A Stereoselective Synthesis of Lentiginosine

Article abstract of DOI:10.1021/acs.joc.5b02804

A concise stereoselective synthesis of (-)-lentiginosine, an iminosugar endowed with an interesting proapoptotic activity, has been accomplished using an enantiopure pyrroline N-oxide building block derived from d-tartaric acid. Key steps are a totally diastereoselective nucleophilic addition to the cyclic nitrone followed by a combination of two simultaneous and two tandem reactions occurring under the same conditions in a single laboratory operation. Natural (+)-lentiginosine can be synthesized by the same method but starting from l-tartaric acid. (Chemical Equation Presented).

Full text of DOI:10.1021/acs.joc.5b02804

Article
pubs.acs.org/joc
A Stereoselective Synthesis of Lentiginosine
Franca M. Cordero,* Carolina Vurchio, and Alberto Brandi*
̀
Dipartimento di Chimica “Ugo Schiff”, Universita di Firenze, via della Lastruccia 13, 50019 Sesto Fiorentino (FI), Italy
S
* Supporting Information
ABSTRACT: A concise stereoselective synthesis of (−)-lenti-
ginosine, an iminosugar endowed with an interesting
proapoptotic activity, has been accomplished using an
enantiopure pyrroline N-oxide building block derived from
D-tartaric acid. Key steps are a totally diastereoselective
nucleophilic addition to the cyclic nitrone followed by a
combination of two simultaneous and two tandem reactions
occurring under the same conditions in a single laboratory
operation. Natural (+)-lentiginosine can be synthesized by the
same method but starting from L-tartaric acid.
1,3-dipolar cycloaddition of methylenecyclopropane and but-3-
en-1-ol with dialkoxypyrroline N-oxide 2 and ent-2 (R =
TBDPS, tBu, Bz) in turn derived from D- and L-tartaric acid,
respectively (Figure 1).2,6 Intermediates 3 and 4 were exploited
INTRODUCTION
■
In 1990, Elbein et al. isolated two 1,2-dihydroxyindolizidine
alkaloids from the leaves of Astragalus lentiginosus and identified
their structures as (+)-lentiginosine [(+)-1]1 and its 2-epimer.
The (1S,2S,8aS) absolute configuration of natural lentiginosine
initially guessed on the basis of biosynthetic consideration1 was
later validated through the total synthesis of both the
enantiomers and comparison of their glycosidase-inhibitory
activity.2 In fact, (+)-1 is a potent and selective inhibitor of
amyloglucosidases despite having only two hydroxyl groups. A
recent study showed that natural (+)-1 inhibits in vitro ATPase
and chaperone activities of heat shock protein 90 (Hsp90) by
interacting with the protein middle domain.3
Figure 1. Previous synthetic strategies to (−)-lentiginosine [(−)-1]
and 7-substituted derivatives.
Synthetic (−)-lentiginosine [(−)-1] was shown to possess an
interesting biological activity albeit different from its natural
enantiomer. In particular, (−)-1 has a good caspase-dependent
proapoptotic activity against different cancer cell lines, with low
toxicity toward normal cells.4 The remarkable and specific
biological activities of both the enantiomers 1, in combination
with the low abundance in nature of (+)-1 (approximately 10
mg from 1 kg of dried plant material) and challenging
indolizidine framework featuring three contiguous stereo-
centers, has prompted a multitude of stereoselective synthetic
approaches to 1 and derivatives.5 The inexpensive chiral pool
compound tartaric acid is one of the most frequently used
starting materials because it is readily available in both
enantiomers and has the correct configuration for two of the
three lentiginosine stereocenters.
as building blocks in the synthesis of various 7-substituted
lentiginosine derivatives7 and were converted to 1 through a
suitable deoxygenation/deprotection sequence.2,6a To support
the ongoing drug development program, a shorter synthesis of
1 was desirable. Here we describe an improved synthetic
strategy based on the nucleophilic addition of a suitable
Grignard reagent8 to the same nitrone 2 followed by a domino-
based approach to the indolizidine framework. The synthesis of
(−)-lentiginosine is described here, but natural (+)-lentigino-
sine can be analogously prepared starting from L-tartaric acid.
