Acid-Catalyzed Regiodivergent Annulation of 4-Hydroxycoumarins with Isoprene: Entry to Pyranocoumarins and Pyranochromones

Article abstract of DOI:10.1002/ejoc.201901154

An acid-catalyzed regiodivergent formal [3+3] annulation of 4-hydroxycoumarins with isoprene is developed. A variety of pyranocoumarins were obtained exclusively in the presence of strong Br?nsted acid, while varying to Lewis acid delivered pyranochromones as main products. The protocol also features high atom-economy, wide substrate scope, easy scale-up, and good applications in natural product synthesis.

Full text of DOI:10.1002/ejoc.201901154

DOI: 10.1002/ejoc.201901154
Communication
Regiodivergent Annulation | Very Important Paper |
Acid-Catalyzed Regiodivergent Annulation of
4-Hydroxycoumarins with Isoprene: Entry to Pyranocoumarins
and Pyranochromones
Ying Li,[a,b][‡] Yan-Cheng Hu,[b][‡] Hao Zheng,[b] Ding-Wei Ji,[b] Yu-Feng Cong,*[a] and
Qing-An Chen*[b]
Abstract: An acid-catalyzed regiodivergent formal [3+3] annu- ered pyranochromones as main products. The protocol also fea-
lation of 4-hydroxycoumarins with isoprene is developed. A va-
riety of pyranocoumarins were obtained exclusively in the pres-
ence of strong Brønsted acid, while varying to Lewis acid deliv-
tures high atom-economy, wide substrate scope, easy scale-up,
and good applications in natural product synthesis.
Pyranocoumarins are ubiquitous in nature, especially in vari-
ous medicinal plants.[1] Cyclized prenylated coumarins such as
pterophyllin III, bothrioclinin, and nor-bothrioclinin are the rep-
resentatives that were primarily isolated from Ekeberyia ptero-
phylla and Bothriocline laxa (Scheme 1).[2] Besides, the com-
pounds with such framework often exhibit some intriguing bio-
logical activities. For instance, cyclocoumarol, a hemiketal form
of anticoagulant drug warfarin, was recently proved to be a
selective inhibitor of cyclooxygenase-2 (Scheme 1).[3] Ethulia-
coumarin A and isoethuliacoumarin A, two active principles of
the Egyptian plant Ethulia conyzoides, have been shown to pos-
sess significant molluscicidal activity (Scheme 1).[4] Ferprenin,
derived from Ferula communis, could inhibit vitamin K epoxide
reductase complex subunit 1 (VKORC1), a rate-limiting enzyme
for vitamin K recycling (Scheme 1).[5] Therefore, considerable
efforts have been devoted to the assembly of pyranocoumarins
over the past decades.
4-Hydroxycoumarin, which can be biosynthesized on a large-
scale in Escherichia coli,[6] is generally served as starting material
to prepare pyranocoumarins.[7] For example, this motif could be
formed via two-step reactions of 4-hydroxycoumarin and prenyl
bromide (Scheme 2a).[8] On the other hand, the annulation of
4-hydroxycoumarin with unsaturated precursors has recently
emerged as a powerful means to access pyranocoumarin skele-
tons. Propargylic alcohols,[9] α,ꢀ-unsaturated ketones/alde-
hydes,[10] and nitroalkenes[11] were well explored in such trans-
Scheme 1. Representative natural products containing pyranocoumarin scaf-
fold.
