AMBER SUPPRESSION IN THE ARCHAEBACTERRJMHAL OFERAX VOLCANIIbyJosephine YauHon. B.Sc., The University of Waterloo, 1991.A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTERS OF SCIENCEinTHE FACULTY OF GRADUATE STUDIESDepartment of Biochemistry and Molecular BiologyWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAFebruary 1994© Josephine Yau, 1994In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)__________________Department of Pcc OLE(LLL&L, OLO yThe University of British ColumbiaVancouver, CanadaDateDE-6 (2188)11ABSTRACTThe purpose of this project was to test whether amber suppression can occurin Haloferax volcanii or not, and if so, to construct a H. volcanii strain that cansuppress amber mutations. To achieve this goal, a putative amber suppressor wasconstructed from the tyrosine transfer RNA of H. volcanii. Its ability to recognize anamber stop codon and to restore the wild type function of an amber mutated genewas tested.The gene coding for tyrosine tRNA of H. volcanii DS2 was cloned andsequenced. Site-directed mutagenesis was carried out to change the anticodon of thetRNA gene from GUA to CUA, which recognizes the tyrosine codon UAU andUAC, and the amber stop codon UAG respectively.The hisC gene of wild type H. volcanii was obtained and recloned intopGEM7(-). Site directed mutagenesis was carried out to change the DNA sequence ofits first tyrosine codon (TAC) to an amber stop codon (TAG).I attempted to replace the wild type hisC gene in the H. volcanii WFD11genome with the hisC gene carrying the amber mutation. However, although theconstruct carrying the hisC(Am) gene and an antibiotic resistance marker integratedinto the genome at the correct place, displacement of the wild type gene throughreverse recombination did not occur.An attempt to test amber suppression in H. volcanii was carried out. ThehisC(Am) gene was introduced into the genome of a mutant strain of H. volcaniiWR256 (his, argj with the antibiotic selection marker mevinolin. Thetransformants were then transformed again with a plasmid that carries the putativeamber suppressor (the tRNATY1 gene with the mutated amber anticodon) and theother antibiotic selection marker novobiocin. Transformants were then selectedwith both antibiotics and then tested for restoration of histidine auxotrophy. Alltransformants still required histidine for growth. Southern hybridization showedthat the hisC(Am) gene was not integrated into the genome. Mevinolin resistancein the transformants was due to a double crossover recombination event of theantibiotic resistance gene into the genome to replace the wild type gene. Therefore Iwas not able to conclude whether amber suppression can occur in H.volcanii or not.111ivTABLE OF CONTENTSABSTRACT iiTABLE OF CONTENTS ivLIST OF TABLES viLIST OF FIGURES viiABBREVIATIONS viiiACKNOWLEDGEMENTS xiDEDICATIONS xiiI. INTRODUCTION 11.1. Purpose of the project 11.2. The genetic code 11.3. Transfer ribonucleic acid 41.4. Mutations 61.5. Suppression 61.6. Amber suppression 81.7. Archaebacteria 101.8. The experimental approach 17II. MATERIALS AND METHODS 182.1. Bacterial strains, plasmid constructions and oligonucleotide sequences 182.2. Media and culture conditions 182.3. General techiques of molecular biology 212.3.1. Preparation of plasmid DNA 212.3.2. Preparation of double stranded DNA for sequencing 212.3.3. Restriction endonuclease digestion of DNA 212.3.4. Gel electrophoresis 222.3.5. DNA restriction fragment preparation 222.3.6. Oligonucleotide purification 222.3.7. Dephosphorylation of identical cohesive termini of vectorDNA 222.3.8. Ligations 222.3.9. Competent cells and transformation 232.3.10. Preparation of single-stranded DNA 23V2.3.11. Generation of deletions by exonuclease.242.3.12. Sequencing 242.3.13. Labelling of oligonucleotide probes 242.3.14. Southern hybridization 242.4. Molecular biology techniques for HALOFERAX VOLCANII 252.4.1. Competent cells and transformation 252.4.2. Preparation of genomic DNA 262.4.3. Preparation of plasmid DNA 262.4.4. Shuttle vectors for Hvo 272.5. Mutagenesis 272.6. Autoradiograph scanning 28Ill. RESULTS 293.1. Cloning of the tyrosine tRNA gene 293.2. The tyrosine tRNA sequence 323.3. Mutagenesis of the tyrosine tRNA gene 363.4. The hisC gene 363.5. Mutagenesis of thehisC gene 413.6. Strategy for construction of hisC(Am) strain 413.7. Test for incorporarion of the pHC12M construct 453.8. Test for natural amber suppression 483.9. Test for differential growth rate 483.1O.Plasmid construction 493.11.Test for amber suppression in Haloferax volcanii 49IV. DISCUSSION4.1. The tyrosine tRNA 544.2. Modification pattern 564.3, The hisC gene 574.4. Gene replacement experiments 584.5. Plasmids construction 594.6. Test for amber suppression in Haloferax volcanii 604.7. Conclusion 61V. REFERENCES 63viLIST OF TABLESTable 1: The genetic code .2Table 2: Bacterial strains, plasmid constructions and oligonucleotidesequences 19Table 3: Construction of plasmids that carry either the hisC(Arn) gene or theputative amber suppressor in pGOT vectors 50Table 4: Attempts to construct a plasmid which contains both the ambersuppressor and the hisC(Am) gene 50Table 5: Control experiment to test for incorporation of the putative ambersuppressor into pHC12 51viiLIST OF FIGURESFigure 1: The structure of transfer RNA 5Figure 2: Gene replacement in Haloferax volcanii 16Figure 3: Tyrosine transfer RNA sequence 30Figure 4: Southern hybridization of wild type genomic restricted DNA to detectfor presence of tyr tRNA gene 31Figure 5: Restriction mapping of plasmid pTl 33Figure 6: The plasmid pTl 34Figure 7: DNA sequence of the tyrosine tRNA gene 35Figure 8: Plasmid construction of pT2.l, pT2.2 and pT3 37Figure 9: Site- directed mutagenesis of tyrosine tRNA 38Figure 10: DNA and derived amino acid sequences of the H. volcanii hisC gene.39Figure 11: Construction of plasmid pHC and pHC12 40Figure 12: Site-directed mutagenesis of the hisC gene 42Figure 13: Strategy for constructing the his C(Am)H. volcanii strain 43Figure 14: Southern hybridization to test for the incorporation of pHC12M intoWFD11 46Figure 15: Predicted result of Southern hybridization when the construct isintegrated into the hisC locus 47Figure 16: Southern hybridization to test for the incorporation of pHC12M intoWR256 53Figure 17: Comparison of halobacterial tRNA 5’-flanking regions 55viiiABBREVIATIONSA adenosineATP adenosine triphosphatebop bacteriorhodopsinbp base pairdegrees centigrade (Celcius)C cytosineCsC1 cesium chlorideD dihydroxyuridinedH2O distilled, sterile waterDNA deoxyribonucleic acidDNase deoxyribonucleasedNTP deoxynucleoside triphosphateEDTA ethylene diamine tetraacetic acidG guanosinehis histidineH v o Haloferax volcaniikbp kilobaseM molarmev mevinolinmevr mevinolin resistant.d microlitreixml millilitremRNA messenger ribonucleic acidnov novobiocinflovr novobiocin resistantO.D. optical densityoligo oligonucleotidep plasmidPEG polyethylene glycolPNK polynucleotide kinasepmol picomoleRNA ribonucleic acidRNase ribonucleaserpm revolutions per minuterRNA ribosomal RNAS Svedberg unit of sedimentation coefficientSDS sodium dodecyl sulphatess single-strandedsu suppressorT thymidineTE 10mM Tris-Ci, 1mM EDTAtyr tyrosinetRNA transfer ribonucleic acidU uridineu unitura uracilP pseudouridineYT 8g/1 Bacto trytone, 5g/1 Bacto yeast extract, 5g/1 NaC1xxiACKNOWLEDGMENTSI thank the members of my supervisory committee, Dr. R. T. A. McGillivrayand Dr. P. Candido, for their help and advice during the course of my studies. I alsothank my supervisor, Dr. P. P. Dennis, for his advice and the opportunity to work inhis lab.I thank the members of the lab (Luc Bissonette, Janet Chow, Peter Durovic,Phalgun Joshi, Daiqing Liao, Shanthini Mylvaganam, Simon Potter and Janet Yee)for their support and advice. I also thank Deidre de Jong-Wong for her technicalhelp. Special thanks to Mark M. K. Lee for helping me with the computer scanning.I also thank the M. Smith lab, especially CM-Yip Ho for advice and help with sitedirected mutagenesis. Last, but not least, I would like to thank Dr. R. Redfield for herhelp and advice.I would also like to acknowledge the financial support of the Natural Scienceand Engineering Research Council.DEDICATIONThis thesis is dedicated to my parents, Paul and Molly Yau, and to Eric Leung, fortheir support and encouragement.xii1INTRODUCTION1.1 PURPOSE OF THE PROJECTThe goal of this project was to study amber suppression in thearchaebacterium H. volcanii. Amber suppression has been widely used in E. coli forstudying gene function and expression by providing a system to isolate conditionalmutants in an essential gene. To understand amber suppression, some generalbackground on the genetic code, transcription and translation, transfer RNAproperties and functions, and suppression of mutations will be discussed in thefollowing sections.Haloferax volcanii is a member of the archaebacteria, a group only recentlyrecognized as a separate line of evolutionary descent from a common primordialancestor, apart from the eubacteria and eukaryotes (Woose et at., 1990). The study ofarchaebacteria helps us to better understand the universal ancestor and theevolution of cells. Although recent advances in the development of archaebacterialgenetics have led to a better understanding of the unique phenotypic and geneticcharacteristics of the group, our knowledge is still very limited. The development ofan amber suppressing archaebacterial strain would be useful for the study ofarchaebacterial gene functions and the manipulation of genetic material in thefuture.1.2 THE GENETIC CODEThe genetic code (Table 1) uses the nucleotide sequence of genomic DNA tospecify the order of amino acids in protein produced by a cell or organism (Crick,1966). The vast array of different proteins possess both structural and catalyticproperties that determine the cell’s unique abilities, relating to metabolism, growthU C A CUUU1 UCU UAU1 UGU1 Uj Phe UAC j Tyr UGC j CysU UUA 1 L UCAer UAA* Stop UGA* Stop AUUG J U UCG UAG* Stop UGG Trp CCUU CCU CAU1 . CGU UCUC CCC CAC j CGC CC CUA Leu CCA Pro CAA 1 CGA Arg ACUG CCG CAG j Gin CGG GAUU 1 ACU AAU 1 AGU 1 UAUC He ACC ? j Asn AGC j Ser cA AUA i ACA Thr AAA 1 AGA 1 AAUGt Met ACG AAG] Lys AGG j Arg GGUU GCU GAUl GGU UGUC GCC GAC J ‘ GGC CG GUA Val GCA Ala CAA 1 GGA Gly AGUGt GCG GAG J Glu GGG GTABLE 1: The Genetic CodeTaken from Watson et at (1987), p.43723and division. The characteristic features used to encode protein sequenceinformation were first elucidated for the organism E. coli and are now recognized tobe nearly universal and therefore generally applicable to all other organisms.The code consists of triplets; three successive nucleotide base pairs in DNAspecify one of the twenty different amino acids found in protein. Triplet codons arenon-overlapping, read sequentially in a 5’ to 3’ direction, and when read in sequencespecify the order of amino acids from the N (amino) to the C (carboxyl) terminus ofthe protein. The genetic code is degenerate; using the standard bases found in DNAand RNA, there are sixty-four different triplet combinations, but there are onlytwenty amino acids found in protein. Examination of the codon assignmentsindicates that the third base in the triplet is least important and often carries little orno information. They are given the name wobble bases. For example, the aminoacid glycine is specified by four codons which differ only in the third position: GGG,GGC, GGA, and GGU. These properties help to minimize the deleterious effects ofmutations that occur at the third codon position and simplify the decoding process.The genetic information carried in DNA is not decoded directly. Instead thesequence from the coding strand of the DNA is first transcribed onto messengerRNA. The mRNA is then decoded on the ribosomes by a series of interactions withamino acylated transfer RNA molecules (Hoagland et at., 1957). Four of the sixtyfour codons are used as signals to punctuate the code. Aside from specifying theamino acid methionine, the codon AUG is also recognized by a special initiatormethionine tRNA to signal the site within the mRNA where translation is to begin(Adams and Capecchi,1966). Three codons, UAA (ochre), UAG (amber) and UGA(opal), do not specify amino acids and are not recognized by any tRNA. Instead theysignal the end of translation and are recognized by proteins called releasing factors.4They are the sites where translation stops and the newly synthesized polypeptide isreleased from the ribosome.1.3. TRANSFER RIBONUCLEIC