Biomimetic approach to the catalytic enantioselective synthesis of tetracyclic isochroman

Polyketide oligomers containing the structure of tetracyclic isochroman comprise a large class of natural products with diverse activity. However, a general and stereoselective method towards the rapid construction of this structure remains challenging due to the inherent instability and complex stereochemistry of polyketide. By mimicking the biosynthetic pathway of this structurally diverse set of natural products, we herein develop an asymmetric hetero-Diels–Alder reaction of in-situ generated isochromene and ortho-quinonemethide. A broad range of tetracyclic isochroman frameworks are prepared in good yields and excellent stereoinduction (up to 95% ee) from readily available α-propargyl benzyl alcohols and 2-(hydroxylmethyl) phenols under mild conditions. This direct enantioselective cascade reaction is achieved by a Au(I)/chiral Sc(III) bimetallic catalytic system. Experimental studies indicate that the key hetero-Diels-Alder reaction involves a stepwise pathway, and the steric hindrance between in-situ generated isochromene and t-Bu group of Sc(III)/N,N’-dioxide complex is responsible for the enantioselectivity in the hetero-Diels–Alder reaction step.

This work takes advantage and is a very nice illustration of cooperative catalysis with two catalysts individually activating two substrates in two separate catalytic cycles making possible a highly challenging transformation which otherwise would not have been possible or at least have taken several more steps. In addition, it is the first synthesis of linearly fused chromans instead of the previously achieved synthesis of spirocyclic chromans. Taken together this manuscript is strongly recommended for publication in Nature Communications.
Reviewer #2: Remarks to the Author: In this manuscript, the authors realized the construction of tetracyclic isochromans through the biomimetic approach. The HDA reaction between o-QM and isochromene, which are in-suit generated by using Au(I) and Sc(III)-N,N′-dioxide complex respectively, supplies a series of tetracyclic products in good results. These two catalysts show good compatibility in this design. However, the narrowed substrates scope has been presented in this work and the detective of mechanism is not enough as well. We believe the manuscript could be published only if the authors could supply more information as required. The main arguments are listed below: 1. The manuscript needs polishing since there are too many mistakes presented. The color of handwriting should be the same black but there are red characters in the paper. In line 86, "bronsted" should be "Brønsted"; In line 102, "L-Pit-BuMe2" should be modified to maintain consistency with that in other parts; In line 103, "CHCl3" should be "CH2Cl2"; In line 117, "ee" should be italic and such careless mistakes should be double checked full text; In line 179 and 182, "N,N'" should be italic; In line 207, 213, 218, "°C" should be checked. … 2. On the other hands, the SI should be carefully checked words by words as well. In many cases, the 1H NMR experiments are carried out at 400 MHz but the odd J value are presented which is not possible; Since the products are chiral white solids, then, the melting points together with the optical rotations must be supplied; The HRMS data in cases like 4d, 4f, 4g, 4h, 4m, 4q, 4r, 4s, 4v, 4w, 4y, 6 are not convincing enough since the great deviation; In copies of NMR spectra, the structure of compounds and the serial number of compounds should be fully presented but not part of them. Besides, since the d.r. values are determined by 1H NMR, they should be clearly exhibited in spectra but you have missed unfortunately. In addition, in case 4u, 13C NMR should keep one decimal as others. In HPLC spectra, the impurities are common in spectra which make us doubt about the accuracy of yields. You are required to pure all compounds and supply newly prepared HPLC spectra. In cases like 4d, 4f, 4m, 4n, 4r, 4w, 5, 6, the racemates are presented not even racemic or wrong data, you need a careful check for all these mistakes.
3. As for substrate scope, only ortho-hydroxybenzyl alcohols and α-propargyl benzyl alcohols have been tested in you work. You are asked to test substrates like ortho-aminobenzyl alcohols and αpropargyl benzyl amines. On the other hands, only six membered ring system have been examined, how about 5, 7, 8, or larger ring system? 4. The mechanism needs more details. By the way, other catalyst systems should be examined as well to show the superiority of your system. Chiral phosphoric acid, chiral Au(I) system, other ligands like Box and so on are required in your work. 5. Org. Chem. Front. 2014, 1, 298；Chin. J. Chem. 2021 and Aldrichimica ACTA 2020, 53, 3 should be cited.
Reviewer #3: Remarks to the Author: This manuscript, by Li, Liu and coworkers, reported an asymmetric hetero-Diels-Alder reaction between in-situ generated isochromene and ortho-quinonemethide via a Au(I)/chiral Sc(III) bimetallic catalytic system. The authors carried out the experimental studies and disclosed that a stepwise pathway and the enantioselectivity of this reaction was controlled by proposed in-situ generated Au(I)-isochromene complex and Sc(III)/N,N'-dioxide complex. The reaction scope was investigated and gave several tetracyclic isochroman frameworks fromα-propargyl benzyl alcohols and 2-(hydroxylmethyl) phenols.

