Enantioselective total syntheses of (+)-stemofoline and three congeners based on a biogenetic hypothesis

The powerful insecticidal and multi-drug-resistance-reversing activities displayed by the stemofoline group of alkaloids render them promising lead structures for further development as commercial agents in agriculture and medicine. However, concise, enantioselective total syntheses of stemofoline alkaloids remain a formidable challenge due to their structural complexity. We disclose herein the enantioselective total syntheses of four stemofoline alkaloids, including (+)-stemofoline, (+)-isostemofoline, (+)-stemoburkilline, and (+)-(11S,12R)-dihydrostemofoline, in just 19 steps. Our strategy relies on a biogenetic hypothesis, which postulates that stemoburkilline and dihydrostemofolines are biogenetic precursors of stemofoline and isostemofoline. Other highlights of our approach are the use of Horner–Wadsworth–Emmons reaction to connect the two segments of the molecule, an improved protocol allowing gram-scale access to the tetracyclic cage-type core, and a Cu-catalyzed direct and versatile nucleophilic alkylation reaction on an anti-Bredt iminium ion. The synthetic techniques that we developed could also be extended to the preparation of other Stemona alkaloids.

i) Stemofoline and its analogs are important natural products with insecticidal and anticancer activity and they are very challenging target molecules for total synthesis.
ii) Very few total syntheses of the closely related natural products were achieved. These prior total syntheses are elegant and remarkable, but there are also drawbacks in these syntheses. For example, the formation of the di-oxygenated tetrasubstituted double bond has been a long-standing challenge.
iii) The synthetic approach described by Huang and co-workers are very different from the prior arts, therefore justifies for novelty. iv) They have developed a good solution to construct the problematic di-oxygenated tetrasubstituted double bond. Their protocol involves an HWE reaction followed by a bromine oxidation and a TBAF-promoted elimination and substitution. The HWE reaction allowed them to access two dihydrostemofoline analogs (8 and 9) and stemoburkilline (7), which could be further converted to stemofoline and isostemofoline.
This result also provides evidence to support the proposed biogenetic hypothesis. This protocol will likely be adapted to make many other natural products with such a tetrasubstituted double bond in the stemona alkaloid family. v) A remarkable bridgehead bromide substitution via a sequence of elimination and addition is worth noting as well. This substitution allows the introduction of various alkyl group at the bridgehead carbon, which could be used to make other stemofoloine analogs with varied alkyl group at this position. vi) Additionally, the authors identified a nice conjugate addition to solve the C10 stereochemistry problem, which troubled the previous syntheses.
With the above strength said, there are also a few minor weaknesses.
i) Since the dihydrostemofoline analogs 8 and 9 and stemoburkilline (7) are inevitable intermediates on the way to stemofoline and isostemofoline is a minor product from the last step, the word "divergent" in the manuscript is not very accurate. 2 ii) Is the stereochemistry of 23-1 and 23-2 different at C9 or C10? Why both can be converted to 24? An explanation is needed here.
iii) Did the authors try any aerobic oxidation of 7, 8, or 9 to produce stemofoline?
iv) The writing of this manuscript needs to be polished.
Reviewer #2 (Remarks to the Author): The ms describes the asymmetric syntheses of four stemofoline-type alkaloids and a related diastereomer. The ms describes a substantial body of work aimed at synthesizing the tetracyclic core unit of the stemafoline ring system and then developing methodology to construct and install the butenolide onto the tetracyclic core. The linkage between the teracyclic core and the butenolide moiety incorporates an ene diether motif.
Analytical and spectral data are consistent with structures of compounds. and alkaloids 1 and 2 is "oxidative cyclization". Oxidative cyclization is applicable to 7 --> 1 and 2; however, for 8/9 ----> 1 and 2, the connection is "oxidation" (dehydrogenation).  (Table 1a) most likely is higher energy demanding. Do the authors think that the reaction might  Table 2 can be summarized in the text and Table 2 can be moved to SI. Pg S10: Compound 23a, but on page S68/69 it is numbered 23b.

