(-)-(1R,2R,7S,8aR)-1,2,7-trihydroxyindolizidine ((-)-7S-OH-lentiginosine): Synthesis and proapoptotic activity

Article abstract of DOI:10.1002/cplu.201100069

An improved approach for the preparation of enantiopure 3,4- bis-tert-butoxypyrroline N-oxides is presented. Etherification of 1-benzylpyrrolidine-3,4-diol with tBuOAc/HClO₄ and subsequent N-debenzylation and pyrrolidine oxidation with oxone affords the cyclic nitrone reliably and in superior yield. The enantiomer derived from D-tartaric acid was exploited in a modified synthesis of (-)-7S-OH-lentiginosine. The activity of this trihydroxy indolizidine in inducing the apoptosis of tumour cell lines of lymphoid and epithelial origin is examined.

Full text of DOI:10.1002/cplu.201100069

DOI: 10.1002/cplu.201100069
(À)-(1R,2R,7S,8aR)-1,2,7-Trihydroxyindolizidine ((À)-7S-OH-
Lentiginosine): Synthesis and Proapoptotic Activity
Franca M. Cordero,*[a] Paola Bonanno,[a] Bhushan B. Khairnar,[a] Francesca Cardona,[a]
Alberto Brandi,[a] Beatrice Macchi,*[b] Antonella Minutolo,[c] Sandro Grelli,[c] and
Antonio Mastino[d]
Dedicated to Professor Barry M. Trost on the occasion of his 70th birthday
An improved approach for the preparation of enantiopure 3,4-
bis-tert-butoxypyrroline N-oxides is presented. Etherification of
1-benzylpyrrolidine-3,4-diol with tBuOAc/HClO4 and subse-
quent N-debenzylation and pyrrolidine oxidation with oxone
affords the cyclic nitrone reliably and in superior yield. The
enantiomer derived from d-tartaric acid was exploited in a
modified synthesis of (À)-7S-OH-lentiginosine. The activity of
this trihydroxy indolizidine in inducing the apoptosis of
tumour cell lines of lymphoid and epithelial origin is examined.
Introduction
Iminosugars are a class of natural products characterised by a
polyhydroxylated monocyclic or bicyclic structure containing a
nitrogen atom in the ring. These compounds can inhibit glyco-
sidase enzymes involved in posttranslational processing of gly-
coproteins and lysosomal catabolism of glycoconjugates. The
glycosidase inhibition activity of these compounds is related to
their structural resemblance to the sugar moiety, the natural
substrate of glycosidases. For this peculiar bioactivity, iminosu-
gars have received growing attention as potential therapeutic
agents for several pathologies such as tumour metastasis, viral
infections, diabetes, and genetic disorders.[1,2]
same bioactivity we decided to examine another indolizidine,
(À)-(1R,2R,7S,8aR)-1,2,7-trihydroxyindolizidine ((À)-7S-OH-lenti-
ginosine, 3), as a proapoptotic agent. The enantiomer of 3 is a
non-natural amyloglucosidase inhibitor,[6] that, in its protected
form, is an advanced intermediate in two syntheses of lentigi-
nosine,[7] and has been employed to produce lentiginosine
conjugates.[8] The important biological profile of these com-
pounds in both enantiomeric forms suggested, at first, the im-
provement of the synthesis of 3, and consequently of 2, to
render them more available for biological tests and other ap-
plications. Herein, we report a revisited synthesis of (À)-7S-OH-
lentiginosine 3, which benefits of a novel, efficient, and practi-
Among the natural indolizidine iminosugars, (+)-lentigino-
sine[3] (1) is one of the more recently discovered, but has al-
ready attracted much interest because of the potent glycosi-
dase inhibition activity despite its simple dihydroxylated struc-
ture, and also because of the debate that arose about its abso-
lute configuration.[4]
[a] Prof. F. M. Cordero, Dr. P. Bonanno, Dr. B. B. Khairnar, Dr. F. Cardona,
Prof. A. Brandi
Department of Chemistry “Ugo Schiff” and HeteroBioLab
University of Florence
Via della Lastruccia 13, 50019 Sesto Fiorentino, Florence (Italy)
Fax: (+39)0554573531
E-mail: franca.cordero@unifi.it
[b] Dr. B. Macchi
Department of Neuroscience, University of Rome Tor Vergata
Via Montpellier 1 Rome 00133 and IRCCS S. Lucia
Via Ardeatina 300 00100 Rome (Italy)
Fax: (+39)0620427282
E-mail: macchi@med.uniroma2.it
[c] Dr. A. Minutolo, Prof. S. Grelli
Department of Experimental Medicine and Biochemical Sciences
University of Rome Tor Vergata
Via Montpellier 1, Roma 00133 (Italy)
[d] Prof. A. Mastino
Department of Life Sciences “M. Malpighi”
University of Messina, Via F. Stagno d’Alcontres 31, Messina 98166 (Italy)
and
IRCCS Centro Neurolesi “Bonino-Pulejo”
University of Messina, Salita dello Sperone 31, Messina 98168 (Italy)
Moreover, recently it was found that the enantiomeric
(À)-lentiginosine (2) is a potent proapoptotic agent against dif-
ferent strains of cancer cells, but with low cytotoxicity towards
normal cells.[5] In search of new candidates which display the
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201100069.
