BIOSYNTHESIS OF CYCLIC PEPTIDE NATURAL PRODUCTS IN MUSHROOMS By Robert Michael Sgambelluri A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Biochemistry & Molecular Biology – Doctor of Philosophy 2017 ABSTRACT BIOSYNTHESIS OF CYCLIC PEPTIDE NATURAL PRODUCTS IN MUSHROOMS By Robert Michael Sgambelluri Cyclic peptide compounds possess properties that make them attractive candidates in the development of new drugs and therapeutics. Mushrooms in the genera Amanita and Galerina produce cyclic peptides using a biosynthetic pathway that is combinatorial by nature, and involves an unidentified, core set of tailoring enzymes that synthesize cyclic peptides from precursor peptides encoded in the genome. The products of this pathway are collectively referred to as cycloamanides, and include amatoxins, phallotoxins, peptides with immunosuppressant activities, and many other uncharacterized compounds. This work aims to describe cycloamanide biosynthesis and its capacity for cyclic peptide production, and to harness the pathway as a means to design and synthesize bioactive peptides and novel compounds. The genomes of Amanita bisporigera and A. phalloides were sequenced and genes encoding cycloamanides were identified. Based on the number of genes identified and their sequences, the two species are shown to have a combined capacity to synthesize at least 51 unique cycloamanides. Using these genomic data to predict the structures of uncharacterized cycloamanides, two new cyclic peptides, CylE and CylF, were identified in A. phalloides by mass spectrometry. Two species of Lepiota mushrooms, previously not known to produce cycloamanides, were also analyzed and shown to contain amatoxins, the toxic cycloamanides responsible for fatal mushroom poisonings. The mushroom Galerina marginata, which also produces amatoxins, was used as a model orgasnism for studying cycloamanide biosynthesis due to its culturability. Three enzymes involved in the biosynthesis of cycloamanides were identified in gene knockout studies: a predicted flavin-containing monooxygenase (FMO), P450 monooxygenase, and prolyl oligopeptidase (POP). The gene encoding a specific predicted prolyl oligopeptidase (POPB) was cloned and expressed in Saccharomyces cerevisiae for further characterization, and in vitro studies revealed that the enzyme is bifunctional, catalyzing both a hydrolysis reaction and the key cyclization step in cycloamanide biosynthesis. The utility of POPB as a general catalyst for peptide cyclization was explored by defining its subtrate preferences and limitations. POPB was shown to be highly versatile, catalyzing cyclization of diverse peptide sequences ranging from 8-16 residues in length and sequences containing modified amino acids in addition to the proteinogenic twenty. A method for the use of POPB for the production of combinatorial cyclic peptide libraries is also presented. A total of 100 cyclic peptides, including both novel compounds and bioactive cycloamanides, were produced in these studies and demonstrate the applications of POPB in biotechnology. Copyright by ROBERT MICHAEL SGAMBELLURI 2017 TABLE OF CONTENTS LIST OF TABLES ........................................................................................................... viii LIST OF FIGURES ........................................................................................................... ix KEY TO ABBREVIATIONS .......................................................................................... xiii CHAPTER 1 INTRODUCTION ........................................................................................1 1.1 Cyclic Peptides ................................................................................................2 1.2 Cycloamanides ................................................................................................5 1.3 Ribosomal Biosynthesis of Cycloamanides ....................................................8 WORKS CITED ...................................................................................................11 CHAPTER 2 DETECTION AND PROFILING OF AMATOXINS IN LEPIOTA MUSHROOMS ..................................................................................................................16 2.1 Abstract .........................................................................................................17 2.2 Introduction ...................................................................................................18 2.3 Methods .........................................................................................................21 2.3.1 Mushroom Collection and Identification .........................................21 2.3.2 Toxin Extraction and LCMS ............................................................22 2.4 Results ...........................................................................................................23 2.4.1 Toxins in Amanita and Galerina Mushrooms .................................23 2.4.2 Toxins in Lepiota Mushrooms .........................................................24 2.5 Discussion .....................................................................................................27 APPENDIX ..........................................................................................................28 WORKS CITED ...................................................................................................31 CHAPTER 3 GENOMIC CAPACITY FOR CYCLOAMANIDE BIOSYNTHESIS IN AMANITA MUSHROOMS .........................................................................................34 3.1 Abstract .........................................................................................................35 3.2 Introduction ...................................................................................................36 3.3 Methods .........................................................................................................37 3.3.1 Genomics and Transcriptomics........................................................37 3.3.2 LC/MS/MS of Predicted Cycloamanides.........................................37 3.4 Results ...........................................................................................................39 3.4.1 MSDIN Genes in Amanita bisporigera and A. phalloides ..............39 3.4.2 New Cycloamanides in Amanita phalloides ....................................39 3.5 Discussion .....................................................................................................43 APPENDIX ..........................................................................................................45 WORKS CITED ...................................................................................................49 v CHAPTER 4 CHARACTERIZATION OF AMANITIN BIOSYNTHESIS IN GALERINA MARGINATA ..............................................................................................52 4.1 Abstract .........................................................................................................53 4.2 Introduction ...................................................................................................54 4.3 Methods .........................................................................................................57 4.3.1 Galerina Growth and Toxin Analysis ..............................................57 4.3.2 Galerina Transformation and Gene Knockouts ...............................58 4.3.3 Purification of an Amanitin Intermediate and NMR .......................58 4.3.4 Analysis of Gene Expression by RT-PCR .......................................59 4.4 Results ...........................................................................................................60 4.4.1 Time Course of Amanitin Production ..............................................60 4.4.2 Genes Involved in Amanitin Biosynthesis .......................................61 4.4.3 Role of a P450 Monooxygenase in Amanitin Biosynthesis.............64 4.4.4 Regulation of Biosynthetic Genes ...................................................66 4.5 Discussion .....................................................................................................68 APPENDIX ..........................................................................................................70 WORKS CITED ...................................................................................................78 CHAPTER 5 BIOCHEMICAL CHARACTERIZATION OF PROLYL OLIGOPEPTIDASE B AS A PEPTIDE MACROCYCLASE .........................................81 5.1 Abstract .........................................................................................................82 5.2 Introduction ...................................................................................................83 5.3 Methods .........................................................................................................85 5.3.1 Protein Expression and Purification.................................................85 5.3.2 Enzyme Assays ................................................................................86 5.3.3 Product Purification and NMR Spectroscopy ..................................86 5.4 Results ...........................................................................................................88 5.4.1 Preparation of Recombinant GmPOPB ...........................................88 5.4.2 GmPOPB Catalyzes Peptide Macrocyclization ...............................89 5.4.3 GmPOPB is a Bifunctional Enzyme ................................................91 5.4.4 Residues Involved in Macrocyclization ...........................................94 5.5 Discussion .....................................................................................................96 APPENDIX ..........................................................................................................97 WORKS CITED .................................................................................................104 CHAPTER 6 VERSATILITY OF PROLYL OLIGOPEPTIDASE B IN PEPTIDE MACROCYCLIZATION ................................................................................................108 6.1 Abstract .......................................................................................................109 6.2 Introduction .................................................................................................110 6.3 Methods .......................................................................................................112 6.3.1 DNA Constructs .............................................................................112 6.3.2 Preparation of POPB Substrates ....................................................112 6.3.3 Cyclization Assays and LCMS ......................................................113 6.3.4 Library Preparation and Analysis ..................................................113 6.4 Results .........................................................................................................115 6.4.1 Enzyme and Substrate Preparation ................................................115 vi 6.4.2 Amino Acid Preferences for Cyclization .......................................115 6.4.3 Cyclization of Sequences Containing Unusual Amino Acids .......118 6.4.4 Core Domain Length Requirement ................................................120 6.4.5 Synthesis of Naturally Occurring Cycloamanides .........................120 6.4.6 Cyclization of the Phalloidin Sequence with D-threonine .............123 6.4.7 Cyclic Peptide Library Production.................................................123 6.5 Discussion ...................................................................................................127 APPENDIX ........................................................................................................128 WORKS CITED .................................................................................................145 vii LIST OF TABLES Table 2.1: Compounds Identified in Extracts of Amanita, Galerina and Lepiota Mushrooms .......................................................................................................................24 Table 2.2: α-Amanitin Concentrations in Mushrooms .................................................26 Table 3.1: Cycloamanide Sequences in the Genomes of A. bisporigera and A. phalloides ......................................................................................................................40 Table S3.1: List of Core Domains Identified Among MSDIN Transcripts from RNAseq of Amanita bisporigera ......................................................................................46 Table S3.2: Alphabetical List of All Unique MSDIN Core Sequences Identified to Date ...............................................................................................................................47 Table S4.1: Table of all 2D NMR Correlations Observed in the Amanitin Intermediate .....................................................................................................................76 Table 5.1: GmPOPB Cyclization Kinetic Constants with 35mer and 25mer GmAMA1 Substrates.......................................................................................................93 Table 5.2: Differentially Conserved Residues between POPA and POPB .................95 Table 6.1: Tolerance of POPB for Amino Acid Substitutions in the Core Region of AMA1..........................................................................................................................117 Table 6.2: Cyclization of Naturally Occurring Cycloamanides.................................122 Table S6.1: Compared Cyclization Yields of 35mer and 25mer Substrates ............129 Table S6.2: Alphabetical List of Cyclic Peptides Produced with POPB ...................143 Table S6.3: Alphabetical List of Peptides Not Efficiently Cyclized by POPB .........144 viii LIST OF FIGURES Figure 1.1: Macrocyclic Bonds in Cyclic Peptides. .........................................................2 Figure 1.2: RiPP Precursor Peptide Structure................................................................4 Figure 1.3: Amatoxin and Phallotoxin Structure............................................................6 Figure 1.4: Other Cycloamanide Compounds.................................................................7 Figure 1.5: Southern Blot Analysis of Amanita Mushrooms. ........................................9 Figure 1.6: Primary Structure of the Cycloamanide Precursor Peptides. ...................9 Figure 1.7: Compared Sequences of Amanitin Precursors from Amanita and Galerina .....................................................................................................................10 Figure 2.1: Amatoxin and Phallotoxin Families of Bicyclic Peptides .........................19 Figure 2.2: Mushroom Species Analyzed in this Study for Toxin Content ................20 Figure 2.3: Toxin Profiles of Amanita, Galerina, and Lepiota Mushrooms ................25 Figure S2.1: ITS Sequences from Species of Lepiota Mushrooms...............................29 Figure 3.1: MS/MS Analysis of Cycloamanide E ..........................................................41 Figure 3.2: MS/MS Analysis of Cycloamanide F ..........................................................42 Figure 3.3: Amino Acid Frequencies in MSDIN Core Domain Sequences ................44 Figure 4.1: Basidiocarps of Galerina marginata ............................................................54 Figure 4.2: Steps in the Biosynthesis of α-Amanitin from the GmAMA1 Precursor Peptide.............................................................................................................56 Figure 4.3: Genes Adjacent to GmAMA1 in the Galerina marginata Genome with Relevant Predicted Functions.................................................................................56 Figure 4.4: Culture of Galerina marginata Mycelium. .................................................60 Figure 4.5: Time Course of α-Amanitin Production in Galerina marginata Cultures.............................................................................................................................61 ix Figure 4.6: POPB Knockout in Galerina and Effects on α-Amanitin Production. ....62 Figure 4.7: FMO Knockout in Galerina and Effects on α-Amanitin Production.......63 Figure 4.8: P450 Knockout in Galerina and Effects on α-Amanitin Production. ......63 Figure 4.9: Compared Structures of α-Amanitin and the Pathway Intermediate Purified from the P450(-) Strain of G. marginata . .......................................................64 Figure 4.10: Compared 1H-13C HSQC Spectra of α-Amanitin and the Intermediate . ...................................................................................................................65 Figure 4.11: Expression of Genes Involved in Amanitin Biosynthesis........................67 Figure S4.1: 1H Spectra of α-Amanitin and the Pathway Intermediate Purified from a P450(-) Strain of G. marginata............................................................................71 Figure S4.2: 2D COSY Spectrum of the Amanitin Intermediate ................................72 Figure S4.3: 2D TOCSY Spectrum of the Amanitin Intermediate .............................72 Figure S4.4: 2D ROESY Spectrum of the Amanitin Intermediate .............................73 Figure S4.5: 1H-13C HSQC Spectrum of the Amanitin Intermediate .........................73 Figure S4.6: 1H-13C HMBC Spectrum of the Amanitin Intermediate ........................74 Figure S4.7: Key HMBC Correlations Observed in the Amanitin Intermediate for Structure Determination ...........................................................................................75 Figure 5.1: Crystal Structure of Prolyl Oligopeptidase from Porcine Brain. ............84 Figure 5.2: Purification of Recombinant GmPOPB Expressed in Yeast ....................88 Figure 5.3: Time Course of Conversion of GmAMA1 to cyclo-IWGIGCNP .............89 Figure 5.4: Amide Bond Couplings in HMBC Spectrum of cyclo-IWGIGCNP ........90 Figure 5.5: Reaction Products from GmPOPB Activity on GmAMA1 ......................92 Figure 5.6: Two-Step Nonprocessive Reaction Catalyzed by POPB on the α-Amanitin Precursor Peptide ........................................................................................92 Figure 5.7: Kinetic Analysis of GmPOPB......................................................................93 Figure 5.8: Hyopthetical Mechanism for Macrocyclization Catalyzed by POPB .....96 x Figure S5.1: 2D gCOSY Spectrum of cyclo-IWGIGCNP ............................................98 Figure S5.2: 2D TOCSY Spectrum of cyclo-IWGIGCNP ...........................................98 Figure S5.3: Amide Region from TOCSY Spectrum of cyclo-IWGIGCNP...............99 Figure S5.4: 2D ROESY Spectrum of cyclo-IWGIGCNP .........................................100 Figure S5.5: 2D 1H-13C HSQC Spectrum of cyclo-IWGIGCNP................................100 Figure S5.6: Formation of cyclo-IWGIGCNP Product from 35mer and 25mer Substrates........................................................................................................................101 Figure S5.7: Multiple Sequence Alignment of POPB and Other Prolyl Oligopeptidases ..............................................................................................................102 Figure 6.1: Expression and Purification of the Amanitin Precursor Peptide ..........116 Figure 6.2: Cyclization of Peptides Containing Unusual Amino Acids ....................119 Figure 6.3: POPB Products Produced from Substrates with Varying Core Domain Lengths ............................................................................................................................121 Figure 6.4: LCMS Comparing POPB Products Produced from Substrates Containing the PHD Sequence with Either L-Thr or D-Thr .....................................124 Figure 6.5: Scheme for Generating Mixed Cyclic Peptide Libraries ........................125 Figure S6.1: Effect of Single Amino Acid Substitutions on Cyclization by POPB at Position 1 of the AMA1 Core Domain .....................................................................129 Figure S6.2: Effect of Single Amino Acid Substitutions on Cyclization by POPB at Position 2 of the AMA1 Core Domain .....................................................................130 Figure S6.3: Effect of Single Amino Acid Substitutions on Cyclization by POPB at Position 3 of the AMA1 Core Domain .....................................................................131 Figure S6.4: Effect of Single Amino Acid Substitutions on Cyclization by POPB at Position 4 of the AMA1 Core Domain .....................................................................132 Figure S6.5: Effect of Single Amino Acid Substitutions on Cyclization by POPB at Position 5 of the AMA1 Core Domain .....................................................................133 Figure S6.6: Effect of Single Amino Acid Substitutions on Cyclization by POPB at Position 6 of the AMA1 Core Domain .....................................................................134 xi Figure S6.7: Effect of Single Amino Acid Substitutions on Cyclization by POPB at Position 7 of the AMA1 Core Domain .....................................................................135 Figure S6.8: LCMS Traces of Naturally Occurring Cycloamanide Core Regions Cyclized by POPB ..........................................................................................................136 Figure S6.9: Simultaneous Production of Ten Cyclic Peptides Using POPB ...........137 Figure S6.10: Batch Production of Cyclic Peptides Using POPB ..............................138 xii KEY TO ABBREVIATIONS ACN Acetonitrile, CH3CN BCA Bicinchoninic acid BLAST Basic local alignment search tool BSA Bovine serum albumin CTAB Cetyltrimethyl ammonium bromide DQF-COSY Double quantum filtered correlation spectroscopy DMSO Dimethyl sulfoxide DTT Dithiothreitol EIC Extracted ion chromatogram ESI Electrospray ionization