formations. 4-Hydroxycoumarin and prenal could undergo an-
nulation to deliver pyranocoumarin (Scheme 2a).[10d,12] How-
ever, these approaches often generated stoichiometric by-prod-
ucts, and most of the precursors also required additional multi-
step synthesis. In this context, it is highly desirable to exploit
an atom- and step-economical protocol to construct pyrano-
coumarins from low-cost reagents.[13]
Isoprene is an important C5 conjugated diene in industry
that can be alternatively produced by fermentation of renewa-
ble feedstock.[14] Considering pterophyllin III, bothrioclinin, and
nor-bothrioclinin also contain C5 motif, we envisaged that iso-
prene may serve as a precursor for the synthesis of such struc-
tures. However, both pyranocoumarins 3 and pyranochromones
4 are likely to be formed in the process.[15] Thus, how to tune
the regioselectivity is a challenging issue.[16] Herein, an efficient
acid-catalyzed formal [3+3] annulation of 4-hydroxycoumarin
with isoprene is developed, and notably the regioselectivity can
be manipulated through the choice of acid catalysts
(Scheme 2b). When using strong Brønsted acid [2,4-
(NO2)2C6H3SO3H] as catalyst, the reaction exclusively furnished
pyranocoumarins 3. In comparison, the selectivity switched to
[a] College of Chemistry, Chemical Engineering and Environmental Engineer-
ing, Liaoning Shihua University,
Fushun 113001, China
E-mail: yufengcong02@163.com
[b] Dalian Institute of Chemical Physics, Chinese Academy of Sciences
457 Zhongshan Road, Dalian 116023, China
E-mail: qachen@dicp.ac.cn
http://www.lbcs.dicp.ac.cn
[‡] These authors contributed equally to this work.
Supporting information and ORCID(s) from the author(s) for this article are
available on the WWW under https://doi.org/10.1002/ejoc.201901154.
Eur. J. Org. Chem. 2019, 6510–6514
6510
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Scheme 2. The annulations of 4-hydroxycoumarin with diverse unsaturated precursors.
pyranochromones 4 by varying the catalyst to Lewis acid
Sm(OTf)3.
Table 1. Optimization of the reaction conditions.[a]
4-Hydroxycoumarin 1a and isoprene 2 were selected as
model substrates to test our initial hypothesis (Table 1). When
the reaction was performed in DCE at 90 °C using p-toluenesulf-
onic acid (TsOH, acid 1, 25 mol-%) as catalyst, both pyrano-
coumarin 3a and pyranochromone 4a were afforded with 1:1
rr (regioselectivity ratio) (Table 1, entry 1). Other aromatic sul-
fonic acids including 4-chlorobenzenesulfonic acid (acid 2), 4-
nitrobenzenesulfonic acid (acid 3), and 2,4-dinitrobenzenesul-
fonic acid (acid 4) were further screened (Table 1, entries 2–4).
The pKa values of acids 1–4 in acetonitrile are 8.60, 7.30, 6.73,
and 3.00, respectively.[17] Increasing the catalyst acidity clearly
led to an improvement in the regioselectivity of 3a. The yield
of 3a reached up to 92 % and no 4a was detected at all in the
presence of 2,4-(NO2)2C6H3SO3H (acid 4). An examination of the
solvents showed that DMF completely suppressed the reaction,
while dioxane gave the best result (Table 1, entries 5–7). De-
creasing the amount of catalyst loading to 10 mol-% also exclu-
sively furnished 3a in a high yield (88 %) (Table 1, entry 8). To
our surprise, when varying the catalyst to Lewis acid Sm(OTf)3,
the main product switched to pyranochromone 4a albeit with
a low regioselectivity (Table 1, entry 9). Gratifyingly, the utiliza-
tion of DCE as solvent could increase the yield (52 %) and regio-
selectivity of 4a (5:1) (Table 1, entry 10). However, other metal
triflates such as Zn(OTf)2, Sc(OTf)3, and Yb(OTf)3 all led to infe-
rior results (Table 1, entries 11–13).
Subsequently, the optimized conditions were used to investi-
gate the substrate scope for this divergent protocol (Table 2,
Table 3). For Brønsted acid catalysis (Table 2), the desired
pyranocoumarin 3a was isolated in 81 % yield after submitting
4-hydroxycoumarin 1a and isoprene to the standard conditions.
The electronic properties and positions of the substituents on
[a] Reaction conditions: 1a (0.2 mmol), 2 (0.6 mmol), catalyst (x mol-%), sol-
vent (0.5 mL), 90 °C, 24 h. [b] Yield was determined by HPLC using naphthal-
ene as the internal standard. [c] pKa values of acids in acetonitrile were given.
N.R. = No reaction.
