Pictet-Spengler Route to (R)-PZQ

Published by MatTodd on 29 November 2007 - 1:21pm

The Pictet-Spengler reaction is used in the current industrial synthesis of rac-PZQ, and these reactions have been looked at on the Synaptic Leap here.
 
The PS has been used in the synthesis of PZQ, and we used it as the final step in our solid phase synthesis of PZQ, but both these syntheses were racemic.
 
Catalytic, asymmetric
General aim is this:

We have now started an Electronic Lab Notebook for the raw data for this specific aspect of the project here. Comments can be left here or there.
Essentially we need to start with literature catalysts for the catalytic, asymmetric Pictet-Spengler reaction. We therefore need to assemble a list of these catalysts here (literature help needed from anyone please!):
1996 Nakagawa
1998 Nakagawa
2004 Jacobsen
2007 Jacobsen
2009 Franzen
2010 Wu et al.
2010 Jacobsen
Need more candidate catalysts!
 
We are looking for people to help screen known catalysts for the relevant reaction. If your lab has any such catalysts we can send you starting material to screen. Substantial contributors to the project will be invited to be authors on the resulting papers.
Log of labs contacted for help with screening: (Important note: declining a request to participate will never be criticised - people may have very valid reasons for not wanting to participate, or may just be too busy. This log is here purely for record, to stimulate readers to suggest alternative/additional collaborating labs, and to avoid duplication of effort. If you know of a lab working on the asymmetric Pictet-Spengler reaction please feel free to contact them and leave a note of the request here.)
 
 
Diastereoselective
Wayne Best of Epichem has suggested an alternative approach based on a PS reaction of a chiral phenylethanolamine. This is an interesting idea, since 1 to 2 (below) might work well with asymmetric induction, and the products resemble PZQ. The use of a chiral starting material, in the generation of some analogue of PZQ that contains an extra stereocentre, could be an interesting project, but it's not clear these analogs would be active.  Anyone have a suggestion for a suitable commercially-available chiral starting material? Anyone want to try this?
Pictet-Spengler route to PZQ?

A chiral auxiliary-mediated PS was reported by Ma et al. (J. Chem. Res. 2004, 2004, 186-187 - paper doesn't seem to have a DOI?) where the key asymmetric step is shown below.
Literature PS Approach to PZQ

Other examples: Comins 1991
The use of an auxiliary is of course atom-inefficient, however, and likely to be expensive.

Comments

Michael Wolfle's picture

Another class of chiral Broensted acids, which are able to catalyze a Pictet-Spengler cyclisation in an enantioselective manner, are BINOL-phosphoric acids 6.

Enantioselective Broensted acid-catalyzed reactions with derivatives of those phosphoric acids are one hot spot in recent literature and here are some examples for asymmetric Pictet-Spenger reactions.
2006 List
2008 van Maarseveen, Hiemstra 

The problem of the less acidity of the proton donors (compared to sulfonic acids) we are dealing with could be solved by using strong acidic N-triflylphosphoamides 7.

2006 Yamamoto
2008 Rueping
2009 Nakamura
2010 Rueping

 

quintus's picture

Any asymmetric synthesis will probably be expensive. If you are aiming for very large quantities of product this still requires, even at 1 - 5 mol%, lots of catalyst, expensive! And you need to dispose of it from the product, probably to very low levels <10ppm?
The resolution approach should not be discarded, I would go for that route it may be cheaper in the long run.
 

Just thinking a bit latterly, what about the idea of using a chiral auxiliary, with a traceless linker, on the benzene ring. There's lots of traceless linker methodology in the solid-phase literature that might be adapated. So for example a silyl group on the benzene ring either chiral itself or with a homochiral substituent... That might be enough to give some asymmetry in the Pictet–Spengler, but even if it didn't the product would be a diastereomeric mix (and hence maybe separable). Then clip off the traceless auxiliary. All just hand-waving of course.

MatTodd's picture

Sure, why not. An interesting idea. The auxiliary would need to be close to the site of interest in the transition state. So two comments immediately, now you've stuck your head over the battlements...:
a) Do you know of any example of this kind of auxiliary?
b) Without completely pre-judging the economics, the extra steps involved, as well as a need for a homochiral component that is probably not naturally available, would be negatives to this approach?

What about:

1. PS of phenylalanine, followed by isolation of appropriate diastereomer
2. Decarboxylation of resulting alpha aminoacid using 2-cyclohexon-1-one (See A)

(A) M. Hashimoto, Y. Eda, Y. Osanai, T. Iwai, and S. Aoki
Chemistry Letters 893-896 (1986)

MatTodd's picture

That's a neat suggestion. You mean this:

where the cyclization of that acyliminium ion is guided by the stereocentre? Issue: the carboxylic acid would need to be protected during the synthesis of this P-S precursor, right?

I don't know if you are still interested in this approach, but I have a few general comments about a possible enantioselective PS. Enantioselective PS reactions are getting popular in the field of organocatalysis, and you have most of the notable examples covered there. However, when a simple phenyl group is the pendant nucleophile, I don't think that the current state of organocatalysis offers any solution. IIRC, nearly all published organocatalytic examples involve the nucleophilic tryptamine or dopamine moieties (I will do a lit search at work to check my claim).

The typical reaction conditions for simple phenylethylamine such as neat conc. H2SO4 or (stoich?) BF3OEt2 are very extreme, and do seem to be necessary (given the acid study MW42-1 to 5). I find it interesting that the N-Triflyl phosphoramide does not even promote the reaction. A more acidic version of this catalyst (DOI:10.1021/ja8041542) has been reported, but this may not be acidic enough. Product inhibition may also play a role here.

Instead of strong Bronsted acids, I wonder if a strong Lewis acid may promote the key step of this reaction catalytically. Unactivated phenyl moieties work fine in some polycyclizations using SnCl4 (DOI: 10.1021/ol062378t). Enantioselective approaches are available via BINOL-Sn complexes. The question is whether Sn will bind to the wrong lewis basic site. I think that oxophilic early transition metals have the best chance, given the amide functionality present, so this may be worth a try. As an alternative, lanthanide chlorides and triflates can offer increased functional group tolerance but attenuated reactivity.

MatTodd's picture

After the stuff we've been doing on the resolution, and with a changeover in personnel, we'll shortly be coming back to the Pictet-Spengler. Thanks, Mike for the suggestions above. Comments/suggestions/discussion can still be here, but we've assembled a wiki page where we're iteratively posting a summary of results contained within the online lab book, and we have a Mendeley page where we're assembling the relevant papers. As with everything on this project, anyone can help and modify any of this.
Mat
 

 Not sure how relevant these are, but here are some more references for the PS route/catalysts.  
Li (2008)
Kaufman (2004)
 
If there is a promising BINOL type catalyst, how does incorporation into a MOF sound?
 
Wu (2005)

MatTodd's picture

Thanks Kat. That Li paper is on catalytic asymmetric hydrog, not the P-S. Hydrog is certainly of interest for PZQ, but in a different context. The Kaufman one is a review on how to make a particular set of alkaloids, which look similar to PZQ, but I think that extra nitrogen in PZQ makes a big difference. From a quick skim they are looking specifically at those compounds, rather than something more like PZQ.
The MOF idea is interesting. The example in the Wu paper is organometallic additions, a very common reaction for catalyst evaluation, but which is no good for a PS. If you wanted to use MOFs you'd need to find an example where MOFs have been used with Bronsted acids in the structures. Ideally in an asymmetric environment, or in an environment that could be made to be asymmetric. Non-trivial I suspect, but that's what your search needs to focus on.

MatTodd's picture

...is on the draft paper on this openwetware page. You could use these lead papers to find more recent papers that are relevant. What we need to do is contact these groups, eventually, and ask if they could evaluate their catalysts (since that is quicker/more efficient than us making all of them from scratch). Murray is currently bringing through starting materials for the PS reactions and making one promising catalyst that has given a very nice-looking result on the PS for the racemic reaction.

Again, not sure of the relevance, but I think I'm getting closer.
 
2010 Berkessel
 
Also, I'm not sure how to add it to the wiki if it is relevant...
 

MatTodd's picture

That looks more like it. Do they use them for the PS? Either way, strong chiral Bronsted acids have got to be relevant for us here, even though I question whether that should work at all based on the mechanism of the PS - what's the enantiodetermining step, and where's the acid there?
 
But absolutely - if we get a lead result we may then need to think about changes, and we need to be aware of what changes will make the acid stronger. For the wiki, get an account on openwetware, then click the edit button to see how the page is constructed. A wiki is a learning exercise in itself because the source is always available.