224
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 77, 224 – 233

Indolizidine Alkaloids
cal approach to the key intermediate 3,4-bis-tert-butoxypyrro-
line N-oxide (4a, R=tBu), which is also a valuable building
block in organic synthesis. The study of a three-step, one-pot
version of the previous synthesis of the indolizidine skeleton
and the effects of 3 on cell viability and apoptosis of cell lines
of different origin are also described.
Results and Discussion
Synthesis of 3,4-bis-tert-butoxypyrroline N-oxide (4a)
Scheme 2. Previous synthesis of nitrone 4a. py=pyridine, Ts=4-toluenesul-
The reported syntheses of 3 and its enantiomer ent-3 are
based on the 1,3-dipolar cycloaddition (1,3-DC) of an enantio-
pure dihydroxylated pyrroline N-oxide 4 derived from tartaric
acid (Scheme 1).[6,7,9] At present, the most efficient synthesis[7a,c]
fonyl.
ducibility of the reaction is troublesome, because during neu-
tralization a trace of acid can quickly shift the equilibrium from
6 towards the monoprotected diol 8 with loss of selectivity
and yield (Scheme 3). For this reason, a more practical and reli-
able synthesis of the valuable nitrone 4a is highly desirable.
Scheme 1. Retrosynthetic strategy to (À)-7S-OH-lentiginosine 3.
involves the diastereoselective 1,3-DC of (3R,4R)-3,4-bis(tert-
butoxy)-1-pyrroline N-oxide (4a, R=tBu) with 3-buten-1-ol and
subsequent elaboration of the cycloadduct. Accordingly, the
first and most important goal to increase and facilitate the ac-
cessibility of (À)-7S-OH-lentiginosine 3 was to improve the syn-
thesis of nitrone 4a (R=tBu).
Nitrones 4 are very useful and versatile synthetic intermedi-
ates that can be prepared in both enantiomeric forms starting
from the easily available “chiral-pool” compounds l- and d-tar-
taric acid,[10] and have been used in various highly selective re-
actions.[8a,11–15]
Scheme 3. Equilibrium between di- and monoprotected ester.
Recently, tert-butylation of primary and secondary alcohols
by tert-butyl acetate in the presence of a catalytic amount of
HClO4 at room temperature has been reported.[22] Under these
very practical conditions, dimethyl tartaric ester (5, R=Me) is
converted into the di- and mono-tert-butyl ether (8 and 6, R=
Me) in 43% and 41% yield, respectively, after 44 hours. The
major drawback is the need for a highly diluted solution
(0.0025m) to form equimolar amounts of the di- and monoeth-
er, otherwise, the monoether derivative is the principal prod-
uct. As the authors stated, under the reported conditions cyclic
alcohols react with tert-butyl acetate better than with acyclic
acetates;[22] accordingly, we tested this procedure on a cyclic
derivative of tartaric acid such as imide 9.
A reaction performed on 1 g of imide 9[23] in a 0.1m solution
of tBuOAc in the presence of a catalytic amount of HClO4
(0.1 equiv) afforded, after 18 hours, the ethers 10 and 11 in a
3:1 ratio with 91% overall yield (Scheme 4). This excellent
result suggested the feasibility of a complementary approach
to nitrone 4a that involves oxidation of a protected dihydroxy
pyrrolidine in turn obtained by reduction of a suitable imide of
tartaric acid. This method has been previously applied to the
synthesis of nitrones 4 protected as methoxymethyl, silyl
ethers, or benzoates[19a,3,6,8a] (4; R=MOM, TBDPS, TBDMS, and
Bz), but not to the tert-butyl derivate. Unfortunately, reduction
of imide 10 with LiAlH4 in THF at reflux afforded the corre-
sponding benzyl pyrrolidine 12 in very low yield (17%), along
with 1-benzyl-3,4-di-tert-butoxy-1H-pyrrole (Scheme 4). The ar-
omatization of some succinimide derivatives induced by LiAlH4
was already reported, and could be avoided by using milder
Several protecting groups have been used to mask the two
trans-related hydroxy moieties (Me,[16] Bn,[16,17] tBu,[16,18]
MOM,[19,7b] TBDPS,[3] TBDMS,[6,20] Bz[8a]). Among them, the tert-
butyl group is the most convenient because of its high stability
under many different reaction conditions, the easy and trace-
less hydrolysis, and its bulkiness that can induce superior dia-
stereofacial control compared to smaller protecting groups.
