Letter pubs.acs.org/OrgLett
Stereoselective Sequence toward Biologically Active Fused Alkylidenecyclobutanes Andreas N. Baumann, Michael Eisold, and Dorian Didier* Department of Chemistry and Pharmacy, Ludwig-Maximilians-University, Butenandtstraße 5-13, 81377 Munich, Germany S Supporting Information *
ABSTRACT: Combining an efficient preparation of cyclobutenylmetal species, high-yielding cross-coupling reactions, and highly diastereoselective [4 + 2]-cycloaddition led to opening a new route toward the synthesis of fused alkylidenecyclobutanes containing up to five consecutive stereocenters. New complex architectures, analogues to protoilludane skeletons, were obtained in a very efficient manner and with a minimum number of steps starting from commercial sources and were tested for their cytotoxicity against leukemia cell lines HL60.
A
Scheme 1. Retrosynthetic Approach
mong small, strained ring systems, unsaturated fourmembered rings have fueled curiosity in the organic chemistry community as their selective formation remains a synthetic challenge. Over the past decade, alongside great advances in homogeneous catalysis, a growing interest in regioand stereoselective methods has emerged,1 featuring the importance and potential applications of such architectures.2 However, the study of cyclobutenes (CBs) and alkylidenecyclobutanes (ACBs) has undeniably suffered from the restricted number of strategies allowing their preparation. Recently, we have demonstrated the great ability of in situ generated cyclobutenylmetal species to undergo a subsequent cross-coupling reaction toward the formation of decorated cyclobutenes.3 Alternatively, a stereoselective double boron homologation led to new embedded allylboron reagents that subsequently reacted with a variety of aldehydes to stereoselectively furnish enantioenriched ACBs in good to excellent yields.4 We envisioned that opening a new and straightforward access to vinylcyclobutenes could ultimately unravel a path toward the diastereoselective synthesis of fused ACBs via a simple [4 + 2]-cycloaddition,5 starting from readily available building blocks (Scheme 1), and leading expediently to direct heterocyclic analogues of protoilludanes, a family of sesquiterpenoids.6 Based on the cyclization strategy pioneered by Negishi, substituted homopropargyl bromides were employed to access the key cyclobutenylmetal intermediate (Scheme 2) through a successive deprotonation/carbometalation/π-cyclization sequence.7 When an organozinc reagent ([M] = ZnCl) was employed for carbometalation, the possibility of a one-pot Negishi crosscoupling was explored to directly access vinylcyclobutenes in a one-pot sequence. At first, (Z)-alkyl-substituted alkenyl iodides 3a−c were employed, giving the corresponding (Z)-4a−c with good yields and retention of the double-bond configuration. (E)-3d and the aryl-subsituted (Z)-3e also afforded the © 2017 American Chemical Society
Scheme 2. Formation of the Key Cyclobutenylmetal Intermediate 2
corresponding dienes (E)-4d and (Z)-4e with excellent yields up to 98% (Scheme 3). Alternatively, iodolysis of the intermediate cyclobutenylmetal 2-[M] gave access to corresponding iodocyclobutenes 2-I, which was used as the cross-coupling partner in the presence of Received: March 10, 2017 Published: April 4, 2017 2114
DOI: 10.1021/acs.orglett.7b00724 Org. Lett. 2017, 19, 2114−2117
Letter
Organic Letters
4). A known model for the prediction of the stereochemical outcome of the [4 + 2]-cycloaddition can be proposed in which
Scheme 3. Vinylcyclobutenes 4
Scheme 4. [4 + 2]-Cycloaddition toward Fused ACBs 6
orbital overlaps in the transition state [TS] are involved, leading to the major formation of the endo products 6a−d (dr >99:1).8 Changing the R3 chain to a methyl or phenyl substituent (4l and 4r) allowed for performing the cycloaddition within shorter reaction times, which we explained by a decrease in steric hindrance at the reacting carbon. Compounds 6e−g were obtained in high yields up to 85% (dr >99:1). On the basis of this successful experiment, we envisioned that the sequence could be shortened by using the crude dienes resulting from the cross-coupling reactions. Previously optimized conditions showed a complete conversion of the starting materials with a minimum amount of impurities, allowing for the direct use of the vinylcyclobutenes after simple extraction from the palladium salts. After the solvent was switched to toluene, the appropriate dienophile was added to the mixture and taken up to 90 °C for 18 h in a sealed tube. When a cis-iodoalkene 3 (R4 = alkyl chain) was employed for the Negishi cross-coupling, subsequent Diels−Alder reactions with maleic anhydride led to fused ACBs 6h,i with moderate yields (40−48%), while N-methylmaleimide gave 6j,k with good yields up to 60% over three consecutive steps (cyclization−Negishi−Diels−Alder). In all cases, full control of the diastereoselectivity was observed (de >97%). Crosscoupling of 1-phenylvinylboronic acid (R3 = Ph) gave access to 6v after a two-step sequence (Scheme 5). Interestingly, using cyclic alkenylboronic acids in situ generated via a Shapiro reaction from the corresponding tosyl hydrazones furnished bicyclic dienes that were engaged without purification in the Diels−Alder cycloaddition to give tetracyclic fused ACBs 6l,m in moderate yields but with excellent diastereomeric ratios over three consecutive steps (Shapiro− Suzuki−Diels−Alder).9 Initializing the sequence by direct deprotonation of 2,3-dihydrofuran followed by transmetalation to zinc and Negishi cross-coupling furnished the tetracyclic ACB 6n after cycloaddition in 47% yield.10 Alternatively, starting with trans-alkenylboronic acid in a Suzuki crosscoupling of cyclobutenyl iodides for the diene synthesis followed by a [4 + 2]-cycloaddition led to obtaining epimer structures 6o−u in reasonable yields in a two-step sequence.