ACIDAll tRNAs share some common sequence and structural features and can befolded into a compact L-shaped configuration (Holley et al., 1965; Rich andRajBhandary, 1976) (figure 1). Transfer RNAs contain a large number of unusualbases such as dihydrouracil, pseudouracil, 4-thiouracil, methyl or dimethyl guanineand methyl adenine, which are normally not found in other RNA species. Thesebases are generated by enzymatic modification after transcription and are found inthe ioop region of the tRNA structure. Their functions are still not fully understood(reviewed in Singer and Kroger, 1979). Nishimira (1979) suggests that the modifiednucleosides in the anticodon region may stabilize and enhance certain anticodoncodon base pairing interactions or act as wobble bases. They might also be involvedin stabilizing the conformation of the tRNA or its binding to the ribosome, or mightenhance its resistance to degradation by ribonuclease or the specificity of recognitionof aminoacyl-tRNA synthetase.All tRNAs contain four stems: the acceptor stem, the D stem, the anticodonstem, and the TNIC stem. In some species, an extra arm 3 to 15 bases long may bepresent between the TcC stem and the anticodon stem. The anticodon loop containsthe anticodon which recognizes and base pairs with codons during transcription.The end terminal CCA region of the acceptor stem is the site where thecorresponding amino acid is attached by aminoacyl synthetases (reviewed in Saks eta!., 1994). The amino acyl tRNAs read the triplet codon on the mRNA by basepairing with the complementary tRNA anticodon loop, and sequentially insertamino acids onto the growing peptide.5FIG 1: The Structure of Transfer RNAa. Primary structure of yeast phenylalnine tRNAb. The compact L-shape molecule of tRNATaken from Darnell et at (1986), p.112alAmino acdAm.no acid a(mS.ITCG a.m0 loon•1Amino acid arm5 endmG.TdCG lo —C4-Y end.niodon loQAjcodonCodonCC0CC61.4 MUTATIONSA mutation is a change in any DNA base pair and can occur in many forms:substitution, deletion or insertion. Within coding regions, synonymous mutationsrepresent third position substitution between codons specifying a single amino acid;they do not affect the phenotype of the organism, and are often not detected. Non-synonymous mutations are alterations that cause an amino acid replacement in theencoded protein. The resulting polypeptide product is often still functional or partlyfunctional. Insertions and deletions are often deleterious since they often change thereading frame on the messenger RNA. This produces a protein with an aberrantamino acid sequence beyond the site of the frameshift, giving rise to a dysfunctionalproduct. However, when the number of inserted or deleted bases is a multiple ofthree, the reading frame is not disturbed, and the resulting protein contains eitherfewer or extra amino acids at the site of the deletion or insertion.Nonsense mutations are usually recovered less frequently than the othermutations (Garen, 1968). This type of mutation changes an amino acid specificcodon to a chain terminating codon, and results in premature termination oftranslation and production of a truncated polypeptide product. The length of thepolypeptide translated depends on the position of the mutation within the gene.Incomplete polypeptide fragments generated by nonsense mutations are usuallynon-functional because they do not contain all the essential structure required foractivity. Since the products of nonsense mutations are usually non-functional, theyare very useful in the elucidation of biochemical pathways and protein functions.1.5 SUPPRESSION7Reversion of a mutation in a gene is often a result of a back mutation tospecify the original amino acid. In other cases, it involves second site mutationswhich can be either intragenic or intergenic. Intergenic suppression can be a result ofmissense or nonsense suppression.A missense mutation can be suppressed by a second missense mutationwithin the same gene (Hill, 1975). When the first missense mutation alters the threedimensional structure of the protein, which is essential for its activity, a secondmissense mutation may restore the three dimensional structure and hence thebiological activity of the protein. Missense suppression is usually inefficient andonly partial in its restoration of function. Therefore it is not commonly used ingenetic analysis.In nonsense suppression, a mutant tRNA molecule has an anticodoncomplementary to a chain terminating codon ( reviewed in Lewin, 1974). Themutant tRNA then inserts a functional amino acid at the site of the nonsensemutation to prevent premature termination. The mutation is suppressed and thewild type phenotype will be restored. Suppressors which utilize the UGA, UAA, andUAG stop codons are known as opal, ochre and amber suppressors, respectively(Hirsh, 1971; Garen, 1968; Engelhardt et al., 1965).The nature of opal suppression is still not fully understood, since its mode ofaction is different from the model proposed for the other nonsense suppressors(Hirsh, 1971; Hirsh and Gold, 1971). The tryptophan tRNA suppressor is notmutated at the anticodon region but at position 24 of the tRNA, where a G isreplaced by an A. This affects the three dimensional structure of the tRNA, whichallows moderate suppression of the opal UAG codon. Normal tryptophan tRNAwith anticodon CCA can also read the opal stop codon UGA but at a much lower8frequency.The ochre suppressor is much weaker than the other two suppressors (Garen,1968). The ochre codon UAA is the most frequently used stop codon in vivo and itssuppression is clearly deleterious. Cells that carry the ochre suppressor usually growpoorly and therefore an efficient ochre suppressor cannot be isolated.1.6 AMBER SUPPRESSIONThe amber suppressor system is the most widely used and best understoodnonsense suppressor system. The suppression of amber mutations is an intergenicnonsense suppression which occurs at the level of translation of mRNA(Engelhardt et at., 1965). The codons UAU and UAC specify the amino acid tyrosine,whereas the amber stop codon UAG specifies termination of messenger ENAtranslation. In a non-permissive system (one lacking an amber suppressor tRNA), achange in the third base of the tyrosine codon from a U or a C to a G results inpremature termination of translation which gives rise to an incomplete and usuallynon-functional protein product. However, in a permissive system, the minorspecies of the tyrosine tRNA is mutated in the anticodon from GUA to CUA (theamber suppressor). It can now recognize the UAG stop codon and insert tyrosine atthat position, preventing termination. A complete protein product is then formed(Goodman et at., 1968).In E. coli, amber suppressors can be derived from single base mutations in theanticodon region of several different species of tRNA. They can be formed bymutating the third base in the anticodon of glutamine (CUC), glutamic acid (CUG)and lysine (CULT) tRNAs, or the second base of leucine (CAA) and serine (CGA)9tRNAs. However, amber mutants formed by a mutation in the first base position ofthe tyrosine anticodon (su3) are most widely used and understood. The systemutilizes two species of genes which code for tyrosyl-tRNA, a major and a minor one.The major gene is responsible for the synthesis of most of the tyrosyl-tRNA. Theminor one consists of two copies of tyrosyl-tRNA genes and either one or both ofthem is mutated to form a suppressor (Russel et al., 1969). In this way, the systemstill has an abundant amount of tyrosyl-tRNA for normal translation. The minorspecies is mapped close to the attachment site of phage 80; therefore it can bepackaged into the phage to promote excess synthesis of the minor tRNA (Abelson etat., 1970; Smith et at., 1970). Lewin (1974) summarized the efficiency of 12 ambersuppression systems and concluded that the amber suppressor su3 showed the mostefficient suppression as measured in the suppression of four different genes (T4head protein, alkaline phosphatase, 13-galactosidase and ornithine transcarbamylase).Amber suppression has been widely studied and utilized in E. coti foranalyzing gene expression. Stretton and Brenner (1965) used the amber system todemonstrate that genes and proteins are colinear. Abelson et at. (1970) and Smith etat. (1970) isolated mutants of tyrosyl-tRNA to study the relationship betweensequence change and the functional defects in the molecule. They were able to drawconclusions regarding the structural and functional roles of particular sequences oftRNAs.Amber suppression provides a convenient genetic system by which theexpression of an amber mutated gene can be turned “on” or “off” easily (Gestelandet at., 1967). This provides a way for conditional mutants in an essential gene to beisolated. Suppression competes with chain termination. Each suppressor gene ischaracterized by the relative frequency of amino acid insertion compared totermination, which ranges from less than 1% to about 60% (Smith et at., 1966).101.7 ARCHAEBACTERIAArchaebacteria are a group of organisms that exhibit prokaryotic-like cellstructure and organization but at the same time possess a number of eukaryoticfeatures (Dennis, 1986). The discovery and studies of archaebacteria provide us witha new perspective on early events in the evolution of cells, and help us tounderstand better the universal ancestor and to develop a more accurate concept ofeukaryotic origins. Before the recognition of the archaebacterial kingdom, life onearth was divided into two primary kingdoms: the eukaryotes and the prokaryotes,which includes the eubacteria and the archaebacteria. The archaebacteria are nowbelieved to be more closely related to the eukaryotes (Woese, 1981), and comprisesone of the three newly defined domains: the eukaryotes (Eucarya), eubacteria(Bacteria) and archaebacteria (Archaea) (Woose et a!., 1990).Archaebacteria propagate at biological extremes of temperature, pH and saltconcentration. The group is diverged into two branches: the methanogenichalophilic and the sulfur-dependent thermophilic branch. The former is composedof groups with two distinct phenotypes: anaerobic methane producers and aerobichalophiles. Halophiles grow at slightly elevated temperatures of 350C to 500C andsalt concentrations of 1.5 to 5M NaCl. They have a sequence complexity comparableto that ofE. coli, with a high G+C content and their genome sizes range between5x10 to io bp (Moore and McCarthy, 1969a). In Halobacterium halobium, the G+Ccontent of the chromosomal DNA and the satellite DNA are 66 to 68 mole percentand 57 to 60 mole percent, respectively (Moore and McCarthy, 1969b). Halophilesalso possess a restriction-modification system similar to those found in eubacteria(Daniels and Wais, 1984). Their metabolism and physiology are similar to those ofeubacteria, with some unusual features such as the synthesis of isoleucine frompyruvate and acetyl-CoA instead of from threonine (Eikmarins and Thauer, 1984).11The structure and catalytic properties of many of their enzymes closely resemblethose of either the eubacteria or the eukaryotes.The halophiles possess many unique genetic characteristics. Thechromosomal and extrachromosomal DNA of H. halobium are genetically andphysically very unstable. Plasmid rearrangements are frequent and complex, andinsertions and deletions occur at high frequency. This instability is due to thepresence of repetitive sequences and abundant insertion elements. There are at least500 repetitive sequences in H. halobium (Doolittle, 1985), and there are alsoabundant multiple repetitive sequences in Haloferax volcanii.Comparative analysis of both 5S rRNA and 16S small subunit RNA showsthat archaebacteria form a unique and coherent phylogenetic group as reviewed inDennis (1993). Analysis of the 16S rRNA shows that archaebacteria appear to bemore closely related to the common ancestor than either the eubacteria or theeukaryotes (Woese et a!, 1983). The RNA polymerase found in archaebacteria mostclosely resembles the eukaryotic ENA polymerase II and III in amino acid sequence.Only a single polymerase has been found in any archaebacterium (Zillig et a!, 1985)whereas eukaryotes possess three nuclear RNA polymerases.In archaebacteria, three rRNAs are present: 5S, 16S, and 23S. The ribosomesdissociate into two components: 30S and 50S. Ribosomal RNA sequences areextensively diversified within the archaebacteria; however, there is a cleardistinction between those of the archaebacterial kingdom and those of eubacteriaand eukaryotes.Although