EVALUATION OF POTASSIUM STRESS R ESPONSES AND ID EN TIFIC ATIO N OF NOVEL RT-qPCR REFERENCE GENES IN THE H A LO ARCH AEO N, HALOARCULA M A R ISM O R TU I by M atthew W. Jensen B.Sc., University o f N orthern British Columbia, 2009 THESIS SUBMITTED IN PARTIAL FULFILLM ENT OF THE REQUIREM ENTS FOR THE DEGREE OF M A STER OF SCIENCE IN M ATHEM ATICAL, CO M PU TER AND PHYSICAL SCIENCES (CHEM ISTRY) UNIVERSITY OF NORTHERN BRITISH COLUMBIA August 2012 © M atthew W. Jensen, 2012 1+1 Library and Archives Canada Bibliotheque et Archives Canada Published Heritage Branch Direction du Patrimoine de I'edition 395 Wellington Street Ottawa ON K1A0N4 Canada 395, rue Wellington Ottawa ON K1A 0N4 Canada Your file Votre reference ISBN: 978-0-494-94109-6 Our file Notre reference ISBN: 978-0-494-94109-6 NOTICE: AVIS: The author has granted a non­ exclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distrbute and sell theses worldwide, for commercial or non­ commercial purposes, in microform, paper, electronic and/or any other formats. L'auteur a accorde une licence non exclusive permettant a la Bibliotheque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par I'lnternet, preter, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats. The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. L'auteur conserve la propriete du droit d'auteur et des droits moraux qui protege cette these. Ni la these ni des extraits substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation. In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis. Conform em ent a la loi canadienne sur la protection de la vie privee, quelques formulaires secondaires ont ete enleves de cette these. W hile these forms may be included in the document page count, their removal does not represent any loss of content from the thesis. Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant. Canada ABSTRACT Growth characteristics and stress responses in the halophilic archaeon, H aloarcula marismortui, have been poorly investigated and know ledge o f the effects o f extracellular potassium concentration on halophilic growth is limited. W e report the evaluation o f cellular generation times across a range o f extracellular potassium concentrations to assess the organisms responses to extreme potassium stress. O ur results show H aloarcula marismortui exhibits an optimal generation time o f 4.19 ± 0.14 hours at an extracellular KC1 concentration o f lOOmM. This corresponds to an intracellular K f concentration o f 2.02M as determined through the use o f Induction-Coupled Plasma M ass Spectrometry. Additionally, the validation o f several candidate reference genes for use with RT-qPCR studies is reported. Five reference genes (16S rRNA, rpoB, pykA, polA , and rpoA) have been confirmed as being stably expressed in accordance with the M inimal Inform ation for the Publication o f Quantitative PCR Experim ents (M IQE Guidelines) across several unique halophilic growth conditions. TABLE OF CONTENTS Pg Abstract ii Table o f Contents iii List o f Tables vi List o f Figures vii List o f Symbols and Abbreviations x Dedication xiii Acknowledgem ent xiv Chapter One Introduction 1.1 The Archaea: Historical Perspectives Behind New Life on Earth 1 1.1.1 The Procaryote Ideology 1 1.1.2 The Birth o f the Three Domain System 3 1.2 A B rief Overview o f Archaeal Physiological Hallmarks 6 1.2.1 Archaeal M embrane Lipids 7 1.2.2 Archaeal Cell W all Composition 8 1.2.3 Archaeal D NA-Dependent RNA Polymerases 9 1.2.4 Archaeal Transfer-RNAs 10 1.2.5 Archaeal Ribosomal RNA Base M odifications 11 1.2.6 M ethanogenesis 13 1.3 Phylogenetic Structure o f the Archaeal Domain and its Kingdoms 15 1.3.1 Crenarchaeota 16 1.3.2 Euryarchaeota 17 1.3.3 Korarchaeo ta 19 1.3.4 Nanoarchaeota 20 1.3.5 Thaumarchaeota 21 1.3.6 Evidence for a Sixth, Un-Nam ed Archaeal Kingdom 22 1.4 A B rief Introduction to Haloarcula marismortui 22 1.4.1 Isolation and Growth o f the Laboratory Strain 22 iii 1.4.2 Haloarcula marismortui as a M odel Organism 23 1.4.3 Novel Photo-Active Rhodopsins 24 1.5 Objectives 25 Chapter Two Evaluation o f Potassium Stress Responses in the H alophilic Archaeon, H aloarcula marim sm ortui 2.1 Introduction 27 2.1.1 Growth Characteristics o f Halophilic Species 27 2.1.2 Ion Transport in Haloarcula m arism ortui and Related Species 28 2.1.2.1 M echanism s o f Osmoregulation 28 2.1.2.2 M aintenance o f a Potassium G radient 29 2.1.3 Study Specific Objectives 30 2.2 M ethods 31 2.2.1 Preparation o f Haloarcula m arism ortui Cell Cultures 31 2.2.2 Determination o f Cellular G eneration Times 32 2.2.3 Determination o f Cell Density 33 2.2.4 Determ ination o f Average Cell Volume 33 2.2.5 Evaluation o f Intracellular Ion Concentration 34 2.3 Results 35 2.4 Discussion 40 2.5 Conclusion 47 Chapter Three Identification of Novel RT-qPCR Reference Genes in H aloarcula marismortui. 3.1 Introduction 49 3.1.1 A B rief History o f RT-qPCR 49 3.1.2 The M IQE Guidelines 50 3.1.2.1 Sample Preparation 51 3.1.2.2 RNA Quantification and Quality A ssessm ent 52 3.1.2.3 Design o f Primers and Probes 53 3.1.2.4 Reverse Transcription: Synthesis o f cDNA 53 3.1.2.5 The qPCR Assay 54 IV 3.1.2.6 Controls 54 3.1.2.7 Analysis o f Data 55 3.1.3 Considerations for RT-qPCR Studies in Archaeal Systems 55 3.1.3.1 Archaeal Introns and Intron Processing 55 3.1.3.2 mRNA Characteristics in Archaeal Systems 57 3.1.4 Previous Studies Utilizing RT-qPCR in Archaeal Systems 58 3.1.5 Study Specific Objectives 59 3.1.5.1 Specific Consideration for RT-qPCR Studies in Haloarcula m arism ortui 60 3.2 M ethods 61 3.2.1 Identification o f Candidate Reference G enes and Design o f RT-qPCR Primers 61 3.2.2 Preparation o f Haloarcula m arism ortui Cultures 63 3.2.3 Extraction and Assessment o f RN A and Subsequent cDNA Synthesis 63 3.2.4 Optim ization o f RT-qPCR Reactions and Confirm ation o f Controls 64 3.2.5 Assessment o f RT-qPCR Efficiency 65 3.2.6 Analysis o f Candidate Reference Gene Stability 66 3.3 Results 66 3.4 Discussion 74 3.5 Conclusion 78 Chapter 4 Conclusion 4.1 Assessment o f Potassium Stress Responses in Haloarcula m arism ortui 81 4.2 Future Direction for the Evaluation o f Potassium Stress Responses in Halophilic Archaea 83 4.3 Evaluation o f Novel RT-qPCR Reference Genes in Haloarcula marismortui 84 4.4 Future Direction for the A ssessm ent o f N ovel RT-qPCR Reference Genes in Halophilic Archaea 96 4.5 References 88 Appendix 105 v List o f Tables Pg Table 1.1. Post-transcriptional base modifications found in the archaeal ribosom al RNAs. The presence (+) and absence (-) o f each m odification is indicated for each o f the 5S, 16S, and 23S rRNAs. 11 Table 3.1. Candidate reference genes selected on the basis o f prior use in RT-qPCR experimentation or probability o f uniform expression expected due to location o f gene product involvement in metabolic pathways. 61 Table 3.2. Primer pairs for qPCR assays o f subset #1 candidate reference genes. All primers were designed using the Beacon Designer 7 software package (Premier BioSoft) and supplied by Integrated DNA Technologies. 62 Table 3.3. Candidate reference genes identified as being minimally variable by Halobacterium salinarum NRC-1 micro array data obtained from the Gaggle database. 78 vi List o f Figures Pg Figure 1.1 Pyrrolysine identified in m ethylam ine m ethyltransferases o f the order M ethanosarcina. The naturally occurring am ino acid is used to charge a dedicated tRNA via pyrrolysyl-tRNA synthetase. 15 Figure 2.1. Exponential growth o f H aloarcula m arism ortui under varying extracellular potassium concentrations. Curves were constructed by plotting ODeoo against growth time. Optical density measurements w ere obtained once per generation time as determined by an experimental test curve for each condition. 23% S.W. MGM contains lOOmM KC1. Error bars are representative o f standard deviation obtained from a triplicate o f biological triplicates. Curves w ere fit using the exponential growth equation. The resulting generation times can be found in Figure 2.2. 35 Figure 2.2 Average generation times obtained from the exponential growth curves (Figure 2.1) vs. extracellular potassium concentration. Error bars represent standard error obtained from the determination o f cellular generation times via the exponential growth equation. 36 Figure 2.3. A standard curve relating cell culture density to optical density at 600nm. Cells were counted via standard haem ocytom eter as described in Section 2.2.4. A line o f best fit was applied to the curve to obtain an equation (shown on graph) relating cell culture density in cells/mL to the measured O D 6oo- Error bars indicate standard deviation. 37 Figure 2.4. Intracellular concentrations o f K + obtained via trace metal analysis with ICP-MS. Concentrations were obtained by lysing cells grown to balanced growth under 8 , 100, or 720mM KC1 in 5mL 1 % H N O 3 . The cell density standard curve (Figure 2.3) was used in conjunction with the optical density o f the culture to determine the number o f cells lysed. The num ber o f moles o f ion per cell were determined from ion concentrations obtained using ICP-MS and subsequently with the 1.509fL cellular volume described above to determine intracellular ion concentrations. 38 Figure 2.5. Intracellular concentrations o f Li+, R b f and Cs+ obtained via trace metal analysis with ICP-MS. Concentrations were obtained by lysing cells grown under lOOmM LiCl, RbCl, or CsCl in 5mL 1% H N O 3 . Alternative ion cultures were inoculated 1:100 with cells grown to balanced growth under standard conditions then incubated at 45°C until mid exponential growth was achieved. The cell density standard curve (Figure 2.3) was used in conjunction with the optical density o f the cultures to determine the number o f cells that were lysed. Concentrations obtained by ICP-MS were used to determine the number o f moles o f ion per cell, which in turn was used with the previously determined cellular volume (1.509fL) to determine intracellular ion concentrations. Intracellular ion concentrations observed with growth under 8mM KC1 is provided for a com parison o f concentrations typical o f limiting potassium conditions. 39 vii Figure 2.6. Total intracellular ion concentration o f monovalent cations exam ined in Har. marismortui after growth on lOOmM concentrations o f the alternative monovalent ions o f interest as described in Figure 2.5. Total ion concentrations are given as the sum o f all individual ion concentrations. Intracellular ion concentrations observed with growth under 8mM KC1 is provided for a com parison o f concentrations typical o f limiting potassium conditions. 39 Figure 3.1. Representative RNA purity assessm ent gel obtained by running RNA extracted from two biological replicates o f Har. marismortui cells grow n in m edia containing 720mM KC1 prior to, and after, digestion with DNase 1. Lanes: L-RNA Ladder; l-720m M KC1(1), untreated; 2-720m M KC1(2), untreated; 3-720mM KC1(1), DNase digested; 4-720mM KC1(2), D N ase digested. N um bers in parentheses behind experimental test conditions indicate biological replicate number. The disruption observed in the fragments in lane 5 was caused by a solid piece o f agarose located in the gel immediately in front o f the well. 68 Figure 3.2. M icro-Capillary Electrophoresis gels used for the evaluation o f RN A integrity as produced by B ioR ad’s Experion system using the m anufacturer’s suggested protocol. The 50bp marker is added to each sample as part o f the protocol and is used by the software to properly align all lanes to the RNA ladder. Lanes in A were not aligned by the software due to the poor resolution of the ladder. All extracted RNA samples used in B produced an RQI value o f 9 or higher (M IQE recom mended RQI - 7). All extracted R N A samples used in A, with the exception o f the 20mM KC1(1) sample, produced crisp fragments w ith little apparent degradation and were thus assumed to have an RQI value o f 7 or greater. The 20m M KC1(1) sample was re-run in B to confirm integrity. The 720mM KC1 triplicate was ran twice to confirm reproducibility. A: Lanes: L - Ladder; 1 - 45°C(1); 2 - 45°C(2); 3 45°C(3); 4 - 37°C(1); 5 - 37°C(2); 6 - 37°C(3); 7 - 55°C(1); 8 - 55°C(2); 9 - 55°C(3); 10 - 20mM KC1(1); 11 - 20mM KC1(2); 12 - 20mM KC1(3). B: Lanes: L - Ladder; 1 - 720mM KC1(1); 2 - 720mM KC1(2); 3 - 720mM KC1(3); 4 - 720mM KC1(1); 5 720mM KC1(2); 6 - 720mM KC1(3); 7 - 20mM KC1(1). Numbers in parentheses behind experimental conditions indicate biological replicate number. 69 Figure 3.3. Representative RT-qPCR standard curve constructed using the RNA polymerase beta sub-unit (rpoB) gene primers. Standards were prepared by pooling cDNA synthesized from RNA extracted from Har. marismortui cells under each test condition. Assays were run using a 1/10 serial dilution of pooled cD N A template standards ranging from lOng to 0.0 ln g o f cDNA per assay reaction covering three logarithmic steps as prescribed by the M IQE Guidelines. Standard deviations above 0.5 were considered unacceptable. 70 viii Figure 3.4. Differential expression o f the Har. marismortui RNA polym erase beta sub-unit gene (rpoB) across a range o f extracellular potassium concentrations and growth temperatures A direct com parison o f three replicate RT-qPCR assays is shown. Replicate assays were conducted on the same samples to assess if the observed variation between assays is due to assay preparation or perform ance. Expression is shown as fold change relative to the first biological replicate under standard conditions (23% S.W. MGM; 45°C-1). X-axis labels indicate first (-1) and second (-2) biological replicate for each test condition. Each bar is representative o f an average result obtained from a technical triplicate within each independent assay. Technical triplicates producing a standard deviation below 0.5 were considered acceptable. Note: The drastic change in expression between the 720mM KC1 biological replicates (rpoB-3) is due to an error in assay preparation and not a change in differential expression between assays. 71 Figure 3.5. Comparison o f RNA polym erase beta sub-unit (rpoB) expression obtained from three independent cDNA syntheses using a single RNA sample as a template. RNA was extracted from a third biological replicate o f Har. m arism ortui cells grown under standard conditions (23% S.W. M GM ; 45°C). RNA was extracted after cells were in balanced growth. cDNA synthesis was conducted using random hexameric primers and M -M uLV RT (New England Biolabs) as per m anufacturer’s recom mended protocol. Relative fold expression between cDNA samples is shown relative to 0. 72 Figure 3.6. Expression o f each candidate reference genes under each test condition relative to the average expression o f that gene across all test condition. Figure was produced using relative expression values produced by the qBasep!us 173 software suite. 73 Figure 3.7. Graphical representation o f the GeNorm M-values obtained using the qBaseplus software package173. Genes producing values o f 0.5 or lower are considered to be stably expressed and may be used as reference genes in qPCR assays as per the MIQE Guidelines. 73 Figure A .I. Scanning electron micrograph o f micro-crystalline salt structure. Crystals were formed during an attem pt to image whole Haloarcula m arism ortui cells. Cells were pelleted by centrifugation and excess media was rem oved via pipette. Cell pellets were thinly spread across an imaging disk then subm erged in liquid nitrogen for 30 seconds to solidify any rem aining liquid material. The frozen ells were immediately gold plated under vacuum for 55 seconds at 45 mA before imaging. Photograph was obtained using a Philips XL30 scanning electron microscope by Mr. Erwin Rehl, Departm ent o f Chemistry, University o f N orthern British Columbia. 105 IX List o f Sym bols and Abbreviations 23% S.W. M GM 23% Salt W ater M odified Growth M edia ATP Adenosine triphosphate BHB Bulge-Helix-Bulge motif, found at archaeal intron boundaries BLAST Basic Local Alignm ent Search Tool bp Base pairs, referring to nucleic acid bases CBF5 Centromere-Binding Factor 5 cbiG A gene encoding Cabalamin biosynthesis protein cbiH Gene encoding Precorrin-3B C l 7-methyltransferase cbiK Gene encoding Cobalt chelase thioredoxin cbiJ Gene encoding Cobalt-precorrin-6Y C5-methyltransferase cbiT Gene encoding Precorrin-8W decarboxylase cDNA Complementary Deoxyribonucleic Acid cobN A gene encoding Cobalamin biosynthesis protein Ct Threshold Cycle Cq Quantitative Cycle DLR Dynamic Linear Range DNA Deoxyribonucleic Acid DNase Deoxyribonuclease dNTP Deoxyribonucleotide triphosphate dpa Gene encoding a signal recognition particle receptor EDTA Ethylenediaminetetraacetic Acid etfB l Electron transfer flavoprotein beta sub-unit G+C Guanine + Cytosine, refers to nucleic acid base pairs gapB Gene encoding the G APDH protein GAPDH Glyceraldehyde-3-phosphate dehydrogenase Har. marismortui Halophilic archaeon native to the Dead Sea K+ Potassium Ion Kb Kilo-base pairs; l,000bp KC1 Potassium Chloride; chemical formula mA milliampere Mb Mega-base pairs; 1,000,OOObp MIQE Minimal Inform ation for publication o f Quantitative PCR Experim ents ml M illi-litre, unit o f measurem ent; volume mM Milli-molar, unit o f measurement; concentration M -M uLV RT M oloney M urine Leukem ia Virus Reverse Transcriptase MOPS 3-(N-M orpholino)-propanesulfonic acid mRNA M essenger Ribonucleic Acid M-value Statistical measure o f reference gene stability NaCl Sodium Chloride, chem ical formula NCBI National Center for Biotechnology Information Nanogram, unit o f measurement; mass No-RT No Reverse Transcription NRC-1 Sub-species o f Halobacterium salm arum NTC No-Template Control OD 600 Optical Density at a wavelength o f 600 nanometres oligo-dT Deoxythymidine oligonucleotide PCR Polymerase Chain Reaction p o lA l Gene encoding the DNA polym erase II small sub-unit polA2 Gene encoding the DNA polym erase II large sub-unit Poly(A) Poly-adenylated Poly(T) Poly-thym idinylated Pre-mRNA Pre-messenger RNA, m RNA prior to intron processing pykA Gene encoding the pyruvate kinase protein qPCR Quantitative Polymerase Chain Reaction (lacking RT step) RAP-PCR RNA Arbitrarily Primed Polymerase Chain Reaction RNA Ribonucleic Acid RNase Ribonuclease RNeasy RNA extraction kit; Qiagen corporation rpm Rotations per minute rpoA Gene encoding the RNA polym erase alpha sub-unit rpoB Gene encoding the RN A polym erase beta sub-unit RQI Relative Quality Index RT Reverse Transcription xi RTL Buffer supplied as part o f RNeasy Kit; Qiagen corporation RT-qPCR Reverse-Transcription Quantitative Polymerase Chain Reaction rRNA Ribosomal RNA SucC Gene encoding Succinyl-CoA synthetase beta chain SucD Gene encoding Succinyl-CoA synthetase alpha chain Sun Gene encoding Cytosine-C5-m ethylase Ta Annealing Temperature TAE Tris-Acetate-EDTA trkD Gene encoding a m em brane bound potassium ion transporter trkH Gene encoding a m em brane bound potassium ion transporter Trk System A group o f proteins responsible for low-affinity potassium ion transport tRNA Transfer Ribonucleic Acid tu p l Gene encoding tu p l-lik e transcriptional repressor UCSC University o f California, Santa Cruz pL M icro-litre, unit o f measurement; volume uM M icro-molar, unit o f measurement; concentration UV Ultraviolet % (v/v) Percent com position by volume zim Gene encoding CTAG modification m ethyl’s Dedication To my mother, Rachel, and father Willie, your love and encouragement has always been unwavering. You have always supported me in every possible way and stood behind me while I strive to achieve my goals. I truly can not express how grateful I am for this. To my brother, Cory, I have always appreciated your company, advice, and random visits that typically end up with me unexpectedly dropping w hatever I ’m doing at the time. Y ou have always given me tremendous support and I thank you for making sure I rarely have time for a dull moment. To my father-in-law, Tony, mother-in-law, Diane, and sisters-in-law, Em ily and M aggie, I thank you for your love and support over the past few years and thank you for welcom ing me into your family as I set out toward achieving a new goal. I m ost appreciate your support for your daughter and sister. It has made my countless nights in the lab easier knowing she has always had somebody to talk to while we go without seeing each other for days on end. Finally, to my wife, Amanda, your love and encouragement has been greater than any other. I can not put into words the gratitude I have for the support you have given me. If it w eren’t for you I may not have continued this far. I dedicate this to all o f you. I would not be w here I am today with out all o f your support. Thank you. And Cory, I ’m still not going to get a trade. Acknowledgem ent First and foremost, I would like to thank my supervisor, Dr. Andrea Gorrell, whom I am very fortunate to have been able to study under. I have com e to greatly appreciate her mentorship and am grateful for her willingness and patience in allowing me to learn from my own mistakes even when learning has occasionally become quite costly. I am also grateful to her for allowing me to make key decisions in the direction and progression o f my research. She has taught me much and has been an am azing role model. I would also like to thank my fellow graduate student, Carly Reinheimer, for her assistance with the initial planning and data collection o f this project. To two amazing undergraduate students, Scott M atlock and John Gorman, thank you for your assistance as well. Y our desire to get involved with my research and eagerness to learn what you could from me has been truly inspirational. Though the generosity o f the above four individuals was instrum ental to the completion o f this project, I also thank the other members o f the Gorrell Lab, past and present, for several years o f good conversation and fond memories. Additionally I would like to thank Dr. Sarah Gray for both sitting on my supervisory committee and providing me with the opportunity to learn RT-qPCR, a technique that I am sure will prove to be highly valuable throughout my career. I am also grateful to Jordie Fraser for taking the time to sit down and talk about the intricacies o f RTqPCR whenever I had become unsure of, or overwhelm ed with what felt like, at times, an unending project. Learning from your “ nearly militant approach”, as Sarah has at times so eloquently referred to it, has shown me the importance o f even the most intricate o f details while working with this technique. I am also grateful to Dr. Dezene H uber for sitting on m y supervisory com m ittee and providing valuable feedback about my project w henever we had the chance to talk. To Alida Hall, an individual whom I have come to regard as an amazing mentor, colleague, and friend, I would like to thank you for the sound advice you have provided me with over the past several years and appreciate your willingness to sit dow n and talk about anything and everything. To my closest friends: M eaghan, D ave and Jess, Andrew and Stephanie, Bamy, Jon and Carmen, (M )lke, Tom, Charles, Laura, Anna, Erwin, Randi, Stephane, J.D., and Darla:, as well as my future sister-in-law, Jess, and future brother-in-law, Anthony, I thank you all for your friendship and the support you have given me over the past several years. And finally, to all o f my family friends in Fort St. John whom I could always count on for support or a bed to crash on: M ike and Orlanda, Dan and Erin, Danny, Tyler and others, I have always looked forward to com ing home for weekend visits. Thank you all and to the numerous others that have com e into m y life throughout my years at UNBC. xv Chapter One Introduction “-a new scientific truth does not triumph by convincing its opponents a n d m aking them see the light, but rather because its opponents eventually die, a n d a new generation grows up that is fa m ilia r with it. ” - M ax K arl Ernst Ludwig Planck, 1858 - 1947 "Convincing ourselves was not the problem . Convincing others was. It w ould be a hard sell. F or reasons I could not understand at the time, literally a ll biologists believed in “the prokaryote. ” A n d it was not your typical scientific b e lie f - always open to question. This was dogma, unshakable doctrine! ” - Carl R. W oese, 1928 - 1.1 The Archaea: Historical Perspectives Behind New Life on Earth 1.1.1 The Procaryote Ideology The primary distinction between cellular organism s has long been the identification o f an organism as either a procaryote or a eucaryote. The phylogenetic classification system has reflected this distinction since the publication o f Roger Stanier and C.B. van N iel’s paper entitled “The Concept o f a Bacterium” 1 in 1962. This now iconic article conveyed the confusion among the microbiological disciplines that persisted due to a lack o f sufficient taxonomical classification among the cellular organisms. Stanier and van N iel1 state, “Any good biologist finds it intellectually distressing to devote his life to the study o f a group that cannot be readily and satisfactorily defined in the biological terms; and the abiding intellectual scandal o f bacteriology has been the absence o f a clear concept o f a bacterium .” The pair go on to credit their predecessor, Edouard Chatton: 1 “For a long time, biologists have intuitively recognized that the cell structure o f bacteria and blue-green algae is different from that o f other organism s, and should be characterized as “prim itive” ; but a satisfactory description o f the difference has proved remarkably elusive. The revolutionary advances in our knowledge o f cellular organization w hich have followed the introduction o f new techniques during the past 15 years have changed this situation. It is now clear that among organisms there arc two different organizational patterns o f cells, which Chatton (1937) called, with singular prescience, the eucaryotic and procaryotic type. The distinctive property o f bacteria and blue-green algae is the procaryotic nature o f their cells. It is on this basis that they can be clearly segregated from all other protists (namely, other algae, protozoa, and fungi), which have eucaryotic cells.” 