FMO Flavin-containing monooxygenase HSQC Heteronuclear single-quantum correlation spectroscopy HMBC Heteronuclear multiple-bond correlation spectroscopy HPLC High-performance liquid chromatography IPTG Isopropyl β-D-1-thiogalactopyranoside ITS Internal transcribed spacer LCMS Liquid chromatography - mass spectrometry LC/MS/MS Liquid chromatography - tandem mass spectrometry MBP Maltose-binding protein MS Mass spectrometry NOE Nuclear Overhauser effect xiii NRPS Nonribosomal peptide synthetase OATP Organic anion-transporting polypeptide PCR Polymerase chain reaction PDA Potato Dextrose Agar PDB Protein Data Bank POP Prolyl oligopeptidase PTM Post-translational modification RiPP Ribosomally synthesized and post-translationally modified peptide ROESY Rotating frame nuclear Overhauser effect spectroscopy RT-PCR Reverse transcription polymerase chain reaction SDS-PAGE Sodium dodecyl sulfate - polyacrylamide gel electrophoresis tblastn BLAST query of protein sequence against translated nucleotide TBS Tris-buffered saline TLC Thin layer chromatography TOCSY Total correlation spectroscopy xiv CHAPTER 1 INTRODUCTION 1 1.1 Cyclic Peptides Cyclic peptides are compounds composed of amino acids with covalent linkages forming macrocyclic ring structures [1]. Familiar examples include the immunosuppressant cyclosporine from fungi [2], the peptide hormone oxytocin [3,4], and the antibiotic daptomycin from Streptomyces roseosporus [5]. Macrocycles may be formed by linkages between the amino and carboxyl termini, between amino acid side-chains, or by a linkage between an N/C-terminus and side-chain (Figure 1.1). Cyclic peptides are produced by prokaryotes and eukaryotes and are abundant and diverse in nature, ranging in size from the 6-11 residue cyanobactins found in cyanobacteria [6] to the 35-78 residue bacteriocins found in bacteria [7]. Figure 1.1: Macrocylic Bonds in Cyclic Peptides. The characteristic ring structures of cyclic peptides arise from covalent bonds linking amino acid side-chains, the N- and Ctermini, or a side-chain to an N/C-terminus. Macrocyclic bonds afford peptides with three attributes than result in a high potential for bioactivity. First, cyclic peptides bind to targets with high affinity that results from their increased structural rigidity and a reduced entropic penalty to binding [8]. Second, cyclic 2 peptides are generally stable against proteolysis since macrocyclic bonds involving the N- or Ctermini prevent exoprotease activity, and because any macrocyclic bond is conformationally restrictive. Fairlie et al (2000) [9] compared the conformations of 266 bound protease substrates and inhibitors from structures in the Protein Data Bank (PDB) and found that all adopted the same extended conformation. The structural constraints imposed by cyclization in peptides often prevents this extended conformation that is required for protease activity [1,9]. Finally, enhanced membrane permeability is often displayed by cyclic peptides, and is proposed to result from their ability to more readily adopt conformations that bury polar backbone and side-chain groups in intramolecular interactions [10]. This is exemplified in cyclosporine A, which contains backbone amides and carbonyls that form intramolecular hydrogen bonds in nonpolar solvent [11] but that are solvent-exposed in water [12], affording cell permeability and broad solubility in different solvents [13]. These attributes that often result in bioactivity make cyclic peptides desirable candidates in the development of new therapeutics and research tools [14]. Since 2006, nine cyclic peptides have been approved for clinical use and at least 20 are currently being evaluated for the treatment of infections, cancer, metabolic disorders, blood disorders, and cardiovascular disease [15]. Currently, more than 40 cyclic peptides are in clinical use and most are derived from natural products [13,15]. Despite their usefulness, however, organic synthesis of these compounds remains difficult and expensive [16]. Nature has overcome these challenges through the use of nonribosomal peptide synthetase (NRPS) enzymes [17], as well as recently discovered ribosomal pathways, in which genetically encoded precursor peptides (Figure 1.2) serve as substrates for promiscuous tailoring enzymes resulting in natural libraries of cyclic peptides [18]. Peptide natural products arising from ribosomal pathways are referred to as ribosomally synthesized and 3 post-translationally modified peptides, or RiPPs [19]. Macrocyclic RiPP natural products are now known to be widespread throughout nature, and examples include the microcins [20] and thiopeptides [21] in bacteria, cyanobactins in cyanobacteria [6], conotoxins in conesnails [22], cyclotides in plants [23], and ustiloxins [24] and cycloamanides [25] in fungi. While an extensive number of ribosomal cyclic peptides with useful bioactivities have been discovered, many of the enzymes involved in their biosynthesis remain uncharacterized and elucidation of these pathways could provide efficient strategies for producing these compounds in sufficient quantities. In addition, since RiPP tailoring enzymes display broad substrate preferences and the amino acid sequences of the products can be easily manipulated at the DNA level, RiPP pathways are well suited for use in synthetic biology and the production of novel compounds. Figure 1.2: RiPP Precursor Peptide Structure. Ribosomal peptide precursors are generally composed of leader, core, and recognition or follower domains. The core peptide contains the amino acids found in the final RiPP product. 4 1.2 Cycloamanides Cycloamanides are cyclic peptides produced by mushrooms in the genus Amanita and include amatoxins, the causative agents of fatal mushroom poisonings [26]. Amatoxins bind to and inhibit eukaryotic RNA polymerase II [27] and are often employed as experimental tools for the study of their target. α-Amanitin, the most abundant and potent of the amatoxins, displays an oral LD50 of 0.1 mg/kg in rats [26] and the crystal structure of RNA polymerase II with bound amanitin has been solved [28]. Stability of the amatoxins against cooking and the digestive tract, as well as rapid uptake into hepatocytes through the OATP transporter protein [29] contribute to the toxicity of these compounds. Phallotoxins comprise another family of cycloamanides, and include phalloidin, which binds with high affinity to actin [30]. Fluorescent conjugates of phalloidin are used in cell imaging [31], and the compound is one of the most widely used tools in chemical biology. Structurally, the amatoxins are bicyclic octapeptides with the amino acid sequence IWGIGC(N/D)P, and the phallotoxins are bicyclic heptapeptides with the sequence AWL(V/A)(D/T)CP (Figure 1.3). Both the amatoxins and phallotoxins contain cyclic backbones, with N- and C-termini linked ‘head-to-tail’ by an additional peptide bond [26]. Other modifications found in both toxin families include side-chain hydroxylations and a unique “tryptathionine” [32] linkage between the side-chains of tryptophan and cysteine, a modification that has not been seen in other natural products, resulting in the overall bicyclic structure. Phallotoxins also contain a D-configured aspartic acid or threonine residue. Combined, the amatoxins and phallotoxins comprise 16 known compounds, and diversity among each class arises from differences in both amino acid sequence and side-chain hydroxylation patterns [26]. 5 Figure 1.3: Amatoxin and Phallotoxin Structure. α-Amanitin (A) and phalloidin (B) are bicyclic peptides with cyclic backbones, side-chain hydroxylations (shown in red), and a tryptophan-cysteine linkage (shown in blue). Other cycloamanides that have been isolated include the virotoxins [33], cycloamanides A through D [34], antamanide [35], and amanexitide [36] (Figure 1.4). Virotoxins are similar in structure to the phallotoxins but are monocyclic and contain a tryptophan residue methylsulfonyl modification. The other known cycloamanides, including antamanide and amanexitide, range from six to ten amino acids in length and contain cyclic backbones and unmodified side-chains. All of the post-translational modifications present among the amatoxins, phallotoxins, and other cycloamanides are difficult or currently not possible to achieve in peptides using synthetic chemistry, with the exception of incorporating D-amino acids into peptides. Therefore, insights into cycloamanide biosynthesis and characterization of the enzymes involved may provide new routes and synthetic strategies to these modifications. 6 Figure 1.4: Other Cycloamanide Compounds. Structures of cycloamanides A through D (CyaA-D), antamanide (ANT), and viroisin (a virotoxin) are shown. 7 1.3 Ribosomal Biosynthesis of Cycloamanides Although the amatoxins and phallotoxins were intially isolated in the 1930s and 1940s [26], no details regarding their biosynthesis were known until 2007, when they were identified as ribosomal peptides. Using a BLAST query for the amino acid sequences of α-amanitin (IWGIGCNP) and the phallotoxin phallacidin (AWLVDCP), Hallen et al. (2007) [25] identified sequences in the genome of the poisonous mushroom Amanita bisporigera that could encode the toxins. The sequences were located within longer open reading frames with conserved upstream and downstream sequences, and each encoding a translation product 34 or 35 amino acids in length. The sequences were shown by Southern blot analysis to be present only in species of mushrooms that produce these toxins (Figure 1.5) and were named AMA1 and PHA1 for αamanitin and phallacidin, respectively. Targeting conserved regions of the sequences, PCR and additional BLAST searches revealed 14 additional sequences, all encoding predicted oligopeptides beginning with the sequence ‘MSDIN.’ It was concluded that the amatoxins and phallotoxins were products of a ribosomal biosynthesis pathway with a conserved ‘MSDIN’ gene family encoding precursor peptides to the toxins and to the other cycloamanides. The sequence structure of the cycloamanide precursor peptides is shown in Figure 1.6, as revealed through a multiple sequence alignment of the MSDIN sequences identified in A. bisporigera [37]. The sequences consist of conserved N-terminal (10mer) and C-terminal (17mer) regions that flank an internal hypervariable sequence that contains the amino acids found in the final cyclic peptide products. All known cycloamanides contain at least one proline residue [26], and each internal sequence invariably starts and ends with proline. The degree of conservation among the MSDIN sequences strongly implies that cycloamanide biosynthesis is 8 combinatorial, and that after translation the precursors function as ‘scaffolds,’ with recognition sequences for the same core biochemical machinery and tailoring enzymes resulting in a variety of cyclic peptides. Figure 1.5: Southern Blot Analysis of Amanita Mushrooms. Lanes 1-4 contain genomic DNA from α-amanitin and phallacidin producing species, and Lanes 5-13 from non-toxic species. A, probed with AMA1 cDNA. B, probed with fragment of β-tubulin gene. C, probed with PHA1 DNA. D, stained with ethidium bromide. Reprinted with permission from Hallen et al., 2007 [25]. Copyright © 2007 National Academy of Sciences. Figure 1.6: Primary Structure of the Cycloamanide Precursor Peptides. The image shows a WebLogo representation from a multiple sequence alignment of the 19 translated MSDIN sequences identified in A. bisporigera, with the degree of conservation at each position indicated by the height of each residue. The amino acid sequences of the cycloamanides are located internally and are flanked by conserved N- and C-terminal domains. Reprinted with permission from Luo et al., 2009 [44]. Copyright © 2009 ASBMB. 9 Species in the mushroom genus Galerina such as G. marginata are also known to produce amatoxins [38,39,40]. Unlike A. bisporigera, which contains at least 19 MSDIN sequences, only two genes (GmAma1-1 and GmAma1-2) are found in the G. marginata genome, both encoding the α-amanitin sequence [41]. The precursor peptide sequences in Galerina diverge slightly from those in Amanita, especially in the C-terminal domain, and Figure 1.7 shows an alignment of the two AMA1 sequences. Although cycloamanide biosynthesis appears to be more limited in Galerina, G. marginata may serve as an excellent model organism for characterization of the pathway since its genome and transcriptome have been fully sequenced and annotated [41], and unlike Amanita spp. it can be cultured in the laboratory [42,43]. Figure 1.7: Compared Sequences of Amanitin Precursors from Amanita and Galerina. Sequences are from G. marginata (GmAMA1, top) and A. bisporigera (AbAMA1, bottom). Divergent residues are highlighted in red and the internal α-amanitin sequence is underlined. While some of the genes identified in A. bisporigera contained sequences for known cycloamanides, the majority are predicted to encode previously undiscovered cycloamanides and new natural products. Therefore, a detailed understanding of cycloamanide biosynthesis may provide a means to access new bioactive natural products, as well as characterized cycloamanides that are present in mushrooms in low abundance. 10 WORKS CITED 11 WORKS CITED 1. Craik DJ. (2006). Seamless proteins tie up their loose ends. Science 311(5767): 1563-1564. 2. Laupacis A, Keown PA, Ulan RA, McKenzie N, and Stiller CR (1982). Cyclosporin A: a powerful immunosuppressant. Can. Med. Assoc. J. 126(9): 1041-1046. 3. Du Vigneaud V, Ressler C, and Trippett S. (1953). The sequence of amino acids in oxytocin, with a proposal for the structure of oxytocin. J. Biol. Chem. 205(2): 949-957. 4. Gimpl G, and Fahrenhol F. (2001). The oxytocin receptor system: structure, function, and regulation. Physiol. Rev. 81(2): 629-683. 5. Baltz RH. (2009). Daptomycin: mechanisms of action and resistance, and biosynthetic engineering. Curr. Opin. Chem. Biol. 13(2): 144-151. 6. Sivonen K, Leikoski N, Fewer DP, and Jokela J. (2010). Cyanobactins - ribosomal cyclic peptides produced by cyanobacteria. Appl. Microbiol. Biotechnol. 86(5): 1213-1225. 7. Jack RW, Tagg JR, and Ray B. (1995). Bacteriocins of gram-positive bacteria. Microbiol. Rev. 59(2): 171-200, 8. Millward SW, Fiacco S, Austin RJ, and Roberts RW. (2007). Design of cyclic peptides that bind protein surfaces with antibody-like affinity. ACS Chem. Biol. 2(9): 625-634. 9. Fairlie DP, Tyndall JD, Reid RC, Wong, AK, Abbenante, G, Scanlon, MJ, March, DR, Bergman, DA, Chai, CLL, and Burkett, B.A. (2000). Conformational selection of inhibitors and substrates by proteolytic enzymes: Implications for drug design and and polypeptide processing. J. Med. Chem. 43(7): 1271-1281. 10. Rezai T, Yu B, Millhauser GL, Jacobson MP, and Lokey RS. (2006). Testing the conformational hypothesis of passive membrane permeability using synthetic cyclic peptide diastereomers. J. Am. Chem. Soc. 128(8): 2510-2511. 11. Witek J, Keller BG, Blatter M, Meissner A, Wagner T, and Riniker S. (2016). Kinetic models of cyclosporin A in polar and apolar environments reveal multiple congruent conformational states. J. Chem. Inf. Model. 56(8): 1547-1562. 12. Ko SY and Dalvit C. (1992). Conformation of cyclosporin A in polar solvents. Int. J. Pept. Protein Res.40(5): 380-382. 12 13. Naylor MR, Bockus AT, Bianco MJ, and Lokey RS. (2017). Cyclic peptide natural products chart the frontier of oral bioavailability in the pursuit of undruggable targets. Curr. Opin. Chem. Biol. 38: 141-147. 14. Driggers EM, Hale SP, Lee J, and Terret NK. (2008). The exploration of macrocycles for drug discovery - an underexploted structural class. Nat. Rev. Drug. Discov. 7(7): 608624. 15. Zorzi A, Deyle K, and Heinis C. (2017). Cyclic peptide therapeutics: past, present and future. Curr. Opin. Chem. Biol. 38: 24-29. 16. Marsault E, and Peterson ML. (2011). Macrocycles are great cycles: applications, opportunities, and challenges of synthetic macrocycles in drug discovery. J. Med. Chem. 54(7): 1961-2004. 17. Felnagle EA, Jackson EE, Chan YA, Podevels AM, McMahon MD, and Thomas MG. (2008). Nonribosomal peptide synthetases involves in the production of medically relevent natural products. Mol. Pharm. 5(2): 191-211. 18. Oman TJ, and van der Donk, W.A. (2010). Follow the leader: the use of leader peptides to guide natural product biosynthesis. Nat. Chem. Biol. 6(1): 9-18. 19. Arnison PG et al. (2013). Ribosomally synthesized and post-translationally modified peptide natural products: overview and recommendations for a universal nomenclature. Nat. Prod. Rep. 30(1): 108-160. 20. Duquesne S, Destoumieux-Garzon D, Peduzzi J, and Rebuffat S. (2007). Microcins, geneencoded antibacterial peptide from enterobacteria. Nat. Prod. Rep. 24(4): 708-734. 21. Liao R, Duan L, Lei C, Pan H, Ding Y, Zhang Q, Chen D, Shen B, Yu Y, and Liu W. (2009). Thiopeptide biosynthesis featuring ribosomally synthesized precursor peptides and conserved posttranslational modifications. Chem. Biol. 16(2): 141-147. 22. Duda TF Jr, and Palumbi SR. (1999). Molecular genetics of ecological diversification: duplication and rapid evolution of toxin genes of the venomous gastropod Conus. Proc. Natl. Acad. Sci. U.S.A. 96(12): 6820-6823. 23. Craik DJ, Cemazar M, Wang CK, and Daly NL. (2006). The cyclotide family of circular miniproteins: nature’s combinatorial peptide template. Biopolymers 84(3): 250-266. 24. Tsukui T, Nagano N, Umemura M, Kumagai T, Terai G, Machida M, and Asai K. (2015). Ustiloxins, fungal cyclic peptides, are ribosomally synthesized in Ustilaginoidea virens. Bioinformatics 31(7): 981-985. 13 25. Hallen HE, Luo H, Scott-Craig JS, and Walton JD. (2007). Gene family encoding the major toxins of lethal Amanita mushrooms. Proc. Natl. Acad. Sci. U.S.A. 104(48): 1909719101. 26. Wieland, T. (1986). Peptides of Poisonous Amanita Mushrooms. Springer: New York. 27. Lindell TJ, Weinberg F, Morris PW, Roeder RG, and Rutter WJ. (1970). Specific inhibition of nuclear RNA polymerase II by alpha-amanitin. Science 170(3956): 447-449. 28. Bushnell DA, Cramer P, Kornberg RD. (2001). Structural basis of transcription: αAmanitin-RNA polymerase II cocrystal at 2.8 Å resolution. Proc. Natl. Acad. Sci. U.S.A. 99(3): 1218-1222. 29. Letschert K, Faulstich H, Keller D, and Keppler D. (2006). Molecular characterization and inhibition of amanitin uptake into human hepatocytes. Toxicol. Sci. 91(1): 140-149. 30. Lengsfield AM, Löw I, Wieland T, Dancker P, and Hasselbach W. (1974). Interaction of phalloidin with actin. Proc. Natl. Acad. Sci. U.S.A. 71(7): 2803-2807. 31. Wulf E, Deboben A, Bautz FA, Faulstich H, and Wieland T. (1979). Fluorescent phallotoxin, a tool for visualization of cellular actin. Proc. Natl. Acad. Sci. U.S.A. 76(9): 4498-4502. 32. May JP, and Perrin DM. (2007). Tryptathionine bridges in peptide synthesis. Biopolymers 88(5): 714-724. 33. Faulstich H, Buku A, Bodenmuller H, and Wieland T. (1980). Virotoxins: actin-binding cyclic peptides of Amanita virosa mushrooms. Biochemistry 19(14): 3334-3343. 34. Wieczorek Z, Siemion IZ, Zimecki M, Bolewska-Pedyczak E, and Wieland T. (1993). Immunosuppressive activity in the series of cycloamanide peptides from mushrooms. Peptides 14(1):1-5. 35. Wieland T. (1968). The discovery, isolation, elucidation of structure, and synthesis of antamanide. Angew. Chem. Int. Ed. Engl. 7(3): 204-208. 36. Xue JH, Ping W, Chi YL, Xu LX, and Wei XY. (2011). Cyclopeptides from Amanita exitialis. Nat. Prod. Bioprospect. 1(1): 52-56. 37. Walton JD, Hallen-Adams HE, and Luo H. (2010). Ribosomal biosynthesis of the cyclic peptide toxins of Amanita mushrooms. Biopolymers 94(5): 659-664. 38. Johnson BE, Preston JF, and Kimbrough JW. (1976). Quantitation of amanitins in Galerina autumnalis. Mycologia 68(6): 1248-1253. 14 39. Muraoka S, Fukamachi N, Mizumoto K, and Shinozawa T. (1999). Detection and identification of amanitins in the wood-rotting fungi Galerina fasciculata and Galerina helvoliceps. Appl. Environ. Microbiol. 65(9): 4207-4210. 40. Enjalbert F, Cassanas G, Rapior S, Renault C, and Chaumont JP. (2004). Amatoxins in wood-rotting Galerina marginata. Mycologia 96(4): 720-729. 41. Luo H, Hallen-Adams HE, Scott-Craig JS, and Walton JD. (2012). Ribosomal biosynthesis of α-amanitin in Galerina marginata. Fungal Genet. Biol. 49(2): 123-129. 42. Benedict RG, Tyler VE Jr, Brady LR, and Weber LJ. (1966). Fermentative production of amanita toxins by a strain of Galerina marginata. J. Bacteriol. 91(3): 1380-1381. 43. Muraoka S, and Shinozawa T. (2000). Effective production of amanitins by two-step cultivation of the basidiomycete, Galerina fasciculata GF-060. J. Biosci. Bioeng. 89(1): 73-76. 44. Luo H, Hallen-Adams HE, and Walton JD. (2009). Processing of the phalloidin proprotein by prolyl oligopeptidase from the mushroom Conocybe albipes. J. Biol. Chem. 284(27): 18070-18077. 15 CHAPTER 2 DETECTION AND PROFILING OF AMATOXINS IN LEPIOTA MUSHROOMS Note: The content in this chapter has been previously published. Some text has been modified from the original. Copyright © 2014 by the authors; licensee MDPI, Basel, Switzerland. Citation: Sgambelluri RM, Epis S, Sassera D, Luo H, Angelos ER, and Walton JD. (2014). Profiling of amatoxins and phallotoxins in the genus Lepiota by liquid chromatography combined with UV absorbance and mass spectrometry. Toxins. 6(8): 2336-2347. Author Contributions: Fungal specimens were collected and identified by Sara Epis and Davide Sassera. Evan R. Angelos and Hong Luo confirmed taxonomic identifications with ITS sequencing. 16 2.1 Abstract Ingestion of mushrooms in the genus Lepiota can result in fatal poisonings. Although clinical symptoms and low resolution methods indicated that toxicity is due to the presence of amatoxins, the toxin composition of Lepiota mushrooms has not been analyzed by modern high resolution techniques. The spectrum of peptide toxins present in five species of Lepiota were analyzed by liquid chromatography-mass spectrometry (LCMS). Field taxonomic identifications were confirmed by sequencing of the internal transcribed spacer (ITS) regions. Extracts of other poisonous mushrooms with previously characterized and well defined toxin profiles, including Amanita phalloides, A. virosa, and Galerina marginata, were analyzed for comparison. The compounds α-amanitin, β-amanitin, amanin, and amaninamide were detected in all isolates of L. brunneoincarnata, and α-amanitin and γ-amanitin were detected in all isolates of L. josserandii. Phallotoxins were not detected in either species. No amatoxins or phallotoxins were detected in L. clypeolaria, L. cristata, or L. echinacea. 