Total Synthesis of Stemoamide, 9a- epi -Stemoamide, and 9a,10- epi -Stemoamide: Divergent Stereochemistry of the Final Methylation Steps

Article abstract of DOI:10.1055/s-0040-1707201

Total syntheses of stemoamide, 9a- epi -stemoamide, and 9a,10- epi -stemoamide by a convergent A + B ring-forming strategy is reported. The synthesis required a diastereoselective late-stage methylation of the ABC stemoamide core that successfully enabled access to three of the four possible diastereomeric structures. For the natural stemoamide series, the diastereoselectivity can be rationalized both by kinetic and thermodynamic arguments, whereas for the natural 9a- epi -stemoamide series, the kinetic selectivity is explained by the prepyramidalization of the relevant enolate.

Full text of DOI:10.1055/s-0040-1707201

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
© Georg Thieme Verlag Stuttgart · New York
2020, 31, A–F
letter
A
en
J. H. Siitonen et al.
Letter
Synlett
Total Synthesis of Stemoamide, 9a-epi-Stemoamide, and 9a,10-
epi-Stemoamide: Divergent Stereochemistry of the Final Methyla-
tion Steps
Juha H. Siitonen1,a,b
0
0
0
0-0
0
0
2-9
8
1
6-
6
3
7
X
stereochemistry of alkylation:
Dániel Csókásb
Imre Pápai*b,b
Petri M. Pihko*a
0
0
0
0-
0
0
0
2-4
1
5
0-4
7
7
X
natural stemoamide skeleton
9a-epimer
H
H
O
H
Me
10
0
0
0
0-
0
0
0
2-4
9
7
8-0
3
6
5
H
O
10
9
H
10
9aN
H
H
φ
9
8
φ
8
9a
N
O
0
0
0
0-0
0
0
3-
0
1
2
6-0
9
7
4
N
O
O
11
9a
1
11
O
O
H
O
H
O
H
H
a Department of Chemistry and Nanoscience Centre, University
of Jyvaskyla, 40014 Jyväskylä, Finland
MeI
MeI
natural stemoamide
different outcome
different reasons
petri.pihko@jyu.fi
b Institute of Organic Chemistry, Research Centre for Natural
Sciences, Magyar tudósok körútja 2, 1117 Budapest, Hungary
papai.imre@ttk.mta.hu
Received: 16.04.2020
a) anti-Selective alkylation used in the total synthesis of stemoamide (2)
Accepted after revision: 15.06.2020
Published online: 27.07.2020
DOI: 10.1055/s-0040-1707201; Art ID: st-2020-b0218-l
substituents anti
O
O
Me
C
N
9aN
H
H
H
H
10
[base]
H
A
O
O
A
O
then MeI
B
Abstract Total syntheses of stemoamide, 9a-epi-stemoamide, and
9a,10-epi-stemoamide by a convergent A + B ring-forming strategy is
reported. The synthesis required a diastereoselective late-stage methyl-
ation of the ABC stemoamide core that successfully enabled access to
three of the four possible diastereomeric structures. For the natural ste-
moamide series, the diastereoselectivity can be rationalized both by ki-
netic and thermodynamic arguments, whereas for the natural 9a-epi-
stemoamide series, the kinetic selectivity is explained by the prepyra-
midalization of the relevant enolate.
O
H
1
dr >20:1
2
ref. 3a,d,e,4
b) Will the alkylation proceed with the same selectivity with 9a-epi-
norstemoamide (9a-epi-2)?
anti or syn?
O
O
Me
C
N
9a
H
H
H
H
10
[base]
N
H
H
O
O
A
O
A
then MeI
B
O
9a-epi-1
9a-epi-2
Key words total synthesis, Stemona alkaloids, cyclic stereocontrol,
Mukaiyama–Michael reaction, DFT calculation
c) Compared to norstemoamide (1), 9a-epi-norstemoamide (9a-
epi-1) does not offer any obvious stereochemical bias for the
methylation
Traditional Chinese medicine uses Stemona plants to
treat various diseases and illnesses. Many structurally di-
verse alkaloids have been isolated from these Stemona
plants, with the alkaloid stemoamide (2) being the key rep-
resentative members of the natural product family.2,3 Sever-
al total and formal syntheses of stemoamide (2) have been
disclosed in the literature.4 Many of these syntheses rely on
a late-stage stereoselective C10 methylation of the ring A of
norstemoamide (1). This methylation is experimentally
known to yield natural stemoamide (2) in a stereoselective
fashion with the C10 methyl group disposed anti to the adja-
cent B ring system (Scheme 1, a).4a,d,e,5 A literature prece-
dent also suggests that methylation of 9a-epi-norstemoam-
ide (9a-epi-1) should proceed with the same sense of selec-
tivity (anti, Scheme 1, b).6 The observed stereoselectivity is
somewhat surprising since the stemoamide nucleus does
not possess any obvious bias, especially in the case of the
9a-epi-norstemoamide (Scheme 1 c).7 Herein, we report the
convex
H
H
O
H
10
10
9a
N
O
O
N
O
9a
O
H
concave
H
O
H
1
9a-epi-1
Scheme 1 Methylation of different stemoamide scaffolds at C10. a)
Methylation of norstemoamide skeleton (1) at C10 gives the anti-alkyla-
tion product, stemoamide (2).3,4a,d,e b) and c) The selectivity of the
methylation of 9a-epi-norstemoamide is less predictable as no steric
bias is obvious.
total syntheses of stemoamide (2), 9a-epi-stemoamide (9a-
epi-2), and 9a,10-epi-stemoamide (9a,10-epi-2), and pro-
vide a rationalization for the sense of diastereoselection.