In our previous work toward (−)-1, (+)-1, and lentiginosine
derivatives, we reported on two synthetic approaches based on
Received: December 10, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b02804
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Bu) face of nitrone 10,12 setting up the third stereocenter with
the correct relative and absolute configuration. If kept in
CDCl3, hydroxylamine 11 is slowly oxidized to the correspond-
ing cyclic nitrones but can be safely stored neat at low
temperature for several weeks.
The key direct cyclization of 11 to indolizidine 13 was
initially attempted using Zn/HCl in THF/H2O. These
reductive acidic conditions have been previously applied to
the one-pot cyclization of nitro dimethoxyacetals,13 but in the
case of 11 the reaction breaks off after the N−O bond
reduction. Hydrolysis of the relatively stable 1,3-dioxane group
was observed only in traces after protracted heating. The use of
TFA or acetone to favor aldehyde deprotection led to complex
decomposition mixtures. The lack of product formation was
attributed to the fast rate of Zn consumption in comparison
with the relatively slow acetal hydrolysis. A portionwise
addition of an excess of Zn was not beneficial. Consequently,
a catalytic reducing system was tested.
RESULTS AND DISCUSSION
Our new retrosynthesis of lentiginosine is shown in Scheme 1.
It was envisaged that simultaneous N−O bond reductive
■
Scheme 1. New Retrosynthetic Analysis of (−)-Lentiginosine
[(−)-1]
Catalytic hydrogenation, both on Pd/C and Pd(OH)2/C, is
commonly used in N−O bond hydrogenolysis and reductive
amination.14 All the more, it can be carried out in the presence
of an acid, therefore encompassing all the requirements for the
domino-based process to 13. After ample investigation, the
optimum conditions for the reductive cyclization reaction were
to heat a mixture of hydroxylamine 11 (0.02 M), Pd/C (10 mol
%) and trimethylsilyl chloride (TMSCl, 3 molar equiv) in
methanol at 50 °C for 44 h under a hydrogen atmosphere (1
atm)15 (Scheme 3). Under these conditions, indolizidine 13
cleavage and acetal hydrolysis followed by a spontaneous
intramolecular condensation of amino aldehyde 6 and
reduction of the formed indolizidinium ion could be used to
get 5 in a single laboratory operation starting from hydroxyl-
amine 7. This in turn should be available with high
diastereoselectivity from nitrone 2 and a Grignard reagent ω-
functionalized with an acetal group.
Acetal-substituted Grignard reagents can difficult to be
prepare;9 accordingly, the well-known Grignard reagent 910
derived from 2-(3-chloropropyl)-5,5-dimethyl-1,3-dioxan (8)
was chosen to establish the feasibility of the novel synthetic
approach. The chloroacetal 8 is readily available in large scale
by a two-step procedure employing 2,3-dihydrofuran and 2,2-
dimethyl-1,3-propanediol as inexpensive starting materials
(65% overall yield).11 The corresponding Grignard reagent
was generated according to the procedure of Forbes et al.10 and
immediately reacted with bis-tert-butoxypyrroline N-oxide 10,
in turn prepared by a five-step route starting from D-tartaric
acid (46% overall yield).6b
Scheme 3. Synthesis of (−)-Lentiginosine [(−)-1]
The addition reaction afforded hydroxylamine 11 as a sole
diastereoisomer in 90% yield after purification on silica gel
(Scheme 2). As expected, the bulky tert-butoxy group on
nitrone C-3 favors the exclusive addition of 9 to the anti (3-Ot-
was smoothly formed in 56% yield after chromatography
purification. Analogous results were obtained using Pd(OH)2/
C as catalyst. Under the reported reaction conditions, tert-butyl
ether hydrolysis was never detected.16
Scheme 2. Stereoselective Nucleophilic Addition of 9 to
Nitrone 3
The effect of a different acetal protecting group on the yield
of the domino-based process was also explored. In particular,
when the cyclic acetal was replaced with the more acid-labile
diethyl acetal, catalytic hydrogenation of the corresponding N-
hydroxypyrrolidine 7 (R1 = t-Bu; R2 = Et) occurred faster than
that for 11 (30 °C, 23 h compared to 50 °C, 44 h) but,
unfortunately, gave indolizidine 13 in lower yield (49 compared
to 56%).17 The nature of the acetal group of 7 seems to be
crucial for the success of the reaction, because it must secure
the best timing of the proceeding of the sequential steps of the
reaction, the first being, necessarily to avoid side products, the
reduction of the hydroxylamine group. Hydrolysis of the tert-
butyl ethers with TFA6a completed the synthesis of (−)-1
(Scheme 3).
