Characterisation of the Tailoring and Transport
Enzymes involved in the Microcystin Biosynthesis
Pathway
Leanne A. Pearson
A thesis submitted in fulfilment of the requirements for the degree of Doctor of
Philosophy
School of Biotechnology and Biomolecular Science
The University of New South Wales
Sydney, Australia
July, 2006 ACKNOWLEDGEMENTS
Many thanks to all whom contributed to this work.
I would like to extend my sincere thanks to Tom Börner and Elke Dittmann for allowing me the opportunity to conduct part of my research in their laboratories in
Berlin, and for donating many of the cyanobacterial strains and DNA samples used throughout this study. I would also like to thank The CRC for Water Quality, The
Australian Research Council, and The Deutscher Akademisher Austausch Dienst for funding this research. My thanks also go out to Anne Poljak and Mark Raftery from the
Bioanalytical Mass Spectrometry Facility (UNSW) for their help with the MALDI-TOF work. I would also like to thank Wendy Glenn for her assistance with the ultra centrifuge and FPLC, and Andy Netting for his help editing this manuscript. Many thanks to Michelle Moffitt, Bradley Moore, and Hans von Döhren for their advice regarding the biochemistry of McyI. Likewise, I wish to thank Tony George for his advice regarding the expression and characterisation of McyH. Many thanks to Ian
McFarlane for his advice and co-supervision over the years. A big thank you to my family and friends for their encouragement and support throughout my university career, and to all the members of the BGGM for making the lab a great place to work.
Finally, I wish to thank my supervisors Brett Neilan and Kevin Barrow for their wisdom, patience and inspiration. It has been an honour and a privilege working with you.
2 TABLE OF CONTENTS
ABSTRACT 8
CHAPTER 1. INTRODUCTION 9
1.0 Cyanobacteria 10
1.1 Cyanotoxins 11
1.2 Secondary metabolite biosynthesis 15 1.2.0 Non-ribosomal peptide biosynthesis 16 1.2.1 Polyketide biosynthesis 17 1.2.2 Mixed NRP/PK biosynthesis 18 1.2.3 Combinatorial biosynthesis 19 1.2.4 NRPS and PKS associated tailoring and transport enzymes 20 1.2.5 Biosynthesis of microcystin and nodularin 33 1.2.6 Regulation of microcystin biosynthesis 37
1.3 Aims 40
CHAPTER 2. MATERIALS AND METHODS 41
2.0 Bacterial strains and Culturing 42 2.0.0 Cyanobacteria 42 2.0.1 Escherichia coli 42
2.1 Nucleic Acid Extraction 43 2.1.0 Extraction of genomic DNA 43 2.1.1 Extraction of plasmid DNA 44
2.2 Polymerase chain reaction (PCR) 44
2.3 Purification of Nucleic Acids 46 2.3.0 Ethanol and isopropanol precipitation of DNA solutions 46 2.3.1 Column purification 46
2.4 Automated Sequencing and Analysis 47
2.5 Agarose gel electrophoresis 47
2.6 Modification of DNA fragments 48 2.6.0 Digestion of DNA with restriction endonucleases 48 2.6.1 End-filling DNA overhangs 48 2.6.2 Dephosphorylation of DNA with shrimp alkaline phosphatase (SAP) 49 2.6.3 Ligation of DNA fragments 49
2.7 Genetic manipulation of bacteria 50 2.7.0 Preparation of competent E. coli cells and transformation 50
2.8 Total Protein Assays 51 2.8.0 Bio-Rad Dc protein assay 51 2.8.1 Folin protein assay 51
2.9 Protein Electrophoresis 51
3 2.10 Ammonium sulphate precipitation of proteins 52
2.11 Immunodetection of proteins with Ni-NTA conjugates 52
CHAPTER 3. CHARACTERISATION OF THE ABC TRANSPORTER, MCYH 53
3.0 OVERVIEW 54
3.1 EXPERIMENTAL DESIGN AND METHODOLOGY 57
3.1.0 Screening various strains of cyanobacteria for mcyH orthologues 57
3.1.1 Sequence analysis 57
3.1.2 Phylogenetic analysis 58
3.1.3 Heterologous expression and purification of McyH and component peptides 59 Engineering the McyH expression constructs for the expression of full-length McyH 59 Engineering the McyH expression constructs for the expression of the individual membrane and ATPase domains of McyH 60 Engineering expression constructs for the expression of N-terminally truncated McyH peptides 61 Over expression of recombinant peptides 63 Purification of recombinant peptides 63
3.1.4 Construction of everted membrane vesicles from E. coli expression strains 64
3.1.5 Enzyme assays 65 ATPase assays 65 Microcystin transport in everted membrane vesicles 66
3.1.6 Mutagenesis of mcyH 67
3.1.7 Regulation of McyH expression 68 Regulation of McyH expression under different light conditions 69 Regulation of McyH expression in different mcy mutants 71
3.2 RESULTS 72
3.2.0 Distribution of mcyH orthologues in various species of cyanobacteria 72
3.2.1 Sequence analysis of mcyH 72
3.2.2 Over-expression and Purification of histidine-tagged McyH peptides 75
3.2.3 Enzyme assays 78 ATPase assays 78 Microcystin transport in everted membrane vesicles 78
3.2.4 Mutagenesis of mcyH 79
3.2.5 Regulation of McyH expression 79 Regulation of McyH expression under different light conditions 79 Regulation of mcyH expression in mcy knock-out mutants 79
3.3 DISCUSSION 81 4 CHAPTER 4. CHARACTERISATION OF THE 2-HYDROXYACID DEHYDROGENASE, MCYI 92
4.0 OVERVIEW 93
4.1 EXPERIMENTAL DESIGN 97
4.1.0 Screening various strains of cyanobacteria for mcyI orthologues 97
4.1.1 Sequence analysis 97
4.1.2 Phylogenetic analysis 98
4.1.3 Complementation of E. coli auxotrophs with mcyI 98 Construction of the complementation plasmids pDRIVE(T7/mcyI) and pIN-III(lac/mcyI) 98 Complementation experiments 99
4.1.4 Heterologous expression and purification of McyI 100 Engineering the pET30(mcyI) expression construct 100 Overexpression of histidine-tagged McyI 100 Purification of histidine-tagged McyI 101
4.1.5 Determining the subunit organisation of native McyI 102
4.1.6 Enzyme assays 102 Dehydratase assays 102 Chemical synthesis of 2-Hydroxy-3-methylsuccinic acid 103 Oxidoreductase assays 104 End-product analysis 105 Determining the optimal pH of the McyI OAA reductase assay 106 Determining the optimal temperature and thermostability of the McyI OAA reductase assay 106 McyI specificity 106 McyI Kinetic analysis 107 Inhibition assays 107 Cell extract studies 108
4.2 RESULTS 109
4.2.0 Distribution of mcyI orthologues in various species of cyanobacteria 109
4.2.1 Sequence analysis of mcyI 109
4.2.2 Mutant complementation studies 113
4.2.3 Over-expression and Purification of histidine-tagged McyI 114
4.2.4 Subunit composition of McyI 115
4.2.5 Biochemical and kinetic properties of McyI 115 [MeDha7]microcystin-LR hydratase assays 115 Oxidoreductase assays 115 End-product analysis 117 Inhibition assays 117
4.3 DISCUSSION 118
5 CHAPTER 5. CHARACTERISATION OF THE O-METHYLTRANSFERASE, MCYJ 134
5.0 OVERVIEW 135
5.1 EXPERIMENTAL DESIGN 138
5.1.0 Screening various strains of cyanobacteria for mcyJ orthologues 138
5.1.1 Sequence analysis 138
5.1.2 Phylogenetic analysis 139
5.1.3 Heterologous expression and purification of McyJ 139 Engineering the pET30(mcyJ) expression construct 139 Overexpression of histidine-tagged McyJ 140 Purification of histidine-tagged McyJ 140
5.1.4 Enzyme assays 141
5.2 RESULTS 143
5.2.0 Distribution of mcyJ orthologues in various species of cyanobacteria 143
5.2.1 Sequence analysis of mcyJ 143
5.2.2 Overexpression and Purification of histidine-tagged McyJ 146
5.2.3 Enzyme assays 147
5.3 DISCUSSION 148
CHAPTER 6. GENERAL DISCUSSION 156
6.0 Major results 157 6.0.0 McyH 157 6.0.1 McyI 159 6.0.2 McyJ 160
6.1 Overall significance 162
REFERENCES 166
6 APPENDIX 198
A1. Media for culturing E. coli 198 A1.0 Luria Bertaini medium 198 A1.1 Tryptone phosphate medium 198 A1.2 Superbroth 198 A1.3 M9 salts 199
A2. Sequence alignments for the generation of phylogenetic trees 200 A2.0 McyH homologous sequence alignment 200 A2.1 McyH ATPase domain homologous sequence alignment 200 A2.2 McyI homologous sequence alignment 200 A2.3 McyJ homologous sequence alignment 200
7 ABSTRACT
The cyanobacterium Microcystis aeruginosa is widely known for its production of the potent hepatotoxin microcystin. This cyclic heptapeptide is synthesised non-ribosomally by the thiotemplate function of a large, modular enzyme complex encoded within the 55 kb microcystin synthetase (mcy) gene cluster. The mcy gene cluster also encodes several stand-alone enzymes, putatively involved in the tailoring and export of microcystin.
This thesis describes the characterisation of the Adda O-methyltransferase, McyJ, the
2-hydroxy-3-methylsuccinic acid dehydrogenase, McyI, and the ABC transporter,
McyH. A combination of bioinformatic, molecular, and biochemical approaches have been used to elucidate the structure, function, regulation and evolution of these microcystin synthetase gene cluster encoded enzymes. Extensive sequence analyses are reported, including phylogenetic and structural studies. The distribution of mcyH, mcyI and mcyJ orthologues in different species of cyanobacteria has been investigated via genetic screening with M. aeruginosa specific, and degenerate oligonucleotide primers.
McyH, McyI and McyJ have been heterologosly over-expressed in E. coli and enzymatically assayed. Finally, an McyH antibody has been engineered and used to investigate the regulation of the McyH ABC transporter in wild-type (WT)
M. aeruginosa, and in various non-toxic engineered mutant strains. The results of these experiments are discussed with respect to the roles of McyH, McyI and McyJ in microcystin biosynthesis, and their relevance to the fields of water quality management and rational drug design and production.
8 CHAPTER 1.
INTRODUCTION
9 1.0 Cyanobacteria
The cyanobacteria or “blue-green algae” as they are commonly termed, comprise a diverse group of oxygenic photosynthetic bacteria that possess the ability to synthesise chlorophyll-a and the phycobilin protein, phycocyanin. Stretching back into the fossil record as far as 3 500 Ma, these microbes are believed to be responsible for the evolution of Earth’s early oxygenic atmosphere (Schopf, 2000).
The cyanobacteria inhabit a wide range of aquatic and terrestrial environments from the hot springs of Yellowstone, Montana to the melt water ponds of Antarctica (Lau et al,
2005; Jungblut et al, 2005). Certain species of cyanobacteria also form complex symbiotic relationships with fungi and higher eukaryotes, for example, several Nostoc sp. are commonly associated with lichens, while Prochloron sp. may be ecto- or endosymbionts of marine ascidians (Paulsrud et al, 1998; Maruyama et al, 2003).
The cyanobacteria display incredible morphological diversity and may be unicellular, colonial, or filamentous. Many cyanobacteria are capable of forming specialised cells known as heterocysts, which enable the fixation of atmospheric nitrogen (Wolk, 1996).
Some species may also form spore-like akinetes, which allow them to survive under prolonged periods of environmental stress (Thiel and Wolk, 1983).
In recent years, cyanobacteria belonging to the genera Spirulina and Aphanizomenon have been recognised as a potentially beneficial health food supplements, and are distributed widely across the globe (Saker et al, 2005). However, cyanobacteria are far better known for their adverse affects on human health. Many aquatic, bloom-forming species of cyanobacteria are capable of producing biologically active secondary
10 metabolites, which are highly toxic to many eukaryotic organisms. The release of cyanotoxins into recreational and drinking water supplies has obvious and far-reaching implications for public health and the environment. Hence, considerable research efforts have been directed towards the control and removal of cyanobacteria and their toxins from our waterways.
While the toxic cyanobacteria are generally considered a nuisance, they have recently been recognised as excellent candidates for rational drug design. The manipulation of cyanobacterial secondary metabolite pathways via the addition, deletion or rearrangement of genes, has the potential to produce a virtually limitless combination of novel biologically active compounds including neuro-active and chemotherapeutic drugs.
1.1 Cyanotoxins
The small biologically active secondary metabolites produced by many strains of cyanobacteria are collectively known as cyanotoxins. In addition to being responsible for numerous cases of animal morbidity and mortality, certain cyanotoxins are recognised as promoters of human cancer (Yu et al, 2001). The cyanotoxins span five different classes: the neurotoxins, hepatotoxins, cytotoxins, and irritant toxins
(lipopolysaccharides). While all cyanotoxins can produce adverse symptoms in humans, the neurotoxins and hepatotoxins are of greatest concern to public health.
The alkaloid toxins, saxitoxin, anatoxin, and cylindrospermopsin are produced by various filamentous species of cyanobacteria including Anabaena sp., Aphanizomenon sp., Lyngbya sp., Cylindrospermopsis raciborskii and Planktothrix (Humpage et al,
11 1993). Saxitoxins are tricyclic compounds containing two guanidine groups (Figure
S1.0). Their toxicity is mediated by the blockage of voltage-gated sodium channels and disruption of neuronal transmission. Acute toxic doses of saxitoxin ultimately result in paralysis and respiratory failure (Strichartz et al, 1986). Three common anatoxins have been isolated from cyanobacteria. Anatoxin-a and homoanatoxin-a are described as secondary amines (Devlin et al, 1977). Anatoxin-a(s) is a unique phosphate ester of a cyclic N-hydroxyguanidine structure (Matsunaga et al, 1989) (Figure S1.0). This group of alkaloids disrupt acetylcholinesterase activity, causing muscle cell over-excitation and symptoms of cramping, convulsion and respiratory paralysis (Carmichael, 1994).
Cylindrospermopsin is a tetra-cyclic alkaloid containing a uracil moiety attached to a sulphated guanidine moiety (Figure S1.0). This toxin is activated by cytochrome P-450
(Norris et al, 2002) causing a multitude of cytotoxic effects including DNA strand breakage, and disruption of DNA and protein biosynthesis (Shen et al, 2002). In mammals, cylindrospermopsin poisoning can cause liver, kidney, thymus and heart damage (Terao et al, 1994; Wiegand and Pflugmacher, 2005).
The hepatotoxic microcystins comprise the largest and most structurally diverse group of cyanobacterial toxins. Over sixty-five isoforms of microcystin varying by degree of methylation, hydroxylation, epimerisation, peptide sequence and toxicity have been identified (Sivonen and Jones, 1999; Dittmann and Wiegand, 2006). Underlying the extraordinary heterogeneity present among the microcystins is their common cyclic structure and possession of several rare, highly conserved amino acid moieties.
Collectively, the microcystins may be described as monocyclic heptapeptides containing both D- and L-amino acids plus N-methyldehydroalanine and a unique ß-
12 amino acid side-group, 3-amino-9-methoxy-2-6,8-trymethyl-10-phenyldeca-4,6-dienoic acid (Adda) (Botes et al, 1985) (Figure 1.0). Microcystins have been isolated from the genera Microcystis, Anabaena, Oscillatoria, Planktothrix, Chroococcus and Nostoc, and have been associated with various human and animal poisonings around the globe.
The nodularins are a small group of pentapeptides that closely resemble microcystin in toxicity and structure [(Adda), D-glutamic acid (D-Glu), N-methyldehydrobutyrine
(MeDhb), D-erythro-β-methylaspartic acid (D-MeAsp), and L-arginine (L-Arg)]
(Rinehart et al, 1988) (Figure 1.0). Upon ingestion, microcystins and nodularins are transported to the liver by organic anion transport proteins where they exert their toxicity via inhibition of protein phosphatases 1 and 2A (Runnegar et al, 1991;
Runnegar et al, 1995; Dawson et al, 1998). Inhibition of protein phospatases can lead to excessive phosphorylation of structural filaments, subsequent cyto-skeletal degradation and breakdown of hepatic ultra structure (Eriksson et al, 1990; Sahin et al, 1996).
Retraction of hepatocytes from neighbouring cells and sinusoidal capillaries causes blood to become pooled in the liver tissues. This ultimately results in local tissue damage, organ failure and haemorrhagic shock (Sahin et al, 1996). Microcystin and nodularin doses of as little as 50-70 µg per kilogram of body weight have been reported to produce rapid death in both native and domestic animals (Runnegar, 1988; Rinehart et al, 1994). Several human fatalities have also been reported following acute microcystin poisoning (Jochimsen et al, 1998).
The revelation that cyanobacterial hepatotoxins cause protein phosphatase inhibition has raised the disturbing possibility that human exposure to non-lethal doses of these compounds may contribute to the development of cancer (Yoshizawa et al, 1990;
13 Nishiwaki-Matsushima et al, 1991). Several laboratory studies have indicated that chronic exposure to microcystins can indeed promote skin and liver tumours in rats and mice (Falconer, 1991; Nishiwaki-Matsushima et al, 1992). Epidemiological data suggest that similar long-term effects such as hepatocellular carcinoma may also be observed in humans (Yu, 1989; Yu, 1995). Such results highlight the need for sensitive and rapid detection methods and stringent monitoring of cyanobacterial hepatotoxins in drinking water supplies.
Whilst the true function of cyanotoxins remains a mystery, several putative extracellular and intracellular functions have been suggested for these secondary metabolites. There is compelling evidence that toxins such as microcystin and anatoxin may confer similar advantages to cyanobacteria as secondary metabolites do to vascular plants. For example, they may act as feeding deterrents to other bacteria and higher order grazers such as zooplankton (cladeoceans and rotifers) (DeMott et al, 1991), or they may function as attractants to beneficial organisms including heterotrophic bacteria and protozoa (Pearl and Millie, 1996). An allelopathic function has also been suggested for the microcystins due to their ability to inhibit the growth of various algal species
(Keating, 1978).
While most of the putative functions of cyanotoxins require the export of these compounds from the cell, a cyanotoxin transport system has yet to be identified.
Release of toxins into the extracellular environment is thus generally attributed to cell aging and lysis (Rapala et al, 1997). However, the finding that microcystin is present in the mucilaginous sheath and cell wall of M. aeruginosa (Shi et al, 1995), suggests the possibility of active secretion of toxin to these areas. Due to the chemical nature and
14 subcellular localisation of microcystin, several putative intracellular functions have also been proposed for this toxin. For example, the concentration of microcystin localised on the photosynthetic membranes of the thylakoid, as evidenced by in situ gold hybridisation studies (Shi et al, 1995; Young et al, 2005), has led to speculation that the toxin may play a role in photosynthesis. Alternatively, due to its ability to bind and chelate ferrous ions, microcystin has also been hypothesised to function as an antioxidant, protecting the cell from free radical damage during periods of high intracellular iron concentration (Utkilen and Gjolme, 1995).
Although the precise function of the cyanotoxins is yet to be elucidated, current research focusing on the biosynthesis and regulation of these secondary metabolites, promises to shed some light on this subject.
1.2 Secondary metabolite biosynthesis
Not all proteins are synthesised on the ribosome. Many low molecular weight peptides of fungal and bacterial origin are synthesised on large modular enzyme complexes known as peptide synthetases (PSs) (Neilan et al, 1999; Walsh, 2004). Compounds produced in this manner often have bioactive properties including antibiotic, anti- tumour, immunosuppressant and surfactant activities. Some non-ribosomally-produced compounds are also highly toxic to humans and other eukaryotic organisms. The cyanobacterial hepatoxins, microcystin and nodularin are classic examples of such non- ribosomal peptide (NRP) toxins.
15 1.2.0 Non-ribosomal peptide biosynthesis
Non-ribosomal peptides represent a large, structurally diverse class of compounds that differ from their ribosomal counterparts in several ways. While the production of ribosomal peptides is restricted to 20 amino acid building blocks, several hundred amino acids are found incorporated via non-ribosomal peptide synthetases (Marahiel et al, 1997). For example, D-amino acids are commonly observed in NRPs, as are other unusual structural elements such as glycosylated and N-methylated residues
(Schauwecker et al, 2000). Non-ribosomal peptides frequently display a branched or cyclic structure.
NRPs are produced by non-ribosomal peptide synthetases (NRPSs), large modular enzymes with several specialised domains: the adenylation (A) domain, the thiolation
(T) domain, and the condensation (C) domain. The A-domain selects the cognate amino acid and activates it as an amino acyl adenylate (Dieckmann et al, 1995; May et al,
2002). The activated amino acid is thioesterified via covalent linkage to the thiol group present on a 4’-phosphopantetheine carrier, which is posttranslationally attached to the peptide synthetase T-domain [or peptidyl carrier protein (PCP)]. The final stage of elongation involves the C-domain, which catalyses peptide bond formation (Keating et al, 2002). Together, these three constituents form a module: a section of the NRPS chain responsible for the incorporation of a single amino acid building block of the growing polypeptide chain. Modules can be defined as initiation or elongation modules.
Initiation modules, which constitute the first step of non-ribosomal peptide biosynthesis, are comprised of an A-domain and a PCP. Elongation modules in addition, contain a C- domain for peptide bond formation (Finking and Marahiel, 2004). During peptide biosynthesis, the growing peptide chain is transferred from one module to the next until
16 it reaches the PCP-domain of the terminal module. The terminal module typically contains a thioesterase (TE) domain, which is involved in the cyclisation and release of the final peptide product (Schneider and Marahiel, 1998) (Figure S1.1).
In addition to the enzymes required for the biosynthesis of the NRP, several tailoring enzymes may be involved in its modification. Tailoring enzymes such as oxidoreductases, epimerises, methyltransferases, and glycosylases (reviewed below), further contribute to the structural and functional diversity of non-ribosomal peptides.
The gene clusters encoding NRP biosynthesis enzymes are usually arranged in large operons or gene clusters of 20 kb or more. The arrangement of encoded modules usually adheres to the “co-linearity rule”, which suggests that the order of catalytic processes involved in the biosynthesis of a non-ribosomal metabolite is generally the same as the order of the genes which encode these catalytic enzymes (Kleinkauf and von Döhren , 1996).
1.2.1 Polyketide biosynthesis
Polyketides (PKs) represent one of the largest and most structurally diverse groups of secondary metabolites. These compounds are produced by a myriad of organisms including various microorganisms, fungi, plants, algae and dinoflagellates (Rein and
Borrone, 1999). PKs are synthesised by polyketide synthases (PKSs) in a mechanism analogous to fatty acid biosynthesis (Cane and Walsh, 1999). Three types of bacterial
PKSs are known to date. Type I PKSs are multifunctional enzymes that are organised into modules, each of which harbours a set of distinct, non-iteratively acting activities responsible for the catalysis of one cycle of polyketide chain elongation. Type II PKSs
17 are multienzyme complexes that carry a single set of iteratively acting activities. Type
III PKSs, also known as chalcone synthase-like PKSs, are homodimeric enzymes that essentially are iteratively acting condensing enzymes (Shen, 2003).
Like NRPSs, PKSs are comprised of several enzymatic domains. The acyltransferase
(AT) domain is involved in substrate recognition. As observed in fatty acid synthesis, the substrate usually takes the form of malonyl-CoA, however methyl-, ethyl-, or propyl-, malonyl-CoA units may also be utilised by PKSs (Katz, 1997). In type I and II
PKSs, the AT-domain transfers the substrate to a holo acyl carrier protein (ACP), whereas type III PKSs act directly on the acyl CoA substrates independently of ACPs
(Shen, 2003). The substrate is then relocated to an active site on a ketosynthase (KS) domain (Finking and Marahiel, 2004). Post-translational modification of the PKS from the apo to the holo form is catalysed by a phosphopantatheinyltransferase (PPTase)
(Lambalot et al, 1996). Successive elongation steps are followed by macrocyclisation, which is catalysed by a thioesterase (TE) domain (Finking and Marahiel, 2004). The
PKS is then free to partake in further rounds of synthesis (Figure S1.2).
Additional domains may also be involved in PK synthesis, such as ketoreductase (KR), dehydratase (DH), and enolreductase (ER) domains (Katz et al, 1997). These tailoring domains contribute further to the extraordinary structural variation observed among
PKs.
1.2.2 Mixed NRP/PK biosynthesis
Secondary metabolite pathways occasionally incorporate a mixture of NRP and PKS biosynthetic steps. One of the first mixed NRPS/PKS pathways to be identified was that
18 of the immunosuppressant rapamycin. This macrolide drug is synthesised by three polyketide synthases, RapS1, RapS2 and RapS3, and a single peptide synthetase, RapP
(Molnar et al, 1996). Bleomycin, streptothricin, yersiniabactin and lovastatin are similarly produced by mixed NRPS/PKS pathways (Du et al, 2000; Fernandez-Moreno et al, 1997; Miller et al, 2002; Hendrickson et al, 1999).
NRPSs and PKSs may occasionally be encoded within a single ORF. Such hybrid
NRPS/PKS enzymes have been associated with the biosynthesis gene clusters of the electron transport inhibitor myxothiazol (Silakowski et al, 1999), and the cyanotoxins microcystin and nodularin (Tillett et al, 2000; Moffitt et al, 2004).
1.2.3 Combinatorial biosynthesis
Recent advances in molecular biology have enabled the genetic manipulation of secondary metabolite pathways to produce “unnatural” or “hybrid” natural products.
Preliminary experiments by Hopwood et al (1985) combined the genes from different
Streptomyces antibiotic biosynthesis pathways to produce a novel group of hybrid compounds. The rational design of novel biologically active compounds through combinatorial biosynthesis is currently a widely practiced, rapidly progressing field of science (Weissmann and Leadlay, 2005). While most combinatorial biosynthesis experiments have targeted the early steps in PK/NRP biosynthesis, increasing attention is being directed towards the post-biosynthetic tailoring enzymes, many of which have innate substrate flexibility (Rix et al, 2002).
19 1.2.4 NRPS and PKS associated tailoring and transport enzymes
The structural and functional diversity present among secondary metabolites is largely due to pre- or post-synthetic modifications introduced by tailoring enzymes including: oxidoreductases (eg. oxygenases and ketoreductases), group transferases (eg. glycosyltransferases, methyltransferases and acyltransferses), epimerases (eg. racemases and epimerisation modules), cyclases, and halogenases. Such enzymes may be embedded within PK/NRP modules and act in cis during peptide elongation, or may alternatively function as separate subunits or stand-alone enzymes and act in trans.
A brief overview of some important tailoring enzymes is given below. Where possible, attention has been given to stand-alone enzymes.
Oxidoreductases
Oxidoreductase reactions are among the most commonly observed tailoring reactions in
NRP and PK biosynthsis pathways. A diverse range of enzymes catalyse these reactions, including oxidases, oxygenases, peroxidases, reductases and dehydrogenases.
Oxidoreductase reactions can have a dramatic impact on the stereo-electronic and physical-chemico properties of molecules. They can generate or remove chiral centres, introduce highly reactive functional groups, inter-convert H-bond donor/acceptor sites, and change overall solubility (Rix et al, 2002).
Oxygenases significantly contribute to the enormous structural diversity observed among secondary metabolites. Such enzymes can create hydroxy groups, epoxides, dioxetanes or peroxides (Rix et al, 2002). For example, a single cytochrome P-450 monoxygenase OleP, from Streptomyces antibioticus catalyses the epoxidation of 8,8a-
20 dihydroxy-6-deoxyerythronolide-B, during oleandomycin biosynthesis, while two oxigenases TcmG and TcmH, are required for the biosynthesis of the highly hydroxylated cyclohexenone moiety of tetracenomycin-C (Shah et al, 2000; Shen and
Hutchinson, 1994).
Monoxygenases typically generate hydroxy groups and expoxides, while dioxygenases usually create dioxetane or peroxide species. For example, the recently characterised dioxygenase encoded within the vancomycin antibiotic biosynthesis gene cluster of various actinomycete species, operates via a C2-peroxy intermediate. This enzyme designated DpgC, is responsible for the conversion of 3,5-dihydroxyphenylacetyl-CoA
(DPA-CoA) to 3,5-dihydroxyphenylglyoxylate, the penultimate intermediate of the crucial structural amino acid monomer 3,5-dihydroxyphenylglycine. The DpgC oxygenase belongs to a select group of oxidoreductases that do not require the presence of metal ions or cofactors (Chen et al, 2001).
Ketoreductases also feature widely in secondary metabolite pathways. These enzymes, which are involved in the biosynthesis of both the PK derived aglycon moieties and the deoxysugar building blocks of complex natural products, catalyse the conversion of ketones to secondary alcohols (Rix et al, 2002). Ketoreductases are commonly integrated within the modules of PKSs, for example, as in erythromycin biosynthesis
(Donadio and Katz, 1992). However several stand-alone enzymes have also been identified. While post-PKS ketoreductases are quite rare, these enzymes are excellent candidates for combinatorial biosynthesis of novel compounds. Hence, several research groups have focused on their characterisation.
21 A pharmaceutically relevant ketoreductase DauE, features in the biosynthesis of the daunomycin and aclacinomycin-A antibiotics. This post-PKS tailoring enzyme catalyses the NADPH dependent reduction of the 7-oxo compound aklaviketone, to form aklavinone, the last common intermediate in the daunomycin and aclacinomycin-
A pathways. The utility of DauE resides in its flexible substrate specificity, as evidenced by its capacity to reduce artificial substrates maggiemycin and 7- oxodaunomycinone to rhodomycinone and daunomycinone, respectively (Dickens and
Strohl, 1996).
Another example of a post-PKS ketoreductase is the tetrahydroxynaphthalene reductase
(THNR) involved in the biosynthesis of fungal melanin. This enzyme, isolated from the rice pathogen Magnaporthe grisea, catalyses the NADPH-dependent reductions of the aromatic rings of 1,3,6,8-tetrahydroxynaphthalene to (+)-scytalone, and 1,3,8- trihydroxynaphthalene to (-)-vermelone (Vidal-Cros et al, 1994).
Epimerases
D-amino acids are a common feature of many non-ribosomal peptides. For example, the common tripeptide precursor of penicillin and cephalosporin is L-δ-aminoadipyl-L- cysteinyl-D-valine, while four of seven constituents in the glycosylated backbone of the vancomycin antibiotics are D-amino acids (Stachelhaus and Walsh, 2000). These non- proteinogenic building blocks are important for creating architectural diversity, providing resistance to proteolysis, and imposing stereochemical constraints (Luo et al,
2002). Occasionally, stand-alone enzymes generate the D-amino acids observed in non- ribosomal pathways, however in most characterised systems, L-amino acids are
22 epimerised during the peptide elongation process by embedded epimerisation (E) domains (Linne et al, 2001).
E-domains represent a class of cofactor independent amino acid epimerases that catalyse the interconversion of D- and L- amino acids in many secondary metabolite pathways. Such domains are approximately 50 KDa in size and are located adjacent to
A-domains at the C-terminal end of NRPS modules. During peptide synthesis, the A- domain activates readily available L-amino acids, which are subsequently installed as aminoacyl- or peptydl-S-PCP acyl enzyme intermediates (Walsh et al, 2001). The E- domain then converts the L-amino acids to their corresponding D-isomers via de- or re- protonation of the α-carbon atom on the bound acyl intermediate (Linne et al, 2001).
While E-domains are capable of producing a mixture of D- and L-isomers, there is evidence that C-domains immediately down-stream of epimerases are D-specific
(Stachelhaus and Walsh, 2000). Hence, only D-isomers may be incorporated by NRPS modules possessing E-domains.
In the gramicidin biosynthesis cluster, six functional epimerisation domains are responsible for the production of the D-Leu and D-Val residues found in the mature cyclic peptide (Kessler et al, 2003). Conversely, the microcystin biosynthesis gene cluster contains a single E-domain responsible for the conversion of L-Ala to D-Ala during peptide synthesis (Tillett et al, 2000).
The conversion of L-amino acids to D-isomers is also achieved by stand-alone enzymes, alternatively referred to as racemases. While racemases feature commonly in primary metabolic processes such as peptidoglycan biosynthesis (Walsh, 1989), they
23 have been associated with a few NRP pathways. A notable example includes the cyclosporin biosynthesis pathway, where the D-Ala monomer is provided to the first A- domain by a dedicated alanine racemase (Hoffmann et al, 1994). Similarly, an aspartate racemase in the microcystin biosynthetic pathway presents D-Asp to the second module on the McyB peptide synthetase (Sielaff et al, 2003).
Methyltransferases
A variety of non-ribosomally produced compounds contain methylated amino acid residues. For example, the coumarin antibiotic coumermycin-A, contains at least eight methyl groups (Li et al, 2002), while seven of the eleven amino acids in cyclosporin are methylated (Weber et al, 1994). Methylation can greatly alter the structure and biological activity of secondary metabolites. For example, O- and N-methylations remove hydrogen bond donor sites and increase molecular lipophilicity, while C- methylations can markedly influence stereochemistry and introduce chiral centres (Rix et al, 2002). Methylation can also make NRPs and mixed PKs less susceptible to proteolytic breakdown.
Methylation is important to the toxicity of the cyanobacterial hepatotoxins microcystin and nodularin. Previous studies have shown that natural microcystin variants lacking the
Adda O-methylation have reduced inhibition of protein phosphatases (Namikoshi et al,
1992). Furthermore, the recent disruption of the mcyJ O-methyltransferase in P. aghardii resulted in the production of a de-methylated microcystin variant with similarly reduced toxicity (Christiansen et al, 2003).
24 N-, C- and O- methyltransferases are cofactor dependent, meaning they require the presence of a methyl donor group for catalysis. The requisite cofactor usually takes the form of the highly reactive sulfonium ion, S-adenosyl methionine (SAM). Methylation reactions involving SAM yield S-adenosylhomocysteine in addition to the methylated acceptor (Voet and Voet, 1995). Most methyltransferases are inhibited by the SAM analogue, sinefungin (Ikeda and Omura, 1995). This nucleoside antibiotic has been used to probe the functions of methyltransferases from several secondary metabolite pathways including, cyclosporin, oleandomycin, and avermectin B1a pathways (Velkov and Lawen, 2003; Huh et al, 2004).
Methyltransferases associated with secondary metabolite pathways are often embedded within the modules of PKSs and NRPSs. However, several examples of stand-alone methyltransferases have also been reported in the literature. While methylated secondary metabolite pathways typically utilise only 1-2 methyltransferase enzymes, several NRP and PK systems are known to incorporate several of these tailoring enzymes. Furthermore, N-, C-, and O- methyltransferases may exist in any given combination within a gene cluster.
N-methyltransferases are often inserted within the A domains of NRP modules. Such cis-acting methyltransferases incorporate methyl groups during the peptide chain elongation process. During N-methylation, the S-methyl group of SAM is transferred to the α-nitrogen of the thioesterified amino acid, releasing S-adenosyl-L-homocysteine as a reaction product (Lawen, and Zocher, 1990). In cyclosporin biosynthesis, seven N- methylation reactions occur. The N-methyltransferases involved in this biosynthetic process form integral domains within the cyclosporin synthetase (Lewin and Zocher,
25 1990). Interestingly, only un-methylated precursor amino acids are accepted by cyclosporin synthase. These amino acids are methylated while bound to the enzyme as thioesters (Zocher et al, 1986).
NRP and PK associated N-methyltransferases are occasionally encoded by individual
ORFs, adjacent or distal to the primary biosynthesis genes. Such stand-alone N- methyltransferases methylate amino acid residues that have already been incorporated into the NRP pre-peptide. For example, the N-methyltransferase encoded by ORF 16 within the chloroeremomycin gene cluster, acts in trans to convert the D-Leu residue of the heptapeptide aglycone1 into the N-Me-D-Leu residue of the mature antibiotic
(O’Brien et al, 2000).
While C-methyltransferases are much less common than N-methyltransferases, they have been identified in several NRP and PK pathways. During C-methylation, the S- methyl group of SAM is transferred to a C atom on a carbon-containing functional group such as an olefin2 or ring carbon. In the case of the polyketide anti-tumour drug mithramycin, the methylation at aromatic C-7 has been shown to be essential for biological activity (Rodriguez et al, 2004). C-methylations are also critical to the biosynthesis of many fungal metabolites including, lovastatin and fusarin (Hendrickson et al, 1999; Song et al, 2004). In these examples, the C-methyltransferases are embedded within NRP or PKS modules. C-methyltransferase PKS domains in fungi share a high degree of sequence similarity (>40%), and have been used in the design of oligonucleotide primers to locate the biosynthesis gene clusters of several PKs,
1The non-sugar compound remaining after replacement of the glycosyl group from a glycoside by a hydrogen atom. 2 Also known as alkenes, olefins are unsaturated hydrocarbons that have a double bond between at least two carbon atoms. 26 including the fusarin and squalestatin S1 synthase clusters (Song et al, 2004; Cox et al,
2004).
O-methyltransferases are often stand-alone enzymes that act in the final stages of NRPS and PKS pathways. Examples include those found in the saframycin and daunorubicin biosynthesis pathways. PKS-embedded O-methyltransferase domains have also been described. These usually occur between AT and KR, or AT and ACP domains, as seen in myxothiazol and soraphen-A synthases (Silakowski et al, 1999; Ligon et al, 2002), although other fusions are also possible. Tetracenomycin synthase contains three O- methyltransferases. Two stand-alone enzymes, TcmO and TcmP, methylate the C-8 and
C-9 hydroxyl groups of the prepeptide, respectively, while a third embedded O- methyltransferase TcmN, is responsible for the C-3 methylation. Interestingly, the
TcmN O-methyltransferase is fused to a cyclase/dehydratase enzyme (Summers et al,
1992; Decker et al, 1993).
Glycosyltransferases
Glycosylation is a common feature of many non-ribosomally-produced substances. The addition of glycosyl residues to the aglycones of PKs can dramatically influence their structural and stereoelectric properties. Furthermore, glycosylation is critical to the biological activity of many clinically important drugs, such as the glycopeptide and macrolide antibiotics (Losey et al, 2001).
The pattern of glycosylation within the teicoplanin glycopeptides is central to their antimicrobial activity (Boneca et al, 2003). The recent discovery of the teicoplanin biosynthesis gene cluster in Actinoplanes teichomyceticus, has revealed that two
27 glycosyltransferases, encoded by orf1 and orf10, catalyse the transfer of UDP-(N- acetyl)-glucosamine onto 3-chloro-β -hydroxytyrosine-6 and 4-hydroxyphenylglycine-4 of the teicoplanin heptapeptide, respectively (Li et al, 2004). The utilisation of UDP- activated sugars is common among glycosyltransferases. The donor substrates of such post PKS tailoring enzymes may alternatively take the form of TDP- or GDP-sugar co- substrates (Rix et al, 2002).
The orf10-encoded teicoplanin glycosyltransferase is able to utilise an alternative unacetylated substrate, UDP-glucosamine (Li et al, 2004). Such substrate flexibility is wide-spread among these tailoring enzymes. In addition, glycosyltransferases may tolerate several different acceptor groups (usually alcohols or phenols). These enzymes are therefore, excellent candidates for the rational design of compounds with novel and improved bioactivities.
The glycosyltransferase superfamily is incredibly diverse. However, depending on the configuration of the anomeric3 functional group of the glycosyl donor molecule, and of the resulting glyco-conjugate, all known glycosyltransferases can be divided into two major types: retaining glycosyltransferases and inverting glycosyltransferases.
Retaining glycosyltransferases transfer sugar residues while preserving anomeric configuration, while inverting glycosyltransferases transfer sugar residues with the inversion of anomeric configuration (Kapitonov and Yu, 1999). The majority of glycosyltransferases involved in secondary metabolite pathways fall into the latter category.
3 An anomeric carbon is the new stereocenter created in forming the cyclic structure of a monosaccharide. 28 Glycosyltransferase-mediated reactions are thought to proceed through an oxocarbonium-ion-like transition state. In the case of the inverting glycosyltransferases, the catalytic mechanism is putatively initiated by general base activation of the acceptor hydroxyl group and subsequent nucleophilic attack at the anomeric sugar carbon
(Unligil and Rini, 2000).
In the final stages of vancomycin biosynthesis, two consecutive glycosyltransferases add sugar moieties to the pre-glycopeptide aglycone. GtfE initially transfers D-glucose to the phenolic hydroxyl of the OH-Phe-Gly side-chain to form a DVV intermediate. A second enzyme, GtfD, subsequently attaches L-vancosamine to the glucose O2’ hydroxyl oxygen, generating the disaccharide vancomycin chain (Mulichak et al, 2004).
The identity and position of the sugar constituents in compounds such as vancomycin, can increase overall solubility and mediate important binding interactions between such drugs and their targets (Mulichak et al, 2004). Hence, glycosylation may prove to be an invaluable tool for rationally designing the next generation of antibiotics.
ABC Transporters
ABC transporters represent one of the largest, most highly conserved protein superfamilies in existence (Saurin et al, 1999). These proteins, found in bacteria, eukaryotes and archaea (Jovell et al, 1996), are responsible for the ATP-dependent transport of a vast range of solutes (allocrites) across intracellular and cell surface biological membranes (Jones and George, 1999).
ABC transporter genes have been associated with many NRP and PK biosynthesis gene clusters including amphotericin-B (Caffrey et al, 2001), rapamycin (Schwecke et al,
29 1995), bleomycin (Shen et al, 2001), frenolicin (Bibb et al, 1994), bacitracin
(Neumuller et al, 2001), yersiniabactin (Rakin et al, 1996), and exochelin (Zhu et al,
1998). However, few of these transporters have been functionally characterised. The cyclic dipeptide fungal toxins sirodesmin and gliotoxin, and the PKS fungal toxin fumonisin, each have ABC transporters encoded within their biosynthesis gene clusters.
These transporters have been shown to confer self-resistance to the producing organisms (Gardiner et al, 2004; Proctor et al, 2003). Interestingly, disruption of the sirodesmin, gliotoxin, and fumonisin transporters did not abolish the production, or export, of their corresponding allocrites. These results suggested that alternative
(distally encoded) transporters may exist for these toxins. In contrast, disruption of the
ABC transporter encoded within the exochelin biosynthesis gene cluster resulted in abolished biosynthesis of the siderophore (Zhu et al, 1998). It is thus believed that the export of exochelin occurs via a sole, specific transporter, and that biosynthesis and export of the siderophore are tightly coupled.
While most NRP/PK gene clusters encode a single ABC transporter, several examples of multiple transporters have been reported in the literature. For example, the biosynthesis gene clusters for amphotericin B and nystatin contain two heterologous
ABC transporters, AmphG and AmphH, in the case of amphotericin synthase, and
NysG and NysH, for nystatin synthase. While bacterial ABC transporters usually function as homodimers, it is hypothesised that AmphG/H and NisG/H associate as heterodimers to provide an export pathway for their respective antifungal macrolides
(Caffrey et al, 2001; Fjaervik and Zotchev, 2005).
30 While few NRP/PK-associated ABC transporters have been structurally characterised, the superfamily shares many common features. ABC transporters are minimally comprised of a highly conserved ABC-ATPase and at least one cognate, but much less conserved, membrane domain. These may be considered the energy generator and transport pathway, respectively (Holland and Blight, 1999). Most ABC-ATPases have an approximate molecular mass of 27 KDa and share an overall amino acid sequence identity in excess of 30 percent (Holland and Blight, 1999). This identity is concentrated in several key regions making ABC-ATPases readily recognisable by at least two highly conserved motifs unique to this family. The Walker A site and the hydrophobic Walker B site are two such motifs, conserved presumably because of their functional role in nucleotide (ATP) binding (Walker et al, 1982). The Walker B site incorporates a highly distinctive signature motif [DE(A/P)TSALD or similar], which in most transporters does not tolerate amino acid changes with respect to ATPase activity
(Schneider and Hunke, 1998).
Immediately upstream of the Walker B site lies a highly conserved sequence motif
(beginning LSGGQ) called the linker peptide. As most mutations in this region abolish
ATP hydrolysis but have little or no effect on nucleotide binding, the linker peptide is thought to play a role in coupling ATP hydrolysis to substrate translocation (Ames and
Lecar, 1992; Schneider and Hunke, 1998). A second slightly less conserved sequence designated the "switch", is located C-terminal to the Walker B motif. Sequence homology and mutational analyses of this region has provided evidence that it plays a key role in propagating conformational changes triggered by ATP hydrolysis (Schneider and Hunke, 1998). Compared to their cognate ATPase domains, the membrane domains of ABC transporters display very little or no primary sequence homology. Whilst the
31 prokaryotic import systems do possess a short conserved "EAA" motif (EAA-G-I-LP) in their cytoplasmic loop (Dassa and Hofnung, 1985), the exporters display no significant sequence conservation (Quentin et al, 1999).
