Advances in
MICROBIAL PHYSIOLOGY VOLUME 38
This Page Intentionally Left Blank
Advances in
MICROBIAL PHYSIOLOGY Ed...
58 downloads
899 Views
15MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Advances in
MICROBIAL PHYSIOLOGY VOLUME 38
This Page Intentionally Left Blank
Advances in
MICROBIAL PHYSIOLOGY Edited by
R. K. POOLE Department of Molecular Biology and Biotechnology The Krebs Institute f o r Biomolecular Research The University of Sheffield Firth Court, Western Bank Sheffield SlO 2TN, UK
Volume 38
ACADEMIC PRESS San Diego London Boston New York Sydney Tokyo Toronto
This book is printed on acid-free paper. Copyright 0 1997 by ACADEMIC PRESS All Rights Reserved.
No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Academic Press, Inc. 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.apnet.com Academic Press Limited 24-28 Oval Road, London NWl 7DX, UK ISBN 0-12-027738-7
A catalogue record for this book is available from the British Library
Typeset by Technical Typesetters, Ashford, Kent, UK Printed in Great Britain by Hartnolls Ltd, Bodmin, Cornwall
9697 98 99 00 01 EB 9 8 7 6 5 4 3 2 1
Contents
CONTRIBUTORS TO VOLUME 38
........................
vii
Hydrophobins: Proteins that Change the Nature of the Fungal Surface Joseph G . H.Wessels 1. 2. 3. 4. 5. 6. 7.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Identity of hydrophobins . . . . . . . . . . . . . . . . . . . . . . . . . Rodlets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Surface activities of hydrophobins . . . . . . . . . . . . . . . . . . . . Formation of emergent structures . . . . . . . . . . . . . . . . . . . . . Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 4 10 13 19 34 36 36 36
Structure-function Analysis of the Bacterial Aromatic Ring-hydroxylating Dioxygenases Clive S . Butler and Jeremy R . Mason
................................ ........................... Structure of ring-hydroxylating dioxygenases . . . . . . . . . . . . . . Electron transport system . . . . . . . . . . . . . . . . . . . . . . . . . The catalytic terminal oxygenase component . . . . . . . . . . . . . . . Coordination of the iron-sulphur clusters . . . . . . . . . . . . . . . . . The catalytic non-haem iron centre . . . . . . . . . . . . . . . . . . . . Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction
2. Bacterial oxygenases
3.
4.
5. 6. 7.
8.
47 48 50 51 60 61 72 75 76 76
vi
CONTENTS
Thiol Template Peptide Synthesis Systems in Bacteria and Fungi Rainer Zocher and Ullrich Keller Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The peptide synthetase domain . . . . . . . . . . . . . . . . . . . . . . Enzymesystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Peptide synthetases from fungi . . . . . . . . . . . . . . . . . . . . . . Prokaryotic peptide sythetase systems . . . . . . . . . . . . . . . . Future prospects of peptide synthetase research . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. 2. 3. 4. 5. 6.
. .
86 88 94 96 . 111 . 122 124 124
Microbial Dehalogenation of Halogenated Alkanoic Acids. Alcohols and Alkanes J . Howard Slater. Alan T. Bull and David J . Hardman 1. 2. 3. 4. 5.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Dehalogenation of halogenated alkanoic acids . . . . . . . . . . . . . . 135 Dehalogenation of halogenated alcohols . . . . . . . . . . . . . . . . . 151 Dehalogenation of halogenated alkanes . . . . . . . . . . . . . . . . . .159 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
Metal-Microbe Interactions: Contemporary Approaches T. J . Beveridge. M . N . Hughes. H . Lee. K . T. Leung. R . K . Poole. I . Savvaidis. S . Silver and J .T.Trevors Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 Microorganisms and metals: their essentialchemistry . . . . . . . . . . 179 Complexation of metal ions in media and cellular milieu . . . . . . . . 185 Analysis for total metal and for metal species . . . . . . . . . . . . . . 190 Spectroscopic techniques in the study of metals and microorganisnls . . 205 Molecular and genetic methods . . . . . . . . . . . . . . . . . . . . . . 210 214 7. Case studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 8. Conclusions and outlook . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
1. 2. 3. 4. 5. 6.
Author Index SubjectIndex
.................................. ..................................
245 263
Contributors to Volume 38
T. J. BEVERIDGE, Department of Microbiology, College of Biological Sciences, University of Guelph, Guelph, Canada NIG 2W1 Alan T. BULL,Research School of Biosciences, University of Kent at Canterbury, Canterbury, Kent CT2 7NJ, UK Clive S. BUTLER,School of Biological Sciences, Molecular and Microbiology Sector, University of East Anglia, Norwich NR4 7TJ, UK David J. HARDMAN,Carbury Heme Ltd, Research and Development Centre, Canterbury, Kent CT2 7PD, UK M. N. HUGHES,Department of Chemistry, King’s College London, Strand, London WC2R 2LS, UK Ullrich KELLER,Institut fur Biochemie und Molekulare Biologie, Technische Universitat Berlin, FranklinstraPe 29, D- 10587 Berlin-Charlottenburg, Germany H. LEE,Department of Environmental Biology, Ontario Agricultural College, University of Guelph, Guelph, Canada NIG 2W1
K. T. LEUNG,Department of Environmental Biology, Ontario Agricultural College, University of Guelph, Guelph, Canada N1G 2W1 Jeremy R. MASON,Division of Life Sciences, King’s College London, Campden Hill Road, London W8 7AH, UK R. K. POOLE,Department of Molecular Biology and Biotechnology, Krebs Institute for Biomolecular Research, The University of Sheffield, Firth Court, Western Bank, Sheffield S10 2TN, UK I. SAVVAIDIS, Department of Microbiology, University of Ioannina Medical School, Post Box 1186,45110 Ioannina, Greece
S . SILVER,College of MedicineDepartment of Microbiology and Immunology, University of Illinois, M-C 790,835 S. Wolcott Ave, Chicago, l L 60612, USA
viii
CONTRIBUTORS TO VOLUME 38
J. Howard SLATER,Molecular Ecology Research Unit, School of Pure and Applied Biology, University of Wales, PO Box 915, Cardiff CFI 3TL, UK
J. T.TREVORS, Department of Environmental Biology, Ontario Agricultural College, University of Guelph, Canada, NlG 2W1 Joseph G. H. WESSELS,Department of Plant Biology, Groningen Biomolecular Sciences and Biotechnology Institute, University of Groningen, Kerklaan 30, 975 1 NN Haren, The Netherlands Institut fur Biochemie und Molekulare Biologie, Technische Rainer ZOCHER, Universitat Berlin, FranklinstraPe 29, D- 10587 Berlin-Charlottenburg,Germany
Hydrophobins: Proteins that Change the Nature of the Fungal Surface Joseph G. H. Wessels Department of Plant Biology, Groningen Biomolecular Sciences and B i o t e c h l o g y Institute, University of Groningen, Kerklaan 30, 9751 NN Haren, The Netherlands 1. Introduction . . . . . . . . . . . . 2. Identity of hydrophobins . . . . . . 3. Rodlets . . . . . . . . . . . . . . 4. Surface activities of hydrophobins . 4.1. SC3 hydrophobin . . . . . . . 4.2. Cerato-ulmin . . . . . . . . . 5. Formation of emergent structures 5.1. Formation of aerial hyphae . . 5.2. Formation of fruit bodies . . . 5.3. Formation of conidia . . . . . 5.4. Pathogenesis. . . . . . . . . 5.5. Symbiosis . . . . . . . . . . 6. Technology . . . . . . . . . . . . 7. Conclusions . . . . . . . . . . . . Acknowledgements , . . . . . . . References . . . . . . . . . . . .
. ... .. . . ... .. . . ... ... ....... . ... ... . . .. ... . . .. , . , . . . . .. , . . . . .. . . . . . , , . . . . . . . . . . . . .. . . . . . . , ,
. ..... . . ..... . . . .. .. . ....... . . . . .. . . , .. , . , , , , . . , . , , . , . . . . . . . . . . ,
,
,
,
.. .
. . . . . . . . .. . .. . ... , .. . . . . . . . . . . . . . . . . . . . .. .. , .. .. . . . . . . . . . . . . . . .
. . . . . . . . . . . 1 . . . . . . . . . . .4 . . . . . . . . . . . . 10 . . . . . . . . . . . . 13 . . . . . . . . . . . . 13 . . . , , . . . . . . 18
. . . .
,
. .. . . . .. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . ..
. . . . .
. . . . .
. . . . .
. . . . .
. . . . . . . . . . . . . .. . . . .. . . . ..
. . 19 . . 19 . . 22
..
. . . . .
27
. 29 . 33 . 34
.
36
. 36
.......
36
1. INTRODUCTION Although fungi, with the exception of yeasts, are multicellular, their mechanisms of growth and development are quite distinct from those of plants and animals. In the multicellular fungi, the fundamental growth unit is the hypha, which may or may not be divided into cells. The hypha is essentially a tubular structure containing all the typical eukaryotic cytoplasmic components surrounded by a rigid wall. Hyphae are highly polarized and grow only at one end, where a new wall is ADVANCES IN MICROBIALPHYSIOLOGY VOL 38 ISBN 0-12-027738-7
Copyright 0 1997 Academic Press Limited All rights of r e p d u d i o n in any form reserved
2
JOSEPH G.H.WESSELS
deposited in such a way that the tubular shape is generated, despite the presence of a high internal hydrostatic pressure (turgor) (Wessels, 1986,1990,1993a). Hyphae regularly branch and give rise to a mycelium that forms a colony. The colony grows radially at its periphery,whcre apical extcnsion and branching occur. The mycclium thus colonizes a substrate, maintaining a constant ratio between the total length of hyphae and the number of tips, a ratio known as the “hyphal growth unit” (Trinci. 1974; Trinci et nl., 1994). Another uscful concept developed by Trinci and co-workers (see Tritici el ul., 1994) is that extension of tips of individual hyphae is supported by a certain volume of protoplasm; this mycelial region involved in tip growth is called thc “peripheral growth zone” of the colony. In principle, hyphae appear designed for unlimited transport of water, nutrients and cytoplasmic components. In the zygomycetes the hyphae are not subdivided by septa so that the cytoplasm (with many nuclci) is essentially contained in a continuousbranched tubular system, making it difficult in these organisms to apply the concept of cellularity. In the second major group of the fungi, the ascomycetes, septa are present but these contain large pores that do not seem to obstruct movement of organelles such as nuclei and mitochondria. Only in thc third group. the basidiomycetes, do septa divide the hyphae into separate compartments. The small septa1 pore and the elabori~temembranous structures (parenthosomcs) that cover it (Moore. 1985) effectively prevent passage of nuclei and mitochondria, but apparently do allow for transport of water and nutrients. In fact. members of the basidiomycetes have a great capacity for long-rangetranslocation (Jennings, 1984). However, in situationswhere nuclci have to bc cxchangcd in these basidiomycetes, as in mating interactions, thc scpti are dissolved so that nuclei and other cytoplasmic components can move freely through the hyphal tubes (Raper, 1966; Wessels, 1978). Because hyphae not only grow but also sccrctc cnzymcs at their apices (Wosten cr ul., 1991). the fungal mycelium is ideally suited for growing into solid organic substrates and for degrading the constitucnt polymcrs from within (Wessels, 1993a). Thc colonization is Ihcilituted by the fact that, in the local absence of nutrients, hyphal growth can be sustained by the transport of water and nutrients from a food base (Jennings. 1984, 1994; Kayner et a[., 1995). In this way the mycelium can explore large arc~sthat conlain only isolated patches of nutrients. Another manifestation of the ability of the mycelium to transport water and assimilates to hyphae that are unable to acquire nutrients, is the Occurrence of emergent growth. For instance, hyphae may give up assimilatingnutrients and grow into the air, causing the “mouldy” appearance of many fungi. A function of the hydrophobic felt-like mats that arc often produced is not obvious but may be prevention of water loss from the substrate. Certain aerial hyphae, however, may differentiateinto spore-bearingstalksthat at their apices form sporangiacontaining sporangiospores (many zygomycetes) or sterigmnta that bud off conidiospores (many ascomycetes). Alternatively,aerial hyphae iit thcir apices may break up into oidiosporcs.
HYDROPHOBINS
3
All these structures serve vegetative reproduction. For sexual reproduction, many fungi form multihyphal fruit bodies within which meiosis occurs and meiospores are formed. As asexual spores, these can also be dispersed through the air or otherwise disseminated. Particularly in members of the basidiomycetes, the fruit bodies (mushrooms and brackets) can attain large sizes and their morphogenesis has attracted much attention both from a purely scientific and, for the edible species, from a commercial point of view (Wessels, 1993b). These large multicellular structures are not formed by cell divisions within meristems, as in plants, but they are formed by individual hyphae that grow at their tips and seem to “know” how to organize themselves into a distinct multihyphal structure. It is clear that, for the elaboration of these aerial fruit bodies, massive transport of water and assimilates from the assimilative substrate mycelium is required. When the substrate is exhausted, components of the substrate mycelium itself may be broken down and breakdown products reused for the construction of these emergent structures (Wessels and Sietsma, 1979; Wessels, 1993b). After spores have been dispersed, they must find a substrate in order to germinate and to produce a new mycelium. For saprotrophs this poses no problems, provided the spore lands in an area where dead organic material and enough moisture are available. For biotrophs, however, it is often necessary for a spore to attach to, and to germinate on, the bare surface of the host before infection structures can be formed and the host is penetrated to form an assimilative mycelium. This is particularly clear in parasitic relationships with plants and animals (Cole and Hoch, 1991), but is also evident in the mutually beneficial associations with plants, the mycorrhizas (Harley and Smith, 1983; Read, 1991). In addition, a large number of fungi, an estimated 20% of all species, have evolved as lichens, aerial structures in which the fungus obtains its assimilates from symbiotic algae and cyanobacteria (Honneger, 1993). This brief overview of fungal biology serves as an introduction to understanding the roles played by hydrophobins. These proteins were discovered while searching for genes expressed during emergent growth in Schizophyllurn cornmurre. AS in many homobasidiomycetes, the primary mycelium that grows from a meiospore forms aerial hyphae but, after mating of two compatible primary mycelia, a secondary mycelium is formed that produces fruit bodies in addition. The cDNAs of a number of abundant mRNAs appearing during emergence of the aerial structures of primary and secondary mycelia of S. commune were cloned (Mulder and Wessels, 1986). The most abundantly expressed genes were sequenced, revealing that at least four of the ten cloned genes encoded similar small cysteine-rich hydrophobic proteins (Schuren and Wessels, 1990; Wessels et al., 1995). At that time these proteins were totally unknown. Eventually, the product of one of these genes (SC3) was found in the walls of aerial hyphae, while the abundant product of another (SC4) was found in walls of hyphae that make up fruit bodies (Wessels et al., 1991a,b). The proteins were present in these walls as complexes, insoluble in a hot solution of 2% sodium dodecylsulfate (SDS), that
4
JOSEPH G. H. WESSELS
could be dissociated into monomers only by treatments with pure formic acid or trifluoroacetic acid, although monomers of these proteins were present in the medium of still cultures. Because of the abundance of hydrophobic residues and their presence in walls, we dubbed these proteins “hydrophobins”, a term used earlier to denote any substance conferring hydrophobicity to a microbial surface (Rosenberg and Kjelleberg, 1986). Around the same time, Stringer et al. (1991) found a gene in Aspergillus nidulans with homology to the S. cuinmune hydrophobin genes. Disruption of this gene caused a phenotype of wettable conidiospores from which the so-called rodlet layer was missing. This indicated hydrophobins as an essential component of hydrophobic rodlet layers, generally observed on fungal spores. We then showed that a single purified hydrophobin from S. cuininune (SC3) could form such a hydrophobic rodlet layer in vitm by self-assembly at a water-air interface (Wosten et ab, 1993), and that such a layer was formed at the surface of aerial hyphae (Wosten et al., 1994b).It was also found that this hydrophobin could mediate strong attachment of S. cuintnune hyphae to solid hydrophobic surfaces (Wosten et al., 1994a). In the meantime, hydrophobin-like proteins were found in all fungi examined (de Vries et al., 1993), while anonymous genes highly expressed in fungi during a variety of developmental processes turned out to encode proteins with clear homology to the S. cuminune hydrophobins (Fig. 1).
2. IDENTITY OF HYDROPHOBINS It is noteworthy that most of the hydrophobins listed i n Fig. 1 were found by sequencing cDNAs representing mRNAs abundantly expressed during certain stages of fungal development without knowing anything about the encoded proteins. Only ABHI, CoH1, cerato-ulmin and cryparin were first identified as proteins, and their genes then cloned by polymerase chain reaction (PCR) using degenerate primers based on determined N-terminal amino-acid sequences. In retrospect, the late discovery of these abundantly occurring proteins is understandable because many occur as SDS-insoluble complexes that can be dissociated into monomers only by using concentrated formic acid or trifluoroacetic acid (de Vries et af., 1993), agents not in common use for protein extraction. In principle, these proteins could have been seen when examining proteins present in media from standing cultures, but only after handling such media with special care because the hydrophobins easily aggregate upon exposure to air forming insoluble complexes. Precisely for this reason, cerato-ulmin (CU) and cryparin (CRYP) were detected early because, on shaking, these Class I1 hydrophobins formed a milky turbidity that could be dissolved in SDS. Yet, the fact that hydrophobin sequences are so readily found in screens for developmentally regulated sequences indicates that they are derived from the most abundantly expressed fungal genes. Indeed, the
5
HYDROPHOBINS
i t rl'
I'
Neumspm cmssa
Figure 1 Dendrogram of similarities between aligned hydrophobins obtained by the CLUSTAL programme of the PClGENE programs package, version 6.60 (Higgins and Sharp, 1988). Numbers in superscript indicate references where sequence information was gublished orrefer to unpublished data. 'hchuren and Wessels (1990);Wessels et al. (1991a). )Wessels et al. (1995). 3)S.A. Asgeirsd6ttir and L.A. Casselton (unpublished). 4)Martinet a1. (1995); Tagu et al. (1996). ')Lugones et at. (1996); de Groot et al. (1996). @StLeger et al. (1 992). 7)Talbotel at. (1 993). 8)Strjngerand Timberlake ( 1 995). 9)Strjngeret al. (1991); J. Rhodes and W.E. Timberlake, cited in Stringer and Timberlake (1995). '')Pam et a/. (1994); Thau et al. (1994). ")Bell-Pedersen et nl. (1992); Lauter el 01. (1992); Templeton et al. (1995). 12)Yaguchiet al. (1993); Bowden et at. (1993); Stringer and Timberlake (1993). 13)Loraet al. (1994). 14)Zhanget at. (1994); Carpenter et al. (1 992). 15)Nakari-Set;ilaet al. (1996). 16?. Nakari-Seala and M. Penttila (unpublished). Amino-acid sequences at the N-terminal end located before the first cysteine residue were omitted in the comparison since these include the signal sequence for secretion and in only eight cases (SC3, SC4, CoH1, ABHI, RodA, Eas, CU and CRYP) is the N-terminus of the mature protein known. However, aligning the whole protein sequence, including the signal sequence, results qualitatively in the same type of dendrogram, only the distances become larger. The overall identity of all sequences is only 4.3%, the overall similarity 1.7%.
SC3 and SC4 genes of S. coininune were shown to produce 1% and 3.596, respectively, of the mRNA mass at the tiine of emergent growth [Mulder and Wessels, 1986), while the record is probably set by the mRNA for cryparin that amounted to 25% of the mRNA mass (Zhang ef nl., 1994). The sequence diversity of hydrophobins (Fig. 1) means that isolation of
hydrophobin genes on the basis of sequence homology is mostly impossible. For instance, the four hydrophobin genes cloned from S. coininune do not crosshybridize (Mulder and Wessels, 1986), even under non-stringent conditions. Only in the case of related species has nucleic acid homology been used to isolate a hydrophobin gene that fulfils a similar function: the hydrophobin gene that is responsible for formation of rodlets on conidia of Aspergillus fumigatus was isolated on the basis of its homology to the KodA gene of A. nidulans (Parta et al.,
6
JOSEPH G. H. WESSELS
1994; Thau et al., 1994). This state of affairs means that it is generally unknown how many hydrophobin genes exist in a given fungal species, but the identification of multiple genes in species, such as S. cornrnune,I? tinctorius and A. nidulans, just by screening cDNA libraries, indicates that, in most studied species, only the most abundantly expressed hydrophobin genes may have been identified. Of the (putative) hydrophobins listed in Fig. 1, only SC3 (Wosten et al., 1993) and SC4 (this laboratory, unpublished data) from S. commune, ABHl from Agaricus bisporus (Lugones et al., 1996), CoHl from Coprinus cinereus (S.A. Asgeirsd6ttir and L.A. Casselton, unpublished data), CU from Ophiostorna ulmi (Takai and Richards, 1978; Russo et al., 1982) and CRYP from Cryphonectria parasitica (Carpenter et al., 1992) have been physically isolated and their properties studied. Wessels (1992) noted that the remarkable property of interfacial self-assembly exhibited by the SC3 hydrophobin of S. commune (see below) was earlier observed with CU (Takai and Richards, 1978; Russo etal., 1982)andCRYP (Carpenter et al., 1992). When the amino-acid sequence of CU became available (Yaguchi et al., 1993), Stringer and Timberlake (1993) noted the sequence homology to known hydrophobins. However, whereas interfacial self-assembly of, for instance, SC3, SC4 and ABHl hydrophobins results in aggregates that are highly insoluble in water, organic solvents and 2% SDS, the aggregates formed by CU and CRYP were found to be unstable in water, and soluble in aqueous ethanol and 2% SDS. In addition, they display a hydropathy pattern that is clearly different from that of hydrophobins like SC3 (Fig. 2). Therefore, Wessels (1994) proposed a distinction between Class I hydrophobins that form highly insoluble assemblages and Class I1 hydrophobins that form less stable assemblages (e.g. soluble in 60% ethanol or 2% SDS), a distinction supported by the alignment dendrogram shown in Fig. 1. In the Class I hydrophobins, the cysteine doublets are followed by a stretch of hydrophilic amino acids whereas, in Class TI hydrophobins, hydrophobic residues immediately follow the cysteine doublets (Fig. 2). Also, fewer amino acids separate the third and fourth cysteine residue in Class I1 hydrophobins than in Class I hydrophobins. Whether this grouping is correct can only be decided after isolation and characterization of all the listed hydrophobins. However, because most of the hydrophobins tabulated in Fig. 1 have not yet been physically isolated, they can be only tentatively grouped as Class I and Class I1 hydrophobins on the basis of similarities in hydropathy patterns and solubility characteristics of assemblages. It would not be surprising if some of these hydrophobins exhibit solubility characteristics intermediate between the two classes now distinguished. On the basis of the available information on hydrophobins, they would seem to have the following characteristics:
1. Hydrophobins are small proteins (100 k 25 amino acids) that are moderately hydrophobic. The hydrophobicity indices (Kyte and Doolittle, 1982) for mature proteins vary from 0.01 (RodA) to 0.60 (SC3). The overall hydrophobicity thus varies widely.
HYDROPHOBINS
7
ichi:ophvllum cummune SC3”
irhizophvl/um commune SCI”
Werorhizium onisupliue SSCA*’
Veuruspora cmssa Eaa”’
bfagnuyorrhe griseu MPG 1 ’I
Figure 2 Comparison of hydropathy patterns of selected hydrophobins (SC3, SCI, SC4, SSGA, Eas, RodA, MPGI, CU and CRYP) (for references, see Fig. 1). The patterns were determined using the parameters of Kyte and Doolittle (1982). A six amino-acid window was used and plotted against position in the deduced amino-acid sequence. The hydropathy patterns were then aligned around the first and second cysteine doublet, and around the fourth and eighth cysteine residue leaving gaps in the sequences where the hydrophobic regions (above the lines) alternate with hydrophilic regions. The hydrophobic amino-terminal sequences serve as signal sequences for secretion. The amino termini for the mature hydrophobins, when known, are indicated by arrows. Note that the first seven hydrophobins (Class I) have similar hydropathy patterns, which deviate from those of the last two hydrophobins (Class 11). (Modified from Wessels, 1994, with permission from the publisher.)
JOSEPH G. H.WESSELS
8 2.
3.
4.
5.
6.
Hydrophobins are all secreted as suggested by the presence of signal sequences. This was actually shown for those hydrophobins in which the amino terminus of the mature protein was determined (arrows in Fig. 2). Hydrophobins have a conserved spacing of eight cysteine residues: X2-38-C-X5-9-C-C-X 1~-~9-c-x~~2~-c-x~~~c-c-x~~~ g-C-X*-13 in which X signifies any other amino acid, except for tryptophan, which has been reported only in HydPtl, while methionine has been found only in HydPtl, HydPt2, Eas, SSGA and MPG1. Asparagine mostly follows the first cysteine doublet. Of course, the numbers of amino acids that separate the cysteine residues may change as more hydrophobins are sequenced but the recurrent hydropathy patterns around the sequence C-X5-9-C-€ in the amino-terminal and carboxy-terminal halves of the molecule are remarkable (Fig. 2). (Note that in the putative QID3 protein listed in Fig. 1, serine substitutes for the second cysteine residue.) Hydrophobins have poor amino-acid homology. For instance, the SC1, SC3 and SC4 hydrophobins, all produced by S. cornmune, are only 39% identical. However, many of the differences concern conservative substitutions so that the similarity between these hydrophobins becomes 80%. If the RodA hydrophobin ofA. nidulaiis and the Eas hydrophobin of N. c r a m are also taken into account, the identity between the five hydrophobins drops to 11% and the similarity to 34%. The similarities between the hydrophobins therefore become most clear when both the conserved spacings of cysteine residues and the hydropathy patterns are compared (Fig. 2). Hydrophobins have the capacity to assemble into an amphipathic protein film when confronted with a hydrophilic-hydrophobic interface, such as between water and air. As indicated above for the Class I hydrophobins, this was shown only for the hydrophobins SC3, SC4, CoHl and ABHl. However, the hydrophobins Eas (Templeton et al., 1995), MPGl (Talbot et al., 1993), RodA(Stringeret al., 1991) and HYPl (Partaetal., 1994; Thau et al., 1994) have all been shown to be part of, or constitute, the hot SDS-insoluble hydrophobic rodlet layer on conidiospores and thus most probably had gone through the interfacial self-assembly process. For the putative Class I1 hydrophobins, interfacial self-assembly has clearly been established for CU (Takai and Richards, 1978;Russo et al., 1982; Richards, 1993), CRYP (Carpenter etal., 1992) and HFBl (Nakari-Seda et af., 1996). As far as is known, all hydrophobins are present as assemblages on the surfaces of emergent hyphal structures.
These criteria delimit the hydrophobins from other cysteine-rich proteins of fungal or other origins. It has been suggested (St Leger el al., 1992; Templeton et al., 1994) that hydrophobins may be related to proteins exhibiting the so-called toxin-agglutinin fold (Drenth et al., 1980; Andersen et al., 1993). For the agglutinins belonging to the chitin-binding family (Raikhel et al., 1993) disulphide
HYDROPHOBINS
9
bridges occur between C1-424, C2-C5, C 3 4 6 and C7<8 within the 30-43 amino-acid motifs containing eight cysteine residues. For Class I hydrophobin SC3, it was found that no free SH groups were present in either the monomeric or assembled form, and thus that all cysteincs were probably involved in disulphide bridges (de Vries et al., 1993). For the Class I1 hydrophobin CU, Yaguchi et al. (1993) determined that disulphide bridges occur between Cl-C2, C3-124, C 5 X 6 and C7-428, emphasizing the recurrent motif in hydrophobins already suggested by the spacing of cysteine residues and hydropathy pattern (Fig. 2), and pointing to two-domain proteins. If this is the pattern of disulphide bridges found in all hydrophobins, then there is a clear difference in structure between hydrophobins and other high-cysteine proteins. Nevertheless, a comparison between agglutinins and hydrophobins is interesting since lectin activity has been demonstrated for cryparin (Carpenter et al., 1992). In view of the presence of assembled hydrophobins on the surface of walls that mainly consist of polysaccharides (Wessels and Sietsma, 1981), it would not be surprising if a lectin-like binding were involved in anchoring hydrophobins to the wall. Small extracellular proteins often contain 7-8 cysteine residues. Notable examples are the snake toxins and chitin-binding lectins (Drenth et al., 1980) referred to above, the thionins (Bohlmann and Apel, 1991), the extracellular lipid transfer proteins (Sterk et aL, 1991), the so-called defensins (Terras et al., 1995) and the hydrophobic parts of bimodular proteins (Castonguay ef al., 1994), all from plants. None of these have the specific spacings of cysteine residues as noted for the hydrophobins nor have they been shown to be active in self-assembly. For the same reasons, there is no compelling evidence at the moment to associate the hydrophobins with other secreted fungal polypeptides, such as toxins (Wnendt et al., 1994) or peptide elicitors (Nespoulous et al., 1992; van den Ackerveken et al., 1993; Rohe et al., 1995) as was proposed by Templeton el al. (1994) and Sticklen and Bolyard (1994). However, it should be mentioned that the NIP1 elicitor of Rhynchosporiurn secalis has the first eight of its ten cysteine residues exactly spaced as in hydrophobins (Rohe et al., 1995), emphasizing the possibility that hydrophobins or derivatives of hydrophobins may act as elicitors of the plant defence reaction (Wessels, 1994). There is an imminent need for probing the three-dimensional structure of hydrophobins before and after interfacial assembly. Whereas the assembly of Class I1 hydrophobins appears reversible, the assembly of the known Class I hydrophobins must be accompanied by a very stable conformational change, since it cannot be reversed except by using solvents like formic acid and trifluoroacetic acid (TFA). It should also be noted that these proteins are very robust and functionally unaffected by treatments with these aggressive reagents, since their dissociation and assembly can be repeated many times (Wosten et al., 1993). It is only through structural analyses that we can begin to understand the interfacial assembly of these proteins into an amphipathic film with a rough hydrophobic surface of rodlets, and
10
JOSEPH G. H. WESSELS
their remarkable property to attach to surfaces and to reverse their wettability (see below).
3. RODLETS The structures of concern here are 5-10 nm thick rodlets of varying length organized in bundles or fascicles, in which individual rodlets are laid down in parallel fashion within a single fascicle (Fig. 3). They were first seen by freeze+tching of conidiospores of ascomycetes (Sassen et al., 1967; Hess et aZ., 1968). They were also detected on hyphae and spores of zygomycetes (Cole et al., 1979; Hobot and Gull, 1981) and basidiomycetes (Bronchart and Demoulin, 1971; Wessels et al., 1972; Gerin et al., 1994). Honneger (1991) found similar rodlets lining the air spaces within thalli of many lichens. Wessels et al. (1972) found rodlets on hyphae of Schizuphyllurn cuinrnune decorating the outside of the outer wall layer of (1-3)-a-glucan (S-glucan). Because this glucan was the only component in the untreated wall that showed crystallinity by X-ray diffraction, they made the erroneous suggestion that the rodlets were an aspect of the crystalline S-glucan. Hashimoto etal. (1976) detached the rodlet layer from microconidia of Trichuphyton mentagruphytes, a dermatophyte, and they were the first to publish a careful chemical analysis. The rodlet layer was found to be remarkably resistant to dissolution by most common organic solvents, cell-wall
Figure 3 Morphological appearance of assembled SC3 hydrophobin. (a) Freezefracturing and shadowing shows the typical rodlets of assembled SC3 at the surface of aerial hyphae. (b) and (c) Similar rodlets are observed after drying down solutions of pure SC3 and shadowing the preparations; at 3.5 pg cm-I the rodlets are shorter (b) than at 0.35 pg cni-' (c). Bar represents 100 nm. Arrows indicate direction of shadowing. (From Wosten et ol., 1994b. with permission from the publisher.)
HYDROPHOBINS
11
lytic enzymes, detergents, mild acids and alkali treatments, but was solubilized in boiling 1 M NaOH. Protein (SO-S5%) and glucomannan (7-10%) were found to be the major components of the rodlet layer, in spite of its resistance to degradation with proteases such as trypsin and pepsin. Beever et al. (1979), analysing the chemical nature of the rodlet layer of Neurospora crassa macroconidia, also noted that this layer was extremely insoluble, and likewise found mainly protein (91%) and a small amount of carbohydrate and lipid. They regarded the small amount of lipid and the rough surface of the rodlet layer responsible for the non-wettability of the surface of the conidia. Similar properties were found for the rodlet layer on conidia of Aspergillus niger (Cole et al., 1979) and Aspergillus nidulans (ClaverieMartin et al., 1986), while Hobot and Gull (1981) appear to be the first to show a protein subunit as part of the rodlet layer. Working with the zygomycete, Syncephalastrurn racemosum, they found that alkali extraction removed the rodlet layer from the spores and that a 70% ethanol supernatant of such an extract contained a glycoprotein that ran at approximately 12 kDa on SDS-polyacrylamide gel electrophoresis (SDS-PAGE).By using the method of dissociation of assembled hydrophobin withTFA(de Vries etul., 1993), Templeton et al. (1995) have recently shown that the rodlet layer isolated from N. crassa conidia is predominantly composed of one protein, the product of the hydrophobin gene eas. Similarly, Bidochka et al. (1995a) found a small protein as constituent of the rodlet layer of spores of the entomopathogenic Beauveria bassiana. N-terminal sequencing of the protein showed the typical C-X,-C-C-N motif present in hydrophobins. Freezeetching has revealed similar rodlet layers on the spores of aerobic bacteria (Holt and Leadbetter, 1969; Aronson and Fitz-James, 1976) and very distinctly on aerial hyphae and spores of Streptornyces species (Wildermuth et al., 1971; Williams et al., 1972). It would be extremely interesting to know whether these structures, which probably fulfil the same functions as the rodlet layers of fungi, are composed of hydrophobins or if these rodlets are built of a different material and represent an example of convergent evolution. Unfortunately, very little is know about the chemistry of these prokaryotic rodlet layers. Smucker and Pfister (1978) examined the chemistry of rodlets in the actinomycete Streptornyces coelicolor and suggested that they were composed of a polysaccharide complex most similar to chitin. However, the rodlets on the spore coats of aerobic eubacteria (e.g. Bacillus cereus) were considered to consist of cysteine-rich protein that resisted SDS extraction (Aronson and Fitz-James, 1976). As mentioned in the Introduction, a strong indication for involvement of hydrophobins in the formation of the rodlet layer on fungal spores came from experiments in which anonymous genes, isolated on the basis of high expression during conidiogenesis, were inactivated by targeted mutations. This was first shown for Rod4 in Aspergillus nidulans (Stringer et al., 1991) and subsequently for eas in Neurosporu crassa (Bell-Pedersen et al., 1992; Lauter et al., 1992). In these cases the targeted mutation not only removed the rodlet layer from the spores but also caused the easily wettable phenotype preventing the spores from being
12
JOSEPH G. H. WESSELS
dispersed through the air. The known allele of N. crassa called eas (easily wettable) with the same phenotype (Selitrennikoff, 1976; Beever and Dempsey, 1978) could be complemented by the isolated genes ccg-2 (Bell-Pedersen era[., 1992) and bli-7 (Lauter el al., 1992), both allelic to the ens gene. The homologue of Rod4 in the human pathogen Aspergillmfurnigutm was isolated and found to complement the targeted mutation rodletless in A. nidulans (Parta et al., 1994) whereas a targeted mutation in A. furnigatus itself showed the same phenotype as rodletness in A. nidulans (Thau el al., 1994). Significantly, the spores of this mutant were as pathogenic for mice as those from the wild-type strain. The aforementioned genetic experiments only show that hydrophobins are necessary for the formation of rodlets and probably constitute at least one component of the rodlet layer. It has now been shown, however, that single hydrophobins can form rodlet layers in vitro. Purified monomers of SC3 hydrophobin of Schizophyllutn cotnrnune in aqueous solution immediately assembled around air bubbles or oil droplets coating these with an SDS-insolublefilm (Wosten etal., 1993,1994a). Upon freeze-fracturing, the side facing the hydrophobic phase showed rodlets that could not be distinguished from those seen on aerial hyphae. Simply drying down an aqueous solution of SC3 on a Formvar grid for electron microscopy, exposing the hydrophobin to a water-air interface, also revealed the rodlet layer after shadowing (Fig 3; Wosten et ul., 1993, 1994b). Significantly, the hydrophobicity of the air-exposed surface of the SC3 film was as high (water contact angles of about llOo) as that of the surface of aerial hyphae. This process of interfacial self-assembly of a single hydrophobin into a hydrophobic rodlet layer provides for a remarkably simple mechanism by which hyphae and spores obtain a hydrophobic layer at their surface because it is at this surface that the secreted hydrophobin monomers reach the water-air interface and assemble into an amphipathic film (Wosten et al., 1994b). Apart from SC3, the SC4 hydrophobin of S. corninune also forms a hydrophobic rodlet layer in vitro (this laboratory, unpublished data), as does the ABHl hydrophobin of A. bispoms (Lugones et al., 1996). CoHl also shows interfacial self-assembly (S.A. Asgeirsdbttir, unpublished data) but, as far as is known, none of the other Class I hydrophobins listed i n Fig. 1 has been tested for self-assembly and formation of rodlets in vitro. Disruption of Rod4 in A. nidulans caused complete absence of rodlets on conidiospores and an easily wettable phenotype resulted (Stringer et al., 1991). Yet, a more recent study (Stringer and Timberlake, 1995) showed that such conidia still contain a hydrophobin (DewA) on their surfaces. Apparently DewA alone is not able to produce rodlets on the spores. Disruption of only DewA caused the spores to become wetted by a solution containing both 0.2% SDS and 50 m~ ethylenediamine tetraacetic acid (EDTA), while disruption of both Rod4 and DewA caused a higher hydrophilicity of the spores than disruption of RoaX alone. After removing a hydrophobin-containing rodlet layer from conidia of Beauveria Oassiana (Bidochka et al., 1995a) an SDS-insoluble but formic-acid-extractable protein of low molecular mass
HYDROPHOBINS
13
(15.4 kDa after oxidation with performic acid) remained (Bidochka et al., 1995b). Since N-terminal amino-acid sequencing proceeded for 24 amino acids only, it was impossible to know whether or not this protein represented a hydrophobin. If so, it was apparently not present in a rodlet layer. Although the Class 11hydrophobins cerato-ulmin and cryparin are secreted into the culture medium, they also cover aerial structures in Ophiostorna ulrni (Takai and Hiratsuka, 1980; Svircev et al., 1988) and Cryphonectria (Carpenter et al., 1992), respectively. The Occurrence of rodlets on surfaces of these fungi was not reported. Therefore, it is uncertain at the moment whether rodlets are a general aspect of assembled hydrophobins. Interfacial self-assembly resulting in a rodlet layer at a water-air interface, as evident for the SC3 hydrophobin (Wosten et al., 1993), would only be expected to occur in aerial structures. Indeed, submerged hyphae of S. commune, which produce hydrophobins, secrete these into the medium as monomers (Wessels et al., 1991a,b; Wosten et al., 1994b). Also, on submerged produced conidia of A. nidulans, rodlets were not detected, although the R o d transcript was produced, suggesting diffusion of the hydrophobin into the medium (Stringer el al., 1991). Mufioz et al. (I 995) found submerged conidiospores of Trichoderma harzianum to be hydrophilic but aerial spores were hydrophobic, and only these contained a putative hydrophobin. In contrast, both aerial and submerged conidia of Reauveria bassiana contained hydrophobin (Bidochka et al., 1995a). From the paper it is not entirely clear whether rodlets, and thus the assembled form of the hydrophobin, were actually seen on the submerged spores. Rodlet layers could have formed from hydrophobin monomers assembled on air bubbles swirled into the medium during shaking (Wosten et al., 1993).
4. SURFACE ACTIVITIES OF HYDROPHOBINS
4.1. SC3 Hydrophobin
Most of the Class I hydrophobins listed in Fig. 1 are known to be produced only in specific structures or under particular environmental conditions. For instance, transcription of R o d (Stringer et al., 1991) and DewA (Stringer and Timberlake, 1995) in Aspergillus nidularzs is probably restricted to phialides that produce the conidiospores. Although species like A. nidulans, A. niger, A. oryzae, Neurospora crassa and Perzicillium chrysogerzuriz do produce hydrophobin-like proteins typical for Class I in the culture medium (de Vries et al., 1993; unpublished data), the amounts found were much less than in Schizophyllumn commune.
14
JOSEPH G. H. WESSELS
On account of its abundance, the SC3 hydrophobin could be easily purified from the medium of 5-day-old standing or shaking cultures of the monokaryon of Schizophylluin commune. Although the fungus harbours at least three other hydrophobin genes (Fig. l), the latter are only substantially active in the dikaryon (Mulder and Wessels, 1986) unless the monokaryon carries alleles that induce monokaryotic fruiting (Yli-Mattila et aE., 1989a; see Table 1). SC3 is the most prominently secreted protein of the monokaryon while the dikaryon mainly secretes SC4 together with varying amounts of SC3 (Wessels et al., 1991a). Irrespective of culture conditions, the genes for these hydrophobins become active after 2-3 days of cultivation (Wessels et al., 1987) and the hydrophobins are secreted into the medium. In standing cultures, in which emergent growth occurs, the hydrophobins in the medium accumulate largely in a soluble state; in shaken cultures they mostly occur as very fine insoluble particles because of aggregation at the increased medium-air interface (Wessels el al., 1991a,b;Wosten etal., 1993).In both cases the mycelium can be filtered off on nylon cloth leaving the hydrophobins in the filtrate. They are then purified essentially as described by Wosten et al. (1993,1994b). After mixing with air or heating the medium to lOO"C, to achieve complete aggregation, the aggregates are spun down, treated with concentrated TFA and, after removing TFA by evaporation, the dissociated hydrophobins dissolved in water. After removing particulates, the solution is made 60%in ethanol leaving the monomeric hydrophobins soluble, but polysaccharides and a 15 kDa contaminating protein are precipitated (Wosten et al., 1993).Further purification is achieved by precipitation of the hydrophobin at higher ethanol concentration, dissolution of the precipitated monomers in water and repeating the procedure of interfacial precipitation and TFA dissociation. Using the standard monokaryon 4-39, about 20 mg of purified SC3 is obtained per litre of medium. The protein can be stored for some time in 60% aqueous ethanol keeping it in monomeric form. For assembly, this solution can be diluted with degassed water to lower the ethanol concentration to less than 5%. For experiments involving radioactive SC3, "SO$- is included in the culture medium. SC3 was shown to be a glycoprotein that stained in a periodic acid-Schiff reaction, bound the lectin conconavalin A, and contained mannose. According to its protein sequence, one would predict a molecular mass of 9830 Da but mass spectroscopy showed 14 200 Da (this laboratory, unpublished data). Assuming no other post-translational attachments, the presence of about 23 mannose residues can be calculated, probably linked to the abundantly occurring serine and threonine residues since no putative N-glycosylation site is found in SC3 (Schuren and Wessels, 1990). In SDS-PAGE the SC3 hydrophobin runs slower than expected, namely at a position corresponding to marker proteins of 24 kDa; after oxidation with performic acid, disrupting disulphide bonds by oxidizing cystine to cysteic acid, it migrates at a position corresponding to 28 kDa. When bubbling air or nitrogen gas through an aqueous solution of purified SC3, or simply by shaking or vortexing the solution, a milky suspension is obtained that
HYDROPHOBINS
15
shows irregularly shaped gas vesicles in the light microscope (Wosten et al., 1993). Upon standing, these vesicles float to the surface; by applying a vacuum they collapse, leaving aggregated SC3 that can be easily centrifuged down and which is insoluble in 2% SDS at 100°C. This suggested that the gas vesicles are coated with a highly stable film of assembled SC3. Apparently this is accompanied by a considerable lowering of the surface tension of the water, since the gas vesicles attain various odd shapes. Films of about 10 nm thickness could be visualized by sectioning the precipitate obtained by bursting coated air vesicles, while freezefracturing and direct surface shadowing revealed the typical pattern of rodlets seen on the surface of aerial hyphae and discussed in the previous section. SC3. which had been assembled by drying down an aqueous solution on a hydrophilic glass surface (water contact angle 15") produced a hydrophobic surface corresponding to water contact angles up to 95". Solutionsdried on the surface of the thin mutant of S. commune, a mutant not producing SC3 nor aerial hyphae, and having a hydrophilic surface, produced a surface hydrophobicity of 110". close to the value of 115" found for the surface of the wild-type monokaryon (Wosten et al., 1993). The presence of a water-air interface apparently leads to a stable conformational change that links the SC3 molecules tightly to each other in an amphipathic two-dimensional film, which exposes the typical fascicles of rodlets at its hydrophobic side. Indeed, if only protein and carbohydrate are present and no lipids, then this conformational change must result in an orientation of polar and apolar amino acids to different sides of the film, the mannose residuesprobably being exposed at the hydrophilic side. Because gasses have a hydrophobic character, the ability of an oil suspension in water to provide a suitable interface for SC3 assembly was investigated. Indeed SC3 stabilized oil droplets in water by coating these with a 10 nm thick SDS-insoluble protein film (Wosten et al.. 1994a). Again, the shape of the oil droplets indicated a considerable drop in the surface tension. Freezeetching and shadowing revealed that the amphipathic film exhibited rodlets on its hydrophobic side but had a smooth appearanceat its hydrophilic side. Immersing a hydrophobic sheet of plastic like Teflon into an aqueous solution of SC3 caused SC3 to assemble on the hydrophobic surface. After removal of the Teflon sheet from the solution. water flowed evenly over its surface, indicating that it had become hydrophilic. After drying it was found that water contact angles (0) on the surface had decreased from 108" to 48"and that 5.9 x 10l2molecules SC3/cm2had adsorbed. Treatment with 2% SDS for 10 min at 100°C. removed only 13.5% of the adsorbed hydrophobin, while water contact angles rose slightly to 62". This contrasts with other proteins, such as bovine serum albumin, which also adsorb to plastics but are completely removed by hot 2% SDS. Adsorption of SC3 could occur from very dilute solutions. At 2 pg m1-l. saturation was reached after 16 h but at 20 pg m1-l. saturation of the surface was reached after only 2 min of incubation. It was concluded that the exposure of SC3 to the hydrophilic-hydrophobic interface induces a conformational change in the
16
JOSEPH G . H. WESSELS
hydrophobin monomers leading to their assembly into a stable amphipathic film strongly attached with its apolar groups to the hydrophobic plastic and exposing polar groups at the surface (Wosten et al., 1994a). Materials with lower surface hydrophobicities than Teflon, when immersed in an aqueous SC3 solution, caused fewer molecules of SC3 to assemble at their surfaces (Wosten et al., 1994a). To investigate whether the degree of hydrophobicity was the only factor determining assembly, a continuous hydrophobicity gradient surface, displaying water contact angles ranging from 20" up to 107", was made by coating glass with dichlorodimethylsilane according to Elwing ef al. (1987). The amount of assembled SC3, defined as SC3 on the surface becoming insoluble in hot SDS after immersing the gradients for 16 h in SC3 (2 pg ml-'),sharply increased in the region of the gradient surface displaying advancing water contact angles between 60" and 90", then more slowly to the 107" region, i.e. the hydrophobic end. Here, the absorbed SC3 decreased the advancing water contact angle from 107" on the bare gradient surface to 60" on the protein-coated surface (to 39" before extraction with SDS) (Wosten et al., 1995). It is thus clear that the interfacial tension at the solid-liquid interface (ys~)is the major factor that induces SC3 assembly. The surface activity of SC3 was compared with that of other proteins using the method of axisymmetric drop shape analysis by profile (ADSA-P) (van der Vegt et al., 1996). In this method (Rotenberg et al., 1983; Noordmans and Busscher, 1991), a drop (100 PI) of water or buffer in which the protein is dissolved is placed on the surface of fluoroethylenepropylene (FEP-Teflon), and changes with time with the shape and contact angle of the axisymmetric droplet are recorded, allowing calculations of the interfacial tension changes to the hydrophobic solid (ys~)and to the water-air (vapour) interface (xv). SC3 at 100 pg ml-' buffer (10 m~ KPi, pH7) caused a large drop in the liquid surface tension (~Lv) from 72 to 43 mJ m-2 and even to 32 mJ mW2,when the hydrophobin was dissolved in water. At this low concentration of protein, the drop in surface tensions caused by other proteins examined (bovine serum albumin, human immunoglobin G, chicken egg white lysozyme, bovine pancreatic ribonuclease A, bovine milk a-lactalbumin) was much less, reaching a minimum of 54 IT m-2 with lysozyme. SC3 thus proves to be a powerful surface-active protein, particularly when dissolved in pure water. The kinetics of change in surface activity suggests that the large drop in ~ L Vis mainly caused by a conformational change in the protein. Surprisingly, the surface tension at the solid-liquid interface ( y . ~did ) not decrease, as expected from the measurements of Wosten et al. (1994a, 1995), and observed for all other proteins tested, but slightly increased from about 42 to 43 mJ m-2 with fluidcontact angles increasing from 111" to 122". Adsorption of SC3 to the Teflon thus made the surface more hydrophobic and not hydrophilic as expected. This was explained by assuming that, in this case, the conformational change caused by adsorption of an SC3 layer to the Teflon leads to adsorption of a second layer that exposed its hydrophobic side to the aqueous solution. This second layer would be
HYDROPHOBINS
17
sheared from the bottom layer when removing the Teflon sheet from the aqueous solution so that, after drying of the Teflon, a hydrophilic surface remains as detected by water contact angles (van der Vegt et al., 1996). The atomic composition of microbial surfaces can be analysed by X-ray photoelectron spectroscopy (XPS) (Rouxhet et al., 1994). XPS is based on irradiating a surface with X-rays and analysing the kinetic energy of the photoejected electrons. This provides an elemental surface analysis with an analysed depth in the nanometre range. Since the photoelectron kinetic energy depends on the chemical state of the element, different functional groups can also be distinguished. Between the third and sixth day, water contact angles measured at the surface of an S. commune monokaryon rose from 30" to 125" and XPS showed this to be accompanied by a rise in the N/C ratio from 0.08 to 0.15, and a rise in the S/C ratio from 0.002 to 0.007 (Wosten et al., 1994~).However, these values remained lower than those measured on artificially assembled SC3, possibly owing to absorption of extraneous materials to the hydrophobic hyphae. Measurements on SC3 films assembled in v i m (Wosten et al., 1994c) showed an atomic composition at the hydrophobic side (0 9.5") close to that predicted from the known amino-acid composition. Only the O/C ratios were higher than the values calculated from the polypeptide chain but this could be accommodated by assuming the presence of 11 anhydroinannose molecules in this glycoprotein. XPS of the SC3 film assembled on polytetrafluoroethylene (F'TFE, Teflon), exposing the hydrophilic side (0 48"), showed N/C, S/C and C=O/C ratios, which were significantly lower than those at the hydrophobic side while the N/S ratios were similar. This would indicate an orientation of peptide bonds and amino-acid chains towards thc hydrophobic side and possibly sugar residues oriented towards the hydrophilic side. This emphasizes the amphipathic nature of the SC3 film. Since experimental XPS values obtained at the hydrophobic side of assembled SC3 were similar to those expected for the whole protein, emitted photoelectrons must have originated from all parts of the 10 nm-thick film. However, few electrons were expected to be emitted after excitation from a depth exceeding 5 nm (Rouxhet et al., 1994). Possibly the serrated hydrophobic surface caused by the presence of rodlets in random orientation allowed for emission of photoelectrons from all parts of the film while the smooth hydrophilic surface would prevent photoelectrons from deeper parts from escaping (Wosten etal., 1994~). It has been suggested that the topography of the rodlet surface plays a significant role in decreasing its wettability (Fisher er al., 1978; Beever et al., 1979). If no lipids are attached, it is indeed remarkable that the apolar side chains of hydrophobic amino acids in a hydrophobin such as SC3 could produce surfaces showing water contact angles of 95" when assembled on glass (Wosten et al., 1993) or even 122' when assembled in a double layer on Teflon (van der Vegt et al., 1995), while water contact angles on a bare Parafilm or Teflon surface measure 105-108". Although surface roughness clearly influences surface wettability (Huh and Mason, 1977; Hazlett, 1992), measurements of Busscher et al. (1984) have
JOSEPH G. H.WESSELS
18
indicated that water contact angles increase only if the roughness of a surface is caused by structures exceeding 100 nm in size whereas the hydrophobin rodlets show a periodicity of about 10 nm. Nevertheless, it would be interesting to see whether water contact angles would increase by etching a rodlet pattern on a hydrophobic surface, when compared to a smooth surface of the same material.
4.2. Cerato-ulmin
After Zentmeyer (1942) indicated a toxin of Ophiostoma (Ceratocystis) ulmi as responsible for causing Dutch elm disease, Takai (1974) was the first to propose a small protein produced in the culture medium, named cerato-ulmin (CU) (Fig. l), as the phytotoxin. This was mainly based on a correlation of cerato-ulmin production in shaken cultures and aggressiveness of the isolated strains (Takai, 1974, 1980;Brasier etal., 1990). Although almost half of the amino-acid sequence of cerato-ulmin was already known in 1979 (Stevenson et al., 1979), it was not until 1993 that the complete sequence of the 75 amino acids constituting the protein was reported (Yaguchi et al., 1993). On the basis of this information, Bolyard and Sticklen (1992) assembled a gene that could be expressed in E. coli to give approximately SO pg protein per litre, while Bowden et al. (1993) cloned the gene and determined the complete sequence including the signal sequence for secretion
(Fig. 2). Surprisingly, disruption of the gene for cerato-ulminin an aggressive strain
PB. i il, d
, . . u m t a no e ect on the * patkogenicity of the fungus (Bowden ef al., 1996). The remarkable properties of cerato-ulmin were described by Takai and Richards (1978) and Russo et al. (1982), and more recently sumrnarised by
o
iv
Richards (1993). The protein is very surface active and, as the SC3 hydrophobin, assembles around gas bubbles at concentrations in the nanogram per millilitre range. This property was therefore used to isolate and purify the protein from the
medium in which it accumulated at up to 25 pg rn1-l. Unlike SC3, however, the presumedly amphipathic films that formed were unstable. The milky solution that arose after shaking or bubbling (containing odd-shaped coated air vesicles described as (‘rods” and “fibrils”) turned clear on centrifugation or applying positive pressure. Apparently the assemblages went into solution, although microscopic “units” were still detected and the real solubility of the protein was considered low. Whereas films formed by SC3 are extremely stable to 60% ethanol or 2% SDS, these solvents readily dissolve assembled cerato-ulmin. Cryparin, produced in abundance in the culture medium by the phytopathogen Cryphonectria parasiticu bahavad similarly (Carpenter er al., 1992). Because of their solubility characteristics, these proteins were called Class I1 hydrophobins (Wessels, 1994), and it is interesting to note that their structure and hydropathy pattern are also somewhat different from that of the Class I hydrophobins (Figs 1 and 2). However, how these differences translatc into a causal relationship between protein structure and stability of the assembled films is unknown.
HYDROPHOBINS
19
Both cerato-ulmin (Takai and Hiratsuka, 1980; Svircev et al., 1988) and cryparin (Carpenter et al., 1992) occur in abundance on aerial hyphae and sporulating structures of surface cultures. As the Class I hydrophobins, their function may be related to the emergence of these structures. Only colonies of strains of C. parasitica (Carpenter etal., 1992) and 0.ulmi (Takai, 1980) with a high production of Class I1 hydrophobins appeared fluffy and aerial. Bowden et al. (1996) found that disruption of the cerato-ulmin gene caused a phenotype with sparse production of aerial hyphae and a less hydrophobic surface than wild-type. Supposedly, when present at the wall-air interface, the assembled films of Class I1 hydrophobins would present their hydrophobic side to the air and thus might be washed away less easily than their solubility in water suggests. In addition they might be anchored to the wall polysaccharides by a lectin-like activity as indicated for cryparin (Carpenter et ul., 1992). Whether or not a rodlet structure is present at these hydrophobic surfaces seems not to have been investigated. 5. FORMATION OF EMERGENT STRUCTURES
5.1. Formation of Aerial Hyphae Using in vitro translation of mRNAs, Dons et al. (1984) found the SC3 gene (then not known to encode a hydrophobin and called ZDZO) to be perhaps the most abundantly expressed gene at the time of emergence of aerial hyphae in the monokaryon of Schizuphylluni cumrnune. Subsequently, several observations on the regulation of this gene suggested an involvement in the formation of aerial hyphae. 1. Formation of aerial hyphae by the monokaryon after 2-3 days in surface culture always coincides with a rise in SC3 mRNA (Mulder and Wessels, 1986) owing to transcriptional activation of the gene (Schuren et al., 1993~).
2. The frequently occurring spontaneous and recessive thn mutation, which blocks all emergent growth (i.e. formation of aerial hyphae in the monokaryon and, when homozygous, formation of both aerial hyphae and fruit bodies in the dikaryon) suppresses expression of SC3, indicating that THN is required for expression of SC3 (Wessels et al., 1991b). 3. A heterokaryon containing nuclei with different B-mating-type genes but common A-mating-type genes (a common-MATA heterokaryon), typified by septa1 dissolution, continuous nuclear migration and few aerial hyphae (Raper, 1966),has a very low expression of SC3 (Asgeirsd6ttir etul., 1995). It was earlier found that the same phenotype occurs in a homokaryon (a mycelium with only one genetic type of nucleus) carrying a so-called constitutive mutation in the MATB gene (ImtBCon).This mutation also causes down-regulation of SC3 (Ruiters et al., 1988). Together, these
20
JOSEPH G. H. WESSELS
studies indicate that the MA7B genes are involved in regulation of SC3. However, assigning a role for SC3 in formation of aerial hyphae was problematic because transcription of the gene was not confined to aerial hyphae but also abundantly occurred in submerged growing hyphae even in shaken cultures in which aerial growth could not occur (Wessels et al., 1987). The solution to this problem came with the identification of the product of the SC3 gene as a hydrophobin (Schuren and Wessels, 1990; Wessels et al., 1991a,b). It was then realized that it was not so much cell-type specific gene activation that brought about cell differentiation but rather the deposition of an insoluble form of the hydrophobin in the wall. When it was found that soluble SC3 hydrophobin monomers could assemble at a water-air interface into an SDS-insoluble amphipathic protein film exposing rodlets at its hydrophobic side (Wosten et al., 1993, 1994b),a mechanism as shown in Fig. 4 became obvious. That SC3 is instrumental in the generation of hydrophobic aerial hyphae was proved by disrupting the SC3 gene (Wosten et at., 1994a). The mutant monokaryon, now unable to form SC3 hydrophobin, made few aerial hyphae compared to wild-type, at least in Petri dishes that were tightly sealed. In non-sealed Petri dishes, aerial hyphae did form but these had a hydrophilic surface (van Wetter et ui., 1996). The scheme given in Fig. 4 assumes that, in a hypha that breaches the interface of culture medium and air, the hydrophobin monomers can no longer freely diffuse into the medium but remain in the wall. Those reaching the wall-air interface assemble into an insoluble SC3 monolayer coating the hypha and conferring hydrophobicity to the surface. Here the assembled hydrophobin could be immunolocalized (Wosten et al., 1994b). Whether there are special mechanisms that direct hyphal tips to grow towards the air and cause them to breach the surface is unknown. Anyway, once this happens the hyphal surface, including the tip, is probably quickly covered with assembled SC3 and the hypha is irreversibly determined for growth into the air because of the hydrophobicity of its surface. Note that at the growing apex expansion of the SC3 assemblage could occur by continuous intercalation of SC3 monomers into the SC3 film. Implicit in the above scheme is that hydrophobins are secreted at hyphal tips. A correlation between apical wall growth and protein secretion was shown by Wosten et al. (1991) and more specifically the secretion of SC3 hydrophobin was shown to occur at growing hyphal tips (Wosten et al., 1994b). Recently, we showed that apical wall growth does indeed occur in hyphae emerging into the air (this laboratory, unpublished data). In order to assemble at the outer surface, the hydrophobin monomers have to be translocated over the wall. Even in the absence of a steep diffusion gradient in these aerial hyphae, as in hyphae growing submerged, this could be achieved by a mechanism proposed by Wessels (1990, 1994). According to this “bulk-flow” mechanism, proteins are not so much diffusing through pores in the wall but flow from the inside to the outside of the wall together with visco-elastic pol ysaccharides before these become cross-linked
HYDROPHOBINS
21
Figure 4 Proposed model for the formation of the hydrophobic rodlet monolayer at the surface of aerial hyphae. Hydrophobin monomers are indicated as grey spheres and hydrophobins in self-assembled films as oval structures with black (hydrophobic) and white (hydrophilic) halves to indicate the conformational change they have undergone. The arrows within the hyphae symbolize transport of newly synthesized SC3 hydrophobin monomers to the growing tip, probably via exocytotic vesicles, where the hydrophobin is secreted. In submerged hyphae, the hydrophobin monomers diffuse into the medium. In aerial hyphae the secreted monomers self-assemble when they reach the interface between the hydrophilic wall and hydrophobic air. Here they form an insoluble amphipathic monolayer with the hydrophobic rodlet-decorated side facing the air. SC3 hydrophobin monomers present in the medium may assemble at the surface of the medium or on hydrophobic walls of the culture vessel. (From Wosten et al., 1994b. with permission from the publisher.)
and crystalline to form the rigid wall. By this mechanism the contents of vesicles fusing with the plasma membrane at the very apex would be completely carried to the surface of the wall. However, it is improbable that the amount of SC3 thus secreted would precisely equal the need to cover the hyphal surface with a monolayer of assembled SC3. Wosten et ai. (1994b) showed that drying an SC3 solution on a glass slide caused the generation of a monolayer of assembled SC3 at the water-air interface but, because this layer at the same time destroyed the interface, a variable amount of SC3 monomers - depending on their concentration -remained between the glass and the film. One would therefore expect the presence of monomers of SC3 in the wall, in addition to the assembled SC3 at the surface. However, only the latter could be immunolocalized (Wosten et a]., 1994b). Possibly the wall fabric prevented the antibodies from reaching their target. It is interesting to note that the SC3 hydrophobin of Schizophyllurn commune is related more closely to the CoHl hydrophobin of Coprinus cinereus than to the other hydrophobins (SC1, SC4, SC6) ofS. curnrnune (Fig. 1).As SC3, CoHl seems to be required for formation of aerial hyphae; a mutant that does not form aerial hyphae nor aerial oidia does not express CoHl. Since the two species are not closely related, this would suggest that the relatedness of SC3 and CoHl is based on similar
22
JOSEPH G.H. WESSELS
functions, and that the other hydrophobins of S. commune are tailored for different purposes. If the SC3 hydrophobin were needed only for the emergence of aerial hyphae, one would not expect this protein to be secreted in such large quantities by submerged growing hyphae. Of course, it is possible that this is an artefact of the unnaturai way of cultivating the fungus in a liquid medium (whether gelled with agar or not). Very few fungi live in water (the water moulds belong to the oomycetes and are not true fungi, see e.g. Cavalier-Smith, 1993). Certainly, wood-rotting species such as S. cominune do not grow naturally in water but in wood. A hydrophobin like SC3 could attach a S. commune hypha tightly to a hydrophobic substrate by assembling at the interface of the hydrophilic wall and the hydrophobic solid (Wosten et al., 1994a). Wood contains hydrophobic components such as lignin. It is therefore possible that production of a hydrophobin like SC3, by assimilating substrate hyphae, serves an important function because it attaches the fungus to its natural substrate. A similar function, apart from conferring hydrophobicity to aerial structures, could be assigned to Class I1 hydrophobins produced in abundance in the culture media of Ophiostorna ulnzi and Cryphonectria parasitica, both tree pathogens.
5.2. Formation of Fruit Bodies
The large fruit bodies of the homobasidiomycetes (mushrooms, brackets, toadstools) and of some ascomycetes (discomycetes, such as the morels and truffles) are the most spectacular emergent structures elaborated by fungal hyphae. Their morphogenesis has been studied both from a theoretical and a practical point of view (see Wessels, 1993b), the latter because some of them, such as fruit bodies of Agaricus bisporus, Lentinus edodes and Pleurotus ostreatus are an important agricultural product. Schizuphyllurn commune, though commercially worthless, is by far the easiest to fruit in the laboratory and therefore is genetically the best studied homobasidiomycete (Raper, J.R.,1966; Raper, C.A., 1988). As in most homobasidiomycetes, fruit bodies of S. commune are generally formed on a secondary mycelium, that is a heterokaryon arising after mating two primary mycelia originating from spores (homokaryons that have one nucleus per cell and thus are also called monokaryons). The secondary mycelium mostly contains hyphal compartments with two nuclei of opposite mating type lying closely together, and hence it is called a dikaryon. During nuclear division in apical cells, a clamp connection is formed at the septum that divides the two new cells; clamp connections are thus a diagnostic feature for the secondary mycelium. The whole mating process, the maintenance of the dikaryon, and the formation of fruit bodies is controlled by the mating-type genes (Table 1). Only when the two mating mycelia carry differentMATA and differentMATB genes can sexual morphogenesis proceed. If only the MATB genes are different, nuclei are exchanged, septa are
23
HYDROPHOBINS
Table 1 Gene expression and emergent growth in Schirophyllum commune. m R N A from
Genotype MATAx MATBx MATA# MATB# MatAConmaBCon MATA= M A T W MATAx matBCon MATA# MATB= matACon MATEX MATAx MATBx thn MATA# M A m # thdthn MATAx MATBx$f matAcon MATBc " f b f MATA# MATB#jbflfif MATAx MATBx mfa
SC3
SCi, SC4, SC6, SC7, SC14
+* + +
-*
-
-
-
+ +
-
+ + + +
+ + -
-
-
-
+
Aerial hyphae
+ + +
-
+ +
-
-+ + -+ +
Fruit bodies -
+ +
-
-
-
+
* The + sign indicates an abundance of m R N A s ranging from 0.07% (for SC14) to 1.0% (SC3) and 3.5% (SC4) of the total mRNA mass. The - sign indicates an abundance of less than 0.01%, except for SC4, where it indicates less than 0.1%.
dissolved and nuclear migration ensues. The resulting heterokaryon has a flat morphology (few aerial hyphae) correlated with repression of the SC3 hydrophobin gene (Asgeirsdbttir et al., 1995). If only the MATA genes are different, then primary mycelia fuse - a process independent of the mating-type genes -but nuclei do not migrate. The formation of clamp connections is, however, initiated but not completed; they do not fuse with subterminal cells resulting in so-called pseudoclamps. None of these semicompatible matings produces fruit bodies; this apparently requires the presence of different MATA and MATA genes. Although for simplicity the genetic entities governing mating are referred to as MATA and MATB genes, these entities are rather complex genetic loci (for review, see Kues and Casselton, 1992; Casselton and Kues, 1994). What is called MATA is actually a locus containing a series of different alleles that each, in principle, contain two genes, H D l and HD2 (named after the homeodomain sequences they contain). The interaction was shown to be between an HDI gene in one nucleus and an HD2 gene in the other nucleus belonging to the same allelic series. HDI and HD2 genes within the same nucleus do not interact because they belong to different allelic series. Interactions between HD1 and HD2 gene products are thought to produce a gene-activating regulator because of the presence of homeodomains in the proteins. Specific interactions between HD1 and HD2 proteins of Coprinus cinereus were recently demonstrated in vitro (Banham et al., 1995). It was also known that a rare mutation, indicated as m t A m n in Table 1,
24
JOSEPH G. H. WESSELS
gives a homokaryotic mycelium a phenotype as if it contains two different MATA genes. This was shown to be due to a deletion effectively fusing an HDI and an HD2 gene within a MATA locus in such a way that the recognition sequence is deleted and a presumed DNA-binding domain (HD) from one gene becomes associated with the activating domain of the other gene (Banham et al., 1995). Although MATB genes appear genetically similar toMATAgenes, their interactions seem to involve quite different processes, since genes within these complexes encode multiple pheromones and pheromone receptors (Wendland et al., 1995). As shown in Table 1, formation of normal fruit bodies in which meiosis occurs requires the presence in a heterokaryon of different MATA and different MA7'B genes (MATA# MA7B#), or the presence in a homokaryon of constitutive mutations in both MATA and MA7B (matACo"matBcon).These genetic conditions allow for the accumulation of dikaryotic transcripts (Mulder and Wessels, 1986; Ruiters er al., 1988). Among these are the mRNAs of the hydrophobin genes SCI, SC4 and SC6, as well as the transcripts of two hydrophilic cell-wall proteins SC7 and SC14 (Schuren et al., 1993a). By performing run-on experiments on isolated nuclei, evidence was presented that the appearance of these mRNAs is transcriptionally regulated (Schuren et ul., 1993c), while in the case of the SC4 gene, cis-regulatory elements responsive to products of the MAT genes were indicated (Schuren et aL, 1993b). Other regulatory genes have been implicated in the fruiting process (Table 1). Among these are the THN gene without which no emergent growth occurs in monokaryon and dikaryon, and none of the hydrophobin genes (SC3, SC1, SC4 and SC6) are activated (Wessels et al., 1991b). Another gene found necessary for fruit-body formation is FBF. The mutantfbfwas isolated as a frequently occurring mutation that suppresses fruiting in a rnatACOninatBConstrain but allows for the formation of abundant aerial hyphae (Springer and Wessels, 1989). At the same time the mutation abolishes the formation of all dikaryon-specific transcripts but allows for high expression of the SC3 hydrophobin gene. When homozygously present @$@$) in a MATA# MATE?# mycelium, the phenotype of this secondary mycelium is not a dikaryon because incipient clamp connections fail to fuse with subterminal cells, locking one of the nuclei generated by mitosis in the pseudoclamp. The result is a heterokaryon in which all hyphal compartments, except the apical compartment, are essentially monokaryotic (Springer and Wessels, 1989; Wessels et al., 1995). In this MATA# MATB#@pflf heterokaryon, no transcripts from the SCI, SC4 and SC6 hydrophobin genes, nor other dikaryon-specific transcripts are produced, but the gene for the SC3 hydrophobin is very active (Springer and Wessels, 1989). Finally, monokaryons may carry so-called haploid fruiting alleles (hfu;Table 1). These were interpreted as relaxed versions of genes that normally operate downstream in the regulation cascade from mating-type genes to fruiting genes and therefore allow for fruiting (though not meiosis) in the absence of different mating-type genes (Yli-Mattila et al., 1989a). In these fruiting monokaryons, not only the SC3 hydrophobin gene but also the SCI, SC4 and SC6 hydrophobin genes are expressed (Ruiters et al., 1988; Yli-Mattila et al., 1989a).
HYDROPHOBINS
25
When grown from a mycelial macerate, hydrophobin genes in the dikaryon are switched on 2-3 days after inoculation, the time at which formation of aerial hyphae and fruit bodies is initiated (Mulder and Wessels, 1986). The levels of hydrophobin mRNAs then rise quickly (SC3 and SC4 mRNAs reaching levels of 1 % and 3.5% of the total, respectively), but drop sharply when cultures run out of exogenous nitrogen and carbon. It was known that, under such conditions of starvation, growth of the fruit bodies continues, the substrate mycelium is degraded, and massive transport of water and assimilates occurs from the substrate mycelium into the developing fruit bodies (Wessels and Sietsma, 1979; Ruiters and Wessels, 1989; Wessels, 1993b). It was therefore interesting to see that, although the mRNAs from the SC3, S C I , SC4 and SC6 hydrophobin genes sharply decrease in the mycelium as a whole, their concentrations stay very high in the developing fruit bodies, except for the mRNA of the SC3 hydrophobin gene, which is always expressed at a low level i n the developing fruit bodies (Mulder and Wessels, 1986). This indicates that the hydrophobins SC1, SC4 and SC6 play some role in fruit-body formation but that the SC3 hydrophobin may be less important in this process. With a specific antiserum against SC3, it was recently shown that the SC3 hydrophobin is only present 011 aerial hyphae, including those that cover the fruit bodies, but not on hyphac of the pleutenchyma that make up the major part of the fruit-body tissue (Asgeirsd6ttii et d.,1995). Moreover, a dikaryon hornozygous for a targeted mutation in the SC3 gene produces normal sporulating fruit bodies but any aerial hyphae that form are hydrophilic (van Wetter et al., 1996). Apparently these aerial hyphae do not produce SC4 or SC4 cannot functionally substitute for SC3. Within fruit bodies, differentiation of hyphae occurs, in the sense that the plectenchyma cells produce typical dikaryotic transcripts and proteins. However, the covering aerial hyphae display a more monokaryotic pattern of gene expression, prompting a re-examination of the sites of secretion of these proteins by whole 1995). It was found that the colonies of secondary mycelium (Asgeirsdbttir et d., SC3 hydrophobin is secreted in another region of the colony than the dikaryonspecific proteins, including the SC4 hydrophobin, which are typically formed in areas supporting fruit bodies. It was found that in nascent aerial hyphae expressing SC3, the two nuclei containing different MA7A and MA723 genes are present at some distance from each other. Since in a cominon-MA7A heterokaryon (with interaction between different MA'13 genes) and a MA7Ax inatBConhomokaryon (with constitutive MA723 activity) the SC3 gene is repressed (see above), it was hypothesized that interaction between different MA7B genes resulting in repression of SC3 also occurred in dikaryotic cells with closely juxtaposed nuclei. This explains the absence of SC3 expression in the dikaryotic plectenchyma of the fruit bodies. Possibly the disruption of the binucleate state occurring in incipient aerial hypae interrupts the interaction between different MA7B genes. This could cause a switch froin the production of SC I , SC4 and SC6 hydrophobins (and other dikaryori-specific proteins) to the production of the SC3 hydrophobin (Asgeirslg()s). d6ltir ct ul., 1095; Wcssels el d.,
26
JOSEPH G. H. WESSELS
The hydrophobin genes of S. commune may also be subject to environmental regulation but this issue is not yet clear. Light, which is necessary for fruit-body formation in most strains, increased the abundance of the dikaryon-specific hydrophobin mRNAs but decreased the SC3 mRNA level (Wessels et aL, 1987; Yli-Mattila et al., 1989b). A similar effect was noted for a low carbon dioxide concentration (Wessels et al., 1987). As explained by Wessels (1992), these effects may be (partly) indirect because light and low carbon dioxide are conducive to the formation of fruit bodies in which the dikaryon-specific mRNAs do not decrease asin the vegetative mycelium. It would seem to be essential to evaluate these effects of the environment in the absence of aerial differentiation, i.e. in shaken cultures. What are the functions of the hydrophobins SC 1, SC4 and SC6? The fruit body begins its development with aggregating hyphae that grow upward from the substratum, followed by inward growth of peripheral hypha forming the pit in which the hymenium develops (Raudaskoski and Viitanen, 1982; van der Valk and Marchant, 1978). Since the genes for these hydrophobins are active from the very beginning of fruit-body initiation, the corresponding products were suggested to play a role in the aggregation process, which plausibly involves some surface component of these hyphae (Wessels etaZ., 1991a). Whether or not this is true can hopefully be answered by targeted mutations in these hydrophobin genes and by localization of their products. Using cryo-scanning of fully hydrated frozen fruit-body fragments, hyphae making up the plectenchyma could be seen to be embedded in an extracellular matrix traversed by air channels probably serving gas exchange (Asgeirsdbttir et al., 1995). An antiserum against the dikaryon-specific hydrophilic SC7 localized this protein within the mucilaginous matrix, which binds hyphae together (Schuren et al., 1993a; Asgeirsdbttir et al., 1995). Experiments with an antiserum directed against SC4 localized this hydrophobin at the interface of the extracellular matrix and the air spaces within the fruit bodies (Wessels el al., 1995), whereas freeze-fracture images showed the presence of rodlets at this interface. Given the property of the SC4 hydrophobin to assemble into an SDS-insoluble film at a water-air interface (this laboratory, unpublished data), this would be an appropriate place for SC4 to assemble into a hydrophobic rodlet layer. One function of the SC4 hydrophobin would thus be to provide the air channels in the xerotolerant fruit bodies with a hydrophobic lining, preventing them from becoming water-soaked during recurrent cycles of drying and wetting. Hydrophobins have also been found in fruit bodies of Agaricus bisporus, the edible white button mushroom (Lugones et al., 1996; Fig. 1). The ABHl hydrophobin protein is particularly abundant in the closely interwoven hyphae that make up the skin and the veil of the mushroom. As for SC3, ABHl was found to assemble in vitro at hydrophobic-hydrophilic interfaces into a hydrophobic rodlet layer. Since a rodlet layer of ABHl covers the whole mushroom, ABHl probably has a function similar to that of the SC3 hydrophobin of S. cuinmune,i.e. conferring hydrophobicity to the surface of the fruit body. In addition, rodlets were found
HYDROPHOBINS
27
lining air spaces within the fruit bodies, suggesting that ABHl also provides these air spaces with a hydrophobic lining.
5.3. Formation of Conidia
With a few exceptions, the ascomycetes do not produce large sexual fruit bodies but they do form conspicuous masses of vegetative spores, often conidiospores (conidia) that appear on aerial hyphae (conidiophores), and often at the tips of branches (metulae and phialides). These conidia are generally very hydrophobic, probably an adaptation for aerial dispersal. The molecular genetics of conidiation has been intensively studied in Aspergillus nidulans by W.E. Timberlake and co-workers (for reviews, see Timberlake, 1990, 1993). By complementation of developmental mutants (Clutterbuck, 1969),these workers isolated the genes brL4, a b d and wetA. These turned out to be regulatory genes that work in sequence. An unknown signal, generated when the mycelium has attained a certain developmental stage, turns on brlA. The brL4 mutant forms aerial conidiophore stalks that do not differentiate any further structures and continue to grow. brlA activates abaA, and abaA also activates brlA in a positive feedback. The abaA mutant forms vesicles, metulae and phialides but the latter continue growth without forming conidia. The abaA gen activates wetA, a gene required for maturation of the conidia. The second approach taken was differential screening of a cDNA library. In this way two hydrophobin genes were found, which, after targeted inactivation, both resulted in a wettable phenotype of the conidia. The first gene, r o d , encodes a hydrophobin (Figs 1and 2) involved in the formation of the rodlet layer on conidia, phialides and metulae (Stringer et al., 1991). The rodA mRNA was not found in the developing conidia but accumulates in the phialides. The rod4 gene is activated by brL4 but does not need the activities of abaA and wetA. When the brlA gene was placed under control of a promoter that could be induced experimentally (Adams et al., 1988), conidiophores and conidia were produced in submerged culture accompanied by expression of rod, although rodlets were not formed and the hydrophobin probably diffused into the medium (Stringer et al., 1991). Since rodA appears under the control of brlA, the possibility of brlA producing an activator directly acting on mdA was investigated by performing an elegant experiment in Saccharoinyces cerevisiae (Chang and Timberlake, 1993).In this yeast, brL4, under control of the Gall promoter, activated LmZ, under control of the rod4 promoter. In this mdA promoter, as well as in other genes regulated by brL4, multiple copies of the sequence 5'CIA G/A A G G G G/A were found, apparently mediating binding of the brlA transcription factor. The second hydrophobin gene found in A. nidulans was dewA (Stringer and Timberlake, 1995). DewA was found only on conidia as judged by immunolocalization of an epitope-tagged hydrophobin. However, no rodlets were seen on r o d - d w A + conidia, although DewA contributed to hydrophobicity of the spores.
28
JOSEPH G. H. WESSELS
These conidia were more hydrophobic than rodA- dewA- conidia and r o d ' dewA- conidia could be wetted with a solution containing both 0.2% SDS and 50 m~ EDTA (hut not with solutions containing either of these substances alone). Unlike roo%, expression of dewA requires activity of wetA; activity is only evident after the first spores are produced. This pattern of expression is similar to that of wA, encoding a polyketide synthase, and yA, encoding a laccase, genes required for pigment production in the spores, but likewise not transcribed i n the spores. These proteins are apparently all transported into the developing conidia at a late stage during their maturation, while the product of the rodA gene is incorporated earlier (Stringer and Timberlake, 1995). There were also rodlets found on stalks and vesicles of conidiophores (Stringer et al., 1991), which apparently were not the product of r o d or dewA. So, other hydrophohin genes probably exist in A. nidulans. Since rod4 and dewA are only locally expressed in the conidioptiores, the hydrophobin-like proteins found in theculture medium of wild-typeA. nidulans (de Vries et al., 1993) might be encoded by other hydrophobin genes. However, it is also possible that, in submerged cultures, rod4 and wetA are induced under conditions of C or N starvation as are the rodlet genes of Neurospora crassa (Sokolovsky et al., 1992) and Magnaporthe grisea (Talbot et al., 1993). Another conidiating system in which a hydrophobin has been implicated is N, crassa. The gene for this hydrophobin was independently identified by Bell-Pedersen et al. (1992) as a gene (ccg2) controlled by the circadian rhythm in this fungus (mRNAs reaching high values in the subjective morning) and by Lauter et al. (1992) as a gene (hli-7)induced in dark-grown cultures by blue light. Both groups showed that the gene complemented the previously isolated e m mutant, which has easily wettable conidia (Selitrennikoff, 1976) and lacks rodlets (Beever and Dempsey, 1978). Because ccg-2 and hli-7 are allelic to e m , the gene is referred to as e m (Figs 1 and 2). It appears that the original eas mutant is leaky; expression of mRNA could be detected and some rodlets were found on the spores of the eas mutant (BellPedersen et al., 1992). There is nothing wrong with the coding sequence, but an insertion was found between -lo00 and -1500 base pairs upstream of the transcription start point (Lauter et al., 1992). It is somewhat surprising to find a cis-regulatory element so far upstream. However, in a promoter analysis of eas, Kaldenhoff and Russo (1993) indeed found that a region located between -1498 bp and -1079 bp acted as a positive regulatory element necessary for light activation, and for activation by carbon and nitrogen starvation. Deletion of this region led to complete inactivation of the gene. In addition, they found a positive element for light induction between -429 bp and -380 bp, and a negative regulatory element in the region -595 bp and -429 bp that appeared to inhibit the adjacent lightsensitive element. In surface cultures, eas seems to be precisely regulated (Lauter et al., 1992). Its mRNA becomes abundant at a precise point of conidiospore production but is not found in the conidia; as in A. nidulans conidiogenesis, the rodlet protein is apparently transported into the spore. However, the eas gene is
29
HYDROPHOBINS
also expressed in acon-2 and acon-3 mutants that cannot conidiate. A mutation in
fl, however, blocks expression of eas. Since this mutation also produces aerial
hyphae that are easily wettable, the fl gene may regulate unknown hydrophobin genes in this organism, in addition to eas. The mRNA of eas is also produced when N. crassa is grown in submerged culture under conditions of nitrogen starvation that induce conidiation, even to estimated levels of 10-25% of polyA'mRNA, 50-100 times higher than in surface cultures (Sokolovsky er al., 1992). Also, the MPGl gene, responsible for hydrophobicity of conidia of Magnaporthe grisea, is transcribed in submerged growing mycelium during starvation for carbon or nitrogen (Talbot et al., 1993). Although it has not yet been shown that any of these hydrophobins produce rodlets by interfacial self-assembly, presumably the hydrophobins were secreted into the medium and not assembled on any conidia that formed submerged.
5.4. Pathogenesis
The surfaces of plants and animals can be extremely hydrophobic. Most plant and arthropod surfaces show water contact angles above 100" up to 170°, while the human skin was also found to be rather hydrophobic with water contact angles of 100" (Wosten el al., 1995). However, it should be realized that the wettability of these surfaces is caused not only by their chemical composition but is strongly influenced by topographical structures, for instance, owing to wax morphology and epidermal ridges (Troughton and Hall, 1967; Holloway, 1970; Netting and von Wettstein-Knowles, 1973). These studies have indicated that water contact angles of smooth surfaces could be raised from 90" to 165" solely by the presence of surface roughness. Since water contact angles measured on smooth surfaces of Teflon and Parafilm do not exceed 110" (Wosten el al., 1994a), it is likely that all high-contact angles measured on natural surfaces are partly due to structural features of these surfaces. The extensive literature on adhesion of fungi to the hydrophobic surfaces of plants and arthropod cuticles has beed reviewed (Boucias and Pendland, 1991; Nicholson and Epstein, 1991; Mendgen and Deising, 1993; Clement el al., 1994; Jones, 1994). It is generally thought that hydrophobic interactions occur between the hydrophobic surfaces of airborne spores and the host surface. At least in cases where rodlets are observed on the spores, hydrophobins are probably involved. Also, active adhesion of spores was observed. For instance, upon hydration, spores of the rice blast fungus, Magnaporthe grisea, expel a preformed material from the site of future germ-tube formation by which they tightly adhere to hydrophobic surfaces (Hamer er al., 1988).In addition, the germ tube and the appressonum must adhere tightly to the hydrophobic surface because the infection peg, which penetrates the epidermis, arises from the latter structure. Some fungi can do this by ~~
30
JOSEPH G . H. WESSELS
sheer mechanical force, as demonstrated a century ago by Miyoshi (1895) and more recently for M.grisea by Howard et al. (1991). During nutrient deprivation, the insect pathogen Metarhizium anisopliae produced haustoria and cuticle-degrading enzymes in vitro, and at the same time abundantly transcribed the hydrophobin gene ssg A (St Leger et al., 1992). These authors suggested that the SSGA hydrophobin (Figs 1 and 2) is involved in building the wall of the haustorium and could assist in hydrophobic attachment to the cuticular surface. Talbot et al. (1993) detected abundant transcription of the hydrophobin gene MPGI (Figs 1 and 2 ) during infection of rice plants with Mugnaporthe grisea. MPGI mRNA is highly abundant very early in infection, concomitant with appressorium formation, while a second peak of MPGI mRNA occurs during symptom development. They also performed a gene disruption and observed that the M p g l mutants had a reduced ability to cause disease symptoms, which appears to result from an impaired ability to undergo appressoria formation. Since appressorium formation is triggered in this case by a hydrophobic surface (Hamer et al., 1988), it was assumed that in the M p g l mutant, in the absence of a hydrophobin-mediated contact between the fungal wall and the inducing surface, a morphogenetic signal for appressorium formation is not generated. However, it was recently shown (Talbot et al., 1996) that wild-type and mutant germlings adhere equally well to Teflon, but that the latter grow longer hyphae and make fewer appressoria. It would thus seem that the effect of the MPGl hydrophobin is very specific for generating the signal for appressorium formation and that another hydrophobin may be responsible for attachment of the germlings to the hydrophobic surface. Remarkably, the MPGl hydrophobin was shown to be responsible also for generating the hydrophobic rodlet layer on the conidia of this fungus. Most germlings of plant pathogenic fungi adhere best to hydrophobic surfaces, some clearly exhibit greatest adhesion to hydrophilic surfaces, and a few adhere to both types of surfaces (Nicholson and Epstein, 1991; Nicholson and Kunoh, 1995). Terhune and Hoch (1993) demonstratedthat Urornycesappendiculatus urediospore germlings exhibit best adhesion to hydrophobic substrata. A range of surface hydrophobicities was prepared by treating glass with a variety of silanes. The most hydrophobic surface, determined by measuring surface wettability, allowed the greatest adhesion of germlings to the substrate. Also, small topographic features of the surface were most inducive to appressoria formation when these were hydrophobic. These observations closely parallel the assembly and adhesion of the SC3 hydrophobin of Schizophyllurn commune to a continuous hydrophobicity gradient obtained by coating glass with dichlorodimethylsilane (Wosten et al.. 1995). How could secretion of a hydrophobin attach a hypha to a hydrophobic surface? Wosten et a!. (1994a) found that the SC3 hydrophobin assembles into an SDS-insoluble monolayer on hydrophobic surfaces, making these surfaces wettable. Assuming that the hydrophilic side of the assembled hydrophobin strongly binds to hydrophilic polysaccharide components of the wall, secretion of the SC3
HYDROPHOBINS
31
hydrophobin by a hypha of Schizophyllum commune, forced to grow over Teflon, would thus be expected to glue the hypha to the Teflon. This was shown to be the case and SC3 could be immunolocalized between the hypha and the Teflon. Moreover, attachment of hyphae to the Teflon was reduced in a strain with a targeted disruption of the SC3 gene. However, the wild-type strain did not adhere to the Teflon when immersed in water or liquid medium (this laboratory, unpublished data), although, owing to secretion of SC3 in the surrounding liquid, the Teflon became hydrophilic. It is, therefore, conceivable that a pathogenic fungus first wets a surface with a hydrophobin and then tightly adheres to the hydrophilic side of the amphipathic hydrophobin film on a leaf or insect cuticle surface by means of a hydrophilic mucilage. Mucilages have often been seen as apparently attaching the hypha or appressorium to the host surface (Nicholson and Epstein, 1991). Figure 5 shows a diagram of the two possible ways for a hypha to adhere to a hydrophobic surface by means of a hydrophobin. Figure 5a depicts a situation as observed for hydrophobin-mediated attachment of a S. commune hypha to Teflon under dry conditions. Figure 5b dcpicts a possible means of attachment for a pathogenic fungus under humid conditions, implicating both a hydmphobin and a hydrophilic mucilage. It should be emphasized that the mucilage and the hydrophobin could be simultaneously secreted and that there might be specific lectin-like interactions between the hydrophilic side of the hydrophobin film and the mucilage. The participation of lectins has been suggested by studies on adhesion of Magnaporthe grisea to artificial surfaces (Xiao et al., 1994). If the involved hydrophobins have lectin-like activity, this would explain why the mucilage cannot provide for adherence to any hydrophilic surface but that a hydrophobic surface is required for attachment of the fungus. The beauty of this system is its simplicity. The hydrophobicity of the surface would be sensed by a secreted protein, which by self-assembly serves as an adhesive with an amphipathic character gluing two incompatible surfaces together. As mentioned earlier (Section 4.2), cerato-ulmin produced by Ophiostoina ulrni has been implicated in causing Dutch elm disease. The reason for its assumed role as a phytotoxin was mainly based on correlative evidence and on the ability of isolated cerato-ulmin to cause the wilting syndrome in plant cuttings (Richards, 1993). A compelling reason for assuming a role in wilting was also the occurrence of similar plugging phenomena in the xylem of elm after infection with 0. ufini and application of cerato-ulmin to elm cuttings (Takai and Hiratsuka, 1984). It was plausible to assume that wilting was initiated by assembly of the surface-active cerato-ulmin around air bubbles arising in the xylem fluid under negative pressure (Russo et al., 1982). However, strains of 0. nova-ulmi carrying the pleiotropic mutation cu- did not produce cerato-ulmin, had no aerial hyphae but were nevertheless virulent (Brasier et al., 1995). Disruption of the gene for cerato-ulmin recently showed that indeed cerato-ulmin is not involved in virulence of the fungus (Bowden et al., 1996). Vascular discoloration and foliar wilting in elm seedlings
JOSEPH G. H.WESSELS
32 0
hydrophobin as monomer
ad hydrophobins in assembly
Figure 5 Proposed models for the attachment of hyphae to hydrophobic surfaces. Symbols are as in Fig. 4. In (a) the hypha is growing in air over the hydrophobic substrate and secreted hydrophobin monomers assemble directly at the hyphal surface exposed to air, and between the wall and the hydrophobic substratum, firmly attaching the hypha. This was observed for hyphae of Schizophyilum commune forced to grow over dry Teflon (Wosten et al., 1994a). In (b) it is assumed that free hydrophobin monomers are secreted into the liquid or mucilage surrounding the hyphae, and that the hydrophobin, by assembling on the hydrophobic surface, creates a hydrophilic surface to which the hypha can attach by means of the mucilage. which serves as a hydrophilic adhesive, possibly binding specifically to the primed surface. (From Wessels, 1996, with permission from the publisher.)
are the same for plants inoculated with an aggressive strain of 0.ulmi producing abundant cerato-ulmin and the same strain with the targeted mutation that does not produce cerato-ulmin. Since the mutant produces few aerial hyphae and is easily wettable, cerato-ulmin is probably involved in aerial growth of this fungus. However, it remains possible that 0. ulmi produces another hydrophobin under conditions prevailing in the tree that is responsible for disease symptoms and that cerato-ulmin is just mimicking the effects of this hydrophobin. It should also be noted that a similar Class I1 hydrophobin produced by Cryphonectriu parusitica, cryparin (CRYP; Figs 1 and 2), has not been considered to be a phytotoxin (Carpenter et al., 1992) and that non-pathogenic fungi produce similar hydrophobins (see Fig. 1). Hydrophobicity of the fungal surface has been implicated in fungal infections of humans (Hazen, 1990). This is a vast field that becomes increasingly important because the incidence of immunocompromisedpatients infected by pathogenic and
HYDROPHOBINS
33
opportunistic fungi is rapidly increasing. The involvement of hydrophobins in the infection process has not yet been reported. The availability of the rod4 gene of Aspergillus nidulans has enabled cloning of the corresponding gene from the pathogenic A.fu1nigatu.s (Parta et al., 1994;Thau et al., 1994). However, disruption of the gene, leading to formation of wettable conidia without rodlets, did not lead to a decrease in pathogenicity in mice (Thau et al., 1994). The importance of the hydrophobic rodlet layer probably lies in aerial dispersal of the fungus but other members of the hydrophobin family may still prove to be of importance in pathogenesis. Hydrophobins may play a role in attachment of invading hyphae to host cells or hydrophobic implants. In addition, because invading fungi may release large quantities of specific circulating hydrophobins, these might be used for diagnosis of the mycosis with specific antibodies. Since hydrophobins are small, abundantly secreted proteins that fulfil important roles in fungal development, it is conceivable that plants sense the presence of fungi by having receptors for these proteins. Hydrophobins or derived proteins could thus act as elicitors of the defence response in plants after infection with pathogenic fungi (Wessels, 1994). Known peptide elicitors have some resemblance to hydrophobins, particularly in being rich i n cysteine residues, and it has been suggested that these elicitors are structurally related to hydrophobins (Sticklen and Bolyard, 1994; Templeton et al., 1994).
5.5. Symbiosis
Symbiosis of fungi with other organisms always involves intimate contact of the fungal surface with host tissue. In two cases hydrophobins have now been implicated. F. Martin’s group has studied molecular events occurring during formation of ectomycorrhiza (Martin et al., 1995). During formation of the ectomycorrhizal mantle in the association between Pisolithus tinctorus and the roots of Eucalyptus globulus, they found high expression of two genes encoding hydrophobins (HydPr-l and HydPt-2) (’Ibgu et al., 1996, and Fig. 1). They speculate that these hydrophobins might be involved in aggregation of the hyphae, forming the hyphal tissue around the root, or aid in attachment of hyphae to the root during initial colonization (Martinet al., 1995). Alternatively, they may create the hydrophobic surface of the fungal mantle. Another case of symbiosis in which hydrophobins have been implicated, but not yet isolated, concerns lichens (Honneger, 1993). This is particular interesting because lichens are fungi that have reached the ultimate stage in emergent growth. They always live above the substrate, retrieving organic material from intimately associated algae and cyanobacteria that assimilate carbon dioxide and dinitrogen from the air. Some 20% of the fungi are lichenized and these lichens are the only “vegetation”in about 8% of the land area, particularly tundra and high mountainous areas (Honneger, 1991). Because of their habitat, they are exposed to extreme
34
JOSEPH G. H. WESSELS
cycles of wetting and drying, and temperature changes. The fungal hyphae were seen to make simple contacts with algae, the whole being ensheathed by a matrix decorated on the outside with a hydrophobic rodlet layer, where these associations contacted the air spaces within the thallus (Honneger, 1991). Although a hydrophobin composition of this rodlet was not established, spontaneous self-assembly of secreted hydrophobins could explain how this rodlet layer could extend from the fungus over the algal surface. The possible function of the rodlet layer (and other hydrophobic substances) lining the air spaces could be all-important for the existence of this symbiosis (Honneger, 1993; Honneger and Peter, 1994). First, the layer would collectively shield the apoplast of fungus and photosymbiont from the air spaces, permitting apoplastic transport of water and solutes to and from the symbiont. Second, the lining of the air spaces with a hydrophobic rodlet layer would permit optimal gas exchange and prevent the air channels from becoming soaked with water during wetting after a dry period.
6. TECHNOLOGY
Although this review deals principally with roles of hydrophobins in biological systems, the novel properties of hydrophobins as discussed in Section 4 immediately raise the possibilities of application of hydrophobins in technology. Materials science has a considerable interest in self-assembling molecules from nature (Service, 1994). The abilities of hydrophobins to self-assemble at interfaces into insoluble films are not parallelled by any other known protein and therefore it is likely that applications will be found. In fact, the roles of hydrophobins in fungal growth and development, as discussed in Section 5, give direction to the kind of applications that can be thought of “naturu artis rnagistru”. At the moment only the imagination limits the potential application of hydrophobins. One condition for the successful application of hydrophobins is that they can be cheaply produced in quantity. As mentioned before, hydrophobins are among the most abundant proteins secreted by fungi. Class I hydrophobins appear to be the most promising for application because of the stability of the assembled films. These hydrophobins appear to be particularly abundant in the culture medium of members of the basidiomycetes. For instance, it has been calculated that, in 4-day-old cultures of Schizophyllurn commune, about 15% of the 35Sincorporated into protein goes into synthesis of the SC3 hydrophobin (de Vnes et ul., 1993), while up to 20 mg of SC3 can be easily purified from one litre of culture medium by a simple procedure based on the extraordinary properties of the protein (4.1). Strain selection and optimizing culture conditions could probably enhance the yield as could molecular genetic methods, such as increasing gene dose and heterologous production i n fungi in common use in the fennentation industry.
HYDROPHOBINS
35
On the other hand, it should be realized that quantities needed for certain applications may be small. This is expected from the use that nature makes of an “expensive” product as a protein for changing the wettability of surfaces. Indeed, the very nature of the assembled amphipathic film requires that it is present as a monolayer. The thickness of this monolayer is only about 10 nm and thus very little hydrophobin is required to achieve a drastic change in wettability. From the number of molecules of SC3 adsorbed to Teflon (Wosten et al., 1994a), it can be calculated that about 1.5 mg SC3 hydrophobin suffices to coat 1 m2 of Teflon surface with the effect of decreasing the hydrophobicity of this surface from 110” to 48” water contact angles. Another important lesson from nature is that a fungal species makes different hydrophobins for different purposes. For instance, in S. coininune an SC3 film appears to coat aerial hyphae and to confer water-repellent properties to these structures, whereas air channels in fruit bodies are lined with an assemblage of SC4 hydrophobin. One wonders whether possible biophysical differences in the properties of these films are tuned to different functions, especially because hydrophobins from widely different species may be more closely related than different hydrophobins within one species (Table 1). This is certainly a point tobe considered in any biomimetics before resorting to genetic engineering to tailor a hydrophobin for a specific purpose. Of the many possible applications of hydrophobins, a few of the more obvious ones are listed below. 1. Hydrophobins may be used in tissue engineering (Hubbell, 1995), particularly for coating hydrophobic surfaces to increase their biocompatibility. As already noted, the attachment of the hydrophobin film to hydrophobic surfaces is very strong and the change in surface wettability significant. For instance, the hydrophobins may be used to enhance the biocompatibility of medical implants, including artificial blood vessels and surgical instruments. 2. Hydrophobins may be used as an intermediate to attach cells, proteins, such as antibodies, and small ligands to hydrophobic surfaces, as in biosensors. For instance, researchers at the Department of Bioprocessing and Biomonitoring of TNO, Zeist, The Netherlands, in collaboration with our group, have shown that the SC3 hydrophobin readily coats a hydrophobic gold surface. At the exposed hydrophilic side of the SC3 film, mannose residues can be oxidized with periodic acid without disturbing the binding to the gold, while the generated aldehyde groups can be easily coupled to amino groups of a protein by a Schiff-base reaction. 3. Hydrophobic solids or liquids (oils) can be dispersed in water by coating with a hydrophobin (Wosten et al., 1994a). Oil vesicles coated with a hydrophobin film may be useful for delivery of lipophilic drugs. A glycosylated hydrophobin, such as SC3, would permit easy attachment of targeting antibodies to the outside of such vesicles.
JOSEPH G. H.WESSELS
36
Hydrophobins have unknowingly been ingested by humans for millennia when eating mushrooms and fungus-fermented foods. At least when derived from GRAS fungi (Generally Regarded As Safe), hydrophobins can be considered safe for consumption and can be used in foods and drinks. 5. The property of hydrophobins to coat a surface with a very thin layer (about 10 nm) that nevertheless dramatically changes the nature of this surface promises the use of these proteins in nanotechnology as defined by Thomas (1995). 4.
7. CONCLUSIONS Hydrophobins are a novel class of small secreted cysteine-rich proteins of fungi that assemble into amphipathic films when confronted with hydrophilichydrophobic interfaces. Some hydrophobins form unstable, others extremely stable, amphipathic films. By assembling at a wall-air interface some have been shown to provide for a hydrophobic surface, which has the ultrastructural appearance of rodlets as on aerial hyphae and spores. Some hydrophobins have been shown to assemble into amphipathic films at interfaces between water and oils, or hydrophobic solids, and may be involved in adherence phenomena. It appears that hydrophobins are among the most abundantly produced proteins of fungi, and individual species may contain several genes producing divergent hydrophobins, possibly tailored for specific purposes. Hydrophobins have now been implicated in various developmental processes, such as formation of aerial hyphae, fruit bodies and conidia, and may play essential roles in fungal ecology, including spore dissemination, pathogenesis and symbiosis. The surfactive properties of hydrophobins and the ability of some of them to form very stable insoluble amphipathic films, which change the wettability of surfaces, also makes them good candidates for technical applications.
ACKNOWLEDGEMENTS The author is grateful to all members of his laboratory for stimulating discussions and for permission to quote unpublished results. He is particularly indebted to Dr Han A.B. Wosten for carefully reading the manuscript and to colleagues in other laboratories who were willing to communicate unpublished results.
REFERENCES Adams, T.H., Boylan, M.T. and Timberlake, W.E. (1988) brlA is necessary and sufficient to direct conidiophore development in Aspergillus nidulons. Cell 54, 353-362.
HYDROPHOBINS
37
Andersen, N.H., Cao, B., Rodrfgez-Rornero, A. and Arreguin, B. (1993) Hevein: NMR assignment of solution-state folding for the agglutinin-toxin motif. Biochemistry 32. 1407-1 422. Aronson, A.I. and Fitz-James, P. (1976) Structure and morphogenesis of the bacterial spore coat. Bacferiol.Rev 40,360-402. Asgeirsdbttir, S.A., van Wetter, M.A. and Wessels, J.G.H. (1995) Differential expression of genes under control of the mating-type genes in the secondary mycelium of Schizophylluin commune. Microbiology 141, 1281-1288. Banham, A.H., Asante-Owusu, R.N., Gottgens, B., Thompson, S.A.J., Kingsnorth, C.S., Mellor, E.J.C. and Casselton, L.A. (1995) An N-terminal dimerization domain permits homeodomain proteins to choose compatible partners and initiate sexual development in the mushroom Coprinus cinereus. Plant Cell 7,773-783. Beever, R.E. and Dempsey, G. (1978) Function of rodlets on the hyphae of fungal spores. Nature 272, 608-6 10. Beever, R.E., Redgewell, R.J. and Dempsey, G. (1979) Purification and chemical characterization of the rodlet layer of Neurospora crassa conidia. J. Bacferiol. 140, 1063-1070. Bell-Pedersen, D., Dunlap, J.C. and Loros, J.J. (1992) The Neurospora circadian clockcontrolled gene, ccg-2, is allelic to e m and encodes a fungal hydrophobin required for formation of the conidial rodlet layer. Genes Dev. 6,2382-2394. Bidochka, M.J., St Leger, R.J., Joshi, L. and Roberts, D.W. (1995a) The rodlet layer from aerial and submerged conidia of the entomopathogenic fungus Beauveriu bassianu contains hydrophobin. Mycol. Res. 99,403406. Bidochka, M.J., St Leger, R.J., Joshi, L. and Roberts, D.W. (1995b) An innercell wall protein (cwpl) from conidia of the entomopathogenic fungus Beauveriu bassiana. Microbiology 141,1075-1080. Bohlmann, H. and Apel, K. (1991) Thionins. Annu. Rev. Plant Physiol. Planf Mol. Biol. 42, 227-240. Bolyard, M.G. and Sticklen, M.B. (1992) Expression of a modified Dutch elm disease toxin in Escherichia coli. Mol. Plant-Microbe fnt. 5,520-524. Boucias, D.G. and Pendland, J.C. (1991) Attachment of mycopathogens to cuticle. The initial events of mycoses in arthropode hosts. In: The Fungal Spore and Disease Initiation in Planfs and Animals (G.T. Cole and H.C. Hoch, eds), pp. 101-127. Plenum Press, New York, London. Bowden, C.G., Hintz, W.E., Jeng, R., Hubbes, M. and Horgen, P.A. (1993) Isolation and characterization of the cerato-ulmin toxin gene of the Dutch elm disease pathogen, Ophiostoina ulmi. Curr: Genet. 25,323-329. Bowden, C.G., Smalley, E., Guries, R.P., Temple, B., Hubbes, M. and Horgen, P.A. (1996) Lack of association between cerato-ulmin production and virulence in Ophiostoma nova-ulnii. Mol. Planl-Microbe Int. (in press). Brasier, C.M., Takai, S., Nordin, J.H. and Richards, W.C. (1990) Differences in cerato-ulmin production between the EAN, NAN and non-aggressive subgroups of Ophiostoma rrlmi. Plan1 Pathol. 39,23 1-236. Brasier, C.M., Kirk, S.A. and Tegli, S. (1995) Naturally occurring non-cerato-ulmin producing mutants of Ophiostoma novo-ulmi are pathogenic but lack aerial mycelium. Mycol. Res. 99,436440. Bronchart, R. and Demoulin, V. (1971) Ultrastructure de la paroi des basidiospores de Lycoperdon et de Scleroderma (GastLromycktes) comparke celle de quelques autres spores de champignons. Protoplasma 72, 179-189. Busscher, H.J., van Pelt, A.W.J., de Boer, P., de Jong, H.P. and Arends, J. (1984) The effect of surface roughening of polymers on measuredcontact angles of liquids. Colloids Surf. 9,319-331.
38
JOSEPH G. H.WESSELS
Carpenter, C.E., Mueller, R.J., Kazmierczak, P., Zhang, L., Villalon, D.K. and Van Alfen, N.K. (1992) Effect of a virus on accumulation of a tissue-specific cell-surface protein of the fungus Cryphonectria (Endothia) parasitica. Mol. Plant-Microbe Inf. 4 , 5 5 4 1 . Casselton, L.A. and Kues, U. (1994) Mating-type genes in homobasidiomycetes. In: The Mycofa, Vol. I, Growth, DijFerentiation and Development (J.G.H. Wessels and E Meinhardt, eds), pp. 307-321. Springer Verlag, Berlin. Castonguay, Y., Laberge, S., Nadeau, P and Vdzina, L.-P. (1994) A cold-induced gene from Medicago sativa encodes a bimodular protein similar to developmentally regulated proteins. Planf Mol. Biol. 24,799-804. Cavalier-Smith, T. (1993) Kingdom protozoa and its 18 phyla. Microbiol. Rev. 57,953-994. Chang, Y.C. and Timberlake, W.E. (1993) Identification of Aspergillus brlA response elements (BREs) by genetic selection in yeast. Genetics 133,29-38. Claverie-Martin, F,, Diaz-Torres, M.R. and Geoghegan, M.G. (1986) Chemical composition and electron microscopy of the rodlet layer of Aspergillus nidulans conidia. Curr: Microbiol. 14,221-225. Clement, J.A., Porter, R., Butt, T.M. and Beckett, A. (1994) The role of hydrophobicity in attachment of urediniospores and sporelings of Uromyces viciae-fabae. Mycol. Res. 98, 1217-1228. Clutterbuck, A.J. (1969) A mutational analysis of conidial development in Aspergillus nidulans. Genetics 63,317-327. Cole, G.T. and Hoch, H.C., eds (1991) The Fungal Spore and Disease Initiation in Plants and Animals. Plenum Press, New York. Cole, G.T., Sekiya, M., Kasai, R.L. Yokoyama, T. and Nozawa. Y. (1979) Surface ultrastructure and chemical composition of the cell walls of conidial fungi. Exp. Mycol. 3,132-156. de Groot, P.W.J., Schaap, P.J., Sonnenberg, A.S.M., Visser, J. and van Griensven, L.J.L.D. (1996) The Agaricus bisporus hypA gene encodes a hydrophobin and specifically accumulates in peel tissue of mushroom caps during fruit body development. J. Mol. Biol. 257, 1008-1018. de Vries, O.M.H., Fekkes, M.P., Wosten, H.A.B. and Wessels, J.G.H. (1993) Insoluble hydrophobin complexes in the walls of Schizophyllum commune and other filamentous fungi. Arch. Microbiol. 159,330-335. Dons, J.J.M., Springer J., de Vries S.C. and Wessels, J.G.H. (1984) Molecular cloning of a gene abundantly expressed during fruit-body initiation in Schizophyllum commune. J. Bacteriol. 157,802-805. Drenth,J., Low, B.J., Richardson, J.S. and Wright, C.S. (1980) The toxin-agglutinin fold: a new group of small protein structures organized around a four-disulphide core. J. Bid. Chem. 255,2652-2655. Elwing, H.. Welin, S.,Askendal, A., Nilssom, U. and Lundstrom. L. (1987) A wettability gradient method for studies of macromolecular interactions at the liquid-solid interface. J. Colloid Interface Sci. 119,203-210. Fisher, D.J, Brown, G.A. and Holloway, P.J. (1978) Influence of growth medium on surface and wall lipid of fungal spores. Phyfochemistry 17,85-89. Gerin, PA., Asther, M., Sleytr, U.B. and Rouxhet, P.G. (1994) Detection of rodlets in the outer wall region of conidiospores of Phanerochaete chrysosporium. Can. J. Microbiol. 40,412416. Hamer. J.E., Howard, R.J., Chumleym, F.G. and Valent, B. (1988) Amechanism for surface attachment in spores of a plant pathogenic fungus. Science 239,288-290. Harley, J.L. and Smith, S.E. (1983) Mycorrhizal Symbiosis. Academic Press, London, New York. Hashimoto, T., Wu-Yuan, C.D. and Blumenthal, H.J. (1976) Isolation and characterization
HYDROPHOBINS
39
of the rodlet layer of Trichophyton mentagrophytes microconidid wall. J. Bacten'ol. 127, 1543-1549. Hazen, K.C. (1990) Cell surface hydrophobicity of medically important fungi, especially Candida species. In: Microbial Cell Surface Hydrophobicity (R.J. Doyle and M. Rosenberg, eds), pp. 249-95. American Society of Microbiology, Washington, DC. Hazlett, R.D. (1992) On surface roughness effects in wetting phenomena. J. Adhesion Sci. Technol. 6,625-633. Hess, W.M., Sassen, M.M.A. and Remsen, C.C. (1968) Surface characteristics of Penicillium conidia. Mycologia 60,290-303. Higgins, D.G. and Sharp, P.M. (1988) Clustal: a package for performing multiple sequence alignments on a microcomputer. Gene 73,237-244. Hobot, J.A. and Gull, K. (1981) Structure and biochemistry of the spore surface of Syncephalastrum racemosum. Curr: Microbiol. 5, 183-1 85. Holloway, P.J. (1970) Surface factors affecting the wetting of leaves. Pestic. Sci. 1,156-163. Holt, S.C. and Leadbetter, E.R. (1969) Comparative ultrastructure of selected aerobic spore forming bacteria. Bacteriol. Rev. 33,346-378. Honneger, R. (1991) Functional aspects of the lichen symbiosis. Anna Rev. Plant Physiol. Plant Mol. Biol. 42,553-578. Honneger, R. (1993) Developmental biology of lichens. New Phytol. 125,659-677. Honneger, R. and Peter, M. (1994) Routes of solute translocation and location of water in hetemmerous lichens visualized with cryotechniques in light and electron microscopy. Symbiosis 16, 167-186. Howard, R.J., Ferrari, M.A., Roach, D.H. and Money, N.P. (1991) Penetration of hard substrates by a fungus employing enormous turgor pressures. Proc. NatlAcad. Sci. U.S.A. 88,11281-11284. Hubbell, J.A. (1995) Biomaterials in tissue engineering. Biotechnology 13,565-576. Huh, C. and Mason, S.G. (1977) Effects of surface roughness on wetting (theoretical) J. Colloid Inter- Sci. 60, 11-38. Jennings, D.H. (1984) Water flow through mycelia. In: The Ecology and Physiology of Fungal Mycelia (D.H. Jennings and A.D.M. Rayner, eds), pp. 143-164. Cambridge University Press, Cambridge. Jennings, D.H. (1994) Translocation in mycelia. In: The Mycofa, Vol. I, Growth, DifSerenfiationand Development (J.G.H. Wessels and F. Meinhardt. eds), pp. 163-173. Springer Verlag, Berlin. Jones, E.B.G. (1994) Fungal adhesion. Mycol. Res. 98,961-981. Kaldenhoff, R. and Russo. V.E.A. (1993) Promoter analysis of the bli-7/eas gene. Cum Genet. 24,394-399. Kiies, U. and Casselton, L.A. (1992) Fungal mating type genes - regulators of sexual development. Mycol. Res. 96,993-1006. Kyte. J. and Doolittle, R.F. (1982) A simple method for displaying the hydropathy character of a protein. J. Mul. Biol. 157, 105-132. Lauter, Fa-R., Russo, V.E.A. and Yanovsky, C. (1992) Developmental and light regulation of eas, the structural gene for the rodlet protein of Neumspora. Genes Dev. 6,2373-2381. Lora, J.M., de la Cruz, J., Benitez, T.,Llobell, A. and Pintor-Toro, J.A. (1994) Aputative catabolite-repressed cell wall protein from the mycoparasitic fungus Trichodema harzianum. Mol. Gen. Genet. 242,461-466. Lugones, L.G., Bosscher, J., Scholtmeyer, K., de Vries, O.M.H. and Wessels, J.G.H. (1996) An abundant hydrophobin (ABHl) forms hydrophobic rodlet layers in Agaricus bisporus fruiting bodies. Microbiology 142, 1321-1 329. Martin, F., Laurent, P., de Cavalho. D., Burgess, T., Murphy, P., Nehls, U. and Tagu, D. (1995)
40
JOSEPH G . H. WESSELS
Fungal gene expression during ectomycorrhiza formation. Can. J. Bot. 73, (Suppl. l), S541-SS47. Mendgen, K. and Deising, H. (1993) Infection structures of fungal plant pathogens - a cytological and physiological evaluation. New Phytol. 124, 193-213. Miyoshi, M. (1895) Die Durchbohrung von Membranen durch Pilzfaden. Jahrb. wiss. Bot. 28,269-289. Moore,R.T. (1985) The challenge of the dolipore/parenthosome septum. In: Developmental Biology of Higher Fungi (D. Moore, L.A. Casselton, D.A. Wood and J.C. Frankland, eds), pp. 175-212. Cambridge University Press, Cambridge. Mulder, G.H. and Wessels, J.G.H. (1986) Molecular cloning of RNAs differentially expressed in monokaryons and dikaryons of Schizophyllwn commune. Exp. Mycol. 10, 2 14-227, MuAoz, G.A., Agosin, E., Cotoras, M., San Martin, R. and Volpe, D. (1995) Comparison of aerial and submerged spore properties for Trichoderma harzianum. FEMS Microbiol. Let/. 125,63-70. , T., Aro, N., Kalkkinen, N., Alatalo, E. and Penttila, M. (1996) Genetic and biochemical characterization of the Trichoderma reesei hydrophobin HFB 1. Eul: J. Biochem. 235,248-255. Nespoulous, C., Huet, J.-C. and Pernollet, J.-C. (1992) Structure-function relationships of a and p elicitins, signal proteins involved in the plant-Phytophrhora interaction. Planta 186,551-557. Netting, A.G. and von Wettstein-Knowles, P. (1973) The physico-chemical basis of leaf wettability in wheat. Planta 114,293-309. Nicholson, R.L. and Epstein, L. (1991) Adhesion of fungi to the plant surface: prerequisite for pathogenesis. In: The Fungal Spore and DiseaAe Initialion in Plunrs and Animals (G.T. Cole and H.C. Hoch, eds), pp. 3-23. Plenum Press, New York, London. Nicholson, R. and Kunoh, H. (1995) Early interactions, adhesion, and establishment of the infection count in Erysiphe graminis. Can. J. Bot. 73 (Suppl.), S609-S615. Noordmans, J. and Busscher. H.J. (1991) The influence of droplet volume and contact angle on liquid surface tension measurements by axisymmetric drop shape analysis-profile (ADSA-P) Colloids Su$ 58.239-249. Parta, M., Chang, Y., Rulong, S . , Pinto-DaSilva, P. and Kwon-Chung, K.J. (1994) HYPl, a hydrophobin gene from Aspergillusfumigatus complements the rodletless phenotype in Aspergillus nidulans. In$ Immun. 62,43894395. Raikhel, N.V., Lee, H.4. and Broekaert, W.F. (1993) Structure and function of chitin-binding proteins, Annu. Rev. Plant Physiol. Plant Mol. Biol. 44,591-615. Raper, C.A. (1988) Schizophyllum commune, a model for genetic studies of the basidiomycetes. In: Genetics of Plant Pathogenic Fungi (G.S. Sidhu, ed.), pp. 511-522. Academic Press, London. Raper, J.R. (1966) Genetics of Sexualiv in Higher Fungi. The Ronald Press Co, New York. Raudaskoski, M. and Viitanen, H. (1982) Effects of aeration and light on fruit-body induction in Schizophyllum commune. Trans. Brir. Mycol. Soc. 78, 89-96. Rayner, A.D.M., Griffith, G.S. and Ainsworth, A.M. (1995) Mycelial interconnectedness. In: The Crowing Fungus (N.A.R. Gow and M. Gadd, eds), pp. 21-40. Chapman and Hall, London. Read, D.J. (1991) Mycorrhizas in ecosystems - nature’s response to the “law of the minimum”. In: Frontiers in Mycology (D.L. Hawksworth, ed.), pp. 101-130. CAB International, Wallingford. Richards, W.C. (1993) Cerato-ulmin: a unique wilt toxin of instrumental significance in the development of Dutch elm disease. In: Dutch Elm Disease Research, Cellular and
HYDROPHOB INS
41
Molecular Approaches (M.B. Sticklen and J.L. Sherald, eds), pp. 89-151. Springer Verlag, New York. Rohe, M., Gierlich, A,, Hermann, H., Hahn, M., Schmidt, B., Rosahl. S. and Knogge, W. (1995) The race-specific elicitor, NIP1 , from the barley pathogen, Rhynchosporium secalis, determines avirulence on host plants of the Rrsl resistance genotype. EMBO J. 14,41684177. Rosenberg, M. and Kjelleberg, S . (1986) Hydrophobic interactions: role in bacterial adhesion. Adv. Microb. Ecol. 9, 353-393. Rotenberg, Y., Boruvka, L. and Neumann, A.W. (1983) Determination of surface tension and contact angle from the shapes of axisymmetric fluid interfaces. J. Colloid Interjke Sci. 93,169-183. Rouxhet, P.G., Mozes, N., Dengis, P.B., Dufrene, Y.F., Gerin, P.A. and Genet, M.J. (1994) Application of X-ray photoelectron spectroscopy to microorganisms. Colloids Surf: B 2, 347-369. biters, M.H.J. and Wessels, J.G.H. (1989) In situ localization of specific RNAs in whole fruiting colonies of Schizophyllum commune. J. Gen. Microbiol. 135,1747-1754. Ruiters, M.H.J., Sietsma, J.H. and Wessels, J.G.H. (1988) Expression of dikaryon-specific mRNAs of Schizophyllum commune in relation to incompatibility genes, light and fruiting. Exp. Mycol. 1 2 , 6 0 4 9 . Russo, P.S., Blum, ED.,Ipsen, J.D., Abul-Hajj, Y.J. and Miller, W.G. (1982) The surface activity of the phytotoxin cerato-ulmin. Can. J. Bot. 60, 1414-1422. Sassen, M.M.A., Remsen, C.C. and Hess, W.M. (1967) Fine structure of Penicillium megasporum conidiospores. Protoplasm 64,75-87. Schuren, F.H.J. and Wessels, J.G.H. (1990) Two genes specifically expressed in fruiting dikaryons of Schizophyllum commune: homologies with a gene not regulated by mating type genes. Gene 90,199-205. Schuren,F.H.J.,Asgeirsd6ttir,S.A.,Kothe,E.M.,Scheer, J.H.J. and Wesse1s.J.G.H. (1993a) The Sc7/Sc14 gene family of Schizophyllum commune codes for extracellular proteins specifically expressed during fruit-body formation. J. Cen. Microbiol. 139,2083-2090. Schuren, F.H.J., Harmsen, M.C. and Wessels, J.G.H. (1993b) A homologous gene-reporter system for the basidiomycete Schizophyllum commune based on internally deleted genes. Mol. Gen. Genet. 238,91-96. Schuren, F.H.J., van der Lende, T.R. and Wessels, J.G.H. (1993~)Fruiting genes of Schizophyllum commune are transcriptionally regulated. Mycol. Res. 97,538-542. Selitrennikoff, C.P. (1976) Easily wettable, a new mutant. Neurospora Newsl. 23,23. Service, R.F. (1994) Self-assembly comes together. Nature 265,3 16-3 18, Smucker, R.A. and Pfister, R.M. (1978) Characterization of Streptomyces coelicolor A3(2) aerial spore mozaic. J. Gen. Microbiol. 24,397408. Sokolovsky, V.Y., Lauter, E-R., Miiller-Rober, B., Ricci, M., Schmidhauser, T.J. and Russo, V.E.A. (1992) Nitrogen regulation of blue light-inducible genes in Neurospora crassa. J. Gen. Microbiol. 138,2045-2049. Springer, J. and Wessels, J.G.H. (1989) A frequently occurring mutation that blocks the expression of fruiting genes in Schizophyllum cornmune. Mnl. Gen. Genet. 219,486-488. St Leger, R.J., Staples, R.C. and Roberts, D.W. (1992) Cloning and regulatory analysis of starvation stress gene, ssgA, encoding a hydrophobin-like protein from the entomopathogenic fungus, Metarhizium anisopliae. Gene 120, 119-124. Sterk, Pa, Booij. H., Schellekens, G.A., Van Kammen, A. and de Vries, S.C. (1991) Cell-specific expression of the carrot lipid transfer protein gene. Plant Cell 3,907-921. Stevenson, K.J., Slater, J.A. and Takai, S. (1979) Cerato-ulmin, a wilting toxin of Dutch elm disease fungus. Phytochemistry 18,235-238.
42
JOSEPH G. H. WESSELS
Sticklen, M.B. and Bolyard, M. (1994) Refinement of physiological roles for cerato-ulmin by analogy with other hydrophobins. Trends Microbiol. 2,213-215. Stringer, M.A. and Timberlake, W.E. (1993) Cerato-ulmin, a toxin involved in Dutch elm disease, is a hydrophobin. Plant Cell 5, 145-146. Stringer, M.A. and Timberlake. W.E. (1995) dewA encodes a fungal hydrophobin component of the Aspergillus spore wall. Mot. Microbiol. 1 6 , 3 3 4 . Stringer, M.A., Dean, R.A., Sewall, T.C. and Timberlake, W.E. (1991) Rodlefless, a new Aspergillus developmental mutant induced by directed gene inactivation. Genes Dev. 5, 1161-1171. Svircev, A.M., Jeng, R.S. and Hubbes, M. (1988) Detection of cerato ulmin on aggressive isolates of Ophiostomu ulmi by immunocytochemistry and scanning electron microsCOPY. Phytopathology 78,322-327. Tagu, D., Nasse, B. and Martin, F. (1996) Cloning and characterization of hydrophobins encoding cDNAs from the ectomyconhizal basidiomycete Pisolifhus tinctorius. Gene 168,93-97. Takai, S . (1974) Pathogenicity and cerato-ulmin production in Cerutocystis ulmi. Nature 252,124-126. Takai, S. (1980) Relationship of the production of toxin, cerato-ulmin, to synnemata formation, pathogenicity, mycelial habit and growth of Cerutocystis ubni isolates. Can. J. Bot. 58,658-662. Takai, S. and Hiratsuka, Y. (1980) Accumulation of the material containing the toxin cerato-ulmin on the hyphal surface of Ceratocysfis ulmi. Can. J. Bot. 58,658-662. Takai, S . and Hiratsuka, Y. (1984) Scanning electron microscope observations of internal symptons of white elm following Cerutocystis ulmi infection and cerato-ulmin treatment. Can. J. Bot. 62, 1365-1371. Takai, S. and Richards, W.C. (1978) Cerato-ulmin, a wilting toxin of Cerutocystis ulmi: isolation and some properties of cerato-ulmin from the culture of C. ulmi. Phytoputhof. Z 91,129-146. Talbot, N.J., Ebbole, D.J. and Hamer J.E. (1993) Identification and characterization of MFGI, a gene involved in pathogenicity from the rice blast fungus Mugnuporthe grisea. Plant Cell 5, 1575-1590. Talbot, N.J., Kershaw, M., Wakley, G.E., deVries, O.M.H., Wessels, J.G.H. and Hamer, J.E. (1996) MFGI encodes a fungal hydrophobin involved in surface interactions during infection-related development of the rice blast fungus Mugnuporthe griseu. Plant Cell 8,985-999. Templeton, M.D., Rikkerink, E.H.A. and Beever, R.E. (1 994) Small, cysteine-rich proteins and recognition in fungal-plant interactions. Mol. Plant-Microbe Interact. 7,320-325. Templeton, M.D., Greenwood, D.R. and Beever, R.E. (1995) Solubilization of Neurnsporu crussu rodlet proteins and identification of the predominant protein as the proteolytically processed eus (ccg-2)gene product. Ejip. Mycol. 19, 166-169. Terhune, B.T. and Hoch, H.C. (1993) Substrate hydrophobicity and adhesion of Uromyces urediospores and germlings. Enp. Mycol. 17,241-251. Terras, ERG., Eggermont, K., Kovaleva, V., Raikhel. N.V., Osborn R.W., Kester, A., Rees, S.B.,Torrekens, S . , Van Leuven, F., Vanderleyden, J . , Cammue, B.P.A. and Broekaert. W.F (1995) Small cysteine-rich antifungal proteins from radish: their role in host defence. Plant Cell 7,573-588. Thau, N., Monod, M., Crestani, B., Rolland, C., Tronchin, G., Latg6, J . 2 and Paris, S. (1994) rodletless mutants of A s p e r g i h fumigutus. In$ Immun. 62,43804388. Thomas, D. (1995) Nanotechnology’s many disciplines. Biotechnology 13,439443. Timberlake, W.E. (1990) Molecular genetics of Aspergillus development. Annu. Rev. Genet. 24, 5-36.
HYDROPHOBINS
43
Timberlake, W.E. (1993) Translational triggering and feedback fixation in the control of fungal development. Plant Cell 5, 1453-1460. Trinci, A.P.J. (1974) Astudy ofthekineticsofhyphalextension and branchinitationoffungal mycelia. J. Gen. Micmbiol. 81,225-236. Trinci, A.P.J., Wiebe, M.G. and Kobson, G.D. (1994) The mycelium as an integrated entity. In: 7he Mycota, Vol. I, Growth, Direrentiation and Development (J.G.H. Wessels and F. Meinhardt, eds). pp. 175-193. Springer Verlag, Berlin. Troughton, J.H. and Hall, D.M. (1967) Extracuticular wax and contact angle measurements on wheat (Triticum vulgare L.) Aust. J. biol. Sci. 20,509-525. van den Ackeweken, G.F.J.M., vanKan, J.A.L., Joosten, M.H.A.J.,Muisers, J.M.,Verbakel, H.M. and de Wit, P.J.G.M. (1993) Characterization of two putative pathogenicity genes of the fungal pathogen Cladosporiumfulvum. Mol. Plant-Microbe Interact. 6,210-21 5. van der Valk, P. and Marchant, R. (1978) Hyphal ultrastructure in fruit-body primordia of the basidiomycetes Schizophyllum commune and Coprinus cinereus. Protoplasma 95. 57-72. van der Vegt, W., van der Mei, H.C., Wosten, H.A.B., Wessels, J.G.H. and Busscher, H.J. (1996) A comparison of the surface activity of the fungal hydrophobin SC3p with those of other proteins. Biophys. Chem. 51,253-260. van Wetter, M.-A., Schuren, F.H.J., Schuurs, T.A. and Wessels, J.G.H. (1996) Targeted mutation of the SC3 hydrophobin gene of Schizophyllum commune affects formation of aerial hyphae. FEMS Microbiol. Lett. (in press). Wendland, J., Vaillancourt, L.J., Hegner, J., Lengeler, K.B., Laddison, K.J., Specht, C.A., Raper, C.A. and Kothe, E. (1995) The mating type locus B a l of Schizophyllum commune contains a pheromone receptor and putative pheromone genes. EMBO J . 14,5271-5278. Wessels, J.G.H. (1978) Incompatibility factors and the control of biochemical processes. In: Genetics andktorphogenesis in the Basidiomycetes (M.N. Schwalb and P.G. Miles, eds), pp. 81-104. Academic Press, New York. Wessels, J.G.H. (1986) Cell wall synthesis in apical hyphal growth. Int. Rev. Cytol. 104, 37-79. Wessels, J.G.H. (1990) Role of cell wall architecture in fungal tip growth generation. In: E p Growth ofPlant and Fulgal Cells (I.B. Heath, ed.), pp. 1-29. Academic Press, San Diego. Wessels, J.G.H. (1992) Gene expression during fruiting in Schizophyllum commune. Mycol. Res. 96,609-620. Wessels, J.G.H. (1993a) Wall growth, protein excretion and morphogenesis in fungi. New Phytol. 123,397-413. Wessels, J.G.H. (1993b) Fruiting in the higher fungi. Adv. Micrub. Physiol. 34, 147-202. Wessels, J.G.H. (1994) Developmental regulation of fungal cell wall formation. Annu. Rev. Phytopathol. 32,413437. Wessels, J.G.H. (1996) Fungal hydrophobins: proteins that function at an interface. Trends Plant Sci. 1,9-15. Wessels, J.G.H. and Sietsma, J.H. (1979) Wall structure and growth in Schizophyllum commune. In: Fungal Walls and Hyphal Growth (J.H. Burnett and A.P.J. Trinci, eds), pp. 27-48. Cambridge University Press, Cambridge. Wessels, J.G.H. and Sietsma, J.H. (1981) Fungal walls: a survey. In: Encyclopedia ofPlant Physiology, Vol. 13B (W. Tanner and F.A. Loewus, eds), pp. 352-394. Springer Verlag, Berlin. Wessels, J.G.H., Kreger, D.R.,Marchant,R.,Regensburg,B.A.anddeVries,O.M.H. (1972) Chemical and morphological characterization of the hyphal wall surface of the basidiomycete Schizophyllum commune. Biochim. Biophys. Acta 273,346-358. Wessels, J.G.H., Mulder, G.H. and Springer, J. (1987) Expression of dikaryon-specific and
44
JOSEPH G.H. WESSELS
non-specific mRNAs of Schizophyllum commune in relation to environmental conditions and fruiting. J. Gen. Microbiol. 133,2557-2561. Wessels. J.G.H., de Vries, O.M.H., Asgeirsd6ttir. S.A. and Schuren, F.H.J. (1991a) Hydrophobin genes involved in formation of aerial hyphae and fruit bodies in Schizophyllum. Plant Cell 3,793-799. Wessels, J.G.H. de Vries, O.M.H., Asgeirsdbttir, S.A. and Springer, J. (1991b) The thn mutation of Schizophyllum commune, which suppresses formation of aerial hyphae, affects expression of the SC3 hydrophobin gene. J. Gen. Microbiol. 137,2439-2445. Wessels, J.G.H., Asgeirsd6ttir, S.A., Birkenkamp K.U., de Vries, O.M.H., Lugones, L.G., Schuren, EH.J., Schuurs, T.A., van Wetter, M.A. and Wosten, H.A.B. (1995) Genetic regulation of emergent growth in Schizophyllum commune. Can. J. Bot. 73 (Suppl. l), S2734281. Wildermuth, H., Wehrli, E. and Home, R.W. (1971) The surface structure of spores and aerial mycelium in Streptomyces coelicolor. J. Ultrastrucf. Res. 35, 168-1 80. Williams, S.J. Bradshaw, R.M., Costerton, J.W. and Forge, A. (1972) Fine structure of the spore sheath of some Streptomyces species. J. Gen. Microbiol. 72,249-258. Wnendt, S., Ulbrich, N. and Stahl. U. (1994) Molecular cloning, sequence analysis and expression of the gene encoding an antifungal-protein from Aspergillus giganteus. Cur,: Genet. 25,519-523. Wosten H.A.B., Moukha, S.M., Sietsma, J.H. and Wessels J.G.H. (1991) Localization of growth and secretion of proteins in Aspergillusniger. J. Gen. Microbiol. 137,2017-2023. Wosten, H.A.B.,de Vries, O.M.H. and Wessels, J.G.H. (1993) Interfacial self-assembly of a fungal hydrophobin into a hydrophobic rodlet layer. Plant Cell 5, 1567-1574. Wosten, H.A.B., Schuren F.H.J. and Wessels, J.G.H. (1994a) Interfacial self-assembly of a hydrophobin into an amphipathic membrane mediates fungal attachment to hydrophobic surfaces. EMBO J. 13,5848-5854. Wosten, H.A.B., Asgeirsd6ttir,S.A.,Krook J.H.,Drenth J.H.H. and Wessels, J.G.H. (1994b) The fungal hydrophobin SC3p self-assembles at the surface of aerial hyphae as a protein membrane constituting the hydrophobic rodlet layer. Eur J. Cell Bid. 63, 122-129. Wosten, H,A.B., de Vries, O.M.H., van der Mei, H.C., Busscher, H.J. and Wessels, J.G.H. (1994~)Atomic composition of the hydrophobic and hydrophilic sides of self-assembled SC3p hydrophobin. J. Bacleriol. 176,7085-7086. Wosten,H.A.B.,Ruardy,T.G., vanderMei,H.C.,Busscher,H.J.and Wessels, J.G.H. (1995) Interfacial self-assembly of a Schizophyllum commune hydrophobin into an insoluble amphipathic membrane depends on surface hydrophobicity. Colloids Su$ B: Biointerfaces 5, 189-195. Xiao, J.-Z., Ohshima, A,, Kamakura, T., Ishiama, T. and Yamaguchi, I. (1994) Extracellular glycoprotein(s) associated with cellular differentiation in Magnaporthe grisea. Mol. Plant-Microbe Interacl. 7,639-644. Yaguchi, M., Pusztai-Carey, M., Roy, C., Surewicz, W.K., Carey, P.R., Stevenson, K.J., Richards W.C. and Takai, S. (1993) Amino-acid sequence and spectroscopic studies of Dutch elm disease toxin, cerato-ulmin. In: Dutch Elm Disease Research, Cellulrr and Molecular Approaches (M.B. Sticklen and J.L. Sherald, eds), pp. 152-170. Springer Verlag, New York. Yli-Mattila, T., Ruiters, M.H.J., Wessels, J.G.H. and Raudaskoski, M. (1989a) Effect of inbreeding and light on rnonokaryotic and dikaryotic fruiting in the homobasidiomycete Schizophylltiin commune. Mycol. Res. 93,535-542. Yli-Mattila, T., Ruiters, M.H.J. and Wessels, J.G.H. (1989b) Photoregulation of dikaryonspecific mRNAs and proteins by UV-Alight in Schizophyllum commune. Cur,: Microbiol. 18,289-295. Zentmeyer, G.A. (1942) Toxin formation by Cerastornella ulmi. Science 9 5 , 5 12-513.
HYDROPHOBINS
45
Zhang, L., Villalon, D., Sun, Y., Kazmierczak, P. and van Alfen, N.K. (1994) Virus-
associated down-regulation of the gene encoding cryparin, an abundant cell-surface protein from the chestnut blight fungus Ctyphonectriu parusitica. Gene 139,594.
This Page Intentionally Left Blank
Structure-function Analysis of the Bacterial Aromatic Ring-hydroxylating Dioxygenases Clive S. Butler, and Jeremy R. Mason Division of Laye Sciences, King’s College London,Campden Hill Road, London W8 7AH. UK *Present address: School of Biological Sciences, Molecular and Microbiology Sectol; University of East Anglia, Nonvich NR4 T J , UK 1. 2. 3. 4.
5. 6.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bacterial oxygenases . . . . . . . . . . . . . . . . . . , . . . . , . . . . . . . Structure of ring-hydroxylating dioxygenases . . . . . . . . . . . . . . . . . . . Electron transport system . . . . . . . . . , , . , , . , , . . . . . . . . . . . . 4.1. The reductase component . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. The ferredoxin component . . . . . . . . . . . . . . . . . . . . . . . . . . The catalytic terminal oxygenase component . . . . . . . . . . . . . . . . . . . Coordination of the iron-sulphur clusters . . . , , , . . . . . . . . . . . . . . . 6.1. Evidence that the dioxygenase ISP [2Fe-2SI clusters are of the Rieske type with imidazole ligands . . . . . . . . . . . , . . . . . . . . . . . . . . . . 6.2. Spectroscopic evidence , , . . , , . . , , , . , . , . , . , , , , . . . . 6.3. Comparison of the amino-acid sequences of [ZFe-ZSI Rieske proteins and the dioxygenase ISP a subunits . . . . . . , . . . . . . . . . . . . . . . . 6.4. Site-directed mutagenesis experiments on Rieske iron-sulphur proteins . . The catalytic non-haem iron centre . . . . . . . . . . . . . . . . . . . . , . . . Concluding remarks . . . . . . . . . . . . . . . , . . . , , . . . , . . . , . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . , . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . I
7. 8.
. . 47
..
48
. . 50 . . 51
. . 51 . . 58 . . 60 . . 61 . . 63 . . 63 . . 67 . . 68 . . 72
..
75
. . 76 . . 76
1. INTRODUCTION Soil microorganisms, particularly bacteria of the genus Pseudomonas, are able to metabolize an enormous range of natural and synthetic organic compounds, and are important agents in the degradation and recycling of organic material (Gibson, 1988). The initial step in the aerobic microbial biodegradation of aromatic ADVANCES M MICROBIAL PHYSIOuXiY VOL 38 ISBN 0-12-027738-7
Copyright B 1997 Academic Press Limited All rights of reproduction in any form reserved
40
CLIVE S. BUTLER AND JEREMY R. MASON
compounds is usually the introduction of two hydroxyl groups into the benzene ring, forming cis-dihydrodiols or cis-diol carboxylic acids. This reaction prepares the aromatic compound for further fission and catabolism. The enzymes that catalyse the insertion of molecular oxygen into aromatic substrates are called oxygenases (Hayaishi, 1962). Oxygenases have generated a great deal of interest as they possess the ability to carry out stereo- and regio-specific reactions under mild conditions (Ziffer et al., 1977; Lee et al., 1995). This has led to their use as biocatalysts for the synthesis of chemicals, which are notoriously difficult to produce by conventional chemical methods. Examples of such compounds include cis-glycols (Dalton, 1986) and inositol-related compounds (Ley et al., 1987). Other potentially useful biotransformations are reviewed by Sheldrake (1992) and Hudlicky et al. (1994). In addition, developments in environmental biotechnology have resulted in biodegradation becoming a useful strategy for the removal of hazardous aromatic compounds from soil and ground water (Fetzner and Lingens, 1994). In order to utilize the full potential of oxygenase enzymes, it is important to understand the molecular structure of the proteins. In particular, their manipulation, to effect increased substrate specificity and enhance specific activity, is dependent on an understanding of the environments of the redox centres and substrate binding-si te.
2. BACTERIAL OXYGENASES
The importance of oxygenases in physiological processes can be appreciated when one considers the chemistry of dioxygen. It is generally understood that, while the reaction of the resonance-stabilized benzene ring with oxygen is thermodynamically favoured, there are significant kinetic barriers inherent in these processes. Because molecular oxygen has a triplet electronic ground state, any adiabatic process involving electron transfer from a stable, closed shell (singlet) organic species to dioxygen results initially in the formation of a triplet diradical product. In enzyme-catalysed reactions, this is accomplished by dioxygen reacting with either the unpaired d electrons of a transition metal ion complex or with a stable organic free radical, such as a flavin semiquinone. The resulting activated oxygen species (peroxides, hydroxyl radicals, superoxide anion, singlet oxygen) are highly reactive and, if allowed to gain access to a cell, would result in considerable oxidative damage. Thus, in addition to activation of dioxygen, oxygenase enzymes are also able to constrain the active oxygen species and direct it towards the hydroxylation of the aromatic substrate. The active sites of such enzymes have evolved to channel the free-radical reaction and yield a specific product, rather than the mixture of products, which would result if reactions of this kind were attempted
BACTERIAL AROMATIC R ING-HYDROXYLATING DlOXYGENASES
49
in solution (Halliwell and Gutteridge, 1984). Transition metals, such as iron, serve as catalytic centres in numerous oxygenases; some utilize non-haem Fe(I1) or Fe(II1) (Feig and Lippard, 1994), whereas others are haem enzymes (cytochrome P-450;Raag and Poulos, 1989). In some cases other transition metals are used to bind dioxygen. The tetrahydrobiopterin-dependent phenylalanine hydroxylase from Chrornobacteriuinvioluceum has been shown to utilize copper (CuII) instead of iron for enzymatic activity (Pember et al., 1989) and 3,4-dihydroxyphenylacetate 2,3-dioxygenase has been reported to be a manganese-containing enzyme (Que et al., 1981). Oxygenases may be classified into two groups, the monooxygenases, which incorporate one atom of oxygen into one molecule of substrate and the dioxygenases, which incorporate both atoms of dioxygen into the substrate (Hayaishi, 1962). This general definition fails to account, however, for isopenicillin N synthase, a dioxygenase that catalyses the oxidative ring closure reactions of &(L-a-aminoadipoy1)-L-cysteinyl-D-valine (ACV) to form isopenicillin N, the precursor to all other penicillins (Pang et al., 1984). This reaction results in the complete four-electron reduction of a single equivalent of dioxygen to form two equivalents of H20 (Chen et al., 1989). Monooxygenase-catalysed reactions result in the unincorporated oxygen atom being reduced to water. Thus, these enzymes function as part oxygenase and part oxidase and, as a consequence, are termed mixed-function oxidases (Mason, 1988). It is, however, not always possible to distinguish between a dioxygenase and monooxygenase by activity alone. For example, both toluene and naphthalene dioxygenases can catalyse monooxygenase reactions as in the oxidation of indene and indan to l-indenol and l-indanol, respectively (Wackett et al., 1988; Brand et al., 1992) or desaturation reactions (Gibson et al., 1995). Toluene dioxygenase has also been reported to catalyse the monohydroxylation of phenols to the corresponding catechols (Spain ef al., 1989) and the stereospecific oxidation of aryl alkyl sulphides to sulphoxides (Lee ef al., 1995). Conversely, 4-methoxy benzoate monooxygenase has been shown to dihydroxylate the vinylic side chain of 4-methoxystyrene (Wende et al., 1982). In addition, enzymes such as the 2-0x0-1,2-dihydroquinoline 8-monooxygenase have not been demonstrated to have any dihydroxylation activity, although their structure places them clearly with the aromatic ring-hydroxylating non-haem iron dioxygenases (Rosche ef al., 1995). The dioxygenases are a large group of enzymes that have been divided into distinct groups, owing to the different reactions that they catalyse. The enzymes are involved in the oxidation of aromatic compounds to produce dihydroxy non-aromatic intermediates, catalysing the cleavage of the aromatic nucleus or oxidative ring-closure reactions. The first two of these separate groups are termed the ring-hydroxylating dioxygenases and the ring-fission dioxygenases. The remainder of this chapter will focus on the ring-hydroxylating dioxygenases, referring to ring-cleavage dioxygenases and other related systems only where appropriate (see Harayama et al., 1992, for a detailed review of oxygenases).
50
CLIVE S. BUTLER AND JEREMY R. MASON
3. STRUCTURE OF RING-HYDROXYLATING DIOXYGENASES
The most extensively studied ring-hydroxylating dioxygenase enzyme systems have been isolated from Pseudomnas species. Gibson and co-workers (1968, 1970) were the first to show that a dioxygenase system of Pseudomonas putidu oxidized benzene to generate cis- 1,2-dihydroxy-cyclohexa-3,S-diene (cis-benzene dihydrodiol). Several other ring-hydroxylating dioxygenases have since been isolated and characterized from other strains of Pseudoinonus: pyrazon dioxygenase (Sauber et al., 1977); naphthalene dioxygenase (Ensley et al., 1982; Takizawa et al., 1994; Yang etal., 1994); toluene dioxygenase (Subramanian et al., 1979, 1981, 1985); benzoate 1,2-dioxygenase (Yamaguchi and Fujisawa, 1978, 1980, 1982); phthalate dioxygenase (Batie e f al., 1987); 4-chlorophenylacetate 3,4-dioxygenase (Markus et al., 1986); 2-halobenzoate 1,2-dioxygenase (Fetzner etal., 1992); ortho-halobenzoate 1,2-dioxygenase (Romanov and Hausinger, 1994; Haak et al., 1995); benzene dioxygenase (Axcell and Geary, 1975; Irie et al., 1987); biphenyl dioxygenase (Haddock et al., 1993; Erickson and Mondello, 1993); and 2-nitrotoluene 2,3-dioxygenase (An et al., 1994). Ring-hydroxylating dioxygenases have also been isolated from Acznetobacter (benzoate 1,2-dioxygenase, Neidle et al., 1991), Commonas (4-sulphobenzoate 3,4-dioxygenase, Locher et al., 1991; terephthalate 1,Zdioxygenase, Schlafli et al., 1994; biphenyl dioxygenase, Hurtubise etal., 1995; naphthalene dioxygenase, Goyal and Zylstra, 1996), Alcaligenes (chlorobenzoate 3,4-dioxygenase, Nakatsu and Wyndham, 1993; Nakatsu et al., 1995), Sphingomonus (dibenzofuran 4,4a-dioxygenase, Biinz and Cook, 1993) and Rhodococcus (phthalate dioxygenase, Suemori et al., 1993; biphenyl dioxygenase, Wang et al., 1995). All the enzymes are non-haem iron dioxygenases, which oxidize the aromatic substrates to give cis-dihydrodiols or cis-diol carboxylic acids, although in some cases these products probably dehydrogenate spontaneously to give the corresponding catechol (Locher et al., 1991; Romanov and Hausinger, 1994;Ha& etal., 1995). In addition to molecular oxygen, these enzymes have been shown to have requirements for cofactors, NADH or NADPH, and in the majority of cases, iron (reviewed by Mason and Cammack, 1992). All of the ring-hydroxylating dioxygenases are soluble, multi-component enzymatic systems, comprising two or three separate proteins. A typical arrangement consists of a short electron-transport chain composed of an iron-sulphur flavoprotein, or a flavoprotein and an iron-sulphur ferredoxin, which transfer electrons to a catalytic terminal oxygenase (Fig. 1). The subunit composition of the ring-hydroxylating dioxygenases differs considerably (Fig. 2). If the reductase component contains a [2Fe-2S] cluster, the system tends to lack the ferredoxin component. However, an exception to this is the naphthalene dioxygenase (Ensley et al., 1982), where both the reductase and the ferredoxin are present, and both contain a [2Fe-2S] cluster. Variation is also encountered in the number and size of
51
BACTERIAL AROMATIC RING-HYDROXYIATING DIOXYGENASES
NAD +
,x \
FAD or FMN
Hcduced
Ride
[ZFe-S] Fmedoxin
12Fe-2SJ Fen Ternid orygmaw ISP
Aromatic
L4 Oxidlzcd
xo,+w'
Reduced
subshate
Figure 1 Biochemical organizationof the multi-componentring-hydroxylatingdioxygenases. A flavoprotein reductase accepts electrons from NADH and transfers them via an iron-sulphur cluster or a ferredoxin to a terminal oxygenase, which is a large iron-sulphur protein. The reduced terminal oxygenase catalyses the oxidation of aromatic substrates to cis-dihydroxy products.
the subunits of both the terminal oxygenase and the reductase components. These subunit variations have led to the classification of the three groups of dioxygenases by Batie and co-workers (1992)as Class I, Class I1 and Class I11 (Table 1). However, with the purification and characterization of numerous dioxygenases, it has become clear that this classification based on the enzyme as a whole may be inappropriate, since the individual components exist separately in the cell and in some cases the electron transfer components of one enzyme may complement those of another. This would suggest that each component may be classified better as an enzyme in its own right with a separate EC number.
4. ELECTRON TRANSPORT SYSTEM
4.1. The Reductase Component
The reductase component is the first component in the electron-transfer chain of dioxygenases. It acts as an oxidoreductase, catalysing the transfer of electrons from NADH to a second redox component. The majority of reductases studied use
Table 1 Multi-component hydroxylating dioxygenases.
Reductase Class IA
IA IA IA
IB IB
IB
Enzyme system (reference) Phthalate dioxygenase(1.2) 4-Chlorophenylacetate 3,4dioxygenase (3.4) 4-Sulphobenzoate 3,4-dioxygenase (5) Chlorobenzoate 3,4-dioxygenase (6) 2-Halobenzoate 1,2-dioxygenase (7) Benzoate dioxygenase (8-1 1)
Ferredoxin
Mr
Prosthetic grouP(s)
Mr
34 OOO 35 OOO
FMN [2Fe-2S] FMN [2Fe-2S]
None None
a(48 00014 a(46 000)3
4[2Fe-2S]~, 4Fe 3[2Fe-2S]
36 OOO
FMN [2Fe-2S]
None
M50 000)2
2[2Fe-2s]~,2Fe
42 OOO
nd
None
M51000)n
37 500
FAD [2Fe-2S]
None
a(52 000)3 K20 o w 3
37 500
FAD [2Fe-2S]
None
a(50000)3
3[2Fe-2S], 3Fe
None
p(20 00013 a(55 00016
6[2Fe-S ]R, ?Fe
l-oxo-1,2-dihydroquinoline 38 OOo 8-monooxygenase (1 2)
FAD [2Fe-2S]
Prosthetic group($)
Oxygenase Structure (Mr)
Prosthetic group(s)
nd
3[2Fe-2S], 3Fe
Table 1 continued Reductase Class IIA IIA IIB
IIB
IIB 111
Enzyme system (reference) Pyrazon dioxygenase ( I 3) Dibenzofuran 4,4a-dioxygenase (14) Benzene 1,2-dioxygenase (15-21) Toluene 2,3-dioxygenase (22-26) Biphenyl dioxygenase (27-3 1) Naphthalene dioxygenase (32-36)
Mr
67 000
Prosthetic grouds)
Ferredoxin Mr
Prosthetic group(s)
MOO0
FAD FAD
12000 12OOO
[2Fe-2S] [2Fe-2SJ
87 168Q
FAD
11 939'
[2Fe2s]~
42 942'
FAD
11 90(lQ
[2Fe--2S]~
42970'
FAD
11 98 1'
[2Fe-2sJ~
35 552a
FAD [2Fe-2S]
11 767'
[2Fe-2s]~
(a
Oxygenase Structure (Mr) 180 000 a(45 000)2 P(23 000)2 c1(5 1 105')2 p(22 252')~ ~ ( 5 930")2 0 p(22 0 I3')2 a(51 5 13')~ p(22 085")~ a(49 597'12 p(23 038")~
Prosthetic group(s) [2Fe-2S] 2[2Fe-2S], 2Fc 2[2Fe-x]~,2Fe 2[2Fe-2s]~,2Fe 2[2Fe-2S]~, 2Fe 2[2Fe-x]~,2Fe
R indicates a Rieske-type [2Fe-2S] cluster; nd, not determined; Mr observed from SDS-PAGE or =deduced from amino-acid sequence. References: 1, Batie et al., 1987; 2, Batie and Ballou, 1990,3, Schweizer etal., 1987; 4, Markus et al., 1986; 5, Locher et af., 1991; 6 , Nakatsu and Wyndham, 1993; 7. Fetzner et al., 1992;8, Yamaguchi and Fujisawa, 1978;9, Yamaguchi and Fujisawa, 1980; 10, Yamaguchi and Fujisawa, 1982; 11, Moodie et a].. 1990; 12, Rosche et al., 1995; 13, Sauber etal., 1977; 14, Bunz and Cook, 1993; IS, Axcell and G e q , 1975; 16. Geary et al., 1990; 17, Zamanian and Mason, 1987; 18, Geary et al., 1984; 19, Tan and Mason, 1990; 20. Tan ef al., 1993; 21, Crutcher and Ceary, , 1; 26. Subramanian et al., 1985; 27, 1979; 22, Zylstra and Gibson, 1989; 23, Wackett, 1990; 24, Zylstra et al., 1988; 25, Subramanian et ~ l . 198 Furukawa el a!., 1993; 28, Mondello. 1989; 29, Erickson and Mondello, 1992; 30, Gibson ef al., 1993; 3 1, Taira et al., 1992; 32, Haigler and Gibson, 1 9 9 0 33, ~ Haigler and Gibson, 199Ob; 34, Ensley et al., 1982; 35, Ensley and Haigler, 19% 36, Suen and Gibson, 1993.
54
CLIVE S. BUTLER AND JEREMY R. MASON
Electron Class transport chain
a
Oxygenase
Example
4-sulphobenzoate 3,4-dioxygenase
0
4chlorophenylacetate 3,4-dioxygenase
a
phthalate 4,5-dioxygenase
2-oxo-l ,Z-dihydroquinoline Smonooxygenase
benzoate 1,2-dioxygenase
Q
IIA
IIB
@
111
-
.,.,.
Ferredoxin
benzene 1,Pdioxygenase naphthalene 1,Bdioxygenase
0
Reductase
0 [2Fe-2SI
0
dibenzofuran 4,4a-dioxygenase
...... Reductase FMN,[2Fe-2S] [2Fe-2S] R Ferredoxin
[2F&!S]~,Fe(ll), Oxygenase (Ispa 1
ISPP
Figure 2 Schematic diagram to show the subunit composition of hydroxylatiiig bacterial dioxygenases.
BACTERIAL AROMATIC RING-HYDROXYLATING DIOXYGENASES
55
NADH as the preferred reductant. Benzene and toluene dioxygenase are specific for NADH (Axcell and Geary, 1975; Subramanian et al., 1981), whereas NADPH served as an electron donor for naphthalene dioxygenase, but its activity was 43% of that observed with NADH (Ensley et al., 1982). In contrast, the 2-nitrotoluene 2,3-dioxygenase has been demonstrated to be fully active in the presence of either NADH or NADPH (An et al., 1994). The stereospecificity of hydride removal from NADH during catalysis by the reductases has been examined by Cook and co-workers (Schlafli et al., 1995). Dioxygenases classified as members of Classes IA, IB and I11 were shown to be pro-R specific, whereas the reductases of the Class I1 enzymes dibenzofuran dioxygenase (Class HA) and benzene dioxygenase (Class IIB) were demonstrated to be pro-S specific (Table 2). The active site of the oxygenase component contains a single electron acceptor in the form of a [2Fe-2S] cluster. Since NADH can transfer both of its electrons simultaneously only as a hydride, the reductase is required to mediate this transition. Thus the reductases contain flavin, either as FMN or FAD. Flavins are distinctive among biological coenzyines in that they can accept two electrons, as hydrides from NADH, and transfer these electrons one at a time via the semiquinone redox state to other electron-transfer components. All the Class I dioxygenases are two-component systems that have a reductase component with a [2Fe-2S] cluster. This class is subdivided into IA and IB, depending on whether they contain FMN or FAD, respectively. These reductase components have a relative molecular mass (M,) ranging from 34 kDa for phthalate dioxygenase (Batie et al., 1987) to 42 kDa for chlorobenzoate 3,4-dioxygenase (Nakatsu and Wyndham, 1993) and are all monomers. The properties of these iron-sulphur flavoproteins are listed in Table 2. As these flavoproteins contain a [2Fe-2S] cluster, the absorption spectra of the oxidized proteins show absorption maxima at -340 nm and -460 nm. The most detailed analysis has been carried out on the phthalate dioxygenase reductase (Batie et al., 1992; Gassner et al., 1995). The recently determined structure (Batie el al., 1992; Come11 et al., 1992) shows the protein to be comprised of three domains. The FMN domain is at the N-terminus of the chain, and the [2Fe-2S] domain is at the C-terminus. The domains are arranged to bring the FMN and [2Fe-2S] prosthetic groups together near the centre of the molecule, with the flavin adjoining the binding site for the pyridine nucleotide. The [2Fe-2S] domain includes a four-stranded mixed P-sheet and two helices in a fold that resembles the [2Fe-2S] ferredoxin from Anabaena (Batie et al., 1992). The amino-acid sequence of the iron-sulphur flavoprotein of benzoate dioxygenase (Neidle et al., 1991) shows some homology with the plant-type ferredoxins. EPR spectra of the reduced reductases from phthalate, 4-chlorophenylacetate and benzoate dioxygenases show typical features of plant-type [2Fe-2S] clusters, together with a signal around g = 2.004 from the flavosemiquinoneradical (Batie et al., 1987; Schweizer et al., 1987; Altier et al., 1993); thus it appears that the iron-sulphur clusters of the Class I dioxygenases are of the plant type (see Section 6). These iron-sulphur flavoproteins are also capable of transferring
Table 2 Properties of the reductase component of aromatic dioxygenases.
Dioxygenase (reference) Phthalate (1,2) 4-Chloropheny lacetate (3) 4-Sulphobenzoate (4) 2-Halobenzoate (5) Benzoate (6) 4 . m n (7) Dibenzofuran (8) Benzene (9) Toluene (10) Naphthalene (1 1)
Absorption
Stereospecificityb
Component
Prosthetic group(s)
~Lllad(nm)
EPR (BX, gy, g3
Reductase B
FMN [2Fe2SJ FMN [2Fe-2S]
330,420,462,495', 530' 336,394,458
1.900, 1.949,2.008,2.041 1.90, 1.94,2.004,2.03
pm-R pro-R
B B
FMN [2Fe-2S]
-330, -400,465 272,343,457 273,340,402,467 450,475 nd 385,425,455 372,488,475 278,340,420,460,54@
nd nd 1.88, 1.95.2.04 ND ND ND ND ND
nd pro-R nd nd pro-s pro-S nd pro-R
FAD [2Fe-2S] Reductase FAD [2Fe-2S] FAD A2 An FAD R ~ U C ~ ~ W B E D FAD Reductase~o~ FAD RductaseNAp FAD [2Fe-2S]
Shoulder. Data taken from Sch1;ifi et al. (1995). nd, not determined; ND, not detected. References: 1, Batie etal., 1987; 2, Correll et al., 1992; 3, Schweizer etal., 1987; 4 Locher et al., 1991; 5, Fetzner el al., 1992; 6, Yamaguchi and Fujisawa, 1978; 7, Sauber et al., 1977; 8, Biinz and Cook, 1993; 9, G a r y et al., 1990; 10, Subramanian et al., 1981; 11, Haigler and Gibson, 1990a.
a
BACTERIAL AROMATIC RING-HYDROXYLATINGDIOXYGENASES
57
electrons directly from NADH to an artificial electron acceptor. Thus, the reductase components of benzoate, phthalate, 4-chlorophenylacetate and 2-halobenzoate dioxygenasescan reduce cytochrome c directly in the presence NADH (Yamaguchi and Fujisawa, 1978; Batie etal., 1987; Schweizereta1.,1987; Fetzner et al., 1992). Interestingly, studies on the 4-sulphobenzoate 3,4-dioxygenase have identified two possible reductase components (B and C), both capable of transferring electrons to the terminal oxygenase. However, owing to lower activity with component C, component B is regarded as the authentic electron donor (Locher et al., 1991). Classes I1 and 111 contain the three-component dioxygenases. Their reductases contain FAD, which has been reported to be readily lost during purification (Mason and Cammack, 1992). The Class I1 flavoproteins consist of a single polypeptide with an Mr ranging from 42.9 kDa for toluene dioxygenase (Zylstra and Gibson, 1989) to 67 kDa for pyrazon dioxygenase (Sauber et al., 1977). All are monomeric except for the reductase of benzene dioxygenase, which is a dimer (Geary et al., 1990). This class is subdivided into groups A and B. However, they vary only in the structure of the ferredoxin and terminal oxygenase components. The properties of these reductdses are listed in Table 2. As no detailed structures are known for this group of reductases, the probable cofactor binding sites in the reductase proteins can only be inferred from their amino-acid sequences and comparison with other flavoproteins. When the amino-acid sequence of the reductase of benzene dioxygenase (Tan et al., 1993), toluene dioxygenase (Zylstra and Gibson, 1989) and biphenyl dioxygcnase (Kikuchi et al., 1994) are compared with flavincontaining putidaredoxin reductase, which performs an analogous function in the cytochrome P-450-dependent camphor hydroxylase (Koga et al., 1989) or other NADH-dependent flavoproteins like dihydrolipoamide dehydrogenase, it is clear that two highly conserved regions are located at the N terminus and the middle region. Both are centred around a pap-fold, with the consensus sequence:
Gly-Xaa-Gly-Xaa2-GIy-Xaa3-Ala-Xa~-Gly (where Xaa is any amino acid and the subscript represents the number of amino-acid residues). The N-terminus and middle conserved regions are predicted to bind the ADP moiety of FAD and NAD', respectively. Tan and co-workers (Tan ef al., 1993) have identified a third conserved region that is located at the C-terminus of the benzene dioxygenase reductdse and forms a P-sheet (aa 265-275) with the consensus sequence of Thr-Xa%-Ala-Xaa-Gly-Asp. This conserved region is predicted to bind the 0 - 3 group of the ribityl chain of the flavin moiety of FAD (Wierenga et al., 1986; Eggink et al., 1990). Like the Class I enzyme 4-sulphobenzoate 3,4-dioxygenase, dibenzofuran 4,4a-dioxygenase has been shown to possess two reductases (A1 and A2). The two reductases, though physically similar to each other, were shown to have very different N-termini [reductase A1 : Xaa-Gln-Tyr-(Asp or G1y)-Val-Leu-(Ile or A1a)-Val-GIy-Ala-Leu; reductase A2: Met-Arg-Ser-Ala-Asp-Val-Val-Tle-Val-(Arg
58
CLIVE S. BUTLER AND JEREMY R. MASON
or Ala or Gly)], both reductases could complement dioxygenase activity (Biinz and Cook, 1993). The Class I11 enzyme naphthalene dioxygenase also consists of three components. Similar to those found in Class I, the reductase component is an iron-sulphur flavoprotein that can catalyse the reduction of cytochrome c. The absorption spectrum is also similar to the spectra of the iron-sulphur flavoproteins, with absorption maxima at 278,340,420 and 460 nm, and a broad shoulder at 540 nm (Haigler and Gibson, 1990a) (Table 2). Despite the presence of the reductase [2Fe-2S] cluster, this enzyme still requires a ferredoxin in order to transfer electrons to the terminal oxygenase. The reductase has a Mr of 36.3 kDa (Haigler and Gibson, 1990a).
4.2. The ferredoxin component
The ferredoxins are one-electron acceptors, transferring electrons from the reductase to the terminal oxygenase and are found only i n ring-hydroxylating dioxygenases in Classes I1 and 111. They are small acidic proteins with an Mr of approximately 12 kDa as determined by sequence analysis, but appear larger when observed by SDS-PAGE (Geary et al., 1990). Each contains 2 gram-atoms of iron and acid-labile sulphide per mole in the form of a [2Fe-2S] cluster. In addition to transferring electrons from the flavoprotein to the terminal oxygenase, the ferredoxin can catalyse a flavoprotein-dependentreduction of other electron acceptors, such as cytochrome c (Axcell and Geary, 1975; Subramanian ef a!., 1985). Studies on the toluene dioxygenase by Gibson and co-workers have shown that analogous electron-transfer proteins, such as spinach ferredoxin, ferredoxin from Clostridium pasteurianum, putidaredoxin and adrenodoxin cannot replace ferredoxinToL in either cytochrome c reduction or toluene oxidation (Subramanian et al., 1985). Similarly, ferredoxinToL could not substitute for ferredoxinNAp in the naphthalene dioxygenase (Haigler and Gibson, 1990b). However, ferredoxinBED could substitute for ferredoxinmL, maintaining a functional toluene dioxygenase (Tan, 1991) and ferredoxinNAp could replace ferredoxinZNT*, maintaining a functional 2-nitrotoluene 2,3-dioxygenase (An et al., 1994). Thus, ferredoxins are apparently rather specific for the dioxygenase system (class) in which they act. This specificity is reflected in the primary structure of the proteins. The ferredoxins of different ring-hydroxylating dioxygenases possess similarities in both size and amino-acid sequence. FeffedOXinNAp comprises 104 amino acids, whilst both ferredoxinBED and ferredoxinmL have 107 amino acids (Mason and Cammack, 1992). Sequence analysis has revealed that ferredoxinBED and ferredoxinmL are almost identical, showing 99% homology in amino-acid sequence and 75% homology in the nucleic-acid sequence. In contrast, ferredoxinNAp ~~~
*2NT is suggested to be a Class I11 dioxygenase similar to NAP (An et al., 1994).
59
BACTERIAL AROMATIC RING-HYDROXYLATING DIOXYGENASES
has only 34% homology with the amino-acid sequence of ferredoxinBED (Kurkela er al., 1988). The EPR spectrum of the ferredoxin of the pyrazon dioxygenase resembles that of the P-450 monooxygenases, such as putidaredoxin with g-factors of 2.02 and 1.94 (Sauber et al., 1977) (Table 3), indicating the presence of a typical plant-type ferredoxin [2Fe-2S] cluster (see Section 6). In addition, the absorption spectrum displays maxima at 411 nm and 453 nm, typical of the doublet peak in the 410463 nm region of the absorption spectrum exhibited by other well-studied [2Fe-2S] ferredoxins (Bruschi and Guerlesquin, 1988). Similarly, the ferredoxin of dibenzofuran 4,4a-dioxygenase displays absorption maxima at 410 nm and 463 nm (Biinz and Cook, 1993), typical of the plant-type ferredoxins. In contrast, the ferredoxins of the benzene dioxygenase and toluene dioxygenase have [2Fe-2S] clusters with unusual spectroscopic and redox properties (Table 3). FerredoxinToL displays an extinction coefficient at 460 nm that is approximately 70% of the value for the plant-type ferredoxins at 415 nm (Subramanian et al., 1985), and is nearly equivalent to that of the Rieske iron-sulphur proteins (Fee et al., 1986). The midpoint redox potentials of the ferredoxins of benzene dioxygenase (Geary et al., 1984) and toluene dioxygenase (Subramanian et al., 1985) are -155 mV and -109 mV, respectively. These are significantly higher than those of spinach ferredoxin (-420 mV), adrenodoxin (-274 mV) or putidaredoxin (-235 mV), but are closer to those of the more negative members of the Rieske iron-sulphur proteins (-140 to +350 mV) (Cammack, 1984). The resemblance to the Rieske iron-sulphur proteins is also evident from the EPR spectra of ferredoxinToL and ferredoxinBED. The average g-factors of 1.92 and 1.84 for ferredoxinBED and ferredoxinmL, Table 3 Properties of the ferredoxin component of aromatic dioxygenases. ~
~ _ _ _
Dioxygenase (reference)
Component
Pyrazon (1) Di benzofuran
B B
_ _ _ _ _ ~
Prosthetic group
Absorption
EPR
(Amax) (nm)
(gt, gy, gz)
Redox potential (mV)
[2Fe-2S] [2Fe-2S]
411,453 4 10,463
1.94,2.02 nd
nd nd
(2) Benzene (3,4) FeITedoXinBED
[2Fe-2S]
Toluene (5)
[2Fe-2S]
280,320,456 1.834, 1,890, 2.026 277,327,460 1.81, 1.86, 2.01 278,321,455 1.82, 1.905, 2.02 280,325,460 nd
FeiredoxinmL
ortlw-HaloC [2Fe-2S] benzoate (6) Naphthalene Femedoxin~~p [2Fe-2Sl (7)
--155
-109
nd nd
nd, not determined. References: 1 , Snuber el al., 1977; 2, Biinz and Cook, 1993; 3, Geary et al., 1990; 4, Tan et a/.. 1994; 5, Subramanian el al., 1985; 6, Romanov and Hausinger, 1994; 7, Haigler and Gibson, 1990b.
60
CLIVE S. BUTLER AND JEREMY R. MASON
respectively, are more similar to those of the Rieske clusters (gav= 1.91) than those of the plant-type ferredoxins (gav= 1.96). Examination of the amino-acid sequence of the ferredoxins of benzene dioxygenase (Momce ef al., 1988; Tan et al., 1993), toluene dioxygenase (Zylstra and Gibson, 1989), naphthalene dioxygenase (Haigler and Gibson, 1990b) and biphenyl dioxygenase (Erickson and Mondello, 1992) show that two cysteine and two histidine residues are highly conserved. This suggests that the dioxygenase ferredoxin [2Fe-2S] clusters are of the Rieske-type (see Section 6 for the coordination of iron-sulphur clusters). Recently, site-directed mutagenesis studies were performed in our laboratory on the ferredoxin of benzene dioxygenase, in which the two conserved histidine residues were mutated to cysteines, both separately and in combination. All of these mutations prevented the insertion of the cluster in viva These data suggest that ferredoxinBED and ferredoxinToL, and by analogy ferredoxinBpH and ferredoxinNAp, are of the Rieske type.
5. THE CATALYTIC TERMINAL OXYGENASE COMPONENT The catalytic terminal oxygenase component is also an iron-sulphur protein. In addition to the [2Fe-2S] cluster, it contains the substrate binding site and a mononuclear non-haem iron; all are essential for the hydroxylation of an aromatic substrate. This component shows considerable variation in subunit configuration. Class IA oxygenases consist of a large (MI = 46-51 kDa) subunit arranged as a dimer, trimer or tetramer for 4-sulphobenzoate 3,4-dioxygenase (Locher et al., 1991), 4-chlorophenylacetate 3,4-dioxygenase (Schweizer et al., 1987) and phthalate dioxygenase (Batie et al., 1987), respectively. Most of the Class IB oxygenases examined to date consist of a larger (a)and a smaller (MI -20 kDa) (p) subunit in the configuration of a$3. The exception to this is the 2-oxo-l,2-dihydroquinoline 8-monoxygenase that consists of six identical a subunits (Rosche etal., 1995). The Class I1 and 111oxygenases characterized also consist of both a and p subunits but in the configuration a&, e.g. benzene dioxygenase (Zamanian and Mason, 1987), or a&, e.g. biphenyl dioxygenase (Haddock and Gibson, 1995). All the terminal oxygenases have one [2Fe-2S] cluster per ap-dimer or a monomer (Mason and Cammack, 1992).Studies on benzoate 1,Zdioxygenase (Yamaguchi and Fujisawa, 1982) and naphthalene dioxygenase (Suen and Gibson, 1993,1994), have demonstrated that the [2Fe-2S] cluster is situated in the a subunit. Whilst an early study on the toluate 1,2-dioxygenase reported the fi subunit to be involved in substrate recognition (Harayama et al., 1986), more recent studies on the biphenyl dioxygenase have shown that the a subunit is the major component involved in controlling substrate specificity (Furukawa et al., 1993; Erickson and Mondello, 1993; Tan and Cheong, 1994).
BACTERIAL AROMATIC RING-HYDROXYIATING DIOXYGENASES
61
Furukawa and co-workers have also demonstrated that substitution of the biphenyl dioxygenase a subunit with the toluene dioxygenase a subunit resulted in the formation of a functional multicomponent dioxygenase, capable of the oxidation of toluene preferentially (Furukawa et al., 1993). Similarly, the replacement of the iron-sulphur protein (ISP) a subunit of biphenyl dioxygenase with the ISP a subunit of benzene dioxygenase resulted in a hybrid enzyme with greater activity towards benzene than biphenyl (Tan and Cheong, 1994). Site-directed mutagenesis has been employed to determine the region of the biphenyl dioxygenase a subunit involved in controlling substrate specificity. Comparison of the substrate specificity of two biphenyl dioxygenases from Pseudomonas sp. LB400 and Pseudomonas pseudoalcaligenes KF707 revealed that the former had a much broader substrate range than the latter; however, the enzyme from Kl707 had a greater capacity to hydroxylate congeners chlorinated in the double para post‘ion. The replacement of four residues of the LB400 dioxygenase subunit with the corresponding residues from the KF707 enzyme produced an engineered dioxygenase with the broad substrate range of LB400 combined with better activity towards the 4,4‘ congeners (Erickson and Mondello, 1993). The region responsible for this altered activity is shown underlined in Fig. 3. The location of the mononuclear non-haem iron binding site in the Class I1 and 111dioxygenases is still unknown. However, the location of the iron binding site in the a subunits of these dioxygenases would be consistent with the finding that the oxygenase components of phthalate dioxygenase (Batie et al., 1987) and 4-chlorophenylacetate dioxygenase (Markus et al., 1986), with structures a l and a3, respectively, require Fe(I1) for oxygenase activity. The function of the p subunit of these Class I1 and 111dioxygenasesis unknown. The deduced amino-acid sequences of the p subunits of benzene, toluene, naphthalene and benzoate/toluate oxygenases show a low but significant degree of homology to each other with nine conserved amino acids present in each subunit. Five of the nine conserved residues are charged amino acids. It is possible that the charged amino acids play a role in the association between the subunits of the oxygenase component (Neidle et al., 1991). Recently, we have carried out covalent cross-linking experiments on the components of benzene dioxygenase using l-ethy1-3-(3-dimethyl aminopropyl carbodiimide). Results indicated that the /3 subunit may play a role in ferredoxin “docking” with the ISPBED(unpublished results).
6. COORDINATION OF THE IRON-SULPHUR CLUSTERS
All ferredoxin and oxygenase components of ring-hydroxylating dioxygenases contain an iron-sulphur cluster. These specific groups are responsible for the transfer of electrons through the redox chain by the reduction and oxidation of iron. Several different types of [2Fe-2S] clusters are known in proteins (Cammack,
62
CLlVE S.BUTLER AND JEREMY R. MASON
8 0 0 0 O X 0 0 Y o NNQTETTPIRVR~~TSEIWLFDEQAGRIDPRIYTDEDLYQLELER'JFARSWLLLOllETHI~~DYFTT~GEDP ~QTDTSPIRLRRSWNTSEIEALFDE~GRIDPRIYTDEDLYQLELERVFARSWLL~HETQIRKPGDYI~~EDP M S S A I K E V Q G A P V K W V m W T P E I \ I R G L V O Q E K G L L D P R I Y E D P MNYNNKlLVSESGLSQKHLIHGDEELFQHELKTIFA~LFL~DSLIPAffiD~A~IUE NNYNNKILVSESGLSQKHLIHGDEELFQHELKTIFA~LFL~DSLIFAFGD~A~IUE M P R I P V I N T S H L D R I D E L L V D N T E T G E F K L H R S V F T O Q R L Y I G R Q P MLTPEENLLLCRVEGDAPMGQM-MRRHWTPVCLLE-EVSERLFGEDL 0 I 0 0 II l o OY I OYYY Y 0 I 0 VWVRQKDASIAVFLNQCRHRGMRICRSDAGNA~~SYHGWAYDTAGNLI~PY~ESFACL-DKKE----~----WSPL VVWRQKDASIAVFLNPCRHRGMRIC~DAGNA~~SYH~AYDTAGNL~PY~ESFACL-NKKE---~-----WSPL Y V M V R Q K D K S I K V F L N P C R H R G M R I C R S D A G N A K A F T C S Y G P L VIVSRQN~SIKATLNVCRHRGKTLVSV~GNAKGFVCSYHG~~GSNGELQSVPPEKDLYGESLNKKCLG--------LKE VIVSRQNDGSIRAFLNVCKHRGKTLVSVEAGNAKGTVCSYHG~~~SNGELQSVPFEKDLYGESLNKKCLC,--------LKE ILIARNPNGELNAHINACSHRGAQL~HKRGNK'~~~PFHG~FNNSGKLLKVKDPSDAGYS~FNQ~SHD------LKK V-VFRDTDGRVGVHDEYCPHRRVSLIYGR"SG-LHCL-E-------------------MVSEPA
00 0 0 ool 0 0 0 xu0 o l I O I a 0 KARVETYKGLIFANWDENAIDLDTYZGEAKFYIDHMLDRTEAGTEVIPGIQKWVIPCNWKFMEQFCSDMYHAGTTAllLSGI
KARVETYKGLIFANWDENAVDLDTYLGEAKFYMDHMLDRTEAGTWLI~VQKWVIPCNWKFMEQFCSDMY~GTTSHLSGI
Q A R V A T Y K G L V F A N W D V Q A P D L E T Y L G D A R P Y M D V M L D VARVESFHGFIYGCFDQEAPPLMDYLC,DMWYLE~FKHS-GGLELVGPPGKWIKANWKAPAENFVGDAYsVG-~SL
VARVESFHGFIYGCFDQEAPPLHDYLC,DMWYLEPMFKHS-GGLELVGPPOKWIKANWKAPAENFVGDAYWG-WI1V\SSL VARFESYKGFLFGSLNPVDPSLQEFUjETTKIIDMIVGQSL)(XiLEVLRGVSTYTYEGNWKLTAEN-GADCS-A~YA ASNMCQKVKHTAYKTREWGOYVWAmGPQDAIPEFVPPAWAPHEHVRVSIAKAIIPCNWAQILE-GAIDSABSS-SWISSUF
I
0
0 0
0
IAGLPEDLELADLA--PP-KFGKQ----YRASWGGHGSGFYIGDP~I~GFK~SYLTEGPME~ERLGSIERGTKI LAGLPEULEMADLA--PP-TVGKQ----YRAS~HGSGFYVGDPN~LAIMGPK~S~EGPAS~~ER~SVERGSKL LAGIPP~DLSQAQ--IPTK-GNQ----FRAAWGGHGSGWWDEffiSLLVHGPKVTQYWTEGPAAELAHTGMPVRR
beZl todcl bpM ndoB dobeM pht3 bed1 tom bpM ndoB
doxB benA phtl bed21 tom bgM nd02 dobenA pht3
bedcl t d l
RSGESIFSSLAGNAALPPEGAGLQM~SKYGSGMGVLWDF-YSGVHSADLVPELMANGGAKQERLNKEIGDVRARIYRSHLNC
bPM ndoB
RSGESIFSSLAGNAALPPEGAGLQMTSKYGSGMGVLW~-YSGVHSADLVPELMAFGGAKQERLNKEIGDV~RIYRSHLNC ATMHRKEKQAGDT----------I~SAGS~KH~SYGFEHGHMLL~~NPEURPNFP~E~EKFG~SK~I VFARVGGAWLTSKNWLRPSTD~P~QVERTSYGFRYAALHHPIQNAATSEYVRSTVFVAPATALIFPNNLYNVAN~NVPID
be& pht3
Y
0
doxB
O l
ML-EHMTVF~CSFLP-GVNTIRTWHPXGPNEVEVWAFTWDADAPDDlKEEFRR~LRTFSAGGVFEQDDGENWVElQHIL b&l MV- EHMTVFPTCSPLP-GINTVRTWHPRGPNEVEVWAFTWUADAPDDI KEEFRROTLRTFSAGGVFEODIXENWVEIOH I L tom MVGQHMTIFPTCSFLP buM TVFPNNSML-TCS ---GVFKVWNPIDANTTE~WCYAI'JFKDM--PEDLKRilLAVS~QRT~CPACINIESDDNI~NMFTAS3" ndoB TVFFNNSML- I T S - - --GVFKVJHPLVANII'k~.W YAIVCKDM- - P t V l KRRIADSVQRTFGPAGIWPGDDNDNMETASQNG doxE CHSRNLCLYTNIYLMDOtU~QIHVLPPISVNKCLVTIYCIAPVLUIPtAKAHR1HUYCD1tNASGMATPUULtOLPltCQAGY be& DTHTAFYFMI\WGNPDNT~FLC,QQVGIDLDDSYEPLRN~NRFFQDR~~GN~IKGFPNQDI~TMG ph P t3 IA
TFMJIRIWHPKGPNEILVWAFTLVDADAPAEIKECYRRHNIRN~SAGCVFCQDUCENWV~IQKGL
0
0
RGHKARSRPFNAEM-----SMGQTVDNUPIYPGRIShMVYSEPAARGLYAHWLKMMTSPDW~LKATR RGHKARSRPFNAEM-----SMDQTVDNDPVYffiRISNNVYSEEAAHGLYAHWLRMTR
RGYKAKSQFWAQM-----GLGRSQTGHPDFffiNVG-YVYAEEARHGMYHHWMRMMSEFSWATLKP
KKYQSRDSDLLSNL-----GFGEDVYGDAVY~WGKSAIGETSYRGFY~YQAHVSSSNWAEFEHASS~HTELTKTTDR KKYQSRDSDLLSNL-----GFGEDVYGDAVYPGWGKSAIGETSYRGFYRAYQAHVSSSNWAEFE~SSTWHTELTK'ITDR AGIELEWNDHCRGSKHWIYG~DD~EIGLKPAISGIKTEDtiGLYLAQtlQYWLKSMKQAIAAEKEFASR~ENA DRSDERLGASDLAWEFRRVMLDALMFQAGESAIGTGE~IPSRICSFQAIVSKDIDWRDYQARYVWALDD1\NIVAEPUYEVHT
bedcl tom
2 : doxE benR
ph t3
Figure 3 Alignment of the predicted amino-acid sequence from the ISP a subunits of all known classes of dioxygenases. Sequence identities are as follows: bedC1, benzene dioxygenase; todCl ,toluene dioxygenase; bphA, biphenyl dioxygenase; nduB, naphthalene dioxygenase; doxB, dibenzothiophene dioxygenase; benA, benzoate dioxygenase; andpht3, phthalate dioxygenase. Symbols indicate amino acids conserved in six (0)or seven (#) sequences, respectively. The five conserved histidines and the two conserved cysteines are highlighted. The sequence underlined indicates amino acids shown to influence the substrate specificity of biphenyl dioxygenase.
1992). The largest class consists of the plant-type ferredoxin [2Fe-2S] clusters, of which the first known example was the spinach ferredoxin, found in chloroplast membranes (Gibson et al., 1966). Similar proteins have since been found in cyanobacteria (Fukuyama et al., 1980). These ferredoxins are small hydrophilic proteins containing one [2Fe-2S] cluster per molecule, serving as one-electron carriers (Bruschi and Guerlesquin, 1988). The [2Fe-2S] clusters contain two atoms
BACTERIALAROMATIC RING-HYDROXYWTING DIOXYGENASES
63
of iron, bridged by two labile sulphide atoms, and are coordinated to the protein by four sulphide ligands contributed by four cysteine residues. Upon reduction, one of the iron ions becomes high-spin Fe(II), and the net spin of the coupled Fe(II1) and Fe(I1) ions is S = 1/2. Thus, these [2Fe-2S] clusters give an EPR signal in the reduced state, when measured at low temperatures, which has an average g-factor of less than two (Gibson et al., 1966).The plant-type ferredoxins give an EPR signal with agavof approximately 1.96.This type of iron-sulphur cluster is also associated with the reductase components of the Class I and 111dioxygenases. The second class of proteins that contain [2Fe-2S] clusters is the Rieske proteins. These proteins are named after J. S. Rieske, who described a protein isolated from the cytochrome bcl complex of mitochondria (Rieske et al., 1964). Similar proteins have been shown 10 be associated with the b6f complex of the thylakoid membrane of chloroplasts (Riedel etal., 1991) and the plasma membrane of some aerobic and photosynthetic bacteria (Fee et al., 1984;Gabellini and Sebald, 1986). Examination of various physical properties has been used to distinguish between the plant-type and Rieske-type iron-sulphur clusters. The Rieske proteins have red-shifted visible spectra with maximum absorption about two-thirds of those of the plant-type ferredoxins and relatively more positive midpoint redox potentials. The EPR spectra of the reduced proteins have a lower gavvalue, around 1.91, with a characteristic sharp derivative peak at g = 2.02 and a broad trough around g = 1.80 (Trumpower, 1990). The terminal oxygenase component of many of the ring-hydroxylating dioxygenases have both visible and EPR spectra typical of the Rieske-type iron-sulphur clusters (Table 4); as a consequence, the results of the Rieske protein studies have been compared to those of the dioxygenases (Mason and Cammack, 1992). 6.1. Evidence that the dioxygenase ISP 12Fe-2SI clusters are of the Rieske type with imidazole ligands
Many spectroscopic techniques, including electron nuclear double resonance (ENDOR), electron spin-echo envelope modulation (ESEEM), Mossbauer, resonance Raman and X-ray absorption have been used to distinguish between plant-type and Rieske [2Fe-2S] clusters. The use of these spectroscopictechniques, amino-acid sequence analysis and the application of site-directed mutagenesis has provided evidence that two of the ligands to the dioxygenase and Rieske ironsulphur clusters are the nitrogens of histidine residues.
6.2 Spectroscopic evidence
Mossbauer spectroscopy, which depends upon the chemical state and environment of the at o m containing the 57FeMossbauer nuclei, yields characteristic spectra for
Table 4 Properties of the oxygenase component of aromatic dioxygenases.
Dioxygenase (reference) Phthalate (1 ) 4-Chlorophenylacetate(2) 4-Sulphobenzoate (3) 2-Halobenzoate (4) Benzoate (5.6)
b m n (7) Dibenzofuran (8) Benzene (9-1 1) Toluene (12) Napthalene (13.14) Terephthalate (15)
Absorption Component
Prosthetic group(s)
Oxygenase A
4[2Fe-2S] 3[2Fe-2SI
A A Oxygenase
2[2Fe-2S] 3[2Fe-2S] 3[2Fe-2S] 3Fe
A1 C
"BED ISPTQL ISPNAP
z
[2Fe-2S] 2[2Fe-2S] 2[2Fe-2S] 2L2Fe-B) 2[2Fe-2S] 2[2Fe-2SI
( h a d(nm)
EPR (gx, g, gz)
Redox potential (mV)
325,460,560' 325,458,564
1.73, 1.91.2.01 nd
nd nd
327,467,560 279,325,462,550" 325,464,565'
nd nd 1.77, 1.91,2.01, 1.83,4.2,7.5 1.79, 1.91,2.02 nd 1.75, 1.5q2.02 nd 1.80, 1.91,2.01 1.73, 1.91,2.01
nd nd nd
445,545' 450,565 326,450, 550a 326,450,550' 334,462, 5 6 8 420,460, 520"
nd nd -115 nd nd nd
'Shoulder. nd, not determined. References: 1, Batie et al., 1987; 2, Markus et al., 1986; 3, Locher et al., 1991; 4, Fetzner et al.. 1992; 5, Yamaguchi and Fujisawa, 1980; 6, Yamaguchi and Fujisawa 1982; 7, Sauber et al., 1977; 8, Bunz and Cook, 1993; 9. Zamanian and Mason, 1987; 10, Geary et al.. 1990; 1 I , Crutcher and Geary, 1 9 9 ; 12, Subramanian et al., 1979; 13, Ensley and Gibson, 1983; 14, Suen and Gibson, 1993; 15, Schlafli etal., 1994.
BACTERIAL AROMATIC RING-HYDROXYLATING DIOXYGENASES
65
the two types of iron-sulphur cluster, particularly when they are examined in both their oxidized and reduced states. 57Fewas incorporated into the ISPBED component of benzene dioxygenase by growing I! putida ML2 on a 57Fe-containingmedium. The Mossbauer spectra obtained from the ISPBEDpurified subsequently, indicated the presence of [2Fe-2S] clusters similar to those observed in the plant-type ferredoxins (Geary and Dickson, 1981); however, the main difference was the isomer shift of one of the iron sites, which was greater than in the plant-type ferredoxins, but similar to that observed in the Rieske-type protein of Thermus thennophilus (Fee et al., 1984). The Mossbauer parameters were compared with those of the monooxygenase ferredoxin putidaredoxin, which has a [2Fe-2S] cluster with all sulphur ligation (Miincketal., 1972).In the reduced state, the Fe(II1) site of the dioxygenase [2Fe-2S] cluster was similar to the corresponding iron in the ferredoxin; however, the Fe(II) sites were different, as shown by the electricfield gradient and hyperfine tensors. These differences may have two possible explanations: different distortions of the cluster geometry or different ligands. Miinck et al. (1972) considered it likely that the Fe(I1) site had non-sulphur ligands as this iron atom had a larger Mossbauer isomer shift than that in putidaredoxin. Electron-spin relaxation processes are characteristic of the iron-sulphur clusters.The electron-spin relaxation rate of the [2Fe-2S] clusters in the dioxygenases and the Rieske proteins is slower than in the plant-type ferredoxins, but faster than in hydroxylase ferredoxins, such as putidaredoxin and adrenodoxin. This observation was interpreted by Bertrand and co-workers (1987) as differences in the geometry of the [2Fe-2S] clusters. Using ENDOR and ESEEM spectroscopy, Cline et al. (1985) were able to observe hyperfine interactions in the [2Fe-2S] clusters in the phthalate dioxygenase of /? cepacia and in the respiratory Rieske iron-sulphur protein of T thennaphilus. All of the recorded spectra showed resonances attributed to protons that were weakly coupled to the iron-sulphur clusters and in addition, unusually strong-hyperfineresonances were observed, which were assigned to the I4N of two histidines. Further analysis of the phthalate dioxygenase of F? cepacia by Gurbiel et al. (1989) established the nature of the nitrogen-containing ligands to the [2Fe-2S] Rieske-type cluster. Four different samples of the phthalate dioxygenase terminal oxygenase component were isolated and purified from a histidine auxotroph strain of F? cepacia, grown previously on a specific isotopic labelling medium. The X-band ENDOR spectra obtained with the "N-histidine enriched protein at the g, resonance position, showed two sharp doublets (four lines) attributed to a pair of inequivalent "N nitrogen sites. However, as the spectrum at g, corresponds to a distribution of orientations, the signal from a single 15N nitrogen may give rise to more than one pair of peaks. To eliminate this possibility, Gurbiel et al. (1989) recorded another spectrum at g,. This second spectrum corresponded to a unique molecular orientation and again exhibited two doublets, demonstrating conclusively that the ' 'N ENDOR signals represent two magnetically distinct histidine
66
CLIVE S. BUTLER AND JEREMY R. MASON
4
Figure 4 Structure of the Rieske-type [2Fe-2S] cluster of the phthalate dioxygenase ISP as determined by ENDOR spectroscopy. The [2Fe-2S] core and the g,-g, plane lie in the paper. (Adapted from Gurbiel el al., 1989.)
ligands coordinated to the Rieske-type [2Fe-2S] cluster. Q-band (35 GHz) ENDOR experiments by Gurbiel et al. (1989) have demonstrated that the high-frequency ENDOR resonances observed with the T. thermophilus Rieske protein (Cline et al., 1985), assigned previous1 to a strongly coupled nitrogen, were in fact due to protons. Analysis of the 1 YN hyperfine coupling tensors and the 15N quadrupole coupling tensors have indicated a roughly tetrahedral coordination at Fe(II), with the N-Fe-N ligand plane corresponding to the g,-g, plane. These data led Gurbiel and co-workers to propose a model for the structure of the Rieske-type [2Fe-2S] cluster of phthalate dioxygenase, shown in Fig. 4. The model proposed by Gurbieletal. (1989) was in keeping with data previously obtained by Kuila et al. (1987), who determined the resonance Raman spectra of the Rieske proteins of I: thennophilus and phthalate dioxygenase of l? cepacia. Raman spectroscopy examines the vibrational modes of the metal centre. For the plant-type [2Fe-2S] clusters, the modes display a centrosymmetric pattern; however, for the dioxygenase cluster, the spectrum showed additional vibration modes, indicating a major perturbation from centrosymmetry. From these results it was concluded that both nitrogens were ligated to one iron atom (Kuila et al., 1987). X-ray absorption spectroscopy has also been applied to the terminal oxygenase component of phthalate dioxygenase of k? crpacia by Tsang et ul. (1989). In order to study the [2Fe-2S] cluster selectively, the mononuclear iron site was either depleted or reconstituted with cobalt or zinc. The results obtained showed that the iron environment in the Rieske cluster is structurally indistinguishable from that found in the plant-type [2Fe-2S] clusters, thus strongly supporting the suggestion
67
BACTERIAL AROMATIC RING-HYDROXYLATING DIOXYGENASES
that the unusually high reduction potentials for the Rieske clusters are due to electrostatic rather than structural effects. Based on the model by Gwbiel et al. (1989), Tsang et al. (1989) estimated the average Fe-Fe distance to be 2.68 for both the oxidized and reduced clusters, and the bridging and terminal Fe-S distance to be 2.20 8, and 2.31 A,respectively, for the oxidized cluster. It was not possible, however, to estimate the number or distances of any nitrogen ligands. More recently, Britt et al. (1991) and Shergill and Cammack (1994) have conducted ESEEM spectroscopy on the Rieske centres in spinach cytochrome b6f complex, bcl complex and bovine heart mitochondria1 membranes. The ESEEM spectra indicate a nitrogen coordination environment similar to that proposed by Gurbiel et al. (1989) for the phthalate dioxygenase, supporting the suggested coordination of two histidine ligands to the Rieske [2Fe-2S] cluster. Recently, we have employed the two-dimensional ESEEM technique of hyperfine sublevel correlation spectroscopy (HYSCORE) to correlate unambiguously pairs of ESEEM frequencies belonging to different nitrogen nuclei of ISPBED.The results confirm the assignment of parameters to the two histidine ligands but also provide evidence for the involvement of a weakly coupled peptide nitrogen (Shergill et al., 1995).
a
6.3. Comparison of the amino-acid sequences of [ZFe-2Sl Rieske proteins and the dioxygenase ISP a subunits The amino-acid sequence for many iron-sulphur proteins from various organisms has been determined either by direct amino acid sequencing or deduced from the IDBl I 0 1 IICTD 00011OOCOIIIIIIIOU OICIIOI
,
I
03
~ I I O O Ca I m I r o
0 1
I
I 1
aoia D o ~ a o r r o I r r g01 uouaoou
IIOIII UIO
DNA sequence of the respective genes. A comparison of the amino-acid sequences of the ISP a subunit from benzene dioxygenase (Tan et al., 1993), toluene dioxygenase (Zylstra and Gibson, 1989), benzoate dioxygenase (Neidle et al., 1991), naphthalene dioxygenase (Simon et al., 1993), dibenzothiophene dioxygenase* (Denome et al., 1993), biphenyl dioxygenase (Erickson and Mondello, 1992) and phthalate dioxygenase (Nomura et al., 1992) is shown in Fig. 3. The alignment reveals that four cysteines and two histidines are highly conserved across all three classes of dioxygenase and, in addition, a fifth histidine positioned near to the N terminus is conserved amongst six of the dioxygenases sequenced. The amino-acid sequence of other Rieske-type iron-sulphur proteins has been determined. The amino-acid sequence of the Rieske subunit of Rhodopseudoinonas sphaeroides (Gabellini and Sebald, 1986) and Rhodobacter capsulutus (Davidson and Daldal, 1987) have been deduced from their nucleotide sequences. The sequence shows that four cysteines and three histidines are highly conserved, *Dihenzothiophene dioxygenase genes (dax) found in Pseudomonas sp. encode proteins that are 2 99% similar in amino-acid sequence to naphthalene dioxygenase. As the deduced pathway metabolizes naphthalene as well as dihenaothiophene, this enzyme has not yet been classified as a separate dioxygenase.
68
CLlVE S. BUTLER AND JEREMY R. MASON
similarly, the amino-acid sequences of the cytochrome b6fcomplex Rieske protein (Pfefferkorn and Meyer, 1986) and the mitochondria1 C-111 of Saccharoinyces cerevisiae (Beckmann et al., 1987) also contain four conserved cysteine and three conserved histidine residues. The amino-acid composition of the I: thennophilus Rieske protein (Fee etal., 1984), which coordinates two identical [2Fe-2S] clusters per protein molecule, contains four cysteine residues, indicating that only two cysteine residues could coordinate each [2Fe-2S] cluster. When these Rieske protein sequences were compared with the dioxygenase sequences it appeared that two cysteine and two histidine residues were fully conserved giving rise to the following consensus sequence (Mason and Cammack, 1992): Cys-Xaa-His-XaalC17-Cys-Xaaz-His (where Xaa is any amino acid and the subscript represents the number of amino-acid residues). In the conserved regions of the sequence of six of the dioxygenases, each histidine residue is flanked by a single glycine residue either next to it or one residue away. Glycine residues occur in turns of protein secondary structure (Hanukoglu and Gutfinger, 1989); thus it has been suggested that the iron-sulphur cluster is situated in a cleft within the protein (Mason and Cammack, 1992).
6.4. Site-directed mutagenesis experiments on Rieske iron-sulphur proteins
The use of site-directed mutagenesis has helped considerably in determining the structure and function of proteins. To date no directed mutagenesis studies have been reported on the ISPcomponents of the ring-hydroxylatingdioxygenases. Such studies have been carried out on the mitochondrial Rieske protein of S. cerevisiue (Graham and Trumpower, 1991; Graham et al., 1992) and on the Rieske protein of the cytochrome bcl complex of R. cupsulatus (Davidson el al., 1992). Graham and Trumpower (1991) have prepared site-directed mutants of the S. cerevisiae Rieske protein, in which conserved cysteines at positions 159, 164, 178 and 180 were changed to seines, and conserved histidines at positions 161 and 181 were converted to arginines. The mutation of any one of these six fully conserved residues resulted in an inactive iron-sulphur protein, lacking the iron-sulphur cluster. EPR spectra of the Rieske protein mutated at either of these six positions, showed no detectable g, = 1.89 signal. Mutation of each of the six conserved residues was shown by immunological blotting not to prevent post-translational import of the Fe-S protein precursor into mitochondria, where it is processed into a mature protein. In addition, a histidine at position 184, which is conserved only in respiring organisms, was replaced by an arginine residue. This mutation resulted in the correct assembly of the iron-sulphur cluster and the yeast was able to grow on non-fermentable carbon sources. Therefore. it seems that mutation of each of
BACTERIAL AROMATIC RING-HYDROXYLATING DIOXYGENASES
69
the conserved amino-acid residues C159, H161, C164, C178, C180 and H181 affects the insertion or the stability of the Rieske [2Fe-2S] cluster in the protein. Examination of the cytochrome bcl complex purified from one of the site-directed mutants showed that the other eight subunits of the complex are present. The mutant Rieske protein lacking the iron-sulphur cluster was more easily lost from the bcl complex during purification compared to the wild-type complex. From these observations it has been suggested that the insertion of the iron-sulphur protein is the final step in the assembly of the bcl complex and that insertion of the iron-sulphur cluster plays a role in stabilizing the Rieske protein in the complex (Graham and Trumpower, 1991). Yeast cells expressing mutated Rieske proteins were only able to grow fermentatively, owing to the specific loss of function of the cytochrome bcl complex, which suggested that the four conserved cysteines and two histidines were specifically required for the function of the iron-sulphur protein. Consideration of the spectroscopic data (Gurbiel et al., 1989; Kuila et al., 1987; Britt et al., 1991) has led Graham and Trumpower (1991) to suggest that two of the four cysteines, and the two histidines 161 and 181 are the most likely ligands to the [2Fe-2S] cluster. On the basis of geometric and spatial requirements for liganding the iron-sulphur cluster, they proposed that the two coordinating cysteines are C164 and C178, and that C159 and C180 are essential, and may form a disulphide bond that maintains the correct environment for the [2Fe-2S] cluster to be stably inserted into the apoprotein (Fig. 5a). Davidson and co-workers (Davidson et ul., 1992) have also prepared sitedirected mutants of the bcl complex in an overproducing strain of R. capsulutus, in order to assign the amino-acid ligands of the Rieske [2Fe-2S] cluster. Comparison of the amino-acid sequences from numerous bcl complex Rieske subunits identified two highly conserved regions that Davidson and co-workers have assigned as box I and box IT (Fig. 6).Each box contains two conserved cysteines (box I C133 and C138; box I1 CI53 and Cl55) and a conserved histidine (box I H135; box I1 H156). In additiop, another histidine, H159, which is conserved in some of the sequences and reypdced by a glutamine in others, could not be excluded as a potential iron-sulphu; kigand. This amino-acid residue and those in the box regions were all targets fbr site-directed mutagenesis. All of the mutants, except ClSSS, H159A and &9S, lacked an EPR spectrum typical of the Rieske type [2Fe-2S] cluster, d analysis of the SDS-PAGE and Western blots showed that the chromatop re membranes of the mutants either lacked or had drastically decreased levels of the Rieske apoprotein, leading to the conclusion that substitution of any of the six highly conserved residues had led to the loss of the cluster and the degradation of the mutant subunits. These data favoured H135 and H156 as the two imidazole ligands to the iron-sulphur cluster. Of the four cysteines, C155S was the only mutant that contained a [2Fe-2S] cluster, as detected by EPR spectroscopy. This suggested that C155 was not a ligand to the cluster. This was supported by C 155 being located next to a likely nitrogen ligand, H156. Davidson
/
70
CLlVE S. BUTLER AND JEREMY R. MASON
G
COOH
Figure 5 (a) Model for the [2Fe-2S] cluster binding region of the S. cerevisiae Rieske iron-sulphur protein. The conserved cysteines 164 and 178 and histidines 161 and 181 are the proposed ligands to the iron-sulphur cluster. Cysteines 159 and 180 may form a disulphide bond to maintain the correct environment for the [2Fe-2S] cluster to be inserted stably into the apoprotein. NHz and COOH represent the amino and carboxyl ends of the protein. (Adapted from Graham and Tmmpower, 1991.) (b) Model for the [2Fe-2S] cluster binding region of the R. cappsulatus Rieske protein. The proposed ligands are circled, and S-S represents the putative disulphide bridge between the conserved, non-liganding residues. NH2 and COOH represents the amino and carboxyl ends of the protein. (Adapted from Davidson el al., 1992.)
71
BACTERIAL AROMATIC RING-HYDROXYLATING DIOXYGENASES BO . Rat. S.C. N.C.
Mzm. Tbm.
P.d.
B.j. R.V.
R.r. R.S.
R.c.
Sy. N. Ag. Sp.
Pea Tob. Ar.
EWILIGV EWVILIGV QWLIMUiI EWLVULGV EWLWIGV EWLWIGV
CTHLOC CTHLOC CTHLOC CTHLUC
PIAN PIAN PIGE PIGE I PLPN I PLPN
EMLVMIGV OWLWIGI EWLWYAS QWLVMVGV EWLVMWGV EWLVMLGV
CTHLOC CTHLOC CTHLOC CTHLOc C T H m CTHLOC
V I I I V V
CTHLOC
WW WPW WPW WPF WPF WPF WPW
NYGIN DYGIN DYGIN TFGIN TFAIN TYGIN TYGIN
AV A1 A1 AV AV AV AV
CTHLOC CTHLOC
CTHLOC CTHLGC CTHMC CTHMC
CTHLOC CTHLOC
V V V V
AGDFGGYY AGDFGGYY AGDFGGWF AGDYGGWF AGDFGGWF AGDFGGWF
CPCWW
CPCHOS CPCHOS CPCHOS CPCHQS CPCHOS CPCHOS CPCHOS CPCHOS CPCHOS CPCHOS CPCHOS
PIGCGAGDFGGWF P I A H EGNYDGFF PLGH CGGDWGGWF PLGQKAGDPKGDFDGWF PIGGVSGDF'GGWF PMGDKSGDFGGWF N N N N N N
N
A S E NKFM VAE NKFM TAE NKFM ?.A€ N K F I QAE N K F I AAE N K F I KAE NKFL
CPCHOS CPCHOS CPCHOS CPCHOS
1'1
HYIIASGRIRKGPAPLN HYUASGH 1HKGYAYLN HYDISGR I RKCPAPLN tiYDI S3H1 HKGYAPLN HYDISGRIRKGPAPFN HYDISGRIRKGPAPYN
I
HY.TSGR1 RRGPAPQN QY DSSGH I HCGYAPAN QYDASCRVRKGPAPTN HYDSAGRIRKGPAPLN HYDSAGRIRKGPAPEN HYDSAGRIRKGPAPRN QYNAEGKVVXGPAPLS QYDETGKWRCPAPLS QYDETCKWRGPAPLS VYNNWHWHGPAPLS QYNWXRWRGPAPLS QYNNQGRWRGPAPLS QYNAQGRWRGPAPLS
I CPCHOS CPCHOS
BOX I1 153
155 156
159
CYS-THR-HIS-LEU-GLY-CYS VPMGDKSGDFGGWF CYS-PRO-CYS-HIS-GLY-SER HIS
Figure 6 Alignment of the carboxyl termini of all available Rieske sequences showing conserved boxes 1 and 11. The sequence identities are as follows: Bo., bovine; Rat., rat; S.C., Saccharomyces cerevisiae; N.c., Neurospora crassa; Mzm., maize mitochondrion; Tbm., tobacco mitochondrion; Pad., Paracoccus denitrificcms; B.j.. Bradyrhizobium japonicum; R.v., Rhodopseudoinonas viridis; Rx., Rhodospirillum rubrum; R.s., Rhodobacter sphaeroides; R.c., Rhodobacter capsulutus. The lower group shows the available b6fRieske sequences. Sy.,Synechococcus sp. PCC 7009; N., Nosroc; Ag., Agmenellum quadriplicatwn; Sp., spinach; Pea, pea; Tob,, tobacco; Ar.,Arabudopsis. The putative [2Fe-2S] cluster binding region of the R. cupsulatus Rieske protein is shown in boxes I and 11. The position of the non-conserved His adjacent to box XI is indicated by an arrow. (Adapted from Davidson el al., 1992.)
and co-workers [ 1992) have assigned C153 as a ligand by assuming that both boxes I and I1 provide a cysteine and histidine residue as ligands. On this basis there was insufficient data to assign which of C133 and C138 was the potential cysteinyl ligand from box I. Comparing the amino-acid sequences of the Rieske proteins with those of the bacterial dioxygenases, Davidson's group have proposed that the four ligands are box I; C133, H135 and box IT; C153; H156. This assignment is in total agreement with the amino-acid sequence consensus put forward by Mason and Cammack (1992). This model differs slightly from that proposed by Graham and Trumpower (1991) who suggest that C138 (R. cupsulatus numbering) is the box I cysteinyl ligand. In addition to suggesting the ligands to the iron-sulphur cluster, Davidson el al.
72
CLlVE S.BUTLER AND JEREMY R. MASON
(1992) have proposed an internal disulphide bond formed between the two remaining cysteine residues, C138 and C155, as previously suggested for the Rieske protein of S. cerevisiae (Graham and Trumpower, 1991). As the two histidine ligands both coordinate to the Fe(I1) ion of the [2Fe-2S] cluster (Fee et al., 1984; Gurbiel et al., 1989; Britt et al., 1991), and that the Fe-Fe axis is in the plane of the membrane (Salemoetal.,1979),these data, together with the liganding roles attributed to the specific amino-acid residues suggest that a simple loop between boxes I and I1 cannot suffice. For C133, H135, C153 and H156 to act as ligands and lie in the membrane, the intervening region (amino-acid residues 136-155) between the two boxes, must twist back upon itself to bring the appropriate residues parallel to each other (Fig. 5b) (Davidson et al., 1992). The assignment of C159 (S. cerevisiae numbering) as a ligand to the [2Fe-2S] cluster by Davidson et al. (1992), led Graham and co-workers to assess their proposed structure. After more extensive mctagenesis, Graham et al. (1992) isolated a mutant in which threonine at position 160 was changed to alanine, eliminating an electronegative residue proximal to C 159. This change had almost no effect on cytochrome c reductase activity. In contrast, a mutant proximal to C164, in which proline 166 was converted to serine, reduced the cytochrome c reductase activity to less than 20% of the wild-type activity, while exchanging an adjacent hydrophobic residue to another (valine 165to alanine) had very little effect on cytochrome c reductase activity. Thus Graham et al. (1992) still favour their initial speculation that C164 is a ligand rather than C159, as proposed by Davidson et al. (1992).
7 . THE CATALYTIC NON-HAEM IRON CENTRE Protein systems that use the reaction of dioxygen with non-haem Fe(I1) are functionally quite diverse. These highly evolved biological Fe(I1) complexes are active sites for dioxygen chemistry, typically functioning by binding and activating dioxygen in preparation for its insertion into unactivated C-H or C-C bonds. There are three broad classes of 02-reactive non-haem iron-containing enzymes containing mononuclear, dinuclear and polynuclear non-haem iron centres. Probably the most extensively studied class is that of the dinuclear non-haem iron- containing enzymes. Examples from this group of enzymes are methane monooxygenase, haemerythrin and ribonucleotide reductase, of which perhaps methane monooxygenase has been the most comprehensively characterized (reviewed by Feig and Lippard, 1994). The ring-hydroxylating dioxygenases have been reported to contain a mononuclear iron centre (Mason and Cammack, 1992), although very little structural information about this centre has been reported. Consequently, any discussion of its possible binding site and mode of action can be only speculative. Oxygenases containing mononuclear iron centres have been characterized.
BACTERIAL AROMATIC RING-HYDROXYLATING DIOXYGENASES
73
Examples of such oxygenases are the extradiol catechol dioxygenases, 4-methoxybenzoate monooxygenase and isopenicillin N synthase (reviewed by Feig and Lippard, 1994). The ring cleavage dioxygenase, protocatechuate 3,4-dioxygenase from F! aeruginosa has been studied in detail, resulting in the structure of this molecule being determined by X-ray crystallography (Ohlendorf et ul., 1988; Ohlendorf et al., 1994). The coordination geometry of the non-haem iron centre has been described as trigonal bipyramidal with Tyr447 and His462 serving as axial ligands, and Tyr408 His460 and solvent serving as equatorial ligands (Ohlendorf et al., 1994). Resonance Raman investigations into the structures of 4-hydroxyphenylpyruvate dioxygenase (Bradley et al., 1986) and chlorocatechol dioxygenase (Broderick and O’Halloran, 1991) have revealed the presence of enhanced vibrations characteristic of tyrosinate coordination to the iron centre. The structure for isopenicillin N synthase proposed by Orville and co-workers (1992) also shows that the mononuclear iron centre is ligated by histidines but suggests that an aspartate may act as an additional ligand. The activation of dioxygen requires a centre capable of single-electron transfer. Although all of the ring-hydroxylating dioxygenases have a terminal oxygenase with Rieske-type [2Fe-2S] clusters, there is no evidence for the cluster serving as an oxygen binding site; sulphide is prone to oxidative damage (Mason and Cammack, 1992). For the non-haem iron to activate dioxygen for the dihydroxylation of the substrate, its binding site is presumed to be close to that of the substrate. As substrate specificity of the ring-hydroxylating dioxygenases has been shown to be controlled by the a subunit of the ISP component (Furukawa et al., 1993; Erickson and Mondello, 1993), it is thought that the binding site for the non-haem iron is also situated on this subunit. Analysis of the amino-acid sequences of the ISPa subunits of benzene, toluene, benzoateholuate and naphthalene dioxygenases by Neidle and co-workers (1991) identified two histidines and two tyrosines near the middle of the sequence. This led Neidle et al. to speculate that these conserved residues are ligands to the iron centre. However, a more thorough comparison of the amino-acid sequences of all classes of the ring-hydroxylating dioxygenases (Fig. 3) shows that the tyrosine residues are not conserved in all ring-hydroxylating dioxygenases. It is thus possible that the coordination of the mononuclear non-haem iron in these dioxygenases involves the highly conserved asparagine (N208; BDO numbering), aspartate (D219) and glutamate (E214) in addition to the histidines (H222 and H228). Such residues have been shown to be involved in the ligation of iron centres in the soybean lipoxygenase L-1 (Minor er al., 1993), isopenicillin N synthase (Orville et af., 1992) and methane monooxygenase (Rosenzweig et al., 1993). The possible involvement of histidines H222 and H228 in the active site of ISPBEDhas been supportedrecently by site-directed mutagenesis experiments in our laboratory in which all four conserved histidines (Fig. 4)were mutated. All four mutants resulted in a complete loss of activity, although only mutants H98C and H119C gave rise to a !oss of EPR signal, indicative of their role in coordination of the iron-suphur cluster (Fig. 7).
74
CLIVE S. BUTLER AND JEREMY R. MASON
g-factor 2.2 1
310
2.1
330
2.0
1.9
I
I
350
1.8 I
370
1.7
390
Magnetic field (mT)
Figure 7 EPR spectra of dithionite-reduced cell extracts of E. coli cells expressing wild-type and mutant ISPBEDa subunits. Spectra are from wild-type (a) and mutants (b) H222M, (c) H228C, (d) H98C and (e) H119C. Each spectrum has been resolved by the subtraction of the spectrum from the cell extract of E. culi cells not expressing an ISPBEDa subunit. Conditions of measurement: microwave power 2 mW, microwave frequency 9.37 GHz, modulation amplitude 0.5 mT, temperature 30 K, 5 scans.
The mononuclear non-haem iron centres should be observable using EPR spectroscopy though for some types of ligand field, the reaction may become too broad to be detected. The only EPR signals due to the high-spin Fe(II1) centre in the ring-hydroxylating dioxygenases that have been reported are by Altier et a/. (1993) for the benzoate 1,Zclioxygenase. Their results show that the high spin
BACTERIAL AROMATIC RING-HYDROXYLATING DIOXYGENASES
75
Fe(II1) has g-factors of 7.5,4.2 and 1.83. This result is similar to that reported for the iron site of chlorocatechol dioxygenase (g = 9.83 and 4.25) (Broderick and O’Halloran, 1991) and 4-methoxybenzoate monooxygenase with g-factors around 6.0 and 4.3 (Bill et al., 1981; Twilfer et al., 1981). In the case of oxidized, fully active 4-methoxybenzoate monooxygenase,the g = 6 signals shifted upon addition of substrates, whereas the g = 4.3 signals did not. This led the authors to suggest that a change occurred in the microenvironment in the presence of substrate. A recent magnetic circular dichroism study on the mononuclear ferrous active site of phthdate dioxygenase from k? cepacia showed that a change of ligation, a decrease in coordination froin six to five, occurred on substrate binding. This displacement of an iron ligand may prepare the ferrous centre for dioxygen activation (Gassner et al., 1993). In the EPR study on the benzene dioxygenase by Geary and co-workers (1984) the detection of the mononuclear non-haem iron was limited to a minor signal observed at g = 4.3, but no significant changes in this signal were observed on addition of benzene either alone or in combination with dioxygen. The mechanism of dioxygen activation by the ring-hydroxylating dioxygenase reaction remains to be elucidated. An important property of the dioxygen molecule is its propensity to accept electrons in pairs and single electron-transfer reactions are less frequently encountered. Therefore, in mononuclear systems, it is less certain how two-electron redox reactions are accommodated. Since an external reductant. such as NADH or a [2Fe-2S] cluster, is usually present, it is possible that the mononuclear iron centre, cycles through two one-electron transfer steps during each round of catalysis. A mechanism has been proposed by Twilfer ef al. (1985), who envisaged the activation of the dioxygen at the iron centre by successive electron transfers from the reductase, followed by the reaction of the peroxo complex with the substrate: 1e0 2 1eFe(II1) -+ Fe(I1) + Fe(II)*Oz+ Fe(III).02 + Fe(II).O; -+ Fe(III).Oi-
Twilfer et al. (1985) obtained a series of nitrosyl analogues of the oxo-derivatives using nitric oxide. Since NO is an odd-electron species, it can convert Fe(II), which is normally undetectable by EPR spectroscopy, to an EPR-detectable species. EPR signals were observed for high-spin (S = 3/2) Fe(II).NO (or Fe(III).NO-) species, with g-factors gl = 4, gll = 2. The g-factors of signals changed upon binding of substrates (Twilfer et al., 1985).
8. CONCLUDING REMARKS The bacterial aroinatic ring-hydroxylating dioxygenases represent a broad range of multi-component enzymes, which, despite their diverse microbial origins, possess many properties and mechanisms in common. This has enabled their
76
CLIVE S.BUTLER AND JEREMY R. MASON
classification based on structural properties. This classification has been supported by sequence comparisons based on nucleotide and primary amino-acid sequences. Comparison of the low homology sequences has also enabled the identification of highly conserved functional motifs, e.g. iron-sulphur ligation domains, whereas analysis of high homology sequences has allowed the characterization of amino acids involved in more specific functions, such as the control of substrate specificity. This combination of sequence analysis with structure-function studies will prove invaluable for future protein engineering of improved enzymes for both biocatalysis and environmental remediation.
We thank our colleagues Dr J. K. Shergill and Professor R. Cammack for their critical help and advice. Portions of this work were supported in the authors’ laboratory by awards from the Engineering and Physical Sciences Research Council (GFUJ00151), and a studentship from the Biotechnology and Biological Sciences Research Council (CSB).
REFERENCES Altier, D.J., Fox, B.G., Miinck, E. and Lipscomb, J.D. (1993) EPR and Mossbauer characterization of benzoate 1,2-dioxygenase from t? putida. J. Bioinorg. Chem. 52,300. An, D., Gibson, D.T. and Spain, J.C. (1994) Oxidative release of nibite from 2-nitro-toluene by a three-component enzyme system from Pseudomonas sp. strain JS42. J. Eacteriol. 176,7462-7467. Axcell, B.C. and Geary, PJ.(1975) Purification and some properties of a soluble benzeneoxidizing system from a strain of Pseudomonas. Eiochem. J. 146, 173-183. Batie, C.J. and Ballou, D.P. (1990) Phthalate dioxygenase. In: Methods in Enzymology, Vol. 188, Hydrocarbons andMethylotroplzy (M.E. Lidstrom, ed.), pp. 6 1-70. Academic Press, New York. Batie, C.J., LaHaie, E. and Ballou, D.P. (1987) Purification and characterization of phthalate oxygenase and phthalate oxygenase reductase from Pseudomonas cepaciu. J. Biol. Chem. 262, 1510-1518. Batie, C.J., Ballou, D.P. and Correll, C.J. (1992) Phthalate dioxygenase reductase and related flavin-iron-sulphur containing electron transferases. In Chemistry und Biochemistry of Flavoenzymes. (F. Miiller. ed.), pp. 544-554. CRC Press, Boca Raton. Ljungdahl, P.O., Lopez, J.L. and Trunipower, B.L. (1987) Isolation and Beckmann, J.D., characterization of the nuclear gene encoding the Rieske iron-sulfur protein (RIP1) from Succharomyces cerevisiae. J. Biol. Chem. 262,8901-8909. Bertrand, P., Guigliarelli, B., Gayda, J.-P., Beardwood, P. and Gibson, J. (1985) Aligand-field model to describe a new class of 2Fe-2s clusters in proteins and their synthetic analogues. Biochim. Biophys. Acta 831,261-266.
BACTERIAL AROMATIC RING-HYDROXYLATING DIOXYGENASES
77
Bertrand, P.,Cayda, J.-P., Fee, J.A., Kuila, D. and Cammack, R. (1987) Comparison of the spin-lattice relaxation properties of the two classes of [2Fe-2S] clusters in proteins. Biochim. Biophys. Acta 916,24-28. Bill, E., Bernhardt, F.-H. and Trautwein, A.-X. (1981) Mossbauer studies on the active Fe...[2Fe-2SI site of putidamonooxin, its electron transport and dioxygen activation mechanism. Eu,: J. Biochem. 121,3946, Bradley, F.C., Lindstedt, S., Lipscomb, J.D., Que, L., Roe, A.L. and Rundgren, M. (1986) 4-Hydroxyphenylpyruvate dioxygenase is an iron-tyrosinate protein. J. Biol. Chem. 261, 11693-11696, Brand, J.M., Cruden, D.L., Zylstra, G.J. and Gibson, D.T. (1992) Stereospecific hydroxylation of indan by Escherichia coli containing the cloned toluene dioxygenase genes from Pseudornonasputiria F1. Appl. Environ. Microbiol. 58,3407-3409. Britt, R.D., Sauer, K., Klein, M.P., Knaff, D.B., Krianciunas, A., Yu,C.A., Yu, L. and Malkin. R. (1991) Electron Spin EchoEnvelope Modulation Spectroscopy supports the suggested coordination of two histidine ligands to the Rieske FeS-centres of the cytochrome b6f complex of spinach and cytochrome bcl complexes of Rhodospirillum rubrum, Rhodobacter sphaeroides R-26 and bovine heart mitochondria. Biochemistry 30,1892-1901. Broderick, J.B. and O’Halloran, T.V. (1991) Overproduction, purification and characterization of chlorocatechol dioxygenase, a non-heme iron dioxygenase with broad substrate tolerance. Biochemistry 30,7349-7358. Bruschi, M. and Guerlesquin, F. (1988) Structure, function and evolution of bacterial ferredoxins. FEMS Micmbiol. Rev. 54, 155-176. Biinz, P.V. and Cook, A.M. (1993) Dibenzofuran 4.4a-dioxygenase from Sphingomonas sp. strain RW 1: Angular dioxygenation by a three-component enzyme system. J. Bacteriol. 175,6467-6475. Cammack, R. (1984) Midpoint potentials of iron-sulphur proteins-a survey. In: Charge and Field Efsects in Biosystems (M.J. Allen and P.N.R. Usherwood, eds), pp. 41-51. Abacus, Tunbridge Wells. Cammack, R. (1992) Iron-sulfur clusters in enzymes: themes and variations. Adv. Inorg. Chem. 38,281-322. Chen, V.J., Orville, A.M., Harpel, M.R., Frolik, C.A., Surerus, K.K.. Miinck, E. and Lipscomb, J.D. (1989) Spectroscopic studies of isopenicillin N synthase. Amononuclear non-heme Fe2+oxidase with metal coordination sites for small molecules and substrate. J. Biol. Chem. 264,21677-21681. Cline, J.F., Hoffman, B.M., Mims, W.B., LaHaie, E., Ballou, D.P. and Fee, J.A. (1985) Evidence for N coordination to Fe in the [2Fe-2S] clusters of Themus Rieske protein and phthalate dioxygenase from Pseudomonas. J. Biol. Chem. 260,3251-3254. Correll, C.C., Batie, C.J., Ballou, D.P. and Ludwig, M.L. (1992) Phthalate dioxygenase reductase: a modular structure for electron transfer from pyridine nucleotides to [2Fe-2S]. Science 258,1604-1610. Crutcher, S.E. and Geary, P.J. (1979) Properties of the iron-sulphur proteins of the benzene dioxygenase system from Pseudomonos putida. Biochem. J. 177,393-400. Dalton, H. (1986) Biotransformations for industry. SERC Bull. 3.8-9. Davidson, E. and Daldal, F. (1987) Primary structure of the bcl complex of Rhodopseudom n a s cupsulata: nucleotide sequence of the pet operon encoding the Rieske cytochrome b and cytochrome c1 apoproteins. J. Mol. Bid. 195, 13-24. Davidson, E., Ohnishi, T., Atta-Asafo-Adjei, E. and Daldal, F. (1992) Potential ligands to the [2Fe-2S] Rieske cluster of the cytochrome bcl complex of Rhodobacter capsulatus probed by site-directed mutagenesis. Biochemistry 31, 3342-335 1. Denome, S.A., Stanley, D.C., Olson, E.S. and Young, K.D. (1993) Metabolism of
78
CLIVE S. BUTLER AND JEREMY R. MASON
dibenzothiophene and naphthalene in Pseudomonus strains: complete DNA sequence of an upper naphthalene catabolic pathway. J. Bacteriol. 175, 6890-6901. Eggink, G., Engel, H., Vriend, G., Terpstra, P and Witholt, B. (1990) Rubredoxin reductase of Pseudomonas oleovorurzs. Structural relationships to other flavoprotein oxidoreductases based on one NAD and two FAD fingerprints. J. Mol. Biol. 212,135-142. Ensley. B.D. and Gibson, D.T. (1983) Naphthalene dioxygenase: purification and properties of a terminal oxygenase component. J. Bucteriol. 155,505-5 11. Ensley, B.D. and Haigler, B.E. (1990) Naphthalene dioxygenase from Pseudomonas NCIB 9816. In: Methods in Enzymology, Vol. 188, Hydrocarbons and Methylotrophy (M.E. Lidstrom, ed.), pp. 46-52. Academic Press, New York. Ensley, B.D., Gibson, D.T. and LaBorde, A.L. (1982) Oxidation of naphthalene by a multicomponent enzyme system from Pseudoinonas sp. strain NCIB 9816. J. Bacteriol. 149,948-954. Erickson, B.D. and Mondello, F.J. (1992) Nucleotide sequencing and transcriptional mapping of the genes encoding biphenyl dioxygenase, a multicomponent polychlorinatedbiphenyl-degrading enzyme in Pseudomonas strain LB400. J. Bucteriol. 174, 290329 12. Erickson, B.D. and Mondello, F.J. (1993) Enhanced biodegradation of polychlorinated biphenyls after site-directed mutagenesis of a biphenyl dioxygenase gene. Appl. Envimn. Microbiol. 59, 3858-3862. Fee, J.A., Findling, K.L., Yoshida, T., Hille, R., Tarr, G.E., Hearshen, D.O., Dunham. W.R., Day, E.P., Kent, T.A. and Miinck, E. (1984). Purification and characterization of the Rieske iron-sulfur protein from Thermus thermophilus: Evidence for a [2Fe-2S] cluster having non-cysteine ligands. J. Biol. Chem. 259, 124-133. Fee, J.A., Kuila, D., Mather, M.W. and Yoshida, T. (1986) Respiratory proteins from extremely thermophilic bacteria. Biochim. Biophys. Acta 853,153-1 85. Feig, A.L. and Lippard, S.J. (1994) Reactions of non-heme iron (11) centers with dioxygen in biology and chemistry. Chem. Rev. 94,759-805. Fetzner, S. and Lingens, F. (1994) Bacterial dehalogenases: biochemistry, genetics and biotechnological applications. Microbiol. Rev. 58,641-685. Fetzner, S., Muller, R. and Lingens, F. (1992) Purification and some properties of 2-halobenzoate 1,2-dioxygenase, a two-component enzyme system from Pseudomonus cepacia 2CBS. J. Bacteriol. 174,279-290. Fukuyama, K., Hase, T., Matsumoto, S., Tsukihara, T and Katsube, Y.(1980) Structure of S. platensis [2Fe-2S] ferredoxin and evolution of the chloroplast-type ferredoxins. Nature 286,522-524. Furukawa, K., Hirose. J., Suyama, A,, Zaiki, T. and Hayashida, S. (1993) Gene components responsible for discrete substrate specificity in the metabolism of biphenyl (bph operon) and toluene (tod operon) J. Bucteriol. 175,5224-5232. Gabellini, N. and Sebald, W. (1986) Nucleotide sequence and transcription of thefbc operon from Rhodopseidornonm sphaeroides. Evaluation of the deduced amino-acid sequences of the FeS protein, cytochrome b and cytochrome c1. EUKJ. Biochem. 154,569-579. Gassner, G.T., Ballou, D.P.,Landrum, G.A. and Whittaker, J.W. (1993) Magnetic circular dichroism studies on the mononuclear ferrous active site of phthalate dioxygenase from Pseudornonas cepacia show a change of ligation state on substrate binding. Biochemistry 32,48204825. Gassner, G.T., Ludwig, M.L, Gatti, D.L., Correll. C.C and Ballou, D.P. (1995) Flavoprotein structure and mechanism, 7. Structure and mechanism of the iron-sulfur flavoprotein phthalate dioxygenase reductase. FASEB J. 9, 1411-1418. Geary, P.J. and Dickson, D.P.E. (1981) Mossbauer spectroscopic studies of the terminal
BACTERIAL AROMATIC RING-HYDROXYLATINGDIOXYGENASES
79
dioxygenase protein of benzene dioxygenase from Pseudomonas pulida. Biochem. J. 195,199-203. Geary, P.J., Mason, J.R. and Joannou, C.L. (1990) Benzene dioxygenase from Pseudomonas putida ML2 (NCIB 12190) In: Methods in Enzymology, Vol 188, Hydrocarbons and Methylotrophy (M.E. Lidstrom, ed.),pp. 52-60. Academic Press, New York. Geary, PJ., Saboowalla, F.,Patil. D. and Cammack, R. (1984) An investigation of the iron-sulphur proteins of benzene dioxygenase from Pseudomonas putida by electronspin-resonance spectroscopy. Biochem. J. 217,667673. Gibson, D.T. (1988) Microbial metabolism of aromatic hydrocarbons and the carbon cycle In: Microbial Metabolism and the Carbon Cycle (S.R. Hagedrn, R.S. Hanson and D.A. Kunz, eds), pp. 33-58. Harwood Academic Publishers, New York. Gibson, J.F., Hall, D.O., Thornley, J.H.M. and Whatley, F.R. (1966) The iron complex in spinach ferredoxin. Proc. Nut1 Acad. Sci. 56,987-990. Gibson, D.T., Koch, J.R. and Kallio. R.E. (1968) Oxidative degradation of aromatic hydrocarbons by microorganisms. I. Enzymatic formation of catechol from benzene. Biochemistry 7,2653-2662. Gibson, D.T., Maseles, F.C. and Kallio, R.E. (1970) Incorporation ofoxygen-18 into benzene by Pseudomonas putida. Biochemistry 9,163 1-1 635. Gibson, D.T., Cruden, D.L., Haddock, J.D., Zylstra, G.J. and Brand, J.M. (1993) Oxidation of polychlorinated biphenyls by Pseudomonas sp. strain LB400 and Pseudomonas pseudoalcaligenes KF707. J. Bacteriol. 175,45614564. Gibson, D.T., Resnick, S.M., Lee, K., Brand, J.M., Torok, D.S., Wackett, L.P., Schocken, M.J. and Haigler, B.E. (1995) Desaturation, dioxygenation and monooxygenation reactions catalyzed by naphthalene dioxygenase from Pseudomonas sp strain 9816-4. J. Bacteriol. 177,2615-2621. Goyal, A.K. and Zylstra, G.J. (1996) Molecular cloning of novel genes for polycyclic aromatic hydrocarbon degradation from Comamonas testosteroni GZ39. Appl. Environ. Microbial. 62,230. Graham, L.A. and Trumpower, B.L. (1991) Mutational analysis of the mitochondrial Rieske iron-sulphur protein of Saccharomyces cerevisiae. J. Biol. Chem. 266,22485-22492. Graham, L.A.. Brandt, U., Sargent, J.S. and Trumpower, B.L. (1992) Mutational analysis of assembly and function of the iron-sulphur protein of the cytochrome bcl complex in Saccharoinyces cerevisiae. J. Bioenerg. Biomemb. 25, 245-251. Gurbiel, R.J., Batie, C.J., Sivaraja, M., True, A.E., Fee, J.A. Hoffman, B.M. and Ballou, D.P. (1989) Electron-nuclear resonance spectroscopy of 15N-enriched phthalate dioxygenase from Pseudomonas cepacia proves that two histidines are coordinated to the [2Fe-2S] Rieske-type clusters. Biochemistry 28,4861-4871. Haak, B., Fekner, S. and Lingens, E (1995) Cloning, nucleotide sequence and expression of the plasmid-encoded genes for the two-component 2-halobenzoate 1.2-dioxygenase from Pseudomonas cepacia 2CBS. J. Bacteriol. 177,667675. Haddock, J.D. and Gibson, D.T. (1995). Purification and characterization of the oxygenase component of biphenyl 2,3-dioxygenase from Pseudumonas sp. Strain LB400. J. Bact -eriol.177.5 834-5 839. Haddock, J.D., Nadim, L.M. and Gibson, D.T. (1993) Oxidation of biphenyl by a multicomponent enzyme system from Pseudomonas sp. strain LB400. J. Bacteriol. 175, 395400. Haigler, B.E. and Gibson, D.T. (19904 Purification and properties of NADH-ferredoxinNAp reductase, a component of naphthalene dioxygenase from Pseudomonas sp. strain NCIB 9816. J. Bacteriol. 172, 457464. Haigler, B.E. and Gibson, D.T. (1990b) Purification and properties of ferredoxinNAp, a
80
CLIVE S. BUTLER AND JEREMY R. MASON
component of naphthalene dioxygcnase from Pseudomonas sp. strain NCIB 9816. J. Bacteriol. 172,465468. Halliwell, B. and Gutteridge, J.M.C. (1984) Oxygen toxicity, oxygen radicals, transition metals and disease. Biochem. J. 21Y, 1-14. Hanukoglu, I. and Gutfinger, T. (1989) cDNA Sequence of adrenodoxin reductaseidentification of NADP-binding sites in oxidoreductases. Eus J. Biochem. 180,479-484. Harayama, S., Rekik, M. and Timmis, K.N. (1986) Genetic analysis of a relaxed substrate specificity aromatic ring dioxygenase, toluate 1,Zdioxygenase, encoded by TOLplasmid pWW0 of Pseudomonaspulida. Mol. Gen. Genet. 202,226-234. Harayama. S . Kok, M. and Neidle, E.L. (1992) Functional and evolutionary relationships among diverse oxygenases. Annu. Rev. Microbiol. 46,565601. Hayaishi, 0. (1962) Oxygenases. Academic Press, New York, London. Hudlicky, T., Mandel, M., Rouden, J., Lee, R.S., Bachman, B., Dudding, T., Yost, K.J. and Merola, J.S. (1994) Microbial oxidation of aromatics in enantiocontrolled synthesis. Part 1 . Expedient and general asymmetric synthesis of inositols and carbohydrates via an unusual oxidation of a polarized diene with potassium permanganate. J. Chem. SOC. Perkin. Trans. I 1553-1567. Hurtubise, Y., Barriault, D., Powlowski, J. and Sylvestre, M. (1995) Purification and characterization of the Comamonas testosteroni B-356 biphenyl dioxygenase components. J. Bacteriol. 177, 6610-6618. h e . S., Doi, S., Yonfuji, T., Takagi, M. and Yano, K. (1987) Nucleotide sequencing and characterization of the genes encoding benzene oxidation enzymes of Pseudomoncrr putida. J. Bacteriol. 169,5174-5179. Kikuchi, Y., Nagata. Y, Hinata, M., Kimbara, K., Fukuda, M., Yano, K. and Takagi, M. (1994) Identification of the bphA4 gene encoding ferredoxin reductase involved in biphenyl and polychlorinated biphenyl degradation in Pseuciomonas sp. strain KKS 102. J. Bacteriol. 176, 1689-1694. Koga, H., Yamaguchi, E., Matsunaga, K., Aramaki, H. and Horiuchi, T. (1989) Cloning and nucleotide sequences of NADH-putidaredoxin reductase gene ( c a d ) and putidaredoxin gene (carnB) involved in cytochrome P-450cam hydroxylase of Pseudomonas p u t i h . J. Biocheni. 106, 831-836. Kuila, D., Fee, J.A., Schoonover, J.R., Woodruff, W.H., Batie, C.J. and Ballou, D.P. (1987) Resonance Raman spectra of the [2Fe-2S] clusters of the Rieske protein from Thermus and phthalate dioxygenase. J. Am. Chem. Soc. 109:1559-1561. Kurkela, S.,Lehvaslaiho, H., Palva, E.T. and Teeri, T.H. (1988) Cloning, nucleotide sequence and characterization of genes encoding naphthalene dioxygenase of fseudomonasputida strain NCIB9816. Gene 73,355-362. Lee, L., Brand, J.M. and Gibson, D.T. (1995) Stereospecific sulfoxidation by toluene and naphthalene dioxygenases. Biochern. Eiophys. Res. Commun. 212,9-15. Ley, S.V., Sternfeld, F, and Taylor, S. (1987) Microbial oxidation in synthesis: a six step preparation of (&)-pinit01from benzene. Tetrahedron Lett. 28,225-226. Lucher, H.H., Leisinger, T.and Cook, A.M. (1991) 4-Sulphobenzoate 3.4-dioxygenase. Purification and properties of a desulphonative two-component enzyme system from Comamonas testosteroni T-2. Biochem. J . 274, 833-842. Markus, A., Krekel, D. and Lingens, F. (1 986) Purification and some properties of component A of the 4-chlorophenylacetate 3,4-dioxygenase from Pseudomonas species strain CBS. J. Biol. Chem. 261, 12883-12888. Mason, J.R. (1988) Oxygenase catalysed hydroxylation of aromatic compounds: simple chemistry by complex enzymes. In!. Ind. Biotechnol. 8. 19-24. Mason, J.R. and Cammack, R. (1992) The electron-transport proteins of hydroxylating bacterial dioxygenases. Annu. Rev. Microbiol. 46,277-305.
BACTERIAL AROMATIC RING-HYDROXYLATING DIOXYGENASES
81
Minor, W., Steczko, J., Bolin, J.T., Otwinowski, Z. and Axelrod, B. (1993) Crystallographic determination of the active site iron and its ligands in soybean lipoxygenase L-1. Biochemistry 32,6320-6323. Mondello, F.J. (1989) Cloning and expression in Escherichia coli of the Pseudomonas strain LB400 genes encoding polychlorinated biphenyl degradation. J. Bacteriol. 171, 17251732. Moodie, F.D.L., Woodland, M.P. and Mason, J.R. (1990) The reductase component of the chromosomally encoded benzoate dioxygenase from Pseudomonas putida C- 1 is immunologically homologous with a product of the plasmid encoded XylD gene (toluate dioxygenase) from Pseudomonas putida mt-2. FEMS Microbiol. Lett. 71, 163-1 68. Morrice, N., Geary. P., Cammack, R., Harris, A., Beg, F. and Aitken, A. (1988) Primary structure of protein B from Pseudomonas puiida, member of a new class of 2Fe-2S ferredoxins. FEBS Lett. 231,336-340. Miinck, E., Debrunner, P.G., Tsibris, J.C.M. and Gunsalus, I.C.(1972) Mossbauer parameters of putidaredoxin and its selenium analogue. Biochemistry 11, 855-863. Nakatsu, C.H. and Wyndham, R.C. (1993) Cloning and expression of the transposable chlorobenzoa1e-3.4-dioxygenasegenes of Akaligenes sp. strain BR60. Appl. Environ. Micmbiol. 59, 3625-3633. Nakatsu, C.H., Straus, N.A. and Wyndham, R.C. (1995) The nucleotide sequence of the Tn5271 chlorobenzoate (cbaAU) dioxygenase unites the Class IAoxygenases in a single lineage. Microbiology 141,485495, Neidle, E.L., Hartnett, C., Ornston, L.N., Bairoch, A,, Rekik, M. and Harayama, S. (1991) Nucleotide sequences of Acinetobacter calcoaceticus benABC genes for benzoate 1,2dioxygenase reveal evolutionary relationships among multicomponent oxygenases. J. Bacteriol. 173, 5385-5395. Nomura, Y., Nakagawa, M., Ogawa, N., Harashima, S. and Oshima, Y.(1992) Genes in PHT plasmid encoding the initial degradation pathway of phthalate in Pseudomonas putida. J. Ferment. Bioeng. 74,333-344. Ohlendorf, D.H., Lipscomb, J.D. and Weber, P.C. (1988) Structure and assembly of protocatechuate 3,4-dioxygenase. Noture 336.403405. Ohlendorf, D.H., Orville, A.M. and Lipscomb, J.D. (1994) Structure of protocatechuate 3,4-dioxygenase from Pseutlomonns oeruginosn at 2.14 A resolution. J. Mol. Biol. 244, 586-608. Orville, A.M., Chen, V.J., Kriauciunas, A,, Harpel, M.R., Fox, B.G., Miinck, E. and Lipscomb, J.D. (1992) Thiolate ligation of the active site Fe2+of isopenicih N synthase derives from substrate rather than endogenous cysteine: spectroscopic studies of site-specific cys > ser mutated enzymes. Biochemistry 31,46024612, Pang, C.-P., Chakravarti, B., Adlington, R.M., Ting, H.-H., White, R.L., Jayatilake, G.S., Baldwin, J.E. and Abraham, E.P. (1984) Purification of isopenicillin N synthase. Biochem. J. 222,789-795. Pember, S.O., Johnson, K.A., Villafranca,J.J. and Benkovic, S.J.(1989) Mechanistic studies on phenylalanine hydroxylase from Chromobacterium violaceum. Evidence for the formation of an enzyme-oxygen complex. Biochemistry 28,2124-2130. Pfefferkom, B. and Meyer, H.E. (1986) N-terminal amino-acid sequence of the Rieske iron-sulfur protein from the cytochrome bd'complex of spinach thylakoids. FEBS Lett. 206,233-237. Que, L. Jr, Widom, J. and Crawford, R.L. (1981) 3.4-dihydroxyphenylacetate 2,3dioxygenase: a manganese(I1) dioxygenase from Bacillus brevis. J. Uiol. Chem. 256, 10941-10944. Raag, R. and Poulos, T.L. (1989) Crystal structure of the carbon monoxide-substratecytochrome P-450CAM ternary complex. Biochemistry 28,7586-7592.
82
CLIVE S . BUTLER AND JEREMY R. MASON
Riedel, A., Rutherford, A.W., Hauska, G., Muller, A. and Nitschke, W. (1991) Chloroplast Rieske protein-EPR study on its spectral characteristics, relaxation and orientation properties. J. Biol. Chem. 266, 17838-17844. Rieske, J.S., Zaugg, W.S. and Hansen, R.E. (1964) Properties of a new oxidation-reduction component of the respiratory chain as studied by electron paramagnetic resonance spectroscopy. J. Biol. Chein 239,3017-3022. Romanov, V. and Hausinger, R.P. (1994) Pseudomonas aeruginosa 142 uses a threecomponent ortho-halobenzoate 1,2-dioxygenase for the metabolism of 2.4-dichloro- and 2-chlorobenzoate. J. Bacleriol. 176, 3368-3374. Rosche, B., Tshisuaka, B., Fetzner, S. and Lingens, F. (1995) 2-0~0-1,2-dihydroquinoline 8-monooxygenase, a two-component enzyme system from Pseudomonas putida 86. J. Biol. Chem. 270,17836-17842. Rosenzweig, A.C., Frederick, C.A., Lippard, S.J. and Nordlund, P. (1993) Crystal structure of a bacterial non-haem iron hydroxylase that catalyses the biological oxidation of methane. Ncrture 366,537-543. Salemo, J.C., Blum, H. and Ohnishi, T. (1979) The orientation of iron-sulphur clusters and a spin-coupled ubiquinone pair in the mitochondria1 membrane. Biochim. Biophys. Acta 547,270-281. Sauber, K., Frohner, C., Rosenberg, G, Eberspocher, I . and Lingens, F. (1977) Purification and properties of pyrazon dioxygenase from pyrazon-degrading bacteria. E m J. Biochem. 74,89-97. Schlafli, H.R., Baker, D.P., Leisinger, T. and Cook, A.M. (1995) Stereospecificity of hydride removal from NADH by reductases of multi-component non-heme iron oxygenase systems. J. Bacteriol. 177, 831-834. Schlafli, H.R., Weiss, M.A., Leisinger, T. and Cook, A.M. (1994) Terephthalate 1.2dioxygenase system from Comamonas teslosteroni T-2: purification and some properties of the oxygenase component. J. Bncteriol. 176.66444652. Schweizer, D., Markus, A., Seez,M., Ruf, H.H. and Lingens, F. (1987) Purificationand some properties of component B of the 4-chlorophenylacetate 3.4-dioxygenase from Pseudornonns species strain CBS3. J. Biol. Chem. 262,9340-9346. Sheldrake, G.N. ( I 992) Biologically derived arene cis-dihydrodiols as synthetic building blocks. In: Chircrlify in Industry (A.N. Collins, G.N. Sheldrake and J. Crosby, eds), pp. 127-1 66. Wiley, Chichester. Shergill, J.K. and Cammack, R. (1994) ESEEM and ENDOR studies of the Rieske iron-sulphur protein in bovine heart mitochondria1 membranes. Biochim. Biophys. Acta 1185,3542. Shergill, J.K., Joannou, C.L.. Bratt, P.J., Mason, J.R. arid Cammack, R. (1995) Coordination of the Rieske-type [2Fe-2S] cluster of the terminal iron-sulphur protein of Pseudomoms pufida benzene 1,2-dioxygenase, studied by one- and two-dimensional electron spinecho envelope modulation spectroscopy. Biochemistry 34,16533-16542. Simon, M.J., Osslund,T.D., Saunders, R., Ensley, B.D., Suggs, S.,Harcourt, A., Suen, W.-C., Cmden, D.L., Gibson, D.T. and Zylstra, G.J. (1993) Sequences of the genes encoding naphthalene dioxygenase in Pseudomonas putida strains G7 and NCIB 9816-4. Gene 127,31-37. Spain, J.C., Zylstra, G.J., Blake, C.K. and Gibson, D.T. (1989) Monohydroxylation of phenol and 2,5-dichlol-ophenol by toluene dioxygenase in Pseudornonas putida F1. Appf. Environ. Microbiol. 55,2648-2652. Subramanian, V., Te-Ning, L., Yeh, W.K. and Gibson, D.T. (1979) Toluene dioxygenase: purification of an iron-sulfur protein by affinity chromatography. Biochem. Biophys. Res. Commnn. 91, 1131-1139. Subramanian, V., Liu,T.-N., Yeh, W.-K., Narro, M. and Gibson, D.T. (1981) Purification and
BACTERIAL AROMATIC RING-HYDROXYLATING DIOXYGENASES
83
properties of NADH-ferredoxinToLreductase. A component of toluene dioxygenase from Pseudomonas purida. J. Biol. &hem. 256,2723-2730. Subramanian, V., Liu,T.-N., Yeh, W.-K., Serdar,C.M., Wackett, L.P. andGibson, D.T. (1985) Purification and properties of ferredoxinmL. A component of toluene dioxygenase from Pseudoinonas putida F1. J. Biol. Chein. 260,2355-2363. Suemori, A., Kurane, R. and Tomizuka, N. (1993) Purification and properties of phthalate oxygenase from Rhodococcus eryrhropolis S-1 . Biosci. Biotechnol. Biochem. 57, 1482-1 486. Suen, W.-C. and Gibson, D.T. (1993) Isolation and preliminary characterization of the subunits of the terminal component of naphthalene dioxygenase from Pseudomonas putida NCIB 9816-4. J. Bacferiol. 175,5877-5881. Suen, W.-C. and Gibson, D.T. (1994) Recombinant Escherichia coli strains synthesize active forms of naphthalene dioxygenase and its individual a and p subunits. Gene 143,67-71, Taira, K., Hirose, J., iiayashida, S. and Furukawa, K. (1992) Analysis of bph operon from the polychlorinated biphenyl-degrading strain of Pseudomonas pseudoalcaligenes KF707. J. Biol. Chem. 267,48444853. Takizawa, N., Kaida, N., Torigoe, S., Moritani, T., Sawada, T., Satoh, S . and Kiyohara, H. (1994) Identification and characterization of genes encoding polycyclic aromatic hydrocarbon dioxygenase and polycyclic aromatic hydrocarbon dihydrodiol dehydrogenase in Pseudotnonus putida OUS82. J. Bacteriol. 176,2444-2449. Tan, H.-M. and Mason, J.R. (1990) Cloning and expression of the plasmid-encoded benzene dioxygenase genes from Pseuhnonasputida ML2. FEMS Microbiol. Lett. 72,259-264. Tan, H.-M. (1991) The benzene dioxygenase genes from Pseudomonasputida ML2: cloning, expression and identification of functional domains. Ph.D. thesis, University of London. Tan,H.-M., Tang, H.-Y., Joannou, C.L., Abdel-Wahab, N.H. and Mason, J.R. (1993) The Pseudomonas putida ML2 plasmid-encoded genes for benzene dioxygenase are unusual in codon usage and low in G + C content. Gene 130,33-39. Tan, H.-M. and Cheong, C.-M. (1994) Substitution of the ISP a subunit of biphenyl dioxygenase from Pseudoinonas results in a modification of the enzyme activity. Biochem. Riophys. Res. Coininun. 204,912-917. Tan H.-M., Joannou, C.L., Cooper, C.E., Butler, C.S., Cammack, R. and Mason, J.R. (1994) The effect of fenedoxinBED over-expression on benzene dioxygenase activity in Pseudornonns puiida ML2. J. Bacieriol. 176,2507-25 12. Trumpower, B.L. (1990) Cytochrome bcl complexes of microorganisms. Microbiol. Rev. 54,101-129. Tsang, H.-Y., Batie, C.J., Ballou, D.P. and Penner-Hahn, J.E. (1989) X-ray absorption spectroscopy of the [2Fe-2S] Rieske clusters in Pseudoinonas cepaciu phthalate dioxygenase. Determination of core dimensions and iron ligation. Biochemistry 28, 7233-7240. Twilfer, H., Bernhardt. F.-H. and Gersonde, K. (1981) An electron-spin-resonance study on the redox-active centers of the 4-methoxybenzoate monooxygenase from Pseudomonas putida. Eu,: J. Biochem. 119,595-602. Twilfer, H., Bernhardt, F.H. and Gersonde, K. (1985) Dioxygen-activating iron centre in putidamonooxin. Electron spin resonance investigation of the nitrosylated putidamonooxin. Eu,: J. Biochern. 147, 171-176. Wackett, L.P. (1990) Tvluene dioxygenase from Psedomonas putida FI. In: Methods in Enzymology, V01188, Hydrocarbons nndMethylotrophy (M.E. Lidstrom. ed.), pp. 3 9 4 6 . Academic Press, New York. Wackett, L.P., Kwart, L.D. and Gibson, D.T. (1988) Benzylic monooxygenation catalysed by toluene dioxygenase from Pseudornonas putida. Biochemistry 27. 1360-1 367. Wang, Y., Gamon, J., Labbe, D., Bergeron, H. and Lau, P.C.K. (1995) Sequence and
84
CLIVE S. BUTLER AND JEREMY R. MASON
expression of the bpdClC2BADE genes involved in the initial steps of biphenylkhlorobiphenyl degradation by Rhodococcus sp M5. Gene 164,117-122. Wende, I?, Pfleger, K. and Bernhardt, EH. (1982) Dioxygen activation by putidamonooxin: substrate-modulated reaction of activated dioxygen. Biochem. Biophys. Res. Commun. 104,527-532. Wierenga, R.K., Terpstra, P. and Hol, W.G.J. (1986) Prediction of the occurrence of the ADP-binding pap-fold in proteins, using an amino-acid sequence fingerprint. J. Mol. Biol. 187. 101-107. Yamaguchi, M. and Fujisawa, H. (1978) Characterization of NADH-cytochrome creductase, a component in benzoate 1,2-dioxygenase system from Pseudomonas arvilla C-1 .J. Biol. Chem. 253,8848-8853. Yamaguchi, M. and Fujisawa, H. (1980) Purification and characterization of an oxygenase component in benzoate 1,2-dioxygenase system from Pseudomonas arvilla C-1, J. Biol. Chem. 255,5058-5063. Yamaguchi, M . and Fujisawa, H. (1982) Subunit structure of oxygenase component in benzoate-l,2-dioxygenasesystem from Pseudomonns arvilln C-1. J. Biol. Chem. 257, 12497-12502. Yang, Y., Chen, R.F. and Shiaris, M.P. (1994) Metabolism of naphthalene, fluorene and phenanthrene: preliminary characterization of a cloned gene cluster from Pseudoinonas pu/ida NCIB 9816. J. Bacteriol. 176,2158-2164. Zamanian, M. and Mason, J.R. (1987) Benzene dioxygenase in Pseudomonas putida. Subunit composition and immuno-cross-reactivity with other aromatic dioxygenases. Biochem. J. 244,611616. Ziffer, H., Kabuto, K., Gibson, D.T., Kobal, V.M. and Jerina, D.M. (1977) The absolute stereochemistry of several cis-dihydrodiols microbially produced from substituted benzenes. Tetrahedron 33,2491-2496. Zylstra, G.J. and Gibson, D.T. (1989) Toluene degradation by Pseudomonas putida F1, Nucleotide sequence of todCl C2BADE genes and their expression in Escherichia coli. J. Biol. Chern. 264, 14940-14946. Zylstra, G.J., McCombie, W.R., Gibson, D.T. and Finctte, B.A. (1988) Toluene degradation by Pseudomonas putida F1: Genetic organization of the lod operon. Appl. Environ. Microbiol. 54, 1498-1503.
Thiol Template Peptide Synthesis Systems in Bacteria and Fungi Rainer Zocher and Ullrich Keller lnstitut fur Biocheinie und Molekulare Biologie. Technische Universitiit Berlin. Franklinstrafle 29. 0-10587Berlin.Charlottenburg. Germany
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 2. The peptide synthetase domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 2.1. Peptide synthetases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 2.2. Motifs of the carboxyl-adenylate-formingdomain . . . . . . . . . . . . . . . . 90 2.3. Modules in the activation domain . . . . . . . . . . . . . . . . . . . . . . . . 90 2.4. The N-methylation module . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 2.5. Acyltransfer and epimerization modules . . . . . . . . . . . . . . . . . . . . . 91 2.6. Thioesterase modules in peptide synthetase genes . . . . . . . . . . . . . . . 92 2.7. Properties of amino-acid activating domains . . . . . . . . . . . . . . . . . . . 93 3 . Enzymesystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 3.1. Organization of activation domains in prokaryotes and eukaryotes . . . . . . . 94 4. Peptide synthetases from fungi . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 96 4.1 . 6(La-Aminoadipyl)-cysteinyl-o-valine . . . . . . . . . . . . . . . . . . . . . . 4.2. Enniatins and beauvericin . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 4.3. Cyclosporin and related peptides . . . . . . . . . . . . . . . . . . . . . . . 105 4.4. Ergot peptide alkaloids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 5 . Prokaryotic peptide synthetase systems . . . . . . . . . . . . . . . . . . . . . . 111 111 5.1, Acyi peptide lactones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Surfactin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 5.3. Bialaphos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 6. Future prospects of peptide synthetase research . . . . . . . . . . . . . . . . . . 122 6.1. Domain exchange in thiol template peptide synthesis systems . . . . . . . . 123 6.2. Combinatorial approaches in future peptide synthesis development . . . . . 123 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 124 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abbreviations: Abu. or-aminobutyric acid; ACMS. actinomycin synthetase; ACP. acyl carrier protein; ACV. &(L-a-aminoadipyl)-cysteinyl-D-valine; ACVS. 6-(~-aaminoadipy1)-cysteinyl-D-valinesynthetase; AdoHCy. S-adenosyl-L-homocysteine; ADVANCES IN MICROBIAL PHYSIOLOGY VOL 38 ISBN 0-12-027738-7
CopyrightQ 1997 Academic Press Limited All rights of reproduction in any form reserved
86
RAINER ZOCHER AND ULLRICH KELLER
AdoMet, S-adenosyl-L-methionine; Bmt, (4R)-4-[(E)-2-butenyl]-4-methyl-~threonine; Cysyn, cyclosporin synthetase; D-Hiv, D-2-hydroxyisovaleric acid; DMPT, desmethylphosphinothcin; Esyn, enniatin synthetase; FAS, fatty acid synthase; GSI, gramicidin S synthetase I; GSII, gramicidin S synthetase 11; MeLeu, methyl leucine; 4-MHA, 4-methyl-3-hydroxyanthranilicacid; Orf, open reading frame; PKSI, polyketide synthetase type I; PKSII, polyketide synthetase type 11; Pt, phosphinothricin; Ptt, phosphinothricin tripeptide; Sar, sarcosine; TE, thioesterase; TYI, tyrocidine synthetase I.
1. INTRODUCTION
Bacteria and fungi produce numerous peptides as secondary metabolites, which are valuable as antibiotics, cytostatics. immunosuppressants, enzyme inhibitors and effectors acting on various cellular targets. It was speculated rather early that the mechanism of synthesis must differ from that of protein synthesis, since the vast majority of microbial peptides often contain unusual amino acids that are not found in proteins. Many of the non-proteinogenic amino acids have unique structures with respect to their carbon skeletons, presence of double bonds, unusual functional groups, aromatic character, D-configuration at a-C or presence of methyl groups at the amino groups involved in peptide-bond formation (Kurahashi, 1974; Kleinkauf and von Dohren, 1987, 1990). Also, the fact that such peptides in the producing organisms are very often formed as homologous series instead of single compounds indicates the broad specificity of enzymes responsible for incorporation of the relevant amino acids into peptides, which is in contrast to the specificity observed in protein synthesis. Nevertheless, the normal proteinogenic amino acids are present in most of the peptides of microbial origin, which indicates that the process of microbial peptide synthesis may have evolved in parallel to that of protein synthesis. The discovery of the activation with ATP of both acetate and amino acids led Lipmann, in the early 1950s, to postulate arelationship of peptide/protein synthesis with fatty acid synthesis, which have in common the mechanism of activation of building blocks and the direction of chain growth (head growth). In this model, peptide synthesis takes place on a template characterized by various condensation domains (Lipmann, 1954). The deciphering of the genetic code and later the elucidation of the ribosomal system eventually made it clear that protein synthesis mechanistically is very different from this model, especially in the nature of the template. However, the discovery that the cyclodecapeptide antibiotic gramicidin S is synthesized by enzymes instead of ribosomes independent of RNA (Berg et a]., 1965) confirmed the idea of a protein template directing the incorporation of amino acids into polymers (Lipmann, 1971, 1973). The first such templates with analogy to fatty acid synthase were found to be gramicidin S synthetase and later
87
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
GS I I
L-Om
L-Pro
L-Val
L-Pro
Activation
ESH+ aa + 2)
E.S.H.~~-AMP
MZ+
ATP -T-C
E.s .H.aa-~~~ 4 PPi
E
~ +- AMP ~
~
Figure 1 Structure of gramicidin S and its assembly by the protein thiol templates gramicidin synthetase I and I1 (GSI and GSII). Each of the five domains of the multifunctional enzyme system activates its corresponding amino acid as adenylate, subsequently binds it to a specific thiol in thioester linkage as shown in lower part of the figure. GSI epimerizes thioester bound Phe to D-Phe prior to transfer of this residue to GSII where condensation of the amino acids takes place. Two pentapeptide chains are finally dimerized in head-to-tail condensation.
tyrocidine synthetase, which polymerize amino acids in a stepwise fashion into covalently bound peptidyl intermediates (Gevers et al., 1969; Roskoski et al., 1970). Gramicidin S synthetase I (GS1)-at that time with an estimated molecular mass of 100 kDa-activates the starter amino acid L-phenylalanine as adenylate and binds it in thioester linkage. Similarly, the 280 kDa gramicidin S synthetase I1 (GSII) recruits the residual four amino acids of gramicidin S as thioesters (Fig. 1). After epimerization of phenylalanine, the various amino acids are polymerized into a pentapeptide chain of the sequential order as shown in Fig. 1. Head-to-tail condensation of two of these chains yields gramicidin S. The same mechanism of formation was also shown with tyrocidine, which is structurally similar to gramicidin S (Fig. l), but has more different amino-acid positions and therefore requires more enzymes for its assembly (Lipmann, 1973). The presence of 4'-phospho- pantetheine in these enzymes strongly supported the thiol template model in which the enzyme is composed of activating domains rather than of subunits, with each responsible for activation of one single amino acid. Each domain has an Mr equivalent of more than 70 000, which reflects that GSII is able to activate four amino acids in a defined sequence (Lipmann, 1973). An essential
88
RAINER ZOCHER AND ULLRICH KELLER
feature in every domain is the presence of a specific (peripheral) thiol group to which the corresponding amino acid becomes attached in thioester linkage. A central sulfhydryl was proposed to reside on the 4’-phosphopantetheine group, which, like the swinging arm of acyl carrier protein (ACP) in fatty acid synthesis, was to carry the growing peptide chain from one reaction centre to the next (Lynen, 1972). Thus, the analogy to fatty acid synthase (FAS), which acts upon acyl-coenzyme A (CoA) thioesters with subsequent covalent substrate binding is apparent except that the peptide synthetases consist of an array of similar domains each directing the incorporation of its cognate amino acid into the growing peptide chain. In contrast, in fatty acid synthesis, it is always the same extender unit (ie. malonyl coenzyme A) that contributes to fatty acyl chain growth. Accordingly, FAS has only two thiol groups, one on the P-ketoacyl synthase and the other on the ACP (reviewed by Hopwood and Sherman, 1990). Structural and functional studies, which became possible through the cloning and sequencing of genes encoding peptide synthetases, have led to a refinement of the thiol template mechanism of peptide formation (Schlumbohm et al., 1991; Marahiel, 1992; Stachelhaus and Marahiel, 1995a,b). Here we review the current state of research in this field with respect to the basic structural and functional molecular features of the building blocks of peptide synthesis systems. These molecular concepts will be illustrated in the course of the text by the description of selected examples of bacterial and fungal peptide synthesis systems. Finally, aspects of future developments of peptide synthesis systems such as the creation of recombinant enzymes through directed or combinatorial approaches will be discussed.
2. THE PEPTIDE SYNTHETASE DOMAIN
2.1. Peptide Synthetases The first peptide synthetase genes to be cloned and sequenced were those of GSI (Kratzschmar el al., 1989), tyrocidine synthetase I (TYI) (Weckermann et al., 1988), GSII (Krause et al., 1985; Turgay etal., 1992), and &(L-a-aminoadipyl)-Lcysteinyl-D-valine synthetase (ACVS) (Diez et al., 1990; Smith et al., 1990a,b; Gutierrez etal., 1991; MacCabe etal., 1991). The sequences ofthemulti-functional proteins activating more than one amino acid consist of repeating units, each encompassing a length of approximately 1000 amino acids corresponding to a molecular mass equivalent of 120 kDa. Each of these units show considerable sequence conservation to each other and the number of these units equals the number of the amino acids serving as substrates of the corresponding enzymes. Accordingly, single amino-acid activating enzymes such as TYI or GSI each consist
89
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
of only one unit. These findings lend compelling evidence for the view that each unit is responsible for the assembly of a single amino acid in the process of peptide formation. The sequence conservation in the units is particularly high in a 600 amino-acid-long central part (designated amino-acid activation domain) and is less, but still significantly, conserved in the interdomain sequences (spacers), which comprise some further 500 amino acids. Interpretation of the gene sequence data also revealed that the peptide synthetase proteins were much larger in size than had been estimated from previous mass determinations when no suitable MI markers were available. Thus, GSI and GSII were found to have molecular masses of 121 kDa and 510 kDa, respectively (Kratzschmar et al., 1989; Turgay et al., 1992).Inspection of the highly conserved region not only in the peptide synthetases mentioned above but also in a steadily increasing number of genes that have been cloned in the last 5 years has led to the identification of characteristic amino-acid sequence motifs (Marahiel, 1992; Stachelhaus and Marahiel, 1995a).Figure 2 shows that these motifs are spread over the entire length of the amino-acid activation domain and apparently play a role in functioning of the synthetase. Their sequence conservation is high enough to allow the construction of highly specific oligonucleotide primers for screening of new
peptide synthetase activation domain
I
I methyltransferase module -450 aa
thioester adenylylcarboxylate domain
racemuatiod transfer epimerization acy’
module
module
7 I I l l r n -l ---.LIT= C D E
- r - - v r - -I 200
400
I
u
I
600
I
I 800
I
I
1000
I
I
1200 aa
Figure 2 Assembly of the most highly conserved sequence motifs in peptide synthetase domains. Motifs A-E are highly conserved in adenylate forming enzymes. A(LKAGGAYVPID), B(YSGTTGXPKGV), C(GELCJGGXGXARGYL), D(YXTGD), E(VKIRGXRIELGEIE), F(DNFYXLGGHSL). Motif F represents the attachment site for the 4’-phosphopantetheine cofactor (thioestcr module). Motifs I-IV are conserved in peptide synthetase domains harbouring epimerase activity. I(AYXTEXND1LLTAXG). II(EGHGREXIIE), III(RTVGWFTSMYPXXLD), IV(FNYLGQFD). The acyltransfer module is characterized by the consensus HHXXXGD (“spacer or His motif’) which is found in acyltransferases. The methyltransferase module is present in peptide synthetase domains catalysing synthesis of N-methylated peptide bonds and has sequence motifs common with methyltransferases. The sequences of the various motifs are given in the one letter aa code.
90
RAINER ZOCHER AND ULLRICH KELLER
peptide synthetase genes in various organisms (Borchert et al., 1992; Turgay and Marahiel, 1994). Interestingly, five of the six strongest conserved motifs were also seen in a number of acyl CoA ligases such as coumarate ligase (Knobloch and Hahlbrock, 1975; Lozoya et al., 1988) and carboxyl-adenylate ligases, such as luciferase (de Wet et al., 1987), but not the aminoacyl-tFWAsynthetases indicating the existence of a distinct superfamily of adenylyl-carboxylate-formingenzymes (Turgay el al., 1992).
2.2. Motifs of the Carboxyl-adenylate-forming Domain Most of the work to elucidate the significance of the various conserved motifs in adenylate-forming domains (Fig. 2) has been done by studying peptide synthetase domains carrying site-directed mutations in various core motifs or by specific labelling of active site-located residues with ATP analogues or structurally related substrates. These analyses have been mostly done with TYI and to a limited extent with GSII. Experimental evidence was obtained that motif boxes B-E (Fig. 2) are involved in ATP binding and formation of carboxyl-adenylate. As suggested from its structuralresemblance with the Walker type A(phosphate-binding loop) (Walker etal., 1982), mutations in the glycine-rich sequence in motif B (GXXXXXGKT/S), such as replacement of the conserved lysine by unrelated amino acids, had drastic effects on the ability of the enzyme to activate D- or L-phenylalanine as adenylate. Similarly, mutations in the conserved aspartate in motif D resulted in loss of catalytic activity of the same reaction indicating a significant role of motifs B and D in ATP binding (Gocht and Marahiel, 1994). A specific role of motifs C and E in ATP binding and adenylation became clear through site-specific labelling of TYI with 8-azido-ATP and with fluorescein isothiocyanate, respectively (PavelaVrancic et al., 1994a,b). In other experiments it was shown that replacement of the conserved glycine in motif E (KIRCXRIEL) of the proline domain of GSII significantly affected the proline activation reaction (Tohika et al., 1993).
2.3. Modules in the Activation Domain
Analysis of peptide fragments carrying radioactively labelled substrate amino acids, such as leucine or valine (obtained after proteolytic cleavage of charged GSII), revealed that the motif F is that site of the amino-acid activating domain of peptide synthetases where the amino acid is attached to the enzyme in thioester linkage. Instead of the presence of a conserved cysteine, as was postulated in the previous model of the protein thiol template concept, the F consensus sequence GG IUD S WI shows similarity to the attachment site of the 4'-phosphopantetheine cofactor in numerous fatty acid and polyketide synthases (Hopwood and Sherman, 1990; Katz and Donadio, 1993). This sequence motif possesses a conserved senne
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
91
leading to the view that each amino-acid activating domain of the peptide synthetase harbours a 4’-phosphopantetheine arm,which functions both in aminoacid attachment and peptidyl transfer (Schlumbohmetal., 1991; Stein etal., 1994). Replacements of the corresponding serine residues by alanine in various peptide synthetase domains, such as in surfactin synthetases (D’Souza et al., 1993) or in TYI (Gocht and Marahiel, 1994), abolished thioester formation activity while leaving the adenylation activity of the enzymes unaffected. These findings required a revision of the previous model of peptide synthetases in which each activating domain possesses a 4’-phosphopantetheine cofactor as carrier for both amino acid and peptidyl intermediates. This is in contrast to the situation in fatty acid or polyketide synthase (PKS) enzyme systems where 4’-phosphopantetheine is attached to ACP only as a single swinging arm (Hopwood and Sherman, 1990).
2.4. The NMethylation Module
Numerous peptides contain N-methylated amino acids in their peptide chains (Billich and Zocher, 1990). Biochemical evidence has indicated that, in the case of enniatin, cyclosporin and actinomycin biosynthesis, the N-methylation takes place after covalent binding of the amino acid on the surface of the corresponding peptide synthetases with S-adenosyl-L-methionine (AdoMet) as substrate (Zocher et al., 1982, 1986; Keller, 1987). Haese et al. (1993) eventually showed that the amino-acid activating domain of enniatin synthetase (Esyn), the enzyme catalysing cyclohexadepsipeptide synthesis from branched-chain amino acids and D-2hydroxyisovalerate, contains an additional stretch of 450 amino acids inserted into a position between the conserved sequences E and F, the thioester formation module (Fig. 2). The 450 amino-acid module contains characteristic motifs with similarity to conserved sequences of various methyltransferases (Haese et al., 1993).No such 450 amino-acid insertion is seen in the hydroxy-acid activating domain of Esyn. These findings clearly assign the N-methylation function to the 450 amino-acid insertion. The N-methylation module was later seen also in the sequence of the cyclosporin synthetase gene, where it is present in 7 of the 11 amino-acid activation domains (Weber et al., 1994). The cycloundecapeptide cyclosporin A contains seven N-methylnted amino acids. Their relative position in the cyclopeptide is consistent with the order of the N-methylation modules in the peptide synthetase sequence.
2.5. Acyltransfer and Epimerization Modules
The observation of conserved motifs in a 350-amino-acid stretch distal to the amino-acid activation domains of peptide synthetases with the potential to elongate and epimerize amino acid or peptidyl residues, has led to the tentative assignment
92
RAINER ZOCHER AND ULLRICH KELLER
of these motifs to acyl transfer and epimerization functions in these enzymes puma eta!., 1993; de CrCcy-Lagard et al., 1995; Stachelhaus and Marahiel, 1995a; Stein et al., 1995). A set of four highly conserved sequences is proposed to be involved in racemizatiodepimerization reactions. These motifs, designated I-IV (Fig. 2), were found only in peptide synthetase domains known to be involved in the incorporation of D-amino acids in the corresponding peptides. A HHXXXDG motif, previously named “spacer motif’ (Stachelhaus and Marahiel, 1995a) and which pervades all modules catalysing peptide transfer or epimerization, is also present in the Tn9 chloramphenicol acetyltransferase (CAT) family as well as in members of the dihydrolipoamide acyltransferase (E2p) family, which catalyse acyl transfers (Guest, 1987). The “spacer motif’, also called the “His motif’, is always located to the carboxyterminal side of activation domains in peptide synthetases at a constant distance ( 170-190 amino acids) from the 4’-phosphopantetheine attachment site (F), except in those rarer cases where it precedes an activation domain of peptide synthetases that receive acyl groups from another enzyme, such as in the case of surfactin synthetase E1A (see below). The second histidine in the “His motif’ has been suggested to have a catalytic role as a general base catalysed in the deprotonation events required for acyl transfer reactions as well as in the epimerization of covalently bound amino acids or peptides (de CrCcy-Lagard et al., 1995). The probable involvement of proton acceptor group was shown in epimerization reactions catalysed by peptidyl epimerase functions, such as of actinomycin synthetase I1 (see below). Stachelhaus and Marahiel (1995b) presented evidence that the 350-amino-acid stretch harbouring the putative four racemization motifs (I-IV, Fig. 2) is in fact responsible for amino-acid racemization. Deletion mutants of GSI lacking this region do not epimerize covalently bound phenylalanine, while other functions of the synthetase, such as adenylation and thioesterification of the substrate, were unaffected.
2.6. Thioesterase Modules in Peptide Synthetase Genes
Analysis of open reading frames (orfs) in several peptidc-synthetase gene clusters revealed the presence of thioesterase (TE) genes with similarity to predicted vertebrate fatty acyl thioesterase I1 enzymes from duck and rat, which can transfer and thus catalyse release of fatty acids from fatty acid synthase. Such orfs are olfl and orf2 of the bialaphos synthesis gene cluster of Srreptoinyces hygroscopicus and grsT, associated with gramicidin S synthetase genes grsA and grsB in Bacillus brevis (Kratzschmar et al., 1989; Raibaud et al., 1991). By analogy to fatty acid synthesis, such thioesterase genes probably catalyse, in peptide synthesis, release of the completed product from the synthetase. In the case of ACVS from various sources, a thioesterase module was found in the C-terminal region of the gene as an integral part of the protein (Smith etal., 1990a; MacCabe etal., 1991a,b).These data further support the prediction made by Lipmann (1971) that peptide and
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
93
fatty-acid enzyme complexes share similarities with each other and may have common evolutionary origins. Meanwhile, cloning and sequencing of numerous polyketide synthase genes has revealed that polyketide antibiotics are also synthesized by multi-functional enzymes similar to FAS (Hutchinson and Fuji, 1995). All of these processes require cleavage of thioester bonds; therefore, thioesterase motifs are found either as an integral part of the polyketide synthases or as separate proteins. How product release (and cyclization) is accomplished in those cases of peptide synthetases where no thioesterase module is known, is unclear. It is also unclear in some cases whether thioesterase genes, which are associated with FAS genes, are really necessary for fatty acid synthesis (discussed by Hutchinson and Fuji, 1995).This point should be considered also in thiol template peptide synthesis.
2.7. Properties of Amino-acid Activating Domains
As shown in Fig. 1, the activation of amino acids proceeds in two steps. First, the amino acid is activated as adenylate, essentially as in the case of proteinogenic amino acids by the aminoacyl-tRNA synthetases. In the second step, the amino acid is covalently bound in thioester linkage by nucleophilic attack of the 4'-phosphopantetheine-SH to the acyladenylate. As shown by mutational analysis (Stachelhaus and Marahiel, 1995b) or by blocking the thioester formation module with sulfhydryl-directed agents (Zocher e l al., 1982), the adenylation reaction is independent of the presence of the thioester module. In this way the non-ribosomal system resembles the amino-acid activation by aminoacyl-tRNA synthetases, which activate their cognate amino acids as adenylates even in the absence of tRNA (Schimmel and Soll, 1979). In some cases inactivation or omission of the thioester formation module has led to significant reduction in adenylation reactions (Pfeifer etal., 1995), which might be due to structural constraints in the activation domain arising by deletion of such a long space-filling cofactor as 4'-phosphopantetheine. The 4'-phosphopantetheine also plays an important role in the epimerization reactions in peptide synthetases because the epimerized amino acids or peptides are attached to this cofactor in thioester linkage. The chiral centre inverted is the a-carbon adjacent to the thioestercarbonyl of the relevant residue. Interestingly, single arnino-acid activating enzymes, such as tyrocidine synthetase I or gramicidine synthetase 11, activate both L- and D-phenylalanine, and racemize these amino acids in their antipodal products (Gocht and Marahiel, 1994; Stein et al.. 1995). This is not true in the case of peptidyl epimerization. where the L-enantiomers are always activated and epimerization most probably takes place after peptide-bond formation (see the examples for actinomycin and surfactin below). Generally, one observes a rather strict stereospecificity in the activation reactions of the nonribosomal system, as in protein synthesis. By contrast, the specificity in terms of structural homology of substrates is lower than in protein synthesis, although in the mixtures of homologous series of peptides produced by microorganisms mostly
94
RAINER ZOCHER AND ULLRICH KELLER
conservative exchanges of amino acids occur. These are, of course, much more frequent than in protein synthesis, where misactivations are eliminated by proofreading mechanisms. How such proof-reading mechanisms in non-ribosomal peptide syntheses would work is not known.
3. ENZYME SYSTEMS
3.1. Organization of Activation Domains in Prokaryotes and Eukaryotes
Principally, peptide synthetase systems are defined as the arrangement of various amino-acid activation domains in the form of a multi-enzyme or a multi-enzyme complex. This has been shown by comparisons of various enzyme systems with their corresponding genes (Fig. 3). The order of the various activation domains is mirrored in the sequence of the peptide synthesized. In prokaryotes, one generally observes that the domains are distributed over more than one polypeptide chain with the exception of the single chain ACVS, which, however, is common to both the lower eukaryotes and prokaryotes (Aharonowitz et al., 1993). By contrast, all eukaryotic peptide synthetases, yet known, always consist of a single polypeptide chain encoded by an intronless gene. This single polypeptide chain harbours the various adenylate formation domains, thioester and additional modules necessary for the synthesis of a given product. This parallels observations concerning the modular organization of fatty acid synthases in prokaryotes and eukaryotes, which differ from each other in their structural and functional organization (Hopwood and Sherman, 1990).While in E. coli the FAS complex consists of ten different subunits each harbouring a distinct catalytic function, FAS from vertebrates is one polypeptide chain, and that of yeast consists of two polypeptide chains (Hopwood and Sherman, 1990).Similarly, in polyketide syntheses, which strongly resembles fatty acid syntheses, two types of polyketide syntheses have been described. While PKSI systems, as i n the case of erythromycin, consist of giant multi-enzymes containing the ACPdomain, P-ketoacyl synthase domain, sites for ketoreduction, dehydration, enoylreduction and thioesterase located on one polypeptide chain, PKSII systems consist of many separate largely monofunctional enzymes (Roberts et al., 1993). In contrast to the different FAS systems in prokaryotes and eukaryotes, no correlation of PKSI and PKSII systems with their occurrence in prokaryotes and eukaryotes is seen. The only exception is 6-methyl salicylic acid synthase of fenicilliuin patuluin, which is of PKSII-type but is represented by one single polypeptide chain (Hutchinson and Fuji, 1995). Besides determining the sequence of the peptide synthesized, the sequential order of activation domains and their accompanying modules is also important for
95
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
Arrangement of Domains in Peptide Synthesis
ml I
ProkarVoteS
I I
1 I
Pro Val Orn Leu
Gramicidin S
Leu Asp D-Leu E
Glu Leu D-Leu E
Surlactin
Eukarvotes
m]
Enniatin B
I I I I
I I II I I I
I I
I
I
D-AlaI Leu Me1 Leu Me( Val Me] Bmt Me Abu Gly Me1 Leu Me1 Val Leu lMel Ala Cyclosporin A ACV
Figure 3 Organization of peptide synthetase domains in various multi-functional peptide synthesis systems. The order and organization of peptide synthetase domains in several thiol template peptide synthesis systems is shown as deduced from DNA sequences of the corresponding enzymes and from analysis of their enzymatic activities. ACV. 6-(L-a-aminoadipyl)-cysteinyl-D-valine synthetase. Symbols in the adenylate forming domains indicate the comesponding activated amino acids: Aad, 8-L-a-amino adipic acid; D-Hiv, D-2-hydroxyisovaleric acid; Abu, a-amino-L-butync acid; Bmt, 4-(E)-butenyl-4methyl-L-threonine. The symbols E and TE denote presence of epimerase modules and thioesterase module, respectively. Me, N-methyltransferase.
the class of product, i.e. linear or cyclic, homodetic or heterodetic, etc. In any case, acyl transfer modules such as the “spacer motif’ have to be present between each activation domain of a biosynthetic sequence to enable condensation of the amino-acid residues. In the case of prokaryotic peptide synthetases, which receive acyl or peptidyl intermediates from another multi-functional enzyme, the same motif has also to be present in front of the first amino-acid activation domain, in order to enable the acylation of the first amino acid. The structural principles of the peptide synthetase domains outlined above would allow one to make similar predictions for the synthesis of other peptide classes such as those containing N-methylated peptide bonds. Whether such predictions would really lead to successful synthesis is dependent on many factors, most important among which are programming and timing of events in the assembly of the peptide. Most important may also be the priming of the reaction, the reactivity of the peptidyl intermediates and eventually the size of the cavity in the multi-enzyme where condensation and cyclization reactions take place, which may play a significant role for the chain length of the product. In the case of polyketide synthesis, there
96
RAINER ZOCHER AND ULLRICH KELLER
are examples of the effect of exchanging domains from different biosynthesis systems with consequences for the nature and length of the polyketide chain formed by engineeredPKS complexes (Hutchinson and Fuji, 1995;Tsoi and Khosla, 1995). In peptide synthesis systems, factors governing the programming of chain growth are as yet largely unknown. Most of the information on the reactions of peptide synthetase domains can therefore be obtained only by combination of both genetic and biochemical investigations of the reactions with enzymes in their wild-type and mutated form. In the following, the principles presented above of the structures of peptide synthetase domains will be illustrated by description of some selected, important peptide synthesis systems, the enzymology of which is well developed. These systems include prokaryotic as well as eukaryotic peptide synthetases responsible for the synthesis of cyclic and linear peptides, as well as N-methylcyclodepsipeptides.
4. PEPTIDE SYNTHETASES FROM FUNGI 4.1. 6-(L-a-Aminoadipyl)-cysteinyl-D-valine
&(L-a-Aminoadipyl)-cysteinyl-D-valine(ACV) is the common precursor of the penicillins and cephalosporins (Fig. 4). The peptide is assembled by ACV synthase (ACVS), which has been isolated from a representative number of fungi and bacteria. Based on biochemical investigations in the cases of the enzymes from A. nidulans and S. clavuligerus (van Liempt etal., 1989;Jensen et al., 1990; MacCabe et al., 1991a; Schwecke et al., 1992) and also considering the sequences of a number of ACVS genes (see below), it is clear that this enzyme is composed of three peptide synthetdse domains lying on one polypeptide chain of 420 kDa (Aharonowitz e f al., 1993).
SH
Figure 4 Structure of 6-(L-a-aminoadipyl)-cysteinyl-D-valine. 6-(L-a-aminoadipyl)cysteinyl-D-valineis the common precursor of the penicillins and cephalosponns.
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
97
4.1.1. ACVS Activates a- and 8-Carboxyl Groups ofAmino-acid Substrates Biochemical work has revealed that three independent sites (domains) are responsible for the activation of a-L-aminoadipic acid, L-cysteine and L-valine. While the first activation step of the latter two amino acids is clearly via the corresponding adenylates, conflicting results have been obtained regarding the a-L-aminoadipic acid. The enzyme of A. nidulans catalysed adenylation of a-aminoadipic acid as measured by aminoadipic acid-dependent ATP-pyrophosphateexchange, while the Streptomnyces clavuligerus ACVS did not (van Liempt et al., 1989; Schwecke et al., 1992). In the case of glutathione y-L-glutamyl-L-cysteinyl-glycine, the y-carboxylic group of glutamate is activated as phosphoryl-carboxylate from ATP with formation of ADP and the a-carboxyl group of y-glutamyl-L-cysteine is activated in the same way (reviewed by Meister, 1988). Thus, aminoadipic acid activation differs in its mechanism from the normal 0-carboxyl or peptide activation found in other systems of enzymatic peptide synthesis, such as of glutathione or muramyl peptide in bacterial cell-wall formation (Lipmann, 1980). However, covalent binding to ACVS through thioester linkage is observed with all of the three amino acids in all cases examined and 4'-phosphopantetheine has been shown to be present in several ACVSs tested (Baldwin etal., 1990, 1991). 4.1.2. Epirnerization in ACV Synthesis The fact that all of the valine attached to the enzyme in the activation step has the L-configuration indicates that epimerization of the L-valine is later during the events of peptide formation or in the peptide-bound state (van Liempt et al., 1989; Baldwin et al., 1991). Possibly, a thioesterase would be active in the release of the LLD peptide but not of the LLL intermediate, if the latter exists. Investigations of the reaction mechanism in terms of analysis of covalently bound peptidyl intermediates have not yet been reported. Recent findings, however, indicate that without L-aminoadipic acid, ACVS from Cephalosporiurn acrernoniurn forms and releases L-cysteinyl-D-valineas the product (Shiau et al., 1995b).Formation of the same product was enhanced in the additional presence of glutamate, which was activated by the enzyme but not incorporated in the product. Glutamate is a structural analogue of aminoadipate. The implications from these results are that the peptide bond between cysteine and valine is formed prior to epimerization, possibly in the peptide-bound state and also prior to the formation of the peptide bond between aminoadipate and cysteine. This mechanism would indicate a novel mechanism of peptide formation in thiol template peptide synthesis systems concerning the timing of peptide-bond formation events, which is not correlated with the order of the activation domains on the ACVS polypeptide chain. This mechanism would contrast with the previous model of ACV synthesis (Fig. 5). Interestingly, the cysteinyl-D-valine product was released from the enzyme
98
RAINER ZOCHER AND ULLRICH KELLER
I E I [TEl
Figure 5 Two possible schemes for events of peptide assembly catalysed by &@-aarninoadipy1)-cysteinyl-D-valine synthetase (ACVS). (a) Activation and peptide-bond formation steps in timely correlation with the order of peptide synthetasedomainson the ACVS polypeptide chain. (b) The start of the peptide-bond formation steps at the internal domain position with subsequent amino terminal peptide acylation. Note that in the threedimensional structure of ACVS, the Aad domain possibly could be vicinal in space to the Val domain. For details, see text. Aad, 6-L-a-aminoadipic acid; E, epimerase; TE, thioesterase.
indicating that the releasing enzyme activity in ACVS might recognize peptides shorter than ACV. Shiau et al. (1995a) have also shown that structural analogues of L-cys-D-Val such as 0-methyl-serinyl-D-valineare also efficiently formed and released from the synthetase together with a minor amount of 0-methyl-seryl-Lvaline. The latter finding would indicate that release of end product catalysed by the thioesterase module would not require strict stereospecificity at the a-carbon of valine in the peptidyl-thioester. 4.1.3.
ACVS Genes and Sequence Similarities of Peptide Synthetase Domains
Much of the knowledge on ACVS stems from sequence analyses of ACVS genes, which have been obtained from a variety of sources, such as actinomycetes, Gram-negative bacteria and fungi (Smith et al., 1990a,b; Coque et al., 1991; Diez et al., 1990, Tobin et al., 1990; Gutierrez et al., 1991; MacCabe et al., 1991b). In all cases, one intron-less orf has been found encoding a polypeptide of molecular mass around 420 kDa. This has facilitated determination of the conservation of fungal and prokaryotic peptide synthetase domains responsible for aminoadipic acid, cysteine and valine activation. Construction of phylogenetic trees on the basis
THlOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
99
of sequence similarities has clearly placed each of the three domains of ACVS in a separate branch regardless of their prokaryotic or eukaryotic origin (Turgay ef al., 1992). This might indicate horizontal gene transfer of ACVS domain genes between prokaryotes and eukaryotes, as has been discussed by Smith etal. (1990a). Similarly, genes encoding enzymes converting ACV to p-lactams, such as isopenicillin N-synthase. have led to similar conclusions of horizontal gene transfer of other p-lactam biosynthesis genes (Weigel et al., 1988). Interestingly, different fl-lactambiosynthesis genes involved in the formation of ACV and its modification have been found to be clustered in fungi and bacteria, which would support the horizontal gene transfer hypothesis (Smith ef al., 1990a; Gutierrez ef al., 1991; Martin and Gutierrez, 1995). Whether these hypotheseses will also lead to an evolutionary tree of peptide synthetase domains will be seen in the future.
4.2. Enniatins and Beauvericin
Enniatins are cyclohexadepsipeptides produced by various strains of the genus Fusariuin (Plattner et al., 1948). As shown in Fig. 6, enniatins consist of three residues of a branched chain N-methyl amino acid and D-2-hydroxyisovaleric acid (D-Hiv), arranged in an alternate fashion. Enniatins have antibiotic activity against various bacteria, exhibit immunomodulatory properties (Simon-Lavoine and Forgeot, 1979), and are potent inhibitors of mammalian cholesterol acyl transferase (Tomoda et al., 1992). Besides, enniatins are well known for their behaviour as ionophors with high specificity for potassium ions (Wipf et al., 1968). Enniatinproducing Fusaria are plant pathogens and enniatins were postulated to play a role as wilt toxin during infections of plants (Walton, 1990). Interestingly, enniatins like
Figure 6 Structures of enniatins and of beauvericin. Enniatin A: R1 = R 2 = R3 = sec-butyl; enniatin A l : R1 = iso-propyl, R2 = R3 = sec-butyl; enniatin B: R1 = R2 = R3 = iso-propyl; enniatin B1; R1 = R2 = iso-propyl, R3 = sec-butyl; beauvericin: R1 = R2 = R3 =
benzyl.
100
RAINER ZOCHER AND ULLRICH KELLER
the structurally related cyclodepsipeptides beauvericin and bassianolide also exhibit entornopathogenicproperties (Grove and Pople, 1980). 4.2.1. Enniatin Synthetase: Structure and Function Enniatins are synthesized by the multi-functional enzyme enniatin synthetase (Esyn), which was the first characterized N-methyl-cyclopeptide synthetase isolated from Fusarium scirpi (Zocher et al., 1982). Sequencingof the Esyn gene has revealed that the enzyme is one single polypeptide chain of 347 kDa (Haese et al., 1993). This is consistent with earlier biochemical investigations in which all catalytic functions of the enzyme were located on one polypeptide chain. Esyn synthesizes the enniatin molecule from its primary precursors D-Hiv and a branched-chain amino acid like L-valine or L-isoleucine in an ATP-dependent manner. AdoMet donates the methyl group in the N-methylated peptide bonds. As in other peptide synthetases 4'-phosphopantetheine is present as the prosthetic group. Dissecting the biosynthetic process in the individual steps catalysed by the individual domains of Esyn has revealed the picture of reaction steps shown in Fig. 7. Activation of D-Hiv and the branched-chain L-amino acids (e.g. L-valine) proceeds through adenylation and thioester formation. Characteristically, the thioesterified amino acid is methylated with AdoMet, and thus methylation takes place prior to peptide-bond formation and subsequent cyclization reactions. Omission of AdoMet in the in vitro system leads to formation of desmethyl enniatins (Zocher et al., 1982; Billich and Zocher, 1987a). Therefore, Esyn and other N-methylcyclopeptide synthetases can be considered as hybrid systems between peptide synthetases and N-methyltransferases. 4.2.2. Mechanism of N-Methylation of Amino A c i h in Thiol Template Peptide Synthesis A characteristic property of all N-methyltransferasesthat have been studied so far is their sensitivity to inhibition by S-adenosyl-L-homocysteine(AdoHCy), which is a reaction product of the methylation reaction derived from the methyl donor AdoMet. The antibiotic sinefungin, which is structurally related to AdoMet, is a potent inhibitor of a variety of methylases. Billich and Zocher (1987a) tested the effect of sinefungin and AdoHCy on product formation catalysed by Esyn, and found that sinefungin inhibits the methylation reaction of Esyn but allowed synthesis of desrnethyl enniatin even if present in excess amounts. Kinetic analysis showed a competitive inhibition pattern with respect to AdoMet, indicating direct competition of sinefungin with the AdoMet-binding site. By contrast, AdoHCy not only blocked formation of enniatin (methylated product) but also that of the unmethylated product. Kinetic analysis of the desmethyl enniatin synthesis revealed that, with respect to AdoMet, AdoHCy is a partial competitive inhibitor.
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
Substrate activation 1
ATP + 0 - H i v
-
+E
( D - H ; ~ AMP)E
21 ATP + L-Val + E Cj ( L - V a l Thioester formation
3) (D-Hiv
-
AMPIE
4 ) (L-Val - A M P ) E
+
D-Hiv
+
L-Val
101
+ ppi
-
AMPIE
-
S-E + A M P
+ PPi
S-E + A M P
N-Methylation
5) L-Val
-
AdoMet
S-E-
L-MeVal
- S-E
Pep tide-bond forma tion
6) D-HIv
-S -
L-MeVal
1
S’
E-f
D-Hiv-L-MeVal
HS, S’
-
Ester-bond formation
7) 0-Hiv-L-MeVal
-
S-E
Cvcl i Z 8 t i O f l
cyclo-[ o - H i v - ~ - M e V a l ]+- ~E Enniatin B
Figure 7 Scheme of partial reactions leading to enniatin catalysed by enniatin synthetase.
These results indicate that Esyn must harbour a discrete binding site for the inhibitor AdoHCy but not for sinefungin. Thus, blocking the N-methyltransferase function of Esyn through AdoHCy results in inhibition of the peptide-bond formation ability of Esyn (R. Zocher, unpublished). These data suggest that the active sites for depsipeptide formation and for the methylation step in the Esyn system are not independent of each other as was confirmed later by the analysis of the sequence of Esyn (see below).
4.2.3. Substrate Specificity ofEsyn Owing to the relatively broad substrate specificity of Esyn for amino and hydroxy
102
RAINER ZOCHER AND ULLRICH KELLER
acids, a variety of different enniatins can be synthesized by Esyn if appropriate concentrations of substrates (depending on the various K , values) are used (R. Zocher, unpublished). Nevertheless, Esyns from different Fusarium strains differ in their amino-acid specificity, i.e. they can have different K , values for each of the various branched-chain amino acids (Pieper et al., 1992). For example, the enzyme from the enniatin A producer E sarnbucinum exhibits high affinity for the substrate amino acids L - ~ and U L-Ile. By contrast, Esyn from the enniatin B producer R Zateritiurn preferably accepts L-Val, the constituent amino acid of enniatin B and therefore strongly resembles the Esyn from R scirpi. The reasons for the altered substrate specificity among Esyns may lie in mutations in the amino-acid binding sites of the polypeptide chains. 4.2.4. Mulecular Structure of Esyn Monoclonal antibodies directed to the multi-enzyme Esyn were used to map the catalytic sites of the enzyme (Billich et al., 1987). The antibodies could be divided into three groups based on their influence on catalytic functions. Members of group one exclusively inhibited L-Val thioester formation, while members of group two interfered with D-Hiv thioester formation. Antibodies of group three inhibited both L-Val thioester and D-Hiv thioester formation as well as the N-methyltransferase. From these findings, it was concluded that the two domains of Esyn containing the two catalytically binding sites are situated very close to each other in the three-dimensional structure of the enzyme. The immunochemical data with group three antibodies indicate that this is also the case for the N-methyltransferase site, which is in the vicinity of the acyl and aminoacyl binding sites. Titration of Esyn with radioactive AdoMet revealed binding of one mole AdoMet per mole Esyn (Billich and Zocher, 1987a) indicating the presence of one methylase unit per enzyme molecule. Interestingly, the adenylation reactions for D-Hiv and L-Val were not affected by the monoclonal antibodies, indicating that these reactions have different sites on the multi-enzyme at some distance to the thiol sites confirming the modular structure of peptide synthetase domains. 4.2.5. Structure of the Esyn Gene Analysis of the sequence of the Esyn gene (esynl)of Fusarium scirpi revealed an open reading frame of 9393 bp encoding a 347 kDa polypeptide that contains two highly conserved peptide synthetase domains (designated EA and EB) (see Fig. 8) (Haese eta!., 1993).The domain EA lying on the amino-terminal side of the protein could be identified as the D-Hiv binding site and EB as the L-amino-acid binding site on the carboxytenninal side. In contrast to domain EA, domain EB is interrupted by insertion of a 434 amino-acid portion (M-segment) between motifs E and F (Fig. 2). The M-segment contains a sequence with similarity to a motif
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
103
Figure 8 Structure of the enniatin synthetase as deduced from the gene sequence and biochemical characterizations.EA, D-Hiv-activationdomain; EB, L-Val-activation domain; M, methyltransferase; S, 4'-phosphopantetheine binding site.
apparently conserved within a number of methyl transferases (Haese et al., 1993). This portion of Esyn shows no homology to any region of known peptide synthetases (Haese et al., 1993). Tt was possible to express the M-segment in E. coli and to identify this protein as the methylase of Esyn by its binding properties for AdoMet. Three deletion mutants of this protein were shown to be inactive with respect to AdoMet binding (Haese et af., 1994). Similarly, a variety of other functional recombinant Esyn protein fragments could be expressed leading to the identification of the substrate activation sites for D-Hiv and L-Val (Pieper et al., 1995). A peculiar property of esynl is that it harbours two 4'-phosphopantetheine binding motifs in domain EB, one of which may represent a waiting position for peptidol units during chain growth.
4.2.6. Mechanism of Depsipeptide Formation Esyn consists of the two peptide synthetase domains EA and EB, but assembles three amino acids and three D-2-hydroxy acids in the final product enniatin. Studies on the mechanism of depsipeptide (enniatin B) formation revealed that the enniatin molecule is synthesized by three successive condensations of enzyme-bound (thioesterified) dipeptidols with each other (Zocher et al., 1983). This implies that, as in the fatty acid synthetases, Esyn contains a specific thiol group (waiting position) that picks up the intermediates of enniatin synthesis, i.e. the dipeptidol, tetrapeptidol and hexapeptidol to allow depsipeptide chain elongation (Fig. 9). After reaction of the thioester-bound N-methyl amino acid with the covalently bound D-Hiv (domain EA), the formed dipeptidol is transferred to the waiting position and attacked by the hydroxyl group of the newly formed dipeptidol yielding a tetrapeptidol. After transthiolation to the waiting position, the tetrapeptidol is attacked by the next new dipeptidol to fonn a hexapeptidol, which yields enniatin in the final condensation reaction. The presence of a second thioester formation module in the domain EB is consistent with this model and possibly this represents the waiting position. Apparently, the length of the growing depsipeptide chain is determined by the space provided by a putative cyclization cavity (Fig. 9).
104 (a) Cycle I
Pl-Hiv + P2-MeVal
-
( P2-MeVal-Hiv + P3-SH
RAINER ZOCHER AND ULLRICH KELLER
PI-SH + P2-MeVal-Hiv -+P2-SH + P3-MeVal-Hiv
P1-Hiv + P2-MeVal -* PI-SH + P2-MeVal-Hiv P2-MeVal-Hiv + P3-MeVal-Hiv-PP-(MeVal-Hiv)n + P3-SH PZ-(MeVal-Hiv)n + P3-SH --+ P3-(MeVal-Hiv)z + P2-MeVal-Hiv P1-Hiv + P2-MeVal --+Pl-SH P2-MeVal-Hiv + P3-(MeVal-Hiv)t --+ PZ-(Meval-Hiv)~+ P3-SH PZ-(MeVal-Hiv)s + P3-SH -~+ P2-SH + P3-(MeVal-Hiv)s + P3-SH + enniatin Cyclization P3-{MeVal-HIV)3 -
Figure 9 (a) The scheme of events of dipeptidol condensations on enniatin synthetase and the role of the additional 4'-phosphopantetheine containing thioester module as the waiting position. P1, P2, P3 = 4'-pliosphopantetheine groups. (b) Model of arrangement of catatytic sites of enniatin synthetase. Cy = cyclization cavity; EA, D-Hiv-activation domain; EB, L-Val-activation domain; M, methyltransferase.
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
105
4.2.7. Biosynthesis of Beauvericin Beauvericin is a homologue of enniatins in which the branched-chain N-methyl amino-acid position always contains N-methyl-L-phenylalanine(see Fig. 6). Beauvericin synthetase catalysing beauvericin synthesis from L-Phe and D-Hiv under consumption of ATP and AdoMet has been isolated from the fungus Beauveria bassiana (Peeters et al., 1983). The enzyme strongly resembles Esyn with respect to its molecular size and the reaction mechanism. The main differences between both enzymes lie in the substrate specificity. Beauvericin synthetase exhibits a high specificity for aromatic substrate amino acids of the phenylalanine type, whereas Esyn is unable to incorporate such compounds.
4.3. Cyclosporin and Related Peptides
Cyclosporin is a cyclic undecapeptide with anti-inflammatory, immunosuppressive, antifungal, and antiparasitic properties (Borel, 1986). It is used world-wide in transplantation surgery and in the treatment of autoimmune diseases (Kahan, 1984; Schindler, 1985). The structure of cyclosporins is shown in Fig. 10. Besides the unusual amino acids a-aminobutyric acid (Abu), D-alanine and 4-(E)-butenyl4-methyl-~-threonine(Bmt), it also contains a number of N-methylated peptide bonds. Cyclosporins are produced by the fungus Beauveria nivea as a main component of 25 naturally occurring cyclosporins, mainly differing in positions 1, 2,4,5,7 and 11. Furthermore, unmethylated peptide bonds may occur in positions 1,4,6,9, 10 and 11 of the cyclosporin ring (Traber et al., 1987). 4.3.1. Biosynthesis of Cyclosporins Cyclosporins are synthesized via a thiol template mechanism, which, owing to the N-methylating steps, has strong resemblance to that of enniatin synthesis. This became evident from studies in vivo of the biosynthesis of cyclosporins (Zocher et al., 1984). In the initial cell-free studies on cyclosporin synthesis performed in the authors’ laboratory, a high molecular-weight-enzymefraction was obtained capable of activating all constituents of the undecapeptide and carrying out specific N-methylation reactions (Zocher ef al., 1986). Attempts to synthesize cyclosporin failed enzymatically. However, the enzyme synthesized the diketopiperazine c(D-Ala-N-MeLeu),representing a partial sequence of cyclosporin, which revealed the significance of the enzyme. Cell-free total synthesis of the cyclosporin molecule was finally established with an enzyme fraction obtained from a cyclosporin high producer strain (Billich and Zocher, 1987b). A number of different naturally occurring cyclosporins were synthesized by the multi-enzyme that has been designated cyclosporin synthetase (Cysyn). Further purification and characterization of
106
RAINER ZOCHER A N D ULLRICH KELLER
Figure 10 Structure of cyclosporin A. Bmt, 4-(E)-butenyl-4-methyl-~-threonine; Abu, a-amino-L-butyric acid.
this multi-enzyme supported earlier findings that Cysyn strongly resembles the well-characterized Esyn system with respect to amino-acid activation and N-methylation reactions (Lawen and Zocher, 1990). Furthermore, 4'- phosphopantetheine was detected as a covalently bound cofactor in Cysyn. Like Esyn the multi-enzyme CySyn is unable to carry out racemization reactions. D-Ala is directly incorporated into the D-Ala position of cyclosporin. The D-Ala moiety is racemized from L-Ala by a specific alanine racemase, which plays a key role in cyclosporin biosynthesis (Hoffmann et al., 1994). 4.3.2. Mechanism of Cyclosparin Synthesis Studies on the mechanism of cyclosporin formation were reported by Lawen et al. (1994).The authors were able to isolate four enzyme-bound intermediate peptides of cyclosporin biosynthesis from a complex mixture of unidentified peptides. All four isolated peptides canied alanine as the N-terminal amino acid. One of the
THlOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
107
peptides represented the nonapeptide H-D-Ala-MeLeu-MeLeu-MeVal-MeBmtAbu-Sar-MeLeu-Val-OH.From these findings, the authors conclude that D-alanine provides the “starter amino acid” of cyclosporin synthesis leading to formation of a linear precursor undecapeptide with an N-terminal D-Ala. Cyclo-sporin is formed in a final cyclization step.
4.3.3. Substrate Specificity of Cysyn From the spectrum of naturally occumng cyclosporins, it seems obvious that some of the peptide synthetase domains of Cysyn have a rather broad substrate specificity and others, such as that responsible for Bmt, have a lower specificity, allowing incorporation of homologue substrates. Therefore, it is not surprising that a number of new immunosuppressive cyclosporins could be synthesized in vitro (Lawen et al., 1989). Results of studies on the substrate specificity of the different binding sites of Cysyn at the cell-free level (Lawen and Traber, 1993) agreed with findings from studies in vivo (Kobe1 and Traber, 1982; Traber et al., 1989). 4.3.4. Molecular Structure of Cysyn Cyclosporin synthetase is encoded by a giant 45.8 kb open reading frame. The predicted gene product is a polypeptide of about 1600 kDa containing 11 peptide synthetase domains, of which 7 are homologous to the EB domain of enniatin synthetase carrying the integrated N-methyltransferase module (Weber et al., 1994).From the arrangement of the domains and the fact that domains in peptide synthetases are colinear to the arrangement of amino acids in the peptides to be synthesized, the authors conclude that the 5’-terminal domain is responsible for D-Ala activation. The last domain at the 3‘-end would represent the L-Ala-activating pepride synthetase domain (see Fig. 3). This assumption was supported by the finding that a Cysyn fragment of 130 kDa could be isolated, which is capable of activating L-Ala. Edman degradation of this protein yielded a sequence with an N-terminus in the position of amino acid 13 601 of Cysyn (Weber et al., 1994). Like Esyn and other peptide synthetases, the peptide synthetase domains of Cysyn contain all highly conserved motifs and modules described as characteristic for this class of enzymes (see above).
4.3.5. Cyclosporiiz-related Peptolide SDZ 214-1 03 The peptolide SDZ 214-103 is a cyclosporin-related undecapeptide lactone produced by the fungus Cylindrotrichurn oligospennum (Dreyfuss et al., 1988). It is a peptidolactone analogue of the immunosuppressive undecapeptide cyclosporin, carrying a D-2-hydroxyisovaleratemoiety in position 8 of the ring system
108
RAINER ZOCHER AND ULLRICH KELLER
instead of D-Ala (see Fig. 10). This compound is synthesized by a multi-functional enzyme from its precursor amino acids and D-Hiv with consumption of ATP and AdoMet. The enzyme has been described to be a single polypeptide chain with a molecular mass of about 1400 kDa and strongly resembles Cysyn ( h w e n et al., 1991). Studies of the substrate specificities of peptolide synthetase revealed that most sites of this multi-enzyme appear to have narrower specificities than those of Cysyn. The D-2-hydroxy-acid position (corresponding to position 8 of the cyclosporin ring) can be occupied by a large range of substrates varying from D-lactic to D-2-hydroxyisocaproic acid.
4.4. Ergot peptide alkaloids
Ergot alkaloids of the peptide type are produced by the ergot fungus Claviceps purpurea. They consist of D-lysergic acid attached in amide-like fashion to a vipeptide arranged into a unique cyclol structure as in ergotamine (Fig. 11). The cyclol structure results from the modification of the amino acid adjacent to the lysergyl moiety into an a-hydroxy-a-amino acid. This.modification is believed to take place after the assembly of the D-lySergyhipeptide and most probably after the formation of the corresponding proline lactam (Fig. 11) (reviewed by Stadler, 1982; Kobel and Sanglier, 1986). Introduction of the hydroxyl group to the a-position of the amino acid is followed by spontaneous non-enzymatic closure of the cyclol ring (Hofmann et ul., 1963). The ergot cyclol peptide alkaloids-also called ergopeptines-constitute a family of compounds that differ from each other by substitutionsof amino acids in the first and second position of the peptide moiety adjacent to D-lysergic acid. Common to all naturally occurring ergopeptines is D-lysergic acid and proline, the latter located at the carboxy terminal position of the peptide chain. In the last two decades, members of an additional new class of D-lysergylpeptides have been isolated, which differ from the ergopeptines in that the cyclol structure is missing. They contain a proline lactam ring and therefore are called ergopeptams (Stadler, 1982; Kobel and Sanglier, 1986). In contrast to the ergopeptines, the proline here has the D-configuration. Most probably, these compounds arise from spontaneous isomerization of the corresponding L-prolinecontaining stereoisomer, e.g. D-lysergyl-L-alanyl-L-phenylalanyl-L-pro1in lactam (Fig. II), which can not be further converted to the corresponding ergopeptine. 4.4.1
I
D-Lysergylpeptide Assembly
Investigations of a putative ergopeptine synthetase indicate that the compound is synthesized by a non-ribosomal mechanism. However, this enzyme proved to be unable to incorporate free D-lySefgiC acid into the ergopeptine. Instead, it converts and incorporates biogenetic precursors of D-lysergic acid such as elymoclavin
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
109
Figure 1 1 Biosynthetic relationship of structures of D-lysergyl-L-alanyl-L-phenylalanyl-L-proline lactani and ergotamine. Ergotanline (right) is a representative meniber of the ergopeptine family. The various groups of naturally occuming ergopeptines differ from each other by exchpnge of the amino acids in the two positions adjacent to the tetracyclic D-lysergic acid. Exchange is exclusively with branched-chain amino acids or phenylalanine. D-lysergyl-L-alanyl-L-phenylalanyl-~-proline lactam (left) is the immediate precursor of ergotarnine and is assembled by D-lySergyl peptide synthetase (see text).
(Maier et ul., 1983). This contrasts with the finding of a D-lysergic acid-activating enzyme that had been isolated from C. purpurea, suggesting that D-lysergic acid might be a free intermediate in ergopeptine synthesis. The enzyme is a single polypeptide chain of 62 kDa in its denatured form, which catalyses the ATPpyrophosphate exchange dependent on D-lysergic acid but, surprisingly, is unable to form a thioester with D-lySergiC acid or its structural analogue dihydrolysergic acid (Keller et al., 1984b, 1988). Therefore, it has been tentatively considered as a D-lysergic acid-AMP ligase acting in a similar fashion to actinomycin synthetase I and related proteins of prokaryotic origin (Keller, 1995). Keller et al. (1988) eventually reported a cell-free system of D-lysergyl peptide synthesis that catalysed formation of D-lysergyl-L-alanyl-L-phenylalanyl-L-pro1ine lactam from D-lysergic acid, alanine, phenylalanine and proline, which is the immediate precursor of ergotamine (see Fig. 11). The enzyme that catalyses formation of the D-lysergyl peptide was partially purified and shown to be a 500-550 kDa multi-enzyme complex under native conditions. Labelling the enzyme with radioactive substrate amino acids and dihydrolysergic acid enabled identification of the components of the enzyme complex by fluorography of SDS-PAGE gels. Radioactive dihydrolysergic acid (a reactive structural analogue of D-lySergiC acid) labelled a 160 kDa fragment, while radioactive phenylalanine labelled a 370 kDa fragment, that most probably also contains the binding sites for alanine and proline. Recent investigationswith antibodies raisedagainst the 62 kDa D-lysergic acid-activating enzyme previously isolated from Cluviceps purpurea revealed strong cross-reaction against various bands in the 100-200 kDa range, among them the 160 kDa fragment carrying the site of the D-lysergic acid thioester formation module (B. Riederer and U. Keller, manuscript submitted).The available data suggest that the D-lysergic acid-activating enzyme most probably arises through proteolytic degradation of the 160kDa fragment, which possesses an intact
110
RAINER ZOCHER AND ULLRICH KELLER
peptide synthetase domain. The 62 kDa is only able to activate D-lysergic acid as adenylate and not as thioester which indicates that the thioester formation is lost (Keller etal., 1988).In addition, degradation of the 370 kDa fragment was observed in partially purified protein fractions, where smaller fragments still activated individual amino acids as thioester, such as phenylalanine (M. Han and U. Keller, unpublished data). These data clearly indicate that proteolytic fragmentation of peptide synthetases can result in the formation of intact separate peptide synthetase domains still activating their cognate amino acids. From the data, therefore, it is most likely that the two fragments of the D-lysergylpeptide synthetase result from fragmentation of a 500-530 kDa single polypeptide as shown in Fig. 12. Similarly, results obtained in the case of HC-toxin synthesis indicate that HC-toxin synthetase from Cochliobolus carboneuin is a 550 kDa protein as estimated from the length of the 15.7 kb gene (Scott-Craig et al., 1992). However, purification and analysis of the enzyme revealed that it is always present as two fragments of about 300 kDa activating and epimerizing three of the four amino acids of HC-toxin (Scott-Craig et al., 1992).The previous isolation of a phage clone from a C. purpurea genomic
LSA
Ala
Phe
Pro
Fragmentation
160 kDa
t.’” I I I ....
S ‘LSA
62kDa
I-:::
I1 S
‘Ala
111 I S
‘Phe
D-Lysergyl peptide lactam
IV I
s‘Pt-O
D-Lysergic acid activating enzyme
LSA-AMP Figure 12 Scheme of domain assembly of D-lysergyl peptide synthetase from Cluviceps purpureu. The enzyme system catalysing D-lysergyl peptide synthesis (see Fig. 11) consists of two protein fragments of 370 kDa and 160kDa, which harbour sites for thioester formation of the amino acids and D-lysergic acid, respectively. The 62 kDa fragment represents the previously isolated D-lysergic acid activating enzyme of C. purpurea, which most probably is derived from the D-lysergic-acidactivating domain of D-lysergyl peptide synthetase.LSA,
D-lysergic acid.
THIOL TEMPLATE PEPTiDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
111
library containing four peptide synthetase domains in tandem raises the possibility that the ergot peptide synthetase also consists of one polypeptide chain with four activation domains for D-lysergic acid and the three amino acids of the cyclol nucleus (s. Riederer and U. Keller, unpublished). However, comparison of D-lysergylpeptide synthetase sequences with the DNA sequences are necessary to confirm the identity of that gene. The possible existence of one single orf for this enzyme would fit with the observation that eukaryotic peptide synthetases always consist of one polypeptide chain.
5. PROKARYOTIC PEPTIDE SYNTHETASE SYSTEMS
5.1. Acyl Peptide Lactones Much progress i n the enzymology of prokaryotic peptide synthetases has been achieved in the field of the acyl peptide lactone synthetases. Acyl peptide lactones consist of peptide lactone rings to which are attached aromatic or aliphatic side groups in an amide-like fashion. Acyl peptide lactones with aromatic side groups are mostly produced by streptomycetes whereas the various fatty acyl peptide lactones are mostly formed by Bacillus species (Vater, 1989). Both classes of compounds contain L- and D-amino acids in their chains. In addition, the streptomycete acyl peptide lactones contain N-methyl amino acids. No N-methyl peptides have been found in Bacillus yet. Examples of the acylpeptide lactones with aromatic side chains are the actinomycins (bicyclic pentapetide lactones), the quinoxaline antibiotic (monocylic octadepsipeptides), mikamycin BI (monocyclic hexapeptide lactones) and mikamycin BII (monocyclic heptapeptide lactones) antibiotics (structures reviewed by Okumura, 1983; Keller, 1995). Many of these compounds have therapeutic value as antibiotics or cytostatics. The enzymatic steps in the biosyntheses of the different aromatic acyl peptide lactones appear to be similar. 5.1.1. Actinomycin Biosynthesis as a Model of Aromatic Acyl Peptide Lactone Biosynthesis The large numbers of different actinomycins arise by substitutions in several positions of their pentapeptide lactone rings with homologous amino and imino acids (Katz, 1968; Meienhofer and Atherton, 1973). Actinomycins (e.g. actinomycin D) are bicyclic and originate from a monocyclic precursor, in which one pentapeptide lactone ring is attached to 4-methyl-3-hydroxyanthranificacid (4MHA). 4-MHApentapeptide lactone is the prototype of the whole class of aromatic acyl peptide lactones, although only short-lived in the cell because of dimerization
112
RAINER ZOCHER AND ULLRICH KELLER
MeVal
MeVal
MeVal
I
2 x
I
Thr
I
Thr I
&"*; CH3 OH
4-MHA pentapeptide lactone
Actinomycin D
Figure 13 Synthesis of actinomycin from its immediate precursor4-methyl-3-hydroxyanthranilic acid pentapeptide lactone. Oxidative condensation of 4-methyl-3-hydroxyanthranilic acid (4-MHA) pentapeptide lactone to actinomycin is enzymatic (see Keller, 1995). Sar, sarcosine (N-methyl-glycine); MeVal (N-methyl-L-valine).
through oxidative condensation of two such molecules immediately after its formation (Fig. 13). It is assembled from 4-MHA and the five amino acids constituting the peptide ring and S-adenosyl-L-methionine in an ATP-dependent manner. Synthesis is accomplished by a set of three peptide synthetases, which contain a total of six activation domains, two of which contain an N-methylation module (Fig. 14). 5.1.2. Actinornycin Synthelase I
Actinomycin synthetase I (ACMSI), the first enzyme involved in actinomycin synthesis, is a 4-MHA-AMP ligase, which catalyses the synthesis of adenylyl-4MHA from 4-MHA and ATP (Keller et al., 1984a; Keller and Schlumbohm, 1992). From its properties, it appears as merely consisting of an adenylate forming domain with a size of 45 kDa. The enzyme has broad substrate specificity with respect to various structurally related benzene carboxylic acids. However, heteroaromatic carboxylic acids, such as pyridine, quinoline or quinoxaline carboxylic acids are not activated (Glund et al., 1990; Schlumbohm and Keller, 1990; Keller and Schlumbohm, 1992). Structurally related benzene carboxylic acids with hydroxy groups in the 3-position or methyl group in the 4-position of the benzene nucleus serve as efficient substrates in peptide synthesis in vivo and in vitro. For example, strong evidence for the involvement of ACMSI in actinomycin biosynthesis came
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
AdoMet
___)
Gly
Pro
Val Thr
AdoMet
-
Actinomycin Synthetase 111
113
MeVal
s
Sar
4
3
2
___)
Actinomycin Synthetose II Thr
1
NH
I
4-MHA
Actinomycin Synthetose I
Figure 14 Assembly of 4-methyl-3-hydroxyanthranilic acid pentapeptide lactone by actinomycin synthetases. Actinomycin synthetases activate, modify and polymerize the constituent amino acids and 4-methyl-3-hydroxyanthranilicacid (CMHA). Sar, sarcosine (N-methyl-glycine); MeVal (N-methyl-L-valine); AdoMet, S-adenosyl-L-methionine (methyl group donor).
from the finding that feeding structural analogues of 4-MHA, such as 4-methyl-3hydroxybenzoic acid, 3-hydroxybenzoic acid, p-toluic acid or p-aminobenzoic acid, to cells of actinomycin-producingS. chrysomallus and S.antibioticus resulted in the formation of new compounds instead of actinomycin (Keller, 1984). The new compounds were monocyclic acylpentapeptide lactones (actinomycin half molecules) containing the administered analogue instead of 4-MHA. 5.1.3. Actinomycin Synthetases II and III ACMSII and ACMSIII have been isolated from actinomycin-producingStreptomyces chrysomallus. ACMSII, a 280 kDa multi-enzyme, activates L-threonine and L-valine (occupying positions 1 and 2 in the ring) as thioesters via the corresponding adenylates (Keller, 1987; Stindl and Keller, 1993). ACMSIII activates proline,
114
RAINER ZOCHER AND ULLRICH KELLER
glycine and valine, which are in positions 3 , 4 and 5 of the pentapeptide lactone ring of actinomycin D, and has a molecular mass of 480 kDa (Keller, 1987; Stindl and Keller, to be published) (Fig. 14). In the presence of S-adenosyl-L-methionine, thioester-bound glycine and valine are N-methylated, which yields covalently bound sarcosine and N-methyl-L-valine,respectively. This indicates that ACMSIII has two N-methylation modules necessary for the N-methylation of covalently bound glycine and valine, respectively (Fig. 14). Several lines of evidence indicate that these modules are organized like the N-methylation module in the amino-acid domain of enniatin synthetase. First, the reaction catalysed by the two enzymes proceeds by the same mechanism. Second, monoclonal antibodies directed against the N-methyltransferdse domain of enniatin synthetase strongly cross-reacted with ACMSIII but not with ACMSII, confirming the presence of N-methyltransferase domain(s) in ACMSIII (A. Billich, R. Zocher and U. Keller, unpublished). Both ACMSII and ACMSIII contain 4'-phosphopantetheine as a covalently bound cofactor. 5.1.4.
Total Cell-free Synthesis of Acyl Pentapetide Lactone Biosynthesis
An enzymic total synthesis of acyl pentapeptide lactone analogues of 4-MHA pentapeptide lactone has been achieved as yet only in a system of permeabilized cells of S. chrysonmllus (A. Stindl and U. Keller, unpublished data). The various steps of peptide synthesis are summarized in Fig. 15. In this system, the incorporation into the complete product of all constituents could be demonstrated, which was not observed in cell-free fractions containing the three ACMSs at much higher concentration than in the permeabilized cell system. These findings indicate that functional activity of the synthetase complex strongly depends on the integrity of the cellular structure possibly through membrane association of the ACMS complex. Analysis of the individual synthetases with respect to their reactivity, such as in adenylate formation, thioester formation, N-methylation, peptide-bond formation and epimerization, indicate that they are functionally intact (Keller and Schlumbohm, 1992; Stindl and Keller, 1993,1994). Apparently, impairment of the correct positioning of the three synthetases in the complex andor of final steps in product release interfere with formation of the acyl pentapeptide lactone. 5.1.5. Priming of Reactions in initiation and Elongation of Peptide Synthesis on ACMSII ACMSII contains three distinct sites for covalent binding of 4-MHA, L-threonine and L-valine as thioesters (Keller, 1987;Stindl and Keller, 1993,1994). ACMSII is not able to activate 4-MHA as adenylate. Instead, ACMSI delivers adenylyl-4MHA to ACMSII, which binds 4-MHA as thioester (see Fig. 15). L-Threonine and L-valine are also bound as thioesters. However, prior to this, ACMSII activates
115
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA A N D FUNGI
280 kDa
45 kDa 4MHA
4MHA-AMP7
L-2
Thr
+
Val
+
1
4 Val
GlY
4
AdoMet
AdoMet
4 Ro
480 kDa Figure 15 Organization of the actinomycin synthetase multi-enzyme complex and events of acylpentapeptide chain growth. For explanations of abbreviations, see Fig. 14.
these two amino acids as adenylates. This indicates the presence of two complete peptide synthetase domains on ACMSII for threonine and valine, and one additional thioester formation module for 4-MHA most probably separate from the Thr and Val domain. Functionally, the charging of ACMSII with 4-MHAcan be described as priming of the events taking place on the multi-functional enzyme. Two different initiation reactions can take place in the presence and absence of 4-MHA, respectively (Fig. 16). In the presence of 4-MHA (or its analogue p-toluic acid used in the experiments) peptide synthesis starts with formation of 4-MHA-threonine, the latter reacting further with L-valine to yield a mixture of 4-MHA-threonyl-~-valine
116
(a)
e I : -
RAINER ZOCHER AND ULLRICH KELLER
S -4MHA
4W-AMP Thr,Val, ATP
SH
(b) SH
Val - Thr - 4MHA LDVal- m-4m
es"
Thr ,Val, ATP
s~
S-Thr
sr Val
\D-Val
- Thr
-
Thr
Figure 16 Effect of priming the initiation reaction of acyl peptide lactone on actinomycin synthetase 11. For explanation of abbreviations, see Fig. 14. (a) Initiation in the presence of 4-MHA-adenylate (supplied by ACMSI). (b) Initiation in the absence of 4-MHAadenylate. For details see text.
and 4-MHA-L-threonyl-D-valine. If 4-MHA is absent in this reaction, ACMSII forms a mixture of threonyl-L-valine and threonyl-D-valine. However, the formation of these two dipeptides gives considerably lower yield than the yield of 4-MHA dipeptides, which are formed when 4-MHA is present. Thus, priming of the reaction by acylation of threonine with 4-MHA strongly influences the reactivity and the substrate specificity of the enzyme(s).
5.1.6. Epirnerizatioiz of Amino Acid in the Peptide-bound State ACMSII activates L-valine but not D-valine. Epimenzation of valine takes place after peptide-bond formation with threonine leading to 4-MHA-~-Thr-~-val (Stindl and Keller, 1994). During inversion at the a-carbon of L-valine in the covalently bound acyl dipeptide, a proton is abstracted that goes into solvent. This is strong evidence for the intermediacy of a carbanion structure at the chiral centre. Epimerizdtion proceeds independently of addition of cofactors, such as NAD, FAD or pyridoxalphosphate. Spectral analysis of ACMSII indicates the absence of any tightly bound cofactors in the enzyme. Thus, this mechanism of peptide epimerization is novel and may operate in the synthesis of a large number of nonribosomally as well as of ribosomally made peptides, which have D-amino acids at positions other than the amino-terminal end (Stindl and Keller, 1994). In the case of peptides with D-amino acids as amino-terminal residues of the growing peptide
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
117
chain, the activating enzymes, such as gramicidin S synthetase I, racemize equally well the L- or D-isomer of the respective amino acid (Stein et al., 1995). Other systems, such as cyclosporin synthetase, do not possess racemase function and use D-amino acids as substrates that have been epimerized prior to activation. Stindl and Keller (1994) also propose an unknown proton acceptor group that is present on ACMSII, catalysing proton abstraction during inversion in a one- or two-base mechanism, which might be identical to the conserved His in the abovementioned “His motif’ of peptide synthetase domains. Meanwhile, a similar mechanism of epimerization of amino acid in the “peptide bound state” was also proposed for the epimerization of valine in 6-(L-a-aminoadipyl)-L-cysteinyl-D-vdinesynthesis catalysed by ACV synthetases (Shiau et al., 1995a,b). 5.1.7. Peptide-bond Formation Catalysed by ACMSIII Purified ACMSTII catalyses formation of prolylsarcosine and sarcosyl-N-methylL-valine. The latter two dipeptides cyclize spontaneously to the corresponding diketopiperazincs and are released from the enzyme. These partial reactions give relatively high yields. In the additional presence of ACMSI and 11, acylpeptide chain intermediates most probably representing acyltri-, tetra- and pentapeptide were identified as covalently bound intermediates. The whole sequence of events in acyl pentapeptide (lactone) synthesis based on these data is shown in Fig. 15 (U. Keller and A. Stindl, unpublished data). An enzyme fraction that catalyses the final acyl pentdpeptide lactone has not yet been isolated.
5.2. Surfactin Surfactin is a lipopeptide lactone composed of seven amino acids in the order Glu-Leu-D-Leu-Val-Asp-D-Leu-t-Leu in a heptapeptide ring to which is attached a P-hydroxy fatty acid (Fig. 17). The compound has been shown to be synthesized cell-free by a thiol template mechanism (Kluge et al., 1988; Ullrich et al., 1991; Menkhaus et al., 1993; Zuber et al., 1993). The multi-enzyme system responsible for the assembly of the compound has been resolved into its components and reconstituted from the latter ones into the functionally activecomplex by Menkhaus et al. (1993). These authors identified the enzyme fractions by their ability to catalyse the activation of the various amino-acid constituents of surfactin. Enzyme El consists of two multi-lunctional peptide chains of460 (Ela) and 435 kDa (Elb). Ela is responsible for adenylation and thioesterification of L-Glu, L-Leu (positions 1-3) and El b for adenylation and thioesterification of L-Val, L - A s ~and L - ~ U (positions 4-6). The 160 kDa enzyme E2 is responsible for activation of L - ~ U (position 7), while enzyme E3 has activity of an acyl transferase incorporating
118
RAINER ZOCHER AND ULLRICH KELLER
Leu I
- :0
D-Leu I ASP
Val I D-Leu I Leu I
Glu I
co I
CHZ I
CHL
2
Figure 17 Structure of surfactin.
L-hydroxy fatty acid into peptide from L-hydroxy fatty acyl CoA thioester. E3 has a mass of 40 kDa (Fig. 18).
5.2.1. Initiation of Surfactin Synthesis Menkhaus et al. (1993) have shown that the acyl transferase enzyme E3 of surfactin synthetase acylates L - G thioesterified ~ to El a with an L-P-hydroxy fatty acyl residue with consumption of L-P-hydroxy fatty acyl CoA thioester. The resultant product is L-P-hydroxy fatty acyl glutamate covalently bound to enzyme Ela. D-P-HY~~oxY fatty acyl CoA was a poor substrate under these conditions indicating specificity of the acyltransferase with respect to optical configuration. By contrast, the enzyme has broad specificity with respect to fatty acyl chain length. Whether the enzyme directly acylates Glu on enzyme Ela or transfers the fatty acyl chain prior to that step to a thiol group of Ela is not known. The whole mechanism of initiation differs from that of actinomycin synthesis in that ACMSI is not an acyl transferase (Stindl and Keller, 1993).
5.2.2. Elongation and Epiinerization Reactions in Amino-acid Positions of Surfactin Enzyme Ela forms thioesters with Glu and L - ~ inUa molar ratio of 1:2.Enzyme Elb binds L - A s ~L-Val , and L-Leu in a molar ratio of 1:l. Enzyme E2 binds L - ~ u . Determinations of the optical configurations of all the amino acids covalently bound to the enzymes revealed that all of the leucines were of the L-configuration. Thus, Menkhaus et al. (1993) argue that the enzymes are lacking racemization
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
Glu
Leu
Leu
1
1
1
I
s\Glu
I
I
I
'\Leu
I
I El
119
EnzymeEla
'\D-Leu
Leu
+]
Enzyme E2
Figure 18 Enzyme organization of surfactin biosynthesis as deduced from gene sequences and biochemical data.
functions and that epimerization reactions take place in the course of the condensatiozdelongation reactions. Possibly these take place at the stage of peptidyl intermediates analogous to the situation encountered in the biosynthesis of actinomycin. Analysis of the 4'-phosphopantetheine contents in the various fractions indicated that this cofactor was present as a prosthetic group in enzymes Ela, Elb and E2. Details of the mechanisms of termination and release of the end product surfactin are not yet available.
5.2.3. Structure-ficnctiun Relationships in Surfactin Synthetases More information of the structure-function relationships of the surfactin synthetases became available by investigating surfactin synthetases carrying sitespecific mutations in the reaction centres of the various activation domains. The complete DNA sequence of the surfactin biosynthesis operon had been obtained previously by several groups who showed that part of it is important for competence establishment in Bacillus subtilis (Nakano et al., 1991;Tognoni el al., 1995) (see Fig. 3). This gene locus contains sequences coding for seven peptide synthetase
120
RAINER ZOCHER AND ULLRICH KELLER
domains distributed over a set of three large and one small transcriptionally coupled orfs designated s $ M , s$AB, s$AC and @AD, respectively (Cosmina et al., 1993; Fuma et al.,1993). Ser-to-Ala substitutions in the thioester formation motifs of the first four domains (i.e. all three motifs in s $ M and one in s$AB) of the cluster were introduced to see whether these had influence on surfactin production and competence development in vivo.Each substitution abolished surfactin production, while competence development remained unaffected (D’Souza et al., 1993). More detailed genetic analyses revealed that some unknown function of the Val-activating domain influences competence development. This will not be discussed here further. Further investigation of the products of mutated s.f genes showed that the corresponding enzymes were indeed lacking the expected functions such as thioester formation in Ser-to-Ala-substituted domains 2 and 3 of s $ M , which enabled assignment of EIa to the gene product of that or$ Further mutational analyses revealed that slfAB and srfAC encode enzymes Elb and E2, respectively (Vollenbroichet al., 1994). Interestingly, slfAD from its sequence apparently codes for a thioesterase like the gene product of grsT (Kratzschmar et al., 1989), which implicates the function of a thioesterase i n the terminatiodcyclization reactions of the biosynthetis of surfactin. Such an enzyme activity has not yet been characterized from surfactin-producingB. subtilis (Vollenbroich el al., 1994).
5.3. Bialaphos Bialaphos, a linear tripeptide consisting of two alanine (Ala) residues and an unusual amino acid called phosphinothricin (Pt) in the order Pt-Ala-Ala is synthesized by Streptornyces hygroscopicus and Streptornyces viridochromogenes (Fig. 19). Antibacterial activity of Pt is due to its release from the Pt-tripeptide through the action of intracellular peptidases after transport into bacterial cells (Diddens et al., 1976). The compound is active as an inhibitor of glutamine synthetases, which leads to cell death through glutamine starvation (Bayer et al., 1972). In plants, the same type of inhibition leads to accumulation of ammonium ions, which is toxic to the plant cell and therefore makes Pt a most effective herbicide. In addition, isolation of the bialaphos resistance gene from a bialaphosresistant Streptoniycrs strain has led to the construction of herbicide-resistant plants, removing the constraints i n using phosphinothricin tripeptide (Ptt) as a total herbicide. 5.3.1,
Isolation und Chamcterizatioiz of Peptide Synthetase Domain-related
orfs
Cloning and partial sequencing of the cluster of bialaphos biosynthesis genes has
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI CH3
CH3
O=b--OH
O=b-OH
121
1
CO-Ale-Ala Phosphinothricin (Pt)
Phosphlnothricyl-alanyl-alanine (Ptt, Bialaphos)
Figure 19 Structure of phosphinothricin and phosphinothricin-tripeptide.
identified many of the genes involved in the steps of Pt synthesis, and which are assembled in one large gene cluster in Streptomyces hygroscopicus and Streptomyces viridochrornogenes (Murakami ‘et al., 1986; Hara et al., 1991). By contrast, little is known about the process of non-ribosomal peptide-bond formation in bialaphos synthesis at both the biochemical and genetic level. Remarkably, genes that are involved in the steps of condensation of alanine with the immediate precursor of Pt, i.e. desmethyl-phosphinothricin (DMPT) or N-acetyl-DMPT, have been identified in disruption mutants and most of these mutations mapped to the bialaphos gene cluster (Hara et al., 1988). Sequencing of one particular clone mapping close to the bialaphos resistance gene bar revealed orfs ORFl and ORF2 with similarity to thioesterase genes in vertebrates (Raibaud et al., 1991). Wohlleben and co-workers (1992) isolated from the same region of the genome of S. viridochrornogenes an additional orf with similarity to various peptide synthetases. The deduced amino-acid sequence revealed a peptide synthetase domain encompassing all six highly conserved motifs necessary for adenylate and thioester formation. The gene termedphsA would encode a protein of 660 amino acids. From its size it would represent a single amino-acid activating domain. The presence of a single amino-acid activating enzyme and also single genes for thioesterases, indicates that Pt-tripeptide synthesis takes place on a multi-enzyme complex instead of one single polypeptide chain, as is the case of ACV synthesis (Raibaud et al., 1991). Consequently, one has to postulate additional orfs coding for the missing two activation domains because a total of three domains for the assembly of the Ptt-molecule has to be expected. These orfs could possibly be represented by the mutants isolated by Hara et al. (1 988). The cloning of these genes has not been reported yet.
5.3.2. PhsB is an Alanine-activating Peptide Synthetase Biochemical investigation on Ptt-peptide synthesis have shown the presence of an
122
RAINER ZOCHER AND ULLRICH KELLER
alanine-activating enzyme produced in coordinately elevated levels in mutants of
S. viridochromgenes, which had been previously selected for higher level Pt-tripeptide formation in a strain-improvement program (U. Keller and B. Riederer. manuscript submitted). The enzyme was purified to apparent homogeneity and identified by radioactive substrate labelling and autoradiography in SDS-PAGEgels. In the denatured form, it has an Mrof 147 OOO. Thus, it is clearly distinct from thephsA gene product, which would represent a polypeptide chain of Mr 80 OOO. The protein, which activates and binds alanine covalently as thioester, was designated PhsB. Like other peptide synthetases, it was shown to contain 4'-phosphopantetheine as cofactor. Attempts to detect a second alanine-activating enzyme from S. viridochmmogenes were unsuccessful. It cannot be ruled out that PhsB harbours two alanine-activating domains or is the degradation product of a still larger enzyme harbouring the complete set of two alanine-activatingdomains. Interestingly, PhsB also activates structurally related amino acids, such as leucine, valine and isoleucine, but not glutamate or Pt (U. Keller and B. Riederer, manuscript submitted). Leucine has been shown previously to be a component of a Pt-tripeptide-related protein called phosalacine from Kifasofosporiaphosalucime (Omura el al.. 1984). Phosalacine contains Pt, alanine and leucine in the order Pt-Ala-Leu. It cannot be excluded that Pt-Ala-Leu is synthesized by homologues of PhsA and PhsB under physiological conditions, where high leucine levels are present. Cloning of the gene encoding PhsB will enable us to see whether it contains one or two alanine-activating domains. The existence of PhsB as an alanine-activating enzyme also indicates that PhsA is most probably responsible for Pt activation. Meanwhile, biochemical work on Pt activation in S. viridochmmogenes has revealed the preFnce of an enzyme of molecular mass about 80 kDa, which activates acetyl-Pt and to a lesser extent Pt (U. Keller, unpublished). This enzyme is of the expected size of the phsA gene product. It will be interesting to test this enzyme with the substrate N-acetyl-DMPT to see whether this is an even better substrate than Pt or N-acetyl-Pt. These and the other data available on genes of Ptt synthetases and associated orfs characterize the Ptt biosynthesis system as typical for the organization of bacterial multi-enzyme peptide synthetases in which the various domains are spread over separate enzymes instead of one single enzyme polypeptide chain as in the eukaryotes.
6. FUTURE PROSPECTS OF PEPTIDE SYNTHETASE RESEARCH
Future developments in peptide synthesis research will aim at clarifying the mechanisms of initiation, elongation, epimerization and termination events on peptide synthetase. Parallel work will aim at the elucidation of the structures of peptide synthetase domains by biochemical techniques, such as cross-linking of
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
123
motifs, mutational analysis of reaction sites and by elucidation of crystal structures of various domains by X-ray crystallography. This mechanistical and structural information will help to understand the factors determining substrate specificity and to circumvent possible constraints in combining peptide synthetase domains into novel enzymes by recombinant DNA techniques, which will aid in the design of new bioactive compounds. The success of such manipulations has been demonstrated i n several examples of polyketide synthase systems (Hutchinson and Fuji, 1995; Tsoi and Khosla, 1995).
6.1. Domain Exchange in Thiol Template Peptide Synthesis Systems
Construction of peptide synthetase-containing activation domains with altered substrate specificity or containing activation domains from other peptide synthesizing systems should allow us to synthesize new products in vivo and in vitm. Such an approach has been realized recently (Stachelhaus et al., 1995) when domaincoding regions of bacterial and fungal origin were combined with each other in hybrid genes that encoded peptide synthetases producing peptides with modified amino-acid sequences. In fact, replacement of the Leu-activation domain in the slfAC gene in B. subtilis by the cys domain of ACVS of Penicillium chrysogenum led to the expression of the engineered gene, as revealed by the formation of the corresponding protein as well as in the formation in vivo of a novel surfactin molecule containing cysteine instead of leucine in the 7-position of the peptide lactone ring. This approach is promising for the future development of new and valuable compounds.
6.2. Combinatorial Approaches in Future Peptide Synthesis Development The need for new compounds will require extensive screening programs using high numbers of samples of both natural and synthetic origin. Combinatorial approaches in synthetic chemistry allow the high throughput screening of millions of compounds. It is desirable to introduce such combinatorial approaches into biosynthesis systems by creating domain assemblies, which provide the best possibility to create libraries of compounds. Attempts to combine domains or modules of different PKSII synthases for combinatorial purposes with the aim of establishing libraries have been reported by Khosla, Hopwood and their co-workers (reviewed in Tsoi and Khosla, 1995). In the case of peptide syntheti;e systems, such approaches could be realized soon, provided that the number of different domains is large enough to create a pool of many different enzymes and detect a useful compound. Peptide synthetase systems lack the various modification steps met in PKS systems,
124
RAINER ZOCHER AND ULLRICH KELLER
such as ketoreduction, dehydration and enoylreduction, the programming of which are responsible for the great diversity in the product spectrum of PKS systems. Much more promising for future developments in this field, therefore, is the possibility of improving the production and structures of existing valuable peptides, such as the p-lactams, immunosuppressants and anti-tumor agents by domain exchange and directed mutagenesis affecting substrate specificity.
ACKNOWLEDGEMENTS The authors wish to thank A. Haese for valuable discussion. The help of M. Krause, F. Schauwecker and W. Weckwerth in drawing the figures as well as M. Glinski and B. Gorhardt for correcting the manuscript is gratefully acknowledged. Work done in the authors’ laboratory has been supported by the Deutsche Forschungsgemeinschaft, the Fonds der Chemischen Industrie, the European Community and the Erwin Riesch Foundation.
REFERENCES Aharonowitz, Y,, Bergmeyer, J., Cantoral, J.M., Cohen, G., Demain, A.L., Fink, U., Kinghorn, J., Kleinkauf, H., McCabe, A., Palissa, H., Pfeifer, E., Schwecke, T., van Liempt, H., von Dohren, H., Wolfe, S. and Zhang, J. (1993) &(L-a-aminoadipyl)-Lcysteinyl-D-valine synthetases, the multi-enzyme integrating the four primary reactions in B-lactam biosynthesis, as a model peptide synthetase. Bio-Technology 11,807-810. Baldwin, J.E., Bird, J.W., Field. R.A., O’Callaghan. N.M. and Schofield, C.J. (1990) Isolation and partial charactensation of ACV synthetase from Cephalosporium acremonium and Streptomyces clavuligerus. J. Anlibiotics 43, 1055-1057. Baldwin, J.E.,Bird, J.W.,Field, R.A., O’Callaghan,T.N.M.,Schofield,C.J.andWillis,A.C. (1991) Isolation and partial characterisation of ACV synthetase from Cephulosporium acremoiiium and Slreptomyces clavuligerus: evidence for the presence of phosphopantothenate in ACV synthetase. J. Anlibiorics 44,241-248. Bayer, E., Gugel, K.H., Hagele, K., Hagenmaier, H., Jessipow, S., Konig, W.A. andZaner, H. (1972) Stoffwechselproduktevon Mikroorganismen. Phosphinothricinund Phosphinothricyl-Alanyl-Alanin. Helv Chim. Acta 55,224-239. Berg, T.L., Fr~holm,L.O. and Laland, S. (1965) The biosynthesis of gramicidin S in a cell-free system. Biochern. J. 96.43-52. Billich, A. and Zocher, R. (1 987a) N-Methyltransferase function of the multi-functional enzyme enniatin synthetase. Biochemistry 26, 8417-8423. Billich, A. and Zocher, R. (1987b) Enzymatic synthesis of cyclosponn A. J. Biol. Chem. 262,17258-1 7259. Billich, A. and Zocher, R. (1990) Formation of N-methylated peptide bonds in peptides and peptidols. In: Biochemislry of PepfineAntibiotics (H. Kleinkauf and H. von Dohren, eds), pp. 57-79. Walter de Gruyter, Berlin. Billich, A., Zocher, R., Kleinkauf, H., Braun, D.G., Lavanchy, D. and Hochkeppel, H.K.
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
125
(1987) Monoclonal antibodies to the multi-functional enniatin synthetase. Biol. Chem Hoppe-Seyler 368,521-529. Borel, J.F. (1986) Ciclosporin and its future. frog. Allergy 38.9-18. Borchert, S., Patil, S.S. and Marahiel, M.A. (1992) Identification of putative multifunctional peptide synthetase genes using highly conserved oligonucleotide sequences derived from known synthetases. FEMS Microbiol. Lett. 92,175-180. Coque, J.J.R., Martin, J.F., Calzada, J.G. and Liras, P. (1991) The cephamycin biosynthetic genes pcbAB, encoding a large multidomain peptide synthetase pcbC of Nocardia lactumduruns, are clustered together in an organization different from the same genes in Acremonium chrysogenum and Penicillium chrysogenum. Mol. Microbiol. 5,1125-1 133. Cosmina, P., Rodriguez, F., de Ferra, F., Perego, M., Venema, C. and van Sinderen, D. (1993) Sequence and analysis of the genetic locus responsible for surfactin synthesis in Bacillus subtilis. Mol. Microbiol. 8, 82 1-83 1. De Cdcy-Lagard, V., Marlikre, P. and Saurin, W. (1995) Multi-enzymatic non ribosomal peptide biosynthesis: identification of the functional domains catalysing peptide elongation and epinierization. C.R. Acad. Sci. furis, Sci. Ke/Lve Sci. 318,927-936. de Wet, J.R., Wood, K.V..DeLuca, M., Helinski, D.R. and Subramani, S. (1987) Firefly luciferase gene: structure and expression in mammalian cells. Mol. Cell. Biol. 7,725-737. Diddens, H., Zlhner, H., Kraas, E., Gohring, W. and Jung, C. (1976) On the transport of tripeptide antibiotic in bacteria. Eu,: J. Biochem. 66, 11-23. Diez, B., Gutierrez, S., Barredo, J.L., van Solingen, P., van der Voort, L.H.M. and Martin, J.F. (1990) The cluster of penicillin biosynthetic genes. Identification and characterisation of the pcbAB gene encoding the a-aminoadipyl-cysteinyl-valinesynthetase and linkage to thepchC and penDE genes. J. Biol. Chem. 265,16358-16365. Dreyfuss, M.M., Schreyer, M., Tscherter, H. and Wenger, R. (1988) European Pat. Appl. 0296 123 A2. D’Souza, C., Nakano, M.M., Coibell, N. and Zuber, P. (1993) Aminoacylation site mutations in amino-acid-activating domains of surfactin synthetase: effects on surfactin production and competence development in Bacillus sublilis. J. Bacteriol. 175,3502-35 10. Fuma, S., Fujishima, Y., Corbell, N., D’Souza, C., Nakano, M.M., Zuber, P. and Yamane, K. (1993) Nucleotide sequence of S’-portion of srfA that contains the region required for competence establishment in Bacillus subtilis. Nucl. Acids Res. 21,93-97. Gevers, W., Kleinkauf, H. and Lipmann, F, (1969) Peptidyl transfers in gramicidin S biosynthesis from enzyme-bound thioester intermediates. f roc. Nut1 Acad. Sci. USA 63, 1335-1 345. Glund, K., Schlumbohni, W., Bapat, M. and Keller, U. (1990) Biosynthesis of quinoxaline antibiotics: Purification and characterisation of the quinoxaline-2-carboxylic acid activating enzyme from Sfrepfomycestriosfinicus. Biochemistry 29,3522-3527. Gocht, M. and Marahiel, M.A. (1994) Analysis of core sequences in the D-Phe activating domain of the multi-functiona1peplide synthetase TycA by site-directed mutagenesis. J. Bacteriol. 176, 2654-2662. Grove, J.F. and Pople, M. (1980) The insecticidal activity of beauvericin and the enniatin complex. Mycopafhologiu 70, 103-105. Guest, J.R. (1987) Functional implications of structural homologies between chloramphenicol acetyltransferases and dihydrolipoamide acetyltransferases. FEMS Microbiol. Len. 44,417-422. Gutierrez, S., Diez, B., Montenegro, E. and Martin, J.F. (1991) Characterisation of the Cephulosporiurn acreinortiurn pcbAB gene encoding a a-aminoadipyl-cysteinyl-valine synthetase, a large niultidomain peptide synthetase: linkage to thepcbC gene as a cluster of early cephalosporin biosynthetic genes and evidence of multiple functional domains. J. Bacteriol. 173,2354-2365.
126
RAINER ZOCHER AND ULLRICH KELLER
Haese, A., Schubert, M., Hemnann, M. and Zocher, R. (1993) Molecular characterisation of the enniatin synthetase gene encoding a multi-functional enzyme catalysing N-methyldepsipetide formation in Fusariurn scirpi. Mol. Microbiol. 7 , 905-914. Haese, A., Pieper, R., von Ostrowski, T. and Zocher, R. (1994) Bacterial expression of catalytical active fragments of the multi-functional enzyme enniatin synthetase. J. Mol. Biol. 243, 116-122. Hara, O.,Anzai, H., Imai, S., Kumada, Y.,Murakami, T., Itoh, R., Takano, E., Satoh, A. and Nagaoka, K. (1 988) The bialaphos biosynthetic genes of Streplomyces hygroscopicus: cloning and analysis of genes involved in the alanylation step. J. Anribiotics41,538-547. Hara, O., Murakami, T., Imai, S., Anzai. H., Itoh, R., Kumada, Y., Takano, E., Satoh, E., Atsuyuki, S., Nagaka, K. and Thompson, C.J. (1991) The bialaphos biosynthetic genes of Streptomyces viridochrornogenes: cloning, heterospecific expression and comparison with genes of Streptomyces hygroscopicus. J. Gen. Microbiol. 137,35 1-359. Hoffmann, K., Schneider-Schener, E., Kleinkauf, H. and Zocher, R. (1994)Purification and characterisation of eukaryotic alanine racemase acting as a key enzyme in cyclosporin biosynthesis. J. Riol. Chem. 269, 12710-12714. Hofmann, A., Ott, H., Griot, R., Stadler, P.A. and Frey, A.J. (1963) Die Synthese und Stereochemie des Ergotamins. Hel. Chim. Acta 46,2306-2328. Hopwood, D.A. and Sherman, D.H. (1990) Molecular genetics of polyketides and its comparison to fatty acid biosynthesis. Ann. Rev. Genet. 2 4 , 3 7 4 6 . Hutchinson, C.R. and Fuji, I. (1995) Polyketide synthase gene manipulation: structurefunction approach in engineering novel antibiotics. Annu. Rev. Microbiol. 49: 201-238. Jensen, S.E., Wong, A,, Rollins, M.J. and Westlake, D.W.S. (1990) Purification and partial synthetase from Strepcharacterisation of 6-(L-or-aminoadipyl)-L-cysteinyl-D-vahe tomyces clavuligerus. J. Bacterial. 172,7269-727 1 . Kahan, B.D. (ed.) (1 984) Cyclosporin: Biological Activity and Clinical Applications. Grune and Straton Inc., Orlando, FL. Katz, E. (1968) Actinomycin. In: Antibiotics I1 (D. Gottlieb and PD. Shaw, eds), pp, 276-341. Springer-Verlag, New York. Katz, L. and Donadio, S . (1993) Polyketide synthesis: prospects for hybrid antibiotics. Annu. Rev. Microbiol. 47,875-912. Keller, U. (1984) Acyl pentapeptide lactone synthesis in actinomycin-producing streptomycetes by feeding with structural analogues of 4-methyl-3-hydroxyanthranilicacid (4-MHA) J. Biol. Chem. 259,8226-823 1. Keller, U. (1987) Actinomycin synthetases: multi-functional enzymes responsible for the synthesis of the peptide chains of actinomycin. J . B i d . Chem. 262,5852-5856. Keller, U. (1995) Peptidolactones. In: Generics and Biochemisfry ofAntibioric Production (L.C. Vining, ed.), pp. 71-94, Heinemann-Butterworths, Toronto, New York. Keller, U. and Schlumbohm, W. (1992) Purification and characterisation of actinomycin synthetase I, a 4-methyl-3-hydroxyanthranilic acid: AMP ligase from Strepromyces chrysomallus. J. B i d . Chem. 267, 11745-1 1752. Keller, U., Kleinkauf, H. and Zocher, R. (1 984a) 4-Methyl-3-hydroxyanthranilic acid (4-MHA) activating enzyme from actinomycin producing Streptomyces chrysomallus. Biochemistry 23, 1479-1484. Keller, u., Zocher, R., Krengel, U. and Kleinkauf, H. (1984b) D-Lysergic acid activating enzyme from the ergot fungus Claviceps purpurea. Biochem. J. 218,857-862. Keller, U., Han, M. and Stoffler-Meilicke, M. (1988)D-Lysergic acid activation and cell-free synthesis of D-lysergyl peptides in enzyme fractions from the ergot fungus Claviceps purpurea. Biochemistry 27, 61644170. Kleinkauf, H. and von Dohren, H. (1987) Biosynthesis of peptide antibiotics. Ann. Rev. Microbiol. 41,259-289.
THlOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
127
Kleinkauf, H. and von Dohren, H. (1990) Nonribosomal biosynthesis of peptide antibiotics. Eur J. Biochrm. 192, 1-5. Kluge, B., Vater, J., Salnikow, J. and Eckart, K. (1988) Studies on the biosynthesis of surfactin, a lipopeptide antibiotic from Bacillus subtilis ATCC 21 332. FEBS Lett. 231, 107-110. Knobloch, K.-H. and Hahlbrock, K. (1975) Isoenzymes of p-coumarate: CoA ligase from cell suspension cultures of glycine max. Eur: J. Biochem. 52,311-320. Kobel, H. and Traber, R. (1982) Directed biosynthesis of cyclosporins. EUKAppl. Microbiol. Biotechnol. 14,237-240. Kobel, H. and Sanglier, J.-J. (1986) Ergot alkaloids. In: Biotechnology (H.-J. Rehm and G. Reed, eds), Vol. 4, pp. 569-609. VCH Verlagsgesellschaft, Weinheim. Kriitzschmar, J., Krause, M. and Marahiel., M.A. (1989) Gramicidin S biosynthesis operon containing the structural genes grsA and grsB has an open reading frame encoding a protein homologous to fatty acid thioesterases. J. Bucteriul. 162,5422-5429. Krause, M., Marahiel, M.A., von Dohren, H. and Kleinkauf, H. (1985) Molecular cloning of an ornithine-activating fragment of the gramicidin S synthetase I1 gene from Bacillus subrilis and its expression i n Escherichia coli. J. Bucteriol. 162, 1120-1125. Kurahashi, K. (1 974) Biosynthesis of small peptides. Annu. Rev. Biochem. 43,445-459. Lawen, A,, Traber, R., Geyl, D., Zocher, R. and Kleinkauf, H. (1989) Cell-free biosynthesis of new cyclospodns. J. Antibiotics 42, 1283-1289. Lawen, A. and Zocher, R. (1990) Cyclosporin synthetase - the most complex peptide synthesising multi-enzyme polypeptide so far described. J. Biol. Chem. 265, 1135511360. Lawen, A,, Traber. R. and Geyl, D. (1991) In virro biosynthesis of [Thr2, Leu', D-Hiv', Leu 101-cyclosporin, cyclosporin-related peptolide, with immunosuppressive activity by a multi-enzyme polypeptide. J. Biol. Chem. 266, 15567-15570. Lawen, A. and Traber, R. (1993) Substrate specificities of cyclosporin synthetase and peptolide SDL 214-103 synthetase. J. Biol. Chem. 268,20452-20465. Lawen, A,, Dittmann, J., Wenger, R.M. and Kleinkauf, H. (1994) Mechanism of cyclosporin A biosynthesis. J. Biol. Chem. 269,2841-2846. Lipmann, F. (1954) In: The Mechanism ofEnzyrne Action (W.D.M. Elroy and B. Glass, eds), p. 599. Johns Hopkins University Press, Baltimore. Lipmann, F. (1971) Attempts to map a process evolution of peptide biosynthesis. Science 173,1435-1441. Lipmann, E (1973) Nonribosomal polypeptide synthesis on polyenzyme templates. Acc. Chem. Res. 6,361-367. Lipmann, F. (1980) Bacterial production of antibiotic polypeptides by thiol-linked synthesis on protein templates. Adv. Micmbiol. Physiol. 21,227-266. Lozoya, E., Hoffinann, H. and Douglas, C. (1988) Primary structure and catalytic properties of isoenzymes encoded by the two 4-coumarate: CoA ligase genes in parsley. Eur J. Biochem. 176,661-667. Lynen, F. (1972) Enzyme systems for fatty acid synthesis. In: Current Trends in the Biochemistry of Lipids (J. Ganguly and R.M.S. Smellie, eds), pp. 5-26. Academic Press, New York. MacCabe, A.P., van Liempt, I]., Palissa, H., Unkles, S.E., Riach, M.B.R., Pfeifer, E., von Dohren, H. and Kinghorn, J.R. (1991a) &(L-a-aminoadipyl)-L-cysteinyI-D-vdine synthetase from Aspergillus nidulans. Molecular characterisation of the ucvA gene encoding the first enzyme of the penicillin biosynthetic pathway. J. Biol. Chern. 266.12646-12654. MacCabe, A.P., Riach, M.B.R. and Kinghorn, J.R. (1991b) Identification and expression of the ACV synthetase gene. J. Biotech. 17,91-97.
128
RAINER ZOCHER AND ULLRICH KELLER
Maier, W., Erge, D. and Groger, D. (1983) Further studies on cell-free biosynthesis of ergotamine in Clavicepspurpurea. FEMS Microbiol. Lett. 20,233-236. Marahiel, M. (1992) Multidomain enzymes involved in peptide synthesis. FEBS Lell. 307, 4043. Martin, J.F. and Gutierrez, S. (1995) Genes for P-lactam antibiotic biosynthesis. Antonie van Leeuwenhoek. Int. J. Microbiol. 67, 181-200. Meienhofer, J. and Atherton, E. (1973) Structure-activity relationships in the actinomycins. Adv. Appl. Microbiol. 16,203-300. Meister. A. (1988) Glutathione metabolism and its selective modification. J. Biol. Chem. 263,17205-17208. Menkhaus, M., Ullrich, C., Kluge, B., Vater, J., Vollenbroich, D. and Kamp, R.M. (1993) Structural and functional organization of the surfactin synthetase multi-enzyme system. J. Biol. Chem. 268,7678-7684. Murakami, T., Anzai, H.. Imai, S., Satoh, A., Nagaoka, K. and Thompson, C.J. (1986) The bialaphos biosynthetic genes of Streptomyces hygroscopicus: molecular cloning and characterisation of the gene cluster. Mol. Cen. Genef.205.42-50. Nakano, M.M., Magnuson, R., Myers, A., Curry, J., Grossmann, A.D. and Zuber, P. (1991) S ~ A is an operon required for surfactin production, competence development and efficient sporulation in Bacillus subtilis. J. Bacteriol. 173, 1770-1778. Okumura, Y.(1983) Peptidolactones. In: Biochemistry and Genetic Regulation ofCommercially Important Antibiotics (L.C. Vining, ed.), pp. 147-178. Addison-Wesley Publishing Company, Reading, MA. Omura, S., Hinotozowa, K., Imamura, N. and Murata. M. (1984) The structure of phosalacine, a new herbicidal antibiotic containing phosphinothricin. J. Antibiotics 37, 939-940. Pavela-Vrancic, M., Pfeifer, E., van Liempt, H., Schafer, H.-J., von Dohren, H. and Kleinkauf, H. (1994a) ATP binding in peptide synthetases: determination of contact sites of the adenine moiety by photoaffinity labelling of tyrocidine synthetase I with 2-azidoadenosine triphosphate. Biochemistry 33,62764283. Pavela-Vrancic, M., Pfeifer, E., Schroder, W.. von Dohren. H. and Kleinkauf, H. (1994b) Identification of Ihe ATP-binding site in tyrocidine synthetase I by selective modification with fluorescein 5’-isothiocyanate. J. Biol. Chem. 269, 14962-14966. Peeters, H., Zocher, R., Madry, N., Oelrichs, P.B., Kleinkauf, H. and Kraepelin, G. (1983) Cell-free synthesis of the depsipeptide beauvericin. J. Antibiotics 12, 1983-1987. Pfeifer, E., Pavela-Vrancic, M., von Dohren, H. and Kleinkauf, H. (1995) Characterisation of tyrocidine synthetase I (TYI): requirement of posttranslational modification for peptide biosynthesis. Biochemistry 34,7450-7459. Pieper, R., Kleinkauf, H. and Zocher, R. (1992) Enniatin synthetases from different Fusaria exhibiting distinct amino-acid specificities. J. Antibiotics 45, 1273-1277. Pieper, R., Haese. A., Schroder, W. and Zocher, R. (1995) Arrangement of catalytic sites in the multi-functional enzyme enniatin synthetase. Eul: J. Biochem. 230, 119-126. Plattner, P.A., Nager, U. and Boller, A. (1948) Welkstoffe und Antibiotika. 7. Mitteilung iiber die Isolierung neuartiger Antibiotika aus Fusurien. Helv. Chirn. Acta 31,594402. Raibaud, A., Zalacain, M., Holt, T.G., Tizard, R. and Thompson, C.J. (1991) Nucleotide sequence analysis reveals linked N-acetyl hydrolase, thioesterase, transport and regulatory genes encoded by the bialaphos biosynthetic gene cluster of Streptomyces hygroscopicus. J. Bacteriol. 173,4454-4463. Roberts, G.A., Staunton, J. and Leadlay, P. (1993) Heterologous expression in Escherichia coli of an intact multi-enzyme component of the erythromycin-producing polyketide synthase. Eur: J. Biochem. 214,305-311.
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
129
Roskoski. Jr., R., Kleinkauf, H.. Gevers, W. and Lipmann, E (1970) Isolation of enzymebound peptide intermediates in tyrocidine synthesis. Biochemistry 9,48464851. Schimmel, P. and Sol], D. (1979) Aminoacyl-tRNA synthetases: general features and recognition of transfer-RNAs. Ann. Rev. Biochem. 48,601-648. Schindler, R. (ed.) (1 985) Cyclosporin in Autoimmune Diseuses. Springer Verlag, Berlin. Schlumbohm, W. and Keller, U. (1990) Chromophore activating enzyme involved in the biosynthesis of the mikamycin B antibiotic etamycin from Streptumyces griseoviridus. J. Biol. Chem. 265,2156-2161. Schlumbohm, W., Stein, T., Ullrich, C., Vater, J., Krause, M., Marahiel, M.A.. Kruft, V. and Wittmann-Liebold, B. (1991) An active serine is involved in covalent substrate aminoacid binding at each reaction centre of gramicidin S synthetase. J. Biol. Chem. 266, 23135-23 141. Schwecke, T., Aharonowitz, Y.,Palissa, H., von Diihren, H., Kleinkauf, El. and van Liempt, H. (1992) Enzymatic characterisation of the multi-functional enzyme 6-(L-a-aminoadipyl)-L-cysteinyl-D-vahe synthetase from Streptomyces clavuligerus. Eu,: J. Biochem. 205,687-694. Scott-Craig,J.S., Panaccione, D.G., Pocard, J.-A. and Walton, J.D. (1992) The cyclic peptide synthetase catalysing 1IC-toxin production in the filamentous fungus Cuchliubolus curboneurn is encoded by a 15.7-kilobase open reading frame. J. Biol. Chem. 267, 26044-26049. Shiau, C.Y., Baldwin, J.E., Byford, M.F., Sobey, W.J. and Schofield, C.J. (1995a) ~-(L-(wAminoadipy1)-L-cysteinyl-D-valine synthetase: the order of peptide-bond formation and timing of the epinienzation reaction. FEBS Len. 358.97-100. Shiau. C.Y.,Baldwin, J.E., Byford, M.F. and Schofield, C.J. (1995b) &(L-a-Amino-adipyl)L-cysteinyl-D-valinesynthetase: isolation of L-cysteinyl-D-valine, a “shunt” product and implications for the order of peptide-bond formation. FEBS Lett. 373,303-306. Simon-Lavoine, N. and Forgeot, M. (1979) German Patent 2851629. Smith, D.J., Burnham, M.K.R., Bull, J.H., Hodgson, J.E., Ward, J.M., Browne, J., Barton, B., Alison, J.E. and Turner, G. (1990a) p-lactam antibiotic biosynthetic genes have been conserved in clusters in prokqotes and eukaryotes. E M 3 0 J. 9,741-747. Smith, D.J., Alison, J.E. and Turner, G. (1990b) The multi-functional peptide synthetase performing the first step of penicillin biosynthesis in Penicillium chrysogenum is a 421 073 dalton protein similar to Bacillus hrevis peptide antibiotic synthetases. EMBO J . 9, 2743-2750. Stachelhaus, T. and Marahiel, M.A. (199%) Modular structure of genes encoding multifunctional peptide synthetases required for non-ribosomal peptide synthesis. FEMS Microbiol. Lett. 125,3-14. Stachelhaus, T. and Marahiel, M.A. (1995b) Modular structure of peptide synthetases revealed by dissection of the multi-functional enzyme GrsA. J. Bid. Chem. 270, 6 163-6 169. Stachelhaus, T., Schneider, A. and Mal-ahiel, M.A. (1995) Rational design of peptide antibiotics by targeted replacement of bacterial and fungal domains. Science 269,69-72. Stadler, P.A. (1982) New results of ergot alkaloid research. Planfu Medicu 46, 131-144. Stein, T., Vater, J., Kmft, V., Wittmann-Liebold, B., Franke, P., Panico, M., Dowell, R.M. and Morris, H.R. (1994) Detection of 4‘-phosphopantetheine at the thioester binding site for L-valine of gramicidin S synthetase 11. FEBS Lett. 340,3944. Stein, T., Kluge, B., Vater, J., Franke, P., Otto, A. and Wittmann-Liebold, B. (1995) Gramicidin S synthetase I (pheny lalanine racemase), a prototype of amino acid racemases containing the cofactor 4’-phosphopantetheine. Biochemistry 34,46334662. Stindl, A. and Keller, U. (1993) The initiation of peptide formation in the biosynthesis of actinomycin. 1.Biol. Chem. 268, 10612-10620.
130
RAINER ZOCHER AND ULLRICH KELLER
Stindl, A. and Keller, U. (1994) Epimerization of the D-vdine portion in the peptide chain of actinomycin. Biochemistry 33,9358-9364. Tobin,M.B., Kovacevic, S., Madduri, K., Hoskins. J.A., Skatrud,P.L.,Vining,L.C., Stuttard, C. and Miller, J.R. (1990) Localisation of the lysine E-aminotransferase (lut) and 6-(L-a-aminoadipyl)-L-cysteineyl-D-va~ine synthetase @cbAB) genes from Streptomyces clavuligerus. Production of LAT activity in Escherichia coli. J. M e d Chern. 33, 2321-2323. Tognoni, A,, Franchi, E., Magistrelli, C., Columbo, E., Cosmina, P. and Grandi, G. (1995) A putative new peptide synthetase operon in Bacillus subrilis: partial characterisation. Microbiology 141,645-648. Tohika, K., Hori, K., Kurotsu, T., Kanda, M. and Saito, Y. (1993) Effect of single base substitutions at glycine-870 codon of gramicidin S synthetase 2 gene on proline activation. J. Biochein. 114, 522-527. Tomoda, H.. Huang, X.-H., Cao, J., Nishida, H., Nagao, R., Okuda, S., Tanaka, H. and Omura, S. (1992) Inhibition of acyl-CoA: cholesterol acyltransferase activity by cyclodepsipeptide antibiotics. J. Antibiotics 45, 1626-1632. Traber, R., Hofmann H., Loosli, H.R., Ponelle, M. and von Wartburg, A. (1987) Neue Cyclosporine BUS Tolypocladiunz irflufum. Die Cyclosporine K-2. Helv. Chim. Actu 70, 13-36. Traber, R.. Hofmann, H. and Kobe], H. (1989) Cyclosporins-new analogues by precursor directed biosynthesis. J. Antibiotics 42.59 1-597. Tsoi, C.J. and Khosla, C. (1995)Combinatonal biosynthesis of “unnatural” natural products: the polyketide example. Chern. and Bid. 2, 355-362. Turgay, K. and Marahiel, M.A. (1 994) Ageneral approach for identifying and cloning peptide synthetase genes. Peptide Res. 7,238-241. Turgay, K., Krause, M. and Marahiel M.A. (1992) Four homologous domains in the primary structure of GrsB are related to domains in a superfamily of adenylate-forming enzymes. Mol. Microbiol. 6,529-546. Ullrich, C., Kluge, B., Palacz, Z. and Vater, J . (1991) Cell-free biosynthesis of surfactin, a cyclic lipopeptide produced by Bacillus srrbtilis. Biochemistry 30, 65036508. Van Liempt, H., von Dohren, H. and Kleinkauf, H. (1989) &-(L-a-aminoadipyl)-L-cysteinylD-valine synthetase from Aspergillus nidulai?~.J. Biol. Chern. 264, 3680-3684. Vater, J. (1989) Lipopeptides, an interesting class of microbial secondary metabolites, In: Biological Active Molecules ( U P Schlunegger, ed.), pp. 27-38. Springer-Verlag, Berlin. Vollenbroich, D., Mehta, N., Zuber, P.,Vater, J. and Kamp, R.M. (1994) Analysis of surfactin synthetase subunits in sfrA mutants of Bacillus subtilis OKB105. J. Bacteriol. 176, 395-400. Walker, J.E., Saraste, M., Runswick, M.J. and Gay, N.J. (1982) Distantly related sequences in the a-and P-subunits of ATPsynthase, myosin and other ATP-requiring enzymes and a common nucleotide binding fold. EMBO J . 1,945-951. Walton, J.D. (1990) Peptide phytotoxins from plant pathogenic fungi. In: Biochemistry of Peptide Antibiotics (H. Kleinkauf and H. von Dohren, eds), pp. 179-203. W. de Gruyter, Berlin. Weber, G., Schorgendorfer, K., Schneider-Scherzer, E. and Leitner, E. (1994) The peptide synthetase catalysing cyclosporin production in Tulypocladium niveum is encoded by a giant 45.8-kilobase open reading frame. Cum Genet. 26, 120-125. Weckermann, R., FuhrbaR, R. and Marahiel, M.A. (1988) Complete nucleotide sequence of the tycA gene coding the tyrocidine synthetase I from Bacillus brevis. Nucl. Acids Res. 16,11841. Weigel, B.J., Burgett, S.G.,Chen, V.J., Skatrud,P.L..Frolik, C.A.,Queener, S.W.andhgolja,
THIOL TEMPLATE PEPTIDE SYNTHESIS SYSTEMS IN BACTERIA AND FUNGI
131
T.D. (1988) Cloning and expression in Esclzerichiu coli of isopenicillin N synthase. J. Bucteriol. 170, 3817-3826. Wipf, H.K., Pioda, L.A.R., Stefanac, Z. and Simon, W. (1968) Komplexe yon Enniatinen und anderen Antibiotika niit Alkalimetallionen. Helv. Chim. Actu 51,377-381. Wohlleben, W., Alijah. R., Dorendorf, J., Hillemann, D., Nussbaumer, B. and Pelzer, S. (1992) Identification and characterisation of phosphinothricin-tripeptide biosynthetic genes in Streplomyces viridochromogenes.Gene 115, 127-132. Zocher, R., Keller, U. and Kleinkauf, H. (1982) Enniatin synthetase, a novel type of multi-functional enzyme catalysing depsipeptide synthesis in Fusan'um oxysporum. Biochemislry 21,4348. Zocher, R., Keller, U. and Kleinkauf, H. (1983) Mechanism of depsipeptide formation catalysed by enniatin synthetase. Biochern. Biophys. Res. Cornrnun. 110,292-299. Zocher, R., Madry, N., Peeters. H. and Kleinkauf, H. (1984) Biosynthesis of cyclosponn A. Phytochernistry (Oxb) 23, 549-55 1 . Zocher, R., Nihira, T., Paul, E., Madry, N., Peeters, H., Kleinkauf, H. and Keller, U.(1986) Biosynthesis of cyclosporin A: Partial purification and properties of a multi-functional enzyme from Tolyoclndium injlnrrim. Biochemistry 25, 550-553. Zuber, P., Nakano, M.M. and Malahiel, M.A. (1993) Peptide antibiotics, In: Bacillus subtilis and Other Cmn-posi/ive Bacrerio (A.L. Sonenshein, J.A. Hoch and R. Losick, eds), pp. 897-91 6. Anierican Society of Microbiology, Washington, DC.
This Page Intentionally Left Blank
Microbial Dehalogenation of Halogenated Alkanoic Acids, Alcohols and Alkanes J. Howard Slater', Alan T. Bull' and David J. Hardman3 1
Molecular Ecology Research Unit, School of Pure and Applied Biology, University of Wales, PO Box 915,Cardiff CFI 3T4 UK 2 Research School of Biosciences, University of Kent at Canterbury, Canterbury, Kent CT2 7NJ, UK 3 Cadbury Herne Lrd, Research and Development Centre, Canterbury, Kent CT2 7PD, UK
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Dehalogenation of halogenated alkanoic acids . . . . . . . . 2.1. 2-Haloalkanoic acid hydrolytic dehalogenases . . . . . . 2.2. Regulation of dehalogenase synthesis . . . . . . . . . . 2.3. Genetic organization of 2 H M hydrolytic dehalogenases 3. Dehalogenation of halogenated alcohols . . . . . . . . . . . 3.1. A pathway involving the formation of H M s . . . . . . . 3.2. Haloalcohol hydrolytic dehalogenases . . . . . . . . . . 3.3. Other systems for the transformation of haloalcohols . . 4. Dehalogenation of halogenated alkanes . , . . . . . . . . . . 4.1. Haloalkane hydrolytic dehalogenases . . . . . . . . . . 4.2. Oxygenase-type haloalkane dehalogenases . . . . . . . 4.3. Cofactor-dependent dehalogenases . . . . . . . . . . . 5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . References , . , . , . . . . . . . . . . . . . . . . . . . . .
... . .
.. .. .. . . . . . . . . . .
. . . .
. . ,
.
.
.
. . . . . .. . . . . . . .. . . . . . . . . . . .. . . .. . ... . . .. . . .. .... . . . . . . . . . . . .
..
. . . 134 . . . . . 135 . . . . . 136 . . . . . 146
. ...
149
. . .. . . . . .. .. . . . . . . .. . . .. . , .. . . .. .... .. ..
152 152 159 159 160 164 165 166 166
~
. 151
. .
.
. . .
. . .
Abbreviations: 1BP, 1-brornopropanol; CA, chloroacetone; ZCE, 2-chloroethanol; lCP, 1 -chloropropanol; 2CPD, 2-chloro-l,3-propandiol,2-chloropropan1,301; 3CPD, 3-chloro-1,2-propandiol,3-chloropropan-l,2-01; 13DBP, 1,3-dibromopropanol; DCA, dichloroacetic acid; 13DCA, 1,3-dichIoroacetone;DCM, dichlorornethane; 13DCP, 1,3-dichloro-2-propanol,1,3- dichloropropan-2-01; 23DCP, 2,3-dichloro-I-propanol, 2,3-dichloropropan-1-01; 22DCPA, 2,2- dichloropropionic acid; DMSO, diinethylsulfoxide; 24DNP, 2,4-dinitrophenol; ECH, epichlorohydrin; GDL, glycidol; GSH, glutathione; HAA, halogenated alkanoic acid; 2HAA. 2-halogenated alkanoic acid; MBA, monobromoacetic acid; 2MBPA, ADVANCES IN MICROBIALPHYSIOLOGY VOL 38 ISBN 0-12-027738-7
Copyright Q 1997 Academic Press Limited All rights of reproduction in any form reserved
134
J. HOWARD SLATER eta/.
2-monobromopropionic acid; MCA, monochloroacetic acid: 2MCPA, 2-monochloropropionic acid; MFA, monofluoroacetic acid; NEM, N-ethylmaleimide; PAGE, polyacrylamide gel electrophoresis; PMS, p-mercunbenzoate sulfonate; SDS, sodium dodecyl sulfate; TOL, toluene.
1. INTRODUCTION
About one hundred years ago, investigations began into the effect of halogenated compounds on the physiology and biochemistry of microbes. W. J. Penfold found that the growth of Bacterium coli (Escherich) and Bacterium lactis aerogenes was inhibited by chlorinated and brominated compounds, such as monochloroacetate (MCA), 2-monobromopropionic acid (2MBPA) and monochlorohydrin, and showed that the pattern of sugar fermentations was altered. Growth in the presence of these compounds led to the selection of mutants resistant to the toxic effects of halogenated compounds (Penfold, 1913). It was established that halogenated analogues of intermediary metabolites were toxic because they inhibited key reactions of central metabolism. For example, fluoroacetate (MFA) inhibited the tricarboxylic acid cycle (TCA) because the lethal synthesis of fluorocitrate generated by the activity of the citrate-condensing enzyme inhibited the next TCA cycle enzyme, aconitase (Peters, 1952). The removal of halogens, particulzrly fluorine and chlorine, was identified as one mechanism to relieve the inhibitory effects of these compounds and, moreover, provide novel carbon and energy sources for growth. Den Dooren de Jong (1926) showed that bromopropionate and bromosuccinate were used as the sole growth substrates by some bacteria. Subsequently many halogenated compounds have been shown to be degraded by a variety of microorganisms (Hardman, 1991; Janssen et al., 1994; Slater, 1994). However, not all halogenated organic compounds can be degraded and there are many that are resistant to microbial attack. In general the greater the number of halogens per molecule, the more difficult it is to isolate microbes that dehalogenate and grow on the compound (Commandeur and Parsons, 1994). This fact is of considerable environmental importance since many man-made (xenobiotic) compounds released into the biosphere during the last 50 years are multi-halogenated: indeed, their usefulness as pesticides, dielectrics, flame retardants, preservatives or whatever, depends on their chemical and biochemical inertness. More recently, the application of dehalogenating enzymes as catalysts for the resolution of racemic mixtures and the production of chiral intermediates in chemical syntheses has achieved significance (Taylor, 1985,1988; Hasan etal., 1991; Kasai etal., 1992). How biological systems, in particular microbes, handle halogenated compounds is an important and intriguing topic. Knowledge of the mechanisms of dehalogenation is not only intrinsically important for basic principles of biochemistry, but is
MICROBIAL DEHALOGENATION
135
crucial in delineating what products and processes are environmentally and industrially acceptable. This review is concerned with the catabolism of halogenated aliphatic compounds in the series: alkanoic acids, alcohols and alkanes. Microorganismsremove halogens from halogenated aliphatic compounds by the activity of enzymes generally called dehalogenases. Microbes that produce dehalogenases are widely distributed in nature, apparently having evolved to degrade naturally-occurring halogenated compounds in order either to exploit them as carbon sources for growth, or as a means of protection against the toxicity of these compounds (Fowden, 1968; Slater, 1994).
2. DEHALOGENATION OF HALOGENATED ALKANOIC ACIDS The discovery of dehalogenases was the result of the pioneering work of Jensen who isolated bacteria and fungi that grew on halogenated alkanoic acids (HAAs) and who was responsible for introducing the term dehalogenase (Jensen, 1951, 1957, 1959, 1960, 1963). He was the first to assay dehalogenases in cell-free systems and it was this work that promoted almost all subsequent studies, which have concentrated on HAAs substituted in the C2 position (2HAAs). Some microbes can grow on 3-halogenated alkanoic acids, but the biochemistry and enzymology is far from clear (Bollag and Alexander, 1971; Hughes, 1988). Interest in 2HAA biodegradation was stimulated by the introduction of the herbicide Dalapon (2,2-dichloropropionic acid; 22DCPA) and the isolation of many soil microorganisms that grew on 22DCPA as the sole carbon and energy source, thus explaining the rapid removal of Dalapon from soil (Magee andcolmer, 1959; Hirsch and Alexander, 1960; MacGregor, 1963; Kearney et al., 1964,1965; Kearney, 1966; Burge, 1969; Foy, 1975; Senior et al., 1976). Degradation of the highly toxic MFA by soil pseudomonads provided another stimulus for the characterization of new dehalogenases (Goldman, 1965; Kelly, 1965; Tonomura et al., 1965; Goldman andMilne, 1966; Goldman etal., 1968;Lien etal., 1979). Most research has focused on aerobic microbes. 2HAA dehalogenases calalyse the removal of halogens from organic compounds thereby forming simple compounds which are either intermediates of central metabolism or can be converted to intermediary metabolites. The general hydrolytic mechanism for mono-substituted alkanoic acids is as follows, where X represents the halogen atom: R - (CHX) - COOH + H20 + R - (CHOH) - COOH + H+ + Xand for di-substituted alkanoic acids: R - (CX2) - COOH + H20 R - (CO) - COOH + 2H+ + 2XGenerally 2HAA dehalogenases are inducible enzymes with low affinities for
136
J. HOWARD SLATER et
a/.
substrates (K, values in the millimolar range) and pH optima around pH 9.0. Typically these enzymes have broad substrate specificities and do not dehalogenate acids substituted other than at the C2 position. The stability of the carbon-halogen bond increases with increasing halogen electronegativity and this probably explains the greater ease of isolating microbes growing on chlorinated compounds compared with fluorinated acids (Hoffman, 1950). Principally studies have concentrated on halogenated acetic, propionic and butyric acids, and much less is known about the characteristics of enzymes that attack C5 or haloalkanoic acids of longer chain lengths. Enzymes that dehalogenate both D- and L-, or D- or L-stereoisomers are known. Frequently more than one dehalogenase is expressed by any given microbe (Hardman and Slater, 1981a,b). Some of the general properties of dehalogenases have been reviewed recently by Slater (19941, Janssen et al. (1994) and Slater et al. (1995) and will not be repeated here.
2.1. 2-Haloalkanoic Acid Hydrolytic Dehalogenases
Historically dehalogenases were grouped on the basis of substrate affinities, reaction kinetics, proposed or known catalytic mechanisms, and sensitivities to inhibitory compounds (Little and Williams, 1971; Hardman, 1991; Slater, 1994). We have recently updated a scheme for classifying dehalogenases (Table 1)(Slater et al., 1995) in order to accommodate the current range of known enzymes and their properties (cf. the earlier classifications of Weightman et al., 1982, and Hardman, 1991). Protein and nucleotide sequence data are defining significant similarities and differences between dehalogenases, and, in due course, accurate descriptions at these levels will determine evolutionary relationships (Leisinger and Bader 1993; Janssen et al., 1994; Slater, 1994). Presently the classification and naming of dehalogenases are based primarily on catalytic properties, with subgroups based on other factors, such as substrate specificities, and, where available, nucleotide or amino-acid sequence information.On this basis four classes of 2HAA hydrolytic dehalogenases are recognized (Table 1).
2.1.1. Class IL 2HAA Hydrolytic Dehalogenases These dehalogenases remove halides from L-2-haloalkanoic acids, inverting the product configuration with respect to the substrate, and react to sulfhydryl-blocking reagents to varying degrees. Goldman et al. (1968) postulated two mechanisms to explain the inversion of substrate configuration: an electron-donating group within the enzyme's active site yields a hydroxyl which displaces the halogen by a nucleophilic substitution; or 0 a carboxyl group within the active site acts as the nucleophile displacing the halogen with subsequent hydrolysis of the ester which is formed.
137
MICROBIAL DEHALOGENATION
1
Figure 1 Two proposed dehalogenation mechanisms (from Slater, 1994). A third mechanism, proposed by Little and Williams (1971), was a general base catalysis involving a histidine residue as the nucleophile (Fig. 1). Whatever the exact mechanism may be, presently there are 12detailed examples of this class of enzyme (Table I). The nucleotide sequences of nine of the 2HAA hydrolytic dehalogenases have been reported, clearly demonstrating nucleotide sequence similarities and defining this class as a coherent group of relatedenzymes (Table 2). Class 1L dehalogenases probably include the halidohydrolases I and I1 of the dichloroacetate (DCA)-degrading pseudomonad described by Goldman et al. (1968) and the dehalogenase of fseudumonas dehalogenans NCIMB 9061 described by Little and Williams (1971), although this cannot be asserted without the
Table 1 Classification o f 2-haloalkanoic acid hydrolytic dehydrogenases.
Class
1L
Organism
Enzyme
Gene
Number of amino acids Native Subunit Number from Calculated mol. mass mol. mass of nucleotide subunit mol. (kDa) (kDa) subunits sequence mass(kDa)
Reference
Pseudomom species CBS3
DehCI
dehCI
42.0
28.0
2
227
25.4
Pseudomoms species CBS3
DehCII
dehCii
64.0
29.0
2
229
25.7
Pseudomoms cepacia MBA4 Pseudomoms species 109 Pseudomoms putida AJl Xanthobacter autotmphicus a10 Pseudomoms species Y L Moraxella species B Rhizobium species
HdlIVa
hdlNa
45.0
23.0
2
23 1
25.9
DehHlO9
dehHI09
34.0
25 .O
2
224
25.2
Klages et al., 1983; Morsberger et al.. 1991; Schneider et al., 1991, 1993 Klages ef al., 1983; Morsberger etal., 1991; Schneider et al., 1991, 1993 Tsang et al., 1988; Murdiyatmo et al., 1992 Kawasaki et al.. 1994
HadL
hadL
79.0
26.0
2 (3?)
227
25.7
Jones et al., 1992
DhlB
dhlB
36.0
28.0
1
253
27.4
van der Ploeg et a[.,199I
L-DEXYL
LdexYL
54.0
27.0
2
232
26.2
Nardie-Dei et al., 1994
DehH-2
dehH-2
225
25.3
Kawasaki et al., 1992
280
30.9
Cairns, 1994; Cairns and Cooper, personal communication
DehI/HadL
26.0 60.0
2
Table 1 continued.
Class 1D
2R
21
Organism
Enzyme
Gene
Number of amino acids Native Subunit Number from Calculated mol. mass mol. mass of nucleotide subunit mol. (kDa) (kDa) subunits sequence mass(kDa)
hadD
Pseudomoms putidaAJ1 Rhizobium species
HadD
Pseudomoms putida PP3 A lcaligenes xylosooxidans Unidentified isolate K37 Pseudomonas put& PP3 Pseudomoms 113 Rhizobium species
DehI
dehl
DhlC
dehlC
DehI
dehl
DehII
dehll
DehIIVHadD
58.0 46.0
300
33.6
2
266
29.4
2
296
32.7
2
296
32.7
Barth et al., 1992; Smith et al., 1990 Cairns, 1994; Cairns and Cooper, personal communication Thomas, 1990; Topping, 1992 Brokamp and Schmidt, 1991 Murdiyatmo, 1991
52.0 68.0
DehII
33.0
4
Reference
Topping, 1992 35.0
2
Motosugi etal., 1982a,b Leigh et a [ . , 1988
Table 2 Amino-acid identity between Class 1L 2HAA hydrolytic dehalogenases (after Leisinger and Bader, 1993).
Organism Pseudomonas species CBS3 Pseudomonas species CBS3 I! putida All Xanrhobacter aurotrophicus GJlO I? putida 109 I? cepacia MBA4 Mornxella species B Rhizobium species Pseudomonas species Moraxella species P. putida AJ 1 Rhizobium species
Enzyme Class
Gene
dehCI dehCII hadL dhlB deh109 MlWa dehH2 hadL(R) LdexYL dehHI hadD hadD(R)
1L
dehCI
100
1L
dehCII
36
100
1L 1L
hadL dhlB
38 42
51 42
100 43
67 35
37 49
40 44
38
53
36 49 20 56
12 19
16 13
16 19
15 14
1L 1L
1L
100
1L 1L
1D 1D
100 100 33
100
39
51
100
19 21
19 17
12 19
100
44
57
100 16
100 23
100
MICROBIAL DEHALOGENATION
141
benefit of nucleotide sequence data. Furthermore, the I? dehalogenuns enzyme may be identical to halidohydrolase I. All three enzymes dechlorinated L-2-monochloropropionic acid (2MCPA) yielding D-lactate as the product but they did not dechlorinate D-2MCPA. Halidohydrolase I1 had a limited ability to use D,L2MBPA, but the product configurations are unknown and it seems unlikely that the D-isomer was dechlorinated. All three enzymes were unaffected by sulfhydrylblocking reagents, such as p-chloromercuribenmate and N-ethylmaleimide. Halidohydrolase I differed from halidohydrolase 11in that MCA was the main substrate for the former, and DCA was the principal substrate for the latter. The dehalogenase I of a Rhizobium species isolated on Dalapon was considered to be similar to these three enzymes (Berry et al., 1976, 1979; Allison el al., 1983; Leigh er al., 1988). Pseudomonas species CBS3 was isolated on 4-chlorobenzoate but synthesized two 2HAA dehalogenases, DehCI and DehCII, which dehalogenated L-2MCPAbut not D-2MCPA (Klages et al., 1983; Morsberger et al., 1991). Protein analyses showed that both of these enzymes were dimeric proteins with overall molecular masses of 41 and 64 kDa, and subunit molecular masses of about 28 and 29 kDa, respectively (Klages et al., 1983; Morsberger efal., 1991). Schneider et al. (1991) reported that the dehC1 gene (located on a 1.1 kb SinaI - SstI fragment) yielded a protein of 227 amino acids with a molecular mass of 25.4 kDa. The first 23 amino acids of the proposed sequence corresponded exactly with the N-terminal amino acid sequence from DehCl. The gene was preceded by a promotor of the -lo/-35 consensus sequence type and appeared to be negatively regulated by the enzyme’s substrates. The dehCII gene was similar and coded for a protein of 229 amino acids (molecular mass of 25.7 kDa) with the first eight amino acids of the N-terminus corresponding exactly with the sequence predicted from the nucleotide sequence. There was 45% nucleotide sequence homology, which corresponded to 37.5% amino acid sequence identity, and over 70% amino acid similarity (Table 2 ) . In this case there appeared to be a close evolutionary relationship between the dehCI and dehCIIgenes, which suggested a common origin from an ancestral gene. The observed pattern may have resulted from a gene duplication in the ancestral Pseudoinonasspecies CBS3 followed by separate but parallel evolutionary events. Alternatively, the genes may have evolved in parallel in another host organism and been transfei~edto Pseudoinonas species CBS3. Either way there was a close, and maintained, structural relationship between these two genes. In addition, there was some sequence homology between the N-terminal amino acid sequences of DehCI and DehCII, including a common AsplO, and a region containing the Aspl 24 residue in the haloalkane dehalogenase of Xanthobacter autotrophicus GJ 10, which was involved in the nucleophilic substitution catalysed by this enzyme (Section4.1). Schneider etal. (1993) used site-directedmutagenesis to replace the Aspl 0 with alanine, completely eliminating dehalogenase activity and concluded that AsplO was H nucleophiiic residue involved in the active site. The inference was that there was a defining relationship between these two major classes of dehalogenases i n the crucial region of the active site, suggesting a deep
142
J. HOWARD SLATER
eta/.
and relatively conserved evolutionary link within this region of the proteins. However, Janssen et al. (1994) questioned the direct role of Asp10 in the active site since “the proposed alignment neglects the fact that the sequence is determined largely by the position of the Asp124 in the nucleophile elbow between the p strand 5 and strand 3 and by the presence of Trp125 involved in the leaving-group stabilisation (halide binding)” and suggested that other Asp residues, Asp15 and Asp175, were involved in the active site. Tsang et al. (I 988) isolated Pseudomunus cepucia MBA4 from a batch culture growing on inonobromoaceticacid (MBA) and showed that under these conditions the organism synthesized two dehalogenases, 111 and IVa. The former was not studied in any detail, but the latter was found to be a Class 1Lenzyme (Murdiyatmo, 1991; Asmara, 1991; Murdiyatmo et al., 1992; Asmara et al., 1993). The sequence of the gene, hdllVa, for DehIVa (located on a 1.6 kb genornic fragment) predicted a gene product of 231 amino acids and a molecular mass of 25.9 kDa, corresponding to proteins observed by SDS-PAGE of about 23 kDa and by gel filtration of about 45 kDa. The sequence contained two regions which indicated -101-35 promotors, and it was suggested that the gene was under positive regulation. Unquestionably, this dehalogenase was closely related to others in the class: for DehIVa and DehCI there was 67% amino acid identity and 81% similarity, and for DehIVa and DehCII the corresponding values were 37% and 56%, respectively. Other relationships between Class 1L enzymes, based on amino acid identities, are shown in Table 2. Asmara et al. (1993), on the basis of chemical modification, random and site-directed mutagenesis, suggested that two amino acid residues, His20 and Arg42, were the key active site residues, with Asp18 also implicated in water activation by His20, again providing a possible link with the haloalkane dehalogenases (Janssen et al., 1989; Franken et al., 1991). Dehalogenase I (HadL) of a Hhizobiuin species has, for the moment, been included in Class 1 L on the basis of its overall catalytic properties. However, the protein showed minimal identity with Ha& of Pseudoinonasputida AJ1 (Table 2) and its subunit size appeared to be significantly different from the mean value of 231 for the eight dehalogenases that were unequivocally members of this class (Table 1). Similarly, DehH-2 of Moraxella species B has been included in Class 1Lon the basis of amino acid sequence information (Table 2) (Kawasaki et al., 1981a,b,c, 1992). This organism was isolated with MFA as the growth substrate and the two dehalogenases it contained dehalogenated only haloacetic acids; its response to stereo-specific compounds is not known. This example serves to reinforce the difficulty of using substrate specificities as the method of classification. The sequence data strongly suggested that this enzyme should be in Class lL, but it has evolved substrate specificities that are more limited and exclude C3 alkanoic acids. Basic information on the other microbial dehalogenases in Class 1L will not be discussed in detail, but may be found in the earlier reviews of Leisinger and Bader (1993), Slater (1994), Janssen et al. (1994) and Slater etal. (1995). The information
MICROBIAL DEHALOGENATION
143
presently available shows that this is an important and coherent class of 2HAA dehalogenases. The similarity of subunit sizes in terms of amino acid residues is striking (Table 1),although there may be interesting differences with respect to the number of subunits in the functional proteins. All the biochemical information strongly indicates a uniform class of proteins at the functional and mechanistic levels. In addition, there are suggestions of relationships with haloalkane dehalogenases, which should be of considerable interest in the future as evolutionary relationships are resolved (Section 4.1). 2.1.2. Class I D 2HAA Hydrolytic Dehalogenases The key feature of this class of 2HAA hydrolytic dehalogenases is their ability to dehalogenate selectively D-isomeric substrates, such as D-2MCPA, with inversion of product configuration. Many groups have tried, with little success, to isolate microbes that produce this class of dehalogenase; the most detailed work by the Zeneca plc (formerly ICI plc) group, involved Pseudomonas pufida AJl (Taylor, 1985, 1988; Barth, 1988; Smith e f al., 1989a,b, 1990; Barth et al., 1992). This industrial research group was interested in exploiting D-specific dehalogenase properties in a commercial process using L-2MCPA as the starting point for the synthesis of a novel herbicide. With a racemic mixture of D,L-~MCPA(manufactured cheaply by simple chemical routes) as the starting material, 50% of the final product did not have the required biological activity, and so half the synthesis was a waste of manufacturing time and resources. An initial treatment of the inexpensive racemic mixture to remove the unwanted D-2MCPA selectively ensured the efficient manufacture of a product with full biological activity, whilst the added manufacturing costs of the dehalogenase treatment step were offset by the improved unit costs of the L-specific process. Pseudornonasputida AJ 1, isolated from soil pre-enriched with racemic 2MCPA, contained two dehalogenases. Both enzymes were highly stereospecific: HadL was active against L-2MCPA (Section 2.1,l), whilst HadD dehalogenated D-2MCPA only (Table 1). Both native enzymes were probably tetramers with HadD having a molecular mass of 135 kDa (Smith et al., 1990), and HadLa mass of 79 kDa (Jones et al., 1992). The two genes, ha& and ha&, were closely linked on the genome. The hadD gene encoded a polypeptide of 33.6 kDa, and the first 21 amino acids deduced from the nucleotide sequence corresponded to the N-terminal analysis of the protein. Apart from the dehalogenase 111 from the Rhizobiurn species (see below), this enzyme had no significant homology with any other dehalogenase represented i n nucleotide or protein databases (Table 2). There was no -lo/-35 promotor preceding the structural gene. Recently, Cairns and Cooper at the University of Leicester have examined the Dalapon-degrading Khizobiuriz species (Section 3.1) containing dehalogenases I and 111, and have suggested that these enzymes corresponded to HadL and HadD
144
J. HOWARD SLATER eta/.
of r! putida AJI, respectively (Cairns, 1994; R.A. Cooper, personal cornmunication). The equivalent enzyme to HadD in Rhizobium, dehalogenase 111,had a native molecular mass of 58 kDa and a subunit molecular mass of 28 kDa, suggesting a dimeric structure. Nucleotide sequence data indicated a polypeptide of 29.4 kDa and 266 amino-acid residues. Sequence analysis suggested 23% identity with the amino-acid sequence of HadD, but no other relationships were found (Table 2). This was a rather low level of identity and there were other significant differences, such as its dimeric structure compared with the tetramer of HadD. Furthermore, the Rhizobiunz D-specific dehalogenase I11 subunit size was smaller than the H a m subunit size of P puridu AJ1 (Table 1). It is possible that these two D-Specific enzymes were related but it seems more likely that this was only at the functional level. 2.1.3. Class 21 2 H A A Hydrolyric Dehalogenases Enzymes of this class differ from Class 1 enzymes by their ability to dehalogenate both L- and D-isomers by a mechanism that inverts substrate-product configurations. This in turn distinguishes them from Class 2R enzymes (Section 2.1.4). Class 21 dehalogenases are unaffected by sulfhydryl-blocking reagents unlike Class 2R enzymes (Table 1). Motosugi et al. (1982a,b) isolated Pseudornonus species 113 which grew on both D- and L-2MCPA and synthesized a single dehalogenase which dehalogenated the L-isomer more rapidly than the D-isomer (D-isomer at 72% of the L-isomer rate). The dehalogenase had a molecular mass of 68 kDa and subunits of 35 kDa suggesting that the protein was a dimer. In a series of interesting experiments, Hasan et al. (1991) demonstrated that lyophilized preparations of the purified dehalogenase, when dissolved in organic solvents, such as anhydrous dimethyl sulfoxide (DMSO) or toluene, catalysed the dehalogenation of long-chain haloalkanoic acids and aromatic substituted haloalkanoic acids. For example, 2-bromohexadecanoic acid, which was not dehalogenated in aqueous solution, was dechlorinated in these solvents at 2.7 times the rate of 2MCPA. In solvents, L-2MCPA was only dehalogenated at 6% of the rate in aqueous solutions, leading to the suggestion that activity against higher chain length acids was due to their greater solubility in organic solvents. Pseudornonns putidu PP3 evolved in a stable microbial community growing continuously on the herbicide Dalapon and was found to synthesize two 2HAA dehalogenases (Slater et al., 1976, 1979; Senior et al., 1976; Senior, 1977). Dehalogenase 11 (DehII) (previously termed the fraction I1 dehalogenase) was active against a similar range of HAAs as the other enzyme, dehalogenase I (DehI) (Section 2. I .4), but differed by being unaffected by sulfhydryl-blocking reagents (Weightman el al., 1979, 1982; Weightman, 1981). The native molecular mass of DehII was 52 kDa (Topping, 1992). The differences were such that it seemed very
MICROBIAL DEHALOGENATION
145
unlikely that the two enzymes were related mechanistically and, therefore, originated from different ancestral genes, a situation closer to the unrelated HadL and HadD of l? putida AJI, rather than the two related enzymes of Pseudomoms species CBS3. So far DehII has not been sequenced and so comparisons with other dehalogenases remain tentative. However, it may be closely related to dehalogenase I1 of the Rhizobium species discussed in Sections 2.1.1 and 2.1.2, although the Rhizobium enzyme showed some sensitivity towards thiol reagents. 2.1.4.
Class 2R 2HAA Hydrolytic Dehalogenases
Enzymes of this class differ from Class 1 enzymes in their ability to dehalogenate both L- and D-isomers by a mechanism that retains substrate-product configurations, thus separating them from Class 21 enzymes (Section 3.3). These dehalogenases are strongly inhibited by sulfhydryl-blocking reagents unlike Class 21 dehalogenases (Table 1). Dehalogenase I (DehI) (formerly fraction I dehalogenase) (Slater et al., 1979; Weightman et al., 1979) was estimated to have a molecular mass of 46 kDa as the native protein. However, the nucleotide sequence for d e N predicted that the polypeptide had a molecular mass of 32.7 kDa which corresponded to SDS gel estimates of 33 kDa, suggesting possibly that the DehI protein was a dimer in the active state (Thomas, 1990; Topping, 1992; S.J. Hope and J.H. Slater, unpublished observations). The N-terminal sequence of the purified DehI has not been determined to date, but Murdiyatmo (199 1) purified an enzyme called HdlV from an unidentified isolate, strain K37, and determined the first 13 N-terminal amino acids. The sequence corresponded exactly with that of a protein encoded by the putative dehl open reading frame beginning at the second methionine residue. In the nucleotide sequence between codons specifying the first and second methionine residues (the DehI protein start), there was a strong Shine-Dalgarno sequence region separated by eight bases from the initiation codon, a distance considered to be optimal for transcription (Gold, 1988). Another Shine-Dalgarno sequence ten bases upstream from the codon specifying the first methionine was also located, and so the precise polypeptide sequence remains to be resolved. Recently we have shown that the amino acid sequence for DehI is identical to the sequence for DhlC fromAZcaligeizesxylosooxidnns (J.H. Slater, S.J. Hope, M.R. Lewis, A.W. Thomas, A.W. Topping, S.D. Greenaway and D.J. Hardman, unpublished observations). Unlike the genetic organization for all other dehalogenases described to date, a -12/-24 type promotor motif was identified upstream from dehl, These promotors are recognized by the 054family of sigma factors of RNA polymerases and have been associated with the expression of many metabolic functions in Gram-negative and Gram-positive bacteria (Dixon, 1986; Thony and Hennecke, 1989; Merrick, 1993). This confirmed earlier observations that expression of dehl was dependent
146
J. HOWARD SLATER eta/.
on the presence of a functional d4polymerase, since it was not expressed in rpoN mutants of t.i puticla suggesting that dehl was under positive regulatory control (Thomas et al., 1992b).
2.2. Regulation of Dehalogenase Synthesis
Very little is known about dehalogenase regulatory mechanisms at the molecular level, although the picture is becoming clearer with respect to the genetic organization of the regulation system for DehI. We observed, in gene transfer experiments (Section 2.3.1) that, when hybrid plasmids carrying dehl were transferred to other strains oft? putidu, l? aeruginosa or E. coli, dehl was regulated in a manner that was identical to the parental strain (Beeching et al., 1983; Slater et al., 1985; Thomas, 1990; Thomas et al., 1992a). Subsequently, mapping of an 11.6 kb EcoRI-G fragment from a derivative TOL plasmid containing the dehl gene, by transposon mutagenesis and complementation analyses, identified a region that was crucial to the expression of dehl. Topping ef al. (1995) demonstrated that dehl was adjacent to its regulatory gene, dehRI (Fig. 2). Analysis of dehRI showed that the gene had an open reading frame, which encoded a 64 kDa protein containing 571 amino acid residues. About 60 bases upstream from the putative translation start of dehR, was a highly conserved -35/-10-type promotor motif. A DNAbinding helix-turn-helix motif was identified at the C-terminus (Dodd and Egan, 1990). Comparison of the dehRI amino acid sequence with other d4-dependent activators showed that there was significant amino acid conservation within a central domain of approximately 230 amino acids, whereas there was little amino-acid sequence homology in the carboxyl- and amino-terminal domains. The conserved central region, termed the C-domain by Drummond et al. (1986), was responsible for interactions with RNA polymerase and bound ATP (Thony and Hennecke, 1989; Huala and Ausubel, 1989; Kustu et al., 1991; North et al., 1993) and may be expected to show significant similarities between different d4dependent activators. A cladograin showing the relationships between a number of d4-dependent activators is shown in Fig. 3 . This family of activators includes various nitrogenase regulators (NifA-like) and other nitrogen regulators (NtrC-like). This group contained only two activators for genes associated with biodegradative pathways, namely, DmpR, which regulated dimethylphenol catabolism (Shingler et al., 1993; Fernandez el al., 1994),and XylR, which regulated methylbenzoatecatabolism (de Lorenzo et ul., 1991). However, in the C-domain region of 229 amino acids, DehRI showed only 25.6% similarity with XylR and 26.1% similarity with DmpR. DehR1 did not cluster with either the NifA or NtrC groups; instead it had greatest C-domain similarity (48%) with the putative product of the hyuR sequence from a Pseudornonas species, a plasmid-borne regulatory gene involved in L-amino acid biosynthesis (Watabe et a/., 1992). Despite the similarity of DehRl to NifA-like and
0
I
2.00
4.00
6.W
8.00
10.00
12.00
I
I
I
I
I
I
H M l
PSI
3.80
4.41
4.69
4.91
6.69
6.91
Figure2 Physical map showing the arrangement of the DEHtransposon containing the structural gene dehl and its regulator gene dehRI, showing the key endonuclease restriction sites (after Topping et al., 1995).
148
J. HOWARD SLATER
I I
eta/.
NifA- like
NtrC-like
Figure 3 Cladogram illustratin the relationships between the derived DehRI aminoacid sequence and those of other J4-dependent activators. The cladogram was produced after an alignment of the conserved central regions (see text) of the selected activators using CLUSTALV (Higgins el a/., 1992) and manual editing. Sequence similarities were determined by the Dayhoff PAM matrix method and Neighbor-Joining using the PHYLIP package (Felsenstein, 1989). The following sequences were obtained from the GenBank database unless noted otherwise: AcoR-A.e; acetoin catabolic regulator from Alculigenes eutrophus (P28614-SWISSPROT database); AcoR-C.m, acetoin catabolic regulator from Clostridiurn magnu/n (L3 1844); DctD-R.I, C4-dicarboxylate transport regulator from Rhizobiirm /egu/ninosunrm (X06253); DehRI-ttp, dehalogenase regulator from Pseudomonus pufida (U237 16); DmpR-Psp, dimethylphenol catabolic regulator from Pseudomonas sp. CF600 (X68033); FixD-R.m, nitrogenase regulator from Rhizobium meliloti (X03065); GlnBXc, glutamine metabolic regulator from Escherichiu coli (S67014); HoxA-A.e, hydrogenase regulator from A. eurrophus (M64593);HydG-Ec, hydrogenase regulator from E. coli (M28369); HydG-S.t, hydrogenase regulator from Salmonella typhimurium (M64988); “HyuR’-Rsp, hydantoin catabolic regulator from Pseudomoms
MICROBIAL DEHALOGENATION
149
NtrC-like activators, the dehl gene was not activated in putatively trans-acting plasmid constructions with these nitrogen metabolism activators (A.W. Topping, unpublished observations), suggesting that the significant sequence differences at the termini, particularly the N-terminus, resulted in effector specificity and defined the regulatory capability of DehRI. The G + C content of dehRI was much lower than that expected for a Pseudumunas species (51.7% compared with about 63%). Murdiyatmo et al. (1992) and Jones el al. (1992) also reported lower G + C ratios for dehalogenase structural genes, suggesting that these genes originated in other species and were transferred to Pseudoinonas species later in their evolution, perhaps by plasmid transfer or as a component of an appropriate transposon (Section 2.3.2).
2.3. Genetic Organization of 2HAA Hydrolytic Dehalogenases
2.3.1. Plasrnids Frequently, dehalogenase genes are located on plasmids (Slater and Bull, 1982; Reanney et al., 1983). Kawasaki and his colleagues showed that the MFA- and MCA-utilizing capability in Moruxella species B was associated with a plasmid (Kawasaki el al., 1981b,c,d, 1982, 1983a,b, 1984, 1985, 1992). Strain B contained a single large plasmid, p U 0 1, with a size of 66 kb and encoded the genes for both Deh-H1 and Deli-H2. In some mutants, for example, strain 86, in which Deh-H2 was lost, there was an accompanying decrease in the size of the remaining plasmid, whilst in other strains, for example strain 123,complete loss of the plasmid resulted in the loss of both dehalogenases. Brokamp and Schmidt (1991) demonstrated the transfer of plasmid pFL40, which carried the Class 2R 2HAA dchalogenase in species NS671 (QOl265-SWISSPROT); LuxO-Vh,luminescence regulator from vibrio harveyi(L26221);NifA-Azb.v, nitrogenaseregulatorfrom Azofobacter vineZandii(Y00.554); NifA-Azxc, nitrogenase regulator form Azorhizobium caulinodans (X08014); NifA-K.p, nitrogenase regulator from Klebsiella pneumoniae (X02616);NifA-R.E. nitrogenase regulator from R. legiiminosarum (LI 1084); NifA-Rhd.c, nitrogenase regulator from Rhodabacter capsulatus (X07567); NtrC-Agt, nitrogen regulator from Agrobacterium tumefaciens (J03678); NtlC-Azs.b, nitrogen regulator from Azospirillum brasiiiense (X67684); NtrCB K ~nitrogen , regulator from Bradyrhizobium parasponia (M14227); NtrC-E.c, nitrogen regulator form E. coli (X05173); NtrC-K.p, nitrogen regulator from K. pneumoniae (X02617); NtK-R.m, nitrogen regulator from R. meliloti (M15810); NtrC-TJ nitrogen regulator from Thiobacillrrsferrooxiduns(L18975); PilR-Pa, fimbriae expression regulator from Pseudomonos aei-uginosa (Q00934-SWISSPROT); RocR-B.s, ornithine aminotransferase regulator from Bacillus subtilis(L22006);vrR,E.c, tryptophan biosynthetic regulator from E. coli (M12114); VnfA-Azb.v, nitrogenase regulator from A. vinelandii (M26752); XylR-Pp., methylbenzoate catabolic regulator from P. putida (P06519-SWISSPROT) (from Topping et al., 1995).
150
J. HOWARD SLATER eta/.
Alcaligenes xylosooxidans ABlV to Pseudoinonas fluorexems and other soil bacteria in sterile soil microcosm experiments. Hardman (1982) showed that almost all novel, freshly isolated 2HAA-utilizing strains of soil bacteria contained large plasmids, varying in size from about 150 kb to 300 kb (Hardman etal., 1986).Spontaneous loss or curing of the plasmidresulted in loss of the ability to grow on 2HAAs. Beeching et al. (1983) showed that it was possible to mobilize the dehl gene on to suitable target plasmids such as RP4 and RP45, and we have come to realize that this property was associated with the transposon structure carrying the dehl gene (Section 2.3.2). 2.3.2. The DEH Transposon Many 2HAAs are toxic and their uptake into cells causes cessation of growth and cell death if exposure is prolonged. Active synthesis of dehalogenases removes the inhibitory problem and growth resumes (Slater et al., 1979). In selection experiments originally designed to produce mutants with impaired uptake mechanisms, it was found that mutants of Pseudoinonas putida PP3 resistant to concentrations of MCA or DCA as high as 100 m~ were obtained at very high frequencies. For example, at 42 m~ MCA, resistant mutants appeared at a frequency of 3.2 x lo-' and, at 42 mM DCA, the frequency was 5.3 x (Slater etal., 1985). The mutants were stable and on analysis five classes were defined (Table 3), all of which had reduced capabilities for the uptake of 2HAAs (Slater et al., 1985). On the basis of these observations, we proposed that the mutants were the result of the loss of one or more transposons carrying DchI and DehII, and their associated uptake systems, which impaired the uptake of 2HAAs leading to the resistant phenotypes. Subsequently the transposon, DEH, carrying dehl and dehRI, was characterized (Thomas, 1990; Thomas et al., 1992a,b,c; Topping, 1992). DEH was an unusual transposon, since it transferred to target DNAs as an element of varying size, ranging from 6 to 13 kb. To date there is no evidence to confirm that dehll was also located on a transposon, and it is possible that mutations involving the loss of this gene were associated with the movement of DEH and topological interactions between DEH and the dehll gene which resulted in changes of expression of both genes. Furthermore, changes in the growth environment, in particular the presence of toxic 2HAAs, resulted in high-frequency mutations which switched off one or both genes, and other conditions resulted in the expression of dehllbeing switched back on (Thomas et al., 1992a; Hope and Slater, 1995; Slater and Hope, 1995). R putida PP3 and its associated mutants (Table 3) routinely formed papillae in colonies grown under appropriate conditions as cryptic dehalogenase genes were switched on (Thomas et al., 1992a; Hope and Slater, 1995; Slater and Hope, 1995). Movement of DEH under these conditions led to major rearrangements of the DNA as shown by altered restriction enzyme digest patterns (Thomas et al., 1992a). In other systems, loss of Deh-H2 was associated with a discrete piece of DNA
151
MICROBIAL DEHALOGENATION
Table 3 Characteristicsof various W A Rmutants of Pseudornonusputida PP3 isolated by growth in the presence of 42 mM DCA and succinate.
Strain or mutant class
Specific growth rate' (h-')
PP3 PP40 PP411 PP4 12 PP4 120 PP42
0.3 1
0.00 0.14 0.23 0.04
0.26
Response Dehalogenase complement to DCA Permease (85 mM) DehI DehII activityb S R R R R R
+ +
-
-
+
+
-
+ + +
8000 100 1900 2100 100 657
%Total mutants -
16 54 7 3
20
a Average
values based on three independent determinations,except for the growth rates of PP412 mutants (four detenninations) and PP4120 mutants (six determinations). Determined from the initial rate of uptake of [ 14C]MCAwith rates expressed as counts per minute of radioactivity uptake per minute of incubation per 1 mg of cell protein (from Slater, 1994).
with a constant size of 5.4 kb and it was suggested that this gene might be located on a transposon (Kawasaki et al., 1983a,b, 1984, 1985). Kawasaki et al. (1992) showed that the dehH2 coding region was sandwiched between two repeated sequences of about 1.8 kb, and suggested that these homologous sequences might be implicated in the excision of the gene and/or its transposition. Topping (1992) also found that there were short repeated, homologous regions within the DEH transposon. Janssen et al. (1994) suggested that the overexpression of DhlB in mutants of Xmthobacter autotrophicus was due to the insertion of a small DNA fragment upstream of the dhlts gene. Clearly a great deal of further information about the mobility of dehalogenase genes is needed to shed light on their expression and control, and the interaction between different dehalogenase genes.
3. DEHALOGENATION OF HALOGENATED ALCOHOLS Dehalogenation of haloalcohols has been known since Castro and Bartnicki (1965) showed that 3-bromopropanol was biodegradable. These workers isolated a Flavobacteriurn species which grew on 2,3-dibromopropan-l-ol and found that it converted halohydrins to epoxides (Castro and Bartnicki, 1968; Bartnicki and Castro, 1969). At present it appears that inicroorganisms have evolved two main pathways for the mineralization of haloalcohols.
152
J. HOWARD SIATER eta/.
3.1. A Pathway Involving the Formation of HAAs
The first pathway is mediated by enzymes that accept halosubstituted molecules as their substrates, with dehalogenation taking place only after the haloalcohols have been oxidized to the corresponding haloalkanoic acids. This mechanism was described for the degradation of 2-chloroallylalcohol(van der Waarde et al., 1993), and was found in the Gram-negative bacterium CElr (Stucki and Leisinger, 1983) and Pseudornorzas putida U2 (Strotmann er al., 1990). In Gram-negative bacteria, the haloalcohol pathway was part of the core pathway described for the degradation of 1,2-dichloroethaneby Xanfhobacter autotrophicus GJlO (Janssen ef al., 1985) (Section 4.1). Here catabolism of the disubstituted haloalkane resulted in the formation of 2-chloroethanol and, thereafter, enzyme recruitment resulted in the conversion of the halogenated alcohol to MCA, which in turn was dehalogenated to glycolate by a 2HAA hydrolytic dehalogenase. For example, Stucki and Leisinger (1983) showed that strain CElr synthesized alcohol and aldehyde dehydrogenases, which specifically converted 2-chloroethanol to 2-chloroaldehyde and thence MCA. The 2-chloroethanol-degrading strains expressed the same catalytic pathway as Xarzthobacter autufrophicus GJ10, without the first hydrolytic haloalkane dehalogenase, and constitutively expressed an MCAspecific hydrolytic dehalogenase. A comparison of the enzymes from the pseudomonads and the Xanfhobacter strain at a molecular level would be very interesting in order to determine the exact relationships between these pathways. The possibility that they are closely related and that acquisitive evolution has led to the construction of this pathway was supported by the observation that the structural genes for the haloalkane dehalogenase and the aldehyde dehydrogenase for the 1,2-dichloro-ethane pathway were encoded on a 200 kb plasmid (Tardif et al., 1991). In a related pathway, Rhodotorula glutinis catalysed the asymmetric reduction of prochiral chloroacetone to chiral l-chloro-2-propanol using an NAD(P)H-dependent alcohol dehydrogenase (Weijers ef al., 1992), which suggested that these enzyme systems are related and widely distributed in nature.
3.2. Haloalcohol Hydrolytic Dehalogenases The alternative pathway involves specific haloalcohol dehalogenases. When enrichment cultures were established to select for halohydrin-degrading bacteria, the competent isolates normally catalysed the direct dehalogenation of vicinal haloalcohols via a process that is basically hydrolytic, but in which the actual dehalogenation step involved an internal rearrangement of the substrate and a simultaneous elimination of a proton, with the halide resulting in the formation of an epoxide (Fig. 4). This mechanism, first describedby Castroand Bartnicki (1965) in the 2,3-dibromo-1-propanol-degrading Flavobacteriurn species was catalysed
153
MICROBIAL DEHALOGENATION
CH,CI-CHOH-CH,CI 1,3-dichloro-2-pmpnol CH,OH-CHOH-CH,CI 3-c hl oro-l,Zpropandiol
- /"\ + H,O
-HCI
(a)
-HCI .
(C)
CH,CI epicNomhydrin
(b)
0
+HO 4C%OH 2 (d) glycidol
CH,OH-CHOH-CH,OH gly c e d Figure 4 Pathway for the catabolism of 1,3-dichloro-2-propanol(DCP). (a) and (c), haloalcohol dehalogenases; (b) and (d), epoxide hydrolases.
by enzymes termed halohydrin epoxidases (Castro and Bartnicki, 1968). Subsequently, these enzymes have been variously described as haloalcohol dehalogenases and haloalcohol halogen-halide lyases (van den Wijngaard et al., 1989, 1991) and halohydrin halogen-halide lyases (Nagasawa et al., 1992). In the complete pathway, the epoxide resulting from the dehalogenation reaction is hydrolysed by an epoxide hydrolase to the corresponding alcohol (Fig. 4). Haloalcohol dehalogenases showed a restricted range of activities being active only when the halogen is vicinal to a hydroxyl group or to a keto group, such as in chloroacetone (van den Wijngaard et al., 1991). The most commonly used substrates for the isolation of haloalcohol-competent strains were 1,3-dichloro-2propanol (13DCP) (Nakamura et al., 1992), 3-chloro-l,2-propandiol (3CPD) (Suzuki and Kasai, 1991; Suzuki eral.. 1992) and epichlorohydrin (ECH) (van den Wijngaard el a/., 1989).The dehalogenation reaction was reversible, which allowed the interconversion ofvic-haloalcohols and epoxides. This reversibility also means that the enzymes catalyse the trans-halogenation of haloalcohols. The pathway for the biodegradation of 13DCP mimics the chemical degradation of the substrate when exposed to solutions of high pH (Fig. 4).In all of the systems reported to date, two different epoxide hydrolases were required to open the ECH and glycidol ring structures resulting from the dehalogenation of 13DCP and 3CPD, respectively. However, when a strain lacked the epoxide hydrolases, for example, Arthrobacter species AD2 (van der Wijngaard et al., 1989), chemical hydrolysis of the dehalogenation products still enabled it to utilize the metabolites for growth, albeit at a much reduced rate. In common with other aliphatic biodehalogenation systems, multiple dehalogenases are frequently observed in isolates. The most common profile is two dehalogenases, although isolates with only a single enzyme are known (Fig. 5).
154
J. HOWARD SIATER
1
2
eta/.
3
Dehalogenase D Dehalogenase C
Dehalogenase B Dehalogenase A
i
Dehalogenase A with different subunit
combinations
Figure 5 Zymogram of the common haloalcohol dehalogenase forms as separated by activity PAGE. Examples: 1, Agrobacterium tumefaciens HK7 (Bull et al., 1992), Pseudomonas species AD1 (van den Wijngaardet al., 1989);2, Arthrobacterhistidinolovorans (Bull e l al., 1992). Arthrobacter AD2 (van den Wijngaard el al.. 1989);3 “Arthrobacter erithii” (Assis et al., 1996). Corynebacteriurnspecies N-1074 (Nagasawa et al., 1992).
Five alcohol dehalogenases have been described in detail from P s e u d o m m s species AD1 and Arthrobacter species AD2 (van den Wijngaard et al., 1989), Colynebacterium species N-1074 (Nagasawa et ul., 1992) and another pseudomonad capable of growth on 2,3-dichloro-l-propanol(23DCP)(Kasai et al., 1990, 1992).In our laboratory, we have identified six electrophoreticallydistinct enzymes from species of Pseudomonas, Agrobacteriurn, Arthrobacter and Rhodococcus (Bull etal., 1992; Assis et al., 1996; A.J. Cotton, M. Huxley, D.J. Scherr, A.T. Bull, J.H. Slater and D.J. Hardman, unpublished observations). DNA profiling studies indicated that Pseudoinonas species AD1 may in fact be a strain ofAgrobacterium tumefaciens (A.J. Cotton, S.D. Greenaway, A.T. Bull, J.H. Slater and D.J. Hardman, unpublished observations). In comparison with the other classes of dehalogenases, the biochemical and molecular data are limited (Table 4). The current information suggests that there is limited diversity among these enzymes isolated from organisms across a wide geographical distribution. The
MICROBIAL DEHALOGENATION
155
available data suggest that there are only two classes of haloalcohol dehalogenases, defined ds Class 4 s and Class 4C (Slater et al., 1995) (Table 4). Class 45 enzymes are structurally simpler than Class 4C enzymes, being simple dimeric proteins formed from a single polypeptide subunit; Class 4C enzymes are composed of at least two different polypeptides. Class 4C enzymes exhibit a narrower substrate profile that Class 4 s enzymes, but they also have chiral selective properties. In the same way that the 2HAA dehalogenases are grouped according to their relative activities towards different substrates, so can the haloalcohol enzymes. The Class 4C enzymes, as represented by haloalcohol dehalogenase Ib from Corynebacteriurnspecies N-1074 (Nakamura et al., 1992) and DehA from “Arthrobacter erithii” (Assis et al., 1996), showed little activity towards 3CPD, but high activity towards 13DCP. In contrast, other enzymes, such as the enzyme from Alcaligenes species DS-S-7G (Suzuki et aE., 1992), had no activity towards 13DCP and greatest activity towards 3CPD. Such differences are analogous to the different specificities of the 2HAA dehalogenases towards mono- and di-halogenated acetic acids (Hardman and Slater, 1981a). Activity towards 23DCP also served to differentiate these enzymes (Kasai et al., 1990, 1992; A.J. Cotton, A.T. Bull, D.J. Hardman and J.H. Slater, unpublished observations), whereas Pseudomonas species AD 1 and Arthrobacter species AD2 (van den Wijngaard et al., 1989), Corynebacteriurn species N-1074 (Nagasawa et al., 1992) and Flavobacteriurn species (Castro and Bartnicki, 1968) had little or no activity towards 23DCP. Dehalogenation of 2CPD has not been reported. Detailed biochemical information is available for only four of the haloalcohol dehalogenases. The enzyme from Arthrobacter species AD2 was purified and characterized (van der Wijngaard et at., 1991). This Class 4 s dehalogenase (molecular mass 65-69 kDa) was a dimer consisting of two identical subunits of mass 29 kDa. The enzyme was iinmunologically distinct from that of Pseudoinonas species AD1 and also demonstrated very different substrate profiles. The haloalcohol dehalogenase I, from Corynebacteriurn species N-1074 (Nagasawa et al., 1992)was catalytically similar to the AD2 enzyme and, despite the fact that enzyme I, was a hoinotetrainer (molecular mass 105 kDa), its subunits showed a similar molecular mass (28 kDa) and a very Similar N-terminal amino acid structure to the AD2 dehalogenase. In contrast to the similarities between the dehalogenase 1, and AD2 enzymes, the second enzyme, Ib, in Corynebacteriurnspecies N-1074 was markedy different from the I, dehalogenase. Again the presence of genetically, immunologically and mechanistically different dehalogenases within the same organism has been described previously for 2HAA dehalogenases (Weightman et al., 1982). Dehalogenase Ib, a representative of Class 4C, was a tetrameric protein (molecular mass 115 kDa) forined from two different polypeptide subunits (32 and 35 m a ) . On PAGE, a ladder of five dehalogenase-active bands was created as a consequence of the formation of homo- and hetero-tetramers from different combinations of the
Table 4 Properties of selected haloalcohol dehalogenases.
Organism Pseudomonas species AD1 Dehalogenase Enzyme class Molecular mass
4s
orb>
Subunits Subunit molecular mass (ma) Relative substrate activity 13DCP 100.00 23DCP 0.00 3CPD 19.00 1CP 13DCA CA 13DBP 134 OOO.00 1BP 2CE 13DCPaffinity constant
Alcaligenes
Arthrobacter
Corynebacterium species
species AD2
N- 1074 Ia
Ib
4s 65-69
4s 105
4c 115-118
DehA 4c 200
2 29
4 28
4 32 and 35
6 3 1.5 and 34
100.00
100.00 0.30 37.50 22.70
100.00 0.09 0.92 18.50
12 500.00
161.00
60.00
8.50
0.26 2.44
0.13 1.03
1.20 0.11
9.00
3.13
148.00
236.00
10.00 11.00 356.00 10.00 31 000.00 3780.00
N-1074
species OS-K-29
100.00 10.00 33.00
species DS-K-S-38
100.00 47.00 106.00
“Arthrobacter erithii H l Oa”
100.00 0.00 0.10 27.00 72.00
(Km)(m)
v m (Fr;nOl
min- mg-’) Stereopsecific pH optimum Temperature optimum (“C)
No
No 8.5 50
No 8.0-9.0 55
Yes 45
Yes
Yes
Yes 8.5-9.5 50
Table 4 continued.
Organism Pseudomonas species AD1
Arthrobacter species AD2
Inhibitors (% max. activity) Mercuric Slight chloride NEM None MCA (ki) (mM) 0.05 Reference van der van der Wijngaard et Wijngaard et al. (1989) af. (1991)
Corynebactenum species N- 1074
N-1074
26.0
80.4
Pseudomonas species OS-K-29
A lcaligenes species DS-K-S-38
“Arthrobacter erirhii H 1Oa”
100.0
None Nagasawa et al. Nakamun el 01. (1992) (1992,1994)
Kasai et nl. ( 1992)
Suzuki et al. ( 1992)
Assis ef nl. (1 996)
13DCP, 1,3-dichloro-2-propanol;23DCP, 2,3-dichlorc+I -propanol; 3CPD. 3-chloro-1.2-propandiol; 1CP 1-chloropropanol; 13DCA, 1.3-dichloroacetone; CA, chloroacetone; 13DBP. 1,3-dibromopropanol; I BP, I-bromopropanol;2CE, 2-chloroethanol; NEM. N-ethylmaleimide; MCA, rnonochloroacetic acid.
158
J. HOWARD SLATER eta/.
two subunits (4:O; 3:l; 2:2; 1:3; 0:4) (Nakamura et al., 1992). A second example of a Class 4C haloalcohol dehalogenase, DehA, was purified and characterized by Assis et al. (1996) from “Arfhrubacfererithii” HlOa. This enzyme also was composed of two different subunits (molecular masses 31.5 and 34 kDa) and had a molecular weight of 200 kDa, suggesting that it was a hexameric protein. The Corynebacteriurn I b and the Arthrobacter DehA enzymes were the first dehalogenases reported to be composed of two different polypeptides. The Class 4C haloalcohol dehalogenases appeared to have a greater affinity for 13DCP than the Class 4 s enzymes. However, Fauzi et al. (1996) isolated oligotrophic bacteria growing on very low concentrations of 13DCP.These bacteria possessed haloalcohol dehalogenases that belonged to Class 4s (in terms of electrophoretic mobilities) but demonstrated K, values in the order of 0.1 mM, more akin to the values for Class 4C enzymes. The main functional characteristic of Class 4C haloalcohol dehalogenases was their enantioselective catalytic activity which may be related to their relative complexity. The similarity witli the complex structure of the D-specific 2HAA dehalogenase which was a tetrameric protein and was structurally significantly different to the dimeric L- and D,L-2-haloalkanoic acid dehalogenases, may be significant (Section 2). The enantioselective dehalogenation of 13DCP, leading to the formation of R-ECH, has commercial interest for the synthesis of chiral pharmaceuticals, such as P-adrenergic blockers, platelet-activating factor, vitamins, pheromones, agrochemicals and ferro-electric crystals. Most of the reported methods for the production of chiral epoxides using dehalogenating bacteria were based on the enantioselective degradation of racemic mixtures. The process patented by Zeneca plc for the production of L-2-~nonochloropropionate(Section 2.1.2), utilized a D-specific dehalogenase to remove selectively the D-enantiomer from the racemic mixture. In the case of the chiral epoxides, either 23DCP (Kasai et al., 1990,1992) or 2-chloro-1,3-propandiol (2CPD) (Suzuki and Kasai, 1991; Suzuki et al., 1992) have been used. The unassimilated haloalcohol accumulated in the growth medium and was subsequently converted chemically to the epoxide by addition of hydroxide. Using this approach, both isomers of ECH and glycidol (GDL) were obtained in high enantioineric excess (>99%). The disadvantage of enantiomer resolution based on enantioselective biodegradation was that the yield of the desired product was less than 35%. From an industrial point of view, the production of optically active compounds by enantioselective microbial transformation of prochiral starting materials is more attractive, since a quantitative yield of the desired enantiomer can be obtained. Nakamura et al. (1992, 1994) studied such a transformation of 13DCP to R-2CPD by Corynebacterium species N-1074. Using this method, they obtained a molar conversion yield of 97.3% but the enantiomeric excess of the R-CPD produced was too low (87.3%) to be of commercial interest. The characterization of the haloalcohol dehalogenases is not as advanced as the other groups of aliphatic dehalogenases. It is apparent, however, that there is
MICROBIAL DEHALOGENATION
159
diversity of structure and function, which is comparable to that of haloalkane dehalogenases (Section 4).
3.3. Other Systems for the Transformation of Haloalcohols
A novel enzymatic mechanism responsible for the catalysis of the NAD-dependent oxidative dechlorination of K-3CPD to acetic and formic acids was recently described in Alcaligenes species DS-S-7G (Suzuki ef al., 1994). The enzyme (Enzyme 1) that demonstrated enantiospecificity was characterized and shown to be a dimeric flavoprotein (molecular mass 70 kDa) composed of two different polypeptides (58 and 16 kDa). This protein was associated with a second protein (Enzyme 2), a dimer (86 kDa) comprised of two subunits (33 and 53 kDa). which, whilst unable to catalyse the dehalogenation reaction itself, promoted a 4-5-fold increase in the haloalcohol dehalogenation activity when combined with Enzyme 1. Although Enzyme 1 was capable of the direct conversion of 3CPD, this was the least efficient pathway. The Enzyme 1 and Enzyme 2 complex, the more effective “dehalogenase”, converted the 3CPD to acetic and formic acids via hydroxyacetone and formaldehyde intermediates at rates that were 4-5 times more rapid than Enzyme 1 alone. The Km of the “combined” enzyme for 3CPD was 322 pM with a Vmx of 3.34 pmol min-’ mg-’.
4. DEHALOGENATION OF HALOGENATED ALKANES
Haloalkanes are significant environmental compounds, occurring as both natural products (Gschwend etal., 1985)and as xenobiotic compounds (Keith andTelliard, 1979). Their biodegradability was uncertain until Omori and Alexander (1978a,b) reported that about 1% of soil microorganisms were able to utilize 1,9-dichlorononane under aerobic conditions. Brunner et a/. (1980) showed that dichloromethane was fully mineralized by microbial populations, and Wilson and Wilson (1985) showed that trichloroethylene was completely mineralized by soil populations when also provided with methane under aerobic conditions. Murphy and Perry (1983, 1984) isolated fungi with an ability to use 1-chlorooctadecane and I-chlorohexadecane aerobically, and also showed that the fatty acids of these organisms were heavily halogcnated. Under anaerobic conditions, methanogenic bacteria can degrade a wide range of halogenated alkanes, such as chloroform, carbon tetrachloride and dichloroethylene (Bouwer et al., 1981; Bouwer and McCarty, 1983a,b). Polychlorinated compounds were reductively dechlorinated to less heavily chlorinated products. Belay and Daniels (1987) showed that halogenated compounds were toxic to methanogenic bacteria and that brominated compounds were more toxic than
160
J. HOWARD SLATER
eta/.
chlorinated compounds. Vogel and McCarty (1985) proposed a series of reductive dehalogenations to mineralize tetrachloroethylene completely via trichloroethylene, dichloroethylene and vinyl chloride, a route that was subsequently supported by others (Fathepure et al., 1987; Mikesell and Boyd, 1990). Fathepure and Boyd (1988) suggested that electrons released during the reduction of carbon dioxide to methane were transferred instead to the halogenated alkanes leading to the elimination of the halogen; certainly methanogenically active consortia degraded tetrachloroethylene more rapidly (Fathepure er al., 1987). The microbiological evidence is for a wide diversity of haloalkane dehalogenation mechanisms under aerobic conditions, and there seems to be a much wider diversity of systems and mechanisms involved in haloalkane dehalogenation than for either halogenated alkanoic acids or alcohols. Haloalkane degradation involving NADH-linked reactions (Omori and Alexander, 1978a,b), oxygenases (Hartmans et al., 1985, 1986; Yokota et al., 1986), glutathione (GSH)-dependent reactions (Kohler-Staub and Leisinger, 1985) (Section 4.3), as well as hydrolytic mechanisms (Janssen et al., 1985; Yokota et al., 1986) have been proposed and demonstrated. In general terms the pH activity profiles are broad with an optimum that is slightly lower than that seen for the dehalogenation of haloalkanoic acids, and the enzymes show broad substrate specificities. The most detailed knowledge is available for the hydrolytic mechanisms found in aerobic microbes. These mechanisms have been discovered and studied in Xanthobacter autotrophicus GJlO (Janssen et al., 1985, 1989; Keuning el al., 1985), various Corynebacteriwn species (Yokota et al., 1986), Arthrobacter species (Scholtz et al., 1988a), Rhodococcus species (Sallis et al., 1990) and Ancylobacter aquaticus (van den Wijngaard et al., 1992).
4.1 Haloalkane Hydrolytic Dehalogenases On the basis of substrate specificity, these enzymes presently can be divided into two classes (Slater et al., 1995). 4.1.1. Class 3R Haloalkane Hydrolytic Dehalogenases This class of enzymes found in Gram-negative bacteria shows a restricted range of substrate specificities and has been most extensively studied in Xanthobacrer autotrophicus GJlO (Janssen et al., 1989; Keuning et al., 1985). This organism, which was isolated from enrichment cultures on dichloroethane, also synthesized a Class 1L 2HAA hydrolytic dehalogenase (Section 2.1.1). The hydrolytic haloalkane dehalogenase was heat-labile and acted principally on C 1 4 4 substituted alkanes, such as dichloroethane, bromoethane, 1-chloropropane, l-chlorobutane, 1,3-dichloropropane and 3-chloropropene, yielding the corresponding alcohols,
161
MICROBIAL DEHALOGENATION
CI-$CI-CH,CI 1,2-&chloroet haw
+ HO ,
CH$I-CHO + NADH lchloroacctaldehyde
+
CH,OH-CH,CI l-chlwoethallol
+
+ NADH CH,CI-COOH rnonochloroacetic acid
H+
+
HCI
-
NAD+
+ HO , +N AP
H+
+ HO ,
:
CH,OH-COOH glycolate
+
HCI
Figure 6 Pathway of haloalkane catabolism.
such as ethanol, I-propanol and 1-butanol, as reaction products. The enzyme was expressed constitutively and the enzyme accounted for 2-3% of the total cellular protein content. Janssen et nl. (1985) proposed a simple pathway for haloalkane metabolism (Fig. 6 ) . The purified enzyme had a molecular mass of 36 kDa and functioned without any cofactors (Keuning et al., 1985). dehalogenating l-chloroalkanes up to C4 and l-bromoalkanes up to C10. The substrate affinities of the enzyme were in the millimolar range; for example, the K,,, for dichloroethane was 1.1 m ~and , the pH optimum was 8.2. Janssen et al. (1988) proposed that the enzyme catalysed a nucleophilic substitution with water, and it was also observed that the reaction was strongly inhibited by thiol-blocking reagents, possibly implicating a cysteine residue at the active site (Keuning et al., 1985). Superficially there appeared to be some relationship with 2HAA hydrolytic dehalogenases, but no immunological cross-reactions were demonstrated (Keuning ef al., 1985). Janssen et al. (1989) determined the nucleotide sequence, finding a gene that was predicted to encode for a polypeptide of 310 amino acid residues with a molecular mass of 35.1 kDa. This suggested that the native enzyme was composed of one polypeptide subunit. The three-dimensional structure of the enzyme was determined and shown to be composed of two domains (Rozeboom et al., 1988; Franken et al., 1991; Verschueren ef al., 1993a,b,c). The main domain was composed of a central eight-stranded P-sheet surrounded by a-helices, a structure which is common to many hydrolytic proteins and leads to the concept of the general a / p hydrolase. Surmounting the main domain was a second, termed the cap domain, which was composed of five a-helices linked by loops. The active site was located between the two domains and formed an internal hydrophobic cavity. X-ray crystallography at low temperatures (Verschueren et al., 1993~)revealed three key amino acid residues at the active site: Asp124, which functioned as the nucleophile; His289,
162
J. HOWARD SLATER eta/.
which was involved in the hydrolysis of the enzyme-substrate covalent intermediate; and Asp260, which stabilized the positive charge, which developed on His289 (Fig. 7). The key role of Asp1 24 was demonstrated by site-directed mutagenesis, since replacement of this residue with alanine, glycine or glutamic acid inactivated the enzyme (Pries ef al., 1994).The halide binding was achievedvia two tryptophan residues, namely, Trpl25 located in the main domain, and Trp175 located in the cap domain. Clearly the enryme is a dehalogenase and not a general broadspectrum hydrolase, which functions as a hydrolytic dehalogenase in a fortuitous manner. Van den Wijngaard et al. (1992) isolated a number of facultative methylotrophs, including various Ancylobacter aquaticus strains and Xanthobacter autotrophicus GJ11, growing on dichloroethane by virtue of a constitutive haloalkane dehalogenase. Sequence data (N-terminal amino acid analyses and polymerase chain reaction (PCR)-amplified sequences of the putative dhlA gene from these strains) showed that the enzymes were identical to the dichloroethane dehalogenase from X.aututropliicus GJlO (Janssen et al., 1985). Tardif et al. (1991) demonstrated that this dehalogenase was plasmid-encoded and therefore the evidence for horizontal gene transmission was strong (Section 2.3.1). Janssen et al. (1994) recently suggested that the 2HAA dehalogenase DehHl from Moraxellu species B, despite limited sequence similarity, may be closely related to the haloalkane dehalogenase of X. aufotrophicus GJ10. The evidence relates to the sequence similarity around the nucleophilic aspartate within the active site, and the overall structure (i.e. main and cap domains) and catalytic mechanisms. However, similar relationships can also be defined between the Xunthornoms enzyme and other enzymes, so that the relationship may reflect common ancestry with a general dfi hydrolase. These enzymes are: tetrachlorocyclohexadiene hydrolase (LinB) from I? paucirnobilis UT26; 2-hydroxy-muconic semialdehyde hydrolase (DmpD) from P srudornonas species CF6OO (Norlund and Shingler, 1990); 2-hydroxy-6-oxo-6-phenylhexa-2,4-dienoate hydrolase (BphD) from I? putida KKS 102 (Kimbara et al., 1989); 4-methyl-2-hydroxymuconic semialdehyde hydrolase (TodF) from f? putidu (Menn et al., 1991); dienelactone hydrolase (TcbE) from Pseudomonus species P51 (van der Meer et af., 1991); dienelactone hydrolase (TdfE) from A. eutrophicus (Perkins et al., 1990); and Figure 7 Catalytic mechanism of haloalkane dehalogenase. (a) The carboxylate group of Asp124 acts as a nucleophile, which displaces the halide from the substrate. The leaving group is stabilized by two hydrogens bound to the ring nitrogens of Trpl25 and Trp175. The covalent alkyl-enzyme intermediate is hydrolysed by a water molecule that is activated by His289, which acts as a base catalyst. (b) Attack by water at the carbonyl function of the ester results in hydrolysis of the intermediate, leading to incorporation of oxygen from water in the enzyme. Asp260, the third residue of the catalytic triad, may stabilize the temporary positive charge on His289. (c) After hydrolysis of the covalent intermediate, the alcohol, halide ion, and proton leave the active site. (Reproduced, with permission, from Janssen et al. Annual Review of Microbiology, Volume 48,O 1994, by Annual Reviews Inc.)
163
MICROBIAL DEHALOGENATION
(a)
<\
WX-
OTO Y-
&Tg175 H
164
J. HOWARD SIATER
et 6'1.
dieneiactone hydrolase (ClcD) from Pseudomoms species B 13 (Frantz and Chakrabarty, 1987). The relationship will depend on how far back along the evolutionary pathway one cares to take the analysis and the definition of an ancestral gene. However, this does not eliminate the possibility that there were ancestral 2HAA dehalogenases or haloalkane dehalogenases of greater significance to the subsequent evolution of these proteins than the limited retention of common sequences within the active site. Nevertheless, the further provision of sequence information will be fascinating and essential in resolving the evolutionary relationship between ail classes of dehalogenases.
4.1.2. Class 3B Haloalkane Hydrolytic Dehalogenases The second class of hydrolytic dehalogenases is exemplified by enzymes found in various rhodococci and other Gram-positive bacteria. Arthrobacter species HA1 utilized 18 different 1-chloro-, I-bromo- and I-iodoalkanes for growth, but not 1-fluoroalkanes (Scholtz et al., 1987). This organism synthesized three dehalogenases: an inducible 1-bromoalkane debrominase; an inducible 1-chlorohexane dehalogenase with a wide substrate specificity (50 substrates including C1 and C3 alkanes); and an inducible dehalogenase, which functioned with compounds with achain length greater than C9. The 1-chlorohexanedehalogenase was a monomeric protein with a molecular mass of 37 kDa and a pH optimum of 9.5. Yokota et al. (1986, 1987) isolated various strains of Corynebacterium with abilities to grow on a wide range of mono- and di-substituted chloroalkanes. Some compounds were dehalogenated only under aerobic conditions suggesting that there were at least two types of dehalogenation reaction in these organisms, namely hydrolase- and oxygenase-type mechanisms (Section 4.2). The hydrolytic dehalogenase from Corynebacleriuin species m15-3 was a single polypeptide with a molecular mass of 33-36 kDa. The enzyme hydrolysedC2 to C12 mono- and di-substituted alkanes, and appeared to be similar to the hydrolytic dehalogenase from Arfhrobucfer species HA 1 and Rhodococcus erythropolis Y2 (Sallis et al., 1990; G. Nyandoroh, A.T. Bull and D.J.Hardman, unpublished observations). Indeed the first 18 N-terminal amino acids from these three enzymes were identical, but there was no homology with the haloalkane dehalogenase from Xanrhobacter autofrophicus GJlO (Section 4.1.1). The detailed sequence information for this group is awaited with interest, since the relationship with Class 3R hydrolytic dehalogenases remains unclear.
4.2. Oxygenase-type Haloalkane Dehalogenases
Methane monooxygenases, which normally convert methane to methanol via an NADH- and 02-dependent reaction, can also catalyse the cometabolism of certain
MICROBIAL DEHALOGENATION
165
halogenated alkanes (Colby et at., 1977; Higgins ef af.. 1979; Stirling and Dalton, 1980; Patel et al., 1982; Bouwer and McCarty, 1985; Imai et al., 1986). A d i e l d et al. (1995) re-examined Rhodococcus erythropolis Y2 and found that it synthesized a second dehalogenase of the oxygenase type, which was induced by C7 to C16 I-halolakanes and by n-alkanes. The enzyme was membrane-associated and inhibited by p-mercuribenzoate sulfonate (PMS), 2,4-dinitrophenol(24DNP) and azide. The activity was much more labile than that of hydrolytic dehalogenase. The distribution of these type of dehalogenases is unknown as is their possible relationship with hydrolytic systems.
4.3. Cofactor-dependent Dehalogenases Brunner et al. (1980) isolated a bacterium, strain DM1, which grew well on 2-5 rn dichloromethane (DCM) and found that under aerobic conditions, resting cell suspensions (but not cell-free extracts) dechlorinated DCM and produced formaldehyde. Subsequently, Stucki el al. (1981) isolated a species of Hyphornicrobiurn DM2 and showed that, in cell-free extracts, GSH was an obligate requirement for activity. The reaction was similar to that reported previously in the liver (Heppel and Porterfield, 1948; Ahmed and Anders, 1976, 1978). These enzymes converted DCM to formaldehyde and two HCI stoichiometrically by nucleophilic substitution of the halogens with hydroxyl groups. The S-chloromethyl GSH conjugate was an intermediate produced by GSH-transferase and was non-enzymatically converted to S-hydroxymethyl GSH, which in turn split to yield formaldehyde and GSH. The purified enzyme from H.yphomicrobium species DM2 was a hexameric protein of molecular mass 195 kDa and subunits of 33 kDa (Kohler-Staub and Leisinger, 1985). The GSH-dependent dehalogenases from four other facultative methylotrophs were found to be similar: the first 15 amino acid residues at the N-terminus were identical and the proteins were immunologically identical (Kohler-Staub et al., 1986). Galli and Leisinger (1988) showed that the genes for the enzymes were plasmid-encoded and this probably explained the common distribution in different genera. La Roche and Leisinger (1990) sequenced the DCM dehalogenase gene, dcrnA, from Hyphomicmbium species DM4 and showed that it coded for a polypeptide of 287 amino acid residues giving a molecular mass of 37.4 kDa. Immediately upstream of the postulated -lo/-35 promotor for dcr77.A was a 1.3 kb region, which, when deleted, abolished induction by DCM and led to the constitutive synthesis of DcmA. This suggested that the regulatory gene for dcmA was in close proximity to the structural gene (Section 2.3). Interestingly, there were three highly conserved regions in the dcrd gene, which corresponded to regions in the dimeric GSH S-transferases from maize, a helminth, rat and man, suggestive of a supragene family.
166
J. HOWARD SLATER
et 6'1.
5. CONCLUSIONS In the last few years, many of the molecular details of dehalogenating mechanisms have become clearer. Similarities and differences at amino acid and DNA sequence levels are now providing valuable information about the evolutionary history between different classes of dehalogenases and different dehalogenation mechanisms. We also look forward to much greater clarity on their wider relationship to other proteins with different catalytic activities. Molecular details of the regulatory mechanisms involved in the expression of dehalogenases is also becoming available and here the wider evolutionary links, in this case with the control of nitrogen metabolism, are already apparent to some extent. At the physiological and ecological levels, molecular techniques are providing powerful tools to evaluate the distribution of dehalogenase genes in widely different populations and communities growing in diverse ecological habitats. As these studies develop, the history of the patterns of distribution and the role of transposons and plasmids in effecting horizontal gene transfer will become clearer. The process of gene cryptification and decryptification will be well worth exploring with dehalogenase systems. The practical consequences and importance of these fundamental studies also should not be underestimated. Already several major industrial processes at the level of hundreds of tonnes of product per annum are fully operational. These are leading to more efficient process design and much lower unit costs of production for high-volume products. Application of dehalogenating systems to established processes is providing significant improvements in product quality and environmental acceptability, and the idea that man can develop environmentally enhanced products is one consequence of these applied developments (Hardman et al., 1995a,b).
REFERENCES Ahmed, A.E. and Anders, M.W. (1976) Metabolism of dihalomethanes to formaldehyde and inorganic halide. 1. In v i m studies. Drug Met. Disp. 4,357-361. Ahmed, A.E. and Anders, M.W. (1978) Metabolism of dihalomethanes to formaldehyde and inorganic halide. 11. Studies on the mechanism of the reaction. Biochern. Pharmacol. 27, 202 1-2025, Allison, N., Skinner, A.J. and Cooper, R.A. (1983) The dehalogenases of a 2,2-dichloropropionate-degrading bacterium. J. Gen. Microbiol. 129, 1283-1293. Armfield, S.J., Sallis, P.J., Baker, P.B., Bull, A.T. and Hardman, D.J. (1995) Biodegradation of haloalkanes by Rhodococcus erythropolis Y2: the presence of an oxygenase-type dehalogenase complements a halidohydrolase activity. Biodegradation 6 , 237-246. Asmara, W. (1 991) Moleculru. biology of two 2-haloacid halidohydrolases. Ph.D. thesis, University of Kent at Canterbury, Canterbury.
MICROBIAL DEHALOGENATION
167
Asmara, W., Murdiyatmo. U., Baines, A.J., Bull. A.T. and Hardman, D.J. (1993) Protein engineering of the 2-haloacid halidohydrolase IVa from Pseudomonus cepuciu MBA4. Biochem. J. 292,69-74. Assis, H.M.S., Sallis, P.J.. Bull, A.T. and Hardman, D.J. (1996) Biochemical characterisation of a haloalcohol dehalogenase from Arthrobacler erithii HIOA. Enzyme Micmb. Tech. (submitted). Barth, P.T. (1988) Genetic stability and expression. In: The Release of Genetically EngineeredMicroorganisms(M. Sussman, C.H. Collins, F.A. Skinner and D.D. StewartTull. eds) pp. 239-240. Academic Press, London. Barth, P.T., Bolton, L. and Thomson, J.C. (1992) Cloning and partial sequencing of an operon encoding two Pseuriornonus pufidu haloalkanoate dehalogenases of opposite stereospecificity. J. Bacleriol. 174,2612-2619. Bartnicki, E.W. and Castro, C.E. (1969) Biodehalogenation. The pathway for transhalogenation and the stereochemistry of epoxide formation from halohydrins. Biochemistry 8, 4677-4680. Beeching, J.R., Weightman, A.J. and Slater, J.H. (1983) The formation of an R-prime carrying the fraction I dehalogenase gene from Pseudomonusputidu PP3 using the Inc P plasrnid R6844. J. Gen. Microbiol. 129,2071-2078. Belay, N. and Daniels, L. (1987) Production of ethane, ethylene and acetylene from halogenated hydrocarbons by methanogenic bacteria. Appi. Environ. Mi&biol. 53, 1604-1610. Berry, E.K.M., Skinner, A.J. and Cooper, R.A. (1976) The bacterial degradation of Dalapon. Proc. SOC.Gen. Microbiol. 4, 38-39. Berry, E.K.M., Allison, N. and Skinner, A.J. (1979) Degradation of the selective herbicide 22DCPA (Dalapon) by a soil bacterium. J. Gen. Microbiol. 110,3945. Bollag, J.M. and Alexander, M. (1971) Bacterial dehalogenation of chlorinated aliphatic acids. Soil Biol. Biochem. 3,241-243. Bouwer, E.J. and McCm-ty, P.L. (1983a) Transformation of 1- and 2-carbon halogenated aliphatic organic compounds under methanogenic conditions. Appl. Environ. Microbiol. 45,1286-1294. Bouwer, E.J. and McCarty, P.L. (1983b) Transformation of halogenated organic compounds under denitrification conditions. Appl. Environ. Microbiol. 45, 1295-1299. Bouwer, E.J. and McCarty, P.L. (1985) Ethylene dibromide transformation under methanogenic conditions. Appl. Environ. Microbiol. 50.527-528. Bouwer, E.J., Rittman, B.E. and McCarty, P.L. (1981) Anaerobic degradation ofhalogenated 1- and 2-carbon organic compounds. Environ. Sci. Technol. 15,596-599. Brokamp, A. and Schmidt, F.R.J. (1991) Survival of Alcaligenes xylosooxidms degrading 2,2-dichloropropionate and horizontal transfer of its halidohydrolase gene in a soil microcosm. CiirK Microbiol. 22,299-306. Brunner, W., Staub, D. and Leisinger, T. (1980) Bacterial degradation of dichloromethane. Appl. Environ. Microbiol. 40,950-958. Bull, A.T., Hardman, D.J. and Stubbs, B.M. (1992) Dehalogenation of organohalogencontaining compounds. European Patent 92303681.8,S April 1992. Burge, W.D. (1969) Populations of Dalapon-decomposing bacteria in soil as influenced by additions of Dalapon or other carbon sources. Appl. Microbiol. 17,545-550. Cairns, S.S. (1994) The cloning and analysis of Rhizobium dehalogenase genes. Ph.D. thesis, University of Leicester, Leicester. Castro, C.E. and B'artnicki, E.W. (1965) Biological cleavage of carbon-halogen bonds in the metabolism of 3-broniopropanol. Biochim. Biophys. Acta 100,384-392. Castro, C.E. and Bailnicki, E.W. (1968) Biodehalogenation. Epoxidation of halohydrins,
168
J. HOWARD SLATER eta/.
epoxide opening, and transdehalogenation by a Flavobacterium species. Biuchemistry 7,3213-3218. Colby, J., Stirling, D.I. and Dalton, H. (1977) The soluble methane monooxygenase of Methylococcus capsulatus (Bath), its ability to oxygenate n-alkanes, n-alkenes, ethers and alicyclic aromatic and heterocyclic compounds. Biochem. J. 165, 395-402. Commandeur, L.C.M. and Parsons, J.R. (1994) Biodegradation of halogenated aromatic compounds In: Biochemisrry of Microbial Degradation (C. Ratledge, ed.)pp. 423458. Kluwer Academic Publishers, Dordrecht. de Lorenzo, V., Herrero, M.. Metze, M. and Timmis, K.N. (1991) An upstream XylR and IHF-induced nucleoprotein complex regulates the dependent Pu promoter of TOL plasmid. EMBO J. 10, 1159-1167. Den Dooren de Jong, L.E. (1926) &idrage Tot de Kennis van HetMineralisatiepmces. Nijgh van Ditmar, Rotterdam. Dixon, R.A. (1 986) The xylABC promoter from the Pseudomonas putida TOL plasmid is activated by nitrogen regulatory genes in Escherichia coli. Mol. Gen. Genet. 203, 129-136. Dodd, I.B. and Egan, J.B. (1990) Improved detection of helix-turn-helix DNA-binding motifs in protein sequences. Nucl. Acid Res. 18,5019-5026. Drummond, M., Whitty, P. and Wootton, J. (1986) Sequence and domain relationships of ntK and nifA from Klebsiella pneumonine: homologies to other regulatory proteins. EMBO J. 5,441-447. Fathepure, B.Z. and Boyd, S.A. (1988) Dependence of tetrachloroethylene dechlonnation on methanogenic substrate consumption by Mehmosarcina sp. strain DCM. Appl. Environ. Microbiol. 54, 2970-2980. Fathepure, B.Z., Nengu, J.P. and Boyd, S.A. (1987) Anaerobic bacteria that dechlorinate perchloroethylene. Appl. Environ. Microbiol. 53,267 1-2674. Fauzi, A.M., Hardman, D.J. and Bull, A.T. (1996) Biodegradation of low concentrations of 1,3-dichloropropano1 by mono- and mixed cultures of bacteria. Appl. Microbiol. Biotechnol. (submitted). Felsenstein, J. (1989) PHYLIP-Phylogeny Inference Package (version 3.2). Cladistics 5, 164-166. Fernandez, S., Shingler, V. and de Lorenzo, V. (1994) Cross regulation by XylR and DmpR activators of PJeudomonaspu!ida suggests that transcriptional control of biodcgradative operons evolves independently of catabolic genes. J. Bacteriol. 176,5052-5058. Fowden, L. (1968) The occurrence and metabolism of carbon-halogen compounds. Pmc. Roy. Soc., Series B 171,5-18. Foy, C.L. (1 975) The chlorinated aliphatic acids. In: Herbicides, Chemistry, Degradation andModeofAction (P.C. Kearney and D.D. Kaufman, eds) pp. 399-452. Marcel Dekker, New York. Franken, S.M., RoLeboom, H.J., Kalk, K.H. and Dijkstra, B.W. (1991) Crystal structure of haloalkane dehalogenase: an enzyme to detoxify halogenated alkanes. EMBO J . 10, 1297-1302. Frantz, B. and Chakrabarty, A.M. (1987) Organization and nucleotide sequence determination of a gene cluster involved in 3-chlorocatechol degradation. Proc. Null Acad. Sci. USA 84,4460-4464. Galli, R. and Leisinger, T. (1988) Plasmid analysis and cloning of the dichloromethaneutilization genes of Methylobacterium species DM4. J. Gen. Microbiol. 134,943-952. Gold, L. (1988) Post transcriptional regulatory mechanisms in Escherichia coli. Annu. Rev. Biochern. 57, 199-233. Goldman, P. (1 965) The enzymic cleavage of C-F bond in fluoroacetate. J. Biol. Chem. 240, 3434-3438.
MICROBIAL DEHALOGENATION
169
Goldman, P. and Milne, G.W.A. (1966) Carbon-fluorine bond cleavage, Studies on the mechanism of the defluorination of fluoroacetate. J. Biol. Chern. 241,5557-5559. Goldman, P., Milne, G.W.A. and Keister, D.B. (1968) Carbon-halogen bond cleavage. 111. Studies on bacterial halidohydrolases. J. Biol. Chern. 243,428-434. Gschwend, P.M., MacFarlane, J.K. and Newman, K.A. (1985) Volatile halogenated organic compounds released to seawater from temperate marine macroalgae. Science 227, 1033-1 035, Hardman, D.J. (1982) Dehalogenases in soil bacteria. Ph.D. thesis, University of Warwick, Coventry. Hardman, D.J. (1991) Biotransformation of halogenated compounds. Crit. Rev. Biolechnol. 11, 1-10. Hardman, D.J. and Slater, J.H. (1981a) Dehalogenases in soil bacteria. J. Gen. Microbiof. 123, 117-128. Hardman. D.J. and Slater, J.H. (1981b) The dehalogenase complement of a soil pseudomonad grown in closed and open cultures on haloalkanoic acids. J. Gen. Microbiol. 127, 399-405. Hardman, D.J., Bates, R., Huxley, M., Slater, J.H. and Bull AT (1995a) Clean technologies for paper chemicals: bioprocessing of wet-strength resins. Sixth International Conference on Biotechnology in the Pulp and Poper Induslry, Vienna, Austria, p. 56. Hardman, D.J., Huxley, M. and Slater, J.H. (1995b) Bioprocessing of products of the chemical industry. In: UNEP Internntionol Workshop “Biofechnologyfor Cleaner Production” (B. Kothuis, ed.) p. 27-28, Noordwijerhout. Hardman, D.I., Gowland, P.C. and Slater, J.H. (1986) Plasmid-encoded dehalogenase genes in Pseudoinonas species. Appl. Environ. Microbiol. 5 1 , 4 4 5 1 . Hartmans, S . , de Bont, J.A.M., Tramper, J. and Luyben, K.C.A.M. (1985) Bacterial degradation of vinyl chloride. Biotechnai. Lett. 7, 383-388. Hartmans, S., Schniuckle, A,, Cook, A.M. and Leisinger,T. (1986) Methyl chloridenaturally occurring toxicant and C l growth substrate. J. Gen. Microbiol. 132, 1139-1142. Hasan, A.K.M.Q., Takadn, H., Esaki, N. and Soda, K. (1991) Catalytic action of L-2-halo acid dehalogenase on long-chain L-2-haloalkanoic acids in organic solvents. Biotechnol. Bioeng. 38, 1114-1117. Heppel, L.A. and Porterfield, V.A. (1948) Enzymatic dehalogenation of certain brominated and chlorinated compounds. J. Biol. Chem. 176, 763-769. Higgins, D.G., Bleasby, A.J. and Fuchs, R. (1992) CLUSTALV: improved software for multiple sequence alignment. CABIOS 8, 189-191. Higgins, I.J., Hammond, R.C., Sariaslani, F.S., Best, D., Davies, M.M.. Tryhorn, S.E. and Taylor, F. (1979) Biotransformation of hydrocarbons and related compounds by whole organism suspension of methane-grown Methylosinus trichosporiwn OB3b. Biochem. Biophys. Res. Commun. 89,671-677. Hirsch. P. and Alexander, M. (1960) Microbial decomposition of halogenated propionic and acetic acids. Can. J. Micrabid. 6,241-249. Hoffman, EW. (1 950) Aliphatic fluorides. 11. 1 -halogeno-co-fluoroalkanes.J. Org. Chem. 15,425-434. Hope, S.J. and Slater, J.H. (1995) Cryptic dehalogenase and chloroamidase genes in Pseudoonionas yutido and the influence of environmental conditions on their expression. Arch. Microbiol. 163,57-64. Huala, E. and Ausubel, EM. (1989) The central domain of Rhizobiwn rneliloti nifA is sufficient to itctivate transcription from the R. rneliloti nifH promoter. J. Bucteriol. 171, 3354-3365. Hughes, S. (1988) Microbial growth on 3-chloropropionic acid. Ph.D. thesis, University of Wales, Cudiff.
170
J. HOWARD SLATER eta/.
I d , T., Takigawa, H., Nakagawa, S., Shen, G.-J., Kodarma, T. and Minoda, Y. (1986) Microbial oxidation of hydrocarbons and related compounds by whole cell suspensions of methane-oxidizing bacterium H-2. Appl. Environ. Microbiol. 52, 1403-1406. Janssen, D.B., Scheper, A., Dijkhuizen, L. and Witholt, B . (1985) Degradation of halogenated aliphatic compounds by Xanlhobacfer aurotrophicus GJ10. Appl. Environ. Micmbiol. 49,613-617. Janssen, D.B., Gerritse, J., Brackman, J., Kalk, C., Jager, D. and Witholt. B. (1988) Purification and characterization of a bacterial dehalogenase with activity toward halogenated alkanes, alcohols and ethers. Eur: J. Biochem. 171,67-72. Janssen, D.B., Pries, F., van der Ploeg, J., Kazemier, B., Terpstra, P. and Witholt, B. (1989) Cloning of 1,2-dichloroethane degradation genes of Xanthobacter aurotrophicus GJlO and expression and sequencing of the dhlA gene. J. Bacteriol. 171,6791-6799. Janssen, D.B., Pries, E and wan der Ploeg, J.R. (1994) Genetics and biochemistry of dehalogenahg enzymes. Annu. Rev. Microbiol. 48,163-191. Jensen. H.L. (195 1) Decomposition of chlorosubstituted aliphatic acids by soil bacteria. Can. J. Microbiol. 3, 151-164. Jensen, H.L. (1957) Decomposition of chloroorganic acids by fungi. Nufure 180, 1416. Jensen, H.L. (1959) Decomposition of chlorine-substituted organic acids by fungi. AcfaAgri. Scand. 9,421-434. Jensen, H.L. (1960) Decomposition of chloroacetates and chloropropionates by bacteria. Acfa Agri. Scond. 10.83-103. Jensen, H.L. ( I 963) Carbon nutrition of some microorganisms decomposing halogensubstituted aliphatic acids. AcfoAgri. Scand. 13, 404-412. Jones, D.H.A., Barth, P.T., Byrom, D. and Thomas, C.M. (1992) Nucleotide sequence of the structural gene encoding a 2-haloalkanoic acid dehalogenase of Pseudomonas pufida strain AJ1 and purification of the encoded protein. J. Gen. Microbiol. 138,675-683. Kasai, N., Tsujimura, K., Unoura, K. and Suzuki, T. (1990) Degradation of 2,3-dichloro-lpropanol by a Pseudomonos species. Agric. Biol. Chem. 54,3185-3190. Kasai, N., Tsujimura, K., Unoura, K. and Suzuki, T.(1992) Preparation of (S)-2,3-dichloro1-propanol by Pseudomanas sp. and its use in the synthesis of (R)-epichlorohydrin. J. Ind. Microbiol. 9,97-101. Kawasaki, H., Tone, H. and Tonomura, K. (1981a) Plasmid-determined dehalogenation of haloacetates in Moraxella species. Agric. Biol. Chern. 45, 29-34. Kawasaki, H., Tone, H. and Tononiura, K. (1981b) Purification and properties of haloacetate halidohydrolase specified by plasmid from Moraxella species strain B. Agric. Biol. Chem. 4 5 , 3 5 4 2 . Kawasaki, H.. Miyoshi, K. and Tonomura, K. (1981~)Purification, crystallisation and properties of haloacetate halidohydrolase from Pseudomonos species. Agric. Biol. Chem. 45,543-544. Kawasaki, H., Yahara, €3. and Tononiura, K. (1981d) Isolation and characterization of plasnud pUOl mediating dehalogenation of haloacetate and mercury resistance in Moraxella species B. Agric. Biol. Chem. 45, 1477-1481. Kawasaki, H., Hayashi, S., Yahara, H., Minanii. F.and Tonomura, K. (1982) Plasmid pU02 determining haloacetate dehalogenase and mercury resistance in Pseudomonas species. J. Fermen!. Techno/. 60,5-11. Kawasaki, H., Yahara, H. and Tonomura, K. (1983a) Cleavage maps of dehalogenation plasmid pUOl and its deletion derivative harboured in Moruxella species. Agric. Biol. Chem. 47,1639-1641. Kawasaki. H., Yanase, N., Yahaia. H. and Tononiura, K. (1983b) Molecular modification of a dehalogenation plasnrid originnting from a Moraxella species in a foreign host. Agric. Biol. Chein. 47. 1643-1645.
MICROBIAL DEHALOGENATION
171
Kawasaki, H., Yah'ara, H. and Tonomura, K. (1984) Cloning and expression in Escherichiu coliof the haloacetate dehalogenation genes from Morarella plasmid pUOl. Agric. Biol. Chem. 48,2627-2632. Kawasaki, H., Takao, M., Koiso, A. and Tonomura, K. (1985) Genetic rearrangement of plasmids: in vivo recombination between a dehalogenation plasmid and multipleresistance plasmid RP4 in Pseudomonos species. Appl. Environ. Microbiol. 49, 15441546. Kawasaki, H., Tsuda, K., Matsushita, I. and Tonomura, K. (1992) Lack of homology between two haloacetate dehalogenase genes encoded on a plasmid from Moruxellu species strain B. J. Cen. Microbiol. 138, 1317-1323. Kawasaki, H., Toyama, T., Maeda, T., Nishino, H.. and Tonomura, K. (1994) Cloning and sequence analysis of a plasmid-encoded 2-haloacid dehalogenase gene from Pseudom n a s putida no. 109. Biosci. Biotechnol. Biochem. 58, 160-163. Kearney, P.C. (1966) Metabolism of herbicides in soils. Adv. Chem. Ser. 60,250-262. Keamey, P.C., Kaufman, D.D. and Beall, M.L. (1964) Enzymatic dehalogenation of 2,2-dichloropropionate. Biochem. Biophys. Res. Commun. 14.29-33. Keamey, P.C., Harris, C.I.. Kaufnian, D.D. and Sheets, T.J. (1965) Behaviour and fate of chlorinated aliphatic acids in soils. Adv. Pest Cont. Res. 6, 1-30. Keith, L.H. and Telliard, W.A. (1979) Priority pollutants I: a perspective view. Envimn. Sci. Technol. 13,416-423. Kelly, M. ( I 965) Isolation of bacteria able to nietabolise fluoroacetate or fluornacetamide. Nature 208, 809-810. Keuning, S., Jnnssen, D.B. and Witholt, B. (1985) Purification and characterization of hydrolytic haloalkane dehalogenase from Xanthobacter autoirophicus GJ10. J. Bacter i d . 163.635-639. Kimbara, K., Hashimoto, T., Fukuda, M., Koana, T., Takagi, M., Oishi, M. and Vano, K. (1989) Cloning and sequencing of two tandem genes involved in degradation of 2,3dihydroxybiphenyl to benzoic acid in the polychlorinated biphenyl-degrading soil bacterium Pseudomonas sp. strain KKS 102. J. Bacteriol. 171,2740-2747. Klages, U., Krauss, S. and Lingens, F. (1983) 2-haloacid dehalogenase from a 4-chlorobenzoate-degrading Pseudonzonas species CBS3. Hoppe-Seyler 's Z. Physiol. Chem. 364, 529-5 35. Kohler-Staub, D. and Leisinger. T. (1985) Dichloromethane dehalogenase of Hyphomicrobiurn species strain DM2. J. Bocteriol. 162, 676-681. Kohler-Staub, D., Hartmans, S., Gilli, R., Suter, E and Leisinger, T. (1986) Evidence for identical dichloioniethlrne dehalogenases in different methylotrophic bacteria. J. Gen. Microbiol. 132,2837-2843. Kustu, S., North, A.K. and Weiss, D.S. (1991) Prokaryotic transcriptional enhancers and enhancer-binding proteins. Trends Biochern. Sci. 16,397-402. La Roche, S.D. and Leisinger, T. (1 990) Sequence analysis and expression of the bacterial dichloromethane dehalogenase structural gene, a member of the glutathione S-transferase supragene family. J. Bacteriol. 172, 164-1 7 1. Leigh, J.A,. Skinner, A.J. and Cooper, R.A. (1988) Partial purification, stereospecificity and stoichiometry of three dehalogenases from a Rhizobium species. FEMS Microbiol. Lett. 49,353-356. Leisinger, T. and Bader, R. (1 993) Microbial dehalogenation of synthetic organohalogen compounds: hydrolytic dehalogenases. Chimia 47, 116-121. Lien,B.C.,Cole, A.L.J.,Walker, J.R.L. and Peters, J.A. (1979)Effectofscdiumfluoroacetate ("Compound 1080") on the soil microflora. Soil Biol. Biochem. 11, 13-18. Little, M. and Williams, P.A. (1971) A bacterial halidohydrolase. Its purification, some
172
J. HOWARD SLATER
eta/.
properties and its modification by specific amino acid reagents. Eur: J. Biochem. 21, 99-109. MacGregor, A.N. (1963) The decomposition of dichloropropionate by soil micro-organisms, J. Gen. Microbiol. 30,497-501. Magee, L.A. and Colmer, A.R. (1959) Decomposition of 2,2-dichloropropionic acid by soil bacteria. Can. J. Micr-obiol. 5,255-260. Menn, F.-M., Zylstra, G.J. and Gibson, D.T. (1991) Location and sequence of the todF gene hydrolase in Pseudomonas putidu F1. encoding 2-hydroxy-6-oxohepta-2,4-dienoate Gene 104,91-94. Merrick, M.J. (1 993) In a class of its own-the RNApolymerase sigma factor d4(2) Mol. Microbiol. 10,903-909. Mikesell, M.D. and Boyd, S.A. (1990) Dechlorination of chloroform by Methanosarcina strains. Appl. Environ. Microbiol. 56, 1198-1201, Morsberger, E-M., Muller, R., Otto, M.K., Lingens, E and Klube, K.D. (1991) Purification and characterization of 2-halocarboxylic acid dehalogenase I1 from Pseudomonas species CBS3. Biol. Chem. Hoppe-Seyler 372,915-922. Motosugi, K., Esaki, N. and Soda, K. (1982a) Purification and properties of 2-haloacid dehalogenase from Pseudomonas putida. Agric. Biol. Chem. 46,837-838. Motosugi. K., Esaki, N. and Soda, K. (1982b) Bacterial assimilation of D- and L-2-chloropropionates and occurrence of a new dehalogenase. Arch. Microbiol. 131,179-183. Murdiyatmo, U. (1991) Molecular genetic analysis of a 2-haloacid halidohydrolase structural gene. P1i.D. thesis, University of Kent at Canterbury, Canterbury. Murdiyatmo, U., Asmara, W., Tsang, J.S.H., Baines, A.J., Bull, A.T. and Hardman, D.J. (1992) Molecular biology of the 2-haloacid halidohydrolase from Pseudomonas cepacia ~ 87-93. MBA4. B ~ o c I w IJ?. 284, Murphy, G.L. and Perry, J.J. (1983) Incorporation of chlorinated alkanes into fatty acids of hydrocarbon-utilizing mycobacteria. J. Bacleriol. 156, 1158-1 164. Murphy, G.L. and Perry, J.J. (1984) Assimilation of chlorinated alkanes and fatty acids. Appl. Environ. Microbiol. 160, 1171-1 174. Nagasawa, T., Nakamura. T., Yu,F., Watanabe, I. and Yamada, H. (1992) Purification and characterization of halohydrin hydrogen-halide lyase from a recombinant Escherichia coli containing the gene from a Corynebacterirrrnspecies. Appl. Microbiol. Biotechnol. 36,478482. Nakamura. T., Nagasawa, T., Yu, F., Watanabe, I. and Yamada, H. (1992) Resolution and some properties of enzymes involved in enantioselective transformation of 1.3-dichloro2-propano110 (R)-3-chloro-l,2-propanediol by Corynebacrerium species strain N-1074. J. Bacteriol. 174, 7613-7619. Nakamura, T., Nagasawa, T., Yu, F., Watanabe, I. and Yamada, H. (1994) Characterization of a novel enantioselective halohydrin hydrogen-halide-lyase. Appl. Environ. Microbiol. 60,1297-1301. Nardi-Die, V., Kurihara, T., Okamura, O., Liu, J.Q., Koshikawa, H.. Ozaki, H., Terashima, Y., Esaki, N. and Soda, K. (1994) Comparative studies of genes encoding thermostable L-2-halo acid dehalogenase from Pserrdornonus species strain YL, other dehalogenases, and two related hypothetical proteins from Escherichia coli. Appl. Environ. Microbiol. 60,3375-3380. Norlund, I. and Shingler, V. (1990) Nucleotide sequence of the mera-cleavage pathway enzymes 2-hydroxymuconic semialdehyde hydrolase from Pseudomonas CF600. Biochim. Biophys. Act0 1049,227-230. North, A.K.,Klose, K.E., Stedman, K.M. and Kustu, S. (1993)Prokaryoticenhancer-binding proteins reflect eukaryote-like modularity: the puzzle of nitrogen regulatory protein C. J. Bacter-id. 175,4267-4273.
MICROBIAL DEHALOGENATION
173
Omori, T. and Alexander, M. (1978a) Bacterial and spontaneous dehalogenation of organic compounds. Appl. Environ. Microbiol. 35.5 12-5 16. Omori, T. and Alexander, M. (1 978b) Bacterial dehalogenation of halogenated alkanes and fatty acids. Appl. Environ. Microbiol. 35, 867-871. Patel, R.N., Hou, C.T., Laskin, A.J. and Felix, A. (1982) Microbial oxidation of hydrocarbons: properties of a soluble methane monooxygenase from a facultative methaneutilizing organism, Methylobacterium species strain CRL-26. Appl. Environ. Microbiol. 44,1130-1137. Penfold, W.J. (1913) The inhibitory selective action on bacteria of bodies related to monochloracetic acid. (A contribution to the theory of cell intoxication) f. Hyg. 13, 35-48. Perkins, E.J., Gordon, M.P., Caceres, 0.and Lurquin, P.F. (1990) Organisation and sequence analysis of the 2.4-dichlorophenol hydroxylase dichlorocatechol oxidative operons of plasmid pJP4. J. Bacferiol. 172,2352-2359. Peters, R.A. (1952) Lethal synthesis. Proc. Roy. Soc. Lond. Ser B 139, 143-167. Pries, F., Kingma, J., Pentenga, M., van Pouderoyen, G., Jeronimus-Stratingh, C.M., Bruins, A.P. and Janssen, D.B. (1994) Site-directed mutagenesis and oxygen isotope incorporation studies of the nucleophilic aspartate of haloalkane dehalogenase. Biochemistry 33, 1242- 1247. Reanney, D.C., Gowland, I?C. and Slater, J.H. (1983) Genetic interactions among microbial communities. In: Microbes in Their Natural Environments (J.H. Slater, R. Whittenbury and J.W.T. Wimpenny, eds), pp. 379-421, Cambridge University Press, Cambridge. Rozeboom, H.J., Kingma. J., Janssen, D.B. and Dijkstra, B. (1988) Crystallization of haloalkane dehalogenase from Xanthobacrer aufofrophicusGJlO. J. Mol. Biol. 200, 611-612. Sallis, P.J., Armfield, S.J., Bull, A.T. and Hardman, D.J. ( I 990) Isolation and characterisation of a haloalkane halidohydrolase from Rhodococcrrs erythropolis Y2. J. Gen. Microbiol. 136,115-120. Schneider, B., MUller, R., Frank, R. and Lingens, F. (1991) Complete nucleotide sequences and comparison of the structural genes of two 2-haloalkanoic acid dehalogenases from Pseudomonas species strain CBS3. J. Bacteriol. 173,1530-1535. Schneider, B., Miiller, R., Frank, R. and Lingens, F. (1993) Site directed mutagenesis of the 2-haloalkanoic acid dehalogenase I gene from Pseudomonas sp. strain CBS3 and its effect on catalytic activities. Biol. Chem. Hoppe-Seyler 374,489-496. Scholtz, R., Leisinger, T., Suter, F. and Cook, A.M. (1987) Characterization of l-chlorohexane halidohydrolase, a dehalogenase of wide substrate range from an Arthrobacter species. J. Bacteriol. 169,5016-5021. Scholtz, R., Messi, E, Leisinger, T. and Cook, A.M. (1988) Three dehalogenases and physiological restraints in the biodegradation of haloalkanes by Arfhrobucferspecies strain HAl. Appl. Envimn. Microbiol. 54,3034-3038. Senior, E. (1977) Characterisation of a microbial community growing on the herbicide Dalapon. Ph.D. thesis, University of Kent at Canterbury, Canterbury. Senior, E., Bull, A.T. and Slater, J.H. (1976) Enzyme evolution in a microbial community growing on the herbicide Dalapon. Nafure 263,476479. Shingler, V., B'artilson, M. and Moore, T. (1993) Cloning and nucleotide sequence of the gene encoding the positive regulator (DmpR) of phenol catabolic pathway encoded by pVI150 and identification of DmpR as a member of the NtrC family of transcriptional activators. J. Bacteriol. 175, 1596-1 604. Slater, J.H. (1994) Microbial dehalogenadon of haloaliphatic compounds. In: Biochemistry ofMicrobial Degradation (C. Ratledge, ed.), pp. 379421. Kluwer Academic Publishers, Dordrecht .
174
J. HOWARD SLATER eta/.
Slater, J.H. and Bull, A.T. (1 982) Environmental microbiology: biodegradation. Phil. Trans. Roy. SOC.Lonci. 297,515-597. Slater, J.H. and Hope, S.J. (1995) Interactions between populations of Pseudomonasputidu leading to the expression of a cryptic dehalogenase gene (dehll) World J. Microbiol. Biotechnol. 11, 186-192. Slater, J.H., Weightman, A.J., Senior, E. and Bull, A.T. (1976) The dehalogenases from Pseudomonas putida. Pmc. SOC. Gen. Microbiol. 3, 103. Slater, J.H., Lovatt, D.,Weightman, A.J., Senior, E. and Bull, A.T. (1979) The growth of Pseudomonaspu/ida on chlorinated aliphatic acids and its dehalogenase activity. J. Gen. Microbiol. 114, 125-136. Slater, J.H., Weightman, A.J. and Hall, B.G. (1985) Dehalogenase genes of Pseudomoms putida PP3 on chromosonially located transposable elements. Mol. Biol. Evol. 2, 557-567. Slater, J.H., Bull, A.T. and Hardman, D.J. (1995) Microbial dehalogenation. Biodegradation 6,181-189. Smith, J.M., Harrison, K. and Colby, J. (1989a) Purification and characterization of D-2-haloacid dehalogenase from Psedomonas putida AJ1123. J. Gen. Microbiol. 136, 881-886. Smith, J.M., Harrison, K. and Colby. J. (1990) Purification and characterization of D-2-haloacid dehalogenase from Pseiuiornonasputida strain AJ1/23. J. Gen. Microbiol. 136.881-886. Smith, J.M., Harrison, K., Colby, J. and Taylor, S.C. (1989b) Determination of D-2-halopropionate dehalogenase activity from Pseudomonas putida strain AJ1/23 by ion chromatography. FEMS Microbiol. Lett. 57.71-74, Stirling, D.1. and Dalton, H. (1980) Oxidation of dimethyl ether, methyl formate and bromoethane by Methylococcus capsuiotus (Bath). J. Gen. Microbiol. 116,277-283. Strotmann, U.J., Pentenga and Janssen, D.B. (1990) Degradation of 2-chloroethanol by wild type and mutants of Pseudomonas purido US2. Arch. Microbiol. 154,294-300. Stucki, G.R. and Leisinger, T. (1983) Bacterial degradation of 2-chloroethanol proceeds via chloroacetic acid. FEMS Microbiol. Let[. 16, 123-126. Stucki, G.R., Galli, R., Ebersold, H.-R. and Leisinger, T. (1981) Dehalogenation of dichloromethane by cell extracts of Hyphomicrobium DM2. Arch. Microbiol. 130, 366-37 1 Suzuki, T. and Kasai, N. (1991) A novel method for the generation of (R)- and (S)-3-chloro1,2-propanediol by stereospecific dehalogenating bacteria and their use in the preparation of (R)- and (S)-glycidol. Bioorg. Med. Chern. Lett. 1, 343-346. Suzuki, T., Kasai, R., Yamamoto, R. and Minamiura, N. (1992) Isolation of a bacterium assimilating (R)-3-chloro-1,Z-propandiol and production of (S)-3-chloro-l,2-propandiol using microbial resolution. J. Ferment. Bioengin. 73,443448. Suzuki, T., Kasai, N., Yamamoto, R. and Minamiura, N. (1994) A novel enzymatic dehalogenation of (R)-3-chloro-l,2-propanediolin Aicaligenes sp. DS-S-7G. Appl. Microbiol. Biotechnol. 42,270-279. Tardif, G., Greer, C.W., LabbC, D. and Lau, P.C.K. (1991) Involvement of a large plasrnid in the degradation of 1,2-dichloroethane by Xanthobacter aurolmphicus. Appl. Envimn. Microbiol. 57, 1853-1857. Taylor, S.C. (1985) D-2-Haloalkanoic acid halidohydrolase. European patent no. 179603. Taylor, S.C. (1988) D-2-Haloalkanoic acid halidohydrolase. USA patent no. 47585 18. Thomas, A.W. (1 990) Analysis of a mobile genetic element from Pseudomonasputida which encodes dehalogenase functions. Ph.D. thesis, University of Wales, Cardiff. Thomas, A.W.. Slater+J.H. and Weightman, A.J. (1992a) The dehalogenase gene dehlfrom I
MICROBIAL DEHALOGENATION
175
Pseudoimnasputida PP3 is carried on an unusual genetic element designated DEH. J. Bacferiol. 174, 1932-1940. Thomas, A.W., Topping, A.W., Slater, J.H. and Weightman, A.J. (1992b) Localization and functional analysis of structural and regulatory dehalogenase genes carried on DEHfrom Pseudoinonaspufida PP3. J. Bacteriol. 174, 1941-1947. Thomas, A.W., Lewington, J., Hope, S., Topping, A.W., Weightman, A.J. and Slater, J.H. (1992~)Environmentally directed mutations in the dehalogenase system of Pseudomonasputida strain PP3. Arch. Microbiol. 158, 176-182. Thony, B. and Hennecke, H. (1989) The -24/-12 promoter comes of age. FEMSMicrobiol. Rev. 63,341-358. Tonomura, N., Futsi, F., Tanabe, 0. and Yamaoka, T, (1965) Defluorination of monofluoroacetate by bacteria. 1. Isolation of bacteria and their activity of defluorination. Agric. Biol. Biochem. 29,124-128. Topping, A.W. (1 992) An investigation into the transposition and dehalogenase functions of DEH from Pseudomonaspufida strain PP3. Ph.D. thesis, University of Wales, Cardiff. Topping, A.W., Thomas, A.W., Slater, J.H. and Weightman, A.J. (1995) The nucleotide sequence of a transposable haloalkanoic acid dehalogenase regulatory gene (DehRl) from Pseudomonos yirtida strain PP3 and its relationship with d4-dependent activators. Biodegradation 6, 247-255. Tsang, J.S.H., Sallis, P.J., Bull, A.T. and Hardman, D.J. (1988) A monobromoacetate dehalogenase from Pseudoornunas cepacia MBA4. Arch. Microbiol. 150,441446. van den Wijngaard, A.J., Janssen, D. and Witholt, B. (1989) Degradation of epichlorohydrin and halohydrins by bacterial cultures isolated from freshwater sediments. J. Gen. Microbiol. 135,2199-2208. van den Wijngaard, A.J., Reuvekamp, P.T.W. and Janssen, D.B. (1991) Purification and characterisation of haloalcohol dehalogenase from Arthrobacfer sp. strain AD2. J. Bacteriol. 173, 124-1 29. van den Wijngaard, A.J., van der Kamp, K.W.H.J., van der Ploeg, J., Pries, F., Kazemier, B. and Janssen, D.B. (1992) Degradation of 1,2-dichloroethane by Ancylobacter aquaticus and other facultative niethylotrophic bacteria. Appl. Environ. Microbioi. 58, 976-983. van der Meer, J.R., Eggen, R.I.L., Zehnder, A.J.B. and de Vos, W.M. (1991) Sequence analysis of the Pseudomoms sp. strain P51 tcb gene cluster which encodes metabolism of chlorinated catechols: evidence for specialisation of catechol 1,2-dioxygenases for chlorinated substrates. J. Bacreriol. 173,2425-2434. van der Ploeg, J., van Hall, G. and Janssen, D.B. (1991) Characterization of the haloacid dehalogenase from Xanthobacter au/otropiiicus GJ 10 and sequencing of the dhlB gene. J. Bacferiol. 173,1925-1933. van der Waarde, J.J., Kok, R. and Janssen. D.B. (1993) Degradation of 2-chloroallyl-alcohol by a Pseudoinonas species. Appl. Environ. Microbiol. 59, 528-535. Verschueren, K.H.G., Franken, S.M., Rozeboom, H.J., Kalk, K.H. and Dijkstra, B.W. (1993a) Refined X-ray structures of haloalkane dehalogenase at pH 6.2 and pH 8.2 and implications for the reaction mechanism. J. Mol. Biol. 232, 856-872. Verschueren, K.H.G., SeljCe, F., Rozeboom. H.J., Kalk, K.H. and Dijkstra, B.W. (1993b) Crystallographic analysis of the catalytic mechanism of haloalkane dehalogenase. Nafure 363,693498, Verschueren, K . l I.G., Kingma, J., Rozeboom, H.J., Kalk, K.H., Janssen, D.B. and Dijkstra, B.W. (1993c) Crystallographic and fluorescence studies of the interaction of haloalkane dehalogenase with halide ions. Studies with halide compounds reveal a halide binding site in the active site. Biochemisrry 32,9031-9037. Vogel, T.M. and McCarty, P.L. (1985) Bio~ansforniationof tetrachloroethylene, di-
176
J . HOWARD SLATER eta/.
chloroethylene, vinylchloride and carbon dioxide under methanogenic conditions. Appf. Envimn. Microbiol. 49,1080-1083. Watabe, K., Ishikawa, T., Mukohara, Y. and Nakamura, H. (1992)Cloning and sequencing of the genes involved in conversion of 5-substituted hydantoins to corresponding L-amino acids from the native plasmid of Pseudomonas sp. strain NS671. J. Bacteriol. 174,
962-969.
Weightman, A.J. (1981)The catabolism of halogenated alkanoic acids by Pseudomom pufida strains: characterisation of dehalogenase enzymes and associated functions. Ph.D. thesis, University of Warwick, Coventry. Weightman, A.J., Slater, J.H. and Bull, A.T. (1979)The partial purification of two dehalogenases from Pseudomonas putidn PP3. FEMS Micmbiol. Lett. 6,231-234. Weightman, A.J., Weightman, A.L. and Slater, J.H. (1982)Stereospecificity of 2-monochloropropionate dehalogenation by the two dehalogenases of Pseudomonasputida PP3: evidence for two different dehalogenation mechanisms. J. Gen. Micmbiol. 128,1755-
1762.
Weijers, C.A.G.M., Litjens. M.J.J. and de Bont, J.A.M. (1992)Synthesis of optically pure 1.2-epoxypropane by microbial asymmetric reduction of chloroacetone. Appl. Microbiol. Biotechnol. 38,297-300. Wilson, J.T. and Wilson, B.H. (1985)Biotransformation of trichloroethylene in soil. Appl. Envimn. Microbiol. 49,242-243. Yokota, T., Fuse, H., Omori, T. and Minoda, Y. (1986)Microbial dehalogenation of haloalkanes by oxygenase or halidohydrolase. Agric. Biol. Chem. SO, 453460. Yokota, T., Omori, T. and Kodama, T. (1987)Purification and properties of haloalkane dehalogenase from Corynebacrerium species strain m15-3. J. Bacteriol. 160,4049-
4054.
Metal-Microbe Interactions: Contemporary Approaches T. J. Beveridge'*, M. N. Hughes', H . 9 e 3 , K. T. Leung3 5 R. K. Poole4', 1. Savvaidis5, S. Silver and J. T. Trevors 1
Department of Microbiology, College of Biological Science, University of Guelph, Guelph, Canada N1G 2 WI 2 Department of Chernistiy, King's Coltege London, Strand, London WC2R 2LS, UK 3 Department of Environmental Biology, Ontario Agricultural College, University of Guelph, Guelph, Canada NIG 2WI 4 Department of Molecular Biology and Biotechnology, Krebs Institute for Biomolecular Research, The University of Shefield, Firth Court,WesternBank, Shefield SlO 2TN, UK 5 Department of Microbiology, University of Ioannina Medical School, Post Box 1186, 45110 loannina, Greece 6 College of Medicine/Department of Microbiology and Immunology, University of Illinois, M-C 790, 835 S. WolcottAve, Chicago, IL 60612, USA 1. Introduction . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Microorganisms and metals: their essential chemistry . . . . . . . . . . . . 2.1. Metals: the good, the bad and the indifferent. . . . . . . . . . . . . . 2.2. Properties of metals and their compounds . . . . . . . . . . . . . . . 3. Complexation of metal ions in media and cellular milieu . . . . . . . . . . . 3.1. Speciation, bioavailability and reactivity . . . . . . . . . . . . . . . . . 3.2. Metal limitation of growth as an experimental tool . . . . . . . . . . . 4. Analysis for total metal and for metal species . . . . . . . . . . . . . . . . . 4.1. Atomic absorption and emission spectroscopy . . . . . . . . . . . . . 4.2. Inductively coupled plasma-mass spectrometry . . . . . . . . . . . . . 4.3. Voltammetry . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Metal ion-selective electrodes . . . . . . . . . . . . . . . . . . . . . 4.5. Neutron activation analysis . . . . . . . . . . . . . . . . . . . . . . . . 4.6. Ion chromatography . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.7. Chromatography . . . . . . , . . . . . . . . . . . . . . . . . . . . . 4.8. Dye and proton displacement: semiquantitative methods of assessing metal uptake by microbial surfaces . . . . . . . . . . . . . . . . . . I
I
.
I
. . . .. . ... . . . ... .. . ... .. . . . . ... .. . ., . .. . ... . . .
178 179 179 183 185 185 188 190 192 194 1 94 195 196 197 198
..
198
.
*Contributorsare listed alphabetically 'Corresponding author ADVANCES IN MICROBIAL PHYSIOLOGY VOL 38 ISBN 0-12-027738-7
Copyright 8 1997 Academic Press Limited All rights of reproduction in any form reserved
178
5.
6.
7.
8.
T.J. BEVERIDGE e t a /.
4.9. Isotope transport assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1O.Transmission electron microscopy, energy dispersive X-ray spectroscopy. and selected-area electron diffraction . . . . . . . . . . . . . . . . . . . . . Spectroscopic techniques in the study of metals and microorganisms . . . . . . . 5.1. Characterization of metal-binding sites . . . . . . . . . . . . . . . . . . . . 5.2. Electronic spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. Vibrational spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4. Electron spin resonance spectroscopy . . . . . . . . . . . . . . . . . . . . . 5.5. Nuclear magnetic resonance spectroscopy . . . . . . . . . . . . . . . . . 5.6. Mossbauer spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . Molecular and genetic methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1. The power of molecular tools . . . . . . . . . . . . . . . . . . . . . . . . . 6.2. Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3. Gene probing/polymerase chain reaction . . . . . . . . . . . . . . . . . . . 6.4. Isolation of metal-tolerant and metal-sensitive bacteria: examples . . . . . . Case studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1. Aluminium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.lron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3. Copper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4. Zinc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5. Nickel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.6 Toxic metals and metalloids . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
199 201 205 205 205 206 208 . 208 209 210 210 211 212 213 214 214 216 221 223 224 225 231 232 232
Abbreviatioizs: CAS. chrome azurol S ; DRIFT. diffuse reflectance infrared spectroscopy; EDS. energy-dispersive X-ray spectroscopy; ESR. electron spin resonance (spectroscopy); FTIR. Fourier transform infrared (spectroscopy); ICPAES. inductively coupled plasma atomic emission spectrometry; ICPMS. inductively coupled plasma-mass spectromehy NBD.DF0. 7'.nitrobenz.2.oxa.l.3.diazol e. desferroxamine; NMR. nuclear magnetic resonance (spectroscopy); NTA. nitrilotriacetic acid; PCR. polymerase chain reaction: SAED. selected-area electron diffraction; TEM. transmission electron microscopy.
.
1 INTRODUCTION
Certain metal ions and their coordination complexes play a variety of fundamental roles in microbial growth and metabolism. while other metallic species are toxic to microbial life. The importance of these metal-microbe interactions has been recognized increasingly over the past decade and many aspects of this subject have been discussed in reviews and monographs over this time period (e.g. Hughes and Poole. 1989; Beveridge and Doyle. 1989; Gadd. 1992; Silver and Walderhaug. 1992; Poole and Hughes. 1996).
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
179
In the present article, it is hoped to assess critically the methods currently in use to study interactions between metals and microorganisms. In doing so, we aim to demonstrate the necessity of adopting a multidisciplinary approach to tackle this increasingly important and diverse area of microbiology. Commonly used methods for the determination of metal concentrations in cells (and in cell compartments or fractions), i.e. those of atomic spectroscopy, will not be considered in detail. It is probable that the issue of major concern over the application of atomic spectroscopy to metal analysis is still the preparation of samples for analysis rather than the analytical methodology itself. Radiochemical techniques are of particular value in studying the uptake and localization of metals, and, at present, sometimes offer the only realistic way of studying variations in concentrations of metals in cellular compartments (e.g. Lin et ul., 1993a,b). These methods are discussed in Section 4.9, and useful isotopes are listed in Table 2. In instances where metal deposits are formed, these may be observed by electron microscopy, although more sophisticated methods are required for identification purposes. Attention will also be given to the use of spectroscopic methods that may provide information about the ligands, geometries and, where relevant, the electronic status of metals coordinated at binding sites, sometimes in living cells. The use of non-invasive techniques is of significant mechanistic potential in that they may allow the direct study of ionic and metabolic processes in essentially unperturbed cells. Finally. applications of molecular genetics to the elucidation of aspects of metal-microbe interactions will be considered. The use of these various techniques will be illustrated in a number of case studies in Section 7 or elsewhere in the text. Studies on purified metalloenzymes and other metal-containing molecules of microbial origin will not generally be included. However, studies on extracellular precipitates produced by microbial activity or on the complexation of metal ions in supernatant solutions or growth media by release of extracellular ligands will be discussed.
2. MICROORGANISMS AND METALS: THEIR ESSENTIAL CHEMISTRY
2.1. Metals: The Good, the Bad and the Indifferent
Many workers in this field have attempted to classify metals according to their biological functions and effects. Pirt (1975) subdivided those elements that may be required for growth but this classification now requires revision. From physiological and genetic viewpoints, metals may be divided into “good ions” and “bad ions”. This terminology was initiated nearly 20 years ago (Silver, 1978; Summers and
180
T.J. BEVERIDGE eta/.
Silver, 1978) and periodically new summaries have appeared (e.g. Silver, 1982, Silver et al., 1989), most recently in separate reports on good cations (Silver, 1996) and mostly bad ions (Silver and Phung, 1996). In this review, we extend the terminology of Silver and others, and divide metals into three groups in terms of their microbiological activity. These are: (1) essential metals, namely those with known, specific biological functions; (2) toxic metals; and (3) metals adventitiously present in the cell, such as rubidium, cesium or titanium, which may accidentally contribute in an ill-defined sense to the functioning of the cell. Metals in the third class may ultimately be reclassified as essential elements, although the probability of this occurring is decreasing. These three groups we refer to as the good, the bad and the indifferent. 2.1.1. Essential Metals: The Good Metals presently recognized to have essential functions in microorganisms are Na, K, Mg, Ca, V, Mn, Fe, Co, Ni, Cu, Zn, Mo and W (see Fig. 1). The essentiality of Cr is still in dispute. Potassium and magnesium are usually regarded as intracellular bulk elements. Calcium is also required at relatively high concentrations in some cells, while the other metals can be considered as trace or ultra-trace elements. The essentiality of some of these metals has only been recognized comparatively recently. Tungsten, the most recent addition to the list, appears to be essential (as tungstate) in hyperthermophilic bacteria, such as Pymcoccusfuriosus, which are found in submarine hydrothermal vents, such as “black smokers”, and are able to grow at temperatures up to 130°C (Adams, 1993). Concentrations of tungstate are very high in these environments and it is possible that tungstate is replacing molybdate. Each of these essential elements may become toxic at high concentrations; indeed Cu(1I) compounds are notoriously toxic to microorganisms under certain conditions. A variety of roles may be associated with these essential metals (Hughes and Poole, 1989). The transition elements and zinc will usually be firmly bound to biological macromolecules, particularly proteins, and be involved in such processes as enzyme catalysis of reactions involving oxidatiordreductionor hydrolysis, the transport and storage of small molecules, such as dioxygen, and recognition. The Group 1 and 2 cations are also involved in enzyme catalysis, particularly in the case of Mg2+,but also have a variety of other roles, such as in trigger and control mechanisms, structure stabilization, charge neutralization and the control of osmotic pressure. The monovalent cations are of particular importance in the last two roles because of their very weak binding to ligands. Transport mechanisms for entry of essential metal species into microbial ceils have received considerable attention at physiological, biochemical and genetic levels (Borst-Pauels, 1981; Rosen and Silver 1987; Gadd, 1988, 1990; Gadd and White, 1989; Silver et al., 1989; Jones and Gadd, 1990; Tisa and Rosen, 1990).
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
181
Inward transport across the selectively permeable membranes of prokaryotic and eukaryotic microorganisms is essential for the controlled provision of metals, metalloids and inorganic ions. Small, uncharged and lipophilic inorganic molecules, such as dioxygen and dinitrogen, probably enter by passive diffusion down a concentration gradient. However, the transport of charged ions requires coupling of the translocation event to metabolic energy. This may be provided in the form of ATP, or cotransport with another ion (generally H+ and/or Na'), which is itself driven by a transmembrane potential. Membrane transport is achieved by integral membrane proteins; most of those studied are organized as numerous transmembrane helices, which, in a poorly understood way, provide the channel for ion movement. Many transport systems require additional, peripherally associated proteins for substrate binding or ATPhydrolysis. Study of the components of these transporters has been greatly Facilitated by the tools of molecular genetics, which allow, for example, identification of the structural genes, prediction of protein sequence and organization, overexpression of transport proteins and site-specific mutagenesis of individual amino-acid residues. The best understood transport systems are those for iron in bacteria (Neilands, 1989). The low solubility ofFe(II1) requires that the metal be bound and transported by small ligands, called siderophores, which have extraordinarily high affinity for their substrate. These are excreted from the cell, bind Fe(III), and in Gram-negative bacteria are then transported back across outer and inner membranes to the cell interior, where Fe(I1) is released. The study of siderophores has been stimulated by their significance in pathogenicity, environmental microbiology and biotechnology. The uptake of iron, probably like the transport of virtually all ions, is tightly regulated and is the sum of the action of several different transporters with different kinetic properties, The transport of potassium, another essential metal ion, is also characterized by an apparent superfluity of transport systems, which achieve the high intracellular K+ levels required for maintenance of turgor pressure. Sodium and calcium are actively expelled by most bacteria to maintain low intracellular concentrations of these ions. In the case of Na', transmembrane gradients set up by various Na+-translocatingmechanisms (e.g. ATPases and respiration-driven Na' pumps) are used to drive transport of other solutes; thus Na', like H+,is a coupling ion. Transport systems for the divalent cations Mg2+,Mn2+,Zn2+and Ni2+have been identified but are somewhat less well understood at present. Molybdenum is transported as molybdate (MOO:-) by nitrogen-fixing bacteria and enterobacteria that require the metal for anaerobic respiratory enzymes. Copper is highly reactive and its transport is integrated with intracellular systems for copper homeostasis and efflux. In addition to the membrane transporter systems required for metal ion uptake from the environment, eukaryotic microorganisms present the additional complexity of extensive intracellular compartmentation. Ion transport in and out of organelles is not only essential for organelle function but may also represent a means of regulating metal ion concentration in the cytoplasm. The predominant mechanism for these solute movements is the generation, by the activity of an
182
T.J. BEVERIDGE eta/.
H+-translocating ATPase, of an Htgradient, which “drives” the transport of other ions. A spectacular result of the ability of microorganisms to recover an inorganic element from the environment is the synthesis by diatoms of silicaceous walls of remarkable beauty.
2.1.2. Toxic Metals: The Bad The metals classed as toxic have no known essential functions, and include Ag, Cd, Sn, Au, Hg, T1, Pb, Bi and Al. Some of these metals have a long history of toxic effects. The silver cation is probably the most toxic species of all, subject to the limitations described later. Germanium, arsenic, antimony and selenium are sometimes included amongst the toxic metals, although strictly they are defined as metalloids, having properties between those of metals and non-metals. Their chemistry is anionic rather than cationic and so these metalloid elements will exert rather different effects from cationic toxic metal species. Aluminium is a recent recruit to this group of toxic metals. Its new-found notoriety is the direct consequence of the solubilization of aluminium from aluminosilicates in soil by acid rain. Soluble Al(II1) species are toxic to soil bacteria generally and specifically to rhizobial bacteria, with major consequences for crop yields in horticulture and farming (His et al., 1993). Aluminium differs significantly in chemistry from the other toxic elements. It is a hard metal in its general chemical character, in contrast to the traditional toxic metals that are all polarizable, chemically “soft” elements. The term “soft metals” carries more precise chemical information than does the expression “heavy metals”. Heavy metals are defined arbitrarily as those elements with a density greater than 5 and represent a very diverse group of metals. These classifications have been discussed by Gadd (1992). Aluminium, as a hard metal, coordinates to oxygen donor ligands, which are often negatively charged, while the other toxic elements bind preferentially to soft donor atoms such as sulfur ligands. The soft metals generally bind strongly to ligands, to an extent many orders of magnitude greater than do the essential metals. They exert toxic effects by several mechanisms, including the displacement of essential metals from their normal binding sites on biological molecules. It is essential to note that the toxicity of a metal depends upon its chemical speciation: thus organo derivatives of metals are usually more toxic than their inorganic compounds, as organo derivatives are liposoluble and bioavailable. Microorganisms have always been exposed to toxic metals. However, pollution of the environment by metals including radionuclides has increased dramatically in recent times, largely as a result of industrial activity, although agricultural products and sewage disposal also contribute (Babich and Stotzky, 1985; Gadd, 1990). The ability of microorganisms to grow in the presence of high metal concentrations may result from specific mechanisms of resistance, for example,
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
183
the rapid pumping out from the cell of toxic cations such as Cd2'. However, environmental factors, such as pH, the presence of certain anions and cations, particulate and soluble organic matter can also reduce or eliminate toxicity (Gadd and Griffiths, 1978). Tolerance may result from the intrinsic properties of the microorganism, such as the possession of an impermeable cell wall, the production of extracellular polysaccharide or the lack of specific metal transport systems (Gadd and Griffiths, 1978; Babich and Stotzky, 1985). Complexation or precipitation of metals exterior to cells can result in detoxification. These precipitation and mineral deposition reactions also occur on an immensely greater scale and are of great importance in biogeochemical cycles, for example, in microfossil and mineral formation (Beveridge, 1989a,b; Ferris et al., 1989; Mullen et al., 1989; McLean and Beveridge, 1990), iron and manganese deposition (Ghiorse, 1984), uranium and silver mineralization (Milodowski et al., 1990; Lovley et a/., 1991) and gold deposition (Watterson, 1991; Southam and Beveridge, 1994). 2.1.3. Metals without K~ZOMJFI Biological Effects: The Indifferent Cations of metals such as Rb, Cs, Sr and various transition metals (for example, titanium) are sometimes accumulated into cells by non-specific physicochemical interactions as well as specific mechanisms of sequestration or transport (Rosen and Silver, 1987; Gadd, 1988; Beveridge, 1989a,b) and may exert some indirect biological effects, for example, in replacing K ' in charge neutralization roles by Rb'or Cs' (Avery etal., 1991). The presence of these metals in cells reflects local geological or environmental circumstances. The contamination of soil by 137cs+ and hence its uptake by soiI bacteria is a direct consequence of radioactive fall-out from the Chernobyl nuclear reactor. 2.2. Properties of Metals and Their Compounds
The metallic elements currently recognized to be essentiaI for microbial activity are found in the s and d blocks of the periodic tdbk (Fig. 1). Figure 1 also shows the location of toxic metals; some of these occur in the d block but other particularly well-known examples are in thep block. The special electronic and nuclear features of each of these three groups of metals determine which instrumental techniques, if any, may be applied in the study of their interactions with microorganisms. General aspects of the solution chemistry of metal ions and complexes are discussed fully in many texts on inorganic chemistry. 2.2.1. The s Block Metals
The Group 1 and Group 2 elements readily lose the ns' and ns2 outer electrons giving M' and M" cations, respectively. These cations, especially those of Group
T.J. BEVERIDGE et ti'/.
184 r--1
I UIBa
Rb Sr
Y
Hf
ct 01 LJ Zr
TI
Figure 1 The periodic table. Essential elements are boxed in solid lines, toxic elements are encircled, and non-metals are hatched. Li and B are examples of indifferent elements. The essentiality of chromium is still in dispute.
1, bind ligands weakly owing to their poor polarizing power, which is proportional to the ratio of charge to radius. Thus, the cations Na' and K+ bind ligands exceptionally weakly, as shown quantitatively by values of formation constants. The cations Mg2+and Ca2' are more effective in binding ligands owing to their increased charge and smaller ionic radii. The biological chemistry of these four cations has been difficult to study, partly owing to their weak interactions with ligands and also to the absence of readily accessible instrumental methods. Concentrations of these metals are usually determined by atomic emission (Group 1) and absorption (Group 2) techniques, or by radiochemical methods in some cases. However, all four cations have nuclei with non-zero spins and so may in principle be studied by nuclear magnetic resonance (NMR) methods.
2.2.2. The d Block Metals These are the transition metals, defined as those giving compounds in which the metal may have partly filled d orbitals. The transition metals bind ligands much more strongly than do the s block elements and are well known for forming a wide range of coordination complexes. A number of properties arise from their partly filled d orbitals: they have coloured complexes (owing to d-d transitions) and their compounds are often paramagnetic, owing to the presence of unpaired electrons. In general, concentrations of the transition metals are more difficult to determine in biological systems than those of the Group 1and 2 elements in view of their very low values. These metals have useful electronic (and, sometimes, nuclear) properties that allow the application of additional techniques. Under some conditions, compounds of the transition metals are precipitated by microbial action and may be located by electron microscopy and X-ray methods (see Section 4.10).
2.2.3. The p Block Elements Metallic character increases down a group in the periodic table and decreases across
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
185
the periodic table. Thus, in the p block, full metallic character is sometimes only found in elements at the bottom of the group. This distinction is shown in Fig. 1. Elements such as arsenic are intermediate between metals and non-metals in their chemistry, and are sometimes called metalloids: the solution chemistry of these elements is dominated by the formation and reactions of oxyanions. Another feature of the p block elements is the inert pair effect, which results in the availability of two oxidation states, the group oxidation state and an oxidation state of two less. The stability of the lower oxidation state becomes greater as the group is descended, and will be reflected in values of reduction potentials. This will be important in the chemistry of arsenic and selenium. The p block metals and metalloids are often seriously toxic elements. They are chemically soft and so bind to sites much more strongly than do essential metals and displace them. Compounds of metalloids will interfere particularly in oxyanion biochemistry, notably that of phosphate. The competition with phosphate is shown in that these toxic effects are often alleviated by increasing phosphate concentrations in growth medium.
3. COMPLEXATION OF METAL IONS IN MEDIA AND CELLULAR MILIEU
3.1. Speciation, Bioavailability and Reactivity
This topic underlies many aspects of metal-microbe interaction (Bernhard et al., 1986; Batley, 1989; Hughes and Poole, 1991), but will only be considered here in general terms. Microorganisms are usually cultured in either complex or chemically defined media. Complex media contain ingredients, such as yeast extract, proteose peptone and amino acids, which bind metal ions (Collins and Stotzky, 1989; Ramamoorthy and Kushner, 1975).The composition of the medium can thus have a significant effect on the bioavailability and reactivity of a metal. Accordingly, the toxicity of a given metal to a given microorganism may vary on changing thegrowth medium (Babich andStotzky, 1983; Birdetal., 1985;Hughes andPoole, 1991). In complex media, a potentially toxic metal ion may be complexed strongly and so be unavailable for uptake. In contrast, in a minimal medium, that same metal ion may be fully available to exert its toxic effects. Ideally it would be useful to compare the toxicity of a metal using a fixed medium in which the bioavailability of the metal is known from consideration of the relevant formation constants. In practice, this is an unlikely option as specific microorganisms may have varied requirements in terms of growth media, while growth of the organisms may resuIt in significant changes in the medium, either through depletion of an essential component or through the synthesis and release of metal-binding ligands. Consequently, the speciation of the metal may change with cell growth, In one example,
186
T.J. BEVERIDGE eta/.
speciation of Cu(1I) changed during growth of Pseudomonas testostemni from a 1:3 Cu:NH3 complex to a 1:2 species, owing to the decrease in the medium of the concentration of the ammonium ion as it was utilized by the microorganism (Ahmad el al., 1992). This effect was noticed from observation of a colour change in the medium during growth and comparison with the d-d spectra of various aquaammine complexes of Cu(1I). The interaction of a metal cation with a buffer may affect its bioavailability (Hughes and Poole, 1991). Buffers that exhibit “negligible” metal-binding properties (Good et al., 1986) include the sulfonic acids: MES, pK, = 6.15; PIPES, PI& = 6.80; TES, pKa = 7.5; and HEPES, pKa = 7.55 (Gueffroy, 1990). In theory, the computation of the total speciation of the metal in a growth medium is possible if the appropriate formation constants are available. Such programmes (Allison et al., 1990) have been used to calculate the distribution of chemical species in natural and synthetic media (Farell et al., 1990, 1993). Computer modelling of metal speciation in growth media, conducted in parallel with toxicity tests, may be helpful in interpreting biological effects of heavy metals (Farell e f al., 1993). Such a correlation of toxicity in terms of metal speciation shows that free heavy-metal ions are often most toxic to microorganisms such as the cyanobacteria (Jardim and Pearson, 1985) and methanogens (Jarrell et al., 1987). Simpler approaches may also be helpful; thus, it may be possible to estimate the concentration (pM) of the free metal ion (i.e. the hydrated cation) or some dominant metal species, and to correlate this concentration with toxicity. However, it should be noted that a minor species may be responsible for any toxic effects. If the metal ion M is in excess over the ligand L and this ligand binds strongly, then [Mf,J = [MtOlaJ- [Ltotal].If the ligand is in excess and unprotonated, then [L]= [ML] + [L], [ML] [M,] and the pM value may be calculated from the formation constant Ki,defined below:
-
K1= [MLl/[Ml[Ll = [MtI/[Ml[L, - Mil, pM = -log [MI = log K1 + log ([L - MJ/[M,]) If the ligand is protonated, the calculation becomes L,= ML + L + HL’ + H2L2+ ...HnL”’ (Hughes and Poole, 1991). The calculation of pM in complex media will not be straightforward or accurate owing to the multiplicity of metal-ligand interactions. The pH inay also be important in determining the bioavailability of metal ions in a medium. Aqueous solutions of trivalent metal cations and small divalent metal cations may be acidic, owing to ionization of the coordinated water molecules. The resulting hydroxo complexes undergo dimerization and polymerization, through the formation of 0x0 bridges, leading to precipitation (Collins and Stotzky, 1989; Hughes and Poole, 1991). Some of these reactions are illustrated in the following equations. At higher pH values, some precipitates redissolve, giving anionic hydroxometallate complexes, e.g. Zn(0H)f.
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
M(H2O)F -----+ M(HzO)5(OH)'
M(H20)5(OH)'
187
+ Ht
- - - + M(H20)j(OH)2 + Ht
2M(H20)5(OH)' - - - --+ (H20)-,MOM(H20):
+ H2O
As the pH of solutions containing metal ions is raised, precipitation occurs in the order of the Lewis acidity of the metal ion. Fe(II1) begins to precipitate at about pH 1.5 (unless a strongly complexing ligand is present). In the case of divalent cations, precipitation occurs first with the smallest cation, i.e. with Cu(I1). Precipitation of a cation may lead to its non-availability to microorganisms. Furthermore, the coprecipitation of other metals may occur simultaneously. Frimmel and Geywitz (1987) showed that Pb2', Cu", Zn2t and CdZt coprecipitate with ferric hydroxide at pH 7.5. Lowering the pH to4.5 causes dissolutionof freshly formed ferric hydroxide flocs and remobilization of coprecipitated metals. Changes in pH may also affect the extent of protonation of ligands and therefore metal complexation. Many toxic interactions of heavy metals on fungal cells have been interpreted in terms of pH effects on metal speciation (Newby and Gadd, 1986).Fungal toxicity of Cd(I1) may increase with increasing pH (pH 5-9), possibly as a result of formation of CdOH', which may enter the cells moreeasily than does Cd(I1) (Gadd, 1993). The binding and toxicity of Al(II1) to Escherichia coli was related to pH-dependent speciation and chelation of AI(II1) in the medium (Guida et al., 1991). Inorganic anions may affect metal speciation and availability by forming precipitates. Phosphate, chloride, arsenate and sulfate may cause precipitation of cations with coprecipitation of essential trace metal ions (Hughes and Poole, 1991). Increase in chloride concentration, for example, decreased the toxicity of cadmium towards several filamentous fungi, possibly as a result of the formation of CdC12 coordination complexes (Collins and Stotzky, 1989, see also Section 6.4). Metal speciation is also influenced by cations in the microbial growth medium. Shuttleworth and Unz ( I 991) reported that Ca(I1) and Mg(I1) ions reduced the toxicities of copper, nickel and zinc to filamentous bacteria (Thiothrix species). In general, inorganic cations are considered to protect bacteria and algae against toxic effects of soft cations by competing for metal-binding sites on cell surfaces. In summary, every effort should be made to consider seriously the speciation and biological availability of metal ions in growth media to ensure adequate provision of essential metals and to allow a correct assessment of their biological activities (Gadd and Griffiths, 1978; Bcrnhard et al., 1986; Hughes and Poole, 1991). 3.1.1. Control of Calciuin Concentratioizs Using Photoreactive Ligands Calcium concentrations inside microbial cells can be controlled by the use of
188
T.J. BEVERIDGE c?t a/.
photolabile chelating ligands that change their affinity for calcium after exposure to light of certain wavelengths (for reviews, see Kaplan and Somlyo, 1989; McCray and Trentham, 1989; Kaplan, 1990). Nitr-5 and DM-nitrophen are two caged Ca2+ compounds that release Ca2+upon illumination (Adams et al., 1988; Ellis-Davies and Kaplan, 1988), while diazo-2 takes up Ca2+upon illumination (Adams et al., 1989). Tisa and Adler (1992) have established that calcium ions are involved in E. coli chemotaxis by causing nitr-5/Ca2+tobe taken up by E. coli by electroporation, followed by exposure to 370 nm radiation. This causes release of Ca2+andtumbling of the cells. These caged compounds appear to be important tools in studying the roles of calcium in cellular physiology. Extension to other biologically important metals would be of great value.
3.2. Metal Limitation of Growth as an Experimental Tool
The essentiality of bulk elements is readily determined by omitting the element from microbial growth media in either batch or continuous culture. Numerous studies have been made of the consequences of such deficiencies (Hughes and Poole, 1989). In the case of trace metals, extraction of the metal from the medium before inoculation will be required to demonstrate growth dependency. Many methods have been developed for removal of metals from medium, stemming from the work of the plant physiologist, Steinberg who, in the 1920s, developed painstaking techniques for removing trace elements. In batch cultures of microorganisms, trace metal deficiency is generally reflected in a reduction of growth rate rather than by diminished biomass concentration (pirt, 1975; Fig. 2) but this may be an insensitive measure of deficiency (Archibald and deVoe, 1978). In addition, the carry-over from the inoculum of the metal under study may mask an effect unless several subcultures in metal-deficient medium are made. In chemostat culture, a trace metal deficiency is shown by a gradual decrease in steady-state biomass as growth rate is increased (Pirt, 1975; Fig. 2). Trace elements may be removed by precipitation as insoluble hydroxides, phosphates, sulfides, carbonates, ferrocyanides, by adsorption on alumina and by Seitz filtration (for references, see Hughes and Poole, 1989). Generally, metals must be extracted with organic complexing agents in a non-aqueous phase. However, in some cases, merely adding a chelating agent (such as 2,2'-dipyridyl or 1,lO-phenanthroline) to the growth medium is used to cause metal deficiency (e.g. of iron; Privalle el af.,1993). However, this approach is generally unreliable, since more than one metal may be chelated and meaningful conclusions impossible to draw. When an extracting agent that can itself be removed (e.g. 8-hydroxyquinoline in chloroform) is used to treat medium, it is important to add back to the treated medium those metals (as high-purity reagents) that might have been removed by the chelator even though they are required for growth. Under some
189
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
-.-
A *-.
.. B, . ‘
\
\ -\
i *\
__f
U
Dilution rate (b)
Figure 2 Effects of trace element deficiency on microbial growth in (a) batch culture and (b) chemostat culture. x is biomass in batch culture and 2 is steady-state biomass in a chemostat. Curve A, optimum amount of a trace element; curve B, deficiency of a trace element. Reproduced with permission from Pirt (1975).
circumstances, it may be worthwhile to analyse bottles of reagents to ensure that any with adventitiously high levels of a contaminating metal are discarded. Nitrilotriacetic acid (NTA) is widely used as a complexing agent and in studies of growth-promoting elfects of siderophores (see Section 7.2). However, caution is necessary since NTA can act as a siderophore-like compound for Pseudomonas strains and enlzance growth (Meyer and Hohnadel, 1992). Examples of the application of methods for extraction of metals from growth media are given in Table 1. A particularly effective method for removing trace metals involves the use of an ion-exchange resin, such as Chelex-100. Since such a resin removes all cationic species, it is essential that a solution of trace elements (lacking the metal under study) is added after the Chelex treatment. For example, Ciccognani et a/. (1992) used Chelex-100 resin to treat a defined medium for E. coli, then added all trace elements except copper to give a copper-deficient medium. Iron extraction and limitation has received most attention. Filtration may effectively remove the bulk of the iron as iron phosphates (Hartmann and Braun, 19Sl), but most investigators have used a chelating agent either: (1) to extract iron from the medium prior to inoculation; or (2) as a medium component to “buffer” iron levels in the medium at concentrations that limit growth or physiological processes. Many examples are given in Hughes and Poole (1989). The difficulty of achieving iron limitation varies considerably according to the organism under study. Iron-limited growth of E. coli or Paracoccus denitrificans can be achieved by pre-extracting medium with Chelex (Hubbard et al., 1986) or by maintaining chelators in the medium (Privalle et al., 1993), whereas it has proved extremely
190
T.J. BEVERIDGE et a/.
Table 1 A survey of methods for removing metals from growth media (adapted from Hughes and Poole, 1989). Method
Metal(s)
Extraction with 8-hydroxyquinoline in chloroform Addition of 2,2‘-dipyridyl, EDDA or conalbumin
Fe (but also Cu”, Mn2+,Mont. Zn2t ) Fe
Metal-chelating resins (e.g. Chelex 100)
Fe
Extraction with niycobactin in chloroform 1,5-Diphenylthioc .bazone (dithizone) Coprecipitation with nitroso R salt
cu Other metals Fe3+
CU
Commentdapplications May achieve 3.5 nM Fe None is specific for iron, but 2,2’-dipyridyl addition is easy and convenient May achieve 0.054.3pM
60% decrease in cellular Cu Mg and Zn reduced to < pM Achieves 0.1-0.01 Cu
co
difficult to demonstrate any iron requirement for Lactobacillus plantarum (Archibald, 1983). It is claimed that the iron content of this bacterium is less than two atoms Fe per cell or 0.1 pM intracellular Fe. In contrast,Azotobacter species require unusually high levels of iron and the production of coloured siderophores, indicative of iron limitation, can be visualized in many media. The physiological consequences of iron deficiency are diverse. For example, the technique of iron limitation has been used to investigate the roles of iron in cytochrome structure and function, iron transport mechanisms, alternative pathways for respiratory electron transfer, and pathogenicity (for references, see Hughes and Poole, 1989).
4. ANALYSIS FOR TOTAL METAL AND FOR METAL SPECIES
The use of a number of the techniques discussed in this section will be illustrated in the case studies considered i n Section 7. For these techniques, limited examples only will be discussed here. It should be recalled that well-tried colorimetric methods for the analysis of metals in particular oxidation states are available. The use of colorimetric methods for aluminium is noted in Section 7.1. The reagent ferrozine is specific for Fe(I1) and is widely used for this determination (Stookey,
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
191
1970). The molar absorptivity for the Fe(I1) ferrozine complex is 27 900 M-l ctn-', confirming the sensitivity of ferrozine as an analytical reagent. In contrast the 1,lO phenanthroline complex of Fe(I1) has a molar absorptivity of 11 OOO M-' cm-'. Under some circumstances, colorimetric methods such as these are of great value. They are usually sensitive and are valuable in cases where early indication of the course of an experiment is necessary. A number of fluorescent dyes are available for the determination of calcium, including intracellular concentrations. Thus, Gangola and Rosen (1987) determined the intracellular concentration of free Ca2+in E. coli to be about 90 nM by the use of the fluorescent indicator fura-2. The analytical methods described in this section are often used with a common goal, that is the determination of the extent of metal uptake by microbial cells, either intracellularly or on the cell surface, and sometimes, in the former case, the location of metals in particular cell fractions. In some publications the extent of intracellular accumulation is determined only from measurement of the metal concentration in the supernatant solution. It is assumed that the decrease in the metal concentration in the supernatant solution corresponds to the amount of metal taken up by the cell. This is an unsafe assumption, as metals may precipitate out of solution, or be removed by binding to the walls of flasks and other containers. It is imperative that the metal content of cells is measured directly on an acid digest of the cell pellet, and that a balance is provided of the total metal added in the experiment and the sum of individual metal determinations. The total metal determined in various fractions should equal 100% of the metal added. Any experiments with unacceptable deviations from this should be disregarded. Cells to be analysed for metal uptake are usually washed with buffer to remove metal loosely associated with the cells (which is then analysed), followed by washing with 0.1 M acid to remove metal ions bound to groups on the cell surface, and finally the cells are acid-digested (Section 4.1.2). The use of the acid wash procedure is sometimes regarded as an arbitrary means of analysing for surfacebound metals, but it has been shown ( Z . Saidi, M.N. Hughes and R.K. Poole, unpublished) that a constant amount of metal ion is removed in this washing procedure (for E. coli) when dilute acid in the range 0.01-0.5 M is used. For cell pellets that had been treated with such acid washing conditions, subsequent treatment with 0.5 M acid failed to remove further amounts of metal. Occasionally accounts are published in which EDTA is used to remove surface-bound metal. However, it is well known that EDTA can abstract calcium ions from cell walls and so cause an increase in permeability. This offers a route for metal ions, once surface-bound, to enter the cell. A further complication arises in the fractionation of cells to determine the intracellular location of metal ions. It is not possible to eliminate in a straightforward manner the possibility that metal cations are mobilized during this operation and subsequently bind to new sites of higher affinity, which are exposed during the disruption or fractionation procedures ( e g Kot and Bezkorovainy, 1993). It is
192
T.J. BEVERIDGE eta/.
unwise to draw definitive conclusions on the basis of metal analysis of cell fractions.
4.1. Atomic Absorption and Emission Spectroscopy 4.1.1. Soine Cautionary Points
Certain precautions are necessary when undertaking analysis of metal ions and species at trace levels. Metals such as iron, zinc and copper are ubiquitous contaminants of reagents used in the preparation of media. A copper-free medium prepared from good-quality reagents is still likely to have copper present at 0.3 pM. Treatment of the medium with the chelator diphenylthiocarbazone reduces the copper concentration to 0.1-0.01 pM (Iwasaki eta/., 1980). Again, the analysis of bottles of reagents to identify those with high levels of a contaminating metal may be useful. In any case, the concentrations of metal present in a medium should be determined. Doubly distilled water from all-glass stills is adequate for most purposes. Metals may be leached out of glass storage vessels but the problem can be alleviated by washing the vessels in nitric acid or nitric acidsulfuric acid mixtures and, if necessary, immersing them in acid for several days. Vessels should be then washed many times with best quality distilled water. Plastic containers are less likely to cause problems through leaching of metals into solution, although some years ago certain plastic products were notorious for releasing trace amounts of cadmium. Glass and, to a much lesser extent, plastic, containers also take up metals from solution. At concentrations below micromolar, loss of metal ions to the wall of the container may be considerablewithin 30 min. Landeen etal. (1989) have illustrated this in the case of copper. 4.1.2.
Treatment of Biological Samples
Organic or biological materials in a sample must be destroyed before analysis in order to minimize matrix effects. Wet digestion techniques are most widely used in which the sainple is refluxed in a silicone oil bath (usually at about 110°C with a mixture of concentrated AnalaR quality nitric and sulfuric acids) until the sample is dissolved, evolution of the brown gas nitrogen dioxide has ceased and a clear solution is obtained. In difficult cases, and only exceptionally, perchloric acid may also be added, when great care must be exercised. The digested solution should be made up to a standard volume with distilled water. Aqueous standard solutions of metal ions may be purchased and diluted appropriately. Standards below 10 pgm-' may be unstable owing to loss of metal to the vessel wall, and so should be freshly prepared as required and their concentration checked during the course of an experiment.
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
4.1.3.
193
Atomic Absorption Spectruscopy
Flame atomic absorption spectroscopy (AAS) (with an air-acetylene flame) is a well-established and generally the most convenient analytical technique available for determination of a metal ion in solution, if the analyte concentration is above the milligram per litre range (Welz, 1985; Slavin, 1988). Precision of 1% is usually obtained if the standard solutions are carefully prepared. At concentrations in the milligram per litre level or below, the more sensitive technique of electrothermal AAS is preferred over the flame method. An important advantage of furnace AAS is the small sample size required, while direct analysis of solid samples is also possible The main problem in AAS is that of interference from chemical and biological material in the sample, which is not present in the standard solutions used for calibration. Measurements using spectral lines in the ultraviolet (UV) range may be complicated by absorption of the radiation from the line source by molecular species, thus giving incorrectly high concentrations of metal in the sample. Instruments iisually have one of various “background” correction facilities to cope with this difficulty. Chemical interferences may result from the formation of compounds in the flame that differ in volatility from the original form of the metal. This interference may be reduced by the use of a higher temperature flame or eliminated by addition of a “masking” agent as, for example, in the case of phosphate, which is eliminated by addition of excess strontium or lanthanum compounds. AAS has had many applications in studying the transport and physiological function of metal ions (Section 7; Harel-Bronstein et al., 1995). In addition, it is probably the most common method used for assessing metal-binding by microbial biomass, reflecting its general sensitivity and ease of use. A wide range of metals has been studied. One disadvantage is that the sensitivity of the technique may vary substantially from element to element and so analysis of a solution containing several elements may require different dilutions for each determination.
.
4.1.4.
Aforiiic Eiiiission Spectroscopy
This technique, often in its most basic form as a simple flame photometer, is used especially for Group 1 metals. A more sophisticated form involves the use of inductively coupled plasma atomic emission spectroscopy (ICPAES). This enables the simultaneous or rapid sequential analysis of aqueous mixtures of metal ions, with detection limits for many elements in the parts per billion range, and an unusual degree of freedom from matrix interference effects and a minimum of sample pretreatment. One of its main attributes is its excellent precision (< 1%). Mahan et al. (1989) have used ICPAES to evaluate uptake of Pb, Fe, Cu, As, Mo, V, Se, Mn by several algal strains in a multi-component matrix.
194
T.J. BEVERIDGE eta/.
4.1.5. Atomic Fluorescence Spectrometry
Atomic fluorescence spectrometry (AFS) offers a number of advantages compared to AAS, including high sensitivity, the capability of measuring the concentration of up to 12 elements simultaneously with relative freedom from spectral interference effects, and the long linear range of the calibration curves. A multi-element capability is possible because several light sources can be arranged around a flame or a plasma, and hence the fluorescence of several metals can be detected simultaneously. High sensitivity can be realized by use of a high-intensity light source because the fluorescence signal is proportional to the intensity of the light source. Unfortunately, high-intensity conventional light sources are not easily available for a wide range of metals.
4.2. Inductively Coupled Plasma-Mass Spectrometry
Inductively coupled plasma-inass spectrometry (ICPMS) can be applied to the determination of metal ions in a wide variety of substances, including biological fluids. It is not totally free of interferences and requires some sample preparation. Nevertheless, ICPMS is capable of giving detection limits between 0.01 and 0.1 mgl-' in dilute aqueous solutions, often 10-100 times more sensitive than ICPAES. Because ICPMS relies on a inass spectrometer for separation and detection of elements, reliable information can be obtained with precision and accuracy between 0.1% and 1% (e.g. Shuttleworth and Unz, 1993).
4.3. Voltammetry The study of current-potential relations in an electrolysis cell, where the current is determined by the rate of diffusion of an electroactive species, is called voltammetry. One of the electrochemical techniques used is differential pulse polarography, with detection limits around I 0-7M. Polarographic techniques also provide information on the speciation of a metal in solution as the redox potential of the metal is affected by the ligands. They also allow the ready distinction between oxidation states, for example, of arsenic. Polarography probably offers the best method for determining concentrations of As(TI1) in microbial cultures (Barrett et al., 1993). However, polarography is not sufficiently sensitive for speciation measurements of most elements. The most widely applicable electrochemical technique for trace element speciation is anodic stripping voltammetry (ASV), which involves the reduction of the metallic species at a controlled potential followed by the analytical stage i n which the metal is reoxidized. Owing to the concentration step in ASV, extremely high sensitivity can be obtained so that metal concentrations below lo-" M can be determined. The presence of oxygen in
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
195
analytical solutions is a serious interference in ASV, and care must be taken to remove it completely. Ideally, it is best to use high-quality oxygen-free nitrogen and to maintain a rapid flow of inert gas. Dissolved oxygen can cause an apparent increase in the copper and lead stripping peaks and, in unbuffered solutions, a decrease in the cadmium peak as a result of the consumption of hydrogen ion at the electrode surface. Most metal ions are reducible at the dropping mercury electrode, and multicomponent mixtures can often be analysed by selecting an appropriate electrolyte. The voltammetric behaviour of metal ions depends on the nature of the interaction between the metal ion and the ligands. Voltammetric techniques are ideally suited for characterizing the nature of metal species in solution. They have been widely used for that purpose, for example, in studies of natural water systems (Florence, 1992). Of all trace-element speciation methods available at present, electroanalytical techniques appear to provide the best opportunity for modelling bioavailability of elements and their complexes with organic and inorganic ligands. An application of the technique to studying metal uptake by formaldehydekilled biomass (Klebsiellu pneumoniae) is given by Goncalves et al. (1987). The method has the considerable advantage that the cells need not be separated from the suspending medium for assay of metal removal, making it straightforward to measure uptake at frequent intervals in kinetic studies. Furthermore, the method can be used to observe the time course of metal removal from solutions that contain many metals (I. Savvaidis, M.N. Hughes and R.K. Poole, unpublished). Savvaidis et at. (1992) demonstrated that the method could be used to measure removal of metals from solution by viable Pseuctoinonas cepacia (Fig. 3).
4.4. Metal lon-selective Electrodes
Ion-selcctive electrodes (ISEs) may be used for the determination of both cations and anions and are of wide applicability. Concentrations of metal ions in solution may be determined with a detection limit of around lo4 M. Compared to other methods of instrumental analysis, ion-selective electrodes have special features that enhance their value in metal ion analysis. Measurements can be carried out on volumes as small as 10 ml. The electrodes are simple to use, and usually are not affected by sample colour, turbidity, suspended matter or viscosity. A range of metal ISEs have been used in the determination of metal ions and have proved useful in numerous studies of metal accumulation and binding to micro- organisms. In this context, the use of ISEs is advantageous in that the extracellular free metal-ion concentration can be determined continuously after contact between the metal ions and the biomass. However, it is important to confirm for new experimental systems that the decrease in the extracellular free metal concentration determined by the use of an ISE equals the amount of metal taken up by the cell. This is complicated by the possibility that cells may release
196
T.J. BEVERIDGE
eta/.
-0.5SrnV
-0.56mV
A-
-0.45
-0.75
-0.45
0.045
-0.7s
Potential (mV)
0
0.01
0.02
0.03
0.04
6.05
ICdI (mM)
Figure 3 Use of polarography to determine the uptake of cadmium from solution by live cells without separating cells from the solution. (a) A calibration experiment obtained . The by adding to the polarograph cell increasing amounts of Cd(II), shown as m ~ (b) addition of identical amounts of metal to the polarograph cell, but containingPseudomonas cepaciu (about 5 mg ml-l). (c) The determination of Cd(I1) in the presence ( 0 )or absence (0)of cells. Unpublished data of I. Savvaidis,M.N. Hughes and R.K. Poole.
metal-binding metabolites that could also bind metal ions and so interfere with the determination of the extracellular free metal ion concentration (Cabral, 1992). Therefore, the assay of the extracellular free metal ion concentration using ISEs should be compared with an assay of the total metal ion concentration of the supernatant solution or a direct determination of the total metal content in the cells.
4.5. Neutron Activation Analysis This technique can be used for elemental analysis in general and in particular for trace element analysis. Neutron activation analysis requires access to a nuclear reactor as a source of neutrons. The technique has an excellent sensitivity (lo4 pg) and remarkable specificity, and can be used for multi-elemental analysis. However, because of its relative inaccessibility and high running costs, the technique of neutron activation analysis has been restricted to a few laboratories. The technique has been little used in studies of metals and microorganisms, probably owing to high costs, but some applications involving various organisms and metals are described in the following examples. The distribution of Al, V and Mn in the lichen Ruinulinu stenosporu was studied by Mueller et al. (1987) using thermal neutron
METAL-MICROBE INTERACTIONS:CONTEMPORARY APPROACHES
197
activation analysis. Vanadium was found to range in concentration from 2.1 to 10 mg kg-' and A1 from 900to 4500 mg kg-'. Watanabe et al. (1989) studied the effect of phosphate starvation and the subsequent uptake of phosphate on the distribution and concentrationof phosphate metabolic intermediatesand metals by the alga Heternsigma akashiwo using 31P NMR spectroscopy, electron spin resonance (ESR) spectroscopy and neutron activation. Mn and Co were taken up by the cells, whereas at the same time other metals such as Zn, As, Cu, Mg, K and Ca were excreted, suggesting the possibility of a charge-balancing mechanism. In another study (Augier et al., 1991), the elemental composition of the marine plant Posidoniu oceunica was studied by neutron activation analysis and AAS. Neutronactivation analysis revealed the presence of 23 elements, amongst which were As, Co, Cr, L a , K, Na, Rb, Se and Ag, while AAS showed the presence of Cd, Cu, Hg and Pb.
4.6. Ion Chromatography
In recent years, ion chromatography has become an important complementary technique to that of atomic spectroscopy. The metal species in a sample are separated using ion-exchange columns and an eluant containing a complexing agent, and analysed as eluted from a column by conductivity or optical methods. The metals are identified and quantified by comparison with a range of standards. Detection limits compare well in some cases with those found by flame atomic absorption techniques and may be improved by preconcentration and/or the use of a larger sample. The rapid growth of ion chromatography is due to a number of inherent advantages, including the possibility of identification of a very large number of inorganic and organic ions and of simultaneous quantificationof cations and anions. In addition, the method offers high selectivity and speed (ten ionic species can be detected within 15-20 min). Small sample volumes are needed (< 2 ml) and a broad range of concentrationscan be determined (from 1 ng ml-' to 1000 mg ml-' with no dilution). Complete automation of analysis is possible and in many cases sample pretreatment is unnecessary. The uptake of Mg(I1) and Ca(I1) by a range of bacterial exopolysaccharides (from Klebsiellu aerogenes, Zoogloea ramigera and Enterobacter strains) was studied by ion chromatography (Geddie and Sutherland, 1993). The native acetylatedpolymers showed a selectivityfor Ca(I1) > Mg(I1) > monovalent cations, whereas samples lacking acetyl groups showed a selectivity for monovalent cations > Mg(I1) > Ca(I1). A new ion chromatographictechnique has been described for the determination of tetrathionate and thiosulfate ions in natural samples and microbial cultures (Bak et al., 1993). The sulfur oxyanions were separated on a polymer-coated, silicabased anion-exchangecolumn and directly detected by UV absorption at 216 nm. The concentrations of the three anions could be quantitatively determined in less
198
T.J. BEVERIDGE
eta/.
than 10 min, with detection limits of about 0.6 pmol for tetrathionate, and of 40 and 10 pmol for trithionate and thiosulfate, respectively.
4.7. Chromatography
Chromatographictechniques using either the gas or liquid mode are a useful means of separating mixtures of metal species, whose concentrations can then be determined even at low concentrations by the use of a range of element-specific detection systems. Gas chromatography (GC) has been applied to the analysis of a wide range of metals and organometallic species. However, since only volatile species can be separated, it is necessary in many instances to convert the metal species to volatile forms prior to GC separation. In speciation studies, reduction to hydrides or alkylation are commonly used to convert both inorganic and organometallic species to volatile products. These can then be recovered by purge and trap methods for separation by selective volatilization into a suitable detection system, such as an atomic absorption spectrometer (Van Loon, 1979; Ebdon el al., 1986).GC-atomic absorption systems and applications to trace element speciation have been reviewed (Chau and Wong, 1989; Chau, 1992). It should be noted that the generation of volatile hydrides is an important technique for the determination of elements for which sensitivity is low by the standard atomization techniques. When used in conjunction with a gas or liquid separation chromatographic technique, atomic absorption spectroscopy can be used to detect elements in the presence of complex organic and biological matrices, without the necessity for cumbersome chemical separations. Both emission and fluorescence techniques, however, offer multi-element analysis, although this application is still very much limited by the high cost of the instrumentation. Chromatographic methods used in conjunction with a suitable detection system, such as an atomic absorption spectrometer, have been mostly used in studies on metal speciation of organometallic compounds (organotin, organolead, organomercury, etc.), and trace metal ions such as tin, lead, mercury, selenium and arsenic. Several methods based on coupled chromatographyand AAS for quantitative determinationof different forms of arsenic in a variety of samples have been reviewed (Ebdon et al., 1988).
4.8. Dye and Proton Displacement: Semi-quantitative Methods of Assessing Metal Uptake by Microbial Surfaces
In the first of these simple methods, a cationic dye, such as Janus Green or Methyl Violet, is used to saturate the anionic sites of the microbial cell surface (Savvaidis ef al., 1990). Dye-loaded cells are then incubated in a metal-containing solution and the amount of the dye displaced as a result of the binding of metal ions to the cell wall is measured spectrophotometrically.The dye displacement technique was
199
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
used to screen a wide range of microorganisms for metal ion binding. In general, the order of efficacy of metal ions in displacing the dye was Pb(I1) > Cd(I1) > Cu(I1) > Ni(I1) > Co(I1) and thus, for the transition metals, is in accord with the Irving-Williams series. The method is readily scaled down and allows measurement of metal-induced dye displacement from the cells in a single colony from solid medium. The technique of dye displacement is well suited for the rapid and semi-quantitative assessment of metal ion binding to microbial cell surfaces and furthermore may be used in the absence of conventional analytical facilities (Sawaidis el al., 1990). A similar principle underlies the measurement of proton displacement from cell surfaces by added metal ions. Guida et al. (1991) used a sensitive micro-pH electrode to follow the release of protons when solutions of aluminium salts were added to a suspension of E. coli cells.
4.9. Isotope Transport Assays
Rahoisotopes of metal ions (both essential and toxic) have provided the definitive means of measuring uptake and efflux of these cations, which always occurs through membrane-embedded transport proteins. Table 2 provides a list of cations that have been studied, although there are increasing difficulties with availability and cost of some of these in recent years. Although some of these radionuclides Table 2 Radionuclides of metal cations of biological interest with half-lives and modes of decay. Atomic number
11 12 19 20 25 26 27 28 29
Radionuclide 22Na 28 Mg
42K
4s~a 54M~ "Fe 59Fe %o 63Ni
Modes of decay*
PP-
EC EC
P-
@CU
67cu
30 37 38 48
65zn 86Rb (K probe) (Ca probe) lWCd (Zn probe)
*EC, orbital electron capture y, years; h, hours; d. days.
a-
EC
Half-life
2.6 y 21.3 h 12.4 h 165 d 314 d 2.7 y 45 d 5.3 y 92 Y 12.9 h 61 h 245 d 18.7 d 28 Y 410 d
200
T.J. BEVERIDGE
eta/.
emit soft beta or gamma irradiation of unfamiliar counting properties, it is a generally safe assumption that all can be counted conveniently in readily available liquid scintillation counters. Specific gamma counters or other specialized equipment are not needed, or even advantageous. The general method used in cation transport assays is to grow microbial cells, wash and suspend them in buffer of low and defined cation content, and to add low levels (submicrocurie per millilitre and micromolar) amounts of cations. Small volumes (0.1 to a few millilitres) are removed periodically and filtered through Millipore*-type nitrocellulose filter disks, of 1 cm diameter (the specific chemical composition-nitrocellulose or nylon-and manufacturer may not be of concern but will be tested by the cautious investigator). The filters are washed rapidly with one to several washes of (perhaps) 5 ml each. The washed filters can be dried and counted in standard liquid scintillation fluid or the filters can be placed directly in water-miscible scintillation fluid. For each radionuclide, the optimal counting conditions will be somewhat different. However, high gain ''tritium settings" are frequently best for soft beta- and gamma-emitters. For a few radionuclides (such as 42K and 86Rb,which emit strong beta particles) counting in water by Cerenkov radiation (again with high counter gain) is feasible. By comparing directly the radioactive counts retained on the filter with the total amount present in the volume that was filtered, counts per minute can be directly converted to amount retained. Such assays have been used with many microbes to establish the presence and properties of high1 specific transport systems for K+, Mg2+, Fe3+ and/or Fez+ (anaerobically),Cu or Cu', Mn2+,Ni2+and 2 8 , as well as to measure net efflux of Ca2' and Na' from preloaded cells. Examples of some of these applications are given in Section 7, while studies on the bacterial transport of sodium and calcium are illustrated by the following: Clostridiurn fervidus ("Na', Speelmans et al., 1993); Acetobacterium woodii ("Na', Heise et af., 1993); E. coli ("Na', HarelBronstein et al., 1995; 45Ca2+,Ohyama et al., 1994); Schizosaccharomycespombe (45Ca2+,Ghislain et al., 1990). Because of the short half-life of 42K,86Rb is sometimes used as a probe for studies of potassium transport. However, whereas some bacteria do not distinguish strictly between K' and Rb', others, including E. coli, markedly discriminate in favour of K+ over Rb'. There are also difficulties in studying magnesium as the isotope 28Mg2+is extremely expensive and has an inconveniently short half-life of 21 hours. Accordingly, 63Ni2+has been used as an analogue for Mg2+in studying the three transport systems for magnesium in Salmonella typhiinurium (Snavely et al., 1991). Ni2+ is a competitive inhibitor of Mg2' influx in all three transport systems. It should be noted, however, that Ni2+is nearly always ineffective as an activator of Mg2+-dependentenzymes despite the similarity in ionic radii of the metals. This is due to differences in the rates of ligand substitution at these two cations. Ni2+is very much less labile than is Mg2+so inhibition arises because Ni2+ fails to vacate the binding site for Mg2+over the necessary time period. These complications may be less important in the context of bacterial transport processes.
I+
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
201
4.10. Transmission Electron Microscopy, Energy-dispersive X-ray Spectroscopy, and Selected-area Electron Diffraction
Transmission electron microscopy (TEM), by its very nature, is an effective way of locating and visualizing metallic deposits, which are associated with microorganisms. As the high-energy electron beam of a TEM (usually 60-120 keV) passes through matter, it interacts with the atoms of which the matter is composed. Sometimes these high-energy electrons pass close enough to the electron shells or the nucleus of these atoms so that they are deflected from their normal trajectory; this happens most frequently with high atomic number elements. Since the molecules of cells are predominantly composed of carbon, hydrogen, nitrogen and oxygen (with smaller amounts of phosphorus and sulfur), which are elements with low atomic numbers, there is little deflection of the electron beam and unstained cells of microorganisms are difficult to visualize by this technique. To overcome this problem, electron microscopists usually coat their specimens in “pools” of heavy-metal salts to stain them negatively. Alternatively, chemically complex heavy-metal ions are applied directly to the sample to stain it positively (e.g. in thin-sectioned material). Once stained, structures about 1.0-0.5 nm can be visualized (Beveridge et al., 1994). For our purposes, the important point is that microorganisms cannot easily be visualized by TEM unless heavy elements, such as metals, are present. Therefore, if microbes have interacted with and concentrated heavy metals within their substance, these metallic deposits can readily be seen by TEM because they strongly interact with the electron beam (Fig. 4). The easiest TEM procedure for metal visualization is the “whole mount” technique (Fig. 4). Here, washed cells from suspension are layered on to a TEM grid that has been carbon- and Formvar-coated (Beveridge et al., 1994).Complexation of the metal to the cell must be strong enough to withstand the rigours of washing (for example, by centrifugation or filtration) and the metal deposit must not dissolve during these treatments. High-resistance water (ca 5 megaohm) is best for washing because solutions such as pH buffers and growth medium will contain residual electrolytes that will coat the cell once it dries on the TEM grid. These substances will increase cellular density and will be artefactual. There is usually enough substance to an entire bacterium in a whole mount to produce a TEM image. It will not be detailed but it will be enough to distinguish cell boundaries, which will aid in interpreting the location of a metal deposit (Fig. 4). Whole mounts can give only a general idea of where metal is located within the cellular substance. For greater detail, thin sections are preferred; these are thin enough (ca 60 nm) that inelastic electron scattering (i.e. chromatic aberration; Beveridge et nl., 1994) does not detract from higher resolution structural analysis. Cytoplasmic membranes, cell walls, ribosomes, etc. in their correct cellular alignments can be readily visualized (if they are stained) by this technique (Fig. 5). Unfortunately, for metal analysis, cells cannot be stained with the artificial heavy-metal stains (e.g. uranium, lead, osmium, etc.; Beveridge et al., 1994) that
202
T.J. BEVERIDGE et
a/.
Figure 4 An unstained whole mount of a sulfur-reducing bacterium that has electrondense silicate deposits associated with its surface. Scale bar: 500 nm.
are usually used. In fact, care must be taken as to how the specimen is chemically fixed before embedding in plastic; lipids, proteins, carbohydrates and nucleic acids must be chemically stabilized, the cells must be dehydrated, plastic must be infused into the specimen and then polymerized, and thin sections must be cut. The protocol in one of our laboratories (TJB) is to fix the cells with 2% (v/v) glutaraldehyde (osmium tetroxide cannot be used), to enrobe in 2% (w/v) Noble agar, which is cut into 1 mm cubes, and to process these in either Epon 812 or LR White resins (for more details, see Beveridge et al., 1994; Firtel et al., 1995). Because no stains are used, the agar cubes containing the specimen can be difficult to see in the plastic blocks and this makes trimming of the blocks for sectioning difficult. To overcome this problem, the agar blocks can be lightly dusted with carbon-black (carbon is relatively transparent in the TEM) before they are embedded in plastic. Once sectioned the cells are difficult to see, but the metal deposits “shine” (Fig. 6a). Once the unstained images have been recorded on film, the sections can be taken out of the E M and floated on 2% (w/v) uranyl acetate (Beveridge et al., 1994) to stain cell material. The sections can be re-inserted into the TEM and the cells rephotographed for comparison with the unstained images. Cryofixation is a new technique that preserves microorganisms remarkably well (Graham, 1992) and, although frozen thin sections are possible and show metal deposits accurately, sophisticated equipment and expertise are required, and few
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
203
Figure 5 A stained thin section of a cyanobacterium that has been processed by conventional methods (see Beveridge et al., 1994; Firtel el a[., 1995) so that structural relationships can be easily seen. This cell possesses carboxysomes in the middle of the cytoplasm, concentric photosynthetic membranes towards the periphery of the cytoplasm, a Gram-negative envelope and a paracrystalline S-layer on the envelope’s surface. Metal precipitates can be seen above the cell but, because the preparation has been stained with heavy-metal salts, it is impossible to determine whether or not they are natural minerals. Scale bar: 500 nm.
laboratories are capable of the procedure. Freeze-substitution cannot be used, since osmium tetroxide (an electron-dense fixative) is required in the substitution mixture (Graham and Beveridge, 1990). TEM alone cannot give compositional analyses of these metal deposits nor can it provide the identity of the mineral phase unless additional associated equipment and expertise is used. Energy-dispersive X-ray spectroscopy (EDS) can provide compositional information and this technique relies on the capture of “signature” X-rays that are emitted from the specimen as the electron beam interacts with the cells. Occasionally, the high-energy electrons of the TEM interact so strongly with the electrons of the specimen’s atoms that an electron can be knocked out of its atomic orbit (this becomes a low-energy secondary electron and is used to produce the image made by a scanning electron microscope; SEM). During the rearrangement of the remaining electrons to fill the void left by the exiting secondary electron, X-rays are given off that possess “signature energies” of the constituent element. These X-rays are collected by an EDS detector, which is attached to the
204
T.J. BEVERIDGE
et a/.
Figure 6 (a) An unstained thin section of a bacterium with a silicaceous deposit on its periphery (scale bar: 500 nrn). (b) EDS spectrum of the deposit, which shows it to consist of Si and K. The high Cu peaks are due to the TEM grid and the Cl is from the plastic in which the cell is embedded. (c) SAED pattern of the deposit, which demonstrates that this is a crystalline clay with tightly associated potassium.
TEM column, close to the specimen chamber. In general, X-rays are emitted from an area of the specimen that is twice the area irradiated by the electron beam. A multi-channel analyser separates the energies of all of the collected X-rays and an elemental spectrum is produced according to energy (eV) and peak height (concentration).Each element can be identified by its signature spectral lines (Fig. 6b). Because specimens must be mounted on TEiM grids, it is important that the composition of the TEM grid is chosen with care. Usually, copper or nickel grids are used because they are relatively inexpensive and easily available, but they cannot, of course, be used if one is interested in copper or nickel interactions with microorganisms. In this case, expensive carbon-composite, beryllium or Teflon grids must be used. The combination of TEM-EDS can both locate and characterize fine-grained metal deposits associated with microorganisms. Selected-area electron diffraction (SAED) is another powerful technique associated with TEM. For this procedure, the TEM is converted to an electron diffractometer by turning off the imaging lenses (i.e. objective and projector lenses). The metal deposit is illuminated by the electron beam and the deposit acts
METAL-MICROBE INTERACTIONS:CONTEMPORARY APPROACHES
205
as a diffraction grating. If it is crystalline, the diffracted electrons are imaged as discrete reflections on the TEM screen and recorded (Fig. 6c).Sometimes, if many crystals are present with different orientations to the beam, a powder diagram is detected. If the mineral phase is amorphous, no reflections will be seen. Each crystallinemineral phase has a “signature” atomic (or molecular) lattice that makes identificationpossible. For example, the silicate precipitate seen on the cell in Fig. 6a produces ordered reflections by SAED, proving the precipitate is crystalline, and the EDS spectrum of Fig. 6b shows high Si and K; these criteria are characteristic of an interstratified clay. The combination of TEM-EDS-SAED, therefore, allows a researcher to locate, characterizeand identify fine-grain mineral phases that are associated with the small structures of microbial cells.
5. SPECTROSCOPIC TECHNIQUES IN THE STUDY OF METALS AND MICROORGANISMS
5.1. Characterization of Metal-binding Sites
The use of instrumental techniques is invaluable in helping to characterize the structure and reactivity of metalloenzymes or metal-containing macromolecules isolated from microorganisms.However, this section focuses on the interaction and relationships between metals and whole cells, or sometimes cell fractions. Some of these instrumental techniques result from the electronic and nuclear properties of the metal centre, and give information on the symmetry of the metal-binding site, the identity of the ligands, the nature of the metal-ligand bond and redox state of the metal. The presence of a transition metal ion allows the application of d-d spectroscopy and, for paramagnetic species, ESR spectroscopy. The technique of NMR spectroscopy has been used successfully in studying the dynamic state of inorganic ions and organic metabolites in whole cells, and promises to be of immense importance in future research. 5.2. Electronic Spectroscopy
This topic includes the study of both d-d and charge-transfer electronic transitions associated with coordinated metals. In addition, electronic spectroscopy of the porphyrin group gives invaluable information on the nature and properties of haems, and will be discussed in Section 7.2. The d-d spectra of transition metals provide information on the geometry of the metal site, including any deviation from regularity of structure,but the absorption bands are usually weak in intensity. They may also, for a specific geometry,give some indication of the identity of the ligands
206
T.J. BEVERIDGE eta/.
in terms of their ability to cause splitting of the d orbitals and hence from their place in the spectrochemical series (Hughes, 1981). A straightforward example of this approach involving changes of speciation of Cu(I1) in microbiologicalmedia owing to depletion of ammonia during growth has been given in Section 3.1. An intracellular metal ion may be bound to a number of sites and so the most unambiguous use of d-d spectroscopy is concerned with the study of metal ions bound to cell surfaces. This requires the use of diffuse reflectance techniques, that is, the characterization of spectra obtained from surface-reflected optical radiation. Measurements of diffuse reflectance involve the use of an integrating sphere, usually coated with a highly reflective compound (MgO or MgCOs), which is designed to collect diffusely reflected light on a sensitive photomultiplier. Refiectance d-d spectra of transition metal ions bound to Pseudomunus cell surfaces have been interpreted in terms of the geometry and likely ligand character of the metal-binding sites (Savvaidis etul., 1988 and unpublished work). Both Co(I1) and Ni(I1) have octahedral geometries. The energies of charge-transfer bands (absorptions resulting from the excitation of electrons between ligand orbitals and metal orbitals) will depend upon the ligand if the metal is fixed, and so help to define the ligand.
5.3. Vibrational Spectroscopy Vibrational spectroscopy offers insight into the geometry and bonding arrangements of localized groups of atoms in molecules. A non-linear macromolecule with N atoms has a total of (3N - 6) vibrations, so biological molecules will have complex vibrational spectra with many fundamental vibrational frequencies plus overtone and combination bands. Nevertheless, useful information may be obtained from Raman and infrared (IR) spectroscopy, particularly with the advent of Fourier transform techniques. Fourier transform infrared (FTIR) spectra may be measured using a variety of methods for the presentation of the sample. In the normal transmission cell, the IR beam passes through the sample. Alternatively, in the attenuated total reflectance cell, the beam is passed through a zinc selenide prism and is reflected from its rear face. Spectra of whole cells may be measured readily on a few drops of a thick suspension, although the interpretation of such spectra will be complicated by the intense absorption band due to water. The observed spectrum will be the sum of the absorbances of all the components of the cell, including proteins, nucleic acids, cell walls and membranes, and so will contain broad absorbances that will be difficult to assign in a straightforward manner to specific groups. Nevertheless, strong absorptions in specific regions of the spectrum may be assigned to certain groups with confidence.
METAL-MICROBE INTERACTIONS:CONTEMPORARY APPROACHES
207
It has been demonstrated that the overall FTIR spectra of bacteria may be regarded as a “fingerprint” of a particular bacterium. A computer-based technique for the classification and identification of bacterial samples by FTIR has been established and used successfully for the rapid identification of clinical isolates (Helm et al., 1991). The range 1500-900 cm-’ has been suggested to be the most sensitive region for distinguishing between bacteria, particularly when using first or second derivatives of these spectra. Diffuse reflectance infrared spectra (DRIFT) are recorded on dried concentrated bacteria. Spectra of biofilms on stainless steel discs can also be studied. DRIFT spectra of a variety of bacteria show the presence of the strong amide I and I1 bands between 1690 and 1650, and at 1550 cm-’. respectively. Prominent bands at 3300 and 2950 cm-’ have been assigned to 0-H and C-H stretching frequencies. In other examples, strong carbonyl C-0 stretching frequencies have been seen at 1740 cm-’. FTIR spectra of bacteria, bacteria-polymer mixtures and biofilms have been measured and partly assigned by Nicholls el al.(1985). It is possible that the nature of the binding sites of surface-bound metal ions may be established by observation of shifts in band frequency or change in absorption intensity on binding of the metal ion to the cell surface (R. Hashim, M.N. Hughes and R.K. Poole, unpublished work). Major problems of resolution and sensitivity can be overcome through use of resonance Raman spectroscopy, in which the exciting frequency is tuned into an electronic transition of a chromophore within the sample. This leads to a substantial enhancement of the intensity of the Raman effect. Because of the availability of lasers accessing the entire wavelength range of 190-800 nm, it has become possible to excite selectively the characteristic resonance Raman spectrum of practically every chromophore of biological interest. Raman spectroscopy is particularly well suited for studies in aqueous solution as water does not give a Raman spectrum. Resonance Raman spectroscopy has been used to identify bacteria (for a review, see Nelson and Sperry, 1991), while carotenoid pigments in bacteria, and protein and nucleic acid spectra in other bacteria have been measured. It is also noteworthy that spectra differ with the growth phase of the culture, particularly with respect to contributions from RNA and protein groups. Applications to metal-microbe systems have not yet been substantial. Microbial manganese oxidation was demonstrated at high Mn2+concentrations (5 g 1-I) in bacterial cultures in the presence of a microalga (Greene and Madgwick, 1991). A non-axenic, acid-tolerant microalga, a Chlainydornonas sp., was found to mediate formation of manganite (MnOOH). All algal-bacterial cultures removed Mn2+ from solution, but only those with the highest removal rates formed an insoluble oxide. While the alga was an essential component of the reaction, a Pseudoinonas sp. was found to be primarily responsible for the formation of a manganese precipitate, identified by X-ray diffraction and FTIR spectroscopy. Bremer and Geesey (1991) described a model for monitoring microbiologically induced corrosion of copper. The interactions of bacteria and copper thin films were
208
T.J. BEVERIDGE eta/.
evaluated non-destructively by attenuated total reflectance FTIR. Corrosion of the copper films was associated with the accumulation of exopolysaccharide.
5.4. Electron Spin Resonance Spectroscopy
ESR (or electron paramagnetic resonance, EPR) spectroscopyinvolves the absorption of radiofrequency energy by molecules, ions or atoms possessing electrons with unpaired spins present in a magnetic field. The spectrum shows absorbance of energy from a continuous electromagnetic field against the strength of the magnetic field. The parameter g specifies the magnetic field strength at which absorption occurs. Transition metals may give broad, asymmetric ESR line shapes owing to the g-value anisotropy. The g values give information on the ligands around the paramagnetic ion and the geometry around the metal. Hypefine splitting arises from the interaction of the electron spin with the nuclear spin of the metal. For Cu(I1) with I = 3/2, the spectrum will show four (21+ 1) equally spaced lines. Superhyperfine structure arises from the interaction of the electron spin with the spins of other nuclei. This allows the specific identification of ligand donor atoms, provided that they have a non-zero nuclear spin. A nitrogen donor ligand ( I = 1) will show three lines in the superhyperfine structure. Some donor atoms have a zero nuclear spin and so do not show this effect, but this problem may be overcome by growing microbes on labelled nutrients to produce isotopically substituted molecules where the isoto has a nuclear spin. Well-known examples here involve the use of 57Fe and S In studies on Fe-S centres (Section 7.2). The presence of mononuclear dinitrosyl iron centres in Clostn'diurn sporogenes treated with nitrite or metal nitrosyl complexes is shown readily by their characteristic signals in the ESR spectra of whole cells (Cui et al., 1992).
pe.
5.5. Nuclear Magnetic Resonance Spectroscopy
NMR analysis relies on an inherent pro erty of certain nuclei, namely spin angular momentum. Isotopes of 'H, 13C, 15N, PgF and 31Pall have spin 1/2 nuclei and so give well-defined NMR signals (Sadler, 1986). Unfortunately, all quadrupolar nuclei in asymmetrical environments suffer line-broadening problems, which may make detection impossible. New developments in NMR techniques have allowed high-resolution spectra of solid samples to be obtained, includmg those of large biopolymers. High sensitivity and elimination of line-broadening are achieved using cross-polarizationandmagic angle spinning techniques. Early experiments with microorganisms involved the use of 31PNMR to measure simultaneously various polyphosphates and inorganic phosphate for yeast
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
209
(Shulman et al., 1979). The uptake of sodium, potassium, magnesium, calcium and europium ions has been investigated using 31PNMR (Ogino et al., 1983). Gilboa et al. (1991) have used 23NaNMR to distinguish the true intracellular concentration of free sodium from cell-bound sodium in the halophilic eubacterium Mbrio custicoh The intracellular concentration is only 5-20% of that in the extracellular medium. The binding of Cd(I1) to Citrobacter sp. is dependent on activity of a cell-bound phosphatase, which releases inorganic phosphate to precipitate Cdm) as cell-bound cadmium phosphate, as shown by Ebs and magic angle spinning NMR (Macaskie et al., 1987). Rayner and Sadler (1990) suggested that bacterial resistance to metal ions could be complicated by precipitation reactions in a culture medium (Luria-Bertani) used for growing P. putida after addition of 3 m~ Cd(I1). The precipitate obtained was analysed by elemental analysis, 'H and "P NMR spectroscopy and was shown to consist of cadmium phosphates together with organic material containing valineand glutamate-rich polypeptides. Precipitation significantly reduced the levels of the essential trace elements Fe and Zn in the growth medium.
5.6. Mossbauer Spectroscopy
This form of spectroscopy involves transitions that occur within the nucleus and involves the emission and absorption of gamma-rays. A standard substance is used as an emitter and the compound under study as the absorber. The sources of radiation are the excited nuclei of the same isotope formed in the course of radioactive decay. The tuning of the emitted quanta to the energy required to excite the absorber is achieved by the use of the Doppler effect through control of the relative motion of source and absorber. Thus energy differences in Mossbauer spectroscopy have units of millimetres per second. The only Mossbauer nucleus of importance in biological chemistry is the isotope 57Fe, which may be incorporated into microbial growth media. For example, growth of Thennus thermophilus in a chemically defined medium, in which the naturally occurring Fe was replaced by 57Fe,allowed detailed characterization of the cytochrome Q - U U ~ complex, particularly the ESR-silent a3 haem (for details, see Poole, 1983). Mossbauer spectroscopy has been of special value in characterizing haem proteins, including the nature of oxidized species where the porphyrin may also be oxidized. Oxidation states of iron may be characterized from 57Feisomer shifts. Mossbauer spectroscopy has been used in vivo to characterize the intracellular distribution of iron in E. coli (Matzanke et al., 1989). Two types of iron were detected in the cell spectra: hexacoordinated Fe(TT) and Fe(IIT) high-spin complexes. The Mossbauer parameters could not be attributed to cytochromes, Fe-S proteins or ferric holo-bacterioferritin.
210
T.J. BEVERIDGE eta/.
6. MOLECULAR AND GENETIC METHODS 6.1. The Power of Molecular Tools Molecular methods4efined here as molecular genetics, in v i m cloning, polymerase chain reaction (PCR) amplification and DNA sequencing, as well as related technologies, but not to be confused with classical genetic methodologies-have made a strong impact on studies of microbe-metal interactions, as on essentially all other areas of bioscience. Approximately 10% of the 4400 genes of E. coli are concerned with the physiology of inorganic compounds, largely the cations of metals and the oxyanions of metalloid elements. For a small bacterium, such as Mycoplasma, with perhaps one-ninth of the gene complement of E. coli, the percentage will be higher, since evolutionary pressure can result in loss of genes concerned with amino-acid biosynthesis or carbohydrate metabolism, but inorganic metal cations, such as iron, magnesium and potassium, must be brought into all cells and regulated. The fraction of genes needed for handling inorganic ions is smaller for a eukaryotic microbe, with a larger genome. The physiological functions controlled by genes affecting inorganic ions are few (about five) in number.
1. Uptake of cations or oxyanions across the cell membrane occurs via transport systems that consist generally of 1-5 gene products. Manganese uptake by E. coli and other bacteria appears to require a single protein; the nickel-specific transporter of E. coli involves five proteins (Silver and Walderhaug, 1992). 2. For most cations and oxyanions, highly specific ionlion exchange systems or efflux pumps that couple biochemical energy to net efflux of excessive concentrations of most inorganic ions involve separate and distinct gene products. Using E. coli again as an example, the Cor system of four proteins functions as a cobalt/magnesium exchanger and the Kef systems bring about efflux of potassium down a concentration gradient. A calciudproton exchanger maintains calcium at submicromolar levels using energy from the membrane protonmotive gradient (Silver and Walderhaug, 1992). 3. Additional proteins may be involved as intracellular binding proteins. Examples include ferritin for iron (see Section 7.2) and metallothionein for zinc. 4. Basically all genes governing physiological functions are themselves controlled by regulatory genes, whose products determine rates of mRNA production. These include KdpDE for potassium, Fur for iron and MerR for mercury as well-studied examples. There are several standard patterns of regulation (and many variations on these themes) including: negative regulation, shutting down mRNA synthesis when a nutrient inorganic ion is readily available or when a toxic ion is not present; and positive regulation, i.e. turning on synthesis at times of starvation for a needed nutrient or presence of toxic levels of a toxic metal cation.
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
21 1
5. Enzymes occur with inorganic cations or oxyanions (or small compounds containing inorganic components) as substrates. These enzymes carry out redox chemistry (oxidation or reduction of the inorganic ion) or incorporation into or release from carbon-oxygen-containing complexes. Biosynthetic enzymes synthesizing such complexes are also needed-synthesis of siderophores (Section 7.2.4) for iron chelation is an example. Thus, the 10% estimate for E. coli includes more than 100 genes with currently known products and functions. How may such genes be found? There are two possibilities. First, for most workers, searching available gene (and derived protein) sequence libraries by standard software (the current search protocol is called BLAST, and the program for bringing back information RETRIEVE) is sufficient for keeping up to date both in general and for particular metals of interest. Libraries in Europe, North America and Japan all exchange sequences frequently and are closely networked. Every “genomic” deposit [genomic meaning approximately a “top-down’’ view rather than gene by gene, so that megabases (Mb) are deposited into the library at a time, rather than kilobases (kb) of DNA sequence] has a sizeable number of metal-related genes. These include the first two bacterial genomes totally sequenced (Fleischmann et uE., 1995; Fraser et al., 1995). For many interests, it is sufficient to keep aware of what is being done elsewhere and deposited for general use. Second, for workers concerned with a particular system, for example, iron nutrition or cadmium resistance, it is possible to isolate chromosomal mutants defective in normal physiological processes or plasmids containing genes governing resistances and metabolism (Section 6.4). These mutants will in general have increased sensitivities or resistances to high levels of metal ions, or increased requirements for higher levels of inorganic ions for optimum growth. The approaches to studies of physiology of inorganic metal ions are no different in these regards than those of other aspects of metabolism and physiology. These topics are currently handled well in multi-volume soft-cover laboratory manuals. The important methods include use of transposons that insert randomly into the microbial chromosome to disrupt and “tag” specific genes. After selection for the desired class of properties (for example, the inability to grow on normally low iron concentrations) and characterization of the mutant phenotype (that is physiology of the mutant cell), the disrupted gene can be cloned (that is, “fished out” from among all other bacterial genes) using the transposon as a marker. The functional non-disrupted gene can be identified by complementation of the defect with cloned DNA fragments from the normal cell. Having the gene identified and in hand allows DNA sequencing and use of the procedures that follow. 6.2. Biosensors
Biosensors are the result of “hooking up” an element derived from a biological system, such as a microbe, to a “device” that can amplify a signal to a readily
212
T.J. BEVERIDGE eta/.
measurable scale. Biosensors frequently provide specificity and sensitivity that are not achievable by purely physical4hemical means. There are two most frequent classes of biosensors: the first involves placing an enzyme or other protein providing specificity of substrate recognition on a glass electrode that might be responding to proton concentration (i.e. pH). Such biosensors are widely used in laboratory medicine, since they can be standardized, and placed in machines that measure a wide range of metabolites and physiological ions automatically and in quick sequence. Since such bioelectrodes lack a direct genetic component, we will not consider them here. The second class of biosensor involves “hooking up” regulatory genes that sense, measure and respond to low levels (of inorganic cations or oxyanions in our case) to signal-emitting genes (most often the bacterial lux operon, the products of which emit light, a readily amplified and measured output). The specificity and sensitivity of such a sensor is determined by the physiology of the cell (these are whole cell biosensors, in contrast to the biosensor electrodes that use single purified proteins of biological origin). Mercury-specific biosensor cells emitting light have been patented in Germany and the USA (although the technology is so standard that probably only very narrow patents will survive scrutiny). These involve gene fusions with the l d B luciferase operon of the bacterium Vibrio that are transcribed on exposure of the cell to Hg2+,under control of the promoter of the bacterial mercury-resistance operon (mer) and its positively acting regulatory protein MerR. One report on similar biosensors emitting light in response to added arsenic and cadmium provides many of the general properties of these constructs (Corbisier et aE., 1993). Although biosensor technology is well developed and general, individual problems of specificity, background, sensitivity and toxicity have limited the general use of biosensor cells. We hope that will change over the next few years. 6.3. Gene ProbinZJPolymeraseChain Reaction
The related technologies of gene probing and the polymerase chain reaction have become very useful in environmental microbiology and microbial physiology. This applies equally to metal-microbe interactions as to all other aspects of microbiolo y. Gene probing utilizes the hybridization of DNA (generally radiolabelled with or with fluorescent tags) to non-radioactive total cellular DNA, either extracted from a microbial cell culture or a mixed bacterial population in an environmental sample, or for DNA remaining within a bacterial colony (“colony blotting”), or even with a fluorescent microscope for single cells in a microscope field. For cultures, colonies or environmental samples, the non-radioactive DNA is transferred (“blotted”) on to a nylon membrane, which is readily accessible for hybridization, washing and quantification. DNA/DNA hybridization is referred to as Southern blotting (after E. Southern), whereas hybridization of DNAprobes with cellular RNA is referred to as Northern blotting, a term that has gained full acceptance, although it has no relationship to compass coordinates (or a scientist).
‘4
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
213
A related hybridization technique developed for proteins, using initially 13’1labelled antibody as probe, is referred to as Western blotting. A11 three technologies have been used to analyse metal-related genes, their transcription and the protein products. In one recent example, Nakamura and Silver (1994) demonstrated that 74 of 78 environmental Bacillus isolates (which had been selected for resistance to mercury and organomercurials from the bottom mud of mercury-polluted Minamata Bay, Japan) were positive when Southern blotted with rner operon DNA probes from a well-studied mercury-resistant Bacillus from Massachusetts Bay, USA. The 74 isolates could be subdivided by further blot analysis into six classes based on minor heterogeneity in DNA sequence. PCR technology was also applied to the Bacillus genes for mercury resistance by Nakamura and Silver (1994). PCR involves the use of short synthetic priming oligonucleotides to go, by an autocatalytic process (i.e. geometrically amplifying “chain reaction”), from one or a few copies of a gene or other DNA sequence fragment to larger measurable amounts. By specific positioning of the oligomer primers on the sequence, Nakamura and Silver (1994) demonstrated that all 74 Bacillus strains had genes for mercuric reductase (Hg2+ +Hg(0)) and organomercurial lyase (phenyl-Hg++ benzene + Hg2+)enzymes of the same size, and that the 1.8 kb distance between the two genes was identical in all 74 strains. Many other examples could be cited for each metal cation whose physiological significance has been studied. Studies of gene function under differing physiological conditions (i.e. Northern blotting of starved or stressed cells) or comparing protein activity (by physiological assays) with protein abundance (by Western blot analysis) are powerful tools for studying metal- microbe interactions.
6.4. Isolation of Metal-tolerant and Metal-sensitive Bacteria: Examples There have been apparently countless examples of attempts to isolate bacteria and other microorganisms that are either unusually resistant or sensitive to metal compounds and ions in their environment. Sometimes these studies have been undertaken to isolate new species with unusual tolerances, perhaps as a result of plasmid-borne resistance genes, to investigate mechanisms of metal tolerance, or in an attempt to isolate mutants affected in metal transport and/or detoxification mechanisms. A recent example of the first case is the isolation by Pacheo et at. (1995) of lead-resistant bacteria from soil at a Mexico City freeway. Bacteria were isolated on a solid medium containing Pb at 400 mg ml-’ and further subjected to determination of the minimal inhibitory concentrations of salts of mercury, cadmium and other toxic elements. Such studies require careful consideration of the speciation of the toxic metal in the medium in use. For example, the description of bacteria unusually resistant to silver must be treated with suspicion, since the medium contained high concentrations of chloride ions. This example and others are considered in detail by Hughes and Poole (1991).
214
T.J. BEVERIDGE eta/.
Genetic dissection of copper transport systems in E. coli and Pseudomonas syringae arose from detailed studies of bacteria isolated on selective media. In agricultural areas where copper salts are sprayed on plants for the control of fungal pests, copper-resistantplant pathogenic bacteria have been found (Cooksey, 1993). In Ps.syringae strains that infect tomato in California, four plasmid-borne genes (copABCD) were implicated. Two of these encode abundant periplasmic copperbinding proteins (CopA and CopC), which sequester the metal (up to 11 atoms per polypeptide for CopA), such that the cells turn dramatically blue. The system is inducible, apparently only by copper. Similarly, the copper detoxification system of E. coli (Brown et al., 1992; Cervantes and Gutierrez-Corona, 1994) was also discovered in bacteria exposed to high environmental copper levels, in this case from pigs fed copper sulfate as a growth stimulant in Australia. A plasmid in such strains confers resistance to cupric salts owing to the presence of pco @]asmidborne copper resistance) genes. Several such genes, in an operon pcoABCDRSE, have been identified. The plasmid-encoded resistance mechanism is thought to interact with chromosomal gene products involved in normal copper metabolism. Strains mutated in these copper uptake and transport (cut) genes are copper-sensitive or dependent on small amounts of added copper in the medium, and were isolated by screening for such phenotypes. The design of suitable screens allows one to isolate either metal-resistant (see above) or metal-sensitivemutants. Thus, zinc-sensitive mutants of E. coli have been isolated after transposon mutagenesis by growth of the mutagenized population at toxic but sublethal zinc concentrations. Cycloserine was added to kill actively growing cells, and enrich the recovery of zinc-sensitive mutants (S.J.Beard, M.N. Hughes and R.K. Poole, in preparation). In such studies, assessment of metal tolerance is facilitated by gradient plates in which two opposing wedges of agar are superimposed, only one of which contains the chosen toxic agent. Similar approaches, including the use of gradient plates, for the isolation of copper-resistant and copper-sensitive mutants of Mbrio alginolyticus have been described by Harwood and Gordon (1994). Colonies from mutagenized cultures were screened by filter transfer to copper-supplemented plates containing bromocresol purple: putative copper-sensitive colonies remained purple because of decreased acid production. Selectivity and sensitivity of such screens is often limited by the ingenuity of the investigator.
7. CASE STUDIES 7.1. Aluminium
Current interest in the interaction of aluminium compounds with bacteria results from increasing concern over the effects produced by the acidification of soil by
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
215
acid rain (Flis et al., 1993). Aluminium is a major component of soils, being present in aluminosilicates and a number of other compounds. These aluminium compounds are solubilized to a certain extent under such acid soil conditions and may ultimately be present as polynuclear and mononuclear species together with various complexes containing organic ligands. As noted earlier, the toxicity of aluminium (as one or more of these compounds) to root nodule bacteria is a matter of major concern in agriculture. Aluminium may exert its toxic effects by several routes but investigations tend to be made much more complex because of speciation problems and the difficulty of assigning toxic effects to particular species (Kinraide, 1991). The composition of media used to study aluminium toxicity to microorganisms is of particular importance in view of the possibility that aluminium precipitates such as phosphates may be formed. Toxic effects may arise from interference with phosphate availability or the function of iron. However, formation constants for A?+ are substantially lower than those for Fe3+with the same ligand, probably owing to the extremely small size of A13+,which results in steric constraints in ligand binding, so that aluminium would not normally be able to compete with iron. It should be noted that the toxicity of aluminium may be alleviated by precipitation within the cell or by complexation outside the cell with a secreted polysaccharide. These possibilities have been explored. The cyanobacteriumAnabaena cylindrica accumulates aluminium rapidly via passive diffusion resulting in inhibition of growth, photosynthesis and nitrogen fixation (Pettersson et al., 1985). The use of EDS showed that the aluminium was accumulated into polyphosphate granules and into the cell walls. Growth in a phosphate-rich medium (180 p~ compared to 35 PM) resulted in higher levels of accumulation into the granules and the cell wall. The deposition of the aluminium may be a detoxification mechanism (Pettersson etal., 1985). Similar methods have been used to show thc localization of aluminium in polyphosphate granules in root nodules of soybean plants (Roth et al., 1987). Johnson and Wood (1990) have proposed that aluminium exerts its toxic effects on Rhizobiurn spp. by binding to DNA. They have found that the nucleic acid fraction extracted from lysed Al-sensitive and Al-tolerant Rhizobium strains both contain aluminium, which was determined by carbon furnace AAS. They suggest that A13+is bound to the DNA in both strains but that the Al-sensitive strain lacks a DNArepair system. Binding of A13+to the phosphate groups of DNA will increase its stability, by neutralizing negative charge, and so prevent replication. The measurement of aluminium (and other metals) in specific fractions of bacterial cells may lead to misleading results owing to relocation of the aluminium during the fractionation process. However, this is probably less likely to occur for aluminium than for other metal ions owing to the inertness of Al” with respect to ligand substitution. In a report discounting a role for extracellular polysaccharide in aluminium tolerance of R. leguminosarurn, aluminium concentrations were determined by a colorimetric method using eriochrome cyanine R (Kingsley and Bohlool, 1992). The analysis of aluminium by carbon furnace AAS or nitrous oxide-acetylene flame
216
T.J. BEVERIDGE
eta/.
AAS is time-consuming in view of the acid digestion step for samples. The use of one of the many very sensitive colorimetric methods for aluminium may give similar sensitivity as AAS methods but give results much more rapidly. Provided the chosen method is carefully checked for reproducibility, this may prove to be the method of choice. The appearance of extracellular polysaccharide in a culture medium after rowth of an organism in the presence of aluminium could be checked by the use of A1NMR, which may show the presence of a new aluminium species.
?
7.2. Iron
7.2.1. Introduction Of the metals, only aluminium is more abundant than iron in the earth’s crust. In addition to its two principal oxidation states, ferrous Fe(I1) (d6)and ferric Fe(1II) (d5),iron has other oxidation states, of which ferry1 Fe(1V) is frequently seen in bioinorganic reaction mechanisms. The complexes of Fe(I1) and Fe(II1) readily undergo electron transfer reactions, a property that is critical in understanding the types of analyses that have proved most useful in understanding the interactions between iron and microorganisms.
7.2.2. Assays of Iron Contents of Cells and Media Despite the abundance of iron in the biosphere, microorganisms generally require specialized mechanisms for acquiring iron from their environment, where the solubility of iron is low. Thus, not only is limitation of growth by iron concentration difficult to demonstrate (Section 3.2), but intracellular iron concentrations require sensitive analytical methods. In E. coli, for example, total iron content has been given as 59-168 pg g-’ cells, non-haem iron 3.8-15 pg g-’ cells and haem iron 0.52-0.56 pg g-’ cells (Matzanke et al., 1989). The bulk of the iron is present as a pool of uncertain nature. Contributing to this pool of “other” iron is that stored in ferritin or bacteriofenitin. Both are believed to act as iron storage proteins. Although much has been learned of their structure (e.g. Andrews et al., 1993), this work has been conducted on purified material and is outside the scope of this review. The method most commonly used to determine iron levels in media and cells is AAS (see Section 4.1.3;e.g. Hubbard et al., 1986) or uptake of 59Fe(see Section 7.2.3; Kot and Bezkorovainy, 1993).
METAL-MICROBE INTERACTIONS:CONTEMPORARY APPROACHES
7.2.3.
21 7
Transport of Iron
Aconvenient method for assay of iron transport is the use of 59Fe(see Section 4.9). Details of materials and procedures were carefully considered by Rosenberg (1979). An example of their use is the work of Adams et al. (1990), who used 59Fe 59 2+ to characterize iron acquisition systems in Listeria monocytogenes. Fe and 59 3+ Fe were prepared from 59FeC13and the labelled compounds added to washed cells. Samples were removed at intervals and the labelled cells collected on 59 2+ 59 2+ 0.45 pm filters for counting. L monocytogenes bound both Fe and [ Fe Icit59 3+ rate, but was unable to acquire iron from [ Fe IEDTA or 59FeC13.Iron uptake to levels of 600 pmol mg-' dry wet cells was obtained. A novel method for measuring Fe(II1) and its uptake or mobilization from cells uses NBD-DFO, the 7'-nitrobenz-2-oxa-l,3-diazole(NBD) derivative of the siderophore (see Section 7.2.4) desferrioxamine B (DFO) (Lytton et al., 1992).The compound partitions readily from aqueous media into n-octanol and its fluoresence is quenched on binding Fe(II1). It has been used to monitor iron removal from ferritin or siderophore-Fe(II1) complexes, and has potential for monitoring chelatable iron under conditions of iron-mediated cell damage or iron imbalance. 7.2.4. Siderophores Siderophores are low-molecular-mass compounds that coordinate Fe3+and are excreted under iron-limiting conditions by many microorganisms, particularly bacteria and fungi, for the purpose of obtaining iron from the environment for metabolism. Siderophores fit into two general classes, the phenolkatecholate type and the hydroxamate type. Siderophores can be detected using chemical and microbiological procedures (Neilands, 1984, 1989). The chemical methods are convenient to use, but suffer from lack of sensitivity and specificity. Fekete ct al. (1983) used paper electrophoresis to detect both phenolate-catecholate and hydroxamates produced by microorganisms grown on solid medium. Bioassays for siderophores are orders of magnitudes more sensitive than chemical methods. The drawbacks of these methods are that viable stocks of the organisms must be maintained, and that there is no universal microbial response to siderophores. A "universal" chemical method for detection and determination of siderophores, including those that do not belong to either hydroxamic acid or catecholate type, was described by Schwyn and Neilands (1987). This method exploits the affinity of the dye chrome azurol S (CAS) for Fe3+.The affinity is less than that of the siderophores for Fe3+so that the metal ion is released from CAS, accompanied by a colour change. The amount of CAS released is therefore a measure of the siderophore concentration. A particularly attractive feature of this method is that the dye can be implanted in agar media with several microorganisms, and siderophore-excreting colonies are then identified through the change in dye
218
T.J. BEVERIDGE
eta/.
colour surrounding the colonies. As a cautionary point, it must be remembered that all metal chelators present in the medium, such as EDTA, will contribute to the assay result. Competition for iron between two chelators also forms the basis of a convenient and sensitive plate assay for evaluating the efficacy of siderophores (Minnick et al., 1991). The test siderophore on a filter disk is added to a plate of solid defined medium seeded with the test organism (e.g. Candida albicans) in the presence of the iron chelator ethylenediaminedi(0-hydroxyphenylacetic acid) (EDDA). Zones of growth stimulation around the disks indicate the ability of the peptide to chelate Fe(II1) and render it available for growth. The ability of bacteria to utilize siderophores produced by other organisms can be determined by adding pure siderophore to a culture which has been treated to minimize the iron content (see Section 3.2) and measuring the ability of the complex to enhance growth rate and yield (e.g. Plessner ef al., 1993). It must be borne in mind, however, that exogenous siderophores may serve directly as iron sources but may transfer iron to a chelator or siderophore that is produced by the test organism. Such inter-ligand exchange has been reported in many instances, but is generally slow and may be reflected in a long lag phase of growth (Plessner ef al., 1993). 7.2.5. Analyses in vivo of Iron-containing Proteins (a) Haernproteins. The haem proteins are an almost ubiquitous group of iron-containing proteins in which the metal lies in the plane of a porphyrin ring. The iron is either five- or six-coordinate,depending upon the function of the haem, with the four equatorial positions occupied by the nitrogen atoms of the porphyrin and the one or two axial positions filled with the donor atoms from amino-acid residues in the protein. The identity of these ligands can be determined by spectroscopic techniques, or by comparison of the protein sequence with those of other well-characterized proteins, and confirmation by site-directed mutagenesis. The haem type in a protein is generally determined by reaction with pyridine in alkaline solution (Jones and Poole, 1985);the resulting pyridine haemochrome has pyridine at both axial positions and the haemochromes of different haem types have quite distinct spectral properties, which are independent of the ligation of the haem to the protein. Pyridine haemochromogenscan be formed using crude cell material. High-performance liquid chromatography (HPLC) is increasingly used to separate and identify the types of haem present in, for example, bacterial oxidases (Svensson et al., 1993). Spectrophotometryin the visible region (400-700 nm) is the most common and versatile of methods for the identification, quantification and functional study of cytochromes, and exploits the spectral changes that occur on oxidation and reduction or the binding of ligands, such as carbon monoxide. These measurements
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
21 9
can be made on intact cells when dual-wavelengthor split-beam spectrophotometry is used to observe the haemoprotein spectrum superimposed on the light-scattering signal of the cell suspension. For example, Scott et al. (1981) used difference spectroscopy to determine the patterns of change in amounts of haemoproteins during the cell cycle of E. coli. There are numerous examples in the literature of successful applications of split-beam or dual-wavelength spectrophotometry to identify or quantify haemoproteins in eukaryotic or prokaryotic microorganisms. In addition to the cytochromes, some bacterial globins and other haem proteins can be present at sufficiently high concentrations that they can be assayed spectrophotometrically in viva Routine determinations of haemoprotein identity or concentration in microbial cells have been complemented by more sophisticated methods for studying the reactions of oxidases in vivo with ligands (Poole et al., 1979). The methods based upon light reflectance, rather than transmission, which have been developed for studying tissue metabolism (Chen et al., 1993) have not been much exploited for microbial cells (but see Bashford et al., 1980),yet could find important applications in studying dense microbial cultures or perhaps biofilms. Spectrophotometric methods have been developed that allow measurements of haemoprotein concentrations or redox state in situ in growing cultures, without the need to harvest and concentrate cells. Thirty years ago, Chance (1966) developed methods for obtaining difference spectra of single yeast cells but these methods have been little exploited. (b) Non-haem iron proteins. The iron--sulfur proteins are ubiquitous in the microbial (and non-microbial) world. Because of this and the diversity of microbial processes in which they participate, the iron-sulfur proteins are the most important electron transport proteins. Whereas in the haem proteins, iron is bound by a macrocyclic ligand (porphyrin), the iron-sulfur proteins contain one or more iron-sulfur clusters bound to the protein; this is usually, but not always, via cysteine thiolate groups. The clusters contain two, three or four iron atoms bridged by sulfide. In all cases except the simple rubredoxins, the sulfur is released as H2S on treatment with dilute mineral acids, It is thus often referred to as acid-labile sulfur and assayed as such in quantification of the clusters. These proteins are in some respects harder to study than the haem proteins because they lack distinctive optical properties. If the protein also contains a haem or flavin, this will mask the broad, featurelcss absorption bands of the iron-sulfur cluster. Such proteins are generally studied using ESR spectroscopy. Both [2Fe-2S] and [4Fe-4S] clusters typically give g = 1.Y4 signals in the reduced state and the temperature dependence of the signals gives an indication of cluster type. More complex signals occur when the protein contains two or more iron-sulfur clusters. ESR spectra can be obtained with intact cells. For example, ESR signals from the 2Fe-2S clusters in the ferredoxinBED and ISPEEDcomponents of benzene dioxygenase were quantified in intact cells of Ps. putida ML2 (Tan et al., 1994). Cells were washed and concentrated in ESR tubes to about 2 x 10” cells r n - ’ (around
220
T.J. BEVERIDGE eta/.
10 mg protein ml-’). Samples were reduced with dithionite after a freeze-thaw cycle. The FeS cluster concentration (expressed as millimolar spin) was estimated by subtracting the spectrum of the respective pure protein from the whole cell spectrum until the latter was minimized. Interfering manganese signals in these whole cell samples were broadened (and thus made less intrusive) by the addition of 5 m~ EDTA. In a second example, ESR spectroscopy has been used to determine the pattern of change in FeS clusters during the cell cycle of E. coli (Poole et al., 1981). The technique was sufficiently sensitive that fractions from a zonal centrifugation experiment could be analysed. 7.2.6. Iron Oxidation and Reduction It is well known that bacteria are able to deposit the hydroxides of iron (and manganese) extracellularly. Genera such as Gallionella, Sphaerotilus and k p tothrix have been known since the nineteenth century as “iron bacteria” and their activities are manifest as a yellow or “rusty” coloration in some freshwater environments. Recent awareness of the environmental and biotechnological significance of microbially-mediated metal ion transformations, which include oxidation, reduction, methylation and demethylation, have stimulated interest in the distribution of these capabilities. Thiobacillusferrooxidans is the best known iron-oxidizing bacterium and is responsible for the solubilization of iron pyrites and other sulfide minerals, and contributes to acidification of the environment on a massive scale (for example, acid run-off from disused mines and spoilage heaps). Growth of T.ferrooxidans on solid media is problematic, however, in part due to the release of toxic organic compounds by acid hydrolysis of the agar. An innovative method to overcome these difficulties devised by Johnson and McGinness (199 1) involves a plate having two layers of agarose medium. The lower layer contains an acidophilic heterotroph, which removes these organic compounds; the upper layer contains the acidophilic iron-oxidizers. Ferric iron is almost as powerful an oxidizing agent as oxygen and an abundant potential electron acceptor in the environment. Heterotrophic bacteria, fungi, the roots of higher plants and mitochondria all reduce iron. Indeed, the ability of microorganisms to reduce Fe(II1) should come as no surprise: iron reduction is implicated as an important mechanism for liberating the metal from siderophores, as a prerequisite for iron incorporation into haem, cofactors or magnetosomes, and iron redox changes have important control functions intracellularly. The dissimilatory reduction of Fe(TI1) results in accumulation of significant amounts of Fe(I1) outside the microorganism. It should be noted that reduction of iron can occur indirectly through microbially-mediated changes in the reducing conditions in the organisms’ environment. Early doubts about a biological role for Fe(II1) reduction have been dispelled and electron transfer to Fe(III), which thus acts as a terminal
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
22 1
oxidant, has been clearly demonstrated. Little is known about the enzymes or pathways in these novel forms of metabolism, but iron reduction is easily measured by use of ferrozine, a colorimetric agent specific for Fe(I1) (Stookey, 1970). In the yeast Sacchammyces cerevisiae, iron uptake is accompanied by iron reduction. Fe(II1) salts or complexes are reduced by an inducible plasma membrane reductase; iron is then taken up as free Fe(I1) by an inducible high-affinity carrier. An assay for the ferrireductase activity has been devised (Lesuisse et al., 1995) that exploits the ability of intact cells to reduce a variety of non-permeant femc and non-ferric electron acceptors (e.g. tellurite, ferricyanide, resazurin, nitroprusside and Cu2'), in turn causing colour changes.
7.3. Copper 7.3.1. Copper L.evels in Cells and Growth Media and Studies of Copper Transport Copper is a trace element. It is the strongest Lewis acid of the divalent metal ions, making it one of the more toxic elements. In general, where zinc and the transition metals are present together in excess in a biological system, every binding site would be occupied by Cu(II), to the exclusion of other metal ions, were it not for biological mechanisms that mask Cu(I1) by the provision of ligands, notably metallothioneins, which bind copper preferentially. In yeast and Neurospora crassa, Cu-metallothionein has been extensively studied and details are emerging from genetic studies of how bacteria achieve copper homeostasis. Copper (and silver) have been used for many years to disinfect water. Their biocidal effect is greatest at low concentrations of free chlorine (Yahya et al., 1990) but is reduced by complex formation with phosphates (Landeen et al., 1989).Again, the importance of speciation and the presence of metal-binding ligands can scarcely be overemphasized in both applied and fundamental studies. Detailed studies of copper uptake have generally relied on the use of @Cu(II) (e.g. Gadd et ~ l . 1984) , or AAS (e.g. Cha and Cooksey, 1993). Lin and Kosman (1990) combined these methods to characterize uptake of copper, provided as the bis(histidine) complex of @Cu2+,in strains of S. cerevisiae containing and lacking yeast copper metallothionein. The protocol broadly followed the procedures described in Section 4.8. Cu(I1) is also a substrate in yeast for the reductase Frel, which is required for 50-7096 of the uptake of @Cu by wild-type cells, and perhaps for another reductase in the yeast plasma membrane. These reductase activities can be measured with tetrazoiium salts (Hassett and Kosman, 1995). @Cuhas also been used to study the intracellular distribution of copper following uptake by S. cerevisiae (Lin et aZ., 1993a, b). An alternative approach that offers convenience and continuous monitoring of
222
T.J. BEVERIDGE
t?t a/.
copper uptake is a Cu2+-sensitiveelectrode. Thus, De Rome and Gadd (1987) used a commercially available electrode to monitor removal of copper by S. cerevisiae from solutions containing 10 m~ cu2+. Estimates of total intracellular copper levels are below 0.01 % of dry cell mass. Pirt (1975) quotes a theoretical growth yield of lo5g cells/g element, as for cobalt. The copper requirement for growth is so small that growth is rarely limited even when copper salts are completely omitted from defined media, and complex broth media generally contain adequate levels of copper. Lin and Kosman (1990) report a copper content of 15 nM (determined by electrothermal AAS) for a synthetic yeast medium that was not supplemented with copper. Media based on “yeast nitrogen base” or yeast extract-peptone-dextrose (YPD) contained 150-325 nM copper. Consequently, exploitation of the technique of copper limitation requires that medium is extracted. The chelator 1,5-diphenylthiocarbazone(dithizone) has been used in a number of studies to minimize copper levels: treated medium typically contains 0.14.01 p~ Cu (Hubbard et al., 1986; Hughes and Poole, 1989). The ion-chelating resin Chelex- 100, described in Section 3.2, was used by Ciccognani et al. (1992) in a study of the effects of copper limitation on terminal oxidase function in E. coEi. Copper levels in control and copper-depleted cells measured by graphite furnace AAS were 0.074 and 0.029 nmol Cu mg-’ dry cells, respectively. However, the efficacy of the procedure was more clearly evident in ligand-binding properties of the haem- andcopper-containing terminal quinol oxidase, cytochrome bo’. CO recombination to the oxidase haem was significantly faster in the copper-depleted preparation, presumably owing to lack of the copper atom at the active site that is thought to retard ligand recombination. Other, earlier work (reviewed in Hughes and Poole, 1989) has exploited copper deficiency to study the functions of copper-containing electron transport components. The use of polarographic techniques to monitor directly the uptake of copper by live cells has been discussed in Section 4.3 (Savvaidis et al., 1992). It is important to recognize that the presence of Cu(I1)and other metals in growth media can catalyse the hydrolysis of p-lactam antibiotics, with important consequences for any study that uses these metals and antibiotics together. For example, Beard et ul. (1992) showed that an ampicillin-sensitive strain of E. coli used in attempts to clone genes conferring increased resistance to Zn(IT), Co(I1) or Cu(I1) unexpectedly grew on plates containing ampicillin and these metals. Clearly, double selection for resistance to ampicillin and such metals is hazardous. 7.3.2. Copper-binding Proteins A number of extracellular compounds have been implicated in copper binding, often to very high levels, and so the physiological function of such binding, if any, is probably detoxification of copper. For example, a copper-resistant community of bacteria from activated sludge (Dunn and Bull, 1983) was found to accumulate copper to concentrations of about 30% of dry weight, probably in association with
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
223
extracellular polysaccharide. D.T. Ciccognani, M.N. Hughes and R.K. Poole (unpublished) have sought extracellular copper-binding ligands in spent medium from cultures of E. coli, Alcaligenes xylosoxidans and Pseudomonas aeruginosa using anodic stripping voltammetry (see Section 4.3), but have not obtained evidence for extracellular copper ligands analogous to siderophores (see Section 7.2.4). However, Vibrio alginolyticus produces two copper-inducible extracellular proteins (Harwood-Sears and Gordon, 1990), the analysis of which provides instructive examples. These proteins CuBPl and CuBP2 have been concentrated and partially purified by using a copper-charged immobilized metal ion affinity chromatography column. Copper-complexing activity in fractions from gel-permeation chromatography was assayed with a Cu2+-specificion electrode.
7.3.3.
Other Copper-containing Structures and Proteins
Few copper-containing proteins are amenable to study in intact cells by the techniques described in this review. However, the physiological consequences of copper depletion (see above) or replacing copper with other metals are more easily studied. One example will suffice. Incorporation of copper from the medium into the catalytic site of the Cu-Zn superoxide dismutase of S. cerevisiae limits enzymatic activity. When silver (50 p ~ is)added to the growth medium (containing 4 p~ Cu), growth is partly inhibited and the Cu-Zn superoxide dismutase purified from these cells contains silver at the active site (Ciriolo et al., 1994). Copper is amenable to detection by EDS (see Section 4.10). Iron minerals (Fe& and FeS2) in magnetosomes of a many-celled magnetotactic bacterium have been shown to be accompanied by high levels of copper (0.1-10 atomic %, relative to iron). Copper was quantified by integration of the respective X-ray emission lines in the EDS spectra (Bazylinski et al., 1993).
7.4. Zinc
Zinc is not a transition element and does not function directly in redox processes. There are, however, two important roles for zinc in microbial and other biological processes. In the first, zinc acts as a strong Lewis acid; in these reactions, substrates are either coordinated, polarized and activated, or a zinc-bound water molecule is ionized to Zn(0H) and attacks the substrate. The “catalytic” zinc atom is coordinated to three or four amino acids of the protein, and a water molecule. In the second, zinc has a structural or template role, and is coordinated to fourcysteine residues, or to two cysteines and two histidines. One of the most recently discovered and important of such roles is in zinc fingers, in which the metal stabilizes structural domains in proteins and controls their binding to DNA. Study of these processes
224
T.J. BEVERIDGE eta/.
and others in which zinc is involved is hindered by zinc’s “silent” properties, so that neither electronic nor nuclear spectroscopy can be used to give direct information on the nature of ligand groups around the metal. Zinc is required in minute quantities by most microorganisms (giving a growth yield of 2 x lo4g dry biomass per g Zn; Pirt, 1975), so that studies of zinc deficiency are difficult to perform. Chemically defined media generally contain adequate zinc levels as impurities to allow growth. Studies of the effects of limitation therefore require extraction of the medium (Hughes and Poole, 1989). Zinc transport has received little attention probably as a result of these experimental difficulties. 65Zn(II)was used in early work (Bucheder and Broda, 1974) to demonstrate in E. coli an energy-dependent transport system that was inhibited by cadmium ions. Conversely, Laddaga and Silver (1985) reported active uptake of lWCd(II) and its competitive inhibition by zinc ions. More recently, ‘@Cd(II) uptake studies have revealed two kinetically distinct zinc transport systems (Rosner and Aumercier, 1990).
7.5. Nickel
Since 1979, nickel has become firmly established as an essential metal in several microbial enzymes. Previously, the only established nickel-containing enzyme was urease. Now, nickel joins other metals in the first transition metal series (Mn, Fe, Co, Cu) in having important biological functions. In addition to the well-studied functions for nickel detailed below, diverse roles for nickel that have been suggestedinclude pigment formation and endospore formation (Hughes and Poole, 1989). Nickel enzymes are frequently involved in gas metabolism, particularly in anaerobic bacteria, such as the methanogenic members of the Archaea. The nickel enzymes that have received most attention are urease, hydrogenases, methyl CoM reductase and carbon monoxide dehydrogenase (acetyl-CoA synthase). Many studies of nickel transport utilize 63Ni.For example, Fu and Maier (1991) suspended nickel-free cells of Bradyrhizobiuin japonicum in buffer and added 63NiC12to 1 m~ (0.67 mCi ml-’ of cell suspension). The assay flasks were sampled for 1 h and cells collected on filters. A similar procedure has been described for assay of 6’Ni in Helicobacter pylori, which requires nickel for urease (Mobley et al., 1995). Higher levels of nickel accumulation, e.g. by passive biosorption may be studied by AAS (Holan and Volesky, 1994). Only a few bacterial genes have been shown to be inducible by nickel. One is the hydrogenase of B. japonicum (above), which requires nickel, and another is the celF gene of E. coli encoding phospho-P-glucosidase, although the physiological significance of this is unclear. Insertion of the l d B genes within celF, creating a transcriptional fusion results in Ni-inducible light emission, which is optimal at 20-50 ppm nickel (Guzzo and DuBow, 1994). Such fusions are candidates for metal-responsive biosensors.
METAL-MICROBE INTERACTIONS:CONTEMPORARY APPROACHES
225
Studies of nickel sensitivity and resistance pose no special difficulties, and nickel salts may be added to liquid or solid media (Stoppel and Schlegel, 1995).
7.6. Toxic Metals and Metalloids Research on microbial resistance to various toxic metals and metalloids has demonstrated the versatility of microorganisms to cope with toxic metal ions and stimulated much research activity, particularly in relation to biotechnology. Most work has focused on the mechanisms of microbial resistance to elevated metal levels in natural environments and laboratory media. It has necessitated the application of several well-established and newly evolved methodologies for determining total and bioavailable levels of metals as well as physiological approaches to understanding the microbial responses. In general, resistance mechanisms employed by microorganisms to metals and metalloids can be classified as: (1) efflux of the toxic ions (e.g. arsenic and cadmium); (2) transformation of the toxic compounds to their non-toxic derivatives through degradation or redox reaction (e.g. arsenic, mercury, selenium, tellurium and tin); (3) immobilization of the ions within the cell envelope either by binding to cell surface or precipitation (e.g. germanium, lead, selenium, silver and tellurium); and (4) biomethylation (e.g. mercury and lead). In this section, microbial resistance towards some important metals and metalloids, such as arsenic, cadmium, germanium, lead, mercury, selenium, silver, tellurium and tin will be briefly discussed with selected examples. Attention will be drawn to different analytical methods used to study these systems.
7.6.1. Arsenic The toxicity of arsenic compounds is dependent on the oxidation state of arsenic. For instance, arsenite (HzAsO;) is about 200 times more toxic than arsenate (HzAsOi) to living organisms (Williamsand Silver, 1984). Therefore, oxidation of As(1II) to As(V) can be considered a detoxificationprocess. Two mechanisms are involved in bacterial resistance to arsenic (Cervantes et af., 1994). The first is a chromosomally encoded pathway involving the oxidation of As(I1I) to As(V). An arseniteoxidase that catalysesthis reaction has been purified from Alcaligenes faecalis (Anderson et al., 1992). Although arsenicals can be quantitativelymeasured by either the silver diethyldithiocarbamate assay (Phillips and Taylor, 1976)or inductivelycoupledplasma spectrometry (Mahan et al., 1989), these methods cannot distinguishbetweenthe oxidation states of arsenic. However, polarographic techniques may be used to determine As(II1) specifically (Section
226
T.J. BEVERIDGE eta/.
4.3; Barrett et al., 1993). Arsenite concentration was measured using 2,6-dichloroindophenol as an electron acceptor by Phillips and Taylor (1976), who showed the disappearance of arsenite in a sewage sample containing arsenite-oxidizing bacteria. The second mechanism of bacterial resistance to arsenic is plasmid mediated. This mechanism can be found in both Gram-positive and Gram-negative bacteria, such as Staphylococcus aureus and E. coli, respectively (Cervantes et al., 1994). The arsenic resistance (ars)operon from E. coli plasmid pR773 has been shown to contain five genes (arsRDABC).Both ArsR and ArsD are regulatory proteins. The arsC gene encodes an NADPH-dependent arsenate reductase, which catalyses the reduction of arsenate to arsenite. Active efflux of arsenite from the bacteria is achieved by the ArsA and ArsB proteins. Radioactive 73As was used to study reduction of arsenate and efflux of arsenite by S. aureus. To measure reduction of arsenate, Ji et al. (1994) separated the radiolabelled arsenate and arsenite by thin layer chromatography, and quantified the reaction products with an AMBIS B-scanner. Arsenate reductase activity, as measured by NADPH oxidation, exhibited Michaelis-Menten kinetics with an apparent Km for AsO% of 1 pM and an apparent V,, of 200 pmol min-lmg-' of protein (Ji et al., 1994). Using inhibitors of energy transduction, Broer et al. (1993) showed that the efflux of HzAsO; by S. aureus was an active mechanism.
7.6.2. Cadmium Cadmium has no known essential biological function and is highly toxic to living systems (Gadd and Griffiths, 1978; Trevors etal., 1986). Cd2+exerts its toxic effect by binding to the sulfhydryl groups of proteins and breaking single-strand DNA (Foster, 1983; Mitra and Bernstein, 1978). Various microorganisms, such as S. aureus, B. subtilis, Listeria spp., A. eutrophus, E. coli, Ps. putida, some cyanobacteria as well as some fungi and algae (Ji and Silver, 1995) are resistant to Cd". Several mechanisms for Cd2' detoxification in bacteria are known, and some of these are plasmid mediated. Work in this area has relied greatly on molecular genetic studies of the genes and proteins involved (see Section 6). The most extensively studied mechanism involves efflux of Cd2+by an inner membrane anti-porter cation carrier, CzcA, in A. eutruphus strain CH34. CzcB and CzcC have ancillary roles in the transport process (Nies et al., 1989). CZCD and CZCRhave been identified as regulatory genes of the czc system (Nies, 1992). Recently, another Cd2+-resistanceoperon (ncc) has been cloned from an A. xylusoxidans plasmid (Schmidt and Schlegel, 1994). In Gram-positive bacteria, an active Cd2' efflux system is achieved by a P-type ATPase (CadA) (Silver et al., 1989). In addition, a cysteine-rich metallothionein is responsible for Cd2+resistance in cyanobacterial strains in the genus Synechococcus (Gupta et al., 1993). Many methods have been used to study cadmium resistance. A cadmium
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
227
ion-specific electrode was used to study immobilization of Cd2+ in capsular polysaccharides of K. aerogenes (Bitton and Freihofer, 1978). In this study, the capsular polysaccharides were extracted and found to complex with 9-18% of the Cd2+ added to the culture. Cadmium resistance of K. aerogenes has also been studied by flame atomic spectrophotometry (Aiking et al., 1985), which provides a more accurate analytic measurement of cadmium. With theuse of anodic stripping voltammetry, Macaskie and Dean (1984) showed that 65% of the Cd2+in a growth medium was removed and precipitated as CdS04 inside the cells of a Citrobacter strain. Uptake of Cd2' has been studied using the radioisotope lWCd2+.Laddaga and Silver (1985) reported that uptake of 109Cd2+ by E. coli K-12 was an active process with apparent Km and V- values of 2.1 pM and 0.83 pmol min-lg-' dry cells, respectively. The same approach has been used to study Cd2+uptake in B. subtilis (Laddaga et al., 1985).
7.6.3. Gennuniurn Germanium (Ge) is a valuable metalloid recovered from the processing wastes of coal, zinc and copper-lead-zinc ores. Germanium is used industrially for semiconductor devices, optical lenses and alloys. Although germanium lacks any known function in living organisms (an "indifferent" element; see Section 2.1.3), it has been reported as possessing mutagenic, anti-mutagenic and anti-tumor activities (Lee et al., 1990; Slawson et al., 1992). Little is known about germanium accumulation by microorganisms. Azam and Volcani (1974) showed that uptake of Ge is an energy- and temperature-dependent process in the diatom Nitzschia a h a . Using radioactive "Ge, Chmielowski and Klapcinska (1986) observed a considerable increase in germanium uptake in Ps. putida when bacterial cells were incubated with catechol. It was suggested that Ge was accumulated intracellularly and that a germaniumxatechol complex was taken up by a catechol-inducible transport system. Electron microscopic examination revealed that most of the metal was bound within the cell membrane and only a small amount was found in the cytoplasm (Klapcinska and Chmielowski, 1986). Germanium can also be determined spectrophotometricallyat ,4525 after reaction with a phenylfluorone reagent. Using this approach, Van Dyke et al. (1 989) showed that the apparent Ks and V, values for germanium accumulation by Bacillus cereus NRC 3045 were 4.0 g 1-' and 2.2 mg g-' dry wt h-', respectively. Treatment of B. cereus with toluene (to permeabilize cells) or 2,4-dinitrophenol, resulted in a > 80%decrease in germanium accumulation, inferring an active transport process. EDS revealed that germanium was associated with the cells. In contrast, Ps. stutzeri AG259 accumulate germanium by an energy-independent mechanism (Van Dyke et al., 1990).
228
T.J. BEVERIDGE eta/.
7.6.4. Lead Although toxicity of lead compounds in higher animals, including humans, has been widely studied, relatively little information is available on lead resistance in bacteria. Lead resistance has been found in different bacterial isolates such as S. aureus, Micrococcus luteus, Azotobacter spp., and K.aerogenes (Novick and Roth, 1968; Tornabene and Edwards, 1972; Aiking et al., 198s). Both methylation and immobilization of Pb2+have been suggested as resistance mechanisms. Detoxification of Pb2+by K. aerogenes NCTC 418 has been studied in some detail and the formation of PbS was suggested to be the resistance mechanism (Aiking et al., 1985). AAS has been used to study Pb2+uptake by bacteria. In the study of Tornabene and Edwards (1972), cellular distribution of lead was analysed. After separation and digestion in concentrated nitric acid, different cellular fractions were analysed by AAS. Approximately 99% of the cellular Pb was bound to the cell envelope. Using electron microscopy, Aiking et al. (1985) showed that PbS precipitate was deposited in the cell envelope of K.aerogenes NCTC 418. This method avoids the complication of separating different cell fractions.
7.6.5.
Mercury
Mercury compounds have been used as catalysts in industry as well as disinfectants for clinical purposes. Both mercuric ion (Hg2') and organomercurial compounds are highly toxic to biological systems owing to their strong affinity for thiol groups in proteins. Hence, accumulation of Hg contaminants in nature causes serious environmental and public health problems (Belliveau and Trevors, 1989a; Misra, 1992). Mercury resistance involving enzymatic reduction of Hg2+to volatile elemental mercury and/or decomposition of organomercurialcompounds have been observed in Gram-negative and Gram-positive bacteria (Misra, 1992). Mercury-resistance phenotypes are plasmid mediated and can be divided into two groups. First, broad-spectrum resistance determinants detoxify organomercurial compounds by a two-step process that includes cleavage of the C-Hg bond by an intracellular organomercurial lyase enzyme (MerB) followed by reduction of Hg2+to Hgo by an FAD-containing, NADPH-dependent mercuric reductase (MerA). The second system is a narrow-spectrum pathway, which confers resistance only to inorganic mercury ions via MerA. nierP and merT encode mercuric ion transport proteins. MerP is a Hg2+-bindingprotein located in the periplasmic space. It functions as a shuttle to transfer Hg2+ to the membrane-bound MerT protein, which internalizes the Hg2+into the cytoplasm. The Hg-resistance operon is controlled by the MerR regulatory protein. In addition to studies of the genes required for mercury resistance and their
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
229
individual roles, the MerR protein and its reaction with Hg2+have been intensively studied in vitm, but this work is outside the scope of the present chapter (see O'Halloran et al., 1989; Wright et al., 1990). Similar1 203Hg2+(Sahlman and Jonsson, 1992) has been used to show that MerP is a Hg +-bindingprotein, which binds one Hg2+per MerP monomer. Resistanceto mercuric chloride and phenyl mercuric acetate in a Bacillus cereus strain was encoded on a self-transmissible plasmid pGB130 (Belliveau and Trevors, 1990). A mating system for a Gram-positive bacterium like Bacillus allowed study of the transfer frequency (lo4 to and levels of mercuric reductase activity in the wild-type and transformants. Electrotransformation was also used to introduce this plasmid from B. cereus to B. thuringiensis (Belliveau and Trevors, 1989b).
F
'
7.6.6. Selenium Although selenium is an essential trace element for animals (Gadd, 1993),elevated concentrations of selenium oxyanions are highly toxic to living systems (Painter, 1941). As analogues to sulfur compounds, selenite and selenate inhibit sulfite reductase and sulfate transport systems in bacteria, respectively (Brown and Shrift, 1980; Harrison etal., 1980).Several fungal and bacterial spp., such as a Fusarium sp., Wolinellasuccinogenes, some Salmonella strains and Desulfovibrio desulfuricans, can reduce selenate and/or selenite to elemental selenium and hence detoxify the selenium oxyanions (Gharieb et al., 1995;Tomei et aE., 1992, 1995). Quantitative measurements of selenium can be done by usin either radioactive "Se or fluorometric procedures. Using Na?%e03 and Na27FSe04, Brown and Shrift (1980) showed that selenate ion was transported and assimilated by the same process as was sulfite in S.+phimurim. However, selenite was not transported by the sulfate carrier. Selenium reductase of Desulfovibrio desulfuricans was studied by Tomei et al. (1995). Cell samples were digested with perchloric and nitric acids. After treating with 2,3-diaminonaphthalene and extracting by cyclohexane, fluorescence of samples was recorded (excitation and emission at 369 and 525 nm, respectively). D. desulfuricans was found to reduce 95 and 97% of 1 pM selenate and 100 p~ selenite, respectively. Using electron microscopy and EDS, selenium granules were shown to be deposited in the cytoplasmic fraction of the bacteria. 7.6.7. Silver Soluble silver ions are probably toxic to all microorganisms. Indeed, silver compounds have long been used as anti-microbial agents for treating burns and eye infections of newborn infants (Slawson et al., 1992). A study by Modak and Fox ( I 973) showed that '"Ag was detected intracellularly in f! aeruginosa cells exposed to this isotope, where the majority of the silver
230
T.J. BEVERIDGE eta/.
was complexed to DNA. Binding of '"Ag to RNA and cell envelope was about 1/40 and 1/10, respectively, of the amount bound to DNA. Bacteria that exhibit tolerance to Ag+ have been isolated. For example, Ag"-resistant E. coli and €? stutzeri strains have been isolated from a burn patient and a silver mine, respectively (Gadd et al., 1989; Goddard and Bull, 1989; Starodub and Trevors, 1989). Uptake and accumulation of Ag' have often been associated with silver resistance and detoxification, inferring exclusion or immobilization of Ag' as a mechanism to alleviate Ag' toxicity in cells. There is experimental evidence to suggest a plasmid-encoded Ag' resistance mechanism in an E. coli strain (Starodub and Trevors, 1989). TEM and EDS showed that the Ag+-resistant strain did not accumulate silver, whereas the sensitive strain contained dense silver deposits. Quantitative analysis by AAS showed that the Ag+-sensitiveE.coli strain accumulated about five-fold more Ag than the resistant strain (Starodub and Trevors, 1990). In addition, the Ag+-resistantstrain produced one third more H2S and intracellular acid-labile sulfide than the sensitive strain. Similar Ag' resistance phenomena have been observed in €? stutzeri (Slawson et al., 1992). There is no evidence that Ag' is transformed to Ag(0) by a reductase enzyme.
7.6.8. Tellurium
The metalloid tellurium is used in batteries, alloys and rubber and as a colouring agent in glass. Tellurium compounds are toxic to many microorganisms, particularly Gram-negative bacteria. However, some strains of Corynebucreriurndiphtheriae, Streptococcusfaecalis, S. aureus, A. faecalis and A. denitri'cans, and some members of the family Enterobacteriaceae are resistant to tellurite (Walter and Taylor, 1992). Plasmids from three incompatibility groups have been shown to control tellurium resistance in bacteria (Walter and Taylor, 1992). Unfortunately, the mechanisms of plasmid-mediated tellurite resistance are not known, but increased efflux or reduced uptake have been ruled out as possible mechanisms. Although a tellurite-reducing protein of approximately 53 kDa was purified from 7: thermophilus, it is not related to any of the known tellurite-resistance determinants (Chiong el al., 1988). Investigations of tellurite resistance have been hampered by the lack of a convenient assay to measure the concentration of tellurite and tellurium. Quantitative measurement of the black metallic tellurium was used by Chiong et al. (1988) to estimate tellurite reduction in different cellular fractions of Thennus thermophilus. Electron spectroscopic imaging and X-ray diffraction have also been used to locate metallic tellurium deposition in a teilurite-resistant E. coli mutant (Taylor et al., 1988). Metallic tellurium was shown to be accumulated just inside the inner membrane of the bacteria. An AAS procedure has been developed to detect
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
23 1
tellurium in animal tissues (Cooper, 1971). The detection limit of this method is 2 pg Te g-' tissue. 7.6.9. En Tin compounds are used for protective coating of metals or as anti-fouling agents in marine paints (McDonald and Trevors, 1988). Although many studies on tin toxicity have focused on organotin compounds, inorganic hydrated tin chloride can also be toxic to microbial populations at concentrations above 10 mg 1-' (Hallas and Cooney, 1981). The toxicity of tin compounds is caused in part by inhibition of oxidative phosphorylation and energy metabolism. The form in which tin is present is critical to its toxicity. Trialkyl tin compounds are the most toxic, followed by dialkyl tin. Monoalkyl tin compounds are non-toxic, and tetraalkyl tin is only toxic when converted to trialkyl tin. Bacterial degradation of triorganotin compounds can be considered as a detoxification mechanism, since the conversion of bis(tributy1tin) oxide (TBTO) to monobutyltin derivatives results in lower toxicities (Barug, 1981). Gas chromatography-mass spectrometry (GCMS) was used by Barug (1981) to study the degradation of TBTO. In this study, none of the bacterial and fungal species was able to utilize TBTO as sole carbon source. However, Ps. aeruginosa and some fungal strains degraded TBTO and converted the biocide to a less toxic monobutyltin compound. Using HPLC coupled with a graphite furnace atomic absorption spectrophotometerand gas chromatography coupled with a tin-specific flame photometric detector, Blair et al. (1982) showed that some marine bacterial isolates were able to degrade TBTO and immobilize 3.7-7.7 mg tin g-' dry weight of cells. Radioactive '13Sn has been used to study uptake and accumulation of inorganic tin compounds (Wong et al., 1984). By studying the accumulation of "'Sn in different cellular fractions of an Ankistrodesrnus falcatus strain, these researchers suggested that the uptake of tin was a physicochemical surface adsorption process; 85% of the Il3Sn recovered was in the polysaccharide fraction. Using GCMS analysis, Hallas et al. (1982) showed that inorganic Sn(1V) was rnethylated in sediments. Although evidence has implicated methyltin species as being responsible for increased formation of methylmercury in sediments, the significance of this reaction is presently not clear.
8. CONCLUSIONS AND OUTLOOK Achieving the ultimate objective of a comprehensive understanding of the interactions between metals and microorganisms requires taking a rnulti-
232
T.J. BEVERIDGE eta/.
disciplinary approach. This involves microbiology, particularly microbial physiology, genetics and molecular biology, bioinorganic chemistry, analytical chemistry and the application of instrumental techniques. Our aim in this contribution has been not to review this field but to survey these approaches. We firmly hold the view that the exploration of the effects of any metal ion on microbial physiology is hazardous unless due attention is paid to the underlying chemistry. There are examples in the literature of conclusions that are erroneous or suspect because the speciation and bioavailability of the metal ion under study have not been rigorously considered. Equally important is the ability to control intracellular concentrationsof metals. The last point is illustrated well by the recent use of photolabile compounds for binding calcium. Undoubtedly, such new techniques will find valuable applications in the hands of ingenious investigators. Genetics and molecular biology hold great promise for advancing studies of metal-microbe interactions,particularly those aspects that are currently perceived as especially significant. These include gene regulation by metals (metalcontaining transcription factors and zinc fingers), control of specificity at metalbinding sites and genetic modification of microorganisms for application in biohydrometallurgy.The availability of extensive sequence data-in an increasing number of cases, of the whole genome-will allow hunting for genes and their proteins that display clear binding motifs for metal ions or prosthetic groups. It is clear, however, that microbial physiology coupled with bioinorganic chemistry will remain as key contributors in this field.
Danielle Fortin and Suzanne Schultze-Lam of TJB's laboratory supplied the TEM images and this TEM research was supported by the Natural Science and Engineering Research Council of Canada. RKP and MNH acknowledge support from the Natural Environment Research Council (UK). Research by JTT is supported by the Natural Sciences and Engineering Research Council of Canada in the form of an operating grant.
REFERENCES Adams, M.W.W. (1993) Enzymes and proteins from organisms that grow near and above 100°C.Ann. Rev.Microbial. 47,627-658. Adams, S.R., Kao, J.P.Y., Grynkiewicz, G.,Minta, A. and Tsien, R.Y. (1988) Biologically useful chelators that release Ca2+on illumination.J. Am. Chem. Suc. 110,3212-3220. Adams, S.R.,Kao, J.P.Y. and Tsien, R.Y. (1989) Biologically useful chelators that take up Ca2+on illumination. J. Am. Chern. Suc. 111,7957-7968.
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
237
transport are affected by disruption of cta3, a novel gene encoding Ca2+-ATPase in Schizosaccharomyces pombe. J. Biol. Chem. 265, 18400-18407. Gilboa H., Kogut, M., Chalarnish, S., Regev, R., Avi-Dor, Y.and Russell, N.J. (1991) Use of 23Na nuclear magnetic resonance spectroscopy to determine the true intracellular concentration of free sodium in a halophilic eubacterium. J. Bacteriol. 173,7021-7023. Goddard, P.A. and Bull, A.T. (1989) The isolation and characterization of bacteria capable of accumulating silver. Appl. Microbiol. Biotechnol. 31,308-3 13. Goncalves, M.L.S., Sigg, L., Reutlinger, M. and Stumm, W. (1987) Metal ion binding by biological surfaces. Voltammetric assessment in the presence of bacteria. Sci. Total Environ. 60,105-120. Good, N.E., Winget, G.D., Winter, W., Connolly, T.N., Izawa, S. and Singh, R.M.M. (1986) Hydrogen ion buffers for biological research. Biochemistry 5,467-474. Graham, L.L. (1992) Freeze-substitution studies of bacteria. Electrun Microsc. Rev. 5, 77-103, Graham, L.L. and Beveridge, T.J. (1990) Effect of chemical fixatives on accurate preservation of Escherichia coli and Bacillus subtilis structure in cells prepared by freezesubstitution. J. Bacteriol. 172,2150-2159. Greene, A.C. and Madgwick, J.C. (1991) Microbial formation of manganese oxide. Appl. Environ. Microbiol. 57, 11 14-1 120. Gueffroy, D.E. (1990) Bufers: A Guidefor the Preparation and Use of Bugers in Biological Systems. Calbiochem Corporation, San Diego, CA. Guida, L., Saidi, Z., Hughes, M.N. and Poole, R.K. (1991) Aluminium toxicity and binding to Escherichia coli. Arch. Microbiol. 156,507-512. Gupta, A., Morby, A.P., Turner, J.S., Whitton, B.A. and Robinson, N.J. (1993) Deletion within the metallothionein locus of cadmium-tolerant Synechococcus PCC 6301 involving a highly iterated palindrome (HIP1) Mof. Microbiol. 7, 189-195. Guzzo, A. and DuBow, M.S. (1994) A 1 4 B transcriptional fusion to the cryptic celF gene of Escherichia coli displays increased luminescence in the presence of nickel. Mol. Gen. Genet. 242,455-460. Hallas, L.E. and Cooney, J.J. (1981) Effects of stannic chloride and organotin compounds on estuarine microorganisms. Dev. Ind. Microbiol. 22,529-535. Hallas, L.E., Means, J.C. and Cooney, J.J. (1982) Methylation of tin by estuarine microorganisms. Science 215, 1505-1507. Harel-Bronstein, M., Dibrov, I?, Olami, Y., Pinner, E., Schuldiner, S. and Padan, E. (1995) MHI, a second-site revertant of an Escherichia coli mutant lacking Na+/H+antiporters (bnhaACuthaB), regains Na+ resistance and a capacity to excrete Na+ in a A ~ H + independent fashion. J. Biol. Chem. 270,3816-3822. Harrison, G.I., Laishley, E.J. and Krouse, H.R. (1980) Stable isotope fractionation by Clostridium pasteurianwn. 3. Effect of SeOf- on the physiology and associated sulfur isotope fractionation during SO:- and SO:- reduction. Can. J. Microbiol. 26,952-958. Hartmann, A. and Braun, V. (1981) Iron uptake and iron limited growth of Escherichia coli K-12. Arch. Microbiol. 130,353-356. Harwood, V.J. and Gordon, AS. (1994) Regulation of extracellular copper-binding proteins in copper-resistant and copper-sensitive mutants of Kbrio alginolyticus. Appl. Environ. Microbiol. 60, 1749-1753. Hamood-Sears, V. and Gordon, A S . (1990) Copper-induced production of copper-binding Supernatant proteins by the marine bacterium Kbrio alginolyticus. Appl. Environ. Microbiol. 56, 1327-1332. Hassett, R. and Kosman, D.J. (1995) Evidence for Cu(1I)reduction as a component of copper uptake by Saccharomyces cerevisiae. J. Biol. Chem. 270, 128-134. Heise, R., Muller, V. and Gottschalk, G. (1993) Acetogenesis and ATP synthesis in
238
T.J. BEVERIDGE eta/.
Acetobacteriwn woodii are coupled via a transmembrane primary sodium gradient. FEMS Micmbiol. Lett. 112,261-268. Helm, D., Labischinski, H., Schallehn, G. and Naumann, D. (1991) Classification and identification of bacteria by Fourier-transform infrared spectroscopy. J. Gen. Microbiol. 137,69-79. Holan, Z.R. and Volesky, B. (1994) Biosorption of lead and nickel by biomass of marine algae. Biotech. Bioeng. 43,1001-1009. Hubbard, J.A.M., Lewandowska, K.B., Hughes, M.N. and Poole, R.K. (1986) Effects of iron-limitation of Escherichia coli on growth, the respiratoxy chains and gallium uptake. Arch. Micmbiol. 146, 80-86. Hughes, M.N. (1981) The Inorganic Chemistry of Biological Processes. John Wiley, Chichester. Hughes, M.N. and Poole, R.K.(199 1) Metal speciation and microbial growth-the hard (and soft) facts. J. Cen. Micmbiol. 137,725-734. Hughes, M.N. and Poole, R.K.(1989) Metals and Microorganisms. Chapman and Hall, London. Iwasaki, H., Saigo, T. and Matsubara, T. (1980) Copper as a controlling factor of anaerobic growth under N20 and biosynthesis of N20 reductase in denitrifying bacteria. Plant Cell Physiol. 21, 1573-1584. Jardim, W.F. and Pearson, H.W. (1985) Copper toxicity to cyanobactena and its dependence on extracellular ligand concentration and degradation. Micmb. Ecol. 11,139-148. Jarrell, K.F., Saulnier, M. and Ley, A. (1987) Inhibition of methanogenesis in pure cultures by ammonia, fatty acids and heavy metals, and protection against heavy-metal toxicity by sewage sludge. Can. J. Micmbiol. 33,551-554. Ji, G. and Silver, S. (1995) Bacterial resistance mechanisms for heavy metals of environmental concern. J. Ind Micmbiol. 14,61-75. Ji, G., Garber, E.A.E., Armes, L.G., Chen, C., Fuchs, J.A. and Silver, S. (1994) Arsenate reductase of Staphylococcus aurew plasmid pI258. Biochemistry 33,7294-7299. Johnson, A.C. and Wood, M. (1990) DNA, a possible site of action of aluminium in Rhizobiwn spp. Appl. Environ. Micmbiol. 56,3629-3633. Johnson, D.B. and McGinness, S. (1991) A highly efficient and universal solid medium for growing mesophilic and moderately thermophilic, iron-oxidizing, acidophilic bacteria. J. Micmbiol. Meth. 13, 113-122. Jones, C.W. and Poole, R.K. (1985) The analysis of cytochromes. In: Methods in Microbiology 18 (G. Gottschalk, ed.), pp. 285-328. Academic Press, London. Jones, R.P. and Gadd, G.M. (1990) Ionic nutrition of yeast-physiological mechanisms involved and applications for biotechnology. Enzyme Micmb. Technol. 12,402418. Kaplan, J.H. (1990) Photochemical manipulation of divalent cation levels. Ann. Rev. Physiol. 52,897-914. Kaplan, J.H. and Somlyo, A.P. (1989) Flash photolysis of caged compounds: new tools for cellular physiology. Trends Neurosci. 12.54-59. Kingsley, M.T. and Bohlool, B.B. (1992) Extracellular polysaccharide is not responsible for aluminium tolerance of Rhizobium leguminosarum bv. phaseoli CIAT899. Appl. Environ. Microbiol. 58, 1095-1101. Kinraide, T.B. (1991) Identity of the rhizotoxic aluminium species. Plant Soil 134,167-178. Klapcinska, B. and Chmielowski, J. (1986) Binding of germanium to Pseudomonasputida cells. Appl. Environ. Microbiol. 51, 1144-1147. Kot, E. and Bezkorovainy, A. (1993) Distribution of accumulated iron in BijZdobacterium thermophilum. J. Agric. Food Chem. 41,177-181. Laddaga, R.A. and Silver, S. (1985) Cadrniumuptake in Escherichiu coli K-12. J. Bacteriol. 162, 1100-1105.
METAL-MICROBE INTERACTIONS:CONTEMPORARY APPROACHES
239
Laddaga, R.A., Bessen, R. and Silver, S. (1985) Cadmium-resistant mutant of Bacillus subtilis 168 with reduced cadmium transport. J. Bacteriol. 162, 1106-1110. Landeen, L.K., Yahya, M.T. and Gerba, C.P. (1989) Efficacy of copper and silver ions and reduced levels of free chlorine in inactivation of Legionellapneumophila. Appl. Environ. Microbiol. 55,3045-3050. Lee, H., Trevors, J.T. and Van Dyke, M.I. (1990) Microbial interactions with germanium. Biolechnol. Adv. 8,539-546. Lesuisse, E., Casteras-Simon, M. and Labbe, P. (1995) Ferrireductase activity in Saccharomyces cerevisiae and other fungi: colorimetric assays on agar plates. Anal. Biochem. 226,375-377. Lin, C.-M. and Kosman, D.J. (1990) Copper uptake in wild type and copper metallothioneindeficient Saccharomyces cerevisiae. J. Biol. Chem. 265,9194-9200. Lin, C.-M., Crawford, B.F. and Kosman, D.J. (1993a) Distributionof 64Cuinsaccharomyces cerevisiae: cellular locale and metabolism. J. Gen. Micmbiol. 139, 1605-1615. Lin, C.-M., Crawford B.F. and Kosman, D.J. (1993b) Distribution of @Cuin Saccharomyces cerevisiae: kinetic analysis of partitioning. J. Gen. Microbiol. 139. 1615-1626. Lovley. D.R., Phillips, E.J.P.,Gorby, YA. and Landa, E.R. (1991) Microbial reduction of uranium. Nature 351,413416. Lytton, S.D., Mester, B., Libman, J., Shanzer, A. and Cabanthik, Z.I. (1992) Monitoring of iron(II1) removal from biological sources using a fluorescent siderophore. Anal. Biochem. 205,326-333. Macaskie, L.E. and Dean, A.C.R. (1984) Cadmium accumulation by a Citrobacter sp.. J. Gen. Microbiol. 130,53-62. Macaskie,L.E.,Dean,A.C.R.,Cheetham,A.K.,Jakeman,R.J.B. andSkamoulis,A.J. (1987) Cadmium accumulation by a Citmbacter species. The chemical nature of the accumulated metal precipitate and its location on the bacterial cells. J. Gen. Microbiol. 133, 539-544. Mahan, C.A., Majidi, V. and Holcombe, J.A. (1989) Evaluation of the metal uptake of several algae strains in a multicomponent matrix utilizing inductively coupled plasma emission spectrometry. Anal. Chem. 61,624-627. Matzanke, B.F., Muller, G.I., Bill, E. and Trautwein, A.X. (1989) Iron metabolism of Escherichia coli studied by Mossbauer spectroscopy and biochemical methods. EUKJ. Biochem. 183,371-379. McCray, J.A. and Trentham, D.R. (1989) Properties and uses of photoreactive caged compounds. Ann. Rev. Biophys. Biophys. Chem. 18,239-270. McDonald, L. and Trevors, J.T. (1988) Review of tin resistance, accumulation and transformations by microorganisms. Water Air Soil Pollut. 40,215-221. McLean, R.J.C. and Beveridge, T.J. (1990) Metal-binding capacity of bacterial surfaces and their ability to form mineralized aggregates. In: Micmbial Mineral Recovery (H.L. Ehrlich and C.L. Brierley, eds), pp. 185-222. McGraw-Hill, New York. Meyer, J.-M. and Hohnadel, D. (1992) Use of nitriloacetic acid (NTA) by Pseudomonas species through iron metabolism. Appl. Microbiol. Biotechnol. 37, 114-118. Milodowski, A.E., West, J.M., Peace, J.M., Hyslop, E.K., Basham, I.P. and Hooker, P.J. ( 1990) Uranium-mineralized microorganisms associated with uraniferrous hydrocarbons in southwest Scotland. Nature 347,465467. Minnick,A.A., Eizember,L.E.,McKee, J.A.,Dolence,E.K. andMiller,M.J. (199l)Bioassay for siderophore utilization by Candida albicans. Anal. Biochem. 194,223-229. Misra, K.T. (1992) Bacterial resistances to inorganic mercury salts and organomercurials. Plasmid 27,4-16. Mitra, R.S. and Bemstein, LA. (1978) Single-strand breakage in DNAof Escherichia coli exposed to Cd2+.J. Bacteriol. 133,75-80.
240
T.J. BEVERIDGE eta/.
Mobley, H.L.T., Gamer, R.M. and Bauerfeind, P. (1995) Helicobacter pylori nickeltransport gene n M : synthesis of catalytically active urease in Escherichia coli independent of growth conditions. Mot. Microbiol. 16,97-109. Modak, S.M. and Fox, C.L. (1973) Binding of silver sulfadiazine to the cellular components of Pseudomonas aeruginosa. Biochem Phamtacol. 22,2391-304. Mueller, C.S., Thompson, R.L., Ramelow, G.J., Beck, J.N., Langley, M.P., Young, J.C. and Casserly, D.M. (1987) Distribution of Al, V and Mn in lichens across Calcasieu Parish, Louisiana. WaterAir Soil Pollut. 33, 155-164. Mullen, M.D., Wolf, D.C., Fenis, EG., Beveridge, T.J., Flemming, C.A. and Bailey, G.W. (1989) Bacterial sorption of heavy metals. Appl. Environ. Microbiol. 55,3143-3149. Nakamura, K. and S. Silver. (1994) Molecular analysis of mercury-resistant Bacillus isolates from sediment of Minamata Bay, Japan. Appl. Environ. Microbiol. 60,4596-4599. Neilands, J.B. (1984) Methodology of siderophores. Struct. Bonding 58, 1-24. Neilands, J.B. (1989) Siderophore systems of bacteria and fungi. In: Metal Ions andBacteria (T.J. Beveridge and R.J. Doyle, eds), pp. 141-163. John Wiley and Sons, New York. Nelson, W.H. and Sperry, J.F. (1991) UV resonance Raman spectroscopic detection and identification of bacteria and other microorganisms. In: Modern Techniquesfor Rapid Microbiological Analysis (W.H. Nelson, ed.), pp. 97-143. VCH Publishers, New York. Newby, PH. and Gadd, G.M. (1986) Diffusion assays for determination of the effects of pH on the toxicity of heavy-metal compounds towards filamentous fungi. Environ. Technol. Lett. 7,391-396. Nicholls, P.D., Henson, J.M., Guckert, J.B., Niven, D.E. and White, D.C. (1985) Fourier transfonn-infrared spectroscopic methods for microbial ecology: analysis of bacteria, bacteria-polymer mixtures and biofilms. J. Microbiol. Meth. 4,79-94. Nies, D.H. (1992) Resistance to cadmium, cobalt, zinc and nickel in microbes. Plasmid 27, 17-28. Nies, D.H., Nies, A., Chu, L. and Silver, S. (1989) Expression and nucleotide sequence of a plasmid-determined divalent cation efflux system from Alcaligenes eutrophus. Proc. Nut1 Acud. Sci. USA 86,7351-7355. Novick, R.P. and Roth, C. (1968) Plasmid-linked resistance to inorganic salts in Stuphylococcus aereus. J. Bacteriol. 95, 1335-1342. Ogino, T., den Hollander, J.A. and Shulman, R.G. (1983) 39K, 23Na and 31PNMR studies of iron transport in Sacchuromyces cerevisiae. Proc. Natl Acad. Sci. USA 80,5185-5189. O’Halloran, T.V., Frantz, B., Shin, M.K., Ralston, D.M. and Wright, J.G. (1989) TheMerR heavy metal receptor mediates positive activation in a topologically novel transcription complex, Cell 56, 119-129. Ohyama, T., Igarashi, K. and Kobayashi, H. (1994) Physiological role of the chaA gene in sodium and calcium circulations at a high pH in Escherichia coli. J. Bacteriol. 176, 4311-4315. Pacheo, S.V., Miranda, R. and Cervantes, C. (1995) Inorganic-ion resistance by bacteria isolated from a Mexico City freeway. Antonie van keuwenhoek 67, 333-337. Painter, E.P. (1941) The chemistry and toxicity of selenium compounds, with special reference to the selenium problem. Chem. Rev. 28, 179-213. Pettersson, A., Kunst, L., Bergman, B. and Roomans, G.M. (1985) Accumulation of aluminium by Anabaena cylindrica into polyphosphate granules and cell walls: an X-ray energy-dispersive microanalysis study. J. Gen. Microbiol. 131, 2545-2548. Phillips, S.E. and Taylor, M.L. (1976) Oxidation of arsenite to arsenate by Alcaligenes faecalis. Appl. Environ. Microbial. 32,392-399. Pirt, S.J. (1975) Principles of Microbe and Cell Cultivation. Blackwell, Oxford. Plessner, O., Klapatch, T. and Guerinot, M.L. (1993) Siderophore utilization by Bradyrhizobium japonicum. Appl. Environ. Microbiol. 59,1688-1690.
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
233
Adams, T.J., Vartivarian, S . and Cowart, R.E. (1990) Iron acquisition systems of Listeria monocytogenes. Infect. Immun. 58,2715-2718. Ahmad, A., Hughes, M.N., Nobar, A.M. and Poole, R.K (1992) Change of speciation of Cu(I1) in growth medium due to the uptake of ammonia by P s e h m o n a s testosteroni during growth. Curr. Microbiol. 25,57-59. Aiking, H., Govers, H. and Van Riet, J. (1985) Detoxification of mercury, cadmium and lead in Klebsiella aerogenes NCM3 418 growing in continuous culture. Appl. Environ. Microbiol. 50. 1262-1267. Allison, J.D., Brown, D.S. and Novo-Gradac, K.J. (1990)A GeochemicalAssessment Model for Environmental System: Version 3.0. Environmental Research Laboratory, Office of Research and Development, US Environmental Protection Agency, Athens, GA. Anderson, G.L., Williams, J. and Hille, R. (1992) The purification and characterization of arsenite oxidase from Alcaligenes faecalis, a molybdenum-containing hydroxylase. J. Biol. Chem. 267,23674-23682. Andrews, S.C., Smith, J.M.A, Hawkins, C., Williams, J.M., Harrison, P.M. and Guest, J.R. (1993) Overproduction, purification and characterization of the bacterioferritin of Escherichia coli and a C-terminally extended variant. Eu,: J. Biochem. 213,329-338. Archibald, F.S. (1983) Lactobacillus plantarum, an organism not requiring iron. FEMS Microbiol. Lett. 19, 29-32. Archibald, F.S. and deVoe, LW. (1978) Iron in Neisseria meningitidis: minimum requirements, affects of limitation and characterization of uptake. J. Bacteriol. 1 3 6 , 3 5 4 8 , Augier, H., Ronneau, C., Roucoux, P., Lion, R. and Charlent, 0. (1991) Neutron-activation analysis of the elementary composition of the marine phanerogam Posidoniu oceanica from a reference area in Port Cros National Park (French Mediterranean). Mar Biol. 109, 345-353. Avery, S.V., Codd, G.A. and Gadd, G.M. (1991) Caesium accumulation and interactions with other monovalent cations in the cyanobacterium Synechocystis PCC 6893. J. Gen. Microbiol. 137,405-413. Azam, F. and Volcani, B.E. (1974) Role of silicon in diatom metabolism V1. Active transport of germanic acid in the heterotrophic diatomNitzschia alba. Arch. Microbiol. 101, 1-8. Babich, H. and Stotzky, G. (1983) Influence of chemical speciation on the toxicity of heavy metals to the microbiota. In: Advances in Environmental Science and Technology.Aquatic Toxicology (J.O. Nriagu, ed.),Vol. 13, pp. 1-46. John Wiley and Sons, New York. Babich, H. and Stotzky, G. (1985) Heavy-metal toxicity to microbe-mediated ecologic processes: a review and potential application to regulatory processes. Environ. Res. 36, 111-137. Bak, E, Schuhmann, A. and Jansen, K.H. (1993) Determination of tetrathionate and thiosulfate in natural samples and microbial cultures by a new, fast and sensitive ion chromatographic technique. FEMS Microbiol. k t t . 12,257-264. Barrett, J., Ewart, D.K., Hughes, M.N. and Poole, R.K. (1993) Chemical and biological pathways in the bacterial oxidation of menopyrite. FEMS Microbiol. Rev. 11,5742. Barug, D. (1981) Microbial degradation of bis(tributy1tin) oxide. Chernosphere 10, 11451154. Bashford, C.L., Chance, B., Lloyd, D. and Poole, R.K. (1980) Oscillations of redox states in synchronously dividing cultures of Acanthamoeba castellanii and Schizosaccharomyces pombe. Biophys. J . 29, 1-12. Batley, G.E. (1989) (ed.) Trace Element Speciation: Analytical Methods and Problems. CRC Press, Bwa Raton, FL. Bazylinski, D.A., Garratt-Reed, A.J., Abedi, A. and Frankel, R.B. (1993) Copper association with iron sulfide magnetosomes in a magnetotactic bacterium. Arch. Microbiol. 160, 35-42.
234
T.J. BEVERIDGE eta/.
Beard, S.J., Ciccognani, D.T., Hughes, M.N. and Poole, R.K. (1992) Metal ion-catalysed hydrolysis of ampicillin in microbiological growth media. FEMS Microbiol. Lett. 95, 207-212. Belliveau, B.H. and Trevors, J.T. (1989,) Mercury resistance and detoxification in bacteria. Appl. Organometallic Chem. 3,283-294. Belliveau, B.H. and Trevors, J.T. (1989b) Transformation of Bacillus cereus vegetative cells by electroporation. Appl. Environ. Microbiol. 55, 1649-1652. Belliveau, B.H. and Trevors, J.T. (1990) Mercury resistance determined by a selftransmissible plasmid in Bacillus cereus. Biol. Metals 3, 188-196. Bemhard, M., Brinckman, F.E. and Sadler, P.J. (eds) (1986) The Imporlance of Chemical Speciation in Environmental Processes. Springer-Verlag, Berlin. Beveridge, T.J. (1989a) Role of cellular design in bacterial metal accumulation. Ann. Rev. Microbiol. 43,293-316, Beveridge, T.J. (1989b) Interactions of metal ions with components of bacterial cell walls and their biomineralisation. In: Metul-Microbe fnteractions (R.K. Poole and G.M. Gadd, eds), pp. 65-83. IRL Press, Oxford. Beveridge, T.J. and Doyle, R.J. (eds) (1 989) Metal Ions andBacteria. John Wiley and Sons, New York. Beveridge, T.J., Popkin, T.J. and Cole, R.M. (1994) Electron microscopy. In: Methodsfor General Molecular Bacteriology (P. Gerhardt, ed.), pp. 42-71. American Society of Microbiology, Washington, DC. Bird, N.P., Chambers, J.G., Leech, R.W. and Cummins, D. (1985) Anote on the use of metal species in microbiological tests involving growth media. J. Appl. Bacteriol. 59,353-355. Bitton, G. and Freihofer, V. (1978) Influence of extracellularpolysaccharides on the toxicity of copper and cadmium toward Klebsiella aerogenes. Microb. Ecol. 4, 119-125. Blair, W.R., Olson, G.J., Brinckman, F.E. and Iverson, W.P. (1982) Accumulation and fate of tri-n-butyltin cation in estuarine bacteria. Microb. Ecol. 8,241-251. Borst-Pauels, G.W.F.H. (1981) Ion transport in yeast. Biochim. Biophys. Acta. 650,88-127. Bremer, P.J. and Geesey, G.G.(1991) Laboratory-based model of microbiologically induced corrosion of copper. Appl. Environ. Microbiol. 57, 1956-1962. Broer, S., Ji, G., Broer, A. and Silver, S. (1993) Arsenic efflux governed by the arsenic resistance determinant of Staphylococcus aureus plasmid pI258. J. Bacteriol. 175, 3480-3485. Brown, N.L., Rouch, D.A. and Lee, B.T.O. (1992) Copper resistance determinants in bacteria. Plasmid 27,41-51. Brown, T.A. and Shrift, A. (1980) Assimilation of selenate and selenite by Salmonella typhimuriwn. Can. J. Microbiol. 26,671-675. Bucheder, F. and Broda, E. (1974) Energy-dependent zinc transport in Escherichia coli. E m J. Biochem. 45,555-559. Cabral, J.P.S. (1992) Limitations of the use of ion-selective electrode in the study of the uptake of Cu(1I) by Pseudomonus syringae cells. J. Microb. Meth. 16, 149-156. Cervantes, C. and Gutierrez-Corona, F. (1994) Copper resistance in bacteria and fungi. FEMS Microbiol. Rev. 14, 121-138. Cervantes, C., Ji, G., Ramirez, J.L. and Silver, S. (1994) Resistance to arsenic compounds in microorganisms. FEMS Microbiol. Rev. 15,355-367. Cha, J . 3 . and Cooksey, D.A. (1993) Copper hypersensitivity and uptake in Pseudomonas syringae containing cloned components of the copper resistance operon. Appl. Environ. Microbiol. 59, 1671-1674. Chance, B. (1966) The recording of enzyme reactions within single cells and in tissues of living animals. J. Franklin Inst. 282,349-356. Chau, YK. (1992) Chromatographic techniques in metal speciation. Analyst 117,571-575.
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
235
Chau, Y.K. and Wong, P.T.S. (1989) Applications of gas chromatography to trace element speciation. In: Trace Element Speciation: Analytical Methods and Problem (G.R.Batley, ed.), pp. 219-244. CRC Press, Boca Raton, FL. Chen, S.S.,Yoshiiiara. H., Harada, N., Seiyama,A., Watanabe, M., Kosaka, H., Kawano, S., Fusamoto, H., Kamada, T. and Shiga, T. (1993) Measurement of redox states of mitochondrial cytochrome aa3 in regions of liver lobule by reflectance microspectroscopy. Am. J. Physiol. 264, G3754382. Chiong, M., Barra, G.E., Barra, R. and Vasquez, C. (1988) Purification and biochemical characterization of tellurite-reducing activities from Thermus thermophilus HB8. J. Bacteriol. 170,3269-3273. Chmielowski, J. and Klapcinska, B. (1986) Bioaccumulation of germanium by Pseudom n m putida in the presence of two selected substrates. AppL Environ. Microbiol. 51, 1099-1 103. Ciccognani, D.T., Hughes, M.N. and Poole, R.K. (1992) Carbon monoxide-binding properties of the cytochrome bo quinol oxidase complex in Escherichia coli are changed by copper deficiency in continuous culture. FEMS Microbiol. Lett. 94, 1-6. Ciriolo, M.R., Civitareale, P.,Carri, M.T., DeMartino, A., Galiazzo, F. and Rotilio, G. (1994) Purification and characterization of Ag,Zn-superoxide dismutase from Saccharomyces cerevisiae exposed to silver. J. Biol. Chem. 269,25783-25787. Collins, Y.E. and Stotzky, G. (1989) Factors affecting the toxicity of heavy metals to microbes. In: Metal Ions and Bacteria (T.J. Beveridge and R.J. Doyle, eds.), pp. 31-90. John Wiley and Sons, New York. Cooksey, D.A. (1993) Copper uptake and resistance in bacteria. Mol. Microbiol. 7, 1-5. Cooper, W.C. (1971) Analytical chemistry of tellurium. In: Tellurium (W.C. Cooper, ed.), pp. 281-312. Van Nostrand Reinhold Company, New York. Corbisier, P., Nuyts, G., Ji, G., Mergeay, M. and Silver, S. (1993) l d B gene fusions with the arsenic and cadmium resistance operons of Staphylococcus aureus plasmid p1258. FEMS Microbiol. Lett. 110,231-238. Cui, X.Y., Joannou, C.L., Hughes, M.N. and Cammack, R. (1992) The bacteriocidal effects of transition metal complexes containing NO' on the food spoilage bacterium Clostridium sporogenes. FEMS Microbiol. Lett. 98,67-70. De Rome, L. and Gadd, G.M. (1987) Measurement of copper uptake in Saccharomyces cerevisiae using a Cu2+-selectiveelectrode. FEMS Microbiol. Lett. 43,283-287. Dunn, G.M. and Bull, A.T. (1983) Bioaccumulation of copper by a defined community of activated sludge bacteria. Eu,: J. Appl. Microb. Biotechnol. 17, 30-34. Ebdon, L., Hill, S.J. and Ward, R.W. (1986) Directed coupled chromatography-atomic spectroscopy. A review. Analyst 111, 11 13-1 11 8. Ebdon, L., Hill., S . , Walton., A.P. and Ward., R.W. (1988) Coupled chromatography-atomic spectrometry for arsenic speciation. A comparative study. Analyst 113, 1159-1166. Ellis-Davies, G.C.R. and Kaplan, J.H. (1988) A new class of photolabile chelators for the rapid release of divalent cations: generation of caged Ca and caged Mg. J. Qrg. Chem. 53,1966-1 969. Farrell, R.E., Germida, J.J. and Huang, P.M. (1990) Biotoxicity of mercury as influenced by mercury(1I) speciation. Appl. Environ. Microbiol. 56,3006-3016. Farrell, R.E., Germida, J.J. and Huang, P.M. (1993) Effects of chemical speciation in growth media on the toxicity uf mercury(I1). Appl. Environ, Microbiol. 59, 1507-1514. Fems, EG., Shotyk, W. and Fyfe, W.S. (1989) Mineral formation and decomposition by microorganisms. In: Metal Ions and Bacteria (T.J. Beveridge and R.J. Doyle, eds), pp. 413-441. John Wiley and Sons, New York. Feteke, EA., Spence, J.T. and Emery, T. (1983) A rapid and sensitive paper electrophoresis
236
T.J. BEVERIDGE
eta/.
assay for the detection of microbial siderophores elicited in solid-plating culture. Anal. Biochem. 131,516-519. Firtel, M., Southam. G. and Beveridge, T.J. (1995) Electron microscopy. Protocol 12. In: Archaea-A Laboratory Manual (K.R. Sowers, ed.). Cold Spring Harbor Laboratory Press. Cold Spring Harbor, Ny. Fleischmann, R.D., Adams, M.D., White, O., Clayton, R.A., Kirkness, E.F. et al. (1995) Whole-genome random sequencing and assembly of Haemophilus injluenzae Rd. Science 269,496-5 12. Flis, S.E., Glenn, A.R. and Dilworth, M.J. (1993) The interaction between aluminium and root nodule bacteria: a review, Soil Biol. Biochem. 25,403-417. Florence, T.M. (1992) Trace element speciation by anodic stripping voltammetry. Analyst 117,551-553. Foster, T.J. (1983) Plasmid-determined resistance to antimicrobial drugs and toxic metal ions in bacteria. Microbiol. Rev. 47, 3 6 1 4 9 . Fraser, C.M., Gocayne, J.D., White, O., Adams, M.D., Clayton, R.A. et al. (1995) The minimal gene complement of Mycoplasrna genfilalium. Science 270,397403, Frimmel, F.H. and Geywitz, J. (1987) Direct polarographic recording of metal elimination from aquatic samples by coprecipitation with ferric hydroxide. Sci. Total Environ. 60, 57-65. Fu, C. and Maier, R.J. (1 991) Competitive inhibition of an energy-dependent nickel transport system by divalent cations in Bradyrhizobium japonicum. Appl. Environ. Microbiol. 57, 351 1-3516. Gadd, G.M. (1988) Accumulation of metals by microorganisms and algae. In: Biotechnology-A Comprehensive Treatise. Special Microbial Processes. (H.J. Rehm, ed.), pp. 401433. VCH Verlagsgesellschaft, Weinheim. Gadd, G.M. (1990) Fungi and yeasts for metal accumulation. In: MicrobialMineral Recovery (H.L. Ehrlich and C.L. Brierley, eds), pp. 249-275. McGraw-Hill, New York. Gadd, G.M. (1992) Metals and microorganisms: a problem of definition. FEMS Microbiol. Lett. 100,197-204. Gadd, G.M. (1993) Microbial formation and transformation of organometallic and organometalloid compounds. FEMS Microbiol. Rev. 11,297-3 16. Gadd, G.M. and Griffths, A.J. (1978) Microorganisms and heavy-metal toxicity. Microb. Eco~.4,303-317. Gadd, G.M. and White, C. (1989) Heavy metal and radionuclide accumulation and toxicity in fungi and yeasts. In: Metal-Microbe Interactions (R.K. Poole and G.M. Gadd, eds), pp. 19-38. IRLPress, Oxford. Gadd, G.M., Laurence, O.S., Briscoe, P.A. and Trevors, J.T. (1989) Silver accumulation in Pseseudomonus stutzeri AG259. Biol. Metals 2, 168-173. Gadd, G.M., Stewart, A,, White, C and Mowll, J.L. (1984) Copper uptake by whole cells and protoplasts of a wild-type and copper-resistant strain of Saccharomyces cerevisiae. FEMS Microbiol. Lett. 24,231-234. Gangola, E? and Rosen, B.P. (1987) Maintenance of intracellularcalcium in Escherichiucoli. J. Biol. Chem. 262,12570-12574. Geddie, J.L. and Sutherland, I.W. (1993) Uptake of metals by bacterial polysaccharides. J. Appl. Bacteriol. 74,467472. Gharieb, M.M., Wilkinson, S.C. and Gadd, G.M. (1995) Reduction of selenium oxyanions by unicellular, polymorphic and filamentous fungi: cellular location of reduced selenium and implications for tolerance. J. Ind Microbiol. 14,300-311. Ghiorse, W.C. (1984) Biology of iron- and manganese-depositing bacteria. Ann. Rev Microbiol. 38,515-550. Ghislain, M., Goffeau, A., Halachmi, D. and Eilam, Y. (1990) Calcium homeostasis and
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
241
Poole, R.K. (1983) Bacterial cytochrome oxidases: a structurally and functionally diverse group of electron transfer proteins. Biochim. Biophys. Acta 726,205-243. Poole, R.K. and Hughes, M.N. (1996)An Introduction to InorganicMicrobiology. Chapman and Hall (in preparation). Poole, R.K., Waring, A.J. andchance, B. (1979) The reaction of cytochromeo in Escherichia coli with oxygen. Low-temperature kinetic and spectral studies. Biochem. J . 184, 379-389, Poole, R.K., Scott, R.I. and Blum, H. (1981) Respiratory biogenesis during the cell cycle of aerobically grown Escherichiu coli K12. The accumulation of iron-sulphur clusters and their orientation in the membrane. J. Gen. Microbiol. 124, 181-185. Privalle, C.T., Kong, S.E. and Fridovich, 1. (1993) Induction of manganese-containing superoxide dismutase in Escherichia coli by diamide and 1 ,lo-phenanthroline: sites of transcriptional regulation. Proc. Natl Acad. Sci. USA 90,2310-2314. Ramamoorthy, S. and Kushner, D.J. (1975) Binding of the mercuric and other heavy-metal ions by microbial growth media. Microbiol. Ecol. 2, 162-176. Rayner, M.H. and Sadler, P.J. (1990) Precipitation of cadmium in a bacterial culture medium: Luria-Bertani broth. FEMS Microbiol. Lett. 71,253-258. Rosen, B.P. and Silver, S. (eds) (1987) Ion Transport in Pmkaryotes. Academic Press, San Diego, CA. Rosenberg, H. (1979) Transport of iron into bacterial cells. Meth. Entymol. 56,388-394, Rosner, J.L. and Aumercier, M. (1990) Potentiation by salicylate and salicyl alcohol of cadmium toxicity and accumulation in Escherichia coli. Antimicrob. Agents Chemother. 34,2402-2406. Roth, LE., Dunlap, J.R. and Stacey, G. (1987) Localizations of aluminium in soybean bacteroids and seeds. Appl. Environ. Microbiol. 53,2548-2553. Sadler, P.J. (1986) Multinuclear NMR methods for the in situ characterization of chemical species. In: Importance of Chemical Speciation in Environmental Processes (M. Bernhard, F.E. Brinckman and P.J. Sadler, eds), pp. 563-578. Springer-Verlag, Berlin. Sahlman, L. and Jonsson, B.H. (1992) Purification and properties of the mercuric-ionbinding protein MerP. E m J. Biochem. 205,375-381. Savvaidis, I., Nobar, A.M., Hughes, M.N. and Poole, R.K. (1988) Metal binding to Pseudomonas species, studied by dye displacement and reflectance d-d spectroscopy. Sci. Technol. Lett. 56,563-564. Savvaidis, I., Nobar, A., Hughes, M.N. and Poole, R.K. (1990) Displacement of surfacebound cationic dyes: a method for the rapid and semi-quantitative assay of metal binding to microbial cell surfaces. J. Microb. Meth. 11, 95-106. Savvaidis, I,. Hughes, M.N. and Poole, R.K. (1992) Differential pulse polarography: a method of directly measuring uptake of metal ions by live bacteria without separation of biomass and medium. FEMS Microbiol. Lett. 92, 181-186. Schmidt, T. and Schlegel, H.G. (1994) Combined nickel-cobalt-cadmium resistance encoded by the ncc locus of Alcaligenes xylosoxidans 3 1A. J. Bacteriol. 176,7045-7054. Schwyn, B. and Neilands, J.B. (1987) Universal chemical assay for the detection and determination of siderophores. Anal. Biochem. 160,47-56. Scott, R.I., Poole, R.K. and Chance, B. (1981) Respiratory biogenesis during the cell cycle of aerobically grown Escherichia coli K12. The accumulation and ligand binding of cytochrome 0.J. Gen. Micmbiol. 122,255-261. Shulman R.G., Brown, T.R., Ugurbil, K. and Ogawa, S. (1979) Cellular applications of 31P and I3C NMR. Science 205,160-166. Shuttleworth, K.L. and Unz, R.F. (1991) Influence of metals and metal speciation on the growth of filamentous bacteria. Waer Res. 25, 1177-1 186.
242
T.J. BEVERIDGE
t?t a].
Shuttleworth, K.L. and Unz, R.F. (1993) Sorption of heavy metals to the filamentous bacterium Thiothrix Strain AT. (1993) Appl. Envimn. Microbiol. 59, 1274-1282. Silver, S. (1978) Transport of cations and anions. In: Bacterial Transport (B.P. Rosen, ed.), pp. 221-324. Marcel Dekker, New York. Silver, S. (1982) Bacterial interactions with mineral cations and anions: good ions and bad. In. Biomineralization and Biological Metal Accumulation fP. Westbroek and E.W. DeJong, eds), pp. 439-457. D. Reidel, Dordrecht. Silver, S. (1996) Transport of inorganic cations, In: Escherichia coli and Salmonella: Cellular and MolecuEnr Biology, 2nd edn (EC. Neidhardt et al., eds), pp. 1091-1102. American Society for Microbiology, Washington, Dc. Silver, S. and Phung, L.T. (1996) Bacterial plasmid-mediated heavy-metal resistances: new surprises. Ann. Rev. Microbiol. 50, (in press). Silver, S . and Walderhaug, M. (1992) Gene regulation of plasmid- and chromosomedetermined inorganic ion transport in bacteria. Microbiol. Rev. 56, 195-228. Silver, S., Laddaga, R.A. and Misra, T.K. (1989) Plasmid-determined resistance to metal ions. In: Metal-Microbe Interactions (R.K. Poole and G.M. Gadd, eds), pp. 49-63. IRL Press, Oxford. Slavin, W. (1988) Atomic absorption spectrometry. Meth. Enqrnol., 158, 117-156. Slawson, R.M., Van Dyke, M.I., Lee, H. and Trevors, J.T. (1992) Germanium and silver resistance, accumulation and toxicity in microorganisms. Plasmid 27,72-79. Snavely, M.D., Gravina, S.A., Cheung, T.T. and Miller, C.G. (1991) Magnesium transport in Salmonella typhimurium. J. Biol. Chem.266, 824-829. Southam, G . and Beveridge, T.J. (1994) The in vitro formation of placer gold by bacteria. Geochim. Cosmochim. Acta 58,4527-4530. Speelmans, G., Poolman, B., Abee, T.and Konings, W.N. (1993) Energy transduction in the thermophilic anaerobic bacterium Clostridiumfervidus is exclusively coupled to sodium ions. Proc. Naif Acad. Sci. USA 90,79757979. Starodub, M.E. and Trevors, J.T. (1989) Silver resistance in Escherichia coli R l . J. Med. Microbiol. 29, 101-110. Starodub, M.E. and Trevors, J.T. (1990) Silver accumulation and resistance in Escherichia coli R1. J. Inorg. Biochem. 39,317-325. Stookey, L.L. (1970) Ferrozine-a new spectrophotometric agent for iron. Anal. Chem. 42, 779-781. Stoppel, R.-D. and Schlegel, H.G. (1995) Nickel-resistant bacteria from anthropogenically nickel-polluted and naturally nickel-percolated ecosystems. Appl. Environ. Microbiol. 61,2276-2285. Summers, A.O. and Silver, S. (1978) Microbial transformations of metals. Ann. Rev. Microbiol. 32,637672. Svensson, B., Lubben, M. and Hederstedt, L. (1993) Bacillus subtilis CtaAandCtaB function in haem-A biosynthesis. Mol. Microbiol. 10, 193-201. Tan, H.-M.,Joannou, C.L., Cooper, C.E., Butler, C.S., Cammack, R. and Mason, J.R. (1 994) The effect of ferredoxinBED overexpression on benzene dioxygenase activity in Pseudom o m putida ML2. J. Bucteriol. 176,2507-25 12. Taylor, D.E., Walter, E.G., Sherbume, R. and Bazett-Jones, D.P. (1988) Structure and location of tellurium deposited in Escherichia coli cells harboring tellurite resistance plasmids. J. Ultrastruct. Mol. Struct. Res. 99,18-26. Tisa, L.S. and Adler, J. (1992) Calcium ions are involved in Escherichia coli chemotaxis. Proc. NatlAcad. Sci. USA 89, 11804-11808. Tisa, L.S. and Rosen, B.P. (1990) Transport systems encoded by bacterial plasmids. J. Bioenerg. Biomemb. 22,493-507. Tomei, F.A., Barton, L.L.. Lemanski, C.L. and Zocco, T.G. (1992) Reduction of selenate and
METAL-MICROBE INTERACTIONS: CONTEMPORARY APPROACHES
243
selenite to elemental selenium by Wolinella succinogenes. Can. J. Microbiol. 38, 1328-1333. Tomei, F.A.,Barton, L.L., Lemanski, C.L.,Zocco, T.G., Fink, N.H. and Sillerud, L.O. (1995) Transformation of selenate and selenite to elemental selenium by Desulfovibrio desulfuricans. J. Ind. Microbiol. 14,329-336. Tornabene, T.G. and Edwards, H.W. (1972) Microbial uptake of lead. Science 176, 1334-1335. Trevors, J.T., Stratton, G.W. and Gadd, G.M. (1986) Cadmium transport, resistance and toxicity in bacteria, algae and fungi. Can.J. Microbiol. 32,447-464. Van Dyke, M.I., Lee, H. and Trevors, J.T. (1989) Germanium accumulation by bacteria. Arch. Microbiol. 152,533-538. Van Dyke, M.I., Parker, W., Lee, H. and Trevors, J.T. (1990) Germanium accumulation in Pseudomonas stuizeri AG259.Appl. Microbiol. Biotechnol. 33, I1 6-720. Van Loon, J.C. (1979) Metal speciation by chromatography/atomic spectrometry. Anal. Chem. 51,1139A-1150A. Walter, E.G. and Taylor, D.E. (1992) Plasmid-mediatedresistance to tellurite: expressed and cryptic. Plasmid 2 7 , 5 2 4 . Watanabe, M., Takamatsu, T., Kohata, K. and Kunugi, M. (1989) Luxury phosphate uptake and variation of intracellular metal concentrations in Heterosigma akashiwo. J. Phycol. 25,428-436. Watterson, J.R. (1991) Preliminary evidence for the involvement of budding bacteria in the origin of Alaskan placer gold. Geology 20,315-318. Welz, B. (1985) Atomic Absorption Spectrometry. VCH, Weinheim. Williams, J.W. and Silver, S. (1984) Bacterial resistance and detoxification of heavy metals. Enz. Microb. Technol. 6,530-537. Wong, P.T.S., Maguire, R.J., Chau, Y.K. and Kramar,0. (1984) Uptake and accumulation of inorganic tin by a freshwater algae, Ankistmdesmus falcatus. Can. J. Fish. Aquat. Sci. 41,1570-1574, Wright, J.G., Tsang, H.-T., Penner-Hahn, J.E. and O’Halloran, T.V. (1990) Coordination chemistry of the Hg-MerR metalloregulatory protein: evidence for a novel tridentate Hg-cysteine receptor site. J. Am. Chem. SOC. 112,2434-2435. Yahya, M.T.,Landeen, L.K.,Messina, M.C., Kutz, S.M., Schulze, R. and Gerba, C.P. (1990) Disinfection of bacteria in water systems by using electrolytically generated coppersilver and reduced levels of free chlorine. Can. J. Micmbiol. 36, 109-116.