the general secondary structure of archaebacterial tRNA is similarto the eubacterial and eukaryotic tRNAs (Gupta, 1985), they possess many uniquestructural details. The archaebacterial tRNA genes lack the 3’-CCA terminal end12(Wich et al, 1984; Hui and Dennis, 1985). This sequence is added as a posttranscriptional modification by tRNA terminal transferase. The consensus TTAAmotif of all archaebacterial promoters appear to be related to the TATA box of theeukaryotic ENA polymerase II promoters (Reiter et a!, 1990; Thomm and Wich,1988; Thomm et a!, 1989).Archaebacterial transfer RNAs have a characteristic pattern of post-translational modification which is distinctly different from the correspondingeubacterial and eukaryotic patterns. The modified bases T and m7G foundcommonly in eubacteria and eukaryotes are absent in all archaebacteria examined(Gupta and Woese, 1980). Modified nucleosides present in the H. volcanii tyrosinetRNA include pseudouridine (‘P), 1-methylpseudouridine(mP), 2-methylcytidine(Cm), 5-methylcytidine (m5C) and 1-methylguanosine (m’G) (Gupta, 1984). Atposition 54, instead of the nucleoside T found in most eubacterial and eukaryotictRNAs, pseudouridine is present in most archaebacterial tRNAs. In the tyrosinetRNA of H. volcanii, it is modified to m1’P. 1-Methylpseudouridine and T havesimilar molecular profiles and base pairing properties and could be an example of anevolutionary convergence of structures (Gupta, 1985). The modified nucleosides ‘1’and m1P could be involved in binding tENAs to ribosomes (Sprinzl et a!., 1976). Amethylated G (m1G) is present at position 37. Nishirura (1979) proposed that themodified base right next to the anticodon may help stabilize the codon-anticodoninteraction. At position 56 of all examined archaebacterial tRNAs, the modified basemethylated cytosine Cm is present. Cm may also be present at position 32. Position57 of many H. volcanii tRNAs consists of a methylated inosine, which is also aunique characteristic of archaebacterial tRNAs.With the increasing interest in the study of the molecular biology of thearchaebacteria, new genetic tools and selection systems are needed for the analysis of13gene structure, function and regulation. The halophilic archaebacterium H.volcaniiwas discovered by Benjamin E. Volcani and was later isolated and studied(Mullakhanbhai and Larsen, 1975). The optimum sodium chloride requirement ofthe organism is 1.7M, which is close to that found in the Dead Sea. They have a hightolerance for magnesium chloride and are basically disc-shaped; however, the sizeand shape of the cells vary with the culture conditions and from cell to cell in thesame culture. The cells are very fragile and are easily ruptured by mechanicaltreatments (Muriana,et at, 1987) and can be lysed in a hypotonic solution. The DNAhas a high GC content (63%), which is characteristic of a halobacterium. The cells ofH. volcanii are orange to red in color due to carotenoids, and have a characteristicodor.The H.volcanii DS2 genome consists of a 2920 kbp chromosome and fourplasmids: a 690 kbp pHV4, a 442 kbp pHV3, a 86 kbp pHV1, and a 6.4 kbp pHV2. Byethidium bromide treatment, Charlebois and co-workers (1987) cured the pHV2plasmid from DS2 to generate a new strain, WFD11. This strain has no visibledifference in growth rate or phenotype compared to DS2. It transforms efficientlywith the plasmid pHV2 and an artificial construct, pHV2693. Using the WFD11strain, a PEG-mediated spheroplast transformation system was derived (Cline et al.,1989). The efficiency of uptake and expression was comparable to the efficiency oftransfection of H. halobium with phageH DNA (Cline and Doolittle, 1987).Spheroplasts are generated by addition of EDTA to chelate Mg2 ions. The cellsregenerate when Mg2+ levels in the growth medium are returned to normal.Archaebacteria are insensitive to most commonly used antibiotics (Boch andKandler, 1985). They are found to be sensitive to mevinolin and novobiocin.Spontaneous mutants of H. volcanii resistant to the two antibiotics were isolated.This made the development of vectors with selectable markers possible (Holmes14and Dyall-Smith, 1991; Holmes et al., 1991; Lam and Doolittle, 1989). Novobiocininhibits the activity of eubacterial DNA gyrase by binding to the GyrB subunit andblocking the access of ATP to its binding site on the subunit (Mizuuchi et al, 1978). Itis thought that halobacteria are similarly inhibited (Holmes and Dyall-Smith, 1991).Resistant H. volcanii have been shown to produce a gyrase that binds novobiocinless avidly (Thiara and Cundliff, 1988). Mevinolin is an inhibitor of the enzyme 3 -hydroxyl- 3 methylgiutaryl coenzyme A reductase (Cabrera et al, 1985), which isessential in the mevalonate pathway for synthesis of isoprenoid lipids (Kates et a!,1968). Cells are inhibited due to decreased mevalonate availability instead of ageneralized toxic effect. With the development of an efficient transformation systemand the availability of selectable markers, shuttle vector systems were developed(Lam and Doolittle, 1989; Holmes et a!., 1991). This provides a convenient system tomove DNA back and forth between the H. volcanii and E. coli systems and greatlysimplifies DNA propagation and sequence manipulation.The shuttle vectors contain the origin of replication in plasmids isolated fromdifferent strains of H.volcanii, such as pHK2 from strain Aa 2.2 (Holmes et a!, 1991)and pHV2 from DS2 (Lam and Doolittle, 1989). This provides the essentials forreplication and maintenance in the H. vo!canii system. The selectable antibioticresistant genes (mev” or novr) are present so that transformants can be isolatedefficiently. Fragments of E. coli plasmids constitute the rest of the vector, providingthe necessary sequences for plasmid selection and maintenance in E. coli. A wideselection of shuttle vectors are now available. The newer versions are considerablyreduced in size so that larger fragments of DNA can be transformed efficiently(Bissonette and Dennis, unpublished).The organism H. volcanii was chosen for study because an efficient plasmidtransformation and selection system was already made possible with a number of15cloning vehicles. It seems to be genetically more stable than other members of thehalophiles due to the presence of fewer insertion sequences (Doolittle, 1985). Thegenome of H. volcanii DS2 has been collected as minimally overlapping fragmentsof about 36 kbp produced by partial digestion with MluI (Charlebois et at, 1991). Thecollection covers 96% of the whole genome and is available as cosmid clones (Cohenet a!., 1992). Furthermore, H. volcanii grows readily on defined basal salt media withglycerol and succinate as carbon energy source. Since this species is prototrophic,auxotrophic mutants can be readily isolated.A natural system of genetic exchange (Mevarech and Werczberger, 1985) andgene replacement (Krebset at, 1993) in halobacteria has been discovered. Genetictransfer between H. votcanii cells requires cell-to-cell contact and is insensitive toDNase. Transfer is bi-directional and believed to be through cytoplasmic bridgesbetween the two participants of a mating event (Rosenshine and Mevarech,unpublished result cited in Dennis, 1993). Gene replacement is achieved by DNAuptake and integration by homologous recombination of non-replicating DNA intothe chromosome (reviewed in Dennis, 1993). Linear DNA can be recombined intothe chromosome by two crossing-over events, resulting in gene replacement (Fig2a), whereas circular DNA is integrated into the chromosome through a singlecross-over event, generating a tandem duplication (Fig 2b) (Lam and Doolittle, 1989).Mutant sequences along with a selectable marker on a circular molecule can be putinto the chromosome. The clones are then transferred to non-selective media, andthe chromosome will readily undergo a second recombination event to expel one ofthe duplicate sequences. Ideally, in 50% of all cases, the original inserted DNAfragment will be excised. In the other 50% of the cases, the wild type sequence will beexcised to produce the replacement. Krebs and co - workers (1993) had successfullyutilized this method to introduce deletions into the chromosomal bop gene of H.16-Fa. LinearDNA recombined into the chromosome by two cross-overeventsL///7]y ,iVb. Circular DNA recombined into the chromosome by a single cross-overeventFIG 2. GENE REPLACEMENT IN Haloferax volcaniiProducts of homologous recombination between a chromosomal sequenceand a linear or circular DNA fragment. Homologous DNA sequences arerepresented by the shaded area. Chromosomal DNA is represented by brokenlines. The positions of recombination are indicated by crosses. This figure isadapted from Dennis, 199317halobium. The bop gene activity of the isolated mutants ranged from 0% to 56%activity as compared to the wild type.1.8 THE EXPERIMENTAL APPROACHTo test whether amber suppression can occur in H. volcanii , the tyrosine transferRNA was cloned and its anticodon region changed from GTA to CTA (RNAsequence GUA to CUA) by site-directed mutagenesis, making it complementary tothe amber stop codon UAG. The putative amber suppressor was then tested for itsability to recognize the amber stop codon and to restore wild type function in anamber mutated gene. Originally, the uradil auxotrophy gene was chosen as thecandidate for testing the ability of the altered tRNA to suppress amber mutationssince it has a convenient positive selection system for auxotrophs (Kondo et al,1991). However, after several attempts to clone the gene and reviewing the results ofother experiments performed by Dennis (unpublished), we concluded that the H.volcanii urtr mutants that we obtained from the Doolittle lab exhibit partial activityin the mutant protein and therefore cannot be used in the cloning of the gene.The hisC gene of H. volcanii (Conover and Doolittle, 1990) was then used fortesting the amber suppression system. It contains an open reading frame whichencodes histidinol-phosphate aminotransferase, the eighth enzyme of the histidinebiosynthetic pathway. By generating an amber mutation in the gene and thenputting the two (his(Am) gene and the putative amber suppressor tRNA) togetherin the H. volcanii system, we tested for the possibility of amber suppression in H.volcanii.18MATERIALS AND METHODS2.1 BACTERIAL STRAINS, PLASMID CONSTRUCTION ANDOLIGONUCLEOTIDE SEQUENCESThe bacterial strains, plasmid constructions and oligonucleotide sequencesthat were used are described in Table 2.2.2 MEDIA AND CULTURE CONDITIONSAll E. coli strains were grown either in YT media (5g/l Bacto-yeast extract, Sg/lBacto-tryptone, 5g/l NaCl, pH7.5), 2xYT or minimal salts (M9) media (6g/lNa2HPO4, 3g/l KH2PO4, 0.5g/l NaC1, lg/l NH4C1) supplemented with glucose(0.2%), MgSO4 (2mM), CaC12 (0.1mM) and thiamine (0.5ig/ml). Solid media wereprepared by addition of Bacto-agar (15g/l). All strains were grown at 370 C. Whenrequired, antibiotic concentrations used were: ampicillin (100pg/ml) and kanamycin(50pg/ml).Haloferax volcanii was grown in the basal salt medium SWG (3.32M NaCl,0.1M MgSO4.7H20, 0.08M MgC12.6H2O, 0.07M KC1, 6.8mM CaCl2.2H2O, 4.9mMNaBr, 2.4 mM NaHCO3) supplemented with 0.5% yeast extract (Oxoid). Whenminimal media were prepared, the following components were added to 1L of thesalt media instead of yeast extract: 5 ml 1M NH4C1, 45 ml of 10% glycerol, 5 ml of10% sodium succinate, 1 ml of trace elements and 2 ml of 0.5M K2HPO4(Mevarechand Werczberger, 1985). Solid media were prepared by addition of 15g/l agartechnical (agar no. 3; Oxoid). Histidine or arginine supplements were added at aconcentration of 5Opg/ml when needed. All strains were grown at either 37°C or420C. The antibiotics mevinolin and novobiocin were added at concentrations of50.tM and 0.2 pg/ml respectively when needed.19TABLE 2: BACTERIAL STRAINS. PLASMID CONSTRUCTIONS ANDOLIGONUCLEOTIDE SEQUENCESStrain DescriptionE. coli DH5c F- recAl endAl gyrA96 thi hsdRl7 supE44 relAl 2-HB2151 K12 ora 8(lac-pro) thi/F’ proA+B+ lacZ 8M15JM1O1 supE thi 3(lac-proAB), jF’ traD36 proAB lacIZ8M15JRZ1032 Zbd-2791lysA(61-62)J, duti, ungi, thil, rell,supE44,Tnl 0H. volcanii DS2 wild type Haloferax volcanii consisting of a 2920kbp chromosome and 4 plasmids: pHV1-4WFD1 1 strain DS2 lacking the plasmid pHV2WR256 mutagenised WFD11 his argPlasmid DescriptionpT4C 2.7 kbp SacI-MluI fragment containing hisC genein SmaI site of pGS18 obtained from the Doolittlelab.pHC 2.7 kbp BamHI-EcoRJ fragment containing hisCgene subclone from pT4C in pGEM7(-)pHC12 