Enantioselective synthesis and stereochemical revision of communiols A-C

Article abstract of DOI:10.1016/j.tetlet.2005.07.041

Enantioselective synthesis of the proposed structure of communiol C, an antibacterial tetrahydrofuran derivative produced by Podospora communis, and its stereoisomers revealed that the genuine stereochemistry of communiol C should be 3R, 5R, and 6S. Two o

Full text of DOI:10.1016/j.tetlet.2005.07.041

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 6297–6300
Enantioselective synthesis and stereochemical revision of
communiols A–C
Shigefumi Kuwahara* and Masaru Enomoto
Laboratory of Applied Bioorganic Chemistry, Graduate School of Agricultural Science, Tohoku University,
Tsutsumidori-Amamiyamachi, Aoba-ku, Sendai 981-8555, Japan
Received 4 June 2005; revised 8 July2005; accepted 11 July2005
Available online 27 July2005
Abstract—Enantioselective synthesis of the proposed structure of communiol C, an antibacterial tetrahydrofuran derivative pro-
duced by Podospora communis, and its stereoisomers revealed that the genuine stereochemistryof communiol C should be 3 R,
5R, and 6S. Two other structurallyrelated metabolites of the same microbial origin, communiols A and B, were also synthesized
based on the revised stereochemistry.
Ó 2005 Elsevier Ltd. All rights reserved.
Communiols A–D (1–4, Fig. 1) were recentlyisolated
from the culture broth of the coprophilous (dung-colo-
nizing) fungus, Podospora communis, byGloer and
co-workers as its metabolites exhibiting significant anti-
bacterial activityagainst Bacillus subtilis and Staphylo-
coccus aureus.1 The 2,4-disubstituted tetrahydrofuran
substructure incorporated in 1–3 is relativelyrare as a
structural unit of natural products and displays a char-
acteristic difference in substitution pattern from 2,5-
disubstituted tetrahydrofurans frequently found in
annonaceous acetogenins2 or ionophores.3 To the best
of our knowledge, the most structurallysimilar natural
product seems to be aureonitol,4 a fungal metabolite
possessing a 2,4-dialkadienyl-substituted tetrahydrofu-
ran framework, although it has an additional hydroxyl
substituent at its C3 position. The 3,7-disubstituted
2,8-dioxabicyclo[3.3.0]octane structure contained in
communiol D (4) is also rare, although some related
structural units analogous to but different in substitu-
tion and/or oxidation patterns from the bicyclic portion
of 4 have been found in manynatural products such as
clerodane diterpenoids and fungal metabolites.5 These
structural uniqueness of communiols A–D, coupled with
their interesting biological activity, prompted us to
embark on the synthesis of 1–4. We describe herein,
our studies on the enantioselective synthesis of commun-
iols A–C, which led us to the conclusion that the stereo-
chemistryof communiols A–C should be revised.
O
O
H
H
HO2C
HO2C
HO
OH
OH
Our synthesis of communiol C (3), chosen as our first
synthetic target due to its structural simplicity, began
with the Sharpless asymmetric dihydroxylation6,7 of
known olefinic ester 58,9 using AD-mix-a as the chiral
catalyst (Scheme 1). Exposure of the resulting crude
product consisting of diol 6 and its lactonization prod-
uct 7 to acidic conditions brought about complete con-
communiol A (1)
communiol B (2)
H
O
O
H
O
H
HO2C
H
OH
communiol C (3)
OH
communiol D (4)
1
version of 6 into 7, whose H NMR spectrum was in
Figure 1. Originallyproposed structures for communiols A–D.
good agreement with that reported for an authentic
sample of 7 previouslyprepared from L-glutamic acid.10
The absolute stereochemistryof 7 was confirmed by
Keywords: Communiol; Antibacterial; Enantioselective synthesis;
Asymmetric dihydroxylation; Podospora communis.
22
comparison of its specific rotation value (½aŠ +40.3
D
22
*
(c 2.25, CH2Cl2)) with the literature value (½aŠ
+
Corresponding author. Tel./fax: +81 22 717 8783; e-mail:
skuwahar@biochem.tohoku.ac.jp
46.0 (c 2.0, CH2Cl2)),10 and the enantiomeric exDcess
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.07.041

6298
S. Kuwahara, M. Enomoto / Tetrahedron Letters 46 (2005) 6297–6300
and 9-H2 appeared at d 3.85, 3.34, and 4.06/3.49, respec-
OH
a
tively, while the corresponding peaks of natural com-
muniol C were observed at d 3.90, 3.68, and 4.09/3.41.
In their report on the structural determination of com-
muniols A–D, Gloer and co-workers determined the
trans-relative stereochemistrybetween the C3- and C5-
substituents of communiol C byobserving some clear
diagnostic NOESY correlations, and assigned the abso-
lute configuration at the C6 chiral center to be S by
analogywith the ( S)-absolute configuration of commun-
iol A (1), which in turn was established unambiguously
EtO2C
b
EtO2C
OH
5
6
O
O
e
H
R1
OR2
7: R1 = H, R2 = H
c
8: R1 = H, R2 = TBDPS
d
9: R1 = CH2=CHCH2, R2 = TBDPS
1
R1
bythe modified Mosher method. The relative stereo-
9
O
O
chemistrybetween C5 and C6 of communiol C was,
however, proposed onlyon the basis of Born ꢀs empirical
rule,16 which has been used for determining the relative
g
H
H
HO2C
h
6
3
5
OR2
OR
0
stereochemistrybetween C2 and C1 stereogenic centers
of
10: R1 = OH, R2 = TBDPS
11: R1 = H, R2 = TBDPS
12: R = TBDPS
3: R = H
2-(10-hydroxyalkyl)tetrahydrofuran
system.17,18
f
According to the rule, the C10 signal is observed at ca.
13
74 ppm in C NMR when the C2/C10-relative stereo-
i
chemistryis threo, while that of erythro-isomer appears
at ca. 72 ppm. The observed chemical shift (d 73.7) for
the C6 carbon of natural communiol C led them to pro-
pose its C5/C6-relative stereochemistryto be threo as
represented bystructure 3. To the best of our knowl-
edge, however, Bornꢀs rule has not been applied to the
determination of the C2/C10-relative stereochemistry
of 2,4-disubstituted tetrahydrofurans like communiols
A–C, which made us to suppose that the genuine C5/
C6-relative stereochemistryof communiol C might not
be threo, but erythro. Based on this presumption, we
synthesized the 5,6-erythro-stereoisomer of 3 (i.e., 6-
epi-3 in Scheme 1) from intermediate 11 bya four-step
sequence consisting of deprotection of the silyl protec-
tive group (11 ! 13), the Mitsunobu inversion of the
resulting alcohol to form the corresponding p-nitro-
benzoate derivative (13 ! 14),19 oxidative cleavage of
the double bond of 14 to carboxylic acid 15, and finally,
alkaline hydrolysis of the ester functionality of 15 to fur-
O
O
H
k
H
HO2C
OR
OR
15: R = PNB
6-epi-3: R = H
13: R = H
14: R = PNB
j
l
Scheme 1. Synthesis of the originally proposed structure for com-
muniol C (3) and its C6-epimer (6-epi-3). Reagents and conditions: (a)
AD-mix-a, CH3SO2NH2, t-BuOH/H2O, 0 °C, 12 h; (b) TsOHÆH2O,
CH2Cl2, rt, 1 h (94%, two steps); (c) TBDPSCl, imidazole, DMF, rt,
21 h (87%); (d) LDA, CH2@CHCH2Br, THF/HMPA, ꢀ78 °C, 30 min
(64%); (e) DIBAL, CH2Cl2, ꢀ78 °C, 30 min; (f) Et3SiH, BF3ÆOEt,
CH2Cl2, ꢀ78 to ꢀ5 °C, 8 h (86%, two steps); (g) RuCl3Æ(H2O)n, NaIO4,
H2O/CH3CN/CCl4, rt, 2 h (63%); (h) aq HF, CH3CN, rt, 22 h (67%);
(i) TBAF, THF, rt, 2 days (66%); (j) DEAD, Ph3P, p-NO2C6H4CO2H,
toluene, rt, 2 days (59%); (k) RuCl3Æ(H2O)n, NaIO4, H2O/CH3CN/
CCl4, rt, 2.5 h; (l) aq K2CO3, rt, 24 h (36%, two steps).
