Multidirectional desymmetrization of pluripotent building block en route to diastereoselective synthesis of complex nature-inspired scaffolds

Octahydroindolo[2,3-a]quinolizine ring system forms the basic framework comprised of more than 2000 distinct family members of natural products. Despite the potential applications of this privileged substructure in drug discovery, efficient, atom-economic and modular strategies for its assembly, is underdeveloped. Here we show a one-step build/couple/pair strategy that uniquely allows access to diverse octahydroindolo[2,3-a]quinolizine scaffolds with more than three contiguous chiral centers and broad distribution of molecular shapes via desymmetrization of the oxidative-dearomatization products of phenols. The cascade demonstrates excellent diastereoselectivity, and the enantioselectivity exceeded 99% when amino acids are used as chiral reagents. Furthermore, two diastereoselective reactions for the synthesis of oxocanes and piperazinones, is reported. Phenotypic screening of the octahydroindolo[2,3-a]quinolizine library identifies small molecule probes that selectively suppress mitochondrial membrane potential, ATP contents and elevate the ROS contents in hepatoma cells (Hepa1–6) without altering the immunological activation or reprogramming of T- and B-cells, a promising approach to cancer therapy.

The manuscripts reports the structures of 12 crystal structures to support the reported chemistry. Unfortunately all structures use generic computer generated atom labels which makes reviewing more difficult, however in general the structures establish atom connectivirty and relative configuration of stereocentres (these arev light atom structures and the radiation source is Molydenum which means that differentiating between enantiomers is difficult). There are a few structures for which the data completeness is well below the IUCR requirements of 98.5% (some as low as 76% complete), while this does not mean the structures are wrong it does suggest a bit of sloppiness on behalf of the crystallographer. Structure with CCDC code 1857718 (Compound 12) has been reported in the triclinic space group P-1, with two molecules in the asymmetric unit, this is incorrect, and should be in the space group P21/c with one molecule in the asymmetric unit. I have attached a shelx .res file with the transformation matrix to convert the data to this space group. This structure must be refined in this space group and the resulting cif uploaded to replace the existing one.
The cif provided for compound 24a (1857755) has the opposite configuration to that indicated in Figure 5b, and is also opposite to the ortep inset given for 24a in the paper.
For compound 8a, the ortep diagram given in the paper, and the cif (1857606) provided are of the opposite configuration to that given in the chemical structure diagram for 8a These issues must be attended to and clarified, particularly for 8a and 24a for which the structures provided actually correspond to 8b and 24b respectively. Structure (1857869) has a long aromatic C-H bond of 1.37 Å, it appears that all the H atoms have been allowed to refine freely which is OK usually, but the data for this crystal is low quality and as a result one of the H atoms has drifted. All H atoms attached to carbon should be fixed into idealised positions, the resulting cif should then be uploaded to CCDC to replace the current cif which has CCDC code 1857869.
Reviewer #2 (Remarks to the Author): The manuscript by Al-Tel and coworkers demonstrates the power of synthetic planning to produce a versatile template that can be transformed into structurally diverse and complex products that are comparable to certain nitrogen-containing natural products.
Overall this is a nice illustration of DOS and the products generated are of a highly structurally complex nature-perhaps of a structural complexity unseen by previous efforts. The diastereoselectivities of the bond forming reactions are high. Furthermore, the structural assignments are well supported by X-ray crystallographic studies. That said, it is this reviewer's view that the underlying concept of this work is not novel. Schreiber's review on the build/couple/pair approach (B/C/P)to DOS is now over a decade old. While this work revolves around some elegant synthetic transformations it does not conceptually go into a new direction from the B/C/P approach. Moreover, several of the transformations were unpredicted making the work less general. Therefore, it is not believed that this work will find broad appeal to the chemistry or chemical biological community.
The manuscript is well written and but in light of the above discussion, this is potentially better served as a full article (perhaps in JACS). Given the narrow scope and lower novelty of work this reviewer does not feel that the work will be considered of interest to the larger synthetic chemistry or chemical biology community. As such it is my recommendation that this work not be accepted for publication in Nature Communications given that it is more suited to a specialized journal.
