Racemic synthesis of an intermediate for the formal synthesis of madindoline A and B

ABSTRACT On the basis of the application of the Darzens/ring-expansion process of cyclobutenedione developed previously by our group, a new strategy for the multisubstituted cyclopentene units of madindoline A and madindoline B has been reported in this paper. In light of the strategy, the synthesis of racemic Omura’s intermediate was finished in four steps and 34% overall yield, which furnished a new formal synthetic route to madindolines A and B. GRAPHICAL ABSTRACT


Introduction
Madindoline A (1) and B (2) are bioactive microbial metabolites isolated from Streptomyces nitrosporeus K93-0711 by Omura and coworkers. [1] Detailed biological studies showed they are potent and selective inhibitors of interleukin 6, which is critical for the progression of multiple types of cancer cachexia and tumor cells. However, 1 and 2 could not be obtained from Streptomyces nitrosporeus K93-0711 again due to the mutation of the bacterial strain. [1,2] Therefore, the chemical synthesis becomes the only source of 1 and 2.
Structurally, 1 and 2 possess a rare molecular frame, comprising a 3a-hydroxyfuroindoline and a multiple-substituted cyclopentene unit (Fig. 1). The difference between 1 and 2 is the opposite relative configurations at their C-2′ position. Because of their important bioactivity and the unique structures, 1 and 2 have attracted a number of attentions from the synthetic community. During all synthetic reports on 1 and 2, the asymmetric oxidation ring closure of tryptophol through Sharpless asymmetric epoxidation has been confirmed as an efficient and general approach to the chiral 3a-hydroxyfuroindoline motif.
The synthesis of the multiple-substituted cyclopentene unit is always the major challenge faced in the synthesis of 1 and 2. Therefore, the development of concise synthetic strategies for the multiple-substituted cyclopentene unit becomes the major concern of the synthetic studies of 1 and 2 (Scheme 1).
Omura and coworkers reported the asymmetric construction of the multisubstituted cyclopentene unit first based on the ring-closing olefin metathesis of 3 [1] and later based on an intramolecular acylation of allysilane 4. [3] Kobayashi's group finished the construction of the cyclopentene unit using an intramolecular aldol condensation of triketone 5. [4] Van Vranken and coworkers succeeded in the construction of the cyclopentene unit through the Moore ring contraction process of azidoquinone 6. [5] Tius's group reported an approach to the cyclopentene unit based on Nazarov cyclization of allene ether 7. [6] Tu's group developed a strategy to the cyclopentene unit using AlEt 3 -promoted tandem reductive rearrangement of α -hydroxy epoxides 8. [7] Mikami's group developed a gentle Cu-catalyzed enantioselective desymmetrization process in continuous alkylation of cyclopentene-1,3-dione as an asymmetric strategy to the cyclopentene unit. [8] Mukherjee and coworkers reported an enantioselective alkylative desymmetrization process of cyclopentene-1,3-dione catalyzed by a dihydroquinine-based bifunctional urea derivative as an asymmetric approach to the cyclopentene unit. [9] Although there were lots of strategies reported for the construction of the multiple-substituted cyclopentene unit of 1 and 2, the development of concise strategy to the multiple-substituted cyclopentene unit and the synthetic route of 1 and 2 with good efficiency still remains of interest. Herein, we report a concise synthetic strategy to the multiple-substituted cyclopentene compound 13 and the formal synthesis of 1 and 2 based on a Darzens/ring expansion of cyclobutenedione strategy developed previously during our synthetic studies on linderaspirone A and bi-linderone [10] and asterredione. [11] The multisubstituted cyclopentene compound 13 is an important intermediate in Omura's synthesis of 1 and 2. We anticipate that the synthesis of racemic 13 could be finished in only four steps through the application of the Darzens/ring expansion process of cyclobutenedione 16. Therefore, a new synthetic route to 1 and 2 could be developed (Scheme 2).

Results and discussion
According to our synthetic plan in Scheme 2, our work began with the synthesis of cyclobutenedione 16 [12] through a one-pot process. Dialkyl squarate 17 was submitted for two nucleophilic 1,2-additions stepwise with methyl lithium and n-butyl lithium. cyclobutenedione 16 was formed through the subsequent acid-promoted rearrangement of crude residue obtained from the nucleophilic additions. The nucleophilic addition process was Scheme 2. Our synthetic plan to 1 and 2.
investigated with different addition sequence of organolithium reagents and different dialkyl squarate 17 (17a, R 1 ¼ CH 3 O, and 17b, R 1 ¼ i-PrO, Scheme 3). The results revealed that the addition of CH 3 Li and n-BuLi in turn to diisopropyl squarate and the following acid-promoted rearrangement in dichloromethane provided the desired 16 at the best yield. Thus, we succeeded in the synthesis of cyclobutenedione 16 and the introduction of the methyl group and the butyl group of the final target in only one step. With cyclobutenedione 16 in hand, we next explored the construction of the multiplesubstituted cyclopentene-1,3-dione 14 through the Darzens/ring expansion process (Scheme 4). [10] In the presence of sodium bis(trimethylsilyl)amide (NaHMDS), the expected Darzens/ring expansion of 16 and 15 possessing different substitutents R 4 and R 5 happened smoothly in most cases, generating 14 in good yields through probably an epoxide intermediate 18. The expected 14b was obtained in 80% yield (based on the recovery of 23% of 16). Notably, no product was generated from the reaction of 16 and 15 possessing carboxylic ethyl ester as the R 5 substituent.
After the following reduction of the two carbonyls of 14b and the protection of the two generated hydroxyl groups, racemic Omura's intermediate 13 was readily formed in 53% yield. The spectroscopic properties of 13 are consistent with Omura's report. [1] As a result, a formal synthetic approach to madindolines A and B has been accomplished.

Conclusion
In summary, the construction of the multisubstituted cyclopentene unit of madindolines A and B was achieved through the Darzens / ring expansion process of cyclobutenedione developed previously by our group. In the light of this strategy, the synthesis of racemic Omura's intermediate 13 was finished in only four steps and with 34% overall yield, and a new formal synthetic route to madindoline A and madindoline B has been developed.

Experimental
All of the reactions were performed under an argon atmosphere. Commercial-grade solvents were distilled prior to use. Column chromatography was performed using 200-to 300-mesh silica gel. Thin-layer chromatography (TLC) was performed on silica-gel GF254 plates. High-resolution mass spectra (HRMS) were recorded in electrospray ionization (ESI) mode using a Q-TOF analyzer. Proton and carbon nuclear magnetic resonance spectra ( 1 H NMR, 13 C NMR) were recorded on the basis of the resonating frequencies as follows: 1 H NMR at 400 MHz and 13 C NMR at 100 MHz having the solvent resonance as internal standard ( 1 H NMR, CDCl 3 at 7.26 ppm; 13 C NMR, CDCl 3 at 77.16 ppm. Chemical shifts are reported in parts per million (ppm) with tetramethylsilane (TMS) at 0.00 ppm used as an internal standard and the residual solvent peak as an internal indicator.

Synthesis of compound 14b
A solution of NaHMDS (1.0 M in THF, 2.1 mL, 2.1 mmol) was added dropwise to the solution of 15 (R 4 and R 5 are methyl, 334 mg, 2.0 mmol) in THF (10 mL) at À 78 °C under argon. After 5 min, a solution of 16 (152 mg, 1.0 mmol) in THF (5 mL) was added dropwise. After another 5 min, the reaction was quenched with saturated ammonium chloride solution. The aqueous phase was extracted with EtOAc (3 � 20 mL). The combined organic layer was washed with saturated brine and dried over Na 2 SO 4 . The solvent was removed under reduced pressure. The crude residue was purified by silica-gel chromatography (PE/EtOAc, 20/1) to give 14b as a colorless oil (146 mg, 80%, based on the recovery of 23% of 16) R f ¼ 0.6 (PE/EtOAc ¼ 4/1). 1

Funding
We are grateful for financial support from the National Natural Science Foundation of China (NSF-21002078, 21372184), the State Education Ministry Scientific Research Foundation for the Returned Overseas Chinese Scholars, and Shaanxi Province Technology Foundation for Selected Overseas Chinese.