A specific feature of ABC systems is that the direction of transport may be outward
(export or secretion) or inward (import) with respect to the cytoplasm. The import systems found in both prokaryotes and eukaryotes, are distinguishable from the exporters in a number of ways. Firstly, the ABC proteins of importers are encoded separately from their membrane domains, which are frequently composed of two different polypeptides. Secondly, the molecules to be imported normally require a dedicated binding protein to deliver them to the membrane transport domains (Holland and Blight, 1999).
The most distinguishing feature of the ABC transporter phylogenetic tree is its division into two major branches representing the export (ABC-A) and import (ABC-B) systems
(Saurin et al, 1999) (Figure 1.1). Further phylogenetic analysis reveals a tendency for transporters in both import and export subdivisions to cluster according to allocrite specificity. Thus by simply knowing the primary structure (peptide sequence) of a newly discovered ABC protein it is possible to predict the type of allocrite specified by the system, and the direction of allocrite transport.
ABC transporters of modified cyclic peptides such as pyoverdine and syringomycin, form a distinct branch within the ABC-A1 exporter group. Transporters of other structurally related peptides such as lantibiotics (nisin and subtilin), bacterocins
(lactococcin and pediocin PA-1), and various antibiotics also cluster on the A1 branch.
32 Interestingly, peroxisomal membrane proteins (PMPs) also cluster on the A1 branch.
While this phylogenetic arrangement may seem unusual, the long-chain fatty acid substrates of PMPs, are structurally similar to polyketides, and are produced in an analogous fashion by modular synthase enzymes. While these preliminary results suggest that transporters of NRPs, PKs, and structurally similar compounds are evolutionarily and functionally related, future characterisation of secondary metabolite transporters will hopefully shed further light on this area.
1.2.5 Biosynthesis of microcystin and nodularin
Microcystin and nodularin are synthesised non-ribosomally by the thiotemplate functions of large multifunctional enzyme complexes containing both NRPS and PKS domains. The gene clusters encoding these biosynthetic enzymes, mcyS (microcystin) and ndaS (nodularin), have recently been sequenced and partially characterised in several cyanobacterial species including Microcystis, Anabaena, Planktothrix and
Nodularia (Tillett et al, 2000; Christiansen et al, 2003; Moffitt et al, 2004; Rouhiainen et al, 2004) (Figure 1.2). Such fundamental studies have offered insight into the evolution of cyanotoxin biosynthesis, and have additionally provided much of the ground-work for current PCR-based cyanobacterial detection methods.
The microcystin biosynthesis gene cluster, mcyS, was the first complex metabolite gene cluster, to be fully sequenced from a cyanobacterium. In M. aeruginosa PCC7806, the mcyS gene cluster spans 55 kb and comprises 10 genes arranged in two divergently transcribed operons, mcyA-C and mcyD-J. The larger of the two operons, mcyD-J, encodes a modular PKS (McyD), two hybrid enzymes comprising NRPS and PKS modules (McyE and McyG), and enzymes putatively involved in the tailoring (McyJ, F,
33 and I) and transport (McyH) of the toxin. The smaller operon, mcyA-C encodes three
NRPSs (McyA-C) (Tillett et al, 2000).
The formation of Adda putatively involves enzymes encoded by mcyD-G and J, based on bioinformatic analyses and homology to related enzymes. The hybrid NRPS/PKS enzyme, McyG, constitutes the first step in Adda biosynthesis. It was initially hypothesised that the NRPS module of McyG activates phenylacetate, however, recent biochemical characterisation of the McyG A–PCP didomain has revealed that assorted phenylpropanoids are preferentially activated and loaded onto the PCP (Hicks et al,
2006). Following activation, the phenylpropanoid starter unit is extended by several malonyl-CoA elongation steps and subsequently modified by C-methylation, reduction and dehydration, all catalysed by the PKS modules of McyD, E and G. The aminotransferase domain of McyE then converts the polyketide to a β-amino acid in the final step of Adda biosynthesis (Figure S1.3). The NRPS module of the second hybrid
PKS/NRPS enzyme, McyE, is thought to be involved in the activation and condensation of D-Glu with Adda.
The mcyF ORF was originally predicted to encode a glutamate racemase, responsible for the epimerisation of the L-Glu residue of microcystin (Tillett et al, 2000; Nishizawa et al, 2001). A subsequent study by Sielaf et al (2003), contended this theory and offers evidence that McyF acts exclusively as an Asp racemase. The authors propose that the
D-Glu residue is provided by an L-Glu racemase residing outside the mcyS gene cluster.
Mutagenesis experiments in P. agardhii showed that the production of Adda also involves an O-methylation step catalysed by the putative monofunctional tailoring enzyme, McyJ (Christiansen et al, 2003).
34 The remaining biosynthetic enzymes in the microcystin biosynthesis pathway (NRPSs) are putatively involved in the specific activation, modification and condensation of substrate amino acids onto the linear peptide chain, which is then cyclised to produce microcystin. Firstly, McyA adds L-Ser to the growing chain, followed by the addition of
D-Ala. This step is followed by the addition of L-Leu and D-MeAsp residues (McyB) followed by the addition of L-Arg (McyC), and subsequent cyclisation and release of the final peptide product (Figure S1.4).
An additional stand-alone tailoring enzyme, McyI, is hypothesised to play a role in the modification of microcystin. McyI shows greatest homology to a group of D-3- phosphoglycerate dehydrogenase (D-3-PGDH) enzymes from various archaeal species
(Tillett et al, 2000). In E. coli, D-3-PGDH enzymes are responsible for the first step in the pathway for serine biosynthesis (Sugimoto and Pizer, 1968). The role of this enzyme in microcystin synthesis is unknown, however the MeDha residue in microcystin is produced from L-Ser, therefore McyI may play a role in the production of L-Ser or conversion of L-Ser to MeDha (Tillett et al, 2000).
An ABC transporter-like protein McyH, is proposed to be involved in the transport of microcystin. This transporter may also be responsible for the thylakoid localisation of the toxin (Shi et al, 1995), or for the extrusion of the toxin under certain high-light growth conditions (Kaebernick et al, 2000).
Comparative studies of the mcyS gene clusters from M. aeruginosa, P. agardhii
(Christiansen et al, 2003), and Anabaena sp. (Rouhanien et al, 2004) have noted
35 variation in the arrangement of mcyS genes between these different species of cyanobacteria, although the proposed toxin biosynthetic processes are thought to be similar. The M. aeruginosa and Anabaena sp. mcyS clusters are both arranged into two divergently transcribed operons, however, the arrangement of genes within these operons differs between the two species. In P. agardhii, the mcyS cluster also has a distinctive arrangement and lacks mcyF and mcyI. Furthermore, the P. agardhii mcyS cluster contains an additional gene mcyT, upstream of the central promoter region. This gene is thought to encode a putative type II thioesterase enzyme, which may play an editing role by removing mis-primed amino acids from the NRPS and PKS enzymes.
The characterisation of mcyS in M. aeruginosa, P. agardhii and Anabaena sp. has important implications for understanding the origins and evolution of hepatotoxin biosynthesis in cyanobacteria.
The nodularin biosynthesis gene cluster ndaS, from Nodularia spumigena was also recently sequenced and characterised (Moffitt et al, 2004). The 48 kb cluster consists of nine ORFs (ndaA-I) transcribed from a bidirectional regulatory promoter region. While most of the ndaS encoded genes have homologues in the mcyS cluster, their arrangement adheres more closely to the ‘co-linearity’ rule, which suggests that the order of catalytic processes involved in the biosynthesis of a non-ribosomal metabolite is generally the same as the order of the genes which encode these catalytic enzymes
(Kleinkauf and von Döhren , 1996).
The proposed pathway for nodularin biosynthesis is similar to that proposed for microcystin. Functional assignment of the enzymes was based on bioinformatic analysis and homology to the microcystin synthetase enzymes. The Adda side-chain is produced
36 via a mixed NRP/PKS pathway from a phenylacetate starter unit and several malonyl-
CoA extensions (NdaC, D and F) (Figure S1.5). The NRPS module of the hybrid
NRP/PKS, NdaF, subsequently adds D-Glu to the growing chain. Two NRPS enzymes
NdaA and B complete the peptide by adding the final amino acid residues, L-Thr, D-
MeAsp and L-Arg (S1.6). The NRPS modules responsible for the activation of D-Ala and D-Leu in mcyS (McyA and B) are absent from ndaS since nodularin lacks these amino acids.
The ndaS cluster also encodes several putative mono-functional enzymes that may play a role in the modification and transport of nodularin. Putatively, ndaE encodes an O- methyltransferrase, ndaG encodes an L-Asp/L-Glu racemase, and ndaI encodes an ABC transporter. Also encoded within the ndaS cluster is a D-3-PGDH homologue, NdaH, that shares 71% identity with McyI. It has been hypothesised that NdaH may catalyse the conversion of methyl-threonine to methyl-dehydrobutyrine in the final peptide structure.
1.2.6 Regulation of microcystin biosynthesis
Hepatotoxin production in cyanobacteria is thought to be influenced by a number of different physical and environmental parameters, including nitrogen, phosphorous, trace metals, growth temperature, light, and pH (Sivonen, 1990; Lukac and Aegerter, 1993; van der Westhuizen and Eloff, 1985; Song et al, 1998). However, due to the fact that most regulatory investigations have not been standardised, and the data have not been interpreted against the same specific growth controls, the subject of hepatotoxin regulation remains a somewhat contentious issue.
37 Several batch culture experiments have suggested that increased microcystin production in cyanobacteria is correlated with high levels of nitrogen and phosphorus (Sivonen et al, 1990; Vezie et al, 2002). Conversely, low iron concentrations have been correlated with increased toxin production (Lukac and Aegerter, 1993). While these results suggested that microcystin production is influenced by nutrients and trace metals, the observed toxin fluctuations were probably due to the indirect effects of nitrogen, phosphorous and iron on cell growth rate. Long et al (2001), observed that under nitrogen-limited culture conditions, fast growing M. aeruginosa cells are smaller, of lower mass and contain higher intracellular levels of toxin than slow-growing cells.
These results strongly suggested a positive linear relationship between the microcystin content of cells and their specific growth rate. The generalised model by Long et al for microcystin regulation, may also explain the conflicting results yielded from other batch culture investigations where variables such as temperature, light and pH have been tested.
While most toxin regulation studies have focused on direct measurements of cellular toxin, the description of the mcy gene cluster by Tillett et al (2000) enabled a closer examination of microcystin regulation at the molecular level. Kaebernick et al (2000) used the RNase protection assay to measure the transcription of mcyB and mcyD under a variety of different light conditions. High light intensities and red light were correlated with increased transcription, while blue light led to reduced transcript levels.
Interestingly, the authors observed two light thresholds, between dark and low light (0 and 16 µmol photons m-2 s-1), and medium and high light (31 and 68 µmol photons m-2 s-1), at which a significant increase in transcription occurred. The same group later found that transcription of mcy genes occurs via two polycistronic operons, mcyABC
38 and mcyDEFGHIJ, from a central bidirectional promoter between mcyA and mcyD
(Kaebernick et al, 2002). Interestingly, alternate transcriptional start sites were identified for both operons when cells were cultured under different light intensities.
For example, under low light conditions, the polyketide and tailoring genes mcyD-J, are transcribed as part of a polycistronic message (mcyDEFGHIJ) from a central (mcyD) promoter, while under high light conditions, the genes are transcribed from an alternative up-stream promoter. It is thought that initiation from the alternate promoters under high light conditions may lead to increased transcription, as previously observed for mcyB and mcyD. Many of the tailoring enzymes (mcyF, G, H, I and J) also possess their own individual promoters (Kaebernick et al, 2000).
Several studies have reported an increase in extracellular microcystin content following exposure of cultures to high light conditions (Rapala et al, 1997; Bottcher et al, 2001;
Kaebernick et al, 2000; Wiedner et al, 2003). Wiedner et al (2003), found that on average, extracellular microcystin concentrations were 20 times higher when cells were cultured at 40 µmol photons m-2 s-1 compared to those grown at 10 µmol photons m-2 s-1. However, it is important to note that the extracellular microcystin concentrations at both irradiances accounted for only 2.47 and 0.22% of the total microcystin content, respectively. Kaebernick et al (2001) proposed that microcystin may be constitutively produced under low and medium light intensities, and exported when a higher threshold intensity is reached. The recently identified ABC transporter McyH, may be responsible for this apparent export, although increased cell lysis and leakage of the toxin at higher irradiances has not been ruled out at this stage. Further investigation is warranted.
39 1.3 Aims This project aimed to investigate the stucture, function and regulation of three tailoring enzymes McyH, McyI and McyJ, proximally encoded within the microcystin biosynthesis gene cluster of Microcystis aeruginosa PCC7806. By characterising these enzymes, it was envisaged that a more complete understanding of the microcystin biosynthesis pathway could be gained, and that this knowledge in turn, could be applied to the fields of pharmaceutical design and production, as well as water quality management.
The specific goals of this project were to determine the distribution of McyH, McyI and
McyJ among toxic and non-toxic cyanobacteria, and to investigate their evolution within and beyond the cyanobacterial clade. Additionally, the project aimed to investigate the structure, function and regulation of McyH, McyI and McyJ using bioinformatic techniques combined with mutant studies, heterologous expression and enzymatic analysis.
40 CHAPTER 2.
MATERIALS AND METHODS
41 2.0 Bacterial strains and Culturing
2.0.0 Cyanobacteria
The microcystin-producing strain M. aeruginosa PCC7806, was kindly provided by J.
Weckesser (Freiburg University, Freiburg, Germany) and R. Rippka (Pasteur Institute,
Paris, France). Other cyanobacteria were obtained from The Institut für Biologie
(Humboldt University, Berlin), Microbiology Division, Biocenter (University of
Helsinki, Finland) and the cyanobacterial culture collection (University of New South
Wales, Sydney, Australia).
M. aeruginosa PCC7806 wild-type (WT) and mcyA-, mcyB- and ΔmcyH mutants were grown as batch cultures in BG11 medium (Fluka) and BG11 plus 3 µg/ml chloramphenicol, respectively. Other cyanobacteria were grown as batch cultures in
BG11 medium supplemented with antibiotics where indicated. Cyanobacteria were routinely grown under 16 µmol photons m-2 s-1 white light (unless otherwise specified).
Light intensities were measured using a LI-CORR LI-250 lightmeter (Walz). The optical density (OD750) of cultures was measured using an Ultrospec II (LKB,
Biochrom). For physiological experiments, cells were cultured at 28°C. For production of cyanobacterial biomass (for DNA extractions etc.) cells were cultured at room temperature.
2.0.1 Escherichia coli
E. coli cultures (Table 2.0) were used in cloning, heterologous expression and complementation experiments. Recombinant plasmids were amplified in DH5α,
TOP10F’ or XL1Blue. Heterologous expression of recombinant peptides was performed
42 in the Rosetta(DE3) or Rosetta(DE3)PlysS strains (Novagen). E. coli auxotrophs were kindly provided by D. Clark (Southern Illinois University), Jae-gu Pan (Genofocus,
Korea) and E. Juni (University of Michigan).
E. coli strains were grown in M9 salts (supplemented with the appropriate amino acid,
A1.3), Luria-Bertani (LB), tryptone-phospate (TP), or super broth (SB) medium
(appendix). Where appropriate, media were supplemented with 100 µg/ml ampicillin,
50 µg/ml kanamycin, 34 µg/ml chloramphenicol or 5 µg/ml gentamycin. Where blue/white selection of colonies was required, LB plates were supplemented with 0.5 mM IPTG and 80 µg/ml X-gal. Cultures were grown at 37°C unless otherwise specified. Liquid cultures were grown in conical flasks with constant shaking (150 rpm,
Ratek orbital shaker).
2.1 Nucleic Acid Extraction
2.1.0 Extraction of genomic DNA
DNA was isolated according to the xhanthogenate-SDS (XS) method of Tillett and
Neilan (2000). Briefly, 1 mL of mid to late logarithmic growth phase bacterial cell cultures were harvested by centrifugation and the cell pellets resuspended in 50 µl TER
[10 mM Tris-HCl, pH 7.4; 1 mM EDTA, pH 8; 100 µg/ml RNase A]. To each cell suspension (in a 1.5 mL microcentrifuge tube) was added 750 µl of freshly made XS buffer [1% potassium ethyl xhanthogenate (Fluka); 100 mM Tris-HCl, pH 7.4; 20 mM
EDTA, pH 8; 1% sodium dodecylsulfate; 800 mM ammonium acetate] and the tubes were inverted several times to mix. The tubes were incubated at 70°C for 120 min, vortexed for 10 s and placed on ice for 30 min. Precipitated cell debris was removed by
43 centrifugation at 12 000 g for 10 min and the supernatants carefully transferred to fresh tubes containing 750 µl of isopropanol. Samples were incubated at room temperature for 10 min and the precipitated DNA pelleted by centrifugation for 10 min at 12 000 g.
The DNA pellets were washed once with 70% ethanol, air dried and resuspended in 100
µl TE [10 mM Tris-HCl, pH 7.4; 1 mM EDTA, pH 8].
2.1.1 Extraction of plasmid DNA
Plasmid DNA was purified from E. coli using the Viogene Mini-M Plasmid DNA
Extraction System according to the manufacturers protocol. Typically, 1.5 mL-5 mL of overnight culture was used for extraction of high copy number plasmids.
For extraction of low copy number plasmids, cells were cultured in chloramphenicol prior to plasmid extraction to increase plasmid yield according to the method of
Sambrook et al (1989). Briefly, 600 µl of an overnight culture was inoculated into 10 mL LB (plus antibiotic) and grown at 37°C with shaking for 2.5 h. After addition of 1.7 mg chloramphenicol, cultures were incubated for an additional 16 h at 37°C. Cells were then harvested 5 000 g for 10 min and processed according to the Viogene Mini-M
Plasmid DNA Extraction System protocol.
2.2 Polymerase chain reaction (PCR)
DNA amplifications were performed in 25 µl reactions containing 1 × PCR buffer
(Fischer Biotech), 2.5 mM MgCl2, 130 µM each dNTP. PCR was performed using 1 µl of template DNA at a concentration of approximately 100 ng/µl or a freshly picked bacterial colony (in the case of colony-screening PCR). Oligonucleotide primers
44 (purchased from Sigma Genosys) were added to a final concentration of 0.4 µM for specific primers, or 0.75 to 1.5 µM for degenerate primers. Taq F1 DNA polymerase was diluted in TD buffer [20 mM Tris-HCl, pH 7.6; 30 mM KCl; 1 mM ß- mercarptoethanol; 50% v/v glycerol; 1 mg/ml BSA] to 1 unit/µl and used in the reaction at a final concentration of 0.01 unit/µl.
Thermal cycling was performed in a GeneAmp PCR System 2400 (Perkin Elmer
Corporation) and generally consisted of an initial denaturation step of 94°C for 2 min followed by 30 cycles of DNA denaturation at 94°C for 20 s, primer annealing at x°C
(where x = the lowest Tm of the two primers - 5°C) for 30 s and extension at 72°C for y s (where y = (no. base pairs) 0.06). A final extension step at 72°C for 5 min was also included. The following formula was used for estimating the melting temperature (Tm) of the primers:
Tm (°C) = 2(NA+ NT) + 4(NG+NC)
Where N equals the number of primer adenine (A), thymidine (T), guanidine (G), or cytosine (C) bases.
Where PCR fragments were to be used in cloning experiments, amplifications were performed using Pfu DNA polymerase followed by an A-tailing reaction with Taq DNA polymerase. The initial amplification reaction (25 µl) contained 1× Pfu buffer
(Fermentas), 2 mM MgSO4 130 µM each dNTP, approximately 100 ng/µl DNA template, 0.4 µM to 1.5 µM oligonucleotide primers and 0.6 units of Pfu DNA polymerase (added immediately before thermal cycling). Thermal cycling involved an
45 initial denaturation step of 94°C for 2 min followed by 25-30 cycles of DNA denaturation at 94°C for 20 sec, primer annealing at x°C for 30 s and extension at 72°C for z s (where z = (no. base pairs) 0.12). A final extension step at 72°C for 5 min was also included.
PCR products were column purified and 1-2 µl of the blunt PCR products were added to
10 µl reactions containing 1 µl 10 × Taq F1 reaction buffer, 1 µl 25 mM MgCl2, 0.2 mM dATP and 5 units Taq F1 DNA polymerase. The reactions were incubated at 70°C for 15-30 min, to add A-tails to the 3’ends of PCR fragments. O n e t o t w o microlitres of the reactions w ere used in ligation reactions with pGEM-T easy (Promega) or pDRIVE
(Qiagen) vectors according t o the manufacture’s recommended p r o c e d u r e s.
2.3 Purification of Nucleic Acids
2.3.0 Ethanol and isopropanol precipitation of DNA solutions DNA solutions were precipitated by addition of 1 volume 3 M ammonium acetate and either 2 volumes absolute ethanol, or 1 volume absolute isopropanol. Tubes were vortexed briefly and incubated at –20°C for at least 30 min, then centrifuged 12 000 g for 15 min. DNA pellets were washed with 0.2-1 mL 70% ethanol and centrifuged as above. Supernatants were removed and DNA pellets air dried for at least 15 min, then resuspended in an appropriate volume of TE or nuclease free water.
2.3.1 Column purification
DNA was purified from agarose gels and enzymatic reactions using the Wizard SV Gel and PCR Clean-Up System according to the manufacturers protocol (Promega).
46 2.4 Automated Sequencing and Analysis
Automated sequencing was performed using the PRISM Big Dye kit (Applied
Biosystems). Linear amplification of PCR sequencing fragments was performed in a 20
µl reaction containing 1 µl Big Dye, 3.2 pmol oligonucleotide primer, 3.5 µl 5 × sequence buffer (Applied Biosystems) and 20-50 ng purified PCR product. Thermal cycling was performed in a GeneAmp PCR System 2400 (Perkin Elmer Corporation).
An initial denaturation step at 96°C for 3 min was followed by 30 cycles of 96°C for 10 sec, 50°C for 5 s and 60°C for 4 min. Labelled PCR fragments were precipitated by the addition of 16 ul milli-Q and 64 µl absolute ethanol, incubated at room temperature for
15-30 min and centrifuged at 12 000 g for 20 min. The precipitated sequencing fragments were then washed with 200 µl of 70% ethanol and centrifuged as above. The supernatants were then aspirated and the pellets air dried for at least 15 min at room temperature.
Automated sequencing was performed by the Ramaciotti Centre (University of NSW).
Sequence data was analysed and edited using the Applied Biosystems Autoassembler software.
2.5 Agarose gel electrophoresis
DNA was electrophoresed on 10 mL agarose mini gels [0.8-2% agarose; 1 × TAE (40 mM Tris-HCl, pH 7.8; 5.71% glacial acetic acid; 1 mM EDTA, pH 8)], in a Gel
Electrophoresis Apparatus GNA-100 (Pharmacia) containing 1 × TAE at a current of 85 volts supplied by an LKB-GPS 200/400 power pack (Pharmacia). Prior to
47 electrophoresis, DNA solutions were mixed with DNA loading buffer [80% glycerol;
8% bromophenol blue-xylene cyanole dye (Sigma)]. One hundred and fifty nanograms of a molecular weight standard, GeneRuler DNA Ladder mix (Fermentas) was also loaded on each gel.
2.6 Modification of DNA fragments
2.6.0 Digestion of DNA with restriction endonucleases
Digestion of DNA with restriction endonucleases was generally performed in 50µl reactions containing 1 × endonuclease buffer (supplied with enzyme) and 5 units of restriction endonuclease per µg DNA. The reaction was incubated at 37°C for 1-16 h followed by a heat deactivation step at 65°C for 20 min. Restriction endonucleases were supplied by Roche, Fermentas, Promega and New England Biolabs.
2.6.1 End-filling DNA overhangs
DNA fragments with 5’or 3’ overhangs were made blunt by treatment with T7 DNA polymerase or the Klenow polymerase fragment (Fermentas). For treatment with the
Klenow fragment, reactions were performed in 20 µl containing 1 µg of digested DNA,
1× Klenow reaction buffer (Fermentas), 0.05 mM dNTP mix and 1-5 units Klenow fragment. The reaction was incubated at 37°C for 10 min then stopped by heating at
70°C for 10 min.
For treatment with T7 DNA polymerase, reactions were performed in 20 µl volumes, containing 1 µg of digested DNA, 1 × T7 reaction buffer (Fermentas), 0.4 mM dNTP
48 mix and 2 units T7 DNA polymerase. The reaction was incubated at 37°C for 5 min then stopped by heating at 70°C for 10 min.
2.6.2 Dephosphorylation of DNA with shrimp alkaline phosphatase (SAP)
Dephosphorylation of linear DNA fragments was performed in 25 µl reactions containing 1 × dephosphorylation buffer (or appropriate compatible restriction endonuclease buffer), 0.5-2 µg DNA and 1 unit of shrimp alkaline phosphatase
(Fermentas) per picomole of DNA 5'-termini. The reaction was incubated at 37°C for 30 min then stopped by heating at 65°C for 20 min. Molar conversions of DNA were calculated using the following formula:
1 µg of 1 000 bp DNA = 1.52 pmol
pmol DNA 5’-termini = pmol linear DNA (number of cuts × 2 + 2)
2.6.3 Ligation of DNA fragments
Ligation of DNA fragments was performed in 5-20 µl reactions containing 1 × ligation buffer (Promega) (or appropriate compatible restriction endonuclease buffer supplemented with 1 mM ATP), 1-2 units T4 DNA Ligase for “sticky” ends, or 5 units for blunt ends, 5% PEG 4000 solution (for blunt ends only), 25-300 ng vector DNA and x ng of foreign DNA to be inserted, as calculated by the following equation:
x = conc. vector DNA (ng) × insert size (Kb) × insert:vector
size vector (Kb)
49 The reaction was incubated for 2 h at room temperature or overnight at 4°C then stopped by heating to 65°C for 10 min. The mixture was then used directly in transformation experiments.
2.7 Genetic manipulation of bacteria
2.7.0 Preparation of competent E. coli cells and transformation
E. coli cells were made chemically competent as follows: Fifty millilitres of LB broth was inoculated from a glycerol stock and grown to an optical density (650 nm) of 0.5 at
37°C. Cells were incubated on ice for 15 min then harvested by centrifugation at
3 000 g for 10 min at 4°C. The cell pellet was gently resuspended in 16 mL of ice cold solution 1 [10 mM MES pH 6.2; 100 mM RbCl; 10 mM CaCl2; 50 mM MnCl2, pH 5.8] and incubated on ice for 15 min. Cells were harvested as above and resuspended in 1.6 mL of ice cold solution 2 [10 mM MOPS, pH 6.5; 75 mM CaCl2; 10 mM RbCl; 15% glycerol] and then incubated 15 min on ice. Cells were stored in 50 µl aliquots at
–80°C.
For transformation of plasmid DNA, cells were thawed on ice for approximately 10 min prior to the addition of 0.5-10 µl of cold DNA solution. Cells were incubated on ice a further 15-30 min, heat shocked at 42°C for 1.5 min, then placed on ice for 2 min. Five hundred microlitres of LB was added to the transformation mix which was subsequently incubated at 37°C with shaking for 30-60 min. Twenty to four hundred microlitres of the transformation mix was then spread onto LB plates supplemented with the appropriate antibiotic/s and incubated overnight at 37°C.
50 2.8 Total Protein Assays
2.8.0 Bio-Rad Dc protein assay
Total protein concentrations of samples were routinely measured using the Bio-Rad Dc protein assay (BIO-RAD), in microtitre plates, according to the manufacture’s protocol.
2.8.1 Folin protein assay
Total protein concentrations of samples with high lipid content (eg. vesicles and membranes) were measured using Folin-Ciocalteu reagent. Briefly, reagent A [2%
Na2CO3; 0.4% NaOH; 0.16% sodium tartrate; 1% SDS] was mixed with reagent B [4%
CuSO4.5H2O] at a ratio of 100:1 to create reagent C. BSA standards (0, 20, 40, 60, 80 and 100 µg/mL) were diluted in water from a 100 µg/mL stock solution. Vesicle aliquots of 3, 6 and 9 µL were made up to 1 mL with water.
Three mililitres of reagent C was then added to standards and samples, with mixing after each addition, and samples were incubated at room temperature for 20 min. Zero point three mililitres of diluted Folin's reagent (1:1 with water) was then added to each tube with vigorous mixing after each addition. Tubes were then incubated for 45 min at room temperature. Optical densities at 660 nm were then measured.
2.9 Protein Electrophoresis
Twenty to fifty micrograms of total, soluble and insoluble protein fractions (as determined by BioRad Dc assay) were electrophoresed on 10-12% polyacrylamide gels
51 and either stained with Coomassie brilliant blue solution, or used in immunoblotting experiments.
2.10 Ammonium sulphate precipitation of proteins
Proteins in solution were selectively precipitated by the addition of ammonium sulphate according to the protocol outlined in The Protein Guide, Tips and Techniques
(Promega, 1993). All procedures were performed at 0-4°C.
2.11 Immunodetection of proteins with Ni-NTA conjugates
Twenty to fifty micrograms of protein extract was electrophoresed on 12-15% SDS polyacrylamide gels using the Mini protean apparatus (BIO-RAD). Proteins were then blotted to polyvinylidine fluoride (PVDF) membranes immersed in Tris-glycine transfer buffer [25 mM Tris-Hcl, pH 8.3; 192 mM glycine] for 1h at 100 V using a Mini Trans-
Blot Module (BIO-RAD). Membranes were washed twice for 10 min with PBS buffer
[140 mM NaCl; 2.7 mM KCl; 8 mM Na2HPO4; 18 mM KH2PO4, pH 7.4] and incubated for 1-16 h at room temperature, with gentle shaking, in PBS containing 5% skim milk powder. Membranes were washed 3 times for 10 min with PBS-Tween buffer [140 mM
NaCl; 2.7 mM KCl; 8 mM Na2HPO4; 18 mM KH2PO4, pH 7.4; 0.05% Tween 20] and incubated for 1h at room temperature in TBS-Tween containing 1/1 000 dilution of Ni-
NTA alkaline phosphatase (AP) conjugate stock solution (Qiagen). Membranes were washed 3 times for 10 min with PBS-Tween buffer and stained with 5-bromo-4-chloro-
3-indolyl phosphate/nitro blue tetrazolium (BCIP/NBT) liquid substrate (Sigma). The staining reaction was stopped by rinsing once in milli-Q water, incubating in 3% trichloracetic acid (TCA) for 5 min and washing again in milli-Q.
52 CHAPTER 3.
CHARACTERISATION OF THE ABC
TRANSPORTER, MCYH
53 3.0 OVERVIEW
The seventh open reading frame in the mcyD-J operon of the microcystin biosynthesis gene cluster of Microcystis aeruginosa PCC7806 encodes an 1 617 bp gene, mcyH
(complement 3056-4672 of [AF183408]). Preliminary sequence analysis of the inferred primary peptide sequence of mcyH by Tillett et al (2000), revealed a 38% identity (58% similarity) to an ATP binding cassette (ABC) transporter isolated from Synechocystis sp. PCC6803. While a biological function was unable to be assigned to McyH, it was speculated that the putative transporter might play a role in the thylakoid localisation of microcystin previously observed in M. aeruginosa, or alternatively, in the export of the toxin from the cell.
Release of microcystin into the extracellular environment, has in the past, been attributed to cell ageing and lysis (Sivonen, 1990; Lehtimaeki, 1994). However, subsequent cellular and genetic investigations have since provided evidence in support of a microcystin export pathway. For example, Shi et al (1995) used in situ hybridisation and electron microscopy to demonstrate the presence of microcystin beyond the cytoplasmic membrane in the wall and sheath area of intact M. aeruginosa cells. Furthermore, research on the influence of light on toxin production revealed that extracellular microcystin levels increase under high light culture conditions (Kaebernick et al, 2000; Wiedner et al, 2003). While the reported extracellular toxin levels were generally quite low (~0.22-2.47% total microcystin content), it has been suggested that microcystin is constitutively produced under low and medium light intensities, and is exported when a high light threshold is reached (Kaebernick et al, 2000). The recent discovery of mcyH, associated with the mcyS gene cluster, has led to the hypothesis that
54 this ABC transporter is responsible for the active export of microcystin (Tillett et al,
2000).
Comparative sequence analyses have identified mcyH homologues as ORFs in the mcyS gene clusters of Planktothrix agardhii CYA126 (Christiansen et al, 2003) and
Anabaena sp. 90 (Rouhiainen et al, 2004), and in the nodularin synthetase (ndaS) gene cluster of Nodularia spumigena NSOR10 (Moffitt and Neilan, 2004). The fact that every hepatotoxin gene cluster thus far sequenced encodes a putative ABC transporter, suggests that these enzymes (McyH and NdaI) are critical for hepatotoxin production, and/or the physiology of hepatotoxin-producing strains.
While mcyH is the first ABC transporter-encoding gene to be described in
M. aeruginosa, genome sequencing efforts suggest that nearly fifty percent of all cellular transporters in the cyanobacterium Synechocystis sp. belong to the ABC superfamily (Paulsen et al, 1998). Previously characterised cyanobacterial ABC transporters have been implicated in the transport of an extensive range of allocrites
(molecules transported) including manganese (Bartsevich and Pakrasi, 1995), nitrate
(Omata, 1995), osmoprotectants (Hagemann et al, 1997) and glycolipids (Fiedler et al,
1998).
ABC transporters are commonly found in association with secondary metabolite gene clusters. While several such associations have been identified in cyanobacteria, for example, microcystin (Tillett et al, 2000), nodularin (Moffitt and Neilan, 2004), anabaenopeptilide (Rouhiainen, 2000), nostopeptolide (Hoffmann et al, 2003) and microginin (D. Kramer, pers. comm.), the vast majorities have been discovered in
55 Gram-positive bacteria, such as the streptomycetes, mycobacteria and bacilli. These
ABC transporters are generally thought to confer allocrite resistance to the producing organisms.
This chapter describes the characterisation of the putative microcystin exporter McyH, using a combination of bioinformatic, molecular and biochemical approaches.
Extensive sequence analyses are reported, including phylogenetic and structural considerations. The distribution of mcyH orthologues in different species of cyanobacteria has been investigated, and the complete McyH transporter and several truncated forms have been over-expressed, purified and enzymatically analysed. An
McyH antibody has been engineered and used to investigate the regulation of McyH in wild-type (WT) M. aeruginosa, as well as in various non-toxic engineered mutant strains. Finally, everted membrane vesicles have been used to investigate the McyH- dependent transport of microcystin. The roles of McyH in microcystin production and export are discussed, as is the potential for this research to contribute to the innovation of current water quality protocols.
56 3.1 EXPERIMENTAL DESIGN AND METHODOLOGY
3.1.0 Screening various strains of cyanobacteria for mcyH orthologues
Degenerate oligonucleotide primers mcyHscreenF [5’-
ATTTAGTYGATGTTATTATTC-3'] and mcyHscreenR [5’-
ACCATTAGCAAACATACTGC-3'] were designed to amplify by PCR, an approximately 700 bp sequence within the mcyH ORFs of M. aeruginosa PCC7806, N. spumigena NSOR10 and Anabaena sp. 90. These primers were used to screen chromosomal DNA samples from several strains of toxic and non-toxic cyanobacteria for mcyH orthologues (Table 3.0). The same DNA samples were also screened with the
M. aeruginosa PCC7806 mcyH-specific primers, mcyH(200)F [5’-
TATTCTGGATTGGTATGAATGGCTA-3’] and mcyH(1222)R [5’-
GTAACTGTTCTCTGAGAGTTCCT-3’].
3.1.1 Sequence analysis
Unless otherwise stated, all sequence analyses were performed on the extended mcyH
ORF incorporating an alternative start site 141 bases upstream of that of the published
Genbank sequence [AAF00956].
The ~1 800 bp nucleotide sequence of mcyH was analysed using several different computer programs. The primary amino acid sequence of the corresponding McyH peptide was deduced using Translate (ExPASy), while theoretical physico-chemical parameters (amino-acid compositions, molecular weight, pI, etc.) were determined using ProtParam (ExPASy). Determination of potential membrane-spanning motifs and hydrophobic moments within McyH, was achieved using DAS analysis (ExPASy). The
57 probability of signal peptides within the McyH sequence and the putative sub-cellular localisation of McyH, were determined using SignalP and PSORTb (ExPASy), respectively. Codon usage within the mcyH sequence was assessed with the Kazusa countcodon program and the Rare Codon Caltor (University of California, LA). The percentage similarity and identity scores of McyH and other peptide sequences were determined using the PSI-Blast program [National Centre for Biotechnology
Information (NCBI)]. Conserved domains within McyH were detected using the Blast
CD-search (NCBI), ScanProsite (ExPASy), and FUGUE (Shi et al, 2001).
3.1.2 Phylogenetic analysis
A tri-iterated PSI-Blast search with the McyH sequence, returned ~50 sequences with a score of greater than 250 (Table 3.2). These sequences were subsequently used to generate a multiple sequence alignment (Appendix A2) and corresponding phylogenetic tree (ClustalX). As the closest characterised relatives of McyH, by PSI-Blast analysis, were the peroxisomal membrane proteins from Mus musculus [AAA39958], Gallus gallus [NP_001012615] and Homo sapiens [NP_002849], these reference sequences were also used in the phylogenetic analysis. The E. coli maltose importer MalK
[P02914], was used as an artificial out-group. Phylogenetic trees were generated using the neighbour-joining method (Saitou and Nei, 1987) with gaps removed. Trees were displayed graphically using njplot and AppleWorks6 (Drawing).
A second phylogenetic analysis was performed as above, based on the ATPase domains
(encompassing approximately 30 amino acids upstream of the Walker A site, to the
Walker B site) of McyH and its homologues (Table 3.3, Appendix A3).
58 3.1.3 Heterologous expression and purification of McyH and component peptides
Engineering the McyH expression constructs for the expression of full-length McyH
For heterologous expression of the full-length McyH ABC transporter (that is, the membrane domain plus ATPase) in E. coli, three constructs were engineered: pET30(mcyH), pET32(mcyH), and pET43(mcyH) (Figure 3.0). These constructs incorporated different fusion tags to the N- or C-termini of the McyH peptides. A fourth construct pET30(mcyHL), was also engineered. This construct incorporated an alternative ATG start site 141 bases upstream of that of the published Genbank sequence [AAF00956]. In all cases, the final expression constructs were sequenced with the vector-binding primers T7prom [5’-TAATACGACTCACTATAGGG-3’] and
T7term [5’-GCTAGTTATTGCTCAGCGGT-3’] to verify the encoded peptides were in-frame and free of mutations.
For construction of pET30(mcyH) and pET43(mcyH), mcyH was PCR amplified with the primers abcSF [5’-CAAACTCCATTTTTTCAACA-3’] and abcSR [5’-
GGTGAAGTAGTAGTCATCGT-3’] and cloned into the pGEM-T easy vector
(Promega). The mcyH-containing fragment was then excised with EcoRI and subcloned into the pBluescript vector (Fermentas). After sequence verification, the mcyH- containing fragment was excised with NotI and XhoI and subcloned into the pET30b and pET43b expression vectors (Novagen) to give rise to pET30(mcyH) and pET43(mcyH), respectively (Figures 3.0A and B). The pET30 and pET43 vectors are designed for the cloning and high-level expression of polypeptide sequences fused with
N- and/or C-terminal hexahistidine tags. The pET43 vector additionally encodes a 495 amino acid Nus•Tag™ to enhance the solubility of recombinant proteins.
59 For construction of pET32(mcyH), the mcyH gene was amplified with the primers mcyH(EcoRI)F [5’-TAGGGAATTCATGCTGTTTCTCCTCCTGTT-3’] and mcyH(XhoI)R [5’-CGCTCGAGACAGAGCTTTTGAGGCTCGATAATC-3’] which incorporated restriction sites (underlined) at either end of the PCR product. The resulting PCR product was then column purified, digested with EcoRI and XhoI and subcloned directly into pET32a (Novagen) (Figure 3.0C). The pET32a vector is designed for cloning and high-level expression of polypeptides fused with N- and/or C- terminal hexahistidine tags, plus a Trx•Tag which facilitates disulfide bond formation in the E. coli host cytoplasm, to help maximise the level of soluble, active and properly folded target protein.
For construction of pET30(mcyHL), the mcyH gene was amplified with the primers mcyHL(NdeI)F [5’-CGCGCCATATGCAAACTCCATTTTTTCAACAAG-3’] and mcyH(XhoI)R2 [5’-CGCTCGAGCAGAGCTTTTGAGGCTCGATAATC-3’] which incorporate restriction sites (underlined) at either end of the PCR product. The resulting
1 760 bp PCR product was then column purified, digested with NdeI and XhoI and cloned directly into pET30a (Novagen) (Figure 3.0D).
Engineering the McyH expression constructs for the expression of the individual membrane and ATPase domains of McyH
For expression of the individual membrane domain (MD) and ATP binding cassette
(ABC) domain of McyH, two constructs were engineered. These were designated pET30(md) and pET30(abc), respectively.
60 For construction of pET30(md), the putative membrane domain (MD) of mcyH
(complement 3789-4672 of [AF183408]) was amplified with the oligonucleotide primers abcSF [5’-CAAACTCCATTTTTTCAACA-3’] and Ebony5 [5’-
GACGCTCAACATAACTGGAA-3’] and cloned into pGEM-T easy. The MD encoding fragment was then excised with EcoRI and SpeI and subcloned into the pBluescript vector (Fermentas). Following sequence verification, the fragment was finally excised from pBluescript with NotI and XhoI and ligated into the pET30 expression vector (Figure 3.1A).
For construction of pET30(abc), the DNA fragment encoding the putative ATP binding cassette (ABC) domain of McyH (complement 3056-3860 of [AF183408]) from M. aeruginosa PCC7806, was amplified by PCR with the oligonucleotide primers abcMidF
[5´-GTTATGTTGAGCGTCTATCTG-3’] and abcSR [5´-
ACGATGACTACTACTTCACC-3’]. The resulting 804 bp PCR product was cloned into the pGEM-T easy vector then excised with EcoRI and SpeI and subcloned into pBluescript. After sequence verification, the putative ATPase-encoding fragment was excised from pBluescript with SacI and KpnI and cloned into the pET30b expression vector at the corresponding restriction sites giving rise to the final expression construct, pET30(abc) (Figure 3.1B).
Engineering expression constructs for the expression of N-terminally truncated
McyH peptides
Following analysis of the published McyH peptide sequence with the DAS membrane topology prediction program (Figure 3.3), three constructs were designed to sequentially delete the three predicted major hydrophobic trans-membrane regions of
61 McyH. These constructs were designated pET30(mcyHΔ15), pET30(mcyHΔ57) and pET30(mcyHΔ161) (Figure 3.1C). The nomenclature of the constructs was based on the number of amino acids deleted from the N-teminus of McyH.
For construction of pET30(mcyHΔ15), oligonucleotide primers ΔmcyH(15)F [5’-
TAGGGAATTCAATGCAGTTAATAGCTTCGT-3’] and mcyH(XhoI)R2 were used to amplify an 1 570 bp product corresponding to nucleotides 45-1 617 of the mcyH gene
(i.e omitting amino acids 1-15 in the McyH peptide). The PCR product was digested with EcoRI and XhoI, column purified and ligated directly into pET30 to give rise to pET30(mcyHΔ15).
For construction of pET30(mcyHΔ57), oligonucleotide primers ΔmcyH(57)F [5’-
TAGGGAATTCAATGCCTCTAAGTTTCTGACAA-3’] and mcyH(XhoI)R2 were used to amplify a 1 450 bp product corresponding to nucleotides 171-1 617 of the mcyH gene (i.e amino acids 1-57 in the McyH peptide). The PCR product was cloned directly into pET30 to give rise to pET30(mcyHΔ57).
For construction of pET30(mcyHΔ161), oligonucleotide primers ΔmcyH(161)F [5’-
TAGGGAATTCTTAAGTCGAGAATTAGATAAGA-3’] and mcyH(XhoI)R were used to amplify an 1 130 bp product corresponding to nucleotides 438-1 617 of the mcyH gene (i.e. amino acids 1-161 in the McyH peptide). The PCR product was cloned directly into pET30 (as above) to give rise to pET30(mcyHΔ161).