Reviewer #3 (Remarks to the Author): The step-economical synthesis of complex and sp3-rich natural product-inspired libraries is a challenge to the synthetic organic chemists. Moreover, one step Diversity Oriented Synthesis (DOS) from privileged scaffolds or an intermediate to diverse molecular structures is an art and an important milestone for drug design and discovery. Symmetrical cycloheaxdienones, synthesized via oxidative-dearomatization of phenols, is such as privileged structure employed previously for polycyclic compounds using transition metals and organocatalysis.
In the present manuscript by Al-Tel et al, entitled "Multidirectional Desymmetrization of Pluripotent Building Block en Route to the Asymmetric/Diastereoselective Synthesis of Complex Nature-Inspired Scaffolds" described one-step access to diversely functionalized octahydroindolo[2,3-a]quinolizine scaffolds in high diasteroselctivity via desymmetrization of symmetrical oxocyclohexa-2,5dienylpropanal (2) with tryptamine following the cascade process of Pictet-Spengler and aza-Michael cyclizations. Chiral amines, tryptophane was also employed for the synthesis of their enentioenriched variants with complete transfer of chirality (99% ee). Unsymmetrical cyclohexadienones (9 and 11) were employed to achieve bridged analogues with excellent diasteroselectivity. These complex penta/hexacyclic compounds were achieved in low to moderate chemical yields on average, however, in very good diastereoselectivity. The several key (viz. OsO4 and NBS mediated) organic transformation were performed on octahydroindolo[2,3-a]quinolizine scaffolds to archive rearranged products, such as novel amino-acetals and spiro-polycyclic compounds, respectively. In addition, these privileged cyclohexadienone-aldehydes were serendipitously enabled to access the synthesis of oxocanes and piperazinones with the variantion of amine coupling partners, probably due to remarkable reactivity difference of aromatic and aliphatic amines.
Overall, the developed strategy demonstrated the synthesis of diverse range of scaffolds (37 products), starting from a privileged cyclohaxadienone-aldehydes (4 building blocks) and amines. The reaction conditions are simple and diverse molecular scaffolds were archived by varying the structure of amines. Under acidic condition, amino acids gave respective octahydroindolo[2,3a]quinolizine scaffolds, whereas in the absence of acid, piperazinones were formed. In contrast, aromatic amines led to oxocanes. Although, chemical yields are not much appealing in most of examples, excellent diastereoselectivity, diversification of scaffolds, and simple reaction conditions are quite impressive. Finally, phenotypic screening of the was done with octahydroindolo[2,3a]quinolizine library and authors have claimed that compounds 5c and 26c that effectively suppressed glycolytic production of ATP and membrane potential in hepatoma cell line. I didn't see any structure corresponding to 26c ??
The chemistry described here is interesting and quite appealing to the synthetic as well as medicinal chemists for design and synthesis various sp3 rich molecular scaffolds. Thus, in my opinion, the present manuscript can be accepted after following addition and corrections. 1) Abstract, line 12: "amino acids" were used as chiral reagent, not auxiliaries, needed to be changed.
2) Scheme 3a; Yields for 2a and 2b should be clearly mentioned.
3) Scheme 3c; Isn't the aza-Michael reactions of intermediate, derived from 9 and 11 shows "regioselectivity", not "chemoselectivity" ?? Mechanism part for the explanation of switching of regioselectivity with cyclopenatne and cyclohexane analogues is not so clear. The intermediate XII shows some steric congestion b/w 'H' and cyclopentane-bridged methylene groups?? 4) Page 13; It is mentioned NBS-catalysed, however it is not. It should be changed to NBS-mediated.
5) It is mentioned that ". This serendipitous cascade has turned out to be a general approach to access these complex structures with complete enantioselectivity…..". However, it would be more appropriate to mention "….with complete transfer of enantioselectivity". 6) From the library of octahydroindolo[2,3-a]quinolizine, 5c and 26c are most effective to suppress glycolytic production of ATP and membrane potential in hepatoma cell line, however structure 26c is not shown ?? Special attention is needed to denote the number for hit molecule. The presentation of tabular data of activity of other analogues will be more informatory for medicinal chemists, it should be mentioned. 7) Synthetic procedures and spectral data should be presented for the starting materials (2a, 2b, 9, and 11) in Supplementary Information.

