Synthesis of the Proposed Structure of Cryptoconcatone H

Article abstract of DOI:10.1055/s-0037-1609657

A synthesis of the proposed structure of cryptoconcatone H and one diastereoisomer has been achieved, featuring a tandem deprotection-oxa-Michael addition to establish the tetrahydropyran moiety. Comparison of the NMR spectroscopic data confirms that the original structural assignment was incorrect.

Full text of DOI:10.1055/s-0037-1609657

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
© Georg Thieme Verlag Stuttgart · New York
2018, 29, A–C
letter
A
en
D. Csókás, R. W. Bates
Letter
Synlett
Synthesis of the Proposed Structure of Cryptoconcatone H
Dániel Csókás
OH
O
Me Me
metathesis
Roderick W. Bates*
O
O
O
Grignard
CHO
CHO
Ph
Ph
O
Ph
Division of Chemistry and Biological Chemistry, School of
Physical and Mathematical Sciences, Nanyang Technological
University, 21 Nanyang Link, Singapore 637371
roderick@ntu.edu.sg
H
H
metathesis
aldol
allylation
oxa-Michael
Received: 10.10.2018
Accepted after revision: 05.11.2018
Published online: 11.12.2018
To the best of our knowledge, however, there is no re-
port of the synthesis of the proposed structure 1 of crypto-
concatone H. We wish to report here our synthesis and
comparison with the reported spectroscopic data.
DOI: 10.1055/s-0037-1609657; Art ID: st-2018-d0654-l
Abstract A synthesis of the proposed structure of cryptoconcatone H
and one diastereoisomer has been achieved, featuring a tandem depro-
tection–oxa-Michael addition to establish the tetrahydropyran moiety.
Comparison of the NMR spectroscopic data confirms that the original
structural assignment was incorrect.
The reaction between cinnamaldehyde and the acetyl
thiazolidinone 26 mediated by titanium tetrachloride and
diisopropylethyl amine at –78 °C gave the β-hydroxy car-
bonyl product 3 with a d.r. of 7:1.7 We note that it is import-
ant to use only 1.2 equivalents of the titanium reagent, oth-
erwise the diastereoselectivity is seriously eroded. Facile
displacement of the thiazolidinyl moiety with N,O-dimeth-
ylhydroxylamine gave the Weinreb amide 48 which reacted
efficiently with an excess of allylmagnesium bromide to
give the labile β-hydroxyketone 5 (Scheme 1). This could be
reduced to the anti-diol 6 on treatment with tetrameth-
ylammonium triacetoxyborohydride according to the
method of Evans.9 This reaction proceeded with extremely
high diastereoselectivity as no syn isomer was detectable by
400 MHz 1H NMR spectroscopy. Protection of the diol as its
acetonide 7 under very mildly acidic conditions then al-
lowed selective cross-metathesis with crotonaldehyde at
the less hindered alkene moiety, delivering the unsaturated
aldehyde 8 in good yield.10 As in our synthesis of cyano-
lide,2b,11 tandem deprotection–intramolecular oxa-Michael
addition could then be effected under acidic conditions.
Aqueous hydrochloric acid was found to be an effective cat-
alyst for this transformation. The resulting tetrahydropyran
9 was found to be exclusively the cis isomer by analysis of
the coupling constants of the two protons α to the THP oxy-
gen atom. Although these appeared as multiplets in the
Key words tetrahydropyran, Michael addition, metathesis, stereo-
chemistry, tandem reaction
Recently, Luo et al. reported the isolation, structure, and
biological activity of a series of compounds from Crypto-
carya concinna.1 One of these, cryptoconcatone H, was as-
signed a cis-tetrahydropyran structure 1a (Figure 1). This
compound attracted our interest due to our experience in
tetrahydropyran synthesis and, in particular, our experi-
ence in the synthesis of 2,6-cis tetrahydropyrans by intra-
molecular oxa-Michael addition.2,3 Following Luo’s report,
Pilli et al. challenged the stereochemical assignment based
upon computational analysis of the H NMR data.4,5 They
predicted an alternative structure 2 for cryptoconkatone H
with, in particular, a trans- rather than cis-tetrahydropyran
ring. Indeed, synthesis of one of the predicted structures
1
1
provided material with H NMR spectroscopic data closely
matching that reported for the natural product.4
OH
O
O
O
1a
1
simple H NMR spectrum of tetrahydropyran 9, homonu-
O
H
H
clear decoupling revealed that each had a ca. 11 Hz coupling
constant with one of the respective ring β-protons.12 This is
consistent with an axial-axial arrangement of ring protons.
To begin to establish the lactone moiety, after protection of
the alcohol group, aldehyde 10 was subjected to allylation.
Allylation with both substrate and catalyst control of ste-
reochemistry was assessed. Barbier allylation using either
Zn or In gave modest stereoselectivity in favor of the anti
OH
O
2
O
H
H
Figure 1 Original and reassigned structures of cryptoconcatone H
© Georg Thieme Verlag Stuttgart · New York — Synlett 2018, 29, A–C

B
D. Csókás, R. W. Bates
Letter
Synlett
isomer 11b (Table 1, entries 1 and 2). Barbier allylation can
proceed with high diastereoselectivity under chelation con-
trol,13 this was not the case for aldehyde 10. Use of TiCl4 to
mediate allylation lead to decomposition (Table 1, entry 3).
Surprisingly, both Keck and Carreira allylation using a
Ti(IV)/BINOL combination were also unselective, although
the milder titanium reagents employed did ensure that
there was little or no decomposition (Table 1, entries 4–7).
For Keck allylation (Table 1, entries 6 and 7), no reaction
was observed when allyl tri-n-butyltin was employed as
the nucleophile, while tetraallyltin proved competent.14 Cu-
riously, the same ratio of diastereoisomers was obtained re-
gardless of whether (R)- or (S)-BINOL was employed as the
chiral ligand, indicating overriding substrate control of this
addition. In contrast, Krische allylation using allyl acetate
and an iridium-BINAP catalyst yielded the desired alcohols
11 with better selectivity (Table 1, entries 8 and 9).15 It may
be that the titanium Lewis acids are highly prone to chela-
tion by this particular substrate, resulting in enhancement
of substrate control over reagent control. The softer and
more highly ligated iridium center of the Krische system
may be less prone to this effect and this may be the reason
that this system gives better selectivity. Despite the gener-
ally modest selectivities obtained, the two diastereoisomers
could be chromatographically separated. Acylation of the
alcohol group of diastereoisomer 11a with acryloyl chloride
smoothly gave triene 12a, which underwent ring-closing
metathesis to give lactone 13a. A final desilylation complet-
ed the synthesis and delivered 1a in 70% yield over two
steps. The minor diastereoisomer from the allylation reac-
tion, alcohol 11b, was converted into the diastereoisomeric
structure 1b by an analogous series of transformations.
OH
O
S
O
S
MeNH(OMe)⋅HCl,
imidazole
Ph
N
CHO
+
TiCl4, i-Pr2NEt
S
N
Ph
Ph
S
67%
67%
3
Bn
2
Bn
OH
O
MgCl
OH
O
Me4NBH(OAc)3
AcOH, MeCN
THF –20 °C
98%
Ph
NMeOMe
5
79%
4
Me Me
CHO
OH OH
O
O
Grubbs II (5 mol%)
toluene, N2
Me2C(OMe)2,
ppts, acetone
Ph
Ph
6
77%
70%
7
OR
Me Me
O
O
O
HCl, H2O, THF
86%
9
R = H
TBSCl, imidazole
68%
CHO
Ph
O
Ph
10 R = TBS
H
H
8
allylation
(see Table 1)
OTBS
OTBS
OH
OH
+
Ph
O
H
Ph
O
H
H
H
11a
11b
ClOC
ClOC
68%
75%
NEt3
NEt3
OTBS
O
OTBS
O
O
O
Ph
O
H
Ph
O
H
H
H
12a
12b
Grubbs I (10 mol%)
CH2Cl2, N2
Grubbs I (10 mol%)
CH2Cl2, N2
OR
O
OR
O
O
O
Ph
O
Ph
O
H
H
H
H
13a R = TBS
1a R = H
HF⋅py
13b R = TBS
1b R = H
HF⋅py
70%
70%
(two steps)
(two steps)
Scheme 1 Synthesis of the proposed structure of cryptoconcatone H and an epimer
© Georg Thieme Verlag Stuttgart · New York — Synlett 2018, 29, A–C

C
D. Csókás, R. W. Bates
Letter
Synlett
Table 1 Diastereoselective Allylation of Aldehyde 10
Entry
Catalyst/reagent
Allylating reagent
Temp
(°C)
Time
(h)
Conversion
(%)
Yield
Diastereoselectivity
(%, combined) 11a:11b
1
2
3
4
5
6
7
8
Barbier: Zn, NH4Cl
Barbier: In, AcOH
TiCl4
allyl bromide
allyl bromide
tetraallyltin
RT
ON
ON
0.25
ON
ON
3
100
100
dec.
100
100
100
100
70
84
–
0.5:1
0.4:1
–
–20
0 to RT
0 to RT
0 to RT
–40
–
TiF4 (10 mol%), (R)-BINOL
TiF4 (10 mol%), (R)-BINOL
Ti(Oi-Pr)4, (R)-BINOL
Ti(Oi-Pr)4, (S)-BINOL
allyltrimethylsilane
tetraallyltin
–
0.6:1
0.6:1
0.6:1
0.6:1
8:1
–
tetraallyltin
92
92
62
tetraallyltin
–40
5
(R)-BINAP, [Ir(COD)Cl]2,
4-Cl-3-NO2-BzOH, Cs2CO3
allyl acetate
100
27.5
9
(S)-BINAP, [Ir(COD)Cl]2,
4-Cl-3-NO2-BzOH, Cs2CO3
allyl acetate
100
14
80
–
0.4:1
The 1H NMR spectra of the synthetic lactones 1a and 1b
showed only small differences in the 6.0–7.5 ppm and 3.5–
4.8 ppm regions. Comparison of either of the synthetic iso-
mers with the data reported for the natural product1 shows
a considerable similarity in the 6.0–7.5 ppm region, but a
number of very distinct differences in the region of 3.6–4.8
ppm. This region corresponds to the signals for the protons
α to the oxygen atoms.12 This is entirely consistent with the
proposal of Pill et al.4 that the natural product differs in the
stereochemistry of the tetrahydropyran ring, particularly
that it has 2,6-trans, rather than 2,6-cis stereochemistry.
These differences clearly confirm that the structure origi-
nally assigned to the natural product is incorrect. This re-
sult provides final confirmation of the structural reassign-
ment by Pilli et al. and further illustration of how the oxa-
Michael addition provides a facile and highly stereoselec-
tive route to cis-tetrahydropyrans.
(2) For some recent examples from this laboratory, see: (a) Bates, R.
W.; Wang, K.; Zhou, G.; Kang, D. Z. Synlett 2015, 26, 751.
(b) Bates, R. W.; Lek, T. G. Synthesis 2014, 46, 1731. (c) Bates, R.
W.; Song, P. Synthesis 2010, 2935.
(3) For reviews, see: (a) Nasir, N. M.; Ermanis, K.; Clarke, P. A. Org.
Biomol. Chem. 2012, 12, 3323. (b) Fuwa, H. Heterocycles 2012,
86, 1255. (c) Nising, C. F.; Bräse, S. Chem. Soc. Rev. 2012, 41, 988.
(d) Larossa, I.; Romea, P.; Urpí, F. Tetrahedron 2008, 64, 2683.
(e) Clarke, P. A.; Santos, S. Eur. J. Org. Chem. 2006, 2045. For a
partial review, see: (f) Hu, J.; Bian, M.; Ding, H. Tetrahedron Lett.
2016, 57, 5519.
(4) Della-Felice, F.; Sarotti, A. M.; Pilli, R. A. J. Org. Chem. 2017, 82,
9191.
(5) For a discussion of recent trends in structural revision, see:
Chhetri, B. K.; Lavoie, S.; Sweeney-Jones, A. M.; Kubanek, J. Nat.
Prod. Rep. 2018, 35, 514.
(6) Crimmins, M. T.; Christie, H. S.; Hughes, C. O. Org. Synth. 2011,
88, 364.
(7) Yadav, J. S.; Ganganna, B.; Bhunia, D. C. Synthesis 2012, 44, 1365.
(8) Crimmins, M. T.; King, B. W.; Tabet, E. A.; Chaudhary, K. J. Org.
Chem. 2001, 66, 894.
(9) Evans, D. A.; Chapman, K. T.; Carreira, E. M. J. Am. Chem. Soc.
1988, 110, 3560.
Funding Information
(10) Only a trace of cinnamaldehyde could be detected in the reac-
tion mixture.
We thank Nanyang Technological University for support of this work.
()
(11) For another example of tandem deprotection–cyclisation, see:
Paterson, I.; Haslett, G. W. Org. Lett. 2013, 15, 1338.
(12) See Supporting Information for more details.
(13) For examples, see: Podlech, J.; Maier, T. C. Synthesis 2003, 633;
for an additional example, see ref. 2b.
Supporting Information
Supporting information for this article is available online at
https://doi.org/10.1055/s-0037-1609657.
S
u
p
p
orit
n
gInformati
o
n
S
u
p
p
orti
n
gInformati
o
n
(14) Keck, G. E.; Tarbet, K. H.; Geraci, L. S. J. Am. Chem. Soc. 1993, 115,
8467.
(15) Wang, G.; Krische, M. J. J. Am. Chem. Soc. 2016, 138, 8088.
References
(1) Yang, B.-Y.; Kong, L.-Y.; Wang, X.-B.; Zhang, Y.-M.; Li, R.-J.; Yang,
M.-H.; Luo, J.-G. J. Nat. Prod. 2016, 79, 196.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2018, 29, A–C