UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Evolution and expression of the superoxide dismutase of the archaebacterium halobacterium cutirubrum May, Bruce Pearson 1989

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

Item Metadata


831-UBC_1989_A1 M39.pdf [ 7.05MB ]
JSON: 831-1.0098355.json
JSON-LD: 831-1.0098355-ld.json
RDF/XML (Pretty): 831-1.0098355-rdf.xml
RDF/JSON: 831-1.0098355-rdf.json
Turtle: 831-1.0098355-turtle.txt
N-Triples: 831-1.0098355-rdf-ntriples.txt
Original Record: 831-1.0098355-source.json
Full Text

Full Text

EVOLUTION AND EXPRESSION OF THE SUPEROXIDE DISMUTASE OF THE ARCHAEBACTERIUM HALOBACTERIUM CUTIRUBRUM By BRUCE PEARSON MAY B.Sc. (Hons.), University of Winnipeg, 1983 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES GENETICS PROGRAM We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA JULY, 1989 (c) copyright Bruce Pearson May, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. The University of British Columbia Vancouver, Canada Date DE-6 (2/88) i i Abstract Archaebacteria are a diverse group of bacteria that share a number of fundamental biochemical characteristics that distinguish them from other bacteria (eubacteria) and eucaryotes. They are believed to be an independent lineage that diverged from other organisms more than 3.5x10$ years ago. To study the evolution of oxygen tolerance in the archaebacteria, the superoxide dismutase (SOD) and SOD-encoding gene of the extremely halophilic archaebacterium Halobacterium cutirubrum were investigated. Purification of the SOD activity of this organism resulted in a single Mn-containing enzyme with subunits of Mr 25000 that probably associate as tetramers. It has - maximal activity in 2 M KC1, reflecting adaptation to the high salt concentrations within halobacterial cells. Assays of activity in the presence of SOD inhibitors (cyanide, azide, and hydrogen peroxide) gave a unique pattern of responses. The sequence of 51 of 56 amino acid residues at the amino-terminus was determined and found to be homologous to the amino-terrnini of SODs of eubacteria. The gene, designated sod, that encodes the halobacterial Mn SOD was cloned by using as a probe a mixture of 20mer oligonucleotides corresponding to codons for amino acid residues 27-33. The sod gene and 5' and 3' untranslated regions were located on a Sau3AI genomic DNA fragment of 1127 nucleotides. The deduced amino acid sequence is 200 residues long and has 39 to 42% identity with Mn-containing SODs of eubacteria and mitochondria. This homology may be due to either lateral transfer of the gene between eubacteria and archaebacteria or to high amino acid sequence conservation in the enzyme during the divergent evolution of eubacteria and archaebacteria. Transcription of the gene i i i initiates only 2-3 nucleotides upstream of the translation initiation codon. The 5 ' end of the transcript does not contain a purine-rich Shine-Dalgarno element and the promoter does not contain elements that closely match consensus sequences found in other archaebacterial promoters. Termination of transcription occurs at five consecutive T residues that are preceded by a GC-rich region containing short inverted repeats. The gene is basally expressed in anaerobically grown cells, and is inducible by both oxygen and paraquat, a generator of oxygen radicals. The same transcription initiation site is used in both types of expression, suggesting that one promoter is responsible. In addition to the single copy of the sod gene, the „genome of H. cutirubrum contains a sequence that is very closely related to sod, but does not encode the previously purified S O D of this organism. This sod-like gene (slg) has 8 7 % nucleotide identity with sod in the coding region and the predicted protein has 8 3 % amino acid identity with the Mn S O D protein. It conserves the four amino acid residues that bind the. prosthetic Mn atom, yet no superoxide dismutase activity corresponding to the slg product was detected in cell extracts. The 5 ' and 3 ' flanking regions of sod and slg. are unrelated and s_lg. is not inducible by paraquat. Transcription of slg initiates 1 3 nucleotides upstream of the translation initiation codon and the promoter contains elements (TTCGA and TTAA) that closely match archaebacterial consensus sequences. Termination of transcription occurs in a tract of seven T residues that is preceded by short inverted repeats. The slg gene may be sufficiently divergent from sod to encode a different function. The divergence of slg and sod appears to be the product of an unusual mode of evolution. iv T a b l e of C o n t e n t s A b s t r a c t i i Table of Contents i v L i s t of Tables v i i L i s t of Figures v i i i A c k n o w l e d g e m e n t s x L i s t of Abbreviat ions x i I n t r o d u c t i o n 1 Superoxide Dismutase 1 E a r l y E v o l u t i o n : three lineages 4 Later evolution: the appearance of oxygen 8 A r c h a e b a c t e r i a 8 H a l o p h i l i c Archaebacter ia 1 0 Superoxide Dismutase i n Archaebacteria 1 4 A i m s of This Study 1 5 Mater ia ls and Methods 1 6 M a t e r i a l s 1 6 Culture and Strains of Bacteria 1 6 Assays of S O D and Protein 1 8 Purification of S O D from H . c u t i r u b r u m 1 8 Electrophoresis of Proteins 2 0 M e t a l A n a l y s i s 2 0 A m i n o A c i d Sequencing 2 0 Plasmids and Phages 2 0 V Analysis of DNA 20 Analysis of RNA 2 2 Mobility Shift Assay 24 Results I: Purification and Characterization of SOD from H. cutirubrum 25 Results of the Purification of SOD from H. cutirubrum 2 5 Molecular Weight of Halobacterial SOD 3 2 Response to Inhibitors 3 2 Metal Content 3 8 Salt Dependence of Activity 3 9 Amino Acid Sequence of the N-terminus , 4 2 Results II: Cloning and Characterization of the SOD-encoding Gene (sod) and a sod-like Gene (slg) from H. cutirubrum 4 6 Cloning of sod 4 6 Characterization of sod 46 Estimation of Copies of the sod Gene and Detection of the SQd-like Gene (slg) 5 6 Characterization of the s lg. Gene 6 3 Results III: Analysis of Expression of the sod and slg. Genes 6 7 Response of H. cutirubrum to Paraquat 6 7 Mapping of the 5' Ends of the sod Gene Transcripts 7 2 Mapping of the 3' Ends of the sod Gene Transcripts 7 7 Mapping and Quantitation of the slg. Gene Transcripts 7 7 Comparison of Promoters and Terminators 8 2 v i Discussion and Conclusion 9 0 Evolution of Mn/Fe-containing SOD 9 0 Adaptation of Halobacterial SOD to High Salt Concentrations 9 2 Expression of the sod and slg. Genes 9 2 The Enigma of slg 9 6 Future Research Prospects 102 References 105 v i i List of Tables Table 1. The archaebacterial phenotype. 9 Table 2. Growth characteristics of extremely halophilic archaebacteria. 1 1 Table 3. Purification of SOD from H. cutirubrum. 2 6 Table 4. Sensitivity of halobacterial SOD and bovine Cu-Zn SOD to inhibitors. 3 7 Table 5. Amino acid compositions of the MnSOD from H. cutirubrum. the slg. product from H. cutirubrum. the MnSOD from B_. stearothermophilus. and the MnSOD from E. coli. 9 3 Table 6. Comparison of pairs of related genes from H. halobium and H. cutirubrum. 101 v i i i List of Figures Figure 1. Evolution of the archaebacteria, eubacteria, and eucaryotes from a common ancestor, the hypothetical progenote. 7 Figure 2. Salting-out chromatography on Sepharose CL4B. 2 8 Figure 3. Sulfate-mediated ion exchange chromatography on DEAE Sepharose CL6B. 3 1 Figure 4. Purification of SOD as monitored by SDS-discontinuous gel electrophoresis. 3 4 Figure 5. Analytical gel filtration of H. cutirubrum SOD. 3 6 Figure 6. Salt dependence of enzyme activity. 41 Figure 7. Amino acid sequence of the amino-terminus of SOD from H_. cutirubrum. 44 Figure 8. (A) Sequence of a 20mer oligonucleotide probe mixture and (B) Construction of size fractionated librairies of genomic Sau3Al fragments from H_. cutirubrum. 4 8 Figure 9. The nucleotide sequence of the 571 bp subfragment of the 600 bp Sau3AI fragment. 50 Figure 10. Sequencing strategy of the 1.1 kb (1127 bp) Sau3AI fragment. 5 2 Figure 11. Nucleotide sequence of the 1127 bp Sau3AI fragment containing the sod gene. 5 4 Figure 12. Alignment of the deduced H_. cutirubrum amino acid sequence with that of the MnSOD of Bacillus stearothermophilus. 5 8 Figure 13. Southern blot analysis of H. cutirubrum DNA. 60 Figure 14. Sequencing strategy of the slg. gene found on the 1.8 kb PstI fragment. 6 2 ix Figure 15. Comparison of the nucleotide and amino acid sequences of the slg. and sod genes. 6 5 Figure 16. Induction of halobacterial SOD activity and sod mRNA in response to paraquat. 6 9 Figure 17. Northern blot analysis of RNA from untreated and paraquat-treated cells. 7 1 Figure 18. Strategy and results of mapping the sod transcript. 7 4 Figure 19. Actual results of the mapping of the 5' and 3' ends of the sod transcript. 7 6 Figure 20. Detection of the 3' ends of sod mRNA in RNA from anaerobic cells. 7 9 Figure 21. Strategy and results of mapping the slg. transcript. 8 1 Figure 22. Actual results of mapping the slg. transcript. 8 4 Figure 23. Comparison of 5' and 3' flanking regions of the sod and slg genes. 8 6 Figure 24. Alignment of the sod and slg. promoters with promoters of other genes of halobacteria. 8 8 Figure 25. Conformation of an Fe SOD. 100 X Acknowledgements I thank my supervisor Dr. Patrick P. Dennis for his advice and encouragement throughout my studies. Thanks also to Drs. Ross MacGillivray, Tom Beatty, and Phil Bragg for watching over me and to the MRC for watching over my funding. I am forever indebted to Lawrence "Mad Dog" Shimmin for his boundless knowledge, curiosity, advice, and homogenized milk. Much appreciated is the friendship and help of the members of the lab: Willa Downing, Deidre deJong-Wong, Phalgun Joshi, Peter Durovic, and Janet Yee. Finally and most importantly thanks are given to my parents, my brother, and Diana Mawson ("But Bruce, what does this have to do with the rest of the world") for keeping perspective. x i Abbreviations A adenine (in nucleic acids) or alanine (in proteins) A46O absorbance at 460 nanometers Ala alanine AMV avian myeloblastosis virus Arg arginine Asn asparagine Asp aspartic acid ATP adenosine triphosphate bisTris bis(2-hydroxyethyl)imino-tris(hydroxymethyl)methane b p base pairs BSA bovine serum albumin C cytosine (in nucleic acids) or cysteine (in proteins) Cys cysteine D aspartic acid DETPA diethylenetriaminepentaacetic acid dATP deoxyadenosine triphosphate dCTP deoxycytidine triphosphate dGTP deoxyguanosine . triphosphate DNA deoxyribonucleic acid dNTP deoxynucleotide triphosphate dTTP deoxythjmidine triphosphate E glutamic acid EDTA ethylenediaminetetraacetic acid F phenylalanine FDM formamide dye mix x i i G guanine (in nucleic acids) or glycine (in proteins) g force of gravity or grams Gin glutamine Glu glutamic acid Gly glycine GTP guanosine triphosphate H histidine His histidine I isoleucine lie isoleucine K lysine . k b kilobase pairs kd kilodaltons L leucine Leu leucine Lys lysine M methionine or moles per litre mA milliamperes Met methionine mg milligram .. ml millilitre Mr relative molecular mass (a ratio therefore unitless) mRNA messenger ribonucleic acid |ig microgram ul microlitre N nucleotide (in nucleic acids) or asparagine (in proteins) NADH nicotinamide adenine dinucleotide (reduced) nanogram nmol nanomole nt nucleotide P proline Phe phenylalanine PIPES piperazine-N,N'-bis(2-ethanesulfonic acid) pmol picomole poly dldC copolymer of deoxyinosine and deoxycytidine Pro proline Q glutamine R purine (in nucleic acids) or arginine (in proteins) RNA ribonucleic acid RNase ribonuclease rRNA ribosomal ribonucleic acid S serine or Svedberg unit SDS sodium dodecyl sulfate Ser serine SOD superoxide dismutase SSC standard saline citrate T thymine (in nucleic acids) or threonine (in proteins) Thr threonine Tris tris(hydroxymethyl)aminomethane tRNA transfer ribonucleic acid Trp tryptophan Tyr tyrosine U units of enzyme or uracil V valine x i v Val valine W tryptophan X-gal 5-bromo-4-chloro-3-indolyl-PD-galactoside Y pyrimidine (in nucleic acids) or tyrosine (in proteins) 1 Introduction Superoxide Dismutase Molecular oxygen composes 20% of the earth's present atmosphere, yet some of its derivatives are toxic. By a quantum mechanical quirk the outer two orbitals of a molecule of dioxygen each contain an electron of the same spin. Because of this property molecular oxygen is usually reduced one electron at a time: O 2 + le" O 2 " (superoxide) O 2 ' + le" + 2H+ + H 2 O 2 H 2 O 2 + le" + 1H+ »• H 2 O + OH- (hydroxyl radical) OH- + le- + 1H+ • H2O The intermediate chemical species produced during this process are unstable and react nonspecifically with biomolecules such as nucleic acids (Demple and Linn 1982), proteins (Brot et al. 1981), and lipids (Mead 1976). One intermediate, superoxide, is produced chemically and enzymatically in cells exposed to O 2 (Fridovich 1978). It is not highly reactive itself, but it is able to react with hydrogen peroxide via a metal catalyst to form the highly unstable hydroxyl radical (Haber and Weiss 1934, Koppenol and Butler 1977). This is not a full explanation of its toxicity, however, since some damage by superoxide occurs independently of the presence of hydrogen peroxide (Fridovich 1986a). Any cell which lives in the presence of oxygen must protect itself against the toxic effects of oxygen radicals. There is no single factor that 2 accounts for the spectrum of oxygen tolerance found in living organisms, but several contributing conditions have been described: the absence of cell constituents that are especially sensitive to oxidation (for example thiol groups on proteins), the amount of reducing power available for oxygen scavenging, the rate and mechanism of oxygen consumption, and the presence of protective enzymes that convert toxic oxygen radicals to innocuous products (Morris 1979). One of the most important protective enzymes is superoxide dismutase (SOD) (Fridovich 1986b). The prosthetic metal atom in the enzyme transfers an electron from one superoxide anion to another, forming molecular oxygen and peroxide as products: + SOD 0 „ + 0„ + 2H • 0 „ + HO 2 2 2 2 2 Various catalases and peroxidases are responsible for converting peroxide to water and other harmless products. Although it was isolated in 1939 from erythrocytes, the function of SOD was not determined until 1969, by McCord and Fridovich (1969). The enzyme has since been isolated from a multitude of aerobes, facultative anaerobes, and even some anaerobes (Hewitt and Morris 1975, Steinman 1982). Three types of enzyme have been described: an Fe-containing enzyme found in bacteria and some chloroplasts, a Mn-containing enzyme found in bacteria, mitochondria, and chloroplasts, and a Cu and Zn-containing enzyme found in eucaryotic cytoplasm (Asada et al. 1980, for reviews see Steinman 1982 and Fridovich 1986b). The protein sequences of the Fe and Mn SODs are homologous while the Cu-Zn enzyme shows no homology to the others (Steinman 1980, Brock and Walker 1980). The Fe and Mn SODs have subunits of 20,000-25,000 mol. wt. that associate as dimers or 3 tetramers; the Cu-Zn SODs have subunits of 16,000 mol. wt. that associate as dimers. All three types catalyze the dismutation reaction near the diffusion limit (Klug et al. 1972, McAdam et al. 1977a and 1977b). In practice they have been distinguished from one another in crude extracts of cells by their differing sensitivities to inhibition by cyanide, hydrogen peroxide, and azide (Misra and Fridovich 1978, Weisinger and Fridovich 1973, Asada et al. 1975). Several lines of evidence suggest that SOD is important in oxygen tolerance. First, the various types of SOD have high amino acid sequence conservation: there is 55% identity between the yeast and bovine Cu-Zn SODs (Steinman 1980) and 58% identity between the Mn SODs of Escherichia coli and Bacillus stearothermophilus (Brock and Walker 1980). Such conservation is found in essential enzymes. Second, the SOD activities of rat lung (Crapo and Tierney 1974), alveolar macrophages (Simon et al. 1977), and several bacterial species (Gregory and Fridovich 1973a and 1973b, Puget and Michelson 1974, Privalle and Gregory 1979) are induced in response to increased concentrations of dioxygen. Third, reduction in the amount of Mn SOD in rats, mice, and chickens caused by a Mn-deficient diet results in a compensating increase in the amount of Cu-Zn SOD (deRosa et al. 1980), suggesting that the catalytic function of SOD, not its metal-binding capacity is important. Fourth, the Mn SOD of Escherichia coli is induced by exposure to increased dioxygen or generators of superoxide, for example paraquat or plumbagin (Hassan and Fridovich 1977, 1978). Exposure to low concentrations of superoxide generators leads to induction of resistance against higher concentrations (Hassan and Fridovich 1977). This induction is sensitive to inhibitors of protein synthesis and is probably the result of an increase in the production of SOD. 4 Mutants of E. coli and yeast that have no SOD (Carlioz and Touati 1986) or no mitochondrial SOD (van Loon et al. 1986), respectively, are more sensitive to killing by oxygen under certain growth conditions (lack of amino acids and fermentable carbon source) and are more sensitive to superoxide generators. The mutants of E_. coli also showed enhanced mutagenesis in the presence of oxygen (Farr et al. 