Responses to Reviewer 1 comments:
Comment: Structure with CCDC code 1857718 (Compound 12) has been reported in the triclinic space group P-1, with two molecules in the asymmetric unit, this is incorrect, and should be in the space group P21/c with one molecule in the asymmetric unit. I have attached a shelx .res file with the transformation matrix to convert the data to this space group. This structure must be refined in this space group and the resulting cif uploaded to replace the existing one.
Response: Thank you very much for the positive and supportive comments. We have refined the structure as requested to be in the space group P21/c with one molecule in the asymmetric unit. Furthermore, the original cif file of compound 12 that was deposited in the CCDC is now replaced by the refined cif file for the CCDC 1857718. During the refining process some parameters values were changed and these new values were reported in the corresponding table of compound 12 in the SI (Supplementary Table 6, Page S17-S18). Also the new ORTEP is now included in Figure 3c (Revised manuscript, Page 11).

Comment:
The cif provided for compound 24a (1857755) has the opposite configuration to that indicated in Figure 5b, and is also opposite to the ortep inset given for 24a in the paper.
For compound 8a, the ortep diagram given in the paper, and the cif (1857606) provided are of the opposite configuration to that given in the chemical structure diagram for 8a. These issues must be attended to and clarified, particularly for 8a and 24a for which the structures provided actually correspond to 8b and 24b respectively.
Response: Thank you very much for the helpful comments. After revising the original X-ray data in the Xray machine, we found that the data we reported for compounds 8a and 24a, in the original submission were right however, they belong to compounds 8b and 24b. We mistakenly assigned them to 8a and 24a. Please accept our apology for such a mistake and we would like to thank the reviewer for this important note. Now both typos were corrected in the text (page 12, paragraph 2; page 21, paragraph 2). The ORTEP drawings were now assigned to compounds 8b (Fig. 3b) and 24b (Fig. 5c).
Comment: Structure (1857869) has a long aromatic C-H bond of 1.37 Å, it appears that all the H atoms have been allowed to refine freely which is OK usually, but the data for this crystal is low quality and as a result one of the H atoms has drifted. All H atoms attached to carbon should be fixed into idealised positions, the resulting cif should then be uploaded to CCDC to replace the current cif which has CCDC code 1857869.
Response: Thank you very much for this supportive comment. As suggested, the X-ray structure of compound 19 was refined and the drifted aromatic C-H bond was refined. The refined cif file was uploaded in the CCDC and the old cif file was replaced and the new refined file number is CCDC 1857869. During the refinement process, some parameter values were changed and the new values were placed in the Supplementary Table 10 for compound 19. Furthermore, the new ORTEP is placed in Fig. 4d.

Responses to Reviewer 2 comments:
Comment: The manuscript by Al-Tel and coworkers demonstrates the power of synthetic planning to produce a versatile template that can be transformed into structurally diverse and complex products that are comparable to certain nitrogen-containing natural products.
Overall this is a nice illustration of DOS and the products generated are of a highly structurally complex nature-perhaps of a structural complexity unseen by previous efforts. The diastereoselectivities of the bond forming reactions are high. Furthermore, the structural assignments are well supported by X-ray crystallographic studies. That said, it is this reviewer's view that the underlying concept of this work is not novel. Schreiber's review on the build/couple/pair approach (B/C/P)to DOS is now over a decade old. While this work revolves around some elegant synthetic transformations it does not conceptually go into a new direction from the B/C/P approach. Moreover, several of the transformations were unpredicted making the work less general. Therefore, it is not believed that this work will find broad appeal to the chemistry or chemical biological community.
The manuscript is well written and but in light of the above discussion, this is potentially better served as a full article (perhaps in JACS). Given the narrow scope and lower novelty of work this reviewer does not feel that the work will be considered of interest to the larger synthetic chemistry or chemical biology community. As such it is my recommendation that this work not be accepted for publication in Nature