Eur. J. Org. Chem. 2019, 6510–6514
www.eurjoc.org
6511
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
the phenyl ring had negligible influence on the reaction (3b–
j). For example, electron-donating 6-OMe and 7-OMe hydroxy-
coumarins (1d, 1e) underwent the annulations smoothly to pro-
vide the corresponding products in 70 % and 77 % yield, re-
spectively. Electron-withdrawing groups such as –F, –Cl, and –
Br were compatible with the process as well (3f–j). A higher
temperature (110 °C) was required to achieve good yields in
the cases of benzocoumarins 1k and 1l. Notably, the annulation
of hydroxypyranone 1m with isoprene could take place at
130 °C, resulting in the formation of 3m in 53 % yield. It is not
surprising that prenylated diketone 3n was afforded using 3-
benzyl-4-hydroxycoumarin as substrate. More importantly, this
transformation can be extended to 4-hydroxyquinolinones 1o
and 1p. The resulting pyranoquinolinones 3o and 3p are the
precursors of flindersine and N-methylflindersine.[18] It is note-
worthy that no pyranochromone 4 was observed in all above
case.
Table 3. Substrate scope for pyranochromones synthesis.[a]
Table 2. Substrate scope for pyranocoumarins synthesis.[a]
[a] Reaction conditions: 1 (0.4 mmol), 2 (1.2 mmol), Sm(OTf)3 (10 mol-%), DCE
(1.0 mL), 90 °C, 24 h. Isolated yields of 4 were given. The ratio was determined
by 1H NMR of crude reaction mixture. [b] Pyranocoumarins 3h (53 %), 3m
(40 %) and pyranoquinolinone 3o (53 %) were obtained. [c] Sm(OTf)3
(1.0 equiv.).
4h, as main product. This phenomenon was also observed in
the reaction of hydroxypyranone 1m. It is most likely because
pyranochromones could be rapidly transformed into thermody-
namically stable pyranocoumarins in both cases. The coupling
of 3-benzyl-4-hydroxycoumarin 1n with isoprene also occurred
in the presence of Sm(OTf)3, but the yield of 3n (30 %) was
lower than that in Brønsted acid catalysis (56 %). Naturally oc-
curring alkaloid N-Me khaplofoline (4p) was successfully synthe-
sized from 4-hydroxyquinolinone 1p by this protocol, albeit
with a slightly low yield.[19]
To demonstrate the synthetic utility of this methodology, di-
verse transformations of pyranocoumarin 3a were studied
(Scheme 3). The annulation of 4-hydroxycoumarin with iso-
prene could be easily scaled up to 5.0 mmol without protection
from air or moisture, and a simple crystallization delivered 3a
in 53 % yield. The selective dehydrogenation of 3a by DDQ fur-
nished natural product nor-bothrioclinin 5a in a moderate yield.
The lactone motif of 3a could be efficiently reduced by LiAlH4
to afford diol 6a. The benzene ring of 3a was partially hydro-
genated in the presence of Pd/C at 100 °C, leading to 7a in
66 % yield.[20] Treatment of 3a with KOH yielded ring-opening
product 8a.
[a] Reaction conditions: 1 (0.4 mmol), 2 (1.2 mmol), 2,4-(NO2)2C6H3SO3H
(10 mol-%), dioxane (1.0 mL), 90 °C, 24 h. Only products 3 were observed
and isolated yields were given. [b] 2,4-(NO2)2C6H3SO3H (1.0 equiv.).
For Lewis acid catalysis (Table 3), 4-hydroxycoumarin 1a was
converted to pyranochromone 4a in 60 % yield with 5:1 rr un-
der the optimized conditions. A range of substituents includ-
ing –Me, –OMe, –F, –Cl, and –Br were all well-tolerated in the
process, producing the desired pyranochromones in acceptable
To gain more insights into the competitive formation of pyr-
yields and regioselectivities. Benzocoumarins 1k and 1l were anocoumarin 3a and pyranochromone 4a, some additional ex-
suitable substrates as well. It is noted that Sm(OTf)3-catalyzed periments were conducted at lower temperature within shorter
annulation of 6-Cl hydroxycoumarin 1h with isoprene yielded time (12 h). As shown in Figure 1, pyranochromone 4a was the
pyranocoumarin 3h, rather than the desired pyranochromone main product below 70 °C, while pyranocoumarin 3a became
Eur. J. Org. Chem. 2019, 6510–6514
www.eurjoc.org
6512
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Scheme 3. Synthetic transformations of pyranocoumarin 3a.
