Todd:Pictet-Spengler to PZQ

=Synthesis of (R)-Praziquantel via a Catalytic, Asymmetric Pictet-Spengler Reaction=

Matthew H. Todd, School of Chemistry, The University of Sydney, NSW 2006, Australia Michael Woelfle, School of Chemistry, The University of Sydney, NSW 2006, Australia

(additional authors - add alphabetically. Please consider adding yourself and helping to write this paper, particularly if you have already helped out over at relevant The Synaptic Leap site)

Abstract
(This article is a continually-updated summary of the results to date from the Open Lablog: Pictet-Spengler Route to Praziquantel. The project is open source, meaning anyone can participate. This paper may be added to and edited by anyone. The project, and this page are currently active - when this changes <= these words will be changed. References for this page may be found in full at the Mendeley page)

The Pictet-Spengler approach is a potential method for the enantioselective synthesis of praziquantel (PZQ), the drug used worldwide for the treatment of the neglected tropical disease schistosomiasis. Following the recent identification of routes to enantiopure PZQ by classical resolution, we report here the progress to date on the enantioselective synthesis of PZQ using a Pictet-Spengler reaction. The approach employs a known peptide acetal precursor in an acid-catalyzed cyclization.

Introduction
The anthelmintic drug PZQ (1b, Scheme 1) is widely used in the treatment of schistosomiasis and remains the only viable drug for the mass treatment of this disease. PZQ is synthesized and administered as a racemate, even though the inactive (S)-enantiomer is associated with side effects and is responsible for the highly bitter taste of the pill. Administration of the pure active enantiomer is listed as a priority in the WHO business plan 2008-2013. Production of PZQ as a single enantiomer while keeping the price low is a challenge - preparation of single enantiomers is essentially always a more expensive enterprise than preparation of enantiomers, unless relevant stereochemistry is contained within natural products, which is not the case for PZQ.

An efficient resolution of PZQ was recently developed by an open science mechanism. A contract research organisation also developed a similar resolution in parallel. Resolution approaches are viable candidates for the large-scale preparation of PZQ on purely economic grounds. Yet there remain potentially very efficient approaches based on asymmetric catalysis that would have the advantage of not requiring either disposal or separation and recycling of the inactive enantiomer. The challenge is firstly to demonstrate a path to (R)-PZQ using asymmetric catalysis, but then to optimise the process to ensure the catalyst loading does not make such a route prohibitively expensive.

Besides an alternative separation of enantiomers based on chromatography, there has been a single report each of diastereo- and enantio-elective syntheses of PZQ. These routes are not, however, realistic for the large-scale syntheses of PZQ, partly because they use synthetic routes that are not currently used for the racemate - de novo process optimization for these approaches is not likely to happen. A better approach is to take existing routes to the drug, and make a key step asymmetric. PZQ was originally synthesized by a Reissert process. This route has the disadvantage of requiring a large amount of cyanide, but such a process was for many years used in the production of the drug, and it is possible that it is still used. There are literature reports of catalytic, asymmetric Reissert processes, but surprisingly there are no reports of this reaction being successfully applied to the system required for PZQ - isoquinoline.

While the exact routes used to synthesise PZQ on a ton scale are not clear, it is likely that one of the main suppliers, Shin Poong, employs a published method that uses a Pictet-Spengler (PS) cyclization. The key precursor to the cyclization, and hence the substrate for an asymmetric version of this reaction, is likely available in quantity, and can in any case be prepared by a new and more efficient route than that originally published. A large-scale route to (R)-PZQ is hence a viable possibility via a Pictet-Spengler sequence if a catalyst could be found to effect the required cyclization.

Unfortunately the key cyclization is beyond the current state of the art. There is a small number of reports in the literature of catalytic, asymmetric Pictet-Spengler reactions. In all cases the aromatic ring involved in the cyclization is electron rich, usually by virtue of containing one or more methoxy substituents. To the best of our knowledge there are no reports of catalytic, asymmetric Pictet-Spengler reactions involving a simple phenyl ring. Besides the route described above, other racemic syntheses of PZQ have used the PS reaction.

Besides the substrate (1a) required for the synthesis of PZQ by a PS approach, three other peptide acetal starting materials are worthy of investigation: the benzoyl analog (3a) and the dimethoxy-functionalised analogs of both these structures (2a and 4a). The change from cyclohexanoyl to benzoyl might influence the ease of initial acetal cyclisation to generate an acyliminium ion, and the final product of the reaction, the benzoyl analog of PZQ (3b), may be easier to crystallise/purify. Interestingly this benzoyl analog is more potent as an anthelmintic than PZQ itself, yet is not used as the drug of treatment; regardless, it is possible to convert the benzoyl PZQ analog fairly easily to PZQ. The two methoxy analogs are clearly of interest as they are more likely to participate in PS cyclizations. In fact the 6,7-di(MeO) analog of PZQ (2b) is itself biologically active, so again, the effective production of this molecule is an attractive possible alternative to enantiopure PZQ if the synthesis of (R)-PZQ proves intractable.



Question for consideration: P-S reactions challenging on rings with no EDG's, but how important is the amide in the reaction?

Literature: Examples of Catalytic, Enantioselective Pictet-Spengler Reactions
(Literature resources will be generally assembled here, with full bibliographic information contained in the Mendeley Group associated with this project. However, it is desirable to assemble the full state of the art at Wikipedia, to avoid duplication. There is only currently a brief account of PS reactions on WP, and no page for the asymmetric version of the reaction.)

Chiral Lewis Acids: Diisopinocampheylchloroborane Nakagawa 1996 Ditto Nakagawa 1998 Chlorosilane Lewis Acids + Ketimine substrates - Leighton 2009

Thiourea approaches: Jacobsen 2004 Anion binding Jacobsen 2007 Bronsted Acid + Thiourea Jacobsen 2009 Cationic Polycyclizations Jacobsen 2010

BINOL phosphoric acids: List 2006 Hiemstra 2008 Nakamura 2009 - but Aza-Friedel-Crafts Alkylation of Pyrroles Possible alternative thiophosphoramide: Yamamoto 2008

Systems of interest, not yet applied to PS: Bronsted Acids: N-triflyl phosphoramide Yamamoto 2006, N-triflyl phosphoramide [10.1002/anie.200802139 Rueping 2008], ? Rueping 2010 Diastereoselective polycyclization involving unactivated Ph using SnCl4 Yamamoto 2006 - suggestion here to use this enantioselectively.

Miscellaneous, not 100% relevant: Asymmetric prolinol Michael addition followed by diastereoselective PS - Fisher 2009 Ditto - Zhao 2010

Results
We showed that the synthesis of various intermediates of PZQ analogs can be xxx

First we started with conventional approach via the Shin Poong route [link]

''Racemic PS for all 4 works? Links. Conditions are bad. Di-Meo was catalytic, others not.

Chiral catalysts synthesised to date: Description of which made, with links. [figure of catalysts made]''

''We have been looking at 4 different substrates. Can make all the precursors [Links]''

Preparation of the Ugi-intermediates
The Ugi-multicomponent reaction is ... 4 components (isocyanide, amine, carboxylic acid, formaldehyde) forming a open-chain XXX product in very high yields. [Scheme] The reaction is easy to perform and 3 of the starting materials are cheap and commercially available. However highly reactive isonitrile can't be stored for a long time and has to be synthesised by ourselfes [link synthsis isocyanides]

See: Multistep synthesis of rac-PZQ (Ugi route)

Synthesis of the isocyanides For the synthesis of the isocyanides two procedures were used published in the Doemling patent [#]. Route A uses chloroform as a C1 source in a 1-step procedure but the work-up is more effort because of impurities due to side reactions. The 2-step procedure is easy to perform and the yield is slightly better.



Procedures 2-phenylethyl isocyanide: Route A and Route B

(Note: very smelly compound with a pungent odor)

Procedures 2-(3,4-Dimethoxyphenyl)ethyl isocyanide: Route A and Route B

(Note: Route B works better for this compound)


 * Novel Synthesis of Praziquantel, A. Dömling, Patent Application 2009, WO 2009/11533(A1), Language: German.

Synthesis of the Ugi-intermediates The Ugi-reaction was variied with 2 isocyanides (R1 = H, MeO) and carboxylic acids (benzoic acid, cyclohexane carboxylic acid) obtaining 4 'Ugi-intermediates' in good to very good yields. The procedure was easy to perform - the components were mixed in a 1:1:1:1 stoichiometry and stirred for 24 h at r.t.



Procedures for the Ugi-intermediates of:
 * PZQ
 * Dimethoxy-PZQ
 * N-benzoyl-protected PZQ
 * Dimethoxy-N-benzoyl-PZQ

Synthesis of the catalysts
N,N’-bis-3,5-bis[3,5-bis(trifluoromethyl)phenyl]-thiourea

This achiral version of a 'Jacobsens thiourea-catalyst' is a Bronsted-acid which is used for Screening pretests to evaluate if a reaction works under the choosen conditions without using expensive chiral versions of the catalyst.



Procedure: Synthesis of N,N’-bis[3,5-bis(trifluoromethyl)phenyl-thiourea


 * Acid-free, organocatalytic acetalization, M. Kotke and P. R. Schreiner, Tetrahedron 2006, 62, 2-3, 434-439; doi:10.1016/j.tet.2005.09.079.


 * Synthetic Studies toward Aryl-(4-aryl-4H-[1,2,4]triazole-3-yl)-amine from 1,3-Diarylthiourea as Urea Mimetics, A. Natarajan, Y. Guo, H. Arthanari, G. Wagner, J. A. Halperin and M. Chorev, J. Org. Chem. 2005, 70, 16, 6362–6368; DOI: 10.1021/jo0508189.

(+/-)-BINOL-N-triflyl phosphoramide

Achiral version of a BINOL catalyst with an acidic NH-proton, commonly used for acid-catalysed asymmetric reaction



Procedure: Preparation of (+/-)-BINOL-N-triflyl phosphoramide


 * Design of Chiral N-Triflyl Phosphoramide as a Strong Chiral Brønsted Acid and Its Application to Asymmetric Diels-Alder Reaction, D. Nakashima and H. Yamamoto, J. Am. Chem. Soc. 2006, 128, 30, 9626–9627.

1,1-Binaphthyl-2,2-disulfonate

Strong acidic achiral BINOL catalyst wearing two sulfonic acid groups



Procedure:

[http://www.ourexperiment.org/racemic_pzq/567 1. Step: 1,1’-Binaphthalene-2,2’-diyl-O,O’-bis(N,N’-dimethylthiocarbamate)]

[http://www.ourexperiment.org/racemic_pzq/583 2. Step: 1,1’-Binaphthalene-2,2’-diyl-S,S’-bis(N,N’-dimethylthiocarbamate)]

[http://www.ourexperiment.org/racemic_pzq/587 3. Step: 1,1’-Binaphthalene-2,2’-disulfonic acid]


 * Pyridinium 1,1′-Binaphthyl-2,2′-disulfonates as Highly Effective Chiral Brønsted Acid−Base Combined Salt Catalysts for Enantioselective Mannich-Type Reaction, M. Hatano, T. Maki, K. Moriyama, M. Arinobe and K. Ishihara, J. Am. Chem. Soc. 2008, 130, 16858–16860; DOI: 10.1021/ja806875c.


 * A Powerful Chiral Counteranion Motif for Asymmetric Catalysis, P. García-García, F. Lay, P. García-García, C. Rabalakos, B. List, Angew. Chem. Int. Ed. 2009, 48, 4363 –4366; DOI: 10.1002/anie.200901768.

Pictet-Spengler reactions - Cyclization
The Pictet-Spengler cyclization of the Ugi-intermediates can be cylclisized by strong very Bronsted-acids. The acids used for this cyclisation are conc. sulforic acid [ref. Kim et. al] and methane sulfonic acid [ref. Doemling] were the acids were also used as a solvent (excess of acid) at r.t. or heated at 60°C up to 2 days (?). To find a (chiral) acid which can catalyse this reaction we tested several precursors of chiral Bronsted catalysts and and also looked for the conditions which can be



Acid-mediated and acid-catalysed Pictet-Spengler reactions & Screenings

Synthesis of the dimethoxy-N-benzoyl-derivative of PZQ via Pictet-Spengler reaction (MW54-3)

Synthesis of the N-benzoyl-derivative of PZQ via Pictet-Spengler reaction (MW53-4)

Experiments to the acid-mediated Pictet-Spengler of the PZQ intermediate with various Bronsted-acids (phenyl phosphonic acid, p-toluenic acid, TFA) -> no reaction

Attempts to the acid-mediated Pictet-Spengler cyclization of the ‘Ugi intermediate’ MW29 (MW31)

Cleavage of the dimethoxy acetal of 2-((2,2-dimethoxyethyl)amino)-N-phenethylacetamide hydrochloride MW7 (MW42-1 to MW42-7)

Brosted-acid catalysed reactions:

Attempts to the Bronstedt-acid catalysed Pictet-Spengler cyclization using N,N’-bis[3,5-bis(trifluoromethyl)phenylthiourea (MW44)] -> no reaction

Attempts to the Bronstedt-acid catalysed Pictet-Spengler cyclization using BINOL-N-triflyl phosphoramide (MW41) -> no reaction

Acid-catalyzed Pictet-Spengler reaction with binaphthalenedisulfonic acid (MW56-1 to MW56-4)

Acid-catalyzed Pictet-Spengler reaction with binaphthalenedisulfonic acid (MW56-1 to MW56-4) - Starting material for unactivated comp., cycl for dimethoxy-intermediates (hemi-acetal or fragmentation from MS?)

Acid-catalyzed Pictet-Spengler reaction with methanesulfonic acid (MW56-5 to MW56-8) -> no reaction for unactivated comp, P-S cyclization for the dimethoxy-intermediates

Continuation: Acid-catalyzed Pictet-Spengler reaction with methanesulfonic acid (MW56-9 to MW56-14)

Summary
Results: Use of catalytic vs stoich achiral acid? Results. Use of chiral acids. Little reaction. What is the evidence for any hemiaminals? Starting material – lots in unactivated. Messy NMR, so some reaction.

Discussion of range of acidity – what we have tried compared to what not.

Outlook
What we plan to do: Short term: finish screening Longer term: More catalysts known – List [draw]?, and work with others

What we need: Ideas for catalysts we haven’t tried, conditions and others