all the unprocessed and processed transcripts from this operon are affected.The functional significance of the processing of the chioroplast polycistronic messagesis not known (Barkan, 1989; Gruissem, 1989), however, nuclear genes are undoubtedlyinvolved. Mutant analysis is one method for dissection of this complex nucleo-chloroplast160interaction. Although the primary effect of the mutant lesion in mutants hcf5, hcf6 and hct2appears to be involved in chloroplast mRNA stability or transcription, clearly this questioncan only be addressed after isolation of the gene. Arabidopsis offers the possibility of geneisolation by chromosome walking and analysis of the gene function through transformation.Future work on this and other mutants should allow us to begin to understand nuclearfactors which are involved in chioroplast biogenesis and electron transport.4.5. GENETIC ANALYSIS OF REGENERATION AND TRANSFORMATION OF WILD-TYPE ArabidopsisThe previous sections have dealt with the characterization of the hcf phenotype inArabidopsis. Studies involving the analysis of genes isolated from mutant phenotypes, bethey hcf or others, will have to be evaluated in the context of the mutant phenotype. Thisinvolves transformation of the mutant Arabidopsis lines with A.grobacterium containing thegene of interest. While it is presently technically feasible to introduce foreign genes intoArabidopsis by transformation, the procedures for obtaining regenerated seed bearing plantsare not on par with systems for tobacco and tomato. In addition, the transformationefficiency of the wild-type Columbia ecotype, used in this study, is one of the lowest inArabidopsis regardless of the protocol used.Previous studies on Arabidopsis regeneration have shown considerable ecotypicvariation in response to different tissue culture regimes. In order to isolate an ecotype thathad a high regenerative potetial, several wild-type ecotypes were tested on the same tissueculture regime. One ecotype which has shown promise is a wild-type isolated on theUniversity of British Columbia campus, named Point Grey. The ability of explants of thePoint Grey ecotype to form roots and shoots simultaneously facilitates transfer from sterile161medium into soil. No visible morphological differences have been observed in progeny of over100 regenerated plants from Point Grey cotyledon explants. It was surprising that by usingcotyledon explants infected with Agrobacterium, the rapid regeneration appears to be adetriment for regeneration of transformed plants. However, by using root explants, thePoint Grey ecotype regenerates seed bearing plants fairly efficiently.The different patterns of inheritance on the different shoot inducing media (SIM1and SIM2) are not unexpected. If there is a major gene that affects regeneration, it wouldnot be surprising if the segregation could only be demonstrated on a specific medium,because there is ample evidence that gene expression is highly environmentally-dependant.Two advantages for the use ofArabidopsis in genetic and physiological studies are itsshort generation time and small size. These same features should allow for genetic analysisof the regeneration response in culture. Previous studies crossing ecotypes with high andlow regenerative potentials have shown that inheritance appears to be under the control of asingle gene (Chaudbury and Signer, 1989). However, the results suggest such aninterpretation can only be made in the context of the tissue culture regime used. It is likelythat the response differences between media are due to interactions between genotypes andthe concentrations of hormones used in the media. It is worth noting that mutants ofArabidopsis have been isolated that are deficient in particular hormones, or insensitive toexogenous hormone applications (Klee and Estelle, 1991). However, measurements of theendogenous hormone levels in different wild-type ecotypes has not been undertaken.Previous reports using tomato and cucumber suggest that regeneration response inculture is controlled by few genes, and genetic effects were observed to be additive(Frankenberger et al,. 1981; Koornneef et al,. 1987; Nadolska-Orczyk and Malepszy, 1989).Although in the present study the results on shoot regeneration are equivocal, the data on162root regeneration show a reasonably good fit to Mendelian monogenic inheritance, withefficient regeneration dominant. The least convincing ratios are those of the selfed F2plants, but even here the overall ratio for the F3 generation is close to the expected value.The results suggest that shoot formation is more likely on rooted explants than onunrooted ones. Therefore, it is possible there is a single major genetic determinant in PointGrey that is responsible for both rooting and shooting. Since shooting is generally lessfrequent than rooting, the penetrance of the genetic determinant may simply be lower inshoots. An alternative explanation is that shoot formation is facilitated by roots; thusrooting would be seen as the primary facilitating event for regeneration from tissue culturein the Point Grey system.Since concordance of the cotyledon pairs for rooting on SIM2 is so high in theparental Point Grey and F1 generations, the poor concordance in the F2 and F3 must beaccounted for on a formal genetic level by incomplete penetrance. One major differencebetween parental Point Grey and F1 on the one hand, and F2 and F3 on the other hand, isthat the former are genetically uniform, whereas the latter are expected to show variancethrough genetic segregation. [f the full penetrance of the regeneration genotype in theparent and F1 is caused by the expression of modifiers that strengthen the expression of amajor gene, then the segregation of these modifiers in the F2 and F3 is expected and couldaccount for the incomplete penetrance in these generations. One possibility is that in theparent and F1 the overall genotype excedes some kind of expression threshold, whereas inthe F2 and F3 the threshold is sometimes not met genetically, but can be exceded by theaction of environmental factors. The threshold model also explains the variability in the F3sibships; the threshold is sometimes not exceded even though the desired major gene ispresent. Further exploration of this genetic system will require several generations of163backcrosses to parental lines in order to achieve a level of uniformity in the geneticbackground against which clearer ratios can be seen.164REFERENCESAllen JF, Holmes NG (1986) Electron transport and redox titration.. In: M.F. Hipkins andN.R. Barber (Eds). Photosynthesis, Energy Transduction: A Practical Approach, IRLPress, Oxford, England.Alt J, Herrmann RG (1984) Nucleotide sequence of the gene for pre-apocytochrome fin thespinach plastid chromosome. Curr. Genet. 8: 551-557.Anderson, JM (1982) The significance of grana stacking in chlorophyll b-containingchloroplasts. Photobiochem. Photobiophys. 