Bacterial Systematics Bacterial Systematics Niall A. Logan © 1994 Blackwell Scientific Publications. ISBN: 978-0-632-03775-9 Bacterial Systematics NIALL A. LOGAN BSc PhD Department of Biological Sciences Glasgow Caledonian University OXFORD Blackwell Scientific Publications LONDON EDINBURGH BOSTON MELBOURNE PARIS BERLIN VIENNA © 1994 by Blackwell Scientific Publications Editorial Offices: Osney Mead, Oxford OX2 OEL 25 John Street, London WC1N 2BL 23 Ainslie Place, Edinburgh EH3 6AJ 238 Main Street, Cambridge Massachusetts 02142, USA 54 University Street, Carlton Victoria 3053, Australia Other Editorial Offices: Librairie Arnette SA C rue de Lille 75007 Paris France Blackwell Wissenschafts-Verlag GmbH Dusseldorfer StI. 38 D-10707 Berlin Germany Blackwell MZV Feldgasse 13 A-l238 Wien Austria All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior pennission of the copyright owner. First published 1994 Set by D & N Publishing, Ramsbury, Wilts Printed and bound in Great Britain at The Alden Press, Oxford DISTRIBUTORS Marston Book Services Ltd PO Box 87 Oxford OX2 ODT (Orders: Tel: 0865791155 Fax: 0865 791927 Telex: 8375151 USA Blackwell Scientific Publications, Inc. 238 Main Street, Cambridge, MA 02142 (Orders: Tel: 800 759-6102 617876-7000) Canada Oxford University Press 70 Wynford Drive Don Mills Ontario M3C lJ9 (Orders: Tel: 416 441·2941) Australia Blackwell Scientific Publications Pty Ltd S4 University Street Carlton, Victoria 3053 (Orders: Tel: 03347-5552) A catalogue record for this title is available from the British Library ISBN 0-632-03775-X Library of Congress Cataloging in Publication Data Logan, Niall A. Bacterial systematics/Niall A. Logan. p. cm. Includes bibliographical references and index. ISBN 0-632-03775-X 1. Bacteriology-Classification. L Title- QR81.L64 1994 .589.9'0012-dc20 Contents Preface, vi 1 Introduction, 1 2 Phenotypic characters, 13 3 Genotypic characters, 35 4 Similarity and arrangement, 47 5 Identification, 62 6 Evolution and the archaea, 79 7 The spirochaetes, 97 8 Helical and curved bacteria, 105 9 Gram-negative aerobic bacteria, 115 10 Gram-negative, facultatively and strictly anaerobic bacteria, 143 11 The Gram-positive rods, 165 12 The Gram-positive cocci, 183 13 The mollicutes, 198 14 The actinomycetes, 211 Further reading, 232 Organism index, 243 Subject index, 257 v vi Preface I have long felt the need for an introductory text covering the theory and practice of bacterial classification and identification, and this is my attempt at such a book. It was written at a time of rapid progress in the subject, and while the use of molecular sequencing techniques in evolutionary studies has tended to dominate the literature, it has also revived interest in all the other aspects of bacterial systematics. I thank the authors and publishers who kindly allowed me to adapt their figures for this book, and am most grateful to the follow- ing people who generously provided material for illustrations: Dr G.K. Adak, PHLS Comillunicable Disease Surveillance Centre (Fig. 8.2); Mr J. Bagshaw, bioMerieux UK (Figs S.3a &. 5.5); Dr R.C.W. Berkeley, University of Bristol (Fig. 2.8); Dr P. Berry, Mercia Diag- nostics IFig. 5.3cl; Mr N. Claxton, Becton Dickinson} UK (Fig. 5.3b); Dr P. De Vos, Universiteit Gent (Figs 2.4, 2.5 &. 3.6); Dr M. Gillis, Universiteit Gent (Fig. 6.3); Prof. M. Goodfellow, University of New- castle upon Tyne (Fig. 14.2); Dr A.C. Hill, Medical Research Coun- cil Toxicology Unit (Fig. 13.1); and Dr C.W. Moss, Centers For Disease Control, Atlanta (Figs 2.2 &. 9.8); I also thank Mr J. McWilliams, Glasgow Caledonian University, for the photography. I am indebted to Prof. P.H.A. Sneath, University of Leicester, for explaining to me the taxon-radius model for identification, to Prof. M. Goodfellow for reading the Inanuscript and offering much help- ful comment, and to my wife and children for their forbearance. NIALL A. LOGAN 1 Introduction Not Chaos-like, together crush'd and bruis'd, But as the World, harmoniously confus'd: Where Order in Variety we see, And where, tho' all things differ, all agree. Alexander Pope, 1713, Windsor Forest The development of bacterial classification Michel Adanson's proposal in 1764 for natural classifications based upon many characters and his contempt for 'systems' based upon a selected few brought him into conflict with Carl von Linne (LinnaeusJ, who had devised such systems for plants, animals and minerals. Linnaeus doubted the value of microscopy and was therefore unable to classify the animalcules described by Antonie van Leeuwenhoek and others for want of characters; he placed them all in a class of invertebrates which he called 'Chaos' - the shape of matter before it was reduced to order. Most microscopists of the 17th and 18th centuries did not try to clas- sify the infusion animalcules that they observed and often described so meticulously. With works published in 1773 and 1774 the Danish natural- ist Otto Muller was the first to attempt a systematic arrangement of the animalcules, but he did not make a clear distinction between what we now call protozoa and bacteria. For his last work however, which was published posthumously in 1786, he did create two form genera, Monas and Vibrio, which contained bacteria and