17 2.2 Introduction The amatoxins, such as α-amanitin, are a group of bicyclic peptides produced by some species of mushrooms and account for the majority of fatal mushroom poisonings worldwide [1]. They display potent inhibition of eukaryotic RNA polymerase II, and factors that contribute to their toxicity include resistance to heat and the digestive tract, and active intestinal and cellular uptake [2]. Amatoxin poisoning is clinically manifested as symptoms of gastroenteritis resolving into an asymptomatic period and ultimately followed by fulminant liver failure. In severe cases, liver transplantation is the sole recourse [3]. In clinical settings, amatoxin poisoning is often assumed on the basis of hepatic misfunction subsequent to mushroom ingestion, even in the absence of chemical evidence [1,3]. Structurally, the amatoxins comprise the amino acid sequence Ile-Trp-Gly-Ile-Gly-Cys-Asn/AspPro, cyclized by head-to-tail peptide bonds and also a ‘tryptathionine’ side-chain linkage between tryptophan and cysteine residues. Further diversity among of the amatoxins arises from differences in hydroxylations of the side chains, which include 4-hydroxy-Pro, γ,δ-dihydroxy-Ile, and 6-hydroxy-Trp (Figure 2.1). All of the amatoxins contain a cysteine with a sulfur oxidized to the sulfoxide [4]. The phallotoxins, such as phalloidin and phallacidin, are a related class of bicyclic heptapeptides that also contain tryptathionine. The phallotoxin core sequence is Ala-Trp-Leu-Ala/Val-DAsp/Thr-Cys-Pro, and differences in hydroxylations also generate structural diversity (Figure 2.1) [4]. Phallotoxins bind and stabilize F-actin, and their fluorescent conjugates are used as cytological reagents to delineate the actin cytoskeleton [5]. 18 Figure 2.1: Amatoxin and Phallotoxin Families of Bicyclic Peptides. Numbers in parentheses after the compound names refer to the peak numbers in chromatography traces shown later in the text. Although species in the Amanita and Galerina genera (Figure 2.2A-C) are the most notorious source of amatoxins and account for most fatal mushroom poisonings, numerous deaths have also been attributed to ingestion of Lepiota, a genus of small, saprobic mushrooms distributed worldwide (Figure 2.2D-H) [6-12]. However, in constrast to Amanita spp., there have been relatively few analyses of the toxic composition of Lepiota and none using modern high resolution methods. To date, chemical studies of Lepiota species have been restricted to thin layer chromatography (TLC), which has poor resolution and relies on nonspecific visualization reagents for identification, and the Meixner test. The Meixner test is a qualitative assay developed in 1979 for amatoxins that involves blotting a sample onto paper and addition of concentrated hydrochloric acid [13]. The formation of a blue color upon acid treatment is indicative of amatoxins; however, the method suffers from a high rate of false positives from 19 reactions with other compounds such as substituted indoles, and is no longer considered a reliable assay for amatoxin identification [14,15]. To redress the relative scarcity of information regarding the distribution and abundance of amatoxins and phallotoxins in the clinically significant Lepiota genus, five species of Lepiota were analyzed for toxin content by liquid chromatography-mass spectrometry (LCMS). Figure 2.2: Mushroom Species Analyzed in this Study for Toxin Content. A, Amanita phalloides. B, A. virosa. C, Galerina marginata. D, Lepiota josserandii. All photographs reprinted with permission from Mykoweb (http://www.mykoweb.com). Copyright © 1996-2016, Michael Wood and Fred Stevens. 20 2.3 Methods 2.3.1 Mushroom Collection and Identification. Lepiota brunneoincarnata, L. clypeolaria, L. cristata, L. echinacea, and L. josserandii mushrooms were collected in the Lombardy region of Italy during the period of May 2012 through November 2013 by Sara Epis and Davide Sassera (Department of Veterinary Sciences and Public Health, University of Milan), including multiple isolates from different locations. For comparison, specimens of Amanita phalloides and A. virosa were also collected from Italy and California, USA. All mushrooms were morphologically identified by local expert mycologists with standard taxonomic keys. Galerina marginata was obtained from Centraalbureau voor Schimmelcultures (CBS), Utrecht, Netherlands (catalog number 339.88) and laboratory grown as described by Muraoka and Shinozawa (2000) [16]. All specimens were freeze-dried or dried at room temperature and then stored at -80˚C. Lepiota species identifications were confirmed by sequencing of the internal transcribed spacer (ITS) regions. ITS regions were amplified using primer pairs ITS1 and ITS4 [17]. For template preparation, approximately 1 mg of dried mushroom was homogenized with a tissue grinder in 50 μL of lysis buffer as described in Al Shahni et al. (2009) [18]. The samples were centrifuged at 15,000 x g in a microfuge (Eppendorf 5415D) for 2 min and 1 μL of the supernatant was used as the PCR template. PCR was performed under standard conditions using RedTaq polymerase (Sigma, St. Louis, MO) in a total reaction volume of 20 μL. The DNA products of the reaction were cloned into pGEM-T-Easy vector (Promega, Madison, WI) and sequenced by Sanger technology. Sequences (Figure S2.1) were compared to nucleotide sequences in GenBank. 21 2.3.2 Toxin Extraction and LCMS. The dried fungal tissues were frozen in liquid nitrogen, ground with a mortar and pestle, and suspended in methanol: H2O:0.01 M HCl, 5:4:1 at a concentration of 10 mL/g tissue [19]. Following a one hour incubation at room temperature, the extracts were centrifuged at 10,000 x g for 10 min, and the supernatants were filtered through a 0.22 μm filter (Millex polyvinylidene fluoride, GV4, Thermo Fisher Scientific, Waltham, MA). Samples were stored at -80˚C until analysis. Immediately prior to HPLC fractionation, the extracts were diluted with 20 mM ammonium acetate, pH 5, to a concentration of 20 mg dry weight/mL. The fungal extracts were separated on a reversed-phase Proto 300 C18 column (Higgins Analytical; 5 μm, 250 x 4.6 mm) using an Agilent series 1200 HPLC equipped with a multiwavelength detector. Solvent A was 0.02 M ammonium acetate, pH 5, and solvent B was acetonitrile. Toxins were separated with a stepwise gradient of 10% B for 4 min, 18% B for 6 min, and then a linear gradient from 18% B to 100% B over 20 min at a constant flow rate of 1 mL/min [19]. In each run, the equivalent of 0.6 mg of tissue was injected in a volume of 30 μL, except for the G. marginata extract, for which 3 mg was injected. Mass analysis of the eluate was performed with an Agilent 6120 single quadrupole mass spectrometer in positive ion mode. Ions were generated by electrospray with a capillary voltage setting of 5 kV, a drying gas (nitrogen) temperature of 350˚C, and flow rate of 12 L/min. UV absorbance of the eluate was monitored at 280, 295, and 305 nm, because amatoxins and phallotoxins exhibit an absorbance maximum (λmax) at 295 nm due to the presence of tryptathionine, and the presence of 6-hydroxytryptophan shifts the λmax to 305 nm [4,20]. Quantitation of α-amanitin was based on absorbance at 305 nm and an external standard curve of commercial α-amanitin (Sigma, St. Louis, MO). 22 2.4 Results 2.4.1 Toxins in Amanita and Galerina Mushrooms. The toxin profiles of Galerina marginata, Amanita phalloides, and A. virosa are well characterized and remarkably consistent among reported analyses [4,19,21,22]. Since no standards are commercially available for the majority of amatoxins and phallotoxins, extracts of these mushrooms were analyzed as a benchmark and source of standards for which mass, UV absorbance, and retention times could be compared. G. marginata produces only α-amanitin, β-amanitin, and γ-amanitin in significant quantities [19,23,24], and our extracts contained three prominent peaks with masses corresponding to these compounds. A. phalloides is known to contain significant levels of α-amanitin, β-amanitin, amanin, phallacidin, phallisacin, phallisin, and phalloidin. Using the same separation method as Enjalbert et al. (1992) [19], all seven compounds were observed in A. phalloides extracts with the expected elution order, nominal masses, and absorbance maxima of 305 nm for compounds containing both tryptathionine and 6-hydroxytryptophan or 295 nm for compounds containing only tryptathionine. As reported by Smith et al. (2012) [21], an apparent phallisin analogue (referred to as ‘phallisin II’) with the same mass and UV absorbance as phallisin was also present in the European A. phalloides isolate. A. virosa is unique among other poisonous mushrooms in that it lacks β-amanitin and contains amaninamide, which is structurally similar to α-amanitin but lacks 6-hydroxytryptophan [25]. Amaninamide was detected in the A. virosa extract along with α-amanitin, phallisin II, phallacidin, and phalloidin. β-amanitin was absent as expected. Both Amanita species analyzed contained several additional compounds (compounds 11 through 14) that are suspected to be uncharacterized amatoxins or phallotoxins based on mass range and UV absorbance profiles. 23 2.4.2 Toxins in Lepiota Mushrooms. Lepiota brunneoincarnata and L. josserandii are common Lepiota spp. associated with hospitalizations, and in agreement with the reported clincal features and symptoms of ingestion, amatoxins were detected in both species. On the basis of mass, UV absorbance, and column retention time, all isolates of L. brunneoincarnata contained α-amanitin and β-amanitin, and the corresponding analogues lacking 6-hydroxytryptophan, amaninamide and amanin. In L. josserandii, all isolates contained only α-amanitin and γ-amanitin. As in Galerina, neither species contained phallotoxins. While no case reports identify them specifically, L. clypeolaria, L. cristata, and L. echinacea are often listed as poisonous, however, no amatoxins or phallotoxins were detected in all isolates of these species. Peak Number Compound Expected Mass (Da) Observed Masses (m/z) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 β-amanitin α-amanitin amanin phallisacin γ-amanitin phallisin II amaninamide phallacidin phallisin phalloidin unknown unknown unknown unknown 919.338182 918.354170 903.343267 865.316720 902.359252 804.311240 902.359252 846.321804 804.311240 788.316330 x x x x 920.3 [M+H], 942.4 [M+Na], 958.4 [M+K} 919.3 [M+H], 941.2 [M+Na], 957.2 [M+K} 904.3 [M+H], 926.3 [M+Na], 942.2 [M+K} 863.3 [M+H], 885.3 [M+Na], 901.2 [M+K} 903.4 [M+H], 925.4 [M+Na], 941.3 [M+K} 805.3 [M+H], 827.3 [M+Na], 843.2 [M+K} 903.3 [M+H], 925.3 [M+Na], 941.2 [M+K} 847.3 [M+H], 869.3 [M+Na], 885.3 [M+K} 805.4 [M+H], 827.3 [M+Na], 843.3 [M+K} 789.3 [M+H], 811.3 [M+Na], 827.3 [M+K} 789.2, 811.3, 827.2, 848.3, 889.3, 911.3, 927.2 872.5 , 893.4, 914.5 915.4, 937.4, 953.3, 960.6, 974.4 755.3, 795.3, 811.2, 832.4, 869.5, 891.5 Table 2.1: Compounds Identified in Extracts of Amanita, Galerina, and Lepiota Mushrooms. Compounds are numbered in order of elution time. Observed masses are monoisotopic from singly charged ions. Table 2.1 lists all of the compounds identified in these studies within mushroom extracts and the toxin profiles are compared in Figure 2.3. While the levels of α-amanitin in L. brunneoincarnata 24 (on average, 0.76 mg per gram of dry tissue) were comparable to those in Amanita mushrooms, the L. josserandii isolates contained an average of 4.2 mg α-amanitin per gram of dry weight, which is more than three times higher than Amanita spp. and the highest reported levels of the toxin to date (Table 2.2). Figure 2.3: Toxin Profiles of Amanita, Galerina, and Lepiota Mushrooms. Signals are overlaid UV absorbances at 295 nm (blue) and 305 nm (red). The identities and observed masses for each peak are listed in Table 2.1. Peaks are labeled in order of retention time and shared numbers between traces indicate the same compound. The shift in retention time for compound 2 in the L. josserandii extract is due to column performance and is within the deviations observed for standards. 25 Species Isolate α-amanitin content (mg/g dry weight) A. phalloides (Europe) A. phalloides (USA) A. virosa G. marginata L. brunneoincarnata L. brunneoincarnata L. josserandii L. josserandii L. josserandii 1 1 1 1 1 2 1 2 3 1.33 0.88 1.39 0.57 0.82 0.69 4.24 4.39 3.99 Table 2.2: α-Amanitin Concentrations in Mushrooms. Concentrations were calculated using absorbance at 305 nm and a standard curve of α-amanitin. 26 2.5 Discussion This work details the first high-resolution analysis of cyclic peptide toxins in Lepiota mushrooms. Structural identifications were made on the basis of mass, UV absorbance (including diagnostic differences in λmax), and comparisons to extracts of other mushrooms with well-defined toxin profiles. The results indicate that Lepiota brunneoincarnata and L. josserandii produce amatoxins. Based on α-amanitin quantitation, L. josserandii is over three times more toxic than Amanita species, and ingestion of a single fruiting body could prove fatal. No amatoxins or phallotoxins were observed in extracts of L. clypeolaria, L. cristata, or L. echinacea; however, Lepiota species can be difficult to identify without molecular tools, and therefore none should be considered edible. Two additional Lepiota species, L. chlorophyllum [9] and L. helveola [8], have been specified in case reports as the cause of mushroom poisoning. While their toxin content remains to be determined, their toxicity is likely due to the presence of amatoxins since these compounds have now been confirmed in other Lepiota species. Further characterization and genomic studies of the amatoxin-producing Lepiota species identified in this work may provide important insights into the biosynthesis of amatoxins and other cyclic peptides in mushrooms. 27 APPENDIX 28 APPENDIX Lepiota brunneoincarnata TCCGTAGGTGAACCTGCGGAAGGATCATTATTGAATAAACTTGGTGGGTTGTTGCTGGCTTC TTGGAGCATGTGCACGCTCATCGACTTTATCCATCCACCTGTGCACCTTCTGTAGTCTTCGAA ATGAAAGCGGCTGAGCCTCGATGGGCATTTTGCCCTATCGGATGTGAGGAATGCTTTTGTGA AGGCATGGCTCTCCTCAAAGGCCTGTGATCGTTTCTTGGACTATGTTTTTCCATATACCACAT AGCATGTTGTAGAATGTATCGGTGGGCCTCTGTGCCTATAGAACTCAATACAACTTTCAGCA ACGGATCTCTTGGCTCTCGCATCGATGAAGAACGCAGCGAAATGCGATAAGTAATGTGAATT GCAGAATTCAGTGAATCATCGAATCTTTGAACGCACCTTGCGCTCCTTGGTATTCCGAGGAG CATGCCTGTTTGAGTGTCATTTAATTCTCAACCATGCTGGCTTTGTAAAGGTCAGTTGTGGCT TGGATTGTGGGGGTATTCCTGCGGGTCTCTCTTGAGGTCGGCTCCCCTAAAATGCATTAGCA GAACCGTTTGCGGTCAGTCGCAGGTGTGATAATTATCTACGCCAAAGACCAAGGCTGCTCTC TGTTTGTTCAGCTTCTAATTGTCTCGGGACAAATTTTTTTGAATGTTTGACCTCAAATCAGGT AGGACTACCCGCTGAACTTAAGCATATCAATAAGCGGAGGA Lepiota clypeolaria (synonym L. magnispora) TCCGTAGGTGAACCTGCGGAAGGATCATTATTGAATAACTATGGTGGGTTGTTGCTGGCTTC TTGAAGCATGTGCACACCTGCTGTCTTTATCTATCCCACTGTGCACCATTTGTAGTCTTGGAG GGGGAAGAGCGGTGAAGCTCACATGCCCCCCCTTCCGGGTCTATGTCTTTTCCACAAACATT GTAGTATGTCACAGAATGTAATCAAAGGGTCTTTGTGCCCATAAAACTATATACAACTTTCA GCAACGGATCTCTTGGCTCTCGCATCGATGAAGAACGCAGCGAAATGCGATAAGTAATGTG AATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACCTTGCGCTTCTTGGTATTCCGAG GAGCATGCCTGTTTGAGTGTCATTAAATTCTCAATCCCTTCCAGTATTCTGGTTGTGGCTTGG ATATTGGGGGTTTCTGCAGGCCTTATTATGTTGAGGTCAGCTCCCCTAAAATACATTAGCAG AACTGTTTGCGGTCTGTCACTGGTGTGATAATTATCTGCACCAAGGCTGCTTTCTATCTTGTT CAGCTTCCAACCGTCTTCTTGGAGACAACTATTGAACATTTGACCTCAAATCAGGTAGGACT ACCCGCTGAACTTAAGCATATCAATAAGCGGAGGA Figure S2.1: ITS Sequences from Species of Lepiota Mushrooms. 29 Figure S2.1 (cont’d) Lepiota cristata TCCGTAGGTGAACCTGCGGAAGGATCATTATTGAATAAACTTGGTAGGTTGTAGCTGGCTTT TCGAAGCATGTGCACGCCTACTATCTTTATCCATCCACCTGTGCACCCTTTGTAGTCTTGGAG GACAAGAGCGGCTGACTCCTCGAACGGCTTCTTCTAGCCTTTCGGATGTGAGGGATGCTGTG TGAAAGCACRGCTCTCCTCAATGGCTCGCAATTTCCTCTAGGTCTATGTCTTTTCCATATACC ACATAGTATGTTGTAGAATGCATTATATGGGCCCATGTGCCTATAAAACTCAATACAACTTT CAGCAACGGATCTCTTGGCTCTCGCATCGATGAAGAACGCAGCGAAATGCGATAAGTAATG TGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACCTTGCGCTCCTTGGTATTCCG AGGAGCATGCCTGTTTGAGTGTCACTAAATTCTCAACCACTCCAGCCTTTGCGGGTTGGATG TGGCTTGGATGTTGGGGGTTTCTGCGGGCCTCTCTTTTGAGGTCGGCTCCCCTGAAATGCATT AGCGGAACCGTTTGCGGTCCGTCGCCGGTGTGATAATTATCTACGCCATAGACGAAGGCTGC TCTCTGTATGTTCAGCTTCTAACTGTCCCCTGTGGACAACTTTTTGAACGTTTGACCTCAAAT CAGGYAGGACTACCCGCTGAACTTAAGCATATCAATAAGCGGAGGA Lepiota echinacea TCCGTAGGTGAACCTGCGGAAGGATCATTATTGAATAAACCTGGTGGGCTGTAGCTGGCTCT TCGGAGCATGTGCACRCTCATCCACTTTTATCCATCCACCTGTGCACCATGTGTAGTCTTGGG GGAGAAAGATTTGCGGTCCCGCTGTgGGCTTGTGAAGACGTCCTCTCAATTCTATGTTTTTCA TATACCACRTAGTATGTTGCAGAATGTAtATAACGGGCCTATGTGCCTATAAAACACAATAC AACTTTCAGCAACGGATCTCTTGGCTCTCGCATCGATGAAGAACGCAGCGAAATGCGATAAG TAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACCTTGCGCTCCTTGGT ATTCCGAGGAGCATGCCTGTTTGAGTGTCATTATATTCTCAACCCTTCCCAGTTWTAATGACT TGGGTAAGTGGATTGGATTGTGGGGGCTTGCTGGTCGCTTTACTGCGGTCGGCTCCTCTGAA ATGTATTAGCGGAACTGTTTGCGGTCcCGTCACTGGTGTGATAATTATCTACGCCGAAGACG AAGGCTGCTCTCTATACGTTCAGCTTATAATCAGTCCCCTcTGGtGGACAACTTTTGAAAGTTT GACCTCAAATCAGGTAGGACTACCCGCTGAACTTAAGCATATCAATAAGCGGAGGA Lepiota josserandii (synonym L. subincarnata) TCCGTAGGTGAACCTGCGGAAGGATCATTATTGAATAAACATGGTGGGTTGTCGCTGGCTCC TTGGAGCATGTGCACGCTCATCGTCTTTATCCATCCACCTGTGCACCTTTTGTAGTCTTGGGA AATGAATGCAATGGAACCTCGATAGGTTTTTCAGCCTTTCGGATGTGAGGAATGCTTTGTGA AAGCATGGCTCTTCTCAATAGCCTTGCAATCGTTACTCAGACTATGTTTTTCATACACCATGT AGTATGTTTGCAGAATGTATCAATGGGCCTCTGTGCCTATAAAACTCAATACAACTTTCAGC AACGGATCTCTTGGCTCTCGCATCGATGAAGAACGCAGCGAAATGCGATAMGTAATGTGAA TTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACCTTGCGCTCCTTGGTATTCCGAGG AGCATGCCTGTTTGAGTGTCATTTAATTCTCAACCACAAAGGCTTTTGCGAGCTTTTGTGGAT TGGACGTGGGGGTAACTGCAGGCCTTCCCAGGTCAGCTCCCCTAAAATGCATTAGCGGAACC GTTTGCGGTAACCAGTCGCCAGGTGTGATAATTATCTACGCCAATAGACATGAACTGCTCTC TGTTGTTCTGCTTCAAATTGTCTTGCTAGACAACTTTTGAATGTTTGACCTCAAATCAGGTAG GACTACCCGCTGAACTTAAGCATATCAATAAGCGGAGGA 30 WORKS CITED 31 WORKS CITED 1. Breskinsky A, and Besl H. (1990). A Colour Atlas of Poisonous Fungi; translated by Bisset NG. Wolfe: London, UK. 2. Novello F, Fiume L, and Stirpe F. (1970). Inhibition by α-amanitin of ribonucleic acid polymerase solubilized from rat liver nuclei. Biochem. J. 116(2): 177-180. 3. Ward J, Kapadia K, Brush E, and Salhanick SD. (2013). Amatoxin poisoning: case reports and review of current therapies. J. Emerg. Med. 44(1): 116-121. 4. Wieland, T. (1986). Peptides of Poisonous Amanita Mushrooms. Springer: New York. 5. Lengsfield AM, Löw I, Wieland T, Dancker P, and Hasselbach W. (1974). Interaction of phalloidin with actin. Proc. Natl. Acad. Sci. U.S.A. 71(7): 2803-2807. 6. Haines JH, Lichstein E, and Glickerman D. (1986). A fatal poisoning from an amatoxin containing Lepiota. Mycopathologia 93(1): 15-17. 7. Paydas S, Kocak R, Erturk F, Erken E, Zaksu HS, and Gurcay A. (1990). Poisoning due to amatoxin-containing Lepiota species. Br. J. Clin. Pract. 44(11): 450-453. 8. Ramirez P, Parrilla P, Sanchez Bueno F, Robles R, Pons JA, Bixquert V, Nicolas S, Nunez R, Alegria MS, Miras M, et al. (1993). Fulminant hepatic failure after Lepiota mushroom poisoning. J. Hepatol. 19(1): 51-54. 9. Feinfeld DA, Mofenson HC, Caraccio T, and Kee M. (1994). Poisoning by amatoxincontaining mushrooms in suburban New York – report of four cases. J. Toxicol. Clin. Toxicol. 32(6): 715-721. 10. Meunier BC, Camus CM, Houssin DP, Messner MJ, Gerault AM, and Launois BG. (1995). Liver transplantation after severe poisoning due to amatoxin-containing Lepiota – Report of three cases. J. Toxicol. Clin. Toxicol. 33(2): 165-171. 11. Mottram AR, Lazio MP, and Bryant SM. (2010). Lepiota subincarnata J.E. Lange induced fulminant hepatic failure presenting with pancreatitis. J. Med. Toxicol. 6(2): 155-157. 12. Kervégant M, de Haro L, Patat AM, Pons C, Thomachot L, and Minodier P. (2013). Phalloides syndrome poisoning after ingestion of Lepiota mushrooms. Wilderness Environ. Med. 24(2): 170-172. 13. Meixner A. (1979). Amatoxin-Nachweiss in Pilzen. Z. Mykol. 45(1): 137-139. 32 14. Beutler JA, and Vergeer PP. (1980). Amatoxins in American Mushrooms: Evaluation of the Meixner Test. Mycologia 72(6): 1142-1149. 15. Beuhler M, Lee DC, and Gerkin R. (2004). The Meixner test in the detection of alphaamanitin and false-positive reactions caused by psilocin and 5-substituted tryptamines. Ann. Emerg. Med. 44(2): 114-120. 16. Muraoka S, and Shinozawa T. (2000). Effective production of amanitins by two-step cultivation of the basidiomycete, Galerina fasciculata GF-060. J. Biosci. Bioeng. 89(1): 73-76. 17. White TJ, Bruns T, Lee S, and Taylor J. (1990). Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenies. In PCR Protocols: A Guide to Methods and Applications; Innis MA, Gelfand DH, Sninsky JJ, and White TJ, Eds.; Academic: San Diego, CA, USA; pp. 315-322. 18. Al Shahni, MM, Makimura K, Yamada T, Satoh K, Ishihara K, Sawada T. (2009). Direct colony PCR of several medically important fungi using Ampdirect Plus. Jpn. J. Infect. Dis. 62(2): 164-167. 19. Enjalbert F, Gallion C, Jel F, Monsteil H, and Faulstich HJ. (1992). Simultaneous assay for amatoxins and phallotoxins in Amanita phalloides Fr., by high-performance liquid chromatography. J. Chromatogr. 598(2): 227-236. 20. May JP, and Perrin DM. (2007). Tryptathionine bridges in peptide synthesis. Biopolymers 88(5): 714-724. 21. Clarke DB, Llyod AS, and Robb P. (2012). Application of liquid chromatography coupled to time-of-flight mass spectrometry separation for rapid assessment of toxins in Amanita mushrooms. Anal. Meth. 4(5): 1298-1309. 22. Jansson D, Fredriksson SA, Herrmann A, and Nilsson C. (2012). A concept study on idenfication and attribution profiling of chemical threat agents using liquid chromatography-mass spectrometry applied to Amanita toxins in food. Forensic Sci. Int. 221(1-3): 44-49. 23. Benedict RG, Tyler VE Jr, Brady LR, and Weber LJ. (1966). Fermentative production of amanita toxins by a strain of Galerina marginata. J. Bacteriol. 91(3): 1380-1381. 24. Muraoka S, Fukamachi N, Mizomoto K, and Shinozawa T. (1999). Detection and identification of amanitins in the wood-rotting fingi Galerina fasciculata and Galerina helvoliceps. Appl. Environ. Microbiol. 65(9): 4207-4210. 25. Buku A, Wieland T, Bodenmuller H, and Faulstich H. (1980). Amaninamide, a new toxin of Amanita virosa mushrooms. Experientia 36(1): 33-34. 33 CHAPTER 3 GENOMIC CAPACITY FOR CYCLOAMANIDE BIOSYNTHESIS IN AMANITA MUSHROOMS Note: The content in this chapter has been previously published. Some text has been modified from the original. Copyright © 2016 by the authors. Citation: Pulman JA, Childs KL, Sgambelluri RM, and Walton JD. (2016). Expansion and diversification of the MSDIN family of cyclic peptide genes in the poisonous agarics Amanita phalloides and A. bisporigera. BMC Genomics. 17(1): 1038. Author Contributions: Assembly and annotation of fungal genomes and transcriptomes was done by Jane A. Pulman and Kevin L. Childs. Field collection of fungal specimens and manual annotation of MSDIN sequences was done by Jonathan D. Walton. 34 3.1 Abstract Cycloamanides are cyclic peptide natural products found in Amanita, Galerina, and Lepiota mushrooms and are produced from ribosomally-synthesized precursors. The precursor peptides are encoded by the MSDIN gene family and are composed of conserved N- and C-terminal domains and an internal hypervariable core domain containing the amino acids found in the final peptides. While some MSDIN genes have been identified in genome surveys of Amanita species, the full complement of MSDIN genes in a single species has yet to be reported. Draft genome sequences were obtained for Amanita bisporigera and A. phalloides mushrooms and 31 MSDIN genes were identified in the genome of A. bisporigera and 33 in A. phalloides, with a combined total of 51 unique core domain sequences. RNAseq analysis of A. bisporigera confirmed expression of 19 MSDIN sequences. Extracts of A. phalloides were searched for novel cyclic peptides based on their expected masses and two new compounds, named cycloamanide E and cycloamanide F, were demonstrated by LC/MS/MS. A. bisporigera and A. phalloides together have the genetic capacity to synthesize at least 51 cycloamanides. 35 3.2 Introduction The cycloamanide family of cyclic peptides produced by mushrooms includes the amatoxins, phallotoxins, virotoxins, and other compounds including cycloamanides A through D (CyalA-D), antamanide, and amanexitide [1,2,3,4]. Known bioactivities among the cycloamanides include RNA polymerase II inhibition [5,6], actin binding and stabilization [7,8,9], immunosuppression [10,11], and inhibition of the mammalian liver transporter OATP [12]. Cycloamanides are biosynthesized from precursor peptides encoded by the MSIDN gene family. The precursors are composed of a conserved N-terminal leader peptide domain, a hypervariable core region containing the amino acid sequences of the cyclic peptides, and a conserved C-terminal domain [13]. Discovery of the MSDIN gene family in genome surveys led to identification of 15 unique MSDIN genes in Amanita bisporigera and 4 in other species, suggesting an extensive gene family that gives rise to a large number of natural products [13]. Galerina mushrooms are also known to produce amatoxins [14]; however, only two MSDIN sequences, both encoding αamanitin, were found in the complete genome of G. marginata [15]. The Amanita species A. exitialis, A. fuliginea, A. fuligineoides, A. pallidorosea, A. phalloides, and A. rimosa have also been searched for MSDINs by RNAseq [16] and PCR [17], and 42 MSDIN sequences with 28 unique core domains were found. To date, a total of 36 unique MSDIN sequences have been identified and are predicted to encode natural products. However, studies have been limited to incomplete genome and transcriptome surveys and PCR, and therefore the full complement of MSDIN genes and capacity for cyclic peptide biosynthesis has yet to be determined for a single species. 36 3.3 Methods 3.3.1 Genomics and Transcriptomics. Individual basidiocarps of Amanita bisporigera (Ab) and A. phalloides (Ap) were collected in Ingham County, Michigan, in the summer of 2010, and in Alameda County, California in the winter of 2011, respectively, by Jonathan D. Walton (Department of Plant Biology and Department of Energy-Plant Research Laboratory, Michigan State University). Genomic DNA and total RNA were isolated by organic solvent extraction using cetyltrimethyl ammonium bromide (CTAB), phenol, and chloroform. DNA from each species was sequenced using Illumina MiSeq technology, and RNA from Ab was reversetranscribed and sequenced using Illumina HiSeq. Sequencing was performed by the Michigan State University RTSF Genomics Facility. Assembly and annotation of the Ab and Ap genomes and Ab transcriptome was performed by Jane A. Pulman and Kevin L. Childs (Department of Plant Biology and Center for GenomicsEnabled Plant Science, Michigan State University). High-quality reads from Ab and Ap were selected using Trimmomatic (ver 0.32) [18] and assembled using Velvet (ver 1.2.10) [19]. Gene structural annotations were made using the MAKER pipeline [20,21] and functional annotations using Trinotate (ver 2.0.2) [22]. MSDIN genes were identified by Jonathan D. Walton within assemblies using tblastn with the conserved leader peptide sequence (MSDINATRLP) as query and an e-value cutoff set to 100. Annotations of MSDIN genes were accomplished manually with the aid of MAKER-predicted gene models and protein and transcript alignments with known MSDIN genes. 3.3.2 LC/MS/MS of Predicted Cycloamanides. A lyophilized basidiocarp of Amanita phalloides was ground to a powder in liquid nitrogen and resuspended in 90% ethanol at a 37 concentration of 1 g/50 mL. After stirring for 1 hr at room temperature, the ethanol was removed under vacuum and the resulting residue was dissolved in a water/chloroform (1:1) solution. The aqueous layer was collected and dried under vacuum and the residual oil was redissolved in 50% acetonitrile. This extract was analyzed using a Waters Xevo G2-XS QtoF HPLC/MS/MS system with a 5 uL injection onto a BEH C18 UPLC column (2.1 mm x 50 mm, 1.7 µm particle size; Waters). The column temperature was maintained at 30°C and the flow rate at 0.3 mL/min. Separation was performed with 10 mM ammonium formate in water (solvent A) and acetonitrile (solvent B) with an initial hold at 5% solvent B for 3 min followed by a linear gradient to 99% solvent B over 27 min. The MS settings were electrospray ionization (ESI) in positive mode, 3 kV capillary voltage, 100°C source temperature, 350°C desolvation temperature, 600 L/hr desolvation nitrogen gas flow, and 35 V cone voltage. Data were acquired using an MSe method having two separate acquisition functions, where function 1 was performed with no collision energy and function 2 was performed with a collision energy ramp from 60-100 V. For both functions, the scan range was 50-1500 m/z with a scan rate of 0.2 seconds per function. Data were analyzed using Masslynx (ver 4.1) (Waters) and mMass (ver 5.5.0) [23]. 38 3.4 Results 3.4.1 MSDIN Genes in Amanita bisporigera and A. phalloides. The Ab genome contained 23,572 contigs assembled into 10,390 scaffolds and a total assembly size of 75 Mb with 74X predicted fold coverage. The genome of Ap contained 5,437 contigs assembled into 1,465 scaffolds for an assembly size of 40 Mb and 69X predicted fold coverage. Identification of MSDIN genes required tblastn searches and manual annotation, since none were annotated by the MAKER tool even after the minimum length parameter was reduced to 150 base pairs. A total of 64 MSDIN genes with 51 unique core domain sequences were identified in the Ab and Ap genomes (Table 3.1). Ab contained 31 with 26 unique core domains and Ap contained 33 with 28 unique core domains. Expression at the level of RNA transcript was confirmed for 19 of the unique sequences in Ab by RNAseq (Table S3.1). Only three core domain sequences were common to both genomes, the α-amanitin (IWGIGCNP) and phalloidin (AWLVDCP) sequences and ISDPTAYP. Of the 15 MSDIN sequences that were previously identified in genome surveys of Ab [13], only 6 were present in our isolate. Similarly, genes encoding several cycloamanides previously isolated from Ap (CylA, CylC, CylD and antamanide) [1] were absent in our Ap isolate, suggesting significant intraspecies diversity in the gene family. 3.4.2 New Cycloamanides in Amanita phalloides. An extract of A. phalloides was searched for new cycloamanides using extracted ion chromatograms for the predicted masses of the head-totail cyclic, but otherwise unmodified peptides based on the genomic MSDIN sequences. Extracts of Ap contained two compounds with masses corresponding to the cyclic versions of two MSDIN core domain sequences, SFFFPVP and IVGILGLP. High resolution measurements indicated a m/z of 822.4216 for putative cyclo-SFFFPVP (C45H56N7O8, calculated m/z 822.4190, 39 3.8 ppm error) and a m/z of 763.5118 for cyclo-IVGILGLP (C38H67N8O8, calculated m/z 763.5076, 5.5 ppm error). MS/MS confirmed the sequence of each compound, and by the presence of unambiguous, overlapping fragments that span the entire sequence, the compounds could be deduced as cyclic. The compounds were named cycloamanide E (SFFFPVP) (Figure 3.1) and cycloamanide F (IVGILGLP) (Figure 3.2). A. bisporigera A. phalloides AWLAECP AWLVDCP CIGFLGIP FFWPIIIPP FIWVLWLWLL FNFFRFPYP FSVLSIIPP GLGLIP GLPIIAIIP GLPMVLP GMDPPSPMP GMEPPSPMP IFWPIFAP IFWYIYFP IGRPQLLP IIFEPIIP ILMLAIPP ISDPTAYP IVFLEFYS IWGIGCNP IWWYIYFP LFFPPDFRPP LFYPPDFRPP LSSPMLLP MAFPEFLA MIQRPFYP AWLATCP AWLVDCP FFFPPFFIPP FFPIVFSPP FIFPPFIIPP FMPLAP FNILPFMLPP FNLFRPYP GPVFFAY HFASFIPP IFLAFPIPP IFWFIYFP IILAPIIP IRLPPLFLPP ISDPTAYP IVGILGLP IWGIGCDP IWGIGCNP LFFWFWFLWP LGRPESLP LILLAALGIP LIQRPFAP LPVLPIPLLP LRLPPFMIPP SFFFPIP SFFFPVP TIYYLYFIP VQKPWSRP Table 3.1: Cycloamanide Sequences in the Genomes of A. bisporigera and A. phalloides. 40 A B Peak 1 2 3 3 3 4 4 5 6 7 8 9 10 10 11 12 13 13 13 Ion M b6 b6 b6 b6 b5 b5 b5 b4 b3 b3 b3 b2 b2 b2 b2 im4 im3 im2 Meas. m/z 822.4216 723.3519 675.3496 675.3496 675.3496 626.2978 626.2978 578.2992 479.2286 392.1957 344.1951 332.1603 295.1434 295.1434 245.1277 185.0903 120.0796 120.0796 120.0796 Calc. m/z 822.4185 723.3501 675.3501 675.3501 675.3501 626.2973 626.2973 578.2973 479.2289 392.1969 344.1969 332.1605 295.1441 295.1441 245.1285 185.0921 120.0808 120.0808 120.0808 δ (Da) 0.0031 0.0018 -0.0005 -0.0005 -0.0005 0.0005 0.0005 0.0019 -0.0003 -0.0012 -0.0018 -0.0002 -0.0007 -0.0007 -0.0008 -0.0018 -0.0012 -0.0012 -0.0012 δ (ppm) 3.8 2.5 -0.7 -0.7 -0.7 0.8 0.8 3.3 -0.6 -3.1 -5.2 -0.6 -2.4 -2.4 -3.3 -9.7 -10.0 -10.0 -10.0 Sequence SFFFPVP PSFFFP PVPSFF FPVPSF FFPVPS SFFFP PSFFF VPSFF PSFF FFP FPV PSF FF FF FP PS F F F C Fragment # M ………………......…. V P S F F F P V 5 …………….. V P S F F 6 ……………..…..…. P S F F 12 …………..…...…. P S 9 …………..…….…. P S F 2 …………..……….…. P S F F P 10 ………………….………………...……...…. F F 7 ……………...…...………………………. F F P 11 ………………..……………………………………...…. F P 8 ………..…………………………………………... F P V Figure 3.1: MS/MS Analysis of Cycloamanide E. A, MS/MS spectrum. B, Peak list. Peaks with more than one entry correspond to fragments with more than one possible sequence. C, Overlapping fragments indicating a cyclic structure. The highlighted valine is the same residue in each sequence. 41 A B Peak 1 2 2 2 2 3 3 3 4 5 5 5 6 7 8 9 10 10 Ion M b7 b7 b7 b7 b6 b6 b6 b6 b5 b5 b5 b5 b4 b3 b3 b2 b2 Meas. m/z 763.5118 650.4256 650.4256 650.4256 650.4256 593.4048 593.4048 593.4048 537.3364 480.3174 480.3174 480.3174 440.2896 367.2352 310.2145 268.1665 211.1453 211.1453 Calc. m/z 763.5076 650.4236 650.4236 650.4236 650.4236 593.4021 593.4021 593.4021 537.3395 480.3180 480.3180 480.3180 440.2867 367.2340 310.2125 268.1656 211.1441 211.1441 δ (Da) 0.0042 0.002 0.002 0.002 0.002 0.0027 0.0027 0.0027 -0.0031 -0.0006 -0.0006 -0.0006 0.0029 0.0012 0.002 0.0009 0.0012 0.0012 δ (ppm) 5.5 3.1 3.1 3.1 3.1 4.6 4.6 4.6 -5.8 -1.2 -1.2 -1.2 6.6 3.3 6.4 3.4 5.7 5.7 Sequence C IVGILGLP PIVGILG Fragment # GLPIVGI LGLPIVG ………………......…. M G L P I V G VGILGLP …………….. 9 G L P PIVGIL ……………..…..…. 4 G L P I V G LPIVGI LGLPIV ……………………………...…..…...…. 8 P I V GLPIVG ……………………………………...…….…. 7 P I V G PIVGI .………………………………………………………………………..…. 6 V G LPIVG GLPIV VGILG PIVG PIV GLP PI LP I L G I L G Figure 3.2: MS/MS Analysis of Cycloamanide F. A, MS/MS spectrum. B, Peak list. Peaks with more than one entry correspond to fragments with more than one possible sequence C, Overlapping fragments indicating a cyclic structure. The highlighted glycine is the same residue in each sequence. 42 3.5 Discussion Besides Galerina marginata, which only contains two MSDIN sequences [15], this chapter details the first complete assessment of MSDIN sequences in the genome of a single species. Combined with previous studies in other species, a total of 73 MSDIN sequences with unique core domains have been identified to date (Table S3.2). The core domains range from 6 to 10 amino acids in length and all 20 amino acids are represented at least once. Because only two new cycloamanides were found in Ap extracts and not all were expressed in Ab based on RNAseq data, it is possible that not all MSDIN sequences are precursors to functional natural products. However, comparing the amino acid distribution in the core domain sequences to expected frequencies based on the number of codons for each residue reveals that the sequences are not random and that bulky hydrophobic residues are highly overrepresented (Figure 3.3). This indicates a process for genetic selection and suggests functionality in the products. Cycloamanides E and F were identified in extracts as cyclic peptides with unmodified side-chains using predicted masses based on genomic sequence, and other products of the pathway likely eluded detection due to the presence of additional post-translational modifications. Future studies should be aimed at describing these other post-translational modifications, identifying new cycloamanides and their bioactivities, and describing the genetic mechanisms behind the extensive duplication of the MSDIN genes and hypermutation of their core domains. 43 Figure 3.3: Amino Acid Frequencies in MSDIN Core Domain Sequences. Values are observed frequency minus expected frequency (%) for each amino acid, colored by type. 44 APPENDIX 45 APPENDIX AWLAECP AWLVDCP FNFFRFPYP FSVLSIIPP GLPIIAIIP GMDPPSPMP GMEPPSPMP IFWPIFAP IFWYIYFP IGRPQLLP IIFEPIIP ILMLAIPP ISDPTAYP IVFLEFYS IWGIGCNP IWWYIYFP LSSPMLLP MAFPEFLA MIQRPFYP Table S3.1: List of Core Domains Identified Among MSDIN Transcripts from RNAseq of Amanita bisporigera. 46 Core Sequence AWLAECP AWLALCP AWLATCP AWLTDCP AWLVDCP CIGFLGIP FFFPPFFIPP FFPIVFSPP FFQPPEFRPP FFWPIIIPP FIFPPFIIPP FIWVLWLWLL FLFPPVRLPP FMPLAP FNFFRFPYP FNILPFMLPP FNLFRFPYP FSVLSIIPP FVFVASPP FYQFPDFKYP GAYPPVPMP GFVPILFP GLGLIP GLPIIAIIP GLPMVLP GMDPPSPMP GMEPPSPMP GPVFFAY HFASFIPP HLVRYPP HPFPLGLQP IFLAFPIPP IFWFIYFP IFWPIFAP IGRPQLLP IIFEPIIP IIGILLPP IIIVLGLIIP IILAPIIP IIWAPVVP Species bisporigera, rimosa fuligineoides ocreata, phalloides exitialis bisporigera, exitialis, pallidorosea, phalloides bisporigera phalloides phalloides bisporigera bisporigera phalloides bisporigera bisporigera phalloides bisporigera phalloides phalloides bisporigera exitialis bisporigera bisporigera bisporigera bisporigera bisporigera bisporigera bisporigera bisporigera phalloides phalloides fuligineoides bisporigera phalloides exitialis, phalloides bisporigera bisporigera bisporigera phalloides rimosa phalloides exitialis, fuliginea Table S3.2: Alphabetical List of All Unique MSDIN Core Sequences Identified to Date. 47 Table S3.2 (cont’d) Core Sequence ILMLAILP ILMLAIPP IPGLIPLGIP IRLPPLFLPP ISDPTAYP IVFLEFYS IVGILGLP IWGIGCDP IWGIGCNP IWWYIYFP LFFPPDFRPP LFFWFWFLWP LFLPPVRMPP LFYPPDFRPP LGRPESLP LGRPFAP LILLAALGIP LIQRPFAP LLILSILP LPVLPIPLLP LRLPPFMIPP LSSPMLLP MAFPEFLA MIQRPFYP SFFFPIP SFFFPVP TIYYLYFIP VFSLPVFFP VQKPWSRP VWIGCSP VWIGYSP WLATCP YVVFMSFIPP Species bisporigera bisporigera bisporigera phalloides bisporigera, phalloides bisporigera phalloides exitialis, fuliginea, fuligineoides, pallidorosea, phalloides, rimosa bisporigera, exitialis, fuliginea, fuligineoides, pallidorosea, phalloides, rimosa bisporigera bisporigera, exitialis phalloides bisporigera bisporigera phalloides phalloides phalloides phalloides exitialis phalloides phalloides bisporigera bisporigera bisporigera phalloides phalloides phalloides exitialis phalloides fuliginea exitialis, fuligineoides phalloides bisporigera 48 WORKS CITED 49 WORKS CITED 1. Wieland, T. (1986). Peptides of Poisonous Amanita Mushrooms. Springer: New York. 2. Wieland T. (1983). The toxic peptides from Amanita mushrooms. Int. J. Pept. Protein Res. 22(3): 257-276. 3. Wieland T. (1968). The discovery, isolation, elucidation of structure, and synthesis of antamanide. Angew. Chem. Int. Ed. Engl. 7(3): 204-208. 4. Xue JH, Ping W, Chi YL, Xu LX, and Wei XY. (2011). Cyclopeptides from Amanita exitialis. Nat. Prod. Bioprospect. 1(1): 52-56. 5. Novello F, Fiume L, and Stirpe F. (1970). Inhibition by α-amanitin of ribonucleic acid polymerase solubilized from rat liver nuclei. Biochem. J. 116(2): 177-180. 6. Lindell TJ, Weinberg F, Morris PW, Roeder RG, and Rutter WJ. (1970). Specific inhibition of nuclear RNA polymerase II by alpha-amanitin. Science 170(3956): 447-449. 7. Lengsfield AM, Löw I, Wieland T, Dancker P, and Hasselbach W. (1974). Interaction of phalloidin with actin. Proc. Natl. Acad. Sci. U.S.A. 71(7): 2803-2807. 8. Faulstich H, Buku A, Bodenmüller H, and Wieland T. (1980). Virotoxins: actin-binding cyclic peptides of Amanta virosa mushrooms. Biochemistry 19(14): 3334-3343. 9. Gicquaud C, and Paré M.(1992). Virotoxins polymerize actin and induce membrane fragmentation in cytoplasmic preparations of Amoeba proteus. Biochem. Cell Biol. 70(8): 719-723. 10. Wieczorek Z, Siemion IZ, Zimecki M, Bolewska-Pedyczak E, and Wieland T. (1993). Immunosuppressive activity in the series of cycloamanide peptides from mushrooms. Peptides 14(1):1-5. 11. Siemion IZ, Pedyczak A, Trojnar J, Zimecki M, Weiczorek Z. (1992). Immunosuppressive activity of antamanide and some of its analogues. Peptides 13(6): 1233-1237. 12. Letschert K, Faulstich H, Keller D, and Keppler D. (2006). Molecular characterization and inhibition of amanitin uptake into human hepatocytes. Toxicol. Sci. 91(1): 140-149. 13. Hallen HE, Luo H, Scott-Craig JS, and Walton JD. (2007). Gene family encoding the major toxins of lethal Amanita mushrooms. Proc. Natl. Acad. Sci. U.S.A. 104(48): 1909719101. 50 14. Enjalbert F, Cassanas G, Rapior S, Renault C, and Chaumont JP. (2004). Amatoxins in wood-rotting Galerina marginata. Mycologia 96(4): 720-729. 15. Luo H, Hallen-Adams HE, Scott-Craig JS, and Walton JD. (2012). Ribosomal biosynthesis of α-amanitin in Galerina marginata. Fungal Genet. Biol. 49(2): 123-129. 16. Li P, Deng WQ, Li TH, Song B, and Shen YH. (2013). Illumina-based de novo transcriptome sequencing and analysis of Amanita exitialis basidiocarps. Gene 532(1): 63-71. 17. Li P, Deng W, and Li T. (2014). The molecular diversity of toxin gene families in lethal Amanita mushrooms. Toxicon 83(1): 59-68. 18. Bolger AM, Lohse M, and Usadel B. (2014). Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30(15): 2114-2120. 19. Zerbino DR and Birney E. (2008). Velvet: algorithms for de novo short read assembly using de Bruijn graphs. Genome Res. 18(5): 821-829. 20. Canterel BL, Korf I, Robb SMC, Parra G, Ross E, Moore B, Holt C, Alvarado AS, and Yandell M. (2008). MAKER: An easy-to-use annotation pipeline designed for emerging model organism genomes. Genome Res. 18(1): 188-196. 21. Simão FA, Waterhouse RM, Ioannidis P, Kriventseva EV, and Zdobnov EM. (2015). BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics 31(19): 3210-3212. 22. Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, Adiconis X, Fan L, Raychowdhury R, Zeng Q, Chen Z, Mauceli E, Hacohen N, Gnirke A, Rhind N, di Palma F, Birren BW, Nusbaum C, Lindblad-Toh K, Friedman N, and Regev A. (2011). Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat. Biotechnol. 29(7): 644-652. 23. Niedermeyer TH, and Strohalm M. (2012). mMass as a software tool for the annotation of cyclic peptide tandem mass spectra. PLoS One. 7(9): e44913 51 CHAPTER 4 CHARACTERIZATION OF AMANITIN BIOSYNTHESIS IN GALERINA MARGINATA 52 4.1 Abstract Galerina marginata is a saprobic mushroom that produces the ribosomal bicyclic peptide toxin α-amanitin. Unlike most basidiomycetes, G. marginata is culturable and thus may be useful as a model organism for studying the biosynthesis of amanitin and related compounds. α-Amanitin levels were quantified over time in laboratory grown G. marginata mycelium. On average, amanitin production began after 25 days of growth and peaked after 40 days to 1.39 mg per gram of dry tissue. Candidate biosynthetic genes were identified in the G. marginata genome based on genome clustering with the gene encoding the amanitin precursor peptide, GmAMA1, and included a predicted prolyl oligopeptidase (GmPOPB), flavin-containing monooxygenase (GmFMO), and P450 monooxygenase (GmP450-29). G. marginata strains harboring knockouts for the three candidates were developed and the effects on α-amanitin production were assessed by HPLC. Production of the toxin was abolished in all three mutants, suggesting the involvement of these enzymes in the pathway. In the P450-29(-) strain, an intermediate to α-amanitin accumulated that had a mass corresponding to α-amanitin missing two hydroxylations of the amino acid side-chains. NMR spectroscopy of the purified intermediate indicated the absence of hydroxyl groups at the δ-position of Ile1 and the γ-position of Pro8. Expression patterns of the genes known or hypothesized to be involved in the pathway were characterized by RT-PCR as a potential avenue for identifying additional biosynthetic genes by patterns of co-expression. Transcriptional activation of GmAMA1 correlated with the onset of toxin biosynthesis but no correlation with expression of the other biosynthetic genes was observed. 