Metabolite production in endophytes

Abstract

The present invention relates to nucleic acids encoding amino acid sequences for the biosynthesis of janthitrem in janthitrem producing endophytes. The present invention also relates to constructs and vectors including such nucleic acids, and related polypeptides, regulatory elements and methods.

Claims

1. A recombinant nucleic acid construct comprising a heterologous promoter operably linked to a polynucleotide sequence encoding a JtmD protein having aromatic prenyl transferase activity and comprising an amino acid sequence having at least 95% amino acid sequence identity to the amino acid sequence as set forth in SEQ ID NO: 11.

2. A host cell comprising the recombinant nucleic acid construct of claim 1.

3. The recombinant nucleic acid construct of claim 1, wherein said polynucleotide sequence has at least 95% nucleotide sequence identity to the nucleotide sequence as set forth in SEQ ID NO: 10.

4. The recombinant nucleic acid construct of claim 1, wherein said JtmD protein has the amino acid sequence as set forth in SEQ ID NO: 11.

5. The recombinant nucleic acid construct of claim 1, wherein said polynucleotide sequence has the nucleotide sequence as set forth in SEQ ID NO: 10.

Description

BRIEF DESCRIPTION OF THE DRAWINGS/FIGURES

(1) In the Figures:

(2) FIG. 1: Epoxy-janthitrem I and Lolitrem B. Epoxy-janthitrem I is a paxilline-like indole diterpene that exhibits structural similarity to lolitrem B. Structure, chemical formula (C39H51NO7) and exact mass (645.3665) of 11,12-epoxyjanthitrem G (epoxy-janthitrem I) from Tapper et al. 2004.

(3) FIG. 2: UPGMA phenogram of genetic relationships among endophytes in ryegrass accessions of diverse origins in relation to reference Epichloë species. Genetic identity was measured across 18 SSR loci using the Dice coefficient (Kaur et al, 2015). LpTG-3 and LpTG-4 endophyte strains are genetically distinct from other asexual Epichloë identified including Epichloë festucae var. lolii and LpTG-2 (Hettiarachchige et al. 2015).

(4) FIG. 3: Genome survey sequence analysis was used to determine the presence/absence profiles of the genes responsible for peramine, ergovaline and lolitrem B biosynthesis in endophyte strains representing each of the four taxa observed to form associations with perennial ryegrass (Epichloë festucae var. lolii, LpTG-2, LpTG-3 and LpTG-4). Strains that do not produce lolitrem B have a deletion in the third (ItmE-ItmJ) lolitrem B gene cluster. Adapted from Davidson et al. 2012.

(5) FIG. 4: NEA12 PacBio contig 3 is 247 475 bp in length and has 13 predicted and known genes in four clusters. Cluster 1 (ItmG, ItmS, ItmM, ItmK), Cluster 2 (ItmP, ItmQ, ItmF, ItmC, ItmB), Cluster 3 (PP01, PP02) and Cluster 4 (jtmD and jtmO). Light grey arrows display predicted and known genes and their orientation. The locations of the pks pseudogene, transposase with a MULE domain (PP03), Helitron helicase-like transposable element (TE), and three AT-rich regions are also shown. PP=predicted protein; TE=transposable element; ip=pseudogene.

(6) FIG. 5: Genomes of representative strains of Epichloë sp. endophytes from 4 taxa Epichloë festucae var. lolii (NEA2, NEA6, NEA10), LpTG-2 (NEA11), LpTG-3 (NEA12, AR37, 15310, 15311), LpTG-4 (E1) and FaTG-3 (NEA23) were mapped to NEA12 PacBio contig 3. A c. 177436 bp region (c.70039 bp 247475 bp) of the genome unique to janthitrem producing taxa LpTG-3 and LpTG-4 was identified. Within this region there are two gene clusters containing candidate genes (PP01, PP02, jtmD and jtmJ) predicted to be associated with janthitrem biosynthesis in Epichloë endophytes. Endophyte strains from the taxa LpTG-3 and LpTG-4 all contain candidate genes for janthitrem biosynthesis, while for endophytes from Epichloë festucae var. lolii, LpTG-2 and FaTG-3 this region is absent. DNA reads generated using Illumina sequencing technology were mapped with Gydle ‘nuclear’ aligner version 3.2.1. Reads were mapped with settings: 1 50 (length of overlap); s 25 (sensitivity); k 13 (kmer length); m 6 (maximum number of mis-matches); F 3 (filter settings). Alignments were visualised with Gydle program Vision version 2.6.14.

(7) FIG. 6: In planta expression of NEA12 genome PacBio contig 3 genes. Genomes of representative strains of Epichloë sp. endophytes from LpTG-3 (AR37) and Epichloë festucae var. lolii (SE) were mapped to the 247475 bp NEA12 PacBio contig 3. In planta expression of candidate genes for janthitrem biosynthesis in LpTG-3, LpTG-4 and Epichloë festucae var. lolii was determined using RNA-seq analysis of perennial ryegrass-endophyte association transcriptome data (refer to key below). DNA and RNA reads were mapped with Gydle ‘nuclear’ aligner version 3.2.1. Reads were mapped with settings:150 (length of overlap); s 25 (sensitivity); k 13 (kmer length); m 6 (maximum number of mis-matches); F 3 (filter settings). Alignments were visualised with Gydle program Vision version 2.6.14. Expression of Cluster 2 (ltmP, ltmQ, ltmF, ltmC, ltmB), Cluster 1 (ltmG, ltmS, ltmM, ltmK), Cluster 3 (PP01, PP02) and Cluster 4 (jtmD and jtmO) genes was observed for endophyte strains NEA12 and E1 in planta. Cluster 3 and Cluster 4 genes are not present in the Epichloë festucae var. lolii (SE) genome, expression of these genes was not observed by SE in planta.