In summary, we have developed an improved and concise
stereoselective synthesis of the proapoptotic agent (−)-lentigi-
nosine [(−)-1]. The method involves a completely diaster-
B
DOI: 10.1021/acs.joc.5b02804
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
filtrate was eluted through a column of Amberlyst A26 with MeOH as
eluent. (The Amberlyst A26 column had been previously flushed with
MeOH). MeOH was evaporated under reduced pressure, and the
residue was purified by chromatography on silica gel (eluent: CH2Cl2/
MeOH = 25:1), to yield indolizidine 13 (48 mg, 56%) along with 2,2-
dimethyl-1,3-propanediol (23 mg, 70%). Under the same conditions, a
little bit higher amounts of 11 (270 mg, 0.70 mmol and 400 mg, 1.03
mmol) afforded indolizidine 13 with a slightly lower but reproducible
53% yield (100 and 147 mg, respectively).
eoselective nucleophilic addition to (3R,4R)-3,4-bis(tert-bu-
toxy)-3,4-dihydro-2H-pyrrole-1-oxide (10),6b a useful building
block in turn prepared from D-tartaric acid in five steps with
46% overall yield. The formed hydroxylamine 11 featuring the
requested three stereogenic centers with the correct config-
uration is then converted into (−)-1 in two simple steps with
48% overall yield. The major achievement of this novel
approach is the creation of the indolizidine skeleton by catalytic
hydrogenation of 11 in the presence of a strong protic acid to
trigger a sequence of two simultaneous and two tandem
reactions, i.e., hydroxylamine N−O hydrogenolysis, acetal
hydrolysis, intramolecular condensation, and reduction of the
cyclic iminium ion 13 in a single laboratory operation. This
strategy provides a practical and scalable access to (−)-lenti-
ginosine to further in vitro and in vivo studies (in progress).
The same sequence applied to L-tartaric acid is a novel synthesis
of the natural iminosugar (+)-lentiginosine.
Procedure B. Following the same procedure used to prepare 11,
(2R,3R,4R)-2-(4,4-diethoxybutyl)-3,4-di-tert-butoxypyrrolidin-1-ol (7,
R1 = t-Bu; R2 = Et) was obtained starting from 4-chloro-1,1-
diethoxybutane (1 g, 5.53 mmol) and nitrone 10 (633 mg, 2.76
mmol). Purification by chromatography on silica gel (eluent:
petroleum ether/EtAcOEt = 3:1) afforded hydroxylamine 7 (R1 = t-
Bu; R2 = Et) as a colorless viscous oil in 89% yield (923 mg).
7 (R1 = t-Bu; R2 = Et): Rf = 0.28 (petroleum ether/EtAcOEt = 3:1);
21
1
[α]D = −24 (c = 0.74, CHCl3); H NMR (CDCl3, 400 MHz): δ =
4.48 (pseudo t, J = 5.6 Hz, 1H, OCHO), 3.92 (pseudo dt, J = 2.8; 4.6
Hz, 1H, 4-H), 3.67−3.57 (m, 3H, 3-H + OCHHMe × 2), 3.473 (dq, J
= 9.3; 7.1 Hz, 1H, OCHHMe), 3.470 (dq, J = 9.4; 7.1 Hz, 1H,
OCHHMe), 3.17 (d, J = 4.7 Hz, 2H, 5-H), 2.75 (pseudo q, J = 5.9 Hz,
1H, 2-H), 1.71−1.49 (m, 6H, CH2CH2CH2), 1.19 [s, 9H, C(CH3)3],
1.18 (t, J = 7.1 Hz, 6H, CH2CH3 × 2), 1.16 [s, 9H, C(CH3)3] ppm;
13C NMR (CDCl3, 100 MHz): δ = 102.9 (d, OCHO), 81.7 (d, C-3),
77.0 (d, C-4), 74.2 (s, Me3CO), 73.8 (s, Me3CO), 73.3 (d, C-2), 64.5
(t, C-5), 61.0 (t, CH2Me), 60.8 (t, CH2Me), 33.8 (t, CH2CH2), 30.0
(t, CH2CH2), 29.0 [q, 3C, C(CH3)3], 28.6 [q, 3C, C(CH3)3], 21.7 (t,
CH2CH2), 15.3 (q, 2C, CH2CH3 × 2) ppm; IR (CDCl3): ν = 3580,
3380, 2978, 2873, 1457, 1391, 1367, 1190 cm−1; C20H41NO5 (375.3):
calcd C 63.96, H 11.00, N 3.731; found C 63.55, H 10.80, N 3.48.