The drawback of the tert-butyl ether protection that limits an
even larger use of nitrone 4a (R=tBu) is the tricky introduc-
tion of this bulky group on the two vicinal hydroxy groups.
The reported synthesis of the tert-butyl derivative 4a (R=
tBu) involves the following five-step sequence: a) tert-butyla-
tion of a tartaric ester 5, b) reduction of 6 with LiAlH4, c) tosyla-
tion of the primary hydroxy groups, d) cyclization with hydrox-
ylamine, and e) oxidation of the N-hydroxypyrrolidine
7
(Scheme 2).[16,18] Etherification of the two vicinal hydroxy
groups of 5 (step a) is troublesome, especially with bulky alkyl
groups. tert-Butylation, achieved by treatment the ester 5 in
solution in the presence of a strong acid by bubbling isobu-
tene or with liquid isobutene under pressure, gave the di-tert-
butyl ether of dimethyl tartrate 6 (R=Me) in 39% yield, to-
gether with the monoether 8 in 23% yield.[21] Moreover, repro-
ChemPlusChem 2012, 77, 224 – 233
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
225

F. M. Cordero, B. Macchi et al.
further purification, especially because chromatography on
silica gel resulted in a partial decomposition of 4a. Starting
from l-tartaric acid the enantiomeric, nitrone ent-(+)-4a was
prepared following the same procedure.
In conclusion, nitrone 4a was obtained in five steps starting
from tartaric acid, through the intermediate formation of pyr-
rolidine 12, in 46% overall yield (Scheme 7). Accordingly, com-
pared to the previous synthesis (five steps from tartaric ester,
25–35% overall yield; Scheme 2) a significant improvement in
yield of 4a, and reproducibility of the synthetic process, was
accomplished.
Scheme 4. tert-Butylation of imide 9. Bn=benzyl.
reducing reagents such as borane.[24] However, this possibility
was not explored, because tert-butylation of the dihydroxy pyr-
rolidine 13[23,8a] was even more selective than etherification of
imide 9 and afforded a 5.6:1 mixture of 12 and 14 in high
overall yield (92%; Scheme 5). Analogously to the previous
case, the reaction could be run on gram scale (2.8 g of 13)
using a 0.1m concentration; however, owing to the presence
Scheme 7. New synthesis of nitrone 4a.
Synthesis of (À)-7S-OH-lentiginosine (3)
The cycloaddition/cyclization/reductive ring-opening steps[7,26]
to 1,2,7-trihydroxyindolizidine were then reinvestigated. In the
reported synthesis, the 1,3-DC reaction of 4a with butenol 16
(5 equiv) in benzene at 608C for two days afforded a mixture
of diastereomeric cycloadducts in 10:2:1 ratio with complete
regioselectivity.[7c] The major exo-[4-OtBu] anti-adduct 18 was
then obtained in 74% yield after chromatography (Scheme 8,
step a; Table 1, Entry 1). Similar result can be obtained in
1.5 hours at 1008C. In this case, only two adducts were ob-
served in 5:1 ratio and adduct 18 was isolated in 75% yield
(Scheme 8,step a; Table 1, Entry 2). Cycloadduct 18 was then
converted into indolizidine 23 through activation of the pri-
mary hydroxy group by mesylation and subsequent spontane-
ous cyclization to salt 21 that was directly reduced with H2
Scheme 5. tert-Butylation of pyrrolidine 13.
of the basic nitrogen atom of 13, an excess amount of HClO4
(1.5 equiv) was required. Both the reagent/solvent tBuOAc and
the side product 14 could be recovered and reused (see the
Experimental Section for details). Notably, the etherification of
13 with isobutene in the presence of an acid was not success-
ful in our hands.
Hydrogenolysis of the benzyl group in 12 was quite tricky. In
the presence of a catalytic amount of Pd/C or Pd(OH)2/C and
H2 (1 atm) in MeOH, debenzylation was slow and surprisingly a
partial hydrolysis of the tert-butyl ether occurred with forma-
tion of a significant amount of 4-tert-butoxypyrrolidin-3-ol.