(8) Key to FIG. 6

(9) TABLE-US-00001 Row Genome/Transcriptome Taxon (Strain) Experiment Treatment 1 Genome LpTG-3 (AR37) genome survey n.a. sequence analysis 2 In planta LpTG-3 (NEA12) seedling growth and post imbibition (0 transcriptome maturation hours) 3 In planta LpTG-3 (NEA12) seedling growth and 10 day old transcriptome maturation seedlings (10 days) 4 In planta LpTG-4 (E1) transcriptome atlas leaf transcriptome 5 In planta LpTG-4 (E1) transcriptome atlas stigma transcriptome 6 Genome Epichlo{umlaut over (e)} festucae genome survey n.a. var. lolii (SE) sequence analysis 7 In planta Epichlo{umlaut over (e)} festucae seedling growth and post imbibition (0 transcriptome var. lolii (SE) maturation hours) 8 In planta Epichlo{umlaut over (e)} festucae seedling growth and 10 day old transcriptome var. lolii (SE) maturation seedling (10 days)

(10) FIG. 7: Nucleotide sequence for the PP01 gene (Sequence ID No 1). The coding sequence for the predicted PP01 protein is highlighted in grey (Sequence ID No 2). The complete nucleotide sequence for the PP01 gene was identified by mapping RNA reads from the in planta (Alto-NEA12) transcriptome data described in FIG. 6 followed by extraction of the DNA sequence from NEA12 PacBio contig 3. Nucleotides shown in lowercase were not observed in the analysis of the Alto-NEA12 transcriptome dataset.

(11) FIG. 8: PP01 is predicted to be a cytochrome P450 monoxygenase 387 amino acids in length. Shown here is the alignment of predicted amino acid sequences for PP01 from LpTG-3 strain NEA12 (Sequence ID No 3.) and HirsuteIla minnesotensis (KJZ77225 amino acids 3-379) (Sequence ID No 4). Protein identity: 258/387 (66.7%); Protein similarity: 304/387 (78.6%); Gaps: 10/387 (2.6%). Sequences were aligned using EMBOSS Needle.

(12) FIG. 9: Bootstrap consensus tree generated through Maximum Likelihood analysis of the predicted amino acid sequence of PP01 from LpTG-3 (NEA12) and the top 6 BLASTp hits in the NCBI database. Multiple alignment of complete predicted protein sequences was performed using ClustalW with default parameters. To construct tree topology, maximum likelihood (ML) was used as implemented in MEGA 6 with default parameters and 500 bootstrap replicates. Branches with bootstrap values of greater than 70% from 500 bootstrap replications are marked next to each branch. Genbank accession numbers for each protein sequence is provided in each tree diagram. PP01 exhibits sequence similarity to cytochrome P450 monoxygenases: KJZ77225.1 [68%; Hirsutella minnesotensis 3608]; EQL02233.1 [57%; Ophiocordyceps sinensis CO18]; KND87478.1 [53%; Tolypocladium ophioglossoides CBS 100239]; OAQ66296.1 [50%; Pochonia chlamydosporia 170]; KOM22171.1 [55%; Ophiocordyceps unilateralis]; XP_013947710.1 [48%; Trichoderma atroviride IMI 206040].

(13) FIG. 10. Nucleotide sequence for the PP02 gene (Sequence ID No 5). The coding sequence for the predicted PP02 protein is highlighted in grey (Sequence ID No 6). Start (ATG) and stop (TGA) codon sequences are shown in bold. Untranslated 5′ and 3′ sequences are shown in lowercase. The complete nucleotide sequence for the PP02 gene was identified by mapping RNA reads from the in planta (Alto-NEA12) transcriptome data described in FIG. 6 followed by extraction of the DNA sequence from NEA12 PacBio contig 3.

(14) FIG. 11. PP02 is predicted to be a membrane bound O-acyl transferase (MBOAT) protein 315 amino acids in length. Shown here is the alignment of predicted amino acid sequences for PP02 from LpTG-3 strain NEA12 (Sequence ID No 7) and Oidiodendron maius Zn (KIM95229) (Sequence ID No 8). Protein identity: 110/412 (26.7%); Protein similarity: 165/412 (40.0%); Gaps: 118/412 (28.6%). Within the predicted MBOAT domain (shown in bold) the two sequences exhibit protein identity of 42% (37/89) and protein similarity of 61% (54/89). Sequences were aligned using EMBOSS Needle.

(15) FIG. 12. Bootstrap consensus tree generated through Maximum Likelihood analysis of the predicted amino acid sequence of PP02 from LpTG-3 (NEA12) and the top 5 BLASTp hits in the NCBI database. Multiple alignment of complete predicted protein sequences was performed using ClustalW with default parameters. To construct tree topology, maximum likelihood (ML) was used as implemented in MEGA 6 with default parameters and 500 bootstrap replicates. Branches with bootstrap values of greater than 70% from 500 bootstrap replication are marked next to each branch. Genbank accession numbers for each protein sequence is provided in each tree diagram. PP02 exhibits sequence similarity to MBOAT proteins: KIM95229.1 [33%; Oidiodendron maius Zn]; KZL85868.1[30%; Colletotrichum incanum]; CCX05903.1 [30%; Pyronema omphalodes CBS 100304]; KZP09605.1 [29%; Fibulorhizoctonia sp. CBS 109695]; XP_007593790.1 [31%; Colletotrichum fioriniae PJ7].

(16) FIG. 13. Nucleotide sequence for the jtmD gene (Sequence ID No 9). The coding sequence for the predicted JtmD protein is highlighted in grey (Sequence ID No 10). Start (ATG) and stop (TGA) codon sequences are shown in bold. Untranslated 5′ and 3′ sequences are shown in lowercase. The complete nucleotide sequence for the jtmD gene was identified by mapping RNA reads from the in planta (Alto-NEA12) transcriptome data described in FIG. 6 followed by extraction of the DNA sequence from NEA12 PacBio contig 3.

(17) FIG. 14. JtmD is predicted to be an aromatic prenyl transferase 420 amino acids in length (Sequence ID No 11). JtmD exhibits highest homology to a predicted protein from Ophiocordyceps unilateralis (KOM22681.1) (Sequence ID No 12). Protein identity: 264/420 (62.9%); Protein similarity: 334/420 (79.5%); Gaps: 26/420 (6.2%). Sequences were aligned using EMBOSS Needle.

