WO2018141146A1

WO2018141146A1 - Total chemical synthesis of lasso peptide

Info

Publication number: WO2018141146A1
Application number: PCT/CN2017/087855
Authority: WO
Inventors: Ming Chen
Original assignee: Ming Chen
Priority date: 2017-02-06
Filing date: 2017-06-12
Publication date: 2018-08-09
Also published as: CN106749497B; CN106749497A

Abstract

A cryptand-ionic liquid supported method for the first totally chemical synthesis of lasso peptide BI-32169 and its analogs is described. Cryptand-ionic liquid supported technique has been proved to be a powerful and controllable synthetic tool for producing or modifying chemically engineered lasso-shaped peptides and studying structure-activity relationships of lasso peptides as well. Thus, it is highly desirable that this synthetic strategy with designable multi-linker support could benefit the expansion of diverse lasso peptide libraries for discovery of novel bioactive peptides.

Description

Total chemical synthesis of lasso peptide

TECHNICAL FIELD

The present invention relates to the new strategy using cryptand-ionic liquid complex as a multi-linker support for the chemical synthesis of lasso peptides BI-32169 and its analogs.

BACKGROUND ART

Lasso peptides are an important class of natural products that belong to the family of ribosomally biosynthesized and post-translationally modified peptides (RiPPs) . [1] Various interesting biological activities of these peptides have been reported, such as enzyme inhibition, receptor antagonism and/or antimicrobial activities. [1c] The sequence of lasso peptide exhibits a knotted topology involving an N-terminal lactam-bridged ring (7-9 amino acids) , which is threaded by a exocyclic C-terminal tail (7-15 amino acids) . [1b] It is noteworthy that the antimicrobial lasso peptides, e.g. Microcin J25 and lassomycin, have been found to lost their activities when their conformations were unthreaded. [2] Therefore, the unique knotted feature was proposed to be a key factor for the activity of lasso peptide.

Chemical synthesis, such as solid phase peptide synthesis (SPPS) , is easily automated and provides almost infinite possibilities for peptide design and modification. [3] A significant improvement reported recently is that a lasso peptide scaffold could be mimicked by chemically synthesized peptide-based rotaxane with a crown ether ring. [4] However, until now, lasso peptides can only be isolated from their natural resources or synthesized by recombinant technology and none of them has been prepared successfully by chemical approach. [5] According to the previous study, the fact is mainly due to the great challenge in lasso peptide folding. The conventional method of peptide synthesis commonly offers only one anchor or linker on synthetic support, which is not sufficient for conducting the C-terminal tail to thread through the N-terminal ring of lasso peptide (Figure 1A) .

In our previous work, ionic liquids (ILs) , that are organic salts in liquid states at low temperature or even at room temperature, have been used as an efficient reaction medium in peptide synthesis. [6] Miao et al. described a new method called ionic-liquid-supported peptide synthesis (ILSPS) , in which the imidazolium-based ionic liquid acted as a high-loading and recoverable support for synthesizing oligopeptides. [7] Specially, it is reported that imidazolium-based ionic liquid can form a stable inclusion complex with hollow circular compounds via hydrogen bonds or π-π interactions. [8] These findings give us inspiration to develop a cryptand-ionic liquid complex as support for lasso peptide synthesis.

BRIEF DISCLOSURE OF INVENTION

The present invention relates to a multi-linker strategy for the chemical synthesis of lasso peptide. The multi-linker synthesis strategy is given in Figure 1B, in which three anchors or linkers are introduced for constructing a lasso peptide with knotted structure. The C-terminal tail of lasso peptide (AA hollow circles in Figure 1B) is prolonged after first amino acid is anchored to an initial linker (middle linker in Figure 1B) . The N-terminal ring of lasso peptide (AA solid circles in Figure 1B) is then formed by taking the amino acid anchored to the second linker (left linker in Figure 1B) as starting point. The direction of the ring formation around the C-terminal tail could be controlled by the third linker (right linker in Figure 1B) , chirality of the peptidyl support and the distribution of rigid residues like L-proline in the sequence. The tail sequence then threads through the ring after cleavage of the linkages. The multi-linker strategy required a special support which should provide the following features: containing multiple linkers, no steric hindrance between adjacent linkers and increasing solubility of peptide. To implement the multi-linker supported approach for the synthesis of lasso peptide, we develop a cryptand-ionic liquid complex as support, in which the imidazolium cation of ionic liquid contains the first linker for C-terminal tail formation of lasso peptide and the cryptand furnishes the second and third linkers for N-terminal ring formation for the multi-linker strategy of lasso peptide synthesis.

