A selenium-catalysed para-amination of phenols

Antioxidant enzyme glutathione peroxidase (GPx) decomposes hydroperoxides by utilizing the different redox chemistry of the selenium and sulfur. Here, we report a Se-catalysed para-amination of phenols while, in contrast, the reactions with sulfur donors are stoichiometric. A catalytic amount of phenylselenyl bromide smoothly converts N-aryloxyacetamides to N-acetyl p-aminophenols. When the para position was substituted (for example, with tyrosine), the dearomatization 4,4-disubstituted cyclodienone products were obtained. A combination of experimental and computational studies was conducted and suggested the weaker Se−N bond plays a key role in the completion of the catalytic cycle. Our method extends the selenium-catalysed processes to the functionalisation of aromatic compounds. Finally, we demonstrated the mild nature of the para-amination reaction by generating an AIEgen 2-(2′-hydroxyphenyl)benzothiazole (HBT) product in a fluorogenic fashion in a PBS buffer.

Referee report for Zhao et al. Selenium Enables the Catalytic Intramolecular para-Amino Functionalization of Phenols" (ms. no. NCOMMS-18-10930) This manuscript from Zhao et al. is a rewrite of a previous manuscript submitted to Nature Catalysis entitled "Organoselenium-Catalyzed para-Amidation of Phenols" (ms no. NATCATAL-17070154). Because the authors have chose to recast the manuscript and resubmit to the Nature group, without substantial change, I have reiterated my previous review below. The same conclusion applies here.
This manuscript from Zhao et al. is a follow up to their recently published article in Nature Communications (ref. 34) in which an identical transformation was carried out with a sulfenylating agent in place of a selenylating agent. In that study, the reaction of the aryl hydroxamic esters proceeded unsurprisingly to produce an ortho substituted sulfilimine. Herein, the identical process is executed, but the authors now find that the reaction does not stop at the ortho substituted selenimine, but rather continues on through a second [2,3] rearrangement to afford a para substituted phenol. The authors briefly survey selenylating reagents for their catalytic efficiency and then carry out a standard demonstration of scope. This section is followed by illustrations of routine transformations directed by the amide group as well as a few mechanistic studies that establish the intramolecular nature of the rearrangement. Finally a straightforward computational study substantiates the proposed mechanism and rules out a direct 1,5 shift.
Overall this is an interesting transformation but one that hardly rises to the stature of Nature Catalysis. The focus (unlike the previous article which emphasized the biocompatibility of the transformation) is primarily chemical, and as such is better suited for publication in the Journal of Organic Chemistry or Chemistry European Journal.
Reviewer #3 (Remarks to the Author): This manuscript reports some interesting and publishable results, but I don't think the results are sufficiently significant to warrant publication in Nature C ommunications. I recommend publication in a more specialized journal.
The authors use DFT computations to support their observations and discussion of the experimental results, which is fine, but they provide no justification for their choice of the DFT method and basis sets employed. I suspect they have selected the DFT method on the basis of the publications by other authors on molecules containing selenium. A few key references should be added to the manuscript. The authors refer to free energies and Gibbs free energies. Thermodynamicists have long advocated using the term Gibbs energies and IUPAC has adopted their recommendation. Another trivial point: Angstroms should be written as angstroms to be consistent with SI conventions: joule or J, kelvin or K, etc.

Reviewers' comments:
Reviewer #1 (Remarks to the Author): The manuscript presents an organoselenium-catalyzed para-amidation reaction of phenols. The reaction is proposed to proceed via a double [2,3]-sigmatropic rearrangement, and proves to be applicable to a broad scope of amide and phenol substrates. The manuscript also presents an analogous organosulfur-promoted paraamidation reaction of phenols, which leads to different amidated products bearing the N-S bond. The authors have performed control experiments and computational modeling to probe the reaction pathways, providing some insights on the possible intermediates in both N-S and N-Se mediated rearrangements. The studies are well performed with sufficient experimental information. The reported chemistry is interesting and novel, which is expected to find great use in organic chemistry.
Overall, the work is suitable for publication at Nature Communication. Yet before it can be accepted for publication, the manuscript should be revised to address the following questions/comments adequately.

Response
We thank the reviewer for this valuable question. More details about the fundamental difference between selenium and sulphur as well as their different behaviors in biology are discussed in our new edition. We have added several references in the revised manuscript.
In summary, we discovered an organoselenium-catalysed para-amination of phenols or dienones under mild conditions. The methodology features a broad substrate scope and a high para-selectivity. More importantly, this work reveals significant a difference between the sulfenylation reagents and organoselenium reagents. While experimental and computational studies suggest that both the sulphur and selenium variants proceed through a double [2,3]-sigmatropic rearrangement, the sulfenylation reagents behave as coupling partners while organoselenium reagents can be employed catalytically. Since the larger atomic radius of selenium compared to sulphur, selenium are more polarizable ("softer") than sulphur, allowing selenium intrinsic to be more nucleophilic and electrophilic 61,62 . Compared to sulphur, the larger hybridized orbitals of selenium results in weaker σ overlap 63 . So most bond strength of Se−X is weaker. The differences between sulphur and selenium developed here is reminiscent of their behaviors in biology, illustrates the potential of selenium to enable catalytic processes. Notably, type II and type III reactions reveal a novel organoselenium-catalyzed reaction in C-H functionlization with the unprecedented para-selectivity. For example, the catalytic activity of the native enzyme dramatically reduces when the Sec residue in the type I ID enzyme was replaced by a cysteine (Cys) moiety 64,65 . We expect our present work to stimulate future studies of selenium as an alternative catalytic platform to transition metal-catalysed C-H amination reactions.