We recognized butenolide aldehyde 3, accessible via an
enantioselective Mukaiyama–Michael reaction previously
developed in our group (Scheme 2), as a suitable A-ring
building block from which the stemoamide (2) structure
would emerge by installation of the C ring in a second
Mukaiyama–Michael reaction with a silyloxypyrrole, fol-
lowed by cyclization of the B ring.8
© 2020. Thieme. All rights reserved. Synlett 2020, 31, A–F
Georg Thieme Verlag KG, Rüdigerstraße 14, 70469 Stuttgart, Germany

B
J. H. Siitonen et al.
Letter
Synlett
cat.
Ph
Ph
N
Boc
N
H9a
C
O
H
OTBS
4-NBA
N
10
b) [H], see below
c) TsCl 91%
Boc
O
A
O
H
O
O
A
a) acrolein
65%
d)
13
75%
dr 1:1
TBSO
O
O
H
H
H
O
R
(+)-12
)-12
(+)-3
)-3
a') pyrrolidine/4-NBA,
(
(
)-14
9
OTs
)-epi-14
(
OTs
(
for ( )-3, 64%
[concomitant
racemization]
er of (+)-12
52%
74%
NaBH4, CeCl3·7H2O, MeOH
BH3·THF, THF
50:50
92:8
e) H2, Pd/C
quant.
[diastereomers
separated]
R = CHO
3
11 R = CH2OH
anti
H
O
H
O
H
Me
H
9a
9a
H
H
H
N
R
N
B
N
h) LiHMDS, MeI
g) NaH
25%
R
O
H
O
H
O
O
O
N
O
+
9a
O
O
H
62%
H
H
O
H
O
stemoamide (2)
1
OTs
OTs
[stemoamide branch]
R = Boc 15
R = H
16
R = Boc epi-15
f) TFA
84%
f) TFA
81%
R = H
epi-16
H
H
O
H
H
H
H
Me
H
O
O
10
9a
N
B
h) LiHMDS, MeI
g) NaH
73%
9a
9a
O
N
N
O
O
+
O
O
O
H
70%
Me
H
H
syn H
dr 1:9
i) K2CO3
84%
9a-epi-2
[minor]
9a,10-epi-2
[major]
9a-epi-1
[9a-epi-stemoamide branch]
(from 1:9 to 3:1 dr)
epi-16
[ScXRD]
Scheme 2 Total synthesis of (±)-stemoamide (2), (±)-9a,10-epi-stemoamide (9a,10-epi-2), and (±)-9a-epi-stemoamide (9a-epi-2). Reagents and condi-
tions: a) acrolein (5.0 equiv), (2S,5S)-diphenylpyrrolidine (10, 0.1 equiv), 4-nitrobenzoic acid (0.1 equiv), water (2.0 equiv), DCM, 0 °C, 5 h, 65%. For the
racemic series: a′) acrolein (3.0 equiv), pyrrolidine (0.1 equiv), 4-nitrobenzoic acid (0.1 equiv), water (2.0 equiv), DCM, –50 °C, 5 h, 64%; b) BH3·THF (1.1
equiv), THF, –60 °C, 30 min, 74%; c) TsCl (1.5 equiv), MeNH2·HCl (0.1 equiv), Et3N (1.5 equiv), DCM, 0 °C to rt, 10 min, 91%; d) 7 (2.0 equiv), TBSOTf (0.1
equiv), HFIP (1.0 equiv), DCM, –25 °C, 8 h, 75%, dr ca. 1:1; e) H2, Pd/C, THF, rt, overnight, quant.; f) TFA (2.0 equiv), DCM, rt, 4 h, 84% for 15, 81% for epi-
15; g) NaH (5.0 equiv), THF, 0 °C to rt, 17 h, 25% for 16, 6 h, 73% for epi-16; h) LiHMDS (1.1 equiv) then MeI (5.0 equiv), THF, –78 °C, 3 h, 62% for 1, 70%
for 9a-epi-1; g) K2CO3 (1.0 equiv), MeOH, rt, 48 h, 84% recovery.