a derivative of pHC containing the mutation TACto TAG in codon 12 of hisCpTl 4.5 kbp SmaI fragment containing the H. volcaniityrosine tRNA gene in pGEM7(-)pT2.1 800bp SmaI-EcoRI fragment containing the H.volcanii tyrosine tENA gene in pGEM7(-)20pT2.2 800bp SmaI-EcoRI fragment containing the H.volcanii tyrosine tRNA gene with an alteredanticodon GUA to CUApT3 1.5 kbp Smal fragment in pGEM7(-) containing theputative amber suppressing tyrosine tRNA geneOligonucleotide Description01 TCCGCTCTCCCCGATIT/Ca minus strand sequence complementary toposition 73 to 57 within the tyrosine tRNA gene02 AGAGCAGCCGACTGTAGa plus strand sequence identical to position 21 to 37within the tyrosine tRNA gene03 CTGCTCAAACCGGCTCGa minus strand sequence complementary to theregion 17 bp 3’ downstream of tyr tRNA04 CGATCTAGAGTCGGCTGCTCTa minus strand sequence complementary toposition 41 to 21 within the tyrosine tRNA geneexcept at position 34, where a G is present instead ofaC05(JMH) ACGCTCCCTAGGTACCCGGCCGa plus strand sequence identical to position 26to 47 within the hisC gene except at position 36,where a G is present instead of a C06(JSH) TCGGTCGTCGCGTCCCCAACa plus strand sequence identical to position70 to 89 within the hisC gene2107(JSH2) TGGGCGGTCTrCGGGTAGACa minus strand sequence complementary toposition 200 to 180 within the hisC gene2.3 GENERAL TECHNIQUES OF MOLECULAR BIOLOGYGeneral recombinant DNA techniques were carried out according toSambrook et at. (1982) unless otherwise specified.2.3.1 PREPARATION OF PLASMID DNASmall scale preparation of plasmid DNA was done by the alkaline lysismethod (Sambrook et at., 1982) or one-step miniprep method (Chowdhury, 1991).Large scale DNA was prepared using alkaline lysis (Sambrook et at., 1982), with thefollowing modifications: the supernatant was not filtered through layers ofcheesecloth. The final DNA pellet was dissolved in 97% CsC1 (97g/lOOml dH2O) andbanded in a CsC1-ethidium bromide gradient at 50,000 rpm overnight. The DNAobtained after removal of EtBr was diluted in three volumes dH2O and thenprecipitated with 95% EtOH.2.3.2 PREPARATION OF DOUBLE STRANDED DNA FOR SEQUENCINGDouble stranded DNA was either prepared as described in Saunders andBurke (1990) using CsC1 or with the Magic miniprep DNA purification systemsupplied by Promega. Reactions were carried out according to manufacturer’sprotocol.2.3.3 RESTRICTION ENDONUCLEASE DIGESTION OF DNARestriction enzymes used were purchased from Pharmacia Inc., Bethesda22Research Laboratories (BRL) or New England Biolabs. Digestions were carried outaccording to the instructions of suppliers.2.3.4 GEL ELECTROPHORESISSamples of DNA were separated by electrophoresis on agarose gels (0.7% or1%) in TBE buffer (89mM Tris, 89mM boric acid, 2.5mM Na2EDTA) in the presenceof 0.25 mg/mi ethidium bromide.2.3.5 DNA RESTRICTION FRAGMENT PREPARATIONBands of restricted DNA, stained with ethidium bromide, were excised fromagarose gels and recovered using Sephaglas BandPrep Kit (Pharmacia)2.3.6 OLIGONUCLEOTIDE PURIFICATIONOligonucleotides were purified as described in Sawadogo and Dyke (1991) andthe concentrations were determined by spectrophotometry.2.3.7 DEPHOSPHORYLATION OF IDENTICAL COHESIVE TERMINI OF VECTORDNAIdentical cohesive termini of vector DNA were dephosphorylated withshrimp alkaline phosphatase (SAP) (United States Biochemical) by using 3u ofenzyme per microgram of DNA. The reaction was incubated in SAP buffer (20mMTris-Cl pH8.0, 10mM MgC12) for 1 hour at 37°C to prevent religation. The enzymewas then heat inactivated at 65°C for 15 minutes.2.3.8 LIGATIONSCohesive-end ligations were incubated at 14-160Covernight or 2 hours atroom temperature.23For blunt-end ligations, incubations were at room temperature overnight or 2hours with 5% PEG8000.2.3.9 COMPETENT CELLS AND TRANSFORMATIONCompetent cells (DH5c or HB2151) were made by treatment with CaC12(Sambrook et al, 1982). Frozen competent cells were prepared as above andresuspended in ice-cold 100mM CaC12 containing 15% glycerol. The suspension wasincubated on ice for 1 hour and then aliquoted into lOOp.! stocks. These were thenfrozen in dry ice and stored at -70°C until needed. Transformation of plasmid DNAwas carried out by incubating competent cells with DNA on ice for 30 minutes. Thecells were then heat shocked at 37°C for 2 minutes, plated on selective media andincubated for 12-16 hours.2.3.10 PREPARATION OF SINGLE -STRANDED DNASingle stranded DNA template can be made from a host carrying a phagemidby superinfection using a helper phage. The plasmid pGEM7(-) can be used as aphagemid since it contains the fi origin of replication. A single colony of freshlygrown HB2151 on M9 plates containing the plasmid pGEM7(-) with the appropriateinserts was inoculated into 2xYT media. The culture was grown to mid-exponentialphase. It was then infected with either phage R408 or M13K07 and incubated for 6hours with good aeration. Cells were then spun down and the phage particles andDNA were precipitated in 4% PEG8000 and 0.7M NH4OAc. The pellet wasresuspended in TESDS(lOOmM Tris, 10mM EDTA, 0.1%SDS) and incubated withproteinase K for 30 minutes. The sample was then extracted with phenol andchloroform until no protein interface was observed and then precipitated andwashed with ethanol.242.3.11 GENERATION OF DELETIONS BY EXONUCLEASEExonuclease deletion was performed as described in the Promega protocolsand application guide(1991), p90-98, at 32.50C. Samples were taken every 30 seconds.After treatment with Klenow and dNTPs, the samples were loaded on a 1% agarosegel to determine the efficiency and extent of deletion. The bands were thenrecovered as described in 2.3.5, the ends were religated to form a circular plasmidand then transformed into DH5x.2.3.12 SEQUENCINGSequencing was performed according to the instructions supplied with theDeaza G/A T7 Sequencing Kit (Pharmacia) with the exception that the primer wasadded to the double stranded DNA template before denaturing in 0.2M NaOH.Furthermore, the template/primer/NaOH mixture was denatured by boiling in awater bath for 5 minutes and then quickly quenched on ice.2.3.13 LABELING OF OLIGONUCLEOTIDE AND FRAGMENT PROBESOligonucleotide probes were end-labeled using bacteriophage T4polynucleotide kinase (PNK) and [y-32PJATP (sp. act. 3000 Ci/mmol, 10.tCi/pi).Bacteriophage T4 PNK was inactivated by heating to 65°C for 10 minutes.Restriction fragments purified as in section 2.3.5 were radiolabeled using theRandom Primers DNA Labeling System (BRL). The labeling reaction was carried outat 250C for 1 hour. The enzyme was then inactivated by addition of Sjil of stop bufferand boiled for 5 minutes in a water bath. Approximately bOng of DNA fragmentwas used in each reaction and half of the labeled reaction mixture was used in eachSouthern Hybridization.2.3.14 SOUTHERN HYBRIDIZATION25Southern hybridization was carried out either on Hybond paper or using thedried-down agarose gel.DNA was transferred and probed on Hybond paper as described in theprotocol: Blotting and hybridization protocols for HybondTM membranes suppliedby Amersham. The transfer was carried out overnight. Prehybridization was carriedout at 400C or 65°C and hybridization at 50°C for labeled oligonucleotide probes or65°C for fragment probe respectively. When oligonucleotide probes were used, thethird wash was not carried out.Hybridization with genomic DNA was done in a dried gel. The 1% agarose gelwas dried under vacuum at room temperature for 1 hour and then at 60°C foranother 30 minutes or until the gel is dried. The gel was then denatured for 30minutes in 0.5M NaOH/0.15M NaCl and neutralized twice for 15 minutes in 0.15MTris/0.15M NaCl. The gel was hybridized to the probe by heating to the estimatedTm(melting temperature) of the probe for 30 minutes and then cooled slowly downto room temperature. The hybridized gel was then washed with 6xSSC (175.3g/LNaCl. 88.2g/L Na.citrate.2H20, pH7) to remove any non-specific binding of probe.Autoradiography was then carried out on the membrane or gel.2.4 MOLECULAR BIOLOGY TECHNIQUES FOR HALOFERAX VOLCANII2.4.1 PREPARATION OF COMPETENT CELLS AND TRANSFORMATIONCompetent cells of WFD11 or other strains of halobacteria were prepared bypelleting 1 ml of culture in mid or late exponential phase(O.D.= 0.5-1.5) and thenresuspending in 200p.l spheroplast generation solution (0.8 M NaC1, 27mM KC1,50mM Tris-HC1, 15% glycerol, 15% sucrose). The cells can either be usedimmediately or frozen at -70°C.Spheroplasts were formed by addition of 20il 0.5M EDTA and incubated with26the DNA solution. Then 240ii1 of HFPEG(6m1 PEG600 and 4ml spheroplastgeneration solution) was added and mixed gently until a homogenous solution wasobtained. Cells were regenerated by adding 1 ml of solution R (3.4M NaCl, 175mMMgSO4.7H20, 34mM KC1, 7mM CaCl2.2H20, 50mM Tris-HC1, 15% sucrose) andgrown in SWGR (SWG supplemented with yeast extract and 15% sucrose) for 6hours before plating on appropriate media(Cline et al, 1989). The plates wereincubated at 42°C for 5-14 days.2.4.2 PREPARATION OF GENOMIC DNAThe pellet from 3mls of a H. volcanii culture was resuspended in 0.4 ml lysisbuffer(4OmM Tris, 20mM EDTA and 10mg/mi lysozyme), vortexed vigorously andincubated at 37°C for 30 minutes. Then 0.1 ml of 5% SDS with lOOj.tg/ml RNase wasthen added and incubated for 5 minutes at 45-60°C. The clear, viscous solution wasthen extracted with phenol twice and then with chloroform: isoamyl alcohol once.The aqueous phase was transferred to a fresh tube containing 15.tl of 5M NaCl andmixed well. To precipitate the DNA, cold 95% ethanol was added and mixed gentlyuntil DNA threads were visible. The DNA was spooled onto a sealed pasteur pipetteand rinsed by dribbling 1 ml of 70% ethanol over the pipet tip. The DNA was thendried by standing the tip up for 10 minutes and resuspended by swirling the tip in0.2m1 TE warmed to 37°C.2.4.3 PREPARATION OF PLASMID DNAOne milliliter of H. volcanii culture was pelleted and resuspended in 100jil ofHFNTh (1M NaC1, 50mM Tris, 10mM EDTA, pH8). Cells were lysed by adding 5 ill of5% deoxycholic acid, followed by 200il of alkaline lysis solution II (0.2N NaOH and1% SDS) and solutionifi (294.5 g/L potassium acetate and 115 ml/L acetic acid). It wasthen centrifuged and the supernatant was recovered. DNA was ethanol precipitated27and resuspended in TE. Proteins were differentially precipitated by addingammonium acetate to a final concentration of 2M.2.4.4 SHUTTLE VECTORS FOR H. VOLCANIIShuttle vectors were used throughout the experiments to move DNA backand forth between E. coli and H. volcanii for genetic manipulation and in vivo geneexprssions respectively. The pGOT series of shuttle vectors used are as described inBissonnette and Dennis (unpublished).2.5 MUTAGENESISUracil containing single stranded DNA template was synthesized in RZ1032,and prepared as described in 2.3.10. Mutagenesis (Kunkel, 1985; Kunkel et a!, 1987;Ner et al,1988) was carried out with oligo 4 on tyr tRNA gene and with oligo 5 on hisC gene.One microgram of uracil containing single stranded DNA template wasannealed with 5 pmol of the corresponding phosphorylated oligonucleotide in loxannealing buffer (100mM Tris-HC1 pH8, 500mM NaCl, 100mM MgC12 and 10mMDTT). The mixture was heated to 750C for 10 minutes and cooled slowly to roomtemperature. Extension and ligation were then performed by addition of one-fifthvolume of 5x polymerase mix (100mM Tris-HC1 pH8.8, 10mM DTT, 50mM MgCl2,5mM ATP and 2.5mM each of dATP, dTTP, dGTP, dCTP) and 2U of T4 DNA ligaseand 2 u Klenow. The mixture was left on ice for 5 minutes, then incubated at roomtemperature for 2 hours. Further aliquots of Klenow (2u) and ligase (2u) were addedto the mixture and incubated at 370C for another 2 hours. One-tenth of the reactionmixture was transformed into E.coli DH5oc and colonies formed were individuallypicked and sequenced. Only the G and C sequencing reactions were performed in theinitial screening of the mutagenized clones to avoid redundant sequencing.2.6. AUTORADIOGRAPH SCANNINGAll autoradiographs obtained from Southern hybridizations and sequencingreactions were scanned using a flatbed scanner (Hewlett Packard ScanJet Plus). Theimages were printed with a scanning program onto a laser printer.2829RESULTS3.1 Cloning of tyrosine tRNA geneThe sequence of the tyrosine tRNA was determined by Gupta (1984), (fig3). Toclone the tyrosine tRNA gene, Southern hybridization was carried out on genomicdigests of DS2 DNA to detect the gene. Fragments from the region where a positivesignal was