Total syntheses of the dihydrofuranonecarboxylate natural products gregatin b and e: Gram-scale synthesis of (+)-gregatin b and unambiguous assignment of the stereostructure of (+)-gregatin e

Article abstract of DOI:10.1021/ol5032602

We synthesized the dextrorotatory enantiomers of gregatin B (1.3 g scale) and E, establishing their diene side chains in the last step by Heck coupling with variable iodoolefins. Their counterpart was synthesized in 30% overall yield and with high enantio

Full text of DOI:10.1021/ol5032602

Letter
pubs.acs.org/OrgLett
Total Syntheses of the Dihydrofuranonecarboxylate Natural Products
Gregatin B and E: Gram-Scale Synthesis of (+)-Gregatin B and
Unambiguous Assignment of the Stereostructure of (+)-Gregatin E
Fabian Weber and Reinhard Bruckner*
̈
Institut fur Organische Chemie, Albert-Ludwigs-Universitat, Albertstraße 21, 79104 Freiburg im Breisgau, Germany
̈
̈
S
* Supporting Information
ABSTRACT: We synthesized the dextrorotatory enantiomers of gregatin B (1.3 g scale) and E, establishing their diene side
chains in the last step by Heck coupling with variable iodoolefins. Their counterpart was synthesized in 30% overall yield and
with high enantiopurity (97% ee) from benzyl tiglate, beginning with an asymmetric dihydroxylation. The stereostructure of
gregatin E followed from HPLC comparisons between the four synthetic stereoisomers of this compound and natural gregatin E,
which we isolated from Cadophora gregata.
he gregatins (Figure 1) are secondary metabolites from the
CH2−CHCH−CH2−CHO likewise.5b Analogous routes
to synthetic gregatin E would have constructed its side chain from
trans-H3C−C(Oprotected)H−CH2−CHCH−CHO or, con-
sidering the promise of higher E,E-selectivity, preferably from
trans-H3C−C(Oprotected)H−CHCH−CH−CHO. Making
the latter building block would entail a non-negligible number
of steps, which would have to deal with an aggravating factor: At
the outset of the present study, the side-chain configuration of
gregatin E was unknown. Circumventing these constraints we
now synthesized gregatin B, gregatin E,7 and epi-gregatin E7 by a
different strategy. Its objectives were (1) to establish the diene
moiety without a Z-configuration and (2) to introduce the side-
chain stereocenter as late as possible.
We addressed the mentioned objectives concomitantly as
revealed by our retrosynthetic analysis (Scheme 1). We traced
back the (R,E,E)- or (S,E,E)-configured 5-hydroxyhexa-1,3-diene
side chain of gregatin Eor whatever hexa-1,3-diene a different
gregatin would requireto an E,E-selective8 Heck coupling as its
origin. It would engage an appropriately substituted (E)-1-
iodobut-1-ene and the vinyl substituent of an appropriately
substituted furanone (shown in the upper right-hand corner of
the scheme). The vinyl group should arise from the β-elimination
of a secondary mesylate. The latter was present in every
precursor except in the starting olefin and in the product of its
asymmetric dihydroxylation. This is because these mesylates
represent a protected alcohol of sorts. The furanone core should
be established by a Knoevenagel-type condensation9 (right-hand
part of Scheme 1). It would start from a β-ketoester, which
looked like it originated from a Claisen condensation. In fact this
compound would be made by condensing an activated α-acyloxy-
T
fungus Cadophora gregata.1,2 They are phytotoxins and
Figure 1. The Gregatins A−E have been known for decades.1 Their
original structures were revised twice.3,5
display antimicrobial activity.1 The structures initially attributed
to gregatin A and B1 (and to their congeners C−E1) were
synthesized by Clemo and Pattenden in 1982.3 They turned out
to be distinct from the natural products.3,4 The gregatin formulas
advocated thereupon3 yielded to a synthesis from our group in
2011, which targeted gregatin B.5 Yet our material differed from
the natural product, too. The rerevised scaffold of the gregatins,
which we put forward thereupon, was read from a side product,
which our misled approach to gregatin B had afforded in trace
amounts. This insight allowed reaching the gregatins A−D by
synthesis purposefully5b and assigning their 3D-structures
concomitantly.5b
Accessing the hexadiene side chain of the mentioned gregatins
from trans-H3C−CH2−CH2−CHCH−CHO and the
lithium enolate of Seebach’s pivaldehyde acetal of L-lactic acid6
gave a 93:7 mixture of the respective E,E- and Z,E-isomers.5b
This ratio increased to 98:2 when we processed trans-H3C−
Received: November 10, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ol5032602 | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Vilsmeyer reagent21 proved fastest and cleanest at activating the
mesylated carboxylic acid (2R,3S)-5. Employing it had the
additional benefit that the reaction progress could be monitored
by watching the solid disappear. The resulting carboxylic chloride
acylated the “dianion” of monomethyl malonate generated from
monomethyl malonate and 2 equiv of iPrMgCl in THF.22 It, too,
was a suspended solid. It dissolved gradually after the carboxylic
chloride was added, to proceed toward the β-ketoester (4R,5S)-6
(72% yield).
Scheme 1. Our Second Generation Retrosynthetic Analysis of
the Gregatins
The subsequent Knoevenagel-type condensation9 delivered
the furanone (1′S,5R)-7 under mildly basic conditions
(NaHCO3 in MeOH) in 85% yield (Scheme 2). Again, the
mesylate had not been affected. Initially, the enantiopurity of
(1′S,5R)-7 was 91% ee. This value equaled the ee which had
resulted from the Sharpless dihydroxylation 1 → (2R,3S)-2. To
our delight the ee of (1′S,5R)-7 became 97% after precipitating
crystals of rac-(1′S,5R)-7 (no rotatory power!) from the
supernatant oil. This improvement followed from an analysis
of the follow-up product (R)-10 of such an “optically enriched
specimen” of (1′S,5R)-7 by chiral HPLC.23
β-mesyloxy carboxylic acid with the “dianion” of monomethyl
malonate (bottom of Scheme 1).
Our synthesis began with a (DHQ)2PHAL-mediated
Sharpless dihydroxylation of benzyl tiglate (1; Scheme 2). It
The upper part of Scheme 3 summarizes our efforts at effecting
the β-elimination of 1′-OMs and 2′-H from the secondary
Scheme 2. Part I of Our Second Generation Route to
Gregatins: Obtaining the [(Mesyloxy)ethyl]furanone
(1′S,5R)-7
Scheme 3. Part II of our Second Generation Route to
Gregatins: Forming the Vinyl-Furanone (R)-10 without an
Interfering Ring Enlargement
established the diol (2R,3S)-2 with 91% ee.10 The secondary
hydroxy group was mesylated selectively.11 The acetylation of the
resulting tertiary alcohol (2R,3S)-3 was performed in pure Ac2O
[→ (2R,3S)-4]. AgOTf as a catalyst12 did not work reproducibly
whereas FeCl3 did (other than in a related derivatization,5b we
omitted a cosolvent). In contrast, attempted acetylations using
AcCl in CH2Cl2 with catalytic DMAP effected no acetylation at
all. The benzyl ester moiety in triester (2R,3S)-4 was cleaved
hydrogenolytically in THF13 [→ (2R,3S)-5].
The C2-elongation of the carboxylic acid (2R,3S)-5 to the β-
ketoester (4R,5S)-6 was difficult to realize (Scheme 2). This was
mostly because the mesylate had to be left intact.14 This limited
“allowed” conditions for activating the carboxylic acid (2R,3S)-5
chemoselectively and for engaging the resulting intermediate in a
chemoselective Claisen(-type) condensation. The last-men-
tioned need suggested using stabilized rather than simple
enolates.15 Experimenting accordingly we found that many
activating agents [(COCl)2,16 (COCl)2 + cat. DMF,17 CDI18]
and nucleophile/base pairs (potassium monomethyl malonate/
NEt3 + MgCl2,19 Meldrum’s acid/pyridine20) failed to extend the
acid (2R,3S)-5 to the β-ketoester (4R,5S)-6 or would achieve
that transformation irreproducibly. At long last a suspension of
mesylate (1′S,5R)-7 to reach the vinyl-substituted furanone (R)-
10. We failed no matter which base/solvent combination we
used:24 Either the unchanged mesylate (1′S,5R)-7 was reisolated
(up to quantitative yield) or a ring expansion to the pyranone
1125 occurred (up to 84% yield).26 The conversion of the
mesylate (1′S,5R)-7 into the vinyl-substituted furanone (R)-10
was accomplished via the small detour depicted in the lower part
of Scheme 3. We substitued the mesyloxy group with buffered27
phenyl selenide28 (→ 74% 12). With H2O2 in amine-free THF
we oxidized to the corresponding selenoxide(s). A β-
elimination,29 still in the absence of amines, was complete
within 3 h at rt. It furnished the vinyl-substituted furanone (R)-
10 in 81% yield.
