oxepines b - ACS Publications - American Chemical Society

Mar 27, 2018 - ABSTRACT: Ylide-type reactivity of diazo compounds is exploited in a new way to prepare benzo[b]oxepines thanks to the formation of thr...
3 downloads 5 Views 1MB Size
Letter Cite This: Org. Lett. XXXX, XXX, XXX−XXX

pubs.acs.org/OrgLett

Straightforward Entry toward Highly Substituted 2,3Dihydrobenz[b]oxepines by Ring Expansion of Benzopyryliums with Donor−Acceptor Diazo Compounds Thibaut Courant, Morgane Pasco, and Thomas Lecourt* Normandie Univ, INSA Rouen, UNIROUEN, CNRS, COBRA UMR 6014, 76000 Rouen, France S Supporting Information *

ABSTRACT: Ylide-type reactivity of diazo compounds is exploited in a new way to prepare benzo[b]oxepines thanks to the formation of three chemical bonds and two contiguous and highly substituted stereocenters in a single pot. This cationic reaction cascade first involves addition of a donor−acceptorsubstituted diazo compound to a benzopyrylium. Selective 1,2 migration of the endocyclic C−C bond then results in a ring-expansion and generates a second oxocarbenium that is trapped by a nucleophile added sequentially.

T

Scheme 1. Synthetic approaches toward benzo[b]oxepines

he benzoxepine unit represents a major structural motif found in many natural products and synthetic bioactive compounds.1 More particularly, 3- and 5-arylbenzo[b]oxepines were recently shown to have promising anti-inflammatory properties (Figure 1)2 and to specifically modulate and downregulate estrogen receptors respectively (ERs).3

Figure 1. Selected examples of biologically relevant benzo[b]oxepines.

However, preparation of densely functionalized benzoxepines usually requires the multistep preparation of highly elaborated precursors. In fact, these highly valuable heterocycles are usually formed from ortho-substituted phenol derivatives following transition-metal-catalyzed cyclization or annulation approaches (Scheme 1, eqs 1 and 2).4 Alternative routes based on the rearrangement of fused cyclopropanes have also been reported, but structural diversification following these methods also suffers from similar drawbacks (Scheme 1, eq 3).5 In this context, straightforward approaches giving rise to a large molecular diversity of benzoxepines are strongly appealing,6 and we report herein a new route toward highly substituted 2,3-dihydrobenzo[b]oxepines involving homologation of benzopyryliums with donor−acceptor-substituted diazo compounds. With the advent of organocatalysis, ylide reactivity of diazo compounds has been pushed toward an exquisite level of © XXXX American Chemical Society

sophistication over the past decade.7 These transformations first involve formation of a betaine by addition of the ylide to the polarized double bond of a carbonyl or imine derivative. After cyclization or 1,2 migration reactions, which are driven by spontaneous nitrogen gas evolution, a large diversity of strained heterocycles and homologated carbonyl derivatives are obtained from readily available starting materials, thanks to the formation of two chemical bonds from readily available starting materials.8 Herein, an unprecedented reaction cascade between donor−acceptor substituted diazo compounds and stabilized carbocations has been developed to prepare, in a single pot, benzo[b]oxepines with contiguous and highly Received: March 27, 2018

A

DOI: 10.1021/acs.orglett.8b00984 Org. Lett. XXXX, XXX, XXX−XXX

Letter

Organic Letters Scheme 2. One-Pot Preparation of Benzo[b]oxepines by Homologation of Benzopyryliums with Diazo Compounds

substituted stereocenters (Scheme 2).9,10 This new transformation first involves the formation of benzopyrylium B from chromene A under acid catalysis. Trapping this cationic intermediate with carbenoid C gives adduct D from which 1,2 migration of the endocyclic C−C bond results in ringexpansion and induces the formation of a second oxocarbenium E.11 Sequential addition of a second nucleophile finally delivers benzo[b]oxepines F with a large molecular diversity at positions C2 and C3. Our studies started with 2-methoxy-4-phenyl 2H-chromene 1 as a benzopyrylium precursor and ethyl 2-diazo-2-phenylacetate 2 (Table 1). Addition of TMSOTf (30 mol %) to a solution of Table 1. Homologation of 2H-chromene 1 into Benzo[b]oxepine 3 Figure 2. ORTEP representation of 3.

