SELECTIVE ISOLATION, CHARACTERISATION AND IDENTIFICATION OF STREPTOSPORANGIA Thesissubmitted in accordancewith the requirementsof theUniversity of Newcastleupon Tyne for the Degreeof Doctor of Philosophy by Hong-Joong Kim B. Sc. NEWCASTLE UNIVERSITY LIBRARY ____________________________ 093 51117 X ------------------------------- fn L:L, Iýý:, - L. 51-ý CJ - Departmentof Microbiology, The Medical School,University of Newcastleupon Tyne December1993 CONTENTS ACKNOWLEDGEMENTS Page Number PUBLICATIONS SUMMARY INTRODUCTION A. AIMS 1 B. AN HISTORICAL SURVEY OF THE GENUS STREPTOSPORANGIUM 5 C. NUMERICAL SYSTEMATICS 17 D. MOLECULAR SYSTEMATICS 35 E. CHARACTERISATION OF STREPTOSPORANGIA 41 F. SELECTIVE ISOLATION OF STREPTOSPORANGIA 62 MATERIALS AND METHODS A. SELECTIVE ISOLATION, ENUMERATION AND 75 CHARACTERISATION OF STREPTOSPORANGIA B. NUMERICAL IDENTIFICATION 85 C. SEQUENCING OF 5S RIBOSOMAL RNA 101 D. PYROLYSIS MASS SPECTROMETRY 103 E. RAPID ENZYME TESTS 113 RESULTS A. SELECTIVE ISOLATION, ENUMERATION AND 122 CHARACTERISATION OF STREPTOSPORANGIA B. NUMERICAL IDENTIFICATION OF STREPTOSPORANGIA 142 C. PYROLYSIS MASS SPECTROMETRY 178 D. 5S RIBOSOMAL RNA SEQUENCING 185 E. RAPID ENZYME TESTS 190 DISCUSSION A. SELECTIVE ISOLATION 197 B. CLASSIFICATION 202 C. IDENTIFICATION 208 D. FUTURE STUDIES 215 REFERENCES 220 APPENDICES A. TAXON PROGRAM 286 B. MEDIA AND REAGENTS 292 C. RAW DATA OF PRACTICAL EVALUATION 295 D. RAW DATA OF IDENTIFICATION 297 E. RAW DATA OF RAPID ENZYME TESTS 300 ACKNOWLEDGEMENTS I would like to sincerely thank my supervisor, Professor Michael Goodfellow for his assistance,guidance and patienceduring the course of this study. I am greatly indebted to Dr. Yong-Ha Park of the Genetic Engineering Research Institute in Daejon, Korea for his encouragement, for giving me the opportunity to extend my taxonomic experience and for carrying out the 5S rRNA sequencing studies. I am most grateful to Dr. David Minnikin for his advice and practical assistance in chemotaxonomy as well as to Dr. Alan Ward for his encouragement and discussions on computer-assisted identification. However, I am also indebted to Professor Han of the Department of Biology, Inha University in Inchon, Korea for teaching me the importance of microbiology. Penultimately, I also extend thanks to all in the Department of Microbiology at the University of Newcastle upon Tyne. Finally I am extremely grateful for the never ending support and encouragementof my family and friends. PUBLICATIONS Someof the work presentedin the thesishas been published: Kim, H-J. and Goodfellow, M. (1993). Computer-assistedidentification of Streptosporangium. In Identification of Bacteria: Present Trends-Future Prospects. Proceedingsof the FEMS Meeting in Granada,Spain (abstract). Chun, J., Atalan, E., Kim, S-B., Kim, H-J., Hand, M. E., Trujillo, M. E., Magee, J. G., Manfio, G. P., Ward, A. C. and Goodfellow, M. (1993). Rapid identification of streptomycetesby artificial neural network analysis of pyrolysis mass spectra. FEMS Microbiological Letters (in press). SUMMARY Large numbers of actinomycetes were isolated from composite soil samples using procedures considered to be selective for the isolation of streptosporangia and related sporoactinomycetes from environmental samples. The highest streptosporangial counts were obtained when suspensionsof air-dried soil were heated in the presence of yeast extract for 20 minutes at 4000 then plated onto humic acid vitamins agar supplemented with actidione (50mg/1) and nalidixic acid (30mg/1) and incubated for 4 weeks at 30°C. The highest count, 7.94 ± 1.19 x 104 colony forming units per gram dry soil, were obtained from samples of Ginseng field soil. Representative strains had morphological and chemical properties consistent with their classification in the genus Streptosporangium. Representativeisolates and marker strainsof the genusStreptosporangium were examined for diagnostic features recommended for computer-assisted identification of unknown streptosporangia. Stringent criteria were adopted for positive identifications of both known and unknown strains following a critical evaluation of identification scores obtained for the marker cultures. Sixty-five of the seventy marker strains and twelve of the hundred and thirty six unknown streptosporangia were identified to known streptosporangial taxa. A further nineteen of the isolates were assigned to known taxa using less stringent cut-off points for positive identifications. 5S ribosomalRNA sequenceswere determinedfor nine representativesof the genus Streptosporangium including centrotype strains of two taxa, clusters 1 and 2, circumscribed in a recent numerical phenetic survey and two Ginseng field soil isolates. The primary and secondary structure of the resultant sequenceswere of the type characteristic of Gram-positive bacteria with DNA rich in guanine and cytosine. It was evident from the phylogenetic tree that the genus Streptosporangium is heterogeneous as the type strains of Streptosporangium albidum and Streptosporangium viridogriseum subspecies viridogriseum were sharply separated from the remaining test strains; a result in good agreement with current trends in streptosporangial systematics. Pilot experimentswere designedto determinethe potential of Curie point pyrolysis mass spectrometry and rapid fluorogenic enzyme tests in the classification and identification of streptosporangia. The pyrolysis mass spectral data supported the taxonomic integrity of clusters 1 and 2 and showed that Streptosporangium viridogriseum subspecies viridogriseum had little in common with bona fide members of the genus Streptosporangium. Pyrolysis data also supported the results of the computer-assisted identification exercise as ten isolates assigned to cluster 1 using stringent cut-off criteria were found to be closely related to representatives of cluster 1. There was evidence that some of the conjugated substrates based on the fluorophores 7-amino-4-methylcoumarin and 4-methylumbelliferone have potential as taxonomic markers for the classification of streptosporangia and related actinomycetes. INTRODUCTION A. AIMS Actinomycetes are an unique source of high value bioactive products, notably antibiotics, enzymes, enzyme inhibitors and vitamins. In particular, they account for sixty percent, that is, more than seven thousand of the naturally occurring antibiotics that have been discovered (Table 1, page 2). Approaches to the search for, and discovery of, new bioactive compounds are generally based on screening both naturally occurring actinomycetes and genetically manipulated strains. Current efforts to find the next generation of new antibiotics of therapeutic value are compromised as the probability of discovering new compounds is declining as the number of known antibiotics is increasing (Okami and Hotta, 1988). It is, therefore, important in search and discovery programmes to screen novel and rare actinomycetes in order to raise the probability of finding novel antibiotics (Nolan and Cross, 1988; Bull et al., 1992). Given recent developments in microbial systematics it is now possible to recognise and characterise rare and novel actinomycetes derived from the application of selective isolation procedures by detecting key phenetic taxonomic markers (O'Donnell, 1986,1988; Goodfellow and O'Donnell, 1989; Bull et al., 1992). In addition, molecular taxonomic methods, such as nucleic acid hybridisation, sequencing and fingerprinting techniques, are available for the accurate description of patent strains (Stackebrandt and Goodfellow, 1991). In addition, information in numerical taxonomic databases can be used to design media formulations for the selective isolation of specific fractions of the actinomycete community from soil and other natural habitats (Vickers et al., 1984; Williams et al., 1984a; Williams and Vickers, 1988; Goodfellow and ODonnell, 1989; Bull et al., 1992). Screening methods, which are increasingly target I Table 1 Number of antibiotics producedby membersof selectedactinomycete genera* Genus 1974 1980 1984 1988 1993 Streptomyces 1934 2784 3477 4876 5645 Micromonospora 41 129 269 398 535 Nocardia 45 74 107 262 287 Actinomadura - 16 51 164 248 Actinoplanes 6 40 95 146 169 Streptoverticillium 19 41 64 138 169 Streptosporangium** 7 20 26 39 57 Saccharopolyspora - 4 33 44 55 Dactylosporangium - 4 19 31 40 Amycolatopsis - - - - 23 Kibdelosporangium - - - 7 18 Actinosynnema - - 5 14 17 Microbispora ** 4 6 6 10 15 Streptoalloteichus - 3 4 12 14 14 Kitasatosporia - - - 11 Planobispora ** 10 - - - - 4 Microtetraspora ** - - - - Planomonospora ** 2 - - - - * Data from Berdy (1974,1984), Nisbet (1982) and Berdy database(August, 1993; Data-wm MedhaRtk ye, Sm°iI,kl11e ßeectiens,$rcckl, aºn PaA, Be#tl,wýrtý,, Suwreq, u.k. ), ** Membersof the family Streptosporangiaceae(Goodfellow et al., 1990a). 2 directed, are also important in the searchfor new products (Okami and Hotta, 1988;Bull et al., 1992). There is evidence that the genus Streptosporangiummay become an increasinglyrich sourceof commerciallysignificant products,notably antibiotics (Table 2, page 4). In addition, cystathionine T-lyase has been detectedin a strain of Streptosporangium(Nagasawa et al., 1984). This enzymecatalyses the a, T-elimination reaction of L-cystathionine and also the 7-replacementof L- homoserinein the presenceof various thiol compounds(Kanzaki et al., 1986a). An efficient methodbased on the reactionof 7-replacementhas been developed for the preparationof L-cystathionine(Kanzaki et al., 1986b),a product with potential