This approach was also supported by very informative
previous syntheses utilizing similar Michael–additions to
a) Nitroenolate as a nucleophile Qiu (2013), Pilli (2015)
butenolide A ring starting materials such as 3 and 6
NO2
(Scheme 3).4g,h We should also add that during the course of
NO2
H
CO2R
CO2R
this study, the group of Chida published an elegant and
comprehensive total synthesis of various Stemona alkaloids
with a similar strategy to ours.4a
O
4
O
O
O
[nitroenolate
Michael]
H
H
H
O
O
We first set out to convert the aldehyde functionality of
3 into a leaving group to serve as a handle for the construc-
tion of the azepane B ring later in the synthesis. Towards
this end, aldehyde 3 was reduced with borane and tosylated
using the Tanabe tosylation protocol to give butenolide to-
sylate 12 in 67% yield over two steps (er 92:8, Scheme 2).9
Attempts at using a Luche reduction instead of borane led
to full racemization of the material.8a We then installed the
C-ring lactam unit onto the butenolide tosylate 12 using a
TBSOTf-catalyzed Mukaiyama–Michael reaction with sily-
loxypyrrole 13. Unfortunately, again the stereochemical la-
bility of 5H-furanones such as 12 presented a serious prob-
lem: all attempts at using enantioenriched 12 yielded race-
5
3
or protected
b) Silyloxypyrrole as a nucleophile, Chida (2017), This work
H
O
OSiR3
N
N
O
O
Pg
7
Pg
O
O
[Mukaiyama–
Michael]
H
OLg
OLg
6
8
Scheme 3 Previous synthetic approaches to Stemona alkaloid core re-
lying on conjugate additions to a butenolide4a,e,h
© 2020. Thieme. All rights reserved. Synlett 2020, 31, A–F

C
J. H. Siitonen et al.
Letter
Synlett
mic
Mukaiyama–Michael
adduct
rac-14.10
The
Table 1 Modifications to the Optimized Mukaiyama–Michael Reaction
between Butenolide 12 and Silyloxypyrrole 13
racemization is likely taking place via a reversible silyla-
tion–desilylation of the butenolide. With this setback, we
opted to complete the synthesis with racemic butenolide
tosylate rac-12 to demonstrate the viability of the route. It
is also worth noting that attempts at using alternative leav-
ing groups, such as the corresponding bromide or mesylate,
led to low yields of the desired Michael adduct and multiple
undesired side reactions.
OTBS
N
O
OTBS
NBoc
Boc
H
H
H
H
9aNBoc
H
13 (200 mol%)
O
O
O
O
(
TBSOTf (10 mol%)
HFIP (100 mol%)
DCM, –20 °C
O
O
)
OTs
OTs
OTs
12
14
17
The TBSOTf/HFIP system used in the key A–C ring cou-
pling to form rac-14 was identified after extensive screen-
ing of conditions. Most Lewis acids such as SnCl4, BF3·OEt2,
and Sc(OTf)3 in various solvents, temperatures, and addi-
tion rates merely decomposed the N-Boc-silyloxypyrrole 13
or led to complex mixtures. The decomposition of the N-
Boc-protected silyloxypyrrole 13 with traditional Lewis ac-
ids was also noted in the related total synthesis of Stemona
alkaloids by Chida and co-workers, who elegantly solved
this issue by changing the N-protecting group on the silyl-
oxypyrrole nucleophile.4a,11 The TBSOTf/HFIP-catalyzed
Mukaiyama–Michael reaction between A-ring butenolide
12 and N-Boc silyloxypyrrole 13 yields a mixture of the de-
sired AC ring adduct 14 and the corresponding silyloxypyr-
role 17 (Table 1). To improve the yield of 14, silyloxypyrrole
side product 17 was hydrolyzed to the desired product 14
with a prolonged aqueous quench. The thus obtained AB
ring fragment 14 formed as a close to 1:1 mixture of 9a-epi-
mers, with slight variation between batches, in 70–75%
yields, allowing for access to both stemoamide (2) and its
9a-epimer (9a-epi-2).
Entry Modification to standard protocol
Conversion (%)a Yield (%)b
1
2
3
4
5
6
7
8
9
–
98
13
82
68
63
15
77
18
–
75
no HFIP
10
4 Å MS (50 wt%)
7 (110 mol%)
HFIP (500 mol%)
HFIP (10 mol%)
TfOH (10 mol%) no TBSOTf
65
61
48
12
61
TfOH (10 mol%), no TBSOTf, no HFIP
quenched with Et3N
13
23 (17)
a Based on crude 1H NMR spectrum.
b Isolated yield after flash column chromatography of both diastereomers.
The diastereomeric ratio in all cases was ca. 1:1 and varied slightly based on
quenching.
crystal X-ray structure (see Scheme 2 inset). The deprotect-
ed epi-16 was cyclized with NaH in 4 h to give a 73% yield of
9a-epi-1.