(18) FIG. 15. Bootstrap consensus tree generated through Maximum Likelihood analysis of the predicted amino acid sequence of JtmD from LpTG-3 (NEA12) and the top 11 BLASTp hits in the NCBI database. Multiple alignment of complete predicted protein sequences was performed using ClustalW with default parameters. To construct tree topology, maximum likelihood (ML) was used as implemented in MEGA 6 with default parameters and 500 bootstrap replicates. Branches with bootstrap values of greater than 70% from 500 bootstrap replication are marked next to each branch. Genbank accession numbers for each protein sequence is provided in each tree diagram. JtmD exhibits amino acid sequence identity to aromatic prenyl transferases: KOM22681.1 [67%; O. unilateralis]; AGZ20478.1 [49%; P. janthinellum]; AAK11526.2 [46%; P. paxilli]; KOS22745.1 [50%; E. webers]; CEJ54109.1 [47%; P. brasilianum]; BAU61555.1 [31%; P. simplicissimum]; AGZ20194.1 [31%; P. crustosum]; KZF25225.1 [33%; Xylona heveae TC161]; KGO76903.1 [30%; P. italicum]; KJK61458.1 [31%; Aspergillus parasiticus SU-1]; CAP53937.2[[31%; Aspergillus flavus].

(19) FIG. 16. Nucleotide sequence for the jtmO gene (Sequence ID No 13). The coding sequence for the predicted JtmO protein is highlighted in grey (Sequence ID No 14). Start (ATG) and stop (TAG) codon sequences are shown in bold. Untranslated 5′ and 3′ sequences are shown in lowercase. The complete nucleotide sequence for the jtmO gene was identified by mapping RNA reads from the in planta (Alto-NEA12) transcriptome data described in FIG. 6 followed by extraction of the DNA sequence from NEA12 PacBio contig 3.

(20) FIG. 17. JtmO is predicted to be a FAD-binding oxidoreductase 479 amino acids in length (Sequence ID No 15). JtmO exhibits highest homology to a predicted protein (6-hydroxy-D-nicotine oxidase) from Escovopsis weberi (KOS22754.1) (Sequence ID No 16). Protein identity: 271/481 (56.3%); Protein similarity: 344/481 (71.5%); Gaps: 39/481 (8.1%). Sequences were aligned using EMBOSS Needle.

(21) FIG. 18. Bootstrap consensus tree generated through Maximum Likelihood analysis of the predicted amino acid sequence of JtmO from LpTG-3 (NEA12) and LpTG-4 (E1) and the top 6 BLASTp hits in the NCBI database. Multiple alignment of complete predicted protein sequences was performed using ClustalW with default parameters. To construct tree topology, maximum likelihood (ML) was used as implemented in MEGA 6 with default parameters and 500 bootstrap replicates. Branches with bootstrap values of greater than 70% from 500 bootstrap replication are marked next to each branch. Genbank accession numbers for each protein sequence is provided in each tree diagram. JtmO exhibits amino acid sequence similarity to FAD-binding oxidoreductases: KOS22754.1 [56%; Escovopsis webers]; AGZ20488.1 [52%; P. janthinellum]; ADO29935.1 [49%; P. paxilli]; BAU61564.1 [43%; P. simplicissimum]; AGZ20199.1 [43%; P. crustosum]; EON68203.1 [Coniosporium apollinis].

(22) FIG. 19. LC-ESI-FTMS extracted ion chromatogram of metabolites observed in perennial ryegrass- LpTG-3 associations, collected from 0-20 min in positive ionisation mode (ESI+).

(23) FIG. 20. Proposed pathway for epoxy-janthitrem biosynthesis. The suggested scheme follows the indole-diterpene biosynthetic pathway, illustrating a parsimonious route to epoxy-janthitrem I (11, 12-epoxjanthitrem G) and its variants (epoxy-janthitrems II-IV). All metabolites were observed by LC-MS/MS (FIG. 19).

(24) FIG. 21. Nucleotide sequence of jtmD (Sequence ID No 17). Gene sequences selected for generation of RNAi silencing vectors are highlighted: Gene sequences selected for cassette 2, 3 and 4 are shown in italics (Sequence ID No 18)., underlined (Sequence ID No 19). and in bold respectively (Sequence ID No 20).

(25) FIG. 22. Schematic diagram of gene silencing vectors. To generate the entry clones, gene cassettes [inverted repeats of candidate gene sequences, separated by a 147 bp spacer (cutinase gene intron from M. grisea) and containing attB1 and attB2 sites], were cloned into the pDONR 221 vector using BP clonase (Invitrogen, USA). The Gateway™-enabled destination vector (pEND0002) was constructed through modifications of the T-DNA region of pPZP200 containing hph gene (selectable marker) under the control of trpCP (Aspergillus nidulans tryptophan biosynthesis promoter) and trpCT (A. nidulans tryptophan biosynthesis terminator and first reading frame A [RFA-A] cassette (gateway) under the control of gpdP (A. nidulans glyceraldehyde-3-phosphate dehydrogenase promoter) and trpCT (A. nidulans tryptophan biosynthesis terminator). The final RNA silencing vectors were produced by LR clonase reaction between an entry vector and the pEND002 vector.

(26) FIG. 23. Fungal protoplast regeneration. A. Regeneration of fungal protoplasts without hygromycin selection, assessment of protoplast viability. B. Regeneration of fungal protoplasts transformed with RNA silencing vector on hygromycin selection (arrows indicate individual colonies). C. Recovery of E1 strains carrying an RNA silencing vector on hygromycin selection.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

(27) Identification of Genes for Janthitrem Biosynthesis in LpTG-3 Endophyte Strain NEA12

(28) Whole genome sequence analysis was used to identify candidate genes for janthitrem biosynthesis in the NEA12 genome. The protein sequences LtmE and LtmJ from Standard Endophyte (SE) strain were used as query sequences to search the predicted protein database derived from the NEA12 genome. Using this approach, BLASTp searches yielded 13 putative LtmE protein homologues and 26 putative LtmJ protein homologues in the library of predicted NEA12 proteins.