accommodated the punctiform and elon- gated types. In 1838 Christian Ehrenberg considerably extended Muller's nomen- clature and added the helical bacteria. He was handicapped by the micro- scopes of his day, and many of the groups that he described cannot now be recognized. None the less, he established genus and species names, such as Spirochaeta plicatilis and Spirillum volutans, that are still in use. Subse- quent workers devised simpler classifications, but all these early bacterial systematists based their arrangements upon microscopic morphology, and assumed constancy of form at a time when the theories of spontaneous gen- eration and pleomorphism (i.e. no constancy of form) were still widely held and the germ theory of disease had yet to be proven. In the 1870s Ferdinand Cohn, having satisfied himself that bacterial forms were constant irrespective of environmental conditions, recognized the existence of a wide diversity of bacteria but considered them to form a distinct group that was unrelated to fungi, yct had close affinities with blue-green algae. He arranged bacteria in six form genera, which he believed to be the natural ones, and many provisional species, but appreci- ated that the physiologies, products and pathogenicities of similar-shaped organisms might differ, and so used such properties in subdividing his gen- era. In Bacillus he placed B. subtilis, having recognized its spores as 1 Bacterial Systematics Niall A. Logan © 1994 Blackwell Scientific Publications. ISBN: 978-0-632-03775-9 Chapter 1 Introduction persistent forms, and another sporeformer, B. anthracis, whose life history Robert Koch demonstrated to him in 1876 - thereby proving the germ the- ory of disease. FrOln his studies of infectious diseases, Koch later concluded that the different forms of pathogenic bacteria must be regarded as distinct and constant species. Cohn found it necessary to defend his beliefs, based as they were on careful observation, against those who still tried to prove, using defective methods, that bacteria were merely stages in the development of fungi (after the foundation of medical mycology in the 1840s several diseases, including cholera and measles, had been attributed to fungi by some work- ers), and that changes in environmental conditions altered bacterial mor- phology. In 1882 Edouard Buchner claimed to have converted B. subtilis to B. anthracis by shaking it in media at different temperatures! Pure cultures The concept of bacterial species led to the idea that pure cultures might be obtainable. By 1872 Cohn's co-worker Joseph Schroeter had cultivated pure colonies of chromogenic bacteria, including the violet-pigmented organ- isms now informally referred to as chromobacteria, on a variety of starchy foods, eggs and meat, and in 1878 Joseph Lister obtained a pure culture of a milk-souring organism by dilution. Koch had turned from using animal pas- sage for purifying his pathogenic strains to developing plates of cultivation media solidified by gelatin, which was subsequently replaced by agar, and he demonstrated and published his Inethods in the early 1880s. So began the 'golden age of microbiology' - bacteria could now be isolated routinely by streak dilution culture and, as Perkins observed in 1928, this 'discovery of the principles of pure-culture study resulted in such a sudden burst of inves- tigation that it was a lost month in which a new organism was not described, catalogued, and laid away, very frequently in the wrong grave', and led to the development of the many diverse characterization tests, such as the Voges-Proskauer (1898) test for acetylmethylcarbinol, the methyl-red test (1915) for production of copious acid from glucose, and the tests for cytochrome oxidase (1928) and urea hydrolysis (1946), that were described over the next 80 years. The various later workers who developed classifications drawn up by Cohn continued to regard spherical organisms (cocci) as prilnitive forms and they emphasized morphological characters, especially shape and size of cells, arrangement or absence of flagella, and production of spores, in their higher divisions of bacteria. This Linnaean approach caused confusion; B. subtilis and B. anthracis, for example, were put together as sporefonners, or simply as rods by some, and placed in separate genera by others because the former is motile and the latter is not. It was only in the second decade of the twen- tieth century, following the work of OrIa-Jensen, and with the reports of the Society of American Bacteriologists' Committee on Bacterial Classification and Nomenclature in 1917 and 1920, that physiological characters such as 2 aerobic or anaerobic growth became widely used in descriptions of genera. Bergey's Manual Bergey's Manual of Determinative Bacteriology, published in 1923, was written to provide a modern identification key for bacteria but little of it was based on direct experience of the organisms and this first and five subsequent editions, which were very difficult to use in the laboratory unless organisms were already identified to genus level, came to be used more as the authoritative reference works on bacterial classification. In successive editions of Bergey's Manual, and in the absence of a usable fossil