1986). Mutants of Drosophila lacking Cu-Zn (cytoplasmic) SOD are infertile, more sensitive to paraquat, and have reduced longevity (Phillips et al. 1989). Recently, Imlay and Linn (1988) have found that molecular oxygen and SOD are able to prevent NADH from reducing hydrogen peroxide to the hydroxyl radical by scavenging unpaired electrons from the reaction pathway. Neither SOD without dioxygen nor denatured SOD were effective. Thus a scavenging system of oxygen radicals seems to be composed of SOD and dioxygen itself. Despite this evidence, the full physiological role of the enzyme is still being debated (Fee 1982). The finding of normal aerobic growth of the SOD-lacking mutants when they are supplied with amino acids and the presence of SOD in some anaerobes (Hewitt and Morris 1975) indicates that all of the functions of SOD may not be known and that SOD may play a somewhat variable role in oxygen tolerance. Early Evolution: three lineages As more data from geology and fossils have accumulated, a new idea of the origins of life on earth has emerged. The finding of stromatolites, fossilized communities of cyanobacteria, in sedimentary rocks of the Warrawoona group in Australia has pushed the date of the origin of life to before 3.5x10^  years ago, a time only about 1x10^  years after the formation of the planet (Walter 1983). The world inhabited by the first 5 organisms was probably anaerobic, rich in carbon dioxide, and reducing, but what were the biochemical characteristics of the organisms themselves? Fossils can provide only a limited amount of information on biomolecules and metabolic processes. A more recent approach to this problem is to extrapolate the molecular characteristics of extant organisms back to a common ancestor. In 1965 Zuckerkandl and Pauling proposed the use of molecules as chronometers. The ideal chronometer is a protein or nucleic acid sequence for which a multitude of functionally equivalent sequences exists. Changes in the sequence will reflect the intrinsic mutation rate rather than selective forces, and as these changes are fixed randomly over time they will serve to measure the passage of time. Some of the best molecules for this purpose are the rRNAs (Woese 1987). They are functionally constant, found in all organisms, composed of many independent domains, and are easily sequenced. Prior to 1977 it was thought that bacteria were a homogeneous group that composed the first life-forms and, after formation of symbiotic relationships, gave rise to the eucaryotes. An unexpected finding from the comparison of 16S rRNA sequences was that the first life forms apparently branched very early (before the appearance of cyanobacteria 3.5 xlO^ years ago) into three distinct lineages: the eubacteria, the urcaryotes (later to evolve into the eucaryotes), and the archaebacteria (Woese 1987). The eubacteria, which include the cyanobacteria, comprise most procaryotic organisms (fig. 1). The urcaryote is thought to have been an anaerobic ameboid organism that had the molecular characteristics of present eucaryotes. Bacterial endosymbionts eventually formed mitochondria and chloroplasts to give rise to the present eucaryotes (Margulis 1970). The 6 Figure 1. Evolution of the eucaryotes from a common progenote. Distances correspond sequences (Woese 1987). archaebacteria, eubacteria, and ancestor, the hypothetical to divergence of 16S rRNA ] extreme halophiles (incl. _H. cutirubrum) methanogens Thermococcus sulfur-dependent thermoacidophiles > o zr. > m co > Thermotoga > flavobacteria m cr C O 3> o —purple bacteria ( l n c l . 12. c o l l ) ~ gram-positive bacteria ( i n c l . JS. s"tearothermophilus) cyanobacteria green non-sulfur bacteria animals fungi lants c i l iates microspor idia o TO m co f l a g e l l a t e s 8 archaebacteria are a group of diverse procaryotes that is distinguished from the eubacteria by many unusual features (table 1). Later Evolution: the appearance of oxygen Geological evidence indicates that life originated in an anaerobic atmosphere. Non-biological processes of oxygen production such as photodissociation of water would not have produced enough free molecular oxygen to saturate the sinks of ferrous iron (Chapman and Schopf 1983). After the appearance of photosynthetic cyanbbacteria 3.5xlO^  years ago, the earth was to witness the greatest' biological change, and perhaps the first case of pollution, ever produced. Using water as a reductant of carbon dioxide, the cyanobacteria produced a toxic byproduct: molecular oxygen. Between 2 and 3x1 years ago the atmosphere became stably and globally aerobic (Chapman and Schopf 1983). Organisms were forced to either inhabit diminishing anaerobic environments or cope with the new gas. The fact that the protective enzyme SOD has originated twice during evolution indicates that it was probably not present in the common ancestor and originated at a later time, after the accumulation of atmospheric oxygen. The distribution of the enzyme is also significant. One type, the Fe and Mn SODs, is found in the eubacteria and their putative descendants, the mitochondria and chloroplasts (Steinman 1982). The other type, the Cu-Zn SODs, is found in the separate eucaryotic lineage. The relationship of the SODs of archaebacteria to the others was unknown at the beginning of the work described in this thesis (1984-1985). Archaebacteria Within the archaebacteria, the thermophilic, anaerobic phenotype is believed to be ancestral (Woese 1987). Subsequently, two major lineages 9 Table 1. The archaebacterial phenotype: some distinguishing characteristics of the eubacterial, archaebacterial and eucaryotic kingdoms are compared. The data were adapted from Jones et al. 1987 and references therein. The abreviations are kanamycin (Kan), anisomycin (Ani), and pseudouracil(xjr) Characteristics Archaebacteria Eubacteria Eucaryotes Cellular organization Genomes Lipids Ribosomes size of subunits antibiotic sensitivity elongation factors tRNA T y C loop 1-methyl adenine initiator tRNA initial amino acid aminoacyl-tRNA synthetases RNA Polymerase types in each cell subunits procaryotic 106bp branched chain alcohols linked to 2,3-sn-glycerol 30S, 50S K a n R A n i s aEF-2 contains diphthamide l-methyh]/\|/CG present 5'-triphosphate Met pretransfer proofreading 1 9-11 nucleus, organelles procaryotic 10 5-10 7bp 107-1011 bp straight chain —acids linked to— 1,3-sn-glycerol 30S. 50S KanS A n i R EF-G lacks diphthamide AOS, SOS K a n R A n i s EF-2 contains diphthamide ItyCG TtyCG absent present S'-monophosphate N-formyl Met Met posttransfer proofreading pretransfer (higher eucaryotes) and posttransfer (yeast) proofreading 1 5 a -amanitin rifampin mRNA 12 or more, plus transcription factors (eg. Spl) (Subunits of the archaebacterial and eucaryotic polymerases are immunologically cross-reactive.) insensitive insensitive Pol.II is sensitive insensitive sensitive insensitive uncapped uncapped 7-mG cap 10 diverged: the methanogenic-halophilic branch and the sulfur-dependent thermophilic branch (fig. 1). Major differences exist between the ribosomes (Cammarano et al. 1986) and RNA polymerases (Schnabel et al. 1983) of the two groups, substantiating the proposition that a deep division exists. The sulfur-dependent, thermophilic branch is exclusively thermophilic, with species thriving at up to 105° (Jones et al. 1987). Metabolic energy is generated either by oxidation of sulfur or by sulfur respiration of organics. They are isolated from hot springs and volcanic vents, including vents on the sea floor. The methanogenic-halophilic branch is composed of two distinct phenotypes: anaerobic methane producers and aerobic halophiles. The methanogens generate methane by using H2, formate, acetate, methanol, or methylamines to reduce CO2 (Jones et al. 1987). Some are halophilic, but none are aerobic. Halophilic Archaebacteria The halophiles have evolved from one group of methanogens to become aerobic organisms that obtain their metabolic energy from the respiration of carbohydrates and amino acids (table 2). In addition, the closely related species Halobacterium cutirubrum. H_. halobium. and H_. salinarium (now classified as strains of the single species H_- salinarium. Larsen 1984) are able to use the proton-pumping action of bacteriorhodopsin to generate ATP photosynthetically (Stoeckenius and Bogomolni 1982). Halobacteria inhabit salt brines of 1.5-5 M NaCl and grow at temperatures of 35° to 50° (table 2). Because the halobacterial cell contains 4M KC1 and IM NaCl (Christian and Waltho 1962), all enzymes must be adapted to this environment. Generally, halophilic proteins have fewer nonpolar and more acidic residues than their non-halophilic Table 2. Growth characteristics of extremely halophilic archaebacteria. The data were adapted from Jones et al. (1987) and references therein. Species morphology optimal temp. optimal PH optimal NaCl (M) Substrates Halobacterium salinarium (R. halobium. H. cutirubrum) rod 50© 7.2 3.5-4.3 amino acids photosynthesis H. volcanii disk, oval 450 6.8 1.8-2.5 carbohydrates H. saccharovorum rod 50O - 3.5-4.5 carbohydrates H. vallismortis (maris-mortui) pleomorph 40° 7.0 4.3 carbohydrates H. mediterranei rod 350 6.5 3 carbohydrates amino acids H. sodomense rod 4fJO - 1.7-2.5 carbohyrates Halococcus morrhuae coccus 30-370 7.2 3.5-4.5 amino acids Natronobacterium gregorvi rod 370 9.5 3 carbohydrates N. magadii rod 37-400 9.5 3.5 -N. pharaonis rod 450 8.5-9 3.5 organic acids N. occultus coccus 35-400 9.5 3.5-4 carbohydrates 12 counterparts (Lanyi 1974 and 1979, Eisenberg and Wachtel 1987). For example, the glutamate dehydrogenase of H_. marismortui has 20.2 mol% excess of acidic over basic residues compared to 6.4 mol% for the bovine liver enzyme (Leicht et al. 1978). Aspartate and glutamate residues bind 1.89 and 2.08 water molecules per side chain, respectively, compared with 0.32 to 1.22 water molecules bound by other amino acids (Saenger 1987). This property allows the acidic residues to compete with salt ions for structured water and maintain a hydration shell around the protein. The structure of water is less disrupted by K ions than Na ions because the hydration number of K + is half that of Na+; this may be the reason halobacteria selectively accumulate K + rather than Na+ in, their cytoplasm (Lanyi 1974). Salts such as KC1 and NaCl also act to "salt out" proteins (von Hippel and Schleich 1969). That is, they strengthen hydrophobic interactions within and between molecules. When nonpolar residues are exposed to water they cause an entropy decrease by increasing the organization (ie. ordered clusters) of water. Salting-out type salts decrease the structure of water by reducing the average size of clusters of water molecules. Exposure of nonpolar groups to the less structured water will cause a greater decrease of entropy (ie. increase in organization), therefore interaction of nonpolar residues with water is more strongly avoided in the presence of these salts (Lanyi 1974). The increased hydrophobic interactions mean that halophilic proteins require fewer nonpolar residues. These properties create problems when purifying such proteins. In the absence of high concentrations of salt the hydrophobic bonds are weakened and the acidic charges are no longer shielded. The repulsion of like charges in the molecule causes unfolding, loss of enzymatic activity, 13 and sometimes irreversible denaturation (Eisenberg and Wachtel 1987). For this reason, halophilic proteins must be purified in the presence of molar concentrations of salt. The genomes of halobacteria are about the same size as those of eubacteria (4x10*> bp) and many species contain a major fraction of GC-rich DNA and a minor fraction of AT-rich DNA (Pfeifer and Betlach 1985, Kushner 1985). In H_. halobium some of the AT-rich DNA is present on endogenous plasmids (Pfeifer et al. 1981, Kushner 1985). Covalently closed circular DNA has been isolated from H.. halobium and H. volcanii (Kushner 1985, Pfeifer 1988). The plasmid pHHl of H. halobium carries a gene (p-vac) encoding gas vacuole protein (Home et al. 1988), however the functions of other plasmids have not yet been determined. In fact, H_. volcanii could be cured of plasmid pHV2 with no noticeable phenotypic changes (Charlebois et al. 1987). The closely related species H_- halobium, H. cutirubrum and H. salinarium contain families of repetitive elements, some of which have been characterized as transposable elements (Sapienza and Doolittle 1982). They are clustered in the AT-rich fraction of genomic DNA that is composed of large plasmids (150 kb) and 70 kb islands on the chromosome (Pfeifer 1986). The transposable elements rearrange at high frequency, creating a very high mutation rate (10"4 to 10"2) for visible markers such as bacteriorhodopsin and gas vesicle production (DasSarma 1989). It is unknown if other, more essential genes are protected from inactivation by transposition. The previous lack of a transformation system has made functional analysis of halobacterial and other archaebacterial genetic elements difficult. Only about 20 protein-encoding genes have been cloned from 14 halobacteria and comparison of upstream sequences has yielded a sketchy idea of what may serve as a promoter (Gropp et al. 1989). From analysis of Sulfolobus phage genes and other archaebacterial genes, Reiter et al. (1988) have proposed that an AT-rich hexanucleotide (consensus: TTTAAA) about 25 nucleotides upstream of the initiation site is a universally conserved feature of archaebacterial promoters. Footprinting studies of methano-bacterial promoters have shown that RNA polymerase binds from -32 to +18 nucleotides relative to the start site, encompassing the putative promoter element (Thomm and Wich 1988). Some promoters, however, do not closely match the consensus. The recently developed transformation system for H_. volcanii should aid the identification of elements that are responsible for initiation and control of transcription (Cline et al. 1989). Superoxide Dismutase in Archaebacteria Aerobic representatives are found in both of the major subgroups of the archaebacteria: Sulfolobus spp. in the sulfur-dependent thermophilic branch, and Halobacterium spp., Halococcus spp., and Thermoplasma  acidophilum in the methanogenic-halophilic branch (Woese 1987, Jones et al. 1987). To date, SOD activity has been assayed and found in Sulfolobus  acidocaldarius (Searcy and Searcy 1981), H_. halobium (Searcy and Searcy 1981), T. acidophilum (Searcy and Searcy 1981) and Methanobacterium  bryantii (an anaerobe; Kirby et al. 1981). In addition, the SODs of the last two mentioned species have been isolated. They contain Fe as the prosthetic metal and have similar molecular weights (Mr 20000-25000) to eubacterial SODs, but no amino acid sequences were determined so the relationship of the archaebacterial enzymes to other SODs is unknown. 15 Aims of This Study The first aim of this study was to isolate the SOD from an archaebacterium, Halobacterium cutirubrum. and determine its relationship to SODs from the eubacteria and eucaryotes. A Halobacterium was chosen because the halophiles are aerobes that evolved from anaerobic archaebacteria, the methanogens. This represents a path to oxygen tolerance that is separate from eubacteria and eucaryotes. Also, isolation of the halobacterial enzyme allows investigation of its adaptation to high concentrations of salt. After isolating the enzyme, the corresponding gene was then cloned and used to study further the evolution and regulation of expression of SOD. A gene closely related to the SOD-encoding gene was incidentally discovered and found to be diverging from the SOD-encoding gene in an unusual way. The results of this study provide another perspective on the evolution of oxygen tolerance and contribute to the growing understanding of archaebacterial genetics. 1 6 M a t e r i a l s a n d M e t h o d s M a t e r i a l s Hydroxylapat i te was purchased from Calb iochem-Behr ing and Ultrogel A c A 4 4 was from L K B Instruments, Inc.; a l l other chromatography resins were f r o m Pharmacia , Inc. P y r o g a l l o l was purchased f r o m M a l l i n c k r o d t , Inc. B o v i n e C u - Z n S O D was purchased in lyophi l i zed form from Boeringer-M a n n h e i m , Inc. Acry lamide and bis-acrylamide were f rom Eastman Kodak, Inc. and agarose was from Sigma Chemical C o . and Schwartz M a n n Biotech. Components of bacterial culture media were purchased as fo l lows: yeast extract, tryptone, and casamino acids from D i f c o Laboratories; pure amino acids from Sigma Chemical C o . , Eastman K o d a k , and B D H ; D-glucose from B D H ; a m p i c i l l i n f r o m S i g m a C h e m i c a l C o . ; X - g a l (5-bromo-4-chloro-3-i n d o l y l -0D - g a l a c t o s i d e ) f r o m B R L . R a d i o n u c l i d e s were purchased f r o m A m e r s h a m and N e w E n g l a n d Nuclear . Deoxynucleot ide triphosphates and d i d e o x y n u c l e o t i d e triphosphates were purchased f r o m P h a r m a c i a , Inc. Restr ict ion enzymes and D N A modif icat ion enzymes were purchased from P h a r m a c i a , Inc . , B o e r i n g e r - M a n n h e i m , a n d B R L . M o d i f i e d T 7 D N A p o l y m e r a s e (Sequenase) and d i d e o x y n u c l e o t i d e m i x e s were purchased f r o m U n i t e d States B i o c h e m i c a l Corp. Paraquat was purchased from Sigma C h e m i c a l and B D H . H a l o b a c t e r i u m c u t i r u b r u m was obtained from Dr . A l Matheson of the University of V ic tor ia . C u l t u r e a n d S t r a i n s of B a c t e r i a For isolation of S O D or D N A , H . c u t i r u b r u m was grown in rich medium containing salts (231.3 g N a C l , 24.6 g M g S 0 4 - 7 H 2 0 , 2.2 g K C I , and 1.35 g N a 3 citrate per 1 adjusted to p H 7), 0 . 