Table 1 details the effects of modifying the optimized
Mukaiyama–Michael reaction conditions. These screens re-
vealed that the reaction required both HFIP as well as silyl
triflate (Table 1). Without HFIP, or with reduced HFIP load-
ing (10 mol%), the reaction gave low 10% and 12% isolated
yields of 14, respectively (Table 1, entries 2 and 6). The yield
of 14 was also lower when a large excess of HFIP was used
(Table 1, entry 5). When TBSOTf was replaced with triflic
acid, a comparable 61% yield of 14 was obtained (Table 1,
entry 7), but without HFIP, the yield dropped to 13% (Table
1, entry 8). These observations suggested that the combina-
tion of TBSOTf–HFIP promotes the reaction presumably by
forming catalytic amounts of triflic acid in situ. HFIP is also
crucial for the reaction.11d,12
With secured access to key intermediate 14, we pro-
ceeded to construct the final B (azepane) ring. The ,-un-
saturated lactam C ring of 14 was hydrogenated to give two
separable epimeric butanolides 15 and epi-15.13 Relative
configurations of the diastereomers could not be reliably
established from NMR data at this stage. Thankfully, after
N-deprotection of epi-15 with trifluoroacetic acid, the
product lactam epi-16 was crystalline, and its relative ste-
reochemistry could be reliably established from a single-
The natural 9a-epimer 15 was similarly N-deprotected
with trifluoroacetic acid and then cyclized to norstemoam-
ide (1) with NaH. In marked contrast to the epi-16, the cy-
clization of the natural 16 resulted in a lower 25% yield of 1,
taking 17 h to consume tosylate 16. The different cycliza-
tion rates and yields of the natural and 9a-epimeric to-
sylates 16 are in agreement with the work of Cossy and co-
workers, where cyclization of a 1:1 C9a-epimeric mixture of
the bromo analogues of tosylates 16 afforded a product
mixture enriched in the unnatural 9a-epi-norstemoamide
(dr 3:1).14
With both norstemoamide (1) and 9a-epi-1 prepared,
the final C10 methylation was explored. With LiHMDS and
MeI alkylation, norstemoamide (1) gave (±)-stemoamide
1
(2) as a single diastereomer.4e The H NMR, 13C NMR, and
scXRD data for stemoamide (2) were in full agreement with
the previously reported data (Figure 1, a).3,4
Under the same methylation conditions 9a-epi-1 afford-
ed a 9:1 mixture of C10 epimers, with the syn-methylated
9a,10-epi-stemoamide (9a,10-epi-2) being the major prod-
uct. The relative C10 syn configuration was assigned based
on a NOE between H8 and C10 methyl protons, as well as the
vicinal coupling constant 3JHH of 7.7 Hz between H10 and H9,
© 2020. Thieme. All rights reserved. Synlett 2020, 31, A–F

D
J. H. Siitonen et al.
Letter
Synlett
a) Stemoamide (2): C9–C10 stereochemistry is anti
a) Methylation of 9a-epi-1
3JH10-H9 = 13.1 Hz
H
H
Me
9a
9
10
10
O
2.43
N
9a
O
8
H
O
H
H
H
φ = –22.1°
10
9
8
φ
O
9a
N
O
[ScXRD]
11
O
H
H
b) 9a,10-bis-epi-Stemoamide (9a,10-epi-2): C9–C10 stereochemistry is syn
NOE
9a-epi-2
anti-TS (1.5)
3JH10-H9 = 7.7 Hz
H
H
O
10
H
H
O
9a
10
φ = 12.6°
10
9a
N
φ = 25.6°
9a
9
9
8
N
9
O
O
O
11
O
Me
8
Me
8
H
H
H
H
2.50
NOE
NOE
en
c) 9a-epi-Stemoamide (9a-epi-2): C9–C10 stereochemistry is anti
NOE
3JH10-H9 = 11.6 Hz
syn-TS (0.0)
9a,10-epi-2
H
Me
H
O
H
Me
O
10
10
9a
N
9a
9
N
9
O
O
O
O
8
H
8
H
H
H
H
b) Methylation of 1
NOE
NOE
Figure 1 Establishing the C9–C10 configuration of stemoamide diaste-
reomers based on NMR data
2.44
H
9
H
H
10
φ = –21.2°
φ
8
O
9a N
O
11
a value typical for syn configuration. The corresponding 3JHH
is 13.1 Hz for natural stemoamide (2), where the protons
are anti to each other (Figure 1, b).
1
H
O
anti-TS' (0.0)
2
10
φ = 12.6°
The crude 9:1 mixture of 9a,10-epi-2 and 9a-epi-2 could
be equilibrated with K2CO3 in methanol to a 1:3 mixture,
and the major 9a-epi-2 diastereomer isolated.15,16 For 9a-
epi-stemoamide (9a-epi-2), a diagnostic NOE from H9 to the
C10 methyl protons was observed. Additionally, the vicinal
9a
9
11
φ = 30.1°
8
1
2.46
en'
1
3
coupling constant between H9 and H10 hydrogens JH9–H10
was 11.6 Hz, indicative of hydrogens disposed anti (Figure
1, c). Our 1H NMR and 13C NMR data for 9a,10-epi-ste-
moamide (9a,10-epi-2) fully match to those previously re-
ported for 9a-epi-stemoamide (9a-epi-2). In addition, our
data for 9a-epi-stemoamide (9a-epi-2) was not in agree-
ment with the literature data, suggesting that the previous-
ly reported 9a-epi-stemoamide (9a-epi-2) is actually 9a,10-
epi-stemoamide (9a,10-epi-2).6,17
syn-TS' (0.8)
10-epi-2
Figure 2 Transition states computed for the methylation of enolates
derived from 9a-epi-1 and 1. The deviation from the planar structure of
the enolates is measured via the O11C11C10H10 dihedral angle (denoted
as ϕ). Relative stabilities are given in kcal/mol (in parenthesis), selected
C···C and H···H distances in Å. Most of the hydrogen atoms (except
those of the butyrolactone ring) are omitted for clarity.
To rationalize the observed selectivities – the anti selec-
tivity for the stemoamide series and the syn selectivity for
the 9a-epi series – the reactions were examined computa-
tionally. Using a simple model that includes the enolate de-
rived from 9a-epi-1 and MeI, we identified the alkylation
transition states by DFT leading to both the syn and anti
products (see Figure 2, a).18
DFT computations predicted the syn methylation of C10
carbon of the enolate en derived from 9a-epi-1 to be clearly
favored kinetically (syn-TS in Figure 2, a).19 The transition
state corresponding to the anti attack (anti-TS) is 1.5
kcal/mol higher than the transition state for syn attack,
which is in line with the experimentally observed stereose-
lectivity of kinetic methylation (dr 9:1) to deliver 9a,10-epi-
2 as the major product. Inspection of the located transition
states revealed no steric hindrance between the enolate en
and the approaching methyl group. We anticipate that the
syn facial selectivity of methylation of 9a-epi-1 could be as-
sociated with the nonplanar structure of the ground state
enolate en. This enolate features a notably pyramidalized
C10 atom (ϕ = 12.6° in en), predisposing the enolate en to-
wards syn attack. In the syn-TS, the degree of pyramidaliza-
tion is enhanced (ϕ = 25.6°). The anti attack is found to be
less favored as it requires inversion of the pyramidality at
the enolate a carbon (C10, ϕ = –22.1° in anti-TS). In addition,
DFT calculations predict the anti product 9a-epi-2 to be 0.7
kcal/mol more stable than the syn isomer 9a,10-epi-2,
© 2020. Thieme. All rights reserved. Synlett 2020, 31, A–F

E
J. H. Siitonen et al.
Letter
Synlett
which is in good agreement with the isomeric ratio ob-
tained via equilibration (dr 3:1).
(2) (a) Pilli, R. A.; Ferreira de Oliveira, M. C. Nat. Prod. Rep. 2000, 17,
117. (b) Pilli, R. A.; Rosso, G. B.; Ferreira de Oliveira, M. C. Nat.
Prod. Rep. 2010, 27, 1908.
In contrast, computational analysis of the methylation
of enolate en′ derived from norstemoamide (1) points to a
preferential anti attack (Figure 2, b). While the anti path-
way (via anti-TS′) requires inversion of the pyramidality at
the enolate en′  (C10) carbon compared to the ground state
enolate, the resulting energy penalty is less than that of the
steric clashes observed in the diastereomeric syn transition
state: the methyl hydrogens are in close contact with the C1
hydrogens of the lactam C ring (see Figure 2, b, syn-TS′).
The computed free energy difference for the two diastereo-
meric transition states is only 0.8 kcal/mol, implying that
experimentally observed high anti stereoselectivity (dr
>20:1) for stemoamide might be a result of thermodynami-
cally driven equilibration rather than kinetic alkylation. In-
deed, computations predict the anti product stemoamide
(2) to be 5.1 kcal/mol more stable than 10-epi-stemoamide
(10-epi-2). This result is in line with experimental evi-
dence.20
In summary, we have reported total syntheses of ste-
moamide (2), 9a-epi-stemoamide (9a-epi-2), and 9a,10-epi-
stemoamide (9a,10-epi-2).21–24 Our synthetic and computa-
tional work demonstrates how the stereocontrol at the late-
stage methylation of norstemoamides cannot readily be ac-
counted for by steric effects. Instead, for the 9a-episte-
moamide series, enolate pyramidalization25 appears to con-
trol the stereochemistry, whereas for the natural stemoam-
ide series the stereochemistry is likely to be under
thermodynamic control.
(3) Lin, W.-H.; Ye, Y.; Xu, R.-S. J. Nat. Prod. 1992, 55, 571.
(4) For recent total syntheses, see: (a) Yoritate, M.; Takahashi, Y.;
Tajima, H.; Ogihara, C.; Yokoyama, T.; Soda, Y.; Oishi, T.; Sato, T.;
Chida, N. J. Am. Chem. Soc. 2017, 139, 18386. (b) Nakayama, Y.;
Maeda, Y.; Hama, N.; Sato, T.; Chida, N. Synlett 2016, 48, 1647.
(c) Mi, X.; Wang, Y.; Zhu, L.; Wang, R.; Hong, R. Synthesis 2012,
44, 3432. (d) Honda, T.; Matsukawaa, T.; Takahashia, K. Org.
Biomol. Chem. 2011, 9, 673. (e) Torssel, S.; Wanngren, E.; Somfai,
P. J. Org. Chem. 2007, 72, 4246. (f) Li, Z.; Zhang, L.; Qiu, F. G.
Asian J. Org. Chem. 2014, 3, 52. (g) Brito, G. A.; Sarotti, A. M.;
Wipf, P.; Pilli, R. A. Tetrahedron Lett. 2015, 56, 6664.
(5) For the seminal work, see: Yasushi, K.; Koichi, N. Bull. Chem. Soc.
Jpn. 1996, 7, 2063.
(6) Gao, P.; Tong, Z.; Hu, H.; Xu, P.-F.; Liu, W.; Sun, C.; Zhai, H.
Synlett 2009, 2188.
(7) For representative similar methylations of ring-fused butano-
lides, see: (a) Matsumoto, K.; Koyachi, K.; Shindo, M. Tetrahe-
dron 2013, 69, 1043. (b) Evans, M. A.; Morken, J. P. Org. Lett.
2005, 7, 3371. (c) Lee, E.; Yoon, C. H.; Sung, Y.-S.; Kim, Y. K.; Yun,
M.; Kim, S. J. Am. Chem. Soc. 1997, 119, 8391. (d) Jung, M. E.; Im,
G.-Y. J. Tetrahedron Lett. 2003, 49, 4962.
(8) (a) Kemppainen, E. K.; Sahoo, G.; Piisola, A.; Hamza, A.; Kótai, B.;
Pápai, I.; Pihko, P. M. Chem. Eur. J. 2014, 20, 5983.
(b) Kemppainen, E. K.; Sahoo, G.; Valkonen, A.; Pihko, P. M. Org.
Lett. 2012, 14, 1086.
(9) The tosylation proceeded very slowly under standard TsCl/
DMAP/Et3N conditions, see: Yoshida, Y.; Sakakura, Y.; Aso, N.;
Okada, S.; Tanabe, Y. Tetrahedron 1999, 55, 2183.
(10) (a) Duret, P.; Figadère, B.; Hocquemiller, R.; Cavé, A. Tetrahedron
Lett. 1997, 51, 8849. (b) Yu, Q.; Wu, Y.; Wu, Y.-L.; Xia, L.-J. Tang
M.-H. Chirality 2000, 12, 127. (c) Ye, J.-L.; Zhang, Y.-F.; Liu, Y.;
Zhang, J.-Y.; Ruan, Y.-P.; Huang, P.-Q. Org. Chem. Front. 2015, 2,
697.
(11) (a) Casiraghi, G.; Battistini, L.; Curti, C.; Rassu, G.; Zanardi, F.
Chem. Rev. 2011, 111, 3076. (b) Casiraghi, G.; Rassu, G. Synthesis
1995, 607. (c) Jusseau, X.; Chabaud, L.; Guillou, C. Tetrahedron
2014, 16, 2595. (d) Suga, H.; Takemoto, H.; Kakehi, A. Heterocy-
cles 2007, 71, 361.
Funding Information
Financial support from the Academy of Finland (Projects 259532,
297874, 307624) and NKFIH Hungary (Grant No. K-112028) are
gratefully acknowledged.
A
c
a
d
e
m
y
of
F
i
n
l
a
n
d
(2
5
9
5
3
2)A
c
a
d
e
m
y
of
F
i
n
l
a
n
d
(2
9
7
8
7
4)A
c
a
d
e
m
y
of
F
i
n
l
a
n
d
(3
0
7
6
2
4)N
e
m
z
etiKutatásiF
e
j
l
esztésiésI
n
n
o
v
á
c
i
ós
H
i
v
atal(K-1
1
2
0
2
8)
(12) Colomer, I.; Chamberlain, A. E. R.; Haughey, M. B.; Donohoe, T. J.
Nat. Rev. Chem. 2017, 1, 1.
(13) If the N-Boc group of 14 was removed before hydrogenation, a
Acknowledgment
double-bond migration to form
observed.
a tertiary enamide was
We thank Sami Kortet for assistance with X-ray measurements, Esa
Haapaniemi for assistance with NMR measurements, and Johanna Hi-
idenheimo and Elina Kalenius for their assistance with HMRS mea-
surements.
(14) Bogliotti, N.; Dalko, P.; Cossy, J. J. Org. Chem. 2006, 71, 9528.
(15) For a similar equilibration of 9,10-epi-stemoamide, see: Khim,
S.-K.; Schultz, A. G. J. Org. Chem. 2004, 69, 7734.
(16) For the C10 epimers in the 9a-epi series, DFT calculations pre-
dicted that the anti diastereomer 9a-epi-2 is 0.7 kcal/mol more
stable than the syn isomer 9a,10-epi-2, which is in good agree-
ment with the isomeric ratio obtained via equilibration (dr 1:3),
see the Supporting Information.
Supporting Information
Supporting information for this article is available online at
https://doi.org/10.1055/s-0040-1707201.
S
u
p
p
orit
n
gInformati
o
n
S
u
p
p
orti
n
gInformati
o
n
(17) See the Supporting Information for {1H}13C shift comparisons.
(18) For the relevance of using the simple molecular model that does
not involve the Li+ ion in the calculations, see the Supporting
Information.
References and Notes
(19) The DFT calculations were carried out using the B97X-D func-
tional along with the Def2SVP and Def2TZVPP basis sets. The
reported relative stabilities were obtained from solution-phase
Gibbs free energies. For details, see the Supporting Information.
(1) New address: Department of Chemistry, Rice University, 6500
Main Street, Houston, TX 77030, USA.
© 2020. Thieme. All rights reserved. Synlett 2020, 31, A–F

F
J. H. Siitonen et al.
Letter
Synlett
(20) (a) For an experimental evidence for the epimerization at C10,
see: Jacobi, P. A.; Lee, K. J. J. Am. Chem. Soc. 1997, 199, 3409.
(b) In our hands, rapid quench and extraction of the norste-
moamide (1) methylation reaction gave a mixture of diastereo-
mers that was characterized without further purification.
(21) tert-Butyl (RS)-2-oxo-5-{(2R,3R)-5-oxo-2-[3-(tosyloxy)pro-
pyl]tetrahydrofuran-3-yl}-2,5-dihydro-1H-pyrrole-1-carbox-
ylate (14, -14)epi
2.04–2.10 (m, 1 H), 1.87–1.91 (m, 1 H), 1.74 (app. dq, J = 12.3,
10.7 Hz, 1 H), 1.51–1.60 (m, 2 H), 1.33 (d, J = 6.9 Hz, 3 H) ppm.
13C NMR (126 MHz, CDCl3):  = 177.5, 174.2, 77.8, 56.0, 52.9,
40.4, 37.5, 35.0, 30.8, 25.8, 22.7, 14.3 ppm. FTIR (film):  = 3501,
2935, 1764, 1676, 1420, 1190, 1008 cm–1
.
(23) 9a,10-epi-Stemoamide (9a,10-epi-2)
To a solution of 9a-epi-norstemoamide (9a-epi-1, 16.0 mg, 0.76
mmol, 1.0 equiv) in THF (1 mL) at –78 °C was added LiHMDS (84
L, 14 mg, 0.84 mmol, 1.1 equiv, 1.0 M in THF). After 10 min the
reaction mixture was warmed to 0 °C for 10 min and then
recooled to –78 °C. To the stirred yellow suspension methyl
iodide (38.8 L, 54 mg, 0.38 mmol, 5.0 equiv) was added drop-
wise. After 3 h the reaction mixture was quenched with sat. aq
NH4Cl (0.5 mL) and extracted with EtOAc (5 × 2 mL). Combined
organic layers were dried with Na2SO4 and concentrated in
vacuo. Flash chromatography (5% MeOH/EtOAc) gave 9a,10-epi-
stemoamide (9a,10-epi-2) as a white solid (12.0 mg, 70%); mp
127.3–127.9 °C. Rf (10% MeOH/EtOAc) = 0.32 (KMnO4). 1H NMR
(500 MHz, CDCl3):  = 4.51 (td, J = 10.7, 4.8 Hz, 1 H), 3.90 (ddd,
J = 14.6, 6.3, 3.6 Hz, 1 H), 3.65 (ddd, J = 10.4, 7.3, 6.2 Hz, 1 H),
3.04 (ddd, J = 14.3, 10.7, 3.1 Hz, 1 H), 2.81 (app. pent, J = 7.7 Hz,
1 H), 2.52–2.41 (3 H, m), 2.37–2.29 (1 H, m), 2.25 (app. td, J =
7.8, 10.4 Hz, 1 H), 1.98–1.66 (4 H, m), 1.30 (d, J = 7.7 Hz, 3 H)
ppm. 13C NMR (126 MHz, CDCl3):  = 178.2, 174.1, 80.6, 57.9,
52.5, 40.9, 38.6, 30.3, 30.2, 23.3, 21.8, 11.0 ppm. FTIR (film):  =
2940, 1773, 1681, 1208, 1005 cm–1. HRMS (ESI+): m/z [M + H]
calcd for [C12H17NO3H+]: 224.1208; found: 224.1201, Δ = –0.7
mDa.
To a stirred solution of tosylate 12 (100 mg, 0.34 mmol, 1.0
equiv) at –25 °C in DCM (10 mL) was added TBSOTf (8 L, 9 mg,
0.1 equiv) and HFIP (35 L, 57 mg, 1.0 equiv). To this solution
was then added silyloxypyrrole 13 (201 mg, 0.67 mmol, 2.0
equiv) in DCM (0.5 mL), and the resulting solution was stirred
for 8 h. The reaction was quenched with pH 7.0 buffer (2 mL),
and vigorously stirred for 1 h at rt to hydrolyze all silylated
product. The resulting mixture was then extracted with EtOAc
(5 × 3 mL), the combined organic layers dried with Na2SO4, and
concentrated in vacuo. Purification of the residue by Combi-
Flash automated chromatography system (10% EtOAc/hexane to
80% EtOAc/hexane) afforded 14 and epi-14 as a white foam (121
mg, 75%, dr ca. 1:1 with slight variation between batches). Rf
(60% EtOAc/hexane) = 0.19 (ninhydrin, brown). 1H NMR (500
MHz, CDCl3, diastereomers overlapping, integrals were normal-
ized to 1 H for signal at  = 6.21 ppm):  = 7.75–7.82 (m, 2 H),
7.33–7.39 (m, 2 H), 7.07 (dt, J = 6.2, 2.1 Hz, 1 H), 6.29–6.25 (m, 1
H), 4.74 (dt, J = 4.0, 1.8 Hz, 0.5 H), 4.72 (dt, J = 4.0, 1.8 Hz, 0.5 H),
4.42–4.40 (m, 0.5 H), 4.17–4.11 (m, 1 H), 4.08–3.90 (m, 1.5 H),
3.91–3.88 (m, 0.5 H), 3.26–3.22 (m, 0.5 H), 3.22–3.16 (m, 0.5 H),
2.88–2.83 (m, 0.5 H), 2.45 (s, 3 H), 2.43–2.38 (m, 0.5 H), 2.04 (t,
J = 2.5 Hz, 0.5 H), 1.95–1.62 (m, 4 H), 1.57 (s, 3.5 H), 1.56 (s, 3.5
H). 13C NMR (126 MHz, CDCl3, diastereomers overlapping):  =
174.8, 168.2, 168.1, 149.9, 149.8, 145.52, 145.49, 145.2, 145.1,
132.9, 130.5, 130.13, 130.10, 129.8, 128.0, 84.5, 84.3, 80.5, 78.7,
69.52, 69.47, 63.1, 62.7, 40.1, 40.0, 31.8, 31.7, 30.8, 28.3, 28.2,
28.0, 25.1, 25.02, 21.83, 21.82 ppm. FTIR (film):  = 2978
(weak), 2934 (weak), 1769, 1738, 1771, 1353, 1310, 1172, 1153,
(24) 9a-epi-Stemoamide (9a-epi-2)
To a solution of 9a,10-epi-stemoamide (9a,10-epi-2, 6.0 mg, 2.7
mol, 1.0 equiv) in methanol (0.5 ml) was added potassium car-
bonate (3.7 mg, 2.7 mol, 1.0 equiv). The resulting suspension
was stirred at rt for 48 h, concentrated in vacuo and acidified
with 2 M HCl (0.5 mL). The resulting solution was extracted
with DCM (5 × 1 mL) and the combined organic layers dried
with Na2SO4. The solvent was evaporated in vacuo to give a 3:1
mixture of 9a-epi-stemoamide (9a-epi-2) and 9a,10-epi-ste-
moamide (9a,10-epi-2, 5.0 mg, 84% recovery). A sample of 9a-
epi-2 for further NMR analysis was purified by flash column
chromatography (5% MeOH/EtOAc to 8% MeOH/EtOAc). Rf (10%
MeOH/EtOAc) = 0.32 (KMnO4 stain). 1H NMR (500 MHz, CDCl3):
 = 4.30 (ddd, J = 10.9, 10.0, 5.2 Hz, 1 H), 3.89 (ddd, J = 14.6, 6.4,
3.3 Hz, 1 H), 3.58 (td, J = 9.1, 6.6 Hz, 1 H), 3.14 (ddd, J = 14.6, 9.3,
3.4 Hz, 1 H), 2.37–2.52 (m, 3 H), 2.28–2.36 (m, 1 H), 2.24 (ddt,
J = 15.2, 8.0, 3.1 Hz, 1 H), 1.94 (dt, J = 11.0 Hz, 10.2 Hz, 1 H),
1.79–1.88 (m, 2 H), 1.73 (ddt, J = 13.4, 11.1, 6.7 Hz, 1 H), 1.67–
1.61 (m, 1 H), 1.39 (d, J = 7.0 Hz, 3 H) ppm. 13C NMR (126 MHz,
CDCl3):  = 177.6, 174.3, 81.2, 62.2, 55.4, 40.4, 40.0, 30.9, 30.5,
25.0, 22.5, 15.5 ppm. FTIR (film): 2924, 2851, 1774, 1683, 1278,
816, 662, 553 cm–1. HRMS (ESI+): m/z [M + Na] calcd for [C23H29
NO8SNa+]: 502.1506; found: 502.1511, Δ = –0.5 mDa.
(22) Stemoamide (2)
-
To a solution of norstemoamide 1 (5.0 mg, 0.024 mmol, 1.0
equiv) in THF (0.5 mL) LiHMDS (27 L, 4.4 mg, 0.026 mmol, 1.1
equiv, 1.0 M in THF) was added at –78 °C. After 10 min the reac-
tion mixture was warmed to 0 °C for 10 min and then recooled
to –78 °C. To the stirred yellow suspension, methyl iodide (7 L,
17 mg, 0.12 mmol, 5.0 equiv) was added dropwise. After 3 h the
reaction mixture was quenched with sat. aq NH4Cl (0.5 mL) and
extracted with EtOAc (5 × 2 mL). Combined organic layers were
dried with Na2SO4 and concentrated in vacuo. Flash chromatog-
raphy (EtOAc to 5% MeOH/EtOAc) gave (±)-stemoamide (2) as a
white solid (3.3 mg, 62%). Note: X-ray quality crystals were
obtained upon slow evaporation from EtOAc. Spectroscopic
data matched those reported previously.3,4a,b,e Rf (10%
MeOH/EtOAc) = 0.34 (KMnO4). 1H NMR (500 MHz, CDCl3):  =
4.19 (app. td, J = 4.8, 10.7 Hz, 1 H), 4.16 (td partially obstructed,
J = 3.0, 14.8 Hz, 1 H), 4.02 (td, J = 10.7, 6.4 Hz, 1 H), 2.26–2.68
(m, 1 H), 2.60 (dq, J = 12.3, 6.9 Hz, 1 H), 2.40–2.46 (m, 4 H),
1177, 1010 cm–1. HRMS (ESI+): m/z [M + H] calcd for [C12H17
-
NO3H+]: 224.1286; found: 224.1300, Δ = –1.4 mDa.
(25) For an expanded computational study of the effect of enolate
pyramidalization on the stereochemistry of methylation reac-
tions of trans-fused butyrolactones, see: Csókás, D.; Siitonen, J.
H.; Pihko, P. M.; Pápai, I. Org. Lett. 2020, 22, 4597.
© 2020. Thieme. All rights reserved. Synlett 2020, 31, A–F