3: 225-241.Anderson JM (1989) The grana margins of plant thylakoid membranes. Physiol. Plant. 76:243-248.Andreasson L-E, Vanngard T (1988) Electron transport in photosystems I and II. Annu.Rev. Plant Physiol. Plant Mol. Biol. 39: 379-411.Arnon DI (1949) Copper enzymes in isolated chloroplasts. Polyphenol-oxidase in Betavulgaris. Plant Physiol. 15: 1-15.Arnon DI, Tang GM-S (1988) Cytochrome b-559 and proton conductance in oxygenicphotosynthesis. Proc. Nati. Acad. Sci. USA 85: 9524-9528.Artus NN, Somerville C (1988) A mutant ofArabidopsis thaliana that exhibits chlorosis inair but not in atmosphere enriched in 002. Plant Physiol. 87: 83-88.Askins RA, Lambowitz AM (1987) A protein required for splicing group I introns inNeurospora mitochondria is mitochodrial tyrosyl-tRNA synthetase or a derivativethereof. Cell 50: 331-345.Baker B, Coupland G, Federoff N Starlinger P Schell J (1987) Phenotypic assay for excisionof the maize controlling elementAc in tobacco. EMBO J. 6: 1547-1554.Bar-Nun 5, Schantz R, Ohad I (1977) Appearance and composition of chlorophyll-proteincompexes I and II during chioroplast membrane biogenesis in Chamydomonasreinhardi y-1. Biochem. Biophys. Acta 459: 451-467.Barkan A (1988) Proteins encoded by a complex chioroplast transcription unit are eachtranscribed from both monocistronic and polycistronic mRNAs. EMBO J. 7: 2637-2644.Barkan A (1989) Tissue-dependent plastid RNA splicing in maize: Transcripts from fourplastid genes are predominantly unspliced in leaf meristems and roots. Plant Cell 1:437-445.165Barkan A, Martienssen RA (1991) Inactivation of maize transposon Mu suppresses amutant phenotype by activating an outward-reading promoter near the end ofMul.Proc. Natl. Acad. Sci. USA 88: 3502-3506.Barkan A, Miles D, Taylor WC (1986) Chloroplast gene expression in nuclear,photosynthetic mutants in maize. EMBO J. 5: 1421-1427.Bengis C, Nelson N (1975) Purification and properties of the photosystem I reaction centerfrom chloroplasts. J. Biol. Chem. 250: 2783-2788.Bengis C, Nelson N (1977) Subunit structure of chioroplast photosystem I reaction center.J. Biol. Chem. 252: 4564-4569.Bennett J, Shaw EK, Michel H (1988) Cytochrome b6fcomplex is required forphosphorylation of light-harvesting chlorophyll a/b complex II in chioroplastphotosynthetic membranes. Eur. J. Biochem. 171: 95- 100.Bennoun P, Levine RP (1967) Detecting mutants that have impaired photosynthesis bytheir increased level of fluorescence. Plant Physiol. 42: 1284-1287.Blowers AD, Ellmore GS, Klein U, Bogorad L (1990) Transcriptional analysis of endogenousand foreign genes in chioroplast transformants of Chiamydomonas. Plant Cell 2:1059-1070.Bousquet I, Dujardin G, Poyton RO, Slonimski PP (1990) Two group I mitochondrial intronsin the cob-box and coxi genes require the same MRSI /PET157 nuclear gene productfor splicing. Curr. Genet. 18: 117-124.Bradbeer JW, Atkinson YE, Borner T, Hagemann R (1979) Cytoplasmic synthesys of plastidpolypeptides may be controlled by plastid-synthesized RNA. Nature 279: 816-817.Breton J (1982) Hypothesis- the F695 fluorescence of chioroplasts at low temperature isemitted from the primary acceptor in photosystem II. FEBS Lett. 147: 16-20.Brown JS, Anderson: JM, Grimme LH (1982) Antenna chlorophyll a complexes in mutantand developing barley. Photosynth. Res. 3: 279-29 1.Browse J, Somerville C (1991) Glycerolipid biosynthesis: biochemistry and regulation.Annu. Rev. Plant Physiol. Plant Mol. Biol. 42: 467-506.Bruce BD, Malkin R (1988) Subunit stoichiometry of the chloroplast photosystem I complex.J. Biol. Chem. 263: 7302-7306.Bruce BC, Malkin R (1991) Biosynthesis of the chloroplast cytochrome b6fcomplex: Studiesin a photosynthetic mutant ofLemma. Plant Cell 3: 203-212.Burgess DG, Taylor WC (1987) Chloroplast photooxidation affects the accumulation ofcytosolic mRNAs encoding chioroplast proteins in maize. Planta 170: 520-527.V166Camm EL, Green BR (1989) The chlorophyll ab complex, CP29, is associated with thephotosystem II reaction centre core. Biochem. Biophys. Acta 974: 180-184.Cao J, Govindjee (1990) Chlorophyll a fluorescence transient as an indicator of active andinactinve photosystem II in thylakoid membranes. Biochem. Biophys. Acta 1015:180-188.Chang C, Bowman JL, DeJohn AW, Lander ES, Meyerowit.z EM (1988) Restrictionfragment length polymorphism linkage map for Arabidopsis thaliana. Proc. Nati.Acad. Sci. USA 85: 6856-6860.Chaudhury AM, Signer ER (1989) Relative regeneration proficiency ofArabidopsis thalianaecotypes. Plant Cell Rep. 8: 368-369.Chen L-J, Rogers SA, Bennet DC, Hu M-C, Orozco EM Jr. (1990) An in vitro transcriptiontermination system to analyze chloroplast promoters: identification of multiplepromoters for the spinach atpB gene. Curr. Genet. 17: 55-64.Cherniack AD, Garriga G, Kittle JD, Askins RA, Lambowitz AM (1990) Function ofNeurospora mitochodrial tyrosyl-tRNA synthetase in RNA splicing requires anidiosyncratic domain not found in other synthetases. Cell 62: 745-755.Chia CP, DuesingJH, Arntzen CJ (1986) Developmental loss of photosystem II activity andstructure in a chloroplast-encoded tobacco mutant, Lutescens-1. Plant Physiol. 82:19-27.Chitnis PR, Purvis D, Nelson N (1991) Molecular cloning and targeted mutagenesis of thegene psaF encoding subunit HI of photosystem I from the cyanobacteriumSynechocystis sp. PCC 6803. J. Biol. Chem. 266: 20146-20151.Chitnis PR, Reilly PA, Nelson N (1989a) [nsertional inactivation of the gene encodingsubunit II of photosystem I from the cyanobacterium Synechocystis sp. PCC 6903. J.Biol. Chem. 264: 1838 1-18385.Chitnis PR, Reilly PA, Miedel MC, Nelson N (1989b) Structure and targeted mutagenesis ofthe gene encoding 8-kDa subunit of photosystem I from the cyanobacteriumSynechocystis sp. PCC 6803. J. Biol. Chem. 264: 18374-18380.Chory J, Pete C, Feinbaum R, Pratt L, Ausubel F (1989a) Arabidopsis thaliana mutant thatdevelops as a light-grown plant in the absence of light. Cell 58: 99 1-999.Chory J, Pete CA, Ashbaugh M, Saganich R, Pratt L, Ausubel F (1989b) Different roles forphytochrome in etiolated and green plants deduced from characterization ofArabidopsis thaliana mutants. Plant Cell 1:867-880.Chory J, Nagpal P, Pete CA (1991) Phenotypic and genetic analysis of det2, a new mutantthat affects light-regulated seedling development in Arabidopsis. Plant Cell 3: 445-459.167Chow WS, Anderson JM (1987) Photosynthetic responses of Pisum sativum to an increasein irradiance during growth. IL Thylakoid membrane components. Aust. J. PlantPhysioL 14: 9-19.Chow WS, Hope AB (1987) The stoichiometries of supramolecular complexes in thylakoidmembranes from spinach chioroplasts. Aust. J. Plant Physiol. 14: 21-28.Chow WS, Hope AB, Anderson JM (1991) Further studies on quantii’ing photosystem II invivo by flash-induced oxygen yield from leaf disks. Aust. J. Plant Physiol. 18: 397-410. VChua N-H, Matlin K, Bennoun P (1975) A chlorophyll-protein complex lacking inphotosystem I mutants of Chiamydomonas reinhardtii. J. Cell Biol. 67: 36 1-377.Chylla RA, Whitmarsh J (1989) Inactive photosystem II complexes in leaves. Plant Physiol.90: 765-772.Cleland RE, Melis A, Neale PJ (1986) Mechanism of photoinhibition: photochemicalreaction center inactivation in system II of chloroplasts. Photosynth. Res. 9: 79-88.Colbert JT (1988) Molecular biology of phytochrome. Plant Cell Environ. 11: 305-318.Cook WB, Miles D (1988) Transposon mutagenesis of nuclear photosynthetic genes in Zeamays. Photosynth. Res. 18: 33-59.Cook ‘WB, Miles P (1990) Anomalous electron transport activity in a photosystem I-deficientmaize mutant. Photosynth. Res. 24: 81-88.Coughian SJ (1988) Chloroplast thylakoid protein phosphorylation is influenced bymutations in the cytochrome bf complex. Biochem. Biophys. Acta 933: 413-422.Coughlan SJ, Hind G (1986) Purification and characterization of a membrane-boundprotein kinase from spinach thylakoids. J. Biol. Chem. 261: 11378-11385.Cramer WA, Theg SM, Widger WR (1986) On the structure and function of cytochromeb559. Photosynth. Res. 10: 393-403.Crossland LD, Rodermel SR, Bogorad L (1984) Single gene for the large subunit of ribulosebisphosphate carboxylase in maize yields two differently regulated mRNAs. Proc.Nati. Acad. Sci. USA 81: 4060-4064.Croxdale JG, Omasa K (1990) Patterns of chlorophyll fluorescence kinetics in relation togrowth and expansion in cucumber leaves. Plant Physiol. 93: 1083-1088.Damm B, Willmitzer L (1988) Regeneration of fertile plants from protoplasts of differentArabidopsis thaliana genotypes. Mol. Gen. Genet. 213: 15-20.V168Daniell H, Anbudurai PR, Periyannan S, Renganathan M, Bhardwaj R, Kulandeivelu G,Gnanam A (1985) Oxygenic photoreduction of methyl viologen and nicotinamideadenine dinucleotide phosphate without the involvement of photosystem I duringplastid development. Biochem. Biophys. Res. Commun. 26: 1114-1121.Decoster E, Simon M, Hatat D, Faye G (1990) The MSS5I gene product is required fortranslation of the COX1 mRNA in yeast mitochondria. Mol. Gen. Genet. 224: 111-118.Deisenhofer J, Epp 0, Miki K, Huber R, Michel H (1985) Structure of the protein subunitsin the photosynthetic reaction centre ofRhodopseudomonas viridis at 3A resolution.Nature 318: 618-624.Delepelaire P (1984) Partial characterization of the biosysnthesis and integration of thephotosystem II reaction centers in the thylakoid membrane of Chiamydomonasreinhardtii. EMBO J. 3: 70 1-706.Dellaert LMW (1980) Segregation frequencies of radiation-induced viable mutants inArabidopsis thaliana (L.) Heynh.. Theor. Appi. Genet. 57: 137-143.Demmig B, Björkman 0 (1987) Comparison of the effect of excessive light on chlorophyllfluorescence (77k) and photon yield of 02 evolution in leaves of higher plants. Planta171: 171-184.Demmig-Adams B, Adams WW, Heber U, Neimanis S, Winter K, Kruger A, Czygan F-C,Bilger W, Björkman 0 (1990) Inhibition of zeaxanthin formation and of rapidchanges in radiationless energy dissipation by dithiothreitol in spinach leaves andchioroplasts. Plant Physiol. 92: 293-301.Deng X-W, Gruissem W (1987) Control of plastid gene expression during development: thelimited role of transcriptional regulation. Cell 49: 379-387.Deng X-W, Gruissem W (1988) Constitutive transcription and regulation of gene expressionin non-photosynthetic plastids of higher plants. EMBO J. 7: 3301-3308.Deng X-W, Caspar T, Quail PH (1991) cop.!: a regulatory locus involved in light-controlleddevelopment and gene expression inArabidopsis. Genes & Dev. 5: 1172-1182.Dieckmann CL, Tzagoloff A (1985) Assembly of the mitochondrial membrane system. CPB6,a yeast nuclear gene necessary for synthesis of cytochrome b. J. Biol. Chem. 260:1513-1320.Dieckmann CL, Koerner TJ, Tzagoloff A (1984) Assembly of the mitochondrial membranesystem: CBP1, a yeast nuclear gene involved in 5’ end processing of cytochrome bpre-mRNA. J. BioL Chem. 259: 4722-4731.Dietz K-J, Bogorad L (1987) Plastid development in Pisum sativum leaves during greening.Plant Physiol. 85: 816-822.169Duckham SC, Linforth RST, Taylor lB (1991) Abscisic-acid-deficient mutants at the abagene locus ofArabidopsis thaliana are impaired in the epoxidation of zeaxanthin.Plant Cell Envinron. 14: 601-606.Dunn PPJ, Gray JC (1988) Localization and nucleotide sequence of the gene for the 8 kDasubunit of photosystem I in pea and wheat chioroplast DNA. Plant Molec. Biol. 11:311-320.Dunn PPJ, Packman LC, Pappin D, Gray JC (1988) N-terminal amino acid sequenceanalysis of the subunits of pea photosystem I. FEBS Lett. 228: 157-16 1.Edwards GE, Jenkins CLD, Andrews J (1988) CO2 assimilation and activities ofphotosynthetic enzymes in high chlorophyll fluorescence mutants of maize havinglow levels of ribulose-1,5-bisphosphate carboxylase. Plant Physiol. 86: 533-539.Eisermann A, Tiller K, Link G (1990) In vitro transcription and DNA bindingcharacteristics of chloroplast and etioplast extracts from mustard (Sinapsis alba)indicate differential usage of the psbA promoter. EMBO J. 11: 398 1-3987.Estelle MA, Somerville CR (1988) The mutants ofArabidopsis. Trend Genet. 4: 89-93.Evans MCW, Brendenkamp G (1990) The structure and function of the photosystem Ireaction center. Physiol. Plant. 79: 415-420.Feldmann KA (1991) T-DNA insertion mutagenesis inArabidopsis: mutational spectrum.PlantJ. 1: 71-82.Feldmann KA, Marks DM (1987) Agrobacteriurn-mediated transformation of germinatingseeds ofArabidopsis thaliana: A non-tissue culture approach. Mol. Gen. Genet. 208:1-9.Feldmann KA, Marks DM, Christianson ML, Quatrano RS (1989) A dwarf mutant ofArabidopsis generated by T-DNA insertion mutagenesis. Science 243: 135 1-1354.Fenton JM, Crofts AR (1990) Computer aided fluorescence imaging of photosyntheticsystems: Application of video imaging to the study of fluorescence induction in greenplants and photosynthetic bacteria. Photosynth. Res. 26: 59-66.Fish LE, Kuck U, Bogorad L (1985) Two partialy homologous adjacent light-inducible maizechloroplast genes encoding polypeptides of the P700 chlorophyll a protein complex ofphotosystem I. J. Biol. Chem. 260: 1413-1321.Forsburg SL, Guarente L (1989) Communication between mitochondria and the nucleus inregulation of cytochrome genes in the yeast Saccharomyces cerevisiae. Annu. Rev.Cell Biol. 5: 153-180.Fox TD (1986) Nuclear gene products required for translation of specific mitochondriallycoded mRNA’s in yeast. Trend Genet. 2: 97-100.170Foyer C, Furbank R, Harbinson J, Horton P (1990) The mechanisms contributing tophotosynthetic control of electron transport by carbon assimilation in leaves.Photosynth. Res. 25: 83- 100.Frankenberger EA, Hasegawa PM, Tigchelaar EC (1981) Diaflel analysis of shoot-formingcapacity among selected tomato genotypes. Z. Pflanzenphysiol. Bd. 102: 233-242.Gal A, Hauska G, Herrmann R, Ohad I (1990) Interaction between light harvestingchlorophyll-a/b protein (LHCII) kinase and cytochrome b6Ifcomplex. J. Biol. Chem.265: 19742-19749.Gal A, Shahak Y, Schuster G, Ohad I (1987) Specific loss of LHCII phosphorylation in theLemma mutant 1073 lacking the cytochromeb6/fcomplex. FEBS Lett. 2: 205-210.Gamble PE, Mullet JE (1989) Translation and stability of proteins encoded by the plastidpsbA and psbB genes are regulated by a nuclear gene during light-inducedchloroplast development in barley. J. Biol. Chem. 264: 7236-7243.Gamble PE, Sexton TB, Mullet JE (1988) Light-dependent changes inpsbD and psbCtranscripts of barley chloroplasts: accumulation of two transcripts maintains psbDand psbC translation capability in mature chloroplasts. EMBO J. 7: 1289-1297.Ghanotakis DF, Yocum CF (1990) Photosystem II and the oxygen-evolving complex. Annu.Rev Plant Physiol. Plant Mol. Biol. 41: 255-276.Ghirardi ML, Melis A (1988) Chlorophyll b deficiency in soybean mutants. I. Effects onphotosystem stoichiometry and chlorophyll antenna size. Biochem. Biophys. Acta932: 130-137.Ghirardi M L, McCauley SW, Melis A (1986) Photochemical apparatus organization in thethylakoid membrane of Hordeum vulgare type and chlorophyll b-less Chiorina /2mutant. Biochem. Biophys. Acta 851: 33 1-339. -Glazer AN, Melis A (1987) Photochemical reaction centers: Structure, organization, andfunction. Ann. Rev. Plant Physiol. 38: 11-45.Glick RE, Melis A (1988) Minimum photosynthetic unit size in system I and system II ofbarley chioroplasts. Biochem. Biophys. Acta 934L 15 1-155.Glick RE, McCauley SW, Gruissen W, Melis A (1986) Light quality regulates expression ofchioroplast genes and assembly of photosynthetic membrane complexes. Proc. Nati.Acad. Sci. USA 83: 4287-4291.Golbeck JH, Bryant DA (1991) Photosystem I. in (ed. Lee CP) Current Topics inBioenergetics, Vol. 16: Light-Driven Reactions in Bioenergetics. pp. 83-177.Govindjee (1990) Photosystem II heterogeneity: the acceptor side. Photosynth. Res. 25: 151-160.171Green BR (1988) The chlorophyll-protein complexes of higher plant photosyntheticmembranes or just what green band is that:. Photosynth. Res. 15: 3-32.Green BR Pichersky E, Kloppstech K (1991) Chlorophyll a/b-binding proteins: an extendedfamily. Trend Biochem. Sci. 16: 181-186.Grill E, Somerville C (1991) Construction and characterization of a yeast artificialchromosome library ofArabidopsis which is suitable for chromosome walking. Mol.Gen. Genet. 226: 484-490.Grivell LA (1989) Nucleo-mit,ochondrial interactions in yeast mitochondrial biogenesis.Eur. J. Biochem. 182: 477-493.Gruissem, W. (1989) Chloroplast gene expression: How plants turn their plastids on. Cell56: 161-170.Gruissem W, Barkan A, Deng X-W, Stern 1) (1988) Transcriptional and post-transcriptionalcontrol of plastid mRNA levels in higher plants. Trends Genet. 4: 258-263.Guenther JE, Nemson JA, Melis A (1990) Development of photosystem II in dark grownChiamydomonas reinhardtii: A light-dependent conversion of PSIIB, Q3-nonreducingcenters to the PSII, QB-reducing form. Photosynth. Res. 24: 35-46.Hahlbrock K, Scheel D (1989) Physiology and molecular biology of phenyipropanoidmetabolism. Annu. Rev. Plant Physiol. Plant Mol. Biol. 40: 347-369.Haley J, Bogorad L (1989) A 4-kDa maize chioroplast polypeptide associated with thecytochrmoe b6-f complex: subunit 5, encoded by the chloroplast petE gene. Proc.Natl. Acad. Sci. USA 86: 1534-1538.Haflick, R. B. (1989) Proposals for the naming of chloroplast genes. II. Update to thenomenclature of genes for thylakoid membrane polypeptides. Plant Molec. Biol. Rep.7: 266-275.Hansen U-P, Dau H, Bruning B, Fritsch T, Moldaenke C (1991) Linear analysis applied tothe comparative study of the I-D-P phase of chlorophyll fluorescence as induced byactinic PS-IT light, PS-I light and changes in CO2 concentration. Photosynth. Res.28: 119-130.Harle JR (1972) A revision of mutation breeding procedures in Arabidopsis based on a freshanalysis of the mutant sector problem. Can. J. Genet. Cytol. 14: 59-72.Harpster MIT, Mayfield SP, Taylor WC (1984) Effects of pigment-deficient mutants on theaccumulation of photosynthetic proteins in maize. Plant Mol. Biol. 3: 59-7 1.Heinemeyer W, Alt J, Herrmann R (1984) Nucleotide sequence of the clustered genes forapocytochrome b6 and subunit 4 of the cytochrome b6/fcomplex in the spinachplastid chromosome. Curr. Genet. 8: 543-549.172Henning J, Herrmann RG (1986) Chloroplast ATP synthetase of spinach contains ninenonidentical subunit species, six of which are encoded by plastid chromosomes in twooperons in a phylogenetically conserved arrangement. Mol. Gen. Genet. 203: 117-128.Herber U, Kobayashi Y, Leegood RC, Walker DA (1985) Low fluorescence yeild in anaerobicchloroplasts and stimulation of chlorophyll a fluorescence by oxygen and inhibitorsthat block electron flow between photosystems H and I. Proc. R. Soc. London Ser.225: 41-53.Herbert CJ, Labouesse M, Dujardin G, Slonimski PP (1988) The NAM2 proteins from S.cerevisiae and S. doiuglasii are mitochondrial leucyl-tRNA synthetases, and areinvolved in mRNA splicing. EMEO J. 7: 473-483.Herrmann F (1971) Genetic control of pigment-protein complexes I and Ia of the plastidmutant EN:Alba-1 ofAntirrhinun Majus. FEBS Lett. 19: 267-269.Hill R, Bendel R (1960) Function of the two cytochrome components in chioroplasts: aworking hypothesis. Nature 186: 136-137.Hiller RG, Linberg-Moller B, Hoyer-Hansen G (1980) Characterization of six putativephotosystem I mutants in barley. Carlsberg Res. Commun. 45: 3 15-328.Hippler M, Ratajczak R, Haehnel W (1989) Identification of the plastocyanin bindingsubunit of photosystem I. FEBS Lett. 250: 280-284.Hiratsuka J, Shimada H, Whittier R, Ishibashi T, Shakamoto M, Mon M, Kondo C, Honji Y,Sun C, Meng B, Li Y, Kanno A, Nishizawa Y, Hirai A, Shinozaki K, Sugiura M(1989) The complete sequence of the rice (Oryza sativa) chioroplast genome:Intermolecular recombination between distinct tRNA genes accounts for a majorplastid DNA inversion during the evolution of cereals. Mo]. Gen. Genet. 217: 185-194.Hoffman NE, Pichersky E, Malik VS, Kenton K, Cashmore AR (1988) Isolation andsequence of a tomato cDNA clone encoding subunit II of the photosystem I reactioncenter. Plant Molec. Biol. 10: 435-445.Hoj PB, Svendsen I, Scheller HV, Moller BL (1987) Identification of a chloroplast encoded9kDa polypeptide as a 2 (4Fe-4S) protein carrying A and B of photosystem I. J. Biol.Chem. 262: 12676-12684.Holzwarth AR (1990) The functional organization of the antenna in higher plants and greenalgae as studied by time-resolved fluorescence techniques. In: Baltsxheffsky M (ed)Current Research in Photosynthesis. Kiuwer Press Dordrecht FRG. 2: 611-614.Hoyer-Hansen G (1980) Identification of haem-proteins in thylakoid polypeptide patterns ofbarley. Carlsberg Res. Commun. 45: 167-176.173Iba K, Takamiya K-I, Toh Y, Satoh H, Nishimura M (1991) Formation of functionally activechloroplasts is determined at a limited stage of leaf development in virescentmutants of rice. Dev. Genet. 12: 342-348.Ikeuchi M, Inoue Y (1988) A new photosystem H reaction center component (4.8 kDaprotein) encoded by chloroplast genome. FEBS Lett. 241: 99-104.flceuchi M, Takio K, Inoue Y (1989) N-terminal sequencing of photosystem II low-molecular-mass proteins: 5 and 4.1 kDa components of the 02 -evolving core complex fromhigher plants. FEBS Lett. 242: 263-269.Inamine G, Nash B, Weissbach H, Brot N (1985) Light regulation of the synthesis of thelarge subunit of ribulose-1,5-bisphosphate carboxylase in peas: Evidence fortranslational control. Proc. Natl. Acad. Sci. USA 82: 5690-5694.Johnson EM, Sears BB (1990) Structure and expression of cytochrome f in the Oenotheraplatome mutant. Curr. Genet. 17: 529-534.Karlin-Neumann GA, Sun L, Tobin EM (1988) Expression of light-harvesting chlorophylla /b-protein genes is phytochrome-regulated in etiolated Arabidopsis thalianaseedlings. Plant Physiol. 88: 1323-1331.Karuna JP, Poff KL (1989) Mutants ofArabidopsis thaliana with altered phototropism.Planta 178: 400-406.Karuna JP, Best TR, PoffKL (1989) Influence of hook position on phototropic andgravitropic curvature by etiolated hypocotyls ofArabidopsis thaliana. Plant Physiol.90: 376-379.Kawata EE, Cheung AY (1990) Molecular analysis of an aurea photosynthetic mutant(Su/Su) in tobacco: LHCP depletion leads to pleiotropic mutant phenotypes. EMI3OJ. 11: 4197-4203.Kirsch W, Seyer P, Herrmann RG (1986) Nucleotide sequence of the clustered genes for twoP700 chlorophyll a apoproteins of the photosystem I reaction center and theribosomal protein s14 of the spinach plastid chromosome. Curr. Genet. 10: 843-855.KLaffP, Gruissem W (1991) Changes in chloroplast mRNA stability during leafdevelopment. Plant Cell 3: 517-529.Klee H, Estelle M (1991) Molecular genetic approaches to plant hormone biology AnnuRev. Plant Physiol. Plant MoL Biol. 42: 529-55 1.Klein RR, Gamble PE, Mullet JE (1988) Light-dependent accumulation of radiolabeledplastid-encoded chlorophyll a-apoproteins requires chlorophyll a. I. Analysis ofchlorophyll-deficient mutants and phytochrome involvement. Plant Physiol. 88:1246 1256174Klimov VV, Krasnovskii AA (1981) Pheophytin as a primary electron acceptor inphotosystem H reaction centres. Photosynthetica 15: 592-609.Koncz C, Mayerhofer R, Koncz-Kalman Z, Nawrath C, Reiss B, Redei GP, Shell J (1990)Isolation of a gene encoding a novel chloroplast protein by T-DNA tagging inArabidopsis thaliana. EMBO J. 9: 1337-1346.KoornneefM, van Eden J, Hanhart CJ, Stam P, Braaksma FJ, Feenstra WJ (1983) Linkagemap ofArabidopsis thatiana. J. Hered. 74: 265-272.Koornneef M, Hanhart CJ, Martinelli L (1987) A genetic analysis of cell culture traits intomato. Theor. AppI. Genet. 74: 633-641.Krause GH, Weis E (1984) Chlorophyll fluorescence as a tool in plant physiology. II.Interpretation of fluorescence signals. Photosynth. Res. 5: 139-157.Krause GH, Weis E (1991) Chlorophyll fluorescence and photosynthesis: The basics. Annu.Rev. Plant Physiol. Plant Mol. Biol. 42: 3 13-349.Krupinska K, Apel K (1989) Light-induced transformation of etioplasts to chioroplasts ofbarley without transcriptional control of plastid gene expression. Mol. Gen. Genet.219: 467-473.Kuchka MR, Goldschmidt-Clermont M, van Dillewijn, Rochaix J-D (1989) Mutation ot theChiamydomonas nuclear NAC2 locus specifically affects stability of the chioroplastpsbD transcript encoding polypeptide D2 of P511. Cell 58: 869-876.Kyle DJ, Staehelin LA, Arntzen CJ (1983) Lateral mobility of the light-harvesting complexin chioroplast membranes controls exitation energy distribution in higher plants.Arch. Biochem. Biophys. 222: 527-532.Lagoutte B (1988) Cloning and sequence of spinach cDNA clones encoding the 20 kDa PSIpolypeptide. FEBS Lett. 174: 24-29.Lagoutte B, Mathis P (1989) The photosystem I reaction center: Structure andphotochemistry. Photochem. Photobiol. 49: 833-844.Langdale JA, Zelitch I, Miller E, Nelson T (1988) Cell position and light influence C4 versusC3 patterns of photosynthetic gene expression in maize. EMBO J. 7: 3643-3651.Larrinua IM, Muskaavitch KMT, Gubbins EJ, Bogorad L (1983) A detailed restrictionendonuclease site map of the Zea mays plastid genome. Plant Mol Biol 2: 129-140.Last RL, Fink GR (1988) Tryptophan-requiring mutants of the plant Arabidopsis thaliana.Science 240: 305-3 10.175Legocki RP, Verma DPS (1981) Multiple immunoreplica technique: screening for specificproteins with a series of different antibodies using one polyacrylamide gel.Analytical Biochem. 111: 385-392.Lemaire C, Woliman F-A (1989) The chioroplast ATP synthase in Chiamydomonasreinhardtii. I. Characterization of its nine constitutive subunits. J. Biol. Chem. 264:10228-10234.Lemaire C, Woilman F-A (1989) The chioroplast ATP synthase in Chiamydomonasreinhardtii. U. Biochemical studies on its biogenesis using mutants defective inphotophosphorylation. J. Biol. Chem. 264: 10235-10242.Leto KJ, Miles CD (1980) Characterization of three photosystem II mutants in Zea mayslacking a 32,000 dalton lamellar polypeptide. Plant Physiol. 66: 18-24.Leto KJ, Keresztes A, Arntzen DJ (1982) Nuclear involvement in the appearance of achloroplast-encoded 32,000 dalton thylakoid membrane polypeptide integral to thephotosystem II complex. Plant Physiol. 69: 1450-1458.Leto KJ, Bell E, Mcintosh L (1985) Nuclear mutation leads to an accelerated turnover ofchioroplast-encoded 48kd and 34.5 kd polypeptides in thylakoids lacking photosystemII. EMBO J. 4: 1645-1653.Leu S, Weinberg D, Michaels A (1990) Transcription and translation of the chioroplastatpB-gene and assembly of ATP synthase subunit (beta). FEBS Lett. 269: 41-44.Leutwiler IS, Hough-Evans BR, Meyerowitz EM (1984) The DNA ofArabidopsis thaliana.Mol. Gen. Genet. 194: 15-23.Leutwiler LS, Meyerowitz EM, Tobin EM (1986) Structure and expression of three light-harvesting chlorophyll a/b binding proteins in Arabidopsis thaliana. Nucleic AcidsRes. 14: 4051-.Li Y, Sugiura M (1990) Three distinct ribonucleoproteins from tobacco chioroplasts: eachcontains a unique amino terminal acidic domain and two ribonucleoproteinconcensus motifs. EMBO J. 9: 3059-3066.Li N, Warren PV, Golbeck JH, Frank G, Zuber H, Bryant DA (1991) Polypeptidecomposition of the photosystem I complex and the photosystem I core protein fromSynchococcus sp. PCC 6301. Biochem. Biophys. Acta 1059: 2 15-225.Liscum E, Hangarter RP (1991) Arabidopsis mutants lacking blue light-dependentinhibition of hypocotyl elongation. Plant Cell 3: 685-694.Lloyd AM, Barnason AR, Rogers SG, Byrne MC, Fraley RT, Horsch RB (1986)Transformation ofArabidopsis thaliana with Agrobacterium tumefaciens. Science234: 464-466.176Marder JB, Barber J (1989) The molecular anatomy and function of thylakoid proteins.Plant Cell Environ. 12: 595-614.Maroc J, Gamier J (1981) Gel electrophoresis of chloroplast membranes of mutants ofChiamydomonas rheinhardtii which have impaired photosystem H function and lackphotosynthetic cytochromes. Biochem. Biophys. Acta 637: 473-.Martienssen R, Barkan A, Taylor WC, Freeling M (1990) Somatically heritable switches inthe DNA modification ofMu transposable elements monitored with a suppressiblemutant in maize. Genes Dev. 4:331-343.Martienssen RA, Barkan A, Freeling M, Taylor WC (1989) Molecular cloning of a maizegene involved in photosynthetic membrane organization that is regulated byRobertsons Mutator. EMBO J. 6: 1633-1639.Matineau B, Taylor WC (1985) Photosynthetic gene expression and cellular differentiationin developing maize leaves. Plant Physiol. 78: 399-404.Mayes SR, Cook KM, Zhang Z, Barber J (1991) Deletion of the gene encoding thephotosystem 1133 kDa protein from Synechocystis sp. PCC 6803 does not inactivatewater-splitting but increases vulnerability to photoinhibition. Biochem. Biophys.Acta 1060: 1-12.Mayfield SP, Taylor WC (1984) Carotenoid deficient maize seedlings fail to accumulatelight-harvesting chlorophyll a/b binding protein (LHCP) mRNA. Eur. J. Biochem.144: 79-84.Meinke DW (1991) Perspectives on genetic analysis of plant embryogenesis. Plant Cell 3:857-866Melis A (1985) Functional properties of photosystem JIB in spinach chioroplasts. Biochem.Biophys. Acta 808: 334-342.Melis A (1991) Dynamics of photosynthetic membrane composition and function. Biochem.Biophys. Acta 1058: 87-106.Metz JG, Miles D (1982) Use of a nuclear mutant in maize to identitSr components ofphotosystem II. Biochem. Biophys. Acta 681: 95-102.Metz JG, Miles D, Rutherford? (1983) Characterization of nuclear mutants of maize whichlack the cytochrome f/b-563 complex. Plant Physiol. 73: 452-459.Metz JG, Krueger RW, Miles CD (1984) Chlorophyll-protein complexes of a photosystem IImutant of maize. Plant Physiol. 75: 238-241.Meyerowitz EM (1989) Arabidopsis, a usefull weed Cell 56 263-269Miles D (1980) Mutants of higher plants.. Maize Methods in Enzymology 69 3-23177Miles CD, Daniel DJ (1973) A rapid screening technique for photosynthetic mutants inhigher plants. Plant Sci. Lets. 1: 237-240.Miles CD, Daniel DJ (1974) Chloroplast reactions of photosynthetic mutants in Zea mays.Plant Physiol. 53: 589-595. -Miles CD, Markwell JP, Thornber JP (1979) Effect of nuclear mutation in maize onphotosynthetic activity and content of chlorophyll-protein complexes. Plant Physiol.64: 690-694.Moller BR, Smillie RM, Hoyer-Hanson G (1980) A photosystem I mutant in barley(Hordeum vulgare L.). Carsberg Res. Commun. 45:87-99.Morris J, Herrmann RG (1984) Nucleotide sequence of the gene for the P680 chlorophyll aapoprotein of the photosytem II reaction center of spinach. Nucleic Acids Res. 12:2837-2850.Mourad G, Polacco M, Skogen-Hagenson MJ, Morris D, Robertson D (1989) A maternallyinherited mutant ofZea mays L. lacks the cytochrome b/f complex. Curr. Genet. 16:109-116.Mullet JE (1988) Chloroplast development and gene expression. Ann. Rev. Plant Physiol.39: 475-502.Mullet JE, Klein RR (1987) Transcription and RNA stability are important determinants ofhigher plant chloroplast RNA levels. EMBO J. 6: 157 1-1579.Mullet J, Burke JJ, Arntzen C (1980) Chlorophyll proteins of photosytem I. Plant Physiol.65: 8 14-822.Mullet JE, Gamble-Klein P, Klein RR (1990) Chlorophyll regulates accumulation of theplastid-encoded apoproteins CP43 and Dl by increasing apoprotein stability. Proc.Natl. Acad. Sci. USA 87: 4038-4042.Munch 5, Ljungberg U, Steppuhn J, Schneiderbauer A, Neshushtai R, Beyreuther K,Herrmann RG (1988) Nucleotide sequence of cDNAs encoding the entire precursorpo1ypeptides for subunits II and III of the photosystem I reaction center fromspinach. Curr. Genet. 14: 511-518.Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassays with tobaccotissue culture. Plant Physiol. 15: 493-497.Murata N, Miyao M, Hahasida N, Hidaka T, Sugiura M (1988) Identification of a new genein the chloroplast genome encoding a low-molecular-mass polypeptide of photosystemH complex. FEBS Lett. 235: 283-288.178Murray, DL, Kohorn BD (1991) Chloroplasts ofArabidopsis thaliana homozygous for thech-1 locus lack chlorophyll b, lack stable LHCPII and have stacked thylokoids. PlantMolec. Biol. 16: 71-79.Nadolska-Orczyk A, Malepszy S (1989) In vitro culture of Cucumis sativus L. 7. Genescontrolling plant regeneration. Theor. Appi. Genet. 78: 836-840.Nam H-G, Giraudat J, Boer B den, Moonan F, Loos WBD, Hauge BM, Goodman HM (1989)Restriction fragment length polymorphism linkage map ofArabidopsis thaliana.Plant Cell 1:699-705.Namba 0, Satoh K (1987) Isolation of a photosystem II reaction center consisting of D-1, D2 polypeptides and cytochrome b-559. Proc. NatI. Acad. Sci. USA 84: 109-112.Nechushtai R, Nelson N (1985) Biogenesis of photosystem I reaction center during greeningof oat, bean and spinach leaves. Plant Molec. Biol. 4: 377-384.Nechushtai R, Muster P, Binder A, Liveanu V, Nelson N (1983) Photosystem I reactioncenter from the thermophilic cyanobacterium Mastiglcladus laminosus. Proc. NatI.Acad. Sci. USA 80: 1179-1183.Nixon PJ, Gounaris K, Coomber SA, Hunter CN, Dyer TA, Barber J (1989) psbG is not aphotosystem two gene but may be an ndh gene. J. Biol. Chem. 264: 14129-14135.Noctor G, Rees D, Young A, Horton P (1991) The relationship between zeaxanthin, energy-dependent quenching of chlorophyll fluorescence, and trans-thylakoid pH gradient inisolated chioroplasts. Biochem. Biophys. Acta 1057: 320-330.Oelmuller R, Levitan I, Bergfeld R, Rajasekhar VK, Mohr H (1986) Expression of nucleargenes as affected by treatments acting on the plastids. Planta 168: 482-492.Oelmuller R, Schuster C, Mohr H (1988) Physiological characterization of a plastidic signalrequired for nitrate-induced appearance of nitrate and nitrite reductases. Planta174: 75-83.Ohashi K, Tanaka A, Tsuji H (1989) Formation of the photosynthetic electron transportsystem during the early phase of greening in barley leaves. Plant Physiol. 91:409-414.Ohyama K, Fukuzawa H, Kohchi T, Shirai H, Sano T, Sano S, Umesono K, Shiki Y, TakeuchiM, Chang Z, Aota S, Inokuchi H, Ozeki H (1986) Chloroplast gene organizationdeduced from complete sequence of liverwort Marchantia polymorpha chloroplastDNA. Nature 322: 572-574.Olive J, Vallon 0, Wollman F-A, Recouvreur M, Bennoun P (1986) Studies on thecytochromeb6/fcomplex. II. Localization of the complex in the thylakoid membranesfrom spinach and Chlamydomonas reinhardtii by immunocytochemistry and freeze-fracture analysis ofb6/fmutants. Biochem. Biophys. Acta 851: 239-248.179Ortega JM, Hervas M, Losada M (1989) Location of cytochrome b-559 between photosystemI and photosystem II in noncyclic electron transport. Biochem. Biophys. Acta 975:244-251.Pakrasi HB, Diner BA, Williams JGK, Arntsen CJ (1989) Deletion mutagenesis of thecytochrome b559 protein inactivates the reaction center of photosystem II. Plant Cell1: 591-597.Parikh VS, Morgan MM, Scott R, Clements IS, Butow RA (1987) The mitochondrialgenotype can influence nuclear gene expression in yeast. Science 235: 576-580.Parsons TJ, Bradshaw lID, Gordon MP (1989) Systemic accumulation of specific mRNAs.inresponse to wounding in poplar trees. Proc. Natl. Acad. Sci. USA 86: 7895-7899.Patton DA, Meinke DW (1988) High-frequency plant regeneration from cultured cotyledonsofArabidopsis thaliana. Plant Cell Rep. 7:233-237.Patton DA, Franzmann LII, Meinke DW (1991) Mapping genes essential for embryodevelopment in Arabidopsis thaliana. Mol. Gen. Genet. 227: 337-347.Pichersky I, Green BR (1990) The extended family of chlorophyll a/b-binding proteins ofPSI and PSII. Curr Res. in Photosynth. 3:553-556.Piechulla B, Pichersky E, Cashmore AR, Gruissem W (1986) Expression of nuclear andplastid genes for photosynthesis-specific proteins during tomato fruit developmentand ripening. Plant Molec. Biol. 7: 367-376.Redei GP (1975) Arabidopsis as a genetic tool. Annu. Rev. Genet. 9: 111-127.Reilly P, Hulmes JD, Pan Y-CE, Nelson N (1988) Molecular cloning and sequencing of thepsaD gene encoding subunit II of photosystem I of the cyanobacterium, Synechocystissp. PCC 6803. J. Biol. Chem. 265: 17658-17662.Robertson D, Woessner JP, Gillham NW, Boynton JE (1989) Molecular characterization oftwo point mutations in the chloroplast atpB gene of the green alga Chlamydomonasreinhardtii defective in assembly of the ATP synthase complex. J. Biol. Chem. 264:2331-2337.Robertson D, Gillham NW, Boynton JE (1990) Cotranscription on the wild-type chloroplastatpE gene encoding the CF1/C0epsilon subunit with 3’ half of the rps7 gene inChiamydomonas reinhardtii and characterization of frameshift mutations in atpE.Mol. Gen. Genet. 221: 155-163.Rochaix, J-D, Erickson J (1988) Function and assembly of photosystem II: Genetic andmolecular analysis. Trends Biochem. Sci. 13: 56-59.180Rochaix J-D, Kuchka M, Mayfield 5, Schirmer-Rahire M, Girard-Bascou J, Bennoun P(1989) Nuclear and chioroplast mutations affect the synthesis or stability of thechloroplastpsbC gene product in Chiamydomonas reinhardtii. EMBO J. 8: 1013-1021.Rock CD, Zeevaart JAD (1991) The aba mutant ofArabidopsis thaliana is impaired inepoxy-carotenoid biosynthesis. Proc. NatI. Acad. Sci. USA 88: 7496-7499.Rock CD, Barkan A, Taylor WC (1987) The maize plastid psbB-psbF-petB-petD gene cluster:Spliced and unpliced petB and petD RNAs encode alternate products. Curr Genet.12: 69-77.Rodermel SR, Bogorad L (1985) Maize plastid photogenes: Mapping and photoregulation oftranscript levels during light-induced development. J Cell Biol 100: 463-476.Rogers SG, Klee HJ, Horsch RB, Fraley RT (1987) Improved vectors for planttransformation: Expression cassette vectors and new selectable markers. Meth.Enzymol. 153: 253-27 7.Roshchina VV, Karpilova IP, Bozhok GV, Gostimskii SA (1983) Photosynthesis of peamutants with damaged photosystems. Photosynthetica 17: 590-596.Ruhie W, Reilander H, Otto K-D (1983) Chlorophyll-protein-complexes of thylakoids of wild-type and chlorophyll b mutants ofArabidopsis thaliana. Photosynth. Res. 4: 301-3054.Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A Laboratory Manual. (ColdSpring Harbor, NY: Cold Spring Harbor Laboratory).Sanders CE, Melis A, Allen JF. (1989) In vivo phosphorylation of proteins in thecyanobacterium Synechococcus 6301 after chromatic acclimation to photosystem I orphotosystem II light.. Biochim. Biophys. Acta 976: 168-172.Sandmann G (1991) Biosynthesis of cyclic carotenoids: biochemistry and molecular geneticsof the reaction sequence. Physiol. Plant. 83: 186-193.Schagger H, Jagow G von (1987) Tricine-sodium dodecyl sulfate-polyacrylamide gelelectrophoresis for the separation of proteins in the range from 1 to 100 kDa.Analytical Biochem. 166: 368-379.Scheller HV, Moller BL (1990) Photosystem I polypeptides. Physiol. Plant. 78: 484-494.Scheller HV, Hoj PB, Svendsen I, Moller BL (1988) Partial amino acid sequences of twonuclear-encoded photosystem I polypeptides from barley. Biochem. Biophys. Acta933: 501-505.181Scheller JTV, Okkels JS, Nielsen VS, Andersen, B, Moller BL (1991) Identification andcharacterization of photosystem I genes and subunits. J. Cell. Biochem. Suppl. 15A79.Schlodder E, Brettel K (1988) Primary charge separation in closed photosystem II with alifetime of 11 ns. Flash absorption spectroscopy with oxygen evolving photosystem IIcomplexes from Synechococcus. Biochem. Biophys. Acts 933: 22-34.Schmidt R, Willmitzer L (1988) High efficiency A.grobacterium tumefaciens-mediatedtransformation ofArabidopsis thaliana leaf and cotyledon explants. Plant Cell Rep.7: 583-586.Schmidt R, Willmitzer L (1989) The maize autonomous element Activator (Ac) shows aminimal germinal excision frequency of 0.2%-O.5% in transgenic Arabidopsisthaliana plants. Mol. Gen. Genet. 220: 17-24.Schneider T, Dinkins R, Robinson K, Shelihammer J, Meinke DW (1989) An embryo-lethalmutant ofArabidopsis thaliana is a biotin auxotroph. Dev. Biol. 131: 161-167.Schreiber U, Schliwa U, Bilger W (1986) Continuous recording of photochemical and nonphotochemical chlorophyll fluorescence quenching with a new type of modulationfluoromenter. Photosynth. Res. 10: 5 1-62.Schrubar H, Wanner G, Westhoff P (1990) Transcriptional control of plastid geneexpression in greening Sorghum seedlings. Plants 183: 101-111.Senger H (ed) (1987). Blue light responses: phenomena and occurance in plants andmicroorganisms. Vols I and H. CRC Press, Boca Raton, Fl.Sexton TB, Jones JT, Mullet JE (1990) Sequence and transcriptional analysis of the barleyctDNA region upstream ofpsbD-psbC encoding trnK (IJUU), rpsl6, trnQ (UUG),psbK, psbl, and trnS (GCU). Curr. Genet. 17: 445-454.Shatz GH, Brock H, Holzwarth AR (1988) Kinetic and energetic model for the primaryprocesses in photosystem II. Biophys. J. 54: 397-405.Sheen J (1990) Metabolic repression of transcription in higher plants. Plant Cell 2: 1027-1038.Shinozaki K, Fukuzawa M, Tanaka M, Wakasugi T, Harashida N, Matsubayashi T, Zaita N,Chunwongse J, Obokata J, Yamaguchi-Shinozaki K, Ohto C, Torazawa K, Meng BY,Sugita M, Deno H, Kamogashira T, Yamada K, Kusuda J, Takaiwa F, Kate A,Tohdoh N, Shimada H, Sugiura M (1986) The complete nucleotide sequence of thetobacco chloroplast genome: Its gene organization and expression. EMBO J. 5: 2043-2049.182Short TW, Briggs WR (1990) Characterization of a rapid, blue light-mediated change indetectable phosphorylation of a plasma membrane protein from etiolated pea(Pissum sativum L.) seedlings. Plant Physiol. 92: 179-185.Sieburth LE, Berry-Lowe S, Schmidt GW (1991) Chioroplast RNA stability inChiamydomonas: Rapid degradation ofpsbB and psbC transcripts in two nuclearmutants. Plant Cell 3: 175-189.Siefermann-Harms D (1985) Carotenoids in photosynthesis. I. Location in photosyntheticmembranes and light harvesting function. Biochem. Biophys. Acta 811: 325-355.Siefermann-Harms D (1987) The light-harvesting and protective function of carotenoids inphotosynthetic membranes. Physiol. Plant. 69: 56 1-568.Simpson DJ, Wettstein D von (1980) Macromolecular physiology of plastids XIV. viridismutants in barley: Genetic, fluoroscopic and ultrastructural characterization.Carlsberg Res. Commun. 45: 238-314.Simpson DJ, Machold 0, Hoyer-Hansen G, Wettstein D von (1985) Chlorina mutants ofbarley (Hordewn vulare L.). Carlsberg Res. Commun. 50: 223-238.Somersalo S, Aro E-M (1987) Fluorescence induction in pea leaves of different ages.Photosynthetica 21: 29-35.Somerville CR (1986) Analysis of photosynthesis with mutants of higher plants and algae.Annu. Rev. Plant Physiol. 37: 467-507.Somerville CR, Ogren WL (1982) Mutants of the cruciferous plant Arabidopsis thalianalacking glycine decarboxylase activity. Biochem. J. 202: 273-380.Staehelin LA, Arntzen CJ (1983) Regulation of chloroplast membrane function: Proteinphosphorylation changes the spatial organization of membrane components. J. CellBiol. 97: 1327-1337.Steinmuller K, Ellersiek U, Bogorad L (1991) Deletion of the psbGl gene of thecyanobacterium Synechocysts sp. PCC6803 leads to the activation of the crypticpsbG2 gene. Mol. Gen. Genet. 226: 107-112.Steppuhn J, Hermas J, Nechushtai R, Herrmann GS, Herrmann RG (1989) Nucleotidesequences of cDNA clones encoding the entire precursor polypeptide for subunit VIand the plastome-encoded gene for subunit VII of the photosystem I reaction centerfrom spinach. Curr Genet 16: 99-108.Stern DB, Gruissem W (1987) Control of plastid gene expression: 3’ inverted repeats act asmRNA processing and stabilizing elements, but do not terminate transcription. Cell51: 1145-1157.183Stern DB, Jones H, Gruissem W (1989) Function of plastid mRNA 3’ inverted repeats: RNAstabilization and gene-specific protein binding. J. Biol. Chem. 264: 18742-18750.Stern DB, Radwanski ER, Kindle KL (1991) A 3’ stern/loop structure of the Chiamydomonaschioroplast atpB gene regulates mRNA accumulation in vivo. Plant Cell 3: 285-297.Sugiura M (1989) The chioroplast chromosomes in land plants. Annu. Rev. Cell Biol. 5: 51-70.Sugita M, Gruissem W (1987) Developmental, organ-specific, and light-dependentespression of the tomato ribulose-1,5-bisphosphate carboxulase small subunit genefamily. Proc. Nati. Acad. Sd. USA 84: 7104-71083.Svensson P, Andreasson E, Arbertsson P-A (1991) Heterogeneity among photosystem I.Biochem. Biophys. Acta. 1060: 45-50.Takakashi Y, Goldschmidt-Clermont M, Soen S-Y, Franzen LG, Rochaix J-D (1991)Directed chioroplast transformation in Chiamydomonas reinhardtii: insertionalinactivation of the psaC gene encoding the iron sulfur protein destabilizesphotosystem I. EMBO J. 10:2033-2040.Taylor WC (1989) Regulatory interactions between nuclear and plastid genomes. Annu.Rev. Plant Physiol. 40: 2 11-233.Taylor WC, Barkan A, Martienssen RA (1987) Use of nuclear mutants in the analysis ofchioroplast development. Dev. Genet. 8: 305-320.Telfer A, Rivas J, Barber J (1991) B-Carotene within the isolated photosystem II reactioncentre: photooxidation and irreversible bleaching of this chromophore by oxidisedP680. Biochem. Biophys. Acta 1060: 106-114.Thompson WF, White MJ (1991) Physiological and molecular studies of light-regulatednuclear genes in higher plants. Annu. Rev Plant Physiol. Plant Mol. Biol. 42:423-466.Thornber JP (1986) Biochemical characterization and structure of pigement-proteins ofphotosynthetic organisms. End. Plant Physiol. New Series 19: 98-.Thornber JP, Morishige DT, Anandan S, Peter GF (1990) Chlorophyll-carotenoid proteins ofhigher plant thylakoids. In: Scheer H(ed) Chlorophylls. CRC Press Boca Raton pp.550-585.Tjus SE, Andersson B (1991) Extrinsic polypeptides of spinach photosystem I. Photosynth.Res. 27: 209-2 19.Troxier RF, Lin S, Offner GD (1989) Heme regulates expression of phycobiliproteinphotogenes in the unicellular rhotophyte, Cyanidium caldarium. J. Biol. Chem. 264:20596-20601.184TzagoloffA, Myers AM (1986) Genetics of mitochondrial biogenesis. Annu. Rev. Biochem.55: 249-285.Valvekens D, Van Montagu M, Van Lijsebettens M (1988) Agrobacterium tumefaciensmediated transformation ofArabidopsis thaliana root explants by using kanamycinselection. Proc. NatI. Acad. Sci. USA 85: 5536-3340.van Dorssen RJ, Breton J, Plijter JJ, Satoh K, van Gorkom HJ, Amesz J (1987)Spectroscopic properties of the reaction center and of the 47 kDa chlorophyll proteinof photosystem H. Biochem. Biophys. Acta 893: 267-274.van Kooten 0, Snel JFH (1990) The use of chlorophyll fluorescence nomenclature in plantstress physiology. Photosyn Res 25: 147-150.Vermaas WFJ, Williams JGK, Rutherford AW, Mathis P, Arntzen CP (1986) Geneticallyengineered mutant of the cyanobacterium Synechocystis 6803 lacks the photosystemII chlorophyll-binding protein CP-47. Proc. Natl. Acad. Sci. USA 83: 9474-9477.Webber AN, Malkin R (1990) Photosystem I reaction-centre proteins contain leucine zippermotifs. A proposed role in dimer formation. FEBS Lett. 264: 1-4.WesthoffP, Herrmann RG (1988) Complex RNA maturation in chioroplasts. Eur. J.Biochem. 171: 551-564.WesthoffP, Alt J, Nelson N, Herrmann RG (1985) Genes and transcripts for the ATPsynthase CFO(sub) subunits I and II from spinach thylakoid membranes. Mol GenGenet 199: 290-299.WesthoffP, Farchaus JW, Herrmann RG (1986) The gene for the Mr(sub) 10,000phosphoprotein associated with photosystem II is part of the psbB operon of thespinach plastid chromosome. Curr Genet 11: 165-169.White MJ, Green BR (1987) Antibodies to the photosystem I chlorophyll a+b cross-reactwith polypeptides of CP29 and LHCII. Eur. J. Biochem. 163: 545-551.Widger WR, Cramer WA, Herrmann RG, Trebst A (1984) Sequence homology andstructural similarity between cytochrome b of mitochondrial complex Ill and thechloroplastb6/fcomplex: position of the cytochrome b hemes in the membrane.Proc. NatI. Acad. Sci. USA 81: 674-678.Willey DL, Gray JC (1989) Two small open reading frames are co-transcribed with the peachioroplast genes for the polypeptides of cytochrome b-559. Curr. Genet. 15: 2 13-220.Willey DL, Gray JC (1990) An open reading frame encoding a putative haem-bindingpolypeptide is cotranscribed with pea chioroplast gene for apocytochrome f. PlantMolec. Biol. 15: 347-356.185Woliman F-A, Lemaire C (1986) Studies on kinase-controlled state transitions inphotosystem U andb6/fmutants from Chiamydomonas reinhardtii which lackquinone-binding proteins. Biochem. Biophys. Acta 933:85-94.Wynn RM, Malkin R (1988) interaction of plastocyanin with photosystem I. A chemicalcross-linking study of the polypeptide that binds to plastocyanin. Biochemistry 27:5863-5869.Wynn RM, Omaha J, Malkin R (1989) Structural and functional properties of thecyanobacterial photosystem I complex, Biochemistry 28: 5554-5560.Yanofsky M..F, Ma H, Bowman JL, Drews GN, Feldmann KS, Meyerowitz EM (1990) Theprotein encoded by the Arabidopsis homeotic gene agamous resembles transcriptionfactors. Nature 346: 35-39.Yu J, Vermaas WFG (1990) Transcript levels and synthesis of photosystem II componentsin Cyanobacterial mutants with inactivated photosystem II genes. Plant Cell 2: 315-322.Zanetti G, Merati G (1987) Interaction between photosystem I and ferredoxin: Identificationby chemical cross-linking of the polypeptide which binds ferredoxin. Eur. J.Biochem. 169: 143-146.Zilber A, Malkin R (1988) Ferredoxin cross-links to a 22 kfla subunit of photosystem I.Plant Physiol. 88: 810-814.186


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items