(29) The NEA12 genome is expected to have predicted LtmE and ItmJ protein homologues in common with the SE strain. However, candidates for janthitrem production would be unique to LpTG-3 and LpTG-4 genomes. As SE does not produce janthitrems, further analysis was performed to reduce the number of candidates to those present only in LpTG-3 and LpTG-4 endophytes. Each of the 13 putative LtmE protein homologues and 26 putative LtmJ protein homologues were used as a BLASTx query of the predicted SE protein database. A single ItmE NEA12 homologue (g30.t1) was identified in this analysis (Table 1) and therefore the best likely candidate for further investigation. The predicted protein sequence for gene g30.t1 has homology to aromatic prenyl transferases from P. janthinellum (JanD; 49%) and P. paxilli (PaxD; 46%) (Table 2). These genes are associated with synthesis of the indole diterpenes shearinine K and paxilline respectively. The gene g30.t1 is therefore henceforth referred to jtmD.

(30) TABLE-US-00002 TABLE 1 BLASTx analysis of putative LtmE and LtmJ protein homologues from NEA12 to the SE predicted protein database identified g30.t1 as the most likely candidate gene for janthitrem biosynthesis in NEA12. query subject % alignment mis- gap q. q. s. s. bit id id identity length matches opens start end start end evalue score g2.t1 g1806.t1 99.75 403 1 1 1 403 1 403 0 838 g30.t1 g4103.t1 28.05 385 256 12 22 395 10 384 1.00e−37 152 g5701.t1 g6522.t1 86.85 502 25 3 1 483 1 480 0 796 g98.t1 g1890.t1 99.4 332 1 1 1 331 1 332 0 678 g7273.t1 g1977.t1 89.06 466 14 3 1 440 1 455 0 827 g6270.t1 g7010.t1 99.63 537 2 0 1 537 1 537 0 1097

(31) Identification of the Janthitrem Biosynthetic Gene Cluster in the LpTG-3 Genome

(32) The NEA12 genome was sequenced using the PacBio Sequel sequencing platform (PacBio). The contig containing the putative LpTG-3 janthitrem biosynthetic gene cluster was identified using the jtmD gene sequence as a query. The gene content of NEA12 PacBio contig 3 (247 475 kb), containing jtmD, was then annotated using a combination of both Augustus (Stanke and Morgenstern, 2005) gene prediction and manual annotation using the known gene sequences of LTM genes (Young et al., 2005, 2006) and jtmD (Table 2).

(33) NEA12 PacBio contig 3 contains 13 predicted and known genes (FIG. 4). Cluster 1 (ItmG, ItmS, ItmM, ItmK) and Cluster 2 (ItmP, ItmQ, ItmF, ItmC, ItmB) are located at c. 57243-67332 bp and c. 6838 bp-16951 bp respectively (Table 2). The order and orientation of genes within Cluster 1 and Cluster 2 is maintained as compared to the Epichloë festucae var. lolli and Epichloë festucae LTM loci (Young et al., 2006; Saikia et al., 2008). Downstream of ItmK, a polyketide synthase (pks) pseudogene (also described by Young et al., 2005), containing several frame-shift mutations, flanked on the right by an additional AT-rich sequence was observed. The topology of the partial LpTG-3 (NEA12) LTM locus is more similar to that of the Epichloë festucae (FI1) LTM locus than the Epichloë festucae var. lolli (Lp19) which has two retrotransposon relics inserted between ItmK and the pks pseudogene (Saikia et al., 2008).

(34) The pks pseudogene defines the left-hand boundary between sequence in common to LpTG-3 and Epichloë festucae var. lolli (PacBio contig 3: 1 bp-c.70039 bp) and a previously undescribed genome sequence unique to janthitrem producing strains from the taxa LpTG-3 and LpTG-4 (PacBio contig 3: c.70039 bp-247475 bp) (FIG. 4). The right hand boundary to this region is defined by the end of PacBio contig 3 (247475 bp). This region of the NEA12 genome is characterised by 4 genes, a transposase with a MULE domain (159248 bp-163900 bp), a Helitron helicase-like transposable element (170950 bp-175054 bp), and three AT-rich regions (FIG. 5). Two novel gene clusters termed Cluster 3 and Cluster 4, each containing 2 genes, were identified on NEA12 PacBio contig 3 (Table 2; FIG. 4).

(35) The genomes of representative strains of Epichloë sp. endophytes from 4 taxa—Epichloë festucae var. lolii (SE, NEA2, NEA6, NEA10), LpTG-2 (NEA11), LpTG-3 (NEA12, AR37, 15310, 15311), LpTG-4 (E1) and FaTG-3 (NEA23)—were mapped to NEA12 PacBio contig 3. A region unique to janthitrem producing taxa LpTG-3 and LpTG-4 was identified (PacBio contig 3: c.70039 bp-247475 bp) while for endophytes from Epichloë festucae var. lolii, LpTG-2 and FaTG-3 this region was absent (FIG. 5). None of the genes in this region had been previously described in Epichloë endophytes.

