Synthesis of the furo[2,3-b]chromene ring system of hyperaspindols a and B

Article abstract of DOI:10.3762/bjoc.11.29

The synthesis of the unique furo[2,3-b]chromene ring system found in hyperaspidinols A and B, acylphloroglucinols from Hypericum chinense has been achieved in twelve steps. By comparison of the NMR spectra of the synthesized compounds with those of the natural products, a relative stereochemistry is suggested, especially that of the ketal carbon.

Full text of DOI:10.3762/bjoc.11.29

Synthesis of the furo[2,3-b]chromene ring system of
hyperaspindols A and B
Danielle L. Paterson and David Barker*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2015, 11, 265–270.
School of Chemical Sciences, University of Auckland, 23 Symonds St,
Auckland, New Zealand
doi:10.3762/bjoc.11.29
Received: 22 December 2014
Accepted: 04 February 2015
Published: 17 February 2015
Email:
David Barker* - d.barker@auckland.ac.nz
* Corresponding author
Associate Editor: A. Kirschning
Keywords:
© 2015 Paterson and Barker; licensee Beilstein-Institut.
License and terms: see end of document.
acylphloroglucinols; bicyclisation; furo[2,3-b]chromene; fused ketal
Abstract
The synthesis of the unique furo[2,3-b]chromene ring system found in hyperaspidinols A and B, acylphloroglucinols from Hyper-
icum chinense has been achieved in twelve steps. By comparison of the NMR spectra of the synthesized compounds with those of
the natural products, a relative stereochemistry is suggested, especially that of the ketal carbon.
Introduction
Two novel racemic acylphoroglucinols, hyperaspidinols A (1) ities, as well as inhibitory activity on thromboxane A2 and
and B (2) (Figure 1) were recently isolated from the leaves of leukotriene D4 [4]. Acylphloroglucinols are known to act as
Hypericum chinense, a member of the St John’s Wort plants, anti-oxidants, by reducing hydroperoxides and hydrogen
which contains 490 flowering plants [1]. These plants have been peroxide, thereby suppressing the formation of the reactive
used medicinally for treating illnesses such as hepatitis and species [9].
depression, and as topical antimicrobials for wounds and snake
bites [2-5]. There is great interest in secondary metabolites The hyperaspidinols 1 and 2 both possess a highly function-
produced by plants from the Hypericum genus due to the bioac- alised furo[2,3-b]chromene ring system (Figure 2, highlighted
tivity of many compounds that have been isolated from this in blue) and differ only with the nature of the ketone side-chain.
source. A wide variety of compounds have been isolated from Furo[2,3-b]chromenes have not been reported in any other
H. chinense including prenylated acylphloroglucinols such as natural products to date with the closest related system being
chinesins I (3) and II (4), xanthones, flavonoids, terpenoids, the chromeno[2,3-b]chromenes. Compound which contain this
naphthodianthrones, and norlignans, such as hyperione A (5) motif, such as albanol A and australisine A, display potent
and B (6) [1,6-8]. Chinesins I (3) and II (4) are acylphloroglu- bioactive properties including hypotensive, anticancer, antimi-
cinol derivatives which possess antibacterial and antiviral activ- crobial and antimalarial activity [10-12].
265

Beilstein J. Org. Chem. 2015, 11, 265–270.
compounds. The relative stereochemistry of the hyperaspidi-
nols was determined using a variety of NMR spectroscopic
techniques. Whilst the isolation paper reported a weak ROESY
interaction between H-8 and H-10, suggesting a cis configur-
ation between these protons, the relationship of the aryl
substituents could not be determined (Figure 2) [1]. Our present
aim was therefore to prepare furo[2,3-b]chromene 7 which
contains the unique ring system and aryl subsitutents found in
hyperaspidinols 1 and 2 and hopefully use it to determine the
complete relative stereochemistry of the natural products.
Results and Discussion
The proposed route to furo[2,3-b]chromene 7 was based around
the preparation of benzylic ketone 8 which was hoped to under
acidic conditions would undergo cyclisation to give 7
(Figure 2). We have previously reported the formation of a
number of diaryl tetrahydrofuran lignans under acidic or
buffered conditions utilising the high reactivity of electron-rich
benzylic alcohols to assist in the rapid and high yielding forma-
tion of the tetrahydrofuran ring [13-18]. The two methylene-
dioxyphenyl groups in ketone 8 would be added by sequential
addition of aryllithiates to both the carbonyl and aldehyde
groups, derived from the terminal alkene, of a homoallylic
carboxylic acid derivative 9.
Figure 1: Hyperaspidinols A (1) and B (2) and other compounds 3-6
from Hypericum chinense.
Our route to ketone 8 began from salicylaldehyde (10) which
underwent a Horner–Wadsworth–Emmons reaction with
triethylphosphonoacetate to give ester 11 [19] in 94% yield
(Scheme 1). Hydrogenation of the alkene followed by protec-
tion of the phenol gave benzyl ether 12 in 88% yield over two
steps. Initially the preparation of alkene 13 through the allyla-
tion of ester 12 was attempted, however keto-ester 14 resulting
from an additional Claisen condensation was the only isolated
product.
It was therefore decided to functionalise the ester moiety first.
This was achieved by direct conversion of ester 12 into Weinreb
amide 15 in 81% yield, followed by addition of lithiate 16
(formed from 1-bromo-3,4-methylenedioxybenzene and t-butyl-
lithium) to give ketone 17 in 84% yield (Scheme 2). Allylation
of ketone 17 proved problematic with the use of strong bases
such as LDA and LiHMDS, giving none of the desired product.
However, the use of allyl bromide and NaH in the presence of
TBAI, in THF at reflux, gave the desired alkene 18 in 95%
yield. Protection of the ketone in 18 as the cyclic ketal 19 was
achieved in 44% yield using excess ethylene glycol and pTSA
and despite attempting a variety of alternate conditions [20-23]
this yield could not be improved without degradation of both
Figure 2: Hyperaspidinol A (1), target compound 7 and proposed
precursors.
The unique ring system found in the hyperaspidinols combined the starting material 18 and product 19. Dihydroxylation of 19,
with the potent biological activities of similar structures led us followed by oxidative cleavage of the resultant diol gave alde-
to explore routes to the carboheterocyclic skeleton of these hyde 20 in 80% yield over two steps. Addition of lithiate 16 to
266

Beilstein J. Org. Chem. 2015, 11, 265–270.
Scheme 1: Reagents and conditions: (i) triethylphosphonoacetate, DBU, THF, 48 h, 94%; (ii) H2, 10% Pd/C, EtOAc, 3 h, quant; (iii) BnBr, K2CO3,
DMF, 3 h, 88%; (iv) LDA, THF, −78 °C, then allyl bromide, 24 h, 14%.
Scheme 2: Reagents and conditions: (i) H3C(CH3O)NH·HCl, n-BuLi, THF, −78 °C, 4 h, 81%; (ii) 1-bromo-3,4-methylenedioxybenzene, t-BuLi, THF,
−78 °C, 3 h, 84%; (iii) NaH, allyl bromide, TBAI, THF, 65 °C, 20 h, 95%; (iv) ethylene glycol, pTSA, toluene, reflux, 16 h, 44%; (v) OsO4, NMO,
t-BuOH, H2O, THF, 65 h, 80%; (vi) NaIO4, MeOH, H2O, 3 h, quant; (vii) 1-bromo-3,4-methylenedioxybenzene, t-BuLi, THF, −78 °C, 19 h, 87% 1:1 dia-
stereoisomers; (viii) H2, 10% Pd/C, MeOH, 4 h, quant; (ix) 1:1 2 M HCl (aq):THF, 22 h, 7a 52%, 7b 32%.
aldehyde 20 gave alcohol 21 in 87% yield as an inseparable 1:1 ketal protecting group, giving ketone 8, which under the acidic
mixture of diastereoisomers. Hydrogenolysis of the benzyl ether condition immediately cyclized giving furo[2,3-b]chromenes 7a
in 21 gave ketal-diol 22 in quantitative yield. Finally, stirring of and 7b in a 1.6:1 ratio, in an overall 84% yield. Separation of
ketal 22 in a 1:1 2 M HCl (aq):THF resulted in removal of the the diastereoisomers 7a and 7b was achieved using column
267

Beilstein J. Org. Chem. 2015, 11, 265–270.
chromatography and allowed characterisation of the individual pectively, showing an anti-relationship between these protons
isomers.
[25]. In isomer 7b the coupling constant between H-8 and H-9b
(9.5 Hz) represent a ca. 170° dihedral angle between these
Comparison of the 1H and 13C NMR data of isomers 7a and 7b protons whilst the coupling constant between H-9b and H-10
showed strong similarities between 7a and the reported data [1] (4.5 Hz) represent a ca. 40° dihedral angle or syn relationship
for hyperaspidinols A (1) and B (2) whilst isomer 7b showed between these two protons (Figure 5) [26]. Overall this repre-
clear differences in both the chemical shift and multiplicities in sents an anti-relationship between H-8 and H-10 confirming 7b
the furo[2,3-b]chromene rings (see Supporting Information to be the C-10 epimer of 7a.
File 1 for a complete table of NMR data).
Extensive use of 2D NMR techniques, in particular a NOESY,
allowed the complete relative stereochemistry of both 7a and 7b
to be determined (Figure 3). In isomer 7a NOESY correlations
between H-8 and H-10 showed the syn relationship between
these two protons. An additional NOESY correlation between
H-8 and H-2′ showed the syn relationship between these two
groups and thus the trans relationship between the two methyl-
enedioxyphenyl groups at C-10 and C-11.
Figure 4: 3D representation of 7a.
Figure 3: NOESY correlations in isomers 7a and 7b.
The NOESY spectrum of 7b exhibited a NOESY correlation
between H-8 and H-2′ showed the syn relationship between
these two groups but no correlation between H-8 and H-10 was
observed, suggesting 7b to be the C-10 epimer of 7a. A correla-
tion of H-10 to the H-9b, which was on the opposite face to H-8
further suggested this. Further analysis of the conformations of
7a and 7b was achieved by examination of the coupling
constants in the furo[2,3-b]chromene ring. The use of coupling
constants to determine stereochemistry in five-membered rings
is often extremely difficult, due to the large number of confor-
mations available [24-26]. However, with the relatively fixed
geometry for the five membered ring in 7a and 7b due to the
fused chroman ring and the quarternary centre at C-11, analysis
Figure 5: 3D representation of 7b.
of the coupling constants was considered viable. In isomer 7a
the coupling constant between H-9a and both H-8 and H-10 was
6.5 Hz, which corresponds to a ca. 50° dihedral angle between The ratio of diastereoisomers in ketal 22 was 1:1, however in
the protons, placing all three on the same face of the five the cyclised products 7 the ratio of diastereoisomers 7a to 7b
membered ring (Figure 4). Additionally H-9b had coupling was 1.6:1. This suggests partial isomerisation of the stereo-
constants of 10.5 Hz and 12.6 Hz between H-8 and H-10, res- chemistry at C-10 during the formation of the furo[2,3-
268

Beilstein J. Org. Chem. 2015, 11, 265–270.
b]chromene. Indeed, one possible mechanism for the formation provides clues to the relative stereochemistry of all substituents
of isomers 7a and 7b is initial protonation of ketone 8, attack of on the natural products. The route developed allows easy intro-
phenol on the activated carbonyl resulting in formation of duction of alternate aryl substituents at the C-10 and C-11 posi-
chroman-hemiketal 23, where the large C-11 aryl group and tions, whilst replacing salicylaldehyde 10 with a more function-
arylethyl substituent at C-8 adopt an anti relationship alised aldehyde could allow for the preparation of a number of
(Figure 6). Loss of stereochemistry at C-10 is most likely due to highly functionalised furo[2,3-b]chromenes, including the
the formation of a quinone methide, such as 24, which upon natural products 1 and 2.
attack by the C-11 alcohol gives a mixture of isomers with the
isomer 7a being favoured over the more sterically congested 7b.
Supporting Information
We have noticed a similar process in the synthesis of other THF
lignans [15,16]. To test this hypothesis we stirred a sample of
Supporting Information File 1
isomer 7b in 1:1 2 M HCl (aq)/THF and found that after
24 hours obtained a mixture of 7a to 7b with isomer 7a being
predominant. This highlights the reversibility of the ring closure
and preferred formation of isomer 7a.
Experimental procedures, characterisation data of new
compounds and NMR tables of 7a and 7b.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-11-29-S1.pdf]
Supporting Information File 2
1H/13C NMR spectra.
Conclusion
In conclusion, the first synthesis of the furo[2,3-b]chromene
ring system found in hyperaspidinols A (1) and B (2) has been
achieved. Analysis of the NMR of synthetic furo[2,3-
b]chromenes 7a and 7b and comparison to the data of 1 and 2
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-11-29-S2.pdf]
Figure 6: Possible mechanism for the formation of furo[2,3-b]chromenes 7a and 7b.
269

Beilstein J. Org. Chem. 2015, 11, 265–270.
24.Kilpatrick, J. E.; Pitzer, K. S.; Spitzer, R. J. Am. Chem. Soc. 1947, 69,
2483–2488. doi:10.1021/ja01202a069
Acknowledgements
We would like to acknowledge the University of Auckland for
25.Wu, A.; Cremer, D. Int. J. Mol. Sci. 2003, 4, 158–192.
doi:10.3390/i4040158
funding this project.
26.Napolitano, J. G.; Gavin, J. A.; Garcia, C.; Norte, M.; Fernández, J. J.;
Hernández Daranas, A. Chem. – Eur. J. 2011, 17, 6338–6347.
doi:10.1002/chem.201100412
References
1. Wang, W.; Zeng, Y. H.; Osman, K.; Shinde, K.; Rahman, M.;
Gibbons, S.; Mu, Q. J. Nat. Prod. 2010, 73, 1815–1820.
doi:10.1021/np1004483
2. Avato, P. A survey on the hypericum genus: Secondary metabolites
and bioactivity. In Studies in Natural Products Chemistry;
Atta-ur-Rahman, Ed.; Elsevier B.V.: Oxford, 2005; Vol. 30, pp 603–634.
doi:10.1016/S1572-5995(05)80043-2
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
3. Tanaka, N.; Kashiwada, Y.; Kim, S. Y.; Hashida, W.; Sekiya, M.;
Ikeshiro, Y.; Takaishi, Y. J. Nat. Prod. 2009, 72, 1447–1452.
doi:10.1021/np900109y
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
4. Aramaki, Y.; Chiba, K.; Tada, M. Phytochemistry 1995, 38, 1419–1421.
doi:10.1016/0031-9422(94)00862-N
5. Abe, S.; Tanaka, N.; Kobayashi, J. J. Nat. Prod. 2012, 75, 484–488.
doi:10.1021/np200741x
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
6. Tanaka, N.; Kashiwada, Y.; Kim, S.-Y.; Sekiya, M.; Ikeshiro, Y.;
Takaishi, Y. Phytochemistry 2009, 70, 1456–1461.
doi:10.1016/j.phytochem.2009.08.015
(http://www.beilstein-journals.org/bjoc)
7. Tanaka, N.; Abe, S.; Kobayashi, J. Tetrahedron Lett. 2012, 53,
1507–1510. doi:10.1016/j.tetlet.2012.01.052
The definitive version of this article is the electronic one
which can be found at:
8. Tanaka, N.; Abe, S.; Hasegawa, K.; Shiro, M.; Kobayashi, J. Org. Lett.
2011, 13, 5488–5491. doi:10.1021/ol2021548
doi:10.3762/bjoc.11.29
9. Verotta, L. Phytochem. Rev. 2002, 1, 389–407.
doi:10.1023/A:1026069624278
10.Osyanin, V. A.; Osipov, D. V.; Borisov, D. D.; Klimochkin, Yu. N.
Chem. Heterocycl. Compd. 2013, 49, 683–688.
doi:10.1007/s10593-013-1298-x
11.Kikuchi, T.; Nihei, M.; Nagai, H.; Fukushi, H.; Tabata, K.; Suzuki, T.;
Akihisa, T. Chem. Pharm. Bull. 2010, 58, 568–571.
doi:10.1248/cpb.58.568
12.Zhang, Q.-J.; Tang, Y.-B.; Chen, R.-Y.; Yu, D.-Q. Chem. Biodiversity
2007, 4, 1533–1540. doi:10.1002/cbdv.200790133
13.Dickson, B. D.; Dittrich, N.; Barker, D. Tetrahedron Lett. 2012, 53,
4464–4468. doi:10.1016/j.tetlet.2012.06.088
14.Barker, D.; Dickson, B. D.; Dittrich, N.; Rye, C. E. Pure Appl. Chem.
2012, 84, 1557–1565. doi:10.1351/PAC-CON-11-09-27
15.Rye, C. E.; Barker, D. J. Org. Chem. 2011, 76, 6636–6648.
doi:10.1021/jo200968f
16.Rye, C. E.; Barker, D. Synlett 2009, 3315–3319.
doi:10.1055/s-0029-1218363
17.Pilkington, L. I.; Barker, D. J. Org. Chem. 2012, 77, 8156–8166.
doi:10.1021/jo3015006
18.Duhamel, N.; Rye, C. E.; Barker, D. Asian J. Org. Chem. 2013, 2,
491–493. doi:10.1002/ajoc.201300086
19.Barancelli, D. A.; Salles, A. G., Jr.; Taylor, J. G.; Correia, C. R. D.
Org. Lett. 2012, 14, 6036–6039. doi:10.1021/ol302923f
20.Liu, G.; Lu, X. J. Am. Chem. Soc. 2006, 128, 16504–16505.
doi:10.1021/ja0672425
21.Sato, M.; Uchimaru, F. Chem. Pharm. Bull. 1981, 29, 3134–3144.
doi:10.1248/cpb.29.3134
22.Dauben, W. G.; Gerdes, J. M.; Look, G. C. J. Org. Chem. 1986, 51,
4964–4970. doi:10.1021/jo00375a039
23.Firouzabadi, H.; Iranpoor, N.; Karimi, B. Synlett 1999, 321–323.
doi:10.1055/s-1999-2605
270