predominant at higher temperature. With increasing the tem- double bond of A is attacked by C3 of 1 to furnish prenylated
perature (40–90 °C), the yield of pyranocoumarin 3a dramati- diketone/coumarin B, followed by protonation to generate cat-
cally increased from 1 % to 54 %. Meanwhile, the yield of ion C. A final intramolecular cyclization affords pyranochro-
pyranochromone 4a gradually increased (18 % to 37 %) from mone 4a or pyranocoumarin 3a. It is worthwhile to mention
40 to 70 °C and obviously decreased by further raising the tem- that intermediate C to 4a is a reversible pathway, while the
perature. These findings indicated pyranochromone 4a is a ki- route to 3a is irreversible, which can account for the conversion
netic product, whereas pyranocoumarin 3a is a thermodynamic of 4a to 3a in the presence of acid at high temperature.
product. Besides, pyranochromone 4a could be readily trans-
formed into pyranocoumarin 3a under the standard conditions
[Equation (1)].
Scheme 4. Proposed mechanism.
In conclusion, an acid-catalyzed regiodivergent annulation
Figure 1. Temperature effect on the product distributions. Reaction condi-
tions: 1a (0.2 mmol), 2 (0.6 mmol), 2,4-(NO2)2C6H3SO3H (10 mol-%), dioxane
(0.5 mL), T°C, 12 h. Yield was determined by HPLC using naphthalene as the
internal standard.
of biomass-derived 4-hydroxycoumarin and isoprene has been
developed.[21] The strong Brønsted acid [2,4-(NO2)2C6H3SO3H]
exclusively resulted in pyranocoumarins, whereas the selectivity
switched to pyranochromones in the presence of Sm(OTf)3. The
On the basis of these experimental results, a plausible mech- salient features of our protocol also include high atom-econ-
anism was proposed in Scheme 4. Isoprene 2 is initially proto- omy, wide substrate scope, easy scale-up, and good applica-
nated to form active isopentenyl cation A. The terminal C–C tions in natural product synthesis.
Eur. J. Org. Chem. 2019, 6510–6514
www.eurjoc.org
6513
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
K. Deckers, G. N. Hermann, L. Mertens, G. Raabe, D. Enders, Org. Lett.
2014, 16, 5188; i) N. Kausar, P. P. Ghosh, G. Pal, A. R. Das, RSC Adv. 2015,
5, 60199; j) S. Mahato, S. Santra, R. Chatterjee, G. V. Zyryanov, A. Hajra,
A. Majee, Green Chem. 2017, 19, 3282; k) S. Wang, J. Xiao, J. Li, H. Xiang,
C. Wang, X. Chen, R. G. Carter, H. Yang, Chem. Commun. 2017, 53, 4441;
l) W.-C. Hsieh, C.-H. Lin, Y.-J. Yang, D.-Y. Yang, RSC Adv. 2018, 8, 39162.
[11] a) G. Zhang, Y. Zhang, J. Yan, R. Chen, S. Wang, Y. Ma, R. Wang, J. Org.
Chem. 2012, 77, 878; b) S. Singh, A. Srivastava, S. M. Mobin, S. Samanta,
RSC Adv. 2015, 5, 5010; c) S. Biswas, A. Dagar, S. M. Mobin, S. Samanta,
Org. Biomol. Chem. 2016, 14, 1940; d) R. Gurubrahamam, B.-F. Gao, Y. M.
Chen, Y.-T. Chan, M.-K. Tsai, K. Chen, Org. Lett. 2016, 18, 3098; e) A. M.
Jadhav, S. K. Krishnammagari, J. T. Kim, Y. T. Jeong, Tetrahedron 2017, 73,
5163; f) J. Zhang, G. Yin, Y. Du, Z. Yang, Y. Li, L. Chen, J. Org. Chem. 2017,
82, 13594.
Acknowledgments
Financial support from Dalian Institute of Chemical Physics, the
National Natural Science Foundation of China (21702204,
21801239),
Liaoning
Revitalization
Talents
Program
(XLYC1807181), and the “Thousand Youth Talents Plan” is ac-
knowledged. Dedicated to the 70th anniversary of the Dalian
Institute of Chemical Physics, Chinese Academy of Sciences.
Keywords: Regiodivergent annulation · Brønsted acid ·
Lewis acids · Isoprene · 4-Hydroxycoumarin
[12] G. Appendino, G. Cravotto, G. B. Giovenzana, G. Palmisano, J. Nat. Prod.
1999, 62, 1627.
[13] a) B. M. Trost, Science 1991, 254, 1471; b) B. M. Trost, Acc. Chem. Res.
2002, 35, 695; c) T. Newhouse, P. S. Baran, R. W. Hoffmann, Chem. Soc.
Rev. 2009, 38, 3010; d) Y. Hayashi, Chem. Sci. 2016, 7, 866; e) C. J. Li,
Green Chem. 2016, 18, 1836.
[14] a) X. M. Lv, F. Wang, P. P. Zhou, L. D. Ye, W. P. Xie, H. M. Xu, H. W. Yu, Nat.
Commun. 2016, 7, 12851; b) Y. J. Zhou, E. J. Kerkhoven, J. Nielsen, Nat.
Energy 2018, 3, 925.
[15] A H3PO4-promoted annulation of 4-hydroxycoumarin with isoprene was
reported, but gave both pyranocoumarin and pyranochromone as main
products with 1:1 regioselectivity, see: V. K. Ahluwalia, K. K. Arora, I. Muk-
herjee, Heterocycles 1984, 22, 223.
[16] Our previous works on regioselectivity manipulation of reactions of iso-
prene: a) Y.-C. Hu, D.-W. Ji, C.-Y. Zhao, H. Zheng, Q.-A. Chen, Angew. Chem.
Int. Ed. 2019, 58, 5438; Angew. Chem. 2019, 131, 5492; b) D.-W. Ji, Y.-C.
Hu, H. Zheng, C.-Y. Zhao, Q.-A. Chen, V. M. Dong, Chem. Sci. 2019, 10,
6311.
[17] a) A. Kütt, I. Leito, I. Kaljurand, L. Soovali, V. M. Vlasov, L. M. Yagupolskii,
I. A. Koppel, J. Org. Chem. 2006, 71, 2829; b) A. Kütt, T. Rodima, J. Saame,
E. Raamat, V. Maemets, I. Kaljurand, I. A. Koppel, R. Y. Garlyauskayte, Y. L.
Yagupolskii, L. M. Yagupolskii, E. Bernhardt, H. Willner, I. Leito, J. Org.
Chem. 2011, 76, 391; c) W. Schutyser, S.-F. Koelewijn, M. Dusselier, S.
Van de Vyver, J. Thomas, F. Yu, M. J. Carbone, M. Smet, P. Van Puyvelde,
W. Dehaen, B. F. Sels, Green Chem. 2014, 16, 1999.
[1] a) H. M. P. Poumale, R. Hamm, Y. Zang, Y. Shiono, V. Kuete, Coumarins
and Related Compounds from the Medicinal Plants of Africa. In Medicinal
Plant Research in Africa, (Ed.: V. Kuete), Elsevier: Oxford, 2013; pp. 261; b)
S. D. Sarker, L. Nahar, Progress in the Chemistry of Naturally Occurring
Coumarins. In Progress in the Chemistry of Organic Natural Products (Eds.:
A. D. Kinghorn, H. Falk, S. Gibbons, J. i. Kobayashi), Springer International
Publishing: Cham, 2017; pp. 241.
[2] a) F. Bohlmann, C. Zdero, Phytochemistry 1977, 16, 1261; b) D. A. Mulhol-
land, S. E. Iourine, D. A. H. Taylor, F. M. Dean, Phytochemistry 1998, 47,
1641.
