Identification, heterologous production and bioactivity of lentinulin A and dendrothelin A, two natural variants of backbone N-methylated peptide macrocycle omphalotin A

Backbone N-methylation and macrocyclization improve the pharmacological properties of peptides by enhancing their proteolytic stability, membrane permeability and target selectivity. Borosins are backbone N-methylated peptide macrocycles derived from a precursor protein which contains a peptide α-N-methyltransferase domain autocatalytically modifying the core peptide located at its C-terminus. Founding members of borosins are the omphalotins from the mushroom Omphalotus olearius (omphalotins A-I) with nine out of 12 L-amino acids being backbone N-methylated. The omphalotin biosynthetic gene cluster codes for the precursor protein OphMA, the protease prolyloligopeptidase OphP and other proteins that are likely to be involved in other post-translational modifications of the peptide. Mining of available fungal genome sequences revealed the existence of highly homologous gene clusters in the basidiomycetes Lentinula edodes and Dendrothele bispora. The respective borosins, referred to as lentinulins and dendrothelins are naturally produced by L. edodes and D. bispora as shown by analysis of respective mycelial extracts. We produced all three homologous peptide natural products by coexpression of OphMA hybrid proteins and OphP in the yeast Pichia pastoris. The recombinant peptides differ in their nematotoxic activity against the plant pathogen Meloidogyne incognita. Our findings pave the way for the production of borosin peptide natural products and their potential application as novel biopharmaceuticals and biopesticides.


Contents
Supplementary Materials and methods: Chemical synthesis of omphalotin A General information. HPLC grade acetonitrile from Sigma-Aldrich (Buchs, Switzerland) was used for analytical and preparative purification. DMF and DIPEA from Sigma-Aldrich (Buchs, Switzerland) for solid phase peptide synthesis were used without additional purification. Additional commercially available reagents and solvents were purchased from Sigma-Adrich (Buchs, Switzerland), except stated otherwise. Solvents for flash chromatography (EtOAc, MeOH) were of technical grade and used without further purification. Coupling reagents, such as HATU and all Fmoc-amino acids were purchased from Peptides International (Louisville, KY, USA) and Chemimpex (Wood Dale, IL, USA). 2-Chlorotrityl chloride resin (100-200 mesh) was purchased from Novabiochem (Burlington, MA, USA).

Reactions and purification.
Peptides were purified on Jasco reverse phase high performance liquid chromatography (RP-HPLC) preparative apparatus equipped with a dual PU 2087 plus pumps, in-line degasser and the spectra were recorded simultaneously at three different wavelengths: 220 nm, 254 nm and 301 nm with a UV-2077 plus detector.
The mobile phase was composed of high-grade Millipore H2O and acetonitrile containing 0.1% (v/v) TFA. At a flow rate of 10 mL/min using a Phenomenex C4 (10 m, 100 Å, 250 x 21 mm) column. The column was heated to 60°C and preequilibrated at 40% acetonitrile for approximately 10 min before injecting the sample. After 5 min at 40% acetonitrile a linear gradient was run for 40 min until a final 95% acetonitrile gradient was reached. Finally, the columns were flushed for 7 min at 95% acetonitrile. The cyclization reaction was monitored by thin layer chromatography using precoated glass plates Merck (Burlington, MA, USA) and visualized by ninhydrin staining.
The synthesis of the hydrophobic cyclic peptide was performed by established Fmoc manual solid phase peptide synthesis, followed by standard resin cleavage, cyclization reaction and several purification steps 1 . To load the resin, 2-chlorotrityl resin (1 g, 1.46 mmol/g, 1.46 mmol) was supplied with 10 mL of CH2Cl2 and vigorously shaken for 30 min at room temperature. The swollen resin was further rinsed with CH2Cl2 (3 x 10 mL) and supplied with a solution of DIPEA (278.7 L, 1.6 mmol, 1.0 eq.) activated Fmoc-Sar-OH (124.5 mg, 0.4 mmol, 0.274 eq.) in 10 mL of CH2Cl2. The resin was shaken for 2 h at room temperature, before being thoroughly washed with CH2Cl2 (5 x 10 mL), DMF (5 x 10 mL), followed by CH2Cl2 (5 x 10 mL). The resin was further flushed with 20 mL of diethyl ether and dried under nitrogen flux. After determining the resin loading (0.3112 mmol/g), the resin was swollen in 10 mL CH2Cl2 for 30 min, flushed and supplied with 10 mL capping solution CH2Cl2/MeOH/DIPEA (17:2:1) for 5 min at room temperature. This process was repeated once. The resin was washed thoroughly with CH2Cl2 (5 x 10 mL), DMF (5 x 10 mL), followed by CH2Cl2 (5 x 10 mL) and dried under nitrogen flux. Loaded resin was stored in the fridge overnight.

Loading efficiency calculation.
After the addition of the first amino acid, approximately 12.3 mg of dried resin were collected and subjected to 2 mL 2% DBU in DMF, shaken for 15 min at room temperature and filtered through cotton. Subsequent amino acid couplings. The resin was swollen in DMF for 30 min at room temperature, before being supplied with 8 mL of 20% piperidine solution and shaken for 8 min. Fmoc deprotection was repeated once.
Simultaneously, Fmoc amino acid (5 equiv.) and HATU (4.95 equiv.) were dissolved in 8 mL of DMF and activated with DIPEA (10 equiv.) for 3 min. The resin was vigorously washed with DMF and DCM and supplied with the activated amino acid solution. The coupling was carried out at room temperature for 3 hours and washed thoroughly. 10 mL of DMF, acetic anhydride (5 eq.) and NMM (5 eq.) were added to the resin. Capping procedure was carried out for 15 min and resin further washed with DMF (5 x 10 mL), CH2Cl2 (5 x 10 mL), followed by DMF (5 x 10 mL). The resin was dried under reduced pressure for 30 min and stored in the fridge.
Stepwise analysis of peptide couplings. Peptide coupling efficiencies were continuously monitored, by micro cleavage after each coupling step. Approximately 15 mg of dried resin were supplied with 1 mL of HFIP/ CH2Cl2 (1:5) and shaken for 30 min at room temperature. The mixture was filtered through cotton filters and solvents evaporated under reduced pressure. Crude peptides were dissolved in 500 L acetonitrile and analyzed by HPLC and LC-MS analysis.
Cleavage of resin. 600 mg of Fmoc protected peptide resin was swollen in 10 mL of DMF for 30 min.  Colony PCR analysis of transformants of P. pastoris strain GS115 expressing ophMA and ophP alone, and coexpressing ophMA together with ophP or ledP (lane 6). The transformants were streaked on YPD plates, and incubated at 30⁰C for three days. One single colony for each strain was scraped off the plate using a loop and resuspended in 30 µl of 0.2% SDS in an Eppendorf tube, heated to 95°C for 10 min and centrifuged at 11000 x g for 2 min. The supernatant (0.5 µl) was used as template of a PCR reaction following the user guide of DreamTaq™ Green PCR Master Mix (2x) (ThermoFisher Scientific). Oligonucleotides 5-AOX1 and 3-AOX1 (see Supplementary Table 2 for sequences) were used as primers. 5 µl of the PCR reaction were run out on a 1% (w/v) agarose gel and stained with EtBr. Transformant 1, GS115; 2, GS115-pPICZA; 3, GS115-OphMA; 4, GS115-OphP; 5, GS115-OphMA-OphP; 6, GS115-OphMA-LedP. Transformants expressing the hybrid ophMA constructs were analyzed accordingly (data not shown). (B) Immunoblot of soluble protein extracts of respective GS115 transformants assessing the production of OphP and OphMA. Heterologously produced OphMA was N-terminally His-tagged while OphP was fused to a N-terminal StrepII-SUMOSTAR. Anti-His-tag (α-His) and anti-StrepII-tag (α-Strep) were used as primary antibodies, while HRP-conjugated goat anti-mouse IgG served as secondary antibodies. The yellow boxes highlight the concomitant production of OphMA and OphP in three different GS115-OphMA-OphP transformants (1, 2). Transformants expressing the hybrid ophMA and the ledP constructs were analyzed accordingly (data not shown).