To a mixture of hydroxylamine 7 (R1 = t-Bu; R2 = Et, 63.4 mg, 017
mmol) and Pd/C (10% in weight, 18 mg; Pd: 3.4 mg, 10 mol %) in
EtOH (8.4 mL) under H2 atmosphere (balloon) was added TMSCl
(0.121 mL, 0.951 mmol) with a syringe below the surface of the liquid.
After being stirred under H2 atmosphere (balloon) at 30 °C for 23 h,
the mixture was filtered through a pad of Celite and the filtrate was
eluted through a column of Amberlyst A26 with MeOH as eluent.
(The Amberlyst A26 column had been previously flushed with
MeOH). MeOH was evaporated under reduced pressure, and the
residue was purified by chromatography on silica gel (eluent: CH2Cl2/
MeOH = 25:1), to yield indolizidine 13 (22.1 mg, 46%) along with
(2R,3R,4R)-3,4-di-tert-butoxy-2-(4-ethoxybutyl)pyrrolidine (9 mg,
17%).
EXPERIMENTAL SECTION
■
General Information. Rf values refer to TLC on 0.25 mm silica gel
1
plates. H and 13C NMR data are reported in δ (ppm) relative to
CDCl3 (7.26 and 77.0 ppm) or methanol-d4 (3.31 and 49.05 ppm),
and peak assignments were made on the basis of 1H−1H COSY,
HSQC, and HMBC experiments.
(2R,3R,4R)-2-[3-(5,5-Dimethyl-1,3-dioxan-2-yl)propyl]-3,4-di-tert-
butoxypyrrolidin-1-ol (11). A solution of chloroacetal 811 (999.5 mg,
5.19 mmol) and 1,2-dibromoethane (156 mg, 0.83 mmol) in
anhydrous THF (1 mL) was slowly added under stirring to
magnesium turnings (354 mg, 14.57 mmol) under nitrogen
atmosphere at 30 °C. The mixture was maintained at ca. 30 °C by
periodic removal of the warm bath for 1.5 h and then diluted with
THF (3 mL) and cooled to 0 °C. Nitrone 106b (662.5 mg, 2.89 mmol)
in THF (7 mL) was added dropwise over 20 min at 0 °C. The
resulting mixture was allowed to warm to rt and stirred for 2 h further.
A saturated aqueous solution of NH4Cl (9 mL) was added dropwise at
0 °C, the mixture was filtered through a cotton plug, the layers were
separated, and the aqueous layer was extracted with CH2Cl2 (4 × 8
mL). The combined organic layers were dried over Na2SO4, filtered,
and concentrated under reduced pressure. The crude product (1.57 g)
was purified by chromatography on silica gel [eluent: initially
petroleum ether/AcOEt = 3:1 followed in sequence by petroleum
ether/AcOEt = 3:2, AcOEt 100%, AcOEt/MeOH = 25:1, and AcOEt/
MeOH (NH3 1%) = 10:1] to yield hydroxylamine 11 as a colorless
viscous oil (861 mg, 77%) along with unreacted nitrone 10 (93.3 mg,
14%) and chloroacetal 8 (302 mg, 30%).