Reviewer #1
In this manuscript, Huang and co-workers descried their total syntheses of stemofoline and four related analogs including its plausible biogenetic precursors. This work is important and recommended for publication for the following reasons.
i) Stemofoline and its analogs are important natural products with insecticidal and anticancer activity and they are very challenging target molecules for total synthesis.
ii) Very few total syntheses of the closely related natural products were achieved. These prior total syntheses are elegant and remarkable, but there are also drawbacks in these syntheses. For example, the formation of the di-oxygenated tetrasubstituted double bond has been a long-standing challenge.
iii) The synthetic approach described by Huang and co-workers are very different from 5 the prior arts, therefore justifies for novelty. iv) They have developed a good solution to construct the problematic di-oxygenated tetrasubstituted double bond. Their protocol involves an HWE reaction followed by a bromine oxidation and a TBAF-promoted elimination and substitution. The HWE reaction allowed them to access two dihydrostemofoline analogs (8 and 9) and stemoburkilline (7), which could be further converted to stemofoline and isostemofoline.
This result also provides evidence to support the proposed biogenetic hypothesis. This protocol will likely be adapted to make many other natural products with such a tetrasubstituted double bond in the stemona alkaloid family. v) A remarkable bridgehead bromide substitution via a sequence of elimination and addition is worth noting as well. This substitution allows the introduction of various alkyl group at the bridgehead carbon, which could be used to make other stemofoloine analogs with varied alkyl group at this position.
vi) Additionally, the authors identified a nice conjugate addition to solve the C10 stereochemistry problem, which troubled the previous syntheses.
Our response: We are grateful for the visions, comments and questions, which are very helpful for us to improve this article.
With the above strength said, there are also a few minor weaknesses.
1. Since the dihydrostemofoline analogs 8 and 9 and stemoburkilline (7) are inevitable intermediates on the way to stemofoline and isostemofoline is a minor product from the last step, the word "divergent" in the manuscript is not very accurate.
Our response: Indeed, the original meaning of "divergent" synthesis involved the access to two or more products in equal amount. We intend to introduce this term to natural product chemistry/ photochemistry to account for the chemical diversity of natural product, those natural products are formed in non-equal amount. We believe that although living systems can produce each bio-molecules in an enzymatically controlled accurate manner (highly selective), the chemo-diversity of natural products might imply another strategy, namely, some steps of a synthesis deviate of enzymatic control. We shall elaborate this concept in a near future in form of a review/ perspective. Regarding isostemofoline, which is indeed a minor component among stemofoline alkaloid, has been the target of Kende's total synthesis (ref. 48, J. Am. Chem. Soc. 1999, 7431). We are confident that the use of the word "divergent" is not accurate chemically, but bio/ phyto-chemically accurate.  Table   S1). For the oxidation of 8 and 9, several oxidation conditions have been tried including DBU/O 2 (cf. Pg 14 Ln 274 and the Supplementary Information, Table S2).

The writing of this manuscript needs to be polished.
Our response: Thanks for the kind suggestion. The manuscript has been sent for English polishing, but we didn't accept all corrections. The corrected manuscript has been further polished.

Reviewer #2
The ms describes the asymmetric syntheses of four stemofoline-type alkaloids and a 7 related diastereomer. The ms describes a substantial body of work aimed at synthesizing the tetracyclic core unit of the stemafoline ring system and then developing methodology to construct and install the butenolide onto the tetracyclic core. The linkage between the teracyclic core and the butenolide moiety incorporates an ene diether motif.
Analytical and spectral data are consistent with structures of compounds.
Our response: We are grateful for the visions, comments and questions, which are very helpful for us to improve this article.
Comments, questions, and clarifications which the authors might like to address are provided below.
1. Pg1, Abstract; ln 11-15: Sentence starts off in plural form and then changes to singular form.
Our response: Thanks for indicating on this issue. The last sentence changes to plural form: "However, concise, enantioselective total syntheses of stemofoline alkaloids remain a formidable challenge due to their structural complexity." 2. Ln 17: "biogenic" or biogenetic hypothesis?
Our response: Indeed, Yes, in the synthesis of tropinone 16, the cis-17: trans-17 ratio was lower than the previous route, but this approach to cis-17 is shortened by three steps as compared with previous one. According to our observation, trans-17 is the thermodynamically stable diastereomer, which could not be converted into cis-17, the trans-diastereomer not used.  (Table 1a) most likely is higher energy demanding. Do the authors think that the reaction might proceed via an organocopper-type intermediate -e.g., Tropinone-16-Cu(L)-Grignard instead of iminium ion A?
Our response: Thanks for the kind suggestions. Now, Table 1a, b are separated into   Table 1 and Fig. 4.
Thanks for helping analyzing the differences between Kibayashi's ring system and ours.
The generation of an organocopper-type intermediate is an interesting idea that is possible for the specific tropinone ring system as ours involving the formation of a homo-enolate. However, if it occurred, it should not be the main pathway because the reagents used herein for the reactions were Grignard reagents (nucleophiles) instead of electrophiles.
Does the 1.1:1 diastereomeric mixture indicate that the product is epimeric only at the carbinol center, and the "-carbon stereocentre is confirmed on account of the axial 10 "-C-C(OH) bond?
Our response: Thanks for indicating on these issues. First, "adduct 23" changes to "adducts 23-1 and 23-2", and the structure of 23a provided. Second, in the Supplementary Information section, "23b" corrected to "23a". Third, we were unable to distinguish between 23-1 and 23-2 by their NMR spectra. From the experimental results, we deduced that the structures of two diastereomers 23-1 and 23-2 have the stereochemistries shown the scheme. Subjecting of both diastereomers to derivatization with CDI and subsequent base-promoted anti-elimination then generated E-enone ester 24. The following sentences and structural details are added to the main text and Figure   4. Our response: 91.2% is the overall yield for two steps. "91.2%" changes to "91.2% for two steps". Thanks for indicating the problem. 13. Line 223: "The two transformations....83% yield." Not exactly one step. The experimental procedure in SI suggests that the two steps were executed sequentially and involved a work-up step to obtain crude 27 which was used without purification in the lactonization step.
Our response: This involved one operation and two pots. We change it to two steps. In the text, "in a single step" (pg11 ln223) and "in a one-step manner" (pg 12 ln 233) change to "Importantly, neither the hydrogenation or lactonization step required purification to yield 12 of sufficient purity for further reactions." And the total steps