Design and application of α-ketothioesters as 1,2-dicarbonyl-forming reagents

The 1,2-dicarbonyl motif is vital to biomolecules, especially natural products and pharmaceuticals. Conventionally, 1,2-dicarbonyl compounds are prepared via an α-keto acyl chloride. Based on the methods used in nature, a transition-metal-free approach for the synthesis of an α-ketothioester reagent via the combination of an α-hydroxyl ketone, elemental sulfur and a benzyl halide is reported. Mechanistic studies demonstrate that the trisulfur radical anion and the α-carbon radical of the α-hydroxy ketone are involved in this transformation. The dicarbonylation of a broad range of amines and amino acids, and importantly, cross couplings with aryl borates to construct dicarbonyl-carbon bonds are realized under mild conditions by employing this stable and convenient α-ketothioester as a 1,2-dicarbonyl reagent. The dicarbonyl-containing drug indibulin and the natural product polyandrocarpamide C, which possess multiple heteroatoms and active hydrogen functional groups, can be efficiently prepared using the designed 1,2-dicarbonyl reagent.

2. The thioester will remove after the 1,2-dicarbonyl-forming reagents transform to other useful compounds. So, did the authors try to use other electrophilic reagents instead of only alkyl bromides to reduce the coat and toxicity or increase atom economy?
3. The authors demonstrated the practicability of α-ketothioester through many examples (table 3 and 4), but these examples were only simple extension of two kinds of reaction. More types of reactions will lead to more persuasion instead of rigid extension of single reaction.
Reviewer #3 (Remarks to the Author): The article by Jiang and co-workers describes a transition-metal-free approach for the synthesis of an α-ketothioester reagent via the combination of an α-hydroxyl ketone, elemental sulfur and a benzyl halide. This convenient process avoids the use of malodourous thiols and the corresponding products could be easily converted into 1,2-dicarbonyl compounds, which are commonly encountered in pharmaceuticals and biomolecules. The authors have also conducted some preliminary experiments to gain insights into the mechanism of the reaction. This is a good piece of work and I recommend its publication in Nature Communications after minor revisions as noted below.
1. The authors have performed the experiments well with aryl-substituted ketones. What happens with alkyl-or vinyl-substituted 2-hydroxy-ethanones? The 2-hydroxy-esters is also widely available, are they suitable for this transformation? The authors should comment on this in order to provide an idea about the scope of the reaction. If the reaction works with the substrates mentioned above, then a few more examples should be added to demonstrate the scope.
2. The authors have carried out some preliminary investigation about the mechanism in Fig. 3. These results should be described in the text (data is in the figure only). Also in the eq. 1 of Fig. 3, how does it work to generate the 2aa?
3. It takes two steps to generate the final 1,2-dicarbonyl compounds. Did the authors try the cascade or one-pot reaction? It would be potentially more useful if it can be done in one step to generate the corresponding amides and ketones. 4. In SI, some NMR spectrum is not pure, please purify them. For example, 2o, 3x, 7e. 5. In the NMR traces of Supplementary Information, a caption should be included on the NMR spectrum, noting the nucleus being measured, the solvent (formula preferred, e.g. CDCl3 over Chloroform-d), and the field strength.

Responds to the reviewers' comments
Reviewer #1: Q1: This journal is aimed at publishing high impact, cutting edge new developments in Chemistry, and this paper is far too specialised for this. The basic idea has multiple exemplifications and methods, and this one, though novel, is only an iterative improvement and is an extension of expected chemistry. It would be more appropriate submitted to an organic journal.

A1:
This referee raised a reasonable point here that this journal is aimed at publishing high impact, cutting edge new developments in Chemistry. However, we do not agree with this referee that this paper is far too specialized, only an iterative improvement and an extension of expected chemistry. We would like to share our thoughts and rational that justifies the publication of this work on Nature Communications.
1) These novel 1,2-dicarbonyl reagents can be easily constructed from safe, odorless and readily available inorganic elemental sulfur via a transition-metal-free approach. Mechanistic studies demonstrated that the trisulfur radical anion initiated the α-carbon radical of the α-hydroxy ketone in this transformation. This is also a new achievement for radical chemistry.
2) The dicarbonylation of a broad range of N-nucleophiles, O-nucleophiles, C-nucleophiles and even amino acids with sensitive chiral centers. Importantly, cross couplings with commercial available aryl borates to construct dicarbonyl-carbon bonds were first realized under mild conditions with this stable and convenient α-ketothioester, which reveal the unique versatility and applicability of the 1,2-dicarbonyl reagents.
3) To further highlight the applicability of the 1,2-dicarbonyl reagents, pharmaceutically relevant molecules indibulin containing indolyl and pyridyl, natural products polyandrocarpamide C possessing free hydroxyl and a 9,10-phenanthrenequinone derivative with strong electron-rich aromatics were afforded with this reagent.
We believe that these 1,2-dicarbonyl reagents and broad-spectrum method will be of great interest for the field of organic chemistry, biochemistry and medicinal chemistry in general.
Reviewer #2: "recommend its publication" Q1: The authors should make some comments on the representative experimental phenomena: a. In Table 1, why the water is necessary? Why could the TBAB promote the reaction? Did the authors try to use organic base without water and TBAB?
b. Solvent has obvious influence on the reaction, and the authors should make some comments to explain the effect of solvents.
c. When α-hydroxy ketones bear electron-withdrawing substituents on the aromatic ring, the reaction have to changed conditions. This phenomenon indicated that the electronic effect of a-hydroxy ketones had significant influence on the reactivity, and the authors should illuminate the reason (stability or lifetime of radical).  Table).  was afforded in 35% yield, which provided strong evidence of α-carbon radical being involved in this transformation (Figure 3-2). The coupling product 2ab was not observed in the absence of elemental sulfur under the standard conditions, which indicate that sulfur is enssential for radical generation ( Figure   3-3)".

Q2:
The thioester will remove after the 1,2-dicarbonyl-forming reagents transform to other useful compounds. So, did the authors try to use other electrophilic reagents instead of only alkyl bromides to reduce the coat and toxicity or increase atom economy?

A2:
The simplest electrophilic reagent should be H + , which will generate the 1,2-dicarbonyl compound A. Unfortunately, compound A is not a suitable 1,2-dicarbonyl reagent, since -SH group is more difficult to remove than -SR group with strong coordination, strong smell and sensitive unstable property Radical-trapping experiments were conducted through investigating TEMPO under the standard conditions. The coupling product 2ab was afforded in 35% yield, which provided strong evidence of α-carbon radical being involved in this transformation (Figure 3-2). The coupling product 2ab was not observed in the absence of elemental sulfur under the standard conditions, which indicate that sulfur is enssential for radical generation (Figure 3-3)." For side product 2aa, hydration of substrate 1a' generates 1a'', which affords