Application of a rhodium-catalyzed cyclization cycloaddition cascade strategy to the total synthesis of (−)-curcumol

X. X. Zhang*a, R. Y. Y. Kob, X. Q. Xiea, W. P. Qia, P. C. Lia and P. Chiu*b
aCollege of Pharmaceutical Sciences, Zhejiang University of Technology, Hangzhou, Zhejiang 310032, P. R. China. E-mail: zhangxx@zjut.edu.cn
bDepartment of Chemistry and State Key Laboratory of Synthetic Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong, P. R. China. E-mail: pchiu@hku.hk

Received 23rd December 2017 , Accepted 23rd January 2018

First published on 23rd January 2018


A de novo asymmetric total synthesis of (−)-curcumol has been realized, based on a rhodium-catalyzed cyclization–cycloaddition cascade reaction as the key step to construct the tricyclic nucleus.


Introduction

The genus Curcuma is a native plant of Southeast Asia, many species of which have been used for medicinal purposes and have been subjects of phytochemical studies over the years.1 Curcuma longa, or turmeric, is widely distributed in India, Pakistan and Bangladesh, and is cultivated for use as spice, dye and medicine.2 Curcuma zedoaria, or white turmeric, is a related species found in the India–Pakistan–Bangladesh subcontinent as well as China.3 Curcuma wenyujin is a species found in China, whose rhizomes “ezhu” and root tubers “yujin” have been used in traditional Chinese medicine for the clinical treatment of cervical cancer.4

Among the guaiane-type sesquiterpenoids found in these species, isocurcumenol is the most widely distributed,5 along with other hemiacetals (−)-curcumol,6 curcumenol,7 and 4-epi-curcumenol8 (Fig. 1). (−)-Curcumol (1) bearing a 7α-isopropyl side chain was one of the earliest guaianes isolated, characterized and identified in phytochemical studies from the essential oil of Curcuma zedoaria and Curcuma wenyujin.6 The bioactivity studies of (−)-curcumol revealed its varied anti-cancer effects, including the suppression of breast cancer cell metastasis.9 More recently, it has been found to inhibit IGF-1R and activate p38 MAPKs, and showed anti-colorectal cancer activity in xenograft models of nude mice.10 Curcumol has also been found to modulate GABAA receptors.11 The absolute structure of 1 was determined by CD and X-ray crystallography as shown.12


image file: c7qo01150d-f1.tif
Fig. 1 Natural products from Curcuma.

Herein we report the first de novo asymmetric total synthesis of (−)-curcumol,13 featuring an approach that may be amenable for obtaining curcumol analogues for bioactivity and SAR studies.

Our retrosynthetic strategy to the target molecule is shown in Scheme 1. (−)-Curcumol could be obtained from a reductive ring-opening of the oxygen bridge of oxatricyclic ketone 2, in which cyclic hemiacetal formation should be favourable. Ketone 2 could be obtained from 3 through the introduction of the isopropyl group at the C-5 position. This oxatricyclic ketone 3 was the key intermediate that can be constructed by the carbene cyclization cycloaddition cascade reaction14 of diazoketone 4, a reaction which we have exploited previously for the synthesis of other natural products.15


image file: c7qo01150d-s1.tif
Scheme 1 Retrosynthetic analysis of (−)-1.

This synthetic plan hinges on the regioselectivity of the cycloaddition with a tethered allene as the dipolarophile, and we have examined this reactivity through model studies of diazoketones 5 and 6 (Scheme 2).16 The treatment of 5 with the rhodium catalyst resulted in carbonyl ylide formation from the cyclization of the carbonyl group with the electrophilic rhodium carbene. Although both the double bonds of the allenyl unit could potentially react as the dipolarophile for the carbonyl ylide, intramolecular (3 + 2) cycloaddition proceeded with the internal double bond of the allene to give the desired perhydroazulene 7 due to the more efficient and rapid concomitant formation of the intervening five-membered ring. For diazoketone 6, cycloaddition engaged the terminal double bond instead to yield perhydroazulene 8, probably to avoid the formation of the strained bicyclo[5.2.0]nonane.


image file: c7qo01150d-s2.tif
Scheme 2 Model studies (ref. 16).