record, a quasi-evolutionary approach to arrangement, as used for plants and animals, was adopted; in the seventh edition, published in 1957, the photoautotrophs were regarded as the most primitive forms and the rickettsias, to which the viruses were tentatively attached, the most advanced. In the eighth edition, which was published in 1974, it was con- sidered that such an approach was no longer justifiable and the genera, sometimes grouped in families where thought useful, were arranged in 19 parts. Each of these was based upon a few readily determined characters and bore a vernacular name such as 'Gram-positive cocci'; hence, no evo- lutionary relationships were implied. Instead, a tentative subdivision of the Kingdom Procaryotae was given - proposals for the recognition of such a kingdom of anucleate organisms had been made since the late 1930s and were subsequently supported by microscopy, including elec- tron microscopy, and molecular studies. Numerical taxonomy As the variety of methods for characterizing bacteria increased, so bacter- ial systematists suffered more and more from the lack of quantitative approaches to classification. In 1957 Peter Sneath revolutionized the sub- ject with his two papers 'Some Thoughts on Bacterial Classification' and 'The Application of Computers to Taxonomy' (taxonomy being the sci- ence of classification; it is often used as a synonym for classification) in which he described numerical methods of grouping bacteria using the chromobacteria together with some other Gram-negative rods as his example. In subsequent publications, including two books (1963, 1973) written in collaboration with Robert Sokal, Sneath summarized the fundamental position of numerical taxonomy (the grouping by numerical methods of taxonomic units based upon their character states), in principles referred to as neo-Adansonian and which included the need for many and equally- weighted characters. The rapid development and availability of computers enabled the integration of many different pieces and types of data into the classification process, which consequently benefited from greater information content and the more objective recognition of groups or taxa; thus, the realization of Adanson's ideas after 200 years depended, to a great extent, on the advent of electronic computing. Chapter 1 Introduction 3 Chapter 1 Introduction 4 Modern methods The period that saw the development of numerical taxonomy also saw the rise of chemotaxonomy - the application of modern biochemical analyti- cal techniques, principally chromatographic and electrophoretic separa- tion methods, to the study of distributions of specific chemicals such as aluino acids, proteins, sugars and lipids in bacteria. Of particular interest are studies of nucleic acids, the ultimate objective being rapid and direct sequencing. Data from DNA-rRNA reassociation experinlents and protein sequencing, and from methods of inferring and comparing sequences of rRNA molecules, which have evolved very slowly so that the base sequences of many cistrons are highly conserved, have been used like a bac- terial fossil record (but without time units) to facilitate the construction of prokaryote 'phylogenies' (genealogical trees) that are quite different to those implied in the seventh edition of Bergey's Manual. The first of four volumes of Bergey's Manual of Systematic Bacteriol- ogy, published in 1984, was arranged in sections much like the parts of the eighth edition of Bergey's Manual of Determinative Bacteriology, and pro- posed a revision of the higher taxa of prokaryotes consistent with the phy- logenetic information then available. It is becoming clear that most existing bacterial classifications, which are based upon phenotypic charac- ters (observable expressions of the genotype) with a view to providing iden- tification schemes, correlate poorly with the evolutionary relationships that appear to exist between the higher taxa. The problem now facing tax- onomists is the construction of a single and practical scheme capable of incorporating genotypic and phenotypic information. Consequently, polyphasic taxonomic studies, which use ranges of both genomic and phe- notypic approaches, are now widely favoured. Why classify bacteria? Classification, the grouping of things together, is a comlnon and important human activity that has been practised since earliest times. In dealing with large numbers of objects or pieces of infonnation some convenient system of orderly arrangern.ent is needed for the purposes of storage and retrieval, and in any scientific work up-to-date classifications are essential. Books} for example, may be grouped by author, subject, title or a combination of these; without such a system a library would be virtually unusable and the retrieval of information froln it hopelessly inefficient. In bacteriology, classification is a means of summarizing our knowledge of prokaryotes and cataloguing that knowledge. As this information is constantly and rapidly expanding, so classifications evolve and increase in importance, with contemporary schemes reflecting our state of knowledge about the organisms concerned. Such schemes may thenlselves be classi- fied into natural and artificial (or special purpose) types.