A versatile chiral pyrrolidine aldehyde building-block for synthesis and formal synthesis of ent-nakadomarin A

Article abstract of DOI:10.1055/s-0029-1219182

A stable, simple to synthesise and versatile chiral aldehyde building-block has been developed, its reactivity in Wittig, Horner-Wadsworth-Emmons and Grignard reactions investigated, and its use is demonstrated in a highly efficient synthesis of an intermediate in Dixon's synthesis of nakadomarin A. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-0029-1219182

LETTER
559
A Versatile Chiral Pyrrolidine Aldehyde Building-Block for Synthesis and
Formal Synthesis of ent-Nakadomarin A
C
hira
l
Pyrrolidine
o
A
ldehyde
B
b
uilding-Blo
e
a
School of Chemistry, University of Nottingham, Nottingham, NG7 2RD, UK
Fax +44(115)9513564; E-mail: Robert.Stockman@Nottingham.ac.uk
b
c
School of Chemistry, University of East Anglia, Norwich, NR4 7TJ, UK
Department of Chemistry and Biochemistry, University of Texas at Austin, Austin, TX 78712-1167, USA
Received 9 November 2009
Dedicated to Prof. Gerry Pattenden on the occasion of his 70th birthday
An inspection of the literature for pyroglutamate-derived
chiral building-blocks provides three potential answers to
the problem. Several reports of the use of pyroglutamic al-
dehyde have been disclosed.8 In addition, several synthe-
ses of N-Boc pyroglutamic aldehyde have been reported
in the literature.9 However, when we tried to use these ma-
terials in our own laboratories, we found them to be very
unstable (too much so to purify), and both underwent
epimerization upon isolation. A third, potentially less re-
active building-block is that used by Clive in 1998 in a
synthesis of epibatidine.10 The pyroglutamic acid derived
N-Boc hemiaminal aldehyde 2, looked very interesting to
us, because the aldehyde should be no more reactive or
susceptible to epimerization than Garner’s aldehyde, the
nitrogen is protected with a reliable and simple to remove
protecting group, and the lactam motif of the pyroglutam-
ic acid has been transformed into a hemiaminal, which is
a function that could be potentially transformed into either
an enamine or an N-acyl iminium or can be reoxidised
back to a lactone. We thus initially set out to investigate
the range of pyroglutamate-derived hemiaminal alde-
hydes by synthesizing methyl, ethyl, isopropyl and n-
butyl hemiaminals. The synthesis of these is outlined in
Scheme 1.
Abstract: A stable, simple to synthesise and versatile chiral alde-
hyde building-block has been developed, its reactivity in Wittig,
Horner–Wadsworth–Emmons and Grignard reactions investigated,
and its use is demonstrated in a highly efficient synthesis of an in-
termediate in Dixon’s synthesis of nakadomarin A.
Key words: aldehyde, building-block, nakadomarin, manzamine
Garner’s aldehyde1 is a very flexible chiral building-block
that has seen use in over 200 publications since its debut
in the literature in the mid 1980s.2 Garner’s aldehyde has
been particularly useful in natural product synthesis,
where it has been used to generate a wide range of chiral
amine motifs.3 Due to its nature, Garner’s aldehyde func-
tions as a three-carbon building-block and, therefore, if a
chiral amine-containing ring is required in the synthesis,
the ring will need to be synthesized from the functionality
in place – often requiring several steps, protecting group
manipulations and redox operations.
As there are a large number of natural products containing
pyrrolidine rings with substituents at the 1-, 2-, 3- and 5-
positions, we were attracted to the idea of developing a
chiral building-block based on a pyrrolidine ring, such
that this might serve as a versatile entry into many of these
natural product systems. Figure 1 shows some representa-
tive natural products, manzamine A,4 stemonine,5
nirurine6 and berkleyamide A.7
O
O
O
SOCl2
MeOH
Boc2O, DMAP
Et3N, CH2Cl2
NH
CO2H
L-pyroglutamic acid
NH
NBoc
H
H
H
CO2Me
CO2Me
4 (89%)
3 (92%)
H
H
O
O
O
N
N
DIBAL-H, THF
PhMe, –78 °C
O
H
N
H
H
stemonine
O
N
OR
OH
OH
OR
O
H
DIBAL-H
ROH
TsOH (cat.)
O
manzamine A
NBoc
CHO
N
NBoc
H
N
Ph
NBoc
H
H
PhMe
–78° C
H
H
HO
O
H
CO2Me
5 (98%)
CO2Me
2a R = OMe (72%)
2b R = OEt (89%)
6a R = OMe (72%)
6b R = OEt (96%)
6c R = i-Pr (86%)
6d R = n-Bu (92%)
N
H
O
nirurine
berkleyamide A
Figure 1
Scheme 1
SYNLETT 2010, No. 4, pp 0559–0562
0
1.
0
3.
2
0
1
0
Advanced online publication: 11.01.2010
DOI: 10.1055/s-0029-1219182; Art ID: D32009ST
© Georg Thieme Verlag Stuttgart · New York

560
R. A. Stockman et al.
LETTER
Both the methyl (2a) and ethyl (2b) building-blocks were With this excellent result, we decided to investigate a
easily formed and were stable at room temperature over a Horner–Wadsworth–Emmons reaction of 2b (Scheme 3).
number of days; the isopropyl (2c) and n-butyl (2d) build- Phosphonoacetate 12 was deprotonated with sodium hy-
ing-blocks did not undergo complete reduction in the final dride and reacted with aldehyde 2b, giving the olefination
DIBAL-H step. In all cases, the ratio of diastereomers was product 13 in an excellent 84% yield as a single E-isomer.
2:1.
As before, it was found that the lactam functionality could
be reintroduced by treatment of the hemiaminal 13 with
chromium trioxide in acetic acid followed by treatment
with TFA.
We next looked at a typical Wittig reaction of aldehydes
2a and 2b using the protected alcohol-containing phos-
phonium salt 7 (Scheme 2). It was found that the methyl
hemiaminal 2a was relatively unstable under the reaction
conditions, and thus only a very poor (21%) yield of the
olefin product was obtained. In contrast, the ethyl hemi-
aminal 2b was found to be an excellent substrate for the
Wittig olefination, with alkene 8b being formed in 98%
yield. As a comparison, aldehyde 9, which has previously
been reported in the literature,9 gave just 8% yield of the
olefin product 10 when treated under the same reaction
conditions. Indeed, this aldehyde proved to be very unsta-
ble even over a few hours, and is thus not a practical syn-
thetic building-block. Having achieved olefination, we
wished to discover whether the hemiaminal could be reox-
idised to the lactam and found that reoxidation was, in-
deed, facile using chromium trioxide in acetic acid. If
glacial acetic acid was used, the BOC protecting group
was partially deprotected, and thus subsequent treatment
of the resulting mixture with TFA gave lactam 11 in 84%
yield over the two steps. On one occasion it was found that
treatment of 8b with chromium trioxide in a mixture of
O
EtO2C
P
OEt
OEt
OEt
OEt
NBoc
12
NBoc
CHO
2b
H
H
NaH, THF
EtO2C
13 (84%)
1. CrO3, AcOH
2. TFA, CH2Cl2
O
NH
H
EtO2C
14 (58%)
Scheme 3
AcOH and H2O (5:1) led directly to lactam 11 in 90% As the aldehyde functionality of aldehyde 2b was proving
yield, however, this reaction proved to be unreliable, giv- to be a good substrate for olefination chemistry, we decid-
ing yields in the range 44–90%, so for further studies we ed to investigate its potential in Grignard addition chem-
decided to use the reliable two-step procedure.
istry (Scheme 4). Thus, aldehyde 2b was treated with allyl
magnesium chloride in THF at low temperatures. We
found that Grignard addition proceeded smoothly to give
a diastereomeric mixture of alcohols. To gauge the stereo-
selectivity of the reaction, we protected the alcohol func-
tionality of 15 with an acetate group before oxidising 16
to the lactam. This procedure resulted in a 2:1 mixture of
diastereomeric alcohols, suggesting that a reasonable de-
gree of substrate control is afforded by the chiral aldehyde
in the Grignard addition.
OMe
OR
Br
Ph3P
OTBDPS
NBoc
NBoc
CHO
2a
H
H
OTBDPS
KHMDS, PhMe
–40 °C to 0 °C
8a R = Me (21%)
R = H (8%)
OEt
NBoc
OEt
Br
Ph3P
OTBDPS
NBoc
H
H
KHMDS, PhMe
–40 °C to 0 °C
OTBDPS
8b (98%)
CHO
2b
OEt
NBoc
OEt
1. CrO3, AcOH
2. TFA, CH2Cl2
ClMg
NBoc
CHO
2b
H
H
THF, –78 °C to r.t.
O
OH
15 (91%), 2:1 d
NH
H
OTBDPS
11 (84%)
Ac2O,
DMAP, CH2Cl2
OEt
NBoc
OAc
16 (96%)
O
O
O
O
1. CrO3
AcOH
Br
NH
OAc
NH
H
Ph3P
OTBDPS
H
H
NBoc
NBoc
2. TFA
CH2Cl2
H
H
OTBDPS
10 (8%)
KHMDS/PhMe
–40 °C to 0 °C
OAc
17 (43%), 2:1 dr
CHO
9
Scheme 4
Scheme 2
Synlett 2010, No. 4, 559–562 © Thieme Stuttgart · New York

LETTER
Chiral Pyrrolidine Aldehyde Building-Block
561
Having investigated a number of reaction types for alde- In conclusion, we have developed a stable, easily synthe-
hyde substrate 2b, we decided to investigate its suitability sised chiral aldehyde building-block13 for the synthesis of
as a building-block for a synthesis of the manzamine alka- chiral 5-substituted pyrrolidin-2-ones.14 We have also
loids. Dixon used lactam 18 in his seminal recent synthe- used this chiral aldehyde to prepare a useful intermediate
sis of nakadomarin A,11 which is a member of the for the synthesis of the manzamine family of alkaloids, in-
manzamine family of alkaloids (Scheme 5).
cluding nakadomarin A, in very high yields. Further stud-
ies on the reactivity of chiral aldehyde 2b are ongoing and
will be reported in due course.
N
O
O
Acknowledgment
Dixon, 7 steps
(ref. 11)
H
N
The authors thank the Welch Foundation (P.M.), the University of
East Anglia (P.J.M.) and the EPSRC (Advanced Research Fel-
lowship, R.A.S.) for funding.
H
N
H
18
nakadomarin A
References and Notes
(1) (a) Garner, P.; Park, J. M. Org. Synth. 1991, 70, 18.
(b) Garner, P.; Park, J. M.; Malecki, E. J. Org. Chem. 1988,
53, 4395. (c) Garner, P.; Ramakanth, S. J. Org. Chem. 1986,
51, 2609.
(2) Liang, Y. F.; Andersch, J.; Bols, M. J. Chem. Soc., Perkin
Trans. 1 2001, 2136.
Scheme 5
Thus, the lactam 11 was transformed in a three-step pro-
cedure into the 5/8 C/E ring system of the manzamine al-
kaloids, generating a potentially useful building-block for
future synthetic efforts towards these topical natural prod-
ucts (Scheme 6). The first step of the three-step procedure
involved deprotection of the alcohol functionality by
treatment of 11 with HF in acetonitrile. Conversion of the
revealed alcohol 19 into the tosylate 20 was found to pro-
ceed in near quantitative yield using Tanabe’s proce-
dure.12 Eight-membered ring-closure to the 5,8-
azabicycle ent-18 was achieved13 using potassium tert-bu-
toxide in THF at –40 °C in two hours, followed by warm-
ing to room temperature overnight. The reaction
proceeded in an outstanding 98% yield, giving an overall
yield for the synthesis of ent-18 of 49.2% from L-pyro-
glutamic acid. This compares well with Dixons synthesis
of 18, which proceeds in 29.2% yield from the tosylate of
D-pyroglutamic alcohol (which also requires several syn-
thetic steps from pyroglutamic acid). Thus, our synthesis
could also be perceived as a formal synthesis of ent-naka-
domarin A.
(3) For some recent examples of the use of Garner’s aldehyde in
total synthesis, see: (a) Blot, V.; Jacqemard, U.; Reissig,
H.-U.; Kleuser, B. Synthesis 2009, 759. (b) Guaragna, A.;
D’Alonzo, D.; Paolella, C.; Palumbo, G. Tetrahedron Lett.
2009, 50, 2045. (c) Ribes, C.; Falomir, E.; Carda, M.;
Marco, J. A. J. Org. Chem. 2009, 73, 7779. (d) Srivastava,
A. J.; Panda, G. Chem. Eur. J. 2008, 14, 4675.
(e) Passiniemi, M.; Koskinen, A. M. P. Tetrahedron Lett.
2008, 49, 980. (f) Pearson, M. S. M.; Evain, M.; Mathe-
Allainmat, M.; Lebreton, J. Eur. J. Org. Chem. 2007, 27,
4888.
(4) (a) Sakai, R.; Higa, T.; Jefford, C. W.; Bernardinelli, G.
J. Am. Chem. Soc. 1986, 108, 6404. (b) Sakai, R.;
Kohmoto, S.; Higa, T. Tetrahedron Lett. 1987, 28, 5493.
(5) Koyama, H.; Oda, K. J. Chem. Soc. B 1970, 1330.
(6) Petcharee, P.; Bunyapraphatsara, N.; Cordell, G. A.; Cowe,
H. J.; Cox, P. J.; Howie, R. A.; Patt, S. L. J. Chem. Soc.,
Perkin Trans. 1 1986, 1551.
(7) Stierle, A. A.; Stierle, D. B.; Patacini, B. J. Nat. Prod. 2008,
71, 856.
(8) (a) Friedman, T. C.; Kline, T. B.; Wilk, S. Biochemistry
1985, 24, 3907. (b) Meisch, H.-U.; Maus, R. Z.
Naturforsch., C 1983, 38, 563.
O
O
(9) (a) Rassu, G.; Pinna, L.; Spanu, P.; Ulgheri, F.; Casiraghi, G.
Tetrahedron Lett. 1994, 35, 4019. (b) Rassu, G.; Zanardi,
F.; Battistini, L.; Gaetani, E.; Casiraghi, G. J. Med. Chem.
1997, 40, 168. (c) Rassu, G.; Carta, P.; Pinna, L.; Battastini,
L.; Zanardi, F.; Acquotti, D.; Casiraghi, G. Eur. J. Org.
Chem. 1999, 1395.
NH
NBoc
HF, MeCN
H
H
OTBDPS
OH
11
19 (90%)
TsCl, Me3N⋅HCl, Et3N
CH2Cl2
(10) Clive, D. L. J.; Yeh, V. S. C. Tetrahedron Lett. 1998, 39,
4789.
(11) Jakubec, P.; Cockfield, D. M.; Dixon, D. J. J. Am. Chem.
Soc. 2009, 131, 16632.
O
O
(12) Yoshida, Y.; Sakakura, Y.; Aso, N.; Okada, S.; Tanabe, Y.
Tetrahedron 1999, 55, 2183.