(36) TABLE-US-00003 TABLE 2 Sequence analysis of genes and other features identified in NEA12 PacBio contig 3. Position in NEA12 PacBio Top BLASTp Hit contig 3 (bp) Percent Genbank Gene Gene Predicted Homologous Identity Accession ID start end cluster function gene (aa) Organism No. Reference ItmP 6838 7843 2 Cytochrome P450 ItmP 100% Epichloë festucae DQ443465 Young et al, monooxygenase var. lolii 2006 ItmQ 9169 11557 2 Cytochrome P450 ItmQ 100% Epichloë festucae DQ443465 Young et al, monooxygenase var. lolii 2006 ItmF 12830 14082 2 Prenyl transferase ItmF 99% Epichloë festucae DQ443465 Young et al, var. lolii 2006 ItmC 16001 14888 2 Prenyl transferase ItmC 100% Epichloë festucae DQ443465 Young et al, var. lolii 2006 ItmB 16370 16951 2 Integral membrane ItmB 100% Epichloë festucae DQ443465 Young et al, protein var. lolii 2006 ItmG 57243 58343 1 GGPP synthase ItmG  99% Epichloë festucae AY742903 Young at al., var. lolii 2005 ItmS 59651 60554 1 Integral membrane ItmS 100% Epichloë festucae AY742903 Young et al., protein var. lolii 2005 ItmM 61702 63348 1 FAD-dependent ItmM  99% Epichloë festucae AY742903 Young at al., monooxygenase var. lolii 2005 ItmK 65270 67332 1 Cytochrome P450 ItmK  99% Epichloë festucae AY742903 Young at al., monooxygenase var. lolii 2005 ψpks 68047 69091 — Polyketide synthase  73% Fusarium equiseti ALQ32965.1 unpublished (pseudogene) PP01 117514 116031 3 Cytochrome P450 hypothetical  68% Hirsutella KJZ77225 Lai et al., monooxygenase protein minnesotensis 2014 PP02 118533 119536 3 Membrane bound hypothetical  34% Oidiodendron KIM95229 unpublished O-acyl transferase protein maius Zn jtmD 150720 151982 4 Aromatic prenyl hypothetical  68% Ophiocordyceps KOM22681 de Bekker transferase protein unilateralis et al., 2015 PP03 159248 163900 4 Transposase hypothetical  86% Hirsutella KJZ68513 Lai et al., protein minnesotensis 2014 jtmO 164992 166560 4 6-hydroxy-D-nicotine hypothetical  59% Escovopsis KOS22754 unpublished oxidase protein weberi TE 170950 175054 — Transposable  85% Hirsutella KJZ70955 Lai et al., element minnesotensis 2014

(37) Transcript Expression of Genes Located Within PacBio Contig 3

(38) In planta expression of candidate genes for janthitrem biosynthesis in LpTG-3 (NEA12), LpTG-4 (E1) and Epichloë festucae var. lolii (SE) was determined using RNA-seq analysis of perennial ryegrass-endophyte association transcriptome data by mapping the reads generated from two perennial ryegrass-endophyte transcriptome studies to NEA12 PacBio contig 3 (FIG. 6). In study one, transcriptome analysis was performed to study the major changes that occur in host and endophyte transcriptomes during seedling growth and maturation at six timepoints, from post imbibition (0 hours) to 10 day old seedlings (10 days) (Sawbridge, 2016). Transcript expression for genes within NEA12 PacBio contig 3 in perennial ryegrass cultivar Alto-SE and Alto-NEA12 at two time points (0 hours and 10 days) is shown here. In study two, a transcriptome atlas derived from distinct tissue types of perennial ryegrass-endophyte association Impact-E1 was developed (Cogan et al., 2012). Transcript expression for genes within NEA12 PacBio contig 3 in two tissue types, leaf and stigma are shown here.

(39) In addition to the previously defined Cluster 1 and Cluster 2 genes, the genes proposed to be involved in janthitrem biosynthesis, PP01, PP02, jtmD and jtmO are also expressed. As Cluster 3 and Cluster 4 genes are not present in the Epichloë festucae var. lolii (SE) genome, expression of these genes was not observed by SE in planta.

(40) Detailed Description of the Four Gene Clusters on NEA12 PacBio Contig 3

(41) Cluster 1 (LTM1) and Cluster 2 (LTM2)

(42) Core genes for the initial stages of indole-diterpene biosynthesis in Epichloë spp. are present in LpTG-3 endophyte NEA12. Genes ItmG, ItmC and /trnM are predicted to encode a generanyl geranyl diphosphate synthase, a prenyl transferase and a FAD-dependent monooxygenase with 99%, 100%, 99% amino acid sequence identity compared with their respective Ltm homologues in Epichloë festucae var. lolii. The predicted protein product of ItmB (100%), an integral membrane protein, together with ItmM are proposed to catalyse epoxidation and cyclisation of the diterpene skeleton for paspaline biosynthesis. Genes ItmP (100%) and ItmQ (100%) encode cytochrome P450 monooxygenases and complete the collection of 6 genes required for paxilline biosynthesis in Epichloë spp.

(43) Cluster 3 Genes

(44) Cluster 3 (116033 bp-119536 bp) contains 2 genes, predicted gene PP01 (predicted protein 1), a putative cytochrome P450 monoxygenase, and PP02, predicted to be a membrane bound O-acyl transferase (MBOAT) protein (Table 2).

(45) PP01

(46) The nucleotide sequence for the PP01 gene is shown in FIG. 7. PP01 shows homology to a putative cytochrome P450 monoxygenase from HirsuteIla minnesotensis (FIG. 8; KJZ77225.1), an endoparasitic fungi of the soybean cyst nematode (Heterodera glycines). PP01 may have a role in janthitrem biosynthesis, however, PP01 does not have a homolog in any other indole-diterpene gene cluster characterized to date. For example, PP01 does not share sequence similarity with previously described cytochrome P450 monoxygenases (e.g. LtmP, LtmQ/PaxQ/AtmQ, LtmK) involved in indole-diterpene biosynthesis (FIG. 9). The predicted protein sequence of PP01 from E1 (LpTG-4) has 1 amino acid difference (at amino acid 42 D>G) to that of NEA12 (LpTG-3).

(47) PP02

(48) The nucleotide sequence for the PP02 gene is shown in FIG. 10. PP02 is predicted to be a membrane bound O-acyl transferase (MBOAT) protein (FIG. 11; FIG. 12). The predicted protein sequence of PP02 from E1 (LpTG-4) is identical that of NEA12 (LpTG-3). While membrane associated, PP02 is not a transmembrane protein based on prediction analysis with TMHMM.

(49) Cluster 4

(50) Cluster 4 (150720 bp-175051 bp) contains 2 genes, JtmD an aromatic prenyl transferase, and JtmO predicted to encode a FAD-binding oxidoreductase.

(51) JtmD

(52) The nucleotide sequence for the jtmD gene is shown in FIG. 13. JtmD, predicted to be an aromatic prenyl transferase, exhibits highest homology to a predicted protein from Ophiocordyceps unilateralis (63%; FIG. 14). The predicted protein sequence for JtmD also has homology to aromatic prenyl transferases such as those from P. janthinellum (JanD; 49%) and P. paxilli (PaxD; 46%) (FIG. 15; Nicholson et al., 2015). These genes are associated with synthesis of the indole diterpenes shearinine K and paxilline respectively. The predicted protein sequence of JtmD from NEA12 (LpTG-3) is identical that of E1 (LpTG-4).

(53) JtmO

(54) The nucleotide sequence for the jtmO gene is shown in FIG. 16. JtmO exhibits highest homology to a predicted protein (6-hydroxy-D-nicotine oxidase) from Escovopsis weberi (59%; FIG. 17). JtmO also has homology to JanO, predicted to be a FAD-binding oxidoreductase, associated with synthesis of shearinines in P. janthinellum (52%; Nicholson et al., 2015). Genes with similar predicted functions have been identified other indole-diterpene gene clusters (FIG. 18). The JtmO protein product is likely to have a role in the subsequent modification of the indole-diterpene core. The predicted protein sequence of JtmO in NEA12 (LpTG-3) and E1 (LpTG-4) is 97.9% identical. The E1 JtmO predicted protein has a 9 amino acid deletion (aa 12-20) and one amino acid change (T>A at amino acid 326) compared to that of NEA12.

(55) JtmO

(56) The nucleotide sequence for the jtmO gene is shown in FIG. 16. JtmO exhibits highest homology to a predicted protein (6-hydroxy-D-nicotine oxidase) from Escovopsis weberi (59%; FIG. 17). JtmO also has homology to JanO, predicted to be a FAD-binding oxidoreductase, associated with synthesis of shearinines in P. janthinellum (52%; Nicholson et al., 2015). Genes with similar predicted functions have been identified other indole-diterpene gene clusters (FIG. 18). The JtmO protein product is likely to have a role in the subsequent modification of the indole-diterpene core. The predicted protein sequence of JtmO in NEA12 (LpTG-3) and E1 (LpTG-4) is 97.9% identical. The E1 JtmO predicted protein has a 9 amino acid deletion (aa 12-20) and one amino acid change (T>A at amino acid 326) compared to that of NEA12.

(57) JtmD and JtmO have not previously been described in Epichloë endophytes. Homologues of the two genes have been identified in a number of Penicillium species (e.g. P. janthinellum, P. paxilli, P. crustosum) and are often found located side by side. It is interesting to note that in the Escovopsis weberi genome (GenBank: LGSR01000002.1), the two gene homologues identified in this study (JtmD: KOS22745.1; JtmO: KOS22754.1) are also found to be adjacent to each other. Escovopsis sp. are parasitic microfungi that rely on other fungi to be their hosts.

(58) Proposed Biosynthetic Pathway for Janthitrem Production

(59) The work described here provides a genetic basis for janthitrem biosynthesis in Epichloë endophytes, specifically LpTG-3 and LpTG-4. While applicant does not wish to be restricted by theory, it is likely that in addition to these two asexual taxa there is (or once was) at least one ancestral sexual Epichloë species that synthesises janthitrems.

(60) All of the indole-diterpene gene clusters identified to date have a core set of genes for the synthesis of paspaline, and a suite of additional genes that encode multi-functional cytochrome P450 monooxygenases, FAD dependent monooxygenases and prenyl transferases that catalyse various regio- and stereo-specific oxidations on the molecular skeleton to generate a diversity of indole-diterpene products.

(61) Robust liquid chromatography-mass-spectrometry (LC-MS) approaches were employed to targeted key metabolites associated with the biosynthesis of indole-diterpene alkaloids.

(62) The extracted ion chromatograms of these metabolites, isolated in planta from perennial ryegrass-LpTG-3 associations are illustrated in FIG. 19. The observed accurate masses and fragmentation patterns (via LC-MS/MS analysis) are indicated in Table 3.

(63) While applicant does not wish to be restricted by theory, based on the identification and fragmentation of these metabolites, we have proposed a framework for the biosynthesis of the epoxy-janthitrems (FIG. 20). Here, we propose that janthitrem biosynthesis is likely to arise from the synthesis of paspaline to p-paxitriol by LtmP and LtmQ. JtmD and JtmO are required for the initial biosynthesis of janthitrems, followed by PP01 and PP02. LtmF and LtmK are required for the synthesis of the epoxy-janthitrems II and IV.

(64) TABLE-US-00004 TABLE 3 Targeted LC-MS/MS analysis of the proposed metabolites associated with the biosynthesis of epoxy-Janthitrem I and its derivatives (epoxy-Janthitrem II-IV), following the indole-diterpene alkaloid biosynthetic pathway for LpTG-3 endophytes in planta. To identify each metabolite, accurate masses (m/z), retention times (RT) and MSn fragmentation data (LC-MS/MS) were acquired in positive ionisation mode [M + H] using a Thermo Fisher Q-Exactive Plus orbitrap mass spectrometer. Accurate mass and MSn fragmentation results were compared with theoretical masses and fell within the range of 5 ppm difference (Delta ppm). Chemical Theoretical m/z RT Production: LC-MS/MS Formula Mass Delta Metabolite [M + H] (min) 1 2 3 4 [M + H] [M + H] (ppm) Paspaline 422.3034 12.28 130.0651 182.0960 407.2766 C28 H40 O2 N 422.3054 −4.7 13-Desoxypaxilline 420.2534 10.69 130.0651 182.0963 402.2420 C27 H34 O3 N 420.2533 0.09 Paxilline 436.2482 9.85 130.0650 182.0961 346.1795 C27 H34 O4 N 436.2482 −0.15 β-Paxitriol 436.2482 9.67 130.0651 182.0960 335.2132 C27 H36 O4 N 438.2639 2.36 Janthitrem E 604.3637 10.60 222.1276 280.1694 546.3211 589.3346 C37 H50 O6 N 604.3633 0.7 Janthitrem F 646.3735 11.24 222.1277 280.1696 588.3320 631.3459 C39 H52 O7 N 646.3738 −0.5 Janthitrem G 630.3807 11.19 222.1274 392.1917 615.3461 C39 H52 O6 N 630.3789 2.9 Epoxy-janthitrem I 646.3735 11.24 222.1277 280.1696 588.3320 631.3459 C39 H52 O7 N 646.3738 −0.5 Epoxy-janthitrem II 670.4076 12.41 222.1275 280.1692 612.3676 655.3814 C42 H56 O6 N 670.4102 −3.8 Epoxy-janthitrem III 672.423 12.50 222.1274 280.1692 614.3833 657.3969 C42 H58 O6 N 672.4259 −4.3 Epoxy-janthitrem IV 714.4341 12.52 222.1278 280.1694 656.3934 699.4081 C44 H60 O7 N 714.4364 −3.3

