The bidirectional total synthesis of sampsonione P and hyperibone i

Article abstract of DOI:10.1002/anie.201306256

Make it simple: The separation of framework-decorating from framework-constructing steps facilitated the selective introduction of one prenyl group next to allyl groups. The selective epoxidation of the more electron-rich prenyl group led to the efficient formation of the tetrahydrofuran moiety in the title compounds. Spectroscopic analysis of hyperiboneI and comparison with literature data led to a revision of the original structure. Copyright

Full text of DOI:10.1002/anie.201306256

Angewandte
Chemie
DOI: 10.1002/anie.201306256
Natural Product Synthesis
The Bidirectional Total Synthesis of Sampsonione P and
Hyperibone I**
Katharina Lindermayr and Bernd Plietker*
Polycyclic polyprenylated acylphloroglucinols (PPAPs) are
a class of acylphloroglucinol-derived natural products that
currently consist of more than 120 members.[1–4] The common
feature of these natural products is a densely substituted
[3.3.1]bicyclononatrione core that possesses only three hydro-
gen substituents. The lack of hydrogen atoms within this core
sets particular challenges in structure elucidation and the
assignment of absolute and relative configurations at C7.
Within the biosynthesis of these natural products the prenyl-
transferase-initiated cationic cyclization of acylphloroglucine
(I) could lead to mixtures of up to four PPAPs (II–V;
Figure 1).[4–6]
cyclization pairs of trans (or endo) type B PPAPs (III)/cis (or
exo) type A PPAPs (II) and exo type B PPAPs (V)/endo
type A PPAPs (IV) are formed. In most cases mixtures of
these isomers are obtained from natural sources. While the
differentiation between type A and type B PPAPs by spec-
troscopic methods is unproblematic, the exact assignment of
exo or endo configuration is challenging. The PPAP core
possesses only three hydrogen atoms and hence NOE experi-
ments do not give clear information on relative configura-
tions.[7,8] At this point total synthesis aids the structure
elucidation.[9]
The two tetrahydrofuran-annulated type B PPAPs samp-
sonione P (1) and hyperibone I (2) are representative exam-
ples for the pitfalls of PPAP structure analysis. Sampsonio-
ne P was isolated in 2007 from Hypericum sampsonii, a genus
in the family of Hypericaceae.[10] Hyperibone I, together with
eight other PPAPs, was isolated from Hypericum scabrum in
2002.[11] In addition to different positions of the tetrahydro-
furan ring within the tricyclic core, different C7 configura-
tions were proposed (Figure 2). While for sampsonione P (1)
an endo configuration was suggested, the isomeric hyper-
ibone I was proposed to possess an exo configuration (2a).
The latter stereochemical assignment was mainly based upon
comparison with similar compounds and 1D-NMR spectros-
copy.
While in type A PPAPs the bridgehead carbon atom C5 is
substituted by an acyl group, type B PPAPs have a C3 acyl
group. Depending on the course of the prenylase-catalyzed
In their 2006 review, Grossman and Ciochina proposed
that based on characteristic NMR shifts hyperibone I might
Figure 1. Polyvergent biosynthetic pathway to exo- and endo-type
PPAPs.
[*] Dipl.-Chem. K. Lindermayr, Prof. Dr. B. Plietker
Institut fꢀr Organische Chemie, Universitꢁt Stuttgart
Pfaffenwaldring 55, 70569 Stuttgart (Germany)
E-mail: bernd.plietker@oc.uni-stuttgart.de
[**] K.L. thanks the Landesgraduiertenstiftung Baden-Wꢀrttemberg for
a PhD grant.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201306256.
Figure 2. Molecular structures of sampsonione P (1) and hyperibone I
(2a, 2); confirmation and revision.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
better be regarded as an endo type B PPAP (2a).[4] After
careful comparison of NMR data of known PPAP structures,
they found that an exo configuration leads to a difference in
chemical shifts of 0.3–1.2 ppm for the two H6 atoms and the
characteristic shift of C7 is within the range of 41–44 ppm. In
contrast, endo isomers have only a small difference in the
chemical shifts of the H6 atoms (0.0–0.2 ppm) and the
chemical shift of C7 is 45–49 ppm.
Recently we disclosed a short total synthesis of endo
type B PPAPs.[12] The strict separation of framework-con-
structing from framework-decorating steps permitted the
selective variation of the substitution pattern without affect-
ing the general synthetic approach. Based on these results and
the divergence in the stereochemical assignment of hyper-
ibone I we envisioned a total synthesis to be the only way to
fully assign the structure of hyperibone I. Herein we describe
a short divergent nine- and ten-step total synthesis of
sampsonione P and hyperibone I. The NMR spectra for
both PPAPs are in almost perfect agreement with the
reported data of the natural products and confirm Grossmanꢀs
hypothesis that hyperibone I is an endo type B PPAP
(Figure 2).
With regard to the structural differences between samp-
sonione P and hyperibone I, we thought the facile electro-
philic oxidation of an electron-rich prenyl side chain in the
presence of a less electron-rich allyl substituent would be
a suitable regioselectivity switch for the formation of the
tetrahydrofuran ring at a late stage in the total synthesis
(Figure 3).[13] In order to verify this strategy, regio- and
diastereoselective allylation or prenylation at different stages
of the synthesis was necessary.
Starting from acetylacetone (6), (a-allylvinyl) methyl
ketone (7) was prepared in a two-step/one-pot procedure
consisting of an allylation and a deacylating aldol condensa-
tion[14] in good overall yield (Scheme 1). The subsequent
Scheme 1. Bidirectional total synthesis of sampsonione P and hyper-
ibone I: preparation of cyclohexanones 11 and 13. Reagents and
conditions: a) NaH, allyl bromide, ethanol, 08C–RT, 15 h, then K2CO3,
CH2O(aq), RT, 15 h; b) MeMgCl, dimethyl 1,3-acetonedicarboxylate,
methanol, 0–608C, 18 h; c) NaH, MeLi, THF, 08C; 5 h; d) NaH, prenyl
bromide (for 13) or e) allyl bromide (for 11), THF, 08C–RT, 15 h;
f) LiCl, CuI, MeMgBr, Me3SiCl, THF, À788C, 4 h.
tandem Michael addition/Dieckmann condensation using
dimethyl acetonedicarboxylate and regioselective 1,2-addi-
tion[15] gave rise to cyclohexenone 9 in high overall yields. This
starting material represents the point of divergence for the
two syntheses.
Allylation using prenyl bromide under basic conditions
gave the corresponding cyclohexenone 12 in excellent diaste-
reoselectivity in favor of the trans-bis-allylated stereoisomer.
We were pleased to find that the introduction of the sterically
less demanding allyl side chain, which is required for the
synthesis of hyperibone I, proceeded equally well and with
good diastereoselectivity (leading to 10). The 1,4-addition of
MeCu[16] to 10 and 12 furnished the corresponding gem-
dimethyl moiety that is found in many PPAPs.
With these key intermediates in hand we then turned our
attention to the total synthesis of sampsonione P (Scheme 2).
Unexpectedly, the introduction of the allyl side chain proved
to be problematic. Direct allylation using allyl bromide in the
presence of base gave only the undesired allyl vinyl ether.
Metal-catalyzed intermolecular Tsuji–Trost allylation pro-
vided the C-allylation product 15 as a mixture of stereoiso-
mers along with varying amounts of the allyl vinyl ether.
However, transformation of the allyloxycarbonyl derivative
14 in a Pd-catalyzed decarboxylative Tsuji–Trost allylation
Figure 3. Retrosynthesis.
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Scheme 2. Total synthesis of sampsonione P. Reagents and conditions:
a) NaH, allyl chloroformate, DMF, 08C–RT, 3 h, then b) [Pd2dba3], p-
Tol3P, toluene, 608C, 16 h; c) KOtBu, THF, then d) PhC(=O)CN, THF,
08C–RT, 48 h; (e) mCPBA, CH2Cl2, 08C, 30 min; f) Grubbs II cat.
(0.15 equiv), 2-methyl-2-butene, CH2Cl2, 408C, 20 h. mCPBA: meta-
chloroperbenzoic acid.
Scheme 3. Total synthesis of hyperibone I. Reagents and conditions:
a) KOCMe2Et, 1,3-dimesitylimidazolin-2-ylidene hexafluorophosphate
(0.2 equiv), Bu4N[Fe(CO)3(NO)] (0.2 equiv), methyl(2-methyl-3-butene-
2-yl)carbonate, THF/MTBE, RT–808C, 20 h;b) KOtBu, then
c) PhC(=O)CN, THF, 08C–RT, 48 h; d) mCPBA, CH2Cl2, 08C, 2 h;
e) Grubbs II cat. (0.15 equiv), 2-methyl-2-butene, CH2Cl2, 408C, 20 h.
MTBE: tert-butyl methyl ether.
gave the desired cis product 15 in a diastereomeric ratio of
5:1.[17–19] Treatment of 15 with KOtBu in THF at 08C gave
enolate 17, which reacted with benzoylcyanide to give the
desired C-benzoylated [3.3.1]bicyclononatrione core 5 in
good yield. Treatment of 5 with mCPBA at 08C for one
hour led to the chemoselective epoxidation of the prenyl side
chain with concomitant cyclization to give the precursor (18)
to sampsonione P containing the tetrahydrofuran motif.[13]
Oxidation of the allyl side chains was not observed. The
desired product 18 was obtained in good yields albeit in low
diastereoselectivity. The mixture of isomers was subjected to
a cross-metathesis[20] using 2-methyl-2-butene and the Grubb-
s II catalyst to give the corresponding product 1 in good yield
(Scheme 2). At this stage separation of the stereoisomers
proved to be nonproblematic and was achieved by conven-
tional column chromatography. Sampsonione P (1) was
unambiguously identified by comparison of the NMR data
with the reported data.[10,21]
The total synthesis of hyperibone I (2) was tackled
subsequently (Scheme 3). As already observed in the syn-
thesis of sampsonione P, the introduction of a prenyl group
proved to be problematic. In analogy to related transforma-
tions in the total syntheses of epi-clusianone and oblongifo-
lin A,[12] Fe-catalyzed allylic substitution was the only way to
obtain synthetically useful amounts of the desired product 20.
However, only a modest diastereoselectivity of 4.3:1 was
achieved in favor of the cis product with formation of 18 as
a byproduct.[22]
benzoylcyanide to provide bicycle 4 in good overall yield.
Gratifyingly, the subsequent mCPBA-mediated epoxidation
was accompanied by ring opening and cyclization to give the
corresponding tetrahydrofuran derivative 22. Finally cross-
metathesis with 2-methyl-2-butene furnished product 2 in
good overall yield and a diastereomeric ratio of 3.1:1.0.
Product 2 was unequivocally identified as hyperibone I by
comparison with the NMR data of the isolated natural
product.[21]
Although the endo configuration of PPAPs obtained
through our synthetic strategy was proven in our previous
study,[12] we performed in-depth NMR measurements in order
to secure the relative configuration of synthetic 1 and 2
(Figure 4). The use of [D6]benzene as the solvent proved to be
important in order to resolve the relevant proton signals at C6
and C7 in 1 and 2, respectively. The C7 substituent occupies
a pseudo-axial orientation within the cyclohexanone core
which can be assumed based on solution experiments and X-
ray crystallographic analysis of the related hyperibone L.[12] In
case of hyperibone I (2) a strong NOE was observable
between H(eq.,C7) and H(ax.,C6) and a weak NOE to H(eq.,C6)
Furthermore, a strong NOE was visible between H(ax.,C6)
H(eq.,C7) and CH3(ax.,C8). CH3(eq.,C8) showed a strong NOE to
(eq.,C10), H(eq.,C11), and H(C27). Based on these data we propose
.
/
Following the base-promoted intramolecular Claisen
cyclization to give 21, the enolate was trapped using
H
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
[1] J.-A. Richard, R. H. Pouwer, D. Y.-K. Chen, Angew. Chem. 2012,
124, 4612 – 4638; Angew. Chem. Int. Ed. 2012, 51, 4536 – 4561.
[2] N. S. Simpkins, Chem. Commun. 2013, 49, 1042 – 1051.
[3] J. T. Njardarson, Tetrahedron 2011, 67, 7631 – 7666.
[4] R. Ciochina, R. B. Grossman, Chem. Rev. 2006, 106, 3963 – 3986.
[5] P. Adam, D. Arigoni, A. Bacher, W. Eisenreich, J. Med. Chem.
2002, 45, 4786 – 4793.
[6] O. Cuesta-Rubio, H. Velez-Castro, B. A. Frontana-Uribe, J.
Cardenas, Phytochemistry 2001, 57, 279 – 283.
[7] O. Cuesta-Rubio, A. Padron, H. V. Castro, C. Pizza, L. Rastrelli,
J. Nat. Prod. 2001, 64, 973 – 975.
[8] R. B. Grossman, H. Jacobs, Tetrahedron Lett. 2000, 41, 5165 –
5169.
Figure 4. Characteristic NOEs in sampsonione P and hyperibone I
[9] K. C. Nicolaou, S. A. Snyder, Angew. Chem. 2005, 117, 1036 –
1069; Angew. Chem. Int. Ed. 2005, 44, 1012 – 1044.
[10] Z. Y. Xiao, Q. Mu, W. K. P. Shiu, Y. H. Zeng, S. Gibbons, J. Nat.
Prod. 2007, 70, 1779 – 1782.
[11] M. Matsuhisa, Y. Shikishima, Y. Takaishi, G. Honda, M. Ito, Y.
Takeda, H. Shibata, T. Higuti, O. K. Kodzhimatov, O. Ashurme-
tov, J. Nat. Prod. 2002, 65, 290 – 294.
[12] N. Biber, K. Mçws, B. Plietker, Nat. Chem. 2011, 3, 938.
[13] N. M. Ahmad, V. Rodeschini, N. S. Simpkins, S. E. Ward, A. J.
Blake, J. Org. Chem. 2007, 72, 4803 – 4815.
[14] T. Ben Ayed, H. Amri, Synth. Commun. 1995, 25, 3813 – 3819.
[15] S. N. Huckin, L. Weiler, Can. J. Chem. 1974, 52, 1379 – 1380.
[16] M. Amat, N. Llor, B. Checa, E. Molins, J. Bosch, J. Org. Chem.
2010, 75, 178 – 189.
[17] J. Tsuji, I. Minami, I. Shimizu, Tetrahedron Lett. 1983, 24, 1793 –
1796.
[18] B. M. Trost, J. Xu, T. Schmidt, J. Am. Chem. Soc. 2009, 131,
18343 – 18357.
[19] B. M. Trost, J. Xu, J. Am. Chem. Soc. 2005, 127, 2846 – 2847.
[20] D. R. Siegel, S. J. Danishefsky, J. Am. Chem. Soc. 2006, 128,
1048 – 1049.
[21] A comparison of the NMR data of the isolated natural products
and the synthetic products is found in the Supporting Informa-
tions.
(recorded in [D6]benzene).
the revised structure of hyperibone I to possess the relative
configuration shown in Figure 3 and to be an endo type B
PPAP.
In summary, we have presented the first total synthesis of
sampsonione P and hyperibone I in ten and nine steps,
respectively, starting from acetylacetone. The strict separation
of framework-decorating and framework-constructing trans-
formations allowed us to selectively introduce a prenyl group.
Owing to the different rates in the epoxidation of prenyl and
allyl groups, we could form the tetrahydrofuran motif at a late
stage in the synthesis. The final cross-metathesis provided the
desired remaining prenyl side chains in good overall yield.
The key step of the two syntheses is the introduction of the
last allyl or prenyl substituent by either a decarboxylative
Tsuji–Trost allylation or a Fe-catalyzed allylation. Both the
total syntheses and the in-depth spectroscopic investigations
confirmed the reported structure of sampsonione P and the
previously proposed revised structure of hyperibone I.
[22] a) B. Plietker, Angew. Chem. 2006, 118, 1497 – 1501; Angew.
Chem. Int. Ed. 2006, 45, 1469 – 1473; b) B. Plietker, A. Dieskau,
K. Mçws, A. Jatsch, Angew. Chem. 2008, 120, 204 – 207; Angew.
Chem. Int. Ed. 2008, 47, 198 – 201; c) M. Holzwarth, A. Dieskau,
M. Tabassam, B. Plietker, Angew. Chem. 2009, 121, 7387 – 7391;
Angew. Chem. Int. Ed. 2009, 48, 7251 – 7255.
Received: July 18, 2013
Published online: && &&, &&&&
Keywords: allylation · homogeneous catalysis ·
.
natural products · PPAPs · total synthesis
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Natural Product Synthesis
K. Lindermayr,
B. Plietker*
&&&& — &&&&
The Bidirectional Total Synthesis of
Sampsonione P and Hyperibone I
Make it simple: The separation of frame-
work-decorating from framework-con-
structing steps facilitated the selective
introduction of one prenyl group next to
allyl groups. The selective epoxidation of
the more electron-rich prenyl group led to
the efficient formation of the tetrahydro-
furan moiety in the title compounds.
Spectroscopic analysis of hyperibone I
and comparison with literature data led to
a revision of the original structure.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!