of 7 was estimated to be 93.5% byanalyzing the 1H
NMR spectra of the corresponding (R)- and (S)-MTPA
esters.11 After protection of the hydroxyl group of 7 as
its t-butyldiphenylsilyl (TBDPS) ether 8, the lactone
was subjected to well-documented trans-selective alkyl-
ation with allyl bromide,12,13 which afforded a separable
8.3:1 mixture of desired product 9 and the correspond-
ing cis-allylation product. When the protective group
was changed into t-butyldimethylsilyl (TBS), the reac-
tion showed a much lower selectivityof 2.3:1, probably
reflecting the smaller steric bulkiness of TBS as com-
1
nish 6-epi-3. As expected, the H NMR spectrum of 6-
epi-3 was exactlythe same as that of natural communiol
C, which enabled us to establish the relative stereochem-
istryof communiol C as 3,5- trans and 5,6-erythro. Com-
parison of the specific rotation value of 6-epi-3 with that
22
of natural communiol C (½aŠ +3.6 (c 0.24, CH2Cl2)
D
and [a]D ꢀ3.4 (c 0.142, CH2Cl2),1 respectively) as well
as the newlyestablished relative stereochemistryof com-
muniol C led us to the conclusion that Gloerꢀs assign-
ment of the (6S)-absolute configuration for natural
communiol C was correct, but the genuine structure of
communiol C should be revised to ent-6-epi-3 (the enan-
tiomer of 6-epi-3, see Scheme 2).
pared to TBDPS. The lactone 9 was purified bySiO -
2
column chromatographyand then reduced with DIBAL
to afford lactol 10 as an approximately3:1 mixture of
epimers. Reductive removal of the newlygenerated
hydroxyl group of 10 with triethylsilane in the presence
of BF3ÆOEt214 proceeded smoothlyto give tetrahydrofu-
ran derivative 11, the double bond of which was then
cleaved oxidativelyto give carboxylic acid 12.15 Finally,
deprotection of the TBDPS group with aq HF in aceto-
nitrile completed the synthesis of 3, the proposed struc-
ture of communiol C. Direct comparison of the 1H
NMR spectrum of synthetic 3 with that of natural com-
muniol C, however, revealed some clear differences,
especiallyin the chemical shifts for 5-H, 6-H, and 9-
H2. In the synthetic sample 3, the peaks for 5-H, 6-H,
According to the revised stereochemistryof communiol
C, we set about the synthesis of ent-6-epi-3 (Scheme 2).
Byfollowing the same set of reactions as employed for
the preparation of 7 except that AD-mix-b, instead of
AD-mix-a, was used for the asymmetric dihydroxylation
of 5,6 olefinic ester 5 was converted into hydroxy lactone
22
ent-7 (½aŠ ꢀ41.3 (c 2.25, CH2Cl2)), whose enantiomeric
D
excess was determined to be 96% bythe same method as
used for 7. Protection of its hydroxyl group as a TBDPS
ether gave rise to ent-8 as a white crystalline solid. A sin-
gle recrystallization of the solid from hexane/EtOAc

S. Kuwahara, M. Enomoto / Tetrahedron Letters 46 (2005) 6297–6300
6299
CH2Cl2)) were inconsistent with those of natural
communiol A ([a]D ꢀ1.6 (c 0.25, CH2Cl2)) and natural
communiol B ([a]D ꢀ95 (c 0.075, CH2Cl2)), respec-
tively.1 At present, we are unable to clearlyexplain these
discrepancies in specific rotation, but small amounts of
impurities contained in the synthetic and/or natural
samples of communiols A and B might have affected
the observed specific rotation values.
O
a
c
O
O
H
H
5
OR
ent-14 OPNB
ent-7 R = H
ent-8 R = TBDPS
d
b
e
O
H
O
H
HO2C
OHC
OH
16 OPNB
In summary, the enantioselective total syntheses of the
originallyproposed structure ( 3) for communiol C, its
C6-epimer (6-epi-3), and (3R,5R,6S)-stereoisomer (ent-
6-epi-3) were accomplished starting from known olefinic
ester 5 by using the Sharpless asymmetric dihydroxyl-
ation as the source of chirality, which revealed that the
genuine stereochemistryof communiol C should be rep-
resented by ent-6-epi-3. Based on this newlyestablished
stereochemistryof communiol C as well as the assump-
tion that two other structurallyrelated metabolites of
the same microbial origin, communiols A and B, should
share the same stereochemical arrangement as commun-
iol C, ent-8-epi-1 and ent-8-epi-2 were synthesized as
highlyprobable candidates for the genuine structures
ent-6-epi-3
f, g
O
H
O
H
h
HO2C
HO2C
8
5
7
OH
OH
ent-8-epi-1
ent-8-epi-2
Scheme 2. Synthesis of the revised structures for communiols A–C.
Reagents and conditions: (a) (i) AD-mix-b, CH3SO2NH2, t-BuOH/
H2O, 0 °C, 12 h; (ii) TsOHÆH2O, CH2Cl2, rt, 1 h (89%, two steps); (b)
(i) TBDPSCl, imidazole, DMF, rt, 19 h (quant); (ii) recrystallization
from hexane/EtOAc (75%); (c) steps (d)–(f) in Scheme 1 (71%, three
steps); (d) steps (i)–(l) in Scheme 1 (41%, four steps); (e) O3, MeOH,
ꢀ78 °C, 5 min, then Me2S, ꢀ78 °C to rt, 2 h (85%); (f)
Ph3P@CHCO2Et, CH2Cl2, rt, 9 h (82%); (g) aq LiOH, THF, rt, 20 h
(95%); (h) H2, 10% Pd–C, EtOH, rt, 1 h (96%).
1
of communiols A and B, respectively. Although the H
and 13C NMR data of each synthetic sample were exactly
the same as those of the corresponding natural sample,
their specific rotation values showed inexplicable dis-
crepancies. We feel the need for remeasurement of the
specific rotation values of natural communiols A and B.
yielded enantiomerically pure ent-8 (mp 62.5–63 °C),
whose optical integritywas checked byanalyzing the
1H NMR spectra of the corresponding (R)- and (S)-
MTPA esters, which in turn were obtained bytreatment
of the opticallyenriched silyl ether with TBAF followed
Acknowledgements
by( R)- and (S)-MTPA-esterifications of the resulting
22
D
alcohol (ent-7, ½aŠ ꢀ46.8 (c 0.24, CH2Cl2)). The lactone
We are grateful to Professor Gloer (Universityof Iowa)
for valuable discussions and for providing the copies of
the NMR spectra of natural communiols A–C and
related material. We also thank Ms. Yamada (Tohoku
University) for measuring NMR and MS spectra. This
work was supported, in part, bya Grant-in-Aid for
Scientific Research (B) from the Ministryof Education,
Culture, Sports, Science, and Technologyof Japan (No.
16380075).
ent-8 was then converted into ent-6-epi-3 via ent-14 by
the same seven-step sequence as employed for the syn-
thesis of 6-epi-3. The H and 13C NMR spectra of ent-
1
6-epi-3 were identical with those of natural communiol
22
D
C, and its specific rotation value (½aŠ ꢀ2.7 (c 1.155,
CH2Cl2)) was in good agreement with that of natural
communiol C ([a]D ꢀ3.4 (c 0.142, CH2Cl2))1 including
the minus sign. Based on these results, we concluded
that the genuine structure of communiol C should be
ent-6-epi-3 as depicted in Scheme 2.
References and notes
Assuming that the structurallyrelated tetrahydrofuran
derivatives (communiols A and B) of the same microbial
origin should have the same stereochemical arrangement
as communiol C, we started the synthesis of (5S,7R,8S)-
communiol A (ent-8-epi-1) and (5S,7R,8S)-communiol
B (ent-8-epi-2) from ent-14. Ozonolysis of the double
bond of ent-14 gave aldehyde 16, the chain elongation
of which bythe Wittig reaction afforded ent-8-epi-2 after
hydrolysis of the PNB ester group. Catalytic hydrogena-
tion of ent-8-epi-2 completed the synthesis of ent-8-epi-1.
1. Che, Y.; Gloer, J. B.; Scott, J. A.; Malloch, D. Tetra-
hedron Lett. 2004, 45, 6891–6894.
2. Bermejo, A.; Figadere, B.; Zafra-Polo, M.-C.; Barrachina,
I.; Estornell, E.; Cortes, D. Nat. Prod. Rep. 2005, 22, 269–
303.
3. Wierenga, W. In The Total Synthesis of Natural Products;
ApSimon, J., Ed.; Wiley: New York, 1981; Vol. 4, pp 263–
351.
4. Abraham, W.-R.; Arfmann, H.-A. Phytochemistry 1992,
31, 2405–2408.
5. Lorenzo, E.; Alonso, F.; Yus, M. Tetrahedron 2000, 56,
1745–1757, and references cited therein.
6. Kolb, H. C.; VanNiewenhze, M. S.; Sharpless, K. B.
Chem. Rev. 1994, 94, 2483–2547.
7. Avedissian, H.; Sinha, S. C.; Yazbak, A.; Sinha, A.;
Neogi, P.; Sinha, S.; Keinan, E. J. Org. Chem. 2000, 65,
6035–6051.
8. Odinokov, V. N.; Vakhidov, R. R.; Shakhmaev, R. M.;
Zorin, V. V. Khim. Prir. Soedin. 1996, 936–939.
1
The H and 13C NMR spectra of ent-8-epi-1 and ent-8-
epi-2 were exactlythe same as those of natural commun-
iol A and communiol B, respectively, which enabled us
to revise the originallyproposed 7,8- threo-relative
stereochemistryof natural communiols A and B to
7,8-erythro relationship. Curiouslyenough, however,
22
the specific rotation values of ent-8-epi-1 (½aŠ +1.3 (c
D
22
0.22, CH2Cl2)) and ent-8-epi-2 (½aŠ +4.7 (c 1.0,
D

6300
S. Kuwahara, M. Enomoto / Tetrahedron Letters 46 (2005) 6297–6300
`
9. Bruyere, D.; Bouyssi, D.; Balme, G. Tetrahedron 2004, 60,
15. Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K.
B. J. Org. Chem. 1981, 46, 3936–3938.
4007–4017.
10. Larcheveˆque, M.; Lalande, J. Bull. Soc. Chim. Fr. 1987,
116–122.
11. Kuwahara, S.; Ishikawa, J.; Leal, W. S.; Hamade, S.;
Kodama, O. Synthesis 2000, 1930–1935.
12. Sells, T. B.; Nair, V. Tetrahedron 1994, 50, 117–138.
16. Born, L.; Lieb, F.; Lorentzen, J. P.; Moeschler, H.;
Nonfon, M.; So¨llner, R.; Wendisch, D. Planta Med. 1990,
56, 312–316.
´
17. Chevez, D. C.; Acevedo, L. A.; Mata, R. J. Nat. Prod.
1998, 61, 419–421.
´
13. Bouchard, H.; Soulie, J.; Lallemand, J. Y. Tetrahedron
18. Fujimoto, Y.; Murasaki, C.; Shimada, H.; Nishioka, S.;
Kakinuma, K.; Singh, S.; Singh, M.; Gupta, Y. K.; Sahai,
M. Chem. Pharm. Bull. 1994, 42, 1175–1184.
19. Mitsunobu, O. Synthesis 1981, 1–28.
Lett. 1991, 32, 5957–5958.
14. Bruckner, C.; Hiltrud, H.; Reissig, H.-U. J. Org. Chem.
¨
1988, 53, 2450–2456.