(65) Functional Analysis of Candidate Genes Required for Epoxy-Janthitrem I Biosynthesis

(66) RNAi Silencing of the jtmD Gene

(67) Vector Construction

(68) Three candidate gene sequences (95 bp, 129 bp and 432 bp) within jtmD were selected for design of RNAi silencing vectors (FIG. 21). To generate the entry clones, gene cassettes were cloned into the pDONR 221 vector. RNA silencing vectors (FIG. 22) were produced by LR clonase reaction between an entry clones and the Gateway™-enabled destination vector (pEND0002) (Hettiarachchige, 2014).

(69) Isolation of Fungal Protoplasts

(70) Mycelia were harvested, under sterile conditions, by filtration through layers of miracloth lining a funnel and washed 3 times with 30 mL of sterile ddH.sub.2O. Mycelia were washed with 10 mL of OM buffer (1.2M MgSO.sub.4.7H.sub.2O, 10 mM Na.sub.2HPO.sub.4, 100 mM NaH.sub.2PO.sub.4.2H.sub.2O, pH 5.8) and transferred to a sterile 250 mL plastic vessel. Freshly prepared 10 mg/mL Glucanex (30 mL) (Sigma Aldrich) in OM was added and incubated for 18 hrs at 30° C. with gentle shaking (80-100 rpm). The glucanex/protoplast solution (30-50 μL) was examined under a microscope to confirm successful digestion. Protoplasts were filtered through miracloth in a funnel, into 15 mL sterile glass centrifuge tubes (Gentaur, Belgium) and placed on ice. Each tube was carefully overlaid with 2 mL of ST buffer (0.6 M sorbitol, 100 mM Tris-HCl, pH 8.0) and centrifuged (Beckman coulter, Avanti® J-251) (5000 rpm for 5 min at 4° C.). Following centrifugation, protoplasts formed a white layer between the glucanex solution and ST buffer and this layer was carefully removed. STC buffer (1 M sorbitol, 50 mM CaCl.sub.2.2H.sub.2O, 50 mM Tris-HCl, pH 8.0) (5 mL) was added to the protoplast solution in fresh sterile glass tubes. Samples were gently inverted once and centrifuged (5000 rpm for 5 min at 4° C.). Protoplast pellets were pooled with 5 mL of STC buffer and centrifugation was repeated (5000 rpm for 5 min at 4° C.) until only one pellet remained. Excess STC buffer was removed, and the final protoplast pellet was re-suspended in 500 μL of STC buffer. Protoplast concentration was estimated by diluting protoplasts (1/100 and/or 1/1000 with STC buffer) and counting using a Haemocytometer and microscope. Protoplasts were diluted with STC to 1.25×10.sup.8 protoplasts/mL.

(71) PEG-Mediated Fungal Protoplast Transformation

(72) Prior to delivery into fungal protoplasts, the three RNA silencing vectors (FIG. 22) were verified by restriction enzyme digestion and Sanger sequencing (data not shown). High quality plasmid DNA, suitable for transformation into fungal protoplasts was produced, using PureYield™ Plasmid Midiprep System (Promega), according to manufacturers' instructions. Aliquots (80 μL) of diluted protoplasts (1.25×10.sup.8 protoplasts/mL) were prepared on ice. To each aliquot, added; 2 μL 50 mM spermidine, 5 μL 5 mg/mL heparin (prepared in STC buffer), 10 μg plasmid DNA (1 μg/μL, not exceeding 20 μL) and 20 μL 70% (w/v) PEG solution [70% (w/v) PEG 4000, 10 mM Tris-HCl pH 8.0, 10 mM CaCl.sub.2]. Eppendorf tubes were gently mixed and incubated on ice for 30 min. Following the addition of 1.5 mL STC buffer, protoplasts were mixed and centrifuged (Eppendorf, Centrifuge 5424 R) (5000 rpm for 5 min at 4° C.). The supernatant was removed and protoplasts were resuspended in regeneration medium II (RG II, 500 μL) (304 g/L sucrose, 1 g/L KH.sub.2PO.sub.4, 1 g/L NH.sub.4NO.sub.3, 1 g/L NaCl, 0.25 g/L anhydrous MgSO.sub.4, 0.13 g/L CaCl.sub.2.2H.sub.2O, 1 g/L yeast extract, 12 g/L dehydrated potato dextrose, 1 g/L peptone, 1 g/L acid hydrolysate of casein) and incubated overnight (22° C., dark, 45 rpm).

(73) Fungal Protoplast Regeneration

(74) Overnight protoplast solution (200 μL) was incubated with 800 μL 40% (w/v) PEG solution [40% (w/v) PEG 4000, 1M sorbitol, 50 mM Tris-HCl pH 8.0, 50 mM CaCl.sub.2], at room temperature for 15 min. Molten (50° C.) 0.4% RG II (5 mL) (304 g/L sucrose, 1 g/L KH.sub.2PO.sub.4, 1 g/L NH.sub.4NO.sub.3, 1 g/L NaCl, 0.25 g/L anhydrous MgSO.sub.4, 0.13 g/L CaC1.sub.2.2H.sub.2O, 1 g/L yeast extract, 12 g/L dehydrated potato dextrose broth, 1 g/L peptone, 1 g/L acid hydrolysate of casein, 4 g/L agarose) containing 100 μL of the protoplast/PEG mixture was spread evenly across 0.6% RG II agarose petri dishes (304 g/L sucrose, 1 g/L KH.sub.2PO.sub.4, 1 g/L NH.sub.4NO.sub.3, 1 g/L NaCl, 0.25 g/L anhydrous MgSO.sub.4, 0.13 g/L CaCl.sub.2.2H.sub.2O, 1 g/L yeast extract, 12 g/L dehydrated potato dextrose broth, 1 g/L peptone, 1 g/L acid hydrolysate of casein, 6 g/L agarose) containing 100 μg/mL hygromycin B. Representative RG II petri dishes were retained without hygromycin overlay as controls to assess endophyte viability. All petri dishes were incubated at 22° C. in the dark for 4-6 weeks until regeneration was observed (FIG. 23).