1 2d 3 4 5 6 7 8e 9f 10

solvent

TMSOTf (mol %)

2 (equiv)

yieldb (%)

drc

CH2Cl2 CH2Cl2 1,2-DCE Ph−CF3 Ph−CH3 CH2Cl2 CH2Cl2 CH2Cl2 CH2Cl2 CH2Cl2

30 30 30 30 30 30 30 30 30 15

1.2 1.2 1.2 1.2 1.2 1.0 2.0 1.2 1.2 1.2

89 70 72 72 28 70 80 87 83 65

2.5:1 1:1 2.5:1 1.8:1 2.0:1 2.1:1 2.1:1 1.8:1 0.8:1 2.2:1

With trifluoromethanesulfonic acid as catalyst (Table 1, entry 2), a 1:1 mixture of diastereosiomers was obtained in 70% yield. Bronsted acid catalysis is thus strongly detrimental to the stereoselectivity of this transformation. Influence of solvents was then investigated. 1,2-Dichloroethane (1,2-DCE) gave benzoxepine 3 in good yield and identical diastereoselectivity (Table 1, entry 3). Trifluorotoluene delivered 3 in 72% yield, but with a slightly decreased diastereoisomeric ratio (Table 1, entry 4), and toluene proved to be strongly detrimental to this transformation (Table 1, entry 5). The amount of diazo compound was next varied. With 1 equiv of ethyl 2-diazo-2phenylacetate 2 (Table 1, entry 6), 3 was obtained in a reduced 70% yield, but an excess reagent gave a product contaminated with small amounts of impurities (Table 1, entry 7). The diastereoselectivity was decreased when the reaction was carried out at room temperature (Table 1, entry 8). At −40 °C, stereoselectivity was slightly reversed to give 3 as 0.8:1 mixture of diastereoisomers (Table 1, entry 9). Reducing catalyst loading to 15 mol % delivered benzoxepine 3 in a slightly reduced 65% yield (Table 1, entry 10). The scope of this reaction cascade leading to benzo[b]oxepines by homologation of benzopyrylium with donor− acceptor diazo compounds was next investigated under optimal reaction conditions (TMSOTf 30 mol %, CH2Cl2, −15 °C) (Scheme 3). First, methyl and tert-butyl diazophenylacetates gave benzoxepines 4 and 5 with a slightly reduced diastereoselectivity in 73% and 79% yields, respectively. Substitution of the aromatic ring of the 2H-chromene with electron-donating and -withdrawing groups was next considered. With a tert-butyl or methoxy group at position 6, compounds 6 and 7 were obtained in 73% and 81% yields respectively, but a complete loss of stereoselectivity was observed. 7-Chloro-2-methoxy-4-phenyl-2H-chromene delivered 8 in 81% yield with an increased diastereoselectivity of 3.8:1. Substitution of the aromatic moiety of diazo esters with

a

0.1 mmol scale. bIsolated yield. cMeasured by 1H NMR analysis of the crude. dTfOH as the catalyst instead of TMSOTf. eRoom temperature. f−40 °C.

1 and 2 (1.2 equiv) in CH2Cl2 at −15 °C resulted in the disappearance of the bright yellow color and complete nitrogen gas evolution within minutes. After the reaction mixture was quenched with methanol, the residue was purified by silica gel flash chromatography to give benzo[b]oxepine 3 in 89% yield as a 2.5:1 inseparable mixture of diastereoisomers (Table 1, entry 1). Even if formation of the second asymmetric center occurred with moderate stereoselectivity, the diastereoisomeric ratio was increased to 8.8:1 by simple trituration in Et2O. After recrystallization from Et2O, the major stereoisomer was finally obtained in pure form, and the configuration was unambiguously assigned by X-ray diffraction analysis (Figure 2). Stereochemistry of all compounds was next assigned on the basis of strong similarities between characteristic signals in the 1 H NMR spectra. B

DOI: 10.1021/acs.orglett.8b00984 Org. Lett. XXXX, XXX, XXX−XXX

Letter

Organic Letters Scheme 3. Variation of Diazo Compound and Benzopyrylium Precursor

Table 2. Quenching with allyltrimethylsilane

a

entrya

Lewis acid

equiv

yieldb (%)

drc

1 2d 3 4 5 6

TMSOTf BF3.OEt2 SnCl4 TiCl4 TiCl4

2.0 0.3 0.3 0.3 2.0

0 0 0 0 tracesc 84

3.2:1 b

0.1 mmol scale, allyl trimethylsilane (5 equiv). Isolated yield. Detected on the 1H NMR spectrum of the crude. dReaction carried from −15 °C to room temperature. c

Having shown that allylation could be achieved efficiently after ring expansion, trapping with various nucleophiles was next investigated to increase the molecular diversity at position 2 (Scheme 4). Reduction was first achieved with Et3SiH to give Scheme 4. Late-Stage Diversification at Position 2

p-bromo and p-nitro groups furthermore gave benzoxepines 9 and 10 in good yields and high diastereoisomeric ratios. Finally, compound 11 with a phenylethynyl substituent at position 5 was obtained in 52% yield. Trapping the second cationic intermediate with allyltrimethylsilane instead of methanol was next attempted (Table 2), but methyl acetal 3 was still observed in the crude instead of the expected compound 12 (Table 2, entry 1). Adding excess TMSOTf concomitantly to the carbon nucleophile did not give any detectable traces of 12 in the 1H NMR spectrum of the crude, even if the reaction mixture was carried out at room temperature (Table 2, entry 2). Allylation with catalytic BF3· OEt2 or SnCl4 (0.3 equiv) again gave 3 as the sole product (Table 2, entries 3 and 4), but traces of compound 12 were observed with catalytic TiCl4 (Table 2, entry 5). Benzoxepine 12 was finally obtained in 84% yield as a 3.2:1 mixture of diasteroisomers when 2 equiv of TiCl4 was added to the reaction mixture (Table 2, entry 6).

13 in 91% yield. Addition of trimethylsilyl azide and cyanide gave 14 and 15 respectively in excellent yields. Addition of cyanide to the second cationic intermediate occurred without any stereoselectivity, but benzoxepine 14 was obtained in a high 4:1 diastereoisomeric ratio after trapping with azide. Finally, quenching with propargyl alcohol delivered acetal 16 in 86% isolated yield with a reversed diastereoselectivity. Having shown that allylation could be achieved efficiently after ring expansion, trapping with various nucleophiles was next investigated to increase the molecular diversity at position 2 (Scheme 4). Reduction was first achieved with Et3SiH to give 13 in 91% yield. Addition of trimethylsilyl azide and cyanide gave 14 and 15 respectively in excellent yields. Addition of cyanide to the second cationic intermediate occurred without any stereoselectivity, but benzoxepine 14 was obtained in a high C

DOI: 10.1021/acs.orglett.8b00984 Org. Lett. XXXX, XXX, XXX−XXX

Letter

Organic Letters 4:1 diastereoisomeric ratio after trapping with azide. Finally, quenching with propargyl alcohol delivered acetal 16 in 86% isolated yield with a reversed diastereoselectivity. We next investigated the mechanism of this unprecedented transformation (Scheme 5). Plausibly, after Lewis acid

Scheme 6. Control Experiments for Mechanistic Investigations

Scheme 5. Putative Mechanism for the Ring-Expansion and Late-Stage Diversification at Position 2

the increasing steric bulk of the oxygenated nucleophile thus resulted in an almost complete loss of stereoselectivity. During optimization studies, TMSOTf delivered 3 as a 2.5:1 mixture of stereoisomers (Table 1, entry 1), whereas Bronsted acid catalysis peculiarly gave an equimolar amount of stereoisomers (Table 1, entry 2). In this context, we next questioned the reversibility of the formation of 3 from 19 depending on the reaction conditions, i.e., mechanism of the structural diversification at position 2. TiCl4-promoted transketalization of compound 3 with MeOH first delivered a 1.2:1 mixture of diasteroisomers independently from the dr of the starting materials (Scheme 6, eq 3). Sequential addition of a strong Lewis acid, or initial Bronsted acid catalysis, thus promotes activation of methyl acetal 3 that is quantitatively formed during ring-expansion and results in equilibration of 3 into a 1.2:1 thermodynamic mixture of diastereoisomers. Transketalization of 3 should furthermore involve a pure SN1 mechanism with oxocarbenium 19 as the reactive intermediate. Moreover, the 1:3 diastereoisomeric ratio for propargyl acetal 16 would also account for a thermodynamic control of the transketalization process (Scheme 4). Finally, 2-allyl benzoxepine 12 was obtained in identical dr whatever the diastereoisomeric excess of the starting material (Scheme 6, eq 4). TiCl4-mediated addition of carbon nucleophiles to 3 should consequently also involve a dissociative mechanism. In conclusion, ylide reactivity of diazo esters has been exploited herein in a new way to create three chemical bonds and two contiguous and highly substituted stereocenters in a single pot from stabilized cationic substrates. A diverse series of densely functionalized benzo[b]oxepines was thus efficiently obtained by homologation of benzopyryliums with donor− acceptor diazo compounds. Key steps of this reaction cascade

promoted activation of 1 into benzopyrylium 17, addition of diazo compound 2 would give 18. Homologation by 1,2 migration of the endocyclic C,C bond would then deliver a seven-membered ring oxocarbenium 19.12 Finally, closing the catalytic cycle of this ring-expansion would require to form methyl acetal 3 from 19. This process could either involve trapping of 19 by the MeOSiMe3 released initially, or a transketalization process between 1 and the cationic intermediate 19. In order to investigate deeper activation of 1 under catalytic conditions, CD3OD has been used instead of MeOH to quench the reaction mixture (Scheme 6, eq 1). Obtaining benzo[b]oxepine 3 without any isotope incorporation revealed that oxocarbenium 19 was not present anymore in the reaction mixture after completion of the ring-expansion, i.e., that oxocarbenium 19 had been trapped by MeOSiMe3 to give 3 and regenerate the Lewis acid.13 This control experiment furthermore proved that formation of 3 from 19 would be irreversible with TMSOTf, as anticipated from allylation studies (see Table 2, entries 1 and 2). Obtaining compound 3 as a 2.5:1 mixture of diastereoisomers under TMSOTf catalysis (Table 1, entry 1) thus came from the kinetic trapping of oxocarbenium 19 by MeOSiMe3. Under kinetic control, increasing reactivity of the second oxocarbenium by substitution of the aromatic moiety with electron-withdrawing groups increased the diastereosiomeric ratio (Scheme 3, compound 8). On the contrary, stabilization of homologated oxocarbeniums by electron-donating substituents resulted in a complete loss of diastereoselectivity (Scheme 3, compounds 6 and 7). Finally, 2-isopropoxy-4-phenyl 2H-chromene 20 delivered benzo[b]oxepine 21 in 78% yield as a 1.1:1 mixture of diastereoisomers (Scheme 6, eq 2). Under kinetic conditions, D

DOI: 10.1021/acs.orglett.8b00984 Org. Lett. XXXX, XXX, XXX−XXX

Letter

Organic Letters

4696. (d) Casanova, N.; Del Rio, K. P.; Garcia-Fandino, R.; Mascarenas, J. L.; Gulias, M. ACS Catal. 2016, 6, 3349−3353. (e) Mangina, N. S. V. M. R.; Kadiyala, V.; Guduru, R.; Goutham, K.; Sridhar, B.; Karunakar, G. V. Org. Lett. 2017, 19, 282−285. (5) (a) Dean, F. M.; Park, B. K. J. Chem. Soc., Perkin Trans. 1 1976, 1, 1260−1268. (b) Xie, Y.; Zhang, P.; Zhou, L. J. Org. Chem. 2016, 81, 2128−2134. (6) For recent approaches involving late-stage diversifications at position 2, see: (a) Kelley, B. T.; Walters, J. C.; Wengryniuk, S. E. Org. Lett. 2016, 18, 1896−1899. (b) Chen, C.-R.; Lai, Y.-X.; Wu, R.-Y.; Liu, Y.-H.; Lin, Y.-C. ChemCatChem 2016, 8, 2193−2196. (c) Liu, J.; Liu, Y. Org. Lett. 2012, 14, 4742−4745. (7) For reviews, see: (a) Johnston, J. N.; Muchalski, H.; Troyer, T. L. Angew. Chem., Int. Ed. 2010, 49, 2290−2298. (b) Zhang, Y.; Wang, J. Chem. Commun. 2009, 5350−5361. (c) Zhang, Y.; Lu, Z.; Wulff, W. D. Synlett 2009, 2009, 2715−2739. (d) Hashimoto, T.; Maruoka, K. Bull. Chem. Soc. Jpn. 2013, 86, 1217−1230. (8) For leading references, see: (a) Williams, A. L.; Johnston, J. N. J. Am. Chem. Soc. 2004, 126, 1612−1613. (b) Hashimoto, T.; Naganawa, Y.; Maruoka, K. J. Am. Chem. Soc. 2008, 130, 2434−2435. (c) Hu, G.; Huang, L.; Huang, R. H.; Wulff, W. D. J. Am. Chem. Soc. 2009, 131, 15615−15617. (9) For a recent review on homologations with diazo compounds, see Candeias, N. R.; Paterna, R.; Gois, P. M. Chem. Rev. 2016, 116, 2937− 2981. (10) For a recent review on heterocycles synthesis from diazo compounds, see Choi, S.; Ha, S.; Park, C.-M. Chem. Commun. 2017, 53, 6054−6064. (11) For recent ring-expansions with trimethylsilyle diazomethane ending by elimination, see: (a) Stopka, T.; Marzo, L.; Zurro, M.; Janich, S.; Würthwein, E.-U.; Daniliuc, C. G.; Aleman, J.; Garcia Mancheno, O. Angew. Chem., Int. Ed. 2015, 54, 5049−5053. (b) Gini, A.; Bamberger, J.; Luis-Barrera, J. L.; Zurro, M.; Mas-Ballesté, R.; Aleman, J.; Garcia Mancheno, O. Adv. Synth. Catal. 2016, 358, 4049− 4056. (12) Despite being usually favored, 1,2 hydride migration has never been observed. Theoretical studies are currently underway to gain a better understanding of this homologation process that is controlled by multiple parameters (reversibility of the addition of 2 to 17; conformation of adduct 18; formation of a transient benzylic carbocation with a fused cyclopropane from 18 by attack of the endocyclic CC bond, followed by rearrangement into 19). (13) If TMSOMe could not trap 19, i.e., that a transketalization would be involved in activation of 1, a small amount of 19, close to catalyst loading, should remain in the reaction mixture after completion of the ring expansion. Quenching with CD3OD would then result in isotope incorporation at a level also close to catalyst loading. (14) For transformations involving the formation of chiral ion pairs with benzopyryliums, see: (a) Reisman, S. E.; Doyle, A. G.; Jacobsen, E. N. J. Am. Chem. Soc. 2008, 130, 7198−7199. (b) Terada, M.; Yamanaka, T.; Toda, Y. Chem. - Eur. J. 2013, 19, 13658−13662. (15) For general reviews on chiral ion pairs, see: (a) Lacour, J.; Moraleda, D. Chem. Commun. 2009, 7073−7089. (b) Brak, K.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2013, 52, 534−561. (c) Mahlau, M.; List, B. Angew. Chem., Int. Ed. 2013, 52, 518−533.