Luckily, in the presence of AcOH (10 equiv), debenzylation was
complete after 2.5 hours and ether hydrolysis was suppressed
(Scheme 6). After neutralization, 15 was obtained in 87% yield
and its oxidation with oxone[25,8a] afforded crude nitrone 4a in
93% yield. Both crude 15 and 4a were pure according to spec-
troscopic analysis and were used in the next reaction without
Scheme 6. N-debenzylation of pyrrolidine 12 and subsequent oxidation to
nitrone 4a. EDTA=ethylenediaminetetraacetic acid, THF=tetrahydrofuran.
Scheme 8. Synthesis of (À)-7S-OH-lentiginosine (3). Ms=methanesulfonyl,
TFA=trifluoroacetic acid.
226
www.chempluschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 77, 224 – 233

Indolizidine Alkaloids
Table 1. Reagents and conditions of the 1,3-DC/cyclization/NÀO bond
Table 2. Effects of compounds 3 and SN38 on viability, as assessed by
the MTS assay in tumour cell lines.
reduction sequence (see Scheme 8).
Entry Step a
Step b
Step c
Yield of
Cell lines
MOLT-3
Compounds
MAIC5O ÆSD [mm][a]
485Æ16.62
14.3Æ2
23 [%][b]
(À)-7S-OH-lentiginosine
1[a]
16 (5 equiv), C6H6,
608C, 2 days
MsCl, TEA,
dry CH2Cl2
H2 (45 psi), Pd/C,
MeOH, 1 day
64
64
60
SN38
(À)-7S-OH-lentiginosine
396Æ9.09
<0.1
SH-SY5Y
HT-29
2
16 (5 equiv), toluene, MsCl, TEA,
H2, Pd/C, MeOH,
15 h
SN38
1008C, 1.5 h
dry CH2Cl2
(À)-7S-OH-lentiginosine
285Æ35.3
<1
3
17 (4 equiv), RT, 3 days
Zn, AcOH, Et2O
SN38
[a] Reference [7c]: synthesis of ent-3. [b] Overall yield with respect to 4a.
TEA=triethylamine.
(À)-7S-OH-lentiginosine
355.50Æ78.56
<1
MCF-7
M-14
SN38
(À)-7S-OH-lentiginosine
396.53Æ49.02
<1
SN38
without isolation in the presence of a catalytic amount of Pd/C
(Scheme 8, steps b and c; Table 1, Entries 1 and 2).[7c]
[a] MAIC50 value=metabolic activity cytotoxic inhibitory concentration
50%, MTS=3,4-(5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-
sulfophenyl)-2H-tetrazolium salt, SD=standard deviation.
To shorten the process, the use of a dipolarophile already
having a suitable leaving group, such as 3-butenyl tosylate 20
was explored. In this case, a one-pot, three-step reaction could
be carried out (Scheme 8; Table 1, Entry 3). The two-step
domino process consisting of 1,3-DC/intramolecular nucleo-
philic substitution[27] was run, in the presence of an excess
amount of dipolarophile 20 (4 equiv), at room temperature for
three days, because of the thermal instability of salt 22. Acti-
vated Zn, AcOH, and Et2O were then added to reduce the NÀO
bond. Three diastereomeric indolizidine compounds in 5:1:0.2
ratio (60%, 12%, and 2% yield, respectively) were obtained by
flash chromatography. The Zn/AcOH couple was more conven-
ient than H2 over Pd/C for the reductive cleavage of the isoxa-
zolidine NÀO bond, because it does not affect the excess
amount of tosylate 20 that was recovered almost quantitative-
ly. The same process was carried out starting from ent-4a to
obtain the enantiomeric ent-23. In summary, the present one-
pot procedure afforded 23 in similar overall yield (60% vs.
64%), but is more convenient, because is carried out at room
temperature, does not require anhydrous conditions, and is
more atom economical because the excess dipolarophile can
be recovered.
it similar sensitivity to treatment with 3, with MAIC50 values
ranging from 355 to 485 mm. Nevertheless, HT-29 cells were
slightly more sensitive. Conversely, the well-known cytotoxic
chemotherapeutic agent SN38, used as a positive control, was
highly toxic towards all the tested cell lines.
Next, it was verified whether the effect of the tested com-
pound on cell viability could be ascribed to the induction of
cell death by apoptosis. To this aim, cell lines were treated for
24 hours with 3 at concentrations ranging from 1 to 103 mm
and apoptosis was assayed by flow cytometric quantification
of hypodiploid nuclei. The results of apoptosis expressed as a
percentage of hypodiploid nuclei are shown in Figure 1, the
values being the meansÆSD of three independent experi-
ments. The results underline a dose effect response in the
presence of a concentration of 3 from 1 to 103 mm in all the
tested cell lines. However, the assayed cell lines exhibited dif-
ferent susceptibility to compound-induced apoptosis. The
MOLT-3, MCF-7, and M-14 cells were highly susceptible to
apoptosis induced by 3, with differences in comparison with
control cells that were significant at a concentration as low as
10 mm. Conversely, SH-SY5Y and HT-29 cells treated with 3
showed apoptosis levels that were significantly different from
control cells at a concentration of 102 mm or higher. The refer-
ence positive control SN38 was effective in inducing apoptosis
in all the cell lines assayed at a concentration of 10 mm, as ex-
pected.
Eventually, removal of the protecting group of indolizidine
23 by treatment with TFA afforded pure triol 3 in 87% yield.[7c]
In conclusion, the overall new process allows more straightfor-
ward access to (À)-7S-OH-lentiginosine 3 that is very promising
in the light of its interesting biological activity (see below).
Biological tests
To further characterise apoptosis induced by 3, we investi-
gated the effects of compound 3 on both upstream and down-
stream caspases. To this purpose, we treated MOLT-3, SH-SY5Y,
HT-29, MCF-7, and M-14 cells with 3 at 102 mm, that is, at the
minimum compound concentration equally able to induce
apoptosis in all the cell lines. The expression of both pro and
cleaved caspases 3 and caspases 8 after 6 and 24 hours of cul-
ture was investigated by immunoblotting. The results, reported
in Figure 2, show one representative experiment out of three
performed. Data reported in Table 3 refer to results of the
same experiment, quantified by densitometry analysis and ex-
The effect of 3 on both cell viability and apoptosis of tumour
cell lines of different origins was investigated. The cytotoxic
effect of 3 was assayed towards MOLT-3, SH-SY5Y, HT-29, MCF-
7, and M-14 cell lines (Table 2). The concentration values of
compound 3 and 7-ethyl-10-hydroxycamptothecin (SN38) able
to inhibit metabolic activity by 50% (MAIC50 value), detected
through formazan production using a colorimetric kit (MTS),
are reported in Table 2. The values represent the means Æ the
standard deviation (SD) of cumulative values from three differ-
ent experiments. The results indicate that most cell lines exhib-
ChemPlusChem 2012, 77, 224 – 233
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
227

F. M. Cordero, B. Macchi et al.
pressed as OD mmÀ2 after subtraction of the background. Pro-
caspase 3 was constitutively expressed both in untreated and
in treated cells except in MCF-7 and in M-14, where it was ex-
pressed only after treatment with compound 3. Densitometry
analysis values indicate that pro-caspase 3 expression was in-
creased, following 6 hours of treatment, by 2- and 4-fold only
in MCF-7 and in M-14 cells, respectively. Conversely, the
cleaved form of 17/24 KDa was highly expressed with respect
to the untreated control as early as after 6 hours and this ex-
pression was increased after 24 hours by 1.2- to 2-fold in SH-
SY5Y, HT-29, MOLT-3, and M-14 cells. Moreover, pro-caspase 8
expression was constitutive in MOLT-3, SH-SY5Y, MCF-7, and M-
14 cells. At 6 hours following treatment with 3, the expression
of pro-caspase 8 increased from 1.5- to 3-fold in the entire cell
lines except for HT-29 cells, in which a 7-fold increase was ob-
served.
These results highlights that 3, which is an enantiomer of a
synthetic glycosidase inhibitor, is endowed with a well detecta-
ble, caspase-related pro-apoptotic activity on a broad range of
tumour cell lines of lymphoid as well as epithelial origin. The
present study further extends our previous data indicating that
(À)-lentiginosine (2), which is the synthetic enantiomer of a
natural product, was able to induce apoptosis in different
tumour cells in vitro.[5] Although triol 3 [(À)-7S-OH-lentigino-
sine] differs only for the presence of the OH group at C7 with
respect to 2, its biological activity presented two distinct as-
pects. It was able to induce appreciable apoptosis in MOLT-3
cells starting from half the concentration with respect to (À)-
lentiginosine (2). Apoptosis induced by 3 was well detectable
starting from 24 hours after treatment, meanwhile (À)-lentigi-
nosine induced apoptosis as early as 18 hours after treatment.
The reason underlying this
Figure 1. A) Effects of the compound (À)-7S-OH-lentiginosine (3) on apopto-
sis in MOLT-3, SH-SY5Y, HT-29, MCF-7, and M-14 cell lines. Apoptosis was
evaluated as a percentage hypodyploid nuclei by flow cytometry analysis
after 24 hours of incubation with 10 mm 7-ethyl-10-hydroxycamptothecin
(SN38), and 1–1000 mm (À)-7S-OH-lentiginosine (3).*p<0.05; **p<0.001
with respect to control by Tukey’s HSD test.
slightly different biological be-
haviour is not clear. However, we
can notice that the delayed time
of 3 in inducing apoptosis with
respect to (À)-lentiginosine, fits
with the fact that caspase 3 and
caspase 8 expression peaked at
24 hours after treatment. In addi-
tion, maximal cytotoxic activity,
as detected by the MTS assay,
also occurred 24 hours after
treatment. The comparison of re-
sults of the MTS assay with
those of apoptosis suggests that
most of the cytotoxic activity ex-
erted by 3 towards the different
tumour cell lines was attributa-
ble to apoptosis, except for HT-
29 cells that were quite resistant
to apoptosis while more sensi-
tive than the other cell lines to
metabolic activity inhibition. We
Figure 2. A) Caspase 3 and caspase 8 expression in MOLT-3, SH-SY5Y, HT-29, MCF-7, and MCF-14 cells after 6 and
24 hours of incubation with 100 mm (À)-7S-OH-lentiginosine (3). Each blot was loaded with the same amount of
protein as demonstrated by the equal level of b-actin expression. The values of densitometry analysis expressed
as (OD–Bkg) mmÀ2 are shown in Table 3.
must observe that compound 3
is 50–100 times less potent as a
pro-apoptotic agent than etopo-
228
www.chempluschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 77, 224 – 233

Indolizidine Alkaloids
ments were made on the basis of
1H–1H COSY, HMQC, HSQC, and
HMBC experiments. See the Sup-
porting Information for NMR spec-
tra.
Table 3. Values of densitometry analysis expressed as (OD–Bkg) mmÀ2 (see Figure 2).
(OD–Bkg) [mmÀ2
]
Caspase 3
Caspase 8
(À)-7S-OH-lentiginosine [mm]
100
0
0
100
Time [h]
(3S,4S)-1-Benzyl-3,4-di-tert-
Cell lines
MOLT-3
Analysis
6
24
6
24
6
24
6
24
butoxy-2,5-pyrrolidinedione (10)
FL
CL
4151
15
5636
0.56
4100
1245
4156
2566
3125
0
3056
0
5698
0
6258
0
and
(3S,4S)-1-benzyl-3-tert-
butoxy-4-hydroxy-2,5-pyrrolidine-
dione (11): HClO4 (60%, 49.2 mL,
0.4 mmol) was added to a suspen-
sion of imide 9 (1 g, 4.5 mmol) in
tert-butyl acetate (45 mL, 0.1m) at
08C and the reaction mixture was
stirred at RT for 18 hours. The ob-
tained clear solution was treated
with base to give pH 9 by the slow
addition of a saturated aqueous
solution of Na2CO3 (20 mL) and
solid Na2CO3 at 08C. The two
layers were separated and the two
phases treated as follows. The tert-
butyl acetate solution was washed
FL
CL
2241
0
2236
15
2145
78
2845
146
2564
0
2589
0
2785
365
2956
415
SH-SH5Y
HT-29
FL
CL
3256
0
3145
0
2987
546
2145
688
56
0
6
0
417
0
895
0
FL
CL
389
0
658
0
826
54
1001
53
897
0
952
0
1282
0
1655
0
MCF-7
M-14
FL
CL
126
0
256
0
535.25
145
789
389
496
0
563
0
1558
0
1652
0
Bkg=background, CL=chemoluminescence staining, FL=flow cytometry analysis, OD=optical density.
with brine, dried over Na2SO4, filtered, and partially distilled under
reduced pressure at RT to recover the majority of the solvent
(ca. 60–65%) that could be recycled or used for other protection
reaction. The Na2SO4 used to dry the solution was thoroughly
washed with CH2Cl2 and the obtained solution combined with the
distillation residue. The aqueous solution was extracted with
CH2Cl2 (3ꢁ20 mL) and the combined organic layers were washed
with brine (30 mL), dried over Na2SO4, filtered, and concentrated
under reduced pressure. Purification by chromatography on silica
gel (eluent: CH2Cl2/MeOH first 99.5:0.5, then 99:1, and finally 98:2)
of the combined residues afforded 10 (1 g, 68%) as a yellow solid
and 11 (0.29 g, 23%) as a white solid.
side. However, the concentration of 3 necessary to cause a cy-
totoxic effect (MAIC50 value, Table 2) is 40 to 400 times higher
than that of etoposide, accordingly, indolizidine 3 is less cyto-
toxic than etoposide—analogously to the parent (À)-lentigino-
sine (2).[5]
Conclusion
The synthesis of the enantiopure 1,2,7-trihydroxyindolizidine 3
was revisited leading to a sensible overall improvement—par-
ticularly in the preparation of nitrone 4a, which is a key inter-
mediate not only in lentiginosine synthesis. It has been proved
that (À)-7S-OH-lentiginosine 3 is a proapoptotic agent versus
different tumour cell lines, and it has similar potency to that of
the parent compound 2. The results lead us to a more insight-
ful investigation to assess the potential antitumour effect of
the enantiomer of lentiginosine and its derivatives. Taken to-
gether these data demonstrate how these compounds, with a
structure resembling that of sugar mimetic glycosidase inhibi-
tors, endowed with previously unknown pro-apoptotic activity,
could constitute a new distinct class of pro-apoptotic agents.
Actually, recent data have shown how iminosugars, in general,
can be endowed both with cytotoxic and glycosidase inhibi-
tion activity on human cancer cell lines of different origin.[28]
These studies suggest that a more in-deep investigation
should be aimed at assess whether inhibition of glycosidase
might directly be involved in induction of apoptosis by these
compounds.
Compound 10: m.p. 56.2–598C; [a]D24 =À155 (c=0.7, CHCl3);
1H NMR (CDCl3, 400 MHz): d=7.40–7.35 (m, 2H, HPh), 7.33–7.24 (m,
3H, HPh), 4.64 (A part of an AB system, J=14.0 Hz, 1H, CHHN), 4.60
(B part of an AB system, J=14.0 Hz, 1H, CHHN), 4.36 (s, 2H, 3-H+
4-H), 1.30 (s, 18H, CH3 ꢁ6) ppm; 13C NMR (CDCl3, 50 MHz): d=173.3
(s; 2C, C-2+C-5), 135.4 (s; CPh), 129.0 (d; 2C, CHPh), 128.6 (d; 2C,
CHPh), 127.9 (d; CHPh), 76.4 (s; 2C, CMe3), 75.0 (d; 2C, C-3+C-4),
42.3 (t; CH2), 28.5 (q; 6C, CH3) ppm; IR (CDCl3): v=2980, 1718,
1393, 1371, 1118, 1080, 1025 cmÀ1; MS (EI): m/z (%)=333 ([M+],
0.5), 277 (7), 221 (28), 192 (11), 164 (5), 91 (36), 57 (100), 41 (33); el-
emental analysis (%) calcd for C19H27NO4 (333.4)=C 68.44, H 8.16,
N 4.20; found: C 68.57, H 8.02, N 4.17; Rf =0.7 (CH2Cl2/MeOH 99:1).
Compound 11: m.p. 103.8–105.48C; [a]D23 =À132 (c=0.7, CHCl3);
1H NMR (CDCl3, 400 MHz): d=7.40–7.27 (m, 5H, HPh), 4.05 (s, 2H,
CH2), 4.45 (br A part of an AB system, J=5.3 Hz, 1H, 4-H), 4.41 (B
part of an AB system, J=5.3 Hz, 1H, 3-H), 3.17 (br s, OH), 1.32 (s,
9H, CH3 ꢁ3) ppm; 13C NMR (CDCl3, 100 MHz): d=174.6 (s; C=O),
173.1 (s; C=O), 135.0 (s; CPh), 129.0 (d; 2C, CHPh), 128.7 (d; 2C, CHPh),
128.2 (d; CHPh), 76.5 (s; 2C, CMe3), 75.1 (d; CHO), 75.0 (d; CHO),
42.7 (t; CH2), 28.2 (q; 3C, CH3) ppm; IR (CDCl3): v=3577, 2981,
1720, 1394, 1371, 1104, 1078 cmÀ1; MS (EI): m/z (%)=277 ([M+], 1),
262 (6), 221 (32), 192 (8), 164 (5), 91 (55), 71 (34), 57 (100), 41 (33);
elemental analysis (%) calcd for C15H19NO4 (277.3)=C 64.97, H 6.91,
N 5.05; found: C 64.97, H 7.22, N 4.97; Rf =0.36 (CH2Cl2/MeOH 98:2).
Experimental Section
General experimental methods
The Rf values refer to TLC on silica gel plates of 0.25 mm. The NMR
data are reported in d (ppm) from TMS at 258C and signal assign-
ChemPlusChem 2012, 77, 224 – 233
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
229

F. M. Cordero, B. Macchi et al.
(3R,4R)-1-Benzyl-3,4-di-tert-butoxypyrrolidine (12) and (3R,4R)-1-
benzyl-4-tert-butoxy-3-pyrrolidinol (14): HClO4 (60%, 2.4 mL,
21.5 mmol) was added to a mixture of pyrrolidine 13 (2.8 g,
14.3 mmol) in tert-butyl acetate (145 mL, 0.1m) at 08C. The reac-
tion mixture was stirred at RT for 5.5 hours, cooled at 08C and
then treated with base to give pH 9 by the slow addition of a satu-
rated aqueous solution of Na2CO3 (20 mL) and solid Na2CO3. The
two layers were separated and the two phases treated as de-
scribed in the preparation of compounds 10 and 11. The crude
material was purified by chromatography on silica gel (eluent
CH2Cl2/MeOH first 99:1, then 97:3, and finally CH2Cl2/MeOH 95:5) to
afford 12 (3.4 g, 78%) as a yellow waxy solid and 14 (0.51 g, 14%)
as a white solid.
washing with MeOH. The filtrate was concentrated under reduced
pressure and the resulting oil was dissolved in CH2Cl2 (20 mL) and
treated with base to give pH 9 with a saturated aqueous Na2CO3
solution (40 mL) at 08C. The two layers were separated and the
aqueous solution was extracted with CH2Cl2 (3ꢁ40 mL). The com-
bined organic phases were washed with brine (60 mL), dried over
Na2SO4, filtered, and concentrated under reduced pressure to
afford 15 (1.4 g, 87%) as an oil, which was used in the next step
without further purification.
1
Compound 15: H NMR (CDCl3, 400 MHz): d=3.86–3.81 (m, 2H, 3-
H+4-H), 3.09 (dd, J=12.0, 5.3 Hz, 2H, 2-Ha +5-Ha), 2.66 (dd, J=
12.0, 3.5 Hz, 2H, 2-Hb +5-Hb), 2.41 (br s, NH), 1.17 (s, 18H, CH3 ꢁ6)
ppm; 13C NMR (CDCl3, 50 MHz): d=79.1 (d; 2C, C-3+C-4), 73.5 (s;
2C, CMe3), 53.9 (t; 2C, C-2+C-5), 28.6 (q; 6C, CH3) ppm.
Compound 12: [a]D23 =À88 (c=0.4, CHCl3); 1H NMR (CDCl3,
400 MHz): d=7.34–7.20 (m, 5H, HPh), 3.97–3.92 (m, 2H, 3-H+4-H),
3.63 (A part of an AB system, J=13.1 Hz, 1H, CHHPh), 3.49 (B part
of an AB system, J=13.1 Hz, 1H, CHHPh), 2.78 (pseudo dd, J=9.4,
6.4 Hz, 2H, 2-Ha +5-Ha), 2.39 (pseudo dd, J=9.4, 5.1 Hz, 2H, 2-Hb +
5-Hb), 1.16 (s, 18H, CH3 ꢁ6) ppm; 13C NMR (CDCl3, 100 MHz): d=
138.3 (s; CPh), 128.9 (d; 2C, CHPh), 128.1 (d; 2C, CHPh), 126.9 (d;
CHPh), 78.2 (d; 2C, C-3+C-4), 73.5 (s; 2C, CMe3), 60.6 (t; CH2Ph),
60.3 (t; 2C, C-2+C-5), 28.6 (q; 6C, CH3) ppm; IR (CDCl3): v=2977,
2797, 1391, 1366, 1186 cmÀ1; MS (EI): m/z (%)=305 ([M+], 9), 248
(16), 232 (3), 192 (28), 158 (6), 91 (100), 57 (92); elemental analysis
(%) calcd for C19H31NO2 (305.2)=C 74.71, H 10.23, N 4.59; found
C 74.37, H 10.24, N 4.95; Rf =0.33 (CH2Cl2/MeOH 97:3).
(3S,4S)-3,4-Di-tert-butoxypyrrolidine (ent-15): By following the
same procedure as for 15, the enantiomeric pyrrolidine ent-15 was
prepared starting from ent-12.
Compound ent-15: The spectroscopic properties are identical to
those of 15.
(3R,4R)-3,4-Di-tert-butoxy-3,4-dihydro-2H-pyrrole 1-oxide (4a):
NaHCO3 (4 g, 47.6 mmol) was added to a stirred solution of crude
15 (2 g, 9.3 mmol) in a 4:1 mixture of CH3CN/THF (18 mL) and
aqueous Na2EDTA (0.01m, 13 mL). The reaction mixture was cooled
at 08C and oxone (7.6 g, 12 mmol) was added portion wise over
5 hours. The mixture was stirred at 08C for 30 minutes and then di-
luted with EtOAc (30 mL) and deionised H2O (50 mL). The two
phases were separated and the aqueous solution was sequentially
extracted with EtOAc (2ꢁ20 mL) and CH2Cl2 (3ꢁ20 mL). The com-
bined organic layers were washed with brine (50 mL), dried over
Na2SO4, and concentrated under reduced pressