LIBMRY Michigan State University '\ f PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. _—__. DATE DUE DATE DUE DATE DUE ll MSU Is An Affirmative Action/Equal Opportunity Institution chJ-q . ..—.__ MOLECULAR AND DEVELOPMENTAL ASPECTS OF RESPIRATORY COMPLEXES IN HIGHER PLANT MITOCEONDRIA BY Carrie Beth Hiser A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry and MSU-DOE Plant Research Laboratory 1991 ABSTRACT MOLECULAR AND DEVELOPMENTAL ASPECTS OF RESPIRATORY COMPLEXES IN HIGHER PLANT MITOCHONDRIA BY Carrie Beth Hiser Two important features that distinguish higher plant mitochondria from their mammalian counterparts are cytoplasmic male sterility (CMS) and the alternative respiratory pathway. CNS is a mitochondrial defect resulting in a lack of viable pollen but no adverse effects on vegetative growth. Mitochondrial DNA rearrangements involving genes for respiratory enzyme complex subunits have been correlated with CNS in some species. Part 1 of this thesis attempted to discover molecular changes in male-sterile sugarbeet lines, similar to those found in other species, which could be correlated with CMS. Mitochondrial DNA rearrangements which could be CNS-related were found in some of the male-sterile sugarbeet lines investigated. They involved genes encoding respiratory complex subunits and caused altered gene expression in some cases. The alternative respiratory pathway, consisting of a cyanide-resistant alternative terminal oxidase, is a non- energy-conserving bypass around the bc1 complex and cytochrome oxidase in the inner mitochondrial membrane. Part 2 of this thesis involved aspects of the alternative oxidase. The alternative pathway was studied in two pea cultivars, one of which had been reported to lack the pathway. However, both cultivars were found to possess alternative path capacity and identical alternative oxidase proteins. The alternative oxidase was then investigated in potato tuber slices, which developed alternative path capacity upon aging. This increased capacity was correlated with increased levels of a 36kD alternative oxidase protein. The corresponding potato alternative oxidase gene was cloned and sequenced. Comparison of the deduced amino acid sequence of the potato gene to that of Sauromatum guttatum revealed highly conserved regions, two of which were predicted to form membrane-spanning alpha helices. Investigation of potato alternative oxidase gene regulation during aging showed some level of constitutive expression. ACKNOWLEDGEMENTS I must extend my thanks to the collaborators on various parts of these projects who have provided the plant material, DNA, and clones that made this research possible. These kind people are acknowledged by name in the relevant chapters. Here, I would like to thank several people who have provided general assistance: Lee McIntosh for taking me into his lab and guiding me through many projects over the past years; the members of my Guidance Committee for their help and patience, especially Drs. Ferguson-Miller and Tolbert who have collaborated on parts of this work; Roxy Nickels and Kristin Rorrer, for their friendship as well as technical assistance; Larry Smart and Neil Bowlby for computer assistance; Kurt Stepnitz for all the photography herein; and the many members of the McIntosh lab from whom I have learned so much. Thanks are also’due.to the College of Natural Science for financial support in the form of a Biotechnology Fellowship and a Continuing Doctoral Fellowship. iv TABLE OF CONTENTS List Of TableSe O O O I O O O O O O ........ O 0 O O O O O O O O O O O O O O I O O O OPOViii List of Figures............. ...... . ......... . ...... .....p.ix LiSt Of AbbreViationSe O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O OPOXi Introduction to Higher Plant Mitochondria................p.1 Plant Mitochondrial Genomes............................p.1 Organization of Higher Plant Respiratory Chains........p.2 Goals of the Thesis....................................p.4 Literature Cited.......................................p.5 Part 1: Respiratory Complexes and Cytoplasmic Male Sterility Chapter 1-1: Introduction to Cytoplasmic Male 8t.r111ty000000000O0.0000000000000000000000.0......OOOOOOPOO Phenotype and Importance of Cytoplasmic Male Sterility..............................................p.8 Causes of Cytplasmic Male Sterility...................p.10 Cytoplasmic Male Sterility and Alternative Oxidase..p.10 Cytoplasmic Male Sterility and Non-genomic Elements............................................p.11 Cytoplasmic Male Sterility and Mitochondrial DNA Rearrangements Involving Respiratory Complex Subunit Genes.......................................p.12 Literature Cited.....................................p.14 Chapter 1-2: Molecular Aspects of Respiratory Enzymes and Cytoplasmic Male sterility in 8ugarbeet.............p.19 Introduction..........................................p.19 Materials and Methods.................................p.23 Beet Strains........................................p.23 Maize and Sugarbeet Genes Used as Probes............p.24 Isolation of Mitochondria, MtDNA, and MtRNA.........p.27 Cloning of the Si4-CMS Sugarbeet coxII Gene. . . . . . . . .p. 28 Results...............................................p.29 Restriction Digests of C1 Normal and Sfi-CMS MtDNA...............................................p.29 Cloning of the Sill-CMS coxII Gene. . . . . . . . . . . . . . . . . . .p.31 Northern Blots of Beet Root MtRNA...................p.34 Male-Sterile Cytoplasm Induced by y-Irradiation...p.34 Male-Sterile Cytoplasm Derived from Beta maritima..........................................p.38 Gene Expression in Male Floral Tissues..............p.41 Discussion............................................p.42 Restriction Digests of MtDNA and the Silt-CMS coxII Gene..........................................p.42 Expression of Respiratory Enzyme Genes in Beet Root................................................p.45 Acknowledgements......................................p.49 Literature Cited.............................. ...... ..p.49 Part 2: Genetic and Developmental Aspects of Alternative Oxidase Chapter 2-1: Introduction to Alternative Oxidase........p.54 History...............................................p.54 Inhibitors of the Alternative Pathway.................p.54 Properties of the Alternative Pathway.................p.56 Nature of the Alternative Pathway Terminal Oxidase....p.58 Free Radical Hypothesis.............................p.58 Lipoxygenase Mypothesis.............................p.59 Terminal Oxidase Protein Hypothesis.................p.60 Characteristics of the Alternative Oxidase Proteins...p.61 Genetics of the Alternative Pathway...................p.64 Partitioning of Electrons to the Alternative Pathway..p.65 Capacity and Activity...............................p.65 Measurement of Activity.............................p.67 Theories of Electron Partitioning...................p.68 Functions of the Alternative Pathway..................p.70 Function in Thermogenic Aroid Plants................p.70 Potential Functions in Non-aroid Plants.............p.70 Literature Cited......................................p.75 Chapter 2-2: Alternative Oxidase Capacities and Polypeptides in Two Pea Cultivars.......................p.84 Introduction..........................................p.84 Materials and Methods.................................p.85 Sources of Pea Mitochondria.........................p.85 Isolation of Mitochondria by Method A...............p.86 Isolation of Mitochondria by Method B...............p.86 Immunoblotting......................................p.87 Results and Discussion................................p.88 Acknowledgements......... ..... .. ...... ................p.92 Literature Cited......................................p.92 Chapter 2-3: The Alternative Oxidase of Potato is an Integral Membrane Protein Synthesized dc Nbvo During Aging of Tuber Slices...................................p.96 Introduction..........................................p.96 vi Materials and Methods.................................p.99 Aging of Potato Slices and Isolation of Mitochondria........................................p.99 Respiration Assays..................... ............. p.99 Partial Purification of Membrane Proteins and Immunoblotting.....................................p.100 Two-Dimensional Gel Electrophoresis and Blotting...p.101 Results..................................... ..... ....p.101 Discussion...........................................p.112 Acknowledgements.....................................p.116 Literature Cited.....................................p.116 Chapter 2-4: The Potato Alternative Oxidase Gene.......p.120 Introduction.........................................p.120 Materials and Methods................................p.121 Source of Potato Tissue and DNA and Southern Blotting...........................................p.121 Cloning and Sequencing of the Potato Alternative Oxidase Gene.......................................p.121 Manipulation of Sequence Data......................p.122 Polymerase Chain Reactions.........................p.123 Results..............................................p.123 The Potato Alternative Oxidase Gene and Deduced Amino Acid Sequence................................p.123 Comparison of the Deduced Amino Acid Sequences of the Potato and Sauromatum guttatum Alternative Oxidase Genes......................................p.129 Expression of the Potato Alternative Oxidase Gene..p.132 Discussion...........................................p.135 Acknowledgements.....................................p.139 Literature Cited.....................................p.139 amt? Of Th.'1.00000000000000000000000000000000000IOOp01‘1 Part 1: Respiratory Complexes and Cytoplasmic Male Sterility................. ...... .....................p.141 Results and Implications.......... ..... ............p.141 Future Directions..................................p.142 Part 2: Genetic and Developmental Aspects of Alternative Oxidase..................................p.143 Results and Implications..................... ..... .p.143 Future Directions..................................p.145 Literature Cited.....................................p.146 vii LIST OF TABLES Table 2-2-1. Rates of oxygen uptake by isolated mitochondria and percentage capacity of the alternative pathway...0.000.000.0000.0.00.00.00.00 ...... 00p089 viii Figure Normal Figure probed Figure of the Normal LIST OF FIGURES 1-2-1. Location of probes derived from C1 sugarbeet coxII true gene and pseudogene.........p.26 1-2-2. Southern blot of sugarbeet mtDNA with atp6, atp9, and coxII.. ...... ......... ...... p.30 1-2-3. Restriction maps of three fragments Si4-CMS coxII gene and comparison to the C1 COXII gene.0.0.0....OOOOOOOOOOOOOOOOOOOOOOO0.0.00p033 Figure 1-2-4. Northern blots of sugarbeet mtRNA probed with coxII and atp9........................ ...... p.36 Figure 1-2-5. Northern blot of sugarbeet mtRNA probed with atp6 and stained gel........................p.37 Figure 1-2-6. Northern blots of sugarbeet mtRNA probed with atp6 and atp9...............................p.39 Figure 1-2-7. Northern blot of sugarbeet mtRNA probed with coxII gene and pseudogene...................p.40 Figure 2-2-1. Immunoblotting of alternative oxidase proteins in two pea cultivars...........................p.91 Figure 2-3-1. Respiration of potato mitochondria ...... p.102 Figure 2-3-2. Immunoblots of potato alternative oxidase................................................p.105 Figure 2-3-3. Mixing and incubation of mitochondrial samples.0.0....0..OOOOOOOOOOOO0.00000000000000000......p0109 Figure 2-3-4. Two-dimensional gel electrophoresis and immunoblots of alternative oxidase proteins........p.111 Figure 2-4-1. Nucleotide and deduced amino acid sequences of the potato alternative oxidase gene.......p.125 Figure 2-4-2. Comparison of deduced amino acid sequences of the alternative oxidase genes from potato and Sauromatum guttatum................. ....... .p.127 ix Figure 2-4-3. Southern blot of potato genomic DNA ..... p.128 Figure 2-4-3. Membrane helix predictions for alternative oxidase proteins...........................p.131 Figure 2-4-5. Detection of the potato alternative oxidase mRNA by PCR amplification......................p.134 LIST OF ABBREVIATIONS A: adenine (base of DNA) ADP: adenosine diphosphate ATP: adenosine triphosphate BIGCHAP: N,N-bis-(3-D-glucoamidopropyl)-deoxycholate bp: basepairs (of DNA) BSA: bovine serum albumin C: cytosine (base of DNA) cDNA: complementary DNA (reverse-transcribed from RNA) CMS: cytoplasmic male sterility cv: cultivar DEP: diethylpyrocarbonate EDTA: ethylenediaminetetraacetate, disodium salt EPR: electron paramagnetic resonance spectroscopy FCCP: p-trifluoromethoxycarbonyl cyanide G: guanine (base of DNA) HEPES: N-hydroxyethylpiperazine-N'-2-ethanesulfonic acid kb: kilobase pairs (of DNA) kD: kilodaltons MOPS: 3-(n-morpholino)propanesulfonic acid mRNA: messenger RNA mtDNA: mitochondrial DNA mtRNA: mitochondrial RNA NADH: reduced nicotinamide adenine dinucleotide nt: nucleotides (of RNA) ORF: open reading frame PAGE: polyacrylamide gel electrophoresis PCR: polymerase chain reaction PVPP: polyvinylpolypyrrolidone rRNA: ribosomal RNA SDS: sodium duodecyl sulfate (sodium lauryl sulfate) SHAM: salicylhydroxamic acid T: thymidine (base of DNA) TCA cycle: tricarboxylic acid (Krebs, citric acid) cycle Tris: tris(hydroxymethyl)aminomethane tRNA: transfer RNA Tween-20: polyoxyethylenesorbitan monolaurate URF: unidentified open reading frame xi INTRODUCTION TO EIGEER PLANT MITOCHONDRIA Historically, higher plant mitochondria were long considered to be identical to their better-studied mammalian counterparts. Plant mitochondria do share many conserved features with mammalian mitochondria. These include general morphology, types of phospholipids, several membrane transport systems, TCA cycle functions, the phosphorylation mechanism and ATPase complex, and the cytochrome pathway of electron transport (4). However, important differences and features unique to plant mitochondria have since come to light. PLANT MITOCEONDRIAL GENOMES Plant mitochondria possess 10- to 100-fold more mitochondrial DNA (mtDNA) than mammalian or fungal mitochondria (7) . The genome is arranged in a large, circular ‘molecule (master' circle) ‘which. can. convert into smaller subgenomic circles by recombination at repeated sequences (7) . Plant mtDNA appears to encode a limited set of polypeptides similar to those encoded by mammalian mtDNA. These include subunits of the electron transport chain complexes, the ATPase complex, and a few ribosomal proteins, plus 3 rRNAs and many tRNAs (7). Plant mitochondria also contain some chloroplast 1 2 DNA sequences, but these are generally non-functional (10). The plant. mitochondrial genome is recombinationally active. Rearrangements in mtDNA which splice together parts of different genes to create novel reading frames have been correlated with cytoplasmic male sterility (CMS; 1,2) . CMS is a mitochondrial defect resulting in the lack of viable pollen but without adverse effects on vegetative growth. It is agriculturally valuable as well as being one of the few mitochondrially-encoded traits available to study. ORGANIZATION OF HIGHER PLANT RESPIRATORY CHAINS Higher plant mitochondria possess a pathway of electron transport in their inner membrane which is similar to that of mammalian.mitochondria. Reducing equivalents can be fed from the NADH dehydrogenase complex (Complex I) or the succinate dehydrogenase complex (Complex II) into a mobile inner membrane pool of ubiquinone. The electrons can then flow to the cytochrome b-c1 complex (Complex III, cytochrome c reductase complex), and from there to cytochrome oxidase (Complex IV). This classical cytochrome pathway of electron transport conserves energy as an electrochemical gradient across the inner mitochondrial membrane (4). One important difference in the respiratory chain of higher plant mitochondria is the presence of multiple NADH dehydrogenases (4). Unlike mammalian mitochondria which require shuttle systems to import reducing equivalents from 3 the cytosol, plant mitochondria can directly oxidize exogenous NADH by means of an NADH dehydrogenase complex on the outer surface of the inner’ membrane. They can also oxidize endogenous NADH in a rotenone-insensitive manner, which has been generally attributed to a rotenone-resistant inner membrane NADH dehydrogenase in addition to the classical rotenone-sensitive Complex I. HOwever a recent report has proposed that, instead of two separate dehydrogenases, the plant Complex I reduces quinones by twijathways, one of which is rotenone-resistant (9). A second important difference in the higher plant respiratory chain is an alternative electron transport pathway in the inner membrane with a cyanide-resistant terminal oxidase. Consequently, a portion of the respiratory capacity in most plant mitochondria is resistant to inhibitors of the cytochrome pathway (6). This alternative pathway is a non- energy-conserving bypass around two of the sites of energy conservation in the cytochrome pathway: the cytochrome b-c1 complex and cytochrome oxidase (6). Since the alternative pathway does not conserve energy as the cytochrome pathway does, the balance of electron flow between these two pathways is important to the energy levels and general metabolism of the plant cell. 4 GOALS OF THE THESIS This thesis is a study of higher plant mitochondrial electron transport chains, with emphasis on the terminal oxidases: cytochrome oxidase and the alternative oxidase. Part 1 of the thesis deals with cytoplasmic male sterility in sugarbeets. It represents an effort to discover molecular changes in male-sterile sugarbeet lines, similar to the mtDNA rearrangements found in male-sterile maize (2) and petunia (1) , which could be correlated with the CMS phenotype. Emphasis is placed on the mitochondrially-encoded subunits of the cytochrome oxidase and ATPase complex since mtDNA rearrangements in other species have involved these genes for respiratory complex subunits (1,2). Chapter 1-2 demonstrates that such mtDNA rearrangements do occur in the male sterile lines and could be CMS-related in some cases. However, the techniques to further investigate these correlations in male floral tissues were not available, and subsequently a new line of research was pursued. Part 2 of the thesis deals with aspects of the alternative oxidase. In Chapter 2-2, the alternative pathway is investigated in two pea cultivars. Since one cultivar was reported to lack the alternative pathway (8), this seemed to be an ideal system in which to study alternative oxidase function. However, the results of Musgrave et a1. (8) could not be repeated, so another system had to be chosen. Chapters 2-3 and 2-4 deal with the development of alternative path 5 capacity in potato tuber slices. Fresh potato slice mitochondria have almost entirely cyanide-sensitive respiration while mitochondria from 24-hour-aged slices develop alternative path capacity (3, 11) . This phenomenon had been known for years from respiration measurements, but no probes were available to investigate it at a molecular level. With the development of antibodies against the Sauromatum guttatum alternative oxidase (5) , molecular investigation became possible. In Chapter 2-3 these antibodies are used to characterize the potato alternative oxidase and to demonstrate that the increase in alternative path capacity is due to increased levels of this protein. Chapter 2-4 involves the cloning of the potato alternative oxidase rgene and its regulation during aging. LITERATURE CITED 1. Boeshore ML, Hanson MR, Ishar 8 (1985) A variant mitochondrial DNA rearrangement specific to petunia stable sterile somatic hybrids. Plant Mol Biol 4:125-132 2. Dewey RB, Levings C8 III, Timothy DH (1986) Novel recombinations in the maize mitochondrial genome produce a unique transcriptional unit in the Texas male-sterile cytoplasm. Cell 44:438-444 3. Dizengremel P, Lance C (1976) Control of changes in mitochondrial activities during aging of potato slices. Plant Physiol 58:147-151 4. Deuce R, Neuburger M (1989) The uniqueness of plant mitochondria. Ann Rev Plant Physiol Plant Mol Biol 40:371-414 5. Flthon TE, Nickels RL, McIntosh L (1989) Monoclonal antibodies to the alternative oxidase of higher plant mitochondria. Plant Physiol 89:1311-1317 6 6. Lance C, Chauveau M, Disengremel P (1985) The cyanide- resistant pathway of plant mitochondria. In: R Douce, DA Day, eds, Higher Plant Cell Respiration. Springer-Verlag, New York, pp.202-247 7. Levings III C8, Brown CG (1989) Molecular biology of plant mitochondria. Cell 56:171-170 8. Musgrave MB, Murfet IC, Siedow JN (1986) Inheritance of cyanide-resistant respiration in two cultivars of pea (Pisum sativum L.). Plant Cell Environ 9:153-156 9. Boole KL, Dry 18, James AT, liskich JT (1990) The kinetics of NADH oxidation by complex I of isolated plant mitochondria. Physiol Plant 80:75-82 10. Stern DB, Palmer JD (1984) Extensive and widespread homologies between mitochondrial DNA and chloroplast DNA in plants. Proc Nat Acad Sci USA 81:1946-1950 11. Theologis A, Laties CG (1978) Relative contribution of cytochrome-mediated and cyanide-resistant electron transport in fresh and aged slices. Plant Physiol 62:232-237 PART 1: RESPIRATORY COMPLEXES AND CYTOPLASMIC MALE SI'ERILITY CHAPTER 1-1: INTRODUCTION TO CYTOPLASMIC MALE STERILITY PHIOTYPE AND IMPORTANCE OF CYTOPLASMIC MALE STERILITY Cytoplasmic male sterility (CMS) is the abortion of pollen development in anthers specified by cytoplasmically- inherited factors. Gross morphological abnormalities of the male floral structures can occur, such as altered tassels lacking exserted anthers in maize (22) , feminized or petalloid stamens in Nicotiana (17) , or shrunken, discolored anthers in sugarbeet (42) . Within anthers, functional abnormalities cause abortion of pollen production or production of nonviable pollen. A commonly observed feature is premature degradation of the nutritive tapetal layer of the microsporangial wall (6) . The CMS phenotype, however, occurs without apparent detrimental effects on the vegetative or female floral structures. In fact, increased productivity has been reported in male-sterile lines of rapeseed (12). Several types of evidence point to the mitochondrion as the carrier of the CMS-specifying factors (For reviews, see 17 ,23,30.). Ultrastructural studies of anthers at various stages of development have shown that mitochondrial degeneration occurs early in microsporogenesis (20,22,23) . The CMS phenotype segregates independently of the parental 8 9 plastid type in Nicotiana (15) and Petunia (2), implicating the mitochondria instead of the chloroplasts. Restriction fragment analysis of mitochondrial DNA (mtDNA) can distinguish male-sterile cytoplasms from fertile cytoplasms in many species (2,17,34,36,39,43), while no differences in restriction fragment patterns of chloroplast DNA could be correlated with CMS in sunflower (5), pearl millet (43), sugarbeet (35), or Petunia (17). Major differences in mitochondrial translation products between fertile and male- sterile cytoplasms have been observed in maize (13), sorghum (1), Petunia (31), sugarbeet (3), and other species (17). Consequently it is now widely accepted that the mtDNA encodes the CMS phenotype. CMS is an important trait in major crop plant species. CMS is used agriculturally to produce hybrid seed, since any seed produced on the male-sterile (female parent) plant is necessarily hybrid. This is especially useful for crops with small flowers such as sugarbeet ( 42) , for which hand- emasculation (removal of the anthers with a fOrceps before pollen is shed) is very difficult. Although the CMS trait is mitochondrially-encoded, nuclear genes known as restorers of fertility (R, genes) can be introduced into the nuclear background of a male-sterile line to suppress the CMS phenotype (17,23). CMS is therefore a useful system to study mitochondrial-nuclear gene interaction (10,13,17,20,21,22,27,45,48). Because this ‘mitochondrial defect appears to affect the developmental process only in 10 male reproductive tissues, the study of CMS can provide insight into differential gene expression in specific tissues and at different developmental stages (47). CAUSES OF CYTOPLASMIC MALE STERILITY One single cause for CMS in all plants has not been found; rather, a variety of mitochondrial defects may produce CMS. CMS has arisen spontaneously within cultivars in some cases and has been induced with mutagens in others. Alloplasmic CMS, which arises from interspecific crosses (particularly in Nicotiana), has led to the concept that CMS results from incompatibility between the mitochondrial and nuclear genomes (17). However, the nature of CMS seems to vary between species, and different kinds of explanations have been proposed for specific cases. Cytoplasmic Male Sterility and Alternative Oxidase One paper (29) reported that male-sterile lines of soybean, maize, Plantago lanceolata, and Helenium amarum lacked the cyanide-resistant alternative pathway of respiration. In Petunia, suspension cells and immature anthers from male-sterile lines were reported to have less alternative pathway than isonuclear fertile lines (9). However, other authors have been unable to find any correlation between CMS and the alternative pathway in 11 Plantago (44), tobacco (16), rapeseed (12), or soybean (32). Any' possible connection. between. CMS .and. the alternative oxidase therefore remains debatable. Cytoplasmic Male Sterility and Non-Genomic DNA Elements In some species, small DNA elements not part of the mitochondrial genome (plasmids, episomes, or viruses) have been correlated with CMS. Linear plasmid-like DNAs have been found in a male-sterile Sorghum bicolor cytoplasm (7) . A self-replicating double-stranded RNA viral-like agent has been associated. with CMS in.‘Vicia faba (24). In sunflower (Helianthus annuus), the absence of a 1.45kb circular plasmid was correlated with CMS (5). CMS has been extensively investigated in maize, which has three different male-sterile cytoplasms (reviewed in 22) . Two episomal elements 51 and 82 are associated with the CMS-S maize cytoplasm. Early reports suggested that they existed independently of the main mtDNA and that reversion to fertility' was associated. ‘with integration into the mitochondrial genome (25), and a protein specific to CMS-S mitochondria was encoded by an open reading frame (ORF) in 82 (28). However, it was later shown that integration was not strictly correlated with fertility (33) and that the proteins encoded by at least two of the four ORFs on the two episomes are present in both male-sterile and fertile lines (48,49). 12 Cytoplasmic Male Sterility and Mitochondrial DNA Rearrangements Involving Respiratory Complex Subunit Genes Differences in restriction digests of mtDNA from fertile and male sterile lines of several species suggest that mtDNA rearrangements involving genes of respiratory enzyme complex subunits can give rise to CMS. Differences in the mtDNA restriction fragments containing the cytochrome oxidase subunit II gene (coxII) between fertile and male-sterile plants have been.reported.in‘wheat (39). ‘Rearrangements in or near coxII and the ATPase subunit 6 gene (atp6) were found in CMS-C maize (26) . A rearrangement immediately upstream of the cytochrome oxidase subunit I gene (coxI) and involving 81- and SZ-like sequences was reported in CMS-S maize (19), while a rearrangement within coxI leading to the synthesis of a variant subunit was reported in the male-sterile 9E cytoplasm of sorghum (1). In sunflower, mtDNA rearrangements have been reported near the ATPase a subunit (atpA) gene (41). Two cases of CMS-associated mtDNA rearrangements have been investigated in detail. In Petunia somatic hybrids, the start of the.ATPase subunit.9 gene (atp9), parts of coxII, and an unidentified open reading frame (urfS) have recombined to a unique sequence known as pcf (for Petunia CMS-associated fusion) (2). Upon comparing coxII sequences from fertile and male-sterile cytoplasms, it was suggested that pcf may be a processed coxII pseudogene inserted via homologous recombination behind an atp9 gene (37). This rearrangement 13 was reported to be transcribed only in CMS lines (47) . Reversion to fertility in the progeny of unstable male-sterile somatic hybrids was associated with the loss of pot (8). The pct gene encodes a 25kD protein present only in male-sterile cytoplasms, and its abundance is decreased by the presence of the single Petunia R, gene (31) . The R, gene, however, does not affect pcf'transcripts, which suggests that it works at a post-transcriptional level (38). In CMS-T maize, the 3547bp TURF2H3 fragment has arisen by recombinations between atp6, the mitochondrial 268 ribosomal RNA gene, and a chloroplast tRNA-arg gene fragment present in the mitochondrial genome (10). It contains two ORFs, of which the T-urf13 transcripts are unique to CMS-T mitochondria and are altered by the action of restorer genes R,1 and R,2 (10). The CMS-ijlants synthesize a 13ijprote n not seen in fertile or other male-sterile cytoplasms, and are missing a 21kD protein.present.in other cytoplasms (13,14). Synthesis of the 13kD protein is suppressed by the maize R,1 restorer gene (13) . This 13kD polypeptide confers sensitivity to the fungal Bm-T toxin (from Bipolaris maydis race T, the cause of southern corn leaf blight) not only to CMS-T maize :mitochondria, but.a180‘when introduced into.E. coli (4,11) and yeast (18) , presumably by binding the toxin and causing membrane permeablization. This toxin sensitivity trait has been strictly correlated with CMS in CMS-T maize, suggesting that the 13kD polypeptide is also directly responsible for the CMS trait (46). 14 In these cases, the mtDNA rearrangement associated with CMS involves genes for mitochondrially-encoded subunits of respiratory enzyme complexes. The cytochrome oxidase subunits I, II, and III, and the ATPase subunits a, 6, and 9 are encoded in the mtDNA in higher plants (23,30) . Rearrangements involving any of these genes might result in mitochondrial dysfunction(s) that could exhibit a CMS phenotype. It should be noted, however, that the reported CMS-associated rearrangements involve second or partial copies of the respiratory complex genes. In CMS-T maize, the TURF2H3 fragment contains part of atp6, but an intact copy exists elsewhere in the mitochondrial genome (10). The Petunia pcf sequence contains parts of atp9 and coxII, but multiple copies of both genes exist in the mtDNA (37,40) . It is nevertheless possible that the gene product of the rearrangement may interfere or compete with the product( s) of the intact gene(s) , thereby blocking proper assembly of respiratory enzyme complexes or forming inactive or less active complexes. LITERATURE CITED 1. Bailey-Barres J, Hanson DR, Fox TD, Leaver CJ’ (1986) Mitochondrial genome rearrangement leads to extension and relocation of the cytochrome c oxidase subunit I gene in sorghum. Cell 47:567-576 2. Boeshore ML, Hanson MR, Ishar S (1985) A variant mitochondrial DNA rearrangement specific to petunia stable sterile somatic hybrids. Plant Mol Biol 4:125-132 3. Boutry M, Faber A-M, Charbonnier M, Briquet M (1984) Microanalysis of plant mitochondrial protein synthesis products: Detection of variant polypeptides associated with cytoplasmic male sterility. Plant Mol Biol 3:445-452 15 4. Braun CJ, Siedow JN, lilliams ME, Levings III CS (1989) Mutations in the maize mitochondrial T-urf13 gene eliminate sensitivity to a fungal pathotoxin. Proc Nat Acad Sci USA 86:4435-4439 5. Brown CG, Bussey n, DesRosiers L (1985) Analysis of mitochondrial DNA, chloroplast DNA, and double-stranded RNA in fertile and cytoplasmic male-sterile sunflower (Helianthus annuus). Can J Genet Cytol 28:121-129 6. Chapman GP (1987) The tapetum. Int Rev Cytol 107:111-125 7. Chase CD, Pring DR (1986) Properties of the linear N1 and N2 plasmid-like DNAs from mitochondria of cytoplasmic male- sterile Sorghum bicolor. Plant Mol Biol 6:53-64 8. Clark 3, Gafni Y, Ishar S (1988) Loss of CMS-specific mitochondrial DNA arrangement in fertile segregants of Petunia hybrids. Plant Mol Biol 11:249-253 9. Connett MB, MR Hanson (1990) Differential mitochondrial electron transport through the cyanide-sensitive and cyanide- insensitive pathways in isonuclear lines of cytoplasmic male sterile, male fertile, and restored Petunia. Plant Physiol 93:1634-1640 10. Dewey' RS, Levings III CS, Timothy DH (1986) Novel recominations in the maize mitochondrial genome produce a unique transcriptional unit in the Texas male-sterile cytoplasm. Cell 44:439-444 11. Dewey RE, Siedow JN, Timothy DB, Levings III DH (1988) A 13-kilodalton maize mitochondrial protein in E. coli confers sensitivity to Bipolaris maydis toxin. Science 239:293-295 12. Farineau. J, Pascal L, Pelletier C (1990) Study of respiratory and photosynthetic activities in several cytoplasmic hybrids of rapeseed with cytoplasmic male sterility. Plant Physiol Biochem 28:333-342 13. Forde BG, Leaver CJ (1980) Nuclear and cytoplasmic genes controlling synthesis of variant mitochondrial polypeptides in male-sterile maize. Proc Nat Acad Sci USA 77:418-422 14. Forde BG, Oliver RJC, Leaver CJ, Gunn RB, Ramble RJ (1980) Classification of normal and male-sterile cytoplasms in maize. I.E1ectrophoretic analysis of variation in mitochondrially synthesized proteins. Genetics 95:443-450 15. Galun E,.Arsee-Conen P, Fluhr R, Fdelman M, AVIV’D (1982) Cytoplasmic hybridization in Nicotiana: Mitochondrial DNA analysis in progenies resulting from fusion between protoplasts having different organelle constitutions. Mol Gen Genet 186:50-56 16 16. Hakansson G, Glimelius 1:, Bennett HT ( 1990) Respiration in cells and mitochondria of male-fertile and male-sterile Nicotiana spp. Plant Physiol 93:367-373 17. Hanson MR, Conde MF (1985) Functioning and variation of cytoplasmic genomes: Lessons from cytoplasmic-nuclear interactions affecting male fertility in plants. Int Rev Cytol 94:213-267 18. Huang J, Lee S-H, Lin C, Medici R, Hack B, Myers AM (1990) Expression in yeast of the T-URF 13 protein from Texas male- sterile maize mitochondria confers sensitivity to methomyl and to Texas-cytoplasm-specific fungal toxins. EMBO J 9:339-347 19. Isaac PG, Jones VP, Leaver CJ ( 1985) The maize cytochrome c oxidase subunit I gene: Sequence, expression and rearrangement in cytoplasmic male sterile plants. EMBO J 4:1617-1623 20. Izhar S (1978) Cytoplasmic male sterility in petunia. III. Genetic control of microsporogenesis and male fertility restoration. J Hered 69:22-26 21. Kennell JC, lise RP, Pring DR (1987) Influence of nuclear background on transcription of a maize mitochondrial region associated with Texas male sterile cytoplasm. Mol Gen Genet 210:399-406 22. Laughnan JR, Gabay-Laughnan S (1983) Cytoplasmic male sterility in maize. Ann Rev Genet 17:27-48 23. Leaver CJ, Gray Ml ( 1982) Mitochondrial genome organization and expression in higher plants. Ann Rev Plant Physiol 33:373-402 24. Lefebvre A, Scalla R, Pfeiffer P (1990) The double- stranded RNA associated with the "447" cytoplasmic male sterility in Vicia faba is packaged together with its replicase in cytoplasmic membraneous vesicles. Plant Mol Biol 14:477-490 25. Levings III CS, Rim BD, Pring DR, Conde MF, Mans RJ, Laughnan JR, Cabay-Laughnan SJ ( 1980) Cytoplasmic reversion of cms-S in maize: Association with a transpositional event. Science 209:1021-1023 26. Levings III CS, Braun CJ, Dewey RE (1988) Unusual mitochondrial gene mutations. In: Abstracts from the Second International Congress of Plant Molecular Biology, Jerusalem 17 27. Mackenzie SA, Pring DR, Bassett MJ, Chase CD (1988) Mitochondrial DNA rearrangments associated with fertility restoration and cytoplasmic reversion to fertility in cytoplasmic male sterile Phaseolus vulgaris L. Proc Nat Acad Sci USA 85:2714-2717 28. Manson JC, Liddell AD, Leaver CJ, Murray M (1986) A protein specific to mitochondria from S-type male-sterile cytoplasm of maize is encoded by an episomal DNA. EMBO J 5:2775-2780 29. Musgrave ME, Antenevics J, Siedow JN (1986) Is male- sterility in plants related to lack of cyanide-resistant respiration in tissues? Plant Sci 44:7-11 30. Newton MJ (1988) Plant mitochondrial genomes: Organization, expression and variation. Ann Rev Plant Physiol Plant Mol Biol 39:503-532 31. Nivisen HT, Hansen MR (1989) Identification of a mitochondrial protein associated with cytoplasmic male sterility in petunia. Plant Cell 1:1121-1130 32. Obenland D, Hiser C, McIntosh L, Shibles R, Stewart CR (1988) Occurrence of alternative respiratory capacity in soybean and pea. Plant Physiol 88:528-531 33. O'Brien C, Zabala G, lalbet V (1988) Integrated R2 sequence in mitochondria of fertile B37N maize encodes and expresses a 130kD polypeptide similar to that encoded by the $2 episome of S-type male sterile plants. Nuc Acids Res 17:405-422. 34. Powling A ( 1982) Restriction endonuclease analysis of mitochondrial DNA from sugarbeet with normal and male-sterile cytoplasms. Heredity 49:117-120 35. Powling A, Ellis THN (1983) Studies on the organelle genomes of sugarbeet with male-fertile and male-sterile cytoplasms. Theor Appl Genet 65:323-328 36. Pring DR, Levings III CS (1978) Heterogeneity of maize cytoplasmic genomes among male-sterile cytoplasms. Genetics 89:121-136 37. Pruitt M, Hansen MR (1989) Cytochrome oxidase subunit II sequences in Petunia mitochondria: Two intren-containing genes and an intron-less pseudogene associated with cytoplasmic male sterility. Curr Genet 16:281-291 38. Rasmussen J, Hanson MR (1989) A NADH dehydrogenase subunit gene is ce-transcribed with the abnormal Petunia mitochondrial gene associated with cytoplasmic male sterility. Mol Gen Genet 215:332-336 18 39. Ricard B, Lejeune B, Araya A (1986) Studies on wheat mitochondrial DNA organization. Comparison of mitochondrial DNA from normal and cytoplasmic male sterile varieties of wheat. Plant Sci 43:141-149 40. Rothenberg M, Hansen MR (1987) Different transcript abundance of twe‘divergent.ATP synthase subunit 9 genes in.the mitochondrial genome of Petunia hybrida. Mol Gen Genet 209:21-27 41. Siculella L, Palmer JD (1988) Physical and gene organization of mitochondrial DNA in fertile and male sterile sunflower. CMS-associated alterations in structure and transcription of the atpA gene. Nuc Acids Res 16:3787-3799 42. Smith GL (1980) Sugarbeet. In: Hybridization of Crop Plants. American Society of Agronomy-Crop Science Society of America, Madison, WI pp.601-616 43. Smith.RL, Chewhury MRO, Pring DR (1987) Mitochondrial DNA rearrangments in Pennisetum associated with reversion from cytoplasmic male sterility to fertility. Plant Mol Biol 9:277-286 44. Van Dijk H, Ruiper PJC (1989) No evidence for a pleiotropic relationship between male sterility and cyanide- resistant respiration in Plantago lanceolata. Physiol Plant 77:579-586 45. Waltrud LS, Laughnan JR, Gabay SJ, Heeppe DE (1975) Effects of nuclear restorer genes on the cytochrome content of corn pollen mitochondria. Can J Bot 54:2718-2725 46. Wise RP, Pring DR, Gengenbach PG (1987) Mutation to male fertility and toxin insensitivity in Texas (T)-cyteplasm is associated with a frameshift in a mitochondrial open reading frame. Proc Nat Acad Sci USA 84:2858-2862 47. Young HG, Hansen MR (1987) A fused mitochondrial gene associated with cytoplasmic male sterility is developmentally regulated. Cell 50:41-49 48. Babala G, O'Brien-Vedder C, Walbet v (1987) $2 episome of maize mitochondria encodes a 130-kiloda1ton protein found in male sterile and fertile plants. Proc Nat Acad Sci USA 84:7861-7865 49. Zabala G, lalbot V (1988) An 81 episomal gene of maize mitochondria is expressed in male sterile and fertile plants of the S-type cytoplasm. Mol Gen Genet 211:386-392 CHAPTER 1-2: MOLECULAR ASPECTS OF RESPIRATORY ENZYMES AND CYTOPLASMIC MALE STERILITY IN SUGARBEET INTRODUCTION Sugarbeet (Beta vulgaris L.) is an important crop in Europe, the United States, Canada, and the Soviet Union, accounting for about 45% of the world sugar production (29). It is a biennial which produces a fleshy tap root the first year and an aerial flowering stalk the second (29). Because sugarbeet flowers are tiny (approximately 2-4mm in diameter), hand-emasculation of the flowers is an extremely difficult and time-consuming way to produce hybrid seed. Cytoplasmic male sterility (CMS) therefore provides a practical method for hybrid seed production, since any seed produced by the male- sterile (female parent) plant is necessarily hybrid (29). CMS was first reported in sugarbeet by Owen in 1945 (23) . The Owens-type CMS (also called the US-1 or S type) is the only male-sterile cytoplasm in agricultural use, and world sugarbeet production is considered vulnerable to pathogens due to this lack of genetic variability. Just such a situation occurred in 1970 when Southern corn leaf blight wiped out much of the 0.8. corn crop, because 85% of the corn carried the susceptible CMS-T cytoplasm (18) . Four "induced sterile" 19 20 cytoplasms (8,1, 8,2, 8,3, 8,4) were isolated following y- irradiation of the normal fertile cytoplasm (17) , but were reported to possess the Owens-type CMS (20). Alternative cytoplasms are being sought in other, uncultivated Beta species (2, 13) ; a new male-sterile cytoplasm derived from Beta maritima has been recently examined at the molecular level and found to differ from the Owens-type (9,20). Although premature disintegration of the tapetum and abnormalities in the microspores were observed under the electron microscope (30), the molecular defect causing CMS in sugarbeets is unknown. A comparison of polypeptides synthesized by beet leaf mitochondria showed differences between a fertile line and a male-sterile line (3). Low molecular weight mtDNA molecules (minicircles) with no homology to the main mtDNA have been reported, some of which were found only in fertile cytoplasms and some in both fertile and male-sterile cytoplasms (14,21,24,26,31) . However, in one case a minicircle unique to fertile cytoplasms (24) was later found in male-sterile cytoplasms when other beet lines were examined (21) , suggesting that the correlation between the absence of any particular minicircle(s) and CMS in these studies may have been be coincidental. Powling (25) observed differences in the mtDNA restriction patterns which distinguished male-sterile lines from fertile lines but not from each other. Other authors have also reported differences in mtDNA restriction patterns between fertile and male-sterile lines (2,9,13,20) . Variations in chloroplast DNA restriction 21 patterns could not be correlated with the CMS trait ( 26) . One report saw differences in the thylakoid membrane polypeptides between a male-sterile line and a fertile line (15) , but these may have resulted from using different sample preparation procedures for the two lines. In several species, mtDNA rearrangements involving respiratory enzyme complex subunit genes have been strongly implicated in CMS (see Chapter 1-1) . In sugarbeet, two studies have investigated the involvement of respiratory enzymes in CMS. Respiration measurements of mitochondria oxidizing a-ketoglutarate could not distinguish between fertile and male-sterile lines (30), but large variations in the respiration between individual beets of the same cultivar made comparisons between cultivars questionable. No differences in transcription of a gene for subunit II of the NADH:ubiquinone reductase complex were found between fertile and male-sterile lines (33) , suggesting that this complex may not be involved in CMS. However, mtDNA rearrangements in or near genes for respiratory complex subunits have been noted in male-sterile cytoplasms of sugarbeet (20) . The CMS-associated mtDNA rearrangement in Petunia has been proposed to involve a pseudogene of cytochrome oxidase subunit II (27); interestingly, a pseudogene containing the first exon of subunit II was found in the BMC-CMS male-sterile cytoplasm of sugarbeet (9) . This study examined mtDNA restriction patterns and transcript levels for three mitochondrially-encoded genes 22 which have been implicated in CMS-associated mtDNA rearrangements in other species: cytochrome oxidase subunit II (coxII) and ATPase subunits 6 and 9 (atp6 and atp9). All three genes were reported to be present as single copies in the Owens-type sugarbeet cytoplasm, which also included five repeated sequences (4). Therefore the potential for recombination (and possibly rearrangement) exists in sugarbeet mtDNA, and any rearrangement which disrupted any of these three genes would be likely to cause mitochondrial dysfunction. In this study, three types of male-sterile sugarbeet cytoplasms were examined: the naturally-occurring Owens-type, a y-irradiation-induced sterile type (17) reported to be like the Owens type (20), and the new sterile cytoplasm derived from Beta maritima which differs from the Owens type (20). Since differences in mtDNA may simply reflect natural divergence between different beet lines, this study compared the male-sterile lines with their fertile progenitors. Since differences in mtDNA found between a male-sterile line and its fertile progenitor may still be unrelated to CMS, the pattern of gene expression was examined. The molecular defect leading to the production of CMS may, however, be expressed only in the male floral structures or the developing male gametophyte or be expressed to a greater degree in these tissues. Such a situation was suggested for Petunia (32). 23 MATERIALS AND METHODS Beet Strains All sugarbeet roots, anthers, and pollen were obtained from Dr. J. Clair' Theurer, adjunct (professor, USDA. and Michigan State University Department of Crop and Soil Science, East Lansing, Michigan. The following lines were studied: 1. g1_NQ;mal: fertile line carrying the fertile N cytoplasm and the C1 nuclear background (no restorer genes) 2. M: sterile line carrying the Owens-type sterile cytoplasm and the C1 nuclear background (no restorer genes) 3. S: sterile line carrying the Owens-type sterile cytoplasm (no restorer genes) 4. §,4_-_QM_: sterile line carrying the 8,4 cytoplasm in which sterility was induced by y-irradiation (17) , and the C1 nuclear background (no restorer genes) 5. §,4;3,: fertile line carrying the 8,4 sterile cytoplasm but with nuclear restorer genes 6. BM_Q:typ§: fertile line carrying the fertile N cytoplasm and the BM O-type nuclear background (no restorer genes) 7. BMQ;QM§: sterile line carrying the BMC sterile cytoplasm derived from Beta maritima and the BM O-type nuclear background (no restorer genes) 8. BMC-3,: fertile line carrying the BMC sterile cytoplasm but with nuclear restorer genes 24 Maine and Sugarbeet Genes Used as Probes Genes encoding the maize mitochondrial ATPase subunits 6 and 9 (atp6 and atp9, described in 5 and 6) were gifts from Dr. C.S. Levings III, Department of Genetics, North Carolina State University. The maize cytochrome oxidase subunit II gene (coxII; clone pZmEl described in 11) was a gift from Dr. Barbara Sears, Department of Botany and Plant Pathology, Michigan State University. The following fragments of the C1 Normal sugarbeet true coxII gene and coxII pseudogene (9) were prepared by varda Mann and were gifts from Dr. Joseph Hirschberg, The Hebrew University of Jerusalem, Israel. They are shown in Figure 1-2-1. EIQQ§_£1: BamHl fragment from base 350 to base 745 of the pseudogene clone (9). This sequence is immediately 3' to the pseudogene but is not present in or downstream of the true gene and is unique to the pseudogene. Engg_£;: BamHl-Sacl fragment from base 1141 to base 1606 of the coxII gene clone (9) including part of the intron and the entire second exon of the true gene. 'This sequence is present in the true gene but missing from the pseudogene and is therefore specific for the true gene. 32113.11: EcoRl-BamHl fragment from base 1 to base 350 of the pseudogene (9). This sequence is present as part of the first exon of the true gene as well as the pseudogene and therefore hybridizes to both. 25 Figure 1-2-1. Location of probes derived from C1 Normal sugarbeet coxII true gene and pseudogene. Restriction maps were generated from the sequences of Ekstein (9) . Dark boxes represent exons in the true coxII gene and the exon 1-like sequence in the pseudogene. The hatched box in the pseudogene map represents part of the reading frame not homologous to exon 1 of the true coxII gene. Restriction enzymes used: AvaII (A), BamHI (B), BanII (N), EcoRI (E), HincII (C), HindIII (H), ScaI (S). 26 .TN-.. 0.59“. ml; [“4 Z— m—1 m mcooocnmmo S80 F :98 / I L < m econ =on oat. 27 The maize genes and the S,4-CMS 1860bp coxII fragment cloned herein were labelled with 32P by nick-translation (19) . Sugarbeet coxII gene fragments from Dr. Hirschberg were labelled with ”P by random-priming (10). Isolation of Mitochondria, MtDNA, and MtRNA Beet roots were peeled and ground in isolation buffer (400mm sorbitol, 30mM MOPS pH 7.6, 1mM EDTA, 0.2% BSA, 41!!!! cysteine, 0.6% PVPP) . After filtration through Miracloth (Calbiochem), the filtrate was centrifuged at 1400g for 15 minutes, then the supernatant was centrifuged at 16,3009 for 15 minutes to pellet the mitochondria. The mitochondria were resuspended in wash buffer (350mM sucrose, 30mM MOPS pH 7.2, 1 mM EDTA, 0.2% BSA), repelleted as before, resuspended in wash buffer, and layered onto discontinuous sucrose density gradients for further purification (8). To isolate beet root mtDNA, purified mitochondria were treated with lysis buffer (SOmM Tris pH 8.0, 20mM EDTA, 0.5% sarkosyl, 200ug/mL proteinase K) at 37°C for 1 hour. The mitochondrial lysate was layered on 40mL of 4.4M CsCl solution containing 10mg ethidium bromide and centrifuged at 45,000rpm in the Beckman VTi50 rotor for 18 to 20 hours at 18°C. The DNA band was collected, extracted with n-butanol to remove the ethidium bromide, dialyzed against TE buffer (10mM Tris pH 8.0, 1mM EDTA) for 12 to 24 hours, precipitated with 2.5 volumes of absolute ethanol, and resuspended in sterile 28 distilled water. Electrophoresis through 0.9% agarose gels and Southern blotting were performed as described in Maniatis et a1. (19) . Hybridizations were carried out at 37°C in 50% formamide solution as described (19). To isolate beet root mtRNA, purified mitochondria were resuspended in MgCl2 buffer (12.5mM MgClz, 12.5mM Tris pH 7.6, 12.5mM KCl) containing 0.1% SDS, 3% (v/v) DEP, and 5% (v/v) vanadyl ribonucleoside complex and incubated at 14°C for 10 minutes. The solution was then extracted 3 to 4 times with equal volumes of MgCl2 buffer-saturated phenol containing 0.1% 8-hydroxyquinoline, then twice with equal volumes of chloroform. The mitochondrial RNA was ethanol-precipitated, washed twice, and resuspended in RNA buffer (20mM Tris pH 7.6, 1.8mM NaCl, 0.14mM EDTA) . Electrophoresis through 1% agarose gels containing formaldehyde as the denaturant was carried out essentially as described in Maniatis et a1. (19) . Hybridizations were carried out at 37°C in 50% formamide solution as described (19). Cloning of the S,4-CMS Sugarbeet coxII Gene Total mtDNA from S,4-CMS beet roots was restricted with BamHl endonuclease and fragments were ligated into the plasmid pUC19. The mixture was used to transform competent E. coli HB101 cells and ampicillin-resistant colonies were selected. The desired clones were identified using the colony hybridization procedure of Grunstein and Hogness (12) with the 29 maize coxII gene (11) as a probe. DNA from hybridizing colonies was extracted in a mini-plasmid prep procedure, then Southern analysis of the BamH1-digested DNA was used to confirm the presence of the desired insert (19). RESULTS Restriction Digests of C1 Normal and S,4-CMS MtDNA Restriction digests of mtDNA from fertile and male- sterile lines differed when probed with the three maize mitochondrial genes encoding respiratory enzyme subunits. MtDNA from C1 Normal and S,4-CMS beet roots was restricted with BamH1 endonuclease and subjected to Southern analysis using the maize genes atp6, atp9, and coxII as probes (Figure 1-2-2) . With the atp6 probe (left panel), two mtDNA fragments of approximately 7700bp and 1800bp hybridized in the S,4-CMS line. In the C1 Normal line, two fragments of approximately 7700bp and 1950bp hybridized. A 6100bp fragment hybridized in both lines when atp9 was used as the probe (center panel), but an additional 4200bp fragment was also present in the C1 Normal line. Using the maize coxII gene as a probe (Figure 1-2-2, right panel, part A), a fragment of approximately 1860bp hybridized in both lines. However, there were additional fragments of 3300bp in S,4-CMS beets and 1550bp in C1 Normal beets. A 480bp fragment was also present in both lines which 30 atp6 atp9 A B 1 2 12 1 2 1 2 Figure 1-2-2. Southern blot of sugarbeet mtDNA probed with atp6, atp9, and coxII. Sugarbeet root mtDNA was restricted with BamH1 endonuclease, electrophoresed, transferred to nitrocellulose, and probed with the maize atp6 (5; left panel) and atp9 (6; center panel) genes, the entire maize coxII gene (11; right panel, part A), and the 1.6kb BamH1 fragment of the maize coxII gene (11; right panel, part B). Lane 1: S,4-CMS mtDNA; lane 2: C1 Normal mtDNA. 31 electrophoresed off the gel in part A. This fragment is shown in part B, which was probed with a 1.6kb BamH1 fragment of the maize coxII gene that contains the latter part of the intron and the second exon. The 1550bp fragment of C1 Normal did not hybridize to this probe, although it did hybridize to a 0.8kb BamH1 fragment of the maize coxII gene that contains the first exon and the start of the intron (data not shown). Therefore the 1550bp C1 Normal fragment probably contains exon 1 of coxII and/or the start of the intron. All three S,4-CMS fragments hybridized to the 1.6kb maize probe (Figure 1-2-2, right panel, part B), but only the 1860bp fragment hybridized to the 0.8kb maize probe (data not shown). C1 Normal and S,4- CMS mtDNA were also restricted with enzymes other than BamH1 and probed with the coxII gene. No differences in restriction fragment patterns were seen in Sall- or Kpnl-digested mtDNA, but differences were seen when mtDNA was restricted with HindIII (data not shown). Cloning of the S,4-CMS coxII Gene The three BamH1 fragments from S,4-CMS mtDNA which hybridized to the maize coxII probe were cloned individually into the plasmid pUC19 and restriction mapped (Figure 1-2-3) . Two maize coxII gene fragments were used separately as probes to locate the start and end of the cloned S,4-CMS coxII gene: a 0.8kb BamH1 fragment which encodes exon 1 and about half of the intron; and a 1.6kb BamH1 fragment which encodes the 32 Figure 1-2-3. Restriction maps of three fragments of the S,4- CMS coxII gene and comparison to the C1 Normal coxII gene. The restriction map of the C1 Normal gene was generated from the sequence of Ekstein (9) . Dark boxes represent the two exons of the C1 Normal coxII gene. Restriction enzymes used: AvaII (A), BamHI (B), BanII (N), EcoRI (E), HincII (C), HindIII (H), PvuI (P), PvuII (V), PstI (T), ScaI (S). 33 .3; 2:9“. an 2: 9.0.0 38 I wzoima Escrow: gnome 1'1] m < < m 9.20 38 98.0300 92in 6 sesame 288 929% to 18592., 882 L 41] _ _ _ _ _ A _ _ _ u x L _ L m 0 I ._. > a 0 w 2m _/ < z w m m < .|.l_ m... 0 < w econ S66 .9502 5 34 second half of the intron and exon 2. The 1860bp S,4-CMS fragment hybridized to both maize probes (Figure 1-2-2, right panel, part 8; data not shown for the 0.8kb probe), indicating that this beet fragment contains sequences homologous to one of the exons and/or the intron. Comparison to a restriction map generated from the sequence of the C1 Normal coxII gene cloned by Ekstein (9) suggested that the S,4-CMS 1860bp fragment contains exon 1 and the first part of the intron (Figure 1-2-3). The 3300bp and 480bp S,4-CMS fragments hybridized only to the 1.6kb maize probe (Figure 1-2-2, right panel, part B), indicating that these beet fragments contain sequences homologous to exon 2 or the latter part of the intron. Comparison to the C1 Normal coxII map suggested that the 480bp fragment lies between the 1860bp and 3300bp fragments and encodes part of the intron. The 3300bp S,4-CMS fragment contains at least the latter part of the intron and the start of exon 2. Common restriction sites were not found in the region after the start of exon 2, which suggests a mtDNA rearrangement that has one breakpoint at this site has occurred in the S,4-CMS line. Northern Blots of Beet Root MtRNA - ' o - d Expression of the three mitochondrial respiratory enzyme complex genes was examined in the:y-irradiation-induced male- sterile S,4-CMS cytoplasm, which was reported to be like the 35 Owens-type male-sterile cytoplasm (20). Using the 1860bp fragment of the S,4-CMS coxII gene as a probe, different mtRNA transcripts were observed in fertile and male-sterile beet roots (Figure 1-2-4, left panel). Two transcripts of approximately 1700 nucleotides (nt) and 1200nt were observed in the fertile C1 Normal line (lane 1). The Owens-type male- sterile S line (lane 2) and the induced male-sterile S,4-CMS line (lane 3) possessed the same 1700nt transcript as the C1 Normal line, but also possessed a 2200nt transcript instead of the 1200nt transcript. The restored-fertile S,4-R, line (lane 4) possessed the same transcripts as the S,4-CMS line; the presence of the restorer genes apparently did not alter transcript sizes in the induced male-sterile cytoplasm. The maize atp6 and atp9 genes were used to probe mtRNA from the same fertile and male-sterile beet lines. Using the atp9 probe, the same 860nt transcript was found in all lines (Figure 1-2-4, right panel). The amount of transcript appeared less in the S line (lane 2), but this was due to less RNA loaded in that lane as judged by the stained gel (not shown). Using the atp6 probe (Figure 1-2-5, panel A), a strongly-hybridizing 2600nt transcript was seen in the C1 Normal line (lane 1) but little or no signal was detected in the S, S,4-CMS, and S,4-R, lines (lanes 2-4) . From the stained gel (panel B), it is apparent that less RNA or degradation of the RNA in these lanes is not an explanation for this result. Again the presence of R, genes in the S,4-R, line (lane 4) did not restore expression of the gene to the induced male-sterile 36 coxll atp9 1 2 3 4 1 2 3 4 Figure 1-2-4. Northern blots of sugarbeet mtRNA probed with coxII and atp9. Sugarbeet root mtRNA was electrophoresed, transferred to nitrocellulose, and probed with the 1860bp fragment of the S,4-CMS coxII clone (left panel) or the maize atp9 gene (6; right panel). Lane 1: C1 Normal mtRNA; lane 2: S mtRNA; lane 3: S,4-CMS mtRNA; lane 4: S,4-R, mtRNA. 37 1234 < z ‘E Figure 1-2-5. Northern blot of sugarbeet mtRNA probed with atp6 and stained gel. Sugarbeet root mtRNA was electrophoresed, transferred to nitrocellulose, and probed with the maize atp6 gene (5; panel A). Replicate lanes were concurrently electrophoresed, then stained with ethidium bromide (panel B). Lane 1: C1 Normal mtRNA; lane 2: S mtRNA; lane 3: S.4-CMS mtRNA; lane 4: S,4-Rf mtRNA; rRNA: E. coli 1 23S and 16S ribosomal RNA. 38 cytoplasm. Male-Sterile Cytoplasm Derived from Beta maritime The expression of atp6 and atp9 was also examined in the Beta maritime-derived male-sterile cytoplasm, which was reported to differ from the Owens-type cytoplasm (20). When a Northern blot was probed with the maize atp6 gene (Figure 1-2-6, left panel), an approximately 1120nt transcript hybridized in the fertile BM O-type line (lane 1). Two transcripts of 2800nt and 1300nt hybridized from the male- sterile BMC-CMS line (lane 2), while a single 2400nt transcript was present in the restored-fertiLe BMC-R, line (lane 3). In contrast, when probed with the maize atp9 gene (Figure 1-2-6, right.panel) all three beet lines.possessed.the same 600nt transcript. The C1 Normal coxII gene and pseudogene fragments were used to examine coxII expression in the Beta maritima-derived male-sterile cytoplasm, When probe #3 (a sequence present in the true coxII gene but missing from the pseudogene; see Figure 1-2-1) was used, a complex pattern of transcripts was observed (Figure 1-2-7, panel A). In the fertile C1 Normal line, the same 1700nt and 1200nt transcripts hybridized to probe #3 (Figure 1-2-7, panel A, lane 1) that hybridized to the 1860bp fragment of the S,4-CMS coxII gene (Figure 1-2-4, left panel, lane 1). _The Owens-type male-sterile C1-CMS line possessed identical transcripts (Figure 1-2-7, panel A, lane 2) to the C1 Normal line, which argues against coxII 39 atp6 atp9 1 2 3 1 2 3 Figure 1-2-6. Northern blots of sugarbeet mtRNA probed with atp6 and atp9. Sugarbeet root mtRNA was electrophoresed, transferred to nitrocellulose, and probed with the maize atp6 gene (5; left panel) or the maize atp9 gene (6; right panel). Lane 1: BM O-type mtRNA; lane 2: BMC-CMS mtRNA; lane 3: BMC- R, mtRNA. 4o probe3 probe4 1 2 3 4 5 ~1 2 3 4 5 - were" Figure 1-2-7. Northern blot of sugarbeet mtRNA probed with coxII gene and pseudogene. Sugarbeet mtRNA was electrophoresed, transferred to nitrocellulose, and probed first with probe 3 (sequence present in the true coxII gene but not the pseudogene; left panel), then with probe 4 (sequence present in both the true coxII gene and the pseudogene; right panel). Lane 1: C1 Normal mtRNA; lane 2: C1-CMS mtRNA; lane 3: BM O-type mtRNA; lane 4: BMC-CMS mtRNA; lane 5: BMC-R, mtRNA. 41 involvement in CMS in this beet line. In the fertile BM 0- type line, only the 1200nt transcript hybridized (lane 3). Two additional transcripts of approximately 1400nt and 1600nt hybridized in the male-sterile BMC-CMS line (lane 4), but the restored-fertile BMC-R, line (lane 5) possessed only the transcript present in the fertile BM O-type line (lane 3). The same Northern blot was reprobed with probe #4, which contains a sequence which is present in both the true coxII gene and the pseudogene (see Figure 1-2-1). The pattern of transcription found with probe #4 (Figure 1-2-7, panel B) was nearly identical to that found using probe #3 (Figure 1-2-7, panel A) except for the appearance in all lanes of an additional 420nt transcript. This suggests that the 420nt band represents expression of the pseudogene; yet no transcripts hybridized to probe #1, which encodes a region immediately downstream of the pseudogene (see Figure 1-2-1; data not shown). Gene Expression in Male Floral Tissues Examination of the expression of the genes encoding respiratory enzyme complex subunits was attempted using male floral tissues as well as roots, since it is in the male floral tissues that CMS is manifested. However, attempts to isolate intact RNA from buds, anthers, and pollen for Northern blotting were unsuccessful. Total nucleic acids were extracted from floral tissues from C1 Normal and C1-CMS plants 42 and used in slot blots (data not shown). However, differences in the levels of expression of these genes could not be quantified because insufficient amounts of nucleic acids were obtained to permit successful treatment with RNAse-free DNAse to selectively remove the DNA present in the samples. DISCUSSION Restriction Digests of MtDNA and the S,4-SMS coxII gene Differences in mtDNA restriction patterns between fertile C1 Normal and induced-sterile S,4-CMS sugarbeets when probed with the three maize mitochondrial genes indicate that some changes in or near these genes have occurred in the S,4-CMS mtDNA. These differences may be due to rearrangements such as those in maize (7) and Petunia ( 1) , but point mutations which add/delete restriction sites cannot be ruled out. The extent of the rearrangements or mutations cannot be inferred from the restriction digests, and restriction mapping alone cannot show if these differences are related to production of the CMS phenotype in the S,4-CMS line. If these three genes were present as single copies as reported for the Owens-type male- sterile cytoplasm (4) , then a rearrangement involving any one of the three genes which altered transcription could potentially cause mitochondrial dysfunction. However, the S line, which has been reported to be identical to the S,4-CMS line (20), possesses two copies of 43 the coxII gene (Dr. J. Hirschberg, personal communication) . Both copies contain a 480bp BamH1 fragment that contains intron sequences. In addition, copy 1 has two BamH1 fragments of approximately 1860bp; however, one is slightly longer and probably contains exon 1, while the other is slightly shorter and probably contains exon 2. Copy 2 has the longer "1860bp" fragment containing exon 1 plus a 3300bp fragment containing only part of exon 2 (although an open reading frame continues past the presumed site of mtDNA rearrangement) (Dr. J. Hirschberg, personal communication). The presence of these two coxII genes in the S,4-CMS line would explain the results of the Southern blots. The gels shown in Figure 1-2-2 apparently did not distinguish between the two slightly different sizes of "1860bp" fragments. Consequently, the 1860bp fragment seen here appeared to hybridize to both the 0.8kb maize probe containing exon 1 (data not shown) and the 1.6kb maize probe containing exon 2 (right panel, part B). However, probably one "1860bp" fragment (slightly longer and present in both copies) hybridized to the 0.8kb maize probe while the other "1860bp" fragment (slightly shorter and present only in copy 1) hybridized to the 1.6kb maize probe. The latter S,4-CMS fragment is probably equivalent to the 1860bp C1 Normal fragment that hybridized to the 1.6kb maize probe (right panel, part B) . Exon 1 would be carried on the 1550bp fragment in the C1 Normal line and the longer "1860bp" fragment in the S,4-CMS line. Thus the 1860bp fragments seen 44 in Figure 1-2-2 in the C1 Normal and S,4-CMS lines are not equivalent. Differences between the restriction map of S,4-CMS (Figure 1-2-3) and the restriction map of the C1 Normal coxII gene (9) suggest that a mtDNA rearrangement has occurred in the S,4-CMS coxII gene near the start of exon 2, but the changes would require sequencing to delineate precisely. These differences may have been caused by the y-irradiation that created the S,4-CMS cytoplasm. Since they found no differences in digests of mtDNA from the induced-sterile S,4-CMS line and the Owens-type sterile S line when probed with coxII, Mann et a1. (20) concluded that the induced-sterile cytoplasm was identical to the Owens-type sterile cytoplasm. The S,4-CMS line does seem to possess the two copies of the coxII gene that have been found in the S line (Dr. J. Hirschberg, personal communication) . Of the three S,4-CMS coxII fragments cloned herein, the 480bp fragment is the same in both copies; the 1860bp fragment apparently represents the longer "1860bp" fragment containing exon 1 that is also present in both copies. The 3300bp fragment is apparently from copy 2 and therefore contains only part of exon 2. The shorter "1860bp" fragment from copy 1, containing the entire exon 2, was not cloned herein. It is unknown if the S,4-CMS line also possesses the coxII pseudogene present in the C1 Normal line (9). 45 Expression of Respiratory Enzyme Genes in Beet Root In three sugarbeet lines, differences in coxII expression were correlated with the CMS phenotype. The male-sterile BMC- CMS line possessed two coxII transcripts not present in the related fertile BM O-type line. The male-sterile S and S,4- CMS lines possessed a 2200nt coxII transcript not present in the related fertile C1 Normal line, and lacked a 1200nt transcript present in the C1 Normal line. This suggests that alterations in the mtDNA in or near the coxII gene in the sterile cytoplasms (shown for S,4-CMS in Figure 1-2-2; reported for S in 20) are correlated with altered expression which could be CMS-related. However, different types of alterations were seen in the BMC-CMS line (additional coxII transcripts) and the S/S,4-CMS lines (change in the size of one transcript). This could mean that these differences are not necessarily CMS-related; alternatively, any type of change in coxII expression might contribute to CMS, as opposed to one specific alteration being necessary to cause the CMS phenotype. However, the male-sterile C1-CMS line, which was also derived from the C1 Normal line, possessed coxII transcripts identical to those of its fertile progenitor. That result shows that altered coxII expression cannot be the only cause of CMS in sugarbeets, although it may be CMS- related in some male-sterile lines. A 2600nt atp6 transcript was detected in the fertile C1 Normal sugarbeet line but not in the S, S,4-CMS, or S,4-R, 46 lines (Figure 1-2-5) . It is difficult to conceive of a plant existing without the atp6 gene product, an integral part of the mitochondrial ATP-synthesizing complex. Perhaps a low level of expression, not detectable in this experiment but sufficient for survival, was present in the three lines. Although mtDNA restriction patterns were not investigated in the S line, the S,4-CMS line possessed a mtDNA rearrangement in or near the atp6 gene which might have caused decreased expression. The atp6 transcripts present in the BM O-type, BMC-CMS, and BMC-R, lines (Figure 1-2-6, left panel) differed from each other and from the lines examined in Figure 1-2-5. It is therefore possible that the atp6 gene may be involved in the production of the CMS phenotype. No correlation could be made between atp9 expression in the induced male-sterile S,4-CMS line (Figure 1-2-4 panel B) and CMS despite differences in the mtDNA of the S,4-CMS line in or near the atp9 gene (Figure 1-2-2). Possibly the mutation or rearrangement in the S,4-CMS mtDNA was near but not in the atp9 gene, so that its transcription was not altered. This suggests that rearrangements can occur without detectable effects on gene expression, and therefore a mtDNA rearrangement present in a male-sterile line is not necessarily CMS-related. No alterations in atp9 expression were found in the BMC-CMS line either (Figure 1-2-6, right panel), which further suggests that this gene is not involved with CMS in sugarbeets. The effects of restorer genes on mitochondrial gene 47 expression in these experiments were mixed. Restorer genes present in the S,4-R, line did not change the coxII transcript sizes present in male-sterile cytoplasm (S,4-CMS line) back to those present in the fertile cytoplasm (C1 Normal line). This suggests that either the differences in coxII expression are not CMS-related, or that the restorer genes do not act at the level of transcription. They may instead alter translation of the messages. Such a situation was reported in Petunia, in which the single R, gene had no effect on the transcription of the CMS-associated pcf fusion (28) but decreased the abundance of the CMS-associated 25kD protein (22). Restorer genes also did not restore expression of the atp6 gene to the S,4-R, line in this experiment; they did alter the transcripts in the BMC- CMS cytoplasm, but not back.to the same size as in the fertile line (Figure 1-2-6, left panel). This argues against atp6 involvement with CMS in the S,4-CMS line, but leaves open this possibility for the BMC-CMS line. The presence of restorer genes in the BMC-R, line changed the coxII transcript sizes in the BMC cytoplasm from those present in the male-sterile BMC- CMS line to those present in the fertile BM O-type line. This supports the possibility that altered coxII expression in the BMC-CMS line is CMS-related, and that the restorer genes may function by counteracting this altered gene expression. A similar situation was reported in CMS-T maize, in which restorer genes decreased the levels of the CMS-associated T- urf13 transcripts (7,16). Since different R, genes restore fertility to each male-sterile sugarbeet cytoplasm, it is not 48 unexpected that the restorer genes may function in different ways in each case. Comparison of Figure 1-2-7 panels A and B suggests that the coxII pseudogene was expressed in all the sugarbeet lines examined. The 420nt transcript was present in the mtRNA of all lines when hybridized to probe #4, which contains a sequence common to the true coxII gene and the pseudogene, but was absent when the mtRNA was hybridized to probe #3, which contains a sequence present.in the true coxII gene but missing from the pseudogene. These data suggest that the 420nt transcript represents expression of the pseudogene. However, no transcripts hybridized to probe #1, which contains a sequence immediately downstream from the pseudogene. It is possible that the pseudogene transcript ends early in this region so that not enough of the region is transcribed to hybridize on a Northern blot, or perhaps this area is rapidly removed by a 3' processing event. Other sequences for use as pseudogene-specific probes which were unique to the pseudogene but within or upstream from it could not be obtained. The 420nt transcript was present in both fertile and male-sterile lines; this suggests that, unlike the CMS-associated expression of a coxII pseudogene in Petunia (27), expression of the sugarbeet pseudogene is not CMS-related. In summary, it is possible that mtDNA rearrangements in or near the cytochrome oxidase subunit II gene which cause altered cox II gene expression could be related to the production of the CMS phenotype in the S,4-CMS and BMC-CMS 49 sugarbeet lines. No correlation between CMS and atp9 expression, which appeared constant in all lines, could be found. Dramatic differences in atp6 gene expression were seen between lines which could be CMS-related. However, expression of a coxII pseudogene did not appear to correlate with CMS. Further investigation of coxII expression in anthers and pollen, the tissues in which CMS is manifested, was technically not feasible in this study. In situ hybridization would be the most promising technique with which to carry forward this study of coxII and atp6 expression and CMS in male floral tissues of sugarbeet. ACKNOILEDGEMENTS Thanks are due to Dr. Joseph Hirschberg for his advice and sharing of much unpublished data. Parts of Figure 1-2-7 have been submitted for publication in the following paper: Mann v, Ekstein I, Nissen H, Hiser C, McIntosh L, Hirschberg J (1991) The cytochrome oxidase II‘gene in mitochondria of the sugar beet Beta vulgaris L. Plant Mol Biol (in press) LITERATURE CITED 1. Boeshere ML, Hansen MR, Ishar S (1985) A variant mitochondrial DNA rearrangement specific to petunia stable sterile somatic hybrids. Plant Mol Biol 4:125-132 2 . Boutin V, Pannenbecker G, Ecke I, Schewe G, Saumiteu- Laprade P, Jean R, Vernet P, Michaelis G (1987) Cytoplasmic male sterility' and. nuclear’ restorer genes in. a natural population of Beta maritima: genetical and molecular aspects. Theor Appl Genet 73:625-629 50 3. Beutry M, Faber A-M, Charbonnier M, Briquet M (1984) Microanalysis of plant mitochondrial protein synthesis products: Detection of variant polypeptides associated with cytoplasmic male sterility. Plant Mol Biol 3:445-452 4. Brears T, Lensdale DM (1988) The sugarbeet mitochondrial genome: A complex organisation generated by homologous recombination. Mol Gen Genet 214:514-522 5. Dewey RE, Levings III CS, Timothy DH (1985a) Nucleotide sequence of the ATPase subunit 6 gene of maize mitochondria. Plant Physiol 79:914-919 6. Dewey RE, Schuster AM, Levings III CS, Timothy DH (1985b) Nucleotide sequence of Fo-ATPase proteolipid (subunit 9) gene of maize mitochondria. Proc Nat Acad Sci USA 82:1015-1019 7. Dewey RE, Levings III CS, Timothy DH (1986) Novel recombinations in the maize mitochondrial genome produce a unique transcriptional unit in the Texas male-sterile cytoplasm. Cell 44:439-444 8. Deuce R, Christiansen EL, Bonner ND (1972) Preparation of intact plant mitochondria. Biochim Biophys Acta 275:148-160 9. Ekstein I (1990) Molecular analysis of the gene for subunit II of cytochrome oxidase in the fertile and cytoplasmic male- sterile genotypes of the sugar beet Beta vulgaris L. Thesis submitted for the degree of "Master of Science" under the supervision of Dr. Joseph Hirschberg, Department of Biotechnology, The Hebrew University of Jerusalem 10. Feinberg AP, vogelstein B (1982) A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6-13 11. Fox TD, Leaver CJ (1981) The Zea mays mitochondrial gene coding' cytochrome oxidase. subunit. II has an intervening sequence and does not contain TGA codonsQ Cell 26:315-323 12. Grunstein M, Hbgnese DS (1975) Colony hybridization: A method for the isolation of cloned DNAs that contain a specific gene. Proc Nat Acad Sci USA 72:3961-3965 13. Hallden C, Bryngelssen T, Bosemark NO ( 1988) Two new types of cytoplasmic male sterility found in wild Beta beets. Theor Appl Genet 75:561-568 14. Hansen BM, Marcker KA (1984) DNA sequence and transcription of a DNA minicircle isolated from male-fertile sugar beet mitochondria. Nuc Acids Res 12:4747-4756 51 15. Jiayang L, Jigeng L (1986) Chloroplast thylakoid membrane polypeptides and cytoplasmic. male sterility. Curr Genet 10:947-949 16. Kennell JC, Wise RP, Pring DR (1987) Influence of nuclear background on transcription of a maize mitochondrial region associated with Texas male sterile cytoplasm. Mol Gen Genet 210:399-406 17. Kineshita T (1977) Genetic relationship between pollen fertility restoring genes and cytoplasmic factors in male sterile mutants of sugar beets. Jap J Breeding 27:19-27 18. Laughnan JR, Gabay-Laughnan S (1983) Cytoplasmic male sterility in maize. Ann Rev Genet 17:27-48 19. Maniatis T, Fritsch EF, Sambreek J (1982) Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory, New York 20. Mann v, McIntosh L, Theurer C, Hirschberg J (1989) A new cytoplasmic male sterile genotype in the sugar beet Beta vulgaris L.: A molecular analysis. Theor Appl Genet 78:293-297 21. Mikami T, Harada T, Kineshita T (1986) Heterogeneity of circular mitochondrial DNA molecules from sugar beet with normal and male sterile cytoplasms. Curr Genet 10:695-700 22. Nivison HT, Hansen MR (1989) Identification of a mitochondrial protein associated with cytoplasmic male sterility in petunia. Plant Cell 1:1121-1130 23. Owen FV (1945) Cytoplasmically inherited male-sterility in sugar beets. J Agricultural Res 71:423-440 24. Powling A (1981) Species of small DNA molecules found in mitochondria from sugarbeet with normal and male sterile cytoplasms. Mol Gen Genet 183:82-84 25. Powling A (1982) Restriction endonuclease analysis of mitochondrial DNA from sugarbeet.with normal and male-sterile cytoplasms. Heredity 49:117-120 26. Powling A, Ellis THN (1983) Studies on the organelle genomes of sugarbeet with male-fertile and male-sterile cytoplasms. Theor Appl Genet 65:323-328 27. Pruitt M, Hansen MR (1989) Cytochrome oxidase subunit II sequences in Petunia mitochondria: Two intron-containing genes and an intron-less pseudogene associated with cytoplasmic male sterility. Curr Genet 16:281-291 52 28. Rasmussen J, Hanson MR (1989) A NADH dehydrogenase subunit gene is co-transcribed with the abnormal Petunia mitochondrial gene associated with cytoplasmic male sterility. Mol Gen Genet 215:332-336 29. Smith GA (1980) Sugarbeet. In: Hybridization of crop plants. American Society of Agronomy-Crop Science Society of America, Madison, Wisconsin, pp.601-616 30. Stander JR (1973) Electron-microscopic and polarographic studies of cytoplasmic male sterility in Beta vulgaris L. A thesis presented to the faculty of the graduate school of Cornell University for the degree of Doctor of Philosophy 31. Thomas CM (1986) The nucleotide sequence and transcription of minicircular mitochondrial DNA's associated with male- fertile sugar beet mitochondria. Nuc Acids Res 14:9353-9370 32. Young EG, Hanson MR (1987) A fused mitochondrial gene associated with cytoplasmic male sterility is developmentally regulated. Cell 50:41-49 33. Rue Y, Davies DR, Thomas CM (1990) Sugarbeet mitochondria contain an open reading frame showing extensive homology to the subunit 2 gene of the NADH:Ubiquinone reductase complex. Mol Gen Genet 221:195-198 PART 2: GMC AND DEVELOPMENTAL ASPECTS OF ALTERNATIVE OXIDASB 53 CHAPTER 2-1: INTRODUCTION TO ALTERNATIVE OXIDASE HI STORY In addition to cyanide-sensitive respiration mediated by the cytochrome pathway of electron transport, plant mitochondria possess cyanide-resistant respiration. Cyanide- resistant respiration was reported as early as 1929 in sweet peas by Genevois (referenced in 47). Occurrence of cyanide- resistant respiration is widespread among higher plants, as well as being present in many algae, fungi, and even a few protozoa ( 31) . Localization of cyanide-resistant respiration to the mitochondria was initially observed for Arum maculatum by James and Elliot in 1955 (37), and since then has been measured in isolated mitochondria in many plant species (41) . This cyanide-resistant respiration is mediated by what has become known as the alternative pathway of electron transport. INHIBITORS OF THE ALTERNATIVE PATHWAY The alternative pathway was discovered by its resistance to inhibitors of the classical cytochrome pathway such as cyanide, azide, and carbon monoxide (which inhibit cytochrome oxidase) , and antimycin A (which inhibits the cytochrome b-c1 54 55 complex) (18,47). In 1971, Schonbaum et a1. (77) showed that substituted hydroxamic acids were specific inhibitors of the alternative pathway. Salicylhydroxamic acid (SHAM) , which has become the most widely used inhibitor of the alternative pathway (47,79) , shows 50% inhibition in isolated mitochondria at a concentration of 0.26mM (77). The mechanism of hydroxamate inhibition is believed to be competition for the substrate quinone binding site of the alternative oxidase (11,79). SHAM has been shown to also inhibit lipoxygenase (65,78,82), tyrosinase, and.some.other oxidases (56), and SHAM stimulation of peroxidases has been reported (14,83,91) . These other enzyme .activities, ‘which can interfere ‘with alternative path measurements, can be eliminated by density gradient purification of mitochondria (12,58,78,82,91). SHAM has recently been reported to inhibit photosynthesis in green tissues (14) , but alternative pathway respiration can be measured in the dark or the mitochondria can be separated from the chloroplasts by percoll gradient purification (58). SHAM has been reported to inhibit the cytochrome pathway at the high concentrations required for penetration into intact tissues, but not at. the ‘usual concentrations used. with isolated mitochondria (44). Two other strong inhibitors of the alternative pathway have also been found. Propyl gallate inhibits at or near the same site as SHAM and its mechanism of inhibition may be similar 'to ‘that. of Ihydroxamates (given 'their structural similarity), but 50% inhibition is achieved at 2-5uM 56 concentrations (82) . Like SHAM, it also inhibits lipoxygenase and peroxidases (56,65). Disulfiram (tetraethylthiuram disulfide) binds at a different site than SHAM and 50% inhibition is achieved at 5-10uM concentrations (28). Its mechanism of inhibition is uncertain, but may involve metal ion chelation (11) or sulfhydryl reactions (28). Disulfiram has the advantage of not inhibiting lipoxygenase (54), but it cannot be used in intact tissues (28). Besides the three main inhibitors, other compounds such as terpenes and herbicides have been reported to affect the alternative pathway, many of which also affect the cytochrome pathway (47) . Notably, high concentrations of cytokinins strongly inhibit the alternative pathway (16,53), which has led some researchers to postulate involvement of the alternative path in cytokinin responses (61,63) . However, some anticytokinins are also strong inhibitors of the alternative path (16), which suggests that inhibition of the alternative pathway by unphysiologically high concentrations of cytokinins may not be related to their biological activity (47). PROPERTIES 01' THE ALTERNATIVE PATHWAY The alternative pathway has long been considered a side branch from the cytochrome pathway. The branch point could be roughly localized from inhibitor studies to be after complex I and before the cytochrome b-c1 complex, because alternative 57 path respiration caused reduced P/O ratios but was resistant to antimycin A (11,46,47,88). Storey (87) proposed in 1976 that ubiquinone served as the branch point, a hypothesis that received ample support from several EPR studies (reviewed in 11,79) and accounts of partial quinone extraction from mitochondria causing loss of both pathways (15,33) . This view has now become widely accepted (46) . The mechanism of interaction of ubiquinone with the alternative pathway is unknown. Ubiquinone is generally believed to pass electrons directly to an alternative terminal oxidase protein or complex without intermediate electron carriers ( 47) . One report of a flavoprotein intermediate carrier ( 87) has received no further supporting data. To explain differing sensitivities of respiration to inhibitors depending upon the exogenous substrate given to the mitochondria, multiple quinone pools have been proposed which are not in equilibrium with each other and not equally accessible to the alternative pathway (11,47,74,79) . Other authors have maintained that alternative pathway behavior could be explained by a single quinone pool (18,19). Deviations from "ideal Q-pool behavior" may be due to slow or restricted quinone diffusion or non-random distribution of respiratory enzymes preventing complete randomization of the quinone pool (59,66,71). This point remains controversial. Since the alternative pathway branches from the cytochrome pathway at ubiquinone, it bypasses two energy- conserving sites: the b-c1 complex and cytochrome oxidase. 58 Wilson (93,94) has contended that the alternative pathway itself can support energy-linked functions including phosphorylation, but other authors have presented definitive evidence that it cannot support a mitochondrial membrane potential or phosphorylation (1,57,60) . The alternative pathway per se is generally considered non-energy-conserving, although electrons flowing through the NADH dehydrogenase may still contribute to the electrochemical gradient across the inner mitochondrial membrane during cyanide-resistant respiration (11,47,81). The end product of the alternative pathway could be superoxide, peroxide, or water, depending on whether one, two, or four electrons were donated to oxygen (11). Peroxide was once favored as the likely product by some (88) , but the general consensus now is that the end product is water (47,49,59,79) . This was demonstrated when artificial electron donors to the alternative oxidase were found and the oxidase was first solubilized (34,70). NATURE OF THE ALTERNATIVE PATHWAY TERMINAL OXIDASE Free Radical Hypothesis Rustin et a1. (73) proposed that cyanide-resistant respiration did not result from a terminal oxidase, but was rather an artifact of free radical formation. Fatty acid radicals and peroxyradicals can be generated in membranes and 59 can react with reduced quinones, using oxygen in a cyanide- insensitive manner (47,73). Alternative pathway inhibitors were proposed to act as free radical scavengers (73), but the demonstration of a discrete inhibitor binding site in the alternative pathway (84) argues against this. This hypothesis cannot explain why conditions which could promote non- enzymatic fatty acid radical oxidation do not increase alternative path activity (59), nor can it explain induction or regulation of alternative path respiration in certain plants at certain developmental stages (47) . Such observations of alternative oxidase activity suggest an enzymic origin rather than a free radical mechanism (59). Lipoxygenase Hypothesis Lipoxygenase was once proposed to constitute the cyanide- resistant respiration of higher plants (26,65). Its activity can be confused with the alternative oxidase because it consumes oxygen in a cyanide-resistant manner and is inhibited by SHAM and propyl gallate (65). Lipoxygenase activity can.be high in lipid-rich tissues (12), although it is not believed to contribute significantly to oxygen uptake in intact roots and leaves (44). Lipoxygenase was reported to be present in washed. mitochondrial preparations from several different species (82); however, further purification of mitochondria through percoll (12,58,78) and sucrose (82,91) gradients removed lipoxygenase contamination without removing 60 cyanide-resistant, SHAM-sensitive oxygen uptake. Furthermore, lipoxygenase is not inhibited by disulfiram, so this inhibitor can be used to distinguish lipoxygenase from true alternative path respiration (54,79) . It is therefore clear that the alternative oxidase is not lipoxygenase, although care must be taken in preparing mitochondria to avoid contamination by this enzyme. Terminal Oxidase Protein Hypotheses Van Herk originally suggested in 1937 (referenced in 47) that an autooxidizable flavoprotein was the alternative terminal oxidase in Sauromatum guttatum. James and Beevers (36) also concluded that the alternative oxidase was a flavoprotein based on their observations that aroid spadices have a high flavoprotein content which increases as the respiration rate increases. It has been argued, however, that the affinity of the alternative oxidase for oxygen is too low and the kinetics of oxidation are too slow for a flavoprotein (47) . No EPR signals indicative of a flavoprotein alternative oxidase have been found (35,72); indeed, no characteristic spectral signals of any sort have been reported for the alternative oxidase (reviewed in 79). Upon discovery of iron-sulfur proteins, it was proposed that the alternative oxidase could be a non-heme iron protein (5) . The alternative pathway has been reported to be inhibited by iron chelators (summarized in 11) to and release 61 sulfur upon heating (10). However, there appears to be an equal distribution of iron and acid-labile sulfur in mitochondria which do and do not possess alternative pathway capacity (47), and no EPR signals for non-heme iron proteins could be associated with the presence of the alternative oxidase (35,72; reviewed in 11,47). Nevertheless, overwhelming evidence suggests that alternative respiration proceeds via an alternative terminal oxidase protein or protein complex of some sort. Such evidence includes the thermal lability (8,21,80) and trypsin sensitivity (21,68) of the alternative oxidase activity. The existence of discrete inhibitor binding sites (84) and the characteristics of inhibition (28,38,81) also strongly suggest the involvement of protein(s) . Finally, there have been several reports of its solubilization and partial purification (8,21,35,38,?2). CHARACTERISTICS OF THE ALTERMTIVE OXIDASE PROTEIN(S) That the alternative jpathway' consisted. of‘ a quinol oxidase was first suggested by Bonner and Rich in 1978 (7). It has been solublized by detergents from Arum maculatum (8,35,38,72) and Sauromatum guttatum (21). Purification of the alternative oxidaseahas been.difficult.due to its lability (8,80) and proteases present in some aroid spadices (80). Except herein (32), there have been no reports of solubilization or purification of the alternative oxidase from 62 non-aroid plants, in which the levels of alternative oxidase capacity are considerably lower. Metal analyses of solublized alternative oxidases from aroid plants have been scarce and contradictory. In 1978, Huq and Palmer (35) reported that a Lubrol-solublized alternative oxidase activity from A. maculatum contained copper but no other metals. In the same year, Rich (70) reported deoxycholate solublization of the A. maculatum alternative oxidase activity. EPR analyses (70,72) found no evidence for any redox-active iron- or copper- containing species. More recently, Bonner et al. (8) purified the A. maculatum quinol oxidase activity through density gradients, and only iron copurified with the alternative oxidase fraction. Iron may be somehow associated with the alternative oxidase; yet the type and mechanism of association remains to be elucidated. The polypeptide size and composition of the alternative oxidase have recently been investigated. The Coomassie- stained SDS-PAGE gels of Bonner et a1. (8) showed a 34kD protein which copurified with the quinol oxidase activity of A. maculatum. Radiation inactivation analysis of the alternative oxidases of Symplocarpus foetidus and S. guttatum indicated functional molecular weights ranging from 26kD to 38kD (6). Elthon and McIntosh (21) found three polypeptides of 35kb, 36kD, and 37kD which copurified with the S. guttatum alternative path capacity, to which they raised polyclonal and monoclonal antibodies (22,23) . The antibodies which inhibited alternative path respiration also immunoprecipitated these 63 proteins, and induction of the alternative pathway with "calorigen" (see later) was correlated with the appearance of the 35kD and 36kD proteins (22) . The antibodies cross-reacted with proteins of similar sizes from a variety of aroid and non-aroid higher plants (22,23,27,32). The monoclonal antibodies also identified 36.5kD and 37kD proteins associated with alternative path capacity in the fungus Neurospora crassa (45) . The appearance of a 36kD protein has also been correlated with an increase in cyanide-resistant respiration in the yeast Hansenula anomala (55,97). The orientation of the alternative oxidase in the inner mitochondrial membrane has, until recently, been unclear. Schonbaum et a1. (77) believed that hydroxamate inhibitors could not cross the inner mitochondrial membrane, and therefore concluded that the alternative pathway was on the outer face. Dizengremel (15) suggested that the alternative oxidase was located on the outer surface because it was inhibited by concentrations of Triton X-100 that did not affect the cytochrome oxidase. Many authors have placed the alternative oxidase on the inner surface of the inner membrane, on the basis of putative functional links between the alternative oxidase and succinate dehydrogenase (47) or the inner membrane rotenone-insensitive NADH dehydrogenase (81). Recently, Rasmussen et a1. (68) used submitochondrial particles to show that a trypsin-sensitive component of the alternative oxidase is located on the inner surface of the inner mitochondrial membrane of A. maculatum. 64 GENETICS OP THE ALTERNATIVE PATHWAY Several studies suggest that at least one component of the alternative pathway is encoded in the nucleus. Alternative path capacity can be induced by inhibitors of the cytochrome pathway in the yeast Hansenula anomala (55) and this induction is blocked by cycloheximide, an inhibitor of cytoplasmic protein synthesis. The alternative path can be similarly induced in the fungus Neurospora crassa (45); this induction is also blocked by cycloheximide but not by chloramphenicol, an inhibitor of mitochondrial protein synthesis. Induction of alternative path capacity by aging in potato tuber slices is likewise blocked by cycloheximide but not chloramphenicol ( 17) . Actinomycin D, an inhibitor of nuclear transcription, blocks the appearance of CN-resistant respiration in yeast and fungi (11). These results suggest that some component of the alternative pathway is transcribed from a nuclear gene and synthesized in the cytoplasm in response to these induction processes. Mutations in Neurospora have identified a nuclear-encoded structural gene for the alternative oxidase, plus a second nuclear gene that may be regulatory (45). Recently, a nuclear-encoded structural gene for the S. guttatum alternative oxidase has been cloned and sequenced (69). Although the structural genes for the alternative oxidase appear to be nuclear-encoded in several organisms, some authors suggest that the mitochondrial genome can regulate the 65 pathway. Musgrave et a1. (62) reported maternal inheritance of the alternative path in crosses between two pea cultivars, which was attributed to a mitochondrially-encoded regulatory factor (81). Stegink and Siedow (84) also proposed that a mitochondrial "engaging factor" was necessary to couple electron flow from ubiquinone to the alternative oxidase, again based on the work of Musgrave et al. (62). However, the results of’ Musgrave. et .al. (62) have been contradicted (27,64), and no mitochondrially-encoded regulatory factor(s) or gene(s) have been isolated thus far. PARTITIONING OP ELECTRONS TO THE ALTERNATIVE PATHWAY Capacity and Activity Since the alternative pathway is a:non-energy-conserving bypass around the energy-conserving cytochrome pathway, the mechanism(s) of the distribution of electrons between the two pathways may be important to the energy balance of the plant cell. Measurement of the maximum capacity of either pathway is relatively straightforward; respiratory control is removed by uncoupling the mitochondria, then inhibitors are added sequentially to measure how much each inhibits oxygen uptake. The cytochrome path capacity is the difference between uninhibited respiration and respiration in the presence of a cytochrome oxidase inhibitor (generally KCN) . The alternative path capacity is the difference between respiration in the 66 presence of KCN and respiration in the presence of KCN plus an alternative path inhibitor (generally SHAM). This, however, is a measure of the maximum possible rate of electron flow through the alternative pathway using a given exogenous substrate ("capacity"); it.does not measure the actual rate of electron flow in vivo ("activity"). There is usually some portion of respiration insensitive to both KCN and SHAM. In intact tissues, this residual respiration may represent oxygen uptake by other enzymes or non-enzymological processes; in isolated mitochondria, it may be explained by inhibitors not being 100% efficient at the concentrations used (47). The relationship between activity and capacity is unclear. The technique of oxygen isotope discrimination has recently been applied to this question (29) and offers the first non-invasive method of measuring activity. The levels of activity found by Guy et a1. (29) parallelled the capacities measured by traditional inhibitor studies in isolated mitochondria of Saccharomyces cerevisiae , Symplocarpus foetidus, and castor bean, in Asparagus sprengeri mesophyll cells, and in whole alfalfa sprouts. However, limited oxygen diffusion presented a difficulty in the application. of this technique ‘to ‘tissue slices from .S. foetidus and.S. guttatum (29). Guy et a1. (29) also attribute the low values of isotope discrimination reported for carrot and potato storage tissue (48) to limited oxygen diffusion. 67 Measurement of Activity Three methods have been used to measure the partitioning of electrons between the cytochrome and alternative pathways. It has been generally believed that electrons cannot be shunted from the alternative path to the cytochrome path by adding alternative path inhibitors. It should therefore be possible to estimate alternative path activity simply by adding SHAM in the absence of KCN (11,41). The ADP/O ratio method is based on the idea that electron flow through the alternative path will lower the efficiency of oxidative phosphorylation ( 46) . The ADP/O ratio is first measured without inhibitors, then the electron distribution is altered with an inhibitor and the ADP/O ratio remeasured ( 46) . These ADP/O ratios can be used to calculate the fraction of electrons flowing through each pathway. The drawback of this technique is that it requires isolated mitochondria that are respiring in state 3 (not ADP-limited) and able to phosphorylate. It cannot be used for state 4 respiration (ADP- limited), whole cells, or intact tissues (47). The Bahr-Bonner inhibitor titration method (3) measures the rates of respiration over a range of SHAM concentrations with and without KCN. The observed respiration rate at each SHAM concentration in the absence of KCN (V,) is plotted against the respiration rate at each SHAM concentration in the presence of KCN (g[i]) (3). Assuming that cytochrome path respiration is always maximal and not affected by changes in 68 alternative path respiration, the plot should be a straight line described by the equation: Vt = p g[i] + cht (3) . The contribution of the cytochrome path, cht, is obtained by extrapolation of the line to g[i] ==0. 'The slope of the line, p, represents the "engagementfi of’the.alternativejpath.and.can vary from 0 (not engaged; no electrons flowing down the alternative path) to 1 (fully engaged; maximum possible electron flow through the alternative path) (3,4). However, values of p > 1 have been obtained, which Lance et a1. (47) attribute to SHAM losing its specificity. One can do a reverse titration to obtain an "engagement" of the cytochrome path, contrary to the basic assumption of this technique (47); some authors who have done so have found the cytochrome path to be fully engaged as generally assumed (11). IHowever, Moore and Rich (59) have questioned on theoretical grounds the assumption that electron flow through the alternative path does not affect the cytochrome path. Wilson (95) has recently reported that electrons can be shunted from the alternative path to the cytochrome path, implying that inhibitor titrations may not always be valid. Theories of Electron Partitioning After analyzing inhibitor titration data, Bahr and Bonner (4) reported that the cytochrome path was always expressed to its fullest capacity and was unaffected by the presence or absence of alternative path activity. From these 69 observations, the idea evolved that electrons only flow through the alternative path when the cytochrome path is saturated.(such.as by high levels.of substrates or uncoupling) or constricted (such.as.during state 4:respiration or when.KCN is present) (49). Hewever, using an electrode capable of measuring the level of quinone reduction, Dry et a1. (20) recently demonstrated electron flow through the alternative path when the quinone pool was only 35%-40% reduced. This suggests that electrons may be partitioned to the alternative path before the cytochrome path is saturated (20). A different model of electron partitioning posed by DeTroostembergh and Nyns (13) suggested that electron flux through each pathway was proportional to its respective capacity to accept electrons, with no preference given to the cytochrome pathway. This model implies that a constant fraction of mitochondrial respiration does not conserve energy (47), and.was criticized.as failing to fit the observations in plant tissues and mitochondria (49). In light of the quinone electrode data (20), however, the general acceptance of the Bahr-Bonner model to the exclusion of others should perhaps be reexamined. 70 FUNCTIONS OF THE ALTERNATIVE PATHWAY Function in Thermogenic Aroid Plants The alternative pathway has a defined role in the life cycle of thermogenic aroids. These plants produce an inflorescence which can heat to as much as 12-14°C above ambient temperature, volatilizing compounds to attract insect pollinators (52). These respiratory events are triggered by "calorigen", the active component of which is salicylic acid (67), which is produced in the male floral region and moves into the thermogenic tissue. During flowering, there is rapid breakdown of accumulated starch in the inflorescence and rapid respiration which is predominantly cyanide-resistant (52) . In Sauromatum guttatum, the temperature rise has been correlated with increasing alternative path capacity and concomitant shutdown of the cytochrome pathway, forcing electrons through the alternative pathway to generate heat (24). The alternative pathway is therefore used selectively to generate heat in floral tissue during flowering. Potential Functions in Non-aroid Plants Unlike the case of thermogenesis in aroid plants, the function(s) of the alternative pathway in non-aroid higher plants have not been well defined. The innate respiration rates in most plants are too low to produce significant 71 amounts of heat even if the alternative path were operating (81). Therefore other purposes for the alternative pathway have been suggested. An early suggestion for a role of the alternative pathway was to permit respiration in plants that contain cyanogenic glycosides from which cyanide could be released. Hewever, endogenous levels of cyanide are difficult to measure and.may not be appreciable, and cyanogenic plants have specific enzymes which could rapidly metabolize any cyanide that might be formed (46). There has also been no clear correlation between plants that are cyanogenic and the presence of the alternative path (81). Another suggested role for the alternative pathway has been protection against cold stress (reviewed in 43). It has been proposed that the cytochrome pathway is more sensitive to cold-induced changes in membrane fluidity than the alternative pathway, so that during chilling electrons are diverted to the alternative path (39,51) . One group has reported higher alternative path respiration in an arctic plant than in some temperate plant species (51). Higher alternative path capacity in cold-resistant cultivars than in cold-sensitive cultivars was reported for wheat (92) and maize (50), but other authors found no such differences in these species (75,86). Increased alternative path respiration was reported in chilled cucumber hypocotyls (39) and potato callus (30), and in cold-grown rapeseed (76) and maize (85,86). Because alternative path capacity increases when potato 72 tubers are sliced and aged, it has been invoked as part of the "wound-induced respiration" (17,89) . Theologis and Laties (90), however, found that the alternative pathway was not engaged during the respiration of fresh or aged slices in 9 of 12 tubers and fruits tested. The precise role of the alternative pathway in wound responses therefore remains unknown. A possible function of the alternative pathway in responses to pathogen attack has been suggested. Salicylic acid, an inducer of alternative path respiration, increases during pathogen attack and induces pathogenesis-related proteins (Ilya Raskin, personal communication) . However, the results of one report linking the alternative pathway with production of stress metabolites following infection (2) have been questioned (43,49). Alternative path capacity has been shown to increase during the climacteric of ripening fruits. Although it can be elicited by cyanide, it is unknown if the climacteric definitely requires the development of cyanide-resistant respiration (11). Theologis and Laties (89) found that, during the climacteric of banana and avocado slices, the alternative pathway was present but not engaged. However, a recent report attributed increased internal temperature in intact ripening mangoes to a concomitant increase in cyanide- resistant respiration (40). One report showed that the earliest respiration in imbibing seeds was strongly cyanide-resistant while later 73 respiration was cyanide-sensitive (96), which suggested that the alternative pathway contributes something essential to early germination. Other reports showed the reverse correlation (9,82). However, lipoxygenase activity may be a complicating factor in germination studies (41,65). There is as yet no clear evidence that alternative path capacity or activity is required for germination (43). The alternative path.has been suggested to play a role in adaptation of roots to their environment. Lambers (41,43) suggested the decrease in alternative path respiration when roots are transferred from distilled water to high salt solution was in direct parallel to the increase in sorbitol synthesis by the roots to increase the intercellular osmoticum. The increased demand for energy left no substrates for respiration in excess of what the TCA cycle and the cytochrome pathway could process and therefore the alternative pathway was not engaged. Lambers (41) suggested similar reasoning could explain the lack of alternative path activity in taproots accumulating storage carbohydrates. Observations that changes in the nutrient media affected the alternative path in roots led to the idea that it was somehow involved in root ion uptake (reviewed in 43) , but this correlation is still debatable (41,49). Leaf mitochondria actively and preferentially oxidize glycine as part of photorespiration. Consequently, a role has been suggested for the alternative pathway in the removal of excess reducing equivalents during photorespiratory glycine 74 oxidation. One report suggested electrons generated during glycine oxidation in greening soybean cotyledon mitochondria were channelled predominantly through the alternative pathway, although the cytochrome pathway was utilized in lupin mitochondria (25). A. recent. report, however, found. no evidence of a link between glycine oxidation and the alternative pathway in peas (27). Lance and coworkers have championed the idea that the alternative pathway plays a special role in the oxidation of malate (46,47). Malate oxidation.by'malate dehydrogenase was reported to be strongly cyanide-sensitive, while oxidation by malic enzyme was reportedly linked to the alternative pathway (74). Flexibility to oxidize malate independently of the cell energy charge might be useful in C4 and CAM plants to rapidly oxidize large amounts of malate (47). Laties (49), however, has questioned the existence of such a link. From the above observations, two broad ideas about the general function of the alternative pathway have arisen. In the "overflow hypothesis" proposed by Lambers (41,42) , the alternative pathway is engaged only when the cytochrome pathway is saturated. It therefore functions to "waste" any sugars in excess of the immediate needs. This hypothesis was supported by the work of Musgrave et a1. (62), who reported increased productivity under C02 enrichment of hybrid pea plants lacking the alternative path relative to plants possessing it (but see Chapter 2-2) . The suggestion of wasteful oxidation of sucrose, however, has been criticized 75 (19). The second hypothesis suggests the alternative pathway functions to eliminate reducing power and/or keep the TCA cycle operating for the synthesis of intermediates whenever the energy charge of the cell is high enough to inhibit the cytochrome pathway (19,46) . Both ideas are based on the concept that the alternative path is not engaged unless the cytochrome path is saturated or constricted. Recent evidence that electrbns could flow through the alternative path when the quinone pool is only 35-40% reduced (20) suggests that this concept may be an oversimplification. LITERATURE CITED 1. Akimenke FR, Gelovchenke NP, Medentsev AG (1979) The absence of energy conservation coupled with electron transfer via the alternative pathway in cyanide- resistant yeast mitochondria. Biochim Biophys Acta 545:398-403 2. Alves LM, Heister EG, Riseenger JC, Patterson LM III, Kalan EB (1979) Effects of controlled atmospheres on production of sesquiterpenoid stress metabolites by white potato tuber. Possible involvement of cyanide-resistant respiration. Plant Physiol 63:359-362 3. Bahr JT, Bonner ND Jr (1973) Cyanide-insensitive respiration. I.The steady states of skunk cabbage spadix and bean hypocotyl mitochondria. J Biol Chem248:3441-3445 4. Bahr JT, Bonner ID Jr (1973) Cyanide-insensitive respiration. II.Control of the alternate pathway. J Biol Chem 248:3446-3450 5. Bendall DS, Bonner ID Jr (1971) Cyanide-insensitive respiration in plant mitochondria. Plant Physiol 47:236-245 6. Berthold DA, Fluke DJ, Siedow JN (1988) Determination of molecular mass of the aroid alternative oxidase by radiation- inactivation analysis. Biochem J 252:73-77 76 7. Bonner Jr MD, Rich PR (1978) Molecular aspects of cyanide/antimycin resistant respiration. In: G Ducet, C Lance, eds, Plant Mitochondria. Elsevier/North-Holland Biomedical Press, New York, pp.241-247 8. Bonner Jr MD, Clarke SD, Rich PR (1986) Partial purification and characterization of the quinol oxidase activity of Arum maculatum mitochondria. Plant Physiol 80:838-842 9. Burguille PF, Nicholas G (1977) Appearance of alternate pathway cyanide-resistant during germination of seeds of Cicer arietinum. Plant Physiol 60:524-527 10. Chauveau M, Dizengremel P, Lance C (1978) Thermolability of the alternative electron transport pathway in higher plant mitochondria. Physiol Plant 42:214-220 11. Day DA, Arron GP, Laties GG (1980) Nature and control of respiratory pathways in plants: The interaction of cyanide- resistant respiration with the cyanide-sensitive pathway; In: DD Davies, ed, The Biochemistry of Plants, Vol.2. Academic Press, New York, pp.197-241 12. Day DA, Moore AL, Dry IB, liskich JT, Ascen-Bieto J (1988) Regulation of nonphosphorylating electron transport pathways in soybean cotyledon mitochondria and its implications for fat metabolism. Plant Physiol 86:1199-1204 13. DeTroostembergh J-C, Nyns E-J (1978) Kinetics of the respiration of cyanide-insensitive mitochondria from the yeast Saccharomycopsis lipolytica. Eur J Biochem 85:423-432 14. Dietham.R, Miller MG, Shibles R, Stewart CR (1990) Effect of salicylhydroxamic acid on respiration, photosynthesis, and peroxidase activity in various plant tissues. Plant Cell Physiol 31:179-185 15. Dizengremel P (1983) Effect of Triton x-1oo on the cyanide-resistant pathway in plant mitochondria. Physiol Veg 21:743-752 16. Dizengremel P, Chauveau M, Rousseau J (1982) Inhibition by adenine derivatives of the cyanide-insensitive electron transport pathway of plant mitochondria. Plant Physiol 70:585-589 17. Dizengremel P, Lance C (1976) Control of changes in mitochondrial activities during aging of potato slices. Plant Physiol 58:147-151 77 18. Deuce R, Breuquisse R, Jeurnet E-P (1987) Electron transfer and oxidative phosphorylation in plant mitochondria. In: DD Davies, ed, The Biochemistry of Plants, Vol.11. Academic Press, New York, pp.177-211 19. Deuce R, Neuburger M (1989) The uniqueness of plant mitochondria. Ann.Rev Plant Physiol Plant Mol Biol 40:371-414 20. Dry IB, Moore AL, Day DA, Iiskich JT (1989) Regulation of alternative pathway activity in plant.mitochondria: Nonlinear relationship between electron flux and the redox poise of the quinone pool. Arch Biochem Biophys 273:148-157 21. Elthon TE, McIntosh L (1986) Characterization and solubilization of the alternative oxidase of Sauromatum guttatum mitochondria. Plant Physiol 82:1-6 22. Elthon TE, McIntosh L (1987) Identification of the alternative terminal oxidase of higher plant mitochondria. Proc Nat Acad Sci USA 84:8399-8403 23. Elthon TE, Nickels RL, McIntosh L (1989) Monoclonal antibodies to the alternative oxidase of higher plant mitochondria. Plant Physiol 89:1311-1317 24. Elthon TE, Nickels RL, McIntosh L (1989) Mitochondrial events during development of thermogenesis in Sauromatum guttatum (Schott). Planta 180:82-89 25. Gerard J, Dizengremel P (1988) Properties of mitochondria isolated from greening soybean and lupin tissues. Plant Sci 56:1-7 26. Goldstein AH, Andersen JO, McDaniel RG (1981) Cyanide- insensitive and cyanide-sensitive 0% uptake in wheat. II.Gradient-purified mitochondria lack cyanide-insensitive respiration. Plant Physiol 67:594-596 27. Geyal A, Hiser C, Telbert NE, McIntosh L (1991) Progress no.9 cultivar of pea has alternative respiratory capacity and normal glycine metabolism. Plant Cell Physiol 32:90-96 28. Grover SD, Laties CG (1981) Disulfiram inhibition of the alternative respiratory pathway in plant mitochondria. Plant Physiol 68:393-400 29. Guy RD, Berry JA, Fegel ML, Heering TC (1989) Differential fractionation of oxygen isotopes by cyanide-resistant and cyanide-sensitive respiration in plants. Planta 177:483-491 30. Hemrika-Wagner AM, Verscheer EJ, Van Der Plas LE! (1983) Alternative pathway respiration in vivo of potato tuber callus grown at various temperatures. Physiol Plant 59:369-374 78 31. Henry MF, Nyns E-J ( 1975) Cyanide-insensitive respiration. An alternative mitochondrial pathway. Sub-Cell Biochem 4:1-65 32. Hiser C, McIntosh L (1990) Alternative oxidase of potato is an integral membrane protein synthesized de novo during aging of tuber slices. Plant Physiol 93:312-318 33. Huq S, Palmer JM (1978a) The involvement and possible role of quinone in cyanide resistant respiration. In: G Ducet, C Lance, eds, Plant Mitochondria. Elsevier/North-Holland Biomedical Press, New York, pp.225-232 34. Huq S, Palmer JM (1978b) Superoxide and hydrogen peroxide production in cyanide resistant Arum maculatum mitochondria. Plant 861 Lett 11:351-358 35. Huq S, Palmer JM (1978c) Isolation of a cyanide-resistant duroquinol oxidase from Arum maculatum mitochondria. FEBS Lett 95:217-220 36. James NO, Beevers H (1950) The respiration of Arum spadix. A rapid respiration, resistant to cyanide. New Phytol 49:353-374 37. James IO, Elliot DC (1955) Cyanide-resistant mitochondria from the spadix of an Arum. Nature 175:89 38. Kay CJ, Palmer JM (1985) Solubilization of the alternative oxidase of cuckoo-pint (Arum maculatum) mitochondria. Stimulation by high concentrations of ions and effects of inhibitors. Biochem J 228:309-318 39. Kiener CM, Bramlage WJ (1981) Temperature effects on the activity of the alternative pathway in chill-sensitive Cucumis sativus. Plant Physiol 68:1474-1478 40. Kumar S, Patril BC, Sinha SK (1990) Cyanide resistant respiration is involved in temperature rise in ripening mangoes. Biochem Biophys Res Comm 168:818-822 41. Lambers H (1980) The physiological significance of cyanide-resistant respiration in higher plants. Plant Cell Environ 3:293-302 42. Lambers H (1982) Cyanide-resistant respiration: A non- phosphorylating electron transport pathway acting as an energy overflow. Physiol Plant 55:478-485 43. Lambers H (1985) Respiration in intact plants and tissues: Its regulation and dependence on environmental factors, metabolism and invaded organisms. In: R Douce, DA Day, eds, Higher Plant Cell Respiration. Springer-Verlag, New York, pp.418-473 79 44. Lambers H, Day DA, Ascon-Bieto J (1983) Cyanide resistant respiration in roots and leaves. Measurements with intact tissues and isolated mitochondria. Physiol Plant 58:148-154 45. Lambowits AL, Sabourin JR, Bertrand H, Nickels R, McIntosh L (1989) Immunological identification of the alternative oxidase of Neurospora crassa mitochondria. Mol Cell Biol 9:1362-1364 46. Lance C (1981) Cyanide-insensitive respiration in fruits and vegetables. In: J Friend, MJC Rhodes, eds, Recent Advances in the Biochemistry of Fruits and.Vegetables. Academic Press, New York, pp.63-87 47. Lance C, Chauveau M, Dizengremel P (1985) The cyanide- resistant pathway of plant mitochondria. In: R Douce, DA Day, eds, Higher Plant Cell Respiration. Springer-Verlag, New York, pp.202-247 48. Lane GA, Dale M (1956) Fractionation of oxygen isotopes during respiration. Science 123:574-576 49. Laties GG (1982) The cyanide-resistant, alternative path in higher plant respiration. Ann Rev Plant Physiol 33:519-555 50. McCaig'TN, Hill RD (1976) Cyanide-insensitive respiration in wheat: cultivar differences and effects of temperature, carbon dioxide, and oxygen. Can J Bot 55:549-555 51. McNulty AK, Cummins IR (1987) The relationship between respiration and temperature in leaves of the arctic plant Saxifraga cernua. Plant Cell Environ 10:319-325 52. Meeuse BJD, Raskin I (1988) Sexual reproduction in the arum lily family, with emphasis on thermogenicity. Sex Plant Reprod 1:3-15 53. Miller CO (1980) Cytokinin inhibition of respiration in mitochondria from six plant species. Proc Nat Acad Sci USA 77:4731-4735 54. Miller MG, Obendorf RL (1981) Use of tetraethylthiuram disulfide to»discriminate between alternative respiration and lipoxygenase. Plant Physiol 67:962-964 55. Minagawa N, Sakaje S, Romiyama T, Yoshimete A (1990) A 36kDa mitochondrial protein is responsible for cyanide- resistant respiration in Hansenula anomala. FEBS Lett 264:149-152 56. Meller IM, Berczi A, Van Der Plas LHI, Lambers H (1988) Measurement of the activity and capacity of the alternative pathway in intact plant tissues: Identification of problems and possible solutions. Physiol Plant 72:642-649 8O 57. Moore AL, Bonner Jr MD (1982) Measurements of membrane potentials in plant mitochondria with the safranine method. Plant Physiol 70:1271-1276 58. Moore AL, Preudleve MO (1987) Purification of plant mitochondria on silica sol gradients. In: L Packer, R Douce, eds, Methods in Enzymology, Vol.148. Academic Press, New York, pp.415-420 59. Moore AL, Rich PR (1985) Organization of the respiratory chain and oxidative phosphorylation. In: R Douce, DA Day, eds, Higher Plant Cell Respiration. Springer-Verlag, New York, pp.134-172 60. Moore AL, Bonner Jr MD, Rich PR (1978) The determination of the proton-motive force during cyanide-insensitive respiration in plant mitochondria. Arch Biochem Biophys 186:298-306 61. Musgrave ME, Siedow JN (1985) A relationship between plant responses to cytokinins and cyanide-resistant respiration. Physiol Plant 64:161-166 62. Musgrave ME, Mnrfet IC, Siedow JN (1986) Inheritance of cyanide-resistant respiration in two cultivars of pea (Pisum sativum L.) Plant Cell Environ 9:153-156. 63. Musgrave ME, Miller CO, Siedow JN (1987) Do some plant responses to cytokinins involve the cyanide-resistant respiratory pathway? Planta 172:330-335 64. Obenland D, Hiser C, McIntosh L, Shibles R, Stewart CR (1988) Occurrence of alternative respiratory capacity in soybean and pea. Plant Physiol 88:528-531 65. Parrish DJ, Leopold AC (1978) Confounding of alternate respiration by lipoxygenase activity. Plant Physiol 62:470-472 66. Regan CI, Cettingham IR (1985) The kinetics of quinone pools in electron transport. Biochim Biophys Acta 811:13-31 67. Raskin I, Ehmann A, Melander IR, Meeuse BJD (1987) Salicylic acid: A natural inducer of heat production in Arum lilies. Science 237:1601-1602 68. Rasmussen AG, Meller IM, Palmer JM (1989) Component of the alternative oxidase localized to the matrix surface of the inner membrane of plant mitochondria. FEBS Lett 259:311-314 69. Rheads DM, McIntosh L (1991) Isolation and characterization of a cDNA clone encoding an alternative oxidase protein of Sauromatum guttatum (Schott) . Proc Nat Acad Sci USA 88:2122-2126 81 70. Rich PR (1978) Quinol oxidation in Arum maculatum mitochondria and its application to the assay, solubilisation and partial purification of the alternative oxidase. FEBS Lett 96:252-256 71. Rich PR (1981) A generalised model for the equilibration of quinone pools with their biological donors and acceptors in membrane-bound electron transfer chains. FEBS Lett 130:173-178 72. Rich PR, Bonner Jr* MD (1978) Paramagnetic centers associated with higher plant and poky Neurospora crassa mitochondria. In: G Ducet, C Lance, eds, Plant Mitochondria. Elsevier/North-Holland Biomedical Press, New York, pp.61-68 73. Rustin P, Dupont J, Lance C (1983) Oxidative interactions between fatty acid peroxy radicals and quinones: Possible involvement in cyanide-resistant electron transport in plant mitochondria. Arch Biochem Biophys 225:630-639 74. Rustin P, Moreau F, Lance C (1980) Malate oxidation in plant mitochondria via malic enzyme and the cyanide- insensitive electron transport pathway. Plant Physiol 66:457-462 75. Rybka z (1989) Changes in cyanide-sensitive and cyanide- resistant oxygen uptake in crowns of winter wheat seedings during hardening to frost. J Plant Physiol 134:17-19 76. Rychter.AM, Ciesla E, Kacperska A (1988) Participation of the cyanide-resistant pathway in respiration of winter rape leaves as affected by plant cold acclimation. Physiol Plant 73:299-304 77. Schonbaum GR, Bonner ID, Storey BT, Bahr JT (1971) Specific inhibition of the cyanide-insensitive respiratory pathway in plant mitochondria by hydroxamic acids. Plant Physiol 47:124-128 78. Shingles RM, Arron GP, Hill RD (1982) Alternative pathway respiration and lipoxygenase activity in aged potato slice mitochondria. Plant Physiol 69:1435-1438 79. Siedow JN (1982) The nature of the cyanide-resistant pathway in plant mitochondria. In: LL Creasy, G Hrazdina, eds, Cellular and Subcellular Localization in Plant Metabolism. Plenum Press, New York, pp.47-83 80. Siedow JN (1989) Cyanide resistant respiration in plant mitochondria. In: Abstracts of the 19th Linderstrom-Lang Conference on Plant Mitochondria and Respiration. August 11-13, 1989, Stockholm, Sweden 82 81. Siedow JN, Berthold DA (1986) The alternative oxidase: A cyanide-resistant respiratory pathway in higher plants. Physiol Plant 66:569-573 82. Siedow JN, Girvin ME (1980) Alternative respiratory pathway. Its role in seed respiration and its inhibition by propyl gallate. Plant Physiol 65:669-674 83. Spreen Breuwer K, Van valen T, Day DA, Lambers H (1986) Hydroxamate-stimulated 02 uptake in roots of Pisum sativum and Zea mays, mediated by a peroxidase. Plant Physiol 82:236-240 84. Stegink SJ, Siedow JN (1986) Binding of butyl gallate to plant mitochondria. II.Relationship to the presence or absence of the alternative pathway. Plant Physiol 80:196-201 85. Stewart CR, Martin BA, Redling L, Cerwick S (1990a) Respiration and alternative oxidase in corn seedling tissues during germination at different temperatures. Plant Physiol 92:755-760 86. Stewart CR, Martin BA, Redling L, Cerwick S (1990b) Seedling growth, mitochondrial characteristics, and alternative respiratory capacity of corn genotypes differing in cold tolerance. Plant Physiol 92:761-766 87. Storey ET (1976) Respiratory chain of plant mitochondria. XVIII. Point of interaction of the alternative oxidase with the respiratory chain. Plant Physiol 58:521-525 88. Sterey'BT (1980) Electron transport and.energy coupling in plant mitochondria. In: DD Davies, ed, The Biochemistry of Plants, Vol.2. Academic Press, New York, pp.125-195 89. Theologis A, Laties GG (1978) Respiratory contribution of the alternate path during various stages of ripening in avocado and banana fruits. Plant Physiol 62:249-255 90. Theologis A, Laties GG (1980) Membrane lipid breakdown in relation to the wound-induced and cyanide-resistant respiration in tissue slices. A comparative study. Plant Physiol 66:890-896 91. Van Der Plas LHW, Gude H, Wagner MJ (1987) Hydroxamate- activated peroxidases in potato tuber callus. Interaction with the determination of the cytochrome and alternative pathways. Physiol Plant 70:35-45 92. van De Venter HA (1985) Cyanide-resistant respiration and cold resistance in seedlings of maize (Zea mays L.). Ann Bot 56:561-563 83 93. Wilson SB (1978) Cyanide-insensitive oxidation of ascorbate + NNN'N'-tetramethyl-p-phenylenediamine mixture by mung bean (Phaseolus aureus) mitochondria. An energy-linked function. Biochem J 176:129-136 94. Wilson SB (1980) Energy conservation by the plant mitochondrial cyanide-insensitive oxidase . Some additional evidence. Biochem J 190:349-360 95. Wilson SB (1988) The switching of electron flux from the cyanide-insensitive oxidase to the cytochrome pathway in mung- bean (Phaseolus aureus L.) mitochondria. Biochem J 249:301-303 96. Yentur S, Leopold AC (1976) Respiratory transition during seed germination. Plant Physiol 57:274-276 97. Yoshimeto A, Sakaje S, Minagawa N, Komiyama T (1989) Possible role of a 36kDa protein induced by respiratory inhibitors in cyanide-resistant respiration in .Hansenula anomala. J Biochem 105:864-866 CHAPTER 2-2: ALTERNATIVE OXIDASE CAPACITIES AND POLYPEPTIDES IN TWO PEA CULTIVARS INTRODUCTION The existence of the alternative respiratory pathway has been well established in many plant species (4); indeed, it may be ubiquitous among higher plants. This has posed a problem in approaching the study of the alternative oxidase in plants. Unlike the case of Neurospora crassa in which mutants of the alternative oxidase have been characterized ( 7) , no mutants in the alternative pathway are known for higher plants. Thus research has focused on plant systems in which the alternative pathway is developmentally regulated, as in thermogenic aroids ( 9) , or can be manipulated, as by aging (14) or ethylene treatment (15) of tuber tissue. Musgrave and Siedow (10) reported that leaf tissue of the dwarf pea (Pisum sativum L.) cultivar Progress No.9 lacked the alternative pathway, whereas in the Alaska cultivar the pathway was present. The lack of this pathway was reported to be maternally inherited but distinct from the dwarfness trait in crosses between the two cultivars (11) . Lack of the alternative path was proposed as an 84 85 explanation why Progress No.9 seedlings outperformed Alaska seedlings in terms of growth and total dry matter production, especially under CO2 enrichment (12) , as might be predicted from the overflow hypothesis of Lambers (6) . It was also suggested that lack of the alternative pathway was why Progress No.9 produced little ethylene when treated with benzyladenine ‘while Alaska responded by increasing ethylene production; the authors proposed involvement of the alternative path in cytokinin responses (10). Thus these two pea cultivars seemed to constitute an ideal system in which the alternative oxidase could be manipulated and characterized. MATERIALS AND METHODS Sources of Pea Mitochondria Pea seeds of cultivars Alaska and Progress No.9 were obtained from the W. Atlee Burpee Company, Warminster, PA, for isolation of mitochondria by method A, and from Dr. James N. Siedow, Duke University, Durham, NC, for isolation of mitochondria by method B. Seeds were soaked overnight in distilled water, planted in vermiculite, watered once with Hoagland solution, and grown in a dark cabinet at room temperature for 7 to 10 days before harvesting. 86 Isolation of Mitochondria by Method A Etiolated pea seedlings were cut into small pieces and ground in a blender with a twofold volume of isolation buffer (400mM sorbitol, 30mM MOPS pH 7.6, 1mM EDTA, 4mM cysteine, 0.2% BSA, 0.6% PVPP). After filtration through 4 layers of sterile Miracloth (Calbiochem), the homogenate was centrifuged at 1400g for 15 minutes. The resulting supernatant was centrifuged at 16,300g for 15 minutes to pellet the mitochondria. The pellets were resuspended in wash buffer (350mM sucrose, 30mM MOPS pH 7.2, 1mM EDTA, 0.2% BSA) and the suspension was centrifuged at 4300g for 2 minutes. The supernatant was centrifuged at 12,000g for 5 minutes to pellet the mitochondria, which were resuspended in assay buffer (250mM sucrose, 30mM TES pH 7.2). Oxygen uptake by the washed mitochondria was measured in a Rank Brothers electrode apparatus (digital oxygen system model 10) at 25°C in a total volume of 1.0mL of assay buffer plus sample. Protein concentrations were determined by a modified Lowry method (8) . Calculations were performed as described in Table 2-2-1. Isolation of Mitochondria by Method B Washed mitochondria were obtained following the procedures of Musgrave et a1. (11) except for slight variations in the isolation buffer (400mM mannitol, 20mM 87 HEPES pH 7.2, 1mM EDTA, 0.5mM Nazszoy 0.5% BSA, 0.1% PVPP) and the wash buffer (400mM mannitol, 20mM HEPES pH 7.2, 0.1mM EDTA, 0.1% BSA). The mitochondria were further purified by sucrose density gradient centrifugation (1) . Oxygen uptake by the purified mitochondria was measured as described in method A in the assay buffer of Musgrave et a1. (11) with the addition of 1mg/mL BSA. Immunoblotting Samples of whole mitochondria containing 100pg total protein were separated by one-dimensional polyacrylamide gel electrophoresis and transferred to nitrocellulose as described in Elthon and McIntosh (2) . The nitrocellulose filters were first washed for 30 minutes in PBS-Tween (10mM NaHzPO“ 150mM NaCl, pH 7.2 with 0.3% Tween-20). One filter was reacted for 1.5 hours with 20pL of a polyclonal antibody raised against the alternative oxidase of sauromatum guttatum (2) in 10mL of PBS-Tween, followed by a 1 hour incubation with 10uL anti-mouse IgG conjugated to alkaline phosphatase in 10mL PBS-Tween. The other filter was reacted with 20uL of an alternative oxidase monoclonal antibody conjugated to alkaline phosphatase (3) in 10mL of PBS-Tween for 1.5 hours. The filters were then washed for 5 minutes in pH 9.5 buffer (100mM Tris, 100mM NaCl, 5mM MgClz, 0.3% Tween) and reacted for 2 minutes with NBT-BCIP reagent (10mL of pH 9.5 buffer with 3.3mg nitroblue tetrazolium and 1.7mg 88 5-bromo-4-chloro-3-indoyl phosphate). The filters were then washed for 5 minutes in PBS-Tween followed by 10 minutes in PBS-Tween containing 5mM EDTA. RESULTS AND DISCUSSION Respiratory rates for mitochondria isolated from etiolated seedlings of Alaska and Progress No.9 are compiled in Table 2-2-1. Comparison of the respiratory rates using either washed (Part A) or purified (Part B) mitochondria indicates that the capacity for the alternative pathway was low but measurable in both cultivars. In washed mitochondria with NADH as the substrate, the alternative path capacity was between 8%-9% of the uninhibited rate in both cultivars. Since the reported difference in capacities was not found, mitochondria were then purified by the method of the original authors (11). In purified mitochondria with succinate as the substrate, the alternative path capacity in Progress No.9 was somewhat lower than that of Alaska. Musgrave et a1. (11) reported a similar alternative path capacity of 8:0.3% in washed mitochondria from epicotyls of Alaska peas, but found. no alternative jpath capacity in Progress No.9 peas. That NADH is a poor electron donor to the alternative oxidase in isolated nonaroid mitochondria (4) was suggested (Dr. J .N. Siedow, personal communication) as one possible 89 .5685 .o 2.5.95... o; 2 md 66:53:06 :3. :666 :_ 66.6866 _6_.6:o:oozE 6:... 5:68:36 .66 6::. 656236. 66.5 5.3 62666696 66.5 8 cm H :66... 65 6.6 6626> 66956.2 6:6 6.666562 666v m 6956.2 .5 666.8_ 6:65:85: 665.3...n .5296 so 6E6.o.____:. ed o. 6.0 66:_6Eoo :3. :666 :_ 629.66 _6_.6:o:oogE 6:._. .E6:._.6oxe .66 6::. 362.66. 66.5 5.3 £:6E_.6ox6 62. 8 cm H :62: 65 6.6 66:.6> .A66o56$_ 6:6 6.666565. 666V < 6256.2 .6 6366s 6:65:6er 6656636 . :9. .23 + 38.6 66.6 n 3.2 2.6 u 8.63 8.2 H 66.66 26583 2:: 6.62 36.8.: :9. .23 + 36.5 5.6 H 66.: 636 u 8.63 66.6. n 8.66 26583 2;; 6.662 ” ta 6 m n. 36.3 8.6 u 66... 8.6 u 3. 5 6.6 u 8.66 :92 .2... P 6.62 36.8... 186.8666 4. 8... 66.6 n 636 66.6 H 66.6 :92 _25 F 9.662 6 ”< :6: 26m 656565 26m 66.98:: 656m 6550 6565665 .559: 8 86.95:. .6d 655.: .66 665:6:8 :6m>xo 6Eo6oc6: 66 :95 6.6 666m 25$ 2:: 626 29. SE; :_ 656. 65 665:. 29. 2:: :_ 26. 65 66 6656.36.66 663 856.666. 656522 .600“. 2&6 Co :oE666 .656 665666.: 663 26. 66:58:: 65 6:6 65656636 6:6:6mox6 no 66:666.: 65 :_ 66.66665 66; 26. .6550 $63566 65652.6 65 8 366666 6665:6266 6:6 6:63:63... 6656.3. .5 6x86: 59.6 B 6656.". . TN.“ 636... 90 explanation for the discrepancy. However, alternative path capacity was also detected in Progress No.9 using succinate (and malate, data not shown) as substrates. Our laboratory has also measured significant alternative path capacities in potato, tomato, tobacco, and rice using NADH as a substrate (unpublished results). This lack of alternative path capacity in Progress No.9 found by Musgrave et a1. (11) could not later be replicated (Dr. J .N. Siedow, personal communication). Samples of whole mitochondria from both pea cultivars were immunoblotted with antibodies against the S. guttatum alternative oxidase (3) in Figure 2-2-1. No significant differences were apparent between immunostained, proteins from Alaska (lanes A1 and B2) and Progress No.9 (lanes A2 and B3). Using either the polyclonal (Part A) or monoclonal (Part B) antibodies, both pea cultivars had three strongly reactive proteins with estimated molecular weights of 41kD, 39kD, and 38kD. In most aroid species tested, the antibodies identify 3 proteins with molecular weights ranging from 34-37kD, although in Amorphophallus the highest molecular weight polypeptide is greater than 37kD ( 3) and the precursor to the S. guttatum alternative oxidase polypeptides is 42kD (13). Nonaroid plants tested possess 1 or 2 alternative oxidase polypeptides ranging from 35-37kD (3,5). Some lower molecular weight, weakly immunoreactive bands were also seen from both pea cultivars. Low molecular weight bands have also been reported from potato (3,5) , 91 1231234 >416 —84 -' 4' 1' —58 ‘ —48.5 365 I a _ ' -26.5 Figure 2-2-1. Immunoblotting of alternative oxidase proteins. Total mitochondrial proteins from both pea cultivars were electrophoresed, transferred to nitrocellulose, and reacted with polyclonal (Part A) or monoclonal (Part B) antibodies against the alternative oxidase of Sauromatum guttatum as described in "Materials and Methods." In Part A, lane 1: Alaska; lane 2: Progress No.9; lane 3: S. guttatum. In Part B, lane 1: S. guttatum; lane 2: Alaska; lane 3: Progress No.9; lane 4: molecular weight standards in kD. 92 tobacco (3), and Neurospora (7). The three strongly immunoreactive proteins are likely to constitute components of the alternative oxidase of pea. From this study, it is apparent that the dwarf pea cultivar Progress No.9 possesses an alternative respiratory capacity and the same alternative oxidase proteins as the Alaska cultivar. Therefore, these two pea cultivars do not constitute the desired manipulable system to study the alternative pathway. ACKNOWLEDGMENTS Thanks are due to Dr. J .N. Siedow for seeds and advice. Table 2-2-1 has been reproduced from the following paper with permission from the copyright owner, the American Society of Plant Physiologists: Obenland D, Hiser C, McIntosh L, Shibles R, Stewart CR (1988) Occurrence of alternative respiratory capacity in soybean and pea. Plant Physiol 88:528-531 Figure 2-2-1 has been reproduced from the following paper with permission from the copyright owner, the Japanese Society of Plant Physiologists: Geyal A, Hiser C, Tolbert NE, McIntosh L (1991) Progress No.9 cultivar of pea has alternative respiratory capacity and normal glycine metabolism. Plant Cell Physiol 32:90-96 93 LITERATURE CITED 1. Deuce R, Christiansen EL, Bonner WD Jr (1972) Preparation of intact plant mitochondria. Biochim Biophys Acta 275: 148-160 2. Elthon TE, McIntosh L (1987) Identification of the alternative terminal oxidase of higher plant mitochondria. Pro Nat Acad Sci USA 84:8399-8403 3. Elthon TE, Nickels R, McIntosh L (1989) Monoclonal antibodies to the alternative oxidase of higher plant mitochondria. Plant Physiol 88:1311-1317 4. Henry MF, Nyns EJ (1975) Cyanide-insensitive respiration. An alternative mitochondrial pathway. Subcell Biochem 4:1-65 5. Hiser C, McIntosh L (1990) Alternative oxidase of potato is an integral membrane protein synthesized de novo during aging of tuber slices. Plant Physiol 93:312-318 6. Lambers H (1982) Cyanide-resistant respiration: A non- phosphorylating electron transport pathway acting as an energy overflow. Physiol Plant 55:478-485 7. Lambewits AM, Sabeurin JR, Bertrand H, Nickels R, McIntosh L (1989) Immunological identification of the alternative oxidase of Neurospora crassa mitochondria. Mol Cell Biol 9:1362-1364 8. Larson E, Hewlett B, Jagendorf AT (1986) Artificial reductant enhancement of the Lowry method for protein determination. Anal Biochem 155:243-248 9. Meeuse BJD, Raskin I (1988) Sexual reproduction in the arum lily family, with emphasis on thermogenicity. Sex Plant Reprod 1:3-15 10. Musgrave ME, Siedow JN (1985) A relationship between plant responses to cytokinins and cyanide-resistant respiration. Physiol Plant 64:161-166 11. Musgrave ME, Murfet IC, Siedow JN (1986a) Inheritance of cyanide-resistant respiration in two cultivars of pea (Pisum sativum L.). Plant Cell Environ 9:153-156 12. Musgrave ME, Strain BR, Siedow JN (1986b) Response of two pea hybrids to CO2 enrichment: A test of the overflow hypothesis for alternative respiration. Proc Nat Acad Sci USA 83:8157-8161 94 13. Rheads DM, McIntosh L (1991) Isolation and characterization of a cDNA clone encoding an alternative oxidase protein of Sauromatum guttatum (Schott) . Proc Nat Acad Sci USA 88:2122-2126 14. Theologis A, Laties GG (1980) Membrane lipid breakdown in relation to the wound-induced and cyanide-resistant respiration in tissue slices. A comparative study. Plant Physiol 66:890-896 15. Theologis A, Laties GG (1982) Selective enhancement of alternative path capacity in plant storage organs in response to ethylene plus oxygen: a comparative study. Plant Physiol 69:1036-1039 95 Chapter 2-3 is reprinted from the following published paper with the permission of the copyright owner, the American Society of Plant Physiologists: Hiser C, McIntosh L (1990) Alternative oxidase of potato is an integral membrane protein synthesized de.novo during aging of tuber slices. Plant Physiol 93:312-318 CHAPTER 2-3: THE ALTERNATIVE OXIDASE OP POTATO IS AN INTEGRAL MEMBRANE PROTEIN SYNTHESIZED.DE.NOWO DURING AGING OP TUBER SLICES INTRODUCTION In addition to the cyanide-sensitive cytochrome pathway of electron transport, plant mitochondria possess a cyanide- resistant, hydroxamate-sensitive alternative pathway. Both pathways transfer electrons from ubiquinone to oxidases that reduce oxygen to H20 (31) . In the thermogenic inflorescences of species of the.Araceae, electrons.are channeled.through.the alternative pathway to generate heat, which volatilizes compounds to attract insect pollinators (2,9,21,22) . The purpose of the alternative pathway in non-thermogenic species is unknown. One view is that electrons f low through the alternative pathway whenever the cytochrome pathway is saturated or limited (the overflow hypothesis), permitting continued functioning of the TCA cycle to process carbon skeletons even when ATP is not needed and using up excess carbohydrates (reviewed in 15,20,26) . In Neurospora, the alternative oxidase proteins are not constitutively synthesized, but when the synthesis of mitochondrially-encoded components of the cytochrome pathway is inhibited by 96 97 chloramphenicol, the synthesis of the nuclear-encoded alternative oxidase proteins is induced (16) . Studies on the mechanism of electron partitioning between the two pathways in nonaroid higher plants have been hindered by the lack of specific antibodies or genetic probes for the alternative oxidase. However, with the isolation of antibodies to the alternative oxidase (8) , a classic system for alternative pathway induction - aging potato tuber slices - presents itself as a model system for such future studies. The wound respiration of potato (Solanum tuberosum) was described as early as 1887 (referenced in 11) and the cyanide- resistance of aged slice respiration was recognized and investigated_ in the late 19408 and early 1950s (11). Initially, it was suggested that such respiration may be the result of free radical reactions (reviewed in 31) or due to the activity of lipoxygenase ( 23) . However, that the alternative oxidase is a lipoxygenase (which also consumes oxygen and is cyanide-insensitive) associated with membrane lipid breakdown or repair has been ruled out by careful isolation of mitochondria ( 25) . The possibility that cyanide- resistant respiration may be non-enzymatic in nature has also been ruled out through immunological data demonstrating the presence of specific proteins necessary for activity and by analysis of the engagement of the alternative pathway in whole tissues (7,8,9,10). Fresh tissue slices from tubers demonstrate respiration that is sensitive to cyanide while aged potato tuber slices exhibit cyanide-resistant respiration 98 (3,11,29) . However, the induction of alternative pathway respiration in aged slices could be mimicked in fresh slices if the tubers were treated with ethylene, ethanol, acetaldehyde, acetate, or cyanide for 24 hours before slicing (1,13,14,24,27). These results have been interpreted to mean that the alternative pathway is present constitutively in tubers (20). Theologis and Laties (28,30) proposed that membrane integrity is necessary for the functioning of the alternative oxidase and that slicing causes mitochondrial membrane breakdown which is repaired during aging. Changes in mitochondrial membrane protein composition (4) and a requirement for cytoplasmic protein synthesis (3) have been correlated with the rise in alternative respiratory capacity in aging potato slices. Until recently, the lack of an antibody specific to the alternative oxidase has prevented examination of the levels of the alternative oxidase protein(s) itself. We show here that a monoclonal antibody raised against the Sauromatum guttatum alternative oxidase ( 8) can be employed to characterize developmental changes in the potato alternative pathway. The alternative oxidase was partially purified and characterized by two-dimensional gel electrophoresis. It is interesting to note that the relative levels of the protein parallel the rise in alternative oxidase capacity in potato. Similar results have been previously published for S. guttatum (7,8,9) and the fungal alternative oxidase from Neurospora crassa (18). 99 MATERIALS AND METHODS Aging of Potato Slices and Isolation of Mitochondria Potatoes (Solanum tuberosum cv. Russet Burbank) were obtained from Dr. David Douches (Department of Crop and Soil Sciences, Michigan State University). Aging of potato slices for 12 or 24 hours was carried out according to Dizengremel and Lance (3). Aged slices or shredded fresh potatoes were blended in 1.5 volumes of isolation buffer (400mM sorbitol, 30mM MOPS, 1mM EDTA, 4mM cysteine, 0.2% BSA, 0.6% PVPP, pH 7.6). After filtration ‘through four layers of sterile Miracloth (Calbiochem) , the homogenate was centrifuged at 14009 for 5 minutes. The resulting supernatant was centrifuged at 16,3009 for 15 minutes to pellet the mitochondria. The pellets were resuspended in wash buffer (350mM sucrose, 30mM MOPS, 1mM EDTA, 0.2% BSA, pH 7.2) and the suspension centrifuged at 43009 for 2 minutes. The supernatant was centrifuged at 20,0009 for 5 minutes to pellet the mitochondria. The mitochondria were further purified by sucrose density gradient centrifugation (5). Respiration Assays Purified mitochondria were resuspended in assay buffer (SOOmM mannitol, 10mM KZHPO“ 10mM KCl, 5mM M9C12, pH 7.2). Oxygen uptake was measured in a Rank Brothers electrode 100 apparatus (digital oxygen system Model 10) at 25°C in a total volume of 1.0mL. Samples contained 0.1-0.3mg protein, as determined by a modified Lowry method (190). The exogenous substrate used was 10mM succinate with 150nm ADP, then mitochondria were uncoupled with 0.5uM FCCP before respiration measurements were taken. Cytochrome pathway capacity was taken to be that portion of the respiration inhibited by 1mM KCN. Alternative pathway capacity was measured as that portion of the respiration inhibited by 1mM SHAM in the presence of 1mM KCN (9). Residual respiration is the oxygen uptake which remained in the presence of 1mM KCN + 1mM SHAM (17). Partial Purification of Membrane Proteins and Immunoblotting The alternative oxidase and other mitochondrial membrane proteins were solubilized and partially purified as described (6). Samples of mitochondria (400119 protein) or partially purified membrane proteins ( 600119 protein) were electrophoresed and transferred to nitrocellulose as previously described (7) . The nitrocellulose filter was washed for 30 minutes in PBS-Tween (10mM NaHzPO“ 150mM NaCl, pH 7.2 with 0.3% Tween-20) , then reacted with a 1:200 dilution of the monoclonal antibody against the Sauromatum guttatum alternative oxidase (AOA; 8) in PBS-Tween for 1.5 hours. After two 5-minute washes with PBS-Tween, the filter was reacted for 1 hour with a 1:1000 dilution of anti-mouse 101 IgG-alkaline phosphatase conjugate in PBS-Tween (8) . The filter was then washed for 5 minutes with PBS-Tween, then for 5 minutes with pH 9.5 buffer (100mM Tris, 100mM NaCl, 5mM MgCl?,ij 9.5 with 0.3% Tween-20). Color was developed by reaction with NBT-BCIP reagent (3.3mg nitroblue tetrazolium and 1.7mg 5-bromo-4-chloro-3-indoyl phosphate per 10mL of pH 9.5 buffer) until the desired darkness was reached. The filter was then washed for 5 minutes in PBS-Tween followed by 10 minutes in PBS-Tween containing 5mM EDTA. Two-Dimensional Gel Electrophoresis and Blotting Isoelectric focusing of partially purified and solublized aged potato slice mitochondrial membrane proteins (1.47mg protein in 20uL) was performed as described (12) . Electrophoresis in the second dimension through 1.5mm thick SDS-polyacrylamide gradient gels and. immunoblotting’ were performed essentially as described above. Isoelectric focusing and two-dimensional gel electrophoresis of Sauromatum guttatum mitochondrial membrane proteins (60u9 in 20uL) was performed as previously described (12). Blots were stained for total protein using 100uL India ink in 20mLs PBS-Tween. RESULTS Figure 2-3-1 presents the results of respiration assays on mitochondria isolated from fresh, 12-hour-aged, and 102 [::]cyt. - afiflalt §§§ res '__1 80 - C o _ Z c t _. 3 (SO - 6 m: ‘U 2 T c 3 o g ‘40 - :3 4 O 20 — 12-Hr-oged 24-Hr-oged Figure 2-3-1. Respiration of potato mitochondria. Oxygen consumption of mitochondria isolated from fresh, 12-hour-aged, and 24-hour-aged potato disks was measured as described in "Materials and Methodsfl' White bars represent cytochrome pathway capacity, shaded bars represent alternative pathway capacity, and striped bars represent residual respiration. Capacities are expressed as the percentage of the uncoupled, uninhibited respiration rate and are the averages of three experiments with at least three runs per experiment. 103 24-hour-aged potato tissue. The rate of uncoupled respiration (averages of 3 experiments) rose from 109+/-46 nanoatoms oxygen per minute per milligram of protein in fresh potato slice mitochondria to 181+/-32 nanoatoms oxygen per minute per milligram of protein after 12 hours of aging. The rate then declined after 24 hours of aging to 89+/-15 nanoatoms oxygen per minute per milligram of protein. The percentage of the total respiratory capacity that was cyanide-sensitive and mediated by the cytochrome pathway decreased from 84.1% to 42.7% with aging (Figure 2-3-1). The percentage that was cyanide-resistant, SHAM-sensitive, and mediated by the alternative pathway increased from 6.7% to 38.1% . Residual respiration, insensitive to both cyanide and SHAM and not mediated by either pathway, also increased with aging. These results agree with other reports of cyanide-insensitive respiration in potato slices (11,29) and isolated mitochondria (3,25). The differences in alternative oxidase capacity presented in Figure 2-3-1 are reflected in the levels of alternative oxidase protein. A monoclonal antibody raised against the S. guttatum alternative oxidase (8) was used on Western blots of mitochondrial proteins to determine the presence and relative concentration of the alternative oxidase. Mitochondria were isolated from fresh and 24-hour-aged potato tissue and membrane proteins partially purified as described in "Materials and Methods." Whole mitochondria, supernatant fractions (containing BIGCHAP-solublized membrane proteins), 104 Figure 2-3-2. Immunoblots of potato alternative oxidase. Proteins were electrophoresed, transferred to nitrocellulose, and reacted with a monoclonal antibody against the Sauromatum guttatum alternative oxidase (8) as described in "Materials and Methods." Each panel includes lanes of S. guttatum mitochondria (A1, Bl) for comparison, potato mitochondria (A2, B2), deoxy-BIGCHAP-insoluble mitochondrial membrane proteins from the pellet fraction (A3, B3), and deoxy-BIGCHAP- solublized mitochondrial membrane proteins from the supernatant fraction (A4, B4) . Panel A contains proteins from fresh potato tissue; panel B contains proteins from 24-hour- aged potato disks. Molecular weights in kD are shown to the left. 1 2 3 4 375: ' .35,- Figure 2-3-2. 105 1 2 3 4 106 and pellet fractions (containing membrane proteins insoluble in BIGCHAP) are compared in Figure 2-3-2. The alternative oxidase antibody reacted to several low molecular weight (12- 18kD) polypeptides in fresh potato mitochondria (lane A2) , but to one polypeptide of an apparent molecular weight of 36kD (similar to the S. guttatum alternative oxidase) in 24-hour- aged potato mitochondria (lane B2) . The low molecular weight polypeptides were faintly present in aged mitochondria while the high molecular weight polypeptide was absent or present in extremely low concentration in fresh mitochondria. The low molecular weight polypeptides also appeared in the pellet fractions of both fresh (lane A3) and 24-hour-aged (lane B3) membrane protein preparations, but were not present in the supernatant fractions (lanes A4 and B4). These polypeptides were apparently not easily solublized or removed from the mitochondrial membranes. The high molecular weight polypeptide was not present in the supernatant fraction of the fresh membrane protein preparation (lane A4) , but was enriched in the 24-hour-aged membrane protein preparation (lane B4). We believe this high molecular weight polypeptide is the alternative oxidase of potato or one component of it, and the lower molecular weight polypeptides may be degradation products. These data suggest that the increased alternative pathway capacity in aged tissue results from an increased level of alternative oxidase protein. It has been previously shown in S. guttatum that increased alternative pathway capacity is 107 associated with increased levels of the alternative oxidase proteins (7,8,9). Cytoplasmic protein synthesis is required for the development of alternative path capacity in aging potato slices (3). 'Therefore, the increased levels of alternative oxidase protein in aged slice mitochondria may be due to de novo synthesis during aging of the 36kD active alternative oxidase which was not present in mitochondria of whole tubers or non-aged slices. Of course, our experiments do not differentiate between increased synthesis or decreased turnover. To test the remote possibility that the lower molecular weight bands were degradation products caused by exposure of previously sequestered protein or by an SDS-activated protease during preparation of samples for gels, an experiment was performed in which S. guttatum appendix mitochondria containing high amounts of alternative pathway were mixed with potato mitochondria (Figure 2-3-3). Some samples were boiled immediately after the addition of the mitochondria to the sample buffer (left half of figure), while replicate samples were incubated at room temperature for 10 minutes in the sample buffer before boiling (right half of figure). If there were a protease active during sample preparation, then in the incubated samples it might have been expected to degrade the three prominent S. guttatum bands normally seen on western blots probed with the alternative oxidase antibody. However, identical bands were visible in both the incubated and the non-incubated samples. No bands were seen in the S. guttatum 108 Figure 2-3-3. Mixing and incubation of mitochondrial samples. Samples 1 to 6 were boiled immediately upon addition of the mitochondria to the sample buffer. Samples 7 to 12 were incubated for 10 minutes at room temperature after addition of the mitochondria to the sample buffer and before boiling. Proteins were then electrophoresed, transferred to nitrocellulose, and reacted with a monoclonal antibody against the Sauromatum guttatum alternative oxidase (8) . Lanes 1 and 12: molecular weight standards; lanes 2 and 7: fresh potato mitochondria; lanes 3 and 8: mixture of fresh potato and S. guttatum mitochondria; lanes 4 and 9: 24-hour-aged potato disk mitochondria; lanes 5 and 10: mixture of 24-hour-a9ed potato disk and S. guttatum mitochondria; lanes 6 and 11: S. guttatum mitochondria. 109 ngnincubateg' inggbated . 123456789101112 m .— = «fl -~~~‘ . Figure 2-3-3. 110 plus potato samples (lanes 3 and 8) that were not also present in samples of S. guttatum (lanes 6 and 11) or potato (lanes 2,4,7 and 9) mitochondria alone. This experiment suggests that the low molecular weight bands prominent in fresh potato mitochondria are not a result of general degradory processes or an SDS-activated protease during sample treatment. To further characterize the potato alternative oxidase, we separated the mitochondrial membrane polypeptides on two- dimensional gels (Figure 2-3-4). Partially purified mitochondrial membrane proteins were isolated according to Elthon and McIntosh (6) . Two potato membrane polypeptides are immunoreactive with the alternative oxidase antibody (Panel D). These have the same apparent molecular weight but different isoelectric points of approximately 7.1 and 6.7. They may therefore be different isoforms of the same protein. In order to visualize the potato alternative oxidase in immunoblots, high amounts of mitochondrial membrane proteins were loaded onto the gels. Consequently, many potato proteins are present in the India-ink stained.blot (Panel C) and.we are unable to identify with certainty which are the immunoreactive ones. Since much less S. guttatum mitochondrial membrane protein need be loaded onto gels to visualize the alternative oxidase in immunoblots, the alternative oxidase proteins are readily identifiable on both the stained blot (Panel A) and the immunoblot (Panel B). They have apparent molecular weights of 37kD, 36kD, and 35kD and isoelectric points of approximately 7.2 and 7.3. 111 pH -: -o -o -: -a KDA . 97- Figure 2-3-4. TWO-dimensional gel electrophoresis and immunoblots of alternative oxidase proteins. Panels A and B are mitochondria membrane proteins from Sauromatum guttatum; panels C and D are mitochondrial membrane proteins from 24- hour-aged potato disks. Panels B and D were reacted with a monoclonal antibody against the S. guttatum alternative oxidase (8). Panels A and C are the same blots as panels B and D, respectively, after staining for total protein with India ink. 112 DISCUSSION A fundamental question in plant bioenergetics concerns the mechanism(s) for governing the partition of electron flow between the cytochrome pathway and the alternative pathway. Until recently there have been two major hindrances to answering this question: first, the methods for measurement of electron flow into either pathway; and second, the lack of a physical or molecular probe for the alternate oxidase. Measurements of cytochrome and alternative pathway respiration are measurements of the capacities of the pathways obtained through inhibitor titration (reviewed in 17), not measurements of actual activities for the two pathways in vivo. Recently, differential fractionation of oxygen isotopes has been shown to be a viable method for the measurement of the engagement of the alternate and cytochrome pathways (10). It is heartening, for those who do measure capacities, that engagement and capacities in these recent experiments were similar. There is one report for oxygen discrimination for potato tubers (18); however, as pointed.out.by Guy at al.(10), the value is probably strongly influenced by limited oxygen diffusion. Limited oxygen diffusion is still a major hurdle to overcome for many plant organs and tissues when one wishes to measure engagement (10). A molecular approach to the alternative pathway, such as the one described here which employed a monoclonal antibody, should allow a broader understanding of the mechanism of alternative pathway 113 induction and complement knowledge gained from future experiments on the engagement of the electron transport pathways in plant mitochondria. We have used.a monoclonal antibody raised against the S. guttatum alternative oxidase (8) to follow the development of alternative oxidase capacity in aging potato slices. This antibody cross-reacts to the alternative oxidase of a variety of species including potato (8). Our results indicate that the levels of the 36kD alternative oxidase protein parallel alternative pathway capacity in the mitochondria during the aging of potato slices. Increased alternative oxidase capacity in aged slice mitochondria appears to be the result of increased levels of the 36kD protein present. Dizengremel and Lance (3) demonstrated the need for cytoplasmic (but not mitochondrial) protein synthesis to permit the increase in alternative pathway capacity in aging potato slice mitochondria. This suggests that new alternative oxidase proteins may be synthesized during the aging of potato slices, and that the gene(s) for alternative oxidase are nuclear- encoded in potato. The alternative oxidase genes were shown to be nuclear-encoded in S. guttatum (9,11; Rheads and McIntosh, unpublished results) and Neurospora (2,16). Respiration data have previously been interpreted to mean that the alternative pathway is present in whole tubers but destroyed by slicing and restored by aging (20,28,29). Our data suggest that another interpretation is possible: the active alternative oxidase protein (the 36kD protein) is not 114 present in significant amounts in whole tuber or fresh slice mitochondria but is synthesized de novo during aging of potato slices. Cyanide-resistant oxygen uptake previously measured in whole tubers (22) may have been non-mitochondrial in origin. Dizengremel and Lance (4) reported variations in the protein composition of inner mitochondrial membranes with aging of potato slices. No increase in proteins of molecular weights near 36kD was reported (4); however, the alternative oxidase may not have been visible stained with Coomassie blue at the concentrations of mitochondrial protein they loaded on their gels. Two dimensional gel electrophoresis of the in vitro translation products of polysomal RNA from avocado fruits (32) and carrot roots (33) which were treated with ethylene and/or cyanide showed increases in proteins near the molecular weights and pKas of potato and S. guttatum alternative oxidase proteins as determined in our laboratory (7,8; Rheads and McIntosh, unpublished results). Treatment with ethylene and/or cyanide has been shown to induce cyanide- resistant respiratory capacity in potato (1,13,24,27,31) and other storage organs (31). Dizengremel and Lance did report a decrease in the relative amounts of proteins with.molecular weights less than 22kD (4) . We have also found decreased levels of the 12-18kD immunoreactive polypeptides in aged slice mitochondria. Faint low molecular'weight bands similar to those seen in the immunoblots of fresh potato slice mitochondria (Figure 115 2-3-2, lane.A2) have also been seen in other higher plants (8; A Goyal, C Hiser, NE Tolbert, L McIntosh, unpublished data; RL Nickels and L McIntosh, unpublished data) and Neurospora strains which lack the higher molecular weight alternative oxidase protein (16) . There is presently no conclusive explanation for these bands. Our mixing experiment suggests that they are not the results of general degradory processes during sample preparation. It is still possible that they may be the products of proteases present (perhaps to a greater degree in fresh tissue than aged tissue) during mitochondrial isolation. They may be degradation products formed during the membrane breakdown which accompanies slicing (28, 30) . Another possibility is that disruption of the membranes due to slicing does not allow for proper insertion of alternative pathway proteins into a complex and thus they are broken down. The low molecular weight polypeptides may therefore result from turnover of non-complexed alternative oxidase. If this were the case, then one could speculate that some level of alternative oxidase proteins are constitutively made but rapidly turned over unless some factor (induced by wounding and aging) stabilizes them. We conclude that the potato alternative oxidase capacity increases at the same time as the 36kD protein constituent rises in concentration in aging potato slice mitochondria. The levels of 2 to 3 cross-reactive, low molecular weight polypeptides in the mitochondrial membrane concurrently decrease. It is unknown if the synthesis of potato 116 alternative oxidase is transcriptionally and/or post- transcriptionally controlled, or if levels of the cytochrome oxidase may influence the levels of the alternative oxidase. In S. guttatum, the relative levels of mRNA encoding the cytochrome oxidase subunits I and II decrease during the time in which alternative oxidase capacity is increasing, raising the possibility that regulation of cytochrome oxidase and alternative oxidase may be related (9). Experiments investigating the regulation of both cytochrome oxidase and alternative oxidase on transcriptional and protein levels are currently being pursued. ACKNOWLEDGMENTS We ‘would like to thank. Roxy ‘Nickels for her technical assistance, Dr. David Douches for providing us with potatoes, Dr. Andrew'Hanson for his advice and suggestion for the mixing experiment, Dr. Sarah Gilmour for isoelectric focusing of the potato proteins, and Lydia Herdies for the two-dimensional gel electrophoresis and blots of S. guttatum proteins. LITERATURE CITED 1. Day DA, Arron GP, Christofferson RE, Laties GG (1978) Effect of ethylene and carbon dioxide on potato metabolism. Stimulation of tuber and mitochondrial respiration and inducement of the alternative path. Plant Physiol 62:820-825 2. Day DA, Arron GP, Laties GG (1980) Nature and control of respiratory pathways in plants: the interaction of cyanide- resistant respiration with the cyanide-sensitive pathway. In The Biochemistry of Plants, vol.2, Academic Press, New York, pp 197-241 117 3. Dizengremel P, Lance C (1976) Control of changes in mitochondrial activities during aging of potato slices. Plant Physiol 58:147-151 4. Dizengremel P, Cader J-C (1980) Effect of aging on the composition of mitochondrial membranes from potato slices. Phytochemistry 19:211-214 5. Deuce R, Christiansen EL, Bonner WD Jr (1972) Preparation of intact plant mitochondria. Biochim Biophys Acta 275:148-160 6. Elthon TE, McIntosh L (1986) Characterization and solubilization of the alternative oxidase of Sauromatum guttatum mitochondria. Plant Physiol 82:1-6 7. Elthon TE, McIntosh L (1987) Identification of the alternative terminal oxidase of higher plant mitochondria. Proc Nat Acad Sci USA 84:8399-8403 8. Elthon TE, Nickels R, McIntosh L (1989) Monoclonal antibodies to the alternative oxidase of higher plant mitochondria. Plant Physiol 89:1311-1317 9. Elthon TE, Nickels RL, McIntosh L ( 1989) Mitochondrial events during the development of thermogenesis in Sauromatum guttatum (Schott). Planta 180:82-89 10. Guy RD, Berry JA, Fegel ML, Heering TC (1989) Differential fractionation of oxygen isotopes by cyanide-resistant and cyanide-sensitive respiration in plants. Planta 177:483-490 11. Hackett DP, Haas DW, Griffiths SR, Niederpreum DJ (1960) Studies on the development of cyanide-resistant respiration in potato tuber slices. Plant Physiol 35:8-19 12. Hurkman J, Tanaka CK (1986) Solubilization of plant membrane proteins for analysis by two-dimensional gel electrophoresis. Plant Physiol 81:802-806 13. Janes HW, Ryohter A, Frenkel C (1979) Factors influencing the development of cyanide-resistant respiration in potato tissue. Plant Physiol 63:837-840 14. James HW, Wiest SC (1980) Comparison between aging of slices and ethylene treatment of whole white potato tubers. Plant Physiol 66:171-174 15. Lambers H (1980) The physiological significance of cyanide-resistant respiration in higher plants. Plant Cell Environ 3:293-302 118 16. Lambewits AM, Sabeurin JR, Bertrand H, Nickels R, McIntosh L (1988) Immunological identification of the alternative oxidase of Neurospora crassa mitochondria. Mol Cell Biol 9:1362-1364 17. Lance C, Chauveau R, Dizengremel P (1985) The cyanide- resistant pathway of plant mitochondria. In R Douce, DA Day, eds, Higher Plant Cell Respiration. Springer-Verlag, New York, pp 202-239 18. Lane GA, Dole M (1956) Fractionation of oxygen isotopes during respiration. Science 123:574-576 19. Larsen. E, Hewlett. B, Jagenderf .A (1986) .Artificial reductant enhancement of the Lowry method for protein determination. Anal Biochem 155:243-248 20. Laties GG (1982) The cyanide-resistant, alternative path in higher plant respiration. Ann Rev Plant Physiol 33:519-555 21. McIntosh L, Meeuse BJD (1978) Control of the development of cyanide-resistant respiration in Sauromatum guttatum (Araceae) . In G Ducet, C Lance, eds, Plant Mitochondria, Elsevier/North Holland Biomedical Press, Amsterdam, pp 339-345 22. Meeuse BJD (1975) Thermogenic respiration in aroids. Ann Rev Plant Physiol 26:117-126 23. Parrish DJ, Leopold AC (1978) Confounding of alternate respiration by lipoxygenase activity. Plant Physiol 62:470-472 24. Rychter A, James HW, Frenkel C (1978) Cyanide-resistant respiration in freshly cut potato slices. Plant Physiol 61:667-668 25. Shingles RM, Arren.GP, Hill RD (1982) Alternative pathway respiration and lipoxygenase activity in aged potato slice mitochondria. Plant Physiol 69:1435-1438 26. Solemes T (1977) Cyanide-resistant respiration in higher plants. Plant Physiol 28:279-297 27. Theologis A, Laties GG (1975) The mechanism of ethylene and cyanide action in triggering the rise in respiration in potato tubers. Plant Physiol 55:73-78 28. Theologis A, Laties GG (1976) Membrane lipid integrity as a prerequisite element of cyanide-resistant respiration in potato slices. Plant Physiol 57:S-93 29. Theologis A, Laties GG (1978) Relative contribution of cytochrome-mediated and cyanide-resistant electron transport in fresh and aged slices. Plant Physiol 62:232-237 119 30. Theologis A, Laties GG (1980) Membrane lipid breakdown in relation to the ‘wound-induced and cyanide-resistant respiration in tissue slices. Plant Physiol 66:890-896 31. Theologis A, Laties GG (1982) Selective enhancement of alternative path capacity in plant storage organs in response to ethylene plus oxygen:A comparative study. Plant Physiol 69:1036-1039 32. Tucker ML, Laties GG (1984) Interrelationship of gene expression, polysome prevalence, and respiration during ripening of ethylene and/or cyanide-treated avocado fruit. Plant Physiol 74:307-315 33. Tucker’.ML, Laties. GG (1984) Comparative effects of ethylene and cyanide on respiration, polysome prevalence, and gene expression in carrot roots. Plant Physiol 75:342-348 NOTE: After publication of this paper, an explanation for the low molecular weight (12-18kD) bands present in Figure 2-3-2 (lanes A2,A3,BZ,B3) was discovered- These bands reacted when a subsequent western blot was probed only with the secondary antibody (anti-mouse IgG-alkaline phosphatase conjugate). These bands, therefore, do not appear to be related to the potato alternative oxidase. CHAPTER 2-48 THE POTATO ALTERNATIVE OXIDASE GENE INTRODUCTION Plant mitochondria possess two pathways of electron transport from the ubiquinone pool to oxygen: the cyanide- sensitive cytochrome pathway which terminates with the cytochrome c oxidase complex, and the cyanide-resistant, SHAM- sensitive pathway which consists of an alternative terminal oxidase protein(s) (11). The alternative pathway can now be studied at the molecular level since two probes have recently become available: monoclonal antibodies against the Sauromatum guttatum alternative oxidase (5) and the S. guttatum alternative oxidase gene (14). The aging of potato (Solanum tuberosum) tuber slices has been a model system in which to study induction of the alternative pathway. The respiration of fresh potato slice mitochondria is almost entirely cyanide-sensitive, while mitochondria from aerobically-aged slices exhibit cyanide- resistant alternative path respiration (4,16). By reaction with the monoclonal antibody against the S. guttatum alternative oxidase (5), the potato alternative oxidase was shown to be a 36kD polypeptide (8). This polypeptide was not detectable on western blots of mitochondrial membrane proteins 120 121 from fresh potato tissue, but was present in mitochondrial membrane proteins from 24-hour-aged slices ( 8) . Thus the relative levels of the potato alternative oxidase protein appeared to rise in parallel to the alternative oxidase capacity in aging potato slice mitochondria, which suggests de novo synthesis of the protein during the aging process (8). To further examine the potato alternative oxidase, its gene has been cloned and its expression studied at the RNA level. MATERIALS AND METHODS Source of Potato Tissue and DNA and Southern Blotting Russet-type potatoes were obtained from a local market. Potato genomic DNA (cv. Russet Burbank) was a gift from Dr. David Douches, Department of Crop and Soil Science, Michigan State University. Digestion of potato genomic DNA (Sug per lane) with restriction enzymes and Southern blotting followed standard protocols (15). The IRP-labelled (6) plasmid containing the cloned potato alternative oxidase gene was used as the probe. Cloning and Sequencing of the Potato Alternative Oxidase Potato tissue was sliced and aged for 24 hours as described previously (8) and total RNA was isolated (12). 122 Messenger RNA was isolated from.total RNA (2) and.a sample was sent to Stratagene Cloning Systems (11099 North Torrey Pines Road, LaJolla, CA 92037) where a cDNA library was constructed in the Lambda ZapII vector. Following the protocols supplied by Stratagene, the phagemid library was screened using the S. guttatum alternative oxidase gene (14) labelled with 32P (6) as a probe. One positive plaque was obtained after screening approximately 35,000 plaques. The clone was converted from the Lambda Zap II phagemid to the pBluescript plasmid form following the protocols from Stratagene. The insert was subcloned from pBluescript into pUC119 in opposite orientations by standard methods ( 15) for sequencing. Single- stranded DNA was prepared from the clones (17) and deletions were made (3). Both strands of each clone and deletion were sequenced using the Sequenase Version 2.0 kit (United States Biochemical) and the methods described therein. Manipulation of Sequence Data The deduced amino acid sequences of the potato and Sauromatum guttatum (14) genes were derived from the DNA sequences and compared using the EDITBASE program (version 0.61, Purdue Research Foundation and USDA/ARS). Secondary structure predictions were made using the Chou-Fasman secondary structure predictor with Rae-Argos modifications for membrane proteins in the Sequence Analysis programs MCF and AMPHI (Antony R. Crofts, University of Illinois). 123 Polymerase Chain Reactions Total RNA from fresh and aged potato slices was prepared according to Yeh et a1. (18). Reverse-transcription of the RNA and subsequent PCR amplifications were performed as described (9) using 10ug potato total RNA per sample. Controls for PCR amplification used lug genomic DNA or 1ng plasmid carrying the potato alternative oxidase gene. PCR amplification included 1 minute of denaturation at 94°C, 2 minutes of annealing at 55°C, and 3 minutes of extending at 72°C per cycle for 40 cycles. Primers used for PCR were synthesized by Dr. Chris Somerville (Applied Biosystems DNA Synthesizer) and provided by Mary Kakefuda: 1. L115], (upstream primer, located at bases 587-601 of the potato sequence): 5'-CGGGATCCGCGATGATGCTGGAG-3' 2. ME; (downstream primer, located at bases 962-976 of the potato sequence): 5'-CGGAATTCCGGCAGCCGCCAGTA-3' RESULTS The Potato Alternative Oxidase Gene and Deduced Amino Acid Sequence The nucleotide and deduced amino acid sequences of the potato alternative oxidase gene are shown in Figure 2-4-1. The potato gene is 1254bp long with a 1032bp open reading frame (ORF). The sequence is G-C rich (68.9% G-C pairs) and 124 Figure 2-4-1. Sequence of the cloned potato alternative oxidase gene. Nucleotide numbers are shown at the left. The amino acid sequence of the protein, as deduced from the nucleotide sequence, is shown above the nucleotide sequence. 61 121 181 241 301 361 421 481 541 601 661 721 781 841 901 961 1021 1081 1141 1201 125 GTTTTATTGGCGATTTTTTCAGATCTCTCGGCACTCCCCGTTCCCCGTTCCCCGGTTGTC M M S S R F A G T A L R Q CTGTCGCTGGATCTCGCCGGAGATGATGAGCTCCCGGTTCGCCGGCACCGCGCTGAGGCA L G P V L F A S A P G A R A A A E P A Y ACTCGGCCCGGTCCTCTTCGCCTCGGCCCCCGGCGCCCGCGCCGCTGCTGAGCCGGCGTA A L L A G A P A A A P T R A A V W L V R CGCGCTGCTGGCGGGTGCTCCCGCAGCCGCGCCGACGCGCGCGGCCGTCTGGCTGGTGCG F P L S R A A S T M S A P A A P E G E T GTTCCCCCTCTCCCGCGCCGCCAGCACGATGTCGGCGCCGGCCGCGCCGGAGGGGGAGAC A A K G D V D V T K K A E G D T E Q K A GGCCGCGAAGGGGGACGTGGATGTGACGAAGAAGGCGGAGGGGGATACGGAGCAGAAGGC V V S Y W G V P P S R V T K E D G S P W GGTGGTGAGCTACTGGGGCGTGCCGCCTTCTAGGGTCACCAAGGAGGATGGATCCCCGTG R W A C F R P W E A Y E S D M S I D L K GCGTTGGGCCTGCTTCCGGCCATGGGAGGCGTACGAGTCGGACATGTCAATCGATCTGAA K H H A P T T F L D K M A F W T V K S L GAAGCACCACGCTCCCACCACGTTCCTCGACAAGATGGCCTTCTGGACCGTGAAGTCCCT R W P T D I F F Q R R Y G C R A M M L E CCGCTGGCCCACCGACATCTTCTTCCAGAGGCGGTACGGCTGCCGGGCGATGATGCTGGA T V A A V P G M V G G L L L H L K S L R GACGGTGGCGGCGGTGCCGGGGATGGTGGGCGGGTTGCTGCTCCACCTCAAGTCGCTGCG R F E H S G G W I K A L L E E A E N E R CCGGTTCGAGCACAGCGGCGGGTGGATCAAGGCCCTCCTTGAGGAGGCGGAGAACGAGCG M H L M T F M E V S Q P R W Y E R A L V GATGCACCTGATGACCTTCATGGAGGTGTCGCAGCCGCGGTGGTACGAGCGGGCGCTGGT L A V Q G V F F N A Y F L G Y L L S P K GCTGGCGGTGCAGGGCGTCTTCTTCAACGCCTACTTCCTCGGCTACCTCCTCTCCCCCAA F A H R V V G Y L E E E A I H S Y T E F GTTCGCCCACCGGGTGGTGGGCTACCTGGAGGAGGAGGCCATCCACTCCTACACCGAGTT L K E I D K G T I D N V P A P A I A L D CCTCAAGGAGATCGACAAAGGCACCATCGACAACGTGCCCGCGCCCGCCATCGCCCTGGA Y W R L P P G S T L R D V V M V V R A D CTACTGGCGGCTGCCGCCGGGCTCCACCCTCCGCGACGTCGTCATGGTCGTCCGCGCCGA E A H H R D V N H F A S D V H Y Q G M Q CGAGGCCCACCACCGCGACGTCAACCATTTCGCCTCGGACGTCCATTACCAGGGGCATCA L K E A P A P L G Y H * GCTGAAGGAGGCGCCGGCGCCGCTCGGGTACCACTGAGCAGCACCACACTTCCCTGTTGC GGAAGCGGCATGTGAATCCATGGTTGGAAGAGTTTGAAGTGGAAGGTTTGCATTACGCAT TAGCATCATTAATACTGGGTTACTCCATGCTTAAAAAATTTCTACTAAAAAAAA Figure 2-4-1. 126 especially shows a preference for G (47.8%) or C (45.5%) in the third base of each codon. This is comparable to the Sauromatum guttatum gene (14), which has 68.2% G-C pairs and shows a similar third base preference. The potato ORF has 344 codons and could potentially encode a 41kD polypeptide. This is more than adequate to completely encode the observed 36kD potato alternative oxidase protein (8). The potato protein, however, is likely to be made as a larger precursor with a transit peptide that is cleaved upon import into the mitochondria. The S. guttatum alternative oxidase was shown to be made initially as a 42ijprecursor polypeptide (14). By homology to the S. guttatum deduced amino acid sequence which shows a 66-amino-acid transit peptide (Figure 2-4-2) , the potato sequence appears to have a transit peptide of 62 amino acids. Without the proposed transit peptide, the potato deduced amino acid sequence would be 282 amino acids long and could encode a 34kD protein. Figure 2-4-3 shows a Southern blot of potato genomic DNA digested.with several restriction enzymes and probed with the cloned potato alternative oxidase gene. One or a small number of bands hybridized.in.each lane, which suggests that the gene is present in low copy number in the genome. The Southern data also shows that the potato alternative oxidase gene, like those of S. guttatum (14) and Neurospora (10), is encoded in the nucleus. 127 potato 1 MMSSRFAGTAL RQLGPVLFAS APGARAAAEPAYALLAG S. g. 1 .....LV....C...SH.PVPQYL.AL.PT.DT.SS..H. potato 39 APAAAPTRAAVWLV RFPLSRAASTMSAPAAPEG ETAA S. g. 41 CS....AQR.GLWPPSW.SPP.H...L....QDG.K.K.. potato 76 KGDVDVTKKAEGDTEQKAVVSYWGVPPSRVTKEDGSPWRW S. g. 81 GTAGKOPPGEDOGAOKEOOOODOA.COOKOSOOOOOE... potato 116 ACFRPWEAYESDMSIDLKKHHAPTTFLDKMAFWTVKSLRW S. g. 121 T0.0000TOQAOL...OHOOOVOOOIOOOLOLR.DOA... potato 156 PTDIFFQRRYGCRAMMLETVAAVPGMVGGLLLHLKSLRRF S. g. 161 OOOOOOOOOOAOOOOOOOOOOOOOOOOOOVOOOOOOOOOO potato 196 EHSGGWIKALLEEAENERMHLMTFMEVSQPRWYERALVLA S. g. 201 0.0..COROOOOOOOOOOOOOOOOOOOAOOOOO0...... potato 236 VQGVFFNAYFLGYLLSPKFAHRVVGYLEEEAIHSYTEFLK S. g. 241 0.0.0.000...O...OOOOOOOOOOOOOOOOOOOOO0.0 potato 276 EIDKGTIDNVPAPAIALDYWRLPPGSTLRDVVMVVRADEA S. g. 281 DOOSOAOQDCOOOOOOOOOOOOOQOOOOOOOOTOOOOOOO potato 316 HHRDVNHFASDVHYQGHQLKEAPAPLGYH S. g. 3210.0.0....OOOOOODLEOOTTOOOO... Figure 2-4-2. Comparison of deduced amino acid sequences of the alternative oxidase genes from potato and Sauromatum guttatum (S. 9.). Amino acids in the S. guttatum sequence which are identical to the corresponding amino acids in the potato sequence are denoted by dots. Blank spaces in the potato sequence indicate amino acids present in S. guttatum but not in potato. Letters in bold type indicate the likely start sites of the mature proteins. 128 E v s P s 23.1 -' I 9.42— 6.56- 4.37 -. 2.32 - 2.03 - Figure 2-4-3. Southern blot of potato genomic DNA. Potato genomic DNA was restricted, electrophoresed, blotted, and probed with the potato alternative oxidase clone. Restriction enzymes used: E=EcoRI, V=PvuII, B=BstEII, P=PstI, S=SalI. Molecular weight standards in kb at left. 129 Coaparison of the Deduced Amino Acid Sequences of the Potato and sauromatum guttatum Alternative Oxidase Genes Comparison of the deduced amino acid sequences of the potato and S. guttatum alternative oxidase genes (Figure 2-4- 2) shows striking similarities. The potato sequence contains many charged amino acids: 38 basic amino acids (11%) and 38 acidic amino acids (11%), which.is comparable to the number of charged amino acids in the S. guttatum sequence. The potato sequence also appears to possess a transit peptide (amino acids 1-62). Both transit peptides start with a conserved region (12 of 15 amino acids identical) and the transit peptide/mature protein border region (as determined from N- terminal amino acid sequencing of the S. guttatum mature protein; 8) is conserved (7 of 8 amino acids identical). From secondary structure-predicting computer programs, regions in the potato and S. guttatum deduced amino acid sequences are predicted to form highly conserved membrane- spanning helices (Figure 2-4-4). The first membrane helix is predicted to occur between amino acids 168-186 in the potato sequence and is 95% identical to amino acids 173-192 in the S. guttatum sequence. The comparable nucleotide sequences are 93% identical. The second membrane helix, which is 100% identical at the amino acid level and 91% identical at the nucleotide level between the two plants, is predicted to occur at amino acids 228-250 in the potato sequence and amino acids 233-255 in the S. guttatum sequence. The area between these 130 Figure 2-4-4. Membrane helix predictions for alternative oxidase proteins. lPotential to form .alpha helices was predicted for the potato and S. guttatum alternative oxidase deduced amino acid sequences. Abscissas indicate amino acid position (numbered as in Figure 2-4-2). Ordinates indicate helical potential (Pa) with a window of 11 amino acids. Areas of the curves lying above the horizontal lines are predicted to form helices. Solid lines represent helical potentials obtained using Chou-Fasman parameters, while dotted lines represent areas additionally predicted to form membrane helices using Rae-Argos helix parameter modifications for membrane proteins. Arrows indicate the borders between the proposed transit peptides and mature proteins. 44.6 6.36.... com com 8.. »»> .> 1.....4».-.<4‘ ....4.1.1..<1. >1. L... 43.6.6560 23.6.2965. >___ oocoo> 131 a."au“asses;““6136"...gigging.“Haulage". "an“ “agile“; «E66966... $626.8.» 9.686 132 two predicted membrane helices is probably helical but may or may not span the membrane and.is also highly conserved between the two plants, with 94% identical amino acids and 95% identical nucleotides. Areas outside the predicted membrane helix regions are less conserved. The proposed transit peptides have only 32% identical amino acids, and the potato peptide is predicted to contain an additional membrane Ihelix not. present in .S. guttatum (Figure 2-4-4, left of the arrow). The region between the transit peptide and the first proposed membrane helix (amino acids 63-167 in potato) is 59% identical at the amino acid level and 63% identical at the nucleotide level to the corresponding 5: guttatum region (amino acids 67-171). The regions after the second proposed membrane helix to the end of the ORF (amino acids 251-344 in potato and 256-349 in S. guttatum) have 86% identical amino acids.and 90% identical nucleotides. Expression of the Potato Alternative Oxidase Gene The mRNA from the potato alternative oxidase gene appears to be of very low abundance. No reproducible signal was detected on Northern blots using 32P-labelled DNA probes from the potato or S. guttatum genes or ”P-labelled antisense RNA probes generated from the potato clone (data not shown). Therefore, detection of the message was attempted using reverse transcription of the RNA samples to create cDNA 133 followed by PCR amplification (Figure 2-4-5). From the location of the PCR primers in the potato gene sequence, the specific PCR amplification product was predicted to be about 390bp long. When a plasmid carrying the potato alternative oxidase gene was PCR-amplified as a control, the predicted.band.was the only band visible on both the ethidium- bromide stained gel (data not shown) and the Southern blot of the gel after hybridization to the 32P-labelled potato alternative oxidase clone (Figure 2-4-5, lane 1). In the reverse-transcribed, PCR-amplified potato total RNA samples, however, a band of the predicted size was not seen on ethidium bromide-stained gels although several other bands (non- specific amplification products) were visible (data not shown). After Southern hybridization, the predicted band.was visible in fresh, 12-hour-aged, and 24-hour-aged potato samples (Figure 2-4-5, lanes 2-4). Because the PCR technique is very sensitive, small amounts of contaminating DNA in the mRNA.samples could.have given.rise to these signals. IHowever, when genomic DNA was PCR-amplified for comparison (lane 5), the size of the PCR product from genomic DNA was about 150bp larger than the product from reverse-transcribed mRNA. This suggests that the signals in lanes 2-4 were not caused by contaminating DNA, as well as providing evidence for an intron(s) in the potato gene located between the primer binding sites. It therefore appears that the alternative oxidase:mRNA.is present in potato tubers both before and after aging. Since inherent limitations in the PCR technique 134 1419—- 517— 396— 0.. 195— Figure 2-4-5. Detection of the potato alternative oxidase mRNA by PCR amplification. Lane 1: PCR amplification of 1ng of plasmid DNA carrying the potato alternative oxidase gene. Lanes 2-4: PCR amplification of 10ug of reverse-transcribed total RNA from fresh, 12-hour-aged, and 24-hour-aged potato tuber tissue, respectively. Lane 5: PCR amplification of lug of potato genomic DNA. Molecular weight markers in basepairs at left. 135 preclude accurate quantitation of the levels of message in the samples (7), differences in message levels between fresh and aged tissue may exist but cannot be detected. DISCUSSION To further examine the potato alternative oxidase, a cDNA clone containing the potato alternative oxidase gene was isolated and sequenced. Southern blotting (Figure 2-4-3) showed that the gene is present in one or a low number of copies. The presence of multiple bands in some of the restriction digests, given that the Russet Burbank cultivar is tetraploid, could indicate the possibility of more than one allele of the gene. However, the most likely explanation is that the potato alternative oxidase gene contains one or more introns with additional restriction sites not present in the cDNA clone. Multiple bands were seen on the Southern blot when EcoRI, SalI, and PstI, restriction enzymes that do not out within the coding sequence, were used. The PCR amplification product from genomic DNA (Figure 2-4-5, lane 5) was larger than the product from reverse-transcribed mRNA (lanes 2-4), which also suggests the presence of an intron(s) between the primer binding sites in the potato alternative oxidase gene. A recently isolated genomic clone of the S. guttatum alternative oxidase gene has been shown to contain an intron (Rheads and McIntosh, unpublished data). The alternative oxidase gene from potato was the second 136 to be cloned, so it is now possible to compare the deduced amino acid sequences from.potato (a dicot) and S. guttatum (a monocot lily). Such comparisons have revealed areas of remarkable conservation between these distantly-related species. Highly conserved regions may represent functional domains of the protein. The most highly conserved regions are 2 areas strongly predicted to form membrane-spanning helices (Figure 2-4-4). These areas may provide an anchor into the inner mitochondrial membrane. One or both of these conserved membrane helices (and/or other hydrophobic areas) might be involved with quinone binding, since the alternative oxidase is believed to accept electrons directly from ubiquinone (11) , a molecule residing within the mitochondrial membrane. However, quinone binding sites have not yet been defined in either mitochondrial or photosynthetic protein complexes, so no particular likely site can be delineated here. The highly conserved region between the two predicted membrane-spanning helices (amino acids 186-236 in the potato sequence) is also predicted to be helical but not necessarily membrane-spanning. It contains many charged amino acids, and helical wheel diagrams (not shown) suggest that the charged residues lie predominantly on one side of the helix and nonpolar residues on the other side. Such a helix might lie on the surface of the membrane, perhaps interacting with other proteins or smaller molecules via the charged residues. A site important to A. maculatum alternative oxidase activity was localized to the inner face of the mitochondrial membrane 137 (13) . This highly-conserved, potential surface helix in potato and S. guttatum might be a similar site. The alternative oxidase is generally believed to pass electrons to O2 to form H20 (11) , so it may be expected to possess an O2 binding site. Many enzymes which bind O2 contain a.metal cofactor such.as Fe or‘Cu, and the alternative oxidase has been variously reported to contain one or the other of these metals (see Chapter 2-1) . In cytochrome oxidase, cysteine and histidine residues have been proposed to bind metal cofactors (1). If the gene isolated here encodes an O2 binding site, 12 conserved histidines and 2 conserved cysteines in the potato and S. guttatum amino acid sequences could be candidates for ligating a metal cofactor(s). Although general areas of high sequence conservation are probably important to alternative oxidase function, the potato and S. guttatum alternative oxidase sequences are so highly conserved that it is difficult to pinpoint amino acids likely to be crucial to activity. It would be useful to have sequences of similar genes from. more distantly related organisms such as yeast or fungi, as well as more plant sequences for further comparison. The remarkable conservation between the potato and S} guttatum alternative oxidase genes and their deduced amino acid sequences suggests that the gene has been under selective pressure to remain functional in non-thermogenic plants, as well as in the aroid plants in which it has an obvious function. The alternative oxidase, therefore, is likely to 138 provide some vital service(s) to other plants, although the exact nature of this function(s) remains to be elucidated. The alternative oxidase message appears to be of low abundance in potato tubers. Northern blots can detect mRNAs comprising 0.001% of the total mRNA (15), yet appropriate conditions could not be found which yielded signals from total RNA from either fresh or aged tuber tissue. The very sensitive technique of PCR amplification, however, allowed detection of the alternative oxidase message. The message was present in both fresh and aged tuber tissue, which suggests some constitutive expression of the alternative oxidase gene. However, the levels of the alternative oxidase protein increase with aging (8), so some regulation must occur at some level. Unfortunately, the amounts of alternative oxidase message could not be quantified herein, so regulation at the levels of transcription and/or mRNA stability can neither be confirmed nor ruled out. Even if the alternative oxidase gene were constitutively transcribed, post-transcriptional regulation of the rates of translation and/or protein turnover could lead to increased levels of the alternative oxidase protein in aged tissue. The method of regulation of alternative oxidase gene expression in aging potato tuber slices therefore remains unknown. 139 ACKNOWLEDGEMENTS Thanks are due to Mary Kakefuda for suggesting appropriate PCR cycling conditions, and to David Rheads for allowing inclusion of his S. guttatum data in Figures 2-4-2 and 2-4-4 for comparison. LITERATURE CITED 1. Capaldi RA (1990) Structure and function of cytochrome c oxidase. Ann Rev Biochem 59:569-596 2. Cashmere A (1982) The isolation of poly Af messenger RNA form higher plants. In: Edelman et al., eds, Methods in chloroplast molecular biology. Elsevier Biomedical Press, pp.387-392 3. Dale RM, McClure BA, Houchins JP (1985) A rapid single- stranded cloning strategy for producing a sequential series of overlapping clones for use in DNA sequencing: Application to sequencing the corn mitochondrial 18 S rDNA. Plasmid 13:31-40 4. Dizengremel P, Lance C (1976) Control of changes in mitochondrial activities during aging of potato slices. Plant Physiol 58:147-151 5. Elthon TE, Nickels R, McIntosh L (1989) Monoclonal antibodies to the alternative oxidase of higher plant mitochondria. Plant Physiol 98:1311-1317. 6. Feinberg AP, vegelstein B (1982) A technique for radiolabelling DNA restriction fragments to high specific activity. Anal Biochem 132:6-13. 7. Gilliland G, Perrin S, Bunn HF (1990) Competitive PCR for quantitation of mRNA. In: PCR Protocols: A guide to Methods and Applications. Academic Press, New York, pp. 60-69 8. Hiser’C,.McIntesh.L (1990) Alternative oxidase of potato is an integral membrane protein synthesized de novo during aging of tuber slices. Plant Physiol 93:312-318 9. Kawasaki E (1990) Methods for RNA-PCR. In: Perkin-Elmer Spring 1990 Biotechnology Catalog, p.19 140 10. Lambowits AM, Sabeurin JR, Bertrand H, Nickels R, McIntosh L (1988) Immunological identification of the alternative oxidase of Neurospora crassa mitochondria. Mol Cell Biol 9:1362-1364 11. Lance C, Chauveau M, Dizengremel P (1985) The cyanide- resistant pathway of plant mitochondria. In: R Douce, DA Day, eds, Higher Plant Cell Respiration. Springer-Verlag, New York, pp.202-247 12. McIntosh L, Cattolico RA (1978) Preservation of algal and higher plant ribosomal RNA integrity during extraction and electrophoretic quantitation. Anal Biochem 91:600-612 13. Rasmussen AG, Meller IM, Palmer JM (1989) Component of the alternative oxidase localized to the matrix surface of the inner membrane of plant mitochondria. FEBS Lett 259:311-314 14. Rheads DM, McIntosh L (1991) Isolation and characterization of a cDNA clone encoding an alternative oxidase protein of Sauromatum guttatum (Schott) . Proc Nat Acad Sci USA 88:2122-2126 15. Sambroek J, EF Fritsch, Maniatis T (1989) Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory Press, New York 16. Theologis A, Laties GG (1978) Relative contribution of cytochrome-mediated and cyanide-resistant electron transport in fresh and aged slices. Plant Physiol 62:232-237 17. Viera J, Messing J (1987) Production of single-stranded plasmid DNA. In: wu R, Grossman L, eds, Methods in Enzymology 153:3-11 18. Yeh K-W, Juang R-H, Su J-C (1991) A rapid and efficient method for RNA isolation from plants with high carbohydrate content. BRL Focus 13:102-103 SUMMARY OF THESIS This thesis has been an exploration of the molecular and developmental aspects of respiratory enzymes of higher plant mitochondria. It emphasized the two terminal oxidases of the mitochondrial electron transport chain: the cytochrome oxidase, which is similar to the mammalian enzyme; and the alternative oxidase, an additional terminal oxidase present in higher plant mitochondria. PART 1: RESPIRATORY COMPLEXES AND CYTOPLASMIC MALE STERILITY Results and Implications In this section of the thesis, an effort was made to correlate the CMS phenotype with molecular changes in the mtDNA of sugarbeets. The S,4-CMS line in which CMS was induced by y-irradiation and the BMC-CMS line with a male- sterile cytoplasm derived from.Beta maritima were compared to their fertile progenitor cytoplasm and the male-sterile cytoplasm in common agricultural use. The location and expression of three mitochondrial genes (atp6, atp9, coxII) that have been associated with mtDNA rearrangements in CMS 141 142 lines of other species (1,3) were examined in the sugarbeet lines. The atp6 and coxII genes (but not the atp9 gene) were found to be involved in mtDNA rearrangements that caused altered gene expression in both CMS lines. These mtDNA rearrangements may have been related to the production of the CMS phenotype. However, the correlation between these specific mtDNA alterations and the CMS phenotype may have been coincidental, and CMS may have resulted from alterations in another area(s) of the genome. Future Directions To strengthen or disprove the correlation between these specific mtDNA rearrangements and CMS, examination of the expression of the three mitochondrial genes in male floral tissue would be required. Of particular interest would be the tapeta and developing microspores in young anthers, since this is where the CMS phenotype is manifested and tapetal abnormalities have been observed in CMS sugarbeet lines (2). Since it was not possible to isolate intact mtRNA from sugarbeet anthers, in situ hybridization may be the technique of choice to examine gene expression in anthers. Since little is known about the expression of genes encoding subunits of the respiratory complexes in normal fertile anthers, further investigation along this line would be enlightening not only to CMS research, but also to the general studies of floral development and mitochondrial functioning in plants. 143 For further research into the molecular cause(s) of CMS in the S,4-CMS and BMC-CMS sugarbeet lines, antibodies against the protein products encoded by the three genes would be useful for looking for altered size or mobility of these proteins using one- and two-dimensional PAGE. Alterations in the size of the mRNA could still result in normal proteins if the alterations were in non-coding regions. Sequencing of the mtDNA rearrangements in the CMS lines would be necessary to determine if the location of the alteration(s) was within the gene(s) or promoter region(s), where it would be more likely to cause alterations in the protein product(s) than rearrangements further away. PART 2: GENETIC AND DEVELOPMENTAL ASPECTS OF ALTERNATIVE OXIDASE Results and Implications Previous reports which showed two pea cultivars to differ in their alternative oxidase capacity (8) could not be repeated. Similar alternative oxidase capacities and proteins were found for both Alaska and Progress No.9 cultivars. The results of Musgrave et a1. (8) have been widely cited to support the overflow hypothesis for the function of the alternative oxidase (7). The results presented here weaken this support. Research begun as early as 1960 (6) made aging potato 144 tuber slices a classical system in which to study cyanide- resistant respiration. IHowever, until the recent development of antibodies against. the alternative oxidase (4), investigation of the molecular changes occurring during the aging process was impossible. This thesis has built upon the earlier observations by applying molecular probes to the potato slice system. Using the monoclonal antibodies against the alternative oxidase (4), the increased alternative path capacity that developed with aging was correlated with increased levels of alternative oxidase protein. This result suggested de novo synthesis of the alternative oxidase‘ occurred during aging. This thesis also showed that the potato alternative oxidase has two isoforms of different pIs. Cloning and sequencing of the potato alternative oxidase gene revealed regions of deduced amino acid sequence with high homology to the Sauromatum guttatum alternative oxidase (9), some of which. were predicted to form 'membrane-spanning helices. These highly conserved areas may be functional domains of the alternative oxidase. Using the potato alternative oxidase gene cloned herein as a probe, the level of alternative oxidase:mRNA.was found to be very low in potato tubers. Some constitutive expression of the gene during aging was found, but the level of regulation could not be determined. 145 Future Directions The sequencing of the potato gene, the second alternative oxidase gene to be isolated, has allowed comparisons to the S. guttatum gene (9) to find potential functional regions of this enzyme. Comparison to alternative oxidase gene sequences from other higher plants, algae, and fungi, as they become available, will allow further delineation of potentially important regions of the protein. Since the alternative oxidase has been proposed to function during wounding or pathogen attack, the potato clone could be used for in situ hybridizations to examine alternative oxidase expression in these situations. Having a cDNA clone of this protein now allows testing of the function of the alternative oxidase by using antisense constructs to eliminate or reduce alternative oxidase mRNA in transgenic potato plants. Further investigation of the regulation of the potato alternative oxidase gene during aging of tuber slices should include examination of potential co-regulation of the cytochrome oxidase. The shunting of electrons to the alternative oxidase during thermogenesis in aroid plants has been associated with reduced cytochrome path capacity and reduced expression of cytochrome oxidase subunit genes (5). If antibodies to cytochrome oxidase subunits (including the potentially regulatory, nuclearly-encoded subunits) can be obtained, then any fluctuations in the levels of these proteins during aging could be found. Using the antibodies to 146 screen cDNA libraries could provide clones of the corresponding genes, whose regulation could then be examined. These antibodies and clones would be useful tools for molecular research on both terminal oxidases. LITERATURE CITED 1. Boeshere ML, Hanson MR, Izhar S (1985) A variant mitochondrial DNA rearrangement specific to petunia stable sterile somatic hybrids. Plant Mol Biol 4:125-132 2. Chapman GP (1987) The tapetum. Int Rev Cytol 107:111-125 3. Dewey RE, Levings CS III, Timothy DH (1986) Novel recombinations in the maize mitochondrial genome produce a unique transcriptional unit in the Texas male-sterile cytoplasm. Cell 44:439-444 4. Elthon TE, Nickels R, McIntosh L (1989a) Monoclonal antibodies to the alternative oxidase of higher plant mitochondria. Plant Physiol 89:1311-1317 5. Elthon TE, Nickels RL, McIntosh L (1989b) Mitochondrial events during the development of thermogenesis in Sauromatum guttatum (Schott). Planta 180:82-89 6. Hackett DP, Haas DW, Griffiths SR, Niederpreum DJ (1960) Studies on the development of cyanide-resistant respiration in potato tuber slices. Plant Physiol 35:8-19 7. Lambers H (1982) Cyanide-resistant respiration: A non- phosphorylating electron transport pathway acting as an energy overflow. Physiol Plant 55:478-485 8. Musgrave ME, Strain BR, Siedow JN (1986) Response of two pea hybrids to C02 enrichment: A test of the overflow hypothesis for alternative respiration. Proc Nat Acad Sci USA 83:8157-8161 9. Rheads DM, McIntosh L ( 1991) Isolation and characterization of a cDNA clone encoding an alternative oxidase protein of Sauromatum guttatum (Schott). Proc Nat Acad Sci USA 88:2122- 2126 ”11111111111114