[3] A. M. Rayar, N. Lagarde, F. Martin, F. Blanchard, B. Liagre, C. Ferroud, J. F.
Zagury, M. Montes, M. S. I. Veitia, Eur. J. Med. Chem. 2018, 146, 577.
[4] M. M. Kady, L. Brimer, P. Furu, E. Lemmich, H. M. Nielsen, S. T. Thiilborg,
O. Thastrup, S. B. Christensen, Planta Med. 1992, 58, 334.
[5] a) G. Appendino, S. Tagliapietra, G. M. Nano, V. Picci, Phytochemistry
1988, 27, 944; b) M.-S. Louvet, G. Gault, S. Lefebvre, F. Popowycz, M.
Boulven, S. Besse, E. Benoit, V. Lattard, D. Grancher, Phytochemistry 2015,
118, 124.
[6] Y. H. Lin, X. L. Shen, Q. P. Yuan, Y. J. Yan, Nat. Commun. 2013, 4, 2603.
[7] F. G. Medina, J. G. Marrero, M. Macias-Alonso, M. C. Gonzalez, I. Cordova-
Guerrero, A. G. T. Garcia, S. Osegueda-Roblesa, Nat. Prod. Rep. 2015, 32,
1472.
[8] R. Y. Yang, D. Kizer, H. Wu, E. Volckova, X. S. Miao, S. M. Ali, M. Tandon,
R. E. Savage, T. C. Chan, M. A. Ashwell, Bioorg. Med. Chem. 2008, 16, 5635.
[9] a) G. Aridoss, K. K. Laali, Tetrahedron Lett. 2011, 52, 6859; b) Q. Ren, J.
Kang, M. Li, L. Yuan, R. Chen, L. Wang, Eur. J. Org. Chem. 2017, 2017,
5566; c) Y.-P. Han, X.-S. Li, M. Li, X.-Y. Zhu, Y.-M. Liang, Adv. Synth. Catal.
2018, 360, 2796.
[18] a) S. Ahmad, J. Nat. Prod. 1984, 47, 391; b) I. Butenschön, K. Moller, W.
Hansel, J. Med. Chem. 2001, 44, 1249.
[19] a) R. M. Bowman, M. F. Grundon, J. Chem. Soc., C 1966, 1084; b) M. Sekar,
K. J. R. Prasad, J. Nat. Prod. 1998, 61, 294.
[10] a) M. Shanmugasundaram, S. Manikandan, R. Raghunathan, Tetrahedron
2002, 58, 997; b) N. Halland, T. Hansen, K. A. Jorgensen, Angew. Chem.
Int. Ed. 2003, 42, 4955; Angew. Chem. 2003, 115, 5105; c) C.-N. Huang,
P.-Y. Kuo, C.-H. Lin, D.-Y. Yang, Tetrahedron 2007, 63, 10025; d) E. J. Jung,
B. H. Park, Y. R. Lee, Green Chem. 2010, 12, 2003; e) Y. Liu, J. Zhu, J. Qian,
B. Jiang, Z. Xu, J. Org. Chem. 2011, 76, 9096; f) M. S. Singh, G. C. Nandi,
S. Samai, Green Chem. 2012, 14, 447; g) G. Yin, T. Ren, Y. Rao, Y. Zhou, Z.
Li, W. Shu, A. Wu, J. Org. Chem. 2013, 78, 3132; h) D. Hack, P. Chauhan,
[20] a) Z. W. Chen, H. Y. Zeng, H. Gong, H. N. Wang, C. J. Li, Chem. Sci. 2015,
6, 4174; b) Z. H. Qiu, J. S. Li, C. J. Li, Chem. Sci. 2017, 8, 6954.
[21] The reactions of 4-hydroxycoumarin with other conjugated dienes in-
cluding 2,3-dimethylbutadiene, 1,3-cyclohexadiene, and 2-phenylbutadi-
ene were also explored under both standard conditions, and details
were given on page S7 of the supporting information.
Received: August 2, 2019
Eur. J. Org. Chem. 2019, 6510–6514
www.eurjoc.org
6514
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim