A Mild Method for Dehydration of a Base-Sensitive Primary Amide- Toward a Key Intermediate in the Synthesis of Nirmatrelvir

Nirmatrelvir, the active ingredient of the Pfizer drug Paxlovid (Figure 1), is an inhibitor of the SARS-CoV-2 main protease enzyme.¹

A key transformation- the final synthetic step in the synthesis of Nirmatrelvir- is dehydration of a primary amide (Figure 2).The amide starting material is prepared by reaction of the corresponding ester with ammonia.²

Various methods for the dehydration process have been disclosed by Pfizer including use of the Burgess reagent, T3P, trifluoroacetic anhydride or POCl₃ imidazole/pyridine.³ These methods, though effective, generate significant quantities of waste and varying amounts of the carboxylic acid impurity via product or substrate hydrolysis. A recent paper by the Lipshutz group caught my attention. In it they describe a new, sustainable, synthesis of Nirmatrelvir, including a method of generating the nitrile utilizing a recently reported catalytic dehydration reaction mediated by palladium.⁴ The reaction, used to prepare the cyclic glutamine intermediate (A) from a commercially available protected amino acid, is high yielding and proceeds without racemisation (Figure 3).

The chemistry proceeds via an “amide exchange” process using fluoroacetonitrile as a sacrificial water scavenger under palladium catalysis. The methodology was originally published by Naka et al,⁵ in his case using dichloroacetonitrile as a water acceptor. The halogenated nitrile preferentially reacts with amides over other polar functional groups and, with the aid of a suitable Palladium catalyst, drives the dehydration process (Figure 4). The water acceptor was designed to be kinetically reactive in the amide dehydration giving a thermodynamically stable co-product, thus driving the forward reaction and generating the required nitrile product. Compatible with water, alcohols and acids-the method has impressive substrate scope.

Presumably the Lipshuz team chose fluoroacetonitrile as a suitably volatile material without having to use a heavily chlorinated (environmentally unfriendly) reagent. The synthesis and chemistry of fluoroalkyl nitriles has been reviewed by Usachev.⁶ I can’t speak to the toxicity of these materials, but my guess is they’re potentially very nasty. In a later paper, Lipshutz describes carrying out a comprehensive screen of different functionalised acetonitrile derivatives as water acceptors.⁷ Methoxyacetonitrile was found to give similar results to fluoroacetonitrile (with a model substrate). Its not obvious why the fluoro-derivative was used in his synthesis of Nirmatrelvir.

Pd loadings are relatively low (<1 mol%) and the ligand is nothing unusual (Pd(OAc)₂ or Pd(CH₃CN)₄(BF₄)₂). A large excess of the nitrile scavenger is required (4 eqv. of FCH₂CN in the Nirmatrelvir synthesis).

Lipshutz has also published an extension of this method leveraging his well-established aqueous micellar methodology (TPGS-750-M).⁷ The method appears general and synthetically useful.

Shen and Aisa have also published a synthesis of the primary amide precursor via cyanomethylation of dimethyl N-BocGlu in the presence of NdCl₃, followed by one-pot Raney nickel-catalyzed hydrogenation of the cyano group with concomitant cyclization and ammonolysis (Figure 5).⁸ The Neodymium additive increases the cyanomethylation anti/syn ratio from 96:4 to an impressive >99:1. A similar approach has been reported by Chandrasekhar.⁹

Some useful chemistry with an important application.

See you next time

References:

1. The path to Paxlovid: B. Halford, ACS Cent. Sci. 2022, 405-407; Nirmatrelvir plus Ritonavir: first approval: Y. Lamb, Drugs 2022, 82, 585-591
2. Dehydration of amides to nitriles: a review: N. Bhattacharyya et al, Int. J. Chem. Appl. 2012, 4, 295-304; Recent developments in dehydration of primary amides to nitriles: M. Ganesan et al, Chem. Front., 2020, 7, 3792-3814
3. An oral SARS-CoV-2 Mpro inhibitor clinical candidate for the treatment of COVID-19: D. Owen et al, Science 2021, 374, 1586-1593 (supplementary data); US20220062232; WO202125249; Alternate end-game strategies towards Nirmatrelvir synthesis: Defining a continuous flow process for the preparation of an anti-COVID drug: S. Oruganti et al, Tet Lett. 2023, 116, 154344
4. A sustainable synthesis of the SARS-CoV‐2 M^pro inhibitor Nirmatrelvir, the active ingredient in Paxlovid: B. Lipshultz et al, Commun. Chem 2022, 5, 156; A 1‐pot synthesis of the SARS-CoV‐2 M^pro inhibitor Nirmatrelvir, the key ingredient in Paxlovid: ibid Org. Lett. 2022, 24, 9049-9053
5. Acceptor-controlled transfer dehydration of amides to nitriles: H. Naka et al, Org. Lett. 2019, 21, 4767–4770
6. Chemistry of fluoroalkyl cyanides: B. Usachev, Arkivoc, 2020, Part 1, 499-577
7. Dehydration of primary amides in water. Late-stage funtionalization and 1-pot multistep chemoenzymatic processes under micellar catalysis conditions: B. Lipshutz et al, Green chem. 2022, 24, 2853-2858
8. Optimised synthesis of a key intermediate of Nirmatrelvir: J. Shen & H Aisa et al, Org. Process Res. Dev. 2023, 27, 78-83