5 % yeast extract, and 1% casamino acids (based on Krantz and B a l l o u 1973). The salts, yeast extract, and casamino acids were made up separately, autoclaved and added together 17 after cooling. Cultures used for paraquat treatment were grown in a defined medium containing salts (as above), amino acids (215 mg L-alanine, 484 mg L-arginine-HCl, 65 mg L-cysteine-HCl, 225 mg L-aspartate, 60 mg glycine, 220 mg L-isoleucine, 800 mg L-leucine, 1.06 g L-lysine-HCl, 185 mg L-methionine, 130 mg L-phenylalanine, 50 mg L-proline, 300 mg L-serine, 250 mg L-threonine, 200 mg L-tyrosine, 500 mg L-valine per 1), 0.5 mM K phosphate pH 7, 1 mM CaCl2, trace elements (0.1 ml of a solution containing 230 mg FeCl2, 44 mg ZnS04-7H20, 30 mg MnSO^-rJ^O, and 5 mg CuS0 4-5H20 per 100 ml) and 0.1% glycerol (Bayley 1971). The salts and trace elements were made up separately and autoclaved. The amino acids were dissolved in 70 ml of water, the pH was raised to 11 with NaOH to aid dissolution, the volume was adjusted to 90 ml, and the solution was sterilized by filtration. The components were added together after cooling to room temperature and the pH was adjusted to 7. Paraquat was dissolved in water as a 1 M stock and added to cultures to a final concentration of 1 mM. For anaerobic growth the defined medium was supplemented with 0.5% arginine-HCl (pH 7.0), degassed, and purged with N 2 (Hartmann et al. 1980); culture flasks were capped tightly and sealed with several layers of parafilm. Escherichia coli was grown in rich medium (LB: 1% Bacto tryptone, •1 0.5% yeast extract, 1% NaCl; Maniatis et al. 1982) or in minimal medium (M9: 50 mM Na2HP04, 25 mM K H 2 P O 4 , 8.5 mM NaCl, 20 mM N H 4 C I , ImM MgS04, 0.1 mM CaCl, 0.2% glucose; Maniatis et al. 1982). Strain DH1 (F -, endAl, hsdR17 (rfc-» m ^ ) , supE44, thi-1, recAl, gyrA96) was used for library construction (Hanahan 1983). Strain JM83 (ara, A (lac proAB), rpsL, <j>80, lac ZAM15) was used for propagation of plasmids (Messing 1983). Strain JM101 (supE, thi, A(lac pro), thi, supE, [FtraD36, proAB, lacIQZ M15]) 18 was grown only in M9 supplemented with 1 mg/1 thiamine and 0.2% glucose and was used for propagation of Ml3 phage (Messing 1983). When needed for selection of plasmids, ampicillin (sodium salt) was added to a concentration of 100 mg/1. Agar plates contained 1.5% agar and agar overlays for phage contained 0.75% agar. For indication of p-galactosidase activity X-gal was added to 0.004% to cooled (50°) agar media. Assays of SOD and Protein Activity of SOD was measured by the method of Marklund and Marklund (1974). One unit of activity was defined as the amount of enzyme which inhibits the rate of autooxidation of 0.2 mM pyrogallol (measured at 420 nm) by 50% for 2 min in a volume of, 1 ml of buffer containing 50 mM Tris-HCl (pH 8.2) and 1 mM diethylenetriamine-pentaacetic acid (DETPA). In addition, assay buffers contained various concentrations of NaCl and KC1; 2 M KC1 gave optimal activity. Incorporation of 0.5 mM NaCN in assay buffers for crude extracts was necessary to give full sensitivity. Possibly this is due to contaminating respiratory components which produce superoxide. Protein concentrations were estimated by the Peterson modification of the method of Lowry et al (Peterson 1977). Bovine serum albumin (BSA) was used as the standard. Purification of SOD from H. cutirubrum Hydroxylapatite chromatography and gel filtration were performed at 20°. All other steps were performed at 4°, and all buffers contained 1 mM 2-mercaptoethanol. Cells from 4 1 of culture (mid-log phase, A46Q=1.0) were pelleted at 4,000xg for 20 min and then suspended in 40 ml of 3.0 M ammonium sulfate, 50 mM sodium phosphate (pH 7.0). The cells were lysed by five 2-min sonications (0.15 relative output on a Fisher model 19 150 dismembrator). The lysate was centrifuged at 44,000xg for 12 h. The supernatant was dialyzed against lysis buffer for 4 h and recentrifuged at 114,000xg for 2 h. The supernatant was adjusted to 2.5 M ammonium sulfate and adsorbed onto a column (60x2.6 cm) of Sepharose CL-4B that had been equilibrated with 2.25 M ammonium sulfate, 50 mM sodium phosphate (pH 7.0). The column was washed with 700 ml of buffer, and the SOD activity was then eluted with a 1.3 1 decreasing linear ammonium sufate gradient (2.25 M to 1.4 M in phosphate buffer). Fractions containing SOD activity were pooled and applied to a column (35x2.6 cm) of DEAE Sepharose CL-6B which had been equilibrated with 0.8 M ammonium sulfate, 50 mM bis(2-hydroxyethyl)imino-tris(hydroxymethyl)methane (bis-Tris) HC1 (pH 7.0). After the column was washed with 400 ml of buffer, SOD activity was eluted with an 800 ml increasing linear gradient of NaCl (0 to 0.8 M in bis-Tris HCl-ammonium sulfate buffer). Fractions containing SOD activity were pooled and dialyzed against 4 M NaCl, 50 mM Tris-HCl (pH 7.2). The dialyzed enzyme was loaded onto a column (7x1 cm) of hydroxylapatite which had been equilibrated with 4 M NaCl, 50 mM Tris-HCl (pH 7.2). The column was washed consecutively with 10 ml 4 M NaCl, 10 mM sodium phosphate (ph 7.0); 10 ml of 4 M NaCl, 250 mM sodium phosphate (pH 7.0); 10 ml 2 M NaCl, 10 mM sodium phosphate (pH 7.0). The SOD activity was eluted with 2 M NaCl, 300 mM sodium phosphate (pH 7.0). The SOD-containing eluate was dialyzed against 4 M NaCl, 50 mM Tris HC1 (pH 7.2), concentrated to 1.4 ml by dialysis against solid polyethylene glycol 8000, and then chromatographed on a column (95x1.6 cm) of Ultrogel AcA44 in 4 M NaCl, 50 mM Tris HC1 (pH 7.2). 20 Electrophoresis of Proteins Proteins were precipitated in 7% trichloroacetic acid, redissolved in loading buffer containing 2-mercaptoethanol, and electrophoresed on discontinuous sodium dodecyl sulfate-polyacrylamide gels as described (Laemmli 1970). Metal Analysis The metal content of 400 ng enzyme was determined using a Perkin-Elmer model 603 atomic absorption spectrophotometer equipped with an HGA 2200 graphite furnace. The sample was diluted 10-fold and triplicate assays (20 u.l/assay) were carried out for each metal. The wavelengths used were: Cu, 324.7 nm; Zn, 213.9 nm; Fe, 248.3 nm; Mn, 279.5 nm. Assays were performed by Stanya Horsky (U. British Columbia). The limits of detection were 1-10 parts per billion. Amino Acid Sequencing The N-terminal sequence of 30 nmol of purified SOD was determined by automated Edman degradation on a Beckman 190C protein sequencer. Sequencing was performed by Sandy Kielland (U. Victoria). Plasmids and Phages The plasmid pUC13 ( carrying the ampicillin resistance gene and a multiple cloning site in the gene of the a fragment of p-galactosidase) was used for library construction and all subclonings (Messing 1983). The single stranded phages M13mpl8 and M13mpl9 were used to prepare template DNA for sequencing (Messing 1983). Analysis of DNA Isolation of DNA. Total DNA (chromosomal and plasmid) was isolated from H_. cutirubrum as described (Schnabel 1982). Plasmid DNA was isolated from E_. coli by alkaline lysis (Maniatis et al. 1982). 21 Oligonucleotides were synthesized by Tom Atkinson on an Applied Biosystems 380B DNA synthesizer and purified by electrophoresis on polyacryamide sequencing gels (Maxam and Gilbert 1980). Molecular cloning. Molecular cloning procedures (restriction digestions, gel electrophoresis, electroelution of DNA fragments from gels, ligations, transformation of E. coli. colony hybridization, and end-labelling of DNA) were performed according to standard methods (Maniatis et al. 1982). Size-fractionated DNA that was used for library construction was never exposed to ethidium bromide. Southern blotting and probing. Blotting of DNA from agarose gels onto nitrocellulose was performed by the method of Southern (1975) in 20X SSC (IX SSC is 0.15 M NaCl, 0.015 M sodium citrate, pH 7.0). Radioactive probes of DNA fragments were prepared by the random priming method of Feinberg and Vogelstein (1982) except that 50 mM bis-Tris-HCl (pH 6.6) was used instead of Hepes and Tris. Probes were hybridized to blots in 1 ml/ 30 cm^  of 10X Denhardt's solution (IX Denhardt's is 0.2% BSA, 0.2% Ficoll, 0.2% polyvinyl pyrrolidone), 6X SSC, and 0.1% SDS at 37° (oligonucleotide probes) or 68° (longer probes). Blots that were probed with 20mer oligonucleotides were washed stringently in 0.2X SSC at 45° for 15 min. Southern blots probed with longer fragments of DNA were washed stringently in O.lxSSC at 68° for 1 hr. DNA sequencing. For sequencing of cloned DNA, overlapping sets of deletions were created by the method of Henikoff (1984, 1987) using exonuclease III. Cloned DNA was ligated to M13mpl8 or M13mpl9 phage vectors (Yanisch-Perron et al. 1985, Messing 1983), single-stranded template was prepared and sequenced using dideoxynucleotide chain termination with Klenow enzyme (Sanger 1977) or modified T7 D N A 22 polymerase (Tabor and Richardson 1987). Restriction fragments were isolated from polyacrylamide gels by electroelution (Maniatis et al. 1982) and sequenced by chemical degradation (Maxam and Gilbert 1980). The protein sequence deduced from the DNA sequences was compared with other protein sequences using the Fast P computer program (Lipman and Pearson 1985). Analysis of RNA Isolation of RNA. Total cellular RNA was isolated by lysing cell pellets in boiling lysis buffer ( 50 mM NaCl, 50 mM EDTA, 50 mM Na phosphate pH 7.0, 0.25% sodium dodecyl sulfate), extracting twice with phenol-chloroform, twice with chloroform, and precipitating twice with ethanol. The RNA was separated from DNA by precipitation with LiCl (Auffray and Rougeon 1980) or by centrifugation through a CsCl cushion (Chirgwin et al. 1979). Glassware used for RNA isolation was rinsed with concentrated HC1, rinsed with distilled water then ethanol, dried, and baked at 150° for 1 hr. Slot blotting and northern blotting. Samples of RNA for slot blotting were denatured in 2.5 M formaldehyde, 10X SSC at 65°, chilled, and applied to nitrocellulose. Blotting of RNA from formaldehyde-agarose gels onto nitrocellulose was performed as described (Maniatis et al. 1982). Slot blots and Northern blots probed with 20mer oligonucleotides were washed stringently as above. Northern blots probed with longer fragments of DNA were washed stringently in 0.2X SSC at 55° for 30 min. Primer extension. Primer extension of DNA along RNA templates was performed by the method of Newman (1987). One ng of 5' end-labelled primer (20mer oligonucleotide) was hybridized to 10 u,g RNA in 10 ul of 80 mM KC1, 20mM Tris-HCl pH 8.5, 0.5 mM EDTA by heating to 6 5 ° , 23 cooling slowly to 37°, and incubating at 37° for 1 hr. Ten [il of a mixture containing 10 mM MgCl2, 1 mM each dNTP, 10 mM 2-mercaptoethanol, 5 U RNase inhibitor, and 5 U AMV reverse transcriptase were added and the reaction was kept at 37° for 1 hr. The products were ethanol precipitated, redissolved in 5ul FDM (98% formamide, 10 mM EDTA pH 8.0, 2mg/ml each of xylene cyanol and bromophenol blue), boiled for 2 min, and analyzed on polyacrylamide sequencing gels (Maxam and Gilbert 1980). Nuclease SI analysis. Nuclease SI was used to analyze RNA-DNA hybrids (Dennis 1985). An end-labelled DNA probe fragment (10000 to 250000 cpm, 10^  to 10^  cpm/ pmol) was ethanol precipitated with 10 u,g RNA, redissolved in 20 ul hybridization buffer (40 mM PIPES pH 6.8, 400 mM NaCl, 1 mM EDTA, 80% v/v deionized formamide), heated to 80°, and transferred directly to the hybridization temperature (60° to 70°) for 3 hr. Ice cold nuclease SI buffer (300 u.1 containing 280 mM NaCl, 50 mM Na acetate pH 4.6, 4.5 mM ZnSC>4, 20 u,g/ml denatured DNA, and 60 U nuclease SI) was added to the hybrids and incubated at 37° for 30 min. Digested hybrids were isopropanol precipitated and analyzed on polyacrylamide sequencing gels as above. Analysis of RNA capped in vitro. To verify the presence of a triphosphate on the 5' end of an RNA, triphosphate ends of 50 u.g RNA were labelled with [a-32p] GTP using guanylyl transferase as described (Venkatesan and Moss 1981). The capped RNA was hybridized to 400 ng unlabelled DNA probe as above, digested with 60 U nuclease SI in 300 |il nuclease SI buffer as above, then 25 ng RNase A at 20° for 15 min, phenol-chloroform extracted, isopropanol precipitated, and analyzed on polyacrylamide sequencing gels as above. 24 Mobility Shift Assay The mobility shift assay was based on methods described by Hennighausen and Lubon (1987). Cells were grown to A 4 6 0 1-0 and harvested. The cell pellet was rinsed with basal salts (defined medium without amino acids or glycerol) and resuspended in one 70th volume of binding buffer (100 mM KC1, 20% glycerol, 1 mM 2-mercaptoethanol, 10 mM Tris-HCl pH 7.5, 5 mM MgCl2). The cells were lysed by sonication and centrifuged at 12000 g for 10 min. The supernatant was removed and assayed for protein. The extracts were then adjusted to 5 mg protein/ml with binding buffer and stored as frozen aliquots. The 360 bp Pstl-Aatll fragment (fig. 11) containing the 5' region of the gene was end-labelled and used as the probe. The binding mixtures contained IX binding buffer, 4 ul extract, and 0-10 u.g poly dl-dC (Pharmacia) in a total volume of 24 ul. They were incubated for 30 min. at 20°, then 10 ng (in 1 u,l) of probe were added to each mixture and incubated 30 min. at 20°. The mixtures were electrophoresed on a 5% polyacrylamide gel in 50 mM Tris-acetate pH 7.5, 1 mM EDTA at 30 mA. The gel was dried and autoradiographed. 25 Results I: Purification and Characterization of SOD from H. cutirubrum Results of the purification of SOD from H_. cutirubrum When halobacterial cells were lysed in 4M NaCl and assayed for SOD activity in 2M KC1 with 0.5 mM NaCN they were found to contain 4.2 U/mg of soluble protein. Crude extracts of aerobically grown Escherichia coli cells contained 26 U/mg, using the same assay system minus the KC1. The value for E_. coli agrees with other studies (Touati 1983). The low activity obtained for H_. cutirubrum can be explained by the low solubility of oxygen in its high salt environment creating less oxidative stress. Cells of H. cutirubrum contain a SOD activity which migrates as a single peak in a variety of chromatography assays. The single responsible protein has been purified to about 90% homogeneity by the procedure described in Materials and Methods. The results are summarized in Table 3 and Figure 4. The purification procedure utilizes techniques that allow high concentrations of salt to be maintained. Salting out chromatography on Sepharose CL-4B, an agarose resin, has been used by several investigators for fractionating halophilic proteins (Leicht and Pundak 1981, Mevarech et al. 1976, Pater and Pater 1977). The basis of the adsorption of proteins to the resin is not fully known; it has been attributed to hydrophobic interactions (Mevarech et al. 1976) and surface effects (von der Haar 1978). In the protocol for SOD purification, this step resulted in 6.3-fold purification and 99% recovery of activity. Fortunately, most of the other protein eluted as a large peak at a higher concentration of ammonium sulfate (fig. 2). This peak contained a large amount of dark brown material that was presumed to be ferredoxin. 2 6 Table 3. Purification of SOD from H . cutirubrum Fraction* Vol. Protein SOD Sp. act. Yield Purification (ml) (mg) (U)** (U/mg) (%) (-fold) (NH4)2S04 supernatant 53.0 132 1074 8.1 100 1.0 Salting out on Sepharose CL-4B 76.0 21 1064 5 1 99 6.3 DEAE-Sepharose 47.5 14 1045 7 5 9 7 9.2 Hydroxylapatite 1.4 4.3 946 220 8 8 27 Ultrogel AcA44 9.0 2.5 850 340 7 9 42 * Portions of the first two fractions were dialyzed against 4 M NaCl, 50 mM Tris-HCl (pH 7.2) before assay. ** Enzyme activity was measured in 0.5 M NaCl. 27 Figure 2. Salting-out chromatography on Sepharose CL-4B. Superoxide dismutase was eluted with a decreasing sulfate gradient (dashed line with solid circles), which was monitored by conductivity. Fractions of 7 ml were collected and assayed for SOD activity (solid line with solid triangles) and absorbance at 280 nm (dotted line). 