obtained were then cloned into a vector to obtain a size-fractionatedpartial library. The library was then rescreened to obtain an individual clone whichcarries the gene.Oligonucleotides 1 and 2 were used for detecting for the presence of thetyrosine tRNA gene of H. volcanii. Oligonucleotide 1 is complementary to thetRNA sequence (fig 3) at the 3’ end from position 73 to 57, with a degenerate 3’ end atposition 57. Oligonucleotide 2 is identical to the RNA sequence from position 21 to37, which includes the anticodon region. Genomic DNAs from H. volcanii DS2 andWFD11 were digested with different restriction enzymes and separated on a 1%agarose gel. Southern hybridization was carried out using radiolabeledoligonucleotide 1 to detect the location of the tyrosine tRNA gene (fig 4). Positivesignals were further confirmed by hybridization with oligonucleotide 2. The resultsobtained from the DS2 and WFD11 DNA digests were identical.A Smal genomic fragment of about 1.5kbp in length hybridized to botholigonucleotides. To clone the fragment, DNA from this region of the gel (1.4 to1.6kbp) was eluted, purified, shotgun cloned into the dephosphorylated Smal site ofpGEM7(-) and transformed into E. coli DH5a. Plasmid DNA from the transformantswas analyzed in pools of 4 and screened for the presence of the gene by Southernhybridization using oligonucleotide 2 as probe. Individual transformants from30AOHCCAC-GC-GG-CC-GU-AC-GU-A UAU GGGGC ACXA A 11111 m’IC CUCG CCCCG CmU 1111 C m’PPG GAGC UGCA A U GG-C UC-GC-GG-CA-P(C)Cm AU mTGGUAFIG 3. Tyrosine transfer RNA sequence (adapted from Gupta, 1984)Transfer RNAs of H. volcanii were separated by two-dimensional gelelectrophoresis and sequenced. Gupta (1984) presented the RNA sequence of41 H. volcanii tRNAs including the tyrosine tRNA as shown. They resemblethe general structure and sequence of eubacterial and eukaryotic tENAs.31kbp11.54.7 —2.8 —2.0 —1.7 —1.15 —0.85 —FIG 4: Southern hybridization of restricted wild type genomic DNA to detectthe presence of the tyrosine transfer RNA geneGenomic DNA of wild type Hvo DS2 was digested with various restrictionenzymes: Lanes: 1, MboI; 2, RsaI; 3,BglI; 4,Hindffl; 5,EcoRI; 6,BamFll; 7,PstI; 8Sail; 9, Smal; 10, XhoI; 11, TaqI; 12, NcoI; 13, NarI; 14, Msp1; 15, MiuI;, 16,Asp718. Arrow shows the positive signal at 1.5kbp of the Smal digest.Oligonucleotidel was used as probe.4 5 6 7 8 9 10 11 13 14 15 132positive pools were rescreened in order to obtain the correct clone.3.2 The tyrosine tRNA sequenceThe positive clone pTl was chosen for further study. By restriction mapping(fig 5), the structure of plasmid pTl was analyzed. When the plasmid was digestedwith Smal, a 4.5 kbp band and two bands around 1.5kbp were observed. Whendigested with EcoRI, 2.1 kbp and 5.1 kbp fragments were observed. When theplasmid was digested with both enzymes, a 3 kbp vector band, two 1.5 kbp bands, a0.8 kbp band and a 0.6 kbp band were observed. Southern hybridization showed thatthe tyrosine tRNA gene is contained in the 1.5kbp Smal, the 5.lkbp EcoRI and the0.8kbp EcoRI-SmaI fragments. The plasmid pTl contained three 1.5 kbp fragments inpGEM7(-). The first Smal site was lost since Smal gives blunt ends. The tyrosinetRNA coding region is contained in the middle fragment 38 bp downstream fromthe EcoRI site (fig 6).The DNA sequence of the tyrosine tRNA gene was determined usingoligonucleotide 2 and oligonucleotide 3 as primers (fig 7). At position 15, where thenucleotide is undefined from the published ENA sequence, the DNA sequencecontained a G residue. Furthermore, the methylated inosine at position 57 wasdetermined to have come from the nucleotide A. The putative promoter waspresent as the TTAA box at position -32. The putative terminator was also present asa stretch of Ts followed by As. Upstream consensus promoter elements were alsoidentified at positions -44, -38 and -13. The EcoRI site was present as part of thepromoter element at position -38. The terminal sequence CAA is absent in the DNAsequence of the gene.33Lanes:3 8 12-—--bFIG 5: Restriction mapping of plasmid pTl.(a) Plasmid pTl was digested with different restriction enzymesand fractionated by electrophoresis. Lanes: 1, uncut; 2, Asp718; 3,EcoRI; 4, MiuI; 5, NcoI; 6, Sail; 7 Sau3A; 8, SmaI; 9, SphI; 10, SstI;11, PstI; 12, EcoRI+S maT; 13, Ec0RV+SmaI; 14, MiuI+SmaI; 15,NcoI+SmaI; 16, SaiI+SmaI; 17, Sau3A+SmaI. It was thentransferred to nitrocellulose and probed with 32P labeledoligonucleotide2. The location of the tyrosine tRNA gene withineach digest is shown in (b). Size markers ( DNA cut with PstI)are indicated by arrows from top to bottom: 11.5, 5.1, 4.5, 2.8, 2.0,1.7, 1.2, 0.8, 0.4 kbp.a34FIG 6 :The plasmid pTlI Tyr tENA geneThe plasmid pTl contains three 1.5 kbp fragments cloned fromgenomic SmaI digested DS2 DNA. The tyrosine transfer RNA gene islocated in the middle fragment 32 bases downstream of the EcoRI site,which starts approximately 0.6 kbp downstream of the secondSmalsite. The first SmaI site was lost during religation of blunt ends.Smal EcoRl35—41 —38 —32CGTCGGGCGCTTCGCCGCGfiTGAftftTTCTTRA GTCTGCCCGTGGATT—13 +16 AG AT TCTCT TGA CC GC TCTT AGCTCA GCCTGG CAGAGC A GCCGA C TGTA+75GATCGGCTTGTCCCCCGTTCAAATC6666AGAGCGGRTTTT GCTTGCAAAAFIG 7: DNA Sequence of the tyrosine tRNA geneThe sequence of the tyrosine transfer RNA gene was determined bysequencing using oligo 2 and 3 as primers. The gene is only 73 base pairs inlength. The upstream consensus promoter elements are identified atpositions -44, -38 and -13. The promoter sequence TTAA is present at position-32 and the terminator sequence 1ffI at position 75.363.3 Mutagenesis of tyrosine tRNA geneThe 0.8 kbp SmaI-EcoRI fragment from plasmid pTl was subcloned intopGEM7(-) (pT2.1)(Fig 8a). The smaller fragment permits sequencing through thewhole gene with the universal primer. The possibility of incorporation of wrongnucleotides into the upstream region during site-directed mutagenesis is alsoeliminated. Site-directed mutagenesis (Smith, 1986) was carried out on pT2.1 usingoligonucleotide 4. The anticodon region was mutated from GUA to CUA(fig 9),which in principle should allow the tENA to recognize the amber stop codon TAG(pT2.2; fig 8a). The mutated 0.8 kbp fragment was then excised using HindIII andEcoRI and religated with the upstream 0.6 kbp Smal EcoRI fragment into the HindIIISmal site of pGEM7(-) to reform the entire gene with complete upstream promoterelements. This clone containing the putative tyrosine tRNA amber suppressinggene was designated pT3 (fig 8b).3.4 The hisC geneA clone of the hisC gene of H. volcanii (Conover and Doolittle, 1990) (fig 10)was obtained as pT4C, which contains a 2.7kbp fragment in pBGS18, a kanamycinresistant derivative of pUC18 (Spratt et al, 1986). Since the vector pGEM7(-) was usedthroughout the project, the hisC gene was recloned into the vector for consistencyand convenient manipulation. The plasmid pT4C was cut using BamHI and EcoRIand the 2.7 kbp fragment containing the hisC gene was ligated into the BamHI-EcoRIsite of pGEM7(-) to form pHC(fig 11).37a. plasmid pT2.1/pT2.2 containing 0.8 kpb insert in pGEM7(-)I Tyr tRNA geneb. plasmid pT3 containing 1.5 kpb insert in pGEM7(-)I Tyr tRNA geneFIG 8: Plasmid construction of pT2.l, pT2.2 and pT3Plasmid pT2.1 was constructed by cloning the 0.8 kbp EcoRI-SmaTfragment from pTl into the EcoRI-SmaI site of pGEM7(-). Site-directed mutagenesis was carried out on pT2.1 to change theanticodon region from GTA to CTA, the resultant plasmid wasnamed pT2.2. The mutated 0.8 kbp fragment from pT2.2 wasrecovered by cutting with EcoRI and HindIll and religated to the0.6 kbp SmaI-EcoRI fragment from pTl. The fragment wascloned into the SmaI-HindIII site of pGEM7(-) to form pT3.EcoRlSmal EcoRl38C T .A G C TAGFIG 9: Site-Directed Mutagenesis of tyr tRNAThe two panels are: (a) DNA sequence of the anticodon region of thetyrosine tRNA before site-directed mutagenesis. The DNA sequence ofthe anticodon is GTA, which give rise to tRNAs that recognize thetyrosine codon UAC or UAU; and (b) DNA sequence of the anticodonregion of the tyrosine tRNA after site-directed mutagenesis. The DNAsequence of the anticodon is changed to CTA, which in principleshould give rise to tRNAs that recognize the amber stop codon UAG39AGTCGTTCGGGCGGCCCTCGGCTGACGGCCGTCGGTCGTCGCGTCCCCAACCCGACCCCCTACCGCCACGTCCGACCCGGAGTACGCACCCTTAAGAACCGCGACCCGCATTTTCCGACC+1ATG CAA CCA CGG GAC CTC TCC GCG CAC GCT CCC TAC GTA CCC GGC CGC GGG ACA GAGM Q P R D L S A H A P V P G R G T EGAG CTC GCC CGC GAA CTC GGA ATG GAC CCC GAG GAC CTG ACG AAA CTC TCC TCG AACE V A R E L G M D P E D L T K L S S NGAG AAC CCC CAC GGC CCG AGT CCG AAC GCG GTC GCC GCC ATC GAA GAC GCC GCG CCGE N P H G P S P K A V A A I E D A A PACC GTG AGC GTC TAC CCG AAG ACC GCC CAC ACG GAC CTG ACC GAA CGC CTC GCC GACT V S V Y P K T A H T D L T E R L A DAAG TGG CCC CTC GCA CCC CAA CAG GTG TGG GTG TCT CCC CCC GCG GAC GCC TCT ATCK W G L A P E Q V W V S P G A D G S IGAC TAC CTG ACC CGC GCG GTG CTC GAA CCG GAC GAC CGG ATT CTC GAA CCC GCG CCCD Y L T R A V L E P D D R I L E P A PGGC TTT TCG TAC TAC TCG ATG AGC GCC CGC TAC CAC CAC GGC GAC GCC GTC CAG TACG F S Y Y S M S A R Y H H G D A V Q YGAG GTG TCG AAG GAC GAC GAC TTC GAA CAG ACC GCC GAC CTC GTC CTC GAC GCC TACE V S K D D D F E Q T A D L V L D A YGAC CGC GAG CGC ATG GTC TAC CTC ACA ACG CCC CAC AAC CCC ACC GGT TCC GTG CTCD G E R M V Y L T T P H N P T G S V LCCG CGG GAG GAA CTC GTC GAA CTG GCC GAG TCG GTC GAA GAG CAC ACG CTC CTC GTCP R E E L V E L A E S V E E H T L L VGTC GAC GAG CCC TAC GGC GAG TTC GCC GAG GAG CCG TCG GCC ATC GAC CTC TTG TCGV D E A Y C E F A E E P 5 A I D L L SGAG TAC GAC AAC GTC GCG GCC CTG CGG ACG TTC TCG AAC GCG TAC GGG CTC GCC GGCE Y D N V A A L R T F S K A Y G L A GCTC CGC ATC GGC TAC GCC TGC GTG CCC GAG GCG TGG GCC GAC GCC TAC GCC CGC CTCL R I G Y A C V P E A W A D A Y A R VAAC ACG CCG TTC GCC GCC AGC GAG GTC GCC TGC CGC GCC GCG CTC GCC GCG CTC GACN T P F A A S E V A C R A A L A A L DGAC GAG GAA CAC GTC GAG AAA TCC GTC GAG TCG GCC CGG TGG TCC CGC GAC TAT CTCD E E H V E K S V E S A R W S R D Y LCGC GAA CAC CTC GAC GCG CCG ACG TGG GAA AGC GAG GGC AAC TTC GTC CTC GTC GAGR E H L D A P T W E S E G N F V L V EGTC GGC GAC GCC ACG GCC GTC ACC GAG GCC CCC CAG CGC GAG CGC GTC ATC GTC CGCV G D A T A V T E A A Q R E G V I V RCAC TCC CCC ACC TTC CCC CTG CCG GAG TGC ATC CGC CTC TCC TGC GGC ACG GAA ACCD C G S F G L P E C I R V S C G T E TCAG ACC AAG CGC GCC GTG GAC GTG CTC AAC CGC ATC GTC TCG GAG GTG CCG ACG GCGQ T K R A V D V L N R I V S E V P T ATGA GAGACGACGACACCGGCACGCCCGGCACCGGAAAGACCACGGCGACCGAGCCGGTCGCCGCCGACCTCGACCendTCGACGTGGTCCACCTCAACCGACTCGTGAAAGACGAGGFIG 10: DNA and derived amino acid sequences of the H. volcanii hisC geneThe putative promoter TTAA is identified in bold and the first tyrosinecodon (TAC) is underlined. The figure is adapted from Conover andDoolittle, 1990.40• HIsC geneScale: I I1kbFIG 11: Construction of plasmid pHC and pHC12Plasmid pHC12 was constructed by cloning the 2.7 kbp BamHI-EcoR[fragment from pT4C into the BamHI-EcoRI site of pGEM7(-). Sitedirected mutagenesis was carried out on pHC to change the 12th codonfrom TAC to TAG. The resultant plasmid is named pHC12AfihI NruI NotI413.5 Mutagenesis of the hisC geneThe first tyrosine triplet in the hisC gene occurs at codon position 12. Thiscodon was chosen for mutation to an amber stop codon because termination willoccur close to the site of initiation, giving rise to a very short polypeptide productwhich should be totally non-functional. Furthermore, a tyrosine codon was chosenbecause suppression with the altered tRNA will give rise to the wild type aminoacid sequence. Figure 12 showed the result of site-directed mutagenesis on theplasmid pHC. Oligonucleotide 5 was used to change the first tyrosine codon to anamber codon by changing the C residue at position 36 to a G. The resulting plasmidwas named pHC12.3.6 Strategy for construction of hisC(Am) strainWe wished to construct a hisC(Am) H. volcanii strain by displacing the wildtype hisC gene from WFD11 with the hisC(Am) mutant gene so that the resultantconstruct would contain only one copy of the gene. Since H. volcanii is relativelyactive in recombination, this will reduce the possibility of restoration of function byrecombination between two copies of mutant hisC genes.The scheme for the gene replacement strategy is shown in figure 13. Tointroduce the hisC(Am) gene into the genome, a non-replicating plasmid carryingthe gene with a selectable marker was constructed. The 2.7 kbp BamHI EcoRIfragment containing the