23
13: Rf = 0.22 (CH2Cl2/MeOH = 20:1); [α]D = −42 (c = 0.58,
CHCl3) [lit. ent-13: [α]D21 = +42.8 (c = 0.48, CHCl3); [α]D25 = +42.0
(c = 0.54, CHCl3)];6a,8c 1H NMR (CDCl3, 400 MHz): δ = 3.77 (ddd, J
= 7.2; 4.0; 1.6 Hz, 1H, 2-H), 3.62 (dd, J = 8.7; 4.0 Hz, 1H, 1-H), 2.95−
2.88 (m, 1H, 5-Ha), 2.90 (dd, J = 9.9; 1.6 Hz, 1H, 3-Ha), 2.41 (dd, J =
9.9; 7.2 Hz, 1H, 3-Hb), 1.94−1.50 (m, 6H, 5-Hb + 6-H + 7-Ha + 8a-H
+ 8-Ha), 1.29−1.10 (partially obscured, m, 2H, 7-Hb + 8-Hb), 1.20 (s,
9H, CH3 × 3), 1.16 (s, 9H, CH3 × 3) ppm; 13C NMR (CDCl3, 100
MHz): δ = 83.8 (d, C-1), 76.9 (d, C-2), 73.7 (s, Me3CO), 73.5 (s,
Me3CO), 67.1 (d, C-8a), 62.2 (t, C-3), 53.6 (t, C-5), 29.3 (q, 3 C, CH3
× 3), 28.7 (q, 3 C, CH3 × 3) and (t, C-8), 24.9 (t, C-6), 24.1 (t, C-7)
ppm. Spectral properties were identical to those reported for ent-136a
(the resonance of protons 7-Hb and 8-Hb, which are partially
obscured by the intense singlets of the two t-Bu groups and were
previously erroneously assigned above 1.47−1.45 ppm,6a,8c could be
25
11: Rf = 0.18 (petroleum ether/AcOEt = 3:1); [α]D = −24 (c =
0.71, CHCl3); 1H NMR (CD3OD, 400 MHz): δ = 4.48−4.44 (m, 1H,
OCHO), 3.93 (pseudo dt, J = 6.8; 3.2 Hz, 1H, 4-H), 3.69−3.63 (m,
1H, 3-H), 3.56 (A part of an AB system, J = 11.0 Hz, 2H, OCHH × 2),
3.45 (B part of an AB system, J = 11.0 Hz, 2H, OCHH × 2), 3.15 (A
part of an ABX system, J = 11.1; 2.8 Hz, 1H, 5-Ha), 3.02 (B part of an
ABX system, J = 11.1; 6.9 Hz, 1H, 5-Hb), 2.61−2.54 (m, 1H, 2-H),
1.66−1.54 (m, 6H, CH2CH2CH2), 1.22 (s, 9H, CH3 × 3), 1.19 (s, 9H,
CH3 × 3), 1.16 (s, 3H, CH3), 0.72 (s, 3H, CH3) ppm; 13C NMR
(CD3OD, 100 MHz): δ = 103.4 (d, OCHO), 82.4 (d, C-3), 78.1 (t,
2C, CH2O × 2), 77.7 (d, C-4), 75.4 (s, Me3CO), 75.0 (s, Me3CO),
74.0 (d, C-2), 65.6 (t, C-5), 36.0 (t, CH2), 31.2 (t, CH2), 31.0 (s,
Me2C), 29.6 (q, 3C, CH3 × 3), 29.2 (q, 3C, CH3 × 3), 23.5 (q, CH3),
22.1 (q, CH3), 21.5 (t, CH2) ppm; IR (CDCl3): ν = 3690, 3580, 3378,
2978, 2853, 1472, 1463, 1394, 1366, 1190 cm−1; MS (+ESI): m/z =
388 [M + H]+, 410 [M + Na]+. C21H41NO5 (387.6): calcd C 65.08, H
10.66, N 3.61; found C 65.09, H 10.38, N 3.28.
1
observed by H−1H COSY, HSQC, and TOCSY experiments, see
copies of the corresponding spectra in Supporting Information).
(1R,2R,8aR)-Octahydroindolizine-1,2-diol [(−)-Lentiginosine,
(−)-1]. Indolizidine 13 (164 mg, 0.61 mmol) was dissolved in TFA
(1.9 mL) at 0 °C. The mixture was stirred at rt for 16 h and then
concentrated under reduced pressure. The last traces of TFA were
coevaporated with MeOH, and then the residue was filtered through a
column of of Amberlyst A26 with MeOH as eluent. MeOH was
evaporated under reduced pressure, and the crude product (103 mg)
was purified by chromatography on silica gel [eluent: CH2Cl2/MeOH
(NH3 33%) = 41:8. 1], to yield (−)-lentiginosine (82 mg, 85%).