62 Over expression of recombinant peptides
Expression of recombinant McyH peptides was performed in the Rosetta(DE3)pLysS or
Rosetta(DE3) E. coli host strains (Novagen). Expression cultures were inoculated with a
1% volume of overnight starter culture and grown in LB, tryptone phosphate or superbroth media supplemented with kanamycin (50 µg/mL) and chloramphenicol (34
µg/mL). Cultures were grown with vigorous shaking (160 rpm) at 37°C to an OD650 of
0.6 and either retained at 37°C, or transferred to 30°C or 25°C and incubated for an additional 20 min (or until an OD650 of approximately 1.0 was reached). Expression of
McyH peptides was subsequently induced with 0.01-1 mM IPTG for 0-18 h at 25-37°C.
Cell pellets were harvested via centrifugation (5 000 g, for 10 min at 4°C), washed with
HEPES binding buffer [50 mM HEPES, 300 mM NaCl, pH 7.4], and stored at -80 °C until required.
Purification of recombinant peptides
Cell pellets were thawed on ice, resuspended in 2% of the original culture volume of cold HEPES binding buffer containing 1 mM PMSF and 0.1 % Triton X-100, passed through an 18 gauge needle several times, then briefly sonicated (Branson Sonifier 250, amplitude 25, 50% duty cycle, pulsed). The lysates were cleared via centrifugation
(20 000 g for 30 min at 4°C) and precipitated using 20-60 % ammonium sulphate (see section 2.10). The final pellets were resuspended in 0.4% of the original culture volume of cold HEPES binding buffer containing 20 mM imidazole, filtered through a 22 µm
Millex membrane (Millipore), and applied to 5 mL Hitrap columns (Amersham
Pharmacia Biotech) charged with Ni2+. Recombinant proteins were eluted over a linear gradient of 20-500 mM imidazole, collected in 0.5 mL fractions and analysed via SDS
63 PAGE (10-12% polyacrylamide gels, Coomassie staining) and western blotting (section
2.11).
Where bulk, denatured ABC protein was required for antibody generation, insoluble pellets from the initial lysate were resuspended in HEPES binding buffer containing
10 mM imidazole and 8 M urea. The sample was then briefly sonicated, centrifuged 5 min at 20 000 g, and filtered as described above. Purification was the same as for soluble proteins except all buffers contained 8 M urea.
Purified protein samples intended for enzyme assays, were desalted via size exclusion filtration using 10 KDa (Amicron Ultra) molecular weight cut-off columns and 50 mM
HEPES exchange buffer (pH 7).
3.1.4 Construction of everted membrane vesicles from E. coli expression strains
The method for constructing everted membrane vesicles from E. coli expression strains was adapted from Rosen and Tsuchiya (1979). One litre cultures of Rosetta(DE3) transformed with pET30(mcyHL) or pET30, were grown at 37°C to an OD600 ~ 0.6, induced with 1 mM IPTG for 2 h, then harvested at 5 000 g for 10 min at 4°C. The resulting pellets were washed with 100 mL of ice-cold wash buffer [50 mM Tris-HCl; pH 7.4; 1 mM EDTA; 1% glycerol], then resuspended in 25 mL of ice-cold french press buffer [50 mM Tris-HCl, pH 7.4; 5 mM MgCl2; 0.5 mM PMSF; 0.5 mM DTT; 5% glycerol; and 10 µg/ml DNaseI and RNaseA]. The cold cell suspensions were passed through a french press at 5 000 psi as a slow steady stream. The cell lysates were centrifuged at 10 000 g for 10 min at 4°C to remove unbroken cells, then centrifuged at
40 000 g in a pre-chilled ultracentrifuge for 60 min at 2°C. The supernatants were 64 removed and the vesicle pellets resuspended drop-wise in 900 µL vesicle buffer [50 mM Tris, pH 7.4; 5 mM MgCl2; and 10% glycerol] and worked into a paste using a thin glass rod and a 1 mL pippette. The vesicle mixtures were divided into 40 µL aliquots and stored at -80°C until required. A small aliquot (~10 µl) was retained at -20°C for total protein analysis via the Folin method (section 2.8.1) and SDS PAGE.
3.1.5 Enzyme assays
ATPase assays
The ATPase activities of the purified ABC and truncated peptides (McyHΔ57,
McyHΔ161), as well as the McyHL-containing vesicles, were measured by a colorimetric assay using malachite green reagent (Chan et al, 1986). Assays were performed in 250 µL volumes in 96 well microtitre plates and contained assay buffer
[10 mM Tris-HCl; 10 mM MgCl2; 15% (v/v) glycerol; 10% (v/v) C2H6SO, pH 7-8],
0-15 µM of microcystin-LR and 0-200 µg purified McyH peptide or membrane vesicle protein. The reaction was started by the addition of 10 µl of 5 mM ATP and incubated at 37°C for 5 min. The reaction was stopped by the addition of 200 µl of malachite green solution*. Absorbances were measured at 630 nm. A standard curve was constructed for each assay using 0-175 µM inorganic phosphate (Pi) from potassium phosphate. For each assay, a control reaction without ATP was performed to test for Pi contamination in reagents.
* Malachite green solution was made with 5.72% buffered ammonium molybdate, 2.32% (w/v) polyvinal alcohol (in boiling H2O), 0.0812% (w/v) malachite green and H2O mixed in a ratio of 1:1:2:2, respectively on the day of use and allowed to stand at room temperature for 30 min or until solution turned golden yellow.
65 Microcystin transport in everted membrane vesicles
The uptake of microcystin into everted membrane vesicles was investigated via semi- quantitative MALDI-TOF MS (Wang et al, 1996; Duncan et al, 2005). Each 100 µL assay contained 5 mg of vesicle protein, 0 or 5 mM ATP and 6.7 µM microcystin-LR in transport assay buffer [50 mM Tris, pH 7.4; 5 mM MgCl2]. Reactions were incubated for 60 min at 25°C, then filtered through 10 KDa Microcon (YM-10) molecular weight cut-off centrifugal filter devices (Millipore) at 1 000 g for 30 min. The filtrate and concentrated vesicle mixture were removed to separate tubes and spiked with 0.8 µM
[demethylated Adda, D-Asp3]microcystin-RR which was used as an internal standard for subsequent analysis by MALDI-TOF MS.
Positive-ion mass spectra were recorded using a Voyager DE STR MALDI-TOF mass spectrometer (Applied Biosystems). Half a microlitre of sample was combined with one microlitre of matrix (α-cyano-4-hydroxycinnamic acid in 80% acetonitrile; 0.08% trifluoroacetic acid) and applied directly to the target plate. Analyses were performed in the positive-ion mode, producing mainly singly protonated molecular ions ([M+H]+).
The acceleration voltage was set at 20 kV. All measurements were carried out in the delayed extraction and reflection mode, allowing the determination of monoisotopic mass values (m/z, mass-to-charge ratio). A low mass gate of 450 Da was used to filter out the most intensive matrix ions.
Microcystin-LR to -RR ratios were determined by measuring the relative intensities of their respective peaks (m/z 995 and 1 010). Microcystin-LR and -RR standards (1 µM in assay buffer) were also analysed. The experiment was repeated on separate days.
66 3.1.6 Mutagenesis of mcyH
The mcyH knock-out (ΔmcyH) mutant BsmI-2a was engineered by Michael Hisbergues
(Humboldt University, Berlin) according to Pearson et al (2004). Briefly, mcyH was partially deleted and inactivated via the insertion of a chloramphenicol resistance (CmR) gene cassette. While mcy transcripts were detected in the resulting ΔmcyH mutant strains, they were unable to produce microcystin.
The disruption of mcyH in M. aeruginosa PCC7806, via site-directed mutagenesis
(SDM), was also attempted in this study. For construction of the site directed mutagenesis construct, pDRIVE(mcyH/sdm), oligonucleotide primers abc(mutWa)F [5´-
CGGGCCGAGGTGCTGGAAGGAGG-3´] and abc(mutWa)R [5´-
GGACGAGGTAGAAGTTCTCTA-3´] were used to amplify by PCR a 4 147 bp fragment corresponding to nucleotides -549 to 3 598 of the mcy gene cluster. These primers were designed to introduce a single A/G base pair change in the Walker A- encoding region of mcyH, corresponding to a Lys/Arg substitution in the translated peptide. The resulting PCR fragment was cloned into the pDRIVE vector (Qiagen) giving rise to pDRIVE(mcyH). A 1 414 bp CmR cassette, excised from the plasmid pACYC184 (Fermentas) using BsaAI, was then cloned into a non-coding region of the mcy insert (between mcyJ and dnaN) at a unique MscI restriction site (Figure 3.2). The resulting construct, pDRIVE(mcyH/sdm), was transformed into E. coli XL1-Blue, which was subsequently grown under selection with chloramphenicol (34 µg/ml) and ampicillin (100 µg/ml). Resulting transformants were checked for pDRIVE(mcyH/sdm) via PCR and sequencing using the vector-binding primers MpF and MpR.
67 pDRIVE(mcyH/sdm), was transformed into M. aeruginosa PCC7806 using natural transformation and electroporation. For natural transformation, 10 ml of log-phase culture (6 × 107 cells/ml) was centrifuged then resuspended in 200 µl BG11.
Approximately 10 µg linearised or supercoiled plasmid DNA was added to the resuspended cells which were subsequently incubated at room temperature for 1 h (16
µmol photons m-2 s-1), then spread onto BG11 plates (1% agar). Cells were incubated for 72 h under low light (16 µmol photons m-2 s-1) at 25°C before a chloramphenicol gradient was added to the media. Chloramphenicol gradients were established by underlaying 200 µl of the antibiotic (1 mg/ml) on the plate perimeter and allowing it to diffuse through the agar. Transformants were obtained after 4 weeks incubation under low light at 25°C.
For electroporation, 10 ml of log-phase culture (6 × 107 cells/ml) was centrifuged then washed twice with 1 mM HEPES buffer (pH 7). Following resuspension of the pellet in
100 µl 1 mM HEPES buffer, approximately 10 µg of column purified linearised, or supercoiled plasmid DNA was added to the cells. Cells were electroporated (10 kV/cm,
25 µF, 200 Ω) in Gene Pulser Cuvettes (0.1 cm electrode gap, BioRad) using a Gene
Pulser II (BioRad). After addition of 100 µl 2 × BG11, cells were incubated on ice for
30-60 min then spread onto BG11 plates and grown under a chlorampenicol gradient as above.
3.1.7 Regulation of McyH expression
McyH expression was investigated by analysing total protein extracts from
M. aeruginosa using immunoblotting with McyH primary anibodies. For all experiments, optical densities of cell cultures were measured using an Ultrospec II 68 (LKB, Biochrom), and light intensities were measured using a Walz LI-CORR LI-250 light meter.
Polyclonal antibodies were raised in rabbits by Pineda Antikörper Service (Berlin,
Germany) using 300 to 600 µg of 90-95% pure recombinant ABC protein [expressed from pET30(ABC)]. IgG antibodies from the final sera were precipitated using 40%
(NH4)2SO4, concentrated to 30% of the initial volume and dialysed 24 h in PBS buffer
[140 mM NaCl; 2.7 mM KCl; 10 mM Na2HPO4; 1.8 mM KH2PO4, pH 7.3] at 4°C.
Regulation of McyH expression under different light conditions
The effects of light on McyH expression were investigated by culturing M. aeruginosa under various light conditions and analysing total protein extracts via immunoblotting with McyH anibodies.
An M. aeruginosa PCC7806 starter culture was grown in BG11 under low light (16
-2 -1 µmol photons m s ) to an OD750 of 0.8. The culture was harvested at 5 000 g for 10 min at 4°C, resuspended in fresh medium and incubated for an additional 2 h under low light. The culture was then divided into 6 separate subcultures, which were grown under light conditions a-f, where a = total darkness, b = low light (16 µmol photons m-2 s-1), c = medium light (30 µmol photons m-2 s-1), d = high light (68 µmol photons m-2 s-1), e = low light with a red filter, and f = low light with a blue filter. Aliquots were taken from each culture at 2 h and 18 h, harvested via 5 000 g centrifugation, and stored as frozen pellets at -80°C until required.
69 Cell pellets were thawed on ice and resuspended in ice-cold buffer A (500 mM Tris-
HCl, pH 7.5; 50 mM EDTA; 2 µM PMSF), and lysed via 3 successive rounds of freeze- thawing (liquid nitrogen) and bead beating in a Savant FastPrep (30 sec, max speed) with 200 mg of 0.1 mm zircon silica beads (BioSpec). Samples were centrifuged at 1
000 g for 1 min to sediment glass beads and the resulting supernatants were transferred to a fresh tube and centrifuged at maximum speed for 40 min to pellet insoluble proteins.
Twenty to fifty micrograms of soluble protein extract was electrophoresed on 12-15%
SDS polyacrylamide gels using the Mini protean apparatus (Bio-rad). Proteins were then blotted to Hybond C-extra membranes (Amersham Biosciences) under semi-dry conditions for 1 h at 24 V, using a Trans-Blot SD semi-dry transfer cell (Bio-Rad) through Bjerrum and Schafer-Nielsen transfer buffer (48 mM Tris, pH 9.2; 39 mM glycine; 20% methanol).
After blocking overnight in PBS-T (140 mM NaCl; 2.7 mM KCl; 8 mM Na2HPO4; 18 mM KH2PO4; 0.3% tween) plus skim milk (5% w/v), the membranes were incubated for 1 h with primary McyH antibodies (diluted 1:2 000 to 1:5 000 in blocking buffer).
Membranes were washed 3 times (10 min each) in PBS-T and subsequently incubated for 45 min in anti-rabbit IgG antisera coupled to horseradish peroxidase (Pierce) diluted
1:5 000 in PBS-T. After washing 3 times (10 min each) in PBS-T, blots were developed via chemiluminescence using the Supersignal West Pico chemiluminescent substrate according to the manufactures protocol (Pierce).
70 Regulation of McyH expression in different mcy mutants
The expression of McyH in mutant and WT cell extracts was also analysed via immunoblotting with McyH antibodies as above. The peptide synthetase mutant strains mcyA- (Tillett et al, 2000) and mcyB- (Dittmann et al, 1997), were obtained from Elke
Dittmann (Humboldt University, Berlin).
Wild-type M. aeruginosa and mcyA-, mcyB-, and ΔmcyH mutants were grown under low light to an OD750 of 0.8. Total protein extracts from mid-log phase cultures were prepared, and analysed by immunoblotting with McyH antibodies, as above.
71 3.2 RESULTS
3.2.0 Distribution of mcyH orthologues in various species of cyanobacteria
In order to assess the distribution of mcyH orthologues among toxic and non-toxic cyanobacteria, DNA from 24 different strains was screened by PCR with mcyH-specific and degenerate oligonucleotide primers. The results are presented in Table 3.0. All hepatotoxin-producing strains tested positive for mcyH/ndaI orthologues using both primer sets. Half of the non-toxic cyanobacteria screened also tested positive for mcyH.
All strains possessing mcyH orthologues also tested positive for other mcy genes (as determined by mcyI or mcyJ screening).
3.2.1 Sequence analysis of mcyH
The ~1 800 bp mcyH open reading frame is predicted to encode a 67 KDa peptide with a pI of 5.90. DAS and Kyte-Doolittle topology analyses identified a large, hydrophobic
N-terminal domain with 4 putative transmembrane regions between residues 44-294 of the McyH peptide sequence. The C-terminal domain of the sequence (residues 296-585) was largely hydrophilic (Figure 3.3).
SignalP analysis could not identify a signal peptide within the N-terminal sequence of
McyH, suggesting that it is not a secretory protein. PSORTb predicted a cytoplasmic membrane localisation for McyH. Analysis of McyH codon usage revealed a high proportion of rare E. coli codons, including several repeated and/or consecutive rare codons. Subsequent heterologous expression of the McyH peptide was therefore performed in the Rosetta(DE3) E. coli expression strains (Novagen) containing the
72 pRARE plasmid encoding tRNA genes for “problematic” rarely used E. coli codons
(Table 3.1).
Comparison of the inferred primary peptide (amino acid) sequence of mcyH with other sequences in the NCBI database revealed significant similarity [up to 80% identity (I) and 87% similarity (S), PSI-Blast] to several members of the ABC transporter superfamily. Sequences of particularly high similarity included hypothetical ABC transporters from the microcystin biosynthesis gene clusters of Planktothrix agardhii
(80% I, 89% S), Nodularia spumigena (72% I, 87% S) and Anabaena sp. (72% I, 86%
S) and the nostopeptolide biosynthesis cluster of Nostoc sp. (68% I, 82% S). Other uncharacterised homologues were identified in species belonging to the cyanobacterial genera Nostoc, Thermosynechococcus, Trichodesmium, Synechocystis,
Prochlorococcus, and Synechococcus (37-67% I, 57-86% S) and the plant species
Arabidopsis thaliana (32% I, 53% S) and Oryza sativa (31% I, 50% S). McyH also shared sequence similarity with the peroxisomal membrane proteins of various eukaryotic organisms including Mus musculus (28% I, 49% S), Gallus gallus (28% I,
48% S) and Homo sapiens (27% I, 49% S) (see table 3.2 for accession numbers). Figure
3.4 shows a section of the alignment of McyH and the top 20 peptide sequences with the highest homology (based on the PSI-Blast data). The diagnostic ABC ATPase
Walker A and B motifs (beginning at residues 406 and 517, respectively) are highlighted. A sequence corresponding to the ABC signature motif (LSGGQQ,
Schneider and Hunke, 1998) was also identified at residues 497-504. The EAA motif, characteristic of ABC import systems (Kerppola and Ames, 1992), was not present in
McyH or any other peptides in the alignment.
73 A Blast-CD search identified several conserved domains within the McyH peptide sequence, including a SunT N-terminal double Gly peptidase domain (typically associated with bacteriocin/lantibiotic exporters), an MdlB domain (associated with multidrug transporters) and an FepC domain (associated with siderophore transporters).
A FUGUE homologous structure search using the McyH peptide sequence yielded similar results to the Blast CD search. The highest confidence Z-score (57.95) was obtained for the multi-drug resistance transporter MsbA, from Vibrio cholerae
[Q9KQW9]. A ScanProsite search identified an ABC transporter integral membrane type-1 fused domain profile (ABC_TM1F) between residues 50 to 334 of the McyH peptide sequence, and an ATP-binding cassette domain profile
(ABC_TRANSPORTER_2) between residues 370 and 585. An ABC transporter family signature domain profile (ABC_TRANSPORTER_1) was also identified between residues 497 and 511.
To gain an overview as to the evolutionary relationships between McyH and other ABC transporters, two phylogenetic analyses were performed (Figures 3.5 and 3.6). The first analysis was based on the entire sequence lengths of the McyH homologues, the second analysis was based only on their highly conserved ATPase domains.
The 54 sequences used in the first analysis ranged in length from 489-771 amino acids, with an average length of 619 (Table 3.2). The overall tree partitioned into 6 clusters encompassing groups A-F. Group A was comprised of putative ABC transporters from plants and proteobacteria. Group B was comprised of putative ABC transporters from cyanobacteria, proteobacteia and acidobacteria. Groups C and D were comprised
74 exclusively of putative cyanobacterial ABC transporters. Group E was comprised exclusively of cyanobacterial ABC transporters encoded within, or proximal to, secondary metabolite biosynthesis gene clusters. Group F was comprised of peroxisomal membrane proteins (PMPs) from the phylum Chordata (Figure 3.5).
The 50 sequences used in the second analysis ranged in length from 155-158 amino acids with an average length of 157 (Table 3.3). The un-rooted tree partitioned into two major clusters encompassing the putative cyanobacterial secondary metabolite- associated ABC transporters, and the eukaryotic peroxisomal membrane proteins
(Figure 3.6).
3.2.2 Over-expression and Purification of histidine-tagged McyH peptides
A range of different expression vectors and induction conditions were utilised in an attempt to express McyH and its component peptides in E. coli. McyH peptides were expressed at levels detectable by PAGE (with Coomassie staining) from pET30(mcyHL), pET30(abc), pET30(mcyHΔ15), pET30(mcyHΔ57), and pET30(mcyHΔ161). However, all attempts to express recombinant McyH peptides from the expression vectors pET30(mcyH), pET32(mcyH), pET43(mcyH) and pET30(md) were unsuccessful.
The longer recombinant McyH peptide (McyHL) was expressed from pET30(mcyHL) at low levels for all culture/induction conditions tested, however ~99% of the expressed peptide was present in the insoluble cell fraction. Western blot analysis confirmed that
McyHL was expressed without truncation, at the expected molecular weight of ~65
KDa. The insoluble protein fraction was solubilised in 8 M urea and the recombinant
75 McyHL protein was purified via affinity chromatography under denaturing conditions.
The eluate was of >90% purity and was eluted at concentrations of up to 0.1 mg/mL
(Figure 3.7A).
The recombinant ATP binding cassette (ABC) domain of McyH expressed at moderate levels from pET30(abc) under all culture/induction conditions tested, however >80% of the expressed peptide was present as inclusion bodies. Western blot analysis confirmed that the ATPase was expressed without truncation, at the expected molecular weight of
~40 KDa. Maximum yields of soluble protein were obtained via growth in TP with reduced induction temperatures (25-30°C) and low concentrations of IPTG (0.25-0.50 mM). No differences in solubility or yield were observed for cultures induced at 2-18 h.
Where large yields of protein were required, irrespective of solubility, higher concentrations of IPTG (0.5-1 mM) were optimal.
The recombinant ABC peptide precipitated in the presence of 40% ammonium sulphate at 4°C. The soluble recombinant protein was purified via affinity chromatography and the eluate was of >90% purity and was eluted at concentrations of up to 0.2 mg/mL
(Figure 3.8A). The insoluble protein fraction was solubilised in 8 M urea and the recombinant ABC protein was also purified via affinity chromatography under denaturing conditions. The eluate was of 90% purity and was eluted at concentrations of up to 3 mg/mL (Figure 3.8B). The denatured purified protein was used to raise McyH antibodies in rabbits. These antibodies specifically detected a protein of ~60 KDa in wild-type M. aeruginosa PCC7806 total protein extracts (see below).
76 The recombinant McyHΔ15 peptide was expressed at low levels from pET30(mcyHΔ15) under all culture/induction conditions tested, however 100% of the expressed peptide was present as inclusion bodies. Western blot analysis confirmed that
McyHΔ15 was expressed, at the expected molecular weight of ~59 KDa (Figure 3.7C).
The recombinant McyHΔ57 peptide was expressed at low levels from pET30(mcyHΔ57) when cells were cultured and induced at room temperature with low levels of IPTG (0.15 mM), however >95% of the expressed peptide was present as inclusion bodies. Western blot analysis confirmed that McyHΔ57 was expressed, at the expected molecular weight of ~55 KDa. The recombinant McyHΔ57 peptide precipitated in the presence of 40% ammonium sulphate at 4°C (see materials and methods). The soluble recombinant peptide was purified via affinity chromatography and the protein eluate was of >20% purity and was eluted at concentrations of up to 0.5 mg/mL.
The recombinant McyHΔ161 peptide was expressed at low levels pET30(mcyHΔ161) under all culture/induction conditions tested, with >90% of the expressed peptide present as inclusion bodies. Western blot analysis confirmed that McyHΔ161 was expressed, at the expected molecular weight of ~43 KDa (Figure 3.7C). The recombinant McyHΔ161 peptide precipitated in the presence of 40% ammonium sulphate at 4°C. The protein eluate was of >80% purity and was eluted at concentrations of up to 2.5 mg/mL (Figure 3.7B).
77 3.2.3 Enzyme assays
ATPase assays
The ATPase activities of the purified ABC and McyHΔ161 peptides, and the McyHL- containing vesicles (see below), were measured by a colorimetric assay using Malachite green reagent (Chan et al, 1986). However, the production of Pi in assays containing the purified ABC and McyHΔ161 peptides was equivalent to assays containing heat- denatured control enzyme. Similarly, assays containing McyHL vesicle protein, yielded equivalent levels of Pi to those containing control (pET30) vesicle protein.
The addition of 0-15 µM microcystin had no effect on ATPase activity.
Microcystin transport in everted membrane vesicles
Everted membrane vesicles were isolated from E. coli expression strains transformed with pET30(mcyHL) and pET30 (negative control). Vesicle proteins were analysed by
PAGE (with coomassie staining) and western blotting. McyH was expressed at up to
10% of total vesicle protein (Figure 3.9).
The uptake of microcystin-LR in vesicles was investigated via semi-quantitative
MALDI-TOF MS with an internal microcystin-RR standard. The relative quantity of microcystin-LR in each sample was determined as a ratio of the internal microcystin-RR standard. Membrane vesicles containing McyHL retained, on average,
~56% more microcystin-LR than control vesicles, as judged by their respective LR:RR mass peak % intensity ratios of 100:16 (McyHL) and 100:7 (control) (Figure 3.10).
Samples lacking ATP yielded similar results to those containing control vesicles.
78 3.2.4 Mutagenesis of mcyH
M. aeruginosa PCC7806 was transformed with the site-directed mutagenesis construct, pDRIVE(mcyH/sdm). Transformants isolated after ~ 4 weeks tested positive for chloramphenicol resistance cassettes via PCR, however, antibiotic resistance
(phenotypic and genotypic) was lost when colonies were subcultured onto fresh media.
These mutants were therefore not useful for further characterisation of mcyH.
3.2.5 Regulation of McyH expression
Regulation of McyH expression under different light conditions
The effects of light on McyH expression were investigated by culturing M. aeruginosa under various light conditions and qualitatively analysing total protein extracts via immunoblotting with primary McyH antibodies. Expression of the putative ABC transporter was found to be up-regulated under medium, high and red light conditions, and down-regulated under low and blue light conditions. Expression of McyH was also down-regulated when cells were cultured in total darkness (Figure 3.11A).
Regulation of mcyH expression in mcy knock-out mutants
The expression of McyH in mutant and WT cell extracts was qualitatively analysed via immunoblotting with the McyH antibodies as above. A ~60 KDa protein, corresponding to McyH was specifically detected in wild-type M. aeruginosa and in the mcyA-, and mcyB- mutant strains, however, expression in the mutant strains was much lower than
79 wild-type expression levels (<50%). A 60 KDa protein could not be detected in the
ΔmcyH mutant strains (Figure 3.11B)
80 3.3 DISCUSSION
In order to assess the distribution of mcyH orthologues among toxic and non-toxic cyanobacteria, DNA from 24 different strains was screened by PCR with mcyH-specific and degenerate oligonucleotide primers. A strong correlation between mcyH and hepatotoxin production was observed. Orthologues of mcyH or the nodularin synthetase equivqlent, ndaI, were detected in all microcystin and nodularin-producing strains tested. These results suggest that these ABC-transporters are critical for toxin production and/or the survival of toxin-producing strains.
Interestingly, nearly half of all the non-toxic cyanobacterial strains examined tested positive for mcyH. These results reflected previous studies that identified mcy genes in non-toxic Microcystis strains (Nishizawa et al, 1999; Tillett et al, 2000; Kaebernick et al, 2001). It is likely that these strains have reverted to a non-toxic phenotype due to mutations within their mcy biosynthesis genes, or within mcy regulatory regions. In order to pin-point the exact cause of the various non-toxic phenotypes, it may be necessary to sequence and compare long stretches of the mcy gene cluster from each strain in question. A previous study by Kaebernick et al (2001), reported that M. aeruginosa MR-C is identical to its toxic counterpart MR-D, at 15 genetic loci within the mcy gene cluster. Despite efforts to sequence several different regions within the mcy clusters of these strains, the authors could not identify a single point of mutation.
The fact that false negatives weren’t obtained for the PCR screen with mcyH(200)F/(1222)R and mcyHscreenF/R, suggested that these primer sets are good genetic probes for assessing the potential hepatotoxicity of cyanobacteria in water supplies.
81 The inferred primary peptide sequence of McyH was analysed by several different computer programs in an attempt to gather structural and functional information for subsequent biochemical experiments. The combined sequence data strongly suggested that McyH belongs to the ABC transporter superfamily. In addition to sharing extensive sequence similarity with other ABC transporters, the McyH sequence contains the diagnostic A and B motifs of Walker et al (1982), and the ABC signature motif
(Schneider and Hunke, 1998). The molecular mass of the putative ATP-binding domain of McyH (~26.99 KDa) is also typical for an ABC ATPase, which average 27 KDa
(Holland and Blight, 1999).
Several lines of evidence suggest that McyH functions as an ABC exporter in M. aeruginosa. Firstly, it lacks the EAA motif characteristic of bacterial import systems
(Kerppola and Ames, 1992). Secondly, hydropathy profiling and secondary structure predictions indicated that McyH is a fusion protein with membrane and ABC domains encoded within a single polypeptide. This suggested an export function for the protein since the membrane and ABC domains of exporters are usually fused, while the domains of importers are always separately encoded within individual polypeptides
(Saurin et al, 1999). As the general structure of ABC transporters consists of two membrane spanning domains and two ABC domains (Higgins, 2001), the putative
McyH transporter is predicted to function as a homodimer.
The predicted membrane localisation for McyH, as determined by PSORTb analysis, suggested that the transporter exports molecules from the cytoplasm across the
82 cytoplasmic membrane. McyH may thus be responsible for the observed localisation of microcystin on the wall and sheath of M. aeruginosa cells (Shi et al, 1995).
Previous research indicated that ABC transporters with similar primary peptide sequences have similar physiological functions (Figure 1.1, Saurin et al, 1999). In an attempt to elucidate the function of McyH, various comparative sequence analyses were performed. Interestingly, PSI-Blast analysis identified McyH homologues from toxic and non-toxic species of cyanobacteria, however McyH homologues with the highest score were enzymes associated with cyanobacterial secondary metabolite gene clusters.
McyH also shared sequence similarity with the peroxisomal membrane proteins (PMPs) of various eukaryotic organisms. These ABC transporters are known to transport long- chain fatty acids into the peroxisomes for subsequent degradation into acetyl-CoA
(Lazarow, 1987; Mosser et al, 1993). Such long-chain fatty acids resemble the dienoic acid side-chain moiety (Adda) of microcystin (see below).
A Blast-CD search identified several conserved domains within the McyH peptide sequence, including a SunT N-terminal double Gly peptidase domain (Paik, et al, 1998), an MdlB domain (Parkhill et al, 2001) and an FepC domain (Wyckoff et al, 1999).
These domains are typically associated with bacteriocin/lantibiotic, drug, and siderophore transporters, respectively. The allocrites of such transporters (modified cyclic peptides, non-ribosomal peptides and polyketides), have similar structures, bioactivities and biosynthetic pathways to microcystin. Therefore, these results support the hypothesis that McyH is also involved in microcystin export.
83 The first comprehensive global phylogenetic analysis of ABC transporters which was conducted by Saurin and co-workers in 1999, was based on the highly conserved
ATPase domains of approximately 200 ABC transporters. The authors discovered that the ABC superfamily partitions primarily according to the direction of transport in relation to the cytoplasm, (i.e. import versus export). By eliminating the less conserved membrane domains from their analysis, the authors were also able to demonstrate that
ABC transporters with similar functions clustered together, regardless of their organismic origin.
To gain an overview as to the relative phylogenetic position of McyH among other
ABC transporters, two phylogenetic analyses were performed (Figures 3.5 and 3.6). The first was based on the entire sequence lengths of McyH and its homologues, the second was based on their highly conserved ATPase domains. The sequences used in the analyses had N-terminal membrane domains fused to C-terminal ABC modules, indicating that they belong to the ABC-A1 sub-family (Saurin et al, 1999) and therefore function as exporters. These results concur with initial hydropathy profiling and secondary structure predictions for McyH, which also suggested an export function for the protein.
Nearly 70% of all sequences in the initial (full-length transporter) tree were of cyanobacterial origin, however these sequences were distributed across four phylogenetically distinct clusters, suggesting that the cyanobacterial ABC transporters in the tree evolved from a common ancestor prior to the divergence of this bacterial phylum. McyH partitioned into sub-group E, which was comprised exclusively of putative ABC transporters from secondary metabolite gene clusters. These results
84 suggest that the group E cyanobacterial proteins interact with a similar class of allocrites, presumably cyclic peptides (eg. microcystin, nodularin, nostopeptolide). The fact that the tree encompassed a diverse range of organisms including bacteria, plants and animals, also supports previous findings that suggested ABC transporters originated early in the evolution of life, prior to the divergence of prokaryotes and eukaryotes
(Saurin et al, 1999). The conservation of this superfamily over the millennia is highlighted in Group B, where putative ABC transporters from diverse bacterial orders
(cyanobacteria, proteobacteia and acidobacteria) share a single evolutionary branch.
While none of the sequences in groups A-E have been functionally characterised, it is likely that all putative ABC transporters in this phylogenetic analysis bind a broadly similar class of compounds. As the closest characterised relatives of these proteins, the peroxisomal membrane proteins (sub-group F), transport long-chain fatty acids, sub- groups A-E may transport lipids or other polar molecules, such as branched-chain secondary metabolites.
The second phylogenetic analysis, was based solely on the ATPase domains of McyH and its homologues. Interestingly, the top 50 sequences returned by the PSI-Blast search were comprised exclusively of cyanobacterial secondary metabolite-associated ABC transporters, and eukaryotic peroxisomal membrane proteins that export fatty acids from the cytoplasm into peroxisomes. In the final un-rooted tree, these sequences partitioned into two distinct phylogenetic clusters. The high degree of sequence conservation among the PMPs is highlighted by the fact that sequences from a diverse range of eukaryotes (parasites, insects, fish, birds, mammals etc.) clustered closely together.
85 While cyanobacteria lack peroxisomes, the combined bioinformatic data presented in this study suggest a close evolutionary relationship between the putative hepatotoxin transporters and the eukaryotic PMPs. A possible explanation for this strange observation relates to the evolution of fatty acid and polyketide metabolic pathways.
Type I polyketide synthases, such as those in microcystin synthase, and type I animal fatty acid synthases, are thought to share a common ancestor (Jenke-Kodama et al,
2005). Therefore, it follows that the type I polyketide ABC transporters and type I fatty acid ABC transporters are also evolutionarily linked. While differences in cyanobacterial and peroxisomal membrane architecture are the likely cause of the observed sequence divergence within the MD domains of these ABC transporters, their cognate ATPase domains have retained a high degree of sequence conservation.
The recombinant McyH protein (McyHL), as well as several truncated ABC peptides
(ABC, McyHΔ15, McyHΔ57 and McyHΔ161), were expressed at levels detectable by
SDS-PAGE and either Coomassie blue staining, or western blot analysis. However, despite optimisation of induction and growth conditions during expression, these peptides were extremely prone to inclusion body aggregation. Removal of the hydrophobic membrane domain of the McyH peptide somewhat alleviated this problem in the ABC and McyHΔ161 peptides, however, yields of soluble recombinant protein were generally quite low (0-20% of total recombinant protein).
Despite extensive cloning and experimental optimisation, the 1 614 bp mcyH ORF
[AAF00956] was unable to be expressed under the described conditions. However, a
86 1 755 bp mcyH ORF, incorporating an alternative start site 141 bases upstream of that of the published Genbank sequence, was successfully over-expressed in E. coli. These results suggest that the true mcyH ORF utilises the -141 ATG start codon and not the
Genbank-published start codon.
The ABC peptide was purified via affinity chromatography, and used to raise McyH antibodies in rabbits. These antibodies specifically detected a protein of ~60 KDa corresponding to McyH in wild-type M. aeruginosa PCC7806 total protein extracts.
The soluble McyHL, ABC, McyHΔ57 and McyHΔ161 peptides were similarly purified for enzymatic analysis.
Previous studies have demonstrated that certain ABC transporters display constitutive
ATPase activity in vitro when separated from their cognate membrane domains
(Nikaido et al, 1997). However, the purified, recombinant McyH peptides assayed in this study did not hydrolyse ATP at detectable levels. The loss of ATPase activity in the truncated McyH peptides was probably due to protein misfolding following domain removal. For example, an intact membrane domain may be critical for McyH dimer formation, which has been demonstrated to be important for ATPase activity (Hopfner et al, 2000; Higgins and Linton, 2004). Alternatively, the physico-chemical parameters of the ATPase assay may have been sub-optimal for McyH activity. For example,
McyH may require an alternative metal cofactor such as Co2+ or Mn2+, or may be unstable within the pH range 7-8. Further optimisation of the assay conditions may eleviate these problems.
87 Surprisingly, ATPase activity was also unable to be detected (above background levels) for the full-length McyHL enzyme that was reconstituted into membrane vesicles.
However, as the McyHL vesicles also contained a mixture of E. coli membrane proteins, and as McyHL constituted <10% of vesicle protein, its activity may have been masked by the contaminating host ATPases. Alternatively, McyHL, being a cyanobacterial protein, may not be stable in E. coli membrane vesicles, and hence may have had significantly reduced activity under the described conditions. As concentrations of vesicle protein greater than 0.4 mg/mL caused the formation of a cloudy precipitate in the assay mixture, McyHL-mediated ATPase activity could not be enhanced by simply increasing the concentration of vesicle protein in the assay.
The uptake of microcystin into membrane vesicles was investigated via semi- quantitative MALDI-TOF (Wang et al, 1996; Duncan et al, 2005) using an internal microcystin-RR standard. Everted membrane vesicles containing McyHL retained, on average, 60% more microcystin-LR than control vesicles and vesicles lacking ATP, suggesting that McyHL-mediated ATP-dependent transport was indeed taking place.
While it was impossible to determine kinetic values for microcystin transport using the described methods, the chemical synthesis of a labelled microcystin isotope will enable us to probe the enzymology of this ABC transporter in greater detail in the near future.
The mutation and inactivation of putative ABC transporter genes is the traditional method for determining the function of their encoded proteins. Previous disruption of mcyH, via the partial deletion and the insertion of an antibiotic resistance marker, resulted in the complete loss of toxin production in the ∆mcyH mutant strains (Pearson et al, 2004). As mcyH transcripts were detected in ∆mcyH mutants, the authors
88 concluded that the loss of microcystin production was not caused by polar effects, but was due to the instability of the microcystin synthetase complex, following the removal of one of its components (McyH). Similar results have been recorded for other bacterial secondary metabolite pathways. For example, in a study investigating the biosynthesis of the lantibiotic nisin, Siegers et al (1996) used co-immunoprecipitation and the yeast two-hybrid system to demonstrate that the lanthionine synthetase complex (NisB/C) is anchored to the membrane by the nisin exporter NisT. Subsequent inactivation of nisT via insertional mutagenesis resulted in the complete abolition of nisin biosynthesis.
Similarly, disruption of the ABC transporter ExiT, encoded within the exochelin biosynthesis gene cluster, abolished synthesis of the non-ribosomal peptide siderophore
(Zhu et al, 1998). These results suggested that synthesis and export are tightly coupled in some non-ribosomal peptide systems. In the case of exochelin biosynthesis, this strategy is advantageous for the cell because excess iron-free siderophore would sequester essential iron cofactors from intracellular components (Zhu et al, 1998). Like exochelin, microcystin is a powerful siderophore (Humble, 1994), and its accumulation in the cell might similarly disrupt internal iron metabolism.
In an attempt to create a microcystin-producing mcyH mutant that expressed a stable but non-functional form of the McyH transporter, we engineered a site-directed mutagenesis
(SDM) construct that introduced a single base-pair mutation at the Walker A site.
Previous studies have shown that mutation of the conserved Walker a Lys results in the disruption of ATPase activity and therefore, the loss of allocrite transport (Schneider and Hunke, 1998).
89 While the pDRIVE(mcyH/sdm) vector was successfully transformed into M. aeruginosa
PCC7806 (as judged by the chloramphenicol resistant phenotype/genotype of resulting transformants), stable SDM mutants could not be isolated. A possible explanation for the instability of mcyH mutants may relate to the loss of self-resistance against microcystin. If McyH is responsible for regulating intracellular levels of microcystin, disruption of this ABC transporter may lead to the accumulation of lethal levels of toxin within the cell. Hence the observed death of transformants following prolonged culture periods (more than 4 weeks). A possible strategy for overcoming host-cell lethality in mcyH mutants, is to utilise an inducible mutation, such as a light or temperature- inducible mutation. This strategy would allow characterisation of McyH-mediated transport within a limited time frame, prior to the build up of possibly toxic intracellular levels of microcystin.
Immunoblotting experiments demonstrated that the expression of McyH is up-regulated under medium, high and red light culture conditions, and down-regulated under low and blue light culture conditions, as well as in total darkness (Figure 3.11A). These results reflect those of previous transcriptional regulation studies for mcyH (M. Hisbergues, pers. comm.) and other mcy genes (Kaebernick et al, 2000). It has been suggested that microcystin may be constitutively produced under low and medium light intensities, and is exported only when a certain higher threshold intensity is reached (Kaebernick et al,
2000). The observed upregulation of McyH under high light conditions observed in this study, further supports this theory.
Analysis of the McyH protein in WT and mutant strains of M. aeruginosa PCC7806
90 revealed that expression of the hypothetical ABC transporter is reduced in the non-toxic mcyA- and mcyB- mutants (Figure 3.11B). Instability and degradation of McyH following the disruption of other Mcy components (McyA and McyB) similarly may be the cause of this reduced expression. Alternatively, in the absence of allocrite
(microcystin), expression of the transporter could be down regulated. Similar observations have been recorded for ABC proteins spanning several different classes.
For example, expression of the BcrABC transporter, responsible for extrusion of the non-ribosomal peptide antibiotic bacitracin, was increased several fold when cells were exposed to bacitracin (Neumuller et al, 2001). Similarly, ABC transporters belonging to the multi-drug resistance (MDR) class are also inducible by their own allocrites. For example, expression of the human Mrp2 protein is significantly up-regulated following dosage of patients with the antibacterial drug rifampicin (Fromm et al, 2000).
Taken together, the data here would suggest a role for McyH in both toxin biosynthesis and export. If McyH is in fact responsible for the active transport of microcystin in
M. aeruginosa, as is suggested by the bioinformatic and experimental analyses performed in this study, then this ABC transporter will be the first toxin exporter to be identified in a cyanobacterium. Such a finding raises numerous questions regarding the ecophysiological role of microcystin and other non-ribosomally synthesised secondary metabolites in both toxic and non-toxic cyanobacteria.
91 CHAPTER 4.
CHARACTERISATION OF THE
2-HYDROXYACID DEHYDROGENASE, MCYI
92 4.0 OVERVIEW
The sixth open reading frame in the mcyD-J region of the microcystin biosynthesis gene cluster of Microcystis aeruginosa PCC7806 encodes a 1 014 bp gene, mcyI. Preliminary sequence analysis of the inferred primary peptide sequence of mcyI by Tillett et al
(2000), revealed a 41% identity to the catalytic region of the serA encoded
D-3-phosphoglycerate dehydrogenase (PGDH) [EC 1.1.1.95] from Methanobacterium thermoautotrophicum (Smith et al, 1997). Although this archaeal PGDH homologue has not yet been characterised, extensive research has been carried out on PGDH homologues from mammals, plants and bacteria.
PGDH belongs to the 2-hydroxyacid family of dehydrogenases and catalyses the first committed step in the phosphorylated serine biosynthesis pathway: the oxidation of
3-phosphoglycerate (3-PGA) to 3-phosphohydroxypyruvate (3-PHP), with the concomitant reduction of NAD to NADH. Interestingly, PGDH is also capable of catalysing the interconversion of α-ketoglutarate (αKG) and hydroxyglutaric acid
(HGA) (Zhao and Winkler, 1996). The D-2-hydroxyacid dehydrogenase family also includes D-lactate dehydrogenase (D-LDH) [EC 1.1.1.28] and malate dehydrogenase
(MDH) [EC 1.1.1.37], which catalyse the conversion of lactate to pyruvate and malate
(Mal) to oxaloacetate (OAA), respectively (Figure 4.0). The family members share an overall sequence identity of around 22% and a similarity of approximately 50% (Bell et al, 2002). These enzymes also possess a conserved nucleotide binding motif which preferentially binds to either NAD(H) or NADP(H).
Despite its similarity to PGDH and other 2-hydroxyacid dehydrogenases, McyI was originally predicted to function as a dehydratase enzyme, playing a role in microcystin
93 biosynthesis by catalysing the dehydration of serine to dehydroalanine (Dha) (Tillett et al, 2000). The recently sequenced nodularin gene cluster of Nodularia spumigena
NSOR10 also encodes a 2-hydroxyacid dehydrogenase homologue, NdaH. Like McyI, this enzyme was originally predicted to catalyse a dehydration reaction: the conversion of threonine to dehydrobutyrine (Dhb) (Moffitt and Neilan, 2004). While mcyI has been identified as an open reading frame in the microcystin biosynthesis gene cluster of
Anabaena sp. 90, no such homolog has been identified in Planktothrix agardhii
CYA126. It has been suggested that mcyI may be encoded at an alternative locus in the
P. agardhii CYA126 genome (Christiansen et al, 2003).
Although the microcystin and nodularin biosynthesis pathways may involve serine/threonine dehydration reactions, it is unlikely that these reactions are catalysed by McyI and NdaH, as originally predicted. As mentioned previously, McyI and NdaH are homologous to 2-hydroxyacid dehydrogenases and this family of enzymes is neither structurally or functionally related to the family of dehydratase enzymes known to convert serine to Dha in other secondary metabolite pathways (Sen et al, 1999). While dehydrogenases such as PGDH catalyse redox reactions (i.e. the transfer of electrons involving pyridinium nucleotides), the amino acid dehydratases catalyse dehydration and re-hydration reactions (i.e. the removal or addition of H2O). The chemical nature of these reactions as well as their respective substrates, are very different. Therefore, alternative putative functions for McyI and NdaH in hepatotoxin production needed to be investigated.
As McyI is homologous to PGDH, the first committed step in the phosphorylated serine biosynthesis pathway, it is conceivable that this enzyme may play a role in
94 M. aeruginosa serine metabolism. As McyI is encoded within the mcy gene cluster, and is co-regulated with the polyketide synthase genes mcyD, E and G under normal growth conditions (Kaebernick et al, 2002), the enzyme may direct the steady supply of serine required for microcystin biosynthesis.
An alternative hypothesis is that McyI plays a role in the biosynthesis of the starter unit of Adda. While the structure of Adda suggests priming with phenylacetate, in vivo feeding experiments do not support the direct involvement of this phenylalanine-derived starter unit (Moore et al, 1991; Rinehart et al, 1994). Moffitt et al (2006) have recently demonstrated that the polyketide synthase, McyG, preferentially activates an assortment of phenylpropanoids, particularly, cinnamate, hydrocinnamate and phenyl-lactate. It is thus possible that McyI catalyses a redox reaction leading to the production of one of these compounds (eg. the reduction of phenylpyruvate to phenyl-lactate).
A fourth hypothesis is that McyI is involved in the production of the methylaspartate
(MeAsp) unit of microcystin. Feeding studies with [1,2-13C]acetate suggested that the
MeAsp residues in microcystin and nodularin are synthesised as follows: acetyl-CoA and pyruvic acid condense to 2-hydroxy-2methylsuccinic acid, which is then converted to 2-hydroxy-3-methylsuccinic acid in a manner analogous to the conversion of
(2S)-2-hydroxy-2-isopropylsuccinic acid to (2R, 3S)-2-hydroxy-3-isopropylsuccinic acid in leucine biosynthesis. 2-Hydroxy-3-methylsuccinic acid is then oxidised to
2-oxo-3-methylsuccinic acid and finally transaminated to MeAsp (Moore et al, 1991).
As McyI is a 2-hydroxyacid dehydrogenase homologue, it was hypothesised during this dissertation that this enzyme catalyses the interconversion of 2-hydroxy-3-
95 methylsuccinic acid (3-methylmalate) to 2-oxo-3-methylsuccinic acid (3-methyl oxaloacetate), the penultimate step in methylaspartate biosynthesis (Figure 4.1).
This chapter compares and contrasts McyI and other members of the 2-hydroxyacid dehydrogenase superfamily. A detailed bioinformatic characterisation is described, including comprehensive structural and phylogenetic analyses. Degenerate oligonucleotide primers targeting mcyI have been designed, and used to screen a wide range of toxic and non-toxic strains of cyanobacteria for mcyI orthologues. The recombinant McyI peptide was over-expressed in E. coli, purified and enzymatically characterised using a wide range of proposed substrates. Finally, the ability of mcyI to complement serA (PGDH), ldhA (LDH) and mdh (MDH) mutants has been investigated. The results of these experiments are discussed with respect to the putative role of McyI in microcystin biosynthesis.
96 4.1 EXPERIMENTAL DESIGN
4.1.0 Screening various strains of cyanobacteria for mcyI orthologues
The degenerate oligonucleotide primers mcyIdegenF
[5’-TGTGCGTTATCCTAMTAA-3'] and mcyIdegenR
[5’-GGCTTCTCDCCCTGAAGC-3'] were designed to amplify, via PCR, an approximately 790 bp sequence within the mcyI locus of M. aeruginosa PCC7806 and
Anabaena sp. 90. These primers were used to screen chromosomal DNA samples from several toxic and non-toxic strains of cyanobacteria for mcyI orthologues (Table 4.0).
The same DNA samples were also screened with the M. aeruginosa PCC7806 mcyI- specific primers mcyI(322)F [5’-GGAATGTCAACTACAGCTGTAGCAG-3’] and mcyI(460)R [5’-TTCGAGTTGAATCGGTTGCATCTGA-3’].
4.1.1 Sequence analysis
The 1 011 bp DNA sequence of mcyI (complement of nucleotides 2004-3017, GenBank accession AF183408) was analysed using several different computer programs. The primary amino acid sequence of the corresponding McyI peptide was deduced using
Translate (ExPASy), while theoretical physico-chemical parameters (amino-acid compositions, molecular weight, pI, etc.) were determined using ProtParam (ExPASy).
Determination of potential membrane spanning motifs and hydrophobic moments within McyI was achieved using DAS analysis (ExPASy). The probability of signal peptides within the McyI sequence, and the putative sub-cellular localisation of McyI, were determined using SignalP and PSORTb (ExPASy). Codon usage within the mcyI sequence was assessed with the Kazusa countcodon program and Rare Codon Caltor.
The percentage similarity and identity scores of McyI and other peptide sequences were
97 determined using the PSI-Blast program [National Centre for Biotechnology
Information (NCBI)]. Conserved domains within McyI were detected using the Blast
CD-search (NCBI), ScanProsite (ExPASy), and FUGUE (Shi et al, 2001).
4.1.2 Phylogenetic analysis
A PSI-Blast search with the McyI sequence, returned 26 sequences with Blast scores greater than 200 (Table 4.2). These sequences were subsequently used to generate a multiple sequence alignment (Appendix A4) and corresponding phylogenetic tree
(ClustalX). Four reference sequences from characterised enzymes (PGDH, LDH, MDH and FDH) were also included, as was E. coli NAD-independent D-lactate dehydrogenase (DLD). The latter was designated as an artificial out-group because it is a membrane-bound FAD flavoenzyme and does not belong to the D-isomer specific
2-hydroxyacid dehydrogenase family (Campbell et al, 1984). Phylogenetic trees were generated using the neighbour-joining method with gaps removed. Trees were displayed graphically using njplot and AppleWorks6 (Drawing).
4.1.3 Complementation of E. coli auxotrophs with mcyI
Construction of the complementation plasmids pDRIVE(T7/mcyI) and
pIN-III(lac/mcyI)
For construction of pDRIVE(T7/mcyI), a 1 291 bp fragment containing mcyI was excised from the pET30(mcyI) plasmid (Figure 4.3) with NdeI and NotI and ligated into the pGEM-t easy vector (previously linearised with the same enzymes) to give rise to pGEM(mcyI). A 1 105 bp DNA fragment, containing mcyI was subsequently excised
98 from pGEM(mcyI) with EcoRI and ligated into the pDRIVE vector which had been previously digested with the same enzyme to partially delete lacZ. In the resulting final construct, pDRIVE(T7/mcyI), the transcription of mcyI was driven by the T7 promoter
(Figure 4.2A).
For construction of the pIN-III(lac/mcyI) construct, the DNA fragment containing mcyI including the pET30 N-terminal polyhistidine tag and ribosomal binding site, was excised from pET30(mcyI) with XbaI and HindIII, then ligated into the pIN-III(lppp-5) vector (previously linearised with XbaI and HindIII). The transcription of mcyI in the final construct was driven by the lac promoter (Figure 4.2B).
Expression of McyI in the pDRIVE(T7/mcyI) and pIN-III(lac/mcyI) constructs was inducible with IPTG.
Complementation experiments
The pDRIVE(T7/mcyI) and pIN-III(lac/mcyI) constructs were transferred to several
E. coli strains with mutations in the serA (PGDH), ldhA (LDH) or mdh (MDH) genes
(see Table 2.0). Mutant strains were also transformed with the empty pDRIVE or pIN-
III vectors, which were used as negative controls in all complementation experiments.
Complementation of serA mutants was determined by the ability of the transformants to grow on M9 minimal plates supplemented with glucose (0.4% w/v), appropriate antibiotics, plus 0, 0.25 or 0.5 mM IPTG (+/- 80 µg/ml X-Gal). Complementation of ldhA mutants was determined by the ability of the transformants to grow anaerobically on M9 medium supplemented with the appropriate antibiotic, glucose (0.4% w/v), threonine (50 µg/ml), leucine (50 µg/ml) and vitamin B1 (thiamine, 5 µg/ml) plus 0,
99 0.25 or 0.5 mM IPTG. Complementation of mdh mutants was determined by ability of the transformants to grow on M9 plates supplemented with the appropriate antibiotic,
D/L-Malate (Mal) (0.4% w/v), plus 0, 0.25 or 0.5 mM IPTG.
The expression of recombinant McyI in cells transformed with pIN-III(lac/mcyI) was induced and analysed as described below. Purified McyI was checked for activity (OAA reduction) as described below.
4.1.4 Heterologous expression and purification of McyI
Engineering the pET30(mcyI) expression construct
The oligonucleotide primers mcyIF [5’-ACTACTACTTCACCAAAAACTC-3’] and mcyIR [5’-TAAAAAAGATTCCACGCCTCCGGAT-3’] were used to amplify mcyI from M. aeruginosa PCC7806 chromosomal DNA. The resulting PCR product was cloned into the PCRII-TOPO vector (Invitrogen), excised with EcoRI, gel purified and subcloned into the pET30a expression vector (Figure 4.3). The orientation and sequence of mcyI in pET30a was verified via sequencing with T7prom
[5’-TAAACGACTCACTATAGGG-3’] and T7term
[5’-ACCGCTGAGCAATAACTAGC-3']. The final expression construct pET30(mcyI) was transferred to the E. coli expression host Rosetta(DE3)PlysS which was grown under kanamycin (50 µg/mL) and chloramphenicol (34 µg/mL) selection.
Overexpression of histidine-tagged McyI
Expression cultures were inoculated with 1% overnight starter culture and grown in tryptone phosphate medium supplemented with kanamycin (50 µg/mL) and
100 chloramphenicol (34 µg/mL). Cultures were grown with vigorous shaking (160 rpm) at
37°C to an OD650 of 0.6 and either retained at 37°C or transferred to 30°C or 25°C and incubated for an additional 20 mins (or until an OD650 of approximately 1.0 was reached). Expression of McyI was subsequently induced with 1 mM IPTG for 0-18 hs at
25-37°C. Cell pellets were harvested via centrifugation (5 000 g, for 10 min at 4°C), washed with phosphate buffer [0.5 M NaCl, 20 mM sodium phosphate, pH 7.4] and stored at -80 °C until required.
Purification of histidine-tagged McyI
Cell pellets were thawed on ice, resuspended in 2% of the original culture volume of cold phosphate buffer containing 1 mM PMSF, passed through an 18 gauge needle several times, then briefly sonicated (Branson Sonifier 250, amplitude 25, 50% duty cycle, pulsed). The lysate was cleared via centrifugation (20 000 g for 30 min at 4°C) and precipitated using 40% ammonium sulphate (section 2.10). The final pellet was resuspended in 0.4% of the original culture volume of cold phosphate buffer containing
100 mM imidazole, filtered through a 22 µm membrane, and applied to a 5 mL Hitrap column (Amersham Pharmacia Biotech) charged with Ni2+. The column was washed with 10 column volumes of phosphate buffer containing 100 mM imidazole and eluted with 2 column volumes of phosphate buffer containing 300 mM imidazole. The wash and eluate were collected in 1 mL volumes and analysed via SDS PAGE (12% polyacrylamide gels) and western blotting (section 2.11).
Purified protein samples intended for enzyme assays, were desalted via size exclusion filtration using 10 KDa (Amicron Ultra) molecular weight cut-off columns and 50 mM
HEPES, exchange buffer (pH 7). 101 4.1.5 Determining the subunit organisation of native McyI
The molecular weights of the individual subunits of McyI were estimated by size exclusion chromatography using an AKTA basic 900 series fast protein liquid chromatography (FPLC) apparatus fitted with a Frac-920 fraction collector, a UV detector and a Superdex 200 10/300 column (Amersham Biosciences). The column was equilibrated with wash buffer [50 mM sodium phosphate buffer, pH 7; 150 mM NaCl] then calibrated with ~1 mg each of seven molecular weight standards (Sigma) (Figure
4.8). One milligram of purified McyI was subsequently loaded. All samples were run at a flow rate of 0.5 mL per minute. Eluted proteins were detected spectrophotometrically
(220-280 nm) and collected in 250 µL fractions. Standard curves (log of subunit molecular weight vs. volume eluted) were based on the monomeric molecular weights of the molecular weight standards.
The fractionated McyI subunits were diluted to 5 µg/mL, and checked for activity by performing an OAA reductase assay (see below).
4.1.6 Enzyme assays
Dehydratase assays
In order to test whether McyI was involved in the production of dehydroalanine in
[MeDha7]microcystin-LR biosynthesis, the reverse reaction was attempted in vitro (i.e. the hydration of [MeDha7]microcystin-LR). Each 100 µL reaction contained 1 µg enzyme, 16.75 µM [MeDha7]microcystin-LR, and 0.25 mM ATP in reaction buffer
102 [1 mM DTT, 1 mM PMSF, 100 mM HEPES, pH 7]. Reactions were incubated at 30°C for 1 h and analysed via MALDI-TOF mass spectroscopy.
Positive-ion mass spectra were recorded using a Voyager DE STR MALDI-TOF mass spectrometer (Applied Biosystems). Half a microlitre of sample was combined with
1µL of matrix [α-cyano-4-hydroxycinnamic acid in 80% acetonitrile, 0.08% trifluoroacetic acid] and applied directly to the target plate. Analyses were performed in the positive-ion mode, producing mainly singly protonated molecular ions ([M+H]+).
The acceleration voltage was set at 20 kV. All measurements were carried out in the delayed extraction and reflection mode, allowing the determination of monoisotopic mass values (m/z, mass-to-charge ratio). A low mass gate of 450 Da was used to filter out the most intensive matrix ions.
Chemical synthesis of 2-Hydroxy-3-methylsuccinic acid
2-hydroxy-3-methylsuccinic acid (3-MeMal) was synthesized by reduction of diethyl oxalopropionate [10 g, 0.05 mole (Aldrich)] in 95% ethanol (100 ml) by the addition of a large excess of sodium borohydride (3.8 g, 0.1 mole, 8 fold excess). After stirring the mixture at room temperature for 24 h, 4M HCl (20 ml) was slowly added and the most of the ethanol evaporated under reduced pressure. Saturated sodium chloride solution
(100 ml) was added and the mixture extracted five times with ethyl acetate (100 ml).
The combined extracts were filtered, dried (MgSO4) and evaporated, yielding a viscous
1 oil (4.5 g, ~45%). The H NMR spectrum (CDCl3), showed signals for the threo- isomer
(3-Me δ 1.07 ppm, d J 7.2 Hz ; H-3, m δ 2.9) and the erthro- isomer (H-3, m δ 3.0 ppm) . The ratio of the threo to erythro isomers was 1:6. The other signals of the two isomers (erythro and threo H-2s, and the erythro 3-Me signals) are underneath the CH2
103 resonances (δ 4.1-4.3 ppm) and the CH3 resonances (δ 1.1-1.2 ppm) of the ethyl ester groups.
Hydrolysis of the crude esters (2.0 g) was carried out by refluxing with 4M HCl (50 ml) for 4 h. The cooled mixture was diluted with water (100 ml) and extracted with ethyl acetate (2 × 50 ml) to remove any un-hydrolysed esters and some unsaturated acids formed. The aqueous phase was evaporated to dryness, giving an oil (1.2 g). This was left to stand at -20°C for 72 h after approximately half had crystallized. The oily crystals were washed briefly with cold chloroform, in which the hydroxymethylsuccinic acids are insoluble, leaving a colourless crystalline solid (0.54 g). The 600 MHz 1H NMR
2 ( H2O) now showed only the clearly resolved resonances for the two isomers, threo: erythro in the ratio 1:6. Threo isomer H-2 δ 4.54 ppm, d J 4.05 Hz ; H-3 δ 2.91 ppm , m ; 3-Me δ 1.0 ppm, d J 7.14 Hz. Erythro isomer H-2 δ 4.28 ppm, d J 4.17 Hz ; H-3 δ
2.97 ppm, m ; 3-Me δ 1.05 ppm, d J 7.20 ppm. The latter assignments are identical to those reported for the erythro isomer (Herter et al, 2002). The resonances are also consistent with earlier NMR data, recorded at lower fields and so lacking some of the dispersion reported here for both the free acids and some derivatives (Baht et al, 1985;
Renaud et al, 1987; Kakinuma et al, 1993). No attempts were made to separate these two isomers and they were used as inhibitors in the ratios indicated above.
Oxidoreductase assays
Dehydrogenase and reductase activities of recombinant McyI were analysed in triplicate in 1 mL cuvettes. The production or disappearance of NAD(P)/H was monitored spectrophotometrically at 340 nm using a Cary100 UV spectrophotometer (Varian).
Reactions were initiated by the addition of substrate or enzyme. 104 Dehydrogenase activities of McyI were measured by monitoring the conversion of
NADP to NADPH. Each reaction contained 1 mM NADP, 10 mM substrate (3-MeMal,
D-Mal, L-Mal, D-HGA, D-3PGA) and 50 µg purified McyI in 1 mL assay buffer. The production of NADPH was monitored spectrophotometrically at 340 nm for 0.5-1 min at 37°C.
Reductase activities of McyI were measured by monitoring the conversion of NADPH to NADP. Each reaction contained 0.25 mM NADPH, 5 mM substrate (αKG or OAA) and 2-5 µg purified McyI in 1 mL assay buffer. The disappearance of NADPH was monitored spectrophotometrically at 340 nm for 0.5-1 min at 37°C.
End-product analysis
The end-product of the Mal/OAA oxidoreductase reactions were analysed by nuclear magnetic resonance (NMR) spectroscopy. NMR spectroscopy was carried out on a
Bruker Advance Series DMX-600 spectrometer using an inverse TBI probe. 1H NMR spectra were acquired at 600.13 Mhz with a 90° pulse of 9 µS, 8 992 Hz spectral width,
33 000 data points and a 2 s delay between pulses. Samples were run in 0.5 mL of 95%
2 potassium phosphate buffer pH 7, and 5% H20. The standard Bruker water saturation pulse program was used. The free induction decay was zero filled to 64 K and processed with a line-broadening factor of 1 Hz before Fourier transformation.
105 Determining the optimal pH of the McyI OAA reductase assay
The pH optimum for the McyI OAA reductase assay was determined by monitoring the disappearance of NADPH at 340 nm over a pH range of 5.5 to 9 at 25°C. Each 1 mL reaction contained 0.25 mM NADPH, 0.8 mM OAA and 5 µg purified McyI in 1 mL assay buffer [1.0 mM DTT; 0.1 mM PMSF; 100 mM buffer (MES, pH 5.5-6.5, HEPES, pH 7-8; or Tris-Cl, pH 8.5-9)].
Determining the optimal temperature and thermostability of the McyI OAA reductase assay
The temperature optimum for the McyI OAA reductase assay was determined by monitoring the disappearance of NADPH at 340 nm over a temperature range of 25-
60°C. Each 1 mL reaction contained 0.25 mM NADPH, 0.8 mM OAA and 5 µg purified McyI in 1 mL assay buffer (pH 7). The stability of McyI at different temperatures was determined by incubating the enzyme for 3 min at 25-60°C then chilling it immediately on ice prior to performing the OAA reductase assay at 37°C as described above.
McyI specificity
Various structurally related compounds (lactate, pyruvate, D/L-Ser, L-Phe, oxalic acid, cinnamate, phenylacetate, phenyl-lactate, phenylalanine and diethyloxalpropionate [0.1-
10 mM]) were tested as substrates of McyI in the forward and reverse directions at 37°C for 1 min, pH 7, as described previously.
106 To test wether McyI was able to use unphosphorylated cofactors, the 3-MeMal, Mal,
OAA and αKG oxidoreductase assays were repeated using NAD/H in place of
NADP/H.
McyI Kinetic analysis
Kinetic analysis of the αKG and OAA reductase activities of McyI was carried out in
1 mL reaction mixtures each containing 0.25 mM NADPH, 5 µg purified enzyme and various amounts of substrate in reaction buffer (pH 7). Initial velocities were measured by monitoring the disappearance of NADPH at 340 nm at 37°C for 0.5 min.
Kinetic analysis of the NADPH oxidase activity of McyI was performed in 1 mL reactions each containing 5 µg purified enzyme, 0.8 mM OAA and various amounts of
NADPH in reaction buffer (pH 7). Initial velocities were measured as above.
Inhibition assays
The inhibitory effects of various substances on the OAA and αKG reductase activities of McyI were tested. Assays were performed in 1 mL of reaction buffer (pH 7), containing 0-8 mM inhibitor (L-serine, 3-MeMal, L-Mal, D-Mal, L-Asp, D-Asp,
D-HGA or D-3PGA), 0.25 mM NADPH, 5 µg enzyme, and 0.2 mM OAA or 1 mM
αKG. The disappearance of NADPH was monitored spectrophotometrically at 340 nm for 0.5-1 min at 37°C.
107 Cell extract studies
To investigate the role of McyI in D-MeAsp production during microcystin biosynthesis, the purified enzyme was incubated in a cytoplasmic extract of P. agardhii
CYA126 (known to produce [D-Asp3]microcystin-RR), and analysed by MALDI-TOF
MS. Briefly, 2 mL of late logarithmic culture of P. agardhii CYA126 was harvested via centrifugation (4 000 g, 5 min), resuspended in assay buffer and lysed using 3 successive rounds of freeze/thawing in liquid nitrogen, followed by bead beating
(FastPrep, Savant Bio 101 with 0.1 mm zircon silica beads). The lysate was cleared via centrifugation (10 000 g, 10 min, 4°C) and 100 µL of the resulting supernatant incubated at 30°C for 1 h with 5 µg purified McyI enzyme and 1 mM NADP. The in vitro production of [D-Asp3]microcystin-RR was subsequently analysed using MALDI-
TOF MS as previously described (dehydratase assays section 4.1.6).
108 4.2 RESULTS
4.2.0 Distribution of mcyI orthologues in various species of cyanobacteria
In order to assess the distribution of mcyI orthologues among toxic and non-toxic cyanobacteria, DNA from 24 different strains was screened by PCR with specific and degenerate oligonucleotide primers. The results are presented in Table 4.0. Most microcystin-producing strains (~70%) tested positive for mcyI orthologues using the
Microcystis specific oligonucleotide primers. All microcystin-producing strains, except for P. agardhii CYA126, tested positive for mcyI using the degenerate oligonucleotide primers. Orthologues of mcyI were not detected in the nodularin-producing strains.
Nearly half of all the non-toxic cyanobacteria screened also tested positive for mcyI. All strains possessing mcyI also tested positive for other mcy genes (as determined by mcyA, mcyH or mcyJ screening).
4.2.1 Sequence analysis of mcyI
The 1 014 bp mcyI open reading frame was predicted to encode a 36.71 KDa peptide with a pI of 5.62. While DAS analysis indicated two possible membrane-spanning domains within McyI (between residues 123-125 and 163-164), these were in the low confidence range (Figure 4.4). The Kyte-Doolittle (ProtScale) plot suggested that the
McyI peptide is largely hydrophilic, although several small hydrophobic regions were identified (Figure 4.4A).
SignalP analysis could not identify a signal peptide within the N-terminal sequence of
McyI, suggesting that it is not a secretory protein. PSORTb gave a cytoplasmic localisation score of 9.26 indicating that McyI is a cytoplasmic protein. Codon usage for
109 McyI is shown in Table 4.1. The mcyI sequence contains 32 rare codons including several repeated and/or consecutive rare codons. Subsequent heterologous expression studies were therefore carried out in the Rosetta(DE3)pLysS expression strain containing the pRARE plasmid encoding these rare E. coli tRNA codons.
Comparison of the inferred primary peptide sequence of mcyI with other sequences in the NCBI database revealed significant similarity [up to 72% identity (I) and 84% similarity (S), using PSI-Blast] to putative and experimentally characterised enzymes, mainly PGDHs, of bacterial (mostly cyanobacterial), archaeal and eukaryotic origin.
The top 5 PSI-Blast results were sequences from N. spumigena (71% I, 84% S),
Anabaena sp. (72% I, 84% S), Methanopyrus kandleri (43% I, 64% S), Rubrobacter xylanophilus (42%I, 60% S) and Methanothermobacter thermautotrophicus (41% I,
61% S). A genomic Blast was also performed to identify previously characterised
E. coli homologues. The returned sequences included 2-keto-D-gluconate reductase,
PGDH, LDH, and erythronate-4-phosphate dehydrogenase, although scores were generally quite low (≤32% I and 50% S). BL2seq analysis was also performed for McyI and E. coli PGDH (accession P08328), and returned scores of 31% I and 48% S. The individual domains of McyI (putative nucleotide and substrate domains) were also analysed by BLASTp, however results generally reflected those of the full-length
BlastP search (i.e. no new sequences were returned).
A FUGUE homologous structure search using the McyI peptide sequence yielded similar results to those observed with the Blast searches. High confidence Z-scores were returned for several 2-hydroxyacid dehydrogenases including PGDH from
Mycobacterium tuberculosis (Z score, 69.7; PDB accession, 1ygy), an NAD regulated
110 dehydrogenase from H. sapiens (62.65; 1mx3), D-glycerate dehydrogenase from
Hyphomicrobium methylovorum (49.99; 1gdh), formate dehydrogenase (FDH) from
Pseudomonas sp. 101 (49.99; 2nac), PGDH from E. coli (21.54; 1psd), LDH from
Lactobacillus helveticus (11.95; 2dld), and D-2-hydroxyisocaproate dehydrogenase from Lactobacillus casei (9.44; ldxy).
A ScanProsite pattern search also identified D-isomer specific 2-hydroxyacid dehydrogenase NAD-binding signatures (LIVMA)-(AG)-(IVT)-(LIVMFY)-(AG)-x-G-
(NHKRQGSAC)-(LIV)-G- x(13,14)-(LIVMFT)-x(2)-(FYwCTH)-(DNSTK) at residues
158-186 of the McyI peptide (Figure 4.5). Enzymes possessing these signatures are structurally related and have similar enzymatic activities (Taguchi and Ohta, 1991;
Kochhar et al, 1992; Goldberg et al, 1994).
The sequence alignment of McyI with other cyanobacterial homologues and the E. coli
PGDH, revealed that McyI from M. aeruginosa and Anabaena sp., and NdaH
(N. spumigena) lack approximately 72 amino acids that comprise the corresponding
C-terminal regulatory (ACT) domain, making them, on average, 18% shorter than
E. coli PGDH. Several other sequence variations between McyI and PGDH were also apparent (Figure 4.5). Trp-139 of E. coli PGDH, was not conserved in McyI or NdaH.
In its place was a tyrosine residue (Tyr-142). Another sequence variation between
PGDH and McyI was the Gly-294 substitution for Ala. Glycine-336, Gly-337, Arg-338 and Arg-339 were not present in the McyI sequence, as these residues also form part of the PGDH ACT domain, which McyI lacks.
111 The putative active site of McyI also appeared to be different from that of PGDH. While the Glu-269-His-292 charge-relay system, the core of the active site of PGDH (Grant et al, 1999), was conserved in McyI, Lys-39, Arg-62, Lys-141 and most of the cationic residues which help bind the negatively charged substrate, were not present. Lysine-39, while present in the Anabaena (McyI) and Nodularia (NdaH) sequences, was replaced with serine in M. aeruginosa. Both McyI and NdaH enzymes have a Pro substitution at the Arg-62 position. Most cyanobacterial dehydrogenases contain a Gly at this site, as does PGDH from Arabidopsis thaliana (appendix A4). Lysine-141 was replaced by an
Arg in McyI, NdaH and all other cyanobacterial dehydrogenases examined. The
A. thaliana PGDH also contains an Arg at this position.
To gain an overview as to the relative phylogenetic position of McyI within the
2-hydroxyacid dehydrogenase family, a phylogenetic analysis was performed. Sequence lengths used in the analysis ranged from 306-624 amino acids (Table 4.2). The phylogenetic tree (Figure 4.6) partitioned into three main subgroups. Subgroup A, which contained McyI, was comprised exclusively of sequences from the hepatotoxin biosynthesis gene clusters of cyanobacteria. Subgroup B was comprised of putative dehydrogenases from Arabidopsis thaliana, several species of cyanobacteria, and the thermophylic actinobacterium Rubrobacter xylanophilus. Subgroup C consisted mostly of archaeal proteins, with the exception of two eubacterial sequences,
Thermoanaerobacter tengcongensis (order Clostridiales) and Oceanobacillus iheyensis
(order Bacillales).
Approximately one third of all sequences in the phylogenetic tree possess amino acid
112 C-terminal regulatory domains. While these regulatory domains were present in each organismic lineage they were most prevalent in the plant and archaeal subgroups.
Clustering of ACT-possessing proteins within each organismic lineage was apparent.
The tree also demonstrated that groups B and C shared a common ancestor that had an earlier divergence leading to the group A lineage of cyanobacterial proteins. The distribution of the reference sequences within the tree supported the prediction that the
A/B/C lineage evolved from a PGDH common ancestor rather than an ancestral MDH,
LDH or FDH.
4.2.2 Mutant complementation studies
In an attempt to characterise McyI in vivo, a series of complementation experiments were performed. Two mcyI complementation vectors pDRIVE(T7/mcyI) and pIN-III(lac/mcyI), were transformed into E. coli strains with mutations in either the serA
(PGDH), ldh (LDH) or mdh (MDH) genes (Table 2.0). The serA mutants transformed with pDRIVE(T7/mcyI) and pIN-III(lac/mcyI) were unable to grow on M9 medium lacking serine. Untransformed cells, and cells carrying pDRIVE(T7/mcyI), pIN-
III(lac/mcyI) or the empty backbone pDRIVE and pIN-III control vectors, were all able to grow on M9 medium supplemented with serine.
Interestingly, cells transformed with the pDRIVE(T7/mcyI) construct grew as blue colonies in the presence of X-Gal and IPTG. These results indicate that the lacZ gene can tolerate large in-frame deletions/insertions and still encode functional
β-galactosidase.
113 The ldhA mutants transformed with pDRIVE(T7/mcyI) and pIN-III(lac/mcyI) were unable to grow anaerobically on M9 medium with glucose as a sole carbon source.
Untransformed cells, and cells carrying pDRIVE(T7/mcyI), pIN-III(lac/mcyI) or the empty pDRIVE and pIN-III control vectors, were able to grow on M9 supplemented with pyruvate.
The mdh mutants transformed with the pIN-III(lac/mcyI) were able to grow on M9 with malate as a sole carbon source. However, untransformed cells, and cells carrying the empty pIN-III control vector also grew on M9/malate with equivalent cfu/mL, indicating that the mdh mutation was highly unstable.
McyI purified from the aformentioned E. coli auxotrophs transformed with pIN-
III(lac/mcyI), was able to reduce OAA at comparable rates to McyI purified from the expression strain (Rosetta(DE3)pLysS + pET30(mcyI), see below).
4.2.3 Over-expression and Purification of histidine-tagged McyI
McyI was overexpressed in E. coli with an N-terminal hexa-histidine fusion tag. The recombinant peptide was highly expressed under all culture/induction conditions tested
(4.1.4). Western blot analysis confirmed that McyI was expressed without truncation, at the expected molecular weight of ~40 KDa. Maximum yields of soluble protein (up to
10% total protein) were obtained after overnight induction at 25°C. However, a 2.5 h induction at 30°C also provided sufficient yields (5-10% total protein) for enzyme assays.
114 The recombinant McyI protein precipitated in the presence of 40% ammonium sulphate, and eluted from the Ni2+ charged affinity column at an imidazole concentration of ~300 mM. The recombinant protein eluate was of >95% purity and was eluted at concentrations of 1-2.5 mg/mL (Figure 4.7).
4.2.4 Subunit composition of McyI
The purified McyI protein eluted from the size exclusion column at 27.8 min, corresponding to a molecular weight of 85 KDa (dimeric form). The collected elution fractions (25-30 min) displayed OAA reductase activity of ~12.5-14 µmol/min/mg protein, which were comparable to that of the unfractionated enzyme (Figure 4.8).
4.2.5 Biochemical and kinetic properties of McyI
[MeDha7]microcystin-LR hydratase assays
In order to test whether McyI was involved in the dehydration of Ser to Dha during
[MeDha7]microcystin-LR biosynthesis, the reverse reaction was attempted in vitro.
MALDI-TOF analysis of the assay mixture revealed a mass peak of 994.5 corresponding to [MeDha7]microcystin-LR, however a mass peak of 1 030.53
[994.5+(2×18)] corresponding to [MeSer7]microcystin-LR could not be identified.
Oxidoreductase assays
McyI was stable for more than two weeks at 4°C without significant loss of activity, and for at least six months at –20°C in 15% glycerol. Plots of McyI OAA reductase activity after 1 min incubations at various temperatures were bell-shaped with an optimum of
115 ~45°C at pH 7. However, McyI was relatively unstable at higher temperatures (90% drop in activity following pre-incubation at 50°C). The pH optimum for the McyI OAA reductase activity was 7, however, the enzyme maintained at least 80% of its activity between pH 6.5-7.5 at 25°C (Figure 4.9).
McyI displayed weak 3-MeMal, D-Mal and L-Mal oxidase activities (~0.2-0.7
µM/min/mg protein under the described conditions). However, oxidation of D-HGA or
D-3PGA could not be detected. Due to strong product inhibition and the high concentration of enzyme required for each assay, kinetic constants were unable to be obtained for the oxidase reactions.
app The reduction of OAA was the major in vitro activity observed for McyI, with Vmax of
15.2 µM/min/mg protein at 37° C (Table 4.3, Figure 4.1). However, high concentrations of substrate (>0.8 mM) inhibited the reaction. The enzyme was also able to reduce
app αKG, although the rates of catalysis were much lower (Vmax , ~3.6 µM/min/mg protein). The OAA and αKG reductase reactions both obeyed regular Michaelis-
Menton kinetics (Table 4.3).
McyI was able to utilise phosphorylated and dephosphorylated dinucleotide co-factors although, the rates of catalysis for NADH were around 90% lower than those obtained
app using NADPH (Vmax , ~6.5 µM/min/mg protein) with equivalent amounts of enzyme.
Detailed kinetic data could therefore, not be obtained using NADH.
116 End-product analysis
NMR spectroscopic analysis confirmed that malate was the product of the OAA reductase reaction (Figure S4.C).
Inhibition assays
The OAA reductase activity of McyI was inhibited by D-Mal (I50, 2.2 mM) and
3-MeMal (I50, 0.7 mM) in a negatively cooperative manner. Interestingly, L-Mal did not inhibit the reaction, nor did L-Ser, D-Asp, L-Asp, D-HGA or D-3PGA. The αKG reductase activity of McyI was inhibited by 3-MeMal (I50, 2.3 mM) and D-HGA (I50,
20.5 mM) (Table 4.3, Figure 4.11).
117 4.3 DISCUSSION
In order to assess the role of mcyI in microcystin production, a variety of hepatotoxic and non-toxic strains of cyanobacteria were screened for mcyI orthologues. While there was a positive correlation between microcystin production and the presence of mcyI, the specific primers used in this study (mcyI(322)F/(460)R), failed to detect mcyI in approximately 30% of all toxic strains. These false negative results were probably due to sequence variation between the strains examined.
To overcome problems caused by sequence variation between putative mcyI orthologues, degenerate oligonucleotide primers (mcyIdegenF/R), were designed. With the exception of P. agardhii CYA126, these primers successfully amplified a ~790 bp
PCR fragment from all microcystin-producing strains tested (Table 4.0). While the PCR products were not sequenced, the fact that amplicons were not obtained for
Synechocystis or Nodularia (both of which lack mcyI), suggested that the primers were specific to mcyI and do not amplify ndaH or other 2-hydroxy acid dehydrogenases, such as PGDH.
Interestingly, nearly half of all the non-toxic cyanobacterial strains examined tested positive for mcyI. These results reflected those of the mcyH genetic screening experiments (section 3.0), and those of previous studies that have identified mcy genes in several non-toxic Microcystis strains (Nishizawa et al, 1999; Tillett et al, 2000;
Kaebernick et al, 2001). It is likely that these strains have reverted to a non-toxic phenotype due to mutations within their mcy biosynthesis genes, or within mcy regulatory regions. Previous studies have demonstrated that the mcy gene cluster does not well tolerate mutations with respect to toxin biosynthesis. In fact the mutation of
118 several individual mcy genes (mcyA, B, D, E, F, and H) has led to the production of mutant strains with non-toxic phenotypes (Dittmann et al, 1997; Tillett et al, 2000;
Nishizawa et al, 2001).
It is interesting to note that P. agardhii CYA126 does not possess mcyI, yet it is still capable of producing microcystin (Christiansen et al, 2003). McyI is therefore not critical for all types of microcystin production, but may be required for the biosynthesis of certain microcystin isoforms. The most frequently reported microcystin isoforms contain D-MeAsp at position 3, yet P. agardhii CYA126 only produces the less toxic
[D-Asp3]microcystin variants (Sivonen and Jones, 1999). McyI may therefore be specific to strains producing [D-MeAsp3]microcystin. The fact that mcyI has only been sequenced from strains that are capable of producing [D-MeAsp3]microcystin
(i.e. M. aeruginosa and Anabaena sp. 90), further supports this supposition. If mcyI genes can be identified specifically in strains that produce [D-MeAsp3]microcystin
(eg. P. agardhii CYA127 and CYA128, Luukkainen et al, 1993), a strong argument could be made in support of the involvement of McyI in D-MeAsp biosynthesis. It may also prove valuable to screen nodularin producing cyanobacteria for ndaH homologues, as only a few different isoforms of this toxin have been reported, one of which contains
D-Asp in place of D-MeAsp at position 1 (Namikoshi et al, 1992).
While over 90% of the microcystin-producing cyanobacteria screened in this study tested positive for mcyI orthologues, the fact that P. agardhii CYA126 tested negative, suggests that mcyI is not an ideal genetic locus for predicting strain toxicity in environmental samples and drinking water supplies.
119 The inferred primary peptide sequence of McyI was analysed by several different computer programs in an attempt to gather structural and functional information for subsequent biochemical experiments. Homology-based searches indicated that McyI is a cytosolic enzyme belonging to the 2-hydroxyacid dehydrogenase family. While its closest characterised relative, A. thaliana PGDH, is involved in Ser metabolism, sequence data suggested an alternative function for McyI.
An immediately obvious difference between the McyI peptide sequence and that of bacterial PGDHs, was its lower subunit molecular weight (37 vs 44 KDa). This was largely due to the fact that McyI lacks an ACT domain (residues 338-409 in E. coli
PGDH, Aravind et al, 1999). These regulatory domains have been linked to a wide range of metabolic enzymes that are controlled by amino acids in a concentration- dependant fashion (Schuller, 1995). Aravind et al (1999) postulated that these
C-terminal domains are evolutionarily mobile modules that have been independently fused to a variety of enzymes.
A key region within the ACT domain of PGDH, is the effector-binding site. Within this groove, residues Val-363, Asn-364, Asn-346, His-344 and Thr-352 come together in such a way as to allow hydrogen bonding with Ser (Al-Rabiee et al, 1996). As McyI lacks these residues and has no comparable site for Ser, or any other amino acid to bind, it is highly unlikely that McyI is regulated by an amino acid. An alternative regulatory mechanism, such as transcriptional control, may negate the need for an ACT domain in
McyI and other cyanobacterial enzymes (eg. those present in the phylogenetic tree,
Figure 4.6). The fact that McyI has two promoters: that is, its own individual promoter
120 as well as the mcyD-J promoter, strongly supports this hypothesis (Kaebernick et al,
2000, 2002).
A close comparison of the primary peptide sequence of McyI with other 2-hydroxy acid dehydrogenases in the databases revealed further interesting characteristics of the protein. A large proportion of amino acids in McyI are highly conserved among most of the other dehydrogenases examined, including the NAD(P)-binding signature encompassing residues 158-186 of the McyI sequence. This nucleotide-binding domain
(NBD) is conserved among all 2-hydroxy acid dehydrogenases including PGDH, MDH,
LDH and glycerate dehydrogenase. Residues that are particularly important within the
NBD of PGDH are His-292, Glu-269 and Arg-240. The His-Glu pair probably serves as a proton shuttle during catalysis, while Arg-240 interacts with the carboxylic acid of the substrate during catalysis (Schuller, 1995). These three residues are almost universally conserved among 2-hydroxy acid dehydrogenases (Grant, 1989 and Goldberg, 1994).
The identification of a NBD within McyI suggested that it is an NAD(P)-dependant enzyme with an oxidoreductase function similar to that of PGDH.
In addition to the residues comprising the NAD(P)-binding signature, the functions of a number of amino acids in various 2-hydroxy acid dehydrogenases have been determined by mutagenesis. As PGDH is the closest characterised relative of McyI, the sequences of both proteins were aligned and compared in an attempt to make structural/functional predictions for McyI. McyI and PGDH shared an overall sequence similarity of around 52%. While most of the important residues within PGDH were conserved or identical in McyI, several interesting amino acid substitutions were apparent.
121 The cationic residues that are thought to form an electrostatic environment for the binding of the negatively charged substrate at the active site of PGDH include Lys-39,
Arg-60, Arg-62, Lys-141 and Arg-240. The results of mutating these residues in PGDH showed that Arg-60, Arg-62, Lys-141 and Arg-240 play distinct roles in the binding of the substrate to the active site (Grant, 1999). Arg-240 was conserved in McyI and all other sequences in the alignment. This basic residue plays an important role in anchoring the C1-hydroxyl group of 3-PGA during catalysis. Similar basic residues are present in MDH and LDH. The additional basic residues (Arg-62, and Lys-141) are unique to PGDH and may interact with an acidic group at the distal end of the substrate
(Grant, 1999). The fact that McyI possesses Arg-240, but lacks Arg-62, and Lys-141, suggested that a 2-hydroxy acid other than 3-phosphoglycerate is the substrate of this cyanobacterial enzyme.
In PGDH Arg-60, Arg-62 and Lys-141work in tandem to bind the phosphate group of the substrate (Grant, 1999). In the M. aeruginosa, Anabaena sp. and N. spumigena sequences Arg-60 was conserved, however, Arg-62 and Lys-141 were not. In McyI and
NdaH, Arg-62 was replaced by Pro. This cyclic amino acid, which can markedly influence protein architecture, also lacks the hydrophilic charged side-chain of Arg.
Despite this non-conservative substitution, the secondary structures of McyI, NdaH and
PGDH did not appear to differ at this site, with each protein possessing a helix-loop- sheet composition. The fact that A. thaliana PGDH contains a glycine in place of Arg-
62, as do most of the other cyanobacterial dehydrogenases, suggested that certain substitutions at this position may not be critical to activity, since the A. thaliana PGDH can complement E. coli PGDH (serA) mutants (Ho et al, 1999). Lysine-141 was
122 replaced by Arg in McyI, as well as in most other sequences in the alignment. This is a conservative substitution (both amino acids are positively charged) and probably has little effect on the overall structure and function of the enzyme. In summary, in PGDH
Arg-60, Arg-62 and Lys-141 are critical for binding the phosphate group of the substrate (Grant, 1999). While the overall architecture of McyI and NdaH may not be affected by substitutions in this region, the results suggested that these cyanobacterial enzymes preferentially bind non-phosphorylated 2-hydroxyacids.
In E. coli PGDH, Trp-139 participates in inter-subunit contact near the active site catalytic residues where it fits within a hydrophobic pocket created by Pro-270, Pro-291 and Phe-277 (Grant et al, 2000, 1999). While the hydrophobic binding pocket is generally conserved in McyI, Trp-139 is replaced by tyrosine. This may have several implications for the enzyme’s structure and function. While Trp and Tyr are both aromatic amino acids, Tyr lacks an indole ring and has a phenolic group. It has recently been demonstrated that by removing the indole ring of Trp-139, PGDH dissociates into dimers (Grant, 2000). We therefore hypothesised that McyI functions as a dimer.
Subsequent analysis of McyI by size exclusion chromatography supported this hypothesis. As Trp-139 also plays a key role in the cooperativity of Ser binding and inhibition, we predict that McyI is a non-cooperative enzyme.
Histidine-292 is important for catalysing the interconversion of 3-PHP and 3-PGA in
PGDH. In addition, Glu-269 acts in tandem with His-292 to form a proton shuttle as seen in many dehydrogenases (Grant, 1999). The conservation of these residues within
McyI again supported an oxidoreductase function for this enzyme.
123 In PGDH, Gly-294 and Gly-295 form the link between the substrate-binding and nucleotide-binding domains which form the active site cleft of the enzyme. Mutations in these residues affect the enzyme’s Kcat without appreciably effecting sensitivity to serine
(Grant, 2001). While Gly-295 was conserved in McyI and NdaH, Gly-294 was replaced by Ala. Mutational studies have shown that Gly-294-Ala substitutions do not affect
PGDH’s catalytic capacity (Grant, 2001), and, therefore, this conservative substitution should not influence the Kcat of McyI or NdaH. Indeed, homologous enzymes can tolerate a number of different amino acids in this position as is evidenced by the alignment (Figure 4.5) in which several sequences have either Ser or Glu in the first position and Lys, Arg or Asn at the second position.
In an attempt to understand the evolution of McyI in the context of other 2-hydroxyacid dehydrogenases such as PGDH, MDH and LDH, a phylogenetic tree was constructed encompassing sequences from plants, archaea and eubacteria, including several species of cyanobacteria (Figure 4.6). Interestingly, the cyanobacterial sequences clustered polyphyletically. While the group B cyanobacterial proteins appeared to share a common ancestor with plant PGDHs, the hepatotoxin-associated proteins formed a phylogenetically distinct subgroup (A). McyI and NdaH must have therefore diverged early in the evolution of PGDHs, prior to the divergence of the archaeal, plant and subgroup B cyanobacterial protein lineages.
A similar phylogenetic study by Ali et al (2004), investigating PGDHs from a wide range of organisms, reported that sequences from the order Bacillales also display a polyphyletic distribution. The authors concluded that many lateral gene transfer events, together with drastic insertion/deletion events, occurred during the evolution of PGDH,
124 making the history of this superfamily very complex. The distribution of ACT domains among 2-hydroxy acid dehydrogenases contributes to this complexity. In the phylogenetic tree presented in this study, it appears that the ancestral PGDH possessed an ACT domain, yet this domain was lost in some, but not all, subsequent generations.
It is interesting to note that even proteins from closely related organisms (eg.
Synechocystis sp. PCC6803 and Synechocystis sp. WH8102) may differ when it comes to possession of an ACT domain. These results suggested that alternative regulatory mechanisms such as transcriptional control, must be in place for many of these enzymes.
While none of the reference sequences (PGDH, LDH, MDH, FDH) partitioned within groups A-C, the fact that PGDH from A. thaliana fell within group B, suggested that the group B cyanobacterial sequences, and probably the archael sequences, also function in serine metabolism. Conversely, the distantly related group A cyanobacterial proteins
(McyI and NdaH) are likely to have evolved functions specific to hepatotoxin biosynthesis.
As McyI displayed sequence homology to the previously characterised 2-hydroxy acid dehydrogenases, PGDH and LDH (~48% similarity), a series of complementation experiments were performed in order to test the PGDH, LDH and MDH activities of
McyI in vivo. The inducible McyI expression/complementation construct pDRIVE(T7/mcyI), was transformed into 3 different E. coli strains with mutations in serA, ldhA and mdh. Unexpectedly, mutants transformed with the pDRIVE(T7/mcyI) construct grew as blue colonies in the presence of X-Gal and IPTG, indicating the expression of a β-galactosidase/McyI fusion product with residual β-galactosidase
125 activity despite the partial deletion of β-galactosidase in pDRIVE(T7/mcyI). An alternative expression/complementation construct, pIN-III(lac/mcyI), was therefore designed and used in subsequent complementation studies.
McyI was able to be expressed in, and purified from each of the mutant strains transformed with pIN-III(lac/mcyI). The enzyme was also shown to possess αKG and
OAA reductase activities in vitro. However it was unable to complement the serA or ldhA mutants. As McyI and PGDH are united by their ability to reduce αKG, and are phylogenetically related, it was predicted that McyI might also be able to oxidise
3-PGA in vivo. However, this proved not to be the case. The inability of mcyI to complement the serA mutation was surprising, as previous complementation attempts with more distantly related PGDH homologs have been successful. For example, the gene encoding PGDH in A. thaliana was able to complement serA, despite the relatively low sequence similarity between the E. coli and A. thaliana enzymes (30% identity,
46% similarity) (Ho et al, 1999).
McyI was also unable to complement ldhA mutants. This result was less surprising as
McyI is only distantly related to the E. coli LDH, and was unable to oxidise lactate or reduce pyruvate in vitro. Unfortunately, the mdH mutants were genetically unstable and continually reverted to the wild-type phenotype. We therefore could not assess the ability of mcyI to complement mutations at this locus.
The sequence-based data and genetic screening experiments described above suggested that McyI is a 2-hydroxyacid dehydrogenase with a function specific to microcystin production. However, in order to understand the precise physiological role of this
126 enzyme, biochemical data were required. Therefore, the recombinant protein was overexpressed and purified with the intent of performing detailed in vitro biochemical analyses.
The recombinant McyI protein was expressed at levels detectable by SDS-PAGE (with
Coomassie blue staining) and western blot analysis at all temperatures tested, indicating that temperature, at least within the range of 25-37°C, was not a critical factor for the expression of this protein in its soluble form. The observed solubility of McyI in E. coli, and the monomeric size of the recombinant protein, as determined by SDS-PAGE and size exclusion chromatography, were both in agreement with the bioinformatic predictions that suggested McyI is a hydrophilic, cytoplasmic protein.
Size exclusion chromatography revealed a dimeric subunit organization for McyI. This result was expected given the Trp/Tyr substitution at position 142 within McyI which in
PGDH causes dissociation of the usually tetrameric enzyme into dimers (Grant et al,
2000).
The purified recombinant protein was analysed enzymatically using over 15 different substrates. The hypothesis that McyI dehydrates Ser to Dha in the production of
(MeDha7) microcystin-LR (Tillett et al, 2001) was initially addressed. As many dehydratase enzymes are known to catalyse reversible reactions, and
[MeSer7]microcystin-LR was not available, the reverse reaction: the in vitro hydration of [MeDha7]microcystin-LR to [MeSer7]microcystin-LR, was attempted. However,
Dha7 hydratase activity was not detected under the described assay conditions. These results were not unexpected since the bioinformatic data suggested that McyI is a 2-
127 hydroxy acid dehydrogenase and is totally unrelated to the dehydratase enzymes involved in Dha production in other secondary metabolite pathways, such as siomycin-
A (Tori et al, 1976), nosiheptide (Pascard, 1977), berninamycin (Pearce, 1979), and the lantibiotics nisin (Gross, 1967), subtilin (Gross, 1969) and epidermin (Allgaier et al,
1985).
The recent description of the mcy gene cluster in P. aghardii (Christiansen et al, 2003) has provided additional evidence against the proposed dehydratase role of McyI. While mcyI homologs are absent in this cyanobacterium, it is still capable of incorporating
Dha into the microcystin peptide during toxin biosynthesis (Christiansen et al, 2003).
While the authors discuss the possibility of a distally encoded mcyI homolog, our screening results did not support this hypothesis. The conversion of Ser to Dha is most likely a result of a serine dehydratase encoded elsewhere in the M. aeruginosa genome.
The hypothesis that McyI is involved in Ser production was then perused. While amino acid metabolism in cyanobacteria is a poorly researched topic, there is evidence that these photosynthetic organisms possess both phosphorylated and glycolytic Ser biosynthesis pathways (Bruin et al, 1970; Colman and Norman, 1997). The fact that
M. aeruginosa possesses a PGDH homologue is therefore not surprising. What is interesting however, is the association of this putative primary metabolic enzyme with the microcystin biosynthesis pathway. PGDHs are typically associated and co-regulated with other serine biosynthesis enzymes, for example phosphoserine transaminase and phosphoserine phosphatase. While Trp biosynthesis genes have been located upstream of the daptomycin biosynthesis gene cluster (Baltz et al, 2005), there have been no
128 previous reports of Ser biosynthesis genes encoded proximal to, or within, secondary metabolite gene clusters.
To investigate the remarkable prospect that McyI may be a PGDH, the recombinant enzyme was assayed for 3-PGA dehydrogenase activity in vitro. However, despite extensive experimental variation, PGDH activity was not detected. The lack of PGDH activity also reflected the phylogenetic data presented in this study that suggested that
McyI is evolutionarily distant to PGDH and belongs to a separate clade of enzymes, uniquely present in the hepatotoxic cyanobacteria. The comparative sequence data also supported this notion and predicted distinct functions for McyI and PGDH.
While McyI is therefore clearly not involved in Ser metabolism, it is probable that
M. aeruginosa possesses a true PGDH elsewhere in its genome. Genome sequencing projects and/or genetic screening studies should shed light on this topic in the near future.
A putative role of McyI in the biosynthesis of the starter unit of Adda was then proposed and investigated. While the structure of Adda suggested priming with phenylacetate, in vivo feeding experiments did not support the direct involvement of this phenylalanine-derived starter unit (Moore et al, 1991; Rinehart et al, 1994). Moffitt et al (2006) have recently demonstrated that the polyketide synthase, McyG, preferentially activates an assortment of phenylpropanoids, in particular, cinnamate, hydrocinamate, phenyl-lactate and phenylalanine. It was thus hypothesised during this thesis that McyI may catalyse a redox reaction leading to the production of one of these compounds. To address this hypothesis, a suite of structurally related compounds were
129 tested as substrates for McyI, including cinnamate, phenylacetate, and phenylpyruvate.
No activity was detected using these substrates. It was therefore concluded that McyI is unlikely to be involved in the production of the Adda starter unit. This function may be encoded within an alternative secondary metabolite gene cluster, such as the cyanopeptolin (Martin et al, 1993) or microginin (D. Kramer pers. com.) biosynthesis clusters, which are yet to be characterised.
Bioinformatic data suggested that McyI is a 2-hydroxyacid dehydrogenase, so therefore,
αKG/2-HGA, lactate/pyruvate, and OAA/Mal were tested as substrates for McyI.
Interestingly, McyI was able to utilise αKG/OAA, and Mal as substrates, but not lactate or pyruvate. As with most 2-hydroxyacid dehydrogenases, the major in vitro activity observed for McyI was in the reductase direction, that is, the reduction of OAA to Mal.
app app Rates of catalysis for McyI OAA reductase activity (Km , 0.0973 mM; Vmax , 15 203
app U/mg) were comparable to those for the E. coli PGDH αKG reductase activity [Km ,
app 0.088 mM; Vmax , 11 100 U/mg (Zhao and Winkler, 1996)].
Both phosphorylated and dephosphorylated dinucleotide cofactors were able to be utilised by McyI, yet NADPH was greatly preferred over NADH. Malate dehydrogenases also prefer NADPH, however, the opposite is true for most PGDHs and
L-LDHs. It would be interesting to determine whether the putative cyanobacterial
PGDHs used in this phylogenetic study were also NAD specific. If so, it is likely that the nucleotide binding domain of McyI evolved from a NAD specific form to its current
NADP specific form.
130 As the reduction of OAA was the major observed in vitro activity for McyI, it was predicted that the true physiological substrate of McyI must be an OAA/Mal analogue.
This led to the hypothesis that McyI catalyses the conversion of 3-MeMal to 3-MeOAA, with subsequent transanimation to MeAsp. In order to test this hypothesis, a racemic diastereomeric mixture of 3-MeMal was synthesised and tested as a substrate in an in vitro dehydrogenase assay. A 6:1 mixture of erythro- and threo-3-methyl malates ( i.e. a
6:1 mixture of 2R,3S plus 2S,3R and 2R,3R plus 2S,3S forms) was synthesised by borohydride reduction of commercially available diethyl oxalpropionate which was then hydrolysed by acid to the mixture of hydroxy acids. The acids were used, after neutralisation, as mixtures, so half of the erythro- form (~45% of the total mixture) would have had the expected configuration of the intermediate in the biosynthetic pathway. Unfortunately, attempted synthesis of the unstable 3-MeOAA, to test the reverse reaction (3-MeOAA to 3-MeMal) was not successful. McyI was able to oxidise
3-MeMal at a comparable rate to D- and L-Mal, however, there appeared to be strong end-product inhibition, which significantly reduced the linear range of the assay (<30 sec).
To understand the inhibitory effects of various substrates on the McyI reductase reactions, several inhibition assays were performed. Unlike PGDH, McyI was not allosterically regulated by Ser or by any other compound tested. This result was not surprising since McyI lacks an ACT regulatory domain. 3-MeMal and D-Mal inhibited
McyI OAA reductase activity in a negatively cooperative fashion. 3-MeMal was the strongest inhibitor with an approximate I50 of 0.7 mM. Interestingly, L-Mal had no inhibitory effects, suggesting that McyI, like other 2-hydroxyacid dehydrogenases, preferentially binds substrates with D-conformations. It is therefore likely that the I50 for
131 3-MeMal is overestimated since the 3-MeMal compound was a diastereomeric racemic mixture. These results support the hypothesis that McyI is a 3-MeOAA/3-MeMal oxidoreductase.
The non-proteinogenic amino acid MeAsp is a common intermediate of the mesaconate pathway for (S)-glutamate fermentation in Clostridium spp. (Buckel, 2001) and members of the family Enterobacteriaceae (Kato and Asano, 1997). This unusual amino acid residue also occurs in the lipopeptide antibiotic friulimicin (Heinzelmann et al, 2003) and in the cyanobacterial hepatotoxins microcystin and nodularin (Tillett et al,
2000, Moffitt and Neilan, 2003). In the mesaconate and friulimicin pathways, the route to MeAsp occurs via rearrangement of Glu to 3-MeAsp involving Glu-mutase.
However, genome sequencing efforts have failed to identify glutamate mutase homologues in cyanobacteria.
Labeled-precursor feeding experiments by Moore et al (1991), suggested that the
MeAsp residues in microcystin and nodularin are produced via an alternative biosynthetic route involving the rearrangement of citramalic acid, yet the authors were unable to identify the enzymes involved in this pathway. In this study, we provide evidence that the 2-hydroxyacid dehydrogenase McyI, is central to MeAsp biosynthesis in cyanobacteria and catalyses the interconversion of 3-MeMal to 3-MeOAA. The final step in the MeAsp pathway, the conversion of 3-MeOAA to MeAsp, is likely to be catalysed by a promiscuous Asp aminotransferase, as these enzymes occur in virtually all organisms, including cyanobacteria (Kim et al, 2002).
132 To our knowledge, McyI is the only reported example of a 3-MeMal oxidase. Based on the data presented in this study, it appears that McyI and it’s homologue, NdaH, have diverged from a PGDH ancestor to become NADP dependent enzymes with specific roles in hepatotoxin production. Future mutagenesis experiments involving mcyI and ndaH will undoubtedly offer further insight into the role of these unique cyanobacterial enzymes.
The combined results of this study supported the putative role of McyI in MeAsp production during microcystin biosynthesis. Numerous lines of evidence including sequence-based, molecular, and biochemical analyses, supported the hypothesis that
McyI is a unique 2-hydroxyacid dehydrogenase that catalyses the interconversion of 3-
MeMal and 3-MeOAA.
133 CHAPTER 5.
CHARACTERISATION OF THE
O-METHYLTRANSFERASE, MCYJ
134 5.0 OVERVIEW
The fifth open reading frame in the mcyD-J operon of the microcystin biosynthesis gene cluster of M. aeruginosa encodes an 837 bp gene, mcyJ. Preliminary sequence analysis of the inferred primary peptide sequence of mcyJ by Tillett et al (2000), revealed a 35% identity (51% similarity) to the erythromycin synthase O-methyltransferase EryG, from
Saccharospora erythraea (Haydock et al, 1991). This enzyme catalyses the terminal step in erythromycin biosynthesis, the conversion of erythromycin C to erythromycin A by O-methylation at C3 (Weber et al, 1989; Paulus et al, 1990). Tillett et al (2000) hypothesised a similar function for McyI in microcystin biosynthesis, the O-methylation of the C9 hydroxyl on the polyketide side-chain, Adda.
Methytransferases are ubiquitous across all forms of life. There are at present, at least three structurally defined types of SAM-dependent methyltransferases. Class I, the largest group, is comprised of enzymes that share a common seven-strand twisted
β–sheet structure (Cheng and Blumenthal, 1999; Cheng and Roberts, 2001). Class II is comprised exclusively of the SET (Su(var), Enhancer of zeste, Trithorax) proteins
(Yeates, 2002). Class III is comprised of membrane spanning-enzymes that have a unique C-terminal tripartite consensus motif (Romano and Michaelis, 2001).
A wide range of secondary metabolites utilise methyltransferases in their biosynthesis pathways. Examples include antibiotics, such as vancomycin and amphotericin (van
Wageningen et al, 1998; Caffrey et al, 2001), immunosuppressants, such as rapamycin
(Aparicio et al, 1996), and toxins, including fumonisin and syringomycin (Proctor et al,
2003; Vaillancourt et al, 2005). Methylation can dramatically effect the biological activity of such compounds, and has been demonstrated to influence their
135 stereochemistry, lipophillicity, and resistance to proteolytic breakdown (Rix et al,
2002). The manipulation of PK/NRP methyltransferases may therefore prove to be a valuable tool for engineering compounds with altered bioactivities.
The cyanobacterial hepatotoxins, microcystin and nodularin, are rich in methylated residues. In addition to the O-methylation present on Adda, microcystin and nodularin typically contain six other methylations: three C-methylations on Adda, N- and
C-methylations on methyldehydroalanine/dehydrobutyrine (MeDha/MeDhb), and a
C-methylation in D-methylaspartate (D-MeAsp). In M. aeruginosa, Planktothrix agardhii and Anabaena sp. these methylations are putatively catalysed by the stand- alone O-methyltransferase McyJ, C-methyltransferase domains on McyD, McyE and
McyG, and an N-methyltransferase domain of McyA (Tillett et al, 2000; Christiansen et al, 2003; Rouhiainen et al, 2004 ). In nodularin biosynthesis (Nodularia spumigena), the Adda O-methylation is putatively catalysed by NdaE, while the C- and
N-methylations are catalysed by NdaC, NdaD, NdaF and NdaA (Moffitt and Neilan,
2004).
The putative Adda O-methyltransferases McyJ and NdaE, share around 80% sequence identity. Both enzymes contain several conserved motifs characteristic of methyltransferases, including the glycine-rich Motif I, required for S-adenosyl-L- methionine binding (Kagan and Clarke, 1994; Moffitt and Neilan, 2004). While NdaE has not been functionally characterised, mcyJ has recently been targeted for gene disruption in P. agardhii (Christiansen et al, 2003). The insertional inactivation of mcyJ confirmed the predicted role of the encoded O-methyltransferase in Adda methylation in P. agardhii, however, the authors were unable to elucidate whether
136 McyJ methylates Adda during toxin biosynthesis, or as a post-synthetic tailoring step. It has been hypothesised that the O-methylation of Adda is a co-synthetic step as most natural microcystin variants possess this modification.
Between the years 2000 and 2004, 15 McyJ sequences were submitted to the NCBI database, yet comprehensive sequence-based analyses have not been performed for any of these hypothetical proteins. Questions regarding the topology, phylogeny and enzymatic subclass of the McyJ enzymes remain unanswered. This chapter offers insights into the evolution, structure and function of McyJ. Several strategies have been employed to characterise this Adda O-methyltransferase. The distribution of mcyJ orthologues among toxic and non-toxic cyanobacteria has been assessed, and comparative sequence and phylogenetic analyses for McyJ and other methyltransferases in the databases have been performed. Finally, the recombinant McyJ peptide has been over-expressed in E. coli, purified and enzymatically characterised in vitro. The results of these experiments are discussed with respect to the role of McyJ in microcystin biosynthesis and its potential use in biopharmaceutical applications as a post-synthetic tailoring enzyme.
137 5.1 EXPERIMENTAL DESIGN
5.1.0 Screening various strains of cyanobacteria for mcyJ orthologues
Degenerate oligonucleotide primers mcyJscreenF [5’-GAGYTACAAGTAGAAATT-
3'] and mcyJscreenR [5’-TCTCGWCCCACTCTAGGCAAAC-3'] were designed to amplify by PCR, an approximately 240 bp sequence within the mcyJ (or ndaE) gene of
M. aeruginosa PCC7806, Microcystis sp. MCS3, M. novacekii, Nodularia spumigena
NSOR10 and Anabaena sp. 90. These primers were used to screen chromosomal DNA samples from 24 strains of toxic and non-toxic cyanobacteria for mcyJ orthologues
(Table 5.0). The same DNA samples were also screened with the M. aeruginosa
PCC7806 mcyJ specific primers mcyJ(345)F
[5’-TACAGCCAACCTTCCACCGGGCTGC-3’] and mcyJ(525)R
[5’-GCTAAAGCAGGGTTATCGGACAGGA-3’].
5.1.1 Sequence analysis
The 837 bp nucleotide sequence of mcyJ (GenBank accession AF183408) was analysed using several different computer programs. The primary amino acid sequence of the corresponding McyJ peptide was deduced using Translate (ExPASy), while theoretical physico-chemical parameters (amino-acid compositions, molecular weight, pI, etc.) were determined using ProtParam (ExPASy). Determination of potential membrane spanning motifs and hydrophobic moments within McyJ was achieved using DAS analysis (ExPASy) and the algorithm of Kyte and Doolittle (1982) (DNA Strider 1.0).
Prediction of α-helices and β-sheets within the McyJ secondary structure was achieved using GOR IV (Garnier et al, 1996) and the Hierarchical Neural Network method
(HNN, Guermeur, 1997) methods (ExPASy). The probability of signal peptides within
138 the McyJ sequence, and the putative sub-cellular localisation of McyJ, were investigated using SignalP and PSORTb (ExPASy). Codon usage within the mcyJ sequence was assessed with the Kazusa Countcodon program and Rare Codon Caltor.
The percentage similarity and identity scores of McyJ and other peptide sequences were determined using the PSI-Blast program [National Centre for Biotechnology
Information (NCBI)]. Conserved domains within McyJ were detected using the Blast
CD-search (NCBI), ScanProsite (ExPASy), and FUGUE (Shi et al, 2001).
5.1.2 Phylogenetic analysis
A PSI-Blast search with the McyJ sequence, returned ~50 sequences with a score of greater than 70 (Table 5.2). These sequences were subsequently used to generate a multiple sequence alignment (Appendix A5) and corresponding phylogenetic tree
(ClustalX). A DNA methylase DmtA from Nostoc sp. (Matveyev et al, 2001) was also included in the alignment as an artificial out-group. Phylogenetic trees were generated using the neighbour-joining method (Satou and Nei, 1987) with gaps removed. Trees were displayed graphically using njplot and AppleWorks6 (Drawing).
5.1.3 Heterologous expression and purification of McyJ
Engineering the pET30(mcyJ) expression construct
The oligonucleotide primers mcyJF [5’-GCGTTTTTGCAAGATCAGTTCAAG-3’] and mcyJR [5’-GCCAATCGCTAACAAGAAGTCCCCCCT-3’] were used to amplify mcyJ from M. aeruginosa PCC7806 chromosomal DNA. The resulting PCR product was cloned into the PCRII-TOPO vector (Invitrogen), excised with EcoRV and SacI, gel-purified and subcloned into the pET30a expression vector (Invitrogen) that had been
139 previously linearised with the same restriction enzymes (Figure 5.0). The orientation and sequence of mcyJ in pET30a was verified by sequencing with the vector binding
T7prom [5’-TAAACGACTCACTATAGGG-3’] and T7term
[5’-ACCGCTGAGCAATAACTAGC-3']. The final expression construct pET30(mcyJ) was transformed into the E. coli expression host Rosetta(DE3)PlysS (Invitrogen) which was grown under kanamycin (50 µg/mL) and chloramphenicol (34 µg/mL) selection.
Overexpression of histidine-tagged McyJ
Expression cultures were inoculated with a 1% volume of overnight starter culture and grown in tryptone phosphate medium supplemented with kanamycin (50 µg/mL) and chloramphenicol (34 µg/mL). Cultures were grown with vigorous shaking (160 rpm) at
37°C to an OD650 of 0.6, and either retained at 37°C, or transferred to either 30°C or
25°C and incubated for an additional 20 min (or until an OD650 of approximately 1.0 was reached). Expression of McyJ was induced with 0.1-1 mM IPTG for 0-18 h at 25-
37°C. Cell pellets were harvested via centrifugation (5 000 g, for 10 min at 4°C), washed with phosphate buffer [0.5 M NaCl; 20 mM sodium phosphate, pH 7.4] and stored at -80°C until required.
Purification of histidine-tagged McyJ
Cell pellets were thawed on ice and resuspended in 0.5% of the original culture volume of cold phosphate buffer containing 1 mM PMSF and 20 mM imidazole. The lysate was passed through an 18 gauge needle several times, sonicated briefly (Branson Sonifier
250, amplitude 25, 50% duty cycle, pulsed) and cleared via centrifugation (20 000 g for
30 min at 4°C). A small aliquot of crude lysate was reserved for enzyme assays (see
140 below) and then the remaining lysate was precipitated with 60% ammonium sulphate
(section 2.10). The final pellet was resuspended in 0.1% of the original culture volume of ice-cold phosphate buffer containing 20 mM imidazole, filtered through a 22 µm
Millex membrane (Millipore), and applied to a 1 mL Hitrap column (AP biotech) charged with Ni2+. The column was washed with 10 column volumes of phosphate buffer containing 20 mM imidazole, followed by 10 volumes of phosphate buffer containing 50 mM imidazole, and eluted with 2 column volumes of phosphate buffer containing 300 mM imidazole. The wash and eluate were collected in 0.5 mL fractions and analysed via SDS PAGE (12% polyacrylamide gels) and western blotting (section
2.11).
Purified protein samples intended for enzyme assays, were desalted via size exclusion filtration using a 10 KDa molecular weight cut-off columns (Amicron Ultra) and 50 mM HEPES exchange buffer (pH 7).
5.1.4 Enzyme assays
The demethylated microcystin ([demethylated-Adda, D-Asp3]microcystin-RR) was donated by Elke Dittmann (Humboldt University, Berlin) and was isolated from a
Planktothrix agardhii mcyJ deletion mutant (Christiansen et al, 2003).
The O-methylation of [demethylated-Adda, D-Asp3]microcystin-RR by McyJ was analysed in vitro as follows: Each 25 µL assay contained 50 mM HEPES (pH 7), 5 mM
3 DTT, 5 mM MgCl2, 57 µM [demethylated-Adda, D-Asp ]microcystin-RR, 570 µM
141 S-adenosyl methionine (SAM) and 1.5 µg purified enzyme or 10 µL of crude
Rosetta(DE3)PlysS[pET30(McyJ)] lysate. Reactions were incubated at 25°C overnight
(~18 h) and analysed by MALDI-TOF MS.
Positive-ion mass spectra were initially recorded using a Voyager DE STR MALDI-
TOF mass spectrometer (Applied Biosystems). Half a microlitre of assay mixture was combined with 1µL of matrix (α-cyano-4-hydroxycinnamic acid in 80% acetonitrile,
0.08% trifluoroacetic acid) and applied directly to the target plate. Analyses were performed in the positive-ion mode, producing mainly singly protonated molecular ions
([M+H]+). The acceleration voltage was set at 20 kV. All measurements were carried out in the delayed extraction and reflection mode, allowing the determination of monoisotopic mass values (m/z, mass-to-charge ratio). A low mass gate of 450 Da was used to filter out the most intensive matrix ions.
After the identification of target peaks, the assay mixture was re-analysed using a
Q-Star Pulsar MALDI-TOF MS/MS mass spectrometer (Applied Biosystems) equipped with an o-MALDI interface. Peaks corresponding to protonated [demethylated-Adda,
D-Asp3]microcystin-RR (m/z ~ 1010) and [Adda, D-Asp3]microcystin-RR (m/z ~ 1024) were selected using Q1 and fragmented in q2 using Ar collision gas with an energy sufficient to reduce the precursor ion to ~5% of the base peak intensity. TOF product ion spectra were recorded and compared to characteristic fragment ion signatures in the literature (Welker et al, 2002). All assays were performed in duplicate on separate days.
142 5.2 RESULTS
5.2.0 Distribution of mcyJ orthologues in various species of cyanobacteria
In order to assess the distribution of mcyJ orthologues among toxic and non-toxic cyanobacteria, DNA from 24 different strains was screened by PCR with mcyJ-specific and degenerate oligonucleotide primers. The results are presented in Table 5.0. All microcystin-producing strains tested positive for mcyJ orthologues using both primer sets. The degenerate primers mcyJscreenF/R, were also able to amplify ndaE from all of the nodularin-producing strains. Over half of the non-toxic cyanobacteria screened also tested positive for mcyJ. All strains possessing mcyJ/ndaE also tested positive for other mcy genes (as determined by mcyA, mcyH or mcyI screening).
5.2.1 Sequence analysis of mcyJ
The 837 bp mcyJ open reading frame was predicted to encode a 31.9 KDa peptide with a pI of 5.69. While the KD profile of the McyJ peptide contained several small hydrophobic regions, the DAS topology prediction program suggested that it is not a membrane bound protein (Figure 5.1). SignalP analysis could not identify a signal peptide within the N-terminal sequence of McyJ, suggesting that it is not a secretory protein. PSORTb gave equal localisation scores for all cellular compartments.
Secondary structural predictions suggested that McyJ contains 10-14 α-helices and at least 7-10 β-sheets. Codon usage for mcyJ is given in Table 5.1. The mcyJ sequence contained 24 rare codons including several repeated and/or consecutive rare codons.
Subsequent heterologous expression studies were therefore carried out in the
Rosetta(DE3)pLysS E. coli expression strain containing the pRARE plasmid encoding the E. coli rare tRNA codons.
143 Comparison of the inferred primary peptide sequence of mcyJ with other sequences in the NCBI database revealed significant similarity (up to 98% identity (I) and 99% similarity (S), PSI-Blast) to other putative methyltransferases. The top 5 PSI-Blast results were sequences from the cyanobacteria P. agardhii (87% I, 93% S),
N. spumigena (81% I, 92% S), Anabaena sp. 90 (81% I, 92% S) and Microcystis sp.
MCS3 (98% I, 98% S). The highest scoring non-cyanobacterial sequence was that from the proteobacterium Polyangium cellulosum (39% I, 58% S).
A Blast CD-search identified a conserved SAM-binding domain within the McyJ peptide. Other enzymes possessing this domain include methyltransferases involved in coenzyme metabolism (eg. ubiquinone methyl transferase, EC 2.1.1.64), various DNA and mRNA methyltransferases, hypothetical plant methyltransferases, and cyclopropane fatty acid synthase related methyltransferases (EC 2.1.1.79). The CD-search failed to identify a Class II methyltransferase SET domain, or a Class III ICMT domain within the McyJ peptide sequence.
A FUGUE homologous structure search using the McyJ peptide sequence yielded similar results to the Blast CD search. High confidence Z-scores were obtained for cyclopropane-fatty-acyl-phospholipid synthase 1 from Mycobacterium tuberculosis
(PDB code, [lkph]) and an mRNA capping enzyme from Encephalitozoon cuniculi
[lri5]. High Z-scores were also returned for SAM-dependant methyltransferases from
Thermotoga maritima msb8 [lvlm], Clostridium acetobutylicum atcc824 [ly8c], and
Haemophilus influenzae [lim8]. Surprisingly, a ScanProsite search could not identify any patterns, profiles or motifs within the McyJ peptide sequence.
144 The 22 sequences with the highest homology to McyJ (based on PSI-Blast search) were compiled in a multiple sequence alignment and analysed (Figure 5.2). SAM binding motifs I, (V/I/L)(L/V)(D/E)(V/I)G(G/C)G(T/P)G,
II, (P/G)(Q/T)(F/Y/A)DA(I/V/Y)(F/I)(C/V/L), and
III, LL(R/K)PGG(R/I/L)(L/I)(L/F/I/V)(I/L) were identified within McyJ at residues 64,
128 and 155 of the peptide sequence, respectively, although several substitutions were identified within these consensus residues. The terminal Gly-9 of Motif I was replaced in the cyanobacterial sequences by Ala. In Motif II Tyr-6 and Phe-7 were replaced with
Leu and Thr, respectively. All Microcystis species contained Gln in place of Lys/Arg at position 3 within Motif III, however, the filamentous cyanobacteria retained the conserved Arg at this position. Conversely, all Microcystis species retained the conserved Arg at position 7 within Motif III. However, in the filamentous cyanobacteria, this residue was replaced by Lys. The cyanobacterial sequences also possessed a unique conserved Cys downstream of Motif II and a unique conserved Lys downstream of Motif III.
The tripartite consensus motif of the Class III (ICMT) methyltransferases, was not identified at the C-terminus of McyJ.
To gain an overview as to the evolutionary relationships between McyJ and other methyltransferases, a phylogenetic analysis was performed (Figure 5.3). The 50 sequences used in the analysis ranged in length from 184-380 amino acids with an average length of 282 (Table 5.2). The overall tree partitioned into two clusters encompassing the groups A-D and E-G, respectively. Group A was comprised
145 exclusively of sequences from the hepatotoxin biosynthesis gene clusters of cyanobacteria. Group B was comprised of putative methyltransferases from the phyla chlorobi and proteobacteria. Group C was comprised of PKS-associated methyltransferases from actinobacteria. Group D was mostly comprised of PKS- associated methyltransferases from proteobacteria (with the exception of AdpE from
Anabaena circinalis). Group E was comprised of putative sterol and RNA methyltransferases from several species of fungi. Group F was comprised of putative cyanobacterial methyltransferases, as well as other uncharacterised enzymes from
Cytophaga hutchinsonii and Methylococcus capsulatus. Group G, the largest cluster in the tree, was comprised of putative actinobacterial methyltransferases, most of which are associated with secondary metabolite biosynthesis pathways.
5.2.2 Overexpression and Purification of histidine-tagged McyJ
McyJ was overexpressed in E. coli with an N-terminal hexa-histidine fusion tag. The recombinant peptide expressed at low levels for all culture/induction conditions tested
(section 5.1.3). Western blot analysis confirmed that McyJ was expressed without truncation, at the expected molecular weight of ~35 KDa. Maximum yields of soluble protein (up to 2% total protein) were obtained when induction was carried out at reduced temperatures (25-30°C) with an IPTG concentration of ~0.5 mM.
The recombinant McyJ protein precipitated in the presence of 60% ammonium sulphate under the described conditions (section 2.10), and eluted from the Ni2+ charged affinity column at an imidazole concentration of ~300 mM. The recombinant protein was purified to >50% homogeneity at concentrations of 0.25-0.5 mg/mL in the eluate
(Figure 5.4).
146 5.2.3 Enzyme assays
The O-methylation of [demethylated-Adda, D-Asp3]microcystin-RR by McyJ, was investigated via MALDI-TOF MS/MS. All samples analysed displayed a mass peak with an m/z of 1 010, corresponding to the monoisotopic mass of protonated
[demethylated-Adda, D-Asp3]microcystin-RR (Figure 5.5). Further analysis of the
1 010 peak by MS/MS revealed several microcystin-RR signature peaks (m/z, 70, 213, and 375), plus a fragment with an m/z of 585 (corresponding to Arg + demethylated-
Adda + Glu + H). Samples containing purified McyJ or crude
Rosetta(DE3)PlysS[pET30(McyJ)] lysate displayed an additional peak (m/z, 1024) corresponding to protonated [D-Asp3]microcystin-RR. Fragmentation of this peak by
MS/MS resulted in the production of microcystin-RR signature fragments (m/z, 70 and
135), plus a fragment with an m/z of 599 (corresponding to Arg + Adda + Glu + H)
(Figure 5.6).
147 5.3 DISCUSSION
The distribution of mcyJ orthologues among toxic and non-toxic cyanobacteria, showed a strong correlation between the presence of mcyJ and hepatotoxin production. McyJ or ndaE orthologues were detected in all microcystin and nodularin-producing strains tested. These results are in agreement with the observation that nearly all natural microcystin/nodularin variants isolated thus far contain the Adda O-methylation.
Interestingly, nearly half of all the non-toxic cyanobacterial strains examined tested positive for mcyJ. These results reflect those of the mcyH and mcyI genetic screening experiments, and those of previous studies (Nishizawa et al, 1999; Tillett et al, 2000;
Kaebernick et al, 2001). It is likely that these strains have reverted to a non-toxic phenotype due to mutations within their mcy biosynthesis genes, or within mcy regulatory regions (see discussion sections 3.3 and 4.3).
The fact that false negatives weren’t obtained for the PCR screen with mcyJscreenF/R, suggests that the degenerate primers are good genetic probes for assessing the potential risk of hepatotoxic cyanobacteria in water supplies. However, the recent mutagenesis of mcyJ in P. agardhii, has proven that mcyJ is not critical for toxin production, at least under laboratory growth conditions (Christiansen et al, 2003). The mcyJscreenF/R primers should therefore be used in conjunction with other genetic probes, for example, primers targeting the mcy PKS/NRPS genes, in order to confidently assess the presence of toxigenic cyanobacteria in water supplies.
In an attempt to gather information relating to the topology, phylogeny, cellular localisation and enzymatic subclass of McyJ, the inferred primary peptide sequence of
148 mcyJ was analysed using several different computer programs. While PSORTb could not predict the cellular location of McyJ, N-terminal sequence analysis suggested that
McyJ is a non-secretory protein. Sequence-based topological studies suggested that
McyJ is also not a membrane bound protein, and is thus unlikely to belong to the Class
III group of methyltransferases. This group of enzymes, also known as the isoprenylcysteine carboxyl methyltransferases (ICMTs), are embedded within the cell membrane. Previous primary and secondary structural investigations also suggest that
ICMTs, contain a novel tripartite (RHPxY-hyd-EE ) consensus motif in their C-termini
(Romano and Michaelis, 2001). However, this motif was not identified in the McyJ peptide sequence.
At least 7 β-sheets were predicted within the secondary structure of McyJ, suggesting that it is likely to belong to the Class I methyltransferase family. These enzymes share a common core structure and a mixed seven-stranded sheet referred to as an AdoMet- dependent methyltransferase fold (Cheng and Roberts, 2001). A Blast CD-search also identified the SAM-binding domain of Class I methyltransferases within the McyJ peptide sequence, however the CD-search failed to identify a Class II methyltransferase
SET domain, or a Class III ICMT domain within McyJ.
Comparison of the inferred primary peptide sequence of McyJ with other sequences in the databases identified a wide range of homologous enzymes with putative functions ranging from sterol methylation to rRNA methylation. However, the most abundant group of McyJ homologues was the putative PK methyltranferases. The closest characterised relative of McyJ was the EryG O-methyltransferase of S. erythraeum. This enzyme catalyses the final step in the biosynthesis of erythromycin A, the O-
149 methylation at C3 of the erythromycin C aglycone (Webber et al, 1989, Paulus et al,
1990). The sequence data presented in this thesis support a similar function for McyJ in microcystin biosynthesis.
A close examination of the peptide sequence alignment in Figure 5.2, revealed that
McyJ, NdaE, and their homologues, all possessed the signature Motifs I, II and III described by Kagan and Clarke (1994). These motifs play a role in SAM binding and are common to a wide range of Class I methyltransferases. While RNA methyltransferases usually possess Motif I and/or Motif II, most small molecule
O-methyltransferases, including the caffeic acid, tetracenomycin, midamycin and erythromycin O-methyltransferases, possess all three motifs. This pattern is also commonly observed in protein carboxyl methyltransferases, and lipid methyltransferases (Kagan and Clarke, 1994). The presence of these SAM binding motifs in McyJ and NdaE suggested that these enzymes are dependent on this cofactor methyl-donor.
Motif I, which comprises part of the SAM binding-pocket of methyltransferases, usually contains a conserved Phe-6, plus several other hydrophobic residues. Together, these amino acids form a hydrophobic platform on one side of the purine and ribose rings of SAM. Motif I also features several conserved Gly residues, which form a tight loop responsible for the correct positioning of SAM in the binding-pocket (Cheng et al,
1993). While Phe-6 was conserved in the cyanobacterial sequences, Gly-9 was substituted with Ala. As the Gly-9/Ala substitution was conservative, it is probably of little consequence to the overall structure/function of McyJ.
150 Motif II is usually abundant in aromatic residues and it has been suggested that these residues are involved in binding and stabilising the positively charged sulfonium ion of
SAM (Kagan and Clarke, 1994). While Phe-3 of Motif II was conserved in the cyanobacteria, and Tyr-6 was conservatively substituted with Leu, Phe-7 was replaced by a hydrophyllic Thr residue. Whether or not this substitution affects the efficiency of
SAM-binding in the cyanobacterial enzymes remains to be elucidated.
The role of Motif III in SAM-binding remains a subject of much debate (Gomi et al,
1992; Fujioka, 1992; Kagan and Clarke, 1994), yet it is highly conserved in most methyltransferases, including those analysed in this study. Interestingly, the Lys-3/Gln substitution in Motif III was unique to the Microcystis sequences within the alignment.
Conversely, the Arg-7/Lys substitution was only present in the filamentous cyanobacterial sequences. While the substitution of positively charged Lys with negatively charged Gln could potentially affect substrate binding, the Arg/Lys substitution is conservative and is therefore probably of little consequence to the catalytic activity of these enzymes.
Taken together, the primary and secondary sequence analyses suggested that McyJ is a
SAM-dependent methyltransferase belonging to the Class I family of enzymes. Within this class, McyJ shared many sequence characteristics with the small molecule methyltransferases.
In an attempt to understand the evolution of McyJ in the context of other methyltransferases, a phylogenetic tree was constructed encompassing homologous sequences from a wide range of bacteria and several species of fungi (Figure 5.3). Most
151 sequences within the tree possessed typical SAM binding motifs of Kagan and Clarke
(1994), indicating that they belonged to the Class I family of methyltransferases.
Around 90% of all sequences in the tree originated from organisms known to possess secondary metabolite pathways, and more than half of those sequences were encoded within, or proximal to, NRP/PK biosynthesis gene clusters. These results suggested that the Class I family of methyltransferases partitions primarily according to enzyme function. Interestingly, the tree also contained several methyltransferases from fungal primary metabolic pathways, including, an rRNA methyltransferase from Gibberella zeae and sterol methyltransferases from Pneumocystis species (group E). It is thus conceivable that the PK/NRP methyltransferases evolved from a primary metabolic enzyme in a common ancestor prior to the divergence of bacteria and eukaryotes.
While most sequences within the tree clustered according to organismic lineage, the cyanobacterial sequences were polyphyletic. The group A cyanobacterial proteins
(McyJ and NdaE) appear to have shared a common ancestor with the group B chlorobi and proteobacterial enzymes, while the group F cyanobacterial enzymes were more closely related to the actinobacter PK/NRP methyltransferases. It has been postulated that the ancestral cyanobacterium was toxigenic, and that subsequent non-toxic lineages resulted through the loss of the biosynthesis genes (Rantala et al, 2004). Therefore, the group F methyltransferases may be relics of partially deleted hepatotoxin biosynthesis pathways. Alternatively, the genes encoding the group F methyltransferases may have been introduced via horizontal gene transfer events. The presence of putative transposase genes within the microcystin biosynthesis gene cluster provides a possible avenue for this method of genetic exchange.
152 The distribution of sequences within group A, indicated that the McyJ from Anabaena sp. is more closely related to NdaE than it is to McyJ from P. agardhii, or other
Microcystis species. These results suggested that the ancestral Adda
O-methyltransferase arose prior to the divergence of the hepatotoxic cyanobacteria.
In order to determine whether McyJ methylates Adda during toxin biosynthesis or as a post-synthetic tailoring step, the O-methyltransferase was overexpressed in E. coli, purified and enzymatically analysed. The monomeric size of the recombinant McyJ peptide, as determined by SDS-PAGE, was in agreement with the theoretical molecular weight of the protein (~35 KDa). Due to the presence of a 27 KDa co-eluting protein,
McyJ could only be purified to ~50%. This purity was deemed sufficient for subsequent assays, as E. coli does not contain any homologous O-methyltransferases that are likely to methylate Adda.
The O-methylation of [demethylated-Adda, D-Asp3]microcystin-RR by McyJ, was investigated via MALDI-TOF MS/MS. Mass spectrographs for samples with, and without enzyme, contained mass peaks at 1 010, corresponding to [demethylated-Adda,
D-Asp3]microcystin-RR. However, samples containing McyJ also revealed an additional peak at 1 024. This peak was further analysed by MS/MS and was identified as [Adda, D-Asp3]microcystin-RR. It was therefore concluded that McyJ was able to methylate Adda post-synthetically. Whether or not McyJ is also able to act co-synthetically during Adda biosynthesis remains to be determined. Previous studies have shown that the mutation of mcy biosynthesis genes results in the destabilisation of the microcystin synthetase enzyme complex and the loss of toxin production (Tillett et
153 al, 2000). Yet disruption of McyJ did not abolish toxin production in P. agardhii
(Christiansen et al, 2003). Together with the fact that mcyJ has its own promoter
(Kaebernick et al, 2002), we predict that McyJ is not an integral part of the microcystin biosynthesis mega-enzyme complex and does not methylate Adda as a synthetic step during toxin production (Figure S5.7).
Tailoring enzymes are known to be crucial for the final bioactive properties of numerous natural products (Fawaz and Jones, 1988; Paulus et al, 1990; Dhar and
Rosazza, 2000) and may therefore be excellent candidates for the rational design of novel antibiotics. Methyltransferases constitute a particularly important class of tailoring enzymes. The in vitro and in vivo use of methylation in the biopharmaceutical industry has been harnessed to produce novel biologically active methylated and demethylated compounds. For example methylation of the anticholesterolemic drug lovastatin, has been demonstrated to increase the drugs therapeutic potency (Kennedy et al, 1999), while the demethylation of the antibiotic avilamycin, has been shown to increase its solubility and thereby enhance its pharmacokinetic properties (Weitnauer et al, 2004). Methylation is also critical to the bioactivity of many secondary metabolite toxins including microcystin. Previous studies have shown that natural microcystin variants lacking the Adda O-methylation have reduced inhibition of protein phosphatases (Namakoshi et al, 1992). Furthermore, the recent disruption of the mcyJ
O-methyltransferase in P. aghardii resulted in the production of a demethylated microcystin variant with similarly reduced bioactivity (protein phosphatase inhibition)
(Christiansen et al, 2003). Through the manipulation of mcyJ and other genes within the mcy gene cluster, it may be possible to synthesise demethylated microcystin variants that retain their liver-targeting characteristics but lack toxicity. Such compounds could
154 then be conjugated to immunosuppressant, antimicrobial, antiviral or anticancer drugs and used as targeted liver therapeutics without the numerous side effects of systemically acting medications. The characterisation of tailoring enzymes, such as McyJ, is central to the achievement of such goals.
155 CHAPTER 6.
GENERAL DISCUSSION
156 The main objectives of this project were to characterise the tailoring enzymes McyI and
McyJ, and the membrane-bound transporter, McyH, encoded within the microcystin biosynthesis gene cluster of M. aeruginosa PCC7806. By elucidating the function and regulation of these enzymes, it was envisaged that a more complete understanding of the microcystin biosynthesis pathway could be gained, and that this knowledge, in turn, could be applied to the fields of rational drug design and water quality management.
The primary goals of this project were to elucidate the individual enzymatic functions of McyH, McyI and McyJ and to determine their respective roles in toxin production and/or export. To this end, several different experimental strategies were adopted, including bioinformatic and biochemical investigations, combined with mutagenesis and mutant complementation studies. The project also aimed to investigate the regulation of these enzymes in vitro and in vivo. A final objective was to characterise the distribution of mcyH, mcyI and mcyJ among toxic and non-toxic cyanobacteria, and elucidate their evolutionary origins within, and beyond the order, cyanobacteria.
6.0 Major results
6.0.0 McyH The combined data for McyH strongly suggested that it is an ABC-A1 type microcystin exporter that also plays an integral role in toxin biosynthesis. To our knowledge, McyH is the first ABC transporter to be identified in M. aeruginosa. Secondly, and more significantly, McyH is the first and only characterised example of a cyanotoxin exporter.
157 Genetic screening tests revealed a strong correlation between the presence of mcyH and microcystin production, indicating that the ABC transporter is a critical component of toxin metabolism. Sequence-based analyses revealed that McyH has a classic ABC-A1 architecture, with membrane and ATPase domains encoded within a single polypeptide.
The three ABC transporter signature sequences (Walker A and B, and the linker peptide) were also identified within the primary peptide sequence of McyH.
Comparative sequence analyses for McyH identified several homologous ABC transporters from a range of different organisms. Interestingly, the ATPase domain of
McyH shared high sequence similarity with several eukaryotic peroxisomal membrane proteins (PMPs). These proteins are responsible for the transport of long-chain fatty acids out of the cytoplasm and into peroxisomes. McyH also shared sequence homology with several other ABC transporters known to export bacteriocins, lantibiotics, drugs and siderophores.
The recombinant McyH protein was successfully over-expressed in E. coli from an alternative up-stream start codon, and the purified transporter reconstituted into membrane vesicles and analysed in vitro. The uptake of microcystin-LR into McyH- containing vesicles was confirmed via semi-quantitative MALDI-TOF MS using an internal microcystin-RR standard.
Several truncated McyH peptides were also over-expressed in E. coli. While these peptides did not display any measureable ATPase activity, they were successfully purified and used to raise McyH antibodies in rabbits. These antibodies were subsequently used in immunoblot experiments to measure the differential expression of
158 McyH under various culture conditions, in mutant and wild-type cultures. McyH was shown to be up-regulated in cultures grown under high and red light, and down- regulated in cultures grown under low light and in darkness. The McyH antibodies were similarly used to demonstrate that McyH expression levels in microcystin synthetase peptide synthetase mutants (mcyA- and mcyB-) are much lower than those in wild-type
M .aeruginosa PCC7806.
6.0.1 McyI The combined bioinformatic and biochemical data for McyI supported the hypothesis that it is a 2-hydroxyacid dehydrogenase that catalyses the interconversion of 3-methyl- malate and 3-methyl-oxaloacetate. This reaction is predicted to be essential for the production of microcystin variants containing D-MeAsp at position 3 within the cyclic peptide.
Genetic screening tests identified mcyI orthologues in all toxic Microcystis strains examined. However, the gene was not identified in P. agardhii, indicating that McyI may not be necessary for the production of D-Asp3 types of microcystin, but may be required for the biosynthesis of other microcystin isoforms.
Sequence analyses suggested that McyI belonged to the 2-hydroxy acid dehydrogenase family and had sequence homology with phosphoglycerate dehydrogenase. An NAD- binding signature was also identified within the primary peptide sequence of McyI, indicating that the enzyme has a dinucleotide co-factor requirement.
159 McyI was successfully over-expressed in E. coli and purified to homogeneity. The enzyme was found to possess a dimeric structure, which was found to be essential for catalytic activity. The major observed in vitro activity of McyI was the reduction of
app oxaloacetate to malate, with a Vmax of 15.2 µmol/min/mg protein, however, the enzyme was also able to reduce α-ketoglutarate, and oxidise malate and methyl-malate.
The oxaloacetate and α-ketoglutarate reductase reactions both obeyed regular
Michaelis-Menten kinetics.
McyI was found to be NADPH dependent, although NADH was also reduced by the enzyme at a lower catalytic rate. Methyl-malate was the strongest inhibitor of the oxaloacetate and α-ketoglutarate reductase activities of McyI, suggesting that this rare, modified organic acid is the true substrate of the enzyme.
To our knowledge, McyI is the only known example of a methyl-malate dehydrogenase, and furthermore, is the first example of a 2-hydroxyacid dehydrogenase encoded within a secondary metabolite gene cluster.
6.0.2 McyJ The combined bioinformatic and experimental data for McyJ suggested that this enzyme is a Class I methyltransferase responsible for the post-synthetic methylation of
Adda.
Genetic screening tests identified mcyJ or ndaE orthologues in all hepatotoxic cyanobacteria examined. While these results suggested that the methyltransferase is
160 critical to toxin biosynthesis, previous research has demonstrated that the mutation of mcyJ does not abolish microcystin production in P. agardhii.
The primary and secondary sequence analyses here suggested that McyJ is a SAM- dependent methyltransferase belonging to the Class I family of enzymes. Within this class, McyJ shared many sequence characteristics with the small molecule methyltransferases.
The recombinant McyJ protein was successfully over-expressed in E. coli and purified to >50% homogeneity. The O-methylation of [demethylated-Adda,
D-Asp3]microcystin-RR by the purified McyJ enzyme, was demonstrated by an in vitro enzyme assay and MALDI-TOF MS/MS. It was therefore concluded that McyJ was able to methylate Adda post-synthetically. Whether or not McyJ is also able to act co- synthetically during Adda biosynthesis remains to be determined, when appropriate intermediates become available.
6.0.3 Phylogeny of the Mcy tailoring enzymes
The Mcy tailoring enzymes from M. aeruginosa (and their counterparts from other hepatotoxic cyanobacteria) formed phylogenetically distinct subgroups. These results suggest that McyH, I and J are unique to cyanobacterial secondary metabolism. While most of the phylogenetic trees contained sequences from a wide range of cyanobacteria, the Mcy tailoring enzymes clustered primarily according to putative function, not according to organismic lineage. For example, the McyH-based tree (Figure 3.5) contained numerous cyanobacterial sequences, yet these sequences were distributed across several different phylogenetic subgroups. Remarkably, further analysis of McyH
161 indicated that its closest characterised relatives were the eukaryotic peroxisomal membrane proteins.
Another interesting observation was that the Mcy tailoring enzymes from Anabaena are more closeley related to their Nda homologues in Nodularia rather than their Mcy homologues in Microcystis. These results suggest that hepatotoxin biosynthesis evolved in cyanobacteria prior to the divergence of the orders Nostocales and Chroococales.
Alternatively, the hepatotoxin genes may have been transferred between Anabaena and
Nodularia in a relatively recent evolutionary event. The existence of transposase genes proximal to the hepatotoxin clusters provides a likely avenue such horizontal genetic transfer.
6.1 Overall significance
The active transport of microcystin by McyH raises numerous concerns regarding the management of drinking water supplies. For example, if cells are capable of pumping out low levels of toxin, then the removal of intact blooms from water bodies, prior to cell lysis, may not be an effective means of treatment. While conventional water treatment procedures, such as slow sand filtration and chlorination, can remove or kill cells, they do not remove toxins. Therefore, additional, and often costly, procedures such as activated carbon filtration and ozonisation may be required to guarantee water safety.
The results presented in this thesis suggest that high light intensities stimulate the up- regulation of McyH expression. These results concur with previous findings by
Kaebernick et al (2000) who reported that microcystin is constitutively produced under
162 low and medium light intensities, and is exported when a high light threshold is reached. The implementation of simple strategies, such as the planting of shade trees along dams and river banks, may therefore significantly reduce the export of hepatotoxins from surface-dwelling scums of cyanobacteria.
In addition to facilitating the innovation of current water quality management protocols, the characterisation of McyH may have important implications for the biopharmaceutical industry. The fact that the microcystin structure specifically targets the liver makes it a potentially useful vector for hepato-therapeutic drugs. The development of novel microcystin analogs and conjugates will, like other drugs
(penicillin, cyclosporin, amphotericin etc.), probably require the use of mass fermentation in bacteria or fungi. In order to efficiently and sustainably harvest such compounds, without loss of biomass, an extracellular transporter would be highly desirable, if not critical.
The production of novel microcystin conjugates may additionally require the exploitation of tailoring enzymes such as McyI and McyJ. Methylation has been shown to play a major role in the bioactivity, solubility and resistance to proteolytic break- down, of mixed polyketides and nonribisomal peptides. In the case of microcystin, natural and engineered de-methylated microcystin variants have been shown to have reduced toxicity. The results of this thesis indicate that the methylation of Adda is a post-synthetic step. Therefore, by blocking the activity of McyJ, or alternatively, by mutating the mcyJ gene, we could produce de-methylated, less toxic microcystin analogues without disrupting the other microcystin synthetase enzymes.
163 The exploitation of McyI may similarly facilitate the production of novel microcystin variants with reduced toxicity. Previous toxicological data suggest that natural microcystin variants with D-Asp at position 3 (eg. [D-Asp3]microcystin-LR and [D-
Asp3]microcystin-HtyR) have, on average, a 70% reduced toxicity compared to their D-
MeAsp3 counterparts ( [D-MeAsp3]microcystin-LR and [D-MeAsp3]microcystin-HtyR).
Therefore, if McyI is indeed responsible for the production of MeAsp, as the results in this thesis suggest, then blocking the activity of McyI may enable the directed biosynthesis of novel D-Asp3 variants.
McyJ and McyI may additionally have applications beyond microcystin biosynthesis.
Many stand-alone tailoring enzymes including methyltransferases and dehydrogenases have been shown to have relaxed substrate specificity. Therefore, these enzymes could potentially be used to modify other secondary metabolites including antibiotics and immunosuppressants with problems of resistance due to structural recognition.
In addition to highlighting numerous environmental and biopharmaceutical applications, the characterisation of McyH, McyI and McyJ has provided us with a more complete understanding of the microcystin biosynthesis pathway. When the mcy gene cluster was first described by Tillett et al in 2000, the assignment of all Mcy catalytic functions was based solely on comparative sequence data. While the subsequent mutation of various mcy genes has established their involvement in microcystin biosynthesis, the fact that all mutants (except ΔmcyJ) had non-toxic phenotypes, has prevented the functional analysis of the Mcy enzymes. We therefore adopted a biochemical approach to characterise McyH, McyI and McyJ. The results presented in this dissertation support the original hypotheses by Tillett et al that McyH
164 and McyJ are involved in microcystin transport and methylation, respectively.
However, the most remarkable outcome of this study has been the characterisation of the unique methyl-malate dehydrogenase, McyI. As the putative role of this enzyme in serine dehydration has been disproved, the origin of the Me-Dha1 residue is a new mystery waiting to be solved.
165 REFERENCES
Al-Rabiee, R., Y. Zhang, and G. A. Grant, 1996, The mechanism of velocity modulated
allosteric regulation in D-3-phosphoglycerate dehydrogenase. Site-directed
mutagenesis of effector binding site residues: J Biol Chem, v. 271, p. 23235-8.
Ali, V., T. Hashimoto, Y. Shigeta, and T. Nozaki, 2004, Molecular and biochemical
characterisation of D-phosphoglycerate dehydrogenase from Entamoeba
histolytica. A unique enteric protozoan parasite that possesses both
phosphorylated and nonphosphorylated serine metabolic pathways: Eur J
Biochem, v. 271, p. 2670-81.
Allgaier. H., G Jung, RG Werner, U Schneider, and H. Zähner, 1985, Elucidation of the
Structure of Epidermin, a Ribosomally Synthesized, Tetracyclic Heterodetic
Polypeptide Antibiotic: Angew. Chem. Int. Eng., v. 24, p. 1051-1053.
Ames, G. F., and H. Lecar, 1992, ATP-dependent bacterial transporters and cystic
fibrosis: analogy between channels and transporters: Faseb J, v. 6, p. 2660-6.
Aparicio, J. F., I. Molnar, T. Schwecke, A. Konig, S. F. Haydock, L. E. Khaw, J.
Staunton, and P. F. Leadlay, 1996, Organization of the biosynthetic gene cluster
for rapamycin in Streptomyces hygroscopicus: analysis of the enzymatic
domains in the modular polyketide synthase: Gene, v. 169, p. 9-16.
Aravind, L., and E. V. Koonin, 1999, Gleaning non-trivial structural, functional and
evolutionary information about proteins by iterative database searches: J Mol
Biol, v. 287, p. 1023-40.
Baht, K. S., Dixit, K. N., and Rao, A. S. (1985) Indian J. Chem. 24B, 509-512
Baltz, R. H., M. A. McHenney, C. A. Cantwell, S. W. Queener, and P. J. Solenberg,
1997, Applications of transposition mutagenesis in antibiotic producing
streptomycetes: Antonie Van Leeuwenhoek, v. 71, p. 179-87.
166 Bartsevich, V. V., and H. B. Pakrasi, 1995, Molecular identification of an ABC
transporter complex for manganese: analysis of a cyanobacterial mutant strain
impaired in the photosynthetic oxygen evolution process: Embo J, v. 14, p.
1845-53.
Bell, J. K., P. J. Pease, J. E. Bell, G. A. Grant, and L. J. Banaszak, 2002, De-regulation
of D-3-phosphoglycerate dehydrogenase by domain removal: Eur J Biochem, v.
269, p. 4176-84.
Bibb, M. J., D. H. Sherman, S. Omura, and D. A. Hopwood, 1994, Cloning, sequencing
and deduced functions of a cluster of Streptomyces genes probably encoding
biosynthesis of the polyketide antibiotic frenolicin: Gene, v. 142, p. 31-9.
Boneca, I. G., and G. Chiosis, 2003, Vancomycin resistance: occurrence, mechanisms
and strategies to combat it: Expert Opin Ther Targets, v. 7, p. 311-28.
Botes, D., P. Wessels, H. Kruger, M. Runnegar, S. Santikarn, R. Smith, J. Barna, and D.
Williams, 1985, Structural studies on cyanoginosins-LR, -YR, -YA, and -YM,
peptide toxins from Microcystis aeruginosa. : J. Chem. Soc., v. 1, p. 2747-2748.
Bottcher, G., I. Chorus, S. Ewald, T. Hintze, and N.Walz, 2001, Light-limited growth
and microcystin content of Planktothrix agardhii and Microcystis aeruginosa in
turbidostats.: Cyanotoxins: Berlin, Springer, 115-133 p.
Bruin, W. J., E. B. Nelson, and N. E. Tolbert, 1970, Glycolate Pathway in Green Algae:
Plant Physiol, v. 46, p. 386-391.
Buckel, W., 2001, Unusual enzymes involved in five pathways of glutamate
fermentation: Appl Microbiol Biotechnol, v. 57, p. 263-73.
Caffrey, P., S. Lynch, E. Flood, S. Finnan, and M. Oliynyk, 2001, Amphotericin
biosynthesis in Streptomyces nodosus: deductions from analysis of polyketide
synthase and late genes: Chem Biol, v. 8, p. 713-23.
167 Campbell, H. D., B. L. Rogers, and I. G. Young, 1984, Nucleotide sequence of the
respiratory D-lactate dehydrogenase gene of Escherichia coli: Eur J Biochem, v.
144, p. 367-73.
Carmichael, W. W., 1994, The toxins of cyanobacteria: Sci Am, v. 270, p. 78-86.
Chan, K. M., D. Delfert, and K. D. Junger, 1986, A direct colorimetric assay for Ca2+ -
stimulated ATPase activity: Anal Biochem, v. 157, p. 375-80.
Chatterjee R, Millard CS, Champion K, Clark DP, and D. MI., 2001, Mutation of the
ptsG gene results in increased production of succinate in fermentation of glucose
by Escherichia coli.: Appl Environ Microbiol, v. 67, p. 148-154.
Chen, H., C. C. Tseng, B. K. Hubbard, and C. T. Walsh, 2001, Glycopeptide antibiotic
biosynthesis: enzymatic assembly of the dedicated amino acid monomer (S)-3,5-
dihydroxyphenylglycine: Proc Natl Acad Sci U S A, v. 98, p. 14901-6.
Cheng, X., and R. M. Blumenthal, 1999, S-Adenosylmethionine-Dependent
Methyltransferases: Structures and Functions.: New Jersey, World Scientific.
Cheng, X., S. Kumar, J. Posfai, J. W. Pflugrath, and R. J. Roberts, 1993, Crystal
structure of the HhaI DNA methyltransferase complexed with S-adenosyl-L-
methionine: Cell, v. 74, p. 299-307.
Cheng, X., and R. J. Roberts, 2001, AdoMet-dependent methylation, DNA
methyltransferases and base flipping: Nucleic Acids Res, v. 29, p. 3784-95.
Christiansen, G., J. Fastner, M. Erhard, T. Borner, and E. Dittmann, 2003, Microcystin
biosynthesis in planktothrix: genes, evolution, and manipulation: J Bacteriol, v.
185, p. 564-72.
Colman B, and N. EG, 1997, Serine synthesis in cyanobacteria by a
nonphotorespiratory pathway: Physiol Plant, v. 100, p. 133-136.
168 Cox, R. J., F. Glod, D. Hurley, C. M. Lazarus, T. P. Nicholson, B. A. Rudd, T. J.
Simpson, B. Wilkinson, and Y. Zhang, 2004, Rapid cloning and expression of a
fungal polyketide synthase gene involved in squalestatin biosynthesis: Chem
Commun (Camb), p. 2260-1.
Dassa, E., and M. Hofnung, 1985, Sequence of gene malG in E. coli K12: homologies
between integral membrane components from binding protein-dependent
transport systems.: Embo J, v. 4, p. 2287-2293.
Dawson, R. M., 1998, The toxicology of microcystins: Toxicon, v. 36, p. 953-62.
Decker, H., H. Motamedi, and C. R. Hutchinson, 1993, Nucleotide sequences and
heterologous expression of tcmG and tcmP, biosynthetic genes for
tetracenomycin C in Streptomyces glaucescens: J Bacteriol, v. 175, p. 3876-86.
DeMott, W., Q. Zang, and W. Carmichael, 1991, Effects of toxic cyanobacteria and
purified toxins on the survival and feeding of a copepod and three species of
Daphnia.: Limnol. Oceanog., v. 36, p. 1346-1357.
Devlin, J., O. Edwards, P. Gorham, M. Hunter, R. Pike, and B. Stavric, 1977, Anatoxin-
a, a toxic alkaloid from Anabaena flos-aquae NCR-44h.: Cana. J. of Chem., v.
55, p. 1367-1371.
Dhar, K., and J. P. Rosazza, 2000, Purification and characterisation of Streptomyces
griseus catechol O-methyltransferase: Appl Environ Microbiol, v. 66, p. 4877-
82.
Dickens, M. L., J. Ye, and W. R. Strohl, 1996, Cloning, sequencing, and analysis of
aklaviketone reductase from Streptomyces sp. strain C5: J Bacteriol, v. 178, p.
3384-8.
169 Dieckmann, R., Y. O. Lee, H. van Liempt, H. von Döhren , and H. Kleinkauf, 1995,
Expression of an active adenylate-forming domain of peptide synthetases
corresponding to acyl-CoA-synthetases: FEBS Lett, v. 357, p. 212-6.
Dittmann, E., B. A. Neilan, M. Erhard, H. von Döhren , and T. Borner, 1997, Insertional
mutagenesis of a peptide synthetase gene that is responsible for hepatotoxin
production in the cyanobacterium Microcystis aeruginosa PCC 7806: Mol
Microbiol, v. 26, p. 779-87.
Donadio, S., and L. Katz, 1992, Organization of the enzymatic domains in the
multifunctional polyketide synthase involved in erythromycin formation in
Saccharopolyspora erythraea: Gene, v. 111, p. 51-60.
Du, L., C. Sanchez, M. Chen, D. J. Edwards, and B. Shen, 2000, The biosynthetic gene
cluster for the antitumor drug bleomycin from Streptomyces verticillus
ATCC15003 supporting functional interactions between nonribosomal peptide
synthetases and a polyketide synthase: Chem Biol, v. 7, p. 623-42.
Dubin, P. L., and J. M. Principi, 1989, Optimization of size-exclusion separation of
proteins on a Superose column: J Chromatogr, v. 479, p. 159-64.
Duncan, M. W., G. Matanovic, and A. Cerpa-Poljak, 1993, Quantitative analysis of low
molecular weight compounds of biological interest by matrix-assisted laser
desorption ionization: Rapid Commun Mass Spectrom, v. 7, p. 1090-4.
Eriksson, J. E., D. Toivola, J. A. Meriluoto, H. Karaki, Y. G. Han, and D. Hartshorne,
1990, Hepatocyte deformation induced by cyanobacterial toxins reflects
inhibition of protein phosphatases: Biochem Biophys Res Commun, v. 173, p.
1347-53.
170 Falconer, I. R., and M. T. Runnegar, 1987, Effects of the peptide toxin from Microcystis
aeruginosa on intracellular calcium, pH and membrane integrity in mammalian
cells: Chem Biol Interact, v. 63, p. 215-25.
Fawaz, F., and G. H. Jones, 1988, Actinomycin synthesis in Streptomyces antibioticus.
Purification and properties of a 3-hydroxyanthranilate 4-methyltransferase: J
Biol Chem, v. 263, p. 4602-6.
Fernandez-Moreno, M. A., C. Vallin, and F. Malpartida, 1997, Streptothricin
biosynthesis is catalyzed by enzymes related to nonribosomal peptide bond
formation: J Bacteriol, v. 179, p. 6929-36.
Fiedler, G., M. Arnold, S. Hannus, and I. Maldener, 1998, The DevBCA exporter is
essential for envelope formation in heterocysts of the cyanobacterium Anabaena
sp. strain PCC 7120: Mol Microbiol, v. 27, p. 1193-202.
Finking, R., and M. A. Marahiel, 2004, Biosynthesis of nonribosomal peptides1: Annu
Rev Microbiol, v. 58, p. 453-88.
Fjaervik, E., and S. B. Zotchev, 2005, Biosynthesis of the polyene macrolide antibiotic
nystatin in Streptomyces noursei: Appl Microbiol Biotechnol, v. 67, p. 436-43.
Fromm, M. F., H. M. Kauffmann, P. Fritz, O. Burk, H. K. Kroemer, R. W. Warzok, M.
Eichelbaum, W. Siegmund, and D. Schrenk, 2000, The effect of rifampin
treatment on intestinal expression of human MRP transporters: Am J Pathol, v.
157, p. 1575-80.
Fujioka, M., 1992, Mammalian small molecule methyltransferases: their structural and
functional features: Int J Biochem, v. 24, p. 1917-24.
Gardiner, D. M., A. J. Cozijnsen, L. M. Wilson, M. S. Pedras, and B. J. Howlett, 2004,
The sirodesmin biosynthetic gene cluster of the plant pathogenic fungus
Leptosphaeria maculans: Mol Microbiol, v. 53, p. 1307-18.
171 Garnier, J., J. F. Gibrat, and B. Robson, 1996, GOR method for predicting protein
secondary structure from amino acid sequence: Methods Enzymol, v. 266, p.
540-53.
Goldberg, J. D., T. Yoshida, and P. Brick, 1994, Crystal structure of a NAD-dependent
D-glycerate dehydrogenase at 2.4 A resolution: J Mol Biol, v. 236, p. 1123-40.
Gomi, T., K. Tanihara, T. Date, and M. Fujioka, 1992, Rat guanidinoacetate
methyltransferase: mutation of amino acids within a common sequence motif of
mammalian methyltransferase does not affect catalytic activity but alters
proteolytic susceptibility: Int J Biochem, v. 24, p. 1639-49.
Grant, G. A., 1989, A new family of 2-hydroxyacid dehydrogenases.: Biochem Biophys
Res Commun, v. 165, p. 1371-1374.
Grant, G. A., Z. Hu, and X. Xu, 2001, Specific interactions at the regulatory domain-
substrate binding domain interface influence the cooperativity of inhibition and
effector binding in Escherichia coli D-3-phosphoglycerate dehydrogenase.: J
Biol Chem, v. 276, p. 1078-1083.
Grant, G. A., S. Kim, X. Xu, and Z. Hu, 1999, The contribution of adjacent subunits to
the active sites of D-3-phosphoglycerate dehydrogenase.: J Biol Chem, v. 274,
p. 5357-5361.
Grant, G. A., X. Xu, and Z. Hu, 2000, Removal of the tryptophan 139 side chain in
Escherichia coli D-3-phosphoglycerate dehydrogenase produces a dimeric
enzyme without cooperative effects.: Arch Biochem Biophys, v. 375, p. 171-
174.
Gross, E., and J. L. Morell, 1967, The presence of dehydroalanine in the antibiotic nisin
and its relationship to activity: J Am Chem Soc, v. 89, p. 2791-2.
172 Gross, E., J. L. Morell, and L. C. Craig, 1969, Dehydroalanyllysine: identical COOH-
terminal structures in the peptide antibiotics nisin and subtilin: Proc Natl Acad
Sci U S A, v. 62, p. 952-6.
Hagemann, M., S. Richter, and S. Mikkat, 1997, The ggtA gene encodes a subunit of
the transport system for the osmoprotective compound glucosylglycerol in
Synechocystis sp. strain PCC 6803: J Bacteriol, v. 179, p. 714-20.
Hansen, E. J., and E. Juni, 1979, Properties of mutants of Escherichia coli lacking malic
dehydrogenase and their revertants: J Biol Chem, v. 254, p. 3570-5.
Haydock, S. F., J. A. Dowson, N. Dhillon, G. A. Roberts, J. Cortes, and P. F. Leadlay,
1991, Cloning and sequence analysis of genes involved in erythromycin
biosynthesis in Saccharopolyspora erythraea: sequence similarities between
EryG and a family of S-adenosylmethionine-dependent methyltransferases: Mol
Gen Genet, v. 230, p. 120-8.
Heinzelmann, E., S. Berger, O. Puk, B. Reichenstein, W. Wohlleben, and D. Schwartz,
2003, A glutamate mutase is involved in the biosynthesis of the lipopeptide
antibiotic friulimicin in Actinoplanes friuliensis: Antimicrob Agents Chemother,
v. 47, p. 447-57.
Hendrickson, L., C. R. Davis, C. Roach, D. K. Nguyen, T. Aldrich, P. C. McAda, and
C. D. Reeves, 1999, Lovastatin biosynthesis in Aspergillus terreus:
characterisation of blocked mutants, enzyme activities and a multifunctional
polyketide synthase gene: Chem Biol, v. 6, p. 429-39.
Henning, M., H. Hertel, H. Wall, and J. G. Kohl., 1991, Strain-specific influence of
Microcystis aeruginosa on food ingestion and assimilation of some cladocerans
and copepods.: Int. Rev. Ges. Hydrobiol, v. 76, p. 37-45.
173 Herdman M, Castenholz RW, Iteman I, Waterbury JB, and R. R, 1979, Subsection I:
Order "Chroococcales" In: Bergey's manual of systematic bacteriology. The
archaea, cyanobacteria and deeply branching bacteria.: Bergey's manual of
systematic bacteriology, v. 1: New York, Springer.
Herter, S., Fuchs, G., Bacher, A., and Eisenreich, W. (2002) J. Biol. Chem. 277(23),
20277-20283.
Hicks, L. M., M. C. Moffitt, L. L. Beer, B. Moore, and N. L. Kelleher, 2006, Structural
characterisation of in vitro and in vivo intermediates on the loading module of
microcystin synthetase: ACS Chemical Biology, v. 1, p. 93-102.
Higgins, C. F., 2001, ABC transporters: physiology, structure and mechanism--an
overview: Res Microbiol, v. 152, p. 205-10.
Higgins, C. F., and K. J. Linton, 2001, Structural biology. The xyz of ABC transporters:
Science, v. 293, p. 1782-4.
Higgins, C. F., and K. J. Linton, 2004, The ATP switch model for ABC transporters:
Nat Struct Mol Biol, v. 11, p. 918-26.
Ho, C. L., M. Noji, M. Saito, and K. Saito, 1999, Regulation of serine biosynthesis in
Arabidopsis. Crucial role of plastidic 3-phosphoglycerate dehydrogenase in non-
photosynthetic tissues: J Biol Chem, v. 274, p. 397-402.
Hoffmann, D., J. M. Hevel, R. E. Moore, and B. S. Moore, 2003, Sequence analysis and
biochemical characterisation of the nostopeptolide A biosynthetic gene cluster
from Nostoc sp. GSV224: Gene, v. 311, p. 171-80.
Hoffmann, K., E. Schneider-Scherzer, H. Kleinkauf, and R. Zocher, 1994, Purification
and characterisation of eucaryotic alanine racemase acting as key enzyme in
cyclosporin biosynthesis: J Biol Chem, v. 269, p. 12710-4.
174 Holland, I. B., and M. A. Blight, 1999, ABC-ATPases, adaptable energy generators
fuelling transmembrane movement of a variety of molecules in organisms from
bacteria to humans: J Mol Biol, v. 293, p. 381-99.
Hopfner, K. P., A. Karcher, D. S. Shin, L. Craig, L. M. Arthur, J. P. Carney, and J. A.
Tainer, 2000, Structural biology of Rad50 ATPase: ATP-driven conformational
control in DNA double-strand break repair and the ABC-ATPase superfamily:
Cell, v. 101, p. 789-800.
Hopwood, D. A., F. Malpartida, H. M. Kieser, H. Ikeda, J. Duncan, I. Fujii, B. A. Rudd,
H. G. Floss, and S. Omura, 1985, Production of 'hybrid' antibiotics by genetic
engineering: Nature, v. 314, p. 642-4.
Huh, J. H., D. J. Kim, X. Q. Zhao, M. Li, Y. Y. Jo, T. M. Yoon, S. K. Shin, J. H. Yong,
Y. W. Ryu, Y. Y. Yang, and J. W. Suh, 2004, Widespread activation of
antibiotic biosynthesis by S-adenosylmethionine in streptomycetes: FEMS
Microbiol Lett, v. 238, p. 439-47.
Humble, A., G. Gadd, and G. Codd, 1994, Polarographic analysis of the interactions
between cyanobacterial variants and metal cations.: VIII International
Symposium on Phototrophic Prokaryotes, p. 82.
Humpage, A. R., J. Rositano, P. D. Baker, B. C. Nicholson, D. A. Steffensen, A. H.
Bretag, and R. K. Brown, 1993, Paralytic shellfish poisons from freshwater
blue-green algae: Med J Aust, v. 159, p. 423.
Ikeda, H., and S. Omura, 1995: J. Antibiot., v. 48, p. 95.
Irvine, G. B., 1987, High performance size exclusion chromatography of polypeptides
on a TSK G2000SW column in acidic mobile phases: Journal of
Chromatography, v. 404, p. 215-222.
175 Jackson, A. R., A. McInnes, I. R. Falconer, and M. T. Runnegar, 1984, Clinical and
pathological changes in sheep experimentally poisoned by the blue-green alga
Microcystis aeruginosa: Vet Pathol, v. 21, p. 102-13.
Jackson, A. R. B. M., A.; Falconer, I. R.; Runnegar, M. T. C. Vet Path 1984, 21, 102-
113. Links
Jenke-Kodama, H., A. Sandmann, R. Muller, and E. Dittmann, 2005, Evolutionary
implications of bacterial polyketide synthases: Mol Biol Evol, v. 22, p. 2027-39.
Jochimsen, E. M., W. W. Carmichael, J. S. An, D. M. Cardo, S. T. Cookson, C. E.
Holmes, M. B. Antunes, D. A. de Melo Filho, T. M. Lyra, V. S. Barreto, S. M.
Azevedo, and W. R. Jarvis, 1998, Liver failure and death after exposure to
microcystins at a hemodialysis center in Brazil: N Engl J Med, v. 338, p. 873-8.
Jones, P. M., and A. M. George, 1999, Subunit interactions in ABC transporters:
towards a functional architecture: FEMS Microbiol Lett, v. 179, p. 187-202.
Jovell, R. J., A. J. Macario, and E. C. de Macario, 1996, ABC transporters in Archaea:
two genes encoding homologs of the nucleotide-binding components in the
methanogen Methanosarcina mazei S-6: Gene, v. 174, p. 281-4.
Jungblut, A. D., I. Hawes, D. Mountfort, B. Hitzfeld, D. R. Dietrich, B. P. Burns, and B.
A. Neilan, 2005, Diversity within cyanobacterial mat communities in variable
salinity meltwater ponds of McMurdo Ice Shelf, Antarctica: Environ Microbiol,
v. 7, p. 519-29.
Kaebernick, M., E. Dittmann, T. Borner, and B. A. Neilan, 2002, Multiple alternate
transcripts direct the biosynthesis of microcystin, a cyanobacterial nonribosomal
peptide: Appl Environ Microbiol, v. 68, p. 449-55.
176 Kaebernick, M., B. A. Neilan, T. Borner, and E. Dittmann, 2000, Light and the
transcriptional response of the microcystin biosynthesis gene cluster: Appl
Environ Microbiol, v. 66, p. 3387-92.
Kaebernick, M., T. Rohrlack, K. Christoffersen, and B. A. Neilan, 2001, A spontaneous
mutant of microcystin biosynthesis: genetic characterisation and effect on
Daphnia: Environ Microbiol, v. 3, p. 669-79.
Kagan, R. M., and S. Clarke, 1994, Widespread occurrence of three sequence motifs in
diverse S-adenosylmethionine-dependent methyltransferases suggests a common
structure for these enzymes: Arch Biochem Biophys, v. 310, p. 417-27.
Kakinuma, K., Teresawa, H., Li, H.-Y., Miyazaki, K., and Oshima, T. (1993) Biosci.
Biotech. Biochem. (Tokyo). 57, 1916-1923
Kapitonov, D., and R. Yu, 1999, Conserved domains of glycosyltransferase:
Glycobiology, v. 9, p. 961-978.
Kato, Y., and Y. Asano, 1997, 3-Methylaspartate ammonia-lyase as a marker enzyme of
the mesaconate pathway for (S)-glutamate fermentation in Enterobacteriaceae:
Arch Microbiol, v. 168, p. 457-63.
Katz, L., 1997, Manipulation of Modular Polyketide Synthases: Chem Rev, v. 97, p.
2557-2576.
Keating, K., 1978, Blue-green algal inhibition of diatom growth: transition from
mesotrophic to eutrophic community structure: Science, v. 199, p. 971-973.
Keating, T. A., C. G. Marshall, C. T. Walsh, and A. E. Keating, 2002, The structure of
VibH represents nonribosomal peptide synthetase condensation, cyclization and
epimerization domains: Nat Struct Biol, v. 9, p. 522-6.
177 Kennedy, J., K. Auclair, S. G. Kendrew, C. Park, J. C. Vederas, and C. R. Hutchinson,
1999, Modulation of polyketide synthase activity by accessory proteins during
lovastatin biosynthesis: Science, v. 284, p. 1368-72.
Kerppola, R. E., and G. F. Ames, 1992, Topology of the hydrophobic membrane-bound
components of the histidine periplasmic permease. Comparison with other
members of the family: J Biol Chem, v. 267, p. 2329-36.
Kessler, N., H. Schuhmann, S. Morneweg, U. Linne, and M. A. Marahiel, 2004, The
linear pentadecapeptide gramicidin is assembled by four multimodular
nonribosomal peptide synthetases that comprise 16 modules with 56 catalytic
domains: J Biol Chem, v. 279, p. 7413-9.
Kim, A. D., A. S. Baker, D. Dunaway-Mariano, W. W. Metcalf, B. L. Wanner, and B.
M. Martin, 2002, The 2-aminoethylphosphonate-specific transaminase of the 2-
aminoethylphosphonate degradation pathway: J Bacteriol, v. 184, p. 4134-40.
Kleinkauf, H., and H. Von Döhren , 1996, A nonribosomal system of peptide
biosynthesis: Eur J Biochem, v. 236, p. 335-51.
Kochhar, S., P. E. Hunziker, P. Leong-Morgenthaler, and H. Hottinger, 1992,
Evolutionary relationship of NAD(+)-dependent D-lactate dehydrogenase:
comparison of primary structure of 2-hydroxy acid dehydrogenases: Biochem
Biophys Res Commun, v. 184, p. 60-6.
Lambalot, R. H., A. M. Gehring, R. S. Flugel, P. Zuber, M. LaCelle, M. A. Marahiel, R.
Reid, C. Khosla, and C. T. Walsh, 1996, A new enzyme superfamily - the
phosphopantetheinyl transferases: Chem Biol, v. 3, p. 923-36.
Lau, E., C. Z. Nash, D. R. Vogler, and K. W. Cullings, 2005, Molecular diversity of
cyanobacteria inhabiting coniform structures and surrounding mat in a
Yellowstone hot spring: Astrobiology, v. 5, p. 83-92.
178 Lawen, A., and R. Zocher, 1990, Cyclosporin synthetase. The most complex peptide
synthesizing multienzyme polypeptide so far described.: J Biol Chem, v. 5;265,
p. 11355-11360.
Lazarow, P. B., 1987, The role of peroxisomes in mammalian cellular metabolism: J
Inherit Metab Dis, v. 10 Suppl 1, p. 11-22.
Lehtimaeki, K., K. Sivonen, R. Luukkainen, and S. I. Niemelae., 1994, The effects of
incubation time, temperature, light, salinity, and phosphorus on growth and
heptotoxin production by Nodularia strains.: Arch. Hydrobiol., v. 130, p. 269-
282.
Li, S. M., L. Westrich, J. Schmidt, C. Kuhnt, and L. Heide, 2002, Methyltransferase
genes in Streptomyces rishiriensis: new coumermycin derivatives from gene-
inactivation experiments: Microbiology, v. 148, p. 3317-26.
Li, T. L., F. Huang, S. F. Haydock, T. Mironenko, P. F. Leadlay, and J. B. Spencer,
2004, Biosynthetic gene cluster of the glycopeptide antibiotic teicoplanin:
characterisation of two glycosyltransferases and the key acyltransferase: Chem
Biol, v. 11, p. 107-19.
Ligon, J., S. Hill, J. Beck, R. Zirkle, I. Molnar, J. Zawodny, S. Money, and T. Schupp,
2002, Characterisation of the biosynthetic gene cluster for the antifungal
polyketide soraphen A from Sorangium cellulosum So ce26: Gene, v. 285, p.
257-67.
Linne, U., S. Doekel, and M. A. Marahiel, 2001, Portability of epimerization domain
and role of peptidyl carrier protein on epimerization activity in nonribosomal
peptide synthetases: Biochemistry, v. 40, p. 15824-34.
179 Long, B. M., G. J. Jones, and P. T. Orr, 2001, Cellular microcystin content in N-limited
Microcystis aeruginosa can be predicted from growth rate: Appl Environ
Microbiol, v. 67, p. 278-83.
Losey, H. C., M. W. Peczuh, Z. Chen, U. S. Eggert, S. D. Dong, I. Pelczer, D. Kahne,
and C. T. Walsh, 2001, Tandem action of glycosyltransferases in the maturation
of vancomycin and teicoplanin aglycones: novel glycopeptides: Biochemistry, v.
40, p. 4745-55.
Lukac, M., and R. Aegerter, 1993, Influence of trace metals on growth and toxin
production of Microcystis aeruginosa: Toxicon, v. 31, p. 293-305.
Luo, L., R. M. Kohli, M. Onishi, U. Linne, M. A. Marahiel, and C. T. Walsh, 2002,
Timing of epimerization and condensation reactions in nonribosomal peptide
assembly lines: kinetic analysis of phenylalanine activating elongation modules
of tyrocidine synthetase B: Biochemistry, v. 41, p. 9184-96.
Marahiel, M. A., 1997, Protein templates for the biosynthesis of peptide antibiotics:
Chem Biol, v. 4, p. 561-7.
Martin, C., L. Oberer, T. Ino, W. A. Konig, M. Busch, and J. Weckesser, 1993,
Cyanopeptolins, new depsipeptides from the cyanobacterium Microcystis sp.
PCC 7806: J Antibiot (Tokyo), v. 46, p. 1550-6.
Maruyama, T., E. Hirose, and M. Ishikura, 2003, Ultraviolet-light-absorbing tunic cells
in didemnid ascidians hosting a symbiotic photo-oxygenic prokaryote,
Prochloron: Biol Bull, v. 204, p. 109-13.
Matsunaga, S., R. Moore, and W. Carmichael, 1989, Anatoxin-a(s), a potent
anticholinesterase from Anabaena flos-aquae.: J Am Chem Soc, v. 111, p. 8021-
8023.
180 Matveyev, A. V., K. T. Young, A. Meng, and J. Elhai, 2001, DNA methyltransferases
of the cyanobacterium Anabaena PCC 7120: Nucleic Acids Res, v. 29, p. 1491-
506.
May, J. J., N. Kessler, M. A. Marahiel, and M. T. Stubbs, 2002, Crystal structure of
DhbE, an archetype for aryl acid activating domains of modular nonribosomal
peptide synthetases: Proc Natl Acad Sci U S A, v. 99, p. 12120-5.
Miller, D. A., L. Luo, N. Hillson, T. A. Keating, and C. T. Walsh, 2002, Yersiniabactin
synthetase: a four-protein assembly line producing the nonribosomal
peptide/polyketide hybrid siderophore of Yersinia pestis: Chem Biol, v. 9, p.
333-44.
Moffitt, M. C., S. I. Blackburn, and B. A. Neilan, 2001, rRNA sequences reflect the
ecophysiology and define the toxic cyanobacteria of the genus Nodularia: Int J
Syst Evol Microbiol, v. 51, p. 505-12.