mutated hisC gene from plasmid pHC12 was cloned intothe BamHI EcoRI site of pGOT1 to form pHC12M. The plasmid pGOT1 wasconstructed by Bissonette (unpublished) by cloning a 2.Okbp DNA fragment whichcontains the mevinolin resistance gene of H.volcanii into the XhoI Asp718 site atCCFIG 12: Site-directed mutagenesis of the hisC genePanel (a) shows the DNA sequence of the wild type hisCgene fromposition 31 to 42. Site-directed mutagenesis was carried out to changethe residue C to a G at position 36 so that the twelve codon is changedfrom a tyrosine to an amber stop codon. Panel (b) shows the DNAsequence of the hisC(Am) gene from position 31 to 42 after site-directedmutagenesis. The resultant plasmid is named pHC12.C TAG42CCCATGCATCCCCATGGATCaCCb43hisC (Am) geneTransform WFD1 1mevrpHC12MhisC geneSmalchromosomal DNARec+Integration into chromosomal DNA, select for mev’mevr SmalReverse recombationhisC (Am) in chromosomal DNAColonies are mev5, hisw.t hisC geneFIG 13: Strategy for constructing the his (Am) H. volcanii strain44the Fl origin of pGEM3(-). The plasmid pHC12M was then transformed into thewild type Haloferax volcanii WFD11 and selected for on mevinolin plates. Since thecircular plasmid lacks an origin for replication in H. volcanii, it was expected tointegrate into the bacterial chromosome by a single crossover event as in figure 2.Transformed cells which are mevinolin resistant will carry two copies of the hisCgene and the mevinolin resistance gene in the chromosome. The plasmid canintegrate into one of the two loci, the mevinolin gene locus, or the hisC gene locus.Ideally, 50% of the time it should integrate into the hisC locus, giving rise to atandem repeat of hisC gene flanking the mevinolin resistance gene.By releasing the antibiotic selection, the cells can undergo reverserecombination (Krebs et at, 1993) between the repeated sequences. In the case wherethe construct has gone into the mev locus, one of the mev genes (w.t. or mevr) willbe lost together with the hisC(Am) gene, giving rise to colonies which are stillprototrophic for histidine. Depending on the position of recombination within therepeats, some of the colonies will then lose resistance to mevinolin while the othersremain mevr.In the other case where the construct goes into the hisC gene locus, eithercopy of the hisC gene will be excised together with the mevinolin resistance gene,giving rise to mevinolin sensitive colonies. Half of the colonies will be his, whenthe hisC(Am) gene is lost, and the rest his. These will be H. volcanii strains whichhave lost the wild type hisC gene and now carry only one copy of the gene with anamber mutation in the first tyrosine codon. In other words, the H. volcaniihisC(Am) strain will be his and mevs.Two separate attempts were carried out to displace the wild type hisC gene45together with the antibiotic resistance gene from the genome by removing theantibiotic selection pressure. In the first trial, the pooi of transformed colonies wasgrown in rich medium until stationary phase was reached (1 week). Ten microlitresof the culture was then inoculated into another 2Oml rich medium and grown untilstationary phase was reached again. In the second trial, the transformed colonieswere grown to stationary phase in rich medium with mevinolin. Then they weretransferred to rich media without the antibiotic and grown as above through threemore dilutions. The final cultures in both trials were then spread onto rich plates toisolate individual colonies. The colonies were then tested for histidine auxotropyand mevinolin sensitivity. In trial 1, about 10% of the colonies obtained weremevinolin sensitive. In the second trial, about 50% of the colonies were mevinolinsensitive but in both attempts, no histidine auxotrophs were isolated.3.7 Test for incorporation of the pHC12M constructTo determine whether the non-replicating plasmid pHC12M had actuallyinserted into the hisC locus, Southern hybridization was carried out on genomicDNA digested with Smal obtained from 8 individual mevinolin resistanttransformants of WFD11. The hisC and mev probes used were a 2.5kbpSmal-EcoRIfragment from pHC12 (fig. 11), and a 2.Okbp Asp718-XhoI fragment from pBZL5 (Bissonette and Dennis, unpublished). Figure 14 shows the hybridization pattern ofSmal digested genomic DNA of WFD11 and the mevinolin resistant colonies, Anovel band of approximately 8 kbp in size which hybridizes to both probes wasobserved in the transformants. In addition, the hisC probe hybridized to a 4.7kbpband in both the wild type and the transformants, whereas a band of around 10 to 11kbp was observed using the mev probe. All of the 8 colonies examined showed thesame hybridization pattern. This result is consistent with integration at the hisC46Lanes: 1 2 1 2kbp11.5— 5.1— 4.74.5— 2.8—2.62.5a bFIG14: Southern hybridization to test for the incorporation of pHC12M intoWFD11Genomic DNA of wild type WFD11 digested with Smal (lane 1) andWFD11 transformants with pHC12M digested with Smal (lane 2) arefractionated by electrophoresis, transferred to nitrocellulose, andprobed with &2P labeled restricted fragment containing the hisC gene(a) or the mevinolin resistance gene (b). In(a), a band of 4.7kbp in sizewas observed in both lanes, and a band of about 8 kbp as indicated bythe arrow in lane 2. In b), a band of 10 to 11 kbp in size is observed inboth lanes, and a band of about 8 kbp is observed in lane 2 as indicatedby the arrow.47Sm ala.I IhisC gene probeSmal SmalI Ib.I Im ev probeI 7.7kbp IhisC gemI in ev gen.eFIG 15: Predicted result of Southern hybridization when the construct isintegrated into the hisC locus(a) shows the genomic DNA around the hisC locus in the wild type H. volcaniiWFD11. When the construct pHC12M is correctly integrated into the hisC locus, twocopies of the gene will be present with a mevinolin resistant gene in between(b).Southern hybridization on SmaI digested genomic DNA using hisC as probe willgive a band of undetermined size in both (a) and (b). An additional band of7.7kbp(3.Okbp vector + 2.7 kbp hisC gene + 2.0 kbp mevr gene) as indicated will beobserved in (b). This band should also hybridize to the mev probe.48locus (fig 15). However, it seems that none of the colonies had undergone genedisplacement by excising the wild type hisC gene. Therefore, I was unable toconstruct a hisC(Am) mutant.3.8 Test for natural amber suppressionOne possible reason why the construction of the hisC(Am) strain was notsuccessful could be due to the fact that a natural amber suppressor is already presentin the wild type genome. Therefore even if the hisC(Am) gene was introduced, theresultant colonies would still be his. To test this hypothesis, I transformedWR256(his, arg), a derivative of WFD11 that is auxotrophic for histidine andarginine, with plasmid p24H (hisC(Am) gene in pGOT24). Transformants wereselected on rich mevinolin plates. The resultant colonies were streakedindividually onto minimal plates with mevinolin and arginine and either with orwithout histidine to test for auxotrophy. All colonies required histidine for growth.This indicates that all the colonies were still his, therefore there appears to be nonatural amber suppressor in H. volcanii.3.9 Test for differential growth rateIn order to confirm that the result obtained in the gene replacementexperiments was not due to differential growth of his+ or his strains in the sameculture,WFD11 and WR256 were inoculated into rich media in a ratio of 1:1 in l000xdilution and grown together for 10 days until stationary phase was reached.Individual colonies from the culture were then tested on minimal plates. Out of 50colonies tested, 16 were his+, leading to the conclusion that there is no significant49difference in the growth rate of the two strains.3.10 Plasmid constructionVarious plasmids were constructed during the course of this project fortransformation and for future use. Plasmids that carry the amber suppressor or thehisC(Am) gene with antibiotic resistance marker (mevinolin or novobiocin) wereconstructed (table 3). In all cases, the amber suppressor was purified as a 1.5 kbpfragment from plasmid pT3 (fig 8), and the hisC(Am) gene as a 2.7kbp fragmentfrom pHC12 (fig 11).Various attempts to construct a plasmid which contains both the ambersuppressor and the hisC(Am) gene were carried out (table 4). The fragments wereligated together overnight and transformed into DH5cz. In all cases, notransformants were observed. A control experiment was then carried out with thepHC12-ambersuppressor construct as in table 5, ligations in T4 DNA ligase were carried out in allreactions overnight and then transformed into DH5c. Again, no colonies wereobserved when the ligation mixture was transformed into DH5z. Whentransformed into JM1O1, hundreds of colonies were observed on the plate. However,none of the colonies analyzed contained the 1.5kbp fragment.3.11 Test for amber suppression in Haloferax volcaniiSince the attempts to displace the wild type hisC gene from WFD11 withhisC(Am) gene had failed, an alternative to test whether amber suppression canoccur is to put the hisC(Am) gene into the genome of a his strain and then test for50Plasmid amber sup. hisC(Am) mevr flOVr Hvo on, remarkp24H X X X Bam/Xba/pGOT25p24T X X X Bam/Xba/pGOT24pHC12M X X Bam/Eco/pGOT1p44H X X X Bam/Xba/pGOT44p44T X X X Bam/Kpn/pGOT44pN7 X X nov fragmentfrom pMDS1O(BstYI/Nsil)in pHC12TABLE 3: Construction of plasmids that carry either the hisC(Am) gene or the putativeamber suppressor in pGOT vectorsCrosses indicate presence of the gene/fragment.vector-cut with amber suppressor-cut with hisC(Am)-cut withpGOT25-EcoRI/XbaI BamHI/ XbaI Sau3A/ EcoRIpGOT44-XbaI/ SphI BamHI/ XbaI Sau3A / SphIpGEM7-EcoRI/ XbaI BamHI/ XbaI Sau3A/ EcoRIpHC12-SmaI SmalTABLE 4: Attempts to construct a plasmid which contains both the ambersuppressor and the hisC(Am) geneIn all trials, the vector, hisC(Am) gene and the putative amber suppressorwere cut with the indicated restriction enzymes. All fragments were thenligated overnight in T4 DNA ligase and transformed into DH5a.51ColoniesControl 1 pHC12/SmaI +++Control 2 pHC12/SmaIdephosphorylatedLigation mix. pHC12/SmaIdephosphorylated+1 .5kb fragmentcontaining the amber suppressorTABLE 5: Control experiment for incorporation of the putative ambersuppressor into pHC12The plasmid pHC12 was cut with Smal, dephosphorylated andthen incubated with Smal fragments containing the ambersuppressor in T4 DNA ligase. The controls and the ligationmixture were then transformed into DH5. Hundreds ofcolonies were observed in control 1. No colony was observed inthe other two cases.52restoration of hisC gene function by putting in the putative amber suppressor. Theplasmid pHC12M (hisC(Am) and mevinolin resistance gene with no replicationorigin) was transformed into WR256 (his, argj. Transformants were designatedWR256.H12. WR256.H12 was then transformed with plasmid p44T (pGOT44containing the altered tyrosine tRNA gene). Transformants were selected onmevinolin plus novobiocin plates. Individual transformants were then tested forhistidine auxotrophy by growing restoration of hisC gene function by putting in theputative amber suppressor. The plasmid pHC12M (hisC(Am) and mevinolinresistance gene with no replication origin) was transformed into WR256 (his, argj.Transformants were designated WR256.H12. WR256.H12 was then transformedwith plasmid p44T (pGOT44 containing the altered tyrosine tRNA gene).Transformants were selected on mevinolin plus novobiocin plates. Individualtransformants were then tested for histidine auxotrophy by growing on minimalplates with arginine . All transformants required histidine for growth.Southern hybridization was carried out on genomic DNA digested with Smalto confirm the presence of the hisC(Am) gene in WR256.H12. Probes used were thehisC gene probe and the mev probe as in illustrated in figure 15. Using the hisC geneprobe, a single band of 4.7 kbp was observed in both the wild type DS2 DNA and theWR256.H12. A single band of 10 to 11 kbp was also observed in all lanes when themev gene was used as the probe (fig 17). The result showed that the hisC(Am) geneapparently did not integrate into the genome of WR256.H12, and only one copy ofthe mev gene was present. Therefore no conclusion about amber suppression in H.volcanii can be drawn yet.53Lanes: 1 2 Lanes: 1 2a bFIG 16: Southern hybridization to test for the incorporation of pHC12Minto WR256Genomic DNA of wild type WFD11 digested with SmaI (lane 1) andWR256 transformants with pHC12M digested with SmaI (lane 2) arefractionated by electrophoresis, transferred to nitrocellulose, andprobed with 32P labeled restricted fragment containing the hisC gene (a)or the mevinolin resistant gene (b). In(a), a band of 4.7kbp in size wasobserved in both lanes. In (b), a band of 10 to 11 kbp in size is observedin both lanes.54DISCUSSION4.1 THE TYROSINE TRANSFER RNAThe presence of the H.volcanii tyr tRNA gene was detected by Southernhybridization. It is present on the Smal genomic digest as a 1.5 kbp fragment. WFD11and DS2 are essentially the same, except that WFD11 lacks the plasmid pHV2.Southern hybridization indicates that the location of the tyrosine tENA gene isidentical in both genomes.A size-fractionated Haloferax volcanii genomic library constructed by cloningthe 1.4 to 1.6kbp genomic Smal fragments was constructed. The recognitionsequence of the restriction enzyme Smal is CCCGGG, and the enzyme cuts betweenthe C and G to give blunt ends. The clone containing the plasmid pTl obtained byscreening the library with oligonucleotide 1 and 2 was confirmed to contain thetyrosine tRNA gene by Southern hybridization. However, restriction mappingshowed an insert of 4.5kbp in size. The plasmid pTl is shown in fig. 6. Since bluntend ligation was carried out, three 1.5kbp fragments were cloned into thedephosphorylated Smal site of pGEM7(-) and the first Smal site is lost during thereligation process. The fragment which contains the tyrosine lENA gene lies in themiddle fragment. An EcoRI site is present at O.6kbp from the first Smal site, and is 38bp upstream of the tyrosine lENA coding sequence. The DNA sequence of thisregion is shown in fig 7. The gene is 74 bp long and agrees with the RNA sequenceof the tyrosine transfer RNA (fig 3). Putative upstream promoter elements wereidentified, which show a high sequence homology as compared with the consensussequence of upstream promoter regions of various other H. volcanii tRNA genes(fig17) (Daniels et al, 1986). In the H. volcanii tyrosine tENA gene, the upstream55Upstream region of tyrosine tRMA:+1CGCCGCGaTGA.aTTCTTARGTCT6CCC6TGGATTGA6ATTCTCTTGA CCCa, RGTA.ftftTC6AAACCCCTTTAR6AAAAATCGCCATACGA6AGAGT6CA6ACRGAACGAA. .66(b. CCAAGAAAGGAAAGTCTATTTACCCACCGGCAGTACGAGAGRTTGCRA 66(c. CTCGGARRTCGAAACGGATTAAACTATCCGCGAGAGAGGCARCAATGGAA 6CCd. GTRGTCC..ftTTCGAAA6CTTAAATTGTACCC6GACAACGGAGAGATGCGTCCGAACG6CA6GAFig 17: Comparison of halobacterial tRNA 5’-flanking regionThe upstream promoter region of tyrosine tRNA gene is comparedwith various other tRNA genes already cloned: a. Trp; b. Lys; c.Ser; d.Va!. (adapted from Daniels et al., 1986). The TTAA putative promoteris identified and shown in bold letters. Other 5’-flanking consensussequences are observed as GA rich regions (underlined). The first threenucleotides of the tRNA are also shown. The conserved sequenceblocks GAA, GAA, TTAA and GAGAGA are observed in the upstreamregion nof all the tRNA examined here. Although the distancesseparating these blocks are variable, this conservation in sequence isremarkable.56elements GAA is present at positions -44 and -38, GAGA at position -13; the TTAAbox is also observed at position -32, and is believed to be homologous to the TATAbox in eukaroytes (Hausner et at., 1991; Reiter et at, 1990; Shimmin and Dennis, 1989;Thomm and Wich, 1988). The promoter region has a very high A+T contentcompared to the rest of the genome. RNA polymerase finds and binds to thepromoter region. It unwinds the DNA helix to allow initiation of translation of thetRNAs. Since the A+T base pairs at the TTAA region are comparatively weaker thanG+C base pairs, the region is more susceptible to “melting”, or formation of an opencomplex.The end of the gene is marked by a poly(T) tract at position +75. It consists of astretch of 4 Ts, which is believed to be the termination sequence in archaebacteria(Shimmin and Dennis, 1989). Shimmin and Dennis (1989) showed by Si nucleaseprotection that the 3’ transcript end site of ENAs are located within runs of Tresidues and are often preceded by GC rich sequences. Long T clusters adjacent toyeast tRNA had been shown to efficiently terminate transcription by ENApolymerase Ill (Daniels and Hall, 1985). Since an RNA-DNA hybrid consisting ofpolyribo U and polydeoxy A is very unstable, the ENA chain will be expelled fromthe DNA duplex leading to termination of transcription. Since poly Ts are involvedin termination, there is a strong preference to avoid codon with adjacent T residuesand the TTT codon for phenylalanine is utilized sparingly in H.volcanii. On theother hand, in both non-coding regions and on the minus strand of coding regions,T runs are more prevalent.4.2 MODIFICATION PATTERNModified bases are very common in archaebacterial tRNAs. In the H. volcanii57tRNA, an unidentified nucleoside was observed from ENA sequencing at position15. This is not found in any eubacterial or eukaryotic tENAs. DNA sequencing ofthe cloned gene indicates that it comes from a guanosine. Gupta and Woese (1980)proposed that it is a modified G whose pKa is considerably higher than that of anunmodified G. It could be a 2-amino, 6-oxypyrimidine, a pseudo-cytodine, a 2-aminopurine or some hypermodified base containing an exo-cyclic amino group atthe 2-carbon position of the ring. The function of this modification is not known.An inosine can be formed by removing the NH2 group from a G or an A. Inthe tyrosine tRNA, an A was present at the corresponding position in the DNAsequence. Since modified nucleosides are constantly encountered in tENAs ofdifferent organisms, Nishimura (1979) suggests that they play an important role intRNA functions, but no real proof has been discovered yet.As in the case with eukaryotic tRNAs, the CCA terminal is added to thetRNA molecule post-transcriptionally. Transfer RNA lacking the CCA end has beenfound in B. subtitis and the archaebactium Sulfolobus solfataricus (Vold, 1985). The3’ terminal adenosine can become attached to the corresponding amino acid bycovalently bonding with the carboxyl group of the amino acid. This reaction iscatalyzed by aminoacyl-tRNA synthetase.4.3 THE hisC GENEThe hisC gene encodes histidinol phosphate aminotransferase (Conover andDoolittle, 1990), which is the eighth enzyme of the histidine biosynthetic pathway. Itcatalyzes the conversion of imidazole acetol phosphate to L-histidinol phosphate(Watsons et at, 1987).58The first tyrosine codon was chosen to be mutated to the amber stop codon sothat only a very short polypeptide will be transcribed. This reduces the possibility offorming a partially functional polypeptide product. The DNA sequence of the regionis changed from TAC to TAG, so that on the mRNA the corresponding tyrosinecodon was changed from UAC to UAG. Theoretically, it can now be recognized bythe anticodon CUA of the putative amber suppressor.4.4 GENE REPLACEMENT EXPERIMENTSTwo individual trials of the gene replacement experiment were carried out.In trial 1, only 10% of reverse recombination in the mev gene was observed.Another trial was then carried out where the process was carried out for a longerperiod of time over more generations to get a 50% mev reverse recombination rate.In both cases, 100 colonies were screened. However, no his colony was observed.This could be due to:1) A natural amber suppressor might be already present in H. volcanii so thatcolonies that did undergo gene replacement were still his since the hisC(Am) geneis suppressed.2) There is a significant difference in growth rate between his and hisstrains. During the process of growing the two strains together in rich media, thehis strains might outgrow the his strains. This effect might be able to eliminate allthe his- strains since they were grown together for more than 20 generations.3) The hisC(Am) gene is polar. The hisC(Arn) gene may affect the expression59of any genes downstream of it so that displacement of the wild type gene to leaveone copy of the hisC(Am) gene is unfavorable.In order to test whether a natural amber suppressor is already present in thesystem, the plasmid p24H (hisC(Am) gene in pGOT24) was transformed into WR256(his, argj and transformants were selected on mevinolin plates. WR256 is a hisstrain in which the hisC gene is randomly mutated. If an amber suppressor isalready present, this hisC(Am) gene will be suppressed, giving rise to his colonies.All colonies still required histidine for growth, leading to the conclusion that nonatural amber suppressor is present in H. volcanii.The test for differential growth rate of his and his- strains indicates thatwhen the two are grown together in rich media for a period of time, the ratio ofthem remains constant. Therefore the second reason is not correct since the his+strains will not be able to outgrow the his strains in the same media.However, the 50% loss in mevinolin resistance indicated that reverserecombination did take place. The result from the Southern hybridization showsthat the construct did go into the hisC gene locus. However, displacement of thewild type hisC gene did not occur. Since the gene replacement procedures werecarried out in rich media, selection on histidine prototrophs should not exist.Another Southern hybridization for the genomic DNA of the colonies after the genereplacement experiments would indicate whether one or both copies of the hisCgene was present and might give a better insight on why the experiment was notsuccessful.4.5 PLASMID CONSTRUCTION60Plasmids carrying the putative amber suppressor or the hisC(Am) gene withmevinolin resistance or novobiocin resistance and with or without replicativeorigin in H. volcanii were constructed for further use (table 3).We attempted to construct a plasmid which carries both the amber suppressorand the hisC(Am) gene. By transforming the plasmid into the his- strain WR256, wehope to test whether the modified tRNA could suppress the hisC(Am) gene (table 4).However, we were unable to put the two genes together on the same plasmid. Thereason for this failure is unclear. The genes on the plasmid pGEM7(-) should not beexpressed by E.coli. We could not explain the fact that the ligation mixturepHC12+amber suppressor fragment transformed JM1O1 but not DH5cz. Transformedplasmids obtained from JM1O1 transformants were analyzed and no 1.5kbp Smalfragment was observed. However, we were able to recover the 2.7 kbp hisC genefragment by cutting with BamHI and EcoRI, although the cutting efficiency wasmuch lower than with the original clone, indicating that some rearrangementmight have occurred.4.6 TEST FOR AMBER SUPPRESSION IN HALOFERAX VOLCANIISouthern hybridization indicates that the pHC12M construct did notrecombine into the WR256 genome since only one copy of the hisC gene waspresent. Furthermore, only one copy of the mev gene was present. Since thetransformants were selected on mevinolin plates, the mevinolin resistance genemust have replaced the wild type mev gene in WR256 by two crossover events (fig2). According to Mevarech (personal communication), the mevinolin gene has ahigh tendency to recombine into the genome to give mevinolin resistant colonies.61However, the fact that pHC12M was successfully integrated into WFD11 in allanalyzed colonies in the attempt to construct a hisC(Am)H. volcanii strain indicatesthat this approach is feasible. The problem might lie in the strain WR256 since itwas mutagenised randomly, and no further study on the strain is available. Theother antibiotic resistance gene, novobiocin, could be used in place of the mevinolinresistance gene to repeat the above experiment. In this case, the resultant flOVTcolonies should be transformed again with a plasmid that carries the mev’ markerand the putative amber suppressor. The double transformants should then be testedfor histidine auxotrophy, and hence the ability for amber suppression to occur inHaloferax volcanii.4.7 CONCLUSIONSince not much is known about the H. volcanii system, many difficultieswere encountered during the course of the project. Some further experiments,however, are possible:1. Gene replacement experiment: Southern hybridization to analyze thegenome after the reverse recombination step. This might give further insight as towhy the displacement of the wild type hisC gene was not successful.2. Test for amber suppression: The experiment can be repeated usingnovobiocin instead of mevinolin as the antibiotic selection marker in the firsttransformation since it has less tendency to recombine into the genome to formnovobiocin resistant colonies. However, novobiocin transformants take more