2 8 100 FRACTION NUMBER 29 Ion exchange chromatography in the presence of ammonium sulfate was originated by Leicht and Pundak (1981). The sulfate does not elute ionically bound proteins but rather promotes binding, as evidenced by the higher chloride concentrations required to elute proteins bound in higher sulfate concentrations. This technique may actually be a combination of salting out chromatography and ion exchange chromatography. In the purification of SOD, 0.8 M ammonium sulfate was used because it approximated the ionic strength of 4 M NaCl and because it allowed elution at a moderate NaCl concentration (about 0.5 M; fig. 3). This step resulted in 1.5-fold purification with 98% recovery of activity. Affinity chromatography on hydroxylapatite in the presence of high concentrations of NaCl has previously been used in the purification of translation elongation factors from halobacteria (Kessel and Klink 1981). Sodium chloride rather than KC1 was chosen because of its greater solubility. Binding to the resin may be the result of charge configurations on the protein that interact with the Ca atoms on the hydroxylapatite crystal. Similarly, gel filtration in the presence of high NaCl concentrations has been used often in purification of halophilic proteins. In summary, the high yields of activity throughout the purification indicate that no substantial denaturation occurred after cell lysis. After purification, the enzyme could be stored in 4 M NaCl, 50mM Tris-HCl (pH 7.2) at 4° for over 1 month without loss of activity. The purified enzyme exhibited a specific activity of 340 U/mg in 0.5 M NaCl and 700 U/mg in 2 M KC1 (optimum salt concentration; see below). Crude extracts of cells lysed in 4 M NaCl then assayed in 2 M KC1 contained 4.2 U/mg of soluble protein. From these figures, SOD is calculated to be 0.6% of the soluble protein in a halobacterial cell. 30 Figure 3. Sulfate-mediated ion exchange chromatography on DEAE- Sepharose CL-6B. Superoxide dismutase was eluted with an increasing chloride gradient (dashed line with solid circles). Fractions of 5 ml were collected and assayed for SOD activity (solid line with solid triangles) and absorbance at 280 nm (dotted line). 31 0 50 100 FRACTION NUMBER 32 Like the enzyme from the archaebacterium T_. acidophilum. halobacterial SOD has only 10-20% of the specific activity normally associated with Fe or Mn SODs (Searcy and Searcy 1981). The methanobacterial enzyme, however, has normal specific activity (2060 U/mg; Kirby et al. 1981), showing that low specific activities are not common to all archaebacterial SODs. The presence of SODs with low specific activities in T. acidophilum and H_. cutirubrum may be due to the limited solubility of oxygen in their habitats placing less selective pressure on their oxygen defenses. Molecular Weight of Halobacterial SOD Electrophoresis on denaturing polyacrylamide gels shows enzyme subunits of Mr 25 000 (fig. 4). This value is the same as that found in the archaebacteria T. acidophilum and M_. bryantii (Searcy and Searcy 1981, Kirby et al. 1981). Analytical gel filtration in 4 M NaCl showed an oligomer of Mr 75 000 (fig. 5). The thermoplasma enzyme showed the same results when these techniques were used, however analytical ultracentrifugation indicated that the enzyme was a tetramer of Mr 82 000 rather than a trimer (Searcy and Searcy 1981). Only one trimeric SOD has been unequivocally demonstrated; all other SODs are dimeric or tetrameric (Steinman 1982). The halobacterial enzyme may also be a tetramer, but attempts to confirm this by crosslinking the subunits in high-salt buffer were unsuccessful. Response to Inhibitors The sensitivities of SODs to various inhibitors have been correlated with the prosthetic metals found in the enzymes. The purified enzyme from H_. cutirubrum was tested for sensitivity to various inhibitors and compared with the well characterized bovine Cu-Zn SOD. When assayed in 33 Figure 4. Purification of SOD as monitored by sodium dodecyl sulfate-discontinuous gel electrophoresis. Portions were removed at various stages of the purification procedure, precipitated in 7% trichloroacetic acid, and electrophoresed on a 3%/12% polyacrylamide gel. Lanes: 1, salting-out pool (10u.g); 2, DEAE-Sepharose pool (10u.g); 3, hydroxylapatite pool (5u.g); 4, Ultrogel AcA 44 pool (5u.g); 5, relative molecular mass markers. 34 1 2 3 4 5 M r — 200.0k — 97.4 k M » 68.0k 43.0k 25.7k 18.4k 14.3k 35 Figure 5. Analytical gel filtration of JH- cutirubrum SOD. The elution of SOD from a column (95 by 1.6 cm) of Ultrogel AcA 44 equilibrated with 4 M NaCl, 50 mM Tris HC1 (pH 7.2) was compared with the elution of RNase (Mr 13700), bovine serum albumin (BSA; Mr 67000), and H_. cutirubrum peroxidase-catalase (Mr 110000; Fukumori et al. 1985). 36 1.0 0 I • • ' — — 1 • -4 4.5 5 LOG MOL.WT. 37 Table 4. Sensitivity of halobacterial SOD and bovine copper-zinc SOD to inhibitors. Inhibitor Per cent of activity remaining'1 H. cutirubrum SOD Salt-free Bovine 2 M NaCl Cu-Zn SOD** 1 mM NaCN 10 mM NaN3 109 92 100 79 26 60 0.5 mM H2O2 1 min. preincubation 3 8 5 min. preincubation 3 3 12 7 78 53 * One unit of enzyme was assayed in the presence of each inhibitor. ** The specific activity of the bovine Cu-Zn SOD was 6200 U/mg, in good agreement with published results for the pyrogallol assay with Tris-HC1 buffer (Marklund and Marklund 1974) 38 salt-free buffer or buffer containing 2 M NaCl, the halobacterial SOD displayed an unusual pattern of responses to the inhibitors (table 4). These responses were seen with crude extracts as well as with purified protein and so do not appear to be artifacts of the purification procedure. Like Mn and Fe SODs (Weisinger and Fridovich 1973), the halobacterial enzyme was resistant to cyanide inhibition; however, unlike Fe and Mn SODs (Misra and Fridovich 1978), it was very resistant to azide inhibition. Both Fe and Cu-Zn SODs are considered to be sensitive to inactivation by hydrogen peroxide (Hodgson and Fridovich 1975). The halobacterial enzyme was also sensitive, but the rate of inactivation was much faster than the rate previously seen for Fe and Cu-Zn SODs. Therefore, the results of inhibitor studies do not clearly indicate the metal type of the halobacterial enzyme. Metal Content To identify the metal type, the purified enzyme was subjected to metal analysis by atomic absorption spectroscopy. Contributions by metals in the buffer were reduced by concentrating 400 of enzyme into 65 u,l with a Centricon 10 ultrafilter (Amicon Corp.). Buffer that flowed through the ultrafilter was used as a blank for analysis. The blank showed negligible amounts of Cu, Mn, and Fe, however Zn contamination was too high to allow reliable determination. Analysis of the enzyme showed 0.2 atoms of Mn, <0.03 atoms of Cu, and <0.001 atoms of Fe per subunit above the values in the blank. The enzyme therefore appears to be a Mn SOD. The appearance of the enzyme throughout the purification procedure is consistent with the finding of a small amount of Mn and little or no Cu or Fe. At no time did a pink (indicative of Mn), amber (indicative of Fe), or blue (indicative of Cu) colour comigrate with SOD activity. Concentrated 39 solutions of the enzyme up to 1.6 mg/ml appeared colourless and had no absorbance in the visible region. Although X-ray crystallography has demonstrated that each subunit of a Mn or Fe SOD has a metal binding site (Stallings et al. 1984), the Fe and Mn SODs that have been isolated from eubacteria characteristically have variable metal contents (Steinman 1982). For example, the purification of the Mn SOD of E. coH B yields 0.35 to 0.90 Mn atoms per subunit (Fee et al. 1976). Also, the two archaebacterial enzymes which have been isolated have less than one metal atom per subunit (the SOD of T. acidophilum has 0.4-0.5 Fe atoms and the SOD of M. bryantii has 0.7 Fe atoms) (Searcy and Searcy 1981, Kirby et al. 1981). Therefore there are three possible explanations for the low content of Mn: (i) cell lysis and the purification procedure result in metal loss; (ii) the enzyme in vivo contains little Mn because of low affinity for the metal; (iii) like the thermoplasma and methanobacterial enzymes, the halobacterial SOD protein has some subunits that contain catalytically inactive Zn. The small number of Mn atoms per subunit may explain the low specific activity of the halobacterial enzyme. Subsequent to publication of these results, Salin and Oesterhelt (1988) purified the SOD of the closely related Halobacterium halobium. In this bacterium they found a low level of SOD activity that was induced by oxygenation. The purified enzyme was dimeric, with a specific activity of 833 U/mg and 0.76 atoms of Mn, 0.5 atoms of Zn, and 0.77 atoms of Cu per subunit. It was concluded to be a Mn SOD. Salt Dependence of Activity The optimal activity of halobacterial SOD was determined with respect to NaCl and KCl concentration (fig. 6). Optimal activity occurs around 2 M 40 Figure 6. Salt dependence of SOD activity. Either 5 or 10 u.1 of purified enzyme was assayed in 50 mM Tris HC1, 1 mM DETPA (pH 8.2) containing various concentrations of KC1 (solid circles) or NaCl (solid squares). 41 2.6h 2.4h 0.2 -1100 90 80 H 70 > 60 ^ O 50 < n 40 I < 30 5 20 ^ 10 0 1 2 3 4 SALT CONCENTRATION (M) 42 salt, while at 4 M salt activity remains at about 75% of optimum. The highest specific activity achieved was 700 U/mg in 2 M KC1. At physiological concentrations of KC1, the enzyme had a specific activity of 560 U/mg. Higher activity was seen in KC1 than in NaCl, reflecting adaptation to intracellular conditions. It can be explained by the lower disruption of water structure by K + and, possibly, specific ion-binding sites on the protein (Eisenberg and Wachtel 1987). The enzymes of the citric acid cycle (Aitken and Brown 1969) and the enzymes involved in protein synthesis and ribosome structure (Lanyi 1979) also have higher activity in KC1 than in NaCl. While some halobacterial enzymes are rapidly denatured by exposure to conditions of low ionic strength (Eisenberg and Wachtel 1987), this was not the case with SOD; the enzyme lost little or no activity in the absence of salt for over 8 hr at room temperature. Amino Acid Sequence of the Amino-terminus Automated Edman degradation was used to determine the sequence of 51 of 56 amino acid residues at the amino-terminus; five residues are ambiguous (fig. 7). Over the first 36 residues, the H_. cutirubrum enzyme has 53% identity with the Mn SOD of Bacillus stearothermophilus. 59% identity with the Mn SOD of E. coli B, and 50% identity with the Fe SOD of E. coli B (Steinman 1982). The sequence of 26 residues at the amino-terminus of the fl_- halobium SOD was determined by Salin and Oesterhelt (1988) and comparison with the H_. cutirubrum enzyme reveals 4 differences: Leu at residue 16, Arg at residue 18, Leu at residue 25, and Gly at residue 26. These differences validate the distinction of H_. cutirubrum and H_- halobium strains in the H.. salinarium species. The homology between the eubacterial and halobacterial enzymes at the amino-terminus may indicate that a 43 Figure 7. Amino acid sequence of the amino-terminus of SOD from ¥L. cutirubrum. Ambiguous residues are designated X. Residues which are invariably found in iron and manganese SODs (Steinman 1982) are emphasized. 44 1 Thr Glu Tyr Glu Leu Asp Tyr Asp Ala Leu Glu Gin Val Leu Thr His His Gin Gly X Asp Ala Glu Glu Thr Asn Thr Gly Asp X 10 Pro Pro Leu Pro Tyr Glu Pro His lie Ser Trp His His Asp Thr Val Asn Gly Trp Asn Leu XX Asn X Ala 45 comon ancestral enzyme had arisen before the two groups diverged, however final conclusions about homology must await the availability of a complete amino acid sequence. For this reason the gene encoding SOD was cloned and sequenced. 46 Results II: Cloning and Characterization of the SOD-encoding Gene (sod) and a sod-like Gene (slg) from H_. cutirubrum Cloning of sod The Mn SOD isolated from H_. cutirubrum has subunits of Mr 25000, corresponding to protein monomers of about 200 amino acid residues (May and Dennis 1987). A mixture of oligonucleotides (20mers) complementary to all possible coding sequences for residues 27-33 was synthesized (fig. 8). The probe mixture was 5' end-labelled and used to probe H_. cutirubrum genomic DNA that had been digested with Mbol (fig. 8). The recognition sequence of Mbol is GATC, the same as that for Sau3AI. Three fragments, 350 bp, 600 bp, and 1.1 kb, hybridized to the probe. Fractions of these size classes of Mbol-digested genomic DNA were ligated into the BamHI site of plasmid pUC13 (fig. 8). Recombinant plasmids were used to transform E. coli DH1 and positive clones were identified by colony hybridization using the oligonucleotide probe mixture. Recombinant clones of the three fragments were obtained and each was sequenced. The 1.1 kb clone contains the complete sod gene, the 350 bp fragment is a subfragment of the 1.1 kb fragment, and the 600 bp clone has a sequence that fortuitously matches the 20mer probe. A 571 bp Sau3AI-Pstl subfragment of the 600 bp fragment hybridized to the probe and contained a sequence matching the probe at 17 of 20 positions. The surrounding DNA, however, could not encode the SOD protein (fig. 9). Characterization of sod Overlapping deletions of the 1.1 kb fragment were constructed and both strands were sequenced (fig. 10). The fragment is 1127 bp long and it contains an open reading frame (designated sod) that encodes a protein of 47 Figure 8. (A) Sequence of a 20mer oligonucleotide probe mixture based on all possible codons for amino acid residues 27-33. (B) Construction of size fractionated libraries of genomic Mbol fragments from H. cutirubrum. Genomic DNA (100 u.g) was digested with 200U Mbol and electrophoresed on a 3.5% polyacrylamide gel. Slices of 5 mm were cut out of the gel and electroeluted. One tenth of the DNA in each fraction was electrophoresed on the agarose gel shown. After staining with ethidium bromide and photographing, the DNA from the gel was blotted onto nitrocellulose and probed with the oligonucleotide probe mixture. Fractions which hybridized to the probe (lanes 3,6 and 8) were ligated into the BamHI site of pUC13 and used to transform E. coli DHL 48 Amino Acids Trp His His Asp Thr His His mFNA 5' UGG CAU CAU GAU ACU CAU CAU 3' c c c c c c A 20irer Probe Mix 3' ACC GTA GTA CTA TGA GTA GT 5' G G G G G T C FRACTION NO. ( M r A ^ i n i D N c o o i w o FRACTION NO. -IOOOBP ,- 600BP - 300BP SOUTHERN BLOT / SOD OLIGONUCLEOTIDE PROBE 49 Figure 9. The nucleotide sequence of the 571 bp subfragment of the 600 bp Sau3AI fragment. The binding of the probe to the sequence is boxed. 50 Sau37AI | | | | | G A T C C G T A T C C A A ^ 60 I I I I I CCOGCCAOIXXXmCCGCTGTTO^ 120 I I I I I CAGOSCGTOGGCACGCGCGGOGTGTQ^  180 I I I I I A O T G C C C G C G G C X A G A A A ^ ^ ^ 240 AOCGTGGTGCTGTGAGTGGT TCGTGTCGGCCTCGTTGTACGC OSGCAGCAOGACACTCAOCT X A O ^ G O C G T C T C G C T C A 300 I I I I I I TTGOCAGCACTCTOGGCTAQa^ 360 G G G A C G C G T C C T T O 420 I I I I I T G ^ X C O G C A G A A T O X A ^ 480 I I I I I G T C G A C A C A O O G T C ^ ^ 540 | | Ps t I GAAACAGCAOOGCATOT^^ 51 Figure 10. Sequencing strategy of the 1.1 kb (1127 bp) Sau3AI fragment. The DNA was sequenced by using three methods: chain termination with klenow enzyme (k), chain termination with modified T7 DNA polymerase (Sequenase; s), and chemical degradation (c). The coding region is shown by the black rectangle, the arrow above the coding region indicates the direction of translation, and the bar above the fragment represents 100 bp. The Sau3AI (Mbol) sites are designated S and the PstI (P) and EcoRI (E) sites of the vector are shown to indicate the orientation of the clone. CO C O 53 Figure 11. Nucleotide sequence of the 1127 bp Sau3AI fragment containing the sod gene. The binding site of the 20mer probe mix is indicated above the sequence. The deduced amino acid sequence of the SOD protein is shown below the nucleotide sequence. The calculated molecular weight of the deduced amino acid sequence is 25770 daltons. The three His residues and one Asp residue that are putatively responsible for binding the Mn atom are underlined and important restriction sites are overlined. 