(75) Identification of Transformed Fungal Protoplasts

(76) Individual regenerated colonies were transferred onto petri dishes containing 15% (w/v) potato dextrose agar (PDA) with 100 μg/mL hygromycin selection and incubated (22° C., dark, 10-21 days). Hygromycin resistant colonies were grown in 250 mL sterile culture vessels in PD broth (50 mL) with 100 μg/mL hygromycin (22° C., dark, 150 rpm, 10-21 days) and mycelia were harvested, under sterile conditions, by filtration through layers of miracloth lining a funnel and washed with 30 mL of sterile M9 phosphate buffer (1 g/L NH.sub.4Cl, 11 g/L Na.sub.2HPO.sub.4.7H.sub.2O, 3 g/L KH.sub.2PO.sub.4, 5 g/L NaCl). Washed mycelia was transferred to a Eppendorf tube, lyophilised (24-48 hrs) and DNA extracted using DNeasy Plant Mini Kit (Qiagen, Germany) according to manufacturers' instructions. Transformed individuals were identified by polymerase chain reaction (PCR) for the hygromycin gene (hph; fwd 5′-tgtcgtccatcacagtttgc-3′ (Sequence ID NO 21), rev 5′-gcgccgatggtttctacaaa-3′ (Sequence ID NO 22), and/or the candidate jtmD gene fragments [jtmD (95 bp) fwd 5′-gcctttcttcttgcctgtca-3′ (Sequence ID NO 23), rev 5′-gaccgcctgtgtgttttgaa-3′ (Sequence ID NO 24); jtmD (129 bp) fwd 5′-cacacagcccaagattgcat-3 (Sequence ID NO 25)', rev 5′-tggaagtctatcgccactgg-3′(Sequence ID NO 26), jtmD (432 bp) fwd 5′-ggagttcagtgcatgctcag-3′(Sequence ID NO 27), rev 5′-ggcaagaagaaaggctcacc-3′(Sequence ID NO 28), carried by the RNA silencing vectors. PCR components and cycling conditions using the CFX ConnectTM Real-Time PCR detection system (BioRad) [2xFastStart SYBR Green master mix (Roche), 10 uM forward and reverse primers, 2 μL template DNA, sterile ddH.sub.2O (V.sub.T 10 μL); 95° C. 10 min, (95° C. 30 sec, 60° C. 60 sec, 72° C. 30 sec)×40, 60-95° C. (0.5° C. inc.) 5 min]. The assay included appropriate positive and negative control DNA.

(77) Finally, it is to be understood that various alterations, modifications and/or additions may be made without departing from the spirit of the present invention as outlined herein.

REFERENCES

(78) Babu, J. V. (2009) Bioactive chemicals of importance in endophyte-infected grasses. PhD Thesis, University of Waikato, New Zealand.

(79) Cogan, N. O. I., Shinozuka, H., Sawbridge, T. I., Spangenberg, G. C., Forster, J. W. (2012) Development of a transcriptome atlas for perennial ryegrass (Lolium perenne L.). In ‘Abstracts 7th International Symposium on Molecular Breeding of Forage and Turf’. July 2012, Salt Lake City, Utah, USA. p. 25.

(80) Gallagher, R. T., Latch, G. C., Keogh, R. G. (1980) The janthitrems: fluorescent tremorgenic toxins produced by Penicillium janthinellum isolates from ryegrass pastures. Applied and Environmental Microbiology 39: 272-273.

(81) Hennessy, L. (2015). Epoxy-janthitrems, effects of temperature on in planta expression and their bioactivity against porina larvae. MSc Thesis. University of Waikato, New Zealand.

(82) Nicholson, M. J., Eaton, C. J., Stärkel, C., Tapper, B. A., Cox, M. P., Scott, B. (2015) Molecular cloning and functional analysis of gene clusters for the biosynthesis of indole-diterpenes in Penicillium crustosum and P. Janthinellum. Toxins 7 (8): 2701-2722.

(83) Saikia, S., Nicholson, M. J., Young, C., Parker, E. J., Scott, B.(2008) The genetic basis for indole-diterpene chemical diversity in filamentous fungi. Mycological Research 112 (2): 184-199.

(84) Sawbridge, T. I. (2016) Genomic and Transcriptomic Analysis of Perennial

(85) Ryegrass/Epichlob Endophytes Symbiota In ‘Abstracts Plant and Animal Genome XXIV Conference’. January 2016, San Diego, Calif., USA, W313.

(86) Stanke, M. and Morgenstern, B. (2005) AUGUSTUS: a web server for gene prediction in eukaryotes that allows user-defined constraints. Nucleic Acids Res. 33:465-467.

(87) Young, C. A., Felitti, S., Shields, K., Spangenberg, G., Johnson, R. D., Bryan, G. T., Saikia, S., Scott, B. (2006) A complex gene cluster for indole-diterpene biosynthesis in the grass endophyte Neotyphodium lolii. Fungal Genetic and Biology 43: 679-693.

(88) Young, C. A., Bryant, M. K., Christensen, M. J., Tapper, B. A., Bryan, G. T., Scott, B. (2005) Molecular cloning and genetic analysis of a symbiosis-expressed gene cluster for lolitrem biosynthesis from a mutualistic endophyte of perennial ryegrass. Molecular Genetics and Genomics 274 (1): 13-29.