thantwo weeks to be visible and transformation efficiency is much lower than with the62mevinolin vectors. The mev vectors can then be used to put the putative ambersuppressor into the transformants. Since the mev vector in the secondtransformation contains a H. volcanii origin, recombination of the mev gene toform mev’ colonies should be minimized.During this project, the tyrosine tRNA gene of H. volcanii was cloned andsequenced. A putative amber suppressor and a hisC gene that carries an amber stopcodon were constructed. Although the attempts to construct a hisC(Am) H. volcaniimutant failed and the test for amber suppression was inconclusive, the establishedproducts and procedures of this project should provide the basic foundation for anyfurther investigation of amber suppression in H. volcanii.63REFERENCESAbelson, J.N., Gefter, M. L., Barnett, L., Landy, A. Russell, R.L. and Smith, J.D. (1970).Mutant tyrosine transfer ribonucleic acids. J. Mol. Biol. 47, 15-28Adams, J.M. and Capecchi, M.R. (1966). N-formylmethionyl-sRNA as the initator ofprotein synthesis. Proc. Natl. Acad. Sci. USA 55,147-155Allison, D. S. and Hall, B. D. (1985). Effects of alternations in the 3’ flanking sequenceon in vivo and in vitro expression of the yeast SUP4-o tRNATYr gene. EMBO.4,2657-2664Boch. A and Kandler,O. (1985). Antibiotic sensitivity of archaebacteria, p525-544. InWoese, O.R. and Wolfe, R.S. (ed), The Bacteria-a treatise on structure andfunction, vol 8, The Archaebacteria, Academic Press, N.Y.Cabrera, J. A., Bolds, J., Shields, P.E., Havel, C.M. and Watsons, J. A. (1985).Isoprenoid synthesis in Halobacterium halobium. J. Biol. Chem. 261, 3578-3583Charlebois, R. L., Lam, W. L., Cline, S. W. and Doolittle, W. F. (1987).Characterization of pHV2 from Halobacterium volcanii and its use indemonstrating transformation of an archaebacterium. Proc. Nati. Acad. Sci.84,8530-8534Charlebois, R. L., Schalkwyk, L.C., Hofman, J.D. and Doolittle, W.D. (1991). Detailedphysical map and set of overlapping clones covering the genome of thearchaebacterium Haloferax volcanii DS2. J. Mol. Biol. 222,509-524Chowdhury, K. (1991). One step ‘miniprep’ method for the isolation of plasmidDNA. Nucleic Acid Res. 19, 2792Cline, S. W. and Doolittle, W.F. (1987). Efficient transfection of the archaebacteriumHalobacterium halobium. J. Bacteriol. 169, 1341-1344Clime, S., Schalkwyk, L.C. and Doolittle, W.F. (1989). Transformation of the64archaebacterium Halobacterium volcanii with genomic DNA. J. Bacteriol. 171,4987-4991Cohen, A., Lam, W. L., Charlebois, RI., Doolittle, W. F. and Schalkwyk, L. C. (1992).Localizing genes on the map of the genome of Haloferax volcanii, one of theArchae. Proc. Nati. Acad. Sd. 89, 1602-1606Crick, F. (1966). The genetic code. Cold Spring Harbor Symposia on QuantitativeBiology. vol 8. The Genetic Code. Cold Spring Harbor, L.I., N.Y.Daniels, L.L. and Wais, A.C. (1984). Restriction and modification of halophage S45 inHalobacterium. Curr. Microbiol. 10, 133-136Dennis, P.P. (1986). Molecular biology of archaebacteria. J. Bacteriol. 168, 471-478Dermis, P.P. (1993). The molecular biology of halophilic archaebacteria,p255-288. InVreeland, R.H. and Hochstein, L.I. (ed), The Biology of HalophilicBacteria, CRC Press.Doolittle, W.F. (1985). Genome structure in archaebacteria, p545-564. In Woese, O.R.and Wolfe, R.S. (ed), The Bacteria-a treatise on structure and function,vol 8, The Archaebacteria, Academic Press, N.Y.Eikmanns, B. and Thauer, R.K. (1984). Catalysis of an isotopic exchange between CO2and the carboxyl group of acetate by Methanosarcina barkeri grown on acetate.Arch. Microbiol. 138, 365-370Engelhardt, D.L., Webster, R.E., Wilhelm. R.C. and Zinder, N.D. (1965). In vitrostudies on the mechanism of suppression of a nonsense mutation. Proc. Nat!.Acad. Sci. 54, 1791-1797Garen, A. (1968). Sense and nonsense in the genetic code. Science 160,149-159Gesteland, R.F., Salser, W. and Bolle, A. (1967). In vitro synthesis of T4 lysozymes bysuppression of amber mutations. Proc. Natl. Acad. Sci. 58,2036-204265Goodman, H.M., Abelson J., Landy, A. Brenner, S. and Smith, J.D, (1968). Ambersuppression: a nucleotide change in the anticodon of a tyrosine transfer RNA.Nature. 217, 1019-1024Gupta, R. (1984). Halobacterium volcanii tRNAs. J. Biol. Chem. 259, 9461-9471Gupta, R. (1985). Transfer ribonucleic acids of archaebacteria. p311-343. In Woese,O.R. and Wolfe, R.S. (ed), The Bacteria-a treatise on structure and function,vol 8, The Archaebacteria, Academic Press, N.Y.Gupta, R. and Woese, CR. (1980). Unusual modification patterns in the transferribonucleic acids of archaebacteria. Curr. Microbiol. 4, 245-249Hausner, W., Gerhard, F. and Thomm, M. (1991). Control regions of an archael gene,a TATA box and an initiator element promote cell-free transcription of thetRNAVal gene of Methanococcus vannielii. J. Mol. Biol. 222, 495-508Hill, C. W. (1975). Informational suppression of missense mutation. Cell. 6,419-427Hirsh, D. (1971). Tryptophan transfer RNA as the UGA suppressor. J. Mo!. Biol. 58,439-458Hirsh, D. and Gold, L. (1971). Translation of the UGA triplet in vivo by tryptophantransfer RNA’s. J. Mol. Biol. 58, 459-468Hoagland, M.B., Stephenson, M.L., Scott, J.F., HEcht, L.I. and Zamecnik, P.C. (1957).A soluble ribonucleic acid intermediate in protein synthesis. J. Biol. Chem.231,241-257Ho!ley, R. W., Apgaz, J., Everett, G. A., Madison J.T., Marquisee, M., Merri!, S.H.,Penswich, J. R. and Zamir, A. (1965). Structure of a ribonucleic acid. Science147:1462Holmes, M. L. and Dyall-Smith, M. L. (1991). Mutations in DNA gyrase result innovobiocin resistant in halophilic archaebacteria, J. Bacteriol. 173, 642-64866Holmes, M.L., Nuttall, S.D. and Dyall-Smith, M.L. (1991). Construction and use ofHalobacterial shuttle vectors and further studies on Haloferax DNA gyrase. J.Bacteriol. 173, 3807-3813Hui, I. and Dennis. P (1985). Characteristic of the ribosomal RNA gene clusters inHalobacterium cutirubrum. J. Biol. Chem. 260,899-906Kates, M., Wassef, M.K., and Kushner, D.J. (1968). Radioisotopic studies on thebiosynthesis of the glyceryl diether lipids of Halobacterium cutirubrum. Can.J. Biochem. 46,971-977Kleckner, N. (1981). Transposable elements in prokaryotes. Ann. Rev. Genet. 15,341-404Kondo, S., Yamagishi, A. and Oshima, T. (1991). Positive selection of the uracilauxotrophs of the sulfur-dependent thermophilic archaebacteriumSulfolobus acidocaldarius by use of 5-fluoroorotic acid. J. Bacteriol. 173, 7698-7700Krebs, MY., Mollaaghababa, R. and Khorana, A.G. (1993). Gene replacement inHalobacterium halobium and expression of bacteriorhodopsin mutants. Proc.Natl. Acad. Sci. 90, 1987-1991Kunkel, T.A. (1985). Rapid and efficient site-specific mutagensis without phenotypicselection. Proc. Natl. Acad. Sci. 82, 488-492Kunkel, T.A., Roberts, J.D. and Zakour, R.A. (1987). Rapid and efficient site-specificmutagensis without phenotypic selection. Meth. Enzymol. 154,367-382Lam, W. L. and Doolittle, W. F. (1989). Shuttle vectors for the archaebacteriumHalobacterium volcanii. Proc. Natl. Acad. Sci. 86, 5478-5482Lewin, B. (1974). Gene expression, vol 1, Bacterial genomes, William Clowes & SonsLimited, LondonMevarech, M. and Werczberger, R. (1985). Genetic transfer in Halobacterium67volcanii. J. Bacteriol. 162, 461-462Mizuuchi, K., O’Dea, M. H. and Gellert, M. (1978). DNA gyrase: subunit structureand ATPase activity of the purified enzyme. Proc. Nati. Acad. Sci. 75, 5960-5963Moore, R.L. and McCarthy, B.J. (1969a). Characterization of the deoxyribonucleic acidof various strains of halophilic bacteria. J. Bacteriol. 99, 248-254Moore, R.L. and McCarthy, B.J. (1969b). Base sequence homology and renaturationstudies of the deoxyribonucleic acid of extremely halophilic bacteria. J.Bacteriol. 99, 255-262Mullakhanbhai, M.F. and Larsen, H. (1975). Halobacterium volcanii spec. nov., adead sea halobacterium with a moderate salt requirement. Arch. Microbiol.104,207-214Muriana, F. J. G., Sanchez, M.C., Rodulfo, J. D., Alvarez-Ossorio, M. C. andRelimpio, A. M. (1987). Optimization of the cell envelope of extremelyhalophilic bacteria. J. Biochem. Biophys. Meth. 14,19-28Ner, S.S., Goodin, D.B. and Smith, M. (1988). Laboratory methods: a simple andefficient procedure for generating random point mutations and for codonreplacements using mixed oligonucleotides. DNA 7,127-134Nishimura, 5. (1979). Modified nucleosides in tENA, p59-79. In Schimmel, P.R.,Soll, D. and Abelson, J.N. (ed),Transfer RNA: Structure, Properties, andRecognition, Cold Spring Harbor LaboratoryReiter, W., Hudepohl, U. and Zillig, W. (1990). Mutationalanalysis of anarchaebacterial promoter: essential role of a TATA box for transcriptionefficiency and start-site selection in vitro. Proc. Natl. Acad. Sci. 87, 9509-9513Rich, A. and RajBhandary, U.L. (1976). Transfer ENA: Molecular structure,sequence, and properties. Annu. Rev. Biochem. 45,805-86068Rosenshine and Mevarech, unpublished result cited in Dennis (1993). Themolecular biology of halophilic archaebacteria,p255-288. In Vreeland, R.H.and Hochstein, L.I. (ed), The Biology of Halophilic Bacteria, CRC Press.Russell, R. L., Abelson, J. N., Landy, A., Gefter, M. L., Brenner, S. and Smith, J. D.(1969). Duplicate genes for tyrosine transfer RNA in Escherichia Coli. J. Mol.Biol. 47, 1-13Saks, M. E., Sampsons, J. R. and Abelson, J. N. (1994). The transfer RNA identityproblem: a search for rules. Science 263, 191-197Sambrook, J., Maniatis, T. and Fritsch, E. (1982). Molecular Cloning: A LaboratoryManual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.Saunders, S. and Burke, J.F. (1990). Rapid isolation of miniprep DNA for doublestrand sequencing. Nucleic Acids Res. 18, 4948Sawadogo, M. and Dyke, M.W.V. (1991). A rapid method for the purification ofdeprotected oligonucleotides. Nucleic Acids Res. 19, 674Shimmin, L. C. and Dennis, P. (1989). Characterization of the Lii, Li, LiO and Li2equivalent ribosomal protein gene cluster of the halophilic archaebacteriumHalobacterium cutirubrum. EMBO J. 8, 1225-1235Singer, B. and Kroger, M. (1979). Participation of modified nucleosides in translationand transcription. Prog. Nucleic Acid Res. Mol. Biol. 23, 151-194Smith, J.D., Abelson, J. N., Clark, B.F. C., Goodman, H. M. and Brenner, S. (1966).Studies on amber suppressor tRNA, p479-485. In Cold Spring HarborSymposia on Quantitative Biology. vol 8. The Genetic Code. Cold SpringHarbor, L.I., N.Y.Smith, J.D., Barnett, L. Brenner, S. and Russell, R.L. (1970). More mutant tyrosinetransfer ribonucleic acids. J. Mol. Biol. 54, 1-1469Smith,M. (1986). Site Directed Mutagensis. Phil. Trans. R. Soc. Lond. A317,295-304Spratt, B.G., Hedge, P.J., te Heesen, S., Edelman, A. and Broome-Smith, K. (1986).Kanamycin-resistant vactors that are analogues of plasmids pUC8, pUC9,pEMBL8 and pEMBL9. Gene, 41,337-342Stretton, A.O.W. and Brenner, S. (1965). Molecular consequences of the ambermutation and its suppression. J. Mol. Biol. 12,456-465Thiara, A.S. and Cundliffe, E. (1988). Cloning and characterization of a DNA gyrase Bgene from Streptomyces sphaeroides that confers resistance to novobiocin.EMBO J. 7, 2255-2259Thomm, M. and Wich, G. (1987). An archaebacterial promoter lelment for stableRNA genes with homology to the TATA box of higher eukaryotes. NucleicAcids Res. 16,151-163Thomm,M., Wich, G., Brown, J.W., Frey, G., Sherf,B. and Beckler, G.S. (1989). Anarchaebacterial promoter sequence assigned by RNA polymerase bindingexperiments. Can. J. Microbiol. 35,30-35Watson, J.D., Hopkins, N.H., Roberts, J.W., Steitz, J.A. and Weiner, A.M. (1987).Molecular Biology of the Gene, 4th edition. The Benjamin/CummingsPublishing Company, Inc., Califonia.Wich, G., Jarsch, M. and Bock, A. (1984) Apparent operon for a 5S ribosomal RNAgene and for tRNA genes in the archaebacterium Methanococcus vannielii.Mol. Gen. Genet. 196, 146-151Woese. C. R. (1987) Bacterial evolution. Microbiol. Rev. 51, 221-271Woese, C.R. (1981). Archaebacteria. Sci. Amer. 244, 98-122Woese, C.R., Gutell, R., Gupta, R. and Noller, H.F. (1983). Detailed analysis of thehigher-order structure of 16S-like ribosomal ribonucleic acids. Microbiol. Rev.47, 621-669Woose, C. R., Kandler 0. and Wheelis, M.L. (1990). Towards a natural system oforganisms: Proposal for the domains Archaea, Bacteria and Eucarya. Proc.NatL Acad. Sci. 87, 4576-4579Zillig, W., Stetter, 0., Schnabel, R. and Thomm, M. (1985). DNA-dependent RNApolymerase of the archaebacteria, p499-524. In Woese, 0.R. and Wolfe, R.S.(ed), The Bacteria-a treatise on structure and function, vol 8, TheArchaebacteria, Academic Press, N.Y.70