Unified total syntheses of (-)-medicarpin, (-)-sophoracarpan a, and (±)-kushecarpin a with some structural revisions

Article abstract of DOI:10.1002/anie.201408910

The total syntheses of medicarpin, sophoracarpan A, and kushecarpin A from a common intermediate are achieved by using ortho- and para-quinone methide chemistry. Additionally, the relative stereochemistry of sophoracarpan A and B have been reassigned.

Full text of DOI:10.1002/anie.201408910

Angewandte
Chemie
DOI: 10.1002/anie.201408910
Total Synthesis
Unified Total Syntheses of (À)-Medicarpin, (À)-Sophoracarpan A, and
(Æ)-Kushecarpin A with Some Structural Revisions**
Zhen-Gao Feng, Wen-Ju Bai, and Thomas R. R. Pettus*
Abstract: The total syntheses of medicarpin, sophoracar-
pan A, and kushecarpin A from a common intermediate are
achieved by using ortho- and para-quinone methide chemistry.
Additionally, the relative stereochemistry of sophoracarpan A
and B have been reassigned.
a desire to advance quinone methide (QM) methods, and the
recognition that a general enantioselective strategy for
pterocarpans preparation was needed.[1] We speculated that
sophoracarpan A (2) and kushecarpin A (3) were both
biosynthesized from medicarpin (1), whereas sophoracar-
pan B (5) and kushecarpin C (6) arose from maackiain (4).
However, the differing acetal stereochemistry of compounds
2 and 3 compared with compounds 5 and 6 was puzzling.
For our planned chemical synthesis, we imagined produc-
ing kushecarpin A (3) by an oxidative dearomatization of the
supposed diastereomer of sophoracarpan A (2), the phenolic
ketal 7. Compound 7 would arise from diastereoselective
replacement of the chiral auxiliary (OR*) in compound 8 with
a methoxy residue (Scheme 1). Medicarpin (1), on the other
P
terocarpans constitute the second largest family of plant-
derived isoflavonoids (Figure 1).[1,2] They all share a fifteen
carbon tetracyclic core comprised of a dihydrobenzopyran or
chromane conjoined in cis fashion with a dihydrobenzofuran
Figure 1. Some known and supposed pterocarpans.
(Figure 1). Various oxygen and carbon substituents decorate
the periphery resulting in a broad range of biological
activities.[2] Derivatives exhibiting strong potency toward
both Mycobacterium tuberculosis and Staphylococcus aureus
display unprotected phenols at their C3 and C9 carbon
atoms.[2] Moderate antibacterial activity has been reported for
(À)-medicarpin A (1),[2] in which one of these phenolic
residues is methylated. Whereas the biological activities of
the most recent family additions, sophoracarpan A (2)[3] and
kushecarpin A (3),[4] remain unvetted,[5] they have established
two new subclasses of pterocarpans; one in which the C6
carbon atom is part of an acetal, and another in which the
benzopyran aryl ring is reformulated as an enone that bears
a hydroxy substituent at C1a.[6]
Scheme 1. Strategy for kushecarpin A via 7, a supposed diastereomer
of 2.
hand, could arise from reduction of the chroman ketal 8. This
tetracycle would be built by an intramolecular cyclization of
the free phenol with p-QM found in species 9. This
intermediate would be formed by regioselective methylene
oxidation of the chromane 10 that emerged from the o-QM 11
combining with the enol ether 12. These two starting
components would both arise from benzaldehyde 13.
Our interest in the natural products 2 and 3 originates
from an appreciation of their tunable biological activities,
Sometime ago we had found satisfactory diastereoselec-
tivities in the cycloaddition reaction of enol ether A with
various o-QMs (inset, Scheme 2).[7] We thought that its E-
phenylated analogue 12 would perform similarly. We began
its construction starting from benzaldehyde 15, which was
available from compound 13 in two pots and 61% yield.[8]
Next, the phosphonium chloride salt 17 was constructed in
62% yield from the nonracemic alcohol 16 (99% ee).[9,10] This
salt was then combined with benzaldehyde 15 and sodium
[*] Z. Feng, W. Bai, Prof. T. R. R. Pettus
Department of Chemistry and Biochemistry
University of California Santa Barbara
Santa Barbara, CA, 93106-9510 (USA)
E-mail: pettus@chem.ucsb.edu
[**] T.R.R.P. is grateful for financial support from National Science
Foundation (CHE-0806356) for his o-QM work.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201408910.
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Scheme 3. Initial strategy. a) 18, Pb3O4, AcOH, benzene, reflux; b) 20,
Pd/C, H2, EtOH, RT; c) K2CO3, EtOAc. 46% yield from 18 to 22 over
three steps.
Scheme 2. Synthesis of 18. a) 16, CH2O, TMSCl; PPh3, benzene, reflux,
62% yield. b) 15, NaHMDS, 17, THF, À788C to RT, 12 h, 59% yield
from 16 (2:1, E/Z). c) 14, ether, 12, MeMgBr, À788C to RT, 65%;
d) 18, Pd/C, H2, EtOH, RT, 92% yield; e) 19, BnBr, NaH, DMF, À788C
to RT, 88% yield. TMSCl=trimethylsilyl chloride, Boc=tert-butoxycar-
bonyl, NaHMDS=sodium hexamethyldisilazide. 24% overall yield
from 13 to 18 via 19 in six steps.
À
C O acetal bond that had consequently thwarted oxonium
formation, so we changed the order of events.
The acetal 18, which lacked the additional dihydrofuran
ring, was found to undergo smooth reduction with boron
trifluoride etherate in triethyl silane to produce the corre-
sponding chromane 23 in 60% yield (Scheme 4).[16] HPLC
hexamethyldisilazide to afford the enol ether 12 in 59% yield
as a 2:1 mixture of E and Z isomers.[11] Although our many
attempts to improve this ratio were futile,[12] chromatography
gave the pure E isomer in a 37% overall yield from the
starting alcohol 16. On the other hand, compound 14, which
served as the precursor to o-QM 11, was synthesized in three
steps from the identical compound 13.[7b]
Pure compounds 12 and 14 were then combined with the
assistance of methylmagnesium bromide (À78 to 08C) so as to
release the o-QM 11 in a slow controlled manner. The ensuing
cycloaddition produced the chromane ketal 18 in 65% yield
with a 10:1 diastereomeric ratio (d.r.) as determined by
1H NMR spectroscopy. Whereas the two diastereomers
proved inseparable, hydrogenolysis of the mixture afforded
the bis-phenol 19 that crystallized as a single diastereomer
(75% yield). Its X-ray analysis confirmed the relative and
absolute stereochemistry.[13] Compound 19 was then rebenzy-
lated to return the pure benzyl ether derivative 18 as a single
diastereomer.
Scheme 4. Synthesis of (À)-medicarpin (1). a) 18, BF3·Et2O, Et3SiH,
DCM, 60% yield, 99% ee; b) 23, Pb3O4, AcOH, benzene, reflux; c) Pd/
C, H2, EtOH, RT; K2CO3, EtOAc, 51% yield from 23; d) 18, HCl (1m),
acetone, reflux, 18 h; e) SOCl2, DMF, DCM, 08C; f) Ag2CO3, 4 ꢀ MS,
MeOH, DCM, RT, 12 h; 64% yield three steps from 18 to 25.
After some experimentation, we observed that red lead
(Pb3O4) in acetic acid affected a selective benzylic oxidation
of the methylene within the chromane to produce the desired
benzylic acetate 20 (1:1 d.r., Scheme 3).[14] Subsequent hydro-
genolysis of the two benzyl ethers proceeded in a nearly
quantitative yield over Pd/C in ethanol to give the crude bis-
phenolic acetate 21. Deprotonation with potassium carbonate
in ethyl acetate continued to the presumed p-QM intermedi-
ate that underwent further cyclization to yield the cis-fused
tetracycle 22 as a single diastereomer in 46% yield from the
starting chromane 18.[15] However, we were unable to convert
the acetal 22 into its corresponding methyl acetal 7. In
addition, we were unable to reduce the acetal 22 to afford
medicarpin (1). All acidic/reductive conditions that we tested
led to decomposition of the starting ketal 22. We attributed
these problems to the pseudo-equatorial positioning of the
comparison with a racemic standard, previously prepared
from the corresponding vinyl methyl ether analogue of 12 in
early investigations, showed that compound 23 possessed
99% enantiomeric excess (ee). Further debenzylation and
oxidative cyclization, as previously described, afforded (À)-
medicarpin (1) as a single diastereomer in a 51% yield.
Bolstered by evidence of formation of the desired oxonium
intermediate B, the acetal 18 was next submitted to various
trans-acetalization reactions involving methanol and acid in
the hope that the simplified methyl acetal 25 might arise.
Under the thermodynamic conditions, the methyl acetal
25 arose albeit as a 1:1 mixture of diastereomers. We thus
explored stepwise procedures amenable to acetal formation
in a second kinetically controlled step. Treatment of the acetal
18 with 1m aqueous hydrochloric acid resulted in the epimeric
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
hemiketal 24 along with the recoverable chiral alcohol 16. The
epimeric hemiketal 24 was then converted into its corre-
sponding epimeric chloro derivative by treatment with thionyl
chloride. Its subsequent exposure to AgI in the presence of
methanol resulted in the desired acetal 25 in 84% yield; an
Running low on material, we investigated the conversion
of the sophoracarpan A (7), used as the racemic standard, to
kushecarpin A (1). All of the typical oxidants for resorcinol
dearomatization including various hypervalent iodine
reagents, Pb(OAc)4, and even an oxone procedure failed in
our hands to afford any of the desired cyclohexadienone.[20]
Broadening our scope, we examined Doyle’s dirhodium
caprolactamate (Rh2(cap)4) catalyzed phenol dearomatiza-
tion procedure.[21] Our expectations were lifted upon finding
that the cyclohexadienone adduct (27, Scheme 6) had indeed
1
anti/syn ratio of > 10:1 was obtained, as shown by H NMR
spectroscopy.[17] We speculate that the addition of the
methanol proceeds unimpeded, opposite the axially disposed
aryl substituent, as there are no 1,3-diaxial interactions to
dissuade addition. However, the ee of the diastereomerically
pure compound 25 in large batches was found to have eroded
to just 60%; 88% ee for smaller batches. We thought that this
unfortunate circumstance was due to partial formation of the
glycal-C from the oxonium-B, whereupon protonation
afforded scalemic D, resulting in compound 26 with reduced
optical purity. To circumvent this problem, we investigated
controlled methods that might avoid the hemiketal 24.[18]
After considerable experimentation we found that with
the sequential addition of boron trifluoride etherate and the
appropriate nucleophile, the acetal 18 could be converted into
the iodide 26 or the thioether 27, so as to avoid the hemiketal
24 as well as enantiomeric erosion (Scheme 5). However, the
Scheme 6. Synthesis of kushecarpin A (3). a) PhI(OAc)2, tBuOOH,
DCE, RT, 48% yield; b) Pd/C, H4N+[HCO2]À, EtOH, MW, 1208C,
10 min,18% yield. DCE=1,2-dichloroethane, MW=microwave.
formed in 25% yield. After considering its mechanism, we
revisited the hypervalent iodine procedure using tert-butyl
hydroperoxide as the nucleophile and found that the adduct
27 had formed in a 48% yield in a 3a:1b mixture of
diastereomers about the C1a hydroxy residue.[22] Selective
reduction of the enone within the dienone 27 and cleavage of
the peroxy bond remained. Reductions of similar resorcinol-
derived dienones are known to be problematic leading to
reductive rearomatization so as to return the pre-dearomat-
ized material.[23] We were therefore surprised to discover that
microwave irradiation of enone 22, together with ammonium
formate and palladium on carbon offered (Æ)-kushecarpin A
(3) in 18% isolated yield.[24] Spectroscopic comparison of
coupling constants and resonances between synthetic and
natural kushecarpin A (3) showed them to be identical.
In summary, we have developed a reliable and unified
strategy that is able to assemble nearly all of the pterocarpans
in an enantioselective manner. We completed the first
enantioselective total syntheses of (À)-medicarpin (1) and
(À)-sophoracarpan A (7) in nine steps (4% overall yield) and
ten steps (5% overall yield), respectively, from the chiral
alcohol 16, as well as (Æ)-kushecarpin A (3) in 11 steps and
1.1% overall yield from the benzaldehyde 13. All compounds
were synthesized in a divergent/convergent manner with their
two halves arising from the same commercial benzaldehyde
13. In addition, the relative stereochemistries for sophora-
carpan A (2) and B (5) have both been reassigned. Our effort
touts the utility of ortho-quinone methide Diels–Alder
reactions to form benzopyran rings in a diastereoselective
manner, an oxidative cyclization likely involving a para-
quinone as well as a new IIII/tBuOOH oxidative dearomati-
zation procedure.
Scheme 5. Synthesis of (À)-sophoracarpan A (7). a) 18, TBAI, BF3
Et2O, DCM, 35% yield of 26. b) 18, PhSH, BF3 Et2O, DCM, 85% yield
of 27. c) 27, Hg(TFA)2, DTBMP, 4 ꢀ MS, MeOH, DCM, RT, 84% yield.
d) 25, Pb3O4, AcOH, benzene, reflux; e) Pd/C, H2, EtOH, RT; K2CO3,
EtOAc, 50% yield from 25. TBAI=tetrabutyl ammonium iodide,
TFA=trifluoroacetic acid, DTBMP=2,6-di-tert-butyl-4-methylpyridine,
MS=molecular sieve.
iodide proved quite fragile and ineffective in providing the
acetal 25 using AgI. In contrast, the phenyl thioether 27 was
smoothly converted to the methyl acetal 25 in 85% yield (9:1
d.r.) upon exposure to mercuric trifluoride acetate and
methanol. Yields from reactions employing mercuric acetate
were substantially lower (30%). Debenzylation and oxidative
cyclization of compound 25 afforded the tetracycle 7 in 50%
overall yield and 93% ee. Comparison of the proton and
carbon NMR spectra of synthetic 7 to its supposed diaste-
reomer 2, claimed as natural sophoracarpan A (2), showed
them to be identical. The crystal structure of compound 7[19]
and comparative nOe study further confirmed that sopho-
racarpan A (7) had been misassigned upon its isolation as
compound 2.[3] As similarly unsound nOe arguments were
used for the original assignment of sophoracarpan B (5), we
speculate that its stereochemical assignment is incorrect and it
should be revised to be that shown (inset, Scheme 5).
Received: September 8, 2014
Published online: && &&, &&&&
Keywords: benzopyran acetal · diastereoselective diels-alder ·
.
oxidative dearomatization · pterocarpan · quinone methide
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
[16] W.-J. Bai, J. C. Green, T. R. R. Pettus, J. Org. Chem. 2012, 77,
379 – 387.
[17] T. V. RajanBabu, T. A. Ayers, G. A. Halliday, K. K. You, J. C.
Calabrese, J. Org. Chem. 1997, 62, 6012 – 6028.
[1] A. Goel, A. Kumar, A. Raghuvanshi, Chem. Rev. 2013, 113,
1614 – 1640.
[2] L. Jimꢀnez-Gonzꢁlez, M. ꢂlvarez-Corral, M. MuÇoz-Dorado, I.
Rodrꢃguez-Garcꢃa, Phytochem. Rev. 2007, 7, 125 – 154.
[3] a) T. Kinoshita, K. Ichinose, C. Takahashi, U. Sankawa, Chem.
Pharm. Bull. 1986, 34, 3067 – 3070; b) T. Kinoshita, K. Ichinose,
C. Takahashi, F.-C. Ho, J.-B. Wu, U. Sankawa, Chem. Pharm.
Bull. 1990, 38, 2756 – 2759.
[18] a) R. J. Ferrier, R. W. Hay, N. Vethaviyasar, Carbohydr. Res.
1973, 27, 55 – 61; b) J. D. Bryant, G. E. Keyser, J. R. Barrio, J.
Org. Chem. 1979, 44, 3733 – 3734; c) G. A. Olah, A. Husain, S. C.
Narang, Synthesis 1983, 1983, 896 – 897; d) D. Kahne, S. Walker,
Y. Cheng, D. Van Engen, J. Am. Chem. Soc. 1989, 111, 6881 –
6882; e) D. Crich, S. Sun, J. Am. Chem. Soc. 1998, 120, 435 – 436;
f) X. Ding, F. Kong, J. Carbohydr. Chem. 1998, 17, 915 – 922; g) F.
Pfrengle, D. Lentz, H.-U. Reißig, Angew. Chem. Int. Ed. 2009,
48, 3165 – 3169; Angew. Chem. 2009, 121, 3211 – 3215; h) S.-R.
Lu, Y.-H. Lai, J.-H. Chen, C.-Y. Liu, K.-K. T. Mong, Angew.
Chem. Int. Ed. 2011, 50, 7315 – 7320; Angew. Chem. 2011, 123,
7453 – 7458.
[4] M. Kuroyanagi, T. Arakawa, Y. Hirayama, T. Hayashi, J. Nat.
Prod. 1999, 62, 1595 – 1599.
[5] a) M. Nisar, W. A. Kaleem, I. Khan, A. Adhikari, N. Khan, M. R.
Shah, I. A. Khan, M. Qayum, Samiullah, M. Ismail, A. Aman,
Fitoterapia 2011, 82, 1008 – 1011; b) I. Oh, W.-Y. Yang, S.-C.
Chung, T.-Y. Kim, K.-B. Oh, J. Shin, Arch. Pharm. Res. 2011, 34,
217 – 222.
[6] a) S. Soby, S. Caldera, R. Bates, H. Vanetten, Phytochemistry
1996, 41, 759 – 765; b) A. F. Barrero, E. Cabrera, I. R. Garcia,
Phytochemistry 1998, 48, 187 – 190.
[7] a) C. Selenski, T. R. R. Pettus, J. Org. Chem. 2004, 69, 9196 –
9203; b) Y. Huang, T. R. R. Pettus, Synlett 2008, 1353 – 1356.
[8] a) J. Tummatorn, P. Khorphueng, A. Petsom, N. Muangsin, N.
Chaichit, S. Roengsumran, Tetrahedron 2007, 63, 11878 – 11885;
b) W. D. Vaccaro, R. Sher, H. R. Davis, Jr., Bioorg. Med. Chem.
1998, 6, 1429 – 1437.
[9] S. B. King, K. B. Sharpless, Tetrahedron Lett. 1994, 35, 5611 –
5612.
[10] a) C.-V. T. Vo, T. A. Mitchell, J. W. Bode, J. Am. Chem. Soc.
2011, 133, 14082 – 14089; b) S. G. Pyne, M. J. Hensel, P. L. Fuchs,
J. Am. Chem. Soc. 1982, 104, 5719 – 5728.
[19] CCDC 1022432 (7) contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via www.
ccdc.cam.ac.uk/data_request/cif.
[20] a) R. W. Van De Water, C. Hoarau, T. R. R. Pettus, Tetrahedron
Lett. 2003, 44, 5109 – 5113; b) A. Pelter, R. S. Ward, Tetrahedron
2001, 57, 273 – 282; c) R. Tello-Aburto, A. M. Harned, Org. Lett.
2009, 11, 3998 – 4000; d) P. J. Krawczuk, N. Schçne, P. S. Baran,
Org. Lett. 2009, 11, 4774 – 4776; e) J. L. Frie, C. S. Jeffrey, E. J.
Sorensen, Org. Lett. 2009, 11, 5394 – 5397; f) A. Bꢀrubꢀ, I. Drutu,
J. L. Wood, Org. Lett. 2006, 8, 5421 – 5424; g) M. C. CarreÇo, M.
Gonzꢁlez-Lꢄpez, A. Urbano, Angew. Chem. Int. Ed. 2006, 45,
2737 – 2741; Angew. Chem. 2006, 118, 2803 – 2807.
[21] M. O. Ratnikov, L. E. Farkas, E. C. McLaughlin, G. Chiou, H.
Choi, S. H. El-Khalafy, M. P. Doyle, J. Org. Chem. 2011, 76,
2585 – 2593.
[11] Y. Takashima, Y. Kaneko, Y. Kobayashi, Tetrahedron 2010, 66,
197 – 207.
[12] M. Schlosser, K. F. Christmann, Angew. Chem. Int. Ed. Engl.
1966, 5, 126 – 126; Angew. Chem. 1966, 78, 115 – 115.
[13] CCDC 946677 (19) contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via www.
ccdc.cam.ac.uk/data_request/cif.
[14] T. Kikuchi, M. Nishimura, A. Hoshino, Y. Morita, N. Saito, T.
Honda, Heterocycles 2003, 60, 1469 – 1475.
[15] R. S. Khupse, P. W. Erhardt, Org. Lett. 2008, 10, 5007 – 5010.
[22] We provided this dearomatization method to Prof. Alison J.
Frontier for their total synthesis of Tetrapetalone A: P. N.
Carlsen, T. J. Mann, A. H. Hoveyda, A. J. Frontier, Angew.
Chem. Int. Ed. 2014, 53, 9334 – 9338; Angew. Chem. 2014, 126,
9488 – 9492.
[23] T. A. Wenderski, C. Hoarau, L. Mejorado, T. R. R. Pettus,
Tetrahedron 2010, 66, 5873 – 5883.
[24] C. Ovens, N. G. Martin, D. J. Procter, Org. Lett. 2008, 10, 1441 –
1444.
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Total Synthesis
Z. Feng, W. Bai,
T. R. R. Pettus*
&&&& — &&&&
Unified Total Syntheses of (À)-
Medicarpin, (À)-Sophoracarpan A, and
(Æ)-Kushecarpin A with Some Structural
Revisions
Ortho- and para-quinone methide
chemistry is used to accomplish the total
syntheses of medicarpin, sophoracar-
pan A, and kushecarpin A from
a common intermediate. Additionally, the
relative stereochemistry of sophoracar-
pan A and B have been reassigned. QM=
quinone methide.
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!