NH
N
t-BuOK, t-BuOH
THF, –40 °C to 0 °C
H
H
OTs
(13) Compound 18 has also been reported previously in racemic
form, see: Martin, S. F.; Chen, H.-J.; Courtney, A. K.; Liao,
Y.; Patzel, M.; Ramser, M. N.; Wagman, A. S. Tetrahedron
1996, 52, 7521.
ent-18 (98%)
20 (99%)
Data for compound 20: Rf = 0.3 (EtOAc); [a]D22 +86.6 (c 1.0,
CHCl3); IR (thin film): 1683 (lactam), 1485 (olefin) cm–1; 1H
NMR (300 MHz, CDCl3): d = 5.80 (dt, J = 10.6, 8.5 Hz,
Scheme 6
Synlett 2010, No. 4, 559–562 © Thieme Stuttgart · New York

562
R. A. Stockman et al.
LETTER
1 H), 5.41 (dd, J = 10.6, 6 Hz, 1 H), 4.26 (dt, J = 6.5, 6 Hz,
1 H), 3.42 (t, J = 5.4 Hz, 2 H), 2.50–2.05 (m, 6 H), 1.91–1.70
(m, 2 H), 1.68–1.49 (m, 2 H); 13C NMR (75 MHz, CDCl3):
d = 174.4, 131.9, 130.4, 56.8, 41.0, 30.9, 27.0, 26.4, 25.9,
25.2; MS (CI): m/z (%) = 166 (100) [M + 1]+, 152 (23);
HRMS: m/z [M + H]+ calcd for C10H16NO: 166.1232; found:
166.1232. Anal. Calcd for C10H15NO: C, 72.69; H, 9.15; N,
8.48. Found: C, 72.82; H, 9.28; N, 8.51.
13C NMR (100 MHz, CDCl3): d = 82.52–81.20, 70.71, 59.5,
52.5, 33.4, 28.5, 28.1, 27.3; MS: m/z (%) = 263 (5) [M +
NH4]+, 245 (40), 128 (100); Anal. Calcd for C11H19NO5 [M
+ NH4]+: 263.1601. Found: 263.1604. A solution of 1-(tert-
butyloxycarbonyl)-5-(hydroxy)pyrrolidine-2-carboxylic
acid methyl ester (5; 33.12 g, 0.135 mol) in EtOH (600 mL)
was treated with PTSA·H2O (1.85 g, 9.73 mmol), and the
solution was allowed to stand for 18 h. The solvent was then
removed under reduced pressure and the resulting residue
was partitioned between EtOAc (300 mL) and sat. aq
NaHCO3 (300 mL). The organic layer was separated, dried
over Na2SO4 and evaporated in vacuo to give 1-(tert-
butyloxycarbonyl)-5-(ethoxy)pyrrolidine-2-carboxylic acid
methyl ester (6; 35.90 g, 96%) as a 2:1 mixture of diastereo-
mers, as a colourless oil. Bp 92°C (0.15 Torr); IR: 1755,
1703 cm–1; 1H NMR (400 MHz, CDCl3): d = 5.38–5.20 (m,
1 H), 4.37–4.22 (m, 1 H), 3.72 (s, 3 H), 3.66–3.49 (m, 2 H),
2.48–1.76 (m, 4 H), 1.41 (s, 9 H), 1.08 (t, J = 3.6 Hz, 3 H);
13C NMR (100 MHz, CDCl3): d = 87.3, 64.0, 59.0, 52.07,
51.92, 32.5, 30.6, 27.5, 15.3; MS: m/z (%) = 273 (6) [M +
H]+, 228 (100); Anal. Calcd for C13H24NO5 [M + H]+:
274.1664. Found: 274.1654. A solution of 1-(tert-butyloxy-
carbonyl)-5-(ethoxy)pyrrolidine-2-carboxylic acid methyl
ester (6; 16.3 g, 59.7 mmol) in anhydrous toluene (100 mL)
was cooled to –78 °C under an atmosphere of argon.
DIBAL-H (1 M in toluene, 71.6 mL) was added over a
period of 30 min by cannula (dribbling down the inside of
the flask). The resulting solution was stirred at –78 °C for a
further 6 h before being quenched by the addition of
anhydrous MeOH (10 mL). After warming to 0 °C, a 1 M
aqueous solution of Rochelle’s salt (300 mL) and EtOAc
(100 mL) were added, and the biphasic solution was stirred
vigorously for 2 h. The organic layer was separated, dried
over Na2SO4, and evaporated in vacuo to give aldehyde 2b
(12.95 g, 89%) as a colourless oil after distillation (bp 96–
98 °C, 0.2 Torr); IR: 1739, 1698 cm–1; 1H NMR (300 MHz,
CDCl3): d = 9.55–9.36 (m, 1 H, 5-H), 5.39–5.18 (m, 1 H),
4.35–4.00 (m, 1 H), 3.72–3.48 (m, 2 H), 2.48–2.00 (m, 2 H),
1.98–1.63 (m, 2 H), 1.47–1.38 (m, 9 H), 1.09 (t, J = 7 Hz,
3 H); 13C (100 MHz, CDCl3): d = 200.43, 200.31, 166.70,
88.0, 81.5, 81.2, 65.5, 64.4, 32.0, 28.4, 25.0, 15.5; MS:
m/z (%) = 244 (10) [M + H]+, 198 (62); Anal. Calcd for
C12H22NO4 [M + H]+: 244.1549. Found: 244.1552.
(14) Procedure for the synthesis of chiral aldehyde 2b: A solution
of pyroglutamic acid methyl ester (3; 25.7 g, 0.180 mol) in
anhydrous CH2Cl2 (500 mL) was cooled to 0 °C under an
atmosphere of argon. The reaction solution was treated with
DMAP (2.19 g, 0.018 mol), Et3N (27.0 mL, 0.198 mol) and
(Boc)2O (41.2 g, 0.189 mol). The solution was allowed to
warm to r.t. and stirred for 16 h. HCl (1 M, 300 mL) was
added, and the organic layer was separated, washed with sat.
NaHCO3 (300 mL), dried over Na2SO4, and evaporated in
vacuo. Recrystallisation from hexanes–EtOAc gave N-tert-
butyloxycarbonylpyroglutamic acid methyl ester (4; 38.74 g,
89%) as colourless needles. Mp 58–65 °C; [a]D20 –30 (c
2.06, CHCl3); IR (nujol): 1763, 1703 cm–1; 1H NMR (400
MHz, CDCl3): d = 4.61 (dd, J = 9.4, 3 Hz, 1 H), 3.75–3.74
(m, 3 H), 2.65–2.55 (m, 1 H), 2.50–2.42 (m, 1 H), 2.34–2.23
(m, 1 H), 2.04–1.96 (m, 1 H), 1.45–1.41 (m, 9 H); 13C NMR
(100 MHz, CDCl3): d = 173.5, 172.1, 149.3, 83.8, 59.0, 52.8,
31.4, 28.1, 21.7; MS: m/z (%) = 261 (98) [M + NH4]+, 144
(100); Anal. Calcd for C11H17NO5 [M + NH4]+: 261.1445.
Found: 261.1443. A solution of N-tert-
butyloxycarbonylpyroglutamic acid methyl ester (4; 34.12 g,
0.14 mol) in anhydrous THF (350 mL) was cooled to –78 °C
under an atmosphere of argon. DIBAL-H (1 M in toluene,
150 mL) was added over a period of 1 h. The resulting
solution was stirred at –78 °C for a further 2 h, before being
quenched by the addition of anhydrous MeOH (30 mL).
After warming to 0 °C, a 1 M aqueous solution of Rochelle’s
salt (600 mL) and EtOAc (300 mL) were added and the
biphasic solution was stirred vigorously for 2 h. The organic
layer was separated, dried over Na2SO4 and evaporated to
give 1-(tert-butyloxycarbonyl)-5-(hydroxy)pyrrolidine-2-
carboxylic acid methyl ester (5; 33.17 g, 98%) as a 2:1
mixture of diastereomers, as a colourless oil. 1H NMR (400
MHz, CDCl3): d = 5.63–5.39 (m, 1 H), 4.60–4.17 (m, 1 H),
3.75–3.66 (m, 3 H), 2.64–1.83 (m, 4 H), 1.45–1.36 (m, 9 H);
Synlett 2010, No. 4, 559–562 © Thieme Stuttgart · New York