Suggestions:
1) The authors carried out the model reactions using 2-(Hydroxylmethyl) phenol (2a) and αpropargyl benzyl alcohol (3a). Scheme 2 and 3 show the reaction scope by changing the substituted groups of two reactants. Actually, these adducts have very similar and specific structures. During the mechanism studies, substrates 7r and 7s (Scheme 4c) were tested. Interestingly, 8a was obtained from 7s. This reviewer suggests the authors think about the practicability of this reaction in current version. More information should be provided to guide the potential utilities by other groups. a) how about the substrates bearing alkyl groups on the benzylic position, b) how about the substrates bearing internal alkynes; c) instead of -propargyl benzyl alcohols, how about directly using hexynol or related alkynol without benzene ring. The author should provide both the advantages and the limitations of this reaction.
2) A stepwise reaction pathway was proposed based on the studies of carbon isotope effects for the [4 + 2] reaction between 2b and 3a. In principle, two diastereomers should be produced during the second bond forming step. This review suggests a detailed studies should be performed by direct using 3-methyl-1H-isochromene (7, Scheme 5) as a reactant. This will give insight about the reaction pathway. 3) In figure 1, proposed stereochemical models were given. A crucial question is what's role of the Au-complex. Is there any interaction between Au with the ligand? Is that necessary to in-situ form the Au-complex? No information was provided.
4) The compounds numbering is messy in the manuscript. For examples, 2a, 7, 8 were used twice in different Schemes. The authors should recheck the manuscript carefully and remove these careless errors before submission.

Response to referee 1: Li, Liu and coworkers report the cooperatively catalyzed (4+2)-cycloaddition of in situ generated ortho-quinone methides and enol ethers to produce a broad range of highly valuable tetracyclic isochromans carrying three adjacent stereogenic centers including one tetrasubstituted acetal center. Product yields are typically moderate to good (55-75% for most products), the diastereoselectivity good to very good (10-20:1 d.r.) and the enantioselectivity excellent (ca. 90% ee for most products). As catalysts they employ chiral scandium-N,N-dioxides developed by Feng to promote the dehydration of the starting ortho-hydroxy benzyl alcohols and formation of Lewis acid-bound ortho-quinone methides and a gold-JohnPhos-complex to effect the hydroxylation of the alkyne to the enol ether. The process benefits from the simple and readily available starting materials and the straightforward, one-step generation of complex heterocyclic scaffolds frequently found in polyketide natural products. Moreover, careful mechanistic studies have been undertaken to elucidate the exact reaction mechanism. This work takes advantage and is a very nice illustration of cooperative catalysis with two catalysts individually activating two substrates in two separate catalytic cycles making possible a highly challenging transformation which otherwise would not have been possible or at least have taken several more steps. In addition, it is the first synthesis of linearly fused chromans instead of the previously achieved synthesis of spirocyclic chromans. Taken together this manuscript is strongly recommended for publication in Nature Communications.
A: We greatly appreciate your highly postive comments. A1: Thank you for your carefulness in reviewing our manuscript. All mistakes have been corrected.

Q2-1. On the other hands, the SI should be carefully checked words by words as well. In many cases, the 1H NMR experiments are carried out at 400 MHz but the odd J value are presented which is not possible;
A2-1: Thanks a lot! According to your suggestion, we checked the SI very carefully and revised some mistakes. We also recalculated the J values of 400 MHz 1 H NMR by multiplying difference of chemical shift of keeping three decimal places by 400. It needs to note that original J values were calculated by keeping four decimal places and rounding to one decimal place.

Q2-2. Since the products are chiral white solids, then, the melting points together with the optical rotations must be supplied;
A2-2: The melting points together with the optical rotations were supplied in the revised Supporting Information. 4d, 4f, 4g, 4h, 4m, 4q, 4r, 4s, 4v, 4w, 4y, 6 are not convincing enough since the great deviation;

Q2-4. In copies of NMR spectra, the structure of compounds and the serial number of compounds should be fully presented but not part of them. Besides, since the d.r. values are determined by 1 H NMR, they should be clearly exhibited in spectra but you have missed unfortunately. In addition, in case 4u, 13 C NMR should keep one decimal as others.
A2-4: In the copies of NMR spectra, all of the compounds have been fully presented. All d.r.
values have been exhibited in 1 H NMR spectra. Moreover, 13 C NMR of 4u has kept one decimal as others. cases like 4d, 4f, 4m, 4n, 4r, 4w, 5, 6, the racemates are presented not even racemic or wrong data, you need a careful check for all these mistakes.

A2-5:
We repeated all the experiments which did not supply clean HPLC traces. The products were purified as much as we can. The yields were recalculated and the clean HPLC traces were supplied. Moreover, the racemates of 4d, 4f, 4m, 4n, 4r, 4w, 5, 6 were obtained again and were presented as racemic version in HPLC.

Q3. As for substrate scope, only ortho-hydroxybenzyl alcohols and α-propargyl benzyl alcohols have been tested in you work. You are asked to test substrates like ortho-aminobenzyl alcohols and α-propargyl benzyl amines. On the other hands, only six membered ring system have been examined, how about 5, 7, 8, or larger ring system?
A3: Thank you for your suggestions. According to your suggestions, a variety of ortho-aminobenzyl alcohols, Cbz-protected α-propargyl benzyl amine 3j as well as various alkyl benzyl alcohol derivatives 3k-o were prepared and investigated in the present catalytic reaction system under the standard reaction conditions. Cbz-protected α-propargyl benzyl amine 3j can react with 2h quite well to form the corresponding aza-poliketide oligomer 5a in 63% yield with 22% ee for major isomer. Moreover, 5, 7 and 8 membered ring systems were examined. The reactions between 2h and α-ethynyl benzyl alcohols 3k-l were carried out with L-PiAd as ligand and chiral spiro products 5b-c were isolated in 72-76% yield and 52-55% ee values in 5 membered ring system. 2-(Propynyl)phenol 3m was also tolerated as well under the standard reaction conditions and gave the corresponding chiral spiro product 5d in 31% yield with 37% ee. Alkyl benzyl alcohol 3o reacted quite well with 2h to form the product 5f by 1,4-addition without annulation in 8 membered ring system likely due to the ring strain. However, a variety of ortho-aminobenzyl alcohols could not afford the desired product probably due to the low reactivity of aza-o-QM. It is worth mentioning that the reaction between 3n and 2h was messy toward 7 membered ring system and poliketide oligomer 5e was isolated in only 14% yield with 0% ee after purification. The above results were added in the manuscript.

Q4. The mechanism needs more details. By the way, other catalyst systems should be examined as well to show the superiority of your system. Chiral phosphoric acid, chiral Au(I) system, other ligands like Box and so on are required in your work.
A4. In order to investigate the role of the Au-complex in this enantioselective cascade reaction, we investigated that the hetero-Diels-Alder reaction between o-QMs precursor 2b and 3-methyl-1H-isochromene 10 with or without the participation of Au-complex (Scheme 6, entry 2 and 3). Two catalytic reactions afforded the tetracyclic isochroman poliketide oligomer 4b in similar yield, d.r. and ee values comparing to the cascade reaction between 2b and 3a (Scheme 6, entry 1), which suggests that 10 may be the real intermediate in the [4+2] cycloaddition and Au-complex may not participate in the control of the enantioselectivity. Based on the reported X-ray structure of the N,N'-dioxide Sc III complex and the absolute configuration of the product 4 as well as the result of control experiment, a catalytic model was proposed for the HDA reaction to explain the origin of enantio-and diastereoselectivity of the process (Figure 1). Moreover, chiral phosphoric acid, chiral Au(I) system, other ligands like Box and so on were examined in this reaction. Chiral phosphoric acids and Sc(III)-Box complexes afforded the racemic product in moderate yield. No product was obtained in chiral Au(I) system. However, chiral acetonitrile-Au(I) system afforded the products in little lower ee values (see SI, table S2). Front. 2014, 1, 298；Chin. J. Chem. 2021, 39, 969 and Aldrichimica ACTA 2020, 53, 3 should be cited.

Q1. a) The authors carried out the model reactions using 2-(Hydroxylmethyl) phenol (2a) and α-propargyl benzyl alcohol (3a). Scheme 2 and 3 show the reaction scope by changing the substituted groups of two reactants. Actually, these adducts have very similar and specific structures. During the mechanism studies, substrates 7r and 7s (Scheme 4c) were tested. Interestingly, 8a was obtained from 7s. This reviewer suggests the authors think about the practicability of this reaction in current version. More information should be provided to guide the potential utilities by other groups. a) how about the substrates bearing alkyl groups on the benzylic position, b) how about the substrates bearing internal alkynes; c) instead of
α-propargyl benzyl alcohols, how about directly using hexynol or related alkynol without benzene ring. The author should provide both the advantages and the limitations of this reaction.
A1. Thanks a lot! According to your suggestion, we synthesized ortho-hydroxybenzyl alcohols bearing alkyl groups on the benzylic position 2p-q, internal alkyne 3p-q, hexynol as well as pentynol derivative and investigated the catalytic activity under the optimized reaction conditions. a) The substrates bearing alkyl groups on the benzylic position 2p-q were demonstrated to be acceptors amenable to the reaction protocol, giving rise to the corresponding products 4p-q in 45-82% yield with enantioselectivities up to 75% ee (see scheme 2). b) Internal alkynes 3p-q were also demonstrated to be reactants amenable to the reaction protocol, giving rise to the corresponding products 5g-h in 39-41% yield with 53-55% ee (see scheme 3). c) Hexynols 3s-t are unavailable substrates in this catalytic system. However, pentynol derivative 3r reacted quite well with 2h to form the chiral spiro product 5i in 52 yield with 66% ee. (see scheme 3). The above results were added in the manuscript.

Q2. A stepwise reaction pathway was proposed based on the studies of carbon isotope effects for the [4 + 2] reaction between 2b and 3a. In principle, two diastereomers should be produced during the second bond forming step. This review suggests a detailed study should be performed by direct using 3-methyl-1H-isochromene (7, Scheme 5) as a reactant. This will give insight about the reaction pathway.
A2. According to your suggestion, we investigated that the hetero-Diels-Alder reactions between o-QMs precursor 2b and 3-methyl-1H-isochromene 10 with or without the participation of Au-complex (scheme 6, entry 2 and 3). Two catalytic reactions afforded the tetracyclic isochroman poliketide oligomer 4b in similar yield, d.r. and ee values comparing to the cascade reaction between 2b and 3a (scheme 6, entry 2 and 3 vs entry 1).

Q3. In figure 1, proposed stereochemical models were given. A crucial question is what's role of the Au-complex. Is there any interaction between Au with the ligand? Is that necessary to in-situ form the Au-complex? No information was provided.
A3. Thank you for your suggestion. The result of control experiment in Scheme 6 suggests that Au-complex may not participate in the control of the enantioselectivity and there may be no interaction between the Au-complex and N,N'-dioxide ligand. Based on the reported X-ray structure of the N,N'-dioxide Sc III complex and the absolute configuration of the product 4 as well as the result of control experiment, a catalytic model was proposed for the HDA reaction to explain the origin of enantio-and diastereoselectivity of the process (Figure 1). The Re face of the o-QMs is shielded by the neighboring t-Bu group of the ligand and the enol attack takes place from the Si face of the o-QMs to form the first chiral center in benzylic position of o-QM skeleton ( Figure 1a VS 1b). Subsequent oxygen anion attack to oxonium ion afford the desired optical tetracyclic isochroman from Si face, while the Re face of the oxonium ion is shielded by the N,N'-dioxide (Figure 1c VS 1d). On the other hands, the Au-complex in-situ formed could not transform α-propargyl benzyl alcohol to 3-methyl-1H-isochromene and no reaction occurred (see SI, table S2, entry 9 and 10). Only (acetonitrile)Au-complex could catalyze the reaction. numbering is messy in the manuscript. For examples, 2a, 7, 8 were used twice in different Schemes. The authors should recheck the manuscript carefully and remove these careless errors before submission.