1 A more recent review by Jan Sapp2 points out the work by Edouard Chatton in question, Titres et Travaux Scientifiques3, used the terms “procaryote” and “eucaryote” only once each and not with the “prescience” described by Stanier and van N iel1. Sapp2 elaborates on the more likely origins and evolution o f the procaryotic and eucaryotic super­ kingdoms found in recent taxonomical structure giving prim ary credit to Ernst Haeckel4' 5, Edwind B. Copeland6, and Herbert F. Copeland7,8. By 1938, Herbert F. Copeland, the son o f Edwind B. Copeland, pushed the progression o f this evolution toward the birth o f four natural phylogenetic kingdoms: M onera, Protista, Plantae, and Animalia2' 1. In 1959 the taxonomical model evolved further when W hittaker9 proposed the addition o f a fifth kingdom, Fungi, as part o f his own four kingdom model that classified M onera as a protist sub-kingdom. W hittaker later refined his model to the widely accepted five kingdom 2 (Animalia, Plantae, Fungi, Protista, and M onera) taxonom y10. Stanier and van N iel’s apparent over-exaggeration o f C hatton’s use o f the “procaryote” and “eucaryote” term s1, though suggested prior to W hittaker’s taxonom ic m odel10, have remained highly popular regardless o f their dichotomization o f the kingdom s through the separation o f M onera as procaryotic and the grouping o f the remaining four kingdom s as eucaryotic. 1.1.2 Birth of the Three Domain System In the early 1960’s Carl W oese, a professor o f microbiology at the U niversity o f Illinois at Urbana-Champaign, began a research program that would attem pt to identify the universal phylogenetic tree o f life by clarifying the widely unknown phylogeny w ithin the so-called procaryotic super-kingdom 11’ 12. In a recent, more personal account, W oese states his motivation behind establishing the program was to “ ...restore an evolutionary perspective/spirit to biology” while arguing that the reasoning behind the developm ent o f the procaryote/eucaryote classification system was based on a need for formal classification among cellular organisms rather than irrefutable scientific evidence13. A cknow ledging that proper criteria for the establishment o f phylogenetic relationships had not yet been clearly established, W oese began comparing the structures o f procaryotic ribosom al ribonucleic acids (rRNA) between species, specifically the 5S rRNAs, by characterizing oligomers produced by specific nuclease digests11. This was a necessary approach as com plete sequence determination was highly time consuming and the previously used approach o f assigning phylogenetic relationships on the basis o f protein sequence similarity, as had been done through much o f the eucaryotic super-kingdom, was highly ineffective when applied to the bacteria14. 3 W ith the introduction o f Fredrick Sanger’s m acrom olecular nucleic acid sequencing technique in 196515 (for which W oese personally refers to Sanger as the m ost important figure in twentieth century biology16), W oese and his post-doctoral fellow at the time, George Fox, expanded their approach to include the examination o f larger rRNA sequences. By the early 1970’s W oese and Fox had m anaged to tune the Sanger m ethod to fit their requirem ents13 and had began studying the sequences o f 16S rRNA o f a BlueGreen A lga17, the 16S rRNA o f the true bacteria (eubacteria)18, and the 18S rRNA am ong the apparent bacterial ancestors within the eucaryotic cytoplasm, referred to as the urkaryotes in phylogenetic analysis18' 19. This holistic approach to phylogenetic analysis, based on the evolution o f what has long been argued as being the oldest and m ost evolutionarily stable cellular macromolecule, the ribosome, illustrated two distinct lines o f decent between the eubacteria and the ancestral urkaryotes in addition to a third distinct lineage18. The identification o f this novel lineage lead to the suggestion o f a new kingdom within the procaryotic super-kingdom, the archaebacteria, w hich was solely represented by 18 20 ^ 1 the methanogenic bacteria at the time ’ Though the original publication was released in the October 1977 issue o f the Proceedings o f the National Academy o f Science o f the United States o f Am erica21' 22, the first m ajor public appearance o f W oese and F o x ’s novel kingdom was made on N ovem ber 3, 1977 when major newspapers around the world, including the N ew York Times, published front page articles quoting W oese’s use o f the phrase “third form o f life.” 13,22. In a personal account o f the events following the public announcement a colleague o f Carl W oese, Ralph W olfe, who was involved with aspects o f the discovery and characterization o f archaebacterial species, explains how the proposal o f the archaebacteria as a novel kingdom within the procaryotes was met with substantial resistance.22 W olfe recounts the 4 scientific communities reaction to the news as “ ...negative with disbelief and much hostility, especially among microbiologists.”22 The strong reaction even included a personal phone call from the Nobel Prize recipient, Salvatore Luria, in which W olfe was told to “ ...dissociate [himself) from this nonsense, o r... ruin [his] career!22” W oese, Fox, and their associates would not be discouraged. In 1980, Fox et al XA further reported the use o f 16S rRNA sequence analysis to identify phylogenetic relationships across m ore than 170 species thus providing the first insight into “procaryotic” phylogeny as a whole. The relationships they observed expanded on the need for the novel archaebacterial kingdom. The phylogenetic analysis revealed each included species fell within one o f three prim ary genealogical lineages14. The 16S rRNA sequences within each lineage were nearly identical to one another but differed significantly from those within each o f the other two lines o f decent. This substantially broadened the spectrum o f the new kingdom which now included the m ethanogens, halophiles, and therm oacidophiles14. The identification o f the archaebacteria and phylogenetic categorization o f the species within this new kingdom eventually culm inated a decade later with a proposal by W oese et al2i to establish a novel rank to the existing taxonomic hierarchy. This, they argued, was a necessity as neither the five kingdom taxonom ical model, nor the procaryoteeucaryote dichotomy, could accurately describe the evolutionary relationships that had been observed when comparing molecular sequencing data. The new rank, the domain, would supersede the level o f kingdom and would hold the formal suffix o f -a . W oese et al2i specifically stated the domains w ould be nam ed under the consideration that they would maintain continuity with existing names, would suggest basic characteristics for the overall group, and would avoid any suggestion that the eubacteria and archaebacteria are related to 5 one another. W ith this in mind they proposed the formation o f the dom ains Bacteria, Archaea, and Eucarya and suggested the abandonm ent o f their own term “archaebacteria” as it suggests a relationship between the bacterial and archaeal domains o f life. U nder this system the bacterial domain would consist o f the true, or eu-, bacteria and the urkaryotes while the eucaryal domain would consist o f all species formerly described as eucaryotic under the procaryotic-eucaryotic dichotomy. 1.2 A B rief Overview of Archaeal Physiological Hallmarks In order to substantiate the proposal o f the three domain system described above, the archaea would have to possess hallm ark characteristics that species in neither the bacterial nor eucaryal domains could be found to possess. These w ould be “genuine phenotypic similarities [that reflect] com mon evolutionary origin.”24 A m ong the hallmarks described below are physiological traits possessed by the archaeal dom ain in its entirety. These include, as outlined by W oese and Fox over the past several decades13, 14, 18, 23, 24, branched ether-linked lipids, specific structural characteristics o f the RNA polym erases, domain specific tRNA base modifications, and com position o f the cell wall. Archaeaspecific rRNA base modifications are also discussed as these have been found to be highly unique to several archaeal species. Though m ethanogenesis is not a physiological hallm ark o f the archaeal domain as a whole and does not apply to the research described in the following chapters, the methanogenic archaeal species are the only organisms o f their kind and possess a suite o f coenzymes exclusive to their biochemistry. M oreover, the methanogens were integral to the identification o f the Archaea as a unique dom ain o f life; therefore, methanogenesis has been briefly described. 6 1.2.1 Archaeal M embrane Lipids The first fundamental characteristic that differentiates the A rchaea from the Bacteria and Eucarya at the domain level lies in the structure o f their prim ary membrane lipids13. M any studies over previous years identified “bacterial” species that did not possess the standard ester-linked lipids typical o f the Bacteria. These species instead possessed diether-linked lipid analogues in w hich long, branched isoprenoid carbon chains are attached to glycerol via an ether linkage25'30. As Woese, Fox, and colleagues14’ ,8’ 24 continued to build the archaeal phylogenetic tree, diether-linked lipids becam e an apparent physiological characteristic o f the Archaea. Synthesis o f the isoprenoid side-chains have been shown to occur through head-to-head condensation o f isoprene units which in turn can be used in the synthesis o f acyclic, monocyclic, or bicyclical diether-linked lipids31. The condensation o f two diether lipids has been observed to occur in a sim ilar m anner to produce a glycerol-dialkyl-glercerol tetra-either lipid product30. A lthough ether-linked lipids have been identified in some eucaryal30 and bacterial30,32 species (and alternatively fatty acid side chains have been observed in some archaeal lipids30' 33) to a m inor extent, it is the distinct stereochemistry o f the carbon at position one o f the glycerol backbone o f archaeal lipids that set the Archaea apart from the Bacteria and Eucarya30. The discovery o f ether-linked lipids (prior to the identification o f the archaeal domain) lead to the suggestion that they had evolved as an adaptation to the hostile environments in which the “bacterial” species that produced them thrived26. The production o f an ether-linkage over an ester-linkage would help these organisms adapt to their typically hostile native environments while the presence o f branched isoprenoid chains would be an environmental adaptation to elevated tem peratures26. In more recent years these early suggestions have been confirmed. Lipid profile studies have illustrated changes 7 in proportion o f different ether-linked lipids in the cell m em brane under extreme pressure34 and varying growth tem peratures35. Further studies revealed cyclization o f the isoprenoid side chains yields an increased m em brane transition tem perature36 while increased unsaturation o f the side chains assists with adaptation to extreme cold37. To date, the synthesis o f the primary isoprenoids, isopentenyl diphosphate and dim ethylallyl diphosphate, and their assembly into chains with subsequent attachment to glycerol-1phosphate has been thoroughly examined; however, the biosynthetic steps behind the extensive modification (addition o f sugar to glycerol; varying levels o f desaturation among side chains) remains largely undefined (reviewed by Boucher30). Taken together, etherlinked lipids have been shown to be integral to the survival o f archaeal species in the hostile environments they tend to flourish in though much remains to be learned. 1.2.2 Archaeal Cell W all Com position Another defining physiological characteristic o f the Archaea described by W oese and F ox13, l4’ 24, with evidence from several studies38*46, is a strict lack o f diam inopim elic and muramic acids (i.e. peptidoglycan) in the outer membrane or cell wall. As our knowledge o f the archaeal domain has progressed over the past decades, it has becom e apparent that the Archaea lack a universal cell wall polym er47. Archaeal cell walls are now known to be composed o f various polym ers dependant on the species being observed and can contain lipoprotein39, sulphated48, acid45, or unm odified heteropolysaccharides, glutaminylglycan, m ethanochondroitin, pseudomurein, proteins, glycoproteins, or glycocalyx47. Rather than rigid cell walls archaeal cells tend to possess a proteinaceous exterior surface layer (S-layer) that form regular two-dim ensional arrays across the exterior o f the cell membrane47' 49. Super-structural features and chem ical com position o f archaeal S-layers and cell wall polymers are highly variable between genus and species and have recently been reviewed in great detail by Konig et a t 1. 1.2.3 Archaeal DNA-Dependent RNA Polym erases W oese13 has also stated that the characteristics o f the D NA -dependent RNA polymerases in the Archaea, though they resemble those found in the Eucarya50'53, are a unique physiological characteristic. First described by Zillig et al50 in Sulfolohus acidocaldarius, archaeal RNA polym erases are com prised o f approxim ately ten monom eric components including two larger sub-units that resemble the b and b' subunits o f the eubacteria (though the archaeal large sub-units were smaller than any previously described in the eubacteria) in addition to several sm aller com ponents50. In Archaea, however, these sub-units are arranged in a different com bination than typical o f the eubacteria and the smaller components are analogous to RNA polym erase associated proteins in the Eucarya50. M ore recent work has revealed that archaeal RNA polym erases, which are most similar to the eucaryal RNA polym erase II53, are capable o f in vitro transcription with only the eucaryal-like transcription factors TBP and TFB54. In comparison, the eucaryal RNA polymerase II requires the eucaryal hom ologs o f TBP and TFB in addition to the multimeric transcription factors TFIIE, TFIIF, and TFIIH before in vitro transcription will occur55. The X-ray crystal structures o f two RNA polymerases from the genus Sulfolobus have recently been solved56’ 57 and have revealed novel, genus specific structural components53. The structure and function o f archaeal RNA polym erases53, and archaeal transcription initiation52, including these very recent findings, have also been reviewed in detail. 9 1.2.4 Archaeal Transfer-RNAs Archaeal transfer-RNAs (tRNA) are another physiological feature W oese13 has described as defining the third domain. G enes encoding archaeal tRNAs are am ong the only known archaeal genes to possess introns58'60 which are processed out o f the immature tRNA transcript by a unique endonuclease61. Although tRNA processing is a unique physiological feature in archaeal species, it will not be reviewed here as it is thoroughly discussed with respect to considerations that must be made pertaining to introns and Reverse-Transcription Quantitative Polymerase Chain Reactions (RT-qPCR) in Chapter 3. In addition to their method o f tRNA processing, however, archaeal tRNAs possess a suite o f unique base modifications that have been reviewed by Ramesh G upta as part o f an analysis o f tRNAs in the halophilic archaeon H aloferax volcanii (previously referred to as Halobacterium volcanii)62. The most characteristic o f these modifications is the presence o f an additional pseudouridine OP) residue in the com mon-arm sequence o f each tRNA providing a sequence o f W-tP-C-G rather than the T-^P-C-G sequence com m only observed in eubacterial and eucaryal tRNAs. Some archaeal species, predom inately in the kingdom Crenarchaeota and the more novel species, Nanoarchaeum equitans (Sections 1.3.1 and 1.3.4 respectively), also possess a unique system for processing some tRNAs from the immature tRNA transcript. Several tRNA genes have been identified that are transcribed as two independent immature transcripts then spliced together via a ?ra«v-splicing mechanism to form the mature tRNA63. Although this m echanism is not solely unique to archaeal species63, the domain does appear to have a large abundance o f trans-spliced tRNAs. 10 1.2.5 Archaeal Ribosomal RNA Base M odifications A num ber o f post-transcriptional m odifications have been identified in archaeal rRNAs in the past several decades; however, the specific positions of only a select few have been placed in rRNA sequences. A ccording to the RNA modification D atabase64 there have been 15 modifications identified in archaeal rRNA to date, with the 16S rRNA being the most heavily modified (Table 1.1). Levels o f m odification vary from species to species and have been shown to change in response to cellular stress such as increased culture temperatures65. This illustrates the potential for extreme variation in rRNA m odification between species as well as w ithin a single species dependant on growth conditions. Table 1.1. Post-transcriptional base m odifications found in the archaeal1 ribosomal RNAs. The presence (+) and absence (-) o f each modification is indicated for each o f the 5S, 16S, and 23S rRNAs. Abbr. 16S Modification 5S 23S Reference + 7w-methyladenosine m 6A 65 + + Am 2'-O m ethyladenosine 65 + TV^A^-dimethyladenosine m62A 21 + + irfC 5-methylcytidine 65 + + + 2'-6>-methylcytidine Cm 6 5 ,6 6 + + + Tv-acetylcytidine ac4C 6 5 ,6 6 + ac4Cm iv-acetyl-2'-(9-m ethylcytidine 66 + TV‘-methylguanosine m2G 65 + m7G 7-methylguanosine 67 + + Gm 2'-0-m ethylguanosine 65 + + Pseudouridine Y 68 + + Urn 2 '-O-methyluridine 65 3-(3-amino-3acp3U + 69 carboxypropyl)uridine + m3U 3-methyluridine 65 + 1-methyladenosine"6 70 m'A - - - - “N ot R ep o rted in the RNA M odification D a ta b a se 64 hNo known reports o f m odification in 5S o r 16S rRNA 11 Despite the potential for variability in archaeal post-transcriptional rRNA modifications, most o f which are not evolutionarily conserved, the dim ethylation o f two adjacent adenosines (A1518 and A1519 as per Escherichia coli numbering; all modification positions to follow based on this num bering)69' 71 near the 3'-termini o f the 16S rRNA is highly conserved. Currently the only one known archaeal exception to the latter is the crenarchaeotal species, Sulfolobus solfataricus, which possesses only a single dimethylation modification in the same region65. The equally conserved enzym e72 responsible for both o f these dimethylations is known as K sgA in bacteria and D im l in archaea and eucaryotes72, 73. D im l transfers four methyl groups from four S- adenosylmethionine (SAM) molecules to the two adjacent adenosines converting them to N6,N6-dimethyladenosine. It remains unclear as to whether or not the m ethyl groups are transferred to the adenosines simultaneously or sequentially; however, it is apparent from the KsgA crystal structure73 that the target adenosines enter the enzym e active site separately and only one molecule o f SAM is bound at a time72 which suggests sequential transfers are likely. A second modification o f interest is the previously described 3-(3-amino-3carboxypropyl) uridine (acp3U)69,74 which has been shown to occur in the 16S rRNA o f the halophilic archaeon, H aloferax volcanii at position 966. This unique m odification was thought to exist in tRNA only and had never previously been observed in rRNA sequences69. The H. volcanii acp3U m odification corresponds to a universally conserved modification site, which in bacteria contains an m2G m odification directly adjacent to an m5C modification at position 967. The same site in eucaryotes is typically an evolutionarily conserved uridine residue which is commonly m odified to l-m ethyl-3-(3amino-3-carboxypropyl) pseudouridine; a structural relative to the acp3U modification69. 12 Previous studies have shown that the 3-am ino-3-carboxypropyl group is transferred from SAM to uridine when synthesized in tRN A s75; however, the mechanism behind this transfer and the function o f the resulting m odification is apparently yet to be determined. 1.2.6 M ethanogenesis The methanogens have long been regarded as an ancient lineage o f organism s that originally adapted to an early anaerobic earth20, 21. Although this would appear to be entirely plausible, more recent phylogenetic studies have suggested that the m ethanogens diverged later than previously anticipated as the Thaumarchaeotal lineage76 apparently diverged first (Section 1.3.5). N onetheless, as the first recognized archaeal organism s, the methanogenic archaea comprise several orders within the euryarchaeotal kingdom and possess a highly unique biochemistry. The w ell studied production o f methane is the primary energy yielding metabolism o f the m ethanogenic archaea and has been thoroughly reviewed by Ferry and Kastead77. As described in this review, m ethanogens utilize several cofactors, many o f which can be found in the other domains o f life while others are specific to the methanogenic archaea. The first methanogen-specific cofactor, 2-mercaptoethanesulfonic acid (Coenzyme M), was described by Taylor and W olfe in 197478 and is the smallest known enzymatic cofactor currently know n77. Additional cofactors include methanofuran, 7-m ercaptoheptanoyl-threonine phosphate (Coenzym e B), methanophenazine, 5,6,7,8-tetrahydromethanopterin, 5-hydroxybenz-imidazolyfcobam ide (Factor III), Factor 43 o, Coenzyme F 4 2 0 , and molybdopterin guanine dinucleotide77. The biological production o f methane typically occurs via one o f two prim ary metabolic pathways which involve the fermentation o f acetate or the reduction o f carbon dioxide77. In addition to the primary pathways, species within the orders M ethanosarcina and 13 Methanoccoides are capable o f producing methane through the conversion o f single carbon compounds such as methanol or m ethylam ines77. One o f the most interesting aspects o f methanogenesis has been made apparent through investigations into the m olecular biology o f methane production. These investigations have lead to the discovery o f the novel amino acid, pyrrolysine79 (Figure 1.1) which has been found to be encoded by a UAG codon and is used to charge a specific tRNA directly via the enzyme pyrrolysyl-tRNA synthetase80’ 81. In 2006, M ahapatra et a t 2 conducted a study in which they knocked out the gene encoding the dedicated pyrrolysine tRNA from M ethanosarcina acetivorans, a species capable o f utilizing several substrates for methane production and carbon assimilation including various m ethylam ines. M ahapatra et a t 2 recognized that growth o f their knockout strain on substrates that required the use o f proteins or processes requiring pyrrolysine should be lethal to the organism . W ith this in mind, growth was assessed on media containing the typical M. acetivorans substrates: methanol, acetate, monom ethylam ine, and dimethylamine. The knockout strain was capable o f growth on methanol and acetate over extended growth periods o f several years and these substrates were interchangeable; however, knockout strain growth on mono- or dimethylamine was not observed even after extended six m onth incubation periods82. Additionally, unlike the wild-type strain, the M. acetivorans knockout strain was apparently unable to assimilate nitrogen from mono- or dimethylamines w hile starved o f other nitrogen containing substrates82. M oreover, methyl transfer from m ethylam ines to Coenzyme M was not observed in the knockout strain82. Taken together, this evidence suggests pyrrolysine is required for a M ethanosarcina-specific biochemical process that transfers a methyl group from methylamines to Coenzyme M and may play a role in nitrogen liberation from these compounds under limiting conditions. 14 Figure 1.1. Pyrrolysine identified in m ethylam ine m ethyltransferases o f the order M ethanosarcina. The naturally occurring am ino acid is used to charge a dedicated tR N A via pyrrolysyl-tR N A synthetase. 1.3 Phylogenetic Structure o f the Archaeal Domain and its Kingdoms In addition to W oese and F ox’s proposal for the three domain system 23, they also suggested the formation o f the novel kingdom s Crenarchaeota and Euryarchaeota within the archaeal domain. The euryarchaeotes were to include the three m ethanogenic lineages, the extreme halophiles, the sulfate-reducing species and the therm ophiles o f the genus Thermoplasma and the Thermococcus/Pyroccocus group. Alternatively, the crenarchaeotes would include the therm oacidophiles, the extreme thermophiles, and the sulphur-dependant 23 • 23 23 83 species . Today five kingdoms (Crenarchaeota , Euryarchaeota , Korarchaeota , N anoarchaeotaA, and Thaumarchaeota16) have been identified encom passing m ore than 110 individual archaeal species for which whole genomes have been sequenced and n r annotated ' o / Q"7 . Several hundred additional species have been identified , w hether it be through m olecular ecology studies involving analysis o f environmental samples or isolation via pure culture. Presumably owing to the lack o f the higher-level taxonomic grouping o f phylum, and the lesser number o f species in each, relative to their bacterial and eucaryal counterparts, the archaeal kingdoms are com monly referred to as phyla throughout the literature; however, because they were originally proposed as such by W oese23, they will 15 continue to be referred to as kingdom s below. Each o f the five kingdoms is briefly outlined below including an introduction to a few species o f particular interest. 1.3.1 Crenarchaeota The kingdom Crenarchaeota was aptly named from the G reek word lcren o s\ meaning spring or origin23,76, which was given for the believed physiological resem blance o f the species to the ancestor o f the domain considering their tendency towards hyperthermophilicity23. The kingdom is currently comprised o f the orders Sulfolobales, Acidilobales, Desulfurococcales, and Thermoproteales76' 85, 86 from which 35 whole genomes have been sequenced76,85,86. The kingdom is home to many highly unique species that could each be discussed at length; however, for the purposes o f this introduction, only the genera Sulfolobus and Ignicoccus are included here. The genus Sulfolobus is com prised o f species that have arguably been the most studied o f the archaeal domain. The genus was first proposed in 1972 as a novel bacterial taxonomic group 42 and was reclassified as one o f the original archaeal genera when the archaeal domain was proposed by W oese and F ox14. Species within this genus are described as typically spherical cells that utilize sulphur or other simple organic compounds as prim ary growth substrates42. These species are also characterized as therm ophilic acidophiles with optimal growth typically occurring at temperatures o f 70-75°C and at pHs between 2 and 342. Both Sulfolobus acidocaldarius and Sulfolobus solfataricus can easily be referred to models for the kingdom as these species have been widely used as the “go­ to” organisms for investigations involving the Crenarchaea and have becom e instrum ental in advancing our knowledge o f the crenarchaeotal kingdom. 