A serendipitous synthesis of (+)-gregatin B, second structure revisions of the aspertetronins, gregatins, and graminin A, structure revision of the penicilliols

Article abstract of DOI:10.1021/ol2008318

A (DHQN)₂AQN-promoted asymmetric dihydroxylation (92% ee) of the allyl chloride derived from enynol (E)-13 and an 8-step sequence provided access to the hydroxyethylated furanone (R)-21. Oxidation with MnO₂ furnished 50% furanone (+)

Full text of DOI:10.1021/ol2008318

ORGANIC
LETTERS
2011
Vol. 13, No. 10
2730–2733
A Serendipitous Synthesis of (þ)-Gregatin B,
Second Structure Revisions of the
Aspertetronins, Gregatins, and Graminin A,
Structure Revision of the Penicilliols
€
Heike Burghart-Stoll and Reinhard Bruckner*
€
Institut fu€r Organische Chemie und Biochemie, Albert-Ludwigs-Universitat Freiburg,
Albertstr. 21, 79104 Freiburg im Breisgau, Germany
reinhard.brueckner@organik.chemie.uni-freiburg.de
Received March 30, 2011
ABSTRACT
A (DHQN)2AQN-promoted asymmetric dihydroxylation (92% ee) of the allyl chloride derived from enynol (E)-13 and an 8-step sequence provided
access to the hydroxyethylated furanone (R)-21. Oxidation with MnO2 furnished 50% furanone (þ)-(R)-2a and 2.7% isomeric furanone (þ)-(R)-3a.
(R)-2a possesses the accepted constitution of (þ)-gregatin B but its spectra are different. Surprisingly, (þ)-(R)-3a equals the natural product.
Analogous structure reassignments are due for the gregatins A and CꢀE, the aspertetronins AꢀB, graminin A, and the penicilliols A and B.
(þ)-Aspertetronin A and (ꢀ)-aspertetronin B are nat-
ural products from Aspergillus rugulosus.1 Degradation
and correlation studies led to the conclusion that they are
the furan-2(5H)-ones 1b and d, respectively (Figure 1).1
The optical antipodes of aspertetronin A and B are natural
products as well, isolated from Cephalosporium gregatum2
and called3 (ꢀ)-gregatin A4,5 and (þ)-gregatin D,4 re-
spectively. (þ)-Gregatin B,4 (þ)-gregatin C, and (þ)-
gregatin E were first isolated from the same source,2
named analogously,3 and believed to be the furan-2(5H)-
ones 1a, d, and f, respectively. “Metabolite (þ)-704-II”6
possesses the connectivity 1e of a methyl ether of gregatin
D. (ꢀ)-Graminin A from Cephalosporium gramineum7
was assigned the analogous structure 1c. Rac-1a8ꢀ10 and
rac-1b8 were synthesized early on. The distinctness of
their spectral properties from those of the natural
products sparked the suggestion that the latter are the
(1) Ballantine, J. A.; Ferrito, V.; Hassall, C. H.; Jones, V. I. P.
J. Chem. Soc. 1969, 56–61.
(2) Kobayashi, K; Ui, T. Tetrahedron Lett. 1975, 4119–4122.
(3) Kobayashi, K.; Ui, T. Physiol. Plant Pathol. 1977, 11, 55–60.
(4) Second isolation of this compound (from Aspergillus panamensis):
Anke, H.; Schwab, H.; Achenbach, H. J. Antibiot. 1980, 33, 931–939.
(5) Third isolation (from Phialophora gregata): Taylor, S. L.; Peterson,
R. E.; Gray, L. E. Appl. Environ. Microbiol. 1985, 50, 1328–1329.
(6) First isolation of this compound (from Aspergillus panamensis):
ref 4.
(7) Kobayashi, K.; Ui, T. J. Chem. Soc., Chem. Comm. 1977, 774–
774.
(8) (a) Clemo, N. G.; Pattenden, G. Tetrahedron Lett. 1982, 23, 585–
588. (b) Clemo, N. G.; Pattenden, G. J. Chem. Soc., Perkin Trans. 1 1985,
2407–2411.
Figure 1. Original (1aꢀf,1,2,4,6,7 2g,12 h12), revised (2aꢀf,8,11
3g,[a] h[a]), and re-revised[a] (3aꢀf) structures of gregatins A [(ꢀ)-
b2,4,5], B [(þ)-a2,4], C [(þ)-d2], D [(þ)-d2,4], and E [(þ)-f2], asper-
tetronins A [(þ)-b1] and B [(ꢀ)-d1], graminin A [(ꢀ)-c7], meta-
bolite 704-II [(þ)-e6], and penicilliols A [(þ)-g12] and B [(þ)-h12].
[a] This work.
r
10.1021/ol2008318
Published on Web 04/28/2011
2011 American Chemical Society

furan-3(2H)-ones 2a and b.8,11 The presence of the same
substructure was suggested for family members cꢀf, too.8
(þ)-Penicilliol A (2g) and (þ)-penicilliol B (2h), which were
isolated from Penicillium daleae very recently, were as-
sumed right away to represent furan-3(2H)-ones.12
Takaiwa and Yamashita reported the first syntheses of
racemic13,14 and (þ)-gregatin B14 from lactone carboxylic
acids rac- or (S)-4, respectively (Scheme 1). Treatment of
the derived acyltetronic acids rac- or (S)-6 with diazo-
methane and BF3 etherate gave 11% rac- or (S)-1a and
1.2% of the elusive gregatin B. The latter was optically
inactive using rac-6 and dextrorotatory using (S)-6. These
observations, matching 1H NMR and IR data, and their
While the compounds compiled in Figure 1 exhibit some
pharmacological promise,16 theyhave been studied solittle
let alone variedstructurally that developing a generalizable
synthetic access appeared worthwhile. Our route emerged
from a retrosynthetic analysis exemplified for the purpor-
ted14 structure (S)-2a of natural gregatin B in Scheme 2. We
planned to obtain (S)-2a from furanone (S)-7 by a lithia-
tion/acetylation sequence. Furanone (S)-7 was known not
to be a viable O-methylation product of the “underlying”
ketolactone 1117,18 even if the required transformation
worked well with ketolactone 12.19 We considered (S)-7 as
a C,C-coupling product between a but-1-enyl reagent and a
hydrometalation or -halogenation product of the CtC-
containing ortholactone (2S,3R)-8. Most orthoesters stem
from alcoholyses of imidoester hydrochlorides.20 Hence,
(2S,3R)-8 was to be made by a Pinner cyclization21 of the
known22 dihydroxynitrile (3R,4S)-9 and a subsequent
methanolysis.
Scheme 1. Purported Synthesis14 of the (S)-Enantiomer of the
Revised8 Structure 2a of (þ)-Gregatin B from (S)-4 but in Actual
Fact Racemic Syntheses13,14 of the Re-revised Structure 3aa
Scheme 2. Retrosynthetic Analysis of the Revised8;and See-
mingly Confirmed14,15;Structure (S)-2a of (þ)-Gregatin B
The imidolactone hydrochloride (4R,5S)-14 was ob-
tained as a crystalline compound from dihydroxynitrile
(3R,4S)-9 and HCl gas (Scheme 3, upper half). Methano-
lysis provided the ortholactone (2S,3R)-8 and hydrostan-
nylation of the CꢁC bond23 a 93:7 mixture of stannane
(2S,3R)-16 and its regioisomer. Treatment with NBS24 led
to the corresponding bromoolefin (2S,3R)-15. Coupling25
a Formal total synthesis of gregatin B from D-alanine15
.
synthetic design led to the belief that natural gregatin B is
(S)-2a. This view was corroborated by a formal total
synthesis of gregatin B from D-alanine, which merged into
the Takaiwa/Yamashita route13,14 at intermediate (S)-6.15
However, as we will show, (þ)-gregatin B possesses struc-
ture (R)-3a rather than (S)-2a.
(16) Gregatin B and E act antimicrobially,2 gregatin A-D and
metabolite 704-II antibacterially,4 (þ)-penicilliol A and B inhibit
DNA polymerase.12
(17) Methylation of rac-11 with diazomethane furnished 18% rac-7
as the minor constituent of a separable mixture with 10 (37% yield):
Effenberger, F.; Syed, J. Tetrahedron Asymmetry 1998, 9, 817–825.
ꢀ
(18) Attempted methylation of (R)-11 with Meerweins salt led to
€
decomposition: Kapferer, T. Dissertation; Universitat Freiburg, 2006;
(9) Takeda, K.; Kubo, H.; Koizumi, T.; Yoshii, E. Tetrahedron Lett.
1982, 23, 3175–3178.
pp 132-133.
€
€
(19) Kapferer, T.; Bruckner, R.; Herzig, A.; Konig, W. A. Chem.;
(10) Miyata, O.; Schmidt, R. R. Tetrahedron Lett. 1982, 23, 1793–
1796.
Eur. J. 2005, 11, 2154–2162.
(20) Lebel, H.; Grenon, M. In Science of Synthesis ꢀ Houben-Weyl
Methods of Organic Chemistry, Vol. 22; Charette, A., Ed.; Thieme:
Stuttgart, NY, 2005; pp 669ꢀ748.
(11) The revisions 1af2a and 1af2a were driven by the mismatch
between data of 1 vs. the natural products but based on little positive
evidence because the reference compounds cited in Clemo, N. G.;
Pattenden, G. Tetrahedron Lett. 1982, 23, 589–592 and ref 8b contain
fewer CdO groups than 1 or 2.
(12) Kimura, T.; Takeuchi, T.; Kumamoto-Yonezawa, Y.; Ohashi,
E.; Ohmori, H.; Masutani, C.; Hanaoka, F.; Sugawara, F.; Yoshida, H.;
Mizushina, Y. Bioorg. Med. Chem. 2009, 17, 1811–1816.
(13) Takaiwa, A.; Yamashita, K. Agric. Biol. Chem. 1982, 46, 1721–
1722.
(14) Takaiwa, A.; Yamashita, K. Agric. Biol. Chem. 1984, 48, 2061–
2065.
(15) Matsuo, K.; Kanayama, M.; Xu, J. Y.; Takeuchi, R.; Nishiwaki,
K.; Asaka, Y. Heterocycles 2005, 65, 1609–1614.
(21) Review: (a) DeWolfe, R. H. Synthesis 1974, 153–172. Cycliza-
tion of parent compound: (b) Topchiev, K. S.; Kirmalova, M. L. Dokl.
Akad. Nauk SSSR 1948, 63, 281ꢀ284 (Chem. Abstr. 1949, 43, 2579).
Recent examples: (c) Fleming, F. F.; Wei, G.; Steward, O. W. J. Org.
Chem. 2008, 73, 3674–3679 and ref 39 therein.
€
(22) Burghart-Stoll, H.; Kapferer, T.; Bruckner, R. Org. Lett. 2011,
13, 1016–1019.
(23) ProcedureBetzer, J.-F.; Delaloge, F.; Muller, B.; Pancrazi, A.;
Prunet, J. J. Org. Chem. 1997, 62, 7768–7780.
(24) Procedures: (a) Wipf, P.; Coish, P. D. G. J. Org. Chem. 1999,
64, 5053–5061. (b) Malecka, R. E., J.; Gallagher, W. P. Org. Lett. 2001,
3, 4173–4176.
Org. Lett., Vol. 13, No. 10, 2011
2731

Scheme 3. Synthesis of the Precursor (ꢀ)-(S)-7 of the Hitherto
Accepted Structure 2a of (þ)-Gregatin B (top half); synthesis of
Antipode (þ)-(R)-7 via Diastereomeric Intermediates (bottom
half)28,29
Scheme 4. Elaboration of Furanones (S)- and (R)-735a
a Attribution of structure (R)-3a to (þ)-gregatin B based on total
[a]
synthesis, methyl-singlet 1H-NMR
shift values, and X-ray crystal-
[a]
lography. The solvent was CDCl3 except for the quote from refs 13
and 14 in which case the solvent was CCl4. [b] [R]D20 = ꢀ47.6 (c = 1.1 in
CHCl3). [c] Since progress required an efficient access this sample of (S)-7
was not prepared as specified in Scheme 3 [top half; 1.9% overall yield
from (Z)-13] but by converting (E)-13 into (ꢀ)-(3S,4S)-922 (85% ee) and
proceeding analogously as shown in Scheme 3 (bottom half) for the
transformation of (þ)-(3R,4R)-9 into (R)-7 (details: Supporting Infor-
[a] [R]D20 = þ53.8 (c = 1.03 in CHCl3). [b] [R]D20 = ꢀ49.1 (c = 1.04 in
CHCl3).
[d]
20
mation; 5.1% overall yield).
[R]D = ꢀ184 (c = 0.09 in CHCl3).
[e] Complete synthesis: Supporting Information.
with boronate 1826 completed the hexadienyl side-chain,
Q
delivering (2S,3R)-17. Oxidation with Pr4Nx RuO4
/
from (S)-7, we obtained (S)-2a, which was levorotatory (31%
over the 2 steps; 86% ee33). It differed in this respect and 1H
NMO27 was accompanied by the loss of methanol and
afforded the desired furanone (S)-7. In anticipation of the
need of an HPLC determination of the enantiopurity of
our gregatin-to-be, wepreparedthe enantiomeric furanone
(R)-7, too (Scheme 3, lower half): by subjecting dihydrox-
ynitrile (3R,4R)-922;a diastereomer of the previous start-
ing material (3R,4S)-9;to the same sequence of steps,
which had led to (S)-7.
Scheme 4 supplements the 4-acetylation of furanones (S)-
and (R)-7. This transformation worked poorly in one step
(LDA or RLi; AcX) but satisfyingly if realized in two steps:
(1) successive treatment with LDA and acetaldehyde;30 (2)
oxidation by ca. 40 equiv of “active MnO2”.31,32 Starting
ꢀ
ꢀ
NMR spectroscopically from Takaiwas and Yamashitas
synthetic “(þ)-(S)-2a” (Scheme 1) and from (þ)-gregatin
B. Surprisingly, (S)-21 and MnO2 also gave a little (ꢀ)-(S)-3a
(83% ee33). The latter showed the same 1H NMR spectrum
as (þ)-gregatin B and (!) the mentioned “(þ)-(S)-2a”.34 The
structures of compounds 2a and 3a followed from 1H NMR
similarities between 2a and compound 23 and between 3a and
compound 22 (Table 1). The structures of the reference com-
pounds emerged from X-ray analyses (Scheme 4, bottom).35
(32) This transformation exemplifies a 2-step sequence, for which we
know a single precedent (19f20); the initiation step differed, though:9
€
(25) Procedure: Hanisch, I.; Bruckner, R. Synlett 2000, 374–378. The
addition of charcoal was suggested by Prof. P. Vogel (EPF Lausanne).
(26) Compound 18 was prepared by a cp2ZrHCl-mediated pinacol-
boration of but-1-yne following a procedure by Wang, Y. D.; Kimball,
G.; Prashad, A. S.; Wang, Y. Tetrahedron Lett. 2005, 46, 8777–8780.
(27) Procedure:Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P.
Synthesis 1994, 639–666.
(28) In conjuncture with other findings (ref 29) these syntheses of (S)-
and (R)-7 proved that the asymmetric dihydroxylations en route to
(3R,4S)- and (3R,4R)-922 proceed in accordance with “Sharpless/Norrby
orientations” of the CdC bonds in the transition state.
(33) The enantiopurity of this compound was determined by HPLC
(for details see Supporting Information).
(34) “(þ)-(S)-2a” from Scheme 1 is 3a by the 1H NMR comparisons
in Scheme 4 yet our preparations of (ꢀ)-(S)- and (þ)-(R)-3a show that
the descriptors “(þ)” and “(S)” do not match for any enantiomer of
3a. The easiest way of reconciliating the findings of refs 13 and 14
with ours assumes that their 3a was (1) racemic [cf. our acid-catalyzed
isomerization (ꢀ)-(S)-2afrac-3a; Scheme 5] and (2) dextrorotatory
because of an unnoticed impurity.
€
€
(29) Burghart-Stoll, H.; Bohnke, O.; Bruckner, R. Org. Lett. 2011,
13, 1020–1023.
(30) The conditions for this step were gleaned from analogous
functionalizations of a 4-methoxyfuran-2(5H)-one; these are described
in ref 8b.
(35) CCDC 813695 and 813696 contain the crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via the link www.ccdc.
cam.ac.uk/data_request/cif.
(31) Review: Fatiadi, A. J. Synthesis 1976, 65–104.
2732
Org. Lett., Vol. 13, No. 10, 2011

Table 1. 1H-NMR Juxtaposition of Type-2 and Type-3 Fura-
nones (Structure-Differentiating Motifs and Shifts on Grey
Background)
Scheme 5. Isomerizations of (ꢀ)-(S)-2a Giving (ꢀ)-(S)-3a (left
side) or rac-3a (right side) and Rationalizations for Their Course
Successive treatments with LDA and acetaldehyde and
oxidation with MnO2 turned furanone (R)-7 into (þ)-(R)-
2a (36%; 92% ee33) and (þ)-(R)-3a (2%; 91% ee33).
(þ)-(R)-3a was a solid (mp = 78 °C; ref 2, 80ꢀ81 °C;
20
ref 4, oil) and identical with natural gregatin B {[R]D
=
þ211 (c = 0.08 in CHCl3); ref 2, [R]D = þ207 (c = 0.84 in
CHCl3); ref 4, [R]D = þ205 (c = 0.1 in CHCl3)}.
Further experimentation (Scheme 5) revealed that
(ꢀ)-(S)-2a (86% ee33) became a progenitor of (ꢀ)-(S)-3a
when exposed to MnO2 (150 equiv, CH2Cl2, room temp, 2
Table 1 lists the 1H NMR shifts of the methyl singlets of
various furan-3(2H)-ones from the current study, refs 13
and 14, and the natural products shown in Figure 1. The δ-
values of the highlighted protons cluster in specific ranges.
This suggests that the gregatins A and CꢀE, the asperte-
tronins A and B, graminin A, and the penicilliols A and B
are rather type-3 furanones than, as has been believed
hitherto, type-2 furanones. The kind of structure revision,
which this implies is identical to how the real structure (R)-
3a of (þ)-gregatin relates to the connectivity in (S)-2a.
The emergence of (R)-3a from the current study
(Scheme 3, bottom, þ Scheme 4, center) represents the
first total synthesis of (þ)-gregatin. It relies on an asym-
metric dihydroxylation and a Pinner cyclization/methano-
lysis route to an ortholactone. It ends with a low-yielding
but novel furanone rearrangement. Targeted syntheses of
gregatin B and its congeners are now conceivable.
d) and a progenitor of rac-3a when treated with pTsOH
3
H2O (CH2Cl2, room temp, 1 h). The former conditions
resemble those of our serendipitous routes to (ꢀ)-(S)- and
(þ)-(R)-3a (Scheme 4). The latter conditions are commem-
orative of the BF3 Et2O environment, in which Takaiwa
3
and Yamashita13,14 allegedly obtained “(þ)-(S)-2a“ but in
fact probably rac-3a34 (Scheme 1). Our deliberate isomer-
izations (ꢀ)-(S)-2af3a were allowed to reach completion
as judged by TLC. In the MnO2 case we isolated 25%
(ꢀ)-(S)-3a (79% ee33) and in the pTsOH.H2O case 46%
nearly racemic 3a [2% ee33 favoring the (S)-enantiomer].
Accordingly, MnO2-mediation retains the CquatꢀO bond
while Brønsted acid catalysis breaks it ꢀ as the mechan-
isms imply, which are suggested in Scheme 5: HOQ formed
from H2O, which is omnipresent in MnO231 opens (ꢀ)-(S)-
2a to give 24 by a nucleophilic vinylic substitution via
addition and elimination. Tautomerism, hemiacetal-anion
(“lactolate”) formation, and loss of HOQ establish ring
(ꢀ)-(S)-3a. The protonation-induced isomerization of
(ꢀ)-(S)-2a seems to follow an SN1 course. It starts with
the heterolysis of the CquatꢀO bond. The resulting penta-
dienyl cation 25 is destabilized by a tetraoxygenated sub-
stituent (perhaps less if twisted as drawn). Conformational
changes 26a h 26b and ion capture would give ring rac-3a.
Acknowledgment. Dedicated to Prof. Horst Prinzbach
on the occasion of his 80th birthday. We are grateful to
€
Dr. Manfred Keller (Institut fur Organische Chemie und
Biochemie, Universitat Freiburg) for NMR support and to
€
Dr. Jens Geier (ibid.) for the X-ray analyses.
Supporting Information Available. Procedures, charac-
terizations, copies of NMR spectra. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
Org. Lett., Vol. 13, No. 10, 2011
2733