Moffitt, M. C., and B. A. Neilan, 2001, On the presence of peptide synthetase and
polyketide synthase genes in the cyanobacterial genus Nodularia: FEMS
Microbiol Lett, v. 196, p. 207-14.
Moffitt, M. C., and B. A. Neilan, 2003, Evolutionary affiliations within the superfamily
of ketosynthases reflect complex pathway associations: J Mol Evol, v. 56, p.
446-57.
Moffitt, M. C., and B. A. Neilan, 2004, Characterisation of the nodularin synthetase
gene cluster and proposed theory of the evolution of cyanobacterial
hepatotoxins: Appl Environ Microbiol, v. 70, p. 6353-62.
Molnar, I., J. F. Aparicio, S. F. Haydock, L. E. Khaw, T. Schwecke, A. Konig, J.
Staunton, and P. F. Leadlay, 1996, Organisation of the biosynthetic gene cluster
181 for rapamycin in Streptomyces hygroscopicus: analysis of genes flanking the
polyketide synthase: Gene, v. 169, p. 1-7.
Moore, R., J. Chen, B. Moore, and G. Patterson, 1991, Biosynthesis of Microcystin-LR.
Origin of the carbons in the Adda and Masp Units.: J Am Chem Soc, v. 113, p.
5083-5084.
Mosser, J., A. M. Douar, C. O. Sarde, P. Kioschis, R. Feil, H. Moser, A. M. Poustka, J.
L. Mandel, and P. Aubourg, 1993, Putative X-linked adrenoleukodystrophy gene
shares unexpected homology with ABC transporters: Nature, v. 361, p. 726-30.
Mulichak, A. M., W. Lu, H. C. Losey, C. T. Walsh, and R. M. Garavito, 2004, Crystal
structure of vancosaminyltransferase GtfD from the vancomycin biosynthetic
pathway: interactions with acceptor and nucleotide ligands: Biochemistry, v. 43,
p. 5170-80.
Namikoshi, M., K. Sivonen, W. R. Evans, W. W. Carmichael, L. Rouhiainen, R.
Luukkainen, and K. L. Rinehart, 1992, Structures of three new homotyrosine-
containing microcystins and a new homophenylalanine variant from Anabaena
sp. strain 66: Chem Res Toxicol, v. 5, p. 661-6.
Neilan, B. A., E. Dittmann, L. Rouhiainen, R. A. Bass, V. Schaub, K. Sivonen, and T.
Borner, 1999, Nonribosomal peptide synthesis and toxigenicity of
cyanobacteria: J Bacteriol, v. 181, p. 4089-97.
Neilan, B. A., D. Jacobs, T. Del Dot, L. L. Blackall, P. R. Hawkins, P. T. Cox, and A.
E. Goodman, 1997, rRNA sequences and evolutionary relationships among toxic
and nontoxic cyanobacteria of the genus Microcystis: Int J Syst Bacteriol, v. 47,
p. 693-7.
182 Neumuller, A. M., D. Konz, and M. A. Marahiel, 2001, The two-component regulatory
system BacRS is associated with bacitracin 'self-resistance' of Bacillus
licheniformis ATCC 10716: Eur J Biochem, v. 268, p. 3180-9.
Nikaido, K., P. Q. Liu, and G. F. Ames, 1997, Purification and characterisation of HisP,
the ATP-binding subunit of a traffic ATPase (ABC transporter), the histidine
permease of Salmonella typhimurium. Solubility, dimerization, and ATPase
activity: J Biol Chem, v. 272, p. 27745-52.
Nishiwaki, S., H. Fujiki, M. Suganuma, R. Nishiwaki-Matsushima, and T. Sugimura,
1991, Rapid purification of protein phosphatase 2A from mouse brain by
microcystin-affinity chromatography: FEBS Lett, v. 279, p. 115-8.
Nishiwaki-Matsushima, R., T. Ohta, S. Nishiwaki, M. Suganuma, K. Kohyama, T.
Ishikawa, W. W. Carmichael, and H. Fujiki, 1992, Liver tumor promotion by the
cyanobacterial cyclic peptide toxin microcystin-LR: J Cancer Res Clin Oncol, v.
118, p. 420-4.
Nishizawa, T., M. Asayama, K. Fujii, K. Harada, and M. Shirai, 1999, Genetic analysis
of the peptide synthetase genes for a cyclic heptapeptide microcystin in
Microcystis spp: J Biochem (Tokyo), v. 126, p. 520-9.
Nishizawa, T., M. Asayama, and M. Shirai, 2001, Cyclic heptapeptide microcystin
biosynthesis requires the glutamate racemase gene: Microbiology, v. 147, p.
1235-41.
Norris, R. L., A. A. Seawright, G. R. Shaw, P. Senogles, G. K. Eaglesham, M. J. Smith,
R. K. Chiswell, and M. R. Moore, 2002, Hepatic xenobiotic metabolism of
cylindrospermopsin in vivo in the mouse: Toxicon, v. 40, p. 471-6.
O'Brien. D. P, P. N. Kirkpatrick, S. W. O'Brien, T. Staroske, T. I. Richardson, D. A.
Evans, A. Hopkinson, J. B. Spencer, and D. H. Williams, 2000, Expression and
183 assay of an N-methyltransferase involved in the biosynthesis of a vancomycin
group antibiotic: Chemical Communications, p. 103-104.
Omata, T., 1995, Structure, function and regulation of the nitrate transport system of the
cyanobacterium Synechococcus sp. PCC7942: Plant Cell Physiol, v. 36, p. 207-
13.
Paik, S. H., A. Chakicherla, and J. N. Hansen, 1998, Identification and characterisation
of the structural and transporter genes for, and the chemical and biological
properties of, sublancin 168, a novel lantibiotic produced by Bacillus subtilis
168: J Biol Chem, v. 273, p. 23134-42.
Parkhill, J., G. Dougan, K. D. James, N. R. Thomson, D. Pickard, J. Wain, C. Churcher,
K. L. Mungall, S. D. Bentley, M. T. Holden, M. Sebaihia, S. Baker, D. Basham,
K. Brooks, T. Chillingworth, P. Connerton, A. Cronin, P. Davis, R. M. Davies,
L. Dowd, N. White, J. Farrar, T. Feltwell, N. Hamlin, A. Haque, T. T. Hien, S.
Holroyd, K. Jagels, A. Krogh, T. S. Larsen, S. Leather, S. Moule, P. O'Gaora, C.
Parry, M. Quail, K. Rutherford, M. Simmonds, J. Skelton, K. Stevens, S.
Whitehead, and B. G. Barrell, 2001, Complete genome sequence of a multiple
drug resistant Salmonella enterica serovar Typhi CT18: Nature, v. 413, p. 848-
52.
Pascard, C., A. Ducruix, J. Lunel, and T. Prange, 1977, Highly modified cysteine-
containing antibiotics. Chemical structure and configuration of nosiheptide: J
Am Chem Soc, v. 99, p. 6418-23.
Paulsen, I. T., M. K. Sliwinski, and M. H. Saier, Jr., 1998, Microbial genome analyses:
global comparisons of transport capabilities based on phylogenies, bioenergetics
and substrate specificities: J Mol Biol, v. 277, p. 573-92.
184 Paulsrud, P., and P. Lindblad, 1998, Sequence variation of the tRNA(Leu) intron as a
marker for genetic diversity and specificity of symbiotic cyanobacteria in some
lichens: Appl Environ Microbiol, v. 64, p. 310-5.
Paulus, T. J., J. S. Tuan, V. E. Luebke, G. T. Maine, J. P. DeWitt, and L. Katz, 1990,
Mutation and cloning of eryG, the structural gene for erythromycin O-
methyltransferase from Saccharopolyspora erythraea, and expression of eryG in
Escherichia coli: J Bacteriol, v. 172, p. 2541-6.
Pearce. C. J., and K. L. Rinehart, 1979, Berninamycin biosynthesis. Origin of the
dehydroalanine residues: Journal of the American Chemical Society, v. 10, p.
5069-5070.
Pearl, H., and D. Millie, 1996, Physiological ecology of toxic aquatic cyanobacteria:
Phycologia, v. 35, p. 160-167.
Pearson, L. A., M. Hisbergues, T. Borner, E. Dittmann, and B. A. Neilan, 2004,
Inactivation of an ABC transporter gene, mcyH, results in loss of microcystin
production in the cyanobacterium Microcystis aeruginosa PCC 7806: Appl
Environ Microbiol, v. 70, p. 6370-8.
Proctor, R. H., D. W. Brown, R. D. Plattner, and A. E. Desjardins, 2003, Co-expression
of 15 contiguous genes delineates a fumonisin biosynthetic gene cluster in
Gibberella moniliformis: Fungal Genet Biol, v. 38, p. 237-49.
Promega, 1993, Promega protein guide: Tips and techniques: WI, USA, Promega
Corporation.
Quentin, Y., G. Fichant, and F. Denizot, 1999, Inventory, assembly and analysis of
Bacillus subtilis ABC transport systems: J Mol Biol, v. 287, p. 467-84.
185 Rakin, A., E. Boolgakowa, and J. Heesemann, 1996, Structural and functional
organization of the Yersinia pestis bacteriocin pesticin gene cluster:
Microbiology, v. 142 ( Pt 12), p. 3415-24.
Rantala, A., D. P. Fewer, M. Hisbergues, L. Rouhiainen, J. Vaitomaa, T. Borner, and K.
Sivonen, 2004, Phylogenetic evidence for the early evolution of microcystin
synthesis: Proc Natl Acad Sci U S A, v. 101, p. 568-73.
Rapala, J., K. Sivonen, C. Lyra, and S. I. Niemela, 1997, Variation of microcystins,
cyanobacterial hepatotoxins, in Anabaena spp. as a function of growth stimuli:
Appl Environ Microbiol, v. 63, p. 2206-12.
Regnier, F. E., 1983, High-performance liquid chromatography of proteins: Methods
Enzymol, v. 91, p. 137-90.
Renaud, P., Hurzler, M., and Seebach, D. (1987) Helv. Chim. Acta. 70, 292-298
Rinehart, K., N. Namikoshi, and B. Choi, 1994, Structure and biosynthesis of toxins
from blue-green algae (cyanobacteria): J. App. Phycol., v. 6, p. 159-176.
Rinehart, K. L., V. Kishore, K. C. Bible, R. Sakai, D. W. Sullins, and K. M. Li, 1988,
Didemnins and tunichlorin: novel natural products from the marine tunicate
Trididemnum solidum: J Nat Prod, v. 51, p. 1-21.
Rippka, R., J. Deruelles, J. B. Waterbury, M.Herdman, and R. Y. Stanier, 1979, Generic
Assignments, Strain Histories and Properties of Pure Cultures of Cyanobacteria.:
Journal of General Microbiology, v. 111, p. 1-61.
Rippka, R., and M. Herdman, 1992, Pasteur Culture Collection of Cyanobacterial
Strains in Axenic Culture. Catalogue and taxonomic handbook., v. 1: Paris,
Institute Pasteur.
186 Rix, U., C. Fischer, L. L. Remsing, and J. Rohr, 2002, Modification of post-PKS
tailoring steps through combinatorial biosynthesis: Nat Prod Rep, v. 19, p. 542-
80.
Rodriguez, D., L. M. Quiros, and J. A. Salas, 2004, MtmMII-mediated C-methylation
during biosynthesis of the antitumor drug mithramycin is essential for biological
activity and DNA-drug interaction: J Biol Chem, v. 279, p. 8149-58.
Romano, J. D., and S. Michaelis, 2001, Topological and mutational analysis of
Saccharomyces cerevisiae Ste14p, founding member of the isoprenylcysteine
carboxyl methyltransferase family: Mol Biol Cell, v. 12, p. 1957-71.
Rosen, B. P., and T. Tsuchiya, 1979, Preparation of everted membrane vesicles from
Escherichia coli for the measurement of calcium transport: Methods Enzymol, v.
56, p. 233-41.
Rouhiainen, L., L. Paulin, S. Suomalainen, H. Hyytiainen, W. Buikema, R. Haselkorn,
and K. Sivonen, 2000, Genes encoding synthetases of cyclic depsipeptides,
anabaenopeptilides, in Anabaena strain 90: Mol Microbiol, v. 37, p. 156-67.
Rouhiainen, L., T. Vakkilainen, B. L. Siemer, W. Buikema, R. Haselkorn, and K.
Sivonen, 2004, Genes coding for hepatotoxic heptapeptides (microcystins) in the
cyanobacterium Anabaena strain 90: Appl Environ Microbiol, v. 70, p. 686-92.
Runnegar, M. T., R. G. Gerdes, and I. R. Falconer, 1991, The uptake of the
cyanobacterial hepatotoxin microcystin by isolated rat hepatocytes: Toxicon, v.
29, p. 43-51.
Runnegar, M. T., A. R. Jackson, and I. R. Falconer, 1988, Toxicity to mice and sheep of
a bloom of the cyanobacterium (blue-green alga) Anabaena circinalis: Toxicon,
v. 26, p. 599-602.
187 Sahin, A., F. G. Tencalla, D. R. Dietrich, K. Mez, and H. Naegeli, 1995, Enzymatic
analysis of liver samples from rainbow trout for diagnosis of blue-green algae-
induced toxicosis: Am J Vet Res, v. 56, p. 1110-5.
Saitou, N., and M. Nei, 1987, The neighbor-joining method: a new method for
reconstructing phylogenetic trees: Mol Biol Evol, v. 4, p. 406-25.
Saker, M. L., A. D. Jungblut, B. A. Neilan, D. F. Rawn, and V. M. Vasconcelos, 2005,
Detection of microcystin synthetase genes in health food supplements
containing the freshwater cyanobacterium Aphanizomenon flos-aquae: Toxicon,
v. 46, p. 555-62.
Sambrook, J., E. F. Fritsch, and T. Maniatis, 1989, Molecular cloning: A laboratory
manual: New York, Cold Spring Harbor Laboratory Press.
Saurin, W., M. Hofnung, and E. Dassa, 1999, Getting in or out: early segregation
between importers and exporters in the evolution of ATP-binding cassette
(ABC) transporters: J Mol Evol, v. 48, p. 22-41.
Schauwecker, F., F. Pfennig, N. Grammel, and U. Keller, 2000, Construction and in
vitro analysis of a new bi-modular polypeptide synthetase for synthesis of N-
methylated acyl peptides: Chem Biol, v. 7, p. 287-97.
Schneider, E., and S. Hunke, 1998, ATP-binding-cassette (ABC) transport systems:
functional and structural aspects of the ATP-hydrolyzing subunits/domains:
FEMS Microbiol Rev, v. 22, p. 1-20.
Schopf, J. W., 2000, The Ecology of Cyanobacteria: Their Diversity in Time and Space:
Dordrecht, Kluwer Academic Publishers, 13-35 p.
Schuller, D. J., G. A. Grant, and L. J. Banaszak, 1995, The allosteric ligand site in the
Vmax-type cooperative enzyme phosphoglycerate dehydrogenase: Nat Struct
Biol, v. 2, p. 69-76.
188 Schwecke, T., J. F. Aparicio, I. Molnar, A. Konig, L. E. Khaw, S. F. Haydock, M.
Oliynyk, P. Caffrey, J. Cortes, J. B. Lester, and et al., 1995, The biosynthetic
gene cluster for the polyketide immunosuppressant rapamycin: Proc Natl Acad
Sci U S A, v. 92, p. 7839-43.
Sen, A., A. Narbad, N. Horn, H. Dodd, A. Par, I. Colquhoun, and M. Gasson, 1999,
Post-translational modification of nisin: The involvement of NisB in the
dehydration process.: FEBS, v. 261, p. 524-532.
Shah, S., Q. Xue, L. Tang, J. R. Carney, M. Betlach, and R. McDaniel, 2000, Cloning,
characterisation and heterologous expression of a polyketide synthase and P-450
oxidase involved in the biosynthesis of the antibiotic oleandomycin: J Antibiot
(Tokyo), v. 53, p. 502-8.
Shen, B., 2003, Polyketide biosynthesis beyond the type I, II and III polyketide synthase
paradigms: Curr Opin Chem Biol, v. 7, p. 285-95.
Shen, B., L. Du, C. Sanchez, D. J. Edwards, M. Chen, and J. M. Murrell, 2001, The
biosynthetic gene cluster for the anticancer drug bleomycin from Streptomyces
verticillus ATCC15003 as a model for hybrid peptide-polyketide natural product
biosynthesis: J Ind Microbiol Biotechnol, v. 27, p. 378-85.
Shen, B., and C. R. Hutchinson, 1994, Triple hydroxylation of tetracenomycin A2 to
tetracenomycin C in Streptomyces glaucescens. Overexpression of the tcmG
gene in Streptomyces lividans and characterisation of the tetracenomycin A2
oxygenase: J Biol Chem, v. 269, p. 30726-33.
Shen, X., P. K. Lam, G. R. Shaw, and W. Wickramasinghe, 2002, Genotoxicity
investigation of a cyanobacterial toxin, cylindrospermopsin: Toxicon, v. 40, p.
1499-501.
189 Shi, J., T. L. Blundell, and K. Mizuguchi, 2001, FUGUE: sequence-structure homology
recognition using environment-specific substitution tables and structure-
dependent gap penalties: J Mol Biol, v. 310, p. 243-57.
Shi, L., W. W. Carmichael, and I. Miller, 1995, Immuno-gold localization of
hepatotoxins in cyanobacterial cells: Arch Microbiol, v. 163, p. 7-15.
Siegers, K., S. Heinzmann, and K. D. Entian, 1996, Biosynthesis of lantibiotic nisin.
Posttranslational modification of its prepeptide occurs at a multimeric
membrane-associated lanthionine synthetase complex: J Biol Chem, v. 271, p.
12294-301.
Sielaff, H., E. Dittmann, N. Tandeau De Marsac, C. Bouchier, H. Von Döhren , T.
Borner, and T. Schwecke, 2003, The mcyF gene of the microcystin biosynthetic
gene cluster from Microcystis aeruginosa encodes an aspartate racemase:
Biochem J, v. 373, p. 909-16.
Silakowski, B., H. U. Schairer, H. Ehret, B. Kunze, S. Weinig, G. Nordsiek, P. Brandt,
H. Blocker, G. Hofle, S. Beyer, and R. Muller, 1999, New lessons for
combinatorial biosynthesis from myxobacteria. The myxothiazol biosynthetic
gene cluster of Stigmatella aurantiaca DW4/3-1: J Biol Chem, v. 274, p. 37391-
9.
Sivonen, K., 1990, Effects of light, temperature, nitrate, orthophosphate, and bacteria on
growth of and hepatotoxin production by Oscillatoria agardhii strains: Appl
Environ Microbiol, v. 56, p. 2658-66.
Sivonen, K., K. Himberg, R. Luukkainen, S. I. Niemela, G. K. Poon, and G. A. Codd,
1989, Preliminary characterisation of neurotoxic cyanobacteria blooms and
strains from Finland.: Toxic Assess, v. 4, p. 339–352.
190 Sivonen, K., and G. Jones, 1999, Toxic cyanobacteria in water: A guide to their public
health consequences, monitoring and management, v. 1: New York, E and FN
Spon, 40-111 p.
Sivonen, K., M. Namikoshi, W. R. Evans, W. W. Carmichael, F. Sun, L. Rouhiainen, R.
Luukkainen, and K. L. Rinehart, 1992a, Isolation and characterisation of a
variety of microcystins from seven strains of the cyanobacterial genus
Anabaena: Appl Environ Microbiol, v. 58, p. 2495-500.
Sivonen, K., M. Namikoshi, W. R. Evans, M. Fardig, W. W. Carmichael, and K. L.
Rinehart, 1992b, Three new microcystins, cyclic heptapeptide hepatotoxins,
from Nostoc sp. strain 152: Chem Res Toxicol, v. 5, p. 464-9.
Smith, D. R., L. A. Doucette-Stamm, C. Deloughery, H. Lee, J. Dubois, T. Aldredge, R.
Bashirzadeh, D. Blakely, R. Cook, K. Gilbert, D. Harrison, L. Hoang, P. Keagle,
W. Lumm, B. Pothier, D. Qiu, R. Spadafora, R. Vicaire, Y. Wang, J.
Wierzbowski, R. Gibson, N. Jiwani, A. Caruso, D. Bush, J. N. Reeve, and et al.,
1997, Complete genome sequence of Methanobacterium thermoautotrophicum
deltaH: functional analysis and comparative genomics: J Bacteriol, v. 179, p.
7135-55.
Song, L., T. Sano, R. Li, M. Watanabe, Y. Liu, and K. Kaya, 1998, Microcystin
production of Microcystis viridis (cyanobacteria) under different culture
conditions.: Phycological research, v. 42, p. 19.
Song, Z., R. J. Cox, C. M. Lazarus, and T. T. Simpson, 2004, Fusarin C biosynthesis in
Fusarium moniliforme and Fusarium venenatum: Chembiochem, v. 5, p. 1196-
1203.
191 Stachelhaus, T., and C. T. Walsh, 2000, Mutational analysis of the epimerization
domain in the initiation module PheATE of gramicidin S synthetase:
Biochemistry, v. 39, p. 5775-87.
Starr, R. C., and J. A. Zeikus, 1993, UTEX—the culture collection of algae at the
University of Texas at Austin.: J. Phycol., v. 29, p. 1-106.
Strichartz, G., T. Rando, S. Hall, J. Gitschier, L. Hall, B. Magnani, and C. H. Bay,
1986, On the mechanism by which saxitoxin binds to and blocks sodium
channels: Ann N Y Acad Sci, v. 479, p. 96-112.
Sugimoto, E., and L. I. Pizer, 1968, The mechanism of end product inhibition of serine
biosynthesis. I. Purification and kinetics of phosphoglycerate dehydrogenase: J
Biol Chem, v. 243, p. 2081-9.
Summers, R. G., E. Wendt-Pienkowski, H. Motamedi, and C. R. Hutchinson, 1992,
Nucleotide sequence of the tcmII-tcmIV region of the tetracenomycin C
biosynthetic gene cluster of Streptomyces glaucescens and evidence that the
tcmN gene encodes a multifunctional cyclase-dehydratase-O-methyl transferase:
J Bacteriol, v. 174, p. 1810-20.
Taguchi, H., and T. Ohta, 1991, D-lactate dehydrogenase is a member of the D-isomer-
specific 2-hydroxyacid dehydrogenase family. Cloning, sequencing, and
expression in Escherichia coli of the D-lactate dehydrogenase gene of
Lactobacillus plantarum: J Biol Chem, v. 266, p. 12588-94.
Terao, K., S. Ohmori, K. Igarashi, I. Ohtani, M. F. Watanabe, K. I. Harada, E. Ito, and
M. Watanabe, 1994, Electron microscopic studies on experimental poisoning in
mice induced by cylindrospermopsin isolated from blue-green alga Umezakia
natans: Toxicon, v. 32, p. 833-43.
192 Thiel, T., and C. P. Wolk, 1983, Metabolic activities of isolated akinetes of the
cyanobacterium Nostoc spongiaeforme: J Bacteriol, v. 156, p. 369-74.
Tillett, D., E. Dittmann, M. Erhard, H. von Döhren , T. Borner, and B. A. Neilan, 2000,
Structural organization of microcystin biosynthesis in Microcystis aeruginosa
PCC7806: an integrated peptide-polyketide synthetase system: Chem Biol, v. 7,
p. 753-64.
Tori. K., K. Toruka, K. Okabe, M. Ebata, and H. Otsuka, 1976, Carbon-13 NMR studies
of peptide antibiotics, thiostrepton and siomycin A: the structure relationship
Tetrahedron Lett, v. 17, p. 185-188.
Unligil, U., and J. Rini, 2000, Glycosyltransferase structure and mechanism: Curr.
Opin. Struct. Biol., v. 10, p. 510-7.
Utkilen, H., and N. Gjolme, 1995, Iron-stimulated toxin production in Microcystis
aeruginosa: Appl Environ Microbiol, v. 61, p. 797-800.
Vaillancourt, F. H., J. Yin, and C. T. Walsh, 2005, SyrB2 in syringomycin E
biosynthesis is a nonheme FeII alpha-ketoglutarate- and O2-dependent
halogenase: Proc Natl Acad Sci U S A, v. 102, p. 10111-6. van Wageningen, A. M., P. N. Kirkpatrick, D. H. Williams, B. R. Harris, J. K. Kershaw,
N. J. Lennard, M. Jones, S. J. Jones, and P. J. Solenberg, 1998, Sequencing and
analysis of genes involved in the biosynthesis of a vancomycin group antibiotic:
Chem Biol, v. 5, p. 155-62.
Velkov, T., and A. Lawen, 2003, Mapping and molecular modeling of S-adenosyl-L-
methionine binding sites in N-methyltransferase domains of the multifunctional
polypeptide cyclosporin synthetase: J Biol Chem, v. 278, p. 1137-48.
193 Vezie, C., J. Rapala, J. Vaitomaa, J. Seitsonen, and K. Sivonen, 2002, Effect of nitrogen
and phosphorus on growth of toxic and nontoxic Microcystis strains and on
intracellular microcystin concentrations: Microb Ecol, v. 43, p. 443-54.
Vidal-Cros, A., F. Viviani, G. Labesse, M. Boccara, and M. Gaudry, 1994,
Polyhydroxynaphthalene reductase involved in melanin biosynthesis in
Magnaporthe grisea. Purification, cDNA cloning and sequencing: Eur J
Biochem, v. 219, p. 985-92.
Voet, D., and J. Voet, 1995, Biochemistry: USA, John Wiley and Sons Inc.
Walker, J. E., M. Saraste, M. J. Runswick, and N. J. Gay, 1982, Distantly related
sequences in the alpha- and beta-subunits of ATP synthase, myosin, kinases and
other ATP-requiring enzymes and a common nucleotide binding fold: Embo J,
v. 1, p. 945-51.
Walsh, C. T., 1989, Enzymes in the D-alanine branch of bacterial cell wall
peptidoglycan assembly: J Biol Chem, v. 264, p. 2393-6.
Walsh, C. T., H. Chen, T. A. Keating, B. K. Hubbard, H. C. Losey, L. Luo, C. G.
Marshall, D. A. Miller, and H. M. Patel, 2001, Tailoring enzymes that modify
nonribosomal peptides during and after chain elongation on NRPS assembly
lines: Curr Opin Chem Biol, v. 5, p. 525-34.
Wang, R., D. Sweeney, S. E. Gandy, and S. S. Sisodia, 1996, The profile of soluble
amyloid beta protein in cultured cell media. Detection and quantification of
amyloid beta protein and variants by immunoprecipitation-mass spectrometry: J
Biol Chem, v. 271, p. 31894-902.
Weber, G., K. Schorgendorfer, E. Schneider-Scherzer, and E. Leitner, 1994, The
peptide synthetase catalyzing cyclosporine production in Tolypocladium niveum
194 is encoded by a giant 45.8-kilobase open reading frame: Curr Genet, v. 26, p.
120-5.
Weber, J. M., B. Schoner, and R. Losick, 1989, Identification of a gene required for the
terminal step in erythromycin A biosynthesis in Saccharopolyspora erythraea
(Streptomyces erythreus): Gene, v. 75, p. 235-41.
Weissman, K. J., and P. F. Leadlay. 2005. Combinatorial biosynthesis of reduced
polyketides. Nat Rev Microbiol 3:925-36.
Weitnauer, G., G. Hauser, C. Hofmann, U. Linder, R. Boll, K. Pelz, S. J. Glaser, and A.
Bechthold, 2004, Novel avilamycin derivatives with improved polarity
generated by targeted gene disruption: Chem Biol, v. 11, p. 1403-11.
Welker, M., J. Fastner, M. Erhard, and H. von Döhren , 2002, Applications of MALDI-
TOF MS analysis in cyanotoxin research: Environ Toxicol, v. 17, p. 367-74.
Westhuizen, A. v. d., and J. Eloff, 1985, Effect of temperature on the toxicity and
growth of the blue-green alga Microcystis aeruginosa (UV-006): Planta, v. 163,
p. 55-59.
Wiedner, C., P. M. Visser, J. Fastner, J. S. Metcalf, G. A. Codd, and L. R. Mur, 2003,
Effects of light on the microcystin content of Microcystis strain PCC 7806: Appl
Environ Microbiol, v. 69, p. 1475-81.
Wiegand, C., and S. Pflugmacher, 2005, Ecotoxicological effects of selected
cyanobacterial secondary metabolites: a short review: Toxicol Appl Pharmacol,
v. 203, p. 201-18.
Wolk, C. P., 1996, Heterocyst formation: Annu Rev Genet, v. 30, p. 59-78.
Woodcock, D. M., P. J. Crowther, J. Doherty, S. Jefferson, E. DeCruz, M. Noyer-
Weidner, S. S. Smith, M. Z. Michael, and M. W. Graham, 1989, Quantitative
195 evaluation of Escherichia coli host strains for tolerance to cytosine methylation
in plasmid and phage recombinants: Nucleic Acids Res, v. 17, p. 3469-78.
Wyckoff, E. E., A. M. Valle, S. L. Smith, and S. M. Payne, 1999, A multifunctional
ATP-binding cassette transporter system from Vibrio cholerae transports
vibriobactin and enterobactin: J Bacteriol, v. 181, p. 7588-96.
Yoshizawa, S., R. Matsushima, M. F. Watanabe, K. Harada, A. Ichihara, W. W.
Carmichael, and H. Fujiki, 1990, Inhibition of protein phosphatases by
microcystins and nodularin associated with hepatotoxicity: J Cancer Res Clin
Oncol, v. 116, p. 609-14.
Young, F. M., C. Thomson, J. S. Metcalf, J. M. Lucocq, and G. A. Codd, 2005,
Immunogold localisation of microcystins in cryosectioned cells of Microcystis: J
Struct Biol, v. 151, p. 208-14.
Yu, S., 1989, Drinking water and primary liver cancer.: Primary liver cancer: New
York, China academic publishers.
Yu, S., 1995, Primary prevention of hepatocellular carcinoma: J. Gastroenterol.
Hepatol., v. 10, p. 674-682.
Yu, S., N. Zhao, and X. Zi, 2001, [The relationship between cyanotoxin (microcystin,
MC) in pond-ditch water and primary liver cancer in China]: Zhonghua Zhong
Liu Za Zhi, v. 23, p. 96-9.
Zhao, G., and M. E. Winkler, 1996, A novel alpha-ketoglutarate reductase activity of
the serA-encoded 3-phosphoglycerate dehydrogenase of Escherichia coli K-12
and its possible implications for human 2-hydroxyglutaric aciduria: J Bacteriol,
v. 178, p. 232-9.
Zhu, W., J. E. Arceneaux, M. L. Beggs, B. R. Byers, K. D. Eisenach, and M. D.
Lundrigan, 1998, Exochelin genes in Mycobacterium smegmatis: identification
196 of an ABC transporter and two non-ribosomal peptide synthetase genes: Mol
Microbiol, v. 29, p. 629-39.
Zocher, R., T. Nihira, E. Paul, N. Madry, H. Peeters, H. Kleinkauf, and U. Keller, 1986,
Biosynthesis of cyclosporin A: partial purification and properties of a
multifunctional enzyme from tolypocladium inflatum: Biochemistry, v. 11;25, p.
550-553.
197 APPENDIX
A1. Media for culturing E. coli
A1.0 Luria Bertaini medium
10 g/L NaCl
10 g/L Tryptone
5 g/L Yeast extract
A1.1 Tryptone phosphate medium
8 g/L NaCl
20 g/L Tryptone
15 g/L Yeast extract
2 g/L Na2HPO4
1 g/L KH2PO4
2 g/L Glucose
A1.2 Superbroth
5 g/L NaCl
32 g/L Tryptone
20 g/L Yeast extract
2 g/L Na2HPO4
1 g/L KH2PO4
4 mL Glycerol
198 A1.3 M9 salts
60 g/L Na2HPO4
30 g/L KH2PO4
5 g/L NaCl
10 g/L NH4Cl
199 A2. Sequence alignments for the generation of phylogenetic trees
A2.0 McyH homologous sequence alignment
A2.1 McyH ATPase domain homologous sequence alignment
A2.2 McyI homologous sequence alignment
A2.3 McyJ homologous sequence alignment
200 Table 2.0. E. coli strains used in this study Source or Strain Genotype reference General Cloning Strains - + r DH5α F'/endA1 hsdR17(rK , mK ) supE44 thi-1 recA1 gyrA (Nal ) Woodcock et al, relA1 Δ(laclZYA-argF)U169 DeoR (F-Φ80lacZΔM15) (1989) TOP10F’ F'[lacIq Tn10(tetR)] mcrA Δ(mrr-hsdRMS-mcrBC) Invitrogen, ϕ80lacZΔM15 ΔlacX74 DeoR nupG recA1 araD139 Δ(ara- Carlsbad, CA leu)7697 galU galK rpsL(StrR) endA1 λ- XL-1Blue F'::Tn10 ProA+B+ laclq Δ(lacZ) M15/recA1 endA1 gyrA96 Stratagene, La (Nalr) thi hsdR17 (rK-mK+) supE44 relA1 lac Jolla, CA Expression Strains – – – r Rosetta(DE3) F ompT hsdSB(rB mB ) gal dcm (DE3) pRARE2 (Cam ) Novagen (Merck), Darmstadt, Germany – – – r Rosetta(DE3)PlysS F ompT hsdSB(rB mB ) gal dcm (DE3) pLysSpRARE2 (Cam ) Novagen (Merck), Darmstadt, Germany Auxotrophic Strains X536 HfrOR11 glu V42l- serA13 T3r Ho et al, (1999) NZN111 ldhA::Kanr pfl::Camr of W1485 Chatterjee et al, (2001) W1485 Parent of NZN111, wild type Chatterjee et al, (2001) DC1368 ldhA::Kanr pfl::Camr of LCB320 D. Clark, SI University LCB320 Parent of DC1368, F- thr-1 leu-6 thi-1 lacY1 supE44 tonA21 StrR Chatterjee et al, (2001) 1477 Mdh mut/ phosphoenolpyruvate synthetase mutant Hansen and Juni, (1979) 1120 Parent of 1477 phosphoenolpyruvate synthetase mutant Hansen and Juni, (1979)
201 Table 3.0. Distribution of mcyH orthologues among toxic and non-toxic cyanobacteria mcyHb Other mcy Strain Reference Microcystina genes mcyH(200)F/(1222)R mcyHscreenF/R detectedc Anabaena Herdman et al, - - - - cylindrica 1979 PCC7122 Anabaena sp. Rippka et al, 1979 - + + + PCC7108 M. aeruginosa Tillett et al, 2001 - - - - Bauld 5 M. aeruginosa Tillett et al, 2001 - - - - Bauld B M. aeruginosa Henning et al, - + + + HUB 5.3 1991 M. aeruginosa National Institute + + + + NIES298 for Environmental Studies (NIES) M. aeruginosa Pasteur Culture + + + + PCC7005 Collection (PCC) M. aeruginosa PCC + + + + PCC7804 M. aeruginosa Ripka and + + + + PCC7806 Herdman, 1992 M. aeruginosa Ripka and + + + + PCC7820 Herdman, 1992 M. aeruginosa Jackson et al, + + + + UWOCC MR-A 1984 M. aeruginosa Jackson et al, - + + + UWOCC MR-C 1984 M. aeruginosa Jackson et al, + + + + UWOCC MR-D 1984 M. aeruginosa Starr and Zeikus, - + + + UWOCCBS 1993 M. aeruginosa Starr and Zeikus, + + + + UWOCE7.GC 1993 M. viridis Neilan et al, 1997 + + + + NIES102 M. wesenbergii NIES + + + + NIES107 Microcystis sp. PCC + + + + PCC7840 Microcystis sp. University of + + + + UTEX2664 Texas (UTEX) Culture Collection Microcystis sp. UTEX + + + + UTEX2667 (PCC7941) Nodularia Moffitt and N + + + harveyana Neilan, 2001 PCC7804
202 Nodularia Moffitt and N + + + spumigena Neilan, 2001 NSOR10 Planktothrix Christiansen et al, + + + + agardhii 2003 CYA126 Synechocystis PCC - - - - PCC6803 aProduction of toxins detected by mouse bioassay, HPLC or ELISA. N indicates nodularin producing strains. bmcyJ orthologues detected by PCR using mcyH(200)F/(1222)R or mcyHscreenF/R. cOther mcy genes (mcyJ and mcyI) detected by PCR in this study
203 Table 3.1. Codon usage for mcyH (594 codons). UUU 55.6(33) Phe UCU 23.6(14) Ser UAU 32.0(19) Tyr UGU 1.7( 1) Cys UUC 5.1( 3) UCC 10.1( 6) UAC 5.1( 3) UGC 1.7( 1) UUA 65.7(39) Leu UCA 6.7( 4) UAA 0.0( 0) STOP UGA 1.7( 1) STOP UUG 13.5( 8) UCG 6.7( 4) UAG 0.0( 0) UGG 15.2( 9) Trp
CUU 18.5(11) Leu CCU 13.5( 8) Pro CAU 13.5( 8) His CGU 8.4( 5) Trp CUC 11.8( 7) CCC 5.1( 3) CAC 10.1( 6) CGC 8.4( 5) CUA 10.1( 6) CCA 6.7( 4) CAA 48.8(29) Gln CGA 15.2( 9) Arg CUG 13.5( 8) CCG 5.1( 3) CAG 13.5( 8) CGG 1.7( 1)
AUU 55.6(33) Ile ACU 16.8(10) Thr AAU 40.4(24) Asn AGU 23.6(14) Ser AUC 10.1( 6) ACC 18.5(11) AAC 5.1( 3) AGC 5.1( 3) AUA 11.8( 7) ACA 13.5( 8) AAA 33.7(20) Lys AGA 13.5( 8) Arg AUG 8.4( 5) Met ACG 1.7( 1) AAG 11.8( 7) AGG 3.4( 2)
GUU 37.0(22) Val GCU 28.6(17) Ala GAU 35.4(21) Asp GGU 11.8( 7) Gly GUC 8.4( 5) GCC 23.6(14) GAC 3.4( 2) GGC 13.5( 8) GUA 13.5( 8) GCA 6.7( 4) GAA 55.6(33) Glu GGA 20.2(12) GUG 5.1( 3) GCG 1.7( 1) GAG 13.5( 8) GGG 1.7( 1)
Fields include: nucleotide triplet, frequency of triplet (per thousand) and number of triplets. Rare codons are presented in bold type.
204 Table 3.2. Sequences used to create phylogenetic tree (Figure 3.5) Protein Putative Phylum Order Species abbreviation Functiona Lengthb Accessionc Acidobacteria Solibacterales Solibacter usitatus Ellin6076 569 ZP_00526302 fatty acid Chordata Galliformes Gallus gallus ALD 660 NP_001012615 transport fatty acid Primates Homo sapiens PMP70 659 NP_002849 transport fatty acid Rodentia Mus musculus PMP70 659 AAA39958 transport Cyanobacteria Chroococcales Crocosphaera watsonii WH8501 666 ZP_00517347
Crocosphaera watsonii WH8501 576 ZP_00515389
Crocosphaera watsonii WH8501 568 ZP_00515388
M. aeruginosa 586 BAB12214
Microcystis aeruginosa PCC7806 McyH 538 AAF00956
Synechococcus elongatus PCC7942 665 YP_399641
Synechococcus elongatus PCC6301 665 BAD79092
Synechococcus sp. CC9605 658 ABB36003
Synechococcus sp. CC9902 668 ABB25384 Synechococcus sp. JA-2-3B'a(2-13) 675 YP_476444
Synechococcus sp. JA-3-3Ab 664 YP_474416
Synechococcus sp. RS9917 651 ZP_01081152 Synechococcus sp. WH 5701 675 ZP_01084188
Synechococcus sp. WH 7805 659 ZP_01122731
Synechococcus sp. WH 8102 661 CAE06844 Synechocystis sp. PCC6803 661 BAA10424
Thermosynechococcus elongatus BP-1 616 BAC07776
Nostocales Anabaena sp. 90 592 AAO62579 Anabaena variabilis ATCC29413 669 ABA21240
Anabaena variabilis ATCC29413 611 ABA21245
Nodularia spumigena NdaI 601 AAO64410 Nostoc punctiforme PCC73102 675 ZP_00110694
Nostoc punctiforme PCC73102 663 ZP_00110904
Nostoc punctiforme PCC73102 672 ZP_00110909 aPrimary enzymatic function is given for experimentally characterised proteins only. bLength of amino acid sequence. cNCBI accession number. References can be found in the corresponding database files.
205 Table 3.2. Continued Protein Putative Phylum Order Species abbreviation Functiona Lengthb Accessionc Nostocales Nostoc sp. ATCC53789 NcpC 672 AAO23332 Nostoc sp. GSV224 NosG 663 AAF17285
Nostoc punctiforme PCC73102 672 ZP_00110909
Nostoc sp. ATCC53789 NcpC 672 AAO23332 Nostoc sp. GSV224 NosG 663 AAF17285
Nostoc sp. PCC7120 580 BAB74362
Nostoc sp. PCC7120 557 BAB74349 Oscillatoriales Planktothrix agardhii McyH 584 CAD29796
Trichodesmium erythraeum IMS101 534 ZP_00673060
Prochlorales Prochlorococcus marinus str. MIT 9211 662 ZP_01006232 Prochlorococcus marinus str. MIT 9312 660 ABB49128
Prochlorococcus marinus str. MIT 9313 667 CAE21770
Prochlorococcus marinus str. NATL2A 662 YP_292619 Prochlorococcus marinus subsp. marinus 662 AAP99125 str. CCMP1375 Prochlorococcus marinus subsp. pastoris 660 CAE18524 str. CCMP1986 Proteobacteria Burkholderiales Polaromonas naphthalenivorans CJ2 576 ZP_01021035
Ralstonia eutropha JMP134 570 YP_295855
Rhodoferax ferrireducens DSM15236 576 YP_525056 Maltose Enterobacteriales Escherichia coli MalK 371 P02914 import Methylococcales Methylococcus capsulatus str. Bath 571 YP_114118
Pasteurellales Haemophilus somnus 2336 489 ZP_00131877
Pseudomonadales Pseudomonas fluorescens Pf-5 575 YP_258017 Pseudomonas fluorescens PfO-1 575 ABA72565
Rhizobiales Xanthobacter autotrophicus Py2 606 ZP_01197238
Rhodocyclales Dechloromonas aromatica RCB 603 YP_283274 Streptophyta Brassicales Arabidopsis thaliana 706 NP_175837
Arabidopsis thaliana 514 AAD25615
Poales Oryza sativa (japonica cultivar-group) NdaI 771 BAD81416
Oryza sativa (japonica cultivar-group) 514 NP_913369
206 Table 3.3. Sequences used to create phylogenetic tree (Figure 3.6) Protein Putative Phylum Order Species abbreviation Functiona Lengthb Accessionc Drosophila Arthropoda Diptera 157 AAM50762 melanogaster Chordata Anura Xenopus laevis 157 AAH70657 Carnivora Canis familiaris hALDR 158 XP_534838 Canis familiaris PMP70 157 XP_537064 Canis familiaris ALDP 158 XP_855341 Canis familiaris hALDR 158 XP_864684 Canis familiaris hALDR 158 XP_864706 Cypriniformes Danio rerio 158 XP_694730 Danio rerio hALDR 157 XP_690956 Galliformes Gallus gallus ALDP 157 NP_001012615 Gallus gallus hALDR 158 XP_415938 Primates Homo sapiens 158 1908394A Homo sapiens 158 AAB27045 Homo sapiens 158 AAH25358 Homo sapiens ABCD3 157 AAH68509 Homo sapiens PXMP1 157 CAA58470 Homo sapiens ALDP 158 CAA79922 Homo sapiens 157 CAC15960 Homo sapiens aldrp 158 JC5712 Fatty acid Homo sapiens 158 NP_005155 transport Pan troglodytes ALDP 158 XP_521322 Rodentia Mus musculus PMP70 157 AAA39958 Mus musculus 158 AAH19187 Mus musculus 157 AAH54446 aPrimary enzymatic function is given for experimentally characterised proteins only. bLength of amino acid sequence. cNCBI accession number. References can be found in the corresponding database files.
207 Table 3.3. continued Protein Putative Phylum Order Species abbreviation Functiona Lengthb Accessionc Rodentia Mus musculus 157 BAB23450 Mus musculus 158 BAC38542 Mus musculus 157 BAE31880 Mus musculus 158 BAE36155 Mus musculus PMP68 157 CAA61748 Fatty acid Mus musculus mALDP 158 CAA83917 transport Mus musculus 157 NP_033017 Mus musculus 158 NP_036124 Mus musculus ALDP 158 XP_978393 Rattus Fatty acid 158 NP_203503 norvegicus transport Rattus 158 XP_343841 norvegicus Ruminantia Bos taurus ALDP 158 XP_593152 Bos taurus PMP70 157 XP_604418 Tetraodon Tetraodontiformes 158 CAG01176 nigroviridis Microcystis Cyanobacteria Chroococcales 155 BAB12214 aeruginosa Microcystis aeruginosa McyH 155 AAF00956 PCC7806 Nostocales Anabaena sp. 90 McyH 155 AAO62579 Anabaena variabilis 155 ABA21240 ATCC29413 Anabaena variabilis 155 ABA21245 ATCC29413 Nostoc punctiforme 155 ZP_00110694 PCC73102 Nostoc punctiforme 155 ZP_00110904 PCC73102 Nostoc punctiforme 155 ZP_00110909 PCC73102 Nostoc sp. NcpC 155 AAO23332 ATCC53789 Nostoc sp. NosG 155 AAF17285 GSV224 Caenorhabditis Nematoda Rhabditida 157 CAE57428 briggsae Caenorhabditis Pmp-4 157 NP_503105 elegans
208 aPrimary enzymatic function is given for experimentally characterised proteins only. bLength of amino acid sequence. cNCBI accession number. References can be found in the corresponding database files.
209 Table 4.0. Distribution of mcyI orthologues among toxic and non-toxic cyanobacteria.
mcyIb Other mcy Strain Reference Microcystina genes mcyI(322)F/(460)R mcyIdegenF/R detectedc Anabaena Herdman et al, - - - - cylindrica 1979 PCC7122 Anabaena flos- University of - - - - aquae Texas (UTEX) Culture Collection Anabaena Rippka et al, 1979 - + + + PCC7108 Anabaena sp. 90 Sivonen et al, 1992 + - + + M. aeruginosa Henning et al, - - + + HUB 5.3 1991 M. aeruginosa National Institute + + + + NIES298 for Environmental Studies (NIES) M. aeruginosa Pasteur Culture + + + + PCC7005 Collection (PCC) M. aeruginosa PCC + + + + PCC7804 M. aeruginosa Ripka and + + + + PCC7806 Herdman, 1992 M. aeruginosa Ripka and + - + + PCC7820 Herdman, 1992 M. aeruginosa Jackson et al, 1984 + + + + UWOCC MR-A M. aeruginosa Jackson et al, 1984 - + + + UWOCC MR-C M. aeruginosa Jackson et al, 1984 + + + + UWOCC MR-D M. aeruginosa Star and Zeikus, - + + + UWOC CBS 1993 M. aeruginosa Star and Zeikus, + + + + UWOCE7.GC 1993 M. viridis Neilan et al, 1997 + + + + NIES102 M. wesenbergii NIES + + + + NIES107 Microcystis PCC + - + + PCC7840 Microcystis sp. UTEX + + + + UTEX2664 Microcystisssp. UTEX + + + + UTEX2667 (PCC7941) Nodularia PCC N - - not tested harveyana PCC73104 Nodularia Moffitt and Neilan, N - - + spumigena 2001 NSOR10 210 spumigena 2001 NSOR10 Planktothrix Christiansen et al, + - - + agardhii CYA126 2003 Synechocystis PCC - - - - PCC6803 aProduction of toxins detected by mouse bioassay, HPLC or ELISA. N indicates nodularin producing strains. bmcyI orthologs detected by PCR using mcyI(322)F/(460)R or mcyIdegenF/R. cOther mcy genes (mcyH and mcyJ) detected by PCR in this study
211 Table 4.1. Codon usage for mcyI (338 codons) UUU 29.6(10) Phe UCU 17.8( 6) Ser UAU 20.7( 7) Tyr UGU 8.9( 3) Cys UUC 0.0( 0) UCC 8.9( 3) UAC 0.0( 0) UGC 3.0( 1) UUA 50.3(17) Leu UCA 17.8( 6) UAA 3.0( 1) STOP UGA 0.0( 0) STOP UUG 11.8( 4) UCG 5.9( 2) UAG 0.0( 0) UGG 3.0( 1) Trp
CUU 26.6( 9) Leu CCU 23.7( 8) Pro CAU 8.9( 3) His CGU 3.0( 1) CUC 11.8( 4) CCC 8.9( 3) CAC 8.9( 3) CGC 3.0( 1) CUA 11.8( 4) CCA 11.8( 4) CAA 23.7( 8) Gln CGA 3.0( 1) CUG 11.8( 4) CCG 14.8( 5) CAG 5.9( 2) CGG 8.9( 3)
AUU 44.4(15) Ile ACU 29.6(10) Thr AAU 44.4(15) Asn AGU 8.9( 3) Ser AUC 11.8( 4) ACC 14.8( 5) AAC 14.8( 5) AGC 3.0( 1) AUA 11.8( 4) ACA 20.7( 7) AAA 44.4(15) Lys AGA 5.9( 2) Arg AUG 32.5(11) Met ACG 3.0( 1) AAG 17.8( 6) AGG 3.0( 1)
GUU 29.6(10) Val GCU 29.6(10) Ala GAU 32.5(11) Asp GGU 11.8( 4) Gly GUC 8.9( 3) GCC 17.8( 6) GAC 3.0( 1) GGC 5.9( 2) GUA 23.7( 8) GCA 17.8( 6) GAA 47.3(16) Glu GGA 26.6( 9) GUG 11.8( 4) GCG 20.7( 7) GAG 23.7( 8) GGG 11.8( 4)
Fields include: nucleotide triplet, frequency of triplet (per thousand) and number of triplets. Rare codons are presented in bold.
212 Table 4.2. Sequences used to create phylogenetic tree (Figure 4.6) Protein Organism abbreviation Primary functiona Lengthb Accessionc Anabaena sp. McyI 337 AA062580 Arabidopsis thaliana 602 NP_195146 D-3-phosphoglycerate Arabidopsis thaliana PGDH 624 T52296 dehydrogenase D-3-phosphoglycerate Arabidopsis thaliana PGDH 603 AAM60833 dehydrogenase Archaeoglobus fulgidus 527 NP_069647 Crocosphaera watsonii 525 ZP_00179809 Fermentative D-lactate Escherichia coli LDH (LdhA) 329 AAB51772 dehydrogenase D-3-phosphoglycerate Escherichia coli PGDH (SerA) 410 P08328 dehydrogenase D-lactate dehydrogenase Escherichia coli DLD 571 AAC75194 (NADH independent) F. trinervia MDH D-malate dehydrogenase 385 AAA87008 Methanocaldococcus 524 NP_248012 jannaschii Methanopyrus kandleri 522 NP_613584 Methanothermobacter 525 NP_276105 thermoautotrophicus Microcystis aeruginosa McyI 337 AAF00955 Microcystis aeruginosa McyI 337 AB032549 Nodularia spumigena NdaH 341 AA064409 Nostoc punctiforme 526 ZP_00112058 Nostoc sp. 526 NP_485930 Oceanobacillus 319 NP_693766 iheyensis NAD-dependent formate Pseudomonas sp. FDH 401 P33160 dehydrogenase (FDH) Pyrococcus abyssi 307 NP_126444 Pyrococcus furiosus 306 NP_579123 Rubrobacter 527 ZP_00199880 xylanophilus Synechococcus 529 ZP_00164567 elongatus Synechocystis sp. 554 NP_441198 Synechocystis sp. 528 NP_896628 Thermoanaerobacter 533 NP_624129 tengcongensis Thermosynechococcus 527 NP_681115 elongatus Trichodesmium 527 ZP_00327083 erythraeum aPrimary enzymatic function is given for experimentally characterised proteins only. bLength of amino acid sequence. cNCBI accession. References can be found in the corresponding database files.
213 Table 4.3. Steady-state kinetic and inhibition parameters of reactions catalysed by McyI a app app b Reaction Km Vmax Inhibition
µM µmol/min/mg protein I50 mM
OAA 97.3±1.5 15.2±0.2 3-MeMal: 0.7±0.01, D-Mal: 2.2±0.03 NADPH 35.8±1.8 6.5±0.3 Not tested
αKG 2886.6±116.2 3.6±0.1 3-MeMal: 2.3±0.06, D-HGA: 20.5±0.2 aKinetic analysis of the αKG and OAA reductase activities of McyI was carried out in 1 mL reaction mixtures each containing 0.25 mM NADPH, 5 µg purified enzyme and various amounts of substrate in reaction buffer (pH 7). Kinetic analysis of the NADPH oxidase activity of McyI was performed in 1 mL reactions each containing 5 µg purified enzyme, 0.8 mM OAA and various amounts of NADPH in reaction buffer (pH 7). Initial velocities were measured by monitoring the disappearance of NADPH at 340 nm at 37˚C for 0.5 min. No deviations in linearity were observed on plots of kinetic data. The data presented in the table are from a single experiment, however, each experiment was reproduced at least four times with <10% deviation in results between trials. bAssays were performed in 1 mL of reaction buffer (pH 7), containing 0-8 mM inhibitor (L-serine, 3- MeMal, L-Mal, D-Mal, L-Asp, D-Asp, D-HGA or D-3PGA), 0.25 mM NADPH, 5 µg enzyme, and 0.2 mM OAA or 1 mM αKG. I50 is the concentration of inhibitor required for 50% inhibition in standard enzyme assays with saturating substrate. The data presented as the mean of 2-3 separate experiments performed consecutively.
214 Table 5.0. Distribution of mcyJ orthologues among toxic and non-toxic cyanobacteria
mcyJb Other mcy genes a c Strain Reference Microcystin mcyJ(345)F/(525)R mcyJscreenF/R detected Anabaena cylindrica Herdman et al, 1979 - - - - PCC7122 Anabaena PCC7108 Ripka et al, 1979 - + + +
M. aeruginosa Bauld 5 Tillett et al, 2001 - - - -
M. aeruginosa Bauld B Tillett et al, 2001 - - - -
M. aeruginosa NIES298 National Institute for + + + + Environmental Studies (NIES) M. aeruginosa PCC7005 Pasteur Culture Collection + + + + (PCC) M. aeruginosa PCC7804 PCC + + + +
M. aeruginosa PCC7806 Ripka and Herdman, 1992 + + + +
M. aeruginosa UWOCC Jackson et al, 1984 + + + + MR-A M. aeruginosa UWOCC Jackson et al, 1984 - + + + MR-C M. aeruginosa UWOCC Jackson et al, 1984 + + + + MR-D M. aeruginosa Star and Zeikus, 1993 - + + + UWOCCBS M. aeruginosa Star and Zeikus, 1993 + + + + UWOCE7.GC M. viridis NIES102 Neilan et al, 1997 + + + + M. wesenbergii NIES107 NIES + + + +
Microcystis PCC7840 PCC + + + +
Microcystis sp. University of Texas (UTEX) + + + + UTEX2664 Culture Collection Microcystis sp. UTEX + + + + UTEX2667 (PCC7941) Nodularia harveyana Moffitt and Neilan, 2001 N - + + PCC7804 Nodularia spumigena Sivonen et al, 1989 N - + + HEM Nodularia spumigena Moffitt and Neilan, 2001 N - + + NSOR10 Planktothrix agardhii Christiansen et al, 2003 + + + + CYA126 Synechocystis PCC6803 PCC - - - - aProduction of toxins detected by mouse bioassay, HPLC or ELISA. N indicates nodularin producing strains. bmcyJ orthologues detected by PCR using the specific mcyJ(345)F/(525)R or degenerate mcyJscreenF/R primers. cOther mcy genes (mcyH and mcyI) detected by PCR in this study.
215 Table 5.1. Codon usage for mcyJ (279 codons). UUU 60.9(17) Phe UCU 10.8( 3) Ser UAU 25.1( 7) Tyr UGU 14.3( 4) Cys UUC 7.2( 2) UCC 0.0( 0) UAC 3.6( 1) UGC 0.0( 0) UUA 35.8(10) Leu UCA 7.2( 2) UAA 0.0( 0) STOP UGA 3.6( 1) STOP UUG 21.5( 6) UCG 10.8( 3) UAG 0.0( 0) UGG 25.1( 7) Trp
CUU 21.5( 6) Leu CCU 21.5( 6) Pro CAU 21.5( 6) His CGU 7.2( 2) Trp CUC 14.3( 4) CCC 14.3( 4) CAC 3.6( 1) CGC 3.6( 1) CUA 7.2( 2) CCA 3.6( 1) CAA 35.8(10) Gln CGA 10.8( 3) CUG 3.6( 1) CCG 3.6( 1) CAG 21.5( 6) CGG 7.2( 2)
AUU 32.3( 9) Ile ACU 17.9( 5) Thr AAU 35.8(10) Asn AGU 10.8( 3) Ser AUC 10.8( 3) ACC 3.6( 1) AAC 17.9( 5) AGC 7.2( 2) AUA 7.2( 2) ACA 7.2( 2) AAA 39.4(11) Lys AGA 14.3( 4) Arg AUG 10.8( 3) Met ACG 0.0( 0) AAG 17.9( 5) AGG 7.2( 2)
GUU 28.7( 8) Val GCU 43.0(12) Ala GAU 50.2(14) Asp GGU 25.1( 7) Gly GUC 3.6( 1) GCC 17.9( 5) GAC 10.8( 3) GGC 3.6( 1) GUA 21.5( 6) GCA 10.8( 3) GAA 57.3(16) Glu GGA 21.5( 6) GUG 7.2( 2) GCG 17.9( 5) GAG 10.8( 3) GGG 3.6( 1)
Fields include: nucleotide triplet, frequency of triplet (per thousand) and number of triplets. Rare codons are presented in bold type.
216 Table 5.2. Sequences used to create phylogenetic tree (Figure 5.3 )
Protein Phylum Order Species abbreviation Putative Functiona Lengthb Accessionc Actinobacteria Actinomycetales Aeromicrobium erythreum EryG Erythromycin O-MT 304 AAU93801 Amycolatopsis mediterranei Orf14 Rifamycin B O-MT 372 AAC01738 Frankia sp. EAN1pec 287 ZP_00573171 Lechevalieria aerocolonigenes RebM Rebeccamycin O-MT 283 BAC10678 Saccharopolyspora erythraeum EryG Erythromycin O-MT 306 CAA42929 Butenyl-spinosyn C- Saccharopolyspora pogona BusF MT 275 AAY88923 Saccharopolyspora spinosa SpnF Spinosyn MT 275 AAG23267 Streptomyces avermitilis MA- 4680 AveD Avermectin O-MT 283 NP_822112 Streptomyces cinnamonensis MonE Monensin O-MT 276 AAO65792 Streptomyces griseochromogenes Milbemycin O-MT 281 AAR15334 Mitomycin aziridine Streptomyces lavendulae MitM NMT 283 AAD28459 Streptomyces nogalater SnogM Nogalamycin O-MT 278 AAG42853 Streptomyces roche Polyketide O-MT 270 NP_851450 Streptomyces sp. TP-A0274 StaMB Staurosporine MT 280 BAC55218 Streptomyces violaceusniger NigE Nigericin O-MT 270 ABC84455
Ascomycota Pneumocystidales Pneumocystis carinii Erg6 377 AAK54439 Pneumocystis carinii f. sp. carinii Sterol MT 359 AAK00294 Saccharomycetales Candida glabrata Erg6 Sterol MT 372 AAX73199
Bacteroidetes Sphingobacteriales Cytophaga hutchinsonii 287 ZP_00309576
Chlorobi Chlorobiales Chlorobium limicola 332 ZP_00513091 Chlorobium phaeobacteroides BS1 292 ZP_00531530 Chlorobium phaeobacteroides DSM266 296 ZP_00528886 Chlorobium tepidum 292 AAM71629 Pelodictyon phaeoclathratiforme 269 ZP_00589575 Prosthecochloris aestuarii 292 ZP_00591555 aMT= methyltransferase bLength of amino acid sequence. cNCBI accession number. References can be found in the corresponding database files.
217 Table 5.2. Continued
Protein Phylum Order Species abbreviation Putative Functiona Lengthb Accessionc Cyanobacteria Nostocales Anabaena sp. 90 McyJ Microcystin O-MT 310 AAO62583 Anabaenapeptolide Anabaena circinalis 90 AdpE O-MT 263 CAC01607 Anabaena variabilis ATCC29413 280 ABA24403 Nodularia spumigena NdaE Nodularin O-MT 309 AAO64406 Nostoc sp. PCC7120 DmtA DNA MT 210 AAF75227 Crocosphaera watsonii Chroococcales WH8501 284 ZP_00518612 Microcystis aeruginosa McyJ Microcystin O-MT 184 BAD32824 Microcystis aeruginosa PCC7806 McyJ Microcystin O-MT 278 AAF00954 Microcystis novacekii McyJ Microcystin O-MT 184 BAD32826 Microcystis novacekii McyJ Microcystin O-MT 184 BAD32825 Microcystis sp. MCS3 McyJ Microcystin O-MT 184 BAD32823 Microcystis viridis McyJ Microcystin O-MT 184 BAD32822 Synechococcus sp. JA-2- 3B'a(2-13) 285 ABD03340 Synechococcus sp. JA-3-3Ab 282 ABD00391 Oscillatoriales Planktothrix agardhii McyJ Microcystin O-MT 276 CAD29800
Mycetozoa Dictyosteliida Dictyostelium discoideum 189 XP_643680
Proteobacteria Chromatiales Halorhodospira halophila 300 ZP_01152570 Nitrosococcus oceani ATCC19707 282 ABA57503 Oceanospirillales Oceanospirillum sp. MED92 276 ZP_01165185 Methylococcales Methylococcus capsulatus 305 AAU90751 Polyangium cellulosum SpiK Spirangiene O-MT 263 CAD43452 Myxococcales Polyangium cellulosum ChiG Chivosazol O-MT 312 AAY89054
rRNA adenine Sordariomycetes Hypocreales Gibberella zeae PH-1 dimethylase 380 XP_385916
Spirochaetes Spirochaetales Leptospira interrogans 277 AAN47802
symbiont bacterium of Unknown Theonella swinhoei OnnG Polyketide O-MT 321 AAV97875 aMT= methyltransferase bLength of amino acid sequence. cNCBI accession number. References can be found in the corresponding database files.
218 219 220 EXPORT
Peroxisomal proteins ABC- A1 Elongation factor 3-C UVRA CFTR-C CFTR-N Proteins ABC- Bacteriocins A MDR, TAP, RTX- Modified cyclictoxins Polysaccharides Lantibioticspeptides ABC- Antibiotic resistance class II A2 Elongation factor 3-N Antibiotic resistance class I Haeme Eye-pigment ABC- Iron precursors Ions - B1 siderophores & MolybdenumOligosaccharides metals Polar amino acids FtsE, Abc Glycine-Betane Nitrate Oligopeptides
Phosphonates Nodulation ABC-B Antibiotic resistance class III Branched chain amino acids LivF ABC- Branched chain amino acids, LivG B2 Yhbg Monosaccharides-C Monosaccharides-N
IMPORT
Figure 1.1. Phylogeny of the ABC transporter super-family. Figure adapted from Saurin et al, 1999. The major subdivisions of the tree are ABC-A, import systems, and ABC-B, export systems. A detailed description of the proteins in each sub-group can be found in Saurin et al, 1999.
221 222 223 224 Figure S1.3. Model for the formation of Adda during microcystin biosynthesis and predicted domain structure of McyG, McyD and McyE. Each circle and rectangle represents respectively a PKS or NRPS enzymatic domain. The putative aminotransferase domain is represented by a diamond. The activity of the tailoring ORFs, McyJ and F, are shown as inverted triangles. KS, β- ketoacyl synthase; AT, acyltransferase; ACP, acyl carrier protein; KR, ketoacyl reductase; DH, dehydratase; CM, C-methyltransferase; OM, O-methyltransferase; A, aminoacyl adenylation; C, condensation; AMT, aminotransferase; RC, racemase. The NRPS thiolation motif is shown in black. Reaction order shows transfer and condensation of Adda to D-glutamate (adapted from Tillett et al, 2000).
225 Figure S1.4. Biosynthetic model for microcystin-LR and predicted domain structure of McyE, McyA, McyB, and McyC. Each circle and rectangle represents, respectively, a PKS or NRPS enzymatic domain. The aminotransferase domain is represented by a diamond. The activity of the putative tailoring ORF, McyI, is shown as an inverted triangle. KS, β-ketoacyl synthase; AT, acyltransferase; CM, C-methyltransferase; ACP, acyl carrier protein; A, aminoacyl adenylation; C, condensation; NM, N-methyltransferase; Ep, epimerization; TE, thioesterase; AMT, aminotransferase. The NRPS thiolation motif is shown in black. Aminoacyl activation and condensation order is predicted: L-Z-Adda, L-glutamate, L- methylserine, D-alanine, L-leucine, D-methyl-aspartate, L-arginine, cyclization (adapted from Tillett et al, 2000).
226 NdaE NdaG OM NdaD NdaF Rac NdaC CM CM KR DH KR DH KR CM A PCP KS AT ACP KS AT ACP KS AT ACP KS AT ACP AMT C A PCP C
S S S S S S
O O O O O O
Me Me Me Me + Phenyl-acetate O NH2
+ATP Me Me N Me HO O O
Me O Me Me Me NH O 2 Me Me +Malonyl-CoA O +NADPH+H+ Me +2x SAM +Malonyl-CoA +NADPH+H+ +2x SAM +Malonyl-CoA Me Me + O -H2O +NADPH+H -H2O +Malonyl-CoA +NH2 +2x SAM +L-Glu +ATP
Figure S1.5 Model of the Adda biosynthetic pathway during nodularin biosynthesis catalysed by proteins encoded by ndaC-F. The green circles represent the NRPS adenylation (A), peptidyl carrier protein (PCP) and condensation (C) domains. The blue circles represent the PKS, ketosynthase (KS), acyltransferase (AT), C-methyltransferase (CM), dehydratase (DH), ketoreductase (KR) and acyl carrier protein (ACP) domains. Condensation reactions are indicated in red. The O-methyltransferase (OM) NdaE (pink) catalyses the transfer of the methyl group to the C9 hydroxyl indicated in blue. The racemase NdaG (pink) catalyses the epimerisation of the Glu residue. The aminotransferase (AMT) domain is shown in yellow and catalyses the transfer of the amine group to Adda as indicated in blue (adapted from Moffit and Neilan, 2004).
227 NdaH NdaF NdaA NdaB CM KS AT ACP AMT C A PCP C A NM PCP C A PCP C A PCP TE
S S S S S
O O NH2 O Me O O
Me Me OH NH N HO NH2 N Me N
N N O O HO O
O Me O Me OH
Me O Me
NH2 N N Me N O HO O Me O Me Me O
O O Me N Me Me NH2
N O HO
Me Me O
O Me O Me
NH N 2 HO
Me Me O
O O Me +D-Glu Me NH +ATP 2
Me Me
+L-Thr O Me +ATP +SAM Me Me
O +D-MeAsp +ATP
+L-Arg +ATP
D-Glu O COOH MeDhb
Me Me OMe N N Me H O O NH Me
H Nodularin Me Me N NH O Adda O COOH D-MeAsp L-Arg
NH
HN NH2
Figure S1.6 Model of the peptide biosynthetic pathway during nodularin biosynthesis catalysed by proteins encoded by ndaFHAB. The green circles represent the NRPS adenylation (A) N- methyltransferase (NM), peptidyl carrier protein (PCP) and condensation (C) domains. The blue circles represent the PKS, ketosynthase (KS), acyltransferase (AT), C-methyltransferase (CM) and acyl carrier protein (ACP) domains. Condensation reactions are indicated in red. The racemase NdaG catalyses the epimerisation of the Glu residue. The putative dehydrogenase, NdaH, catalyses the formation of the MeDhb residue following peptide bond formation as indicated in blue. The thioesterase (TE) domain catalyses the cyclisation and release of the peptide (Adapted from Moffit and Neilan, 2004).
228 229 230 231 McyH peptide sequence (A)
McyH peptide sequence (B)
Figure 3.3. Topological predictions for McyH. (A) Kyte-Doolittle hydropathy profile for McyH. Hydrophobic residues have a value greater than zero, hydrophylic residues have a value of less than zero. (B) DAS Transmembrane segment prediction for McyH. Region above the broken, and unbroken lines indicate putative membrane-spanning regions with moderate, and high confidence values, respectively.
232 233 0.05
1000 H. somnus [ZP_00131877] X. autotrophicus [ZP_01197238] 1000 P. fluorescens [YP_258017] 223 1000 P. fluorescens [ABA72565] M. capsulatus [YP_114118] A 1000 A.thaliana [NP_175837] 689 A.thaliana [AAD25615] 1000 O. sativa [BAD81416] 1000 O. sativa [NP_913369] 944 T. erythraeum [ZP_00673060] 347 658 C.watsonii [ZP_00515388] Nostoc sp. [BAB74362] 1000 684 C.watsonii [ZP_00515389] 819 Nostoc sp. [BAB74349] B S.usitatus [ZP_00526302] 998 R. eutropha [YP_295855] 934 D. aromatica [YP_283274] 831 P. naphthalenivorans [ZP_01021035] 1000 R. ferrireducens [YP_525056] 426 T. elongatus [BAC07776] 658 1000 S. elongatus [YP_399641] S. elongatus [BAD79092] 919 1000 Synechococcus sp. [YP_474416] C Synechococcus sp. [YP_476444] 749 C. watsonii [ZP_00517347] 1000 682 Synechocystis sp. [BAA10424] 1000 Synechococcus sp. [ABB25384] Synechococcus sp. [ZP_01084188] 983 Synechococcus sp. [ZP_01122731] 1000 998 P. marinus [CAE21770] 1000 Synechococcus sp. [ZP_01081152] 992 Synechococcus sp. [CAE06844] D 769 P. marinus [AAP99125] 998 1000 P. marinus [ABB49128] 1000 P. marinus [CAE18524] Synechococcus sp. [ABB36003] 1000 P. marinus [YP_292619] 1000 P. marinus [ZP_01006232] 994 P. agardhii [CAD29796]* M. aeruginosa [BAB12214]* 814 1000 M. aeruginosa [AAF00956]* N. spumigena [AAO64410]* E 933 983 Anabaena sp. [AAO62579]* N. punctiforme [ZP_00110694]* 552 Nostoc sp. [AAO23332]* 744 A. variabilis [ABA21240] 1000 827 N. punctiforme [ZP_00110904]* 1000 Nostoc sp. [AAF17285]* A. variabilis [ABA21245] 1000 N. punctiforme [ZP_00110909]* G. gallus [NP_001012615] F 1000 M. musculus [AAA39958] 1000 H. sapiens [NP_002849] E. coli [P02914]
Figure 3.5. Phylogenetic tree encorporating McyH and similar ABC transporters. The phylogenetic tree encompasses putative ABC transporters from cyanobacteria (bold) and other organisms. Asterisks (*) indicate association with a secondary metabolite biosynthesis gene cluster. The maltose importer, MalK, from E. coli [P02914], constitutes the artificial out-group. Reference sequences include PMPs from G. gallus, M. musculus, and H. sapiens. For a detailed list of sequences used, see Table 3.2. Bootstrap values derived from 1 000 replicates are given. The scale at the top of the tree corresponds to 5% divergence between sequences.
234 0.05
1000 M. aeruginosa [BAB12214] M. aeruginosa [AAF00956] 716 N. punctiforme [ZP_00110909] A.variabilis [ABA21245] 886 Anabaena sp. [AAO62579] 297 Nostoc sp. [AAO23332] 783 A. variabilis [ABA21240] 620 N. punctiforme [ZP_00110694] 517 N. punctiforme [ZP_00110904] 968 Nostoc sp. [AAF17285] M. musculus [BAE36155] [NP_036124] 989 [BAC38542] 523 [AAH19187] R.norvegicus [NP_203503] 953 999 C. familiaris [XP_534838] [XP_864706] 935 [XP_864684] 493 H. sapiens [JC5712] 1000 940 H. sapiens [NP_005155] G. gallus [XP_415938] 496 1000 R. norvegicus [XP_343841] M. musculus [CAA83917] 532 [XP_978393] B. taurus [XP_593152] 763 C. familiaris [XP_855341] 1000 397 P. troglodytes [XP_521322] 996 H. sapiens [AAH25358] 967 [CAA79922] 998 [1908394A] [AAB27045] D. rerio [XP_690956] 982 1000 T. nigroviridis [CAG01176] D. rerio [XP_694730] C. briggsae [CAE57428] 1000 C. elegans [NP_503105] D. melanogaster [AAM50762] G. gallus [NP_001012615] 998 X. laevis [AAH70657] 1000 B. taurus [XP_604418] 563 C. familiaris [XP_537064] 607 641 H. sapiens [CAA58470] [CAC15960] 575 [AAH68509] M. musculus [AAA39958] M. musculus [AAH54446] 835 852 M. musculus [CAA61748] 762 [BAB23450] [NP_033017] [BAE31880]
Figure 3.6. Phylogenetic tree encorporating the ATPase domains of McyH and similar ABC transporters. The phylogenetic tree encompasses ATPase domains from putative ABC transporters from cyanobacteria (bold) and other organisms. For a detailed list of sequences used, see Table 3.3. Bootstrap values derived from 1 000 replicates are given. The scale at the top of the tree corresponds to 5% divergence between sequences.
235 236 237 Induced Un-induced
1(H) 2(C) 3(H) 4(C) 5(+) -62 McyHL -48
McyH L -62 -48
Figure 3.9. Coomassie stained polyacrylamide gel of vesicle protein extracts and corresponding western blot using nickel conjugate. Lanes 1-5, McyHL induced, negative control (pET30) induced, McyHL un-induced, negative control un-induced, positive control (McyI). Numbers on right refer to protein molecular weight (KDa).
238 239 240 241 242 243 244 McyI peptide sequence (A)
McyI peptide sequence (B)
Figure 4.4. Topological predictions for McyI. (A) Kyte-Doolittle hydropathy profile for McyI. Hydrophobic residues have a value greater than zero, hydrophylic residues have a value of less than zero. (B) DAS Transmembrane segment prediction for McyI. Region above the broken and unbroken lines indicate putative membrane-spanning regions with moderate and high confidence values, respectively. Possible membrane spanning domains are between residues 123-125 and 163-164.
245 Anabaena sp. MTIIYPPKNF PSKTKNHKVL LIGKMYDEIG EKLLAE.YTN VEIIK.EPKQ HQIHEAIQDV N.spumigena MTAIFP.QFS TVNLKNQKVL LIGRMYDETG EKLLEE.YTN IQILK.DPTK NEINQAIQEA M. aeruginosa MTTTSP.KTL TMDIQKPRVL LIGKMYDEAG QKLLEE.STN IEILS.NPTL EEINEAIQEV E. coli (PGDH) M...... AKV SLEKDKIKFL LVEGVHQKAL ESLRAAGYTN IEFHKGALDD EQLKESIRDA
Lys-39
Anabaena sp. SGVFVRYPTK LDAQAIGLAK NLKVISTSGF GTDAIDIAAA TKRGIVVVNN PGLSTTAVTE N.spumigena SGVFVRYPTK LDAQAIGLAK KLKVISTSGF GTDAIDISVA TKHGVVVVNN PGLSTTAVAE M. aeruginosa SGVFVRYPNK LEASSIRLAK QLKVISTSGF GTDSIDIMAA TEQGVIVVNN PGMSTTSVAE E. coli (PGDH) HFIGLRSRTH LTEDVINAAE KLVAIGCFCI GTNQVDLDAA AKRGIPVFNA PFSNTRSVAE
Arg-60/62
Anabaena sp. HTLSMILALA KKLTFLNQCV KAGNYLIRNQ VQPIQLEGKT LGIVGLGRIG SAVAKICSTA N.spumigena HTICMILALA KKLTFLNQCV KTGNYLIRNQ VQPMQLEGKT LGIVGLGRIG SAVASKCSAA M. aeruginosa HTLSMILALA KKLPFLDHCV KKGNYLIRNQ MQPIQLENKT LGIVGLGRIG TLVAHKCNLA E. coli (PGDH) LVIGELLLLL RGVPEANAKA HRGVWN.KLA AGSFEARGKK LGIIGYGHIG TQLG.ILAES
Trp-139 Lys-141
Anabaena sp. LQMRVLAYDP YVPSGKADTV RATLVQDLDY LLTESDFVSL HPELTDETCE MFDLEAFKKM N.spumigena FQMRVLAYDP YVLPSQAEAV GGTLVENLDY LLAESDFVSL HPELTDETYE MFALEAFKKM M. aeruginosa LGMRVLAYDP YVPASKAETV GATWVKDLDV LLAESDFVSL HTELTSETRE MFNISVFKKM E. coli (PGDH) LGMYVYFYD. ..IENKLPLG NATQVQHLSD LLNMSDVVSL HVPENPSTKN MMGAKEISLM
Anabaena sp. KPSAFLINTS RGKVVRQPDL VTAIREKLIA GAAIDVFEPE PPAINN.... PLYEFDNVIF N.spumigena KPTAFLINTS RGKIVCEQDL VVAIGEKWIS GAAIDVFEPE PPCQDN.... PLYDFDNVIL M. aeruginosa KPTAFLINTS RGKVVCEKDL GIALNQKLIA GCALDVFEPE PPALDN.... PLYNFENVIL E. coli (PGDH) KPGSLLINAS RGTVVDIPAL CDALASKHLA GAAIDVFPTE PATNSDPFTS PLCEFDNVLL
Arg-240 Glu-269
Anabaena sp. SPHLAGVTPE AGMAAALSAA NQILQVLQGE KPPYIIN.PK V...... N.spumigena SPHLAGVTPE AAIAATLSAA NQILQVLQGQ KPPYIVN.PE V...... M. aeruginosa SPHLAGVTPE ASLAAAVSAA NQILQVLQGE KPPYMIN.PE V...... E. coli (PGDH) TPHIGGSTQE AQENIGLEVA GKLIKYSDNG STLSAVNFPE VSLPLHGGRR LMHIHENRPG
His-292 Gly-294/295 Gly-336/337 Arg-338/339
Anabaena sp. .WNS...... N.spumigena .LDNFPHLQ...... M. aeruginosa .WNLF...... E. coli (PGDH) VLTALNKIFA EQGVNIAAQY LQTSAQMGYV VIDIEADEDV AEKALQAMKA IPGTIRARLLY
Arg-405/407 Tyr-410
Figure 4.5. Alignment of McyI (M. aeruginosa [AAF00955], Anabaena sp. [AA062580]), NdaH (N. spumigena [AA064409]) and PGDH (E. coli [P08328]). Structurally important amino acid residues and the NAD-binding signature are indicated (see text for details). Consensus residues are highlighted. Coloured regions define the three domains of PGDH: the nucleotide binding domain (residues 108-294, blue), the substrate binding domain (7-107 and 295-336, red) and the regulatory domain (338-409, yellow).
246 0.05 F. trinervia (MDH) [AAA87008]
992 N. spumigena (NdaH) [AA064409] Anabaena sp. (McyI) [AA062580] 1000 A M.aeruginosa (McyI) [AB032549]
1000 M. aeruginosa (McyI) [AAF00955]
R. xylanthophilus [ZP_00199880]*
A. thaiana (PGDH) [T52296]* 428 1000 A. thaiana [NP_195146]* 692 1000 A. thaiana [AAM60833]* 514 Synechocystis sp.[NP_896628]* B T. elongatus [NP_681115] 1000 T. erythraeum [ZP_00327083] 1000 S. elongatus [NP_681115] 548 522 932 Nostoc sp. [NP_485930]
1000 N. punctiforme [ZP_00112058] 959 C. watsonii [ZP_00179809] 971 955 Synechocystis sp. [NP_441198]
O. iheyensis [NP_693766]
854 A. fulgidus [NP_069647]* 309 T. tengcongensis [NP_624129]*
M. jannaschii [NP_248012]* 474 602 C 953 M. thermoautotrophicus [NP_276105]*
867 M. kandleri [NP_613584]
798 P. abyssi [NP_126444]
1000 P. furiosis [NP_579123] 1000 E. coli (PGDH) [P08328]*
E. coli (LDH) [AAB51772]
Pseudomonas sp. (FDH) [P33160]
E. coli (DLD) [AAC75194]
Figure 4.6. Phylogenetic tree encorporating McyI and similar dehydrogenases. The phylogenetic tree encompasses phosphoglycerate dehydrogenase (PGDH) homologues from cyanobacteria (bold) and other organisms. An E. coli D-lactate dehydrogenase (DLD) constitutes the artificial out-group. Asterisks (*) indicate sequences with ACT domains. For a detailed list of sequences used, see Table 4.2. Bootstrap values derived from 1 000 replicates are given. The scale at the top of the tree corresponds to 5% divergence between sequences.
247 1 2 3 4 5 6 7 8 9 10 66 56 McyI 43 34 27
20
Figure 4.7. Purification of McyI. Elution of polyhistidine-tagged McyI from Ni2+ charged column. Lane 1, Broad-range molecular weight standard (molecular weights, KDa, given). Lanes 2-10, McyI elution fractions 2-10.
248 Figure 4.8 Determination of McyI subunit organisation (A) Elution standard curve for protein molecular weight standards. (B) Chromatogram of purified recombinant McyI. The peak at 27.8 min corresponds to the McyI dimer (85 KDa including His-tag). All samples were run on an AKTA basic 900 series FPLC apparatus fitted with a Superdex 200 10/300 GL column (Amersham Biosciences).
249 250 251 252 253 Legend a) Oxaloacetate methylene resonance, dihydrate form b) Pyruvate, keto form x) NADPH resonances y) NADP+ resonances Malate resonances
Figure S4.10C NMR spectra (1H, 600MHz) of the enzymatic reduction of oxaloacetate to 2 malate are shown in the figure. Spectra were run in 5% H2O, with water saturation so the water region (4.5-5.2 ppm) is not shown. Also the enzyme preparation used contained glycerol, so the region 2.8-3.7 ppm that contains the intense glycerol signals is also omitted. The resolution in this experiment which involves a viscous enzyme solution, is about 3Hz, and the data is shown without any resolution enhancement. The methylene resonance of the keto form of oxaloacetate (δ 3.5ppm, ~90% at this pH) is unfortunately obscured by the glycerol resonances, but the decrease in the dihydrate form present (2.5 ppm, labelled a, ~5% at this pH) is apparent as is the formation of NADP+ as malate is formed. The non-anomeric ribose resonances of NADPH and NADP+ at 3.9-4.5 ppm mostly obscure the H-2 resonance of malate that is formed (δ 4.16 ppm in added standard), but the H-3 resonances are visible.
H3a formed appears as a broad doublet at δ 2.53 ppm, between the methylene of oxaloacetae dihydrate and the double doublets of the 4-protons of NADPH. The intensity of this clearly rises as the NADPH and oxaloacetate resonances decrease in intensity. The H3b resonance of the malate formed is partly covered by the very large pyruvate resonance that is formed continuously by non-enzymatic decarboxylation of oxaloacetate.
254 255 (A) McyJ peptide sequence
(B) McyJ peptide sequence
Figure 5.1. Topological predictions for McyJ. (A) Kyte-Doolittle hydropathy profile for McyJ. Hydrophobic residues have a value greater than zero, hydrophylic residues have a value of less than zero. (B) DAS Transmembrane segment prediction for McyJ. Region above the broken, and unbroken lines indicate putative membrane-spanning regions with moderate, and high confidence values, respectively. No membrane spanning regions were predicted for McyJ.
256 257 0.05
894 Anabaena sp. 90 [AAO62583]* 1000 Nodularia spumigena [AAO64406]* Planktothrix agardhii [CAD29800]* Microcystis viridis [BAD32822]* 877 Microcystis sp. MCS3 [BAD32823]* A 998 Microcystis novacekii [BAD32825]* 907 Microcystis aeruginosa [BAD32824]* Microcystis novacekii [BAD32826]* Microcystis aeruginosa PCC7806 [AAF00954]* Halorhodospira halophila [ZP_01152570]G 1000 Pelodictyon phaeoclathratiforme [ZP_00589575] G 938 1000 Chlorobium limicola [ZP_00513091]G 861 Chlorobium phaeobacteroides [ZP_00531530] G B 1000 Chlorobium tepidum [AAM71629]G 635 725 Chlorobium phaeobacteroides DSM266 [ZP_00528886]G 992 Prosthecochloris aestuarii [ZP_00591555]G Saccharopolyspora erythraeum [CAA42929]* Aeromicrobium erythreum [AAU93801]* C 550 1000 Leptospira interrogans [AAN47802]G 943 Oceanospirillum sp. MED92 [ZP_01165185] G Nitrosococcus oceani ATCC19707 [ABA57503] G Anabaena circinalis 90 [CAC01607]* 473 854 symbiont bacterium of Theonella swinhoei [AAV97875]* D 970 Polyangium cellulosum [CAD43452]* 1000 Polyangium cellulosum [AAY89054]* Dictyostelium discoideum [XP_643680] G 373 Gibberella zeae PH-1 [XP_385916] 1000 Candida glabrata [AAX73199] E 663 Pneumocystis carinii [AAK54439] 1000 Pneumocystis carinii [AAK00294] 979 Anabaena variabilis ATCC29413 [ABA24403]G G 440 1000 Crocosphaera watsonii WH8501 [ZP_00518612] F Synechococcus sp. [ABD03340] 290 1000 Synechococcus sp. [ABD00391] Cytophaga hutchinsonii [ZP_00309576]G 1000 Methylococcus capsulatus [AAU90751] G 1000 Saccharopolyspora pogona [AAY88923]* 164 Saccharopolyspora spinosa [AAG23267]* 890 Streptomyces roche [NP_851450]* 501 Streptomyces cinnamonensis [AAO65792]* 988 Lechevalieria aerocolonigenes [BAC10678]* 989 Amycolatopsis mediterranei [AAC01738]* Streptomyces violaceusniger [ABC84455]* 791 362 G Streptomyces nogalater [AAG42853]* Frankia sp. [ZP_00573171]* 527 372 Streptomyces lavendulae [AAD28459]* 806 Streptomyces sp. TP-A0274 [BAC55218]* 191 Streptomyces avermitilis MA-4680 [NP_822112]* 1000 Streptomyces griseochromogenes [AAR15334]* Nostoc sp. PCC7120 [AAF75227]
Figure 5.3. Phylogenetic distribution of McyJ homologues. The phylogenetic tree encompasses methyltransferase homologues from cyanobacteria (bold) and other organisms. An rRNA methyltransferase from Nostoc sp. [AAF75227] constitutes the artificial out group. Asterisks (*) indicate sequences associated with NRP/PK gene clusters. (G) indicates sequences from uncharacterised proteins obtained from genome sequencing projects. For a detailed list of sequences used, see table 5.2. Bootstrap values derived from 1 000 replicates are given. The scale at the top of the tree corresponds to 5% divergence between sequences.
Figure 4. Phylogenetic distribution of mcyJ homologues. The phylogenetic tree encompasses methyltransferase homologues from cyanobacteria (bold) and other organisms. An rRNA methyltransferase from Nostoc sp. constitutes the artificial out group. Asterixes (*) indicate sequences associated with NRP/PK gene clusters. GSequences from uncharacterised proteins obtained from genome sequencing projects. For a detailed list of sequences used, see table 3. Bootstrap values are derived from 1000 replicates. The scale at the top of the tree corresponds to 5% divergence between sequences. 258 1 2 3 4 5 6 7 55.6 42.7 34.6 McyJ 27.0
20.0
Figure 5.4. SDS PAGE gel of partially purified polyhistidine tagged McyJ. Lane 1, Broad-range molecular weight standard (scale on left indicates molecular weights KDa). Lanes 2-7, McyJ elution fractions 1-6 following affinity chromatography using N 2+ column. Arrow indicates position of recombinant McyJ.
259 % Intensity
[DM-Adda, D-Asp3]microcystin-RR
[D-Asp3]microcystin-RR
(A) McyJ
% Intensity
[DM-Adda, D-Asp3]microcystin-RR
(B) Control
Figure 5.5. O-methylation of [DM-Adda, D-Asp3]microcystin-RR by McyJ. MALDI-TOF mass spectrometrographs of in vitro O-methylation assays. (A) Reaction containing recombinant McyJ. (B) Negative control. Major peaks in (A) and (B) correspond to [DM-Adda, D-Asp3]microcystin-RR (m/z 1 010, 1 011, 1 012). Additional peaks in (A) correspond to [D-Asp3]microcystin-RR (m/z 1024, 1 025, 1 026).
260 261 262 263 264 265