53 4.2 Introduction Cycloamanides such as amatoxins and phallotoxins are known to be synthesized by mushrooms in the Amanita [1], Galerina [2], and Lepiota [3] genera. Culturing of higher fungi in the laboratory is often difficult or unsuccessful due to the complex and poorly understood growth requirements of these organisms. This growth issue remains true for the majority of cycloamanide producers with the exception of Galerina marginata, a saprobic wood-rotting mushroom that is distributed worldwide [4] (Figure 4.1). The ability to culture G. marginata [5] makes this species a potentially useful model organism for studying cycloamanide biosynthesis. The G. marginata genome and transcriptome have previously been fully sequenced and annotated [6] (publicly available at http://jgi.doe.gov) . Unlike Amanita spp., G. marginata is more limited in cycloamanide biosynthesis and only contains two MSDIN genes, GmAMA1-1 and GmAMA1-2, both of which encode the precursor peptide for α-amanitin. The precursor peptides in G. marginata share the same overall structure as those from Amanita, with conserved Figure 4.1: Basidiocarps of Galerina marginata. Reprinted with permission from MykoWeb (http://www.mykoweb.com). Copyright © 1996-2016, Michael Wood. 54 leader and follower peptides and invariable proline residues flanking a core domain, although the leader and follower sequences diverge significantly between Amanita and Galerina (see Figure 1.6). Starting from a 35mer precursor peptide, steps in the biosynthesis of α-amanitin must include proteolysis, four side-chain hydroxylations, a sulfoxidation, tryptathionine formation, and backbone condensation/cyclization (Figure 4.2). While genes in pathways of secondary metabolism are often clustered in fungi [7,8], no conserved cluster is apparent for the cycloamanides. However, each MSDIN in G. marginata is found in close proximity to genes with predicted functions that could be relevant to the pathway. GmAMA1-1 lies just downstream of three genes predicted to encode P450 monooxygenases and adjacent to a predicted prolyl oligopeptidase (POP) and flavin-containing monooxygenase (FMO). GmAMA1-2 also lies adjacent to a predicted P450 monooxygenase (Figure 4.3). The putative monooxygenase enzymes may be responsible for the side-chain hydroxylations seen in α-amanitin. The precursor peptides contain conserved proline residues, and POP enzymes, which hydrolyze peptides at prolines, are likely involved in their processing (see Chapter 5). In addition, two POP genes, POPA and POPB are present in all mushroom species with available genomic data that produce cycloamanides, whereas non-producers only contain POPA [6]. This finding suggests that POPB might play a dedicated role in cycloamanide biosynthesis. The following studies aim to characterize amanitin biosynthesis, including identification of the genes involved in laboratory grown G. marginata. 55 Figure 4.2: Steps in the Biosynthesis of α-Amanitin from the GmAMA1 Precursor Peptide. The order in which the tailoring steps occur is unknown. Figure 4.3: Genes Adjacent to GmAMA1 in the G. marginata Genome with Relevant Predicted Functions. 56 4.3 Methods 4.3.1 Galerina Growth and Toxin Analysis. G. marginata was obtained from Centraalbureau voor Schimmelcultures (CBS), Utrecht, Netherlands (catalog number 339.88) and maintained on potato dextrose agar. For α-amanitin production, a 1x1 cm square of new growth from a potato dextrose agar (PDA) plate was used to inoculate 100 mL of “HSV-5C” medium [5] in a 250 mL flask, and the mycelium was grown with shaking at room temperature. The growth medium contained (per liter) 5 g glucose, 1 g yeast extract, 100 mg NH4Cl, 100 mg KCl, 100 mg CaSO4•2H2O, 1 mg thiamine, and 0.1 mg biotin. The pH of the medium of was adjusted to 5.2 and then autoclaved. The mycelium was harvested from each flask in 1 or 2 day intervals by filtering the culture through miracloth. The mycelium was then frozen and lyophilized, and the dry weight was measured and recorded before storing at -80°C. A total of 17 time points were collected, each in triplicate between 10 days and 50 days of growth. For toxin extraction, the frozen samples were ground in a mortar and pestle, dissolved in methanol:water:0.01 M HCl (5:4:1) at a concentration of 10 mL per gram of tissue, and incubated for 1 hour at room temperature. The extracts were then centrifuged at 10,000 x g for 10 min and the supernatants passed through a 0.22 µm syringe filter. α-Amanitin was quantified by HPLC and an external standard curve of commercial toxin (Sigma-Aldrich). Separation was performed using an Agilent 1200 series HPLC with a multi-wavelength detector and a Proto 300 C18 reverse-phase column (Higgins; 5 µm, 250 x 4.6 mm). Solvent A was 0.02 M ammonium acetate (pH 5) and solvent B was acetonitrile. The HPLC program used was developed by Enjalbert et al., 1992 [9] and was 10% solvent B for 10 min, step to 18% solvent B for 6 min, and then a linear gradient from 18% to 100% solvent B over 20 min at 1 mL/min. For each 57 sample, an equivalent of 3 mg of tissue was analyzed and the area of the absorbance peak corresponding to α-amanitin (based on retention time and UV profile compared to standard and confirmed with ESI-LCMS) was measured at 305 nm. 4.3.2 Galerina Transformation and Gene Knockouts. Targeted gene knockouts in G.marginata were accomplished by Hong Luo (MSU-DOE Plant Research Laboratory, Michigan State University) using an Agrobacterium tumefaciens mediated transformation method developed for the mushroom Laccaria bicolor [10,11]. For each knockout, the T-DNA cassettes contained a hygromycin resistance gene (hph, hygromycin B phosphotransferase) for selection and 1.5 to 1.8 kbp of upstream and downstream genomic sequences for targeted homologous recombination. Knockouts were confirmed in all transformants by PCR and Southern blotting, and the effects on α-amanitin production were assessed in extracts of the transformants by the HPLC method described in section 4.3.1. 4.3.3 Purification of an Amanitin Intermediate and NMR. A pathway intermediate to αamanitin was identified and purified from approximately 3 g (dried) of a G. marginata strain harboring a knockout of a gene predicted to encode a P450 enzyme. The intermediate was purified by reversed-phase HPLC in two steps on a semi-preparative C18 column (25 cm x 10 mm, 5 mm, Supelcosil LC-18). For the first separation, solvent A was 20 mM aqueous ammonium acetate:acetonitrile (90:10, v/v) and solution B was 20 mM ammonium acetate:acetonitrile (76:24, v/v), both adjusted to pH 5 with glacial acetic acid. A step-wise gradient profile was used and consisted of 100% A for 3 min, 43% A for 7 min, and 0% A for 9 min at a constant flow rate of 2 mL/min. The second purification step consisted of a linear gradient of 100% 20 mM ammonium acetate to 100% acetonitrile over 15 min. Both separations 58 were carried out on an Agilent 1200 series HPLC with a multi-wavelength detector. Fractions containing the intermediate were pooled and lyophilized. For NMR experiments, the intermediate was dissolved in DMSO-d6 to a final concentration of 3.8 mM. Spectra were recorded at 25˚C on a Bruker Avance 900 MHz instrument (Max T. Rogers NMR Facility, Michigan State University) with a TCI cryoprobe. 2D DQF-COSY, TOCSY, ROESY, 1 H-13C HSQC, and 1 H-13C HMBC experiments were performed for assignment and structure determination using standard parameters. DMSO solvent was used as the chemical shift reference and spectra of commercial α-amanitin (Sigma-Aldrich) were recorded for comparison. 4.3.4 Analysis of Gene Expression by RT-PCR. Expression of genes involved in α-amanitin biosynthesis, including GmAMA1-1, GmPOPB, and GmFMO, was analyzed by reverse transcription-PCR (RT-PCR) before and after the onset of toxin production. Two β-tubulin genes (GmTUBB1 and GmTUBB2) and the gene encoding POPA were also analyzed as controls. Total RNA was prepared from G. marginata after 10, 20, and 40 days of growth using a RNeasy Plant Kit (Qiagen), and the quality of the RNA was confirmed on an agarose gel by the presence of intact 16S and 23S rRNA bands. cDNA was then synthesized using SuperScript III reverse transcriptase (ThermoFisher) and an oligo-dT primer. PCR was performed with RedTaq Polymerase (Sigma-Aldrich) with primers designed for the genes listed above based on the available genomic and transcript sequences. Amplification of the correct target sequences was confirmed by Sanger sequencing of the PCR products. 59 4.4 Results 4.4.1 Time Course of Amanitin Production. Production of the toxin α-amanitin by Galerina marginata in laboratory cultures was measured in extracts of fungal tissue between 10 and 50 days of growth (Figure 4.4 and Figure 4.5). α-Amanitin production typically began after 20 to 25 days and peaked on day 40 at an average level of 1.39 mg per gram of tissue. This growth duration was used in all subsequent experiments assessing the effects of knockouts of candidate biosynthetic genes, since the difference between abolished versus diminished toxin production phenotypes would be more apparent at higher levels. After 40 days, α-amanitin levels began to decrease and fell to an average of 0.37 mg/g by day 50. No α-amanitin was detectable in the media of these cultures, ruling out secretion of the toxin. Although α-amanitin is highly stable and resistant to proteases [1], the observed disappearence of the toxin is likely the result of turnover and catabolism by the host. Figure 4.4: Culture of Galerina marginata Mycelium. Photo was taken after 20 days of growth. 60 Figure 4.5: Time Course of α-Amanitin Production in Galerina marginata Cultures. Overlaid plots of G. marginata dry biomass produced in grams (Blue) and amount of αamanitin per gram of tissue (Red). Error bars represent the range in toxin levels from measurements of three separate cultures. 4.4.2 Genes Involved in Amanitin Biosynthesis. G. marginata strains harboring knockouts of genes encoding predicted POPB, FMO, and P450 enzymes on the same scaffold as GmAMA1-1 were successfully engineered by Hong Luo (MSU-DOE Plant Research Laboratory, Michigan State University). A knockout of the gene encoding a P450 adjacent to GmAMA1-2 was unsuccessful, but a knockout of a predicted P450-encoding gene separated from GmAMA1-1 by 29 coding sequences (designated GmP450-29) was achieved. The effects on amanitin production were assessed by LCMS from cultures of the strains grown alongside wild-type Galerina for 40 days, where toxin levels peak in the wild-type strain. No detectable levels of α-amanitin or any of the less hydroxylated forms were present in extracts of the POPB(-) and FMO(-) mutants (Figures 4.6 and 4.7). This result, in combination with their close proximity to GmAMA1-1 and 61 predicted functions, establishes their involvement in the pathway. The P450-29(-) mutant also lost the capacity to produce α-amanitin (Figure 4.8), but we observed formation of a new peak in extracts of the mutant that was absent from the wild-type and hypothesized it to be an intermediate to α-amanitin. UV absorbance spectra of amatoxins and phallotoxins contain a peak at 295 nm due to the presence of tryptathionine, and in compounds containing 6hydroxytryptophan this peak is shifted to 305 nm [1,3]. The suspected intermediate displayed stronger absorbance at 305 nm versus 280 nm, suggesting the presence of both tryptathionine and a modified tryptophan. In agreement with the disrupted gene’s predicted fuction as a monooxygenase, the compound also displayed a mass of 886.4 m/z, 32 mass units less than αamanitin and consistent with an intermediate missing two of the four possible hydroxylations that occur in α-amanitin. Figure 4.6: POPB Knockout in Galerina and Effects on α-Amanitin Production. Signals show UV absorbance (305 nm) of toxin extracts from both wild-type (blue) and the mutant (orange) overlaid. 62 Figure 4.7: FMO Knockout in Galerina and Effects on α-Amanitin Production. Signals show UV absorbance (305 nm) of toxin extracts from both wild-type (blue) and the mutant (green) overlaid. Figure 4.8: P450 Knockout in Galerina and Effects on α-Amanitin Production. Signals show UV absorbance (305 nm) of toxin extracts from both wild-type (blue) and the mutant (red) overlaid. 63 4.4.3 Role of a P450 Monooxygenase in Amanitin Biosynthesis. Approximately 0.7 mg of the intermediate that accumulated in the P450-29(-) Galerina mutant was purified and dissolved in DMSO-d6 for structure determination by NMR spectroscopy. The proton spectrum of the intermediate was similar to that of α-amanitin (Figure S4.1), suggesting an overall related structure. All correlations observed in 2D experiments (Figures S4.2-S4.7 and Table S4.1) were consistent with the structure of a previously undescribed amatoxin shown in Figure 4.9, with missing hydroxylations at the δ-position of Ile1 and γ-position of Pro8. Backbone HN-CO correlations and interresidue NOEs in the intermediate indicated a cyclic backbone. Consistent with γ-hydroxylation, only one Hγ proton was assigned to Ile1 and was coupled to a δ-methyl group in the COSY experiment. HSQC with multiplicity-editing also indicated the absence of CH2 groups in Ile1 (Figure 4.10), and the residue was therefore concluded to be unmodified at the δ-position. For Pro8, all seven protons could be Figure 4.9: Compared Structures of α-Amanitin and the Pathway Intermediate Purified from a P450(-) Strain of G. marginata. Hydroxlyations are highlighted in red and those missing from the intermediate are circled. 64 assigned (Figure 4.10), including γCH2 which was absent in α-amanitin. For both α-amanitin and the intermediate, no couplings were observed with the side-chain indole NH of tryptophan, and only three aromatic 1H-13C bonds were present, consistent with tryptophan modified at positions 2 (tryptathionine) and 6 (hydroxylation). The P450 enzyme is therefore proposed to be responsible for δ-hydroxylation of isoleucine and/or γ-hydroxylation of proline in α-amanitin biosynthesis. Figure 4.10: Compared 1H-13C HSQC Spectra of α-Amanitin and the Intermediate. Signals assigned to the δ-position of Ile1 and the γ-position of Pro8 are indicated. 65 4.4.4 Regulation of Biosynthetic Genes. While the POP and monooxygenase enzymes are likely involved in processing of the precursor peptide and hydroxylations, no candidate enzymes have been designated for formation of the unique tryptathionine group seen in the amatoxins. In fungi, biosynthetic genes for many secondary metabolites are transcriptionally co-regulated, sometimes by a single transcriptional activator dedicated to the pathway [12,13]. If transcriptional co-regulation occurs in the pathway for amanitin biosynthesis, then expression profiling and microarray analysis may be an effective approach for identifying the remaining biosynthetic genes. Expression of GmAMA1-1, GmFMO, and GmPOPB was analyzed before (10 day culture) and after (25 day and 45 day cultures) the onset of α-amanitin production in G. marginata (Figure 4.11) by reverse transcription-PCR. Only GmAMA1-1 expression correlated with biosynthesis, with transcripts only detectable after 25 days and after the start of α-amanitin production. GmFMO and GmPOPB showed constitutive expression along with genes encoding β-tubulin and POPA. Transcriptional activation of GmAMA1-1 is likely limiting to the overall pathway, and other approaches will be necessary for identifying the other remaining biosynthetic genes. 66 Figure 4.11: Expression of Genes Involved in Amanitin Biosynthesis. Transcripts were amplified from mRNA by RT-PCR from G. marginata mycelium grown 10 days (before the onset of toxin biosynthesis), and after 25 and 45 days (during toxin biosynthesis). AMA1-1 expression correlates with toxin production while POPB and FMO show constitutive expression along with housekeeping genes encoding β-tubulin and POPA. 67 4.5 Discussion Three enzymes involved in the biosynthesis of α-amanitin, and possibly cycloamanides in general, were identified in these studies. A predicted prolyl oligopeptidase (GmPOPB), a flavincontaining monooxygenase (GmFMO), and a P450 monooxygenase (GmP450-29) that was further shown to function in isoleucine and/or proline hydroxylation. It is unclear whether the P450 is bifunctional and responsible for both hydroxylations, or catalyzes one hydroxylation to provide a suitable substrate for a separate enzyme. Because no informative pathway intermediates accumulated in the POPB(-) and FMO(-) mutants, the predicted functions of these enzymes were originally based on bioinformatics and automated gene functional annotations. Recombinant POPB has since been shown to encode the macrocyclase that converts the precursor peptide to the cyclic intermediate [14] (see Chapter 5), and determining the precise roles of the P450 and FMO enzymes will similarly require recombinant expression and biochemical characterization. The backbone and side-chain to side-chain (tryptathionine) cyclizations seen in α-amanitin are largely uncharacterized modifications in natural products. At the time of this work, one enzyme, PatG, responsible for N- to C-terminal cyclization in the biosynthesis of the cyanobactin family of cyclic peptides, had been characterized [15]. PatG was determined to be a serine proteaserelated enzyme catalyzing peptide bond ligation instead of hydrolysis, and a peptidase such as POPB may share a similar function. Similarly, while the Trp-Cys linkage seen in amatoxins and other cycloamanides is unique to these compounds and the responsible enzyme is unknown, the FMO enzyme may also function in cyclization since successful synthetic routes to tryptathionine have employed tryptophan with a hydroxylated indole that activates the side-chain for thiol 68 addition [16]. The accumulation of the intermediate resulting from the P450 knockout and detectable presence of the other less modified versions of α-amanitin in wild-type Galerina such as γ-amanitin [3] is indicative of the stability of these less hydroxylated forms, and the absence of any accumulated intermediates or relevant compounds in the FMO(-) mutant suggests a function more integral to the structure of the amatoxins such as cyclization. 69 APPENDIX 70 APPENDIX v Figure S4.1: 1H Spectra of α-Amanitin and the Pathway Intermediate Purified from a P450(-) Strain of G. marginata. 71 Figure S4.2: 2D COSY spectrum of the Amanitin Intermediate. Figure S4.3: 2D TOCSY spectrum of the Amanitin Intermediate. 72 Figure S4.4: 2D ROESY spectrum of the Amanitin Intermediate. Figure S4.5: 2D 1H-13C HSQC spectrum of the Amanitin Intermediate. 73 Figure S4.6: 2D 1H-13C HMBC spectrum of the Amanitin Intermediate. 74 Figure S4.7: Key HMBC Correlations Observed in the Amanitin Intermediate for Structure Determination. 75 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Residue Ile1 Ile1 Ile1 Ile1 Ile1 Ile1 Ile1 Ile1 Ile1 Ile1 Ile1 Ile1 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Trp2 Gly3 Gly3 Gly3 Gly3 Gly3 Ile4 Ile4 Ile4 Ile4 Ile4 Ile4 Ile4 Ile4 Ile4 Ile4 Ile4 Ile4 Ile4 Gly5 Nucleus H H H H H H C C C C C C H H H H H H H H C C C C C C C C C C C H H H C C H H H H H H H C C C C C C H Atom HN Hα Hβ Hγ y'CH3 δCH3 Cα Cβ Cγ Cγ' Cδ CO HN Hα Hβ1 Hβ2 1(N)H 4H 5H 7H Cα Cβ C2 C3 C3a C4 C5 C6 C7 C7a CO HN Hα1 Hα2 Cα CO HN Hα Hβ Hγ1 Hγ2 y'CH3 δCH3 Cα Cβ Cγ Cγ' Cδ CO HN ppm 7.86 4.20 1.99 3.69 0.82 0.92 55.87 40.97 65.41 10.73 17.77 170.58 8.04 4.81 3.26 2.68 11.27 7.45 6.58 6.74 53.03 28.04 129.86 111.28 120.66 122.20 110.41 154.66 96.45 138.76 170.34 7.88 4.37 3.27 40.82 170.35 8.45 3.60 1.55 1.53 1.09 0.80 0.82 58.94 34.52 24.97 14.52 10.73 171.50 8.79 TOCSY COSY HSQC HMBC ROESY 1,2,3,4,5,6 1,2 x 83 2,4,6,13,74 1,2,3,4,5,6 1,2,3 7 8,9,10,12 1,4,6,13 1,2,3,4,5,6 2,3,5 8 7,9,10,11,12 x 1,2,3,4,5,6 4,6 9 7,8,10,11 1,2,6,74 1,2,3,4,5 3,5 10 7,8,9 x 2,3,4,6 4,6 11 8,9 1,2,4 x x 2 3,4,5 x x x 3 2,4,5,6 x x x 4 2,3,5,6 x x x 5 2,3,4 x x x 6 3,4 x x x x 2,3,13,14 x 13,14,15,16 13,14 x 12,21,22 1,2,14,15,16 13,14,15,16 13,14,15,16 21 12,22,31 13,15,18,32 13,14,15,16 14,15,16 22 21,23,24,25 13,14,16,18,67 13,14,15,16 14,15,16 22 21,23,24,25 13,15,32,62,67 x x x x x 18,19,20 18,19 26 24,25,28,29,30 14,15,19 18,19,20 18,19 27 25,28,29 18 18,19,20 x 29 25,27,28,30 x x x 14 13,15,16 x x x 15,16 13,14 x x x x 15,16 x x x x 15,16,18 x x x x 15,16,18,19,20 x x x 18 x x x x 19 20 x x x x 18,19,20 x x x 20 18,19 x x x x 18,20 x x x x 14,32 x 32,33,34 32,33 x 31 14,16,33,34,57,62 32,33,34 32,33,34 35 36 32,34,37 32,33,34 33,34 35 36 32,33,37 x x 33,34 x x x x x 33,34,37 x 37,38,39,41,42 37,38 x 36,45 33,34,39,41,42 37,38,39,41,42,43 37,38,39 44 45,47,49 x 37,38,39,41,42,43 38,39,40,41,42 45 44,46,47,49 37 37,50 x 39,40,41,43 46 44,45,47,48 38,39,41,42,43 39,40,41,43 46 44,45,47,48 37 37,38,39,41,42 39,42 47 44,45,46 37,50,51 39,41,43 40,41,43 48 45,46 x x x 38 39,40,41,42 x x x 39 37,38,40,41,42,43 x x x 40,41 39,42,43 x x x 42 38,39,40,41 x x x 43 40,41 x x x x 38,39,50,51,52 x 50,51,52 50,51,52 x 49,53 38,40,41,42,51,52,55 Table S4.1: Table of all 2D NMR Correlations Observed in the Amanitin Intermediate. All nuclei are listed numerically and for each, the couplings observed in each experiment are indicated with numbers indicating the other coupled nuclei. 76 Table S4.1 (cont’d) 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 Residue Gly5 Gly5 Gly5 Gly5 Cys6 Cys6 Cys6 Cys6 Cys6 Cys6 Cys6 Asn7 Asn7 Asn7 Asn7 Asn7 Asn7 Asn7 Asn7 Asn7 Asn7 Pro8 Pro8 Pro8 Pro8 Pro8 Pro8 Pro8 Pro8 Pro8 Pro8 Pro8 Pro8 Nucleus H H C C H H H H C C C H H H H H H C C C C H H H H H H H C C C C C Atom Hα1 Hα2 Cα CO HN Hα Hβ1 Hβ2 Cα Cβ CO HN Hα Hβ1 Hβ2 δNH2(1) δNH2(2) Cα Cβ γCO CO Hα Hβ1 Hβ2 Hγ1 Hγ2 Hδ1 Hδ2 Cα Cβ Cγ Cδ CO ppm 3.87 3.45 42.17 168.10 8.31 4.94 3.06 2.91 49.92 58.80 167.14 TOCSY COSY HSQC HMBC ROESY 50,51,52 50,51,52 53 49,54 42,50,52,55 50,51,52 50,51,52 53 49,54 50,51,55 x x 51,52 50 x x x x 51,52,55 x 55,56,57,58 55,56 x 54 50,51,52,56,57,58,63 55,56,57,58 55,56,57 59 60,61 55,58,62 55,56,57,58 56,57,58 60 59,61 16,32,55,58,62 55,56,57,58 56,57,58 60 59,61 55,56,57 x x 56 57,58 x x x 57,58 56 x x x x 56,57,58,62 x 8.49 62,63,64,65 62,63 x 61,71 16,32,56,57,63 4.77 62,63,64,65 62,63,64,65 68 70 62,64,65,77 3.48 63,64,65 63,64,65 69 68,70,71 62,63,65,66 3.01 63,64,65 63,64,65 69 68,70 63,64,77 8.50 66,67 x x 70 64,67 7.70 66,67 x x 70 15,16,66 50.65 x x 63 64,65 x 33.05 x x 64,65 x x 173.09 x x x 63,64,65,66,67 x 170.01 x x x 62,64 x 4.15 2.31 1.64 1.98 1.81 3.96 3.60 63.15 29.65 24.96 47.30 170.42 72,73,74,75,76,77,78 72,73,74 79 80,83 1,63,73,76,74 72,73,74,75,76,77,78 72,73,74 80 81,82 72,74,76 72,73,74,75,76,77,78 72,73,74,76 80 79,81,83 1,4,72,73,75 72,73,74,75,76,77,78 x 81 79 x 72,73,74,75,76,77,78 78 81 80,82 72,73,77 72,73,74,75,76,77,78 75,76,77,78 82 79,80,81 63,65,75,76 72,73,74,75,76,77,78 76,77,78 82 81 x x x 72 74,75,77 x x x 73,74 72,76,77 x x x 75,76 73,74,77,78 x x x 77,78 73,76 x x x x 1,72,74 x 77 WORKS CITED 78 WORKS CITED 1. Wieland, T. (1986). Peptides of Poisonous Amanita Mushrooms. Springer: New York. 2. Enjalbert F, Cassanas G, Rapior S, Renault C, and Chaumont JP. (2004). Amatoxins in wood-rotting Galerina marginata. Mycologia 96(4): 720-729. 