16 One o f the most interesting groups o f organism s within the Crenarchaea are those com prising the genus, Ignicoccus. These are arguably the most unique o f the entire archaeal domain in terms o f cellular morphology as they are the only known archaeal species that do not possess either a proteinaceous S-layer or any other cell wall polym er47, but rather a cytoplasmic membrane, variable periplasm ic space, and an outer sheath resem bling those o f gram-negative bacteria49' 88. M embers o f this genus are the only species described as obligate chemolithotrophic sulfate reducing organisms within the order D esulfurococcales%%. Cells grow either singly or in pairs and possess several flagella-like QQ QQ appendages . Additionally, at least one Ignicoccus species, Ignicoccus hospitalis , has been shown to grow in the presence o f a parasitic archaeal symbiont84 (Section 1.3.4). 1.3.2 Euryarchaeota The kingdom Eury’archaeota, was nam ed from the Greek word ‘e u ry o s\ meaning diverse76, broad, wide, or spacious23, which was given for the spectrum o f ecological niches these species fill and for the array o f metabolisms they encompass23’ 76. The kingdom is currently com prised o f the orders Archaeoglobales, Halobacteriales, M ethanobacteriales, M ethan-ocellales, Methanococcales, M ethanom icrobiales, Methanopyrales, M ethano- sarcinales, Thermococcales, and Thermoplasmatales from which 75 genom es have been sequenced76’ 85’86. As with the Crenarchaeota, Euryarchaeota is home to several species o f interest; however, not all o f these can be discussed here. Therefore, in addition to the well studied methanogenic species in Section 1.2.6, the discussion o f the euryarchaeotic species below will include two species o f personal interest: Ferroplasm a acidarmanus, chosen for the intriguing environment in which it thrives, and Haloquadratum walsbyi, chosen for its highly unique morphological features. The halophilic archaeon, H aloarcula m arismortui, is 17 discussed in greater detail below (Section 1.4) as it is the subject o f interest in the investigation detailed in the chapters to follow. The acidophilic archaeon, Ferroplasm a acidarmanus strain fe rl, was first isolated from within the biofilms o f the highly acidic, metal rich waters o f the abandoned Iron M ountain mine in northern California89. In 2003 two additional strains, M T16 and M T17 were isolated from a biooxidation pilot plant 90 o f which only the MT17 strain has been well characterized. F. acidarmanus growth can be observed across a pH range o f 0.0-2.5 with optimal growth occurring between pH 0.6 and 1.489' 91. Both the ferl and M T17 strains grow heterotrophically on yeast extract89*91. The two strains differ in that M T17 has an optimal growth temperature o f 39°C 90 whereas ferl grows optimally at 42°C89’ 92. Both strains will oxidize ferrous iron and growth w ill occur in acidic media supplem ented with pyrite (FeS 2 ) sediments and 0.02% yeast extract89,90. This oxidation o f pyrite results in acid production via the reaction F e S 2 + 14 Fe 34 8 H 2 O —>15 Fe2+ + 2 S O 4 2* + 16 H+ 93 and F. acidarmanus mediated pyrite dissolution such as this is thought to contribute to the acidic environments where these organisms are com m only located92. Iron oxidation o f this nature has also been hypothesized to be a m ajor source o f microbial energy generation in low pH environm ents92. The second species o f interest is the halophilic archaeon, Haloquadratum walsbyi. This morphologically unique organism was first described by A.E. W alsby in 1980 in w ater samples taken from hyper-saline pools near the Red Sea94; however, the species was not successfully isolated in pure culture until the middle o f the last decade95, %. These cells grow in a distinct, square morphology with dim ensions o f 2 x 2 x 0 .2 pm and have been observed in the form o f aggregate sheets up to 40 x 4 0pm 95,96. The organism also contains 18 distinct gas vesicles; a morphological feature that W alsby used to first identify the species as a micro-organism in 198094. Long referred to as “W alsby’s Square H aloarchaeon” , Hal. walsbyi was given its current name in 2007 after 16S rRNA sequencing confirm ed it as a novel species o f the family Halobacteriaceae. 1.3.3 Korarchaeota The kingdom Korarchaeota was proposed in 1996 by Bams et a / 83 after m olecular ecology studies revealed the presence o f several species o f a distinct archaeal lineage in environm ental samples obtained from hot springs in Y ellowstone National Park. The novel lineage was named from the Greek words “koros” , meaning young man, or “kore” , meaning young woman, which was given to describe the apparent early divergence o f these organisms from the larger crenarchaeotal and euryarchaeotal kingdom s83. It was immediately recognized that the K orarchaeota would not be able to be confirm ed as a novel lineage without additional studies and, preferably, with the cultivation o f a korarchaeotal species in pure culture83. As o f 2006 a pure korarchaeotal culture had not net been obtained; however, one o f the strains first described by Barns et a/*3 has been highly enriched97. To the best o f our knowledge, no single species has been isolated in pure culture to date. All known korarchaeotal strains have been identified via additional m olecular ecology studies utilizing Korarchaeota specific prim ers for the 16S rR N A 98-" . To date, a single genome has been sequenced using the enriched cultures described above97. This species, Candidatus Korarchaeum cryp to p h ilu m 1 is currently the only confirm ed organism w ithin the kingdom though evidence suggests many more exist. 19 1.3.4 Nanoarchaeota The kingdom Nanoarchaeota (the dw arf archaea) was first proposed in 2002 with the discovery o f a novel archaeal species that could only be co-cultured 84 with the hyperthermophilic crenarchaeotal species, Ignicoccus sp. strain KIN4I 84' 100 (now know n as Ignicoccus hospitalis88), which as described above, belongs to a genus that is highly unique in its own right. The new species, w hich thrives as a parasitic symbiont o f Ignicoccus, was named Nanoarcheaum equitans, m eaning “riding the fire sphere84. N. equitans is currently the only species representing Nanoarchaeota and is one o f the smallest coccoid organism s currently known with a cell diam eter o f 400nm 84 and a genome of ju st 480K b101' 102. The size o f the genome, which lacks genes for various critical metabolic pathw ays (i.e. lipid synthesis103) emphasizes the reliance o f N. equitans on its Ignicoccus host47, 101. This has been further confirmed after analysis o f lipids from both N. equitans and Ignicoccus sp. KIN4I suggests N. equitans does not synthesize its own membrane lipids, but rather obtains lipids via selective uptake from its h o st103. The small genom e also codes for several halftRNA genes which are transcribed separately before being spliced together via the highly unique trora-splicing m echanism 104"106 described above (Section 1.2.4). W ith the publication o f some phylogenetic studies that have suggested Nanoarchaeota is better grouped as a lower-level taxonomic rank w ithin the Euryarchaeota w hen considering an elevated rate o f evolution76' 102, the validity o f this kingdom is still being debated. The kingdom Nanoarchaeota and its lone occupant have been further reviewed by H uber et 1.3.5 Thaumarchaeota From the Greek word ‘thaum as', m eaning w onder76, Thaumarchaeota is the newest o f the five archaeal kingdoms. The formation o f the novel kingdom was proposed in 2008 by Brochier-Arm anet et al76 based on phylogenetic studies that used ribosom al proteins and small subunit rRNA sequences in addition to genomic core gene sequences com mon to either Euryarchaeota, Crenarchaeota, or both. These studies showed the mesophilic, ammonia-oxidizing species, Cenarchaeum sym biosum, which was previously believed to belong to the kingdom Crenarchaeota, actually branched at a node in the phylogenetic tree that placed the organism in a sister group to a clade com prised of the Euryarchaeota and Crenarchaeota kingdom s76. This would suggest the Thaumarchaeota speciated before the speciation o f the two larger kingdom s76; however, the use o f a single species to properly identify a novel kingdom was less than desirable. Several molecular ecology studies have identified a substantial number o f potential thaumarchaeotal genetic signatures in a wide variety o f m esophilic environmental samples (briefly reviewed by Spang et a /87). The large number o f detected ammonia-oxidizing archaeal species has led to suggestions that these organisms may play a substantial role in global nitrogen and carbon cycles76, 87. The isolation 108 and subsequent genome sequencing 109 o f the marine archaeon Nitrosopum ilus maritimus, followed by the isolation 110 and draft genome sequence determ ination 87 o f Nitrososphaera gargensis has helped to resolve the phylogenetic relationships within the proposed Thaumarchaeota kingdom. Additional phylogenetic analysis including these novel species has shown that the Thaumarchaeota are in fact a distinct, and deeply branching, lineage within the archaeal domain; however, the phylogenetic positioning o f the thaumarchaeal kingdom can not yet be unam biguously resolved. 21 1.3.6 Evidence for a Sixth, Un-nam ed Archaeal Kingdom A recent study utilizing the com posite genom e sequence of a currently uncultured archaeon, Candidatus Caldiarchaeum suhterraneum , has suggested the potential for an additional novel kingdom 111. Nunoura et a lU] conducted phylogenetic studies in 2011 that suggested the uncultured species belonged to a novel taxonomic group that branched deeply within the archaeal domain. The genom ic sequence was shown to contain com m onalities with the other archaeal kingdoms as well as some features typical o f the E ucaryaxn. Unfortunately, C. suhterraneum is the only species currently known to exist in this phylogenetic group and it is not yet available as a pure culture. Until additional species related to this organism can be identified via m olecular ecology studies or isolation in pure culture and included in new phylogenetic analyses, the existence of this potentially novel archaeal kingdom can not yet be confirmed. 1.4 A B rief Introduction to H aloarcula m arism ortui 1.4.1 Isolation and Growth o f the Laboratory Strain The current laboratory strain o f Haloarcula m arism ortui was isolated by Ginzburg et a ln l during the 1960’s under the nam e “Halobacterium o f the Dead Sea” . The organism was believed to have been previously identified as Halobacterium m arism ortui by ElzariVolcani during his work at the Dead Sea113; however, the strain was never deposited into a culture collection and was eventually considered lost114. The new strain isolated by Ginzburg et alu 2 was eventually deposited into a culture collection and was later described by Oren et a l 114 who identified the organism as being m orphologically sim ilar to H aloarcula californiae and as having similar properties to those first described by ElzariVolcani for Halobacterium m arism ortui; however, the new organism possessed properties 22 that were exceptions to those described for both o f the latter species leading to the proposal o f the name Haloarcula marismortui (marism ortui meaning “o f the Dead Sea”). Har. marismortui is a m otile 115 (though originally described as non-m otile114), obligatorily aerobic, chemoorganotroph that has the ability to utilize a wide variety o f compounds such as glucose, sucrose, fructose, glycerol, acetate, succinate, and m alate as sole carbon sources114. Cells have been described as pleomorphic, flat disks with measurements o f 1-2 by 2 -3pm 114 with a tendency tow ard the rod shape112. Growth occurs in media containing 1.7 to 5.1M sodium chloride (NaCl) with optimal growth occurring in concentrations o f 3.4 to 3.9M at an optimal tem perature o f 40 to 50°C114. The intracellular potassium ion concentrations within Har. marismortui have been reported to be in excess o f 3M which may be explained by previous reports suggesting that many halophilic archaeal species sequester potassium ions within the cell116. It is believed Har. m arism ortui sequesters the ion and achieves this substantially elevated internal salt concentration as a mechanism o f balancing the osmotic pressure experienced in its native hyper-saline environm ent112. 1,4.2 Haloarcula marismortui as a M odel Organism H aloarcula marismortui is arguably one o f the most important organism s to m odem molecular biology as it is widely regarded as a model for studies pertaining to the eucaryotic ribosome. Attempts to assess the structure o f the highly com plex, large ribosomal sub-unit began in the 1980s with reports o f the successful isolation o f ribosome crystals obtained using ribosomes from Har. marismortui . The native environm ent in which these cells thrived suggested their proteins and nucleic acids should be highly stable in high-salt environments, thus m aking their ribosomes a favourable choice for attempts at 23 crystallization using solutions near saturation. As time progressed, the laboratory o f Thomas A. Steitz also began working tow ard obtaining a high resolution X -ray crystal structure o f the large ribosomal sub-unit and in 19 9 8 the group published a crystal structure with 9 A resolution118; far exceeding the previous resolution o f only 2 0A 119. S teitz’s group followed this up a year later with the release o f a structure with 5 A resolution 120 then, at the turn o f the decade, they published the current 2 . 4 A resolution crystal structure o f the large ribosomal sub-unit121. Steitz was aw arded a N obel Prize for this work in 2 0 0 9 1 1 9 and today the crystal structures obtained from Har. marismortui are widely used as structural models for the eucaryotic ribosome due to their simplistic similarities. 1.4.3 Novel Photo-Active Rhodopsins M ost halophilic archaeal species possess several m icrobial rhodopsins com prising a family o f trans-membrane proteins that utilize a photo-response to mediate ion transport as a means o f harvesting solar energy or act as receptors for phototaxis122' 123. Until recently, haloarchaeal species were believed to possess 4 unique rhodopsins124' 125. These consisted o f a photo-dependent proton pump (bacteriorhodopsin)126, a photo-dependent chloride pump (halorhodopsin)127' 128, and two sensory rhodopsins, one o f which m ediates both attractant and repellent phototaxis (sensory rhodopsin I)129’ 130 and one o f which mediates repellent-specific phototaxis (sensory rhodopsin II)123’ 131. A recent study by Fu et a lm has shown Har. marismortui to be unique from other halophiles in that it possess a novel, highly refined six-rhodopsin system. This system provides Har. marismortui with a more diverse photo-sensing ability in com parison to other halophiles as the absorption spectra o f the six rhodopsins provides a broader distribution o f wavelength maxima utilized. A photo­ driven ion transport system that contains two isochromatic rhodopsins that, as o f yet, 24 appear to be identical in function and level o f expression is also present. A dditionally, Fu et a l]2i report a sensory-like rhodopsin protein containing a shortened cytoplasm ic region o f unknown function that does not induce phototaxis and suggest this novel photo-dependent protein may play a role in photo-adaptation or regulation o f a circadian rhythm. I.5 Objectives Today our expanding knowledge base o f these unique and fascinating organism s is made clearly evident by searching the online PubM ed Central archival database (http://www.ncbi.nlm.nih.gov/pmc/) using the search term “ArchaeaP’ which returns II,00 0 related articles published since January o f 1995. A b rief survey o f these articles suggests very few, if any, studies have been conducted pertaining to the cellular response o f halophilic species to changes in extracellular ion concentrations. More specifically, there appears to be only a single publication that has investigated intracellular potassium ion concentrations in haloarchaea and this publication provided evidence that has contradicted recent, unpublished findings in our lab. Additionally, o f the studies utilizing RT-qPCR techniques as a prim ary tool for quantitative m olecular studies in archaeal species, none appear to adhere to the guidelines for the minimal information for the publication o f quantitative PCR experiments (the M IQE Guidelines; Chapter 3) as published in 2009132. A concerning lack o f confirmed RT-qPCR reference genes have been identified in archaeal species which is likely the largest contributing factor to the latter. In an attem pt to address both o f these concerns, the goal o f the research described in the following chapters is tw o­ fold: 25 1. Assess the cellular response to potassium stress in Haloarcula m arism ortui via determination o f cellular generation tim es and changes in intracellular potassium concentration across a w ide range o f extracellular potassium concentrations; and, 2. Identify multiple candidate RT-qPCR reference genes and confirm the stable expression for each across various growth conditions in order to provide the archaeal scientific com m unity w ith a starting point for reference gene identification in other species. Changes in cellular generation tim e is a key indicator o f cellular stress and will illustrate Har. m arism ortui’s ability to adapt to changes in its native environment. The additional assessment o f the intracellular ion concentrations obtained from cells grown at various extracellular potassium concentrations will not only confirm or deny previously reported intracellular concentrations, but will provide valuable inform ation pertaining to physiological changes that occur while Har. marismortui adapts to its changing environment. M oreover, this could potentially provide insight into changes in ion transporter activity and physiologically optimal ion concentrations. In addition to the latter, the identification o f novel RT-qPCR reference genes will provide the framework for further studies regarding differential gene expression in Har. marismortui. M ore specifically, this will provide the framework for an RT-qPCR-based investigation into the differential expression o f membrane bound ion transport proteins and further define the first o f the two above objectives. 26 Chapter Two Evaluation of Potassium Stress Responses in the Halophilic Archaeon, H aloarcula marismortui. 2.1 Introduction 2.1.1 Growth Characteristics o f Halophilic Species Growth characteristics o f halophilic archaeal species have been only minimally studied. These studies have a tendency tow ard changes in growth characteristics with varying temperature and, to the best o f our knowledge, have not exam ined changes in growth characteristics in response to specific ion stresses. A more recent investigation o f haloarchaeal growth kinetics by Robinson et a / 133 utilized a variety o f species o f the family Halobacteriaceae and found cellular generation times within the family vary from 1.5 to 3 hours and identified several species that have multiple temperature optima. The species most closely related to Haloarcula m arism ortui that was examined in this study was Haloarcula vallismortis, which produced an optimal cellular generation time o f 3.04 ± 0.20 hours at an optimal temperature o f 43 - 49°C 133. The extent to which Har. marismortui growth has been characterized is limited largely to the initial characterization that accompanied the proposal o f the organism as a novel species in 1990114. The focus o f the other studies pertaining to Har. marismortui growth maintain a focus o f examining substrate utilization and these publications appear to be extremely rare. A survey o f the literature has so far revealed only a single publication by Ginzburg et a lU2 stating a cellular generation time o f 5-6 hours for an unknow n Dead Sea isolate, though other such publications may exist. This isolate is described as a highly pleomorphic cell type that demonstrated a tendency toward the rod shape and produced pink coloured cultures that deepened in colour with age112. Though this description does not 27 allow for a definitive identification o f the isolate as Har. marismortui, it does describe this particular species extremely well. Several more recent publications have since cited this report as referring to Har. marismortui. 2.1.2 Ion Transport in Haloarcula m arism ortui and Related Species 2.1.2.1 M echanisms of Osm oregulation Ion transport in Har. m arism ortui has been studied to a far greater extent than the organisms growth characteristics. Ginzburg et a lxn produced what is likely the first report of ion transport in this species when they determ ined the cell w ater volum e ( 1 . 2 2 ± 0 .0 2 m m 3; 1 . 2 2 ± 0.02pL ) 112 and corresponding concentrations o f K+, N a+, and Cl’ w ithin an unknown halophile isolate from the Dead Sea (described above) using prim arily gravimetric methodologies. This investigation exam ined the intracellular concentrations across various stages o f growth and reported intracellular ion concentrations ranging from 3.7 - 5M K+, 0.5 - 3M Na+, and 2.3 - 4.2M Cl’. Ginzburg et a ln2 acknowledged that these concentrations were extreme and solutions o f 4-5M KC1 and 1-3M N aCl can not be prepared due to limitations o f solubility. They then go on to suggest that the activity o f the potassium ion is somehow limited w ithin the cells, which would allow for extremely elevated concentrations. M oreover, the perm eability o f the m em brane 134 lead them to suggest the mobility o f K+ is largely restricted w ithin the cells. Neutron scattering has been recently used to examine cell water m ovem ent in Har. marismortui and revealed a slow-moving w ater com ponent that accounts for approxim ately 76% o f the total cell w ater135. It has been suggested that this slow-moving cell w ater component is arranged around proteins within the cell as a solvation shell that may bind large amounts o f K + in order to maintain the viability o f halophilic proteins. I f this is the 28 case, it would confirm the notion made by Ginzburg et a l '12 regarding restricted K + mobility. As explained in an excellent review by O ren 136 ju st over a decade ago, m icroorganisms in all domains o f life utilize one o f two prim ary mechanisms for survival: (i) Cells sequester salts internally to concentrations that are osm otically equivalent to the extracellular salt concentrations forcing intracellular systems to adapt to, and function in, a hyper-saline environment, or; (ii) Cells maintain a low-salt intracellular environm ent and balance osmotic pressure through the synthesis o f organic solutes, such as glycerol, glycine betaine, or sucrose, eliminating the need for the adaptation o f intracellular systems. By these definitions, the observations described in the paragraphs above obviously indicate Har. marismortui, along with all other halophiles o f the order H alobacteriales126, utilizes the first o f these mechanisms for survival in their native hyper-saline environm ent. 2.1.2.2 M aintenance of a Potassium Gradient M embers o f the order H alobacteriales utilize the proton electrochemical gradient across the cell membrane to drive the expulsion o f N a+ and sequestration o f K + 136. As reviewed by O ren136, this gradient is maintained via respiratory electron transport during aerobic growth or substrate-level phosphorylation through membrane ATPases. In the case o f members o f the family H alobacteriaceae, which includes Har. marismortui, the proton gradient is generated directly via the photosensitive proton pump, bacteriorhodopsin126, lj6. The established proton gradient, as m aintained by any o f the latter mechanisms, is then used in conjunction with N a 4/H + antiporters as a primary mechanism to drive and maintain the N a+ gradient across the cell m em brane136. Additionally, the accumulation o f C f has been shown to occur via the photosensitive halorhodopsin transporter127' 128 and is also 29 believed to occur through a co-transport mechanism with N a+ m ovement back into the cell136. The accumulation o f K+ has been argued to occur via passive diffusion through a uniport system allowing for accum ulation proportional to the m agnitude o f the electrochemical potential across the cell membrane as per the N em st equation136, l37. A study o f K+ transport in the Haloarchaeon, H aloferax volcanii, has shown, however, that the intracellular concentrations o f K + observed in this organism can not be accounted for by passive processes alone and ATP hydrolysis is required to actively transport K + into the cell in order to reach the 3.6M intracellular concentrations that are m aintained by Hfx. volcanii136, 137. An ATP-regulated, low-affmity K+ transporter that is sim ilar to the Trk system found in Escherichia coli has also been docum ented136, 137. Overall, the accumulation o f intracellular K+ as a m echanism o f osmoregulation is m ore energetically favourable than the mechanism o f synthesizing or sequestering organic solutes (A TP:K + costs reviewed by O ren136); however, a trade o ff for an adaptation o f cellular processes to excessive salt concentrations is required. 