This Page Intentionally Left Blank
Author Index Numbers in bold refer to pages on which references are listed at the end of each chapter Abdel-Wahab, N.H., 53,57,59,60,67,83 A M , A., 223,233 A h , T., 2M), 242 Abraham, E.P., 49,81 Abul-Haj, Y.J., 6, 8, 18,31,41 Adams, M.D.. 211,236 Adams, M.W.W., 180,232 Adams, S.R., 188,189,232 Adams, T.H., 27.36 Adams, T.J., 217,233 Adler, J,, 189, 242 Adlington, R.M.,49,81 Agosin, E., 13,40 Aharonowitz, Y., 94,%, 97,124,129 Ahmad, A., 186,233 Ahmed, E.D., 165,166 Aiking, H., 227.228,233 Ainsworth, A.M., 2,40 Aitken, A., 60.81 Alatalo, E., 5, 8,40 Alexander, M., 135, 159,160,167,169,172, 173 Alijah, R.,121,131 Alison, J.E., 88, 92,98,99,129 Allison, J.D., 186,233 Allison, N., 141,166,167 Altier, D.J., 55, 74,76 An, D.,50,55,58,76 Andersen, N.H., 8.37 Anders, M.W., 165,166 Anderson, G.L., 225,233 Andrew, S.C., 216,233 Anzai, H., 121,126,128 Apel, K., 9.37 Aramaki, H., 57, SO Archibald, F.S., 188, 190,233 Arends, J., 17,37 Armes, L.G., 226,238
Armfield, S.J., 160, 164,165,166,173 Am. N., 58.40
Aronson, A.I., 11,37 Arreguin, B., 8,37 Asante-Owusu, R.N., 23,24,37 AsgeirsdMr, S.A., 3, 4. 5 6 , 10. 12, 13, 14, 19, 20,21,41,44 Askendal, A., 16.38 Asmara, W., 138,142,149,166,167,172 Assis, H.M.S., 154, 158, 167 Asther, M., 10,38 Atherton, E.. 111.128 Atsuyuki, S., 121,126 Atto-Asato-Adjei, E., 68,69,70.71,72 Augier, H., 197,233 Aumercier, M., 224,241 Ausubel, EM.,146,169 Avery, S.V., 183.233 Avi-Dw, Y.,209,237 Axcell, B.C., 50,53,58,76 Axelrod, B.,73,81 h a m , F., 227,233 Babich, H., 182, 183, 185,233 Bachman, B., 48,W Bader,R.. 136,137,140,142,171 Bailey, G.W., 183.240 Baines, A.J.. 138, 142,149,167,172 Bairoch, A., SO, 55,61,66,73,81 Bak, F., 197,233 Baker, D.P., 56,82 Baker. P.B., 165,166 Baldwin, J.E., 49,81,97,98, 117, 124,129 Ballou, D.P., SO, 51,53,55,56,57,60,61,64,
65,66,67,69,72,75,76,77,78,79, 80,= Banham, A.H., 23,24,37 Bapat, M., 112,125 Barra, G.E., 230,235
246 Barra, R., 230,235 Bmedo, J.L., 88,98,125 Barrett, J., 194,226, 233 Bamault, D., SO, 80 Barth, P.T., 138, 139, 142, 143, 149, 165,167. 170 Bartilson, M., 146,173 Bartnicki,E.W., 151,152,153,155,167 Barton, B., 88,92,98,99,129 Barton, L.L., 229,242,243 Barug, D., 231,233 Basham, LP., 183,239 Bashford, C.L., 219,233 Bates, R., 166,169 Batie, C.J., 50.51, 53,5556, 57,60, 61, 64.65, 66. 67,69, 72, 76, 77, 79, 80,83 Batley, G.E., 185, 233 Bauerfeind, P., 224, 240 Bayer, E., 120,124 Bazett-Jones, D.P., 230,242 Bazylinski, D.A., 223,233 Beall, M.L., 135,171 Beard, S.J., 214, 222,234 Beardwood, P., 76 Beck, J.N., 196,240 Beckett, A., 29,38 Beckmann, J.D., 68,76 Beeching, J.R., 146,150,167 Beever. R.E., 5, 8,9, 11, 12, 17,28,33,37 Beg, F., 60,81 Belay, N., 159, 167 Belliveau, B.H., 228,229, 234 Bell-Pedersen, D.. 5 , 11, 12, 28,37 Benitez, T., 5 , 39 Benkovic, S.J., 49, 81 Berg, T.L.. 86, 124 Bergeron H., 50,83 Bergman, B., 215,240 Bergmeyer, J., 94,96,124 Bemhard, M.,185, 187,234 Bemhardt, FAI., 49, 75.77, 83,84 Benistein, LA., 226, 239 Berry, E.K.M., 141,167 Bertrand, P.,65,76,77 Bessen, R., 227,239 Best, D., 165.169 Beveridge. T.J., 178, 183, 201, 202,203,234, 236,237,239,240,242 Bezkorovainy, A., 191,216,238 Bidcchka,M.J., 11, 12, 13.37 Bill, E., 75.77, 209,216, 239 Billich, A., 100, 102. 105, 114, 124
AUTHOR INDEX Bird, J.W., 97,124 Bird, N.P., 185,234 Birkenkamp, K.U., 3,5,24,25,26,44 Bitton, G., 227,234 Blair, W.R., 231,234 Blake, C.K., 49,82 Bleasby, A.J., 148, 169 Blum, ED., 6, 8, 18,31,41 Blum, H.,72,82,220,241 Blumenthal, H.J., 10.38 Bohlmann, H., 9.37 Bohlool, B.B., 215,238 Bolin, J.T.,73,81 Bollag, J.M., 135, 167 Boller, A., 99, 128 Bolton, L., 139, 142, 167 Bolyard, M.G., 9, 18, 33,37,42 Booij, H., 9,41 Borchert, S.,90,125 Borel, J.F., 105,125 Borst-Pauels, G.W.F.H., 180, 234 Bomvka, L., 16.41 Bosscher, J., 5.6, 12, 26,39 Boucias, D.G., 29,37 Bouwer, E.J., 159, 165,167 Bowden, C.G., 5, 18, 19, 31,37 Boyd, S.A., 160,168,172 Boylan, M.T., 28,36 Brackman, J., 161, 170 Bradley, EC., 73,77 Bradshaw, R.M., 11,44 Brandt, U., 68, 72.79 Brand, J.M., 48.49, 53,77,79, 80 Brasier, C.M., 18, 31,37 Bratt, P.J., 67, 82 Braun, D.G., 102,124 Braun, V., 189,237 Bremer, P.J., 207, 234 Bnnckman,F.E., 185, 187, 231, 234 Bnscoe, P.A., 230,236 Britt, R.D., 67,69, 72,77 Broda, E., 224,234 Brodexick, J.B., 73, 75,77 Broekaen, W.F., 8,9,40,42 Broer, A., 226,234 Brokamp, A., 139,149,167 Bronchart, R., 10.37 Brown, D.S., 186,233 Brown, G.A., 17,38 Brown, N.L., 214,234 Brown, T.A., 229,234 Brown, T.R., 209,241
AUTHOR INDEX Browne, J., 88. 92,98, 99, 129 Bruins, A.P., 162,173 Brunner, W., 165,167 Bruschi, M., 59, 62.77 Bucheder, F., 224,234 Bull, A.T., 135, 136, 138, 142, 144,145, 149, 150,154, 155, 160,164,165,166, 167,168,169,172,173,174,175, 176,222,235,237 Bull, J.H., 88,92,98,99, 129 Bunz, P.V., 50,53, 56,58,59,74,77 Burgess, T., 5, 33, 39 Burgett, S.C., 99,130 Burge, W.D., 135,167 Burnham,M.K.R., 88,92,98,99,129 Busscher, H.J., 16. 17, 29, 30,37, 40,43, 44 Butler, C.S., 59, 83, 219,242 Butt, T.M., 29.38 Byford, M.E, 97,98, 117, 129 Byrom, D., 138, 143, 149,170 Cabanthik, Z.I., 217,239 Cabral, J..P.S., 196, 234 Caceres, O., 162.173 Cairns, S.S., 138, 139,167 Calzada, J.G., 98, 125 Cammack, R., 50, 53,57,58,59,60,61,63,65, 67, 68,71,72, 73,75,77,79,80,81, 82,83,208,219,235,242 Cammue, B.P.A., 9,42 Cantoral, J.M., 94,%, 124 Cao, B., 8, 37 Cao, J., 99, 130 Carey, P.R., 5,6, 9, 18,44 Carpenter, C.E., 5,6, 8, 9, 13, 18, 19,32,38 Cam, M.T., 223,235 Casselton, L.A., 5, 6, 23, 24,37, 38,39 Casserly, D.M., I%, 240 Casteras-Simon,M.,221,239 Castonguay, Y.,9.38 Castro, C.E., 151, 152, 153, 155,167 Cavalier-Smith, T., 22,38 Cervantes, C., 213,214,225,226,234,240 Cha, J.-S., 221, 234 Chakrabarty, A.M., 164,168 Chakravarti,B., 49,81 Chalatnish, S.. 209,237 Chambers, J.G., 185,234 Chance, B., 219,233,234,241 Chang, Y.C., 5,12,27,33,38,40 Charlent, O., 197,233 Chau, YK.,198,231,234,235,243 Cheetham, A.K., 209,239
247 Chen, C., 226,238 Chen, R.F., 50.84 Chen, S.S., 219,235 Chen, V.J., 49, 73,77,81, 99, 130 Cheong, C.-M., 60,61,83 Cheung, T.T., 200,242 Chiong, M., 230,235 Chmielowski, J., 227,235,238 Chu, L., 226,240 Chumleym, F.G., 29,30,38 Ciccognani, D.T.,189,222,234,235 Ciriolo, M.R., 223,235 Civitareale, P., 223,235 Claverie-Martin, E, 11,38 Clayton, R.A., 211,236 Clement, J.A., 29.38 Cline, J.F., 65, 66, 77 Clutterbuck, A.J., 27.38 Ccdd, G.A.. 183,233 Cohen, G., 94,%, 124 Colby, J., 139, 143, 165, 168, 174 Cole, A.L.J., 135, 171 Cole, G.T.,3, 10, 11.38 Cole, R.M., 201,202,203, 234 Collins, Y.E., 185, 186, 187,235 Colmer, A.R., 135,172 Columbo, E., 119,130 Commandeur, L.C.M., 134,168 Connolly, T.N., 186, 237 Cook, A.M., 50,53,55,56,57,58,59,60,64, 77,80,82, 160, 164,169,173 Cooksey, D.A., 214,221,234,235 Cooney, J.J., 231, 237 Cooper, C.E.. 59,83, 219,242 Cooper,R.A., 139, 141, 144, 166, 167,171 Cooper, W.C., 230,235 Ccque, J.J.R., 98, 125 Corbell, N., 91, 92, 120,125 Corbisier, P., 212,235 Correll, C.C., 55,56,77, 78 Correll, C.J., 51, 55,76 Cosmina, P., 119,120, 125,130 Costerton, J.W., 11.44 Cotoras, M., 13,40 Cotton, A.J., 154, 155 Cowart, R.E., 217,233 Crawford, B.F., 179,221,239 Crawford, R.L., 49,81 Crestani, B.. 5, 6, 8, 12,33,42 Cruden, D.L., 49.53,77,79,82 Crutcher, S.E., 53,64,77 Cui, X.Y., 208,235
248
AUTHOR INDEX
Cummins, D., 185,234 Curry, J.. 119,128 Daldal, F., 67.68.69, 70,71, 72, 77 Dalton. H., 48,77, 165,168,174 Daniels, L., 159,167 Davidson, E.. 67,68,69,70,71,72,77 Davies, M.M.. 165, 169 Day, E.P., 63,65,68,72,78 de Boer, P., 17.37 de Bont, J.A.M., 152, 160,169,176 de Cavalho, D., 5,33,39 De Cky-LagWd, Y.,92,125 de Ferra, F., 120, 125 de G m t , P.W.J., 5 , 3 8 de Jong, H.P., 17,37 de la Cruz, I.,5,39 de Lorenzo, V., 146,168 De Martino. A., 223,235 De Rome, L., 222,235 de Vos, W.M., 162,175 devries, O.M.H., 3,4,5,6,9, 10, 11, 12.13, 14, 17, 19,28,30,34,38,39,42,43,
44 de Vries, S.C., 19,38,41 de Wet, J.R., 90,125 de Wit, P.J.G.M., 9, 43 Dean, A.C.R., 209,227,239 Dean,R.A.,4,5, 8, 11, 12, 13, 27,28,42 Debrunner, P.G., 65,81 Deising, H., 29.40 DeLuca, M., 90,125 Dernain. A.L., 94.96.124 Demoulin, V., 10.37 Dempsey, G., 11, 12, 17,37 Den Dooren de Jong, L.E., 134,168 den Hollander, J.A., 209,240 Dengis, P.B., 17,41 Denome, S.A., 67.77 dev~e, r.w., 188,233 Diaz-Torres, M.R., 11,38 Dibrov, P., 193,200,237 Dickson, D.P.E., 65,78 Diddens, H., 120,125 Diez, B., 88,98,99,125 Dijkhuizen,L., 152, 160,161, 162,170 Dijkstra, B.W., 142, 161,168, 173,175 Dilworlh, M.J., 182,215,236 Dittmann, J., 106,127 Dixon, R.A., 145,168 Dodd, LB., 146,168 Doi, S., 50,W Dolenw, E.K., 218,239
Donamo, S., 90,126 Dons, J.J.M., 19,38 Doolittle, R.F., 6,7,39 Dorendorf. J., 121,131 Douglas, C., 90,127 Dowell. R.M., 91,129 Doyle, R.J., 178, 234 Drenth, J.H.H., 4, 8,9, 10, 12. 13, 14,20,21, 3% 44 Dreyfuss, M.M., 107,125 Drummond, M., 146,168 DSwza, C., 91,92,120,125 DuBow, M.S., 224,237 Dudding. T.. 48, 80 Dufrene, Y.F., 17,41 Dunham, W.R., 63,65,68,72,78 Dunlap, J.C., 5, 11, 12,28,37 Dunlap, J.R., 215,241 Dunn, G.M.,222,235 Ebbole, D.J., 5.8.28, 29, 30.42 Ebdon, L., 198,235 Ebersold, H.-R., 165, 174 Ebersptkher, J., 50, 53, 56.57.59, 64,82 Eckart, K., 117,127 Edwards, H.W.. 228,243 Egan, J.B., 146,168 Eggen. R.I.L., 162, 175 Eggermont, K., 9.42 Eggink, G., 57.78 Eilatn, Y,200,236 Eizember, L.E., 218,239 Ellis-Davies, G.C.R., 188,235 Elwing, H., 16,38 Emery, T., 217,235 Engel, H., 57, 78 Ensley, B.D., 50,53,55, 64,67,78, 82 Epstein, L., 29, 30, 31,40 Erge, D., 109,128 Erickson, B.D., 50,53,60,61,67,73,78 Esaki, N., 134, 138, 139,144,169,172 Ewart, D.K., 194,226,233 Farrell, R.E., 186,235 Fathepure, B.Z., 160,168 Fauzi, A.M., 158,168 Fee, J.A., 59,63,65,66,67,68, 69, 72,77, 78, 79, 80 Feig, A.L., 49,72,73,78 4, 9, 11. 13, 28, 34,38 Fekkes, M.P., Felix, A., 165,173 Felsenstein, J., 148, 168 Femandez, S., 146,168 Ferrari, M.A., 30.39
AUTHOR INDEX Fems, EG., 183,235,240 Feteke, F.A., 217,235 Fetzner, S.,48,49, 50,53,56, 57.60, 64, 78, 79.82 Field, R.A., 97,124 Finding, K.L., 63,65,68,72,78 Finette, B.A., 53, 84 Fink,N.H., 229,243 Fink, U., 94,96,124 Firtel, M., 202,203,236 Fisher, D.J., 17,38 Fitz-James, I?, 11,37 Fleischmann, R.D., 211,236 Flemming, C.A., 183,240 Flis, S.E.,182,215,236 Florence, T.M., 195,236 Forge, A., 11.44 Forgeot, M., 99,129 Foster, T.J., 226, 236 Fowden, L., 135, 168 Fox, B.G., 55, 73,74,76,81 Fox, C.L., 229,240 Foy, C.L., 135, 168 Franchi, E., 119,130 Frank, R., 138, 141,173 Franke, P.,91,93, 117,129 Frankel, R.B., 223,233 Franken, S.M., 142,161,168,175 Frantz, B., 164, 168, 229, 240 Fraser, C.M., 211,236 Frederick, C.A., 73.82 Freihofer, V.,227,234 Frey, A.J., 108, 126 Fridovich, I., 188, 189,241 Frimmel, F.H., 187, 236 Friihner, C., 50, 53,56, 57.59, 64,82 Fr(dholm,L.O., 86,124 Frolik, C.A., 49,77,99,130 Fu, C., 224, 236 Fuchs. J.A., 226,238 Fuchs, R., 148, 169 FiihrbaB, R., 88,130 Fuji, I., 93, 94,96, 123, 126 Fujisawa, H., 50,53,56,57,60,64,84 Fujishima, Y.,92, 120,125 Fukuda, M., 57.80,162,171 Fukuyama, K., 62.78 Fuma, S., 92, 120,125 Furukawa, K., 53,60,61,62,73,78,83 Fusamoto, H., 219,235 Fuse, H., 160,164,176 Futsi, F., 135, 175
249 Fyfe, W.S., 183,235 Gabellini, N., 63,67,78 Gadd, G.M., 178, 180,182, 183, 187,221,222, 226,229,230,233,235,236,238, 240.243 Galiazzo, F., 223,235 Galli, R., 165, 168, 171,174 Gangola, P., 191,236 Garber, E.A.E., 226,238 Garner, R.M., 224,240 Garnon, J.. 50,83 Garratt-Reed, A.J., 223,233 Gassner, G.T., 55,75,78 Gatti, D.L., 5 5 7 8 Gay, N.J., 90, 130 Gayda, J.-P., 65, 76, 77 Geary, P.J., 50,53,56,57,58,59,60,64,65,75, 77,78,79,81 Geddie, J.L., 197, 236 Geesey, G.G.. 207,234 Genet, M.J., 17, 41 Geoghegan, M.G., 11,38 Gerba, C.P.,192, 221,239,243 Gerin, P.A., 10,17,38,41 Gennida, J.J., 186,235 Gemtse, J., 161, 170 Gersonde, K.. 75,83 Gevers, W., 87,125, 129 Geyl, D., 107, 108,127 Geywitz, J., 187,236 Gharieb, M.M., 229,236 Ghiorse, W.C., 183,236 Ghislain, M., 200,236 Gibson, D.T., 47,48,49,50,53,55,56.57,58, 59,60,64.66,77,78,79,80,82,83, 84,162,172 Gibson, J.F., 62, 63,76,79 Gierlich,A., 9,41 Gilboa, H., 209,237 Glenn, A.R., 182,215,236 Glund, K., 112,125 Gocayne, J.D., 211,236 Gocht, M., 90,91,93,125 Goddard, P.A., 230, 237 Goffeau, A., 200,236 Giihring, W., 120,125 Gold, L., 145, 168 Goldman, P., 135, 136, 137,168,169 Goncalves, M.L.S., 195,237 Good, N.E., 186,237 Gorby, Y.A., 183,239 Gordon, A.S., 214,223,237
250 Gordon, M.P., 162,173 Gtittgens, B., 23,24,37 Conschalk, G., 200,237 Covers, H., 227,228,233 Gowland, EC.,149, 150,169,173 Goyal, A.K., 50,79 Graham,L.A., 68,69,70,71,72,79 Graham,L.L., 202,203,237 Grandi, G., 119,130 Gravina. S.A., 200,242 Greenaway, S.D.. 145,154,169 Greene, A.C., 207,237 Greenwood, D.R., 5,8,11,42 Creer, C.W., 152,162,174 Griffith, G.S., 2,40 Griffiths, A.J., 183,187,226,236 Griot, R., 108,126 Croger, D.. 109, 128 Grossmann, A.D., 119,128 Grove, J.F., 100, 125 Grynkiewicz, G., 188,232 Gschwend, P.M., 159,169 Gucken, J.B., 207,240 Gueffroy,D.E., 186,237 Guerinot, M.L., 218.240 Guerlesquin, F., 59, 62,77 Guest, J.R., 92, 125,216,233 Gugel, K.H., 120,124 Guida, L., 187, 199,237 Guigliarelli, B., 76 Gull, K.. 10, 11,39 Gunsalus, LC., 65, 81 Gupla, A,, 226,237 Curbiel, K.J.,65, 66, 67, 69, 72, 79 Curies, R.P., 18, 31,37 Gutfinger, T., 68,80 Gutierrez, S.. 88,98,99, 125,128 Gutierrer-Corona. E, 214,234 Gutteridge, J.M.C., 49,80 Guzzo. A,, 224,237 Haak, B.. 50, 79 Haddock, J.D., 50,53,60,79 Haese, A,. 91, 100, 102, 103,126,128 Hagete, K., 120, 124 Hagenmaier, H., 120, 124 Hahlbrock, K., 90,127 Hahn, M., 9.41 Haigler, B.E., 49,53,56,58,59,60,78,79 Halachmi, D., 200,236 Hall. B.G., 146, 150. 174 Hall, D.M., 29, 43 Hall. D.O., 62, 63, 79
AUTHOR INDEX Hallas. L.E., 231,237 Halliwell, B., 49, SO Hamer, J.E., 5 , 8 , 28, 29.30, 38,42 Hammond, R.C., 165,169 Hansen. R.E., 63,82 Hanukoglu, I., 68,SO Han, M., 109, 110,126 Hara, O., 121,126 Harada, N., 219,235 Harashima, S., 67.81 Harayama, S., 49,50,55,60,61,66,73,80,81 Harcourt, A., 67,82 Hardman, D.J., 134, 136, 138, 142, 145, 149, 150, 154, 155, 158, 160, 164, 165, 166,167,168,169,172,173,174. 175 Harel-Bronstein, M., 193, 200,237 Harley, J.L., 3,38 Harmsen, M.C., 24,41 Harpel, M.R., 49,73,77,81 Harris, A., 60,81 Harris, C.I., 135, 171 Harrison, G.I., 229,237 Harrison, K., 139, 143,174 Harrison, EM., 216,233 Hartmann, A., 189,237 Hartmans. S., 160,165, 169,171 Hartnett, C., 50,55,61,66,73,81 Hanvood, V.J., 214,237 Hanvood-Sears, V., 223,237 Hasan,A.K.M.Q., 134, 144,169 Hase, T., 62.78 Hashim, R., 207,232 Hashimoto, T., 10, 38, 162, 171 Hassett, R., 221, 237 Hausinger, R.P., SO, 59,82 Hauska, G., 63,82 Hawkins, C . .216,233 48, 49, 80 Hayaishi, 0.. Hayashi, S., 149, 170 Hayashida, S., 53,60,61,73,78,83 Hazen, K.C., 32,39 Hazlett, R.D., 17, 39 Hearshen, D.O., 63, 65,68,72,78 Hederstedt, L., 218,242 Hegner, J., 24, 43 Heise,R., 200,237 Helinski, D.R., 90,125 Helm, D., 207,238 Hennecke, H., 145, 146,175 Henson, J.M., 207,240 Heppel, L.A., 165,169
AUTHOR INDEX
Hermann, H., 9,41 Hemero, M., 146,168 Hemnann,M., 91, 100, 102,103,126 Hess, W.M. 10,39,41 Higgins, D.G., 539, 148, 165,169 Hill, S.J., 198, 235 Hille, R., 65,68,72,78, 225,233 Hillemann, D., 121, 131 Hinata, M., 57,W Hinotozowa, K., 122,128 Hintz, W.E., 5, 18, 19,37 Hiratsuka, Y., 13, 19,42 Hirose, J.. 53.60, 61, 73,78.83 Hirsch, P., 135,169 Hobnt, J.A., 10, 11, 39 Hoch,H.C., 3,30,38,42 Hochkeppel, H.K., 102,124 Hodgson, J.E., 88,92,98,99,129 Hoffman, B.M., 65,66,67,69,72,77,79 Hoffman, EW., 136,169 Hoffmann, K., 106,126 Hofmann, A.. 108,126 Hofmann, H., 90,105, 107,127 Hohnadel, D., 189,239 Hol, W.G.J., 57,84 Holan, Z.R., 224,238 Holcombe, J.A., 193, 225.239 Hollander, J.A.. 209, 240 Holloway, P.J., 17, 29,38, 39 Holt. S.C., 11.39 Holt, T.G., 92, 121,128 Honneger, R., 3,10,33,34,39 Hooker, P.J., 183,239 Hope, S.J., 145, 150,169,174 Hopwood,D.A., 88,M, 91,94,126 Horgen, P.A., 5, 18, 19,31,37 Hori, K., 90,130 Horiuchi, T., 57,80 Home, R.W., 11,44 Hoskins, J.A., 98,130 Hou, C.T.. 165,173 Howard, R.J., 29,30,38,39 Huala, E., 146,169 Huang, P.M., 186,235 Huang, X.-H., 99,130 Hubbard, J.A.M., 189,216,222,238 Hubbell, J.A., 35,39 Hubbes, M., 5, 13, 18, 19,31.37,42 Hudlicky, T., 48,M Huet, J.-C., 9, 40 Hughes, M.N., 178, 180, 185, 186, 187, 188, 189,190, 191,194,195, 196, 198.
251 199,206,207,208,213,214,216, 222,224,224,233,234,235,237, 238,241 Hughes, S., 135,169 Huh, C., 17.39 Hurtubise, Y.. 50.80 Hutchinson, C.R., 93,94,%, 123,126 Huxley, M., 154, 166,169 Hyslop, E.K., 183.239 Igarashi, K., 200,240 Imai, S., 121,126,128 Imai, T., 165,170 Imamura, N., 122,128 Ingolia. T.D., 99,130 Ipsen, J.D., 6.8, 18, 31,41 Irie, S., 50, So Ishiama, T., 31.44 Ishikawa, T., 146, 176 Itoh, R.. 121,126 Iverson, W.P., 231,234 Iwasaki, H., 192,238 Izawa, S., 186,237 Jager, D., 161, 170 Jakeman, R.J.B., 209,239 Jansen, K.H., 197,233 Janssen, D.B., 134, 136, 138, 142,151, 152, 153, 154, 155, 160, 161, 162,170, 171,173,174,175 Jardim, W.F., 186,238 Jarrell, K.F., 186, 238 Jayatilake, G.S.,49.81 Jeng, R.,5, 13, 18, 19, 37.42 Jennings, D.H., 2.39 Jensen, H.L., 135,170 Jensen, S.E., 96,126 Jerina, D.M., 48,84 Jeronirnus-Stratingh, C.M., 162,173 Jessipow, S., 120.124 Ji, G., 212, 225, 226,234,235,238 Joannou, C.L., 53,56,57,58,59,60,64.67,79, 82,83,208,219,235,242 Johnson, A.C., 215,238 Johnson, D.B., 220,238 Johnson, K.A., 49.81 Jones, C.W., 218,238 Jones, D.H.A., 138, 143, 149,170 Jones, E.B.C., 29.39 Jones, R.P., 180,238 Jonssnn, B.H., 229,241 Joosten, M.H.A.J., 9, 43 Joshi, L., 11, 12, 13,37 Jung. G., 120,125
252 Kabuto, K., 48, 84 Kahan, B.D., 105,126 Kaida, N., 50.83 Kaldenhoff, R., 28,39 Kalk, C., 170 Kalk, K.H., 142, 161,168,175 KaIkkinen, N., 5, 8,40 Kallio, R.E., 50.79 Kamada, T.,219,235 Kamakura, T.,31,44 Kamp, R.M., 117, 118, 120,128,130 Kanda, M.,90,130 Kao, J.RY., 188, 189,232 Kaplan, J.H., 188,235,238 Kasai,N., 134, 153,154,155,158,170,174 Kasai, R.L., 10, 11,38, 153,155, 158, 159,174 Katsube, Y., 62.78 Katz, E., 111, 126 Katz, L.. 90,126 Kaufman, D.D., 135.171 Kawano, S., 219,235 Kawasaki, H., 138,142, 149,151,170,171 Kazemier, B., 142, 160,161, 162,170,175 Kazmierczak, P., 5,6,8,9, 13, 18, 19,32,38,45 Kearney, P.C., 135,171 Keister, D.B., 135, 136, 137, 169 Keith, L.H., 159,171 Keller, U., 91,93, 100, 103, 105, 109, 111, 112, 113, 114, 118, 122,129,130,131 Kelly, M., 135, 171 Kent, T.A., 63, 65, 68,72,78 Kershaw, M., 30.42 Kester, A., 9.42 Keuning, S., 160, 161,171 Khosla, C., 96. 123,130 Kikucbi, Y.,57.80 Kimbara, K., 57,80, 162,171 Kinghorn, J.R., 88,92,94,%, 124,127 Kingma, J., 161, 162,173,175 Kingsley, M.T., 215, 238 Kingsnorth, C.S., 23,24,37 215,238 Kinraide. T.B., Kirk, S.A., 31,37 Kirkness, E.F., 211,236 Kiyohara, H., 50.83 Kjelleberg, S., 4,41 Klages, U., 138, 141,171 Klapatch, T., 218,240 KIapcinska, B., 227,235,238 Klein, M.P.,67,69, 72.77 Kleinkauf, H., 86, 87,88,90,91,93,94,%, 97, 100,102, 103, 105,106,107, 109,
AUTHOR INDEX 112,124,126,127,128,129,130, 131 Klose, K.E., 146,172 Klube. K.D., 138,141,172 Kluge, B., 93, 117,118,127,128,129,130 Knaff, D.B., 67,69,72,77 Knobloch, K.-H., 90,127 Knogge, W., 9,41 Koana, T.,162,171 Kobal, V.M., 48,84 Kobayashi, H.. 200,240 Kobel, H., 107,108, 127,130 Koch, J.R., SO, 79 Kodama, T., 164. 165,170,176 Koga, H., 57.80 Kogut, M., 209,237 Kohata, K., 197,243 Kohler-Staub, D., 160, 165,171 Koiso, A., 149, 151,171 Kok, M.,49,80 Kok, R., 152,175 Kong, S.E., 188, 189,241 Konig, W.A., 120,124 Konings, W.N., 200,242 Kosaka, H., 219,235 Koshikawa, H., 138,172 Kosman, D.J., 179,221,222,237,239 Kot, E., 191,216,238 Kothe, E.M.,24,26,41,43 Kovacevic, S., 98,130 Kovaleva, V., 9.42 Kraas, E., 120,125 Kraepelin, G., 105.128 Kramar,0.. 231,243 Kratzschar, I., 88,89,92, 120,127 Krause, M.,88, 89,90,92, 120,127,129,130 Krauss, S., 138, 141, 171 Kreger, D.R.. 10,43 Krekel, D., 50,53.61.64, 80 Krengel, U., 109,126 Krianciunas. A., 67,69,72,73,77,81 Krook, J.H., 4, 10, 12, 13, 14, 20, 21, 44 Krouse, H.R., 229,237 Kruft, V., 88,91,129 Kues, U., 23,38,39 Kuila, D., 59,65,66,69,77,78,80 Kumada. Y., 121,126 Kunoh, H., 30.40 Kunst, L., 215,240 Kunugi, M.,197, 243 Kurahashi, K., 86.127 Kurane, R., 50.83
AUTHOR INDEX
Kurihara, T., 138,172 Kurkela, S.. 59.80 Kurotsu, T., 90,130 Kushner, D.J., 185, 241 Kustu, S.. 146, 171, 172 Kutz, S.M., 221,243 Kwart, L.D., 49,83 Kwon-Chug, K.J., 5, 12. 33,40 Kyte, J., 6, 7,39 La Roche, S.D., 165,171 Labbe,D.,50,83,152,162,174 Labbe, P., 221,239 Laberge, S., 9, 38 Labischinski, H., 207.238 LaBorde, A.L., 53.55.78 Laddaga, R.A., 180,224,226,227,238,239, 242 Laddison, K.J., 24, 43 LaHaie, E., 50,53,55,57,61,64,65,66,76,77 Laishley, E.J., 229,237 Laland, S., 86,124 Landa, E.R., 183,239 Landeen, L.K., 192,221,239,243 Landrum, G.A., 75,78 Langley, M.P., 196, 240 Laskin, A.J., 165,173 Latgb. J.-P., 5, 6, 8, 12, 33,42 Lau, P.C.K., 50, 83, 152, 162,174 Laurence, O.S., 230,236 Laurent. P., 5. 33.39 Lauter. E-R., 5, 11, 28, 29, 39,41 Lavanchy, D., 102,124 Lawen, A., 106,107, 108,127 Leadbetter, E.R., 11.39 Leadlay, I?, 94,128 Lee, B.T.O., 214,234 Lee, H., 227,229,230,239,242,243 Lee,H.-I., 8,40 Lee, K., 49,79 Lee, L., 48,49,80 Lee, R.S., 48.80 Leech, R.W., 185,234 Lehvaslaiho, H., 59,80 Leigh, J.A.. 139, 141, 171 82, Leisinger, T., 50, 53, 55,56, 57, 60,64,80, 136,137, 140, 142, 152, 160,164, 165,167,168,169,171,173,174 Leitner, E., 91, 106,130 Lernanski, C.L., 229,242,243 Lengeler, K.B., 24, 43 Lesuisse, E., 221, 239 Lewandowska, K.B., 189,216,222,238
2 53 Lewington, J., 150, 175 Lewis, M.R., 145 Ley, A., 186, 238 Ley, S.V.. 48,80 Librnan, J., 217,239 h e n , B.C., 135,171 Lin, C.-M., 179,221,222,239 Lindstedt, S., 73,77 Lingens, F., 48,49,50,53,55,56,57,59,60, 61,64,79,80,82, 138, 141, 171, 172, 173 Lion, R., 197,233 Lipmann, F., 86, 87, 92, 97,125,127,129 Lippard, S.J., 49,72,73,78,82 Lipscomb, J.D., 49, 73, 76,77,81 Liras, P., 98, 125 Litjens, M.J.J., 152, 176 Little, M., 136, 171 Liu, J.Q., 138, 172 Liu, T.-N., 50,53, 55,56,58,59,82, 83 Ljungdahl,P.O., a,76 Llobell, A., 5 3 9 Lloyd, D., 219,233 Locher, H.H., 50.53,56, 57,60,64,80 Loosli, H.R., 105,130 Lopez, J.L., 68,76 Lora, J.M., 5 3 9 Loros. J.J., 5, 11, 12, 28,37 Lovatt, D., 144, 145, 150, 174 Lovley, D.R., 183, 239 Low, B.J., 8,9, 38 Lomya, E., 90,127 Lubben, M., 218,242 Ludwig, M.L., 55.56, 77,78 Lugones, L.G., 3, $6, 12, 24, 25, 26,39, 44 Lundstrom, L., 16.38 Lurquin, P.F., 162, 173 Luyben, K.C.A.M., 160,169 Lynen, F., 88, 127 Lytton. S.D., 217,239 Macaskie, L.E., 209,227,239 MacCabe, A.P., 88,92,96,98,127 MacFarlane, J.K., 159,169 MacGregor, A.N., 135,172 Madduri, K., 98,130 Madgwick. J.C., 207,237 Madry, N., 91. 105, 128,131 Maeda, T., 138,171 Magee, L.A., 135,172 Magistrelli, C., 119, 130 Magnuson, R., 119,128 Maguire, R.J., 231, 243
254 Mahan, C.A., 193,225,239 Maier, R.J., 224, 236 Maier, W., 109, 128 Majidi, V., 193,225,239 Malkin, R., 67,69, 72,77 Mandel, M., 48, 80 Marahiel, M.A., 88, 89, 90,9492, 93, 117, 120, 123,125,127,129,130,131 Marchant, R., 10, 26,43 Markus, A., 50,53,55,56,57,60,61,64,80,82 Marliere, P., 92. 125 Martin, F., 5, 33,39,42 Martin, J.F., 88,98,99, 125 Maseles, EC., 50,79 Mason, J.R., 49,50,53.56,57,58,59,60,63, 64,67,68,71,72,73,80,81,82,83, 84,219,242 Mason, S.G., 17,39 Mather, M.W., 59,78 Matsubara, T., 192,238 Matsumoto, S., 62.78 Matsunaga, K., 57, SO Matsnshita, I., 138, 142, 149, 151,171 Matzanke, B.F., 209,216,239 McCabe, A., 94,96,124 McCarty,P.L., 159, 160,165,167,175 McCombie, W.R., 53,84 McCray, J.A., 188,239 McDonald, L., 230,239 McGinness, s., 220,238 McKee, J.A., 218,239 Mclean, R.J.C., 183,239 Means, J.C., 231,237 Mehta, N., 120, 130 Meienhofer, J., 111,128 Meister, A,, 97, 128 Mellor, E.J.C., 23, 24,37 Mendgen, K.,29.40 Menkhaus,M., 117,118,128 Menn, E-M., 162,172 Mergeay, M., 212,235 Merola, J.S., 48.80 Menick, M.J., 145,172 Messi, F., 160, 173 Messina, M.C., 221,243 Mester, B., 217,239 Metze, M., 146,168 Meyer, H.E., 68.81 Meyer, J.-M., 189,239 Mikesell, M.D., 160,172 Miller, C.G., 200, 242 Miller, J.R., 98, 130
AUTHOR INDEX Miller, M.J., 218, 239 Miller, W.G., 6, 8, 18, 31.41 Milne, G.W.A., 135, 136, 137,169 Mildowski, A.E., 183,239 Mms, W.B., 65.66.77 Minami, F.,149,170 Minamiura,N., 153, 155,158, 159,174 Minnick, A.A., 218.239 Mincda, Y., 160, 164, 165,170,176 Minor, W., 73.81 Minta, A,, 188,232 Miranda, R., 213,240 Misra, K.T., 228,239 Misra, T.K., 180,226,242 Mitra, R.S., 226,239 Miyoshi, K., 30,40, 142. 149,170 Mobley, H.L.T., 224,240 Modak, S.M., 229,240 Mondello, EJ., 50, 53, 60.61, 67,73,78, 81 Money, N.P., 30.39 Monod, M., 5,6,8, 12, 33,42 Montenegro, E., 88,99,125 Moodie, F.D.L., 53.81 Moore, R.T., 2,40 Moore, T., 146, 173 Morby, A.P., 226,237 Moritani, T., 50, 83 Morrice, N., 60,81 Morns, H.R., 91,129 Morsberger, E-M., 138, 141,172 Motosugi, K., 139, 144,172 Moukha, S.M., 2.4, 15.20.44 Mowll, J.L., 221, 236 Mozes, N., 17,41 Mueller, C.S., 196,240 Mueller, R.J., 5, 6, 8,9, 13, 18, 19, 32,38 Muisers, J.M., 9, 43 Mukohara, Y.,146.176 Mulder, G.H., 3,5, 14, 19,20.24,25,26,40,43 Mullen, M.D., 183,240 Muller, A., 63,82 Muller, G.I., 209, 216, 239 Muller, R., 50,56, 57, 64,78, 138, 141, 172, 173 Muller, V., 200,237 Muller-Rober, B., 28, 29, 41 Munck, E., 49,65,12,73,76,77,78,81 Muiioz. G.A., 13,40 Murakami, T., 121,126,128 Murata, M., 122, 128 Murdiyatmo, U., 138, 139, 142, 145, 149,167, 172 Murphy, G.L., 159,172
AUTHOR INDEX
Murphy, P., 5, 33 Myers, A., 119,128 Nadeau, P.,9.38 Nadim, L.M., 50,79 Nagaka, K., 121,126 Nagao, R., 99,130 Nagaoka, K., 121,126,128 Nagasawa, T.,153, 154, 155. 158,172 Nagata, Y.,57,M Nager, U., 99,128 Nakagawa, M., 67.81 Nakagawa, S., 165,170 Nakamura, H., 146,176 Nakamura, K., 213,240 Nakamura, T., 153, 154, 155, 172 Nakano, M.M., 91, 92, 117, 119, 120, 125, 128, 131 Nakari-Seala, T., 5,8,40 Nakatsu, C.H., 50, 53, 55.81 Nardi-Die, V., 138,172 Narro, M., 50,53,55,56,82 Nasse, B., 5, 33,42 Naumann. D., 207,238 Nehls, U., 5, 33,39 Neidle, EL., 49,50,55,61,66,73,81 Neilands, J.B., 181, 217, 240,241 Nelson, W.H., 207,240 Nengu, J.P., 160, 168 Nespoulous, C., 9,40 Netting, A.G., 29.40 Neumann, A.W.. 16,41 Newby, P.H., 187,240 Newman, K.A., 159,169 Nicholls, P.D., 207,240 Nicholson, R.. 29. 30, 31, 40 Nies, A., 226.240 Nies, D.H., 226, 240 Nihira, T.,91, 105,131 Nilssorn, U., 16, 38 Nishida, H., 99, 130 Nishino, H., 138,172 Nitschke, W., 63,82 Niven, D.E., 207,240 N0bar.A.M.. 186,198, 199,206,233,241 Nomura, Y., 67,81 Noordmans, J., 16.40 Nordin, J.H., 18,37 Nordlund, P., 73,82 Norlund, I., 162, 172 North, A.K., 146,171,172 Novick, R.P., 228, 240 Novo-Gradac, K.J.,186,233
255 Nozawa, Y,10, 11,38 Nussbaumer, B., 121,131 Nuyts, G., 212, 235 Nyandoroh, H.. 164 O’Callaghan, N.M., 97,124 Oelrichs, P.B., 105,128 Ogawa, N., 67,81 Ogawa, S., 209,241 Ogino, T., 209,240 O’Halloran, T.V.,73,75,77,229,240,243 Ohlendorf, D.H., 73,81 Ohnishi. T.,68.72.77, 82 Ohshima, A., 31,44 Ohyama, T., 200,240 Oishi, M.,162, 171 Okamura, O., 138.172 Okuda, S., 99,130 Okumura, Y,111,128 Olami, Y., 193,200,237 Olson, E X , 67,77 Olson, G.J., 231, 234 Omon, T., 159,160,164,172.173,176 Omura, S., 99, 122,128,130 Omston, L.N., 50, 55,61,66,73,81 Orville, A.M., 49,73,77, 81 Osbom, R.W., 9,42 Oshima, Y., 67,81 Osslund, T.D., 67.82 0 t h H., 108.126 Otto, A., 93, 117,129 Otto, M.K., 138, 141, 172 Otwinowslu, Z., 73.81 Ozaki, H., 138,172 Pacheo, S.V., 213,240 Padan, E., 193,200,237 Painter, E.P., 229,240 Palacz, Z., 117, 130 Palissa, H., 88,92, 96.97, 124, 127, 129 Palva, E.T., 59,M Panaccione, D.G., 110,129 Pang, C.-P., 49.81 Panico, M., 91, 129 Paris, S., 5,6, 8, 12, 33, 42 Parker, W., 227,243 Parsons, J.R., 134,168 Parta, M., 5,8, 12,33,40 Patel, R.N., 165,173 Patil, D., 53,59,75,79 Patil, S.S., 90,125 Paul, E., 91, 105,131 Pavela-Vrancic, M., 90,93,128 Pearce, J.M., 183,239
256 Pearson, H.W., 186.238 Peters, H., 91, 105, 128,131 Pelzer, S., 121, 131 Pember, S.O., 49.81 Pendland, J.C., 29,37 Penfold, W.J.. 134.173 Penner-Hahn, J.E., 66,67,83,229,242 Penlenga. M., 152, 162,173, 174 Penttila, M., 5, 8,40 Perego, M., 120,125 Perkins, E.J., 162, 173 Pernollet, J.-C., 9, 40 Perry, J.J., 159, 172 Peter, M., 34.39 Peters, J.A., 135,171 Peters, R.A., 134,173 Pettersson, A., 215,240 Pfefferkom, B., 68.81 Pfeifer, E., 88, 90, 92, 93,94,%, 124,127, 128 Pfister, R.M., 11,41 Pfleger, K., 49,84 Phillips, E.J.P., 183, 239 Phillips, S.E., 225, 226,240 Phung, L.T., 180,242 Pieper, R., 102, 103,126,128 Pinner, E., 193, 200, 237 Pinto-DaSilva, P., 5, 12, 33,40 Pintor-Toro, J.A., 5,39 Pioda, L.A.R., 99,131 Pirt, S.J., 179, 188, 189,222, 224, 240 Plathler, P.A., 99, 128 Plessner, O., 218, 240 PKard, J.-A., 110, 129 Ponelle, M., 105, 130 Poole. R.K., 178, 180, 185, 186, 187, 188, 189, 190, 191, 195, 196, 198,199,206, 207, 209,213,214,216,218,219, 220,222,224,226,233,234,235, 237,238,241 Poolnian, B., 200,242 Popkin, T.J., 201,202,203, 234 Pople, M.,100, 125 Porter, R., 29.38 Porterfield, V.A., 165,169 Poulos, T.L., 49, 81 Powlowski, J., 50,80 Pries, F., 134, 136, 142, 151, 160, 161, 162,170, 173,175 Privalle, C.T., 188, 189, 241 F’usztai-Carey, M., 5, 6, 9, 18, 44 Que, I,., 73. 77 Queener, S.W., 99,130
AUTHOR INDEX Que, L.Jr., 49,77,81 Raag, R., 49,81 Raibaud, A., 92, 121,128 Raikhel, N.V., 8, 9,40, 42 Ralston, D.M., 229.240 Ramamoorthy, S., 185,241 Ramelow, G.J., 196,240 Ramirez, J.L., 225,226, 234 Raper, C.A., 22,24,40,43 Raper, J.R., 2, 19,22,40 Raudaskoski, M., 14,24,26,40,44 Rayner, A.D.M., 2,40 Rayner, M.H., 209,241 Read, D.J., 3,40 Reanney, D.C., 149,173 Redgewell, R.J., 11, 17,37 Rees, S.B., 9,42 Regensburg, B.A., 10,43 Regev, R., 209,237 Rekik, M., 50.55,60,61,46,73,80,81 Remsen, C.C., 10,39,41 Resnick, S.M., 49.79 Reutlinger, M., 195, 237 Reuvekamp, P.T.W., 153, 155,175 Riach, M.B.R.. 88,92,96,98,127 Ricci, M., 28, 29,41 Richards, W.C., 5,6, 8,9, 18, 31,37,40,42,44 Richardson. J.S., 8.9.38 Riedel, A., 63,82 Riederer, B., 111,122 Rieske, J.S., 63,82 Rikkerink, E.H.A., 8, 9.42 Rittman, B E , 159, 167 Roach, D.H., 30.39 Roberts, D.W., 5.8, 11, 12, 13, 30,37,41 Roberts, G.A., 94,128 Robinson, N.J., 226,237 Robson, C.D., 2,43 Rodriguez, E, 120, 125 Rodriguez-Romero,A., 8,37 Roe, A.L., 73,77 Rohe, M., 9,41 Rolland, C.. 5, 6, 8, 12, 33,42 Rollins, M.J., 96,126 Romanov, V., SO, 59.82 Ronneau, C., 197,233 Roomans, G.M., 215,240 Rosahl, S., 9 , 4 1 Rosche, B., 49,53,60,82 Rosen, B.P., 180, 183, 191,236,241,242 Rosenberg, G., 50,53,56,57, 59,&1,82 Rosenberg, H., 217,241
AUTHOR INDEX
Rosenberg, M., 4.41 Rosenzweig, A.C., 73,82 Roskoski, Jr., R., 87,129 Rosner, J.L., 224,241 Rotenberg, Y., 16,41 Roth, C.. 228, 240 Roth, L.E., 215,241 Rotilio, G., 223,235 Rouch, D.A., 214,234 Roucoux, P., 197,233 Rouden, J., 48,W Rouxhet, P.G., 10, 17,38,41 Roy, C.. 5, 6, 9, 18.44 Rozeboom, H.J., 142, 161,173,175 Ruardy, T.G., 16,29,44 Ruf, H.H., 53,55,56,57,60,82 Ruiters, M.H.J., 14, 20,24,25, 26,41,44 Rulong, S., 5, 12,33,40 Rundgren, M., 73.77 Runswick, M.J., 90, 130 Russell, N.J., 209, 237 Russo, P.S., 6, 18, 31,41 Russo, V.E.A., 5 , s . 11, 28, 29,39,41 Rutherford, A.W., 63,82 Saboowalla, F., 53, 59,75,79 Sadler, P.J., 185, 187,208,209, 234 Sahhan, L., 229,241 Said, 2, 187, 191,199,237 Saigo, T., 192,238 Saito, Y, 90,130 Salerno, J.C., 72,82 Sallis, P.J., 138, 142, 154, 158, 160, 164, 165, 166,167,173,175 Salnikow, J., 117,127 San Martin, K.,13,40 Sanglier, J.-J., 108, 127 Saraste, M., 90. 130 Sargent, J.S., 68,72,79 Sariaslani, F.S., 165,169 Sassen, M.M.A., 10,39,41 Satoh, A., 121,126, 128 Satoh, E., 121, 126 Satoh, S., 50, 83 Sauher, K., 50,53, 56,57,59,64,82 Sauer, K., 67,69,72,77 Saulnier, M., 186,238 Saunders, R., 67,82 92,125 Saurin, W., Savvaidis, I., 195, 196, 198, 199,206, 222,241 Sawada. T., 50,83 Schaap, P.J., 5,38 Schafer, H.J., 90,128
257 Schallehn, G., 207,238 Scheer, J.H.J., 24,26,41 Schellekens, G.A., 9,41 Scheper, A., 152,154,160, 161,162,170 Scherr, D.J., 154 Schimmel, P., 93, 129 Schnder, R., 105,129 SchISfli, H.R., 50, 55,56, 64.82 Schlegel, H.G.. 225,226,241,242 Schlumbohm, W., 88,91, 112,114.125,126, 129 Schmidhauser, T.J.,28,29,41 Schmidt, B.,9,41 Schmidt, F.R.J., 139, 149, 167 Schmidt, T., 226,241 Schmuckle, A., 160,169 Schneider, A., 121,129 Schneider, B., 138, 141,173 Schneider-Scherzer, E., 106,126,130 Schocken, M.J., 49, 79 Schofield, C.J., 97. 98, 117, 124, 129 Scholtmeyer, K., 5, 6, 12.26.39 Scholtz, R., 160, 164,173 Schoonover. J.R.,66.69.80 Schorgendorfer, K., 91,106,130 Schreyer, M., 107,125 Schrljder, W., 90, 103,128 Schubert. M.. 91, 100, 102, 103,126 Schuhmann, A.. 197,233 Schuldiner, S., 193,200,237 Schulze, R., 221,243 Schuren, F.H.J., 3,4, 5, 12, 13, 14, 15, 16, 19, 20, 22,24,30. 32, 35, 41, 43, 44 Schuurs, T.A., 3,5, 20.24, 25,26,43,44 Schwecke, T., 94,96,97,124,129 Schweizer, D., 53, 55.56, 57,60,82 Schwyn, B.. 217,241 Scott, R.L, 219, 220, 241 Scott-Craig, J.S., 110, 129 Sebald, W., 63, 67.78 see^, M.. 53, 55, 56, 57,60, 82 Seiyama, A., 219,235 Sekiya, M., 10, 11,38 Selitrennikoff, C.P., 12, 28,41 SeljQ, E 161, 175 Senior, E., 135, 144, 145,150,173,174 Serdar, C.M., 50,53,58,59,83 Service, R.F., 34, 41 Sewall, T.C. 4,5,8, 11, 12, 13,27, 28.42 Shanxr, A., 217,239 Sharp, P.M., 5.39 Sheets, T.J., 135, 171
258
AUTHOR INDEX
Sheldrake, G.N., 48.82 Shen, G.-J., 165, 170 Sherburne, R., 230,242 Shergill, J.K., 67, 82 Sherman, D.H., 88,90,91,94,126 Shiaris, M.P., 50. 84 Shiau, C.Y. 97,98, 117,129 Shiga, T., 219,235 Shin, M.K., 229,240 Shingler, V., 146, 162,168, 172, 173 Shotyk, W., 183,235 Shrift, A,, 229,234 Shulman, R.G., 209,240,241 Shuttleworth, K.L., 187, 194,241.242 Sietsma, J.H., 2, 3,4, 9, 15, 20.24, 25, 41,43,
44 Sigg, L., 195, 237 Sillerud, L.O., 229, 243 Silver, S., 178, 179, 180,183,210,212, 213, 224,225,226,227,234,235,238, 239,240,241,242,243 Simon, M.J., 67.82 Simon, W., 99,131 Simon-Lavoine, N., 99,129 Singh, R.M.M., 186,237 Sivaraja,M., 65,66,67,69, 72,79 Skamoulis, A.J., 209,239 Skatrud, P.L., 98,99,130 Skinner, A.J., 139, 141, 166,167, 171 Slater, J.A., 18,41 Slater, J.H., 134, 135, 136, 142, 144, 145, 146, 147,149, 150, 151, 154, 155,160, 166,167,169,173,174,175,176 Slavin, W., 193,242 Slawson, R.M., 227,229,230,242 Sleytr, U.B.. 10, 38 Smalley, E., 18.31.37 Smith, D.J., 88, 92, 98, 99, 129 Smith, J.M., 139, 143,174 Smith, J.M.A., 216, 233 Smth, S.E., 3,38 Smucker, R.A., 11,41 Snavely. M.D., 200,242 Sobey, W.J., 98, 117. 129 Soda, K., 134, 138. 139, 144,169,172 Sokolovsky, V.Y, 28,29,41 SB11,D., 93,129 Somlyo, A.P., 188,238 Sonnenberg, A.S.M., 5.38 Southam, G., 183,202, 203,236 Spain, J.C., 49, 50, 55, 58. 76, 82 Specht, C.A., 24,43
Speelmans, G., 200,242 Spence, J.T., 217, 235 Sperry, J.F., 207, 240 Springer, J., 3, 13, 14, 19, 20, 24,26,38,41, 43,44 StLeger, R.J., 5 . 8 , 11, 12, 13,30,37,41 Stacey. G.. 215.241 Stachelhaus, T., 88, 92,93, 123, 129 Stadler, P.A., 108, 126, 129 Stahl, li., 9, 44 Stanley, D.C., 67,77 Staples, R.C., 5, 8, 30,41 Starcdub, M.E., 230,242 Staub, D., 165,167 Staunton, J., 94, 128 Steczkn, J., 73.81 Stedman, K.M., 146,172 Stefanac, Z., 99,131 Stein, T., 88,91, 93, 117, 129 Sterk,P.,9,41 Sternfeld, F., 48,80 Stevenson, K.J., 5, 6, 9, 18, 41.44 Stewart, A., 221,236 Sticklen, M.B., 9, 18,33,37,42 Stindl, A,, 113, 114, 116, 117, 118,129,130 Stirling, D.I., 165, 168, 174 Stijffler-Meilicke,M., 109, 110, 126 Stookey, L.L., 190,221,242 Stoppel, R.-D., 225,242 Stotzky, G., 182, 183, 185,186, 187,233,235 Stratton, G.W., 226, 243 Straus, N.A., 50,81 Stringer, M.A., 4, 5,6, 8, 11, 12, 13, 27,28,42 Strotmann, U.J., 152, 174 Stubbs, B.M., 154,167 Stucki, G.R., 152, 165, 174 Stumm, W., 195,237 Stuttard, C., 98, 130 Subramani, S., 90,125 Subramanian, V., 50,53,55,56,58,59,64,82,
83 Suemori, A., 50, 83 Suen, W.-C., 53, 60,64.67, 82,83 SUggS, S., 67.82 Summers, A.O., 179,242 sun,Y.,5,45 Surerus, K.K., 49.77 Surewicz. W.K., 5,6, 9, 18,44 Suler, F.,164, 165,171,173 Sutherland, I.W., 197,236 Suyama, A., 53,60,61,73,78 Suzuki,T., 134,153,154,155,158,159,170,174
AUTHOR INDEX
Svensson, B., 218,242 Svircev, A.M., 13, 19.42 Sylvestre, M., 50,W Tagu, D., 5,53,39,42 Taira,K., 53.83 Takada, H., 134,144,169 Takagi, M., SO, 57, 80, 162, 171 Takai, S., 5,6,8, 9, 13, 18, 19,37,41,42,44 Takamatsu. T., 197,243 Takano, E., 121, 126 Takao, M., 149, 151,171 Takigawa, H., 165,170 Takuziwa, N., SO, 83 Talbot, N.J.. 5, 8, 28, 29, 30,42 Tan, H.-M., 53, 57,58, 59, 60,61, 67,83, 219, 242 Tanabe, O., 135,175 Tanaka, H., 99,130 Tang, H.-Y., 53, 57, 59,60, 67,83 Tardif, G., 152, 162,174 Tam, G.E., 65,68,72,78 Taylor, D.E., 230,242, 243 Taylor, F., 165,169 225, 226,240 Taylor, M.L., Taylor, S., 48,W Taylor, S.C.. 134, 143,174 Teen, T.H., 5 9 , 8 Tegli, S., 31,37 Telliard, W.A., 159,171 Temple, R., 18, 31,37 Templeton, M.D., 5, 8.9, 11, 33, 42 Te-Ning, L., 50,64,82 Terashima, Y , 138, 172 Terhune, B.T., 30,42 Terpstra, P., 57,78,84, 142, 160, 161,170 Terns, F.R.G., 9, 42 Thau, N., 5 , 6 , 8 , 12,33,42 Thomas, A.W., 139, 145, 146, 147, 149, 150, 174,175 Thomas, C.M., 138, 143, 149,170 Thomas, D., 36,42 Thompson, C.J., 92, 121,126,128 Thompson, R.L., 196,240 Thompson, S.A.J., 23,24,37 Thomson, J.C., 139, 142,167 Thony, B., 145,146,175 Thornley, J.H.M., 62, 63, 79 Timberlake, W.E., 4,5, 6, 8, 11, 12, 13, 27, 28, 36,38,42 Timmis, K.N., 60,80, 146, 168 Ting, H.-H., 49, 81 Tisa. L.S., 180, 189,242
259 Tizard, R., 92, 121, 128 Tobin, M.B., 98,130 Tognoni, A., 119,130 Tohika, K., 90,130 Tomei, EA., 229,242, 243 Tomizuka, N., 50,83 Tomoda, H., 99,130 Tone, H., 142,149,170 Tonomura, K., 135, 138, 142, 149, 151,170, 171,175 Topping, A.W., 139, 144, 145,146, 147, 149, 150, 151, 174 Torigce, S., 50,83 Tomabene, T.G.. 228,243 Torok, D.S., 49.79 Torrekens, S.,9,42 Toyama, T.. 138, 171 Traber, R., 105, 107, 108, 127,130 Tramper, J., 160, 169 Trautwein, A.-X., 75,77,209,216,239 Trenlham, D.R., 188,239 Trevors, J.T., 226,227,228,229,230,234,236, 239,242. 243 Trinci, A.P.J., 2. 43 Tronchin, G., 5, 6, 8, 12, 33, 42 Troughton, J.H., 29,43 True, A.E., 65,66,67,69,72,79 Trumpower, B.L., 63,68,69,70,71,72,76,79, 83 Tryhorn, S.E., 165,169 Tsang, H.-Y., 66, 67, 83, 229, 243 Tsang, J.S.H., 138,142, 149,172,175 Tscherter, H., 107,125 Tshisuaka, B., 49, 53, 60, 82 Tsibns, J.C.M., 65,81 Tsien, R.Y., 188, 189, 232 Tsoi, C.J., 96, 123, 130 Tsuda, K., 138, 142, 149, 151,171 Tsujimura, K., 134. 154, 155, 158,170 Tsukihara, T., 62.78 Turgay, K., 88,89,90,130 Turner. G., 88.92, 98,99,129 nuner, J.S., 226,237 Twilfer, H., 75,83 Ugurbil, K., 209,241 Ulbrich, N., 9,44 Ullrich, C., 88,91, 117, 118,128,129,130 Unkles, S.E., 88.92,96,127 Unoum, K., 134, 154, 155, 158,170 Unz, R.F., 187, 194, 241,242 Vaillancourt, L.J., 24, 43 Valent, B., 29, 30, 38
260 van Alfen, N.K., 5,6, 8,9, 13, 18, 19, 3 2 , s . 45 van den Ackerveken, G.F.J.M., 9’43 van den Mjngaard, A.J., 153, 154, 155, 160, 162,175 van der Karnp, K.W.H.J., 160, 162,175 van der Lende, T.R., 19,24,41 van der Meer, J.R., 162,175 van der Mei, H.C., 16, 17, 29,30,43, 44 van der Plceg, J., 134, 136, 138, 142, 151, 160, 161,162,170,175 van der Valk, P., 26,43 van der Vegt, W., 16, 17,43 van der Vmn, L.H.M., 88,98,125 van der Waarde, J.J., 152, 175 Van Dyke, M.I., 227, 229,230,239, 242,243 van Griensven, L.J.L.D., 5.38 van Hall, G., 138, 175 Van Kammen, A,, 9,41 Van Kan, J.A.L., 9.43 Van Leuven, F., 9,42 van Liempt, H., 88,!30,92,94, %,97,124,127, 128,129,130 Van Loon, J.C., 198,243 van Pelt, A.W.J., 17,37 van Pouderoyen, G., 162,173 van Sinderen, D., 120,125 van Solingen, P., 88, 98, 125 van Wetter, M.-A., 3, 5, 19, 20, 23, 24,25,26, 37,43,44 Vanderleyden, J., 9,42 Vano, K., 162,171 VanRiet, J., 227, 228,233 Vartivarian, S., 217,233 Vasquez, C., 230,235 Vater, J., 88. 91, 111, 117, 118, 120, 128, 129, 130 Venema, G., 120,125 Verbakel, H.M., 9,43 Verschueren, K.H.G., 160, 161,175 Vbzina, L.-P., 9, 38 Viitanen, H., 26,40 Villafranca, J.J., 49,81 Villalon, D.K., 5,6, 8, 9, 13, 18, 19, 32,38,45 Mning, L.C., 98, 130 Visser, I., 5, 38 Vogel, T.M.. 135,175 Volcani, B.E., 227,233 Volesky, B., 224,238 Vollenbroich, D., 117, 118, 120, 128,130 Volpe, D., 13,40 von Dohren, H., 86,88,90,93,94,%, 97,124, 126,127,128,129,130
AUTHOR INDEX von Ostrowsla, T.,103,126 von Wartburg, A., 105,130 von Wettstein-Knowles, P., 29, 40 Vnend, G., 57,78 Wackett, L.P., 49,50,53,58,59,79,83 Wakley, G.E.. 30.42 Walderhaug, M., 178,210,242 Walker, J.E., 90.130 Walker, J.R.L., 135,136 Walter, E.G., 230,242 Walton, A.P., 198, 235 Walton, J.D., 99, 110, 129, 130 Wang, Y., 50,83 Ward, J.M., 88, 92, 98, 99, 129 Ward, R.W., 198,235 Waring, A.J., 219,241 Watabe. K., 146,176 Watanabe, I., 153.1.54, 155, 158,172 Watanabe, M.,197,219,235,243 Watterson, J.R., 183,243 Weber, G., 91, 106.130 Weber, P.C., 73, 81 Weckermann, R., 88,130 Wehrli, E., 11, 44 Weigel, B.J., 99, 130 Weightman, A.J., 136, 144, 145, 146, 147, 149, 150, 155,167,174,175, 176 Weijers, C.A.G.M., 152, 176 Weiss, D.S., 146, 171 Weiss, M.A., 50,55, 64,82 Welin, S., 16.38 Welz, B., 193,243 Wende, P., 49,84 Wendland, J., 24.43 Wenger, R.M., 106, 107,125,127 Wessels, J.G.H., 2, 3.4, 5,6,7, 9, 10. 11, 12, 13, 14, 15, 19, 20, 21, 22, 23,24,25, 26, 28, 29, 30, 32, 33, 34, 3.5,37,38, 39,41,42,43,44 West, J.M., 183,239 Westlake. D.W.S., 96,126 Whatley, F.R., 62,63,79 White, C., 180,221,236 White, D.C., 207,240 White, O., 211,236 White, R.L., 49,81 Whittaker, J.W., 75, 78 Whitton, B.A., 226,237 Witty, P., 146, 168 Widom, J., 49,81 Wiebe, M.G., 2,43 Wierenga, R.K., 57,84
AUTHOR INDEX
Wildermuth, H., 11,44 Wilkinson, S.C., 229,236 Williams, J.M., 216,225,233 Williams, J.W., 225, 243 Williams, J., 225, 233 Williams, P.A., 136,171 Williams, S.J., 11, 44 Willis, A.C., 97,124 Wilson, B.H., 159,176 Wilson, J.T., 159, 176 Winget, G.D., 186,237 Winter, W., 186,237 Wipf, H.K., 99,131 Witholt, B., 57.78, 142, 152, 153, 154, 155, 160,161,162,170,171 Withann-Liebold,B., 88,91,93, 117,129 Wnendt, S., 9.44 Wohlleben, W., 121,131 Wolf, D.C., 183,240 Wolfe, S., 94,%, 124 Wong, A., 96,126 Wong, P.T.S., 198,231, 235, 243 Wood, K.V., 90,125 Wood, M., 215,238 Woodland, M.I?, 53,81 Woodruff, W.H., 66.69.80 Wootton,J., 146, 168 Wosten, H.A.B., 2, 3,4, 5, 6, 9, 10, 11, 12, 13. 14, 15, 16,24, 25,26,28, 29.30.32. 34,35,38,43,44 Wright, C.S., 8, 9,38 Wright, J.G., 229,240,243 Wu-Yuan, C.D., 10.38 Wyndham, R.C., 50,53,55,81 Xiao, J.-2.. 31,44 Yaguchi, M., 5,6,9, 18,44 Yahara, H., 149,151,170,171 Yahya, M.T.,192, 221,239,243 Yamada, H., 153,154, 155. 158,172 Yamaguchi, E., 57,80
261 Yamaguchi, I., 31,44 Yamaguchi, M., 50,53,56,57,60,64,84 Yamamoto, R., 153, 155, 158,159,174 Yamane, K., 92, 120,125 Yamaoka, T, 135,175 Yanase, N., 149, 151,170 Yang, Y., 50,84 Yano, K., 50,57,80 Yanovsky, C., 5, 11,28,39 Yeh, W.-K., 50,53,55,56,58,59,64,82,83 Yli-Mattila, T., 14,26,44 Yokota, T., 160, 164,176 Yokoyama, T., 10,11,38 Yorifuji, T.. 50.80 Yoshida, T., 59,63,65,68,72,78 Yoshihara, H., 219, 235 Yost, K.J., 48.80 Young, J.C., 196,240 Young, K.D., 67,77 Yu, C.A., 67,69,72,77 Yu, F., 153, 154, 155, 158,172 Yu, L., 67, 69, 72,77 Zahner, H., 120,124,125 Za~ki. T.. 53,60,61,73,78 Zalacain, M., 91, 92,93,128 Zamanian, M., 53,60,64,84 Zaugg, W.S., 63.82 Zehnder, A.J.B., 162,175 Zentmeyer, G.A., 18,44 Zhang, J., 94, %, 124 Zhang, L., 5.6.8.9, 13. 18,19,32,38,45 Ziffer, H., 48.84 Zocco, T.G., 229, 242,243 Zocher, R., 91, 100, 101,102.103,105,106, 107,109, 112,114,121,124,126, 127,128,131 Zuber, P., 91,92,117. 119,120,125,12%,130, 131 Zylstra. G.J., 49, 50, 53,57, 60,66,77,79, 82, 84, 162,172
This Page Intentionally Left Blank
Subject Index Figure and table references are shown in italic
abaA A. nidulam gene, in conid~ogenesis,27 ABHl hydrophobin, 4 , 5 , 6 in fruit bodies, 26-7 actinomycin synthesis, 91, 111-14 synthetase I, 112-13 synthetase II, 113-14 in amino acid epimerization, 116-17 reaction priming on, 114-16 synthetase 111, 113-14 in peptide bond formation, 117 synthetases in cell-free synthesis, 114, 115 ACVS see S-(L-aIpha-aminoadipyl)cysteinyl-D-valine synthetase (ACVS) acyl peptide lactones, 111 synthesis, actinomycin as model, 111-12 see also actinomycin synthesis acyl transfer, by peptide synthetases, 92 S-adenosyl-L-homocysteine, and enniatin synthetase, 100-1 S-adenosyl-L-methionine, and enniatin synthetase, 100 AES see atomic emission spectroscopy Agancus bisporus. hydrophobins in fruit bodies, 26-7 see also ABHl hydrophobin agglutinins, and hydrophobins, 8-9 Agrobacteriurn turnefaciens, haloalcohol dehalogenases, 154 Alcaligenes, haloalcohol dehalogenases, 156-7 Alcaligenes eutmphus. cadnuum resistance, 226 alcohol dehalogenation, 151-9 dehalogenases, 152-9 classes, 154-5, 158 pathway, 152-3 profiles, 153-4 properties, 156-7 enzymatic oxidative dechlorination, 159
halogenated alkanoic acid formation, 152 algae, lichen symbiosis, hydrophobins in, 3 3 4 alkane dehalogenation, 1 5 9 4 5 class 3B dehalogenases, 164 catabolism pathway, 161 evolutionary relationships, 162, 164 from Xanthobacfer aurotrophicus, 162,163 class 3R dehalogenases, 1W cofactor-dependent dehalogenases, 165 discovery, 159 diversity of mechanisms, 1M) methanogenic bacteria in, 1 5 9 6 0 oxygenase-type dehalogenases, 164-5 alkanoic acid dehalogenation, 135-51 dehalogenases characteristics, 1 3 5 6 class ID, 139. 140, 1 4 3 4 class lL, 136-7, 138,140, 141-3 class 21, 139, 144-5 class 2R, 139, 1 4 5 4 classification, 136,138-9 genetic organization, 146,147, 149-51 synthesis regulation, 146-9 hydrolytic mechanism, 135 aluminium, toxicity, 182, 214-16 &(L-a-aminoadipyl)-cysteinyI-D-valine(ACV), structure, 96 6-(I.-a-aminoadipyl)-cysteiIiyl-D-valine synthetase (ACVS), 96-9 activation sites, 97 epimerization in synthetic action, 97-8 genes, 98-9 Anabaena qlindrico, aluminium accumulation, 215 Ancylobacter aqwficus, dehalogenase, 162 Ankistmdesmwfalcatur, tin accumulation, 23 I anodic stripping voltammetry (ASV), 194-5 antimony, 182 arsenic, 182
264 arsenic (continued) bacterial resistance, 225-6 Arthrobacter dehalogenases haloalcohol, 154, 155, 156-7 haloalkane, 164 ascomycetes, hyphal structure, 2 Aspergillus hydrophobins from, 13 rodlet layer, 11, 12, 13 Aspergillus nidulans ACV synthase from, 96,97 conidiogenesis, molecular genetics, 27-8 hydrophobin gene, 4 atomic absorption spectroscopy, 193 with chromatography, 198 atomic emission spectroscopy, 193 atomic fluorescence spectrometry, 194 axisymmetric drop shape analysis, for hydrophobins, 16 Azotobacter, lead resistance, 228 Bacillus germanium accumulation, 227 mercury resistance, 229 blotting analysis, 213 bacteriofemtin, 216 basi&omycetes fruit bodies, 3 hyphal structure, 2
Beauveria, 105 beauvericin, 105 benzene dioxygenase, 61 ferredoxin, 5 8 , 5 9 4 non-haem iron, 75 reductase, 57 spectroscopic analysis, 65 benzoate dioxygenases, 55,56,57 iron site, 74-5 beta-lactam antibiotics, hydrolysis by metals, 222 bialaphos, 120-2 biocompatibility. and hydrophobins, 35 biosensors, 21 1-12 hydrophobins in, 35 biphenyl dioxygenases. substrate specificity, 61 BLAST, 21 1 blotting hybridization techniques, 212-13 Bradyrhizobium japonicum, nickel metabolism, 224 brlA A. nidulans gene, in conidiogenesis, 27 cadrmum
INDEX bacterial resistance, 2 2 6 7 fungal toxicity, and pH, 187 precipitation, 209 uptake, zinc competition, 224 caesium, radioactive, from Chernobyl, 183 calcium analysis, 191 cellular concentration, and photoreactive ligands, 187-8 as essential metal, 180 Cephalosporium acremonium, ACVS from, 97 cerato-ulmin, 4,6, 13 disulphide bridges, 9 function, 19 hydropathy pattern, 6.7 phytotoxic mechanism, 3 1-2 sequence determination, 18 surface activity, 18 Chelex- 100 for copper metabolism study, 222 for metal removal from m d u m , 189 Chernobyl fall-out, 183 R-3-chloro- 1,2-propandiol, enzymatic dechlonnation, 159 chlorocatechol dioxygenase, 73 iron site, 75 I-chlorohexane dehalogenase, 164 4-chlorophenyl acetate benzoate dioxygenase, 55,56,57 chromatography, of metals, 198 ion chromatography, 197-8 chrome azurol (CAS), in siderophore assay, 217-18 Chviceps purpurea, 108 Cochliobolus carboneum, HC-toxin synthetase from, 110 CoHl hydrophobin, 4.5 in aerial hypha formation, 21-2 conidia, hydrophobins in formation, 27-9 copper binding proteins, 222-3 cellular uptake, 221-2 microbial corrosion. 207-8 in oxygenase catalysis, 49 resistant bacteria, 214 speciation, and cell growth, 186 transport, 181 copper-zinc superoxide dismutase, 223 Corynebacterium dehalogenases haloalcohol, 154, 155, 156-7, 158 haloalkane, 164 cryofrxation, 202-3
265
INDEX cryparin, 4,6, 13 function, 19 hydropathy pattern, 6 , 7 lectin-like activity, 9 cyclosporins, 105 structure, cyclosporin A, 106 synthesis, 91, 105, 106-7 synthetase, 105-6 molecular structure, 107 substrate specificity, 107 see also SDZ 214-103 Cylindrotrichumoligospennum. 107 Dalapon, 135 Pseudomonasputida growing on, 144 defensins, cyskine residues, 9 dehalogenation, microbial of alcohols see alcohol &halogenation of alkanes see alkane dehalogenation of alkanoic acid see alkanoic acid dehalogenation depsipeptide formation, enniatin synthetase in, 103,104 Desuljovibria desulfuricans,selenium resistance, 229 dewA A. nidulans gene, in conidiogenesis,27-8 dibenzofuran dioxygenase ferredoxin, 59 reductases, 57-8 1,3-dichloro-2-propanol, enantioselective dehalogenation. 158 1,2-dichloroethane,Xanthobacter autotrophicus degradation, 152 dichloromehane, microbial dechlorination, 165 2,2-dichloropropionicacid (22DCPA), 135 diffuse reflectance IR spectra (DRJlT), 207 3.4-dihydroxyphenylacetate2,3-dioxygenase,49 dioxygen, chemistry,48 dioxygenases,ring-hydroxylating,49 biochemical organization,50,Sl catalytic non-haem Fe centre, 72-5 catalytic terminal oxygenase component, 60-1,62 classification, 51, 52-3,754 ferredoxin component, 5 8 - 4 9 ferredoxins sequence analysis, 58-9 specificity,58 iron-sulphur clusters, 61-72 amino acid sequence comparisons,67-8,
71 classes, 61-3
ligand analysis, 63 site-dmcted mutagenesis, 68-72 S ~ ~ C ~ ~ O63,657.66 SCO~Y, reductase component,S 1 , 5 5 4 subunit composition, 50,54 disulphide bridges, in hydrophobins, 9 DM-nitrophen, 188 DRIFT spectra, 207 Dutch elm disease, cerato-ulmin in, 18, 31-2 dye displacementmetal analysis, 198-9
crassa gene, in coni&ogenesis.2.8-9 ectomycollhiza, hydrophobin genes in formation, 33 electrochemicalmetal analysis, voltammetry, 194-5 electron microscopy see transmission electron microscopy electron spin resonance (ESR) spectroscopy, 208 enantioselection,microbial, 158 enniatin synthetase, 1 0 0 4 depsipeptide formation, 103,104 gene structure, 102-3 molecular structure, 102 N-methylationmechanism, 1 W 1 structureffunction,100,101 substrate specificity, 101-2 enniatins, 99-100 synthesis, 91 R-epichlorohydrin, 158 epimerization, by peptide synthetases, 92 ergot peptide alkaloids, 108-11 ergotamine, 108,109 Escherichia coli arsenic resistance, 226 chemotaxis, calcium ions in, 188 copper metabolism study, 222 genes in inorganic ion physiology, 210 iron in, 216 Mossbauer spectroscopy,209 iron-limited growth, 189 metal-tolerand-sensitive,isolation, 214 silver nistance, 230 Eucalyptus globulus, Pisolifhus tinctorus association, 33 e m N.
fatty acid synthase, 88 activation domain organization, 94 FBF, in fruit body formation, 24 ferredoxins, of dioxygenases, 58-60 femtin, 216 ferrozine, for iron determination, 190-1
266 flavoproteins in droxygenase system, 50,51 iron-sulphur, of dioxygenase reductases, 55-7 Fourier transform IR spectroscopy, 206-7 fruit bodies of basidiomycetes, 3 formation, 22-7 fungi biology, 1-3 peptide synthases 6-(L-a-aminoadipyl)-cysteinyl-o-valine. 96-9 beauvericin, 105 cyclosporin, 105-7 enniatins, 99-104 ergot peptide alkaloids, 108-11 SDZ 214-103,107-8 fura-2, in calcium analysis, 191 Fusaria, enniatin-producing, 99, 102 gas chromatography, 198 gene probing, 212-13 gene sequence libraries, 21 1 germanium, 182 microbial accumulation, 227 glutathione, in dichloromethane dechlorination, 165 gramicidin S, synthesis, 86-7 gramicidin synthetases, 87 molecular masses, 89 haem proteins analysis in vivo, 218-19 Mbssbauer spectroscopy, 209 halocompound dehalogenation see alcohol dehalogenation; alkane dehalogenation; alkanoic acid dehalogenation halohydrin epoxidases, 153 haploid fruiting alleles (hfa), in monokaryons, 24 HC-toxin, 110 HD genes, in fruit body formation, 2 3 4 herbicides Dalapon, 135 phosphinothricin as, 120 Hefernsigma akashiwo, phosphorus metabolism, 197 His motif, in peptide synthetases, 92 hybridization techniques, 212-13 hydrophobins, 3-45
INDEX in aerial hypha formation, 19-22 cerato-ulmin. sequence determination, 18 in conidiogenesis, 27-9 Aspergillus nidulanr genes, 27-8 Neumspora crassa genes, 28-9 discovery, 3-4 in fruit body formation ABHl gene expression, 26-7 functions, 26-7 SC gene expression, 24-6 identity, 4-10 agglutinin relationships, 8-9 assembly, 9-10 characteristics, 6, 8 cysteine residues, 9 hydropathy patterns, 6, 7 sequence diversity, 5 . 5 4 in pathogenesis, 29-33 Dutch elm disease, 31-2 fungal host adhesion, 29-31.32 human infections, 32-3 as plant defence response elicitors, 33 purification, enhancement, 34 in rodlet formation, genetic experiments, 11-12
sc3 purification, SC3, 14 surface activity experiments, 14-18 surface activities, 13-19 cerato-ulmin, 18-19 in symbiosis, 33-4 in technology, 34-6 applications, 35-6 and hydrophobin properties, 34-5 4-hydroxyphenylpymvate droxygenase, 73 hyphae aerial formation, SC3 hydrophobin in, 19-22 growth, 1-2 transport processes, 2 Hyphomicrnbim, dehalogenase, 165 inductively coupled plasma AES, 193 inductively coupled plasma MS, 194 ion chromatography, 197-8 ion-exchange resins, for metal removal from medium, 189 ion-selective electrodes, 1 9 5 4 cadmium, 226-7 copper, 222 iron, 216-21 aluminium interference, 215 analysis, 190-1
INDEX assays, 216 forms of,216 growth limitation, 189-90 oxidation/reduction, 220-1 in oxygenase catalysis, 49 protein analysis in vivo, 218-20 siderophores, 181,217-18 lransport, 181,217 iron-sulphur clusters of dioxygenases. 61-72 amino acid sequence comparisons, 67-8, 71 classes, 61-3 Iigand analysis, 63 sitedirected mutagenesis, 68-72 Spectro~c~py, 63,65-7.66 iron-sulphur proteins, analysis in vivo, 219-20 isopenicillin N synthase, 49, 73 isotope transport assays, 199-200 Klebsiella aemgenes cadmium resistance, 227 lead resistance. 228 Lactobacillus planrarum, iron requirement, I 9 0 lead, microbial resistance, 213,228 lectin-like activity of cryparin, 9 lectins, cysteine residues, 9 libraries, gene sequence, 211 lichens, 3 algal symbiosis, hydrophobins in, 3 3 4 Listeria monocytogenes, iron transport assay, 217 D-lysergylpeptide assembly, 108-1 1 Magnaporrhe grisea, hydrophobic host adhesion, 29-30 manganese oxidation, microbial, 207 mass spectrometry, inductively coupled plasma, 194 MAT genes in fruit body formation, 22-5 in SC3 regulation, 19-20 meiospores, fungal, 3 membrane proteins, in metal transport, 181 mercury biosensor, 212 microbial resistance, 228-9 Bacillus, 213,229 metallothioneins, copper, 221 metaldmtalloids, 177-241 aluminium, 182,214-16 analysis, 190-205 cell treatment for, 191-2
267 chromatography, 198 colorimetry, 190-1 contamination avoidance, 192 dye displacement, 198-9 inductively coupled plasma-mass spectrometry, 194 ion chromatography, 197-8 ion-selective electrodes, 1 9 5 4 isotope transport assays, 199-200 neutron activation, 196-7 proton displacement, 199 sample treatment, 192 spectroscopy, 1 9 3 4 transmission electron microscopy, 201-5 voltammetry, 1 9 4 5 antimony, 182 arsenic, 1 8 2 , 2 2 5 4 biosensors, 21 1-12 cadmium see cadmium copper see copper detoxification processes, 183 essential, 180-2 gene probing, 212-13 germanium, 182,227 “indifferent”, 183 ion complexation in media, 185-90 and bioavailability, 185, 186-7 calcium concentration control, 187-8 and growth limitation, 188-90 and speciation, 185-7 iron see iron lead, 213,228 mercury see mercury molecular genetics, 210-1 I , 232 molybdenum, 181 nickel, 224-5 potassium see potassium properties, 183-5 d block, 184 p block, 184-5 s block. 183-4 resistanfhlerant hacteria isolation, 213-14 selenium, 182, 229 silver, 229-30 sodium, 181 S p e C t r ~ ~ ~205-9 ~py, analytical techniques, 1 9 3 4 electron spin resonance, 208 electronic, 205-6 metal binding sites, 205 MBssbauer, 209 nuclear magnetic resonance, 208-9
268 metaldmetalloids, spectroscopy (continued) vibrational. 206-8 tellurium, 230-1 terminology, 179-80 tin, 231 toxicity and resistance, 182-3, 225 zinc, 223-4 Metarhizium anisopliae, hydrophobic host adhesion, 30 methane monooxygenases, and haloalkane metabolism, 164-5 4-methoxybenzoate monooxygenase, 49 iron site, spectroscopy, 75 4-methyl-3-hydmxyanthranilic (4-MHA) pentapeptide lactone, 111-12,112 Micrococcus luteus, lead resistance, 228 molybdenum, transport, 181 L-2-monochloropropionic acid, commercial production, 143, IS8 monooxygenases, 49 methane, and haloalkane metabolism, 164-5 4-methoxybenzoate, 49 iron site, spectroscopy, 75 Moraxella plasmids, 149 species B, dehalogenases, 138, 141, 142, 162 Mossbauer spectroscopy, 209 of iron-sulphur clusters in dioxygenases, 64. 65 mycorrhiza see ectornycorrhiza naphthalene dioxygenase, 49 ferredoxin, 58, 5 9 , m reductases, 58 Neumspora cmssa conidiogenesis genetics, 28 spores, rodlet layer, 11 neutron activation analysis, 196-7 nickel, microbial interactions, 224-5 Nitr-5, 188 nitrilotriacetic acid, 189 7'-nitrobenz-2-oxa-l,3-diazde (NBD). in iron analysis, 217 Nitachiu a h , germanium uptake, 227 nuclear magnetic resonance (NMR) spectroscopy, 2 0 W
oils, water dispersion, hydrophobins in, 35 oxygen see dioxygen oxygenase-type dehalogenases, 164-5 oxygenases, bacterial, 48-9
INDEX as biocatalysts, 48 monooxygenases/dioxygenases, 49 see also dioxygenases, ring-hydroxylating; monooxygenases
Paracoccus denitn'ficans. iron-limited growth, 189 peptide synthesis systems, backridfungi, 85-131 activation domain organization, 9 4 4 . 9 5 fungal, 96-1 11 delta-(L-alpha-aminoadipy1)cysteinyl-wvaline, 96-9 beauvericin, 105 cyclosporin, 105-7 enniatins, 99-104 ergot peptide alkaloids, 108-1 1 SDZ 214-103,107-8 peptide synthetase domain, 88-94 acyltransfer/epimerization modules, 91-2 amino acid activation, 9 3 4 modules, in activation domain, 90-1 motifs in carboxyl-adenylate-forming domain, 90 N-methylation module, 91 peptide synthetases, 88-90 thioesterase modules in genes, 92-3 prokaryotic, 111-22 acyl peptide lactone synthetases, 112-17 bialaphos, 120-2 surfactin, 117-20 research prospects, 1 2 2 4 tho1 template model, 86-8 periodic table, 183, I84 pbenylalanine bydroxylase, 49 phosphinothricin, 120, 120-2, I21 PhsB peptide synthetase, 121-2 phthalate dioxygenase, 55,56,57 ferrous active site, 75 spectroscopic analysis, 65-7.66 Pisolithus tinciorus, Eucalyptus globulus association, 33 plastics, hydrophobin adsorption, 15-17 polarography, 194 polyketide synthetases, 93 activation domain organization, 9 4 , 9 5 4 polymerase chain reaction, 212 Posidonia oceanica, trace element analysis, 197 potassium as essential metal, 180 transport, 181 rubidium as analogue, 200
INDEX protocatechuate 3,4-dioxygenase, 73 proton displacement metal analysis, 199 Pseudomonas, 47 dioxygenases, 50 haloalcohol dehalogenases, 154, 1 5 6 7 species 113, dehalogenase, 139,144 species CBS 3, 138, 141 Pseudomonas aeruginosa, silver accumulation, 229-30 Pseudomonas cepacia dehalogenases, 138, 141, 142 phthalate hoxygenase analysis, 65-7,66 Pseudomnas dehalogennns, dehalogenase, 137, 141 Pseudomnas putida benzene dioxygenase analysis, 65 dehalogenases, 139, 140, 143, 144-5 germanium uptake, 227 iron-sulphur cluster analysis, 219-20 Pseudomnas srurzeri, germanium accumulation, 227 Pseudomonas syringae, copper-resistant, 214 putidaredoxin, Mossbauer parameters, 65 pyridine, in haem protein analysis, 218 Pyrococcusfuriosus, tungsten in, 180 Rumaline stenospora, metal analysis, 196-7 RETRIEVE, 21 1 Rhizobium aluminium toxicity, 215 dehalogenases, 138, 139, 141, 142, 143-4 Rhodobacter capsulatus, Rieske proteins amino acid sequence, 67.7 1 sibdirected mulagenesis, 69, 71-2 Rhodococcus rrythmpolis, dehalogenases, 164 oxygenase type, 165 Rhodopseudomoms sphaemides, Rieske proteins, 67 Rieske proteins, 63 amino acid sequences, 67-72 site-directed mutagenesis, 68-72 Sacchammyces cerevisiae, 6&9,70 specm 64 spectroscopic analysis, 65-7 dA. nidulam gene, in conidiogenesis, 27 rodlet layer, 4, 8 fornation, 20,21 and hydrophobin wettability, 17-18 in lichedalgal synibiosis, 34 in pathogenicity, 33 rodlets, 10, 10-13
269 bacterial, 11 genetic experiments, 11-12 hydrophobins in formation, 11-13 isolation from fungal spores, 10-1 1 rubidium in potassium transport studies, 200 S-glucan, and rodlet location, 10 Sacchammyces cerevisiae copper uptake, 221 genetics of conidiogenesis, 27 iron metabolism, 22 1 Rieske proteins amino acid sequence, 68 site-directed mutagenesis, 68-9, 70 SC3 hydrophobin, 5 . 6 in aerial hypha formation, 19-22 discovery, 3-4 hydropathy pattern, 6 , 7 purification, 14 rodlet layer formation, 4 surface activity experiments, 14-18 SC4 hydrophobin, 5 . 6 discovery, 3-4 Schizophyllum commune fruit body formation, 2 2 4 gene expression, 2 3 , 2 3 4 hydrophobin function, 26 hydrophobin genes, 3 hydrophobins from, 14 dendrogram, 5 hydropathy patterns, 6, 7 in rodlet formation, 12-13 hyphal adhesion, 30-1 rodlet layer, 4 rodlets, 10, 12-13 SDZ 214-103, 107-8 selenium, microbial toxicity, 182, 229 siderophores, 181,217-18 silver, microbial toxicity, 229-30 sinefungin, as methylase inhibitor, 100-1 snake toxins, cysteine residues, 9 s d u m , transport, 181 Southern blotting, 212 spacer motif, in peptide synthetases. 92 spectrometry atomic fluorescence, 194 inductively coupled plasma-MS, 194 spectrophotometry, in haem protein analysis, 218-19 spectroscopy atomic absorption, 193 atomic emission, 193
270
INDEX
spectroscopy (continued) energy-dispersive X-ray, for TEM detection, 203-4 of iron-sulphur clusters in dioxygenases, 63, 657,154 for metal-microbe interactions, 205-9 electron spin resonance, 208 electronic, 205-6 metal binding sites, 205 Mossbauer, 209 nuclear magnetic resonance, 208-9 vibrational, 2Ch5-43 spinach ferredoxin, 61 Staphylococcus aureus arsenic resistance, 226 lead resistance, 228 Streptomyces clirysornallus, 113 Streptomyces clavuligerus, ACV synthase from, %,97 Streptomyces hygroscopicus, phosphinothricin from, 120 Streptomyces viridochromogenes, phosphinothricin from, 120 superoxide dismutase, Cu-Zn. 223 surfactin, 117-20 enzymes in assembly, 117-18,119 reactions in amino acid positions, 118-19 structure, 118 synthesis initiation, 118 synthetases, structdfunction, 119-20 symbiosis, fungal, 3 3 4 Syncephulastrum racemsum spores, rodlet layer, 11 Teflon, hydrophobin adsorption, 15-17.35 tellurium, microbial resistance, 230-1 tetrahionate, ion chromatography, 197-8 Thermm thermophilus iron isotope studies, 209 ironsulphur clusters, spectroscopy, 65, 66
Kieske proteins, amino acid sequence, 68 tellurium resistance, 230 Thiobncillusfermoxidans, 220 thioesterase genes, in peptide synthesis, 92 thiol template peptide synthesis model, 86-8 see also peptide synthesis systems, bacterialfungi thionins, cysteine residues, 9 thiosulphate, ion chromatography, 197-8 THN gene, in fruit body formation, 24 thri mutation, and SC3 expression, 19
tin, microbial toxicity, 231
tissue engineering, hydrophobins in, 35 toluene dioxygenase, 49 ferredoxin, 5 8 , 5 9 4 reductase, 57 toxin-agglutinin fold proteins, hydrophobin relationship, 8-9 trace elements, 180 deficiencies, and growth, 188-90 see also metaldmetalloids transition metals in oxygenase catalysis, 49 see also metaldmetalloids transmission electron microscopy for metals, 201-5 energy-dispersive X-ray spectroscopy detection, 2 0 3 4 selected-area diffraction with, 204-5 thin section preparation, 201-3 whole mounts, 201 transport in hyphae, 2 metals, 180-2 iron, 181. 217 isotope assays, 199-200.217 bis(tributy1tin)oxide (TBTO), bacterial degradation, 231 Trichodema hanionum, conidiospores, 13 Mchophyton menmgrophytes, microconidial rodlet layer, 10-11 tungsten, as essential metal, 180 Uromyces appendicularus, hydrophobic adhesion, 30 Vibrio alginolyticus copper-binding proteins, 223 copper-resistant/-sensitive, 214 voltammetry, 194-5
water, disinfection, copper/silver, 221 Western blotting, 213 wetA A. nidulans gene, in conidiogenesis, 27 X-ray absorption spectroscopy, phthalate dioxygenase, 6 7 X-ray photoelectron spectroscopy ( X P S ) , for hydrophobins, 17 Xanthobacfer aurotrophicus dehalogenases alkane, 16&2,163 alkanoic acid, 138. 141 DhlB overexpression, 151
271
INDEX 1,2-dichloroethane degradation, 152 zinc, microbial interactions, 223-4
zinc-sensitive E. cofi,214 zygomycetes, hyphal structure, 2
This Page Intentionally Left Blank