3. Sgambelluri RM, Epis S, Sassera D, Luo H, Angelos ER, and Walton JD. (2014). Profiling of amatoxins and phallotoxins in the genus Lepiota by liquid chromatography combined with UV absorbance and mass spectrometry. Toxins 6(8): 2336-2347. 4. Smith AH. (1953). New Species of Galerina from North America. Mycologia 45(6): 892925. 5. Benedict RG, Tyler VE Jr, Brady LR, and Weber LJ. (1966). Fermentative production of amanita toxins by a strain of Galerina marginata. J. Bacteriol. 91(3): 1380-1381. 6. Luo H, Hallen-Adams HE, Scott-Craig JS, and Walton JD. (2012). Ribosomal biosynthesis of α-amanitin in Galerina marginata. Fungal Genet. Biol. 49(2): 123-129. 7. Anderson MR, Nielsen JB, Kitgaard A, Peterson LM, Zacharisa M, Hansen TJ, Blicher LH, Gotfredsen CH, Larsen TO, Nielsen KF, and Mortensen UH. (2013). Accurate predicion of secondary metabolite gene clusters in filamentous fungi. Proc. Natl. Acad. Sci. U.S.A. 110(1): E99-107. 8. Brakhage AA and Schroeckh V. (2011). Fungal secondary metabolites - strategies to activate silent gene clusters. Fungal Genet. Biol. 48(1): 15-22. 9. Enjalbert F, Gallion C, Jel F, Monsteil H, and Faulstich HJ. (1992). Simultaneous assay for amatoxins and phallotoxins in Amanita phalloides Fr., by high-performance liquid chromatography. J. Chromatogr. 598(2): 227-236. 10. Kemppainen MJ, and Pardo AG. (2010). Gene knockdown by ihpRNA-triggering in the ectomycorrhizal basidiomycete fungus Laccaria bicolor. Bioeng. Bugs. 1(5): 354-358. 11. Kemppainen MJ, and Pardo AG. (2011). Transformation of the mycorrhizal fungus Laccaria bicolor using Agrobacterium tumefaciens. Bioeng. Bugs. 1(5): 354-358. 12. Brakhage AA. (2013). Regulation of fungal secondary metabolism. Nat. Rev. Microbiol. 11(1): 21-32. 79 13. Bergman S, Schumann J, Scherlach K, Lange C, Brakhage AA, and Hertweck C. (2007). Genomics-driven discovery of PKS-NRPS hybrid metabolites from Aspergillus nidulans. Nat. Chem. Biol. 3(4): 213-217. 14. Luo H, Hong SY, Sgambelluri RM, Angelos E, Li X, and Walton JD. (2014). Peptide macrocyclization catalyzed by a prolyl oligopeptidase involved in α-amanitin biosynthesis. Chem. Biol. 21(12): 1610-1617. 15. Lee J, McIntosh J, Hathaway BJ, and Schmidt EW. (2009). Using marine natural products to discover a protease that catalyzes peptide macrocyclization of diverse substrates. J. Am. Chem. Soc. 131(6): 2122-2124. 16. May JP and Perrin DM. (2007). Tryptathionine bridges in peptide synthesis. Biopolymers. 88(5): 714-724. 80 CHAPTER 5 BIOCHEMICAL CHARACTERIZATION OF PROLYL OLIGOPEPTIDASE B AS A PEPTIDE MACROCYCLASE Note: The content in this chapter has been previously published. Some text has been modified from the original. Copyright © Elsevier Ltd All rights reserved Citation: Luo H, Hong SY, Sgambelluri RM, Angelos E, Li X, and Walton JD. (2014). Peptide macrocyclization catalyzed by a prolyl oligopeptidase involved in α-amanitin biosynthesis. Chem. Biol. 21(12): 1610-1617. Author Contributions: Molecular cloning of POPB cDNA was performed by Hong Luo and Sung-Yong Hong. Evan R. Angelos, Xuan Li, and Hong Luo assisted with POPB purification and enzyme assays. 81 5.1 Abstract Amatoxins are ribosomally encoded and post-translationally modified peptides that account for the majority of fatal mushroom poisonings of humans. A representative amatoxin is the bicyclic octapeptide α-amanitin, formed via head-to-tail macrocyclization, which is ribosomally biosynthesized as a 35-amino acid precursor peptide in Amanita spp. and the distantly related mushroom Galerina marginata. POPB, a member of the prolyl oligopeptidase (POP) family of serine proteases, has been proposed to play a role in α-amanitin posttranslational processing; however the exact mechanistic details are not known. Here we show that POPB from G. marginata is bifunctional and catalyzes two nonprocessive reactions with the α-amanitin precursor peptide: hydrolysis at an internal Pro residue to release the 10mer N-terminal sequence, and transpeptidation at a second Pro to produce a cyclic octapeptide composed of the α-amanitin sequence. 82 5.2 Introduction A predicted prolyl oligopeptidase (POP) enzyme, GmPOPB, was previously shown by reverse genetics to be essential for α-amanitin production in the mushroom Galerina marginata (see Chapter 4). In mushrooms, POPB homologs are only present in species that produce cycloamanides [1], suggesting that POPB plays a dedicated role in the biosynthesis of cycloamanides. POPs are large (~80 kDa) serine proteases that hydrolyze peptides at the carboxyl side of proline residues [2]. POPs have been cloned and characterized from bacteria [3,4], archaea [5], insects [6], and mammals [7,8,9], and share a conserved two-domain structure (Figure 5.1). The C-terminal portion of the sequence forms a conserved peptidase domain with an α/β hydrolase fold and contains the serine protease catalytic triad (Ser-His-Asp) [10,11]. The N-terminal portion is more variable, but forms a seven bladed β-propeller domain with a proposed role as a “gating filter” in substrate selection [12,13]. Proline specificity in POPs is achieved with a hydrophobic S1 specificity pocket and by a ring-stacking interaction with the indole side-chain of an active site tryptophan residue [12,14]. POP enzymes in bacteria are believed to carry out housekeeping functions in protein turnover [2]. In the bacterium Kribbella flavida, one POP was shown to be involved in the biosynthesis of lanthipeptides, RiPPs with antimicrobial activity against Gram-positive bacteria [15]. In mammals, the majority of known peptide hormones and neuropeptides contain at least one proline residue [16], and consistent with a role in neuropeptide metabolism, mammalian POP is concentrated in brain tissue [17]. Aberrant levels of serum POP activity are characteristic of a number of psychiatric disorders in humans including depression [18], mania, and schizophrenia [19]. POP inhibitors slow memory loss in Alzheimer’s disease [20] and reverse drug-induced amnesia in rats [21]. A number of POP-specific inhibitors are in clinical trials [22]. 83 Figure 5.1: Crystal Structure of Prolyl Oligopeptidase from Porcine Brain. The active site is found at the interface of peptidase (purple) and β-propeller (blue) domains. Reillustrated with PyMol from the Protein Data Bank (http://www.rcsb.org) structure 1H2W. Cycloamanides are composed of the amino acid sequences from the core domains of their corresponding precursor peptides [23]. POPB likely functions in proteolysis of the precursors at the invariable proline residues that separate the core domain from the leader and follower sequences. Alternatively, POPB could be responsible for backbone macrocyclization of the core domains, since proteaseses from similar RiPP pathways have recently been shown to catalyze this reaction [24,25,26,27]. The following studies aim to characterize POPB from G. marginata in vitro and to define the enzyme’s role in cycloamanide biosynthesis. 84 5.3 Methods 5.3.1 Protein Expression and Purification. GmPOPB cDNA was cloned by Sung Yong Hong (MSU-DOE Plant Research Laboratory, Michigan State University) and inserted into the pESCHIS vector (Agilent Technologies) for expression in Saccharomyces cerevisiae (strain YPH501) with a N-terminal c-myc epitope tag for purification. Transformed yeast cells were first grown overnight in SD medium (0.67% yeast nitrogen base without amino acids, 2% dextrose, and 0.13% amino acid mixture minus histidine for selection), and the following morning this culture was diluted 1:50 with SG media (SD with galactose substituted for dextrose) for induction. Cells were induced for 48 hr at 30°C with shaking. Cells were harvested at 4,000 x g for 10 min and then lysed by grinding in liquid nitrogen. The yeast powder was then resuspended in buffer (20 mM Tris, pH 7.5, 0.4% glycerol, 1 mM EDTA, 2 mM DTT) at 100 mL per liter of culture. Soluble protein was collected at 21,000 x g for 20 min. Recombinant GmPOPB was first purified on anti c-myc agarose (ThermoFisher) and eluted with tris-buffered saline (TBS) containing 1 mg/mL c-myc peptide. Ion-exchange was included as a second purification step on a TSK DEAE-5PW column (Tosoh Bioscience) with a 25 min gradient from 0 to 600 mM NaCl in 20 mM Tris, pH 7.5 on an Agilent 1100 series HPLC system. Working enzyme solution was stored in aliquots at -80°C at 1 mg/mL in 20 mM Tris buffer, pH 7.5, with 2 mM DTT and ~250 mM NaCl. Protein concentrations were measured using bicinchoninic acid (BCA) (Pierce Biotechnology) against bovine serum albumin (BSA) as standard. GmPOPB mutants were prepared using a QuikChange Lightning site-directed mutagenesis kit (Agilent Technologies) and purified in the same manner as wild-type enzyme. 85 5.3.2 Enzyme Assays. POP activity on peptide substrates was assayed in 20 mM Tris HCl (pH 7.5) containing 10 mM dithiothreitol and ~90 µM peptide at 37°C. Chemically synthesized peptides were supplied by Bachem and Elim Biopharmaceuticals. For kinetic studies, each reaction contained 15 ng (0.18 pmol) of enzyme and varying amounts of substrate in 50 µL total volume and triplicate measurements were made for each substrate concentration. Kinetic constants were calculated using nonlinear curve fitting with GraphPad Prism (GraphPad Software). At the end of the incubation, methanol was added to 50% (v/v), the samples were centrifuged at 20,000 x g for 5 min, and the supernatants were dried under vacuum and resuspended in water. Reactions were analyzed by ESI-LCMS using an Agilent 1100 pump system and Agilent 6120 single quadrupole mass spectrometer equipped with a multi-wavelength UV detector. Separation was performed on a reverse-phase C18 column (RS-2546-W185, Higgins Analytical) with a 20 min linear gradient from 20 mM ammonium acetate (pH 5) to 100% acetonitrile at 1 mL/min. UV absorbance was monitored at 220, 250, and 280 nm. 5.3.3 Product Purification and NMR Spectroscopy. For large-scale purification of the cyclic reaction product, 5 µg of GmPOPB protein was incubated with 10 mg GmAMA1 peptide overnight at 37°C. The protein was removed by precipitation with 50% (v/v) methanol and centrifugation, and the supernatants were dried and resuspended in water. The product was purified on a preparative C18 column (Supelcosil LC-18, 25 cm x 10 mm, 5 mm) using the same HPLC method described above at 0.5 mL/min flow rate. Fractions containing product were then dried to yield ~1.2 mg of white powder (57% yield). LCMS indicated isolation of the intended product (correct mass and retention time) and ~88% purity on the basis of absorbance at 280 nm. The purified product was dissolved in DMSO-d6 at 5 mM concentration. NMR spectra were collected at 25°C on a Varian 600 MHz instrument. 1H atoms were assigned with COSY, 86 TOCSY, and ROESY. TOCSY spectra were acquired with an MLEV17 mixing sequence with a mixing time (tm) of 80 ms and ROESY spectra were collected with a tm of 200 ms. (natural abundance) were assigned with HSQC and HMBC. 87 13 C atoms 5.4 Results 5.4.1 Preparation of Recombinant GmPOPB. GmPOPB with an N-terminal c-myc epitope tag was expressed in S. cerevisiae and purified in two steps from cell extracts. The first step was on an anti-c-myc agarose affinity column, and the second was by anion exchange on DEAE. The resulting protein solution gave a single band by SDS-PAGE with the expected molecular weight (~84 kDa) for the 730 residue protein (Figure 5.2). The method yielded an average of ~1.8 mg of recombinant protein after the second purification step from one liter of culture. Even after purification, the protein was highly sensitive to degradation, likely from autoproteolysis, and required flash freezing and storage in aliquots at -80°C. Figure 5.2: Purification of Recombinant GmPOPB Expessed in Yeast. Image shows SDS-PAGE of GmPOPB protein after purification from crude extracts on c-myc agarose (lane 2) followed by purification by anion exchange chromatography (lane 3). 88 5.4.2 GmPOPB Catalyzes Peptide Macrocyclization. Synthetically produced GmAMA1, the 35mer precursor peptide to α-amanitin and hypothesized natural substrate for GmPOPB, was incubated with POPB enzyme and the reaction was monitored by LCMS. Activity on the GmAMA1 peptide was observed (Figure 5.3) with products corresponding to cleavage of the substrate at proline residues flanking the core domain of AMA1. For the product corresponding to the core domain sequence, LCMS indicated a monoisotopic mass of 841.4 m/z, 18 fewer mass units than expected for linearized peptide and suggesting formation of a new peptide bond concomitant with loss of water and cyclo-IWGIGCNP product. Figure 5.3: Time Course of Conversion of GmAMA1 to cyclo-IWGIGCNP. Signals are UV absorbance at 280 nm from HPLC separation of the reaction products between 0 and 90 min of incubation. 89 A large-scale reaction using purified GmPOPB and synthetic GmAMA1 was used to produce ~1.2 mg of the putative cyclic reaction product for NMR experiments. 1H and 13 C atoms in the product were assigned with COSY, TOCSY, ROESY, HSQC, and HMBC experiments (Figures S5.2 through S5.6). A signal was observed in the HMBC experiment corresponding to throughbond coupling between the backbone amide of Ile1 and carbonyl of Pro8, and served as direct detection of the newly formed peptide bond in the product (Figure 5.4). Coupling between the HN of Ile1 and Hα of Pro8 was also observed in the ROESY experiment, consistent with their close proximity upon cyclization. Finally, the free thiol proton from Cys6 was able to be assigned, indicating that the product did not contain an internal thioester, a modification that could result in the same 18 unit mass discrepancy. These studies confirm the formation of cycloIWGIGCNP and a macrocyclization reaction catalyzed GmPOPB. Figure 5.4: Amide Bond Couplings in the HMBC Spectrum of cyclo-IWGIGCNP. In the HN-CO region of the spectrum, a signal (highlighted in red) indicating through-bond coupling between Ile1 and Pro8 residues confirms a cyclized backbone in the reaction product. 90 5.4.3 GmPOPB is a Bifunctional Enzyme. POPB was incubated with excess AMA1 and analyzed before the reaction was complete and the substrate was consumed. AMA1 was converted to a series of products consistent with bifunctional hydrolase and macrocyclase activity by POPB (Figure 5.5). Specifically, the data indicate hydrolysis at the first proline residue (Pro10) and transpeptidation/cyclization at the second (Pro18) in the AMA1 sequence. A small amount of linearized IWGIGCNP peptide was also detectable as product, but less than the limit of detection for UV absorbance and therefore less than 1% of total product composed of the core domain sequence. During the reaction, a truncated 25mer intermediate resulting from hydrolysis at Pro10 and removal of the N-terminal leader sequence accumulated transiently (Figure 5.5). Since no product corresponding to initial activity at Pro18 was observed, the two reaction steps catalyzed by POPB are ordered, with hydrolysis preceding cyclization (Figure 5.6). To determine if cyclization requires concurrent hydrolysis or if the steps are exclusive, GmPOPB was incubated separately with the 25mer intermediate as initial substrate. As with fulllength substrate, cyclo-IWGIGCNP was produced, indicating the reaction steps are nonprocessive (Figure S5.6). The kinetic constants Km, Vmax, and kcat were determined for both full-length and truncated substrates by measuring rates of cyclic product formation at varying substrate concentrations (Figure 5.7 and Table 5.1). Cyclo-IWGIGCNP was produced from both substrates at identical rates of 5.7 sec-1 consistent with cyclization being rate-limiting in the overall reactionand supported by the observed build-up of the 25mer intermediate in time-course assays. Backbone macrocyclization of peptides is catalyzed by three other known enzymes that have been biochemically characterized: PatG, PCY1, and butelase, with turnover rates of 1 hr-1, 2 hr-1, and 17 sec-1 respectively. POPB is comparable in efficiency to butelase. 91 Figure 5.5: Reaction Products from GmPOPB Activity on GmAMA1. Products were analyzed by LCMS before the reaction went to completion. The observed products and intermediate formed indicate an ordered, two-step reaction scheme beginning with Nterminal hydrolysis. Signals are UV absorbance at 280 nm (top) and extracted ion chromatograms (bottom) for the expected monoisotopic masses of substrate and each product. The observed m/z values and charges are indicated. Figure 5.6: Two-step Nonprocessive Reaction Catalyzed by POPB on the αAmanitin Precursor Peptide. 92 Figure 5.7: Kinetic Analysis of GmPOPB. Shown are overlaid GmPOPB saturation curves with 35mer (blue) and 25mer (red) substrates. Table 5.1: GmPOPB Cyclization Kinetic Constants with 35mer and 25mer GmAMA1 Substrates. 93 5.4.4 Residues Involved in Macrocyclization. GmPOPB is 75.5% identical to AbPOPB, the POPB homolog from Amanita bisporigera; 59% identical to GmPOPA, a housekeeping POP uninvolved in cycloamanide biosynthesis [1]; and 37% identical to the two well characterized POP enzymes from porcine brain and muscle tissue. To identify residues or motifs that may be involved in POPB’s capacity for macrocyclization, the sequences of the porcine POPs and POPAs from A. bisporigera and G. marginata were analyzed in a ClustalW2 multiple sequence alignment (Figure S5.7). A high degree of similarity between all six POPs is revealed in the alignment. All six proteins are roughly 750 amino acids in length with no apparent gaps or additional motifs present that are unique to POPB. Consistent with a serine protease mechanism, all the fungal POPs contained serine, aspartic acid, and histidine residues (Ser577, Asp661, and His698 in GmPOPB) that aligned with these same catalytic residues in the porcine POPs [12]. The active site Trp residue shown in crystallization studies with the porcine POPs to stack with proline in the substrate also aligned with Trp residues (Trp619 in GmPOPB). A GmPOPB variant (S577A) lacking the predicted catalytic serine was prepared to test whether the residue is required for initial hydrolysis of AMA1 and also to test its involvement in cyclization. No activity was observed with the GmPOPB(S577A) variant on either the full-length 35mer AMA1 peptide or the 25mer intermediate missing the leader sequence. This supports the classification of POPB as a serine protease and the involvement of Ser577 in the N-C cyclization mechanism. The sequences adjacent to the catalytic Asp and His also contain residues that are differentially conserved between the POPBs and the other POPs (Table 5.2), and these residues are hypothesized to play a role in cyclization. 94 POPA POPB Region 1 (Asp661) Region 2 (His698) ADHDDRVVP NIGDGRVVP -KAGHGMGK SWLGHGMGK Table 5.2: Differentially Conserved Residues Between POPA and POPB. Sequences in POPA (red) and POPB (green) are located adjacent to catalytic Asp and His residues (purple) 95 5.5 Discussion The enzyme POPB was determined to function in both leader peptide removal and head-to-tail macrocyclization during the biosynthesis of α-amanitin and likely all other cycloamanides. The enzyme catalyzed the two-step, non-processive reaction shown with the precursor peptide to αamanitin as substrate. A total of four other enzymes, all predicted serine or asparagine proteases involved in similar RiPP pathways, have been shown to catalyze peptide head-to-tail condensation/macrocyclization: PatG [24], PCY1 [25], butelase [26], and AEP1 [27]. POPB is unique among these enzymes in its bifunctionality. While the other cyclases require the leader sequence to first be removed from the precursor peptide substrate by a separate protease, POPB catalyzes both steps. The catalytic Ser577 residue in POPB was found to be necessary for both hydrolase and cyclase activities, and catalysis is therefore hypothesized to involve a familiar serine protease mechanism in which macrocyclization is achieved through removal of the covalent intermediate via deacylation with the N-terminal amine of bound substrate instead of water (Figure 5.8). Further mutagenesis and structural studies with bound substrates will be necessary for a complete description of the mechanisms utilized by this unusual enzyme. Figure 5.8: Hypothetical Mechanism for Macrocylization Catalyzed by POPB. 96 APPENDIX 97 APPENDIX Figure S5.1: 2D gCOSY Spectrum of cyclo-IWGIGCNP. Figure S5.2: 2D TOCSY Spectrum of cyclo-IWGIGCNP. 98 Figure S5.3: Amide Region from TOCSY Spectrum of cyclo-IWGIGCNP. The 1H assignments are indicated. 99 Figure S5.4: 2D ROESY Spectrum of cyclo-IWGIGCNP. Figure S5.5: 2D 1H-13C HSQC Spectrum of cyclo-IWGIGCNP. 100 Figure S5.6: Formation of cyclo-IWGIGCNP Product from 35mer and 25mer Substrates. Signals are extracted ion chromatograms (EICs) for the expected masses of the peptides. The observed m/z values and charge states (z) are indicated. 101 Figure S5.7: Multiple Sequence Alignment of POPB and Other Prolyl Oligopeptidases. Positions with conserved/identical residues (orange) and non-identical but conserved by amino acid properties (yellow) are highlighted. Residues that are differentially conserved between POPB (green) and homologs without cyclase activity (red) are also indicated, as well as the catalytic Ser-His-Asp triad and tryptophan residue critical for proline specificity (purple). 102 Figure S5.7 (cont’d) 103 WORKS CITED 104 WORKS CITED 1. Luo H, Hallen-Adams HE, Scott-Craig JS, and Walton JD. (2012). Ribosomal biosynthesis of α-amanitin in Galerina marginata. Fungal Genet. Biol. 49(2): 123-129. 2. Polgár L. (2002). The prolyl oligopeptidase family. Cell Mol. Life Sci. 59(2): 349-362. 3. Chevallier S, Goeltz P, Thibault P, Banville D, and Gagnon J. (1992). Characterization of prolyl endopeptidase from Flavobacterium meningosepticum. Complete sequence and localization of the active-site serine. J. Biol. Chem. 267(12): 8192-8199. 4. Kanatani A, Yoshimoto T, Kitazono A, Kokubo T, and Tsuru D. (1993). Prolyl endopeptidase from Aeromonas hydrophila: cloning, sequencing, and expression of the enzyme gene, and characterization of the expressed enzyme. J. Biochem. 113(6): 790796. 5. Harris MN, Madura JD, Ming LJ, and Harwood VJ. (2001). Kinetic and mechanistic studies of prolyl oligopeptidase from the hyperthermophile Pyrococcus furiosus. J. Biol. Chem. 276(22): 19310-19317. 6. Ohtsuki S, Homma K, Kurata S, and Natori S. (1997). Molecular cloning of cDNA for Sarcophaga prolyl endopeptuidase and characterization of the recombinant enzyme produced by an E.coli expression system. Insect Biochem. Mol. Biol. 27(4): 337-3343. 7. Ishino T, Ohtsuki S, Homma Ki, and Natori S. (1998). cDNA cloning of mouse prolyl endopeptidase and its involvement in DNA synthesis by Swiss 3T3 cells. J. Biochem. 123(3): 540-545. 8. Yoshimoto T, Miyazaki K, Haraguchi N, Kitazono A, Kabashima T, and Ito K. (1997). Cloning and expression of the cDNA encoding prolyl oligopeptidase (prolyl endopeptidase) from bovine brain. Biol. Pharm. Bull. 20(10): 1047-1050. 9. Shirasawa Y, Osawa T, and Hirashima A. (1994). Molecular cloning and characterization of prolyl endopeptidase from human T cells. J. Biochem. 115(4): 724-729. 10. Polgár L. (1992). Structural relationship between lipases and peptidases of the prolyl oligopeptidase family. FEBS Lett. 311(3): 281-284. 11. Goosens F, De Meester I, Vanhoof G, Hendricks D, Vriend G, and Scharpe S. (1995). The purification, characterization and analysis of primary and secondary-structure of prolyl oligopeptidase from hyman lymphocytes. Evidence that the enzyme belongs to the alpha/beta hydrolase fold family. Eur J. Biochem. 233(2): 432-441. 105 12. Fülöp V, Böcksei Z, and Polgár L. (1998). Prolyl oligopeptidase: an unusual betapropeller domain regulates proteolysis. Cell. 94(2): 161-170. 13. Fülöp V, Szeltner Z, and Polgár L. (2000). Catalysis of serine oligopeptidases is controlled by a gating filter mechanism. EMBO Rep. 1(3): 277-281. 14. Racys DT, Rea D, Fülöp V, and Wills M. (2010). Inhibition of prolyl oligopeptidase with a synthetic unnatural dipeptide. Bioorg. Med. Chem. 18(13): 4775-4782. 15. Völler GH, Krawczyk B, Ensle P, and Süssmuth RD. (2013). Involvement and unusual substrate specificity of a prolyl oligopeptidase in class III lanthipeptide maturation. J. Am. Chem. Soc. 135(20): 7426-7429. 16. Mentlein R. (1988). Proline residues in the maturation and degradation of peptidehormones and neuropeptides. FEBS Lett. 234(2): 251-256. 17. Myöhänen TT, Venäläinen JI, Garcia-Horsman JA, Piltonen M, Männistö PT. (2008). Cellular and subcellular distribution of rat brain prolyl oligopeptidase and its association with specific neuronal neurotransmitters. J. Comp. Neurol.507(5): 1694-1708. 18. Maes M, Goossens F, Scharpe S, Meltzer HY, D’Hondt P, and Cosyns P. (1994). Lower serum prolyl endopeptidase enzyme activity in major depression: further evidence that peptidases play a role in the pathophysiology of depression. Biol. Psychiatry. 35(8): 545552. 19. Maes M, De Meester I, Scharpe S, Desnyder R, Ranjan R, and Meltzer HY. (1996). Alterations in plasma dipeptidyl peptidase IV enzyme activity in depression and schizophrenia: effects on antidepressants and antipsychotic drugs. Acta Psychiatr. Scand. 93(1): 1-8. 20. Mannisto PT, Venalainen J, Jalkanen A, and Garcia-Horsman JA. (2007). Prolyl oligopeptidase: a potential target for the treatment of cognitive disorders. Drug News Perspect. 20(5): 293-305. 21. Yoshimoto T, Kado K, Matsubara F, Koriyama N, Kaneto H, and Tsura D. (1987). Specific inhibitors for prolyl endopeptidase and their anti-amnesic effect. J. Pharmacobiodyn. 10(12): 730-735. 22. Babkova K, Korabecny J, Soukup O, Nepovimova E, Jun D, and Kuca K. (2017). Prolyl oligopeptidase and its role in the organism: attention to the most promising and clinically relevant inhibitors. Future Med. Chem. 9(10): 1015-1038. 23. Hallen HE, Luo H, Scott-Craig JS, and Walton JD. (2007). Gene family encoding the major toxins of lethal Amanita mushrooms. Proc. Natl. Acad. Sci. U.S.A. 104(48): 1909719101. 106 24. Lee J, McIntosh J, Hathaway BJ, and Schmidt EW. (2009). Using marine natural products to discover a protease that catalyzes peptide macrocyclization of diverse substrates. J. Am. Chem. Soc. 131(6): 2122-2124. 25. Barber CJ, Pujara PT, Reed DW, Chiwocha S, Zhang H, and Covello PS. (2013). The twostep biosynthesis of cyclic peptides from linear precursors in a member of the plant family Caryophyllaceae involves cyclization by a serine protease-like enzyme. J. Biol. Chem. 288(18): 12500-12510. 26. Nguyen GK, Wang S, Qiu Y, Hemu X, Lian Y, and Tam JP. (2014). Butelase 1 is an Asxspecific ligase enabling peptide macrocyclization and synthesis. Nat. Chem. Biol. 10(9): 732-738. 27. Harris KS, et al. (2015). Efficient backbone cyclization of linear peptides by a recombinant asparaginyl endopeptidase. Nat. Commun. 6: 10199. 107 CHAPTER 6 VERSATILITY OF PROLYL OLIGOPEPTIDASE B IN PEPTIDE MACROCYCLIZATION Note: The content in this chapter has been previously published. Some text has been modified from the original. Citation: Sgambelluri RM, Smith MO, and Walton JD. (in press). Versatility of prolyl oligopeptidase B in peptide macrocyclization. ACS Syn. Biol. doi: 10.1021/acssynbio/7b00264 Author Contributions: Preparation of DNA constructs for expression of POPB substrates was performed by Miranda O. Smith. 108 6.1 Abstract Cyclic peptides are promising compounds for new chemical biological tools and therapeutics due to their structural diversity, resistance to proteases, and membrane permeability. Amatoxins, the toxic principles of poisonous mushrooms, are biosynthesized on ribosomes as 35mer precursor peptides, which are ultimately converted to hydroxylated bicyclic octapeptides. The initial cyclization steps, catalyzed by a dedicated prolyl oligopeptidase (POPB), involves removal of the 10-amino acid leader sequence from the percursor peptide and transpeptidation to produce a monocyclic octapeptide intermediate. The utility of POPB as a general catalyst for peptide cyclization was systematically characterized using a range of precursor peptide substrates produced either in E. coli or chemically. Substrates produced in E. coli were expressed either individually or in mixtures produced by codon mutagenesis. A total of 127 novel peptide substrates were tested, of which POPB could cyclize 100. Peptides of 7-16 residues were cyclized at least partially. Synthetic 25mer precursor peptide substrates containing modified amino acids including D-Ala, β-Ala, N-methyl-Ala, and 4-hydroxy-Pro were also successfully cyclized. Although a phalloidin heptapeptide with all L amino acids was not cyclized, partial cyclization was seen when L-Thr at position #5 was replaced with the naturally occurring D amino acid. POPB should have broad applicability as a general catalyst for macrocyclization of peptides containing 7 to at least 16 amino acids, with an optimum of 8-9 residues. 109 6.2 Introduction Due to their structural ridigity and conformational diversities, cyclic peptides often display high affinity binding to target macromolecules, relatively high membrane permeability, and resistance to proteases [1-4]. Nine cyclic peptide drugs have been approved in the past ten years against bacterial and fungal infections, cancer, and gastrointestinal disorders [5]. Recent examples of promising cyclic peptide drug leads include an inhibitor of the RAS oncogene [6]; the modified griselimycins, which have promise against multidrug resistant tuberculosis [7]; a cyclotide that activates the p53 tumor suppressor pathway [8]; and lugdunin, a novel antibiotic from a human commensal bacterium that is active against Staphylococcus aureus [9]. However, synthesis of cyclic peptides remains difficult and expensive compared to linear peptides [10]. Ribosomally biosynthesized cyclic peptides, known as RiPPs, have been described from bacteria, plants, mammals, and fungi [11]. Prior to the discovery of the genes encoding the amatoxins, phallotoxins, and other cyclic peptides from the agaric genus Amanita (collectively known as the cycloamanides), RiPPs were unknown in fungi [12-14]. Amatoxins such as α-amanitin are defining inhibitors of RNA polymerase II, and phallotoxins such as phalloidin bind and stabilize F-actin [15-17]. The amatoxins are highly stable and rapidly absorbed into the bloodstream and into mammalian cells [18]. Cycloamanides are biosynthesized initially as small (33-37 amino acid) precursor peptides encoded by a gene family comprising at least 73 members among different Amanita species [12,19,20]. The conserved structures of the cycloamanide precursor peptides are composed of a 10-amino acid leader, a variable region of 6-10 amino acids which give rise to the mature toxins, and a conserved follower peptide of 17 residues. Although the amino acid content of the variable 110 region in the naturally occurring cycloamanide gene family is biased toward hydrophobic amino acids and especially Pro, all 20 amino acids are present in at least one predicted cycloamanide [20]. Cyclization of the variable region of the cycloamanides occurs in two nonprocessive steps, both catalyzed by a specialized prolyl oligopeptidase, POPB [21]. The amatoxins and phallotoxins, but not the classic monocyclic cycloamanides, are further posttranslationally processed by multiple hydroxylations and formation of a cross-bridge between Cys and Trp called tryptathionine [22]. Additional modifications include sulfoxidation in the amatoxins and epimerization of one amino acid in the phallotoxins [18]. The kinetic efficiency of POPB from Galerina marginata expressed in Saccharomyces cerevisiae is sufficiently high to make it a practical reagent for custom synthesis of cyclic peptides [21]. POPB is comparable in catalytic properties to the peptide macrocyclase butelase 1 from Clitoria ternatea and PCY1 from Saponaria vaccaria [23-26]. Detailed kinetic studies on POPB expressed in E. coli confirmed its high catalytic efficiency as a peptide macrocyclase and showed that release of the follower peptide is the limiting step [23]. Here we explore the utility of POPB as a general catalyst for peptide macrocyclization through characterization of the enzyme’s substrate versatility and limitations on composition and length of the core domain sequence. 111 6.3 Methods 6.3.1 DNA Constructs. A cDNA of the AMA1 precursor peptide gene from G. marginata (GmAMA1) was synthesized from total RNA and cloned into the pMAL-c5x expression vector (containing the gene for maltose binding protein and a Factor Xa protease cleavage site) (New England Biolabs) using the In-Fusion HD cloning system (Clontech). Constructs for expression of AMA1 variants with single amino acid substitutions were prepared by site-directed mutagenesis of the wild-type construct using a QuikChange Lightning kit (Agilent Technologies). Coding sequences for natural substrates and substrates with varying core domain lengths were obtained as synthetic gene fragments (gBlocks, Integrated DNA Technologies) and inserted into the expression vector using In-Fusion. All DNA constructs were verified by Sanger sequencing and transformed into E. coli (BL21-DE3) cells for expression. 6.3.2 Preparation of POPB Substrates. Substrates containing unusual amino acids were produced by solid-phase synthesis by Bachem Americas, Inc. For all other peptide substrates, E. coli cells expressing MBP-peptides were grown with ampicillin selection in Luria broth (LB) supplemented with 2 g/L glucose at 37˚C with shaking and induced at an OD600 of approximately 0.6 with 2 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) for 4 hr at 30˚C. The cultures were harvested by centrifugation at 8,000 x g for 10 min and the pellet was resuspended in buffer (20 mM Tris, pH 7.5, 50 mM NaCl, 1 mM EDTA) and flash frozen. Cells were lysed by thawing at 42˚C in the presence of 1 mg/mL lysozyme and 0.5 mM PMSF. DNase (10 units/mL) was added until the viscosity of the solutions cleared, and insoluble material was removed by centrifugation at 21,000 x g for 20 min. The MBP-peptide fusions were isolated from crude extracts on amylose resin (New England Biolabs) and eluted with 10 mM maltose. Eluates were 112 then concentrated to 100-fold their original volume with Macrosep Advance spin concentrators (Pall Life Sciences) and incubated at room temperature with 8 μg/mL Factor Xa protease (New England Biolabs). After protease treatment, MBP was precipitated with 50% (v/v) methanol and removed by centrifugation, and the peptide solutions were dried under vacuum and redissolved in water containing 2 mM DTT. 6.3.3 Cyclization Assays and LCMS. Cyclization assays were typically performed in 20 mM Tris, pH 7.5 with 25 mM DTT, ~25 μg substrate, and 5 μg enzyme (prepared as described in Section 5.3.1) at 37˚C. Wild-type AMA1 peptide was used as a control. After 4 hr, the enzyme was removed by precipitation with methanol and the reactions were dried under vacuum and redissolved in water. The products were analyzed by LCMS using an Agilent 1200 pump system and an Agilent 6120 single quadrupole instrument in positive ion mode with a 20 min gradient from 20 mM NH4OAc (pH 5) to acetonitrile on a Higgins Proto-300 C18 column. For each substrate, reactions were analyzed before and after addition of POPB by UV/Vis and extracted ion chromatograms (EICs) targeting the expected masses of full-length substrate, the expected cyclic peptide product, and the expected linear form of the core domain. Substrate levels and the relative amounts of cyclic vs. linear product were quantitated by integrating peak areas at OD 280 with a detection limit of 0.15 μmol/L per tryptophan residue. Relative concentrations of Trpnoncontaining peptides were estimated from absorbance at 220 nm. 6.3.4 Library Preparation and Analysis. The plasmid contruct for library production was prepared with an Ultramer ssDNA fragment (Integrated DNA Technologies) that contained a sequence encoding full-length precursor peptide with degenerate codons for the core domain sequence, i.e., XW(G/A)X(G/A)CXP, as well as forward and reverse adaptor sequences for 113 downstream cloning. Complementary strands for the ssDNA mixture were synthesized in a primer extension reaction using T4 polymerase and the resulting products were inserted into the pMAL vector by In-Fusion. BL21(DE3) cells were transformed with the resulting plasmids and 120 colonies were selected for Sanger sequencing. Colonies giving viable sequences were collected and separated into ten groups based on expected product masses, with no duplicate masses present within the same group. Colonies were grown separately overnight and then pooled for growth and induction in 50 mL cultures of LB. For each polyculture, the remaining processing steps from growth to cyclization were identical to those used for preparation of individual precursor peptide substrates. The ten product mixtures were analyzed using LCMS, with EICs for the expected substrate and product (both linear and cyclic) masses. Native AMA1 was included in all experiments as a standard for cyclization efficiency and background. All products concluded to be present within the mixtures gave EIC signals not observed in the background nor in the absence of POPB treatment and corresponded to monoisotopic masses of the correct charge state. 114 6.4 Results 6.4.1 Enzyme and Substrate Preparation. Recombinant POPB enzyme from G. marginata was produced in yeast with an N-terminal myc epitope tag and purified on anti-c-myc agarose followed by anion exchange chromtography, as described previously [21]. As a source of precursor peptide substrates, a strategy was developed for their expression in E. coli. The coding sequence for the amanitin precursor peptide (AMA1) from G. marginata was expressed as a maltose-binding protein (MBP) fusion by cloning into the vector pMAL-c5x. This afforded high stability, yields, and tractability of the precursor peptides. After induction of expression, the MBP fusion proteins were purified from cell extracts on amylose resin. Treatment with Factor Xa protease released the GmAMA1 peptide from the C-terminus of MBP, and the MBP was then removed by precipitation in methanol. LCMS indicated the release of GmAMA1 from the fusion protein upon treatment with Factor Xa and formation of cyclo-IWGIGCNP upon addition of GmPOPB enzyme. Approximately 6 mg of precursor peptide was produced from one liter of bacterial culture (Figure 6.1). 6.4.2 Amino Acid Preferences for Cyclization. Site-directed mutagenesis of the wild-type AMA1 expression construct was used to generate a series of mutants with amino acid substitutions at each position of the core domain (sequence IWGIGCNP), excluding Pro8, which was presumed to be essential for POPB recognition. Reactions contained 5 μg POPB and 25 μg substrate, and ran for 4 hr at 37˚C. The results are summarized in Table 6.1 and the corresponding LCMS chromatograms are shown in Figures S6.1-S6.7. Cyclic products were produced from all 28 substrates. All substitutions to residues #1, #4, #6, and #7 gave yields of 115 >99% of cyclized core domain. Some substitutions at positions #2, #3, and #5 were less tolerated and gave higher yields of linear octamer product resulting from preference of Figure 6.1: Expression and Purification of the Amanitin Precursor Peptide. (A) SDS-PAGE of AMA1 precursor peptide expression and purification as a MBP fusion. (B) LCMS indicating relase of AMA1 peptide by Factor Xa protease and formation of cyclo-IWGIGCNP by POPB. the enzyme for hydrolysis over transpeptidation in the second catalytic step. Decreased yields were observed when residue #2 was changed to polar amino acids Ser or Asn, suggesting a preference for nonpolar residues at this position. POPB tolerated Ala but not Ser, Leu, or Asn at positions #3 and #5. The cyclic product yields for these less preferred substrates ranged from 18% (G3L) to 76% (G3S). While the incubation time used in these assays was intended to allow the reactions to run to completion, detectable amounts of full-length substrate remained in the assays with five of the mutants (G3S, G3L, G3N, G5S, G5L), all of which also gave reduced yields of cyclic product. 116 Position 1 1 1 1 1 2 2 2 2 2 3 3 3 3 3 4 4 4 4 4 5 5 5 5 5 6 6 6 6 6 7 7 7 7 7 AA Type wild type small, nonpolar small, polar large, nonpolar large, polar wild type small, nonpolar small, polar large, nonpolar large, polar wild type small, nonpolar small, polar large, nonpolar large, polar wild type small, nonpolar small, polar large, nonpolar large, polar wild type small, nonpolar small, polar large, nonpolar large, polar wild type small, nonpolar small, polar large, nonpolar large, polar wild type small, nonpolar small, polar large, nonpolar large, polar Residue Rxn Progress (%) Cyclic (%) Ile > 99 > 99 Ala > 99 > 99 Ser > 99 > 99 Leu > 99 > 99 Asn > 99 > 99 Trp > 99 > 99 Ala > 99 > 99 Ser > 99 32 Phe > 99 > 99 Asn > 99 46 Gly > 99 > 99 Ala > 99 > 99 Ser 72 76 Leu 33 18 Asn 85 63 Ile > 99 > 99 Ala > 99 > 99 Ser > 99 > 99 Leu > 99 > 99 Asn > 99 > 99 Gly > 99 > 99 Ala > 99 > 99 Ser 88 74 Leu 92 60 Asn > 99 64 Cys > 99 > 99 Ala > 99 > 99 Ser > 99 > 99 Leu > 99 98 Asn > 99 97 Asn > 99 > 99 Ala > 99 > 99 Ser > 99 > 99 Leu > 99 > 99 Gln > 99 > 99 Table 6.1: Tolerance of POPB for Amino Acid Substituions in the Core Region of AMA1. Wild-type sequnces are coded green, reactions that gave reduced yields of cyclic product are coded pink. Corresponding LCMS traces are shown in Figures S6.1-S6.7. 117 To test whether the reduced cyclization efficiency was due to reduced first-stage hydrolysis, 25mer forms (i.e., without the 10-amino acid leader) of four of the sequences (wild-type, W2S, G3L, and G5S) were tested as substrates. POPB cyclase is as efficient with the 25mer as with the native 35mer [21]. The same efficiencies in cyclization were observed with the 25mer substrates (Table S6.1). Thus, these substitutions resulted in poorer substrates for both hydrolysis and cyclization steps. 6.4.3 Cyclization of Sequences Containing Unusual Amino Acids. Amatoxins and phallotoxins contain up to five hydroxylations. Both groups of toxins have 4-hydroxyproline, which is critical for high affinity binding of α-amanitin to pol II [17]. The amatoxins also contain 6-hydroxytryptophan, which is the preferred site for attachment of antibodies in antibodyamanitin conjugates targeted against cancer cells [27]. It is not known whether the hydroxylations occur before or after cyclization by POPB. In either case, cyclizing the amanitin percursor with the Pro and Trp hydroxylations already in place would facilitate great progress towards the complete in vitro biosynthesis of α-amanitin, which to date has eluded chemical synthesis. Furthermore, the compatability of POPB with unusual amino acids such as Nmethylated amino acids and/or β-amino acids would expand the utility of POPB to make novel cyclic peptides. We chemically synthesized four additional substrates that contained the modified amino acids trans-4-hydroxyproline, 5-hydroxytryptophan, N-methylalanine, and β-alanine (an Fmoc derivative of 6-hydroxytryptophan was not commercially available). These substrates were prepared as the 25mer form lacking the N-terminal leader domain. All four of these substrates were cyclized by POPB (Figure 6.2). Reduced yields were observed from the substrate 118 containing N-methylalanine at position #3 of the core domain, which gave primarily linearized product. After 4 hr, 26% of the substrate containing 4-hydroxyproline remained, indicating that both hydrolysis and transpeptidation of this substrate was less efficient. POP enzymes achieve proline specificity through a ring stacking interaction between Pro and an active site Trp [28], and this interaction might have been adversely affected by the hydroxyl group. The results indicate that POPB can tolerate amino acids beyond the proteinogenic twenty. Figure 6.2: Cyclization of Peptides Containing Unusual Amino Acids. The modified residues are highlighted in red. Synthetic linear 25mers were incubated with POPB and the reactions analyzed by LCMS. Shown are overlaid EICs; substrate (S) signals are shown in green, cyclized core domains (C) in red, and linearized core domains (L) in blue. Values in the table are the amount of product present as cyclized core domain as a percentage of total cyclic + linear products. 119 6.4.4 Core Domain Length Requirement. Naturally occurring cycloamanides in Amanita species contain 6 to 10 amino acids [18,20]. To examine the allowed peptide lengths for cyclization by POPB in vitro, we prepared six precursor peptides in E. coli with core domains ranging from 6 to 16 residues. Core domains with less than 8 residues were prepared by removing amino acids from the wild-type AMA1 sequence. For longer sequences, Gly, Ala, and Val were added due to their small size and passive nature, and Ser was included in the 16mer sequence to avoid possible issues with water insolubility. Cyclization occurred for all tested substrates with longer core domains (9mer, 10mer, 12mer, and 16mer) (Figure 6.3). Longer sequences were less efficiently cyclized, but even the 16mer yielded 42% cyclic product with some unreacted substrate. Hexamer and heptamer core peptides were efficiently processed but only linear products were produced. 6.4.5 Synthesis of Naturally Occurring Cycloamanides. Amanita phalloides and A. bisporigera produce a number of homodetic monocyclic hexa- to decapeptides, of which six have been structurally characterized [18,20]. The known mushroom genomes predict that these fungi produce more than 50 additional cycloamanides [19,20]. We tested cyclization of POPB substrates containing the sequences of several cycloamanides produced by expression in E. coli, as well as sequences for the precursors of β-amanitin (i.e., α-amanitin in which Asp7 replaces Asn7), and two phallotoxins, phallacidin (PHA; core sequence AWLVDCP) and phalloidin (PHD; core sequence AWLATCP). As before, only linearized products from sequences shorter than eight residues were observed (i.e., CyalA, CylB, PHA, and PHD) (Table 6.2, Figure S6.8). The N-terminal leader peptide of the substrate containing the phallacidin (PHA) sequence was hydrolyzed to 120 Figure 6.3: POPB Products Produced from Substrates with Varying Core Domain Lengths. “% Cyclic” is the amount of product produced as cyclized core domain as a percentage of total cyclic + linear product. Shown are overlaid extracted ion chromatograms (EICs) for substrate (S) in green, cyclized core domains (C) in red, and linearized core domains (L) in blue. yield the 25mer, but no further hydrolase or cyclase activity was observed. The inability of POPB to cyclize these shorter sequences was unexpected, since Amanita mushrooms make cyclic hexapeptides (CylA) and heptapeptides (CylB and phallotoxins). Possible explanations are that other steps such as hydroxylation or epimerization occur before, and are required for, cyclization by POPB, or that the enzyme from Galerina has more limited substrate versatility than POPB from Amanita species. All naturally occurring sequences with at least eight residues were cyclized with good yields including the β-amanitin sequence (Table 6.2, Figure S6.8). The decamer antamanide sequence 121 was cyclized slowly, with less than 10% of the substrate being consumed after 4 hour incubation. Overall, these results show that POPB could be useful to produce at least some of the natural cycloamanides, which have immunosuppressant and other biological activities but are currently only available in limited quantities from mushroom extracts [29,30]. The CylD sequence (MLVFLPLP) gave cyclic and no linear product despite the presence of a bulky Leu residue at position #5, which caused reduced yields in the assays with AMA1 single mutants (Table 6.1). This result indicates that amino acid preferences for cyclization cannot be defined strictly by position in the substrate, but are instead influenced by overall sequence. For instance, mutations to the Gly residues in the α-amanitin sequence might have led to reduced yields due to a loss of flexibility in the sequence, while the turn-inducing effect of the internal Pro in the CylD sequence might facilitate cyclization. Table 6.2: Cyclization of Naturally Occurring Cycloamanides. Corresponding LCMS traces are shown in Figure S6.8. Footnotes: aSubstrate containing the PHA sequence was hydrolyzed to the 25mer form but no futher processing occurred. b24 hour incubation. 122 6.4.6 Cyclization of the Phalloidin Sequence with D-threonine. All of the natural phallotoxins contain one D amino acid at position #5, either D-Asp in phallacidin or D-Thr in phalloidin [18]. Introduction of D-amino acids into peptide sequences can improve the efficiency of cyclization [31]. Since no cyclization was observed with the phallotoxin precursor substrates containing all L amino acids (i.e., PHA and PHD), we hypothesized that epimerization might occur biosynthetically prior to cyclization and therefore promote cyclization. A substrate containing the phalloidin sequence (AWLATCP) with D-Thr was produced synthetically. The presence of DThr resulted in formation of a significant level (13%) of the corresponding cyclic product whereas the all L version showed only hydrolysis of the substrate to the linear octapeptide and no cyclization (Figure 6.4). This demonstrates that POPB can cyclize peptides smaller than eight residues, albeit at low efficiency under our standard conditions, and suggests that epimerization in the phallotoxins might occur prior to cyclization, i.e., at the precursor peptide stage or after removal of the leader peptide. 6.4.7 Cyclic Peptide Library Production. As a more rapid strategy for assessing the substrate versatility of POPB, we constructed a model library of cyclic peptides and processed them in batches. A trial experiment was first performed using ten of the previously prepared AMA1 substrates with single substitutions that gave products with different masses and retention times. Inoculating the growth medium with ten individual colonies directly from agar plates resulted in inconsistent formation of the expected products, likely due to unequal growth rates among the E. coli strains or differences in inoculation sizes. Consistent production of all ten products could be obtained by first growing the cultures separately before pooling for induction. After induction, the pooled cultures were processed through POPB cyclization en masse. The overall scheme is 123 Figure 6.4: LCMS Comparing POPB Products Produced from Substrates Containing the PHD Sequence with either L-Thr or D-Thr. Shown are overlaid extracted ion chromatograms (EICs); linear products (L) are shown in blue, cyclic products (C) in red, and the observed m/z values are indicated. illustrated in Figure 6.5. The results showing expression of all ten cyclic peptides produced in a single batch are shown in Figure S6.9. For a randomized library, DNA molecules with degenerate core domains and conserved leader and follower domains were synthesized as single-stranded DNA with mixed nucleotides in the core sequence to encode X-W-(G/A)-X-(G/A)-C-X-P, where X is any amino acid. X was encoded by NNK, where N is any nucleotide, and K is guanine or thymidine. This allows encoding of all possible amino acids but eliminates two of the three stop codons. Positions #3 and #5 were encoded as either Gly or Ala (codon G[C/G]A) to maximize cyclization efficiency, and the Trp2 and Cys6 residues were maintained to permit the future possibility of tryptathionine 124 formation. Complementary strands for the ssDNA template mixture were synthesized in a primer extension reaction, the products were inserted into the same pMAL expression vector used for expression of the individual substrates, and E. coli cells were transformed with the resulting plasmids. Transformants were randomly selected and their plasmid inserts sequenced. Of 120 inserts sequenced, 79 (66%) gave viable sequences encoding potential POPB substrates. The 41 nonviable sequences contained frameshifts, deletions, stop codons, or sequence errors likely introduced during complementary strand synthesis. Figure 6.5: Scheme for Generating Mixed Cyclic Peptide Libraries. 125 The colonies expressing viable substrates were grown separately in overnight starter cultures and then pooled into polycultures for expression in ten separate groups, chosen so that no two cyclic products of the same mass would be present within each final mixture. The remaining preparation steps of protein extraction, column purification, isolation of the 35mer precursor peptide substrates from MBP-fusions, and in vitro cyclization with POPB were then carried out en masse as outlined in Figure 6.5. The resulting peptide mixtures were analyzed by LCMS and extracted ion analysis for the predicted masses of the cyclic peptides. Cyclic products from 58 of the 79 substrates were confirmed within the mixtures (Figures S6.10). All 20 proteinogenic amino acids were represented among the product sequences. The 21 cyclic products that were expected but absent from the mixtures fell into two categories: either they contained charged residues (17 total) or they contained Tyr at the first position (4 total). However, other substrates with these same characteristics were successfully cyclized and therefore no firm rules for POPB substrate requirements could be established. No full-length substrate, 25mer intermediate, or linear product were detected for many of the charged compounds and for none of the peptides containing Tyr. The absence of these compounds in the final cyclized pool might be due to problems during E. coli expression or purification of the precursor peptides and not POPB cyclization. 126 6.5 Discussion Tables S6.2 and S6.3 include a list of the core peptide sequences tested (127 total) and the cyclic peptides successfully produced (100 total) in this study. The pilot library study demonstrated the feasibility of producing cyclic peptides in batches, which in principle could be scaled up to at least hundreds. Additional time could be saved by not prescreening the plasmid inserts by DNA sequencing, or by not growing the strains separately before induction. POPB is a versatile and efficient peptide macrocyclase that could be used to make billions of novel cyclic peptides of 8-16 amino acids including unusual amino acids. Amanitin has recently been shown to be a promising “warhead” in antibody-drug conjugates against colorectal and prostate cancers [27,32], but currently the only source of amanitin is from mushrooms collected in the wild. Our demonstration that key hydroxylations (on Pro and Trp) that occur in native amatoxins and phallotoxins can be preintroduced into the substrates of POPB might also facilitate the development of a synthetic or semisynthetic approach to α-amanitin production. 127 APPENDIX 128 APPENDIX wild-type 35mer 25mer > 99 > 99 W2S 32 ± 3 30 ± 4 G3L 18 ± 5 19 ± 3 G5S 74 ± 4 77 ± 5 Table S6.1: Compared Cyclization Yields of 35mer and 25mer Substrates. Values are percentage of total core domain present as cyclic peptide in the final products. Figure S6.1: Effect of Single Amino Acid Substitions on Cyclization by POPB at Position 1 of the AMA1 Core Domain. Each trace shows overlaid EICs before (top) or after (bottom) POPB treatment. Substrate (S) signals are shown in green, cyclized core domains (C) in red, and linearized core domains (L) in blue. 129 Figure S6.2: Effect of Single Amino Acid Substitions on Cyclization by POPB at Position 2 of the AMA1 Core Domain. Each trace shows overlaid EICs before (top) or after (bottom) POPB treatment. Substrate (S) signals are shown in green, cyclized core domains (C) in red, and linearized core domains (L) in blue. 130 Figure S6.3: Effect of Single Amino Acid Substitions on Cyclization by POPB at Position 3 of the AMA1 Core Domain. Each trace shows overlaid EICs before (top) or after (bottom) POPB treatment. Substrate (S) signals are shown in green, cyclized core domains (C) in red, and linearized core domains (L) in blue. 131 Figure S6.4: Effect of Single Amino Acid Substitions on Cyclization by POPB at Position 4 of the AMA1 Core Domain. Each trace shows overlaid EICs before (top) or after (bottom) POPB treatment. Substrate (S) signals are shown in green, cyclized core domains (C) in red, and linearized core domains (L) in blue. 132 Figure S6.5: Effect of Single Amino Acid Substitions on Cyclization by POPB at Position 5 of the AMA1 Core Domain. Each trace shows overlaid EICs before (top) or after (bottom) POPB treatment. Substrate (S) signals are shown in green, cyclized core domains (C) in red, and linearized core domains (L) in blue. 133 Figure S6.6: Effect of Single Amino Acid Substitions on Cyclization by POPB at Position 6 of the AMA1 Core Domain. Each trace shows overlaid EICs before (top) or after (bottom) POPB treatment. Substrate (S) signals are shown in green, cyclized core domains (C) in red, and linearized core domains (L) in blue. 134 Figure S6.7: Effect of Single Amino Acid Substitions on Cyclization by POPB at Position 7 of the AMA1 Core Domain. Each trace shows overlaid EICs before (top) or after (bottom) POPB treatment. Substrate (S) signals are shown in green, cyclized core domains (C) in red, and linearized core domains (L) in blue. 135 Figure S6.8: LCMS Traces of Naturally Occurring Cycloamanide Core Regions Cyclized by POPB. Substrate (S) signals are shown in green, cyclized core domain (C) in red, and linearized core domains (L) in blue. Truncated 25mer peptide was the final product from the phallacidin (PHA) substrate (signal in purple). 136 Figure S6.9: Simultaneous Production of Ten Cyclic Peptides Using POPB. The peptides were all based on the α-amanitin core sequence and correspond to those shown in Supplementary Figures 6.1-6.7. Shown are overlaid EICs for their expected masses. 137 Figure S6.10: Batch Production of Cyclic Peptides Using POPB. In each batch, six to eight E. coli strains expressing different 35mer precursor peptides were grown and processed en masse through POPB treatment. Sequences numbered were observed in the EICs; sequences encoded in pink were not seen. 138 Figure S6.10 (cont’d) 139 Figure S6.10 (cont’d) 140 Figure S6.10 (cont’d) 141 Figure S6.10 (cont’d) 142 Individual Assays AWGIGCNP AWLA(D-Thr)CP FFVPPAFFPP I(5-hydroxyTrp)GIGCNP IAGIGCNP IFGIGCNP INGIGCNP ISGIGCNP IW(N-methylAla)IGCNP IW(β-Ala)IGCNP IWAIGCNP IWGAGCNP IWGAGIGAGCNP IWGAVSGIGAVSGCNP IWGGIGGCNP IWGIACNP IWGIGANP IWGIGCAP IWGIGCDP IWGIGCLP IWGIGCN(4-hydroxyPro) IWGIGCNP IWGIGCQP IWGIGCSP IWGIGGCNP IWGIGLNP IWGIGNNP IWGIGSNP IWGILCNP IWGINCNP Individual Assays IWGISCNP IWGLGCNP IWGNGCNP IWGSGCNP IWLIGCNP IWNIGCNP IWSIGCNP LWGIGCNP MLGFLPLP MLGFLVLP NWGIGCNP SWGIGCNP Library AWAAGCSP AWADGCRP AWGSGCSP AWGVGCMP CWALGCFP CWAVACAP CWGGGCQP FWGSACFP FWGTGCFP GWGAACCP GWGFGCFP HWGHACVP HWGSGCRP IWAHACVP IWALACVP IWAYGCYP IWGWGWGP Library KWGVACNP LWAQGCYP LWGFACGP LWGLACQP LWGMGCWP LWGSGCSP LWGVACPP MWGMACFP NWGGACSP NWGLACGP QWGAACLP QWGRGCLP RWANACLP RWGHACYP SWACGCSP SWAHGCHP SWAIACLP SWALGCVP SWASGCLP SWGAGCEP SWGQACIP SWGQGCHP SWGTACVP SWGTGCYP TWGAGCQP TWGGGCMP VWAFACAP VWAFGCFP VWAMGCTP VWASACVP Library VWATACRP VWGCGCGP VWGIACTP VWGPGCVP VWGRGCQP WWAGACLP WWGCACLP WWGGGCRP YWASACAP YWAYGCVP YWGQGCSP Table S6.2: Alphabetical List of Cyclic Peptides Produced with POPB. 143 Individual Assays AWLATCP AWLVDCP IWGIGCP IWGIGP SFFFPIP VFFAGP Library DWAPACFP DWARACSP DWGSGCVP EWAAACPP HWGRGCLP IWGEGCWP LWACACKP PWGPACHP RWAAACAP RWALACVP RWATACKP RWGCGCLP RWGLACCP SWARACVP SWGRACKP SWGRGCSP WWAKGCYP YWAIACNP YWAQACGP YWAVACTP YWGVACAP Table S6.3: Alphabetical List of Peptides Not Efficiently Cyclized by POPB. 144 WORKS CITED 145 WORKS CITED 1. Driggers EM, Hale SP, Lee J, and Terrett NK. (2008). The exploration of macrocycles for drug discovery - an underexploited structural class. Nat. Rev. Drug Discov. 7(7): 608624. 2. Rezai T, Yu B, Millhauser GL, Jacobson MP, and Lokey RS. (2006). Testing the conformational hypothesis of passive membrane permeability using synthetic cyclic peptide diastereomers. J. Am. Chem. Soc. 128(8): 2510-2511. 3. Fairlie DP, Tyndall JD, Reid RC, Wong, A.K., Abbenante, G., Scanlon, M.J., March, D.R., Bergman, D.A., Chai, C.L.L., and Burkett, B.A. (2000). Conformational selection of inhibitors and substrates by proteolytic enzymes: Implications for drug design and and polypeptide processing. J. Med. Chem. 43(7): 1271-1281. 4. Millward SW, Fiacco S, Austin RJ, and Roberts RW. (2007). Design of cyclic peptides that bind protein surfaces with antibody-like affinity. ACS Chem. Biol. 2(9): 625-634. 5. Zorzi A, Deyle K, and Heinis C. (2017). Cyclic peptide therapeutics: past, present and future. Curr. Opin. Chem. Biol. 38: 24-29. 6. Upadhyaya P, Qian Z, Selner NG, Clippinger SR, Wu Z, Briesewitz R, and Pei D. (2015). Inhibition of Ras signaling by blocking Ras-effector interactions with cyclic peptides. Angew. Chem. Int. Ed. Engl. 54(26): 7602-7606. 7. Kling A, et al. (2015). Targeting DnaN for tuberculosis therapy using novel griselimycins. Science. 348(6239): 1106-1112. 8. Ji Y, Majumder S, Millard M, Borra R, Bi T, Elnagar AY, Neamati N, Shekhtman A, and Camarero JA. (2013). In vivo activation of the p53 tumor suppressor pathway by an engineered cyclotide. J. Am. Chem. Soc. 135(31): 11623-11633. 9. Zipperer A, et al. (2016). Human commensals producing a novel antibiotic impair pathogen colonization. Nature. 535(7613): 511-516. 10. Marsault E, and Peterson ML. (2011). Macrocycles are great cycles: applications, opportunities, and challenges of synthetic macrocycles in drug discovery. J. Med. Chem. 54(7): 1961-2004. 11. Arnison PG et al. (2013). Ribosomally synthesized and post-translationally modified peptide natural products: overview and recommendations for a universal nomenclature. Nat. Prod. Rep. 30(1): 108-160. 146 12. Hallen HE, Luo H, Scott-Craig JS, and Walton JD. (2007). Gene family encoding the major toxins of lethal Amanita mushrooms. Proc. Natl. Acad. Sci. U.S.A. 104(48): 1909719101. 13. Umemura M, et al. (2014). Characterization of the biosynthetic gene cluster for the ribosomally synthesized cyclic peptide ustiloxin B in Aspergillus flavus. Fungal Genet. Biol. 68: 23-30. 14. Ding W, Liu WQ, Jia Y, Li Y, van der Donk WA, and Zhang Q. (2016). Biosynthetic investigation of phomopsins reveals a widespread pathway for ribosomal natural products in Ascomycetes. Proc. Natl. Acad. Sci. U.S.A. 113(13): 3521-3526. 15. Lindell TJ, Weinberg F, Morris PW, Roeder RG, and Rutter WJ. (1970). Specific inhibition of nuclear RNA polymerase II by alpha-amanitin. Science. 170(3956): 447-449. 16. Lengsfield AM, Löw I, Wieland T, Dancker P, and Hasselbach W. (1974). Interaction of phalloidin with actin. Proc. Natl. Acad. Sci. U.S.A. 71(7): 2803-2807. 17. Bushnell DA, Cramer P, Kornberg RD. (2001). Structural basis of transcription: αAmanitin-RNA polymerase II cocrystal at 2.8 Å resolution. Proc. Natl. Acad. Sci. U.S.A. 99(3): 1218-1222. 18. Wieland, T. (1986). Peptides of Poisonous Amanita Mushrooms. Springer: New York. 19. Li P, Deng W, and Li T. (2014). The molecular diversity of toxin gene families in lethal Amanita mushrooms. Toxicon. 83(1): 59-68. 20. Pulman JA, Childs KL, Sgambelluri RM, and Walton JD. (2016). Expansion and diversification of the MSDIN family of cyclic peptide genes in the poisonous agarics Amanita phalloides and A. bisporigera. BMC Genomics. 17(1): 1038. 21. Luo H, Hong SY, Sgambelluri RM, Angelos E, Li X, and Walton JD. (2014). Peptide macrocyclization catalyzed by a prolyl oligopeptidase involved in α-amanitin biosynthesis. Chem. Biol. 21(12): 1610-1617. 22. May JP and Perrin DM. (2007). Tryptathionine bridges in peptide synthesis. Biopolymers. 88(5): 714-724. 23. Czekster CM and Naismith JH. (2017). Kinetic landscape of a peptide bond-forming prolyl oligopeptidase. Biochemistry. 56(15): 2086-2095. 24. McIntosh JA, Robertson CR, Agarwal V, Nair SK, Bulaj GW, and Schmidt EW. (2010). Circular logic: nonribosomal peptide-like macrocyclization with a ribosomal peptide catalyst. J. Am. Chem. Soc. 132(44): 15499-15501. 147 25. Chekan JR, Estrada P, Covello PS, and Nair SK. (2017). Characterization of the macrocyclase involved in the biosynthesis of RiPP cyclic peptides in plants. Proc. Natl. Acad. Sci. U.S.A. 114(25): 6551-6556. 26. Nguyen GK, Wang S, Qiu Y, Hemu X, Lian Y, and Tam JP. (2014). Butelase 1 is an Asxspecific ligase enabling peptide macrocyclization and synthesis. Nat. Chem. Biol. 10(9): 732-738. 27. Liu Y, Zhang X, Han C, Wan G, Huang X, Ivan C, Jiang D, Rodriquez-Aguayo C, LopezBerestein G, Rao PH, Maru DM, Pahl A, He X, Sood AK, Ellis LM, Anderl J, and Lu X. (2015). TP53 loss creates therapeutic vulnerability in colorectal cancer. Nature. 520(7549): 697-701. 28. Fülöp V, Böckei Z, and Polgár L. (1998). Prolyl oligopeptidase: an unusual beta-propeller domain regulates proteolysis. Cell. 94(2): 161-170. 29. Azzolin L, Antolini N, Calderan A, Ruzza P, Sciacovelli M, Marin O, Mammi S, Bernardi P, and Rasola A. (2011). Antamanide, a derivative of Amanita phalloides, is a novel inhibitor of the mitochondrial permeability transition pore. PLoS One. 6(1): e16280. 30. Wieczorek Z, Siemion IZ, Zimecki M, Bolewska-Pedyczak E, and Wieland T. (1993). Immunosuppressive activity in the series of cycloamanide peptides from mushrooms. Peptides. 14(1):1-5. 31. White CJ and Yudin AK. (2011). Contemporary strategies for peptide macrocyclization. Nat. Chem. 3(7): 509-524. 32. Hechler T, Kulke M, Müller C, Pahl A, and Anderl J. (2014). Amanitin-based antibodydrug conjugates targeting the prostate-specific membrane antigen. Cancer Res. 74(19 Suppl): 664. 148