Results and discussion

Based on these results that the carbene-cyclization cycloaddition reaction yielded the desired frameworks, we proceeded to apply them to the asymmetric synthesis of (−)-curcumol (1). The diastereoselectivity of the cycloaddition of optically-enriched diazoketone 4 with respect to the C3 stereochemistry is a key issue to be investigated in this route.

The optically pure cycloaddition precursor 4 was assembled as shown in Scheme 3 via an Evan's auxiliary strategy.17 The synthesis commenced from the known acid 9,18 which was activated to couple with (S)-4-benzyl-2-oxazolidone. Acylated oxazolidinone 10 underwent methylation to afford 11 as a single diastereomer in 80% yield. The hydrolysis of 11, followed by conversion to Weinreb amide 13, proceeded in good yield. The homologation of 13 with Normant's Grignard reagent (14)19 to produce 15 was nearly quantitative. A one-pot oxidation catalyzed by TEMPO in the presence of sodium hypochlorite and sodium chlorite afforded cleanly acid 16. Finally, chiral diazoketone 4 was synthesized by the activation of 16 and subsequent treatment with diazomethane.


image file: c7qo01150d-s3.tif
Scheme 3 Synthesis of (+)-4. Reactions and conditions: a. (i) i-Butyl chloroformate, Et3N, THF; (ii) n-BuLi/(S)-4-benzyl-2-oxazolidone, THF, −78 °C, 75% yield; b. NaHMDS, MeI, THF, −78 °C, 80% yield; c. LiOH, H2O2, 0 °C, 81% yield; d. MeONHMe-HCl, Et3N, EDCI, HOBt, 95% yield; e. ClMgO(CH2)3MgCl (14), THF, 99% yield; f. (i) cat. TEMPO, NaOCl, NaClO2, NaH2PO4, MeCN, H2O, 90% yield; g. (i) i-butyl chloroformate, Et3N, THF; (ii) CH2N2, 65% yield.

The treatment of diazoketone 4 with dirhodium(II) acetate induced carbene formation and intramolecular cyclization–cycloaddition cascade reaction to give cycloadducts in 92% yield, but as inseparable diastereomers 3 and 17 in a 1[thin space (1/6-em)]:[thin space (1/6-em)]2 ratio, as observed by 1H NMR spectroscopy. On the other hand, their corresponding ketal derivatives 18a and 18b could be easily separated by flash chromatography on silica gel (Scheme 4). Separate deprotection of 18a and 18b under acidic conditions returned the optically pure cycloadducts 3 and 17, respectively, for structure determination.


image file: c7qo01150d-s4.tif
Scheme 4 Carbene cyclization cycloaddition reaction of 4.

The elucidation of the structures of 3 and 17 was based on their 2D NOESY spectra. The spectrum of diastereomer 17 showed an NOE cross-peak between the methyl group and the bridgehead proton, indicating that they were on the same face of the cyclopentane, whereas the spectrum of diastereomer 3 showed no enhancement. Hence, the minor oxatricyclic product 3 had the correct relative stereochemistry for the synthesis of (−)-curcumol.

We further studied this reaction in an attempt to reverse this substrate-based diastereoselectivity in favour of the desired 3, by varying the reaction conditions and by screening different rhodium catalysts (Table 1), including chiral catalysts (Fig. 2). Unfortunately, the preference for 3 was only slightly improved. The diastereoselectivity was affected to some degree by the reaction solvents (Table 1, entries 2–5) and the application of different catalysts, but 17 remained as the major cycloadduct. In fact, using Rh2(S-TCPTTL)4 as the catalyst, the preference for 17 was as high as 50[thin space (1/6-em)]:[thin space (1/6-em)]1 (Table 1, entry 10), a selectivity which could be exploited for the synthesis of 4-epi-curcumenol (Fig. 1). Ultimately, the highest yield for diastereomer 3 was obtained by using dirhodium octanoate as the catalyst in CH2Cl2 at low temperature (Table 1, entry 5).


image file: c7qo01150d-f2.tif
Fig. 2 Chiral rhodium catalysts screened.
Table 1 Optimization of the reaction of (+)-4

image file: c7qo01150d-u1.tif

Entry Conditions Yield of 3 + 17 3[thin space (1/6-em)]:[thin space (1/6-em)]17
1 Cat. Rh2(OAc)4, PhCF3, −27 °C 70% 1[thin space (1/6-em)]:[thin space (1/6-em)]2.0
2 Cat. Rh2(oct)4, PhCF3, −27 °C 82% 1[thin space (1/6-em)]:[thin space (1/6-em)]1.3
3 Cat. Rh2(oct)4, PhF, −27 °C 70% 1[thin space (1/6-em)]:[thin space (1/6-em)]2.3
4 Cat. Rh2(oct)4, PhH, 20 °C 72% 1[thin space (1/6-em)]:[thin space (1/6-em)]2.0
5 Cat. Rh2(oct)4, CH2Cl2, −27 °C 86% 1[thin space (1/6-em)]:[thin space (1/6-em)]1.2
6 Cat. Rh2(S-DOSP)4, CH2Cl2, 0 °C 75% 1[thin space (1/6-em)]:[thin space (1/6-em)]2.8
7 Cat. Rh2(R-DOSP)4, CH2Cl2, 0 °C 87% 1[thin space (1/6-em)]:[thin space (1/6-em)]2.8
8 Cat. Rh2(S-PTTL)4, PhCF3, −27 °C 92% 1[thin space (1/6-em)]:[thin space (1/6-em)]2.0
9 Cat. Rh2(R-PTTL)4, PhCF3, −27 °C 94% 1[thin space (1/6-em)]:[thin space (1/6-em)]2.0
10 Cat. Rh2(S-TCPTTL)4, PhCF3, −27 °C 90% 1[thin space (1/6-em)]:[thin space (1/6-em)]50
11 Cat. Rh2(R-TCPTTL)4, PhCF3, −27 °C 77% 1[thin space (1/6-em)]:[thin space (1/6-em)]6.5
12 Cat. Rh2(S-TFPTTL)4, PhCF3, −27 °C 78% 1[thin space (1/6-em)]:[thin space (1/6-em)]3.0
13 Cat. Rh2(R-TFPTTL)4, PhCF3, −27 °C 70% 1[thin space (1/6-em)]:[thin space (1/6-em)]7.5


To proceed with the synthesis, alkylation at C-5 needed to be accomplished to afford 2. Not surprisingly, the direct alkylation of the enolate of 3 with isopropyl iodide was unsuccessful. We then employed an indirect alkylation strategy via an aldol reaction (Scheme 5). The conversion of 3 to its enol ether 19, and then subjecting it to a Mukaiyama aldol reaction by treatment with SnCl4 and acetaldehyde provided as products hydroxyketone 20 and 21, which incorporated an extra molecule of acetaldehyde. Treatment with an acid converted a mixture of 20 and 21 to enone 22 in high yield, and exclusively as the (E)-diastereomer. Conjugate addition of methyl cuprate to 22 provided the desired isopropylated ketone, but as a 30[thin space (1/6-em)]:[thin space (1/6-em)]70 mixture of C-5-epimers 2 and 2′. An attempt to improve the selectivity for 2 via a deprotonation of the mixture by KHMDS and reprotonation on the less hindered face, using TFA as the proton source at low temperature, slightly increased the ratio of 2[thin space (1/6-em)]:[thin space (1/6-em)]2′ to 50[thin space (1/6-em)]:[thin space (1/6-em)]50.


image file: c7qo01150d-s5.tif
Scheme 5 Synthesis of 1 and 23. Reactions and conditions: a. TESOTf, 2,6-lutidine, CH2Cl2, 0 °C, 99% yield; CH3CHO, SnCl4, −78 °C, 63% yield of 20 + 21; c. p-TsOH-H2O, PhMe, rt, 94% yield; d. Me2CuLi, Et2O, −90 °C, 82% yield; e. KHMDS, then TFA, −90 °C, then SmI2, THF, 81% yield over 2 steps, dr 23[thin space (1/6-em)]:[thin space (1/6-em)]1 = 1[thin space (1/6-em)]:[thin space (1/6-em)]1.

Unfortunately, the diastereomers were not separable, and so a 50[thin space (1/6-em)]:[thin space (1/6-em)]50 mixture of 2 + 2′ was carried forward to reduction with samarium iodide to obtain separable products 23 and (−)-curcumol (1) in a 1[thin space (1/6-em)]:[thin space (1/6-em)]1 ratio. Compound 1 thus obtained showed spectral data that were identical to those recorded for natural (−)-curcumol.20 This synthesis also confirmed the structural assignment for diastereomeric cycloadduct 3. The optical rotation of synthetic curcumol ([α]20D = −12.5 (c = 0.20, CHCl3)) was comparable to that reported for natural (−)-curcumol ([α]30D = −32.26 (c = 2.13, CHCl3)).12

Conclusions

We have accomplished an asymmetric synthesis of the natural product (−)-curcumol, through a rhodium carbene intramolecular cyclization–cycloaddition cascade reaction to construct the characteristic curcumol ring framework. This route has the potential to synthesize derivatives of 1 for SAR studies and possibly related natural products.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We acknowledge the financial support from the National Natural Science Foundation of China (21372203, 21272076), the State Key Laboratory of Synthetic Chemistry, and the University of Hong Kong.

Notes and references

  1. L. Nahar and S. D. Arker, in Medicinal and Aromatic Plants – Industrial Profiles, ed. P. N. Ravindran, K. Nirmal Babu and K. Sivaraman, CRC Press, Boca Raton, 2007, vol. 45, ch. 3, pp. 71–106 CrossRef CAS PubMed; W. Sun, S. Wang, W. Zhao, C. Wu, S. Guo, H. Gao, H. Tao, J. Lu, Y. Wang and X. Chen, Crit. Rev. Food Sci. Nutr., 2017, 57, 1451–1523 CrossRef CAS PubMed.
  2. G. K. Jayaprakasha, L. J. M. Rao and K. K. Sakariah, Trends Food Sci. Technol., 2005, 16, 533–548 CrossRef CAS.
  3. S. N. Garg, A. A. Naquvi, R. P. Bansal, J. R. Bahl and S. Kumar, J. Essent. Oil Res., 2005, 17, 29–31 CrossRef.
  4. C. X. Zhou, L. S. Zhang, F. F. Chen, H. S. Wu, J. X. Mo and L. S. Gan, Fitoterapia, 2017, 121, 141–145 CrossRef CAS PubMed.
  5. H. Hikino, K. Agatsuma and T. Takemoto, Chem. Pharm. Bull., 1969, 17, 959–960 CrossRef CAS; J. W. Bats and S. H. Ohlinger, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1999, 55, 1595–1598 Search PubMed; S. Lakshmi, G. Padmaja and P. Remani, Int. J. Med. Chem., 2011, 253962 Search PubMed.
  6. H. Hikino, K. Meguro, Y. Sakurai and T. Takemoto, Chem. Pharm. Bull., 1965, 13, 1484–1485 CrossRef CAS PubMed.
  7. H. Hikino, Y. Sakurai, S. Numabe and T. Takemoto, Chem. Pharm. Bull., 1968a, 16, 39–42 Search PubMed.
  8. H. Matsuda, T. Morikawa, I. Toguchida, K. Ninomiya and M. Yoshikawa, Heterocycles, 2001, 55, 841–846 CrossRef CAS.
  9. L. Ning, H. Ma, Z. Jiang, L. Chen, L. Li, Q. Chen and H. Qi, Integr. Cancer Ther., 2016, 15, 216–225 CrossRef CAS PubMed.
  10. J. Wang, F. Huang, Z. Bai, B. Chi, J. Wu and X. Chen, Int. J. Mol. Sci., 2015, 16, 19851–19867 CrossRef CAS PubMed.
  11. Y. M. Liu, H. R. Fan, J. Ding, C. Huang, S. Deng, T. Zhu, T. L. Xu, W. H. Ge, W. G. Li and F. Li, Sci. Rep., 2017, 7, 46654 CrossRef PubMed.
  12. S. Inayama, J. F. Gao, K. Harimaya, T. Kawamato, Y. Iitaka and Y. T. Guo, Chem. Pharm. Bull., 1984, 32, 3783–3786 CrossRef CAS PubMed; K. Harimaya, J. F. Gao, T. Ohkura, T. Kawamata, Y. Iitaka, Y. T. Guo and S. Inayama, Chem. Pharm. Bull., 1991, 39, 843–853 CrossRef.
  13. The only other synthesis of curcumol was from its biogenetic precursor, curdione, through a one-step transformation: X. Li, L. Wu, Z. Ji, Y. Harigaya, Y. Konda, M. Iguchi, H. Takahashi and M. Onda, J. Heterocycl. Chem., 1988, 25, 1403–1406 CrossRef CAS.
  14. A. Padwa and M. D. Weingarten, Chem. Rev., 1996, 96, 223–270 CrossRef CAS PubMed; A. Padwa, Chem. Commun., 1998, 1417 RSC; M. P. Doyle, M. A. McKervey and T. Ye, Modern Catalytic Methods for Organic Synthesis with Diazo Compounds, J. Wiley & Sons, New York, 1998, ch. 7, pp. 397–416 Search PubMed; D. M. Hodgson, A. H. Labande and S. Muthusamy, Org. React., 2013, 80, 133–496 Search PubMed; T. Hashimoto and K. Maruoka, Chem. Rev., 2015, 115, 5366–5412 CrossRef PubMed; A. Ford, H. Miel, A. Ring, C. N. Slattery, A. R. Maguire and M. A. McKervey, Chem. Rev., 2015, 115, 9981–10080 CrossRef PubMed.
  15. P. Chiu, B. Chen and K. F. Cheng, Org. Lett., 2001, 3, 1721–1724 CrossRef CAS PubMed; B. Chen, R. Y. Y. Ko, M. S. M. Yuen, K. F. Cheng and P. Chiu, J. Org. Chem., 2003, 68, 4195–4205 CrossRef PubMed; Z. Geng, B. Chen and P. Chiu, Angew. Chem., Int. Ed., 2006, 45, 6197–6201 CrossRef PubMed.
  16. X. Zhang, R. Y. Y. Ko, S. Li, R. Miao and P. Chiu, Synlett, 2006, 1197–1200 CAS.
  17. D. A. Evans, M. D. Ennis and D. J. Mathre, J. Am. Chem. Soc., 1982, 104, 1737–1739 CrossRef CAS.
  18. K. K. D. Amarasinghe and J. Montgomery, J. Am. Chem. Soc., 2002, 124, 9366–9367 CrossRef CAS PubMed.
  19. G. Cahiez, A. Alexakis and J. F. Normant, Tetrahedron Lett., 1978, 19, 3013–3014 CrossRef.
  20. L. Zhou, W. Xu, Y. Chen, J. Zhao, N. Yu, B. Fu and S. You, Catal. Commun., 2012, 28, 191–195 CrossRef CAS.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c7qo01150d

This journal is © the Partner Organisations 2018