rely on a highly selective 1,2 migration that induces ringexpansion and generation of a second oxocarbenium intermediate which was trapped sequentially to give highly substituted heterocycles from readily available starting materials. Enantioselective transformations, based on the formation of chiral ion pairs, are currently under investigation in our laboratory.14,15



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.8b00984. Experimental procedures, characterization data and copies of 1H and 13C NMR spectra for compounds 316, and crystallographic data for compound 3 (PDF) Accession Codes

CCDC 1819307 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing data_ [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Morgane Pasco: 0000-0002-1556-2802 Thomas Lecourt: 0000-0002-2532-3012 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS LABEX SynOrg (ANR-11-LABX-0029) and ANR (JCJC-2013QuatGlcNAc) are gratefully acknowledged for their financial support. T.C. (Normandy University), M.P. (CRUNCH), and T.L. (CNRS) are also supported by individual research fellowships. X-ray diffraction analysis of compound 3 was performed by Dr. Pascal Retailleau (ICSN, UPR 2301 CNRS).



REFERENCES

(1) For selected examples, see: (a) Anzini, M. A.; Cappelli, A.; Vomero, S.; Giorgi, G.; Langer, T.; Hamon, M.; Merahi, N.; Emerit, B. M.; Cagnotto, A.; Skorupska, M.; Mennini, T.; Pinto, J. C. J. Med. Chem. 1995, 38, 2692−2704. (b) Kim, S.; Chin, Y.-W.; Su, B.-N.; Riswan, S.; Kardono, L. B. S.; Afriastini, J. J.; Chai, N. R.; Farnsworth, N. R.; Cordell, G. A.; Swanson, S. M.; Kinghorn, A. D. J. Nat. Prod. 2006, 69, 1769−1775. (2) Herrmann, J. M.; Untergehrer, M.; Jürgenliemk, G.; Heilmann, J.; König, B. Eur. J. Org. Chem. 2014, 2014, 3170−3181. (3) (a) O’Boyle, N. M.; Barrett, I.; Greene, L. M.; Carr, M.; Fayne, D.; Twamley, B.; Knox, A. J. S.; Keely, N. O.; Zisterer, D. M.; Meegan, M. J. J. Med. Chem. 2018, 61, 514−534. (b) Rasolofonjatovo, E.; Provot, O.; Hamze, A.; Rodrigo, J.; Bignon, J.; Wdzieczak-Bakala, J.; Lenoir, C.; Desravines, D.; Dubois, J.; Brion, J.-D.; Alami, M. Eur. J. Med. Chem. 2013, 62, 28−39. (4) For recent examples, see: (a) Sze, E. M. L.; Rao, W.; Koh, M. J.; Chan, P. W. H. Chem. - Eur. J. 2011, 17, 1437−1441. (b) Yu, X.; Lu, X. J. Org. Chem. 2011, 76, 6350−6355. (c) Calder, E. D. D.; Sharif, S. A. I.; McGonagle, F. I.; Sutherland, A. J. Org. Chem. 2015, 80, 4683− E

DOI: 10.1021/acs.orglett.8b00984 Org. Lett. XXXX, XXX, XXX−XXX