54 10 20 30 40 50 I I I I I 1 GATCGQGCX-nT^  61 GAAAAATACITGTAAAOGCTOG^ I I I E s a l I 121 CTGAAACHXXATTCCG^  MetSer I I BS2l I I I 181 GAATJCGAACTXXX^ CCGCTG^  GluTyrGluLeLuProPro]^ 20mer probe 3' ACCGTPGm(r[mGNGTPGT 5' 241 CAGGTGCTCAOGTGGCATX^GGACACGCACCAra GlnValLeuThrTrpiii^HisAspThrrLisr^ I I I I I 301 GCCGAGGAGACACTCGCGGAGAACCGTGA^  MaGluGluThrLeuMaGluAsnArgGluThrGlyAspHisMaSerTh Aatll | | | | 361 CTCGGGGACGTCACGCACAACGGC LeuGlyAspValThrHisAsnGlySerGly^ I I I I I 421 ATCAGCXX£GCXmDGGa]AQ3^ MetSerProMaGlyGlyAspGluProSerGlyMaLeuMaAspArglleAlaM I I I I I 481 TTOGGCTOCTACGAGAACIGGGGGGCa PheGlySerTyrGluAspTrpArc^aGliTPheGluMaM I I I I I 541 TGGGOGCTXXnXI^TCTAa^ TrpMaLeuLeuValTyrAsp^ I I I I I 601 CAOGACGAGGGOGCGCTCTGGGGCAGCCAOO HisAspGluGlyMaLeuTrpGlySerHisProIleLeijMaLe I Avail | | | 661 TOCTACmCEACmC^ SerTyrTyrTyrAspTyrGlyProAsr^ I I I I I 721 GTXXACTGGGACGAGOOCACO^ ValAspTrpAspGluProThrGluArgPheGluGln^ I I I I I 781 CCCGCCCGCGGGGACACCCTGAMCX^ OCACG^  841 TCTATACACAGCOCAGCCAACACQ^ ^ 901 GATACCAGCTXI^ TCGCGGAGGGCTGG^  961 TGATGGTGCGGGCGCrG^  1021 GQGAOGACAGCACa^ CCACCGTCACGGAGOG^ 1081 GAGACAOGGACGOOGAOGCCACGCTGCTGACG^  55 200 amino acids (fig. 11). The 350 bp fragment that hybridized to the probe was found to be a 363 bp subfragment of the 1.1 kb fragment that extended from position 1 to position 363 of the 1.1 kb sequence (fig. 11). At the latter end of the 363 bp clone the MboI/BamHI junction contained GCTC rather than the expected GATG. Therefore, the 363 bp fragment seems to have been generated by aberrent Mbol cutting at the internal GCTC at position 360-363 of the 1.1 kb fragment. It was not reproducibly found in subsequent Southern blots of genomic DNA (see for example the Sau3AI digestion in fig. 13). The first 56 amino acid residues encoded by sod correspond almost perfectly with the sequence obtained from the purified SOD protein (May and Dennis 1987). The gene specifies Ser rather than Thr and Glu rather than Asn at residues 2 and 52, respectively. These discrepancies are apparently due to errors in protein sequencing since both the 363 bp and 1127 bp Mbol (Sau3AI) fragments gave identical results. In addition, the mature SOD enzyme lacks the Met residue at the amino-terminus. The deduced amino acid sequence substantiates previous indications that the H_. cutirubrum enzyme is of the Mn/Fe type: (i) it contains no Cys, (ii) it is relatively abundant in Trp and Tyr, (iii) it has 39.5% identity with the MnSOD of E. coli. 38.3% identity with human mitochondrial MnSOD (Barra et al. 1984), and 42.4 % amino acid identity with the Mn SOD of Bacillus stearothermophilus. a eubacterium (see fig. 2) (Steinman 1982). For comparison, the identity between the B_. stearothermophilus and E. coli enzymes is 57.9%. When the halobacterial enzyme is aligned with the MnSOD of B_. stearothermophilus. the residues responsible for binding the Mn atom are conserved: His26, His81, Aspl63, Hisl67 of the bacillus 56 enzyme (Stallings et al. 1984) and putatively His29, His76, Aspl58, Hisl62 of the halobacterial enzyme (fig. 12). The predicted Mn SOD protein of H_. cutirubrum contains 18 Asp residues, 21 Glu residues, 0 Lys residues, and 7 Arg residues to give a 5.5-fold excess of acidic over basic residues. The experimentally determined pi of the H_. halobium enzyme was 4.98 (Salin and Oesterhelt 1988), comparable to the pi's of other Mn SODs (Fridovich 1986b). Estimation of Copies of the sod Gene and Detection of the sod-Iike Gene (slg) To estimate the number of copies of the sod gene on the halobacterial chromosome the 1127 bp Sau3AI fragment was labelled with 32p a n c j hybridized to genomic DNA that had been cut with various restriction enzymes (fig. 13). Each lane contained two hybridizing fragments: an intense and a less intense band. In the BamHI digestion the fragments were 9.0 kb and 2.1 kb, respectively; in the PstI digestion they were 3.0 kb and 1.8 kb, respectively; in the SstI digestion they were 6.7 kb and 3.6 kb, respectively. Both fragments were cloned from PstI digested DNA and characterized. The intensely hybridizing fragment (3.0 kb) contains the 1127 bp Sau3AI fragment that carries sod while the poorly hybridizing fragment (1.8 kb) does not contain this fragment. The genome of H_. cutirubrum therefore appears to contain a single copy of the sod gene and a single copy of a closely related sequence. The cross-hybridizing region of the 1.8 kb PstI fragment was localized to a position around the internal EcoRI site. Both strands of overlapping subfragments from this region were sequenced (fig. 14). 57 Figure 12. Alignment of the deduced H. cutirubrum amino acid sequence with that of the gram-positive eubacterium Bacillus strearothermophilus (see fig. 2 for evolutionary position). The residues that bind the Mn atom are indicated by arrows. 58 30 fl.. £12. S E Y E L P J P L P Y D Y D A L E P H I S E Q V L T ' W H H D T I I I I I I I I I I I I I I I I £. SL- P F E L P A L P Y P Y D A L E P H I D K E T M N I H H T K H H Q G Y V N G W N D A E E T L A E N R E T G D H A S T I I I I I I I I I H H N T Y V T N L N A A L E G H P D L Q N K S L E E L L S N A G A L G D V T H N G S G H I L H T L F W Q S M S I I I I I I I I I I I I L E A L P E S I R T A V R N N G G G H A N H S L F W T I L S P A G G D E P S G A . L A D R I A A D F G S Y E N W R A E F E I I I I I I I I I I I I I I I P N G G G E P T G E L A D A I N K K F G S F T A F K D E F S A A A S A A S G W A L L V Y D S H S N T L R N V A V D N I I I I I I I I I I K A A A G R F G S G W A W L V V N N G E L E I T S T P N H D E G A L W G S H P I L A L D V W E H S Y Y Y D Y G P D R I I I I I I I I ' I I I I I I I Q D S P I M E G K T P I L G L D V W E H A Y Y L K Y Q N R R A A G S F V D A F F E V V D W D I I I I I I P E Y I A A F W N V V N W D 59 Figure 13. Southern blot analysis of H. cutirubrum DNA. Genomic DNA (1 u.g/lane) was digested with various restriction enzymes and probed with the 1127 bp Sau3AI fragment containing the sod gene. The positions of the 3.0 kb and 1.8 kb PstI fragments are indicated by arrows. bo Ln cn cr i i i B a m H I PstI S a u 3 A I I # SstI 61 Figure 14. Sequencing strategy of the slg gene found on the 1.8 kb PstI fragment. The sequences of both strands of overlapping restriction fragments were determined using chain termination with klenow enzyme and modified T7 DNA polymerase (Sequenase). The restriction sites are: A, Alul; E, EcoRI; P, PstI; R, Rsal; S, Sau3AI; SI, Sail. The direction of translation of slg is shown by the arrow above the coding region (black rectangle). The bar above the figure represents 100 bp. CX CO" CO-LI-< 63 Characterization of the slg. Gene Around the EcoRI site there is an open reading frame with remarkable similarity to the sod gene (fig. 15). The ORF (designated sod-like gene, slg) has 87% nucleotide identity with sod and would encode a protein with 83% amino acid identity with the Mn SOD. That slg. does not encode the purified Mn SOD is shown by the 10 mismatches between the predicted slg. protein and the purified SOD over the amino-terminal 56 amino acid residues. The mismatches include the same two (Ser2 and Glu52) that are found between the predicted sod protein and the purified enzyme. Comparison of slg with sod yields some unusual results. The similarity of the two sequences ends abruptly at the borders of the coding region; in the 5' and 3' non-coding regions there is no apparent homology (fig.15) except for conserved, putative promoter elements (see Results III). Within the coding region, only 7 of 35 amino acid changes are conservative and there are two areas of low nucleotide identity that result in extended regions of non-identical amino acids. The first is between positions 488 and 502 where only 5 of 15 nucleotides are conserved; in the proteins the five amino acid residues encoded by these nucleotides are completely different. The second, somewhat longer region is at the C-terminus between positions 660 and 695 where only 15 of 35 nucleotides are conserved and 12 consecutive amino acid residues in the two proteins are different. Nucleotide differences within the coding regions are randomly distributed among first, second, and third positions of codons (26, 21, and 31 differences, respectively). Transversions outnumber transitions by 47 to 30. To distinguish between the sod and slg genes and their transcripts, two single-sequence, 20mer oligonucleotides were synthesized (fig. 15). 64 Figure 15. Comparison of the nucleotide (slg and sod) and amino acid sequences of the slg and sod genes. Identical residues are indicated by dashes. The EcoRI site of slg is at position 528-533. The binding sites of the sig.-specific 20mer (5'-TCGGCGCTGTTGAGGCCGTC-3') and the sod-specific 20mer (5'-TGGAGGATGTGCCCCGAGCC-3') are indicated by underlining of the appropriate sequence. The amino acid residues that putatively bind the Mn atom are boxed. The sites of transcription initiation (arrows) and termination (open circles) are also indicated. (See Results III for a description of their determination.) 65 s o d CfllCCCOCCIICIICCICCCTCHCITCCflflGrcCfltCICCIfllCflJCCfiftCBTflCftlCflFGflRflflfllftCIIGlflflflCCCICCCCfllCflCrJCCCCCCITCCCCICCCCTCflCCflflCCCfla TGTCCGflGCGCCGCCGflCflCGCCCCGGflGTflTCCGGflCCCCflTCGTGGfiCCflCRGCCflGCGCCGCGflGGflCGCGflTCGCGflTGITCGflGCGTGCGCGCGGCGflCGflGTGflGCCCflCGRfC slg 20 10 60 60 100 120 • H n S E Y E L P P L . P V O V O R L E P H I S E CTGnflRCTGCRTTCCGGRRnCCflCCRTflRGCRGCGCCGRCGIRCGflCflCflCTGTRTGTCCGflRTnCGRflCTCCCflCCGCIGCCGTflCGflCTflCGflCGCGCtCGRRCCRCflCniCRGCGnG CTCflCGflflCRIIIflflCfllGRCGCCGCGTGflTCflCIGflTCCGGTGGflllCCflCCGfllGflGCCRGCflCGflflCTCCCfllCGCIGCCGIflCGRCIRCGflCGCflCICGflflCCRCRCfllCflGTGflG •L. - - 0 H S - -110 160 ^  180 200 220 210 30 10 SO 60 0 U L T U (JS] H O T H H Q G Y U N G U H O R E E T l f l E M f l E T G O H f l S T f l G f l CflGGIGCTCRCGTGGCflTCflCGflCflCCCflCCRTCflGGGCIRCGIGRflCGGCIGGRRCGflCGCCGflGGRGflCRCTCGCGGRGRflCCGTGRGflCCGGCGflCCflCGCCICGflCflGCCGGCGCG CRGGTGGTCflCGTGGCflCCRCGRCflCCCRCCflCCflGflGCTflCGTGGRCGGCCTCflRCflGCGCCGRGGflGRCGCTGGCGGRGflRCCGTGflGflCCGGCGflCCRCGCTTCGflCTGCCGGCGCG - - U - . - Q - - - _ - - S - - 0 - l . - S - - - - - - - - - - - - - - -260 260 300 320 310 360 70 60 90 100 L G O U T H M G S G H I L [5] T Q F U Q S n S P R G G O E P S G R L R O R I R R O CTCGGGGflCGrCflCGCRCRflCGGCTCGGGGCRCRTCCTCCflCflCGCTGTTCTGGCRGTCCRTGRGCCCGGCGGGCGGCGflCGRGCCGTCCGGGGCGCTCGCCGRCCGCflTCGCGGCGGRC CTCGGGGflCGTCRCGCflCflflCGGCTGTGGGCflCTflCCTCCRCRCGRTGTTCTGGGflGCRCRTGflGTCCCGRCGGGGGCGGCGflGCCGTCCGGGGCGCTCGCCGRCCGCflTCGCGGCGGRC - - - - - - - - c - - Y - { _ ] - n - - E H - - - 0 - - G - - - - - - - - - - - - -380 100 120 110 160 160 110 120 130 110 F G S Y E H U R f l E F E R R R S f l R S G U R L L U Y O S H S H T - R H U A U O M TTCGGCTCCTRCGflGRflCTGGCGGGCCGRGTTCGRGGCCGCCGCCRGCGCGGCCflGCGGCTGGGCGCTGCTCGTCTRCGRCTCCCRCflGCRRCRCGCTCCGGRRCGTGGCCGTGGRCnflC • • • • • • •« ««« • • •«• T r C G G C T C C I R C G f l G f l f l C I G C C G G G C T G f l f l l l C G f l G G I G G C G G C C G G C G C G G C C R G C G G C I G G G C G C I G C r C G I C T f l C G f i l C C G G T C G C C f t f t G C f l G C r C C G G f l f l C G I G G C C G T G G f l C f l f l C _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ G - - - - - - - - - - - P U f l _ Q - - - - - - - -500 520 S10 S60 S80 600 ISO 160 170 160 H O E G R L U G S H P 1 L H L [ O ] O U E ( H ] S Y V Y O Y G P O R G S F U O R F F E V CRCGRCGRGGGCGCGCTCTGGGGCRGCCRCCCCRTCCTCGCGCTCGRCGTCTGGGRGCflCTCCTRCTflCTflCGRCTRCGGTCCCGRCCGCGGCflGCTTCGTCGRCGCCTTCTTCGflGGTC . • • CflCGflCGfiGGGCGCGCTCIGGGGCflGCCflCCCCfllCCICGCCCICGRCGTCTGGGRGCRCTCCTflCTflCTflCGflCIflCGGCCCCGflCCGCGGCflGCIICGICGflCGCCTICTICGRGGIG • • 620 610 660 660 700 720 190 200 U O U O E P T E R F E Q A R E R F E OOOOO GTCGRCTGGGRCGRGCCCRCCGRGCGCTTCGflGCRGGCGGCCGRGCGCITCGRGTflflCGCCCCGCCCGCGGGGRCRCCCIGRflRCGCCRCGCIITTTCGCCGTGTRGCGTTCCGRTRGCT flTCGRCTGGGflCCCCflTCGCGGCGflflCTRCGRCGflCGTGGTGTCSCTGTTCGRGTGRCCGRflCflCGCTCCCGTGTTTTTTTCGCGTCRTGGCGGCICflTCRGTGGCTGGCGTGRGTGCGG I - - - P I R R H V O O U U S L - - ooooo 710 760 780 800 620 610 ICTRIflCflCHGCCCRGCCflflCRCCGGTGIGIGGICflCGICtCCCGTGGICGflC «««« «• •«« •« ««««« « « CGRCCGTGIGCICGICflCCGflCCGCGIGCflCGGIGICGGIGflGCGCGICGflRR 660 860 66 The sod-specific 20mer (sequence: 5'-TGGAGGATGTGCCCCGAGCC-3') is complementary to codons 70 to 77 of the sod gene. It matches slg at 16 of 20 positions. The slg-specific 20mer (sequence: 5'-TCGGCGCTGTTGAGGCCGT C-3') is complementary" to codons 38 to 44 of slg. It matches the sod gene at 15 of 20 positions. 67 Results III: Analysis of Expression of the sod and fiJLg. Genes Response of £L- cutirubrum to Paraquat Paraquat (methyl viologen) is an intracellular generator of superoxide and is known to induce SOD in eubacteria (Hassan and Fridovich 1977, 1978). When paraquat was added to aerobically growing H_. cutirubrum to a final concentration of 1 mM, the doubling time of the culture increased from about 7.5 hr to about 11.4 hr. After 27 hr of exposure to the drug, the specific activity of SOD in the culture had increased 3.6-fold (fig. 16a). The enzyme induction was accompanied by an increase in sod mRNA. The sod-specific 20mer oligonucleotide (fig. 11, fig. 15) was 5' end-labelled and used to probe RNA isolated from cells that had been exposed to paraquat for various lengths of time. (The sod-specific 20mer matches slg at 16 of 20 positions and does not hybridize to the cloned slg under the stringent conditions employed in probing RNA; data not shown.) The fraction of total RNA that was sod mRNA increased 5-fold after 27 hr exposure to the drug (fig. 16b). This increase must be due to changes in the rate of synthesis and/or degradation. The length of the sod mRNA was estimated by northern hybridization to be about 680 nucleotides; the length was not altered during paraquat treatment (fig 17). Anaerobically grown cultures contained 2.6 U SOD/mg protein (vs. 4.2-4.9 U/mg in aerobic cultures) and about 90% as much sod mRNA as aerobically grown cultures (fig. 16b). These results indicate that the levels of SOD activity and sod mRNA are responsive to oxygen radicals and that transcription of the sod gene is not fully repressed in the absence of oxygen. 68 Figure 16. Induction of halobacterial SOD activity and sod mRNA in response to paraquat. Samples of a culture of H. cutirubrum were removed at various times after the addition of paraquat (time 0). (A) Cells were lysed by sonication and the SOD activity and protein were measured. (B) Cells were lysed by boiling in detergent, RNA was isolated, bound to nitrocellulose (0.5 u.g/slot) and probed with the end-labelled sod.-specific 20mer. In addition sod mRNA was quantitated in (a) aerobically grown and (b) anaerobically grown H. cutirubrum cells; (c) yeast RNA served as a negative control. 70 Figure 17. Northern blot analysis of RNA from untreated (lane 1) and paraquat-treated (20 hr treatment; lane 2) cells. Total RNA (10 u.g/lane) was run on a 1% agarose gel containing formaldehyde, blotted onto nitrocellulose, and probed with the end-labelled s_p_d-specific 20mer. The positions of RNA size markers are indicated at the left. From top to bottom they are: 1.77 kb, 1.52 kb, 1.28 kb, 0.78 kb, 0.53 kb, 0.40 kb, 0.28 kb. 71 1 2 72 Mapping of the 5' Ends of the sod Gene Transcripts The strategy and results of sod transcript mapping are shown in figure 18. The 1127 bp Sau3AI fragment of DNA that contains the sod gene also contains 174 nucleotide's of 5' untranslated DNA and 350 nucleotides of 3' untranslated DNA. The 5' ends of in vivo sod transcripts were mapped using primer extension and nuclease SI protection. The sod-specific 20mer is complementary to the coding strand of DNA at codons 70-77 (positions 382-401 in fig. 13 and fig. 15) and was used to prime reverse transcription of RNA isolated from untreated and paraquat-treated cells (fig. 19). The major product was 230 nucleotides in length, with a less intense band 229 nucleotides in length. This indicates that for uninduced and paraquat-induced transcripts the major 5' end is only 3 nucleotides upstream of the ATG initiation codon (positions 172-173 in fig. 11 and fig. 15; fig. 23). This result was confirmed by using nuclease SI analysis (fig. 19). A DNA fragment extending from the Aatll site (positions 367-372 in fig. 11) through the 5' untranslated region to the PstI site of the vector was used as the probe. Fragments that were protected by RNA from nuclease SI were 193 to 199 nucleotides in length; this corresponds to a 5' transcript end centered 3 nucleotides upstream of the ATG initiation codon. Furthermore, a 5' triphosphate end on the transcript was verified by nuclease protection analysis of capped RNA. The 5' triphospate ends of RNA from paraquat-treated cells were labelled with [a-32p] GTP by guanylyl transferase and the RNA was hybridized to the AatH-Pstl DNA fragment. The labelled RNA that was protected by the DNA fragment from nuclease SI and RNase A was about 200 nucleotides in length (fig. 19). This corresponds to a triphosphate end about 3 nucleotides upstream of the initiation codon. The fact that the transcript has a triphosphate end and the 73 Figure 18. Strategy and results of mapping the sod transcript. The black rectangle represents the translated region and the dashed line represents the transcript, with a 5' triphosphate end (open circle) and 3' OH end (arrow) indicated. (A) Primer extension from the sod-specific 20mer primer (open box) yielded a 230 nucleotide product. (B) Nuclease SI digestion of the 384 nucleotide probe (Pstl-AatH fragment; top line) left a 197 nucleotide product (bottom line). (C) A capped 200 nucleotide RNA fragment (dashed line) was protected by the Aatll-PstI fragment. (D) Nuclease SI digestion of the 462 nucleotide probe (Avall-EcoRI fragment; top line) left mostly a 133 nucleotide product (bottom line). O DO > yPstl H-Sau3AI O I I I ? [ • Aatll a Avail > —Sau3AI ^ EcoRI <7_ 75 Figure 19. Actual results of the mapping of the 5' and 3' ends of the sod transcript. (A) The sod-specific 20mer was 5' end-labelled, hybridized to 10 u.g RNA from uninduced (lanel) or paraquat-induced (lane 2) cells, and used to prime DNA synthesis to the 5' end of the mRNA. The sequencing ladder was generated by using the same oligonucleotide to prime DNA synthesis from an Ml3 clone. (B) The Aatll-PstI fragment was 5' end-labelled, hybridized to 10 pg RNA from uninduced (lane 1) or paraquat-induced (lane 2) cells and digested with nuclease SI. Lane M contains DNA markers generated by v, partial digestion of the end-labelled Aatll-PstI fragment with Rsal. (C) The 5' triphosphate ends of 50 pg of RNA from paraquat-induced cells were labelled with [cc-32P] QTP by guanylyl transferase. The RNA was hybridized to the AatH-Pstl fragment and digested with nuclease SI and RNase A (lane 1). Control reactions contained no probe (lane 2) and no nuclease SI or RNase A (lane 3). Lane M contains DNA markers as in (B). (D) The Avall-EcoRI fragment was 3' end-labelled, hybridized to 10 pg RNA from uninduced (lane 1) or paraquat-induced (lane 2) cells, and digested with nuclease SI. The sequencing ladder was made by chemically degrading the Avall-EcoRI fragment. The nucleotide sequences around the transcription initiation and termination sites are indicated in (A) and (D), respectively. For (A), control reactions containing no RNA gave no extension products (not shown). For (B) and (D), samples of undigested probe gave single bands of the expected size (not shown). 76 77 fact that no longer products were seen in either nuclease SI analysis or primer extension demonstrates that initiation almost certainly occurs at or near the third nucleotide upstream of the initiation codon (position 172 in fig. 11 and fig. 15; fig. 23). Mapping of the- 3* Ends of the sod Gene Transcripts The 3' ends of sod transcripts were mapped by hybridizing RNA from untreated and paraquat-treated cells to a 3' end-labelled fragment of DNA extending from the Avail site (positions 679-683 in fig. 11) through the 3' untranslated region to the EcoRI site of the vector (fig. 18). After digestion with nuclease SI, about 90% of the remaining fragments were 131 to 135 nucleotides in length (fig. 19). This length corresponds to 3' ends within the sequence of five T residues (positions 813-817 in fig. 11 and fig. 15; fig. 23) that is 36 to 40 bp downstream of the TAA termination codon. Other fragments, corresponding to less abundant 3' ends, were identified about 26, 54, and 64 nucleotides downstream of the termination codon. Evident in all analyses is the increased abundance of sod mRNA in paraquat-treated cells. The same nuclease SI analysis procedure was used to detect the 3' ends of sod mRNA in RNA from anaerobically grown cells (fig. 20), confirming that sod is transcribed in the absence of oxygen, though the amount of mRNA is reduced. Mapping and Quantitation of the slg. Gene Transcripts The strategy and results of slg. transcript mapping are shown in figure 21. As with the sod gene, the transcripts of slg were mapped using primer extension and nuclease SI protection. The sJLg.-specific 20mer is complementary to codons 38 to 44 (positions 286-305 in fig. 15) and was used to prime reverse transcription of RNA from untreated and paraquat-treated cells. The single product was 144 nucleotides long and corresponds 78 Figure 20. Detection of the 3' ends of s_o_d mRNA in RNA from anaerobic cells. The end-labelled Avall-EcoRI fragment was hybridized to lOpg RNA from anaerobically grown cells (lane 2) or aerobically grown cells (lane 3) and digested with nuclease SI. Yeast RNA (lane 1) served as the negative control. The undigested probe is in lane 4. The markers (lane M) are Mspl fragments of pBR322. 79 80 Figure 21. Strategy and results of mapping the slg transcript. (A) Nuclease SI digestion of the 0.8 kb Pstl-EcoRI probe (top line) left a 345 nucleotide product (bottom line). (B) Primer extension from the sl_g.-specific 20mer (open square) yielded a 144 nucleotide product. (C) Nuclease SI digestion of the 1.0 kb EcoRI-PstI probe (top line) left a 283-292 nucleotide product (bottom line). 82 to a 5' end that is 13 nucleotides upstream of the initiation codon (fig. 22; position 162 in fig. 15; fig. 23). More importantly, no increase of slg. mRNA was seen in response to paraquat. For mapping the 5' ends of the transcripts by nuclease SI analysis, a DNA fragment extending from the EcoRI site (positions 508-513 in fig. 15) through the 5' untranslated region to the PstI site at the end of the clone was labelled at the 5' ends and used as the probe (fig. 21). Fragments protected by RNA from nuclease SI were about 345 nucleotides long, corresponding to a 5' transcript end that is about 11 nucleotides upstream of the initiation codon (fig. 22, fig. 23). This agrees, within the margin of error, with the results of primer extension. The 3' ends of the transcripts were mapped by using a 3' end-labelled DNA fragment extending from the EcoRI site through the 3' untranslated region to the PstI site at the end of the clone (fig. 21). Most of the fragments protected from nuclease SI were 283 to 292 nucleotides long (fig. 22). This corresponds to 3' ends of transcripts in the tract of seven T residues that is 18 to 24 nucleotides downstream of the TGA termination codon (fig. 23). In primer extension and in nuclease SI analysis of the 3' end of the transcripts, no increase in slg. mRNA was observed in response to paraquat. Comparison of Promoters and Terminators of the sod and slg Genes The s_Ig. 5' flanking region contains two consensus sequences that are found in the halobacterial 16S rRNA promoters (fig. 23). There is a TTCGA element 53 nucleotides upstream and a TTAA element 25 nucleotides upstream of the transcription initiation site. Also, there is an element (CACGA) resembling TTCGA 35 nucleotides upstream of the transcription 83 Figure 22. Actual results of mapping the slg transcript. (A) The 0.8 kb Pstl-EcoRI fragment was 5' end-labelled, hybridized to 10 pg RNA from yeast (lane Y) or H. cutirubrum (lane H) and digested with nuclease SI. Lane P contains undigested probe and lane M contains DNA markers generated by digestion of pUC13 with Sau3AI. (B) The sig.-specific 20mer primer was 5' end-labelled, hybridized to 10 pg RNA from untreated (lane 1) or paraquat-treated (lane 2) cells, and used to prime DNA synthesis to the 5' end of the mRNA. The sequencing ladder was generated by using the same oligonucleotide to prime DNA synthesis from an Ml3 clone of slg. Control reactions containing no RNA gave no extension products (not shown) (C) The 1.0 kb EcoRI-PstI fragment was 3' end-labelled, hybridized to 10 pg RNA from untreated (lane 1) or paraquat-treated (lane 2) cells, and digested with nuclease SI. The DNA markers (lane M) are Sau3AI fragments of pUC13. The undigested probe gave a single band of the expected size (not shown) 84 85 Figure 23. Comparison of 5' (A) and 3' (B) flanking regions of the sod and slg genes. The sites of transcription initiation and termination are shown above and below the sequences. Putative promoter elements are boxed and inverted repeats are indicated by arrows over the sequences. 86 A sod GATOTXBOGTTCTTO^  slg GTCOGAGOGCCGCGGACACGCOOOGGAGTA^  50 CATACATCATCAAAAATACT^  ACAGCCAGCGCOGCGAGGACGGGATXI5CGATGITQGA 30GTGCGOGOGGC 100 (XX^TCXXXSTCAOGAACDZA^ OGACXGAGTGAGCOCAOGA^  150 CAGOGOCGAOGTADGACACACTGT ATG. T£ACnXATm3GTX-GAnxrAra ATG.. sod TAA CXXCXXXGCXXXGCGGGGACACCCT slg TGA COGAACAC 87 Figure 24. Alignment of the sod and slg. promoters with promoters of other genes of halobacteria. The underlined nucleotides indicate the sites of transcription initiation. The genes are bop, bacterio-opsin (DasSarma et al. 1984); hop, halo-opsin (Blanck and Oesterhelt 1987); brp. bop-related protein (Betlach et al. 1984); p-vac plasmid copy of gas vacuole gene (Home et al. 1988): c-vac. chromosomal copy of gas vacuole gene (Home et al. 1988); Llle, ribosomal protein equivalent to L l l of E . coli (Shimmin and Dennis 1989); Lie, ribosomal protein equivalent to LI of E. coli (Shimmin and Dennis 1989); T ISH1.8, 1.7 kb transcript of insertion element ISH1.8 (Gropp et al. 1989); Tl-8, transcripts of phage <j>Hl (Gropp et al. 1989); RNA Pol., RNA polymerase subunit genes (Leffers et al. 1989); SI2, ribosomal protein S12 (Gropp et al. 1989); Glycopr., glycoprotein of cell surface (Lechner and Sumper 1987). 88 Gene (species) 5 ' Sequence Stable RNA (various species) b_qp (halobium) llQP (halobium)* brp (halobium) p-vac (halobium) c-vac (halobium)* Llle (cutirubrum) Lie (cutirubrum) TISHl. 8 (halobium) Tl, 2,3 (phage 4»H1) T4 (phage <J>H1) (phage <{>H1) (phage <|>H1) (phage <J>H1) (phage <iHl) RNA Pol (halobium) S12 (halobium) Glycopr (halobium) SQ& (cutirubrum) Slg (cutirubrum) consensus T5 T6 T7 T8 T T C CITAANTA GA GTTACACA AG GTTATTTA TT TTTTGATG TC CTTATGTG CG GTTTTCCG GG GTTAAACC CT TTTAAGCC GA GTTATCTG AT TTTATTAT GA TATAAGTT GT CTTOCOGA AC GTTATGAT GA TTTATCTT GA TTTAATAG GG CTTAAATG GG CTTAAGTG CA TTTACCAG AC CATAAGCA CA TTTAACAT TTTAACTG G 20 to 25 bp start site CATATCCTCGTTAGGT^ ^ ATG ATGGCGTGCCGTGTCCTTCCGAACAC ATG CTCX3GTAGTGACGTGTGTGTAT^  ATG ATGCCCGAGTATAGTTAGAGAT GACACTCCCTG^ GTTGCGGGTG-19nt-ATG aSOGGCGGOGGTTTCTOGGAGT ATG CGGGATCACDaTCTG^ AATTGGGTGTCTCGTATCTGCT^  ACTGGGGTTXX^ CGGACATGACAGAGCAG^  AGAOOCCTXICTAAAGT^  (XXXTCGGAAOGAGGAGGCCD3^ AAGATAT GGGCCAAAAACCTCTTTTAGGT^  TGCACGCATGGAAGTCCAC^ ^ GTAGGGQCCTCrAATGTTT^  CIGTGGGCAGCAACIGGCC^ ^ CX^ XX^ GOGGGATACTCGGCTGTATGA^  TGGOOGGGTATAGTCTGGAGC^  GCGOGGAa^OGACAOVCTT^ ATG GAOQOCGCGTGATCACTGATCXIGG-*Start site is unknown, the sequence has been aligned with a closely related sequence whose start site is known. 89 i n i t i a t i o n site. T h e T T A A element is thought to be e s p e c i a l l y w e l l conserved i n archaebacterial promoters (fig. 24). W h e n the 5' region of the s o d gene is examined, these elements are surprisingly not w e l l conserved (f ig. 23). The closest match for the highly conserved element is A T A A that is 24-25 nucleotides upstream of the transcription init iat ion site. There is a T T C G A sequence, but it is over 140 nucleotides upstream of the init iat ion site. Another sequence ( C C G G A ) somewhat resembling T T C G A is located 34-35 nucleotides upstream of the transcription in i t ia t ion site. These unusual results show that the sequences w h i c h promote and correct ly pos i t ion transcription init iat ion have yet to be ful ly elucidated. T e r m i n a t i o n of transcript ion i n halobacteria appears to be more straightforward. Transcripts terminate i n tracts o f T residues that may or may not be preceded by inverted repeats and G C - r i c h sequences (f ig. 23; S h i m m i n and Dennis 1989). The sod gene terminator contains a l l of these elements whereas the slg. terminator is lack ing the G C - r i c h region. Such a terminator structure is reminiscent of rho-independent terminators found i n eubacteria (Holmes et a l . 1983). 90 Discussion and Conclusion Evolution of Mn/Fe-containing Superoxide Dismutase The conservation of" amino acid sequences of Mn SODs in gram positive eubacteria (B_. stearothermophilus). gram negative eubacteria (E_. coli), mitochondria, and halobacteria indicates that this enzyme serves an important function in the survival of the cells. The Mn SODs of the eubacteria E. coli and B. stearothermophilus are more closely related to each other (57.9% identity; Brock and Walker 1980) than to the halobacterial enzyme (39.5% and 42.4% identity, respectively), but clearly all three are homologous. Caution must be exercised in inferring more detailed relationships since Cu-Zn SOD has been found to evolve at very different rates in different organisms (Lee et al. 1985). Based on physical characteristics of the halobacterial enzyme it appears to be homologous to other archaebacterial SODs, but no sequences are available to verify this. The eucaryotes contain a Cu-Zn SOD that is not related to the eubacterial or halobacterial enzymes (Asada et al. 1980, Steinman 1982). When life originated over 3.5x10$ years ago, the atmosphere was anaerobic (Chapman and Schopf 1983). It became globally aerobic about 2x10$ years ago, more than 1.5x10$ years after photosynthetic cyanobacteria had begun generating oxygen. Comparisons of rRNA sequences indicate that cyanobacteria probably appeared after the eubacteria and archaebacteria had diverged (fig. 1). Within this context, there are several ways to explain the fact that halobacteria, and perhaps other archaebacteria have a SOD that is related to the eubacterial enzyme. First, the ancestral Mn SOD gene may have originated prior to the archaebacterial-eubacterial divergence (believed to have occurred before 91 3.5x10$ years ago and before the appearance of cyanobacteria) and would consequently have been passed to both lines of descent. Environments supporting early biological activity might have contained local, fluctuating amounts of oxygen and selected for an early appearance of SOD. Alternatively, the product of the ancestral SOD gene may have performed some other important function prior to the appearance of oxygen. Fee (1982) has previously speculated that SOD may have an important though unknown function in addition to the dismutation of superoxide. Second, the divergence of the halophilic archaebacteria from the eubacteria may have occurred much more recently, after the accumulation of oxygen in the atmosphere and the appearance of the Mn SOD gene in the common ancestor. The phylogeny of Lake (1988) would support this idea. Based on his own analysis of 16S rRNA sequences he has proposed that the halophilic-methanospirillum branch is more closely related to the eubacteria than to the archaebacteria. Third, the Mn SOD gene may have originated in eubacteria after the appearance of atmospheric oxygen, and then been laterally transferred to the archaebacteria (or vice versa). Wide dissemination of the gene encoding Mn SOD is possible because of its positive selective value. The somewhat lower amino acid sequence similarity of the halobacterial enzyme when compared to eubacterial enzymes may be explained by adaptations required for activity in the high concentrations of salt in the halobacterial cytoplasm (see below). Precedents for this type of transfer exist: the gas vacuole and ferredoxin genes of cyanobacteria are believed to have been transferred to halobacteria (Home et al. 1988, Hase et al. 1977) and a Cu-Zn SOD is believed to have been transferred from a eucaryote to a eubacterium 92 (Bannister and Parker 1985). In the past, halobacteria may have obtained from cyanobacteria many genes that encode products necessary for aerobic life. An examination of the evolutionary origins of halobacterial cytochromes could prove instructive. Adaptation of Halobacterial SOD to High Salt Concentrations Table 5 shows the amino acid composition of halobacterial SOD compared with non-halophilic SODs. The halobacterial enzyme has a notable increase in acidic residues and decrease in basic residues, especially lysine. The nonpolar residues have only a slight reduction. The excess of acidic over basic residues is 15 mol%, which is close to the overall value, 17.1 mol%, obtained for total cytoplasmic protein of H_. salinarium (Reistad 1970). Therefore, the halobacterial SOD is similar enough to non-halophilic SODs for homology to be detected, while it shows changes that are characteristic of adaptation to high salt concentrations. Such changes may skew the measurement of the time of divergence of the halobacterial enzyme. Expression of the sod and slg Genes In the eubacterium E. coli there are two SOD-encoding genes: sod A which encodes the MnSOD (Touati 1983), and sodB which encodes the FeSOD (Sakamoto and Touati 1984). Both are subject to complex regulation. The sodB gene is expressed in both anaerobic and aerobic growth conditions (Hassan and Fridovich 1977) and its expression is increased in response to iron in the growth medium (Simons et al. 1976). Expression of sodA seems to be coupled to the respiratory chain and to the redox state of the cell; it is expressed when cells are respiring with oxygen, nitrate, tetramethylamine oxide, dimethyl sulfoxide, or ferricyanide as electron acceptors (Smith and Neidhardt 1983, Miyake 1986, Hassan and Moody 93 Table 5. Amino acid compositions of MnSOD from H. cutirubrum (H.cu. sod), the slg. product from H_. cutirubrum (H.cu. slg), the MnSOD from B_. stearothermophilus (B.st. MnSOD), and the MnSOD from E. coli (E.co. MnSOD). H.cu. sod H.cu. slg B.st MnSOD E.co. MnSOD Asp 18 20 Glu 20 16 Arg 7 5 Lys 0 1 His 12 14 Ala 24 23 lie 5 5 Leu 15 15 Met 1 2* Phe 8 1 Pro 9 9 Trp 8 7 Val 10 14 Asn 8 7 Cys 0 1 Gin 4 4 Gly 16 17 Ser 15 14 Thr 9 7 Tyr 10 1 1 199 199 Acidic 38 (19%) 36 Basic 7( 4%) 6 Non-polar 80 (40%) 82 8 1 1 18 13 6 6 1 2 16 9 8 20 26 9 "6 19 21 2 2 8 11 13 9 6 6 8 10 17 12 0 0 3 6 15 13 11 11 1 1 11 8 7 203 205 (18%) 26 (13%) 24 (12%) (3%) 18 ( 9%) 22 (11%) (41%) 85 (42%) 91 (44%) * Assuming that the N-terminal Met is removed. 94 1987, Schiavone and Hassan 1988, Privalle and Fridovich 1988). It is also inducible by hyperbaric oxygen, redox compounds such as paraquat, heat shock, and elevated manganese concentrations (Gregory and Fridovich 1973a, Hassan and Fridovich 1977, Privalle and Fridovich 1987; Pugh et al., 1984). Touati (1988) has recently used operon and protein fusions to show that regulation of sod A is accomplished by transcriptional and postranscriptional control. Her results with transcriptional control in the presence of ferrous ions are consistent with a putative Fe-containing repressor whose oxidation state determines expression of the gene (Moody and Hassan 1984). The responsive element in the promoter and the actual trans-acting factor have not yet been found. In contrast to the E. coli genes, the halobacterial MnSOD-encoding gene (sod) has a different pattern of expression. Expression of sod during anaerobic, fermentative growth in arginine demonstrates that it is not strictly coupled to the respiratory chain. It is also inducible by the oxidative stress created by paraquat and the induction seems to be entirely due to increased levels of mRNA. The same transcription initiation site is used during uninduced and induced expression so that one promoter appears to be responsible for both types of expression. Analysis of expression during heat shock and changing manganese concentrations may show further regulatory controls. Alternatively, the increased level of mRNA may not be due to activity at the promoter, but rather to changes in the stability of the mRNA. Unfortunately, H_. cutirubrum transports RNA precursors so slowly that pulse-chase experiments are not feasible. Unlike the sod gene, slg. is not induced in response to paraquat. This may be used to argue that it does not encode a SOD activity, but it must be 95 remembered that the sodB gene of E . coli is not induced by paraquat either (Hassan and Fridovich 1977). The extremely high similarity of the sod and slg coding regions and the finding that the slg. mRNA rather than the sod mRNA has the extra leader provide convincing evidence that regulatory elements that are responsible for inducibility of sod mRNA must reside either in the 3' trailer of the mRNA or in the flanking DNA, probably the 5' region. The transcript of slg. has a leader that is 11 to 12 nucleotides longer than the leader of the sod transcript (fig. 23). At the 5' end of the slg transcript there is a purine-rich sequence GTGGA that is complementary at 4 of 5 positions to the 3' end of the 16S rRNA and may provide a recognition signal for. the initiation ^ codon. Similarly, the genes for the ribosomal proteins Lie and LlOe of H_. cutirubrum (transcribed on a single polycistronic mRNA) are both preceded by sequences complementary to the 3' end of the 16S rRNA (Shimmin and Dennis 1989). The s o l gene transcript does not contain such a sequence upstream or downstream of the initiation codon, indicating that halobacterial ribosomes may be able to use either of two mechanisms for initiating translation. One mechanism would involve a Shine-Dalgarno (eubacterial) type of interaction (Shine and Dalgarno 1974) between the 5' end of the mRNA and the 3' end of the 16S rRNA; the other would involve a more eucaryotic system wherein the ribosome simply scans from the 5' end of the mRNA and initiates at the first AUG codon it encounters (Kozak 1978). Transcripts of the sod gene of H_. cutirubrum and the bop (bacterio-opsin; DasSarma et al. 1984) and brp (bop-related protein; Betlach et al. 1984) genes of H_. halobium initiate very close to the initiation codon (fig. 24). However, unlike the transcripts of the other two genes, the sod gene 9 6 transcript has no hairpin loop at the 5' end and no putative Shine-Dalgarno element downstream of the initiation codon. There is no apparent sequence similarity between the upstream regions of sod and the other two genes. The bop and sod genes both contain a sequence of alternating purines and pyrimidines immediately upstream of the initiation site; in the high intracellular salt concentrations these sequences may assume a Z conformation and play a role in gene regulation. (Bacterio-opsin expression is reduced in response to oxygen; Oesterhelt and Stoekenius 1973.) The Enigma of the sod-like Gene (slg) Salin and Oesterhelt (1988) have recently purified the SOD enzyme from H.. halobium (believed to be the same species as H_. cutirubrum) and determined the sequence of 26 amino acid residues at the N-terminus. They used a 17mer oligonucleotide probe based on the hexapeptide Pro-Tyr-Asp-Tyr-Asp-Ala (corresponding to nucleotide positions 129-145 in Figure 17) to clone a genomic fragment that appears to be closely related to the s_lg. gene described here (Salin et al. 1988). Over 26 N-terminal residues the deduced amino acid sequence of the cloned H . halobium gene and the actual sequence of the H . halobium SOD differ at 8 positions. The 5' and 3' flanking regions of the H_. halobium gene are virtually identical to the H_. cutirubrum slg flanking sequences and the coding sequences differ only by (i) a reading frameshift between positions 364 and 462 and (ii) a substitution at position 622-623 to generate an A C G Thr codon in place of the AGC Ser codon reported here. In H_. halobium the amino acid sequence predicted for the region between the frameshifts has anomalously low identity with the B_. stearothermophilus and H_. cutirubrum SODs (9% and 3%, respectively). The reading frame for the H_. cutirubrum slg is shifted one nucleotide in this region and has high amino acid identity with the B_. 97 stearothermophilus and H. cutirubrum SODs (46% and 91%, respectively). The substitution at position 622-623 creates an AGC Ser codon that overlaps a predicted Alul site (recognition sequence AGCT); cutting at this site has confirmed the presence of the AGC Ser codon. Salin et al. (1988) believed the gene that they cloned from H. halobium encoded the Mn SOD that they had previously purified from this organism. However, the number of mismatches between the deduced and actual protein sequences (Salin and Oesterhelt 1988, Salin et al. 1988) indicates that the gene is not the authentic sod gene of H_. halobium. Rather it appears to be the homolog of the slg. of H_. cutirubrum. In each species of archaebacteria (H. cutirubrum. H_. halobium. T. acidophilum. and M. bryantii) examined to date, only one SOD enzyme has been found. Yet, H_. cutirubrum and probably H_. halobium contain an additional sod-like gene that does not encode the single SOD that was isolated from each by independent research groups. (The genome of H_. volcanii also has two elements that hybridize to the sod gene of H. cutirubrum: P. Joshi unpublished) What does the slg. gene encode? One possibility is that it encodes a minor SOD activity that was missed in crude extracts. But the fact that slg mRNA is approximately as abundant as sod mRNA would argue that any SOD activity possessed by the slg product should be easily detectable. In addition, the slg product contains a cysteine residue at position 71, whereas Mn and Fe SODs show a notable lack of cysteine (Steinman 1982). Analysis of amino acid sequence results of the purified H_. cutirubrum Mn SOD indicated that if the s lg product copurified, it composed only 1.1% or less of the purified protein sample. 9 8 Assuming sod is ancestral to slg. another possibility is that the sig. has diverged sufficiently from the sod gene to have acquired a different function. Conservation of the Mn binding residues suggests that the protein would contain a prosthetic metal atom and may function as an oxidoreductase of a different specificity. Or it may simply function as a metal storage protein. Examination of the structure of a homologous Fe SOD shows that the cluster of amino acid residues at the C-terminus that differ between the sod and slg products is located along a side of the pocket occupied by the Mn atom (fig.25; Stallings et al. 1984). How Mn SOD interacts with its substrate is not presently known. Two other pairs of evolutionarily related proteins from H_. halobium have been analyzed (Table 6). The closely related gas vacuole genes (c-vac and p-vac) show changes typical of proteins with a conserved function: most changes occur at the third position of codons and transitions outnumber transversions (Horne et al. 1988). The more distantly related bacterio-opsin and halo-opsin genes (bop and hop) have diverged enough to encode proteins with different functions (Blanck and Oesterhelt 1987). Both bacterio-opsin and halo-opsin are transmembrane proteins that covalently bind retinal at a conserved lysine residue, but the former pumps protons inward (Oesterhelt and Stoekenius 1971) and the latter pumps chloride ions outward (Schobert and Lanyi 1982). Changes between the hop and bop genes are distinctive: differences are evenly distributed among codon positions and transversions outnumber transitions (Table 6). The sod-slg gene pair from H_. cutirubrum has higher nucleotide identity than the vac gene pair yet they have changes that are similar to the bop-hop pair (Table 6). 99 Figure 25. Conformation of an Fe SOD; the Fe and Mn SODs have homologous structures (Stallings et al. 1984). The amino (N) and carboxy (C) termini are indicated. 100 Table 6. Comparison of pairs of related genes from H_. halobium and H_. cutirubrum. The genes are: c-vac and p-vac. gas vacuole genes from H.. halobium (Home et al. 1988); bop and hop, bacterio-opsin and halo-opsin genes from H.. halobium (Blanck and Oesterhelt 1987); and sod and slg. superoxide dismutase and superoxide dismutase-like genes from H. cutirubrum. Changes at Nucleotide Amino acid Transitions/ codon positions: Protein Gene pair identity identity transversions 1st 2nd 3rd Function c-vac. p-vac 81% 97% 1.5 (21:14) 5 3 36 conserved bop., hop. 50% 30% 0.53 (133:249) 141 131 138 changed sod, slg 87% 83% 0.64 (30:47) 26 21 3 1 ? 102 I hypothesize that the vac genes are the product of one mode of evolution and the hop-bop and sod-slg pairs are the product of a second mode. The close relationship of the sod and slg genes indicates that this mode of evolution may commence very soon after gene duplication. The bop-hop example shows that it can produce a family of genes that encode proteins with distinct, but related activities. The process seems similar to the evolution of pseudogenes, but halobacteria somehow maintain expression of the diverging genes. It is unlike the rapid evolution seen in retroviruses, where third position changes in codons predominate and transitions outnumber transversions (Smith et al. 1988) Duplication of genes may be mediated by the transposable elements in the genome or by reverse transcription. The clustered nature of the changes between the sod and slg; coding regions may be due to multiple gene conversion events that restore sequence identity outside the clustered changes. If this were the case, the duplication that produced sod and slg. may be much older than it appears. Why bop and hop have apparently undergone no gene conversion is unknown. Further analysis of related, expressed genes in halobacteria and other organisms may provide more evidence of this unusual evolutionary process and its phylogenetic distribution. Future research prospects Clearly, interest in archaebacterial evolution is shifting from individual genes to cellular processes. One of the most important processes is transcriptional control. Analysis of events that occur at the promoter, for example the binding of repressor proteins, will reveal some of the mechanisms by which archaebacterial gene regulation is accomplished. There are two ways of approaching this problem: analysis in vitro and 103 analysis in vivo. The former has allowed the polymerase-binding sites of methanobacterial promoters to be mapped (Thomm and Wich 1988, Thomm et al. 1989). Unfortunately, the extreme salt concentrations found in halobacterial cells make in vitro experiments much more difficult. A standard mobility shift assay was used to detect proteins in halobacterial extracts that bind to a DNA fragment containing the 5' flanking region of the sod gene (see Materials and Methods for a description). No binding was seen under the conditions employed. A much better approach is the analysis of deleted and otherwise altered promoters in vivo. Recent developments in the transformation of H_. volcanii by exogenously added DNA have made this feasible (Charlebois et al. 1987, Cline et al. 1989). At the time of writing a shuttle vector capable of replication in E. coli and H_. volcanii has been developed and a marker gene encoding mevinolin resistance has been used to detect transformants (Lam and Doolittle 1989). Another method of investigation is the use of mutants that either overproduce or do not produce SOD. Attempts to isolate overproduces by selecting paraquat-resistant mutants of H_- cutirubrum led to the finding that paraquat resistance was not due to excess SOD (May et al. 1989). Mutants occurred at high frequency (1 in 10^  colonies plated) and are probably due to the simple inactivation of a gene or genes. In E. coli and yeast, SOD-deficient mutants have been created by targeted inactivation of SOD-encoding genes (Carlioz and Touati 1986, van Loon et al. 1986). The mutants are more sensitive to oxygen and paraquat, but recently revertants of E. coli have been found that grow normally and still have no SOD activity (Fee et al. 1988). The alteration that allows reversion is at 104 another, unknown locus. There is apparently no simple system that specifically selects SOD-deficient mutants. The function of slg will remain a mystery until its product can be isolated or until the gene can be inactivated. Recent progress in homologous recombination in H_. volcanii has made the latter a possibility (W. L. Lam and W. F. Doolittle, unpublished). 105 References Aitken, D.M. and A.D. Brown (1969) Citrate and glyoxylate cycles in the halophil Halobacterium salinarium. Biochim. Biophys. Acta 177, 351-354. Asada, K., K. Yoshikawa, M.A. Takahashi, Y. Maeda, and K. Enmanji (1975) Superoxide dismutases from a blue-green alga Plectonema boryanum. J. Biol. Chem. 250, 2801-2807. Asada, K., S. Kanematsu, S. Okaka, and T. Hayakawa (1980) Phylogenetic distribution of three types of superoxide dismutase in organisms and in cell organelles, p. 136-153. In Chemical and biochemical aspects of superoxide dismutase, eds. Bannister, J.V. and H.A.O Hill (Elsevier/ North Holland Publishing Co. New York). Auffray, C. and F. Rougeon (1980) Purification of mouse immunoglobulin heavy-chain messenger RNAs from total myeloma tumor RNA. Eur. J. Biochem. 107, 303-314. Bannister, J.V. and M.W. Parker (1985) The presence of a copper/zinc superoxide dismutase in the bacterium Photobacterium leiognathi: a likely case of gene transfer from eukaryotes to prokaryotes. Proc. Natl. Acad. Sci. USA 82, 149-152. Barra, D., M.E. Schinina, M. Simmaco, J.V. Bannister, W.H. Bannister, G. Rotilio, and F. Bossa (1984) The primary structure of human liver manganese superoxide dismutase. J. Biol. Chem. 259, 12595-12601. Bayley, S.T. (1971) Protein synthesis systems from halophilic bacteria, p. 89-100. In Methods in Molecular Biology, vol. 1, eds. Last, J.A. and A.I. Laskin (Marcel Dekker, New York). Betlach, M., J. Friedman, H.W. Boyer, and F. Pfeifer (1984) Characterization of a halobacterial gene affecting bacterio-opsin gene expression. Nuc. Acids Res. 12, 7949-7959: Blanck, A. and D. Oesterhelt (1987) The halo-opsin gene. II. Sequence, primary structure of halorhodopsin and comparison with bacteriorhodopsin. EMBO J. 6, 265-273. Brock, C.J. and J.E. Walker (1980) Superoxide dismutase from Bacillus  stearothermophilus. Complete amino acid sequence of a manganese enzyme. Biochemistry 19, 2873-2882. 106 Brot, N., L. Weissbach, J. Werth, and H. Weissbach (1981) Enzymatic reduction of protein-bound methionine sulfoxide. Proc. Natl. Acad. Sci. USA 78 , 2155-2158. Cammarano, P., A. Teichner, and P. Londei (1986) Intralineage heterogeneity of archaebacterial ribosomes: evidence for two physicochemically distinct ribosome classes within the third urkingdom. Syst. Appl. Microbiol. 7, 137-143. Carlioz, A. and D. Touati (1986) Isolation of superoxide dismutase mutants of Escherichia coli: is superoxide dismutase necessary for aerobic life? EMBO J. 5, 623-630. Chapman, D.J. and J.W. Schopf (1983) Biological and biochemical effects of the development of an aerobic environment, p.302-320. I_n Earth's Earliest Biosphere: its origin and evolution, ed. Schopf, J.W. (Princeton Univ. Press, Princeton). Charlebois, R.L., W.L. Lam, S.W. Cline, and W.F. Doolittle (1987) Characterization of pHV2 from Halobacterium volcanii and its use in demonstrating transformation of an archaebacterium. Proc. Natl. Acad. Sci. USA 84, 8530-8534. Chirgwin, J.M., A.E. Przybyla, R.J. MacDonald, and W.J. Rutter (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18, 5294-5299. Christian, J.H.B. and J.A. Waltho (1962) Solute concentrations within cells of halophilic and non-halophilic bacteria. Biochim. Biophys. Acta 65, 506-508. Cline, S.W., W.L. Lam, R.L. Charlebois, L.C. Schalkwyk, and W.F. Doolittle (1989) Transformation methods for halophilic archaebacteria. Can. J. Microbiol. 35 , 148-152. Crapo, J.D. and D.L. Tierney (1974) Superoxide dismutase and pulmonary oxygen toxicity. Am. J. Physiol. 226, 1401-1407. DasSarma, S. (1989) Mechanisms of genetic variability in Halobacterium  halobium: the purple membrane and gas vacuole mutations. Can. J. Microbiol. 35, 65-72. 107 DasSarma, S., U.L. RajBhandary, and H.G. Khorana (1984) Bacterio-opsin mRNA in wild-type and bacterio-opsin-deficient Halobacterium  halobium strains. Proc. Natl. Acad. Sci. USA 81, 125-129. Demple, B. and S. Linn (1982) 5,6-Saturated thymine lesions in DNA: production by ultraviolet light or hydrogen peroxide. Nuc. Acids Res. 10, 3781-3789. Dennis, P.P. (1985) Multiple promoters for the transcription of the ribosomal RNA gene cluster in Halobacterium cutirubrum. J. Mol. Biol. 186, 457-461. deRosa, G., C.L. Keen, R.M. Leach, and L.S. Hurley (1980) Regulation of superoxide dismutase activity by dietary manganese. J. Nutr. 110, 795-804. Eisenberg, H. and E.J. Wachtel (1987) Structural studies of halophilic proteins, ribosomes, and organelles of bacteria adapted to, extreme salt concentrations. Ann. Rev. Biophys. Biophys. Chem. 16, 69-92. Farr, S.B., R. D'Ari, and D. Touati (1986) Oxygen-dependent mutagenesis in Escherichia coli lacking superoxide dismutase. Proc. Nad. Acad. Sci. USA 83, 8268-8272. Fee, J.A. (1982) Is superoxide important in oxygen poisoning? Trends Biochem. Sci. 7, 84-86. Fee, J.A., E.R. Shapiro, and T.H. Moss (1976) Direct evidence for manganese (III) binding to the manganosuperoxide dismutase of Escherichia coli B. J. Biol. Chem. 251, 6157-6159. Fee. J.A., E.C. Niederhoffer, and C. Naranjo (1988) First description of a variant of E. coli lacking superoxide dismutase activity yet able to grow efficiently on minimal, oxygenated medium, p. 239-243. In. Trace elements in man and animals, eds. Hurley, L., C. Keen, B. Lonnerdal, and R. Rucker (Plenum Publication Corp., New York). Feinberg, A. and B. Vogelstein (1982) A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132, 6-13. Fridovich, I. (1978) The biology of oxygen radicals. Science 2 0 1 , 875-880. 108 Fridovich, I. (1986a) Biological effects of the superoxide radical. Arch. Biochem. Biophys. 247, 1-8. Fridovich, I. (1986b) Superoxide dismutases. Adv. Enzymol. 58, 68-97. Fukumori, Y., T. Fujiwara, Y. Okada-Takahashi, Y. Mukohata, and T. Yamanaka (1985) Purification and properties of a peroxidase from Halobacterium halobium L-33. J. Biochem. 98, 1055-1061. Gregory, E.M. and I. Fridovich (1973a) Induction of superoxide dismutase by molecular oxygen. J. Bacterid. 114, 543-548. Gregory, E.M. and I. Fridovich (1973b) Enzymatic defenses against the toxicity of oxygen and of streptonigrin in Escherichia coli. J. Bacterid. 129, 1574-1583. Gropp, F., P. Palm, and W. Zillig (1989) Expression and regulation of Halobacterium halobium phage <|)H genes. Can J. MicrobioL 35, 182-188. Haber, F. and J. Weiss (1934) The catalytic decomposition of hydrogen peroxide by iron salts. Proc. R. Soc. London Ser. A 147, 332-351. Hanahan, D (1983) Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166, 557-580. Hartmann, R., H.-D. Sickinger, and D. Oesterhelt (1980) Anaerobic growth of halobacteria. Proc. Nad. Acad. Sci. USA 77, 3821-3825. Hase, T., W. Wakabayashi, H. Matsubara, L. Kercher, D. Oesterhelt, K.K. Rao, and D.O. Hall (1977) Halobacterium halobium ferredoxin: a homologous protein to chloroplast-type ferredoxins. FEBS Lett. 77, 308-310. Hassan, H.M. and CS. Moody (1987) Regulation of manganese-containing superoxide dismutase in Escherichia coli: anaerobic induction by nitrate. J. Biol. Chem. 262, 17173-17177. Hassan, H.M. and I. Fridovich (1977) Regulation of the synthesis of superoxide dismutase in Escherichia coli: induction by methyl viologen. J. Biol. Chem. 252, 7667-7672. Hassan, H.M. and I. Fridovich (1978) Superoxide radical and the oxygen enhancement of the toxicity of paraquat in Escherichia coli. J. Biol. Chem. 253, 8143-8148. 109 Henikoff, S. (1984) Unidirectional digestion with exonuclease III creates targeted breakpoints for DNA sequencing. Gene 2 8 , 351-359. Henikoff, S. (1987) Exonuclease III generated deletions for DNA sequence analysis. Promega Nptes (Promega Corp.) No. 8, p. 1-3. Hennighuasen, L. and H. Lubon (1987) Interaction of protein with DNA in vitro. Methods Enzymol. 1 5 2 , 721-735. Hewitt, J. and J.G. Morris (1975) Superoxide dismutase in some obligately anaerobic bacteria. FEBS Lett. 5 0 , 315-318. Hodgson, E.K. and I. Fridovich (1975) The interaction of bovine erythrocyte superoxide dismutase with hydrogen peroxide: inactivation of the enzyme. Biochemistry 1 4 , 5294-5299. Holmes, W., T. Piatt, M. Rosenberg (1983) Termination of transcription in Escherichia coli. Cell 3 2 . 1029-1032. Home, M., C. Englert, and F. Pfeifer (1988) Two genes encoding gas vacuole proteins in Halobacterium halobium. Mol. Gen. Genet. 2 1 3 , 459-464. Imlay, J.A. and S. Linn (1988) DNA damage and oxygen radical toxicity. Science 2 4 0 , 1302-1309. Jones, W.J., D.P. Nagle Jr., and W.B. Whitman (1987) Methanogens and the diversity of archaebacteria. Microbiol. Rev. 5 1 , 135-177. Kessel, M. and F. Klink (1981) Two elongation factors from the extremely halophilic archaebacterium Halobacterium cutirubrum. Eur. J. Biochem. 1 1 4 , 481-486. Kirby, T.W., J.R. Lancaster Jr., and I. Fridovich (1981) Isolation and characterization of the iron-containing superoxide dismutase of Methanobacterium bryantii. Arch. Biochem. Biophys. 2 1 0 , 140-148. Klug, D., J. Rabani, and I. Fridovich (1972) A direct demonstration of the catalytic action of superoxide dismutase through the use of pulse radiolysis. J. Biol. Chem. 2 4 7 , 4839-4845. Koppenol, W.H. and J. Butler (1977) Mechanism of reactions involving singlet oxygen and the superoxide anion. FEBS Lett. 8 3 , 1-16. Kozak, M. (1978) How do eucaryotic ribosomes select initiation regions in messenger RNA? Cell 15, 1109-1123. 110 Krantz, M.J. and C.E. Ballou (1973) Analysis of Halobacterium halobium gas vesicles. J. Bacteriol. 114, 1058-1067. Kushner, D.J. (1985) The Halobacteriaceae, p. 171-214. In. The bacteria, vol. 8, eds. Woese, CR. and R.S. Wolfe (Academic Press, New York). Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227, 680-685. Lake, J.A. (1988) Origin of the eucaryotic nucleus determined by rate invariant analysis of rRNA sequences. Nature (London) 331, 184-186. Lam, W.L. and W.F. Doolittle (1989) Shuttle vectors for the archaebacterium Halobacterium volcanii. Proc. Natl. Acad. Sci. USA, in press. Lanyi, J. (1974) Salt dependent properties of proteins from extremely halophilic bacteria. Bacteriol. Rev. 38, 272-290. Lanyi, J. (1979) Physicochemical aspects of salt dependence in halobacteria, p. 93-107 I_n Strategies of microbes in extreme environments, ed. Shilo, M. (Verlag Chemie, Weinheim, FRG). . Larsen, H. (1984) Family V. Halobacteriaceae, p. 261-267. In Bergey's Manual of Systematic Bacteriology, eds. Krieg, N.R. and J.G. Holt (Williams and Wilkins, Baltimore). Lechner, J. and M. Sumper (1987) The primary structure of a procaryotic glycoprotein: cloning and sequencing of the cell surface glycoprotein gene of halobacteria. J. Biol. Chem. 262, 9724-9729. Lee, Y.M., DJ. Friedman, and F.J. Ayala (1986) Superoxide dismutase: an evolutionary puzzle. Proc. Natl. Acad. Sci. USA. 82, 824-828. Leffers, H., F. Gropp, F. Lottspeich, W. Zillig, and R.A. Garret (1989) Sequence, organization, transcription and evolution of RNA polymerase subunit genes from the archaebacterial extreme halophiles Halobacterium halobium and Halococcus morhuae. J. Mol. Biol. 206, 1-17. Leicht, W., M.M. Werber, and H. Eisenberg (1978) Purification and characterization of glutamate dehydrogenase from Halobacterium of the Dead Sea. Biochemistry 117, 4004-4010. I l l Leicht, W. and S. Pundak (1981) Large scale purification of halophilic enzymes by salting-out mediated chromatography. Anal. Biochem. 1 1 4 , 186-192. Lipman, DJ. and W.RV Pearson (1985) Rapid and sensitive protein similarity searches. Science 2 2 7 , 1435-1441. Maniatis, T., E.F. Fritsch, and J. Sambrook (1982) Molecular Cloning: a laboratory manual (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY). Margulis, L. (1970) Origin of eucaryotic cells. (Yale University Press, New Haven CT). Marklund, S. and G. Marklund (1974) Involvement of the superoxide anion in the autoxidation of pyrogallol and a convenient assay for superoxide dismutase. Eur. J. Biochem. 4 7 , 468-474. Maxam, A. and W. Gilbert (1980) Sequencing end-labelled DNA with base specific chemical cleavages. Methods Enzymol. 65, 499-560. May, B.P. and P.P. Dennis (1987) Superoxide dismutase from the extremely halophilic archaebacterium Halobacterium cutirubrum. J. Bacteriol. 169, 1417-1422. May, B.P., P. Tarn, and P.P. Dennis (1989) The expression of the superoxide dismutase gene in Halobacterium cutirubrum and Halobacterium  volcanii. Can. J. Microbiol. 35, 171-175. McAdam, M.E., R.A. Fox, F. Lavelle, and E.M. Fielden (1977a) A pulse-radiolysis study of the manganese-containing superoxide dismutase from Bacillus stearothermophilus: a kinetic model for the enzyme action. Biochem. J. 165, 71-79. McAdam, M.E., R.A. Fox, F. Lavelle, and E.M. Fielden (1977b) A pulse-radiolysis study of the manganese-containing superoxide dismutase from Bacillus stearothermophilus: further studies on the properties of this enzyme. Biochem. J. 165, 81-87. McCord, J.M. and I. Fridovich (1969) Superoxide dismutase: an enzymatic function for erythrocuprein (hemocuprein). J. Biol. Chem. 2 4 4 , 6049-6055. 112 Mead, J. (1976) Free radical mechanisms of lipid damage and consequences for cellular membranes, p. 51-68. I_n Free radicals in biology, ed. W.A. Pry or (Academic Press, New York). Messing, J. (1983) New, M13 vectors for cloning. Methods Enzymol. 101, 20-79. Mevarech, M., W. Leicht, and M.M. Werber (1976) Hydrophobic chromatography and fractionation of enzymes from extremely halophilic bacteria using decreasing concentration gradients of ammonium sulfate. Biochemistry 15, 2383-2387. Misra, H.P. and I. Fridovich (1978) Inhibition of superoxide dismutases by azide. Arch. Biochem. Biophys. 189, 317-322. Miyake, K. (1986) Effect of nitrate on the level of superoxide dismutase in anaerobically grown Escherichia coli. J. Gen. Appl. Microbiol. 32, 527-533. Moody, C.S. and H.M. Hassan (1984) Anaerobic biosynthesis of the manganese-containing superoxide dismutase in Escherichia coli. J. Biol. Chem. 259, 12821-12825. Morris, J.G. (1979) Nature of oxygen toxicity in anaerobic microorganisms, p. 149-162. In. Strategies of microbial life in extreme environments, ed. M. Shilo (Verlag Chemie, Weinheim, FRG). Newman, A. (1987) Specific accessory sequences in Saccharomyces  cerevisiae introns control assembly of pre-mRNAs into spliceosomes. EMBO J. 6, 3833-3839. Oesterhelt, D. and W. Stoekenius (1971) Rhodopsin-like pigment from the purple membrane of Halobacterium halobium. Nature (New Biol.) 233, 149-152. Oesterhelt, D. and W. Stoekenius (1973) Functions of a new photoreceptor membrane. Proc. Natl. Acad. Sci. USA 70, 2853-2857. Pater, A. and M.M. Pater (1977) Simultaneous separation of halophilic proteins and nucleic acids after adsorption onto agarose gels. Can. J. Biochem. 55, 904-907. 113 Peterson, G . L . (1977) A simplification of the protein assay method of Lowry et al. which is more generally applicable. A n a l . Biochem. 83, 346-356 . Pfeifer, F . (1988) Genetics of halobacteria, p. 105-133. In Halophilic bacteria, ed. Rodriguez-Valera, F. ( C R C Press, Boca Raton, F L ) . Pfeifer, F . and M . Betlach (1985) Genome organization in Halobacterium  h a l o b i u m : a 70 kb island of more (AT) rich D N A in the chromosome. M o l . Gen. Genet. 198, 449-455. Pfeifer, F . , G . Weidinger, and W . Goebel (1981) Characterization of plasmids in halobacteria. J. Bacteriol. 145, 369-374. Phillips, J.P., S.D. Campbell, D . Michaud, M . Charbonneau, and A . J . Hilliker (1989) N u l l mutation of copper/zinc superoxide dismutase in Drosophila confers hypersensitivity to paraquat and reduced longevity. Proc. Natl. Acad. Sci . U S A 86, 2761-2765. Privalle, C T . and E . M . Gregory (1979) Superoxide dismutase and oxygen lethality in Bacteroides fragilis . J. Bacteriol. 138, 139-145. Privalle, C T . and I. Fridovich (1987) Induction of superoxide dismutase in Escherichia coli by heat shock. Proc. Natl. Acad. Sci . U S A 84, 2723-2726. Privalle, C T . and I. Fridovich (1988) Inductions of superoxide dismutases in Escherichia col i under anaerobic conditions. Accumulation of an inactive form of the manganese enzyme. J . B i o l . Chem. 263, 4274-4279. Puget, K . and A . M . Michelson (1974) Isolation of a new copper-containing superoxide dismutase bacteriocuprein. Biochem Biophys. Res. Commun. 58 , 830-838. Pugh, S .Y.R. , J .L . DiGuiseppi, and I. Fridovich (1984) Induction of superoxide dismutases in Escher ichia c o l i by manganese and iron. J. Bacteriol. 160, 137-142. Reistad, R. (1970) O n the composition and nature of the bulk protein of the extremely halophilic bacteria. Arch . Mikrobiol 71, 353-360. Reiter, W . - D . , P. Palm, and W . Zillig (1988) Analysis of transcription in the archaebacterium Sulfolobus indicates that archaebacterial promoters are homologous to eucaryotic pol l l promoters. Nuc. Acids Res. 16, 1-19. 114 Saenger, W. (1987) Structure and dynamics of water surrounding biomolecules. Ann. Rev. Biophys. Biophys. Chem. 16, 93-114. Sakamoto, H. and D. Touati (1984) Cloning of the iron superoxide dismutase gene (sodB) in Escherichia coli K-12. J. Bacteriol. 159, 418-420. Salin, M.L. and D. Oesterhelt (1988) Purification of a manganese-containing superoxide dismutase from Halobacterium halobium. Arch. Biochem. Biophys. 260, 806-810. Salin, M.L., M.V. Duke, D. Oesterhelt, and D.-P. Ma (1988) Cloning and determination of the nucleotide sequence of the Mn-containing superoxide dismutase gene from Halobacterium halobium. Gene 70, 153-159. Sanger, F., S. Nicklen, and A.R. Coulson (1977) DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463-5467. Sapienza, C. and W.F. Doolittle (1982) Unusual physical organization of the halobacterial genome. Nature (London) 295, 384-389. Schiavone, J.R. and H.M. Hassan (1988) The role of redox in the regulation of manganese-containing superoxide dismutase biosynthesis in Escherichia coli. J. Biol. Chem. 263, 4269-4273. Schnabel, H., W. Zillig, M. Pfaffle, R. Schnabel, H. Michel, and H. Delius (1982) Halobacterium halobium phage <J>H1. EMBO J. 1, 87-92. Schnabel, R., M. Thomm, R. Gerardy-Schahn, W. Zillig, K.O. Stetter, and J. Huet (1983) Structural homology between different archaebacterial DNA-dependent RNA polymerases analyzed by immunological comparison of their components. EMBO J. 2, 751-755. Schobert, B. and J.K. Lanyi (1982) Halorhodopsin is a light-driven chloride pump. J. Biol. Chem. 257, 10306-10313. Searcy, K.B. and D.G. Searcy (1981) Superoxide dismutase from the archaebacterium Thermoplasma acidophilum. Biochim. Biophys. Acta 670, 39-46. Shimmin, L.C., and P.P. Dennis (1989) Characterization of the Lll, LI, L10 and LI2 equivalent ribosomal protein gene cluster of the halophilic archaebacterium Halobacterium cutirubrum. EMBO J. 8, 1225-1235. 115 Shine, J. and L. Dalgarno (1974) The 3' terminal sequence of Escherichia  coli 16S ribosomal RNA: complementarity to nonsense triplets and ribosome binding sites. Proc. Natl. Acad. Sci. USA 71, 1342-1346. Simon, L.M., J. Liu, J. Theodore, And E.D. Robin (1977) Effect of hyperoxia, hypoxia, and maturation on superoxide dismutase activity in isolated alveolar macrophages. Am. Rev. Respir. Dis. 115, 279-284. Simons, R.S., P.S. Jackett, M.E.W. Carroll, D.B. Lowrie (1976) Superoxide independence of paraquat toxicity. Toxicol. Appl. Pharmacol. 37, 271-275. Smith, M.W. and F. Neidhardt (1983) Proteins induced by aerobiosis in Escherichia coli. J. Bacteriol. 154, 344-350. Smith, T.F., A. Srinivasan, G. Schochetman, M. Marcus, and G. Myers (1988) The phylogenetic history of immunodeficiency viruses. Nature (London) 333, 573-575. Southern, E.M. (1975) Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98, 503-517. Stallings, W.C, K.A. Partridge, R.K. Strong, and M.L. Ludwig (1984) Manganese and iron superoxide dismutases are structural homologs. J. Biol. Chem. 259, 10695-10699. Steinman, H.M. (1980) The amino acid sequence of copper-zinc superoxide dismutase from bakers' yeast. J. Biol. Chem. 255, 6758-6765. Steinman, H.M. (1982) Superoxide dismutases: protein chemistry and structure-function relationships, p. 11-68. In. Superoxide Dismutase, vol. 1, ed. L.W. Oberly (CRC Press, Boca Raton, FL). Stoeckenius, W. and R.A. Bogomolni (1982) Bacteriorhodopsin and related pigments of halobacteria. Ann. Rev. Biochem. 52, 587-616. Tabor, S. and C.C. Richardson (1987) DNA sequence analysis with modified bacteriophage T7 DNA polymerase. Proc. Natl. Acad. Sci. USA 84, 4767-4771. Thomm, M. and G. Wich (1988) An archaebacterial promoter element for stable RNA genes with homology to the TATA box of higher eucaryotes. Nuc. Acids Res. 16, 151-163. 116 Thomm, M., G. Wich, J.W. Brown, G. Frey, B.A. Sherf, and G.S. Beckler (1989) An archaebacterial promoter sequence assigned by RNA polymerase binding experiments. Can. J. Microbiol. 35, 30-35. Touati, D. (1983) Cloning and mapping of the manganese superoxide dismutase gene (sodA) of Escherichia coli K-12. J. Bacteriol. 155, 1078-1087. Touati, D. (1988) Transcriptional and posttranscriptional regulation of manganese superoxide dismutase biosynthesis in Escherichia coli. studied with operon and protein fusions. J. Bacteriol. 170, 2511-2520. van Loon, A.P.G.M., B. Pesold-Hurt, and G. Schatz (1986) A yeast mutant lacking mitochondrial manganese-superoxide dismutase is hypersensitive to oxygen. Proc. Natl. Acad. Sci. USA 86, 3820-3824. Venkatesan, S. and B. Moss (1981) In vitro transcription of the inverted terminal repetition of the vaccinia virus genome: correspondence of initiation and cap sites. J. Virol. 37, 738-747. von der Haar (1978) The ligand induced solubility shift in salting out chromatography. FEBS Lett. 94, 371-375. von Hippel, P.H. and T. Schleich (1969) The effects of neutral salts on the structure and conformational stability of macromolecules in solution, p. 417-574. I_n Structure and stability of biological macromolecules, vol. 2, ed. Timasheff, S.N. (Marcel Dekker, Inc. New York). Walter, M.R. (1983) Archean stromatolites: evidence of the earth's earliest benthos, p. 187-212. In Earth's earliest biosphere: its origin and evolution, ed. Schopf, J.W. (Princeton University Press, Princeton, NJ). Weisinger, R.A. and I. Fridovich (1973) Superoxide dismutase: organelle specificity. J. Biol. Chem. 248, 3582-3592. Woese, CR. (1987) Bacterial evolution. Microbiol. Rev. 51, 221-271. Yanisch-Perron, C, J. Viera, and J. Messing (1985) Improved M13 phage cloning vectors and host strains: nucleotide sequence of the M13mpl8 and pUC19 vectors. Gene 33, 103-119. Zuckerkandl, E. and L. Pauling (1965) Molecules as documents of evolutionary history. J. Theor. Biol. 8, 357-366. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items