The vinyl-substituted furanone (R)-10 was a Heck coupling
step short of reaching our target gregatins shown in Scheme 4. In
each case the nucleophilic component was the vinylfuranone
(R)-10. The electrophilic components were the (E)-1-iodobut-1-
ene 1430 or the (E)-3-hydroxy-1-iodobut-1-enes31 (S)-16
B
dx.doi.org/10.1021/ol5032602 | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Comprising 9 linear steps and affording a 15% overall yield or
more, our “second generation gregatin syntheses” disclosed in
Schemes 2−4 demonstrate the viability of the novel approach
outlined in Scheme 1. Future work will reveal whether the same
strategy suits synthesizing, and clarifying configurationally, the
following natural products as well: (+)-penicilliol A,38 (+)-pen-
icilliol B,38 (−)-graminin A,39 (+)-cyclogregatin,40 and (+)-huas-
penone A.41
Scheme 4. Part III of our Second Generation Route to
Gregatins: Converting the Vinyl-Furanone (R)-10 into
Gregatin B and the Dextrorotatory Diastereomers of
Gregatin E
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, characterization data, copies of NMR
spectra for all new compounds. HPLC diagrams for the
assignment of the stereostructure of natural gregatin E. This
material is available free of charge via the Internet at http://pubs.
acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: reinhard.brueckner@organik.chemie.uni-freiburg.de.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors thank the Studienstiftung des Deutschen Volkes for
a Ph.D. fellowship for F.W. We thank Ms. Rita M. DaSilva-Freitas
■
(>99% ee) and (R)-16 (>99% ee). The Heck coupling between
vinylfuranone (R)-10 (2.00 g) and 2.0 equiv of (E)-1-iodobut-1-
ene 14 furnished (+)-(R)-gregatin B (15, 1.26 g) in 49% yield
(97% ee). Its side-chain configurations were exclusively E. The
Heck couplings between the vinylfuranone (R)-10 and the (S)-
and (R)-enantiomer of (E)-3-hydroxy-1-iodobut-1-ene (16)
were conducted similarly but on a smaller (70 mg) scale. They
delivered the diastereomers (+)-(5R,5′S)-17 and (+)-(5R,5′R)-
17 of gregatin E in 58% and 61% yield, respectively. Again, the
side-chain configurations were exclusively E.
Whether our Heck coupling product (+)-(5R,5′S)-17 or its
diastereomer (+)-(5R,5′R)-17 equals natural gregatin E was not
revealed unequivocally by their specific rotations because they
are too alike: [α]2D5 = +154.6 (c = 0.78, CHCl3), [α]2D5 = +173.8 (c
= 0.78, CHCl3), and [α]D = +144 (c = 0.18, CHCl3),1a
respectively. Indeed, independent syntheses by another group,
which paralleled our own study, used a completely different
strategy, requiring 16 linear steps,7 which provided the same
diastereomers (+)-(5R,5′S)-17 {[α]1D5 = +165.5 (c = 0.60,
CHCl3)} and (+)-(5R,5′R)-17 {[α]1D5 = +165.7 (c = 0.54,
CHCl3)}.7 Their specific rotations had been even less indicative
than our values. As a consequence the other workers compared
the melting points of their compounds [(+)-(5R,5′S)-17: 87−88
°C;7 (+)-(5R,5′R)-17: 101−102 °C7] with that of the natural
product (mp 89−91 °C1a). Their inference was that
(+)-(5R,5′S)-17 equals gregatin E.32 We felt that this conclusion
was not safe beyond reasonable doubt.
(Universitat Freiburg) for tireless technical assistance and Dipl.-
̈
Biol. Eric Strittmatter (Universitat Freiburg) for his help
̈
cultivating C. gregata and for patiently answering biology
questions. Finally, we thank Prof. N. Kondo (Graduate School
of Agriculture, Hokkaido University, Sapporo) for helpful hints
regarding the cultivation of C. gregata.
REFERENCES
■
(1) (a) (First) Isolation of these compounds: Kobayashi, K.; Ui, T.
Tetrahedron Lett. 1975, 16, 4119−4122. (b) Naming of these
compounds: Kobayashi, K.; Ui, T. Physiol. Plant Pathol. 1977, 11,
55−60.
(2) Cadophora gregata was originally named Cephalosporinum gregatum
[Allington, W. B.; Chamberlain, D. W. Phytopathology 1948, 38, 793−
802; cited according to http://www.indexfungorum.org/Names/
NamesRecord.asp?RecordID=284846 (15.07.14)]. It was first renamed
to Phialophora gregata [Gams, W. Chephalosporinum-artige Schimmel-
pilze, 1971, ISBN: 3437301179, p 199; cited according to http://www.
indexfungorum.org/Names/NamesRecord.asp?RecordID=319845
(15.07.14)] and finally re-renamed to Cadophora gregata (Harrington,
T. C.; McNew, D. L. Mycotaxon 2003, 87, 141−151).
(3) (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.
(4) (a) Takeda, K.; Kubo, H.; Koizumi, T.; Yoshii, E. Tetrahedron Lett.
1982, 23, 3175−3178. (b) Miyata, O.; Schmidt, R. R. Tetrahedron Lett.
1982, 23, 1793−1796.
(5) (a) Burghart-Stoll, H.; Bruckner, R. Org. Lett. 2011, 13, 2730−
̈
We therefore tailored an unquestionable configurational proof
by the following measures: (1) We obtained the two stereo-
isomers of structure 17, which are missing in Scheme 4.33 (2)
The resulting set of all conceivable gregatin E candidates was
differentiated by chiral HPLC.34,35 (3) We cultivated Cadophora
gregata,36,37 extracted the culture broth, and isolated natural
gregatin E. (4) The latter compound and the four gregatin E
candidates were compared by chiral HPLC.35 The result was that
natural gregatin E is (5R,5′S)-17. This is identical with what the
earlier workers had suggested.7
2733. (b) Burghart-Stoll, H.; Bruckner, R. Eur. J. Org. Chem. 2012,
̈
3978−4017.
(6) (a) Seebach, D.; Naef, R. Helv. Chim. Acta 1981, 64, 2704−2708.
(b) Frat
4221−4224.
́
er, G.; Muller, U.; Gunther, W. Tetrahedron Lett. 1981, 22,
̈ ̈
(7) While our study was underway, (+)-gregatin E and (+)-epi-gregatin
E were synthesized: Kusakabe, T.; Kawai, Y.; Kato, K. Org. Lett. 2013, 15,
5102−5105.
(8) Heck couplings of vinyl compounds with 1-iodoalk-1-enes giving
acceptor-free (E,E)-1,3-dienes stereoselectively: (a) Jeffery, T. J. Chem.
C
dx.doi.org/10.1021/ol5032602 | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Soc., Chem. Commun. 1991, 324−325. (b) Blagg, B. S. J.; Jarstfer, M. B.;
Rogers, D. H.; Poulter, C. D. J. Am. Chem. Soc. 2002, 124, 8846−8853.
(c) Kalesse, M.; Chary, K. P.; Quitschalle, M.; Burzlaff, A.; Kasper, C.;
Scheper, T. Chem.Eur. J. 2003, 9, 1129−1136. (d) Pan, D.; Chen, A.;
Su, Y.; Zhou, W.; Li, S.; Jia, W.; Xiao, J.; Liu, Q.; Zhang, L.; Jiao, N.
Angew. Chem. 2008, 120, 4807−4810; Angew. Chem., Int. Ed. 2008, 47,
4729−4732. (e) Knowles, J. P.; O’Connor, V. E.; Whiting, A. Org.
Biomol. Chem. 2011, 9, 1876−1886. (f) Brandt, D.; Bellosta, V.; Cossy, J.
Org. Lett. 2012, 14, 5594−5597. (g) Maillefer-El Houar, S.; Uebelhart,
P.; Linden, A.; Hansen, H.-J. Helv. Chim. Acta 2013, 96, 1488−1541.
(9) (a) The Knoevenagel approach to dihydrofuranonecarboxylates is
from Kato, K.; Nouchi, H.; Ishikura, K.; Takaishi, S.; Motodate, S.;
Tanaka, H.; Okudaira, K.; Mochida, T.; Nishigaki, R.; Shigenobu, K.;
Akita, H. Tetrahedron 2006, 62, 2545−2554. Applications: (b) Reference
5b. (c) Reference 7.
(28) Initial conditions: Hamon, D. P. G.; Tuck, K. L. Tetrahedron 2000,
56, 4829−4835. EtOH was substituted by MeOH to prevent
transesterification.
(29) Initial conditions: Ley, S. V.; Tackett, M. N.; Maddess, M. L.;
Anderson, J. C.; Brennan, P. E.; Cappi, M. W.; Heer, J. P.; Helgen, C.;
Kori, M.; Kouklovsky, C.; Marsden, S. P.; Norman, J.; Osborn, D. P.;
́
Palomero, M. A.; Pavey, J. B. J.; Pinel, C.; Robinson, L. A.; Schnaubelt, J.;
Scott, J. S.; Spilling, C. D.; Watanabe, H.; Wesson, K. E.; Willis, M. C.
Chem.Eur. J. 2009, 15, 2874−2914.
(30) (E)-1-Iodobut-1-ene (16) was prepared from but-1-yne by
hydroalumination and Al/I exchange: Alexakis, A.; Duffault, J. M.
Tetrahedron Lett. 1988, 29, 6243−6246.
(31) The iodobutenols (S)- and (R)-15 were derived in 2 steps from
(S)- and (R)-but-3-yne-2-ol, respectively, by hydrostannylation (Lee,
T.; Kim, S. Tetrahedron: Asymmetry 2003, 14, 1951−1954) and Sn/I
exchange (Yu, L.; Cui, J.; Padakanti, P. K.; Engel, L.; Bagchi, D. P.;
Kotzbauer, P. T.; Tu, Z. Bioorg. Med. Chem. 2012, 20, 4625−4634).
(32) Our mp: 94 °C for (+)-(5R,5′S)-17, 103 °C for (+)-(5R,5′S)-17.
(33) (a) A Heck coupling analogous to b and c of Scheme 4 between
rac-10 and iodobutene rac-16 gave a 25:25:25:25 mixture of (5R,5′S)-
17, (5S,5′S)-17, (5R,5′R)-17, and (5S,5′R)-17. (b) The Heck coupling
b of Scheme 4 starting from vinylfuranone (R)-10 with 97% ee and
iodobutene (S)-16 with 99% ee should give a 98.0:1.5:0.5:0.0 mixture of
(5R,5′S)-17, (5S,5′S)-17, (5R,5′R)-17, and (5S,5′R)-17 (found:
98.7:0.6:0.7:0.0). (c) The Heck coupling c of Scheme 4 starting from
vinylfuranone (R)-10 with 97% ee and iodobutene (R)-16 with 99% ee
should give a 0.5:0.0:98.0:1.5 mixture of (5R,5′S)-17, (5S,5′S)-17,
(5R,5′R)-17, and (5S,5′R)-17 (found: 1.0:0.0:98.6:0.4). (d) Comparing
the 17-isomers from (b) vs (a) allowed identifying (5R,5′S)-17 (most
abundant isomer) and (5S,5′R)-17 (missing isomer). (e) Comparing
the 17-isomers from (c) vs (a) allowed identifying (5R,5′R)-17 (most
abundant isomer) and (5S,5′S)-17 (missing isomer).
(10) For dihydroxylating benzyl tiglate (1) in the presence of AD-mix
α and MeSO2NH2, >98% ee was reported: Shao, H.; Rueter, J. K.;
Goodman, M. J. Org. Chem. 1998, 63, 5240−5244.
(11) Reaction conditions adapted from: Heathcock, C. H.; Pirrung, M.
C.; Young, S. D.; Hagen, J. P.; Jarvi, E. T.; Badertscher, U.; Marki, H. P.;
Montgomery, S. H. J. Am. Chem. Soc. 1984, 106, 8161−8174. According
to the theoretical part NEt3 was used, but according to the experimental
part pyridine was used. We used less pyridine and MsCl.
(12) Das, R.; Chakraborty, D. Synthesis 2011, 1621−1625.
(13) Reaction conditions adapted from an analogous hydrogenolysis
by: Chen, D.-W.; Kubiak, R. J.; Ashley, J. A.; Janda, K. D. J. Chem. Soc.,
Perkin Trans. 1 2001, 2796−2803.
(14) The mesyloxy group in the carboxylic acid (2R,3S)-5 looks
sensitive to a β-elimination or might participate in a Grob fragmentation.
(15) A SciFinder search of 1- or 2-step C2 elongations of carboxylic
acids giving β-ketoesters by acylating malonic acid-based nucleophiles
gave 1159 hits for carboxylic acids with 2 H atoms at C-α, 543 hits for
exactly 1 H atom at C-α, and 131 hits without H atoms at C-α (17. 7.
2014).
(34) Chiralpak OD-3 column; n-heptane/iPrOH (98:2), 1.0 mL/min;
λdetector = 250 nm. The gregatin E candidates eluted in the order
(5R,5′R)-17 (tR = 44 min), (5R,5′S)-17 (tR = 48 min), (5S,5′R)-17 (tR =
69 min), and (5S,5′S)-17 (tR = 139 min).
(16) For example, used by: Xu, L.; Jin, J.; Lal, M.; Daublain, P.;
Newcomb, M. Org. Lett. 2007, 9, 1837−1840.
(35) HPLC diagrams: Supporting Information.
(17) For example, used by: Padwa, A.; Snyder, J. P.; Curtis, E. A.;
Sheehan, S. M.; Worsencroft, K. J.; Kappe, C. O. J. Am. Chem. Soc. 2000,
122, 8155−8167.
(36) We obtained Cadophora gregata (Phialophora gregata f. sp.
adzukicola T96-1,2 MAFF No. 241056) from the NIAS Genebank
Project of the National Institute of Agrobiological Sciences, Japan. This
particular strain was isolated and characterized by Kondo et al.:
(a) Kondo, N.; Fujita, S.; Murata, K.; Ogoshi, A. Plant Dis. 1998, 82,
928−930. (b) Kondo, N.; Shimada, H.; Fujita, S. J. Gen. Plant Pathol.
2009, 75, 181−187.
(18) For example, used by: Shimada, N.; Oohara, T.; Krishnamurthi, J.;
Nambu, H.; Hashimoto, S. Org. Lett. 2011, 13, 6284−6287.
(19) For example, used by: Mai, A.; Massa, S.; Rotili, D.; Simeoni, S.;
Ragno, R.; Botta, G.; Nebbioso, A.; Miceli, M.; Altucci, L.; Brosch, G. J.
Med. Chem. 2006, 49, 6046−6056.
(37) Details of the cultivation of C. gregata, workup, and
chromatography: Supporting Information.
(20) For example, used by: Hannah, D.; Knight, R. L.; Lock, C. J.;
Ortmans, I.; Owen, D. A.; Pegurier, C.; Raphy, G.; Watson, R. J. WO
2008/031556.
(21) For instance: Audoly, L. P. WO 2005/102389.
(22) The initial inspiration was from ref 18.
(23) Chiralpak AD-H column; n-heptane/iPrOH (97:3), 1.0 mL/min;
λdetector = 220 nm. The enantiomers of 10 eluted in the order (S)-10 (tR =
27 min) and (R)-10 (tR = 29 min).
(38) (a) Isolation and structure attribution: 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. (b) Proposed structure revision:
Reference 5b.
(39) (a) Isolation and structure attribution: Kobayashi, K.; Ui, T. J.
Chem. Soc., Chem. Commun. 1977, 774a−774a. (b) Proposed structure
revision: Reference 5b.
(40) (a) Isolation and structure attribution: Anke, H.; Casser, I.;
Schrage, M.; Steglich, W. J. Antibiot. 1988, 17, 1681−1684. (b)
Proposed structure revision: Reference 5b.
(41) (a) Isolation and structure attribution: Zhan, Z.-J.; Jin, J.-P.; Ying,
Y.-M.; Shan, W.-G. Helv. Chim. Acta 2011, 94, 1454−1458. (b)
Proposed structure revision: Reference 5b.
(24) Representative examples: Supporting Information.
(25) Compound 11 might be the first nonannulated and unbridged
ester of a 2,5,6-trialkylpyran-4-one-3-carboxylic acid. (a) Cyclohexane-
annulated esters thereof: Takeuchi, N.; Nakagawa, H.; Kamisato, M.;
Tobinaga, S. Chem. Pharm. Bull. 1980, 28, 2460−2467. (b) Lactones
thereof: Chantegrel, B.; Nadi, A. I.; Gelin, S. Synthesis 1982, 1107−
1109. (c) Bridged dimers thereof: Siddiq, M.; Khan, A. W. J. Park.
Chem. Soc. 1992, 14, 215−218.
(26) This undesired course may start with a β-elimination, which splits
the 1′-H and the 2-O bond. It would deliver the intermediate 8. Being
both an enolate and an enone, 8 might undergo an intramolecular oxa-
Michael addition. If the resulting enolate 9 expels a mesylate anion, the
pyranone 11 would be accomplished.
(27) When forming the selenide 12 the medium must be neutral such
that the base-promoted rearrangement to the pyranone 11 cannot
interfere.
D
dx.doi.org/10.1021/ol5032602 | Org. Lett. XXXX, XXX, XXX−XXX