methyl (2S)-(2-chlorophenyl)(6,7-dihydrothieno[3,2-c]pyridin-5(4H)-yl)acetate

(+)-Clopidogrel (sold as Plavix) is a common pharmaceutical used to help combat the risk of heart disease. The World Health Organisation lists Plavix on its list of 'essential medicines' due to its broad efficacy, low toxicity and low cost (less than 1USD/month in the developing world).

Many industrial routes rely on racemic synthesis and resolution, however syntheses of a single isomer have been reported.

The molecule makes an interesting target as it requires the synthesis of an unusual heterocyclic core, as well as the challenging stereocentre which is prone to racemisation.

Some ground rules:

  • the synthesis must include formation of the heterocyle (the whole thing, both rings), and a way for setting the stereocentre
  • you can start from anything you like (subject to the above condition), but it must be available in the Sigma Aldrich catalogue
  • $\begingroup$ Sorry this is late (again). $\endgroup$
    – NotEvans.
    Nov 19, 2017 at 17:19
  • 1
    $\begingroup$ One answer :( - guess end of year makes people busy... $\endgroup$ Dec 6, 2017 at 13:31
  • $\begingroup$ @NotEvans. when's the next challenge $\endgroup$
    – Laksh
    May 18, 2019 at 10:45

1 Answer 1


This was a tricky synthesis. There were several issues. One is that the aryl glycine is possibly easy to epimerize, though the ortho substitution may help prevent that. One other major issue is that the thiophene is reducing and not compatible with any oxidative conditions and probably not compatible with electrophiles.

The amino acid moiety can be synthesized in several ways to set the stereogenic center. One is epoxide ring opening, but here I opted for an asymmetric Strecker reaction.

The six member ring was created via a [4+2] cycloaddition. This is probably the sketchiest portion of the synthesis.

The thiophene moiety was created using Paal-Knorr reaction with Lawesson reagent (B).

Detailed route:

  1. imine formation
  2. asymmetric Strecker with 5% PHTA (partially hydrolyzed titanium alkoxide) and 5% ligand (L). 1 eq tBuOH, tol. (Org. Lett., 2010, 12, 264.) Also, possible with Hoveyda type ligand: (JACS, 2000, 122, 2657.) This exact substrate is reported with good ee.
  3. Hydrolysis with HCl. Similar route reported in (JACS, 1999, 121, 4284.) Though there, the authors Boc-protect at the end. We'll attempt to trap a different way...
  4. formalin, toluene, pTsOH. reflux. Product imine is A.
  5. LDA; TMSCl on methyl vinyl ketone to make electron rich diene.
  6. Hopefully, a little heat will coax this [4+2] cycloaddition with A to take place.
  7. Tsuji-Trost allylation is established for these types of TMS-protected enol ethers. Conditions are $\ce{Pd_{3}(dba)_{2}}$, DPPE, allyl methyl carbonate, THF. (Chem Lett., 1983, 1325.) We'll carry both diastereomers forward.
  8. $\ce{O3}$; DMS.
  9. Lawesson reagent (B).

route1 route2

  • $\begingroup$ nice one @Zhe I was thinking of asymmetric imine hydrogenation, but the Strecker Rxn kills two birds with one stone $\endgroup$
    – gannex
    Nov 27, 2017 at 0:22
  • $\begingroup$ @gannex I was thinking of that originally, too, but I don't have good access to lit searches, and I was worried about selectivity on the iminium ion... $\endgroup$
    – Zhe
    Nov 27, 2017 at 0:23
  • $\begingroup$ I was thinking the same thing. I was researching it but then I went to sleep. Looks like that was a mistake :-P $\endgroup$
    – gannex
    Nov 27, 2017 at 0:25
  • $\begingroup$ Conceptually a nice idea. I am concerned about the formation of imine A under aqueous acid conditions (40% formalin). How do you isolate it? Might you consider using gaseous formaldehyde in the presence of the amine and diene. I entertained the thought of an isonitrile (from the formamide) as the dienophile. Couldn't find a precedent. If viable, you would have a ring with an extra double bond. After allylation and ozonolysis of the protonated amine (to protect it), the double can be removed by hydrogenation. $\endgroup$
    – user55119
    Nov 27, 2017 at 0:40
  • $\begingroup$ @user55119 Interesting. I also considered just using a solution of formaldehyde. Probably easier. My first attempt at the [4+2] used Danishefsky's diene, but I had problems manipulating double bonds after the cycloaddition. Using an isonitrile is interesting., but we'd need to be careful about epimerizing the stereocenter. $\endgroup$
    – Zhe
    Nov 27, 2017 at 0:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.