Response
We thank the reviewer for this valuable advice. We have renamed the reactions in Table 1-3 as Se-catalysed (referred to Type II and III) and S-mediated (referred to Type I) reactions. Details have been displayed in the revised manuscript.  Table 2 in SI, and little information is provided in Figure 2 despite considerable amount of discussion in the manuscript. The authors are suggested to move these results in supplementary Table 2 in SI to the manuscript for clarity.

Response
We thank the reviewer for this valuable advice. We have move the Supplementary Table 2 in SI to the new manuscript as Table 3.

Revisions Made
(Please refer to page 7, Table 3).

Response
We are very sorry for our negligence of positive charges. We have checked all the compounds and corrected accordingly.

Response
We were really sorry for this mistake. We have checked all the J coupling constants of the compounds and corrected them for accuracy.
Here is another one: "Moreover, the developed reaction conditions were applicable to polycyclic substrate with moderate yield (4l, 54%)." Based on a single example, specifically naphthalenol derived amidated product 4l, the description aforementioned is an inappropriate overstatement for its applicability for polycyclic substrates.

Response
We thank the reviewer for his/her constructive advice that have helped us to improve our manuscript. We have rewritten the introduction and corrected the mistakes. The following revision is provided.

Revisions Made
(Please refer to Page 2, paragraphs 1, 2; Page 3, Fig. 1) Selenium is an essential biological trace element discovered by the Jöns Jacob Berzelius in 1818 1 . The selenium analogue of cysteine, known as selenocysteine 2-4 (Sec), is the main biological form of selenium. The most studied selenoenzyme glutathione peroxidases (GPx) have a Sec residue in its active site which is responsible for decomposing hydroperoxides (Fig. 1a) (Fig.   1b). However, no selenium-catalysed processes for the functionalisation of aromatic compounds have been developed. One challenge might be the electrophilic selenium catalysts (ESC) react with the aryl rings directly, leading to the deactivation of catalyst 37,38 . We thought that a more nucleophilic site, to accommodate with selenium catalyst temporarily, might be helpful for competing with the deactivation. We herein report a strategy to first form an intermediate with an adjacent, redox versatile Se-N bond which undergoes two successive sigmatropic rearrangements to generate the paraamination product and regenerate the selenium catalyst (Fig. 1c).
(Please refer to Page 4, paragraph 4) Moreover, the developed reaction conditions were applicable to polycyclic substrate with moderate yield (4l, 54%).
The reaction condition was applicable to yield aminated naphthol in 54% yield (2l). with a sulfenylating agent in place of a selenylating agent. In that study, the reaction of the aryl hydroxamic esters proceeded unsurprisingly to produce an ortho substituted sulfilimine. Herein, the identical process is executed, but the authors now find that the reaction does not stop at the ortho substituted selenimine, but rather continues on through a second [2,3] rearrangement to afford a para substituted phenol. The authors briefly survey selenylating reagents for their catalytic efficiency and then carry out a standard demonstration of scope. This section is followed by illustrations of routine transformations directed by the amide group as well as a few mechanistic studies that establish the intramolecular nature of the rearrangement. Finally a straightforward computational study substantiates the proposed mechanism and rules out a direct 1,5 shift.
Overall this is an interesting transformation but one that hardly rises to the stature of Nature Catalysis. The focus (unlike the previous article which emphasized the biocompatibility of the transformation) is primarily chemical, and as such is better suited for publication in the Journal of Organic Chemistry or Chemistry European Journal.

Response
We thank the reviewer for encouraging comments on our manuscript. We have rewritten the introduction which emphasises the significance of our work for the development of selenium catalysis on aromatic compounds.
Besides, we also demonstrated the biocompatibility of the para-amination reaction by generating an AIEgen 2-(2'-hydroxyphenyl)benzothiazole (HBT) product in a fluorogenic fashion in a PBS buffer (Fig. 4).

Revisions Made
(Please refer to Page 2, paragraphs 1, 2; Page 3, Fig. 1) Selenium is an essential biological trace element discovered by the Jöns Jacob Berzelius in 1818 1 . The selenium analogue of cysteine, known as selenocysteine 2-4 (Sec), is the main biological form of selenium. The most studied selenoenzyme glutathione peroxidases (GPx) have a Sec residue in its active site which is responsible for decomposing hydroperoxides (Fig. 1a) 5,6 . Besides, the flavin-containing redox enzyme  (Fig.   1b). However, no selenium-catalysed processes for the functionalisation of aromatic compounds have been developed. One challenge might be the electrophilic selenium catalysts (ESC) react with the aryl rings directly, leading to the deactivation of catalyst 37,38 . We thought that a more nucleophilic site, to accommodate with selenium catalyst temporarily, might be helpful for competing with the deactivation. We herein report a strategy to first form an intermediate with an adjacent, redox versatile Se-N bond which undergoes two successive sigmatropic rearrangements to generate the paraamination product and regenerate the selenium catalyst (Fig. 1c).