To demonstrate the application of the novel cryptand-ionic liquid supported strategy for lasso peptide preparation, as an embodiment, we herein present the first successful total synthesis of lasso peptide BI-32169 (Figure 2A) , which is an effective glucagon receptor antagonist from Streptomyces sp. (DSM 14996) . [9] BI-32169 contains a 9-mer N-terminal ring (G¹LPWGCPSD⁹) established via an isopeptide linkage and a 10-mer C-terminal tail (I¹⁰PGWNTPWAC¹⁹) threaded through the ring. A disulfide bridge is then formed between the C-terminal cysteine residue of the tail (C¹⁹) and the cysteine residue in the N-terminal ring (C⁶) (Figure 2B, 2C) . Moreover, since such a bicyclic secondary structure is chiral, the cryptand-ionic liquid supported approach is also a useful synthetic tool for producing and studying the D-retro enantiomers of lasso peptides (Figure 2D) , which has not been reported before.

BRIEF DESCRIPTION OF DRAWINGS

Non-limiting embodiments of the present invention will be described by way of example with reference to the accompanying figures, which are schematic and are not intended to be drawn to scale. In the figures, each identical or nearly identical molecule or component illustrated is typically represented by a single numeral (bold) . For purposes of clarity, not every component is labeled in every figure, nor is every component of each embodiment of the invention shown where illustration is not necessary to allow those of ordinary skill in the art to understand the invention.

Figure 1A-2B are (A) conventional single-linker strategy of lasso peptide synthesis and (B) multi-linker strategy of lasso peptide synthesis. (AA＝amino acid) .

Figure 2A-2D are (A) primary structure of BI-32169 composed of L-amino acids, (B) crystal structure of BI-32169 composed of L-amino acids, (C) secondary structure diagram of BI-32169 composed of L-amino acids and (D) secondary structure diagram of D-retro-inverso BI-32169, which is composed of D-amino acids in a reversed sequence (in italics) .

Figure 3 (left side) is, according to certain embodiments, the complexation between cryptand 2 and ionic liquid 1. The interactions (hydrogen bonds and π-π stacking) are shown in dotted line.

Figure 3A-3C (right side) are, according to certain embodiments, ¹H NMR spectra of (A) ionic liquid 1, (B) equimolar mixture of 2 and 1 and (C) cryptand 2.

Figure 4A-4C are, according to certain embodiments, (A) attempt at coupling of amino acid with cryptand-ionic liquid complex support, (B) attempt at complexation of peptidyl ionic liquid with cryptand and (C) cryptand-ionic liquid supported total synthesis of BI-32169.

Figure 5 is, according to certain embodiments, synthesis of unthreaded topoisomer of BI-32169 (ESI-MS calculated for C₉₅H₁₂₅N₂₃O₂₄S₂ [M+2H] ²⁺: 1019.6595； Found 1019.6560) .

Figure 6 are, according to certain embodiments, HPLC profiles of (a) crude synthesized BI-32169, (b) purified synthesized BI-32169, (c) native BI-32169 and (d) unthreaded topoisomer of BI-32169 synthesized by SPPS method.

Figure 7 is, according to certain embodiments, MS² spectrum of oxidized BI-32169. Four series of fragment ions (singly and doubly protonated fragments) and their corresponding peaks (marked with italic serial numbers) are showed respectively.

Figure 8 is, according to certain embodiments, MS² spectrum of reduced BI-32169. Three series of fragment ions (singly and doubly protonated fragments) and their corresponding peaks (marked with italic serial numbers) are showed respectively.

Figure 9 is, according to certain embodiments, MS² spectrum of reduced unthreaded topoisomer of BI-32169. Three series of fragment ions (singly and doubly protonated fragments) and their corresponding peaks (marked with italic serial numbers) are showed respectively.

Figure 10 is, according to certain embodiments, the protected cryptand assemblies that are axial enantiomers of each other.

DETAILED DESCRIPTION

To implement the cryptand-ionic liquid supported approach for the synthesis of BI-32169, a 1: 1 inclusion complex assembled by the ionic liquid 1, 3-dimethyl-2-hydroxymethylimidazolium tetrakis [3, 5-bis (trifluoromethyl) phenyl] borate (namely [2-HOM-MMIm] [BAr^F ₄] , 1) and a designed derivative of cryptand, (S_a) -5′, 4″-dinitro-dibenzocryptand [222] (2) which encompasses a phenylisopropyl (PhⁱPr) -protected 4′-oxycarbonylmethoxymethyl linker and a tert-butyldiphenylsilyl (TBDPS) -protected 5″-hydroxymethyl linker, was selected as synthetic support. The intermolecular hydrogen bonds and π-π stacking interactions between the hydroxyl-protected cation of ionic liquid (1) and the cavity of cryptand (2) have been found to play pivotal roles in stabilizing the host-guest formation of this cryptand-ionic liquid complex (Figure 3) . Solvent influence was studied beforehand. In our solubility tests, the hydroxyl-protected or the peptide preloaded supports containing [BAr^F ₄] anions are poorly soluble in aqueous solution, and most of the non-polar solvents, e.g. hexane, while they can be dissolved or partially dissolved in N, N-dimethylformamide (DMF) , dimethyl sulfoxide (DMSO) , N-methyl-2-pyrrolidon (NMP) , methanol, 2-methyltetrahydrofuran (2-MeTHF) and even ether. We preferred 2-MeTHF for several reasons. Besides its green and eco-friendly properties, the main one was that the excess reagents and byproducts in each coupling step could be removed by washing the water-immiscible phase of 2-MeTHF with aqueous solutions. Furthermore, the compatibility of 2-MeTHF has been verified by its applications in fluorenylmethyloxycarbonyl (Fmoc) based SPPS. [10]

Comparative Example 1: First possible synthetic route (Figure 4A)

One route was to use the cryptand-ionic liquid complex as support in the complete procedure for the peptide synthesis. However, the synthesis failed from the start maybe because the attachment of the first amino acid to the complex support was sterically hindered by the linkers of cryptand part (2) .

Comparative Example 2: Second possible synthetic route (Figure 4B)

Another route was to synthesize the linear peptide chain of C-terminal tail by a standard ILSPS method ^[7] utilizing ionic liquid (1) as preliminary support and hereafter carry out the complexation of the cryptand (2) with the peptidyl ionic liquid (e.g. 4) . The resulting peptide-loaded complex would be used as integral support for the subsequent coupling reactions. Unfortunately, this route did not work well either. We found that the peptide chain (≥ 4 amino acids) loaded on the ionic liquid support was not conducive to the formation of the cryptand-ionic liquid complex.

Comparative Example 3: Synthesis of unthreaded peptide topoisomer (Figure 5)

Linear peptide GLPWGCPSDIPGWNTPWAC was prepared by stepwise Fmoc-SPPS on an Advanced ChemTech (ACT-396) automated peptide synthesizer on 2-chlorotrityl chloride resin (100-200 mesh, 1.06 mmol/g) in situ activation protocols to couple Fmoc-protected amino acid (4.0 eq. to resin loading) to the resin using PyBOP (4.0 eq. ) as coupling reagent in the presence of N-methylmorpholine (8.0 eq. ) . The Fmoc group was deprotected with 20％piperidine/DMF. The side chain ODmab of residue D⁹ was selectively removed using 2％hydrazine in DMF. The cyclization via isopeptide bond was carried out using PyBOP (4.0 eq. ) and N-methylmorpholine (4.0 eq. ) . Cyclized peptide was cleaved from the resin at room temperature in TFA/phenol/water/TIPS (88: 5: 5: 2) for 3h. Cold diethyl ether was then added to the filtered cleavage mixture and the peptide precipitated out. Pure peptide (0.2 mM) were oxidized by stirring at room temperature in 0.1M NH₄OAc/DMF for 12h after washing with further cold diethyl ether. The oxidized peptide was purified by semipreparative reversed-phase HPLC equipped with a Waters XBridgeTM BEH3000 C18 column (4.6 × 150 mm) at a flow rate of 10.0 ml/min in 0–50％acetonitrile/0.1％TFA gradient and then lyophilized overnight.

Example 4: Total synthesis of lasso peptide BI-32169 (Figure 4C)

We began with the first anchoring of support. The ionic liquid bound Fmoc-cysteine (^tBu) (5) (^tBu＝ tert-butyl) was synthesized by the attachment of the amino acid Fmoc-cysteine (tBu) -OH (protected C¹⁹) onto the ionic liquid (1) . The solvent was then washed with 20％ (m/v) NaCl solution and deionized water. Next, the cysteine (^tBu) -loaded ionic liquid (6) was obtained via an Fmoc-deprotection with tris (2-aminoethyl) amine (TAEA) . After washing again with a phosphate solution (pH＝5.8) , Fmoc-alanine-OH (protected A¹⁸) was coupled to 6 to form the dipeptidyl ionic liquid (7) . The anchoring (esterification) and coupling reactions were both performed efficiently under conditions using the reagent combination of 1-ethyl-3- (3-dimethylaminopropyl) carbodiimide (EDCI) and ethyl (hydroxyimino) cyanoacetate (Oxyma

) without pre-drying treatment. Subsequently, the cryptand (2) was introduced to constitute a cryptand-ionic liquid complex support loaded with Fmoc-alanine-cysteine (^tBu) (8) . Under the above-mentioned conditions for the repeated cycles of deprotection-wash-coupling-wash, the protected amino acid moieties tryptophan (W¹⁷) , proline (P¹⁶) , threonine (T¹⁵) , asparagine (N¹⁴) , tryptophan (W¹³) , glycine (G¹²) , proline (P¹¹) , isoleucine (I¹⁰) , aspartic acid (D⁹) and serine (S⁸) were sequentially coupled to 8, to form 9, in which the hydroxyl side chain of the residue S⁸ was chosen as a suitable anchor group. Thus, we selectively cleaved the acid-sensitive protecting groups, triphenylmethyl (Trt) of serine side chain and PhⁱPr of the oxycarbonylmethoxymethyl linker in support, by treatment with diluted trifluoroacetic acid (TFA) solution and then achieved the second anchoring (esterification) of peptide with complex support (10) .

To generate the third linkage between peptide and support, as in the second anchoring, we also needed to designate an amino acid residue as the third anchor moiety. Tryptophan (W⁴) and proline (P³) could be appropriate alternatives in the light of their positions (opposite to S⁸) in the ring sequence of BI-32169. However, neither of them has a chemically linkable side chain for anchoring to the support. To overcome this challenge, we replaced tryptophan (W⁴) with its derivative, 2-amino-3- (1-carboxyl-indolin-3-yl) propanoic acid (namely 2H, 3H-1-carboxytryptophan, W′⁴) . The moiety W′⁴ was preferred since it could be converted to tryptophan residue in the final cleavage stage. Thereby, starting from the synthesized 10, we elongated the peptide chain through the successive couplings with the amino acids Fmoc-proline-OH (protected P⁷) , Fmoc-cysteine (^tBu) -OH (protected C⁶) , Fmoc-glycine-OH (protected G⁵) followed by Fmoc-2H, 3H-1-carboxytryptophan (pNB) -OH (protected W′⁴) to obtain 11. The pNB and TBDPS protecting groups belonging to W′⁴ and the hydroxymethyl group of cryptand assembly respectively were then removed simultaneously with tetra-n-butylammonium fluoride (TBAF) . Nevertheless, the attempt to join the free side chain of W′⁴ with the free linker of support using EDCI/Oxyma

offered only a low yield of esterification (27％of 12) . Switching to other common combination of coupling reagents, such as N, N′-diisopropylcarbodiimide (DIC) /4-dimethylaminopyridine (DMAP) or benzotriazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate (PyBOP) /N, N-diisopropylethylamine (DIPEA) , did not alleviate this problem as well. The anchoring yield was ultimately improved to 86％by increasing the temperature to 37℃ and conducting esterification twice with EDCI/Oxyma

As all the three linkages between peptide chain and cryptand-ionic liquid complex were fulfilled, the last three amino acid moieties, proline (P³) , leucine (L²) , glycine (G¹) , were assembled to 12, to give the complex support loaded with complete sequence of BI-32169 (13) . In the subsequent, the protecting groups, N-Fmoc of residue G¹ and 4- (N- [1- (4, 4-dimethyl-2, 6-dioxocyclohexylidene) -3-methylbutyl] amino) benzyl (Dmab) of D⁹ side chain, were both eliminated with 2％hydrazine in 2-MeTHF, and the cyclization was achieved via the isopeptide bond formation between the free residues D⁹ and G¹ to afford the N-terminal ring of the lasso peptide.

To realize the lasso scaffold of the peptide, the second and third linkages were first broken in minutes through a simple photolytic cleavage under mild heating conditions (30℃) . After vacuum evaporation, the concentrated solution was transferred into a cleavage cocktail which consists of TFA, phenol, water and triisopropylsilane (TIPS) with a ratio of 88: 5: 5: 2, to remove all the remaining protecting groups in the peptide chain. The reaction mixture was added dropwise in ice-cold hexane and the complex support loaded with unprotected peptide (15) precipitated immediately. It should be noted that cold ether is not applicable for separation due to its high dissolving capability for the compounds with [BAr^F ₄] anion. The lasso peptide was then released from the complex support by 0.1 M sodium hydroxide in water/tetrahydrofuran (1: 3) with argon protection for 6 hours, and the solution was afterwards exposed to air for 12 hours to generate the disulfide bridge between the residues C¹ and C⁶. In this step, utilizing a higher concentration (1.0 M) solution of sodium hydroxide could obviously shorten the liberation time, yet led to an undesirable opening of the N-terminal ring and thus destroyed the lasso conformation of BI-32169. After the oxidation, the solution was neutralized with 0.1 M aqueous citric acid and the solvent was distilled in vacuum. The resulting material was washed with water and ether thoroughly before being lyophilized to produce the crude peptide BI-32169. The results of high-performance liquid chromatography (HPLC) analysis revealed that the crude product exhibited high purity yet the only detectable byproduct was the unthreaded peptide, which might be generated from the incomplete third anchoring (Figure 6) . Finally, the lasso peptide BI-32169 was obtained in a 98.72％pure form, and in an overall yield of 2.47％by preparative scale HPLC. The threaded lasso structure of BI-32169 was characterized by tandem mass spectrometry (MS²) analysis (Figure 7, Figure 8, Figure 9) . [9b]

Example 5: Total synthesis of D-retro-inverso BI-32169

Like the synthesis of BI-32169, D-retro-inverso BI-32169 was prepared via cryptand-ionic liquid supported approach using an (R_a) -enantiomer of 2 as cryptand assembly (16, Figure 10) and corresponding D-amino acids as building units. Finally, the lasso peptide BI-32169 was obtained in a 98.35％pure form, and in an overall yield of 1.97％by preparative scale HPLC.

References (Non-patent citations)

[1] a) M.O. Maksimov, S.J. Pan, A. James Link, Nat. Prod. Rep. 2012, 29, 996-1006； b) M.O. Maksimov, A.J. Link, J. Ind. Microbiol. Biotechnol. 2014, 41, 333–344； c) J.D. Hegemann, M. Zimmermann, X. Xie, M.A. Marahiel, Acc. Chem. Res. 2015, 48, 1909–1919.

[2] a) R. Ducasse, K. -P. Yan, C. Goulard, A. Blond, Y. Li, E. Lescop, E. Guittet, S. Rebuffat, S. Zirah, ChemBioChem 2012, 13, 371–380； b) S. Lear, T. Munshi, A.S. Hudson, C. Hatton, J. Clardy, J.A. Mosely, T.J. Bull, C.S. Sit, S.L. Cobb, Org. Biomol. Chem. 2016, 14, 4534–4541.

[3] R.B. Merrifield, J. Am. Chem. Soc. 1963, 85, 2149–2154.

[4] a) C. Clavel, K. Fournel-Marotte, F. Coutrot, Molecules 2013, 18, 11553–11575； b) F. Saito, J.W. Bode, Chem. Sci. 2017, DOI 10.1039/C7SC00021A.

[5] a) S. Stolze, M. Kaiser, Synthesis (Stuttg) . 2012, 44, 1755–1777； b) Y. Li, S. Zirah, S. Rebuffat, 2015, pp. 81–95.

[6] T. Kühl, M. Chen, K. Teichmann, A. Stark, D. Imhof, Tetrahedron Lett. 2014, 55, 3658–3662.

[7] W. Miao, T.H. Chan, J. Org. Chem. 2005, 70, 3251–3255.

[8] S. Barman, M.N. Roy, RSC Adv. 2016, 6, 76381–76389.

[9] a) O. Potterat, K. Wagner, G. Gemmecker, J. Mack, C. Puder, R. Vettermann, R. Streicher, J. Nat. Prod. 2004, 67, 1528–1531； b) T.A. Knappe, U. Linne, X. Xie, M.A. Marahiel, FEBS Lett. 2010, 584, 785–789； c) H. Nar, A. Schmid, C. Puder, O. Potterat, ChemMedChem 2010, 5, 1689–1692.

[10] Y.E. Jad, G.A. Acosta, T. Govender, H.G. Kruger, A. El-Faham, B.G. de la Torre, F. Albericio, ACS Sustain. Chem. Eng. 2016, 4, 6809–6814.

Claims

A cryptand-ionic liquid supported method for the first totally chemical synthesis of lasso peptide BI-32169 and its analogs, comprising:

using a cryptand-ionic liquid support which provides multiple linkers；

anchoring first C-terminal amino acid to the initial linker of cryptand-ionic liquid support

prolonging the linear chain of lasso peptide by amino acid coupling；

anchoring a side-chain linkable amino acid in the N-terminal ring to second linker of cryptand-ionic liquid support；

forming the N-terminal ring of lasso peptide by taking the amino acid anchored to the second linker of cryptand-ionic liquid support as starting point；

controlling the formation of the N-terminal ring around the C-terminal tail by the third linker and chirality of the peptidyl support to achieve the lasso peptide scaffold by amino acid coupling；

deprotecting all the protecting groups of peptide and cleaving the lasso peptide product from cryptand-ionic liquid support.
The process of claim 1, wherein the cryptand-ionic liquid support comprises a 1: 1 complex between a imidazolium-based ionic liquid containing one chemical linker group and a derivative of cryptand containing two or more chemical linker groups, preferably a 1: 1 inclusion complex assembled by the ionic liquid 1, 3-dimethyl-2-hydroxymethylimidazolium tetrakis [3, 5-bis (trifluoromethyl) phenyl] borate, and the chiral cryptand 5′, 4″-dinitro-dibenzocryptand [222] which encompasses a phenylisopropyl-protected 4′-oxycarbonylmethoxymethyl linker and a tert-butyldiphenylsilyl-protected 5″-hydroxymethyl linker.
The process of claim 1, wherein the analogs of lasso peptide BI-32169 comprise the lasso-shaped peptide derivates of BI-32169 with one or more amino acid residues of BI-32169 substituted to other natural amino acid residues and/or unnatural amino acid residues.
The process of claim 1, wherein the linkers comprise, but not limited to, hydroxyl groups, carboxyl groups, amino groups, halogen groups and/or thiol groups.