2.1.3 Study Specific Objectives The primary aim o f this study is to assess the cellular responses to potassium stress in H aloarcula marismortui as a means o f addressing the lack o f growth characterization for this species. In order to achieve this goal cellular generation times need to be properly assessed. Growth was evaluated across a variety o f conditions that encom pass extreme changes in extracellular potassium concentration. Growth in the presence o f the alternative monovalent cations o f lithium, rubidium, and caesium was also evaluated. Additionally, the intracellular concentrations o f K +, Li+, Rb+, and Cs+ were determined using m odem 30 analytical techniques conducted on cells grown under the extreme extracellular potassium concentrations used for cellular generation time analysis. This was done in an attem pt to gain insight into Har. m arism ortui's ability to cope w ith ion concentration changes in its native environment and provide valuable insight into the organisms ion transport capabilities. These capabilities were further assessed via the determination o f intracellular alternative ion concentrations when potassium is at a minimal extracellular concentration and the alternative ion concentration is elevated. It is our hope that this evaluation o f growth and subsequent determination o f intracellular ion concentrations in the presence o f alternative monovalent ions m ay prove valuable while refining the known m echanism s o f halophilic osmoregulation. 2.2 M ethods 2.2.1 Preparation o f H aloarcula m arism ortui Cell Cultures Har. marismortui cells were grow n in 23% Salt W ater M odified Growth M edia (23% S.W. MGM) as described by Rodriguez-Valera et a ln8’ 139. Cultures were incubated at 45°C with constant shaking at 250rpm. Consistent lighting conditions were m aintained throughout growth to eliminate the possibility o f changes in cellular generation times due to changes in stimulation o f photosensitive m em brane proteins (bacteriorhodopsin126, halorhodopsin127,128, etc). Cells were continuously sub-cultured at m id-exponential growth (OD 600 = 0.4-0.6) as a means o f maintaining continuously doubling cultures. Once cultures had been sub-cultured no less than 3 times, cells were defined as being in “balanced growth” . M edia containing m onovalent ions alternative to potassium was prepared as per the methods out lined by Rodriguez-V alera138' 139 with modifications. H igh-purity NaCl (Fluka; 99.999% by trace metal analysis) was used in the preparation o f the initial 30% salt­ 31 w ater solution and KC1 was excluded. Y east extract and Peptone were added as described and 3.5M salt solutions containing the desired ionic salt (LiCl, RbCl or CsCl) were added with a volume that provided a lOOmM final concentration. M edia was then brought to a pH o f 7.5 using Tris-base and the m edia was adjusted to the final volume using M illi-Q H 2 0 . In the case o f the lOOmM LiCl growth media, the LiCl salt was added before the media was sterile filtered and autoclaved. For the RbCl and CsCl medias, the media w as autoclaved at a reduced volume before adding the 3.5M salt solutions, adjustment o f pH and volume, and sterile filtering. Trace quantities o f potassium in the high-purity NaCl stock leaves a K + concentration in these m edia conservatively estimated at an 8 m M minima. 2.2.2 Determination o f Cellular Generation Times Cells were grown to balanced growth as described above in standard 23% S.W. M GM as well as in m edia containing 8, 20, 220, 520, and 720mM KC1. Potassium concentrations below 8 mM were not obtainable due to trace levels o f potassium in the highpurity NaCl salts used. Cultures were grown as a biological triplicate o f technical replicates (i.e. one triplicate o f cultures per biological replicate). Cell densities for each culture were measured via spectrophotometry at a wavelength o f 600nm (ODgoo). G row th curves were constructed by measuring cell density at least once per generation time as determ ined by an experimental test culture. M easurements were averaged across all replicates then plotted against the growth time. In order to determine cellular generation times (Figure 2.1) the data were fit to the exponential portion o f the curve using the equation: A = A 0ek' W here A - O D 6oo, Ao = Initial O D 60o, k = growth constant, and t - cellular generation time. 32 2.2.3 Determination of Cell Density The cell density o f a pink, rod shaped Halobacterium that had been isolated from the Dead Sea has been previously reported as 1.20g/mL112. A s this description fits that o f Har. marismortui, the technique used in this report was adapted and repeated to confirm this result. Solutions o f NaCl and sucrose were prepared with a final N aCl concentration o f 3.5M and variable sucrose concentrations to vary densities. Cells at m id-exponential growth (O D 6oo = 0.4-0.6) and at saturation (O D 6oo =: >0.7) w ere pelleted independently by centrifugation. Pellets were resuspended in a small volum e o f 23% S.W. M GM to create a high viscosity cell suspension that was layered onto the N aCl/Sucrose solution. The layered suspensions were then centrifuged at 6,000rpm for 30 seconds. The sucrose solution that, upon visual inspection, allowed 50% o f the high viscosity suspension to travel 50% o f the height o f the solution within the microcentrifuge tube was considered to be o f equal density to the cells. Density o f NaCl/Sucrose solutions was determ ined gravimetrically. 2.2.4 Determ ination of Average Cell Volume Using the procedures for constructing a growth curve, optical densities were measured at least once every experimentally determined generation time. Cells were then diluted 50-1000 fold and counted using a standard haemocytometer. A standard curve plotting the number o f cells/mL against the optical density o f a cell culture grown under standard conditions was constructed and a line o f best fit applied (R 2 =0.997; Figure 2.2). To determine the average mass o f a single cell, the num ber o f cells in a given volum e o f culture was determined using the equation y = ( 2 x 1 0 9)x - ( 2 ><1 0 7) where y is the cellular density in terms o f number o f cells/mL o f culture and x is the optical density o f the culture. The volume o f culture was centrifuged at 13,000rpm for 5 minutes and the media was 33 removed completely by pipette. The pellet was centrifuged a second tim e and residual media was aspirated off o f the pellet. A thin residue was observed on the walls on the micro-centrifuge tube containing the pellet after aspiration so a base-line mass for this residue was obtained by aspirating an equivalent volume o f growth m edia independently. This process was completed three tim es using 500, 750, and 1250mL o f cell culture. The average cellular density was then determ ined using the obtained average cellular mass and the density equation (p=m/v). 2.2.5 Evaluation of Intracellular Ion Concentrations Cells in balanced growth in m edia containing 8 mM KC1, lOOmM KC1 (23% S.W. M GM ) 720mM KC1, lOOmM LiCl, lOOmM RbCl, and lOOmM CsCl were centrifuged at 13,000rpm for 5 minutes and all m edia was removed via pipette. Cell pellets were w ashed twice using the 8 mM KC1 media as a means o f washing excess ion away without lysing cells. Pellets were centrifuged a second time and residual m edia was rem oved via pipette. Cells were lysed in 5mL 1% nitric acid ( H N O 3 ; Milli-Q H 2 O ) and sonicated to break up small particles in the lysate. Induction coupled plasm a-m ass spectroscopy (ICP-M S; conducted by University o f N orthern British C olum bia’s Central Equipment Laboratory) was used to determine the concentration o f K , Li+ R b \ and C s+ within each cell lysate. The measured concentrations were used to determine the approxim ate quantity o f ion within a single cell using the standard curve equation for cell number/mL described above, which in turn was used to determine the intracellular concentration for each ion using the previously determined average cellular volume. 34 2.3 Results Haloarcula marismortui growth was observed across the broad range of extracellular potassium concentrations examined. Generation time assessm ent confirms the use o f our standard growth conditions (lOOmM KC1) provide an optimal K concentration and produced a generation time o f 4.19 ± 0.14 hours. Growth occurs at concentrations as low as 8 mM KC1 and as high as 720mM KC1. The cellular generation time remained highly stable across the elevated K concentrations examined; however, as the extracellular K concentration was decreased the generation time increased sharply. Generation times for Har. marismortui growth on alternative monovalent ions was not assessed as growth was observed to become increasingly slow er after each sequential sub-culture (Section 2.4). (1.7 0.t> 0 . 0 .4 0.2 0.1 0 T i m e ( H n t i rs ) Figure 2.1. Exponential growth o f H aloarcula m arism ortui under varying extracellular potassium concentrations. Curves were constructed by plotting O D 6nn against growth tim e. O ptical density m easurem ents w ere obtained once per generation time as determ ined by an experim ental test curve for each condition. 23% S.W. MGM contains lOOmM KC1. Error bars are representative o f standard deviation obtained from a triplicate o f biological triplicates. C urves were fit using the exponential grow th equation. The resulting generation tim es can be found in Figure 2.2. 35 6.5 4.5 3.5 0 100 200 300 400 500 600 700 800 [K+] (mM ) Figure 2.2 A verage generation times obtained from the exponential growth curves (Figure 2.1) vs. extracellular potassium concentration. Error bars represent standard error obtained from the determ ination o f cellular generation tim es via the exponential grow th equation. Using an adaptation o f a previously reported m ethod 112 as described above, the density o f Har. marismortui cells was determined to be 1.20g/mL and confirm s the result reported by Ginzburg et a ln2. Centrifugation o f a layered suspension o f cells resulting in 50% o f cells (by visual estimate) passing through the high-density solution occurred in a solution comprised o f 3.5M NaCl and 20% (w/v) sucrose w hich produced a scale density (determined by weighing varying volum es o f the solution) o f 1.20g/mL. Additionally, the construction o f a standard curve (Figure 2.3) relating a volume-specific cell count to the optical density has allowed for an approximation o f a cellular mass o f 1 .8 1 x l0 '12 ± 0.144*10 '12g. W hen used with the density relationship (p=m /v), these two values have revealed a cellular volume o f 1.509x 10‘15 ± 0.12*10'15L (1.509 ± 0.12 fL) which allows for the determination o f intracellular ion concentrations as described above. 36 1.40E+09 1.20E+09 ? 1.00E+09 ? 8.00E + 08 E in u e 3 6.00E +08 o u 5 y = 2 E + 0 9 x - 2E +07 R2 = 0 .9 9 6 8 4.00E + 08 2.00E + 08 O.OOE+OO 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0D 600 Figure 2.3. A standard curve relating cell culture density to optical density at 600nm. C ells w ere counted via standard haem ocytom eter as described in Section 2.2.4. A line o f best fit w as applied to the curve to obtain an equation (show n on graph) relating cell culture density in cells/m L to the m easured O D 6oo- E rror bars indicate standard deviation. Intracellular ion concentrations were determ ined by trace metal analysis o f cellular lysates using ICP-MS. Typically, w hen Har. marismortui is placed in water, cells lyse completely; however, when the cell pellets w ere lysed in 1% H N O 3 , cellular m aterials were clearly visible in the lysates. To overcome this and ensure the 5mL lysates were suitable for ICP-MS analysis, lysates were sonicated prior to sample loading. Trace m etal analysis was conducted to identify the concentrations o f K +, Li+, R b+, and Cs+ w ithin the cell lysates. The ion concentrations within the 5mL lysates were then used in conjunction w ith the cell density standard curve and cellular volum e described above to determine the intracellular concentrations for each o f the latter ions under various grow th conditions. Figure 2.4 illustrates variation in intracellular concentrations o f K+ w ith respect to changes in extracellular KC1 concentration while Figure 2.5 and 2.6 illustrate variation o f individual 37 and total ion concentrations, respectively, when cells are grown under lOOmM LiCl, RbCl, or CsCl. Sodium ion was not included in these figures as the inability to w ash cell pellets in an N a+-free buffer lead to Na* concentrations that were excessive and highly unreliable (Section 2.4). sc 2.5 C o . cl ^ U 1.5 U ~ © C3 C "S 2 w ss c: OJ o £1- u s 0.5 * 8 m M KC1 l O O m M KC1 720mM KC1 Growth Condition Figure 2.4. Intracellular concentrations o f K+ obtained via trace m etal analysis with ICP-M S. Concentrations w ere obtained by lysing cells grow n to balanced grow th under 8, 100, or 720m M KC1 in 5m L 1% H N 0 3. The cell density standard curve (Figure 2.3) w as used in conjunction with the optical density o f the culture to determ ine the num ber o f cells lysed. The num ber o f m oles o f ion p er cell w ere determ ined from ion concentrations obtained using ICP-M S and subsequently w ith the 1.509fL cellular volum e described above to determ ine intracellular ion concentrations. 38 1.6 1.4 1.2 1 s o U 0.8 s o ■[Li] □ [Rb] ■ [K] D[Cs] 0.6 0.4 0.2 0 8mM KC1 lOOmM LiCl lOOmM RbCl lOOmM CsCl Ion C oncentration Growth C onditions Figure 2.5. Intracellular concentrations o f L i+, R b + and C s ‘ obtained via trace metal analysis w ith ICP-M S. C oncentrations w ere obtained by lysing cells grow n under lOOmM LiCl, RbCl, or C sC l in 5mL 1% H N 0 3. A lternative ion cultures were inoculated 1:100 w ith cells grown to balanced growth under standard conditions then incubated at 45°C until mid exponential grow th w as achieved. The cell density standard curve (Figure 2.3) was used in conjunction with the optical density o f the cultures to determ ine the num ber o f cells that were lysed. Concentrations obtained by ICP-M S were used to determ ine the num ber o f m oles o f ion per cell, w hich in turn w as used with the previously determ ined cellular volum e (1.509fL) to determ ine intracellular ion concentrations. Intracellular ion concentrations observed w ith grow th under 8mM KC1 is provided for a com parison o f concentrations typical o f lim iting potassium conditions. 1.60 1.40 s © s 1.20 u J2 c o 1.00 (J eS 0.80 C3 3 e e 4O> 0.60 s CQ © — ©u 0.40 0.20 0.00 8mM KCI lOOmM LiCl lOOmM R bC l lOOmM CsCl Extracellular Ion and C oncentration Figure 2.6. Total intracellular ion concentration o f m onovalent cations exam ined in Har. m arism ortui after grow th on lOOmM concentrations o f the alternative m onovalent ions o f interest as described in Figure 2.5. Total ion concentrations are given as the sum o f all individual ion concentrations. Intracellular ion concentrations observed with growth under 8m M KCI is provided for a com parison o f concentrations typical o f lim iting potassium conditions. 39 2.4 Discussion Haloarcula marismortui has exhibited an optimal generation tim e o f 4.19 ± 0.14 hours when grown under standard, lOOmM KCI, conditions at 45°C. As the KCI concentration in the media was decreased to the minimally attainable 8mM concentration, the cellular generation time increased to 5.91 ± 0.60 hours. The increase in generation time is likely indicative o f an inability to sequester sufficient potassium to m aintain proper intracellular functions or a result o f changes in cellular energetics. M oving towards the opposite end o f the spectrum, as extracellular KCI was increased to 220mM the generation time increased from optimal to 4.56 ± 0.23 hours and slowed only slightly to 4.69 ± 0.33 hours at KCI concentrations o f 720mM. This suggests the dem and for intracellular K+ is likely being met and the slowed growth is potentially occurring due to ionic stress associated with the elevated total ion concentration and/or sub-optimal function o f intracellular systems due to excessive intracellular K+ concentrations. Both o f the latter explanations for this observed increase in generation time are discussed in detail below. Previous cellular generation times have been determ ined for several m em bers o f the family Halobacteriaceae and have been reported as ranging from 1.5 to 3 hours 133 with some species exhibiting generation times in excess o f 6 hours137, thus placing the results described above on par with the existing literature. Through a crude adaptation o f the centrifugation technique described by Ginzburg et a lU2, the density o f Har. m arism ortui cells were determined. This process involved visually estimating the quantity o f cells that w ere pulled through solutions com prised o f varying concentrations o f sucrose and sodium chloride during centrifugation in order to estimate the density. Since Ginzburg et a lxn com pleted this task with far greater precision, rough confirmation o f their result was only required as the identity o f the species they were 40 using was unknown. Confirm ation o f a cellular density o f 1.20g/mL allowed for the determination o f a cellular volume. Ginzburg et a lu 2 reported the volume o f their isolate as 1.22mm3 (1.22uL) which is extremely large in comparison to the 1.509nm3 (1.509fL) volume determined by the methods described here. Considering human erythrocytes (red blood cells) have been reported as having a 90fL volum e140, the volum e reported by Ginzburg et a lu 2 appears to be severely over estimated. Even though the attem pt at eliminating the mass o f the proteinaceous surface layer, proteins, and m em branes was not done here, but was done by Ginzburg et a lu2, our obtained volum e appears to be a far more reasonable estimate considering Har. m arism ortui cells are substantially sm aller than most eucaryal cells. The intracellular ion concentrations o f Na+ that were observed w ere extrem ely elevated and would not intuitively be expected with the presence o f a w ell-m aintained N a+ concentration gradient. Since the cell pellets were washed w ith the minimal potassium media prior to lyses in nitric acid, it is very possible a small excess o f m edia was held w ithin the pellet and on the pellet surface. It is extremely difficult to obtain an accurate measure o f the intracellular N a4 concentrations using the methods described here as cell pellets must be washed prior to trace metal analysis. All wash steps conducted here were completed using a high-sodium environm ent to prevent cell lysis due to abrupt changes in osmotic pressure, which is likely to skew results towards a higher observed concentration; therefore, the intracellular concentrations o f N a+ have been excluded from Figure 2.4. Although it was not used for this study, it may be possible to utilize a cell wash methodology that relies on polyethylene glycol (PEG) to maintain osm otic pressure. This would reduce the need for incorporating excess sodium chloride into the w ash buffers and may produce slightly more reliable results. 41 Additionally, the variation we observed in the calculated intracellular N a+ concentrations (data not shown) is very likely due to space-charge effects141’ 142 that can be observed when conducting ICP-M S analysis o f lighter ions at high concentrations. These effects occur with the mutual repulsion o f positively charged ions, which forces ions out o f the argon plasm a laterally within the instrument. This results in an overall low er sensitivity and occurs more noticeably at high ion concentrations. Space-charge effects are less apparent with ICP-MS analysis o f heavier atoms due to their slower overall mom entum within the argon plasma. The high concentration o f N aCl in our standard growth media, and potentially within, or on the exterior o f the cell pellets after washing, m akes ICP-MS analysis o f Na* exceedingly difficult and suggests the obtained results are highly unreliable. The intracellular K+ concentration within Har. marismortui cells have now been shown to be 2.02M under standard/optim al growth conditions (Figure 2.4). This concentration was reduced to 1.38M when grown under minimal potassium conditions (8mM KCI), which demonstrates this species ability to scavenge K+ from its environm ent in order to maintain the suitable internal salt concentration that is required for osmoregulation. As suggested above, the increase in generation time for grow th at this concentration o f extracellular KCI could be indicative o f an inability to sequester sufficient potassium to maintain the optimal function o f intracellular systems. The decreased intracellular K+ reported further supports this notion. A change in available cellular energy production could provide a second possible explanation for the observed variation in cellular generation tim e as it is highly plausible that the cell is required to spend additional energy on maintaining a steady potassium gradient. Potassium efflux (leaking; spillage) has previously been shown to occur in a reversed manner through low-affinity potassium transporters such as the Trk symport 42 system in E. coli resulting in a futile cycling o f potassium ions143. A sim ilar occurrence in the hom ologous system in Har. m arism ortui w ould mean additional energy is required to maintain a gradient through the use o f high-affm ity, ATP-driven transporters under limiting extracellular K + concentrations since efflux would occur more readily under these conditions. The increased energy requirem ent to maintain a steep K+ gradient would leave less energy available for other cellular process thus cellular generation tim e would be expected to increase. This is supported by the dramatic increase in generation tim e that was observed under limiting extracellular K+ concentrations and only a m ild increase in generation time under extremely elevated extracellular K + concentrations. The Trk system does not hydrolyze ATP to function, but rather utilizes it for regulatory purposes only (reviewed by O ren et a /136). As pointed out by O ren et a lii6, it is possible that an increased dem and for K* uptake through the Trk system could im m obilize a substantial quantity o f ATP if regulation occurs through the binding o f ATP in a transporter active site or allosteric site. This in turn would limit the concentration o f ATP available for other cellular processes. Oren et a /136 also state that, at the time o f their review on the bioenergetic aspects o f halophilism , a high-affm ity potassium transport system similar to the Kdp system found in E. co //136, 143 had not yet been identified in any o f the haloarchaeal species. A survey o f available literature suggests that although a haloarchaeal high-affmity system such as this remains elusive, its non-existence rem ains to be proven. Therefore, a high-affmity potassium transport system or significant im m obilization o f ATP by the Trk system could still potentially be the cause o f reduced energy stores available for normal intracellular processes which w ould explain the severely increased generation time observed under limiting K ' concentrations. 43 Alternatively, the intracellular K + concentration was observed to increase to 2.45M when grown in the presence o f 720mM KCI. One possibility for this increase could be that Har. marismortui is unable to regulate the internal salt concentration in response to elevated extracellular KCI concentrations and will sequester additional ions if they becom e available, thus producing elevated, sub-optim al internal salt concentrations. Because the latter energy requirement would not be expected to exist when cells are grow ing under high K+ concentrations, the argument for a change in cellular energetics due to potassium efflux is far less plausible in this case. The gradient under these high extracellular K + concentrations is only a fraction o f the strength o f the gradient formed under standard conditions and a far smaller fraction o f the strength o f the gradient form ed under limiting potassium conditions. It is therefore far more probable that the more subtle increase in generation time that was observed can be attributed to sub-optimal function o f intracellular systems due to an inability to regulate K + uptake as described above. Regardless o f which o f these suggestions is actually occurring, an additional stress, w hether intracellular, extracellular, energetic, or a combination thereof, is being exerted on the cells under these conditions resulting in an increased cellular generation time. It is also possible observed increases in generation time under elevated extracellular K+ concentrations is in response to the increased ionic stress being exerted on the cells in this media. Since extracellular N aC l was not decreased when KCI was increased, a constant total ion concentration was not maintained. Countless publications have described the numerous signalling pathways that are affected by ionic stress in both bacterial and eucaryal cells indicating ionic stress such as that placed on cells in this study may have a profound effect on cellular generation time. Further investigation o f ionic stress responses in Har. marismortui is still required. 44 Rubidium has been previously used as an aid to ion transport studies after another halophilic archaeon, Haloferax volcanii, was shown to uptake R b+ during K + starvation. This attribute has been exploited in an attempt to estimate the quantity o f permanently bound K+ within these cells by forcing a maximum quantity o f mobile K+ ions out o f the cell via exchange with R b+ i37. The publication in which this was reported suggests that 50% o f the K+ in Hfx. volcanii is exchangeable with R b+ indicating up to 1.8M concentrations o f K+ is free and non-bound while the rem aining fC (up to 1.5M) is tightly bound within the cells137. The findings outlined above appear to be agreeable with this statement and appear to confirm this finding in a second halophilic species. W hen grown in media containing lOOmM RbCl the intracellular K + concentration dropped to 0.51 M while intracellular R b h increased to 0.65M. Though this would suggest confirm ation o f the previous report, the concentration o f contaminating K+ in the NaCl used for media preparation must also be considered before these findings can be confirmed. Even with the use o f NaCl that has been assessed as 99.999% pure by trace metal analysis, potassium remains present in the stock sodium salts at concentrations 5mg/Kg N aCl and to unknow n concentrations in the majority o f other reagents used. This corresponds to a conservative estimate o f an excess o f 8mM potassium in the growth media. As this is the lowest attainable K + concentration, the possibility that the apparent exchange o f R b+ with mobile K+ ions, as observed in Figure 2.5, is actually an accumulation o f both K+ and Rb+ from the media can not be ruled out. The previous suggestion that the remaining intracellular K + concentration is non-mobile and bound within the cell (as per observations in Hfx. volcanii), though supported by the neutron scattering experiments described in section 2.1.2.1 above, can not be confirmed until a growth media can be prepared with substantially lower K+ concentrations that will allow for improved evaluation o f ion exchange. Although 45 these studies were not conducted in Har. marismortui, it does not appear likely that the authors o f that publication accounted for the presence o f contaminating potassium in their salt stocks. It is instead more likely that Hfx. volcanii and Har. m arism ortui are accumulating both K + and Rb+ without, or with minimal bias as a means o f osmoregulation. A similar result is observed when cells are grown in the presence o f caesium, albeit to a lesser extent. Figure 2.5 shows Cs+ is sequestered internally to concentrations o f 0.32M and the intracellular K+ concentration falls to 0.90M when Har. marismortui is grown in the presence o f lOOmM CsCl. This is again quite likely an unbiased accum ulation o f both ions as a means o f maintaining osmoregulation. The sim ilar chemistries, and potentially similar hydration o f these ions may allow for transport o f C s+ into the cell via established K + transporters. The ATP-regulated K + transporter that is similar to the Trk system found in Escherichia coliu6' 137 may provide additional merit to the latter statement. Previous work has shown that the Trk system, specifically the low-affinity K + uptake protein trkd, is capable o f transporting Rb+ and Cs+ in addition to K + in some bacterial species144' 14S, but only K ' or R b+ in others. The presence o f this system in members o f the family Halobacteriaceae makes the transport o f the alternative ions via this transporter quite plausible. The high chemical and size similarity o f Rb+ to K+ (as observed from the well established trends in the periodic table) explains the near identical intracellular concentrations observed o f these two ions when grown in lOOmM RbCl. The observed intracellular concentration o f Rb+ is slightly higher than that o f K + which is likely due to a larger abundance o f the ion. Alternatively, the lower concentration o f Cs+ in com parison to K+ observed when cells were grown in lOOmM CsCl can be explained by the substantially larger size o f the Cs+ ion making transport o f this ion more difficult. Similarities in the K +, Rb+, and Cs+ hydration shells could also explain the observed patterns in the uptake o f 46 these ions. Previous reports have shown that specific hydration patterns are required within the transport channel in order for potassium transport channels to accurately select K + ions as they are moved across the cell membrane in E. coli146. Therefore, even though the size o f the R b+ and C s+ ions differ from that o f the K + ion (C s+ to a far greater extent) which should be a preliminary selective bias, the potential for these ions to form the proper hydration shell after the initial stripping o f w ater from the ions at the transporter entrance could be the cause for reduced selectivity. The accumulation o f L i+ as an alternative to K + does not appear to occur when grown in lOOmM LiCl; however, Li+ concentrations observed in Figure 2.5 (0.02M ) may not be overly accurate due to the space-charge effect that occurs with lighter atoms during ICP analysis (described above). This may suggest the actual intracellular concentration could possibly be substantially higher. That being said, because a decrease in K + concentration is not observed in these cells in com parison to those solely grow n under 8mM KCI , as was observed with growth in the presence o f RbCl and CsCl (See Figure 2.5), it is unlikely Har. m arism ortui is sequestering Li+ under limiting potassium conditions. Given the available data, it seems more plausible that cells are scavenging K + from the extracellular media when grow n in the presence o f Li+ and the ion is excluded from the cell exclusively presumably due to the substantially sm aller size o f the ion. 2.5 Conclusion The resilience and adaptability o f Haloarcula m arism ortui to respond to changes in its native environment has been clearly demonstrated through the evaluation o f growth characteristics during potassium stress. Cellular generation tim es increase substantially as extracellular potassium reaches minimal concentrations and these changes in generation 47 time are reflected by changes in intracellular ion concentrations. Under minimal K + growth conditions, intracellular K + is reduced by approxim ately 30%, which is presum ably near the lower concentration limit required to maintain internal cellular functions. The drastically increased generation times observed under limiting K+ concentrations has been attributed to a decrease in energy available to maintain cellular processes due to the am ount o f additional ATP that is either being hydrolyzed by a high-affm ity potassium transport system or immobilized by the Trk system during regulation o f that transporter. Alternatively, intracellular K+ concentrations increase by nearly 25% when cells are grown on 720mM KCI. Cellular generation times rem ain steady, but sub-optimal, as K + concentrations increase toward total salt saturation. W e have suggested that the mild increase in observed generation time under these conditions is likely due to poor regulation o f K+ uptake. This is likely to result in sub-optim al operation o f cellular systems as intracellular K+ concentrations begin to affect protein folding and function, but not to a point o f complete detriment. Although total ionic stress could be the result o f the observed increase in generation time, this suggesting needs to be further evaluated. On a final note, since the dry mass o f the cell pellets was not m easured in order to account for the mass o f cellular components, the calculated cellular volum e is likely slightly overestimated. Therefore, the intracellular concentrations reported above are likely to better serve as a report o f the total concentration o f each ion bound by the cell as a whole, and should not be regarded as only the concentration o f free ions in the cell water. 48 Chapter Three Identification o f Novel RT-qPCR Reference G enes in Haloarcula m arism ortui 3.1 Introduction 3.1.1 A B rief History of RT-qPCR Reverse-transcription quantitative Polymerase Chain Reaction (RT-qPCR) is becoming an increasingly popular analytical method for the analysis o f gene expression. As the name suggests, the technique is a modification on the traditional Polym erase Chain Reaction (PCR). In 1992, Higuchi et a /147 modified the PCR in a manner that allowed for the detection o f specific double-stranded DNA am plicons w ithout opening the tube. This was accomplished by conducting the PCR in the presence o f ethidium brom ide and monitoring the increase in fluorescence via fibre optic cable and fluorometer as the reaction progressed. Higuchi et a /148 im proved this method in 1993 by utilizing a video cam era to continuously m onitor the fluorescence increase throughout the PCR. The kinetics o f ethidium bromide fluorescence accumulation were found to be directly related to the number o f DNA template copies in the reaction mixture, thus providing a basis for m odem real-time monitoring o f the PCR. In 1996, Gibson et a lU9 utilized reverse-transcription and a target-specific, dual­ labelled fluorogenic probe containing a reporter dye and a quenching dye described earlier by Livak et a /150 in a real-time mRNA quantification assay. W hen the probe was intact the reporter dye would be quenched due to the close proxim ity o f the quenching dye; however, upon probe hydrolysis via DNA polym erase 5 ’-exonuclease activity151, the quencher w ould be released from the reporter and fluorescence would be observed. This allow ed for real­ time monitoring o f a specific cDNA target sequence during PCR am plification and the 49 subsequent quantification o f a specific mRNA w ithin a sam ple149. M odem RT-qPCR probes, chem istries, methodologies, instruments, and data analysis has since been extensively reviewed by S.A. Bustin152. In 1999, Gygi et a / 153 determ ined that the num ber o f mRNA transcripts in Saccharomyces cerevisiae as obtained by RT-qPCR did not correlate to expressed protein levels. This showed that mRNA levels are not always directly related to the expression o f their translated protein counterpart and thus, one o f the m ajor pitfalls o f RT-qPCR data interpretation was identified. Bustin et a / 154 has extensively reviewed additional pitfalls pertaining to various RT-qPCR methodologies and trends that have been observed in more recent years with respect to the reporting o f RT-qPCR results. O f particular importance is the linearity o f the reverse-transcription step. As explained by Bustin et a / 154, the synthesis o f cDNA via reverse-transcription can be conducted using random, deoxythym idine oligonucleotide (oligo-dT), or target-specific prim ers, each o f which produce differing yields, variety, and specificity. The calculated mRNA copy num ber obtained by RT-qPCR can vary widely based on the chosen methodology. M any o f the remaining major pitfalls associated with this highly sensitive technique stem from the analysis o f obtained data and how it is reported. Bustin et a / 132 attempted to clarify this as described below. 3.1.2 The MIQE Guidelines Since the introduction o f RT-qPCR investigators have been utilizing a diverse range o f reagents, protocols, data analysis methods, and publication formats. As described above, there are pitfalls associated with these methodologies due to typically unacknow ledged variables throughout the RT-qPCR process154. In 2009 Bustin et a ln i observed the enormous lack o f consistency across publications using qPC R data and specifically 50 identified several recurring technical deficiencies limiting reproducibility or assay performance. These deficiencies include inadequate sample collection, preparation, quality, and storage, poor choice and optimization o f prim ers and probes for the PCR, and the generation o f potentially misleading results due to inappropriate analysis o f data. In an attempt to bring consistency to future publications utilizing RT-qPCR, Bustin et al 132 proposed a set o f guidelines for the M inim um Inform ation for publication o f Q uantitative real-time PCR Experiments (MIQE Guidelines). This docum ent clarifies term inology as it applies to the technique and the proper use o f that term inology, as well as proposing an “industry standard” for methods o f prim er design, sample preparation, RNA integrity analysis, controls, reverse-transcription, qPCR assays, data analysis, and publication o f RTqPCR results. If followed, all published data w ould follow a consistent, standard m inim um format constructed from reliable and reproducible data. The MIQE guidelines outline several specific considerations that m ust be taken into account while designing, conducting, and publishing the results o f RT-qPCR assays. These experimental and publication considerations, starting with sample preparation and ending with the analysis o f data (Section 3.1.3.1 - 3.1.2.7), are briefly outlined below as originally proposed by Bustin et al 132 : 3.1.2.1 Sample Preparation The detailed reporting o f tissue sample collection and processing is a necessity. Publications should clearly state w here tissue samples were obtained and w hether or not they were immediately processed. If samples are stored for any length o f time it is necessary to report the method, length, and conditions o f preservation. The extraction o f nucleic acid is a critical step as extraction efficiency depends on a num ber o f factors 51 including the amount o f biom ass initially processed, physiological status, and adequate sample homogenization. The details o f the method o f extraction used should also be clearly reported. The extent o f genomic DNA contamination should be tested and reported along with the threshold cut-off criteria that was used to determ ine the am ount o f tolerable contamination. Reporting o f the type o f DNase and the corresponding reaction conditions used is essential if the RNA sample was treated with such. 3.1.2.2 RNA Quantification and Quality Assessm ent Quantification o f RNA in extracted samples is critical when conducting RT-qPCR assays and it is necessary to report the methods used to measure the concentrations and RNA quality in detail. M ost common methods o f RNA quantification will produce varying results when compared to alternative methods making it difficult to com pare data obtained by one method to those obtained from another. W ith several methods available such as spectrophotometery, micro fluidic analysis, or capillary gel electrophoresis, it is recommended that only a single m ethod be used to m aintain consistency in results. The preferred method utilizes fluorescent RNA-binding dyes which are ideal for the detection o f low RNA concentrations. The quality assessm ent o f RN A templates is also an essential step and can only be bypassed when the quantity o f extracted RNA is not sufficient enough to allow for quality assessment or when extraction and RT-qPCR steps are being perform ed as a continuous, single-tube experiment. Gel electrophoresis evidence of RNA integrity should be reported at minimum. 52 3.1.2.3 Design of Primers and Probes Several in silico tools, such as BLAST, are useful aids in the design o f primers for RT-qPCR assays. Publications must provide prim er sequence information as well as an assessment o f prim er specificity. The structure o f the target nucleic acid should also be considered. Secondary structures may have an im pact on reverse-transcription and PCR efficiency. In silico nucleic acid-folding tools such as m Fold should be utilized when considering the positioning o f primers or probes on the target molecule. This folding structure data should also be made available to reviewers at the tim e o f m anuscript submission. The prim er supplier’s lot information and experimental validation criteria are also required to be provided in publications. Predicted prim er homology to unexpected targets or pseudogenes should be provided to reviewers as aligned sequences; however, direct experimentation must be conducted to validate prim er specificity during assay optimization. Additionally, although in silico tools for the prediction o f prim er annealing temperatures (Ta) exist, the optimum temperature for annealing should still obtained experimentally. Although prim er optimization is not com m only practiced, it is critical as poor optimization can have substantial effects on assay quality. 3.1.2.4 Reverse Transcription: Synthesis of cDNA Reverse transcription can introduce substantial variation in an RT-qPCR assay. A detailed description o f the reagents and protocol used in the formation o f cDNA from RNA must be provided. This description should include the quantity o f RNA that was reversetranscribed, the prim ing strategy, the type o f enzyme used, volum e and temperature o f the reaction, and duration o f the reverse-transcription step. This step should be carried out in 53 duplicate or triplicate and the total RNA concentrations should be identical across replicates. 3.1.2.5 The qPCR Assay The detailed reporting o f RT-qPCR assay inform ation is required in publications. This information must include database accession num bers for all target and reference genes, the locations o f each prim er and probe on the exon, sequences and concentrations o f each oligonucleotide, and the identity, position, and linkages o f any dyes or m odified bases used. Reaction conditions including the identity and concentration o f the polymerase, the exact compositions o f buffers, the M g2+ concentration, and the reaction volum e are also required in publications. In addition to the above investigators m ust identify the instrum ent being used, document the cycling conditions, and report the identities and m anufacturers o f single tubes, strips, or plates as these consumable items can affect thermal cycling. The method o f sealing used should also be reported when using plates. 3.1.2.6 Controls As with any scientific experimentation, the use o f controls in RT-qPCR is required. These should include no-reverse transcription controls during the reverse-transcription step as well as the use o f N TC ’s should be included on each plate or with each sample set w hen running RT-qPCR assays. Conditions for data rejection should be established with these controls in mind. Positive controls will allow for monitoring o f variation between assays and must be used when calibration curves are not being conducted for each run. Investigators must be aware that m ost quantitative RNA data are relative rather than absolute. Standardization o f results is required and can be achieved through the use o f 54 reference genes. The assessm ent o f the validity o f experimental results m ust consider the relative quantification reference and whether or not it is appropriate. 3.1.2.7 Analysis of Data In order for RT-qPCR characteristics must be data assessed. to be These considered characteristics viable, include assay perform ance PCR efficiency, determination o f the linear dynam ic range o f the RT-qPCR assay via the construction o f standard curves, and determ ination o f the limit o f detection. O nce perform ance characteristics are established it is important to recognize appropriate Threshold Cycle values ( C t ) and normalize results in an appropriate manner. Helpful inform ation on all o f the latter can be found in the reviews by Bustin 152 and Bustin et a l 154 m entioned above. 3.1.3 Considerations for R T-qPCR Studies in Archaeal Systems 3.1.3.1 Archaeal Introns and Intron Processing In order to conduct RT-qPCR studies the characteristics o f archaeal RNAs must be considered. The Archaea lack nuclei and their genome and gene structure are most like that o f the Bacteria; however, their DNA replication, transcription, and translation m echanism s are more similar to the Eucarya (reviewed by Klug et a l155). As in the other two domains, introns have been found w ithin exon-coding regions o f archaeal genes. Introns are transcribed with the pre-mRNA transcript by an RNA polym erase most sim ilar to the RNA polymerase II and III found in the eucaryotic domain o f life (reviewed by Langer et a /51 and Hirata et a l5i) before being processed out o f the m ature transcript by endonuclease activity61. Though Group I and Group III introns are yet to be detected, Group II introns have been identified only in the M ethanosarcinaceae which fall into the kingdom 55 E urya rch a eo ta 56. Several rRNA and tRNA genes across the archaeal domain, as well as a single protein-coding gene in a small handful o f archaeal species60, have been found to contain small introns o f 14 to 106 nucleotides which appear to be related to the eucaryotic nuclear tRNA introns155. As in the other domains, archaeal introns are rem oved by endonuclease excision after transcription155. Many archaeal species have introns in their tRNA and rRNA genes, all o f which form a conserved secondary structure com prised o f a small bulge loop followed by a short helix and a second bulge loop on the opposite side o f the base paired intron sequence. Endonuclease excision occurs at a site specific location w ithin either bulge loop o f these bulge-helix-bulge (BHB) m otifs and mutations in which these locations are removed or made inaccessible prevent splicing from occurring entirely61. The addition o f nucleotides to the helix between the bulge loops causing them to be further separated from one another results in reduced cleavage efficiency; however, cleavage accuracy remains intact61. Slippage in the cleavage sites will occur, however, w hen an additional nucleotide is added to one o f the bulge loops though this is a m inor occurrence and cleavage accuracy remains in the majority o f cleavage products61. It is believed the archaeal intron-processing endonuclease requires a structural recognition pattern that allow s for proper alignm ent o f the intron with the enzymes active site and, though cleavage site m easurement appears to play a role in cleavage efficiency, nucleotide recognition m ay be utilized to identify the intron splice sites61. BHB motifs have now been identified at a number o f locations within tRNA genes from a wide array o f archaeal species. The majority of these introns conform to the BHB m otif described above while a small m inority exist as relaxed motifs that can be cleaved by the tRNA processing endonuclease found in Crenarchaeota w hich is thought to be a more prim itive variant59. 56 In addition to the well established tRNA introns, W atanabe et al60 reported the presence o f small introns in the archaeal protein-coding gene homologue, CBF5 (Centromere-binding factor 5), o f four archaeal species in 2002. Computational analysis predicted the introns would form a BHB secondary structure m o tif similar to those found in the archaeal rRNA and tRNA introns with splice sites predicted to be located in either bulge. The ligation o f the splice sites was confirm ed by sequence analysis o f reversetranscribed cDNA confirm ing the presence o f the first known intron in an archaeal proteincoding gene60. 3.1.3.2 mRNA Characteristics in Archaeal Systems M essenger-RNAs found in Bacteria are typically polycistronic and contain a 5'triphosphate group155 that has been shown to protect mRNA from the endoribonuclease, RNase E, thus prolonging mRNA decay157. Bacterial m RNAs have also been shown to contain short polyadenylated (poly(A)) tails on the 3 '-end155 which play a role in mRNA degradation signalling158. On the other side o f the spectrum, the Eucarya possess mRNAs that are typically monocistronic and contain 5'-m ethylated guanosine caps with long, nontemplated poly(A) tails that provide stability to the m essage155. Archaeal mRNAs are often found as polycistronic transcripts155 and several species have been identified as containing poly(A) tails which have been observed to reduce the stability o f the transcript159,160. No 5' modifications have been identified on archaeal m R N A s159 suggesting archaeal m essages are more similar to those found in Bacteria over those found in Eucarya. Currently the mechanism o f mRNA decay in the Archaea is poorly understood and several studies 57 investigating archaeal mRNA stability have provided results showing great variability in mRNA half-life and stability across a range o f species 155. 3.1.4 Previous Studies Utilizing R T-qPCR in Archaeal Systems RT-qPCR is now becom ing a more widely utilized method for the analysis and quantification o f archaeal RNAs. The technique has been used in investigations ranging from the determination o f am m onia-oxidizing archaeal species abundance in soils161 to identifying changes in expression between the multiple rRNA operons o f the halophilic archaeon, H aloarcula m arism ortui162. Since archaeal mRNAs do not possess long poly(A) tails like their eucaryotic counterparts, and the poly(A ) tails available are involved in mRNA decay signalling, the use o f poly(T) prim ers for reverse-transcription is not appropriate. Instead, reverse-transcription has previously been conducted using random hexameric prim ers162 or primers specific for a target gene163 with apparent success. Reverse-transcription has also been conducted in archaeal halophiles as an initial step to RNA Arbitrarily Primed PCR (R A P-PC R )164. This m ethod o f reverse-transcription, which is highly similar to the use o f random hexam eric prim ers, utilized arbitrary 10-m er prim ers with G+C contents o f 70% and 80% 164 which may assist in prim er annealing as the elevated G+C content observed in many archaeal species is accom m odated for. First strand cDNA synthesis was then perform ed164. Though this method was not used for downstream RTqPCR applications it provides an additional alternative that may assist in achieving improved reverse-transcription results. Quantitative reverse-transcriptase PCR studies have been conducted by LopezLopez et a l162 to determine differential expression between Har. m arism ortui’s three rRNA operons. The organism possesses two rRNA operons that are highly similar and a third that 58 contains a substantial num ber o f single nucleotide polym orphism s. Analysis by RT-qPCR revealed that the unique operon was over-expressed as temperature increased suggesting the divergent operon is expressed as an adaptive measure. Lopez-Lopez et a l'62 were able to successfully use random hexam eric primers for reverse-transcription. After determ ining the p o lA l gene (encoding the small DNA polym erase II sub-unit), which had been used in other archaeal RT-qPCR analysis, had variable expression in Har. marismortui, the ratio o f expression between the two similar rRNA operons under a standard laboratory grow th condition was utilized as a reference to norm alize the expression o f the unique operon. This was done based on an argument that rRNAs are the m ost commonly used reference genes in procaryotes and since the two operons were identical it should be possible to norm alize against the expression ratio between the two operons if the ratio is taken to be I 162. To the best o f our knowledge, this is the only study that has conducted a RT-qPCR assay to assess expression o f any kind in Har. marismortui. Although it is ground-breaking, this investigation does not meet the reference gene requirem ent as outlined by the M IQE guidelines. Since Har. marismortui is a widely used model organism for the eucaryotic ribosome, the proper identification o f a useable subset o f reference genes for RT-qPCR assays is going to be invaluable and would benefit the current research being conducted with respect to ion transport in Har. marismortui described above. M oreover, there does not appear to be any reason as to why the minimum standards required by the M IQE guidelines 132 can not be followed up to, and including, the confirm ation o f reference gene stability and the continuous quality control o f samples. 59 3.1.5 Study Specific Objectives Publications released in the last 2 years reporting RT-qPCR assays conducted during investigations pertaining to archaeal species can still be found that do not yet appear to fully conform to the M IQE guidelines since their conception in 2009165' !66. If researchers are in fact conducting assays with these guidelines in mind it is not always apparent in published m aterials as substantial quantities o f inform ation deemed as required for publication by Bustin et al 132 is not always clearly stated. The most obvious missing information is a lack o f reported reference genes or alternative method for standardizing RT-qPCR data. In the event reference genes are being used and their use is being reported, the likelihood o f data being normalized against more than one reference gene remains minimal. Although this lack o f reference genes is not the only concern with regards to published RT-qPCR assay data it is one o f the m ajor concerns. In order to properly address this m atter adequate reference genes need to be identified in a variety o f archaeal species. 3.1.5.1 Specific Considerations for RT-qPCR Studies in H aloarcula m arism ortui The genome o f Har. marismortui is com prised o f nine circular replicons totalling 4.28Mb. These replicons are organized into one large 3.13Mb chromosome, a substantially smaller 288Kb chromosome, and 7 plasmids ranging from 33Kb to 410K b in size167. As with most other archaeal species, Har. m arism ortui contains short introns w ithin its tRNA genes which are spliced out o f the mature mRNA transcript via tRN A endonuclease cleavage at the BHB m otif168. M oreover, the halophilic archaea are unique from most other archaeal species in that the 3'-ends o f their m RNA lack poly(A) tails entirely169 which leaves the mechanism o f halophilic mRNA decay a mystery. 60 As mentioned above, the lack o f poly(A) tails on Har. m a rism o rtu fs m RN A leaves random hexamers or gene specific prim ers as the only employable option for prim ing cDNA synthesis via reverse-transcription. Due to the num ber o f genes being investigated, random hexamers are used as this m ethodology is m ore cost effective. A dditionally, the placement o f probes, if being used, across introns boundaries is not necessary as introns do not appear in protein coding genes as m entioned above. 3.2 Methods 3.2.1 Identification o f Candidate Reference Genes and Design of R T-qPCR Prim ers Candidate reference genes (Table 3.1) were selected on the basis o f their use in other organisms or the likelihood o f a uniform expression pattern with respect to their use in metabolic pathways. Candidates that had not been previously established in other species were selected based on their location within their respective metabolic pathway and on the likelihood o f being uniformly expressed due to the num ber of ‘inputs’ feeding said pathway. Table 3.1. Candidate reference genes selected on the basis o f prior use in RT-qPCR experimentation or probability o f uniform expression expected due to location o f gene product involvement in metabolic pathways. Loci 16S rRNA* Gene Product gapB rm AC2262 GAPDH rpoA rm AC2428 RNA polym erase alpha sub-unit rpoB polA2 rm AC2430 rmAC2691 RNA polym erase beta sub-unit DNA polym erase II large sub-unit pykA rm AC0546 Pyruvate Kinase Gene 16S rRNA* *The Har. marismortui genome contains three rRNA operons, each of which has a single 16S rRNA. Expression of this RNA will be measured as a combined expression of 16S rRNA from operons rmA and rmB as primer pairs can not be made to accurately distinguish between these two regions. The third 16S rRNA on operon rrnC has been previously identified as having differential expression162. 61 All primers were designed using the Beacon D esigner 7 software package (Prem ier BioSoft). The locations o f candidate reference gene sequences were identified using the UCSC Archaeal Genome Browser85, 86, !70, 171. Location information was then used in conjunction with the NCBI Har. marismortui chrom osom e I whole genom e sequence to obtain each specific gene sequence. The BLAST feature170 within the software suite was used to search each nucleotide sequences against the Har. marismortui genom e in order to maximize prim er specificity and prevent annealing to homologous sequences w ithin the genome. The softw are’s secondary structure search option was used to determ ine regions o f secondary structure across each gene. This cross homology and structural inform ation was used to design prim er pairs (Table 3.2) that produce amplicons ranging from 150-250 base pairs in length (G+C = 45-65%) using the Beacon D esigner ‘avoid cross hom ology’ and ‘avoid secondary structure’ options. Longer am plicons were only generated w hen prim er design parameters for the desired lengths could no longer be relaxed further regarding nucleotide runs, G+C content, annealing temperature, and hairpin or prim er dim er binding energies. Table 3.2. Prim er pairs for qPCR assays o f subset #1 candidate reference genes. All prim ers were designed using the Beacon Designer 7 software package (Premier BioSoft) and supplied by Integrated DNA Technologies. G ene H aloarcula Loci 16S rRNA Operon rm A /B 5 '- GGTTGACGACTTTACTCG gapB * rm A C2262 5 '- Sense Prim er A nti-Sense Primer A m plicon Length (bp) 5 '- GTCATAGCCATTGTAGCC 242 AACTACGAGAAGGCAGTC 5 '- CAACTGTGATGGATTCGG 493 GAACTCCAGATTCAGGTC 5 '- CACACCAATCCGAAGTTC 165 rpoA rm A C2428 5 '- rpoB rm A C 2430 5 '- ATCTCTGCGAGTTCCTGT 5 '- GTTGACCGGACAGACTTC 239 polA 2 rmAC2691 5 '- GAACTGGAAGGACAGATG 5 '- GCGTAGTAGACGTTGATG 230 pykA rm A C0546 5 '- GATGCTTGACTCGATGGT 5 '- ATTTCGCCGATTTCAGGG 338 *The g a p B gene was excluded from the study due to the excessive size of the amplicon produced by the qPCR primers. These were the only primers found that fell within allowable primer parameters using the Beacon Designer 7 software. 62 3.2.2 Preparation o f Haloarcula m arism ortui C ell Cultures Har. marismortui cells were grown in 23% S.W. M G M 138' 139 as described in chapter two. Cells were continuously sub-cultured at mid-exponential growth (ODeoo —0.40.6) as a means o f maintaining continuously generation cultures as before. Cells for reference gene assessment were obtained by sub-culturing cells in balanced grow th to media maintained under varying temperature and potassium ion concentrations which included growth in 23% S.W. MGM at 37°C, 45°C, and 55°C and in media containing 20mM KC1 and 720mM KC1 at standard tem perature (45°C). Three biological replicates under each condition were maintained. Cells (OD = 0.4-0.6) were pelleted by centrifugation once balanced growth under each condition had been obtained and were used im mediately for RNA extraction. 3.2.3 Extraction and Assessm ent of RNA and Subsequent cDNA Synthesis As a means o f m inimizing freeze-thaw cycling o f RNA extracts and to preserve RNA integrity, each o f the following procedures were completed sequentially and immediately after one another. RNA was extracted from each biological replicate (above) using the Qiagen RNeasy kit. Cells in mid-exponential, balanced growth were centrifuged at 13,000rpm (1.5mL/2 minute sequential spin; total vol: 4.5mL) and m edia was rem oved completely by pipette. The cell pellet was resuspended and lysed using the RN easy k it’s guanidine thiocyanate containing buffer, RTL, then hom ogenized using a Q iaShredder spin column (Qiagen) as per m anufacturer’s instructions. Ethanol (70% v/v) was added to the homogenization and ran over the RNeasy spin colum n as per the m anufacturer’s protocol then eluted with nuclease-free water. A preliminary assessm ent of collected RN A purity was conducted post purification via agarose gel electrophoresis. RNA samples were 63 denatured by incubation in the presence o f a formaldehyde/form amide cocktail mix containing ethidium bromide (40mM M OPS, Im M EDTA, Im M sodium acetate, 50% (v/v) formamide, 6.5% (v/v) formaldehyde, 0. lug/m L ethidium brom ide) at 75°C for 15 minutes. Denatured samples were run on a 1% (w/v) agarose-Tris/Acetate/EDTA (TAE) gel then visualized over UV light (Figure 3.1). Follow ing preliminary assessment, RNA samples were digested using the Ambion Turbo DNA-free™ kit as per the m anufacturer’s instructions. Purified RNA samples were stored at -80°C for up to 7 days before quantifying and assessing integrity. RN A was quantified via the Q ubit® fluorom etric system (Invitrogen) and integrity was assessed using the Experion M icro-Capillary Electrophoresis system (BioRad). Both methodologies are preferential as per the M IQE G uidelines132. Single strand cDNA synthesis was conducted using random hexam eric prim ers as per Lopez-Lopez et a l162. cDNA synthesis was completed in 20pL reactions containing 500ng RNA using M oloney M urine Leukemia Virus (M -M uLV) reverse transcriptase (New England Biolabs) as per m anufacturer’s instructions. All cDNA was stored at -20°C until required for use in qPCR assays. 3.2.4 Optimization o f RT-qPCR Reactions and Confirm ation of Controls Candidate reference gene prim ers were tested for specificity via the production o f a PCR amplicon and subsequent agarose gel electophoresis analysis to confirm fragm ent sizes. PCR was conducted using GoTaq Green M asterM ix (Promega) using the m anufacturer’s recommended protocol and am plification products were run on a 1% (w/v) agarose-TAE gel (data not shown). An assessm ent o f prim er specificity with respect to splicing variants as required by the M IQE guidelines132 was not required as haloarchaeal 64 genes do not contain introns and alternative splicing does not occur in any archaeal species. The gradient feature on an iQ5 real-tim e detection system (BioRad) was used to determ ine the optimal prim er annealing temperature (Ta) for each prim er set. Individual reaction conditions (50% (v/v) 2x iQ™ SYBR® Green Supermix (Bio-Rad), luM forward primer, luM reverse primer) and quantity o f template used w ere held constant for all qPCR assays. The Ta that produced the lowest obtainable quantification cycle (Cq) value was used as the optimized Ta for each specific prim er set. No Reverse-Transcription (No-RT) and NoTemplate (NTC) controls were included in each run to m onitor for DNA contam ination and to assess the production o f prim er-dim ers172 via a m elt curve conducted on the real-tim e detection system immediately after the qPCR cycle. 3.2.5 Assessm ent o f RT-qPCR Efficiency The dynamic linear range (DLR) and am plification efficiency was assessed through the construction o f a relative standard curve as described by Taylor et a lm . cDNA from each reverse-transcription reaction above was pooled (250ng from each reaction) and a 1/10 dilution series was created to obtain cDNA standards. qPCR assays (standard recipe, above) were carried out on the pooled cDNA for each serial dilution (lOOng, lOng, lng, 0.1 ng, and 0 .0 ln g per reaction) using the optim ized prim er annealing temperatures. The iQ5 software package (Bio-Rad) was used to create standard curves and assess assay efficiency for each primer pair. Curves were required to produce efficiencies o f 90-110% and have a curve fit o f R2 = 0.98 to be considered acceptable as per the M IQE G uidelines132. 65 3.2.6 Analysis o f Candidate Reference Gene Stability The cDNA (lOng; obtained from the linear dynamic range o f the standard curves) from each biological replicate grown under each condition was used as tem plate in qPCR assays and ran on the iQ5 real-time detection system described in section 3.2.4 above. Assays were run in technical triplicates (sets o f three qPCR reactions constructed from the same sample for the purpose o f averaging variation in results produced by the assay) using the optimized prim er conditions determ ined above. Technical triplicates producing a standard deviation o f 0.5 or lower were considered acceptable and were included in expression comparisons. A relative-fold change in expression was established by comparing candidate reference gene expression during growth under each test condition to the average expression observed for that gene across all conditions. This was done using the “relative gene expression” function found in the qBaseplus software suite (B iogazelle)173. The same software was then used to assess reference gene stability using the GeNorm Mvalue as calculated using the GeNorm algarithm 174. Genes producing an M-value o f 0.5 or lower were considered as stably expressed. 3.3 Results Initial assessment o f RNA purity via gel electrophoresis identified substantial genomic DNA contamination within RNA extracts. D igestion with DNase I immediately after RNA extraction resulted in RNA extracts o f substantially higher purity. Figure 3.1 contains a representative gel used for the assessm ent o f RNA purity ran on a 1% agaroseTAE gel after denaturing formaldehyde agarose gels (ran by standard protocols; data not shown) consistently yielded poor resolution o f RNA fragments. 66 RNA integrity was assessed using B ioR ad’s Experion M ircro-Capillary Electrophoresis system (Figure 3.2). This m ethod utilizes a laser to detect RNA fragments that have been ran out on micro-scale gel w ithin a specialized chip and produces a highresolution digital image o f the fragmentation pattern. This is the preferred method o f evaluating RNA integrity stated by the M IQE G uidelines132 as it is A) highly sensitive; and b) produces a relative quality index (RQI) value based on the densitometric ratio between the 23S and 16S rRNA fragments. The M IQE Guidelines states a minimum RQI o f 7 is acceptable; however, all RNA samples used in this study produced an RQI o f 9 or greater. Assessment o f candidate reference gene stability began with the optim ization o f primers (Table 3.3) for the rpoB gene as described above. The rpoB qPCR assay efficiency (E = 106.6%; R2 = 0.998) was assessed via cDNA standard curve and identified an optimal cDNA concentration o f lOng per qPCR reaction (see Figure 3.3). The curve was constructed using only 4 o f the 5 available serial dilutions due to a consistent lack o f fluorescence in the lOOng reaction replicates. This apparent lack o f am plification was observed across multiple assays at this cDNA concentration leading to the data point being eliminated from all future standard curves. Further evaluation of these non-fluorescing reactions by agarose gel electrophoresis (data not shown) revealed am plification was in fact occurring as a single fragment o f expected size was observed upon visualization with ethidium bromide. Data points were only used in curve construction if the standard deviation between the individual replicates o f a given cDNA concentration was 0.5 or lower. 67 Lane: L 1 2 3 4 G en o m ic DNA 23S rRNA 1 6S rR N A F ig u re 3.1. R epresentative RNA purity assessm ent gel obtained by running RN A extracted from tw o biological replicates o f Har. m arism ortui cells grown in m edia containing 720mM KC1 prior to, and after, digestion w ith DNase 1. Lanes: L-RN A Ladder; l-720m M KC1(1), untreated; 2-720mM KC1(2), untreated; 3720mM KC1(1), D N ase digested; 4-720m M KCI(2), D N ase digested. N um bers in parentheses behind experim ental test conditions indicate biological replicate number. T he disruption observed in the fragm ents in lane 4 was caused by a solid piece o f agarose located in the gel im m ediately in front o f the well. 68 B 6000 400° 3000 I 2000 500 250 50 L I 2 3 4 5 6 7 8 9 10 1 1 12 F ig u re 3.2. M icro-Capillary Electrophoresis gels used for the evaluation o f RNA integrity as produced by B ioR ad’s Experion system using the m anufacturer’s suggested protocol. T he 50bp m arker is added to each sam ple as part o f the protocol and is used by the softw are to properly align all lanes to the RN A ladder. Lanes in A were not aligned by the softw are due to the p o o r resolution o f the ladder. All extracted RN A sam ples used in B produced an RQI value o f 9 or higher (M IQ E recom m ended RQI = 7). All extracted RN A sam ples used in A, w ith the exception o f the 20mM KC1(1) sam ple, produced crisp fragments w ith little apparent degradation and w ere thus assum ed to have an RQI value o f 7 or greater. T he 20mM KC1(1) sam ple w as re­ run in B to confirm integrity. The 720mM KC1 triplicate was ran twice to confirm reproducibility. A: Lanes: L - Ladder; 1 - 45°C (1); 2 - 45°C(2); 3 - 45°C (3); 4 - 3 7 °C (I); 5 - 37°C(2); 6 - 37°C (3); 7 - 55°C(1); 8 55°C(2); 9 - 55°C(3); 10 - 20mM KC1(1); 11 - 20m M KC1(2); 12 - 20mM KC1(3). B: Lanes: L - Ladder; 1 720mM KC1(1); 2 - 720m M KC1(2); 3 - 720m M KC1(3); 4 - 720mM KC1(1); 5 - 720m M KC1(2); 6 720mM KC1(3); 7 - 20mM KC1(1). N um bers in parentheses behind experim ental conditions indicate biological replicate number. 69 rpoB Standard Curve 28 26 E = 106.6% 22 R2 = 0 .9 9 8 Slope = -3.173 Y-int. = 2 1 .8 5 0 20 - 2.0 1.5 1.0 -0 .5 0.0 Log Starting Quantity 0 .5 1.0 F ig u re 3.3. Representative RT-qPCR standard curve constructed using the RNA polym erase beta sub-unit (rpoB ) gene prim ers. Standards were prepared by pooling cDNA synthesized from RN A extracted from Har. m arism ortui cells under each test condition. A ssays w ere run using a 1/10 serial dilution o f pooled cD N A template standards ranging from lOng to O.Olng o f cD N A per assay reaction covering three logarithm ic steps as prescribed by the M IQ E G uidelines. Standard deviations above 0.5 were considered unacceptable. A preliminary assessment o f differential gene expression in the rpoB gene using two biological replicates under each test condition showed the expression o f this gene was highly variable between the two replicates. The assay was repeated two additional times in order to obtain a triplicate o f data. Figure 3.4 shows the relative expression change obtained from each o f the three individual assays which appear to produce a consistent result. Standard growth conditions were considered normal expression and were assigned a value o f 1 thus allowing for a premature relative-fold change to be assigned to alternative expression patterns. Direct comparison o f the three assays (Fig 3.4) shows the results for each independent replicate were highly consistent even though substantial variation between biological replicates was observed. 70 t------------------ 1------------------ 1------------------ 1------------------ 1------------------ 1-------------------1-------------------1------------------ 1-------------------r Sam ple Figure 3.4. Differential expression o f the Har. m arism ortui RNA polym erase beta sub-unit gene (rp o B ) across a range o f extracellular potassium concentrations and grow th tem peratures A direct com parison o f three replicate RT-qPCR assays is shown. Replicate assays w ere conducted on the sam e sam ples to assess if the observed variation between assays is due to assay preparation or perform ance. Expression is show n as fold change relative to the first biological replicate under standard conditions (23% S.W. M GM ; 45°C -1). X -axis labels indicate first (-1) and second (-2) biological replicate for each test condition. Each bar is representative o f an average result obtained from a technical triplicate within each independent assay. Technical triplicates producing a standard deviation below 0.5 w ere considered acceptable. Note: The drastic change in expression between the 720mM KC1 biological replicates (rpoB -3) is due to an error in assay preparation and not a change in differential expression betw een assays. A further comparison was conducted in order to assess variation between cDNA synthesis reactions as previous reports have suggested reverse-transcription being primed by random hexamers can substantially over estimate expression levels162. As a means o f assessing variation between reverse-transcription reactions, three separate cDNA syntheses were conducted from a single extracted RNA sample obtained from a third biological replicate grown under standard conditions (23% S.W. MGM; 45°C). A subsequent RT- qPCR assay was conducted to assess differential expression between the three cDNA samples as a means o f determining if the observed variability in the differential expression 71 between biological replicates is due to natural biological variability or due to the use o f random hexamers during prim ing o f the reverse-transcription step. Differential expression between the three cDNA synthesis reactions is shown as a fold change in expression relative to zero in Figure 3.5. Com parison of G ene Expression Results B etw een cDNA Synthesis R eactions 0.12 c o i/t vt « m 2 o LL_ 0) > S a> X 0-04 0.02 0.00 45C -3(2) 45C~3(1) 45C -3(3) S am p le 11 rpoB~ Figure 3.5. Com parison o f RNA polym erase beta sub-unit (rpoB) expression obtained from three independent cD N A syntheses using a single RN A sam ple as a tem plate. R N A w as extracted from a third biological replicate o f Har. m arism ortui cells grown under standard conditions (23% S.W. M G M ; 45°C). RNA w as extracted after cells w ere in balanced growth. cD N A synthesis was conducted using random hexam eric prim ers and M -M uLV RT (New England Biolabs) as p er m anufacturer’s recom m ended protocol. Relative fold expression betw een cD N A sam ples is show n relative to 0. Once a reasonable explanation for the observed variation in gene expression data between biological replicates had been obtained, the stability o f five candidate reference genes (16S rRNA, rpoA, rpoB, polA2, and pykA ) were assessed. RNA was extracted from three newly established biological replicates and cDNA was synthesized as before. Expression data was imported into the qBaseplus software suite173 and used to determ ine the expression o f each candidate reference gene under each condition relative to the average 72 expression for that gene (Figure 3.6). The software was also used to calculate the GeNorm A/-value which is the established acceptable statistical measure o f expression stability 173 (Figure 3.7). 1.600 1.200 1.000 H 16SrR N A 0.800 Q rp o B 0.600 □ pykA 0.400 H polA 2 0.200 D rp o A 0.000 37°C 45°C 55°C 20m M KC1 720m M KC1 Growth Condition F ig u re 3.6. Expression o f each candidate reference genes under each test condition relative to the average expression o f that gene across all test condition. Figure was produced using relative expression values produced by the qBaseplus 173 software suite. 0.5 373 > 04 ^ 0.3 E 1 w 13 0.2 0.1 Candidate Reference Gene F ig u re 3.7. G raphical representation o f the G eN orm A/-values obtained using the qB asepllJS softw are package. Genes producing values o f 0.5 or low er are considered to be stably expressed and m ay be used as reference genes in qPCR assays as per the MIQE Guidelines. 73 3.4 Discussion Although the standard curve produced for the RN A polym erase candidate reference gene meets the requirements o f the M IQE guidelines132 in terms o f the num ber o f logarithmic steps used to construct the curve, it w ould have been ideal to construct the curve using all 5 planned cDNA standards. Curiously, the expected increase in fluorescence o f the qPCR reactions containing lOOng o f cD N A did not rise above the base-line level o f fluorescence. Since auto-quenching has not been observed in any assay utilizing SYBR Green to the best o f our knowledge, the most logical answ er to this phenom enon would simply be that insufficient reagents (dNTPs, prim ers, etc.) were available for cDNA replication to occur. Intuitively, however, this seems unlikely considering the quantity o f primer and dNTPs used in each 25 pL assay are typical o f a 50pL reaction which would be sufficient to amplify several hundred nanogram s o f template. M oreover, non-specific prim er annealing leading to random am plicon production or highly inefficient am plification is also unlikely as all reactions carried out for each standard curve were prepared from a single m aster mix o f reagents. Had prim er specificity caused this issue all reactions used in the construction o f the standard curve would have been non-functional. As per the M IQE G uidelines132, melt curve analysis was conducted alongside each assay and did not produce results that suggest non-specific PCR products had been produced. This consistently repeatable lack o f fluorescence in the lOOng cDNA standards lead to these data points being eliminated from standard curve construction all together. Agarose gel electrophoresis indicated am plification was in fact occurring norm ally as single fragments o f anticipated size were observed suggesting fluorophore quenching is occurring as a result o f a specific reaction condition or potentially due to auto-quenching o f the SYBR Green fluorophore. 74 Preliminary quantification o f the rpoB gene via a triplicate o f RT-qPCR assays revealed consistency o f expression for each test condition within a single biological replicate. W hen gene expression under each unique test condition was com pared to a second independent biological replicate under the same condition, substantial variance was observed (see Figure 3.4). Considering that biological replicate cultures were grown in parallel to one another using media from the same stock and under identical tem perature conditions, the level o f gene expression between tw o biological replicates w ould be expected to be identical. Seeing as this was not the case, it is possible that the observed variability between biological replicates is due to the m ethod o f priming used for cDNA syntheses. Since Lopez-Lopez et a /162 have shown that the use o f random hexam eric primers can result in an overestimation o f mRNA copy num ber by up to 20-fold, the biological variation being observed could potentially be due to how the mRNAs from each biological replicate was prim ed for cDNA synthesis. The observed relative fold change o f mRNA copy num ber could show substantial variation between identically treated replicates simply by changing the position or sequence, o f an annealing hexameric primer. V ariation in the observed level o f expression between biological replicates may also occur if a single template is being prim ed a variable num ber o f times w ithin a reverse-transcription reaction thus making the reaction less linear then it is intended to be. Because either o f these scenarios could be occurring, it was necessary to evaluate candidate reference gene stability based on an average value obtained with the inclusion o f additional biological replicates. After considering the scenarios above, the potential variation between independent reverse-transcription reactions that utilize a com mon RNA template was assessed. Three independent cDNA samples were obtained from a third biological replicate o f Har. marismortui cells grown under standard conditions and expression o f the rpoB gene was 75 assessed for each. As observed in Figure 3.5, minimal variation in gene expression is observed between the cDNA samples. Since the relative fold change in expression between the samples fall within an acceptable standard deviation o f one another, and because the maximum fold change relative to zero is 0.12, we can assume the variation observed in figures 3.4 can be attributed to natural biological variability. Assessm ent o f reference gene stability using the qBaseplus softw are suite173 has revealed that all five o f the candidate reference genes selected initially can be used as reference genes under the conditions used above. Through exam ination o f the relative fold change data observed (Figure 3.6) it is apparent that the rpoA a n d polA 2 genes are the least stably expressed as the expression o f these genes is shown to decrease by approxim ately 25% under some conditions and increase by approxim ately 40% under others. This observation is supported by the G eN orm M-values produced by the qB aseplus softw are173. As per the M IQE G uidelines132, genes producing GeNorm M -values below 0.5 are considered as stably expressed and are appropriate for use as a reference. The GeNorm Mvalues o f rpoA and polA 2 are both below 0.5 suggesting they are appropriate reference genes; however, these are considerably higher then the remaining three candidates which all have GeNorm M-values o f 0.25 or lower. Placing the five candidates in order o f increasing GeNorm M-value ranks these genes by increasing suitability as qPC R reference genes. This makes the total expression o f all 16S rRNA genes (determ ined using the primers in Table 3.2 which amplify a hom ologous region from all 16S rRNA operons) the most ideal reference gene for qPCR studies followed by the rpoB, p ykA , polA 2, then rpoA genes. The pykA gene, encoding pyruvate kinase has, to best o f our knowledge, never before been validated as a reference gene for qPCR studies making it an entirely novel control. 76 In the event any o f the five novel reference genes that have now been properly established are determined to be unacceptable for use as a reference gene under a future growth condition o f interest, a second subset o f candidates (Table 3.3) has been selected using whole genome microarray data from a closely related halophilic archaeon, Halobacterium salinarum NRC-1. These data are openly available through the online Gaggle database and software package175' 177 and contains differential gene expression data obtained by microarray analysis for all 2400 genes in the NRC-1 genom e across 361 independent growth conditions. The logio ratios for each gene provided by the database have been used to determine the coefficient o f variance across all available experimental conditions (data not shown). The genes producing the lowest overall coefficients have been selected as candidate reference genes and RT-qPCR prim ers have been designed using the Beacon Designer 7 software suite as described above. Interestingly, w hen the NRC-1 homologues o f the five genes evaluated in this study are re-evaluated using this m icroarray data they all fall just beyond the coefficient o f variance cut-off that was used to create the second subset o f candidates found in Table 3.3. This suggests that the new subset o f candidates are very likely to be verified as highly appropriate reference genes and may set a precedent for determining candidate reference genes based on expression levels observed via microarray analysis. 77 T ab le 3.3. Candidate reference genes identified as being minimally variable by Halobacterium salinarum NRC-1 micro array data obtained from the G aggle database175' 177 G ene NRC-1 Loci H aloarcula Loci* H aloarcula G ene P ro d u c t sun VNG0499G VNG0508H VNG1514H rm AC0628 Cytosine-C5 -methylase Hypothetical Protein SucC SucD zim CbiT CbiG VNG1541G VNG1542G VNG1543G rm AC0928 rmAC0981 rm AC0472 rm A C0474 rm AC1876 rm AC2998 Hypothetical Protein Succinyl-CoA synthetase beta chain Succinyl-CoA synthetase alpha chain CTAG modification m ethylase VNG1550G VNG1555G VNG1557G rmAC3003 rm AC3008 CbiK VNG1558H rm A C3009 VNG1559H V N G I561C rm A C3010 rm A C 3011 Hypothetical Protein Hypothetical Protein Cobalt chelase thioredoxin Tupl VNG1562H VNG1564H rm A C3012 rmAC3013 Hypothetical Protein T upl-like transcriptional repressor CobN VNG1566G rm A C3019 C biJ VNG1568G rmAC3021 etJBl VNG2150G rm AC3148 Cobalamin biosynthesis protein Cobalt-precorrin-6Y C5methyltransferase Electron transfer flavoprotein beta sub­ unit dpa VNG2462G rm A C 3118 CbiH Precorrin-8W decarboxylase Cobalamin biosynthesis protein Precorrin-3B C l7-m ethyltransferase Signal recognition particle receptor *These loci were identified as Haloarcula marismortui homologues to the provided NRC-1 loci using the BLAST feature on the UCSC Archaeal Genome Browser85- 86-170- ,71. 3.5 C onclusion In summary, we have shown that investigations utilizing RT-qPCR as a tool for studying gene expression in archaeal organisms can conform to the 2009 MIQE G uidelines132 with relatively minimal additional effort and have established an approach to doing so that can be mimicked by other research groups if required. The use o f the increasingly common RNA spin column as a quick and effective method o f RNA extraction from halophilic cells has proven to yield not only large quantities o f RNA, but RNA o f 78 extremely high integrity as shown by the RQI values obtained from the Experion M icroCapillary Electrophoresis system. This was a point o f initial concern due to the presence o f large salt concentrations that could have potentially affected the spin colum n m em brane and thus the efficiency o f the RNA extraction. A prelim inary evaluation o f differential expression o f the rpoB gene, which appears to be highly variable when grow n across a range o f extracellular potassium ion concentrations and growth temperatures, has shown that the methodology o f using random hexam eric primers for the single strand synthesis o f cDNA is highly consistent leaving the variation in cellular RN A levels for this gene to be attributed to biological variability between cellular replicates. A lthough these m ethodologies may need to be adjusted to accommodate the extraction o f RNA from other archaeal species due to the unique environm ents they live in (For example, how well will an RNA spin column work when isolating cells from media at a pH o f 2?), the techniques described above create a definitive starting point for the evaluation o f reference genes in many other organisms. We have now identified and validated five novel reference genes in the halophilic archaeon, Haloarcula marismortui, which we believe is a valuable starting point for the validation o f proper reference genes in many other archaeal species. Although each o f the genes evaluated above have m et a m inim um set o f standards for use as RT-qPCR references, their expression should still be compared to the expression o f genes that would be expected to be highly variable under the growth conditions used to further confirm stability. This will be completed in the very near future in order to provide additional support for these novel reference genes, whether it is truly required or not. Additionally, a precedent has now been established for the use o f largely available microarray expression data to select potential RT-qPCR reference gene candidates. This approach has now been 79 used to identify a second suitable subset o f candidate haloarchaeal reference genes that can be utilized by other research groups to evaluate and validate should the novel reference genes above prove to be unsuitable for a specific growth condition. It is our hope that this investigation will introduce the M IQE guidelines132 to the archaeal research com m unity and demonstrate that these guidelines can, and should be, applied to RT-qPCR studies that are currently lacking standardization o f experim ental design and methodology. 80 Chapter Four Conclusion 4.1 Assessm ent o f Potassium Stress Responses in Haloarcula m arism ortui Substantial progress tow ard better understanding how the halophilic archaeon, Haloarcula marismortui, responds to potassium stresses in its native environm ent has now been made. This has been com pleted by establishing cellular growth characteristics under standard laboratory growth conditions then evaluating changes in these characteristics under extremely elevated and lim iting concentrations o f extracellular potassium . This investigation has shown that Har. m arism ortui exhibits an optim al generation time o f 4.19 ± 0.14 hours at an extracellular K f concentration o f lOOmM. The cellular generation time increases substantially as K* reaches limiting concentrations (8mM) but only slightly as extracellular K+ increases to concentrations nearly an order o f magnitude higher (720mM ) than optimal. We have suggested that the increased generation time observed in Har. marismortui as K+ becomes limiting is very likely a result o f changing energy requirem ents within the cell. Because Har. m arism ortui sequesters K + to intracellular concentrations o f nearly 1.4M when the extracellular concentration is 8mM (Chapter 2; Figure 2.4), an extremely steep ion gradient is form ed that is likely to result in efflux o f K f through lowaffinity potassium channels in the cellular membrane such as the low-affmity Trk potassium transport system. This efflux should result in the im m obilization o f ATP as a regulator o f the Trk system in order to m aintain the critical intracellular K + concentration at adequate levels which, in turn, would result in a reduced energy reserve for normal cellular system function. A reduced energy capacity for normal cell function will result in slowed generation times. 81 Alternatively, as extracellular K + is elevated in excess o f 700% above normal, the cellular generation time increases only slightly as intracellular K* concentrations approach 2.5M (Chapter 2; Figure 2.4) which may suggest a potential inability to sufficiently regulate K+ uptake. I f an uncontrollable uptake o f K f is in fact occurring, an accum ulation o f intracellular salts to concentrations that becom e sub-optim al for cell system function is likely to cause a deterioration in generation time. As other members o f our lab group are demonstrating with currently unpublished data, the proper function o f many halophilic enzymes are drastically affected by elevated K + concentrations in vitro thus m aking this a plausible explanation for the observed increases in generation time. In addition to the valuable inform ation obtained from the aforem entioned growth studies, we have now established a m ethodology for using m odem analytical chemistry instm mentation in the determination o f intracellular ion concentrations. W e have shown this methodology to be highly ineffective when examining ions o f lower m olecular weight such as sodium and lithium. This is due to the nature o f Induction-Coupled Plasma instrumentation and the space-charge effects that occur when high concentrations o f lowmolecular weight ions are introduced into argon plasma. Though this is the case for some ions that we had hoped to evaluate, it appears to produce reliable, or at m inim um plausible, data when examining the concentrations o f higher m olecular weight ions. Using this methodology we have shown that Har. marismortui sequesters K + ions to concentrations several fold higher than that o f the m edia it grows in and this uptake increases or decreases with the extracellular concentration o f the ion. This is presum ably done in order to balance the osmotic pressure experienced in its native environm ent as has been previously suggested112. We have shown that the organism is an excellent scavenger o f K + ions under limiting conditions and will readily uptake R b \ as has been previously reported. 82 Interestingly, we have also demonstrated that Har. m arism ortui will uptake C s+ in place o f K + if these ions are available; however, the use o f alternative m onovalent ions as a replacement for K+ results in extremely slowed, and apparently variable generation times. Additionally, these ions are not sequestered to concentrations observed for potassium under optimal conditions which may further suggest that K+ is required to m aintain cellular energetics in a manner that has not yet been described. 4.2 F u tu re D irection fo r th e E v alu atio n o f P o tassiu m S tress R esponses in H alophilic A rchaea Now that we have identified and validated several RT-qPCR reference genes (Chapter 3 and Section 4.3) a secondary ‘test’ study will be initiated to exam ine the differential expression o f the gene products within the Trk system described in Haloferax volcaniim . Several genes o f particular interest are those that comprise the Trk low-affinity potassium transport system. Several o f these genes are known to encode m em brane bound potassium ion transport proteins145’ 178 which would be expected to dem onstrate highly differential expression patterns under the conditions used to verify our five novel reference genes. Previous studies have shown that the gene products o f the Trk system genes will actively transport rubidium as a substitute to potassium 145 in some species m aking it o f particular interest with regards to the findings described in Chapter 2. Once gene specific RT-qPCR primers have been optimized and assay efficiency has been assessed in accordance with the MIQE guidelines, the differential expression o f the many genes that comprise the Trk system in Har. marismortui will be examined. Differential expression will be determined across a range o f potassium and rubidium ion concentrations in the growth media while using the reference genes identified in this study as endogenous controls. 83 I I 1) Reference gene stability will also be re-evaluated, as per the M IQE Guidelines “, using the methodologies outlined to ensure stability is not affected by growth in rubidium. This secondary study will further dem onstrate the usefulness o f our reference genes and further our knowledge o f H aloarcula m arism ortui responses to potassium stress. In addition to the latter planned investigation, it would be o f interest to m onitor changes in potassium transporter activity in Har. m arism ortui under in vivo conditions across our chosen range o f extracellular K + concentrations. This can be done using an assay such as the FluxOR™ Potassium Ion Channel Assay (Invitrogen) which utilizes the well documented affinity that potassium channels exhibit for Thallium ions. The assay works by introducing thallium ions into the cell via flow along their concentration gradient or via natural active transport. The ions then interact with, and stimulate a m em brane perm eable reporter dye resulting in cytosolic fluorescence. The intensity o f observed fluorescence is then directly proportional to the total activity o f all potassium ion channels present within the cell. This w ould very likely provide outstanding supporting evidence for any differential expression o f ion transporters observed during the RT-qPCR study outlined above. Although this assay is quite costly and the m anufacturer’s w ebsite states it is intended for use w ith mam m alian cells, it would be well w orth attempting to adapt this assay for use with haloarchaeal systems. 4.3 Evaluation of Novel R T-qPCR Reference Genes in Haloarcula m arism ortui We have now shown that the archaeal research community should be able to conform to the MIQE G uidelines132 while adding only minimal additional effort to established RT-qPCR protocols. The purpose o f the 2009 publication by Bustin et al ]n was not to inflict excessive amount o f additional work on investigators conducting quantitative 84 nucleic acid analyses, but to ensure that all investigators are following a set o f standardized rules that maximum the reliability and reproducibility o f data. As this technique can be highly variable due to the potential for many arbitrary threshold and cut o ff values while analyzing data, and due to the numerous techniques and products available for RNA isolation, purification, and quantification, cDNA synthesis, and the qPCR assay itself, it is important that sufficient information regarding m ethodology and experimental design be provided in qPCR publications. After realizing that m ost investigators using RT-qPCR as a technique for studying archaeal organisms do not appear to be conforming to the M IQE Guidelines due to a severe lack o f available reference genes that have been properly validated, we are now prepared to provide the archaeal research com munity with an established protocol that can be adapted as needed to validate their ow n candidate reference genes in various archaeal species. Five novel reference genes have now been validated by this methodology in Har. marismortui, one o f which (pykA; pyruvate kinase) has, to the best o f our knowledge, never before been used as a reference gene in any species. In order o f decreasing expression stability, these genes are 16S rRNA, rpoB, pykA, polA 2, and rpoA w hich code for the 16S ribosomal RNA, the RNA-polymerase beta subunit, pyruvate kinase, the DNA-polym erase II large subunit, and the RNA-polymerase alpha subunit, respectively. We have also shown that existing gene expression data obtained from microarray analysis can be successfully utilized for the initial identification o f reference gene candidates upon initiation o f a reference gene search. In addition to the validation o f our novel reference genes, preliminary studies have been used to show that substantial biological variation in gene expression is observed between independent biological replicates o f Haloarcula marismortui-, a phenom enon that is likely to be observed in other archaeal species. A fter initially attributing this variation to 85 the m ethod o f priming used for the cDNA synthesis reaction, the use o f random hexam eric primers has been shown to produce acceptably consistent results upon com parison o f expression data obtained from multiple cDNA samples synthesized independently from the same RNA extract. This is o f substantia] im portance as m ost archaeal species do not possess poly-adenylated tails on the RNAs eliminating the possibility o f using oligo-dT primers as a cDNA priming m ethod (see Section 3.1.1). Although gene specific primers can be used for cDNA synthesis, this m ethod can becom e extremely laborious and costly when high-throughput is required while evaluating reference genes. As stated in Chapter 3, it is our hope that this investigation will introduce the MIQE guidelines132 to the archaeal research community and dem onstrate that these guidelines can, and should be, applied to RT-qPCR studies that are currently lacking standardization o f experimental design and methodology. 4.4 F u rth e r E v alu atio n of Id en tificatio n of Novel R T -q P C R R eference G enes in H aiophilic A rch aea As m entioned in Section 4.2, the five novel reference genes identified in this study will now be used to assess the differential expression o f several genes within the lowaffinity Trk potassium transport system. This will be part of an ongoing goal to unequivocally validate a number o f useable reference genes in Haloarcula marismortui. The differential expression data for the Trk protein genes will be evaluated as per the methodology used above. The stability o f these genes will then be com pared to that o f the five novel reference genes to further show that the stability o f our validated references is not an artifact o f an insufficient dataset. 86 Additionally, the evaluation o f several o f the subset 2 candidate reference genes (Table 3.3) in Har. marismortui is nearly underway. The validation o f several m ore genes will allow us to produce a wide range o f acceptable haloarchaeal RT-qPCR reference genes that can be readily used or re-evaluated under new growth conditions by other investigators. This will also provide several “go-to” alternatives in the event one or more validated genes are shown to exhibit reduced stability under one or more specific growth conditions. Finally, all novel reference genes that arc validated in Har. marismortui, current and future, will also be evaluated in the closely related halophilic species, H alobacterium salinarum species NRC-1. This would seem appropriate as all o f candidate genes that are about to be evaluated in Har. marismortui were identified using the N R C -1 microarray data obtained from the Gaggle database and software package175' 177. This will not only allow for “proof o f concept”, but may identify reference genes that are universal am ong the haloarchaea. 87 4.5 References 1. Stanier, R.Y. & van Niel, C.B. The concept o f a bacterium. A rchiv fu r M ikrobiologie 42, 17-35 (1962). 2. Sapp, J. The prokaryote-eukaryote dichotomy: meanings in M icrobiology and M olecular Biology Reviews 69, 292-305 (2005). 3. Chatton, E. Titres et Travaux Scientifiques (1906-1937) de Edouard Chatton. (Sette, Sottano, Italy; 1938). 4. Haeckel, E. Generelle m orphologie der organismen. (Reimer, G., 1866). 5. Haeckel, E. The W onders o f Life: A Popular Study o f Biological Philosophy. (Harper and Brothers, New York, NY; 1904). 6. Copeland, E.B. W hat is a plant? Science 65, 388-390 (1927). 7. Copeland, H.F. The kingdom s o f organisms. The Quarterly Review o f Biology 13, 383-420(1938). 8. Copeland, H.F. The Classification o f Low er Organisms. (Pacific Books, Palo Alto, CA; 1956). 9. W hittaker, R.H. On the broad classification o f organism s. The Quarterly Review o f Biology 34, 210-226 (1959). 10. W hittaker, R.H. New concepts o f kingdoms o f organisms. Science 163, 150-160 (1969). 11. Sogin, S.J., Sogin, M.L. & W oese, C.R. Phylogenetic m easurem ent in procaryotes by primary structural characterization. Journal o f M olecular Evolution 1, 173-184 (1972). 12. Woese, C.R., Sogin, M.L. & Sutton, L.A. Procaryote phylogeny. Journal o f M olecular Evolution 3, 293-299 (1974). mythology. 88 13. W oese, C.R. in Archaea. M olecular and Cellular Biology, (ed. R. Cavicchioli) 1-13 (American Society for M icrobiology Press, W ashington, DC; 2007). 14. Fox, G.E. et al. The phylogeny o f prokaryotes. Science 209, 457-463 (1980). 15. Sanger, F., Brownlee, G.G. & Berrell, B.G. A two-dim ensional fractionation procedure for radioactive nucleotides. Journal o f M olecular Biology 13, 373-398 (1965). 16. W oese, C.R. Q & A. Current Biology 15, R 1 11-R 112 (2005). 17. Doolittle, W.F., Woese, C.R., Sogin, M .L., Bonen, L. & Stahl, D. Sequence studies on 16S ribosomal RNA from a blue-green alga. Journal o f M olecular Evolution 4, 307-315 (1975). 18. W oese, C.R. & Fox, G.E. Phylogenetic structure o f the prokaryotic domain: the primary kingdoms. Proceedings o f the N ational A cadem y o f Sciences 74, 5088-5090 (1977). 19. Zablen, L.B., Kissil, M.S., Woese, C.R. & Buetow, D.E. Phylogenetic origin o f the chloroplast and prokaryotic nature o f its ribosom al RNA. Proceedings o f the National Academ y o f Sciences 72, 2418-2422 (1975). 20. Balch, W.E., M agrum, L.J., Fox, G.E., W olfe, R.S. & Woese, C.R. A n ancient divergence among bacteria. Journal o f M olecular Evolution 9, 305-311 (1977). 21. Fox, G.E., M agrum, L.J., Balch, W .E., Wolfe, R.S. & Woese, C.R. Classification o f methanogenic bacteria by 16S ribosomal RNA characterization. Proceedings o f the N ational Academ y o f Sciences 74, 4537-4541 (1977). 22. Wolfe, R.S. in The Prokaryotes, Edn. 3. (eds. M. Dworkin, S. Falkow, E. Rosenberg, K.-H. Schleifer & E. Stackebrandt) 3-9 (Springer-Verlag, N ew York, NY; 2006). 23. Woese, C.R., Kandler, O. & Wheelis, M.L. Towards a natural system o f organism s: proposal for the domains Archaea, Bacteria, and Eucarya. Proceedings o f the National Academy o f Sciences 87, 4576-4579 (1990). 89 24. Woese, C.R., Magrum, L.J. & Fox, G.E. Archaebacteria. Journal o f M olecular Evolution 11, 245-252 (1978). 25. Kates, M. in Ether Lipids: Chem istry and Biology, (eds. F. Snyder & W .J. Baum an) 251-398 (Academic Press, New York, NY; 1972). 26. Langworthy, T.A., Smith, P.F. & M ayberry, W.R. Lipids o f Thermoplasma acidophilum. Journal o f Bacteriology 112, 1193-1200 (1972). 27. Langworthy, T.A., Mayberry, W .R. & Smith, P.F. Long-chain glycerol diether and polyol dialkyl glycerol triether lipids o f Sulfolobus acidocaldarius. Journal o f Bacteriology 119, 106-116 (1974). 28. de Rosa, M., Gambacorta, A. & Bu'Lock, J.D. Extremely thermophilic acidophilic bacteria convergent with Sulfolobus acidocaldarius. Journal o f G eneral M icrobiology 86, 156-164 (1975). 29. de Rosa, M., Gambacorta, A. & Bu'Lock, J.D. The Caldariella group o f extreme therm oacidophile bacteria: direct com parison o f lipids in Sulfolobus, Thermoplasma, and the M T strains. Phytochem istry 15, 143-145 (1976). 30. Boucher, Y. in Archaea. M olecular and Cellular Biology. . (ed. R. Cavicchioli) 341 353 (American Society for M icrobiology Press, W ashington, DC; 2007). 31. de Rosa, M., Gambacorta, A., M inale, L. & Bu'Lock, J.D. Cyclic diether lipids from very thermophilic acidophilic bacteria. Chem ical Communications, 543-544 (1974). 32. Jahnke, L.L. et al. Signature Lipids and Stable Carbon Isotope Analyses o f Octopus Spring Hyperthermophilic Com m unities Compared with Those o f Aquificales Representatives. Applied and Environm ental M icrobiology 67, 5179-5189 (2001). 33. Gattinger, A., Schloter, M. & M unch, J.C. Phospholipid etherlipid and phospholipid fatty acid fingerprints in selected euryarchaeotal monocultures for taxonomic profiling. FEM S Microbiology Letters 213, 133-139 (2002). 34. Kaneshiro, S.M. & Clark, D.S. Pressure effects on the comnposition and therm al behaviour o f lipids from the deep-sea therm ophile M ethanococcus jannaschii. Journal o f Bacteriology 117, 3668-3672 (1995). 90 35. Sprott, G.D., M eloche, M. & Richards, J.C. Proportions o f diether, m acrocyclic diether, and tetraether lipids in M ethanococcus jannaschii grow n at different temperatures. Journal o f Bacteriology 173, 3907-3910 (1991). 36. Gliozzi, A., Paoli, G., de Rosa, M. & Gambacorta, A. Effect o f isoprenoid cyclization on the transition tem perature o f lipids in therm ophilic archaebacteria. Biochim ica et Biophysica Acta 735, 234-242 (1983). 37. Nichols, D.S. et al. Cold Adaptation in the Antarctic Archaeon M ethanococcoides burtonii Involves M embrane Lipid Unsaturation. Journal o f Bacteriology 186, 8508-8515 (2004). 38. Brown, A.D. & Shorey, C.D. Cell envelopes o f two extremely halophilic bacteria. Journal o f Cell Biology 18, 681-689 (1963). 39. Kushner, D.J. & Onishi, H. Absence o f normal cell wall constituents from the outer layers o f Halobacterium cutirubrum. Canadian Journal o f Biochem istry 46, 997998(1968). 40. Brown, A.D. & Cho, K.Y. The walls o f extremely halophilic cocci: gram -positive bacteria lacking muramic acid. Journal o f General M icrobiology 62, 267-270 (1970). 41. Darland, G., Brock, T.D., Samsonoff, W. & Conti, S.F. A therm ophilic, acidophilic m ycoplasma isolated from a coal refuse pile. Science 170, 1416-1418 (1970). 42. Brock, T.D., Brock, K.M., Belly, R.T. & W eiss, R.L. Sulfolobus: a new genus o f sulfur-oxidizing bacteria living at low pH and high temperature. Archiv fu r M ikrobiologie 84 (1972). 43. Reistad, R. Cell wall o f an extremely halophilic coccus. Archiv fu r M ikrobiologie 82,24 -3 0 (1 9 7 2 ). 44. Weiss, R.L. Subunit cell wall o f Sulfolobus acidocaldarius. Journal o f Bacteriology 118,275-284(1974). 45. Kandler, O. & Hippe, H. Lack o f peptidoglycan in the cell walls o f M ethanosarcina barkeri. Archives o f M icrobiology 113, 57-60 (1977). 91 46. Jones, J.B., Bowers, B. & Stadtman, T.C. M ethanococcus vannielii: ultrastructure and sensitivity to detergents and antibiotics. Journal o f Bacteriology 130, 13571363 (1977). 47. Konig, H., Rachel, R. & Claus, H. in Archaea. M olecular and C ellular Biology, (ed. R. Cavicchioli) 315-340 (Am erican Society for M icrobiology Press, W ashington, DC; 2007). 48. Steber, J. & Schleifer, K.H. H alococcus m orrhuae: a sulfated heteropolysaccharide as the structural com ponent o f the bacterial cell wall. Archives o f M icrobiology 105, 173-177 (1975). 49. Rachel, R., W yschkony, I., Riehl, S. & Huber, H. The ultrastructure o f Ignicoccus: evidence for a novel outer membrane and for intracellular vesicle budding in an archaeon. Archaea 1, 9-18 (2002). 50. Zillig, W., Stetter, K.O. & Janekovic, D. D NA-dependent RNA polym erase from the archaebacterium Sulfolobus acidocaldarius. European Journal o f Biochem istry 96, 597-604(1979). 51. Langer, D., Hain, J., Thuriaux, P. & Zillig, W. Transcription in Archaea: smiliarity to that in Eucarya. Proceedings o f the N ational Academ y o f Sciences 92, 5768-5772 (1995). 52. Bartlett, M.S. Determinants o f transcription initiation by archaeal RNA polymerase. Current Opinion in M icrobiology 8 , 677-684 (2005). 53. Hirata, A. & M urakami, K.S. Archaeal RNA polymerase. Current Opinion in Structural Biology 19, 724-731 (2009). 54. Bell, S.D., Jaxel, C., Nadal, M., Kosa, P.F. & Jackson, S.P. Tem perature, template topology, and factor requirements o f archaeal transcription. Proceedings o f the N ational Academ y o f Sciences 95, 15218-15222 (1998). 55. Woychik, N.A. & Hampsey, M. The RNA polym erase II machinery: structure illuminates function. Cell 108, 453-463 (2002). 56. Hirata, A., Klein, B.J. & M urakami, K.S. The X -ray crystal structure o f RNA polymerase from Archaea. Nature 451, 851-854 (2008). 92 57. Korkhin, Y. et al. Evolution o f com plex RNA polymerases: the com plete archaeal RNA polym erase structure. PLoS Biology 7, 0001-0010 (2009). 58. Abelson, J., Trotta, C.R. & Li, H. tRN A splicing. Journal o f Biological C hem istry 273, 12685-12688 (1998). 59. Marck, C. Identification o f BHB splicing m otifs in intron-containing tRNAs from 18 archaea: evolutionary implications. RNA 9, 1516-1531 (2003). 60. W atanabe, Y.-i. et al. Introns in protein-coding genes in archaea. . F E B S Letters 510, 27-30 (2002). 61. Thompson, L.D. & CDaniels, C.J. Recognition o f exon-intron boundaries by the H alobacterium volcanii tRNA intron endonuclease. Journal o f Biological Chemistry 265, 18104-18111 (1990). 62. W oese, C.R., Gupta, R., Hahn, C., M ., Zillig, W. & Tu, J. The phylogenetic relationships o f three sulfur dependent archaebacteria. Systematic a n d A pplied M icrobiology 5, 97-105 (1984). 63. Chan, P.P., Cozen, A.E. & Lowe, T.M . Discovery o f permuted and recently split transfer RNAs in Archaea. Genome Biology 12, R38 (2011). 64. Rozenski, J., Crain, P.F. & M cCloskey, J.A. The RNA m odification database: 1999 update. Nucleic Acids Research 27, 196-197 (1999). 65. Noon, K.R., Bruenger, E. & M cCloskey, J.A. Posttranscriptional m odifications in 16S and 23S rRNAs o f the archaeal hypertherm ophile Sulfolobus sofataricus. Journal o f Bacteriology 180, 2883-2888 (1998). 66. Bruenger, E. et al. 5S rRNA modification in the hyperthermophilic archaea Sulfolobus sofataricus and Pyrodictium occultum. FASEB Journal 7, 196-200 (1993). 67. Zueva, V.S., Mankin, A.S., Bogdanov, A.A., Thurlow, D.L. & Zim m erm ann, R.A. Occurrence and location o f 7-methylguanine residues in sm all-subunit ribosom al RNAs form eubacteria, archaebacteria and eukaryotes. FEBS Letters 188, 233-238 (1985). 93 68. Ofengand, J. & Bakin, A. M apping the nucleotide resolution o f pseudouridine residues in large subunit ribosomal RNAs from representative eukaryotes, prokaryotes, archaebacteria, m itochondria and chloroplasts. Journal o f M olecular Biology 266, 246-268 (1997). 69. Kowalak, J.A. Identities and Phylogenetic Comparisons o f Posttranscriptional M odifications in 16 S Ribosomal RNA from Haloferax volcanii. Journal o f Biological Chemistry 275, 24484-24489 (2000). 70. Kirpekar, F., Hansen, L., Rasmussen, A., Poehlsgaard, J. & Vestcr, B. The Archaeon has Few M odifications in the Central Parts o f its 23S Ribosomal RNA. Journal o f M olecular Biology 348, 563-573 (2005). 71. O'Farrell, H.C., Scarsdale, J.N. & Rife, J.P. Crystal Structure o f KsgA, a Universally Conserved rRNA A denine Dim ethyltransferase in Escherichia coli. Journal o f M olecular Biology 339, 337-353 (2004). 72. O'Farrell, H.C. Recognition o f a com plex substrate by the K sgA /D im l family o f enzymes has been conserved throughout evolution. RNA 12, 725-733 (2006). 73. Pulicherla, N. et al. Structural and Functional Divergence within the D im l/K sgA Family o f rRNA M ethyltransferases. Journal o f M olecular Biology 391, 884-893 (2009). 74. Ohashi, Z., Maeda, M., M cCloskey, J.A. & Nishimura, S. 3(-3-amino-3carboxypropyl)-uridine: a novel m odified nucleoside isolated from Escherichia coli phenylalanine transfer ribonucleic acid. Biochemistry 13, 2620-2625 (1974). 75. Nishimura, S., Taya, Y., Kuchino, Y. & Ohashi, Z. Enzym atic synthesis o f 3-(3amino-3-carboxypropyl) uridine in Esscherichia coli phenylalanine transfer RNA: transfer o f the 3-amino-3-carboxypropyl group from S-adenosylm ethionine. Biochemical and Biophysical Research Communications 57, 702-708 (1974). 76. Brochier-Armanet, C., Boussau, B., Gribaldo, S. & Forterre, P. M esophilic crenarchaeota: proposal for a third archaeal phylum, the Thaumarchaeota. N ature Reviews M icrobiology 6, 245-252 (2008). 77. Ferry, J.G. & Kastead, K.A. in Archaea. M olecular and Cellular Biology, (ed. R. Cavicchioli) 288-314 (American Society for M icrobiology Press, W ashington, DC; 2007). 94 78. Taylor, C.D. & Wolfe, R.S. Structure and methylation o f coenzyme M. Journal o f Biological Chemistry 249, 4879-4885 (1974). 79. Hao, B. A New UAG -Encoded Residue in the Structure o f a M ethanogen Methyltransferase. Science 296, 1462-1466 (2002). 80. Srinivasan, G. Pyrrolysine Encoded by UAG in Archaea: Charging o f a UAGDecoding Specialized tRNA. Science 296, 1459-1462 (2002). 81. Blight, S.K. et al. Direct charging o f tRN A -CUA with pyrrolysine in vitro and in vivo. Nature 431, 333-335 (2004). 82. Mahapatra, A. et al. Characterization o f a M ethanosarcina acetivorans mutant unable to translate UAG as pyrrolysine. M olecular M icrobiology 59, 56-66 (2006). 83. Bams, S.M., Delwiche, C.F., Palmer, J.D. & Pace, N.R. Perspectives on archaeal diversity, therm ophily and m onophyly from environmental rRNA sequences. Proceedings o f the N ational Academ y o f Sciences 93, 9188-9193 (1996). 84. Huber, H. et al. A new phylum o f A rchaea represented by a nonsized hyperthermophilic symbiont. Nature 417, 63-67 (2002). 85. Schneider, K.L., Pollard, K.S., Baertsch, R., Pohl, A. & Lowe, T.M. The UCSC Archaeal Genome Browser. Nucleic Acids Research 34, D407-D410 (2006). 86. Chan, P.P., Holmes, A.D., Smith, A.M., Tran, D. & Lowe, T.M. in Nucleic Acids Research, Vol. 402012). 87. Spang, A. et al. Distinct gene set in two different lineages o f am m onia-oxidizing archaea supports the phylum Thaumarchaeota. Trends in M icrobiology 18, 331-340 ( 2010 ). 88. Paper, W. et al. Ignicoccus hospitalis sp. nov., the host of'N anoarchaeum equitans'. International Journal o f Systematic and Evolutionary M icrobiology 57, 803-808 (2007). 95 89. Edwards, K.J., Bond, P.L., Gihring, T.M. & Banfield, J.F. A n archaeal ironoxidizing extreme acidophile important in acid mine drainage. Science 287, 17961799 (2000). 90. Okibe, N., Gericke, M., Hallberg, K.B. & Johnson, D.B. Enum eration and Characterization o f Acidophilic M icroorganism s Isolated from a Pilot Plant StirredTank Bioleaching Operation. A pplied a n d Environm ental M icrobiology 69, 19361943 (2003). 91. M acalady, J.L. et al. Tetraether-linked m em brane monolayers in Ferroplasm a spp: a key to survival in acid. Extremophiles 8, 411-419 (2004). 92. Bond, P.L., Druschel, G.K. & Banfield, J.F. Comparison o f acid mine drainage microbial com munities in physically and geochemically distinct ecosystems. Applied and Environmental M icrobiology 66, 4962-4971 (2000). 93. Moses, C.O., Nordstrom, D.K., Herman, J.S. & Mills, A.L. A queous pyrite oxidation by dissolved oxygen and by ferric iron. Geochimica et Cosmochimica Acta 5 1, 1561-1571 (1987). 94. Walsby, A.E. A Square Bacterium. N ature (London) 283, 69-71 (1980). 95. Bums, D., Camakaris, H., Janssen, P. & Dyallsmith, M. Cultivation o f W alsby's square haloarchaeon. FEM S M icrobiology Letters 238, 469-473 (2004). 96. Bolhuis, H., Poele, E.M.t. & Rodriguez-Valera, F. Isolation and cultivation o f W alsby's square archaeon. Environm ental M icrobiology 6, 1287-1291 (2004). 97. Elkins, J.G. et al. A korarchaeal genom e reveals insights into the evolution o f the Archaea. Proceedings o f the National Academ y o f Sciences 105, 8102-8107 (2008). 98. Auchtung, T.A., Takacs-Vesbach, C.D. & Cavanaugh, C.M. 16S rRNA Phylogenetic Investigation o f the Candidate Division "Korarchaeota". A pplied and Environmental M icrobiology 72, 5077-5082 (2006). 99. Reigstad, L.J., Jorgensen, S.L. & Schleper, C. Diversity and abundance o f Korarchaeota in terrestrial hot springs o f Iceland and Kamchatka. The ISM E Journal 4, 346-356 (2009). 96 100. Huber, H. et al. Ignicoccus gen. nov., a novel genus o f hypertherm ophilic chemolithoautotrophic Archaea, represented by two new species, Ignicoccus islandicus sp. nov. and Ignicoccus pacificus sp. nov. International Journal o f Systematic and Evolutionary M icrobiology 50, 2093-2100 (2000). 101. W aters, E. The genome o f Nanoarchaeum equitans: Insights into early archaeal evolution and derived parasitism. Proceedings o f the N ational Academ y o f Sciences 100, 12984-12988 (2003). 102. Brochier, C., Gribaldo, S., Zivanovic, Y., Confalonieri, F. & Forterre, P. Genome Biology 6, R42 (2005). 103. Jahn, U., Summons, R., Sturt, H., Grosjean, E. & Huber, H. Com position o f the lipids o f Nanoarchaeum equitans and their origin from its host Ignicoccus sp. strain KIN4/I. Archives o f M icrobiology 182, 404-413 (2004). 104. Randau, L., Munch, R., Hohn, M .J., Jahn, D. & Soil, D. Nanoarchaeum equitans creates functional tRNAs from separate genes for their 5'- and 3'-halves. Nature 433, 537-541 (2005). 105. Randau, L. The heteromeric N anoarchaeum equitans splicing endonuclease cleaves noncanonical bulge-helix-bulge motifs o f joined tRNA halves. Proceedings o f the National Academy o f Sciences 102, 17934-17939 (2005). 106. Randau, L., Pearson, M. & Soil, D. The com plete set o f tRNA species in Nanoarchaeum equitans. FEBS Letters 579, 2945-2947 (2005). 107. Huber, H., Hohn, M.J., Stetter, K.O. & Rachel, R. The phylum Nanoarchaeota: Present knowledge and future perspectives o f a unique form o f life. Research in M icrobiology 154, 165-171 (2003). 108. Konneke, M. et al. Isolation o f an autotrophic am m onia-oxidizing marine archaeon. Nature 437, 543-546 (2005). 109. W alker, C.B. et al. Nitrosopumilus maritimus genome reveals unique mechanisms for nitrification and autotrophy in globally distributed marine crenarchaea. Proceedings o f the National Academ y o f Sciences 107, 8818-8823 (2010). 97 110. Hatzenpichler, R. et al. A m oderately therm ophilic am m onia-oxidizing crenarchaeote from a hot spring. Proceedings o f the National Academ y o f Sciences 105,2134-2139(2008). 111. Nunoura, T. et al. Insights into the evolution o f Archaea and eukaryotic protein modifier systems revealed by the genom e o f a novel archaeal group. Nucleic Acids Research 39, 3204-3223 (2011). 112. Ginzburg, M., Sachs, L. & Ginzburg, B.Z. Ion metabolism in Halohacterium. Journal o f General M icrobiology 55, 187-207 (1970). 113. Volcani, B.E. in Papers collected to com memorate the 70th anniversary o f Dr. Chaim W eizmann. 71-85 (Collective volume. Daneil Sieff Research Institute, Rehovoth; 1944). 114. Oren, A., Ginzburg, M., Ginzburg, B.Z., Hochstein, L.I. & Volcani, B.E. H aloarcula marismortui (Volcani) sp. nov., nom. rev., an extremely halophilic bacterium from the Dead Sea. International Journal o f Bacteriology 40, 209-210 (7990). 115. Pyatibratov, M.G. et al. Alternative flagellar filament types in the haloarchaeon Haloarcula marismortui. Canadian Journal o f M icrobiology 54, 835-844 (2008). 116. Christian, J.H.B. & Waltho, J.A. Solute concentratoins within cells o f halo-philic and non-halophilic bacteria. Biochimica et Biophysica A cta 65, 506-508 (1962). 117. Shevack, A., Gewitz, H.S., Hennemann, B., Yonath, A. & W ittmann, H.G. Characterization and crystallization o f ribosomal particles from Halobacterium marismortui. F E B S Letters 184, 68-71 (1985). 118. Ban, N. et al. A 9A resolution X-ray crystallographic map o f the large ribosom al subunit. Cell 93, 1105-1115 (1998). 119. Zhao, P. The 2009 Nobel Prize in chemistry: Thom as A. Steitz and the structure o f the ribosome. Yale Journal o f Biology a n d M edicine 84, 125-129 (2011). 120. Ban, N. et al. Placement o f protein and RNA structures into a 5A -resolution map o f the 50S ribosomal subunit. Nature 400, 841-847 (1999). 98 121. Ban, N. The Complete Atomic Structure o f the Large Ribosomal Subunit at 2.4 A Resolution. Science 289, 905-920 (2000). 122. Hildebrand, E. & Dencher, N. Two photosystem s controlling behavioural responses o f Halobacterium halobium. N ature 257, 46-48 (1975). 123. Fu, H.Y. et al. A Novel Six-Rhodopsin System in a Single Archaeon. Journal o f Bacteriology 192, 5866-5873 (2010). 124. M ukohata, Y., Ihara, K., Tam ura, T. & Sugiyama, Y. Halobacterial rhodopsins. Journal o f Biochem istry 125, 649-657 (1999). 125. Spudich, J.L. The multitalented M icrobiology 14, 480-487 (2006). 126. Oesterhelt, D. & Stoeckenius, W. Rhodopsin-like protein from the purple m em brane o f Halobacterium halobium. N ature New Biology 233, 149-152 (1971). 127. M atsuno-Yagi, A. & M ukohata, Y. ATP synthesis linked to light-dependent proton uptake in a red mutant strain o f H alobacterium lacking baceriorhodopsin. Archives o f Biochemistry a n d Biophysics 199, 297-303 (1980). 128. Schobert, B. & Lanyi, J.K. Halorhodopsin is a light-driven chloride pump. Journal o f Biological Chemistry 257, 10306-10313 (1982). 129. Bogomolni, R.A. & Spudich, J.L. Identification o f a third rhodopsin-like pigm ent in phototactic Halobacterium halobium. Proceedings o f the N ational A cadem y o f Sciences 79, 6250-6254 (1982). 130. Spudich, J.L. & Bogomolni, R.A. M echanism o f colour discrim ination by bacterial sensory rhodopsin. N ature 312, 509-513 (1984). 131. Takahashi, T., M ochizuki, Y., Kamo, N. & Kobatake, Y. Evidence that the long­ lifetime photointermediate o f s-rhodopsin is a receptor for negative phototaxis in halobacterium halobium . Biochem ical and Biophysical Research Com munications 127, 99-105 (1985). microbial sensory rhodopsins. Trends in a 99 132. Bustin, S.A. et al. The M IQE Guidelines: M inimum Information for Publication o f Quantitative Real-Time PCR Experiments. Clinical Chemistry 55, 611-622 (2009). 133. Robinson, J.L. et al. Growth Kinetics o f Extremely Halophilic A rchaea (Family Halobacteriaceae) as Revealed by Arrhenius Plots. Journal o f Bacteriology 187, 923-929 (2005). 134. Ginzburg, M. The unusual membrane permeability o f two halophilic unicellular organisms. Biochim ica et Biophysica Acta 173, 370-376 (1969). 135. Tehei, M. et al. Neutron scattering reveals extremely slow cell w ater in a Dead Sea organism. Proceedings o f the National Academ y o f Sciences 104, 766-771 (2007). 136. Oren, A. Bioenergetic aspects o f halophilism. M ircobiology and M olecular Biology Reviews 63, 334-348 (1999). 137. Meury, J. & Kohiyama, M. ATP is required for K + active transport in the archaebacterium H aloferax volcanii. Archives o f M icrobiology 151, 530-536 (1989). 138. Rodriguez-Valera, F., Ruiz-Berraquero, F. & Ram os-Corm enzana, A. Isolation o f extremely halophilic bacteria able to grow in defined inorganic m edia with single carbon sources. . Journal o f General M icrobiology 119, 535-538 (1980). 139. Rodriguez-Valera, F., Juez, G. & Kushner, D J. Halobacterium m editerranei spec, nov., a new carbohydrate-utilizing extreme halophile. System A ppl M icrobiol 4, 369-381 (1983). 140. McLaren, C.E., Brittenham, G.M. & Hasselblad, V. Statistical and graphical evaluation o f erythrocyte volum e distributions. Am erican Journal o f Physiology 252, H857-866 (1987). 141. Olesik, J.W. & Dziewatkoski, M.P. Time-resolved measurements o f individual ion cloud signals to investigate space-charge effects in plasm a mass spectrometry. Journal o f the American Society o f Mass Spectrometry 7, 362-367 (1996). 142. Maher, W. et al. M easurement o f trace elements and phosphorus in m arine-anim al and plant tissues by low-volum e microwave digestion and ICP-MS. Atom ic Spectroscopy 22, 361-370 (2001). 100 143. M ulder, M.M., Teixeira de M attos, M .J., Postma, P.W. & Dam, K. Energetic consequences o f multiple potassium uptake system s in Escherichia coli. Biochim ica et Biophysica Acta 851, 223-228 (1986). 144. Bossemeyer, D. et al. Potassium transport protein TrkA of Escherichia coli is a peripheral membrane protein that requires other TrK gene products for attachm ent to the cytoplasmic membrane. Journal o f Biological Chemistry 264, 16403-16410 (1989). 145. Brown, G.R. & Cummings, S.P. Potassium uptake and retention by Oceanomonas baumannii at low w ater activity in the presence o f phenol. F E M S M icrobiology Letters 205, 37-41 (2001). 146. Zhou, Y., M orais-Cabral, J.H., Kaufman, A. & Mackinnon, R. Chem istry o f ion coordination and hydration revealed by a K+ channel-Fab com plex at 2.0A resolution. Nature 414, 43-48 (2001). 147. Higuchi, R., Dollinger, G., Walsh, P.S. & Griffith, R. Simultaneous am plificatoin and detection o f specific DNA sequences. . Biotechnology 10, 413-417 (1992). 148. Higuchi, R., Fockler, C., Dollinger, G. & W atson, R. Kinetic PCR analysis: real­ time monitoring o f DNA am plification reactions. N ature Biotechnology 11, 10261030(1993). 149. Gibson, U.E., Heid, C.A. & W illiams, P.M. A novel method for real time quantitative RT-PCR. Genome Research 6, 995-1001 (1996). 150. Livak, K.J., Flood, S.J.A., M armaro, J., Giusti, W. & Deetz, K. O ligonucleotides with fluorescent dyes at opposite end provide a quenched probe system useful for detecting PCR product and nucleic acid hybridizatoin. Genome Research 4, 357-362 (1995). 151. Holland, P.M., Abramson, R.D., W atson, R. & Gelfand, D.H. Detection o f specific polymerase chain reaction product by utilizing the 5' - 3' exonuclease activity o f Thermus aquaticus DNA polymerase. Proceedings o f the N ational A cadem y o f Sciences 88, 7276-7280 (1991). 152. Bustin, S.A. Absolute quantificatoin o f m RNA using real-time reverse transcription polymerase chain reaction assays. Journal o f M olecular Endocrinology 25, 169-193 ( 2000 ). 101 153. Gygi, S.P., Rochon, Y., Franza, B.R. & Aebersold, R. Correlation betw een protein and mRNA abundance in yeast. M olecular and Cellular Biology 19, 1720-1730 (1999). 154. Bustin, S.A. & Nolan, T. Pitfalls o f quantitative real-time reverse-transcription polym erase chain reaction. Journal o f Biom olecular Techniques 15, 155-166 (2004). 155. Klug, G., Evguenieva-Hackenberg, E., Omer, A.D., Dennis, P.P. & M archfelder, A. in Archaea. M olecular and Cellular Biology, (ed. R. Cavicchioli) 158-174 (American Society for M icrobiology Press, W ashington, DC; 2007). 156. Dai, L. ORF-less and reverse-transcriptase-encoding group II introns in archaebacteria, with a pattern o f hom ing into related group II intron ORFs. RNA 9, 14-19 (2003). 157. M acki, G.A. Ribonuclease E is a 5'-end-dependent endonuclease. N ature 395, 720723 (1998). 158. Kushner, S.R. mRNA Decay in Escherichia coli Comes o f Age. Journal o f Bacteriology 184, 4658-4665 (2002). 159. Brown, J.W. & Reeve, J.N. Polyadenylated, noncapped RNA from the archaebacterium M ethanococcus vanielii. Journal o f Bacteriology 163, 909-917 (1985). 160. Brown, J.W. & Reeve, J.N. Polyadenylated RNA isolated from the archaebacterium H alobacterium halobium. Journal o f Bacteriology 166, 686-688 (1986). 161. Leininger, S. et al. Archaea predom inate among am monia-oxidizing prokaryotes in soils. Nature 442, 806-809 (2006). 162. Lopez-Lopez, A., Benlloch, S., Bonfa, M., Rodriguez-Valera, F. & Mira, A. Intragenomic 16S rDNA Divergence in Haloarcula marismortui Is an A daptation to Different Temperatures. Journal o f M olecular Evolution 65, 687-696 (2007). 163. Rawls, K.S., Yacovone, S.K. & M aupin-Furlow, J.A. GlpR Represses Fructose and Glucose M etabolic Enzymes at the Level o f Transcription in the Haloarchaeon Haloferax volcanii. Journal o f Bacteriology 192, 6251-6260 (2010). 102 164. Bidle, K.A. Differential expression o f genes influenced by changing salinity using RNA arbitrarily primed PCR in the archaeal halophile H aloferax volcanii. Extremophiles 7, 1-7 (2003). 165. Labrenz, M. et al. Relevance o f a crenarchaeotal subcluster related to Candidatus Nitrosopumilus maritimus to am m onia oxidation in the suboxic zone o f the central Baltic Sea. The ISM E Journal 4, 1496-1508 (2010). 166. Lipscomb, G.L. et al. Natural Competence in the Hypertherm ophilic Archaeon Pyrococcus furiosus Facilitates Genetic M anipulation: Construction o f M arkerless Deletions o f Genes Encoding the Two Cytoplasmic Hydrogenases. A pplied and Environm ental M icrobiology 77, 2232-2238 (2011). 167. Baliga, N.S. Genome sequence o f Haloarcula marismortui: A halophilic archaeon from the Dead Sea. Genome Research 14, 2221-2234 (2004). 168. Dennis, P.P., Ziesche, S. & M ylvaganam, S. Transcription analysis o f two diparate rRNA operons in the halophilic archaeon H aloarcula marismortui. Journal o f Bacteriology 180, 4804-4813 (1998). 169. Portnoy, V. et al. RNA polyadenylation in Archaea: not observed in Haloferax while the exosome polynucleotidylates RNA in Sulfolobus. EM BO reports 6, 11881193 (2005). 170. Kent, W.J. BLAT - the BLAST-like alignm ent tool. Genome Research 12, 656-664 ( 2002 ). 171. Karolchik, D. et al. The UCSC Table Brow ser data retrieval tool. Nucleic Acids Research 32, D493-D496 (2004). 172. Taylor, S., Wakem, M., Dijkm an, G., Alsarraj, M. & Nguyen, M. A practical approach to RT-qPCR— Publishing data that conform to the M IQE guidelines. Methods 50, S1-S5 (2010). 173. Hellemans, J., Mortier, G., De Paepe, A., Speleman, F. & Vandesompele, J. qBase relative quantification framework and software for m anagem ent and autom ated analysis o f real-time quantitative PCR data. Genome Biology 8 , R19 (2007). 103 174. Vandesompele, J. et al. A ccurate normalization o f real-time quantitative RT-PCR data by geometric averaging o f m ultiple internal control genes. Genome Biology 3, research 0034-0034.0011 (2002). 175. Shannon, P.T., Reiss, D.J., Bonneau, R. & Baliga, N.S. The Gaggle: A n opensource software system for integrating bioinform atics software and data sources. BM C Bioinformatics 7 (2006). 176. Bare, J.C., Shannon, P.T., Schmid, A.K. & Baliga, N.S. The Firegoose: two-way integration o f diverse data from different bioinform atic web resources with desktop applications. B M C Bioinformatics 8 (2007). 177. Bare, J.C., Kolde, T., Reiss, D.J., Tenenbaum, D. & Baliga, N.S. Integration and visualization o f systems biology data in context o f the genome. BM C Bioinformatics 11 ( 2010 ). 178. M ongodin, E.F. The genom e o f Salinibacter ruber: Convergence and gene exchange among hyperhalophilic bacteria and archaea. Proceedings o f the N ational Academ y o f Sciences 102, 18147-18152 (2005). 104 A ppendix F ig u re A .I. Scanning electron m icrograph o f m icro-crystalline salt structure. Crystals w ere form ed during an attem pt to image w hole H aloarcula m arism ortui cells. Cells w ere pelleted by centrifugation and excess m edia was rem oved via pipette. Cell pellets w ere thinly spread across an im aging disk then subm erged in liquid nitrogen for 30 seconds to solidify any rem aining liquid m aterial. The frozen ells w ere im m ediately gold plated under vacuum for 55 seconds at 45m A before imaging. Photograph w as obtained using a Philips X L30 scanning electron m icroscope by Mr. Erw in Rehl, D epartm ent o f C hem istry, University o f N orthern British Columbia. 105