(1R,2R,8aR)-1,2-Di-tert-butoxyoctahydroindolizine (13). Proce-
dure A. To a mixture of 11 (123 mg, 0.317 mmol) and Pd/C (10%
in weight, 34 mg; Pd: 3.4 mg, 10 mol %) in MeOH (16 mL) under H2
atmosphere (balloon) was added dropwise TMSCl (0.121 mL, 0.951
mmol) at 0 °C. After being stirred under H2 atmosphere (balloon) at
50 °C for 44 h, the mixture was filtered through a pad of Celite and the
C
DOI: 10.1021/acs.joc.5b02804
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
20
(−)-1: Rf = 0.33 [CH2Cl2/MeOH/NH3 (33%) = 41:8:1]; [α]D
=
(10) Forbes, C. P.; Wenteler, G. L.; Wiechers, A. J. Chem. Soc., Perkin
Trans. 1 1977, 2353−2355.
−1.6 (c = 0.92, CH3OH) [lit. [α]D23 = −1.6 (c = 0.24, CH3OH)];2 1H
NMR (CD3OD, 400 MHz): δ = 3.94 (ddd, J = 7.2; 3.5; 1.5 Hz, 1H, 2-
H), 3.59 (dd, J = 8.5; 3.5 Hz, 1H, 1-H), 2.95 (dm, J = 10.9 Hz, 1H, 5-
Ha), 2.85 (dd, J = 10.5, 1.5 Hz, 1H, 3-Ha), 2.52 (dd, J = 10.5; 7.2 Hz,
1H, 3-Hb), 1.98 (pseudo dt, J = 3.3; 11.4 Hz, 1H, 5-Hb), 1.97−1.93
(m, 1H, 8-Ha), 1.85−1.74 (m, 2H, 7-Ha + 8a-H), 1.66−1.48 (m, 2H,
6-H), 1.33−1.17 (m, 2H, 7-Hb + 8-Hb) ppm. The spectral properties
were identical to those previously reported.2
(11) Stowell, J. C.; Polito, M. A. J. Org. Chem. 1992, 57, 2195−2196.
(12) For stereoselective addition reactions of organometallic
compounds to nitrone 10 and ent-10, see: (a) Merino, P.; Tejero,
T.; Revuelta, J.; Romero, P.; Cicchi, S.; Mannucci, V.; Brandi, A.; Goti,
A. Tetrahedron: Asymmetry 2003, 14, 367−379. (b) Pulz, R.; Cicchi, S.;
Brandi, A.; Reissig, H.-U. Eur. J. Org. Chem. 2003, 2003, 1153−1156.
(c) Goti, A.; Cicchi, S.; Mannucci, V.; Cardona, F.; Guarna, F.;
Merino, P.; Tejero, T. Org. Lett. 2003, 5, 4235−4238. (d) Delso, I.;
Marca, E.; Mannucci, V.; Tejero, T.; Goti, A.; Merino, P. Chem. - Eur. J.
2010, 16, 9910−9919. (e) Morozov, D. A.; Kirilyuk, I. A.; Komarov,
D. A.; Goti, A.; Bagryanskaya, I. Y.; Kuratieva, N. V.; Grigor’ev, I. A. J.
Org. Chem. 2012, 77, 10688−10698 and refs 8b, c..
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.5b02804.
(13) Krull, O.; Wunsch, B. Bioorg. Med. Chem. 2004, 12, 1439−1451.
̈
(14) For a selection of recent syntheses of polyhydroxyindolizidines
by reductive amination, see: (a) Pawar, N. J.; Parihar, V. S.; Khan, A.;
Joshi, R.; Dhavale, D. D. J. Med. Chem. 2015, 58, 7820−7832.
(b) Bergeron-Brlek, M.; Meanwell, M.; Britton, R. Nat. Commun.
2015, 6, 6903. (c) Eum, H.; Choi, J.; Cho, C.-G.; Ha, H.-J. Asian J. Org.
Chem. 2015, 4, 1399−1409.
Copies of NMR spectra (PDF)
AUTHOR INFORMATION
Corresponding Authors
*E-mail: franca.cordero@unifi.it.
*E-mail: alberto.brandi@unifi.it.
■
(15) In previous syntheses of polyhydroxyindolizidines by reductive
amination a high pressure of hydrogen (50−1300 psi) was often
utilized. For example, see refs 14a, b.
(16) In some experiments the formation of small amounts of one of
these two byproducts was observed: 3,4-di-tert-butoxy-2-(3-methox-
ypropyl)pyrrolidine likely derived from hydrogenation of the
intermediate methyloxonium ion and 1,2-di-tert-butoxyoctahydroindo-
lizin-5-yl)methanol, whose formation is not readily explained and is
under investigation.
(17) The relatively lower yield can be attributed to the formation of
non-negligible amounts of the byproduct (2R,3R,4R)-3,4-di-tert-
butoxy-2-(4-ethoxybutyl)pyrrolidine (17%) derived from hydrogena-
tion of the intermediate ethyloxonium ion (see also ref 16).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank the Italian Ministry of University and
Research (MIUR-Rome, PRIN20109Z2XRJ) and Ente Cassa di
Risparmio di Firenze (2013/7921) for financial support.
REFERENCES
■
(1) Pastuszak, I.; Molyneux, R. J.; James, L. F.; Elbein, A. D.
Biochemistry 1990, 29, 1886−1891.
(2) Brandi, A.; Cicchi, S.; Cordero, F. M.; Frignoli, R.; Goti, A.;
Picasso, S.; Vogel, P. J. Org. Chem. 1995, 60, 6806−6812.
(3) Dal Piaz, F.; Vassallo, A.; Chini, M. G.; Cordero, F. M.; Cardona,
F.; Pisano, C.; Bifulco, G.; De Tommasi, N.; Brandi, A. PLoS One
2012, 7, e43316.
(4) (a) Macchi, B.; Minutolo, A.; Grelli, S.; Cardona, F.; Cordero, F.
M.; Mastino, A.; Brandi, A. Glycobiology 2010, 20, 500−506.
(b) Minutolo, A.; Grelli, S.; Marino-Merlo, F.; Cordero, F. M.;
Brandi, A.; Macchi, B.; Mastino, A. Cell Death Dis. 2012, 3, e358.
(5) For two recent reviews, see: (a) Cardona, F.; Goti, A.; Brandi, A.
Eur. J. Org. Chem. 2007, 2007, 1551−1565. (b) Cordero, F. M.; Giomi,
D.; Brandi, A. Curr. Top. Med. Chem. 2014, 14, 1294−1307.
(6) (a) Cardona, F.; Goti, A.; Picasso, S.; Vogel, P.; Brandi, A. J.
Carbohydr. Chem. 2000, 19, 585−601. (b) Cordero, F. M.; Bonanno,
P.; Khairnar, B. B.; Cardona, F.; Brandi, A.; Macchi, B.; Minutolo, A.;
Grelli, S.; Mastino, A. ChemPlusChem 2012, 77, 224−233. (c) Cordero,
F. M.; Bonanno, P.; Neudeck, S.; Vurchio, C.; Brandi, A. Adv. Synth.
Catal. 2009, 351, 1155−1161.
(7) (a) Cordero, F. M.; Bonanno, P.; Chioccioli, M.; Gratteri, P.;
Robina, I.; Moreno Vargas, A. J.; Brandi, A. Tetrahedron 2011, 67,
9555−9564. (b) Cordero, F. M.; Vurchio, C.; Macchi, B.; Minutolo,
A.; Brandi, A. ARKIVOC (Gainesville, FL, U. S.) 2014, iii, 215−217.
(c) Vurchio, C.; Cordero, F. M.; Faggi, C.; Macchi, B.; Frezza, C.;
Grelli, S.; Brandi, A. Tetrahedron 2015, 71, 5806−5813.
(8) For previous syntheses of (+)-lentiginosine through organo-
metallic addition to nitrone ent-2 (R1 = MOM, t-Bu), see:
(a) Giovannini, R.; Marcantoni, E.; Petrini, M. J. Org. Chem. 1995,
60, 5706−5707. (b) Cardona, F.; Moreno, G.; Guarna, F.; Vogel, P.;
Schuetz, C.; Merino, P.; Goti, A. J. Org. Chem. 2005, 70, 6552−6555.
(c) Wozniak, Ł.; Staszewska-Krajewska, O.; Michalak, M. Chem.
́
Commun. 2015, 51, 1933−1936.
(9) (a) Feugeas, C.; Normant, H. Bull. Chem. Soc. Fr. 1963, 1441−
1443. (b) Feugeas, C. Bull. Chem. Soc. Fr. 1963, 2568−2579.
(c) Huang, J.-W.; Chen, C.-D.; Leung, M.-k. Tetrahedron Lett. 1999,
40, 8647−8650.
D
DOI: 10.1021/acs.joc.5b02804
J. Org. Chem. XXXX, XXX, XXX−XXX