an alkenylmetal species. Employing alkyl-substituted alkenylboronic acids furnished dienes (E)-4f−h and (E)-m,n with excellent yields. Arylated alkenylboronic acids led to dienes (E)-4i−k with up to 98% yield. Branched alkenylboronic acids underwent a smooth Suzuki cross-coupling reaction, giving 4l in 81% yield. Finally, 2-I was engaged in the Negishi crosscoupling with isopropenylzinc, and the dienes 4r−t were isolated in up to 86% yield. With a solid building block synthesis, [4 + 2]-Diels−Alder cycloadditions were first attempted on isolated cyclobutaneembedded dienes 4. Maleic anhydride 5a, N-methylmaleimide 5b, and N-phenylmaleimide 5c were chosen as commercially available dienophile partners. Having an alkyl chain as the R5 substituent (from (E)-4n,o) led to generation of fused ACBs 6a−d with good yields (up to 65%) and a perfect control of the diastereoselectivity (Scheme 2115
DOI: 10.1021/acs.orglett.7b00724 Org. Lett. 2017, 19, 2114−2117
Letter
Organic Letters Scheme 5. [4 + 2]-Cycloaddition toward Fused Alkylidenecyclobutanes
Scheme 6. Diastereoselective [4 + 2]-Cycloaddition of Chiral Dienes 4
Scheme 7. Cationic ACB Ring Enlargement
We then took a step further by performing a stereoselective [4 + 2]-cycloaddition by initially having a lateral chain on the starting diene. As described in Scheme 6, the cycloaddition should take place on the less hindered diastereotopic face of the cyclobutene, on the opposite side of R2. Two substrates were employed in which R2 = (CH2)2Ph (7a−d) or n-pentyl (7e,f).11 The simple addition of N-methylmaleimide on the crude dienes furnished the corresponding fused ACBs with up to 77% yields over two synthetic steps and with a perfect control of the diastereoselectivity induced by the presence of R2. With the aim of uncovering a straightforward access to protoilludane skeleton analogues, we envisioned that the tricyclic ACB 8 could be transformed by a literature-known sequence using ethylacetoacetate to formally substitute the oxygen atom by a carbon atom.12 However, we surprisingly noted the formation of a new entity resulting from an unprecedented ring-enlargement rearrangement, as confirmed by X-ray analysis (Scheme 7).8,11 We propose to explain this unusual rearrangement by
the initial protonation of the alkylidene group to give a tertiary carbocation 11. The subsequent ring-opening−ring-closing sequence led to the more stable cyclobutane structure 13 through 12. 1,2-Alkyl transposition gives the secondary carbocation 14, which finally undergoes a ring-closing reaction of the carboxylic moiety and furnishes exclusively 9 as a crystalline compound.11 Furthermore, deuteration experiments with DCl supported the initial protonation of the alkylidene as described in Scheme 7. Finally, the utility of such a methodology was featured with bioassays on different tricylic alkylidenecyclobutanes. While a range of bacteria, fungi, and algae seemed to be resistant, 2116
DOI: 10.1021/acs.orglett.7b00724 Org. Lett. 2017, 19, 2114−2117
Organic Letters
■
interesting activities were observed when tested against HL60 (leukemia cell lines). Cytotoxicity of 6f,g, 6r, 6b,c, and 6p was measured via MTT assays (Figure 1). Only a low bioactivity was obtained for a
tetra-alkyl-substituted alkylidene 6f. However, changing the side chain to a phenyl group (6g) drastically improved the activity against HL60 (IC50 = 23 μM). A reasonable cytotoxicity was measured for 6r and 6b,c possessing alkyl chains on the central ring (IC50 = 17−30 μM). Finally, exchanging the previous alkyl group for the p-CF3-phenyl moiety (6p) improved the cytotoxicity to 14 μM. In conclusion, we have demonstrated a very efficient and expedient route to access tri- and tetracyclic fused alkylidenecyclobutanes with perfect control over the diastereoselective outcome and possessing up to five stereogenic centers, one being quaternary. Starting from readily available substrates, targets were obtained in a minimum number of steps, requiring only a single and final purification. Some of the structures showed a specific cytotoxicity against HL60, encouraging us to pursue our investigations further toward potential applications in pharmacology.
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.7b00724. Experimental procedures and spectroscopic characterization (IR, HRMS, 1H and 13C NMR data, and X-ray diffraction data) of all new compounds (PDF)
■
REFERENCES
(1) (a) Kim, S. M.; Park, J. H.; Kang, Y. K.; Chung, Y. K. Angew. Chem., Int. Ed. 2009, 48, 4532−4535. (b) Zhao, J.-F.; Loh, T.-P. Angew. Chem., Int. Ed. 2009, 48, 7232−7235. (c) Suárez-Pantiga, S.; Hernández-Díaz, C.; Rubio, E.; González, J. M. Angew. Chem., Int. Ed. 2012, 51, 11552−11555. (d) Faustino, H.; Alonso, I.; Mascareñas, J. L.; López, F. Angew. Chem., Int. Ed. 2013, 52, 6526−6530. (e) Markham, J. P.; Staben, S. T.; Toste, D. F. J. Am. Chem. Soc. 2005, 127, 9708−9709. (f) Kurahashi, T.; de Meijere, A. Angew. Chem., Int. Ed. 2005, 44, 7881−7884. (g) Chen, L.; Shi, M.; Li, C. Org. Lett. 2008, 10, 5285−5288. (h) Charpenay, M.; Boudhar, A.; Blond, G.; Suffert, J. Angew. Chem., Int. Ed. 2012, 51, 4379−4382. (i) Innitzer, A.; Brecker, L.; Mulzer, J. Org. Lett. 2007, 9, 4431−4434. (2) (a) Boontanonda, P.; Grigg, R. J. J. Chem. Soc., Chem. Commun. 1977, 583−584. (b) Tobe, Y.; Kishida, T.; Yamashita, T.; Kakiuchi, K.; Odaira, Y. Chem. Lett. 1985, 14, 1437−1440. (c) Samuel, S. P.; Niu, T.-q.; Erickson, K. L. J. Am. Chem. Soc. 1989, 111, 1429−1436. (d) Jiang, M.; Shi, M. Org. Lett. 2008, 10, 2239−2242. (e) Crépin, D.; Dawick, J.; Aïssa, C. Angew. Chem., Int. Ed. 2010, 49, 620−623. (3) Eisold, M.; Baumann, A. N.; Kiefl, G. M.; Emmerling, S. T.; Didier, D. Chem. - Eur. J. 2017, 23, 1634−1644. (4) (a) Eisold, M.; Didier, D. Angew. Chem., Int. Ed. 2015, 54, 15884−15887. (b) Baumann, A. N.; Music, A.; Didier, D. Chem. Commun. 2016, 52, 2529−2532. (c) Eisold, M.; Kiefl, G. M.; Didier, D. Org. Lett. 2016, 18, 3022−3025. (5) (a) Nishimura, A.; Tamai, E.; Ohashi, M.; Ogoshi, S. Chem. - Eur. J. 2014, 20, 6613−6617. (b) Qiu, Y.; Yang, B.; Zhu, C.; Bäckvall, J.-E. Angew. Chem., Int. Ed. 2016, 55, 6520−6524. (6) Himmelbauer, M.; Farcet, J.-B.; Gagnepain, J.; Mulzer, J. Eur. J. Org. Chem. 2013, 2013, 8214−8244. (7) (a) Boardman, L. D.; Bagheri, V.; Sawada, H.; Negishi, E.-i. J. Am. Chem. Soc. 1984, 106, 6105−6107. (b) Liu, F.; Negishi, E.-i. Tetrahedron Lett. 1997, 38, 1149−1152. (8) Relative configurations resulting from [4 + 2]-cycloadditions were attested by 2D NMR studies and X-ray diffraction; see the SI. (9) In-situ cross-coupling following the Shapiro reaction was done according to: Passafaro, M. S.; Keay, B. A. Tetrahedron Lett. 1996, 37, 429−432. (10) Hornillos, V.; Giannerini, M.; Vila, C.; Fananas-Mastral, M.; Feringa, B. L. Chem. Sci. 2015, 6, 1394−1398. (11) CCDC 1536131 (7b), CCDC 1536132 (8), and CCDC 1536133 (9) contain the supplementary crystallographic data for this paper. (12) Wilbuer, J.; Schnakenburg, G.; Esser, B. Eur. J. Org. Chem. 2016, 2016, 2404−2412.
Figure 1. IC50 measurements of fused ACBs against HL60.
■
Letter
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Michael Eisold: 0000-0002-2314-3990 Dorian Didier: 0000-0002-6358-1485 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS M.E. and D.D. are grateful to the Chemical Industry Fund (FCI Liebig-fellowship) and A.N.B. to the SFB749 for financial support. Prof. Dr. Paul Knochel (LMU, Munich) is kindly acknowledged for his generous support. We thank Martina Stadler for biological activity measurements and Dr. Peter Mayer for X-ray measurments (LMU, Munich). 2117
DOI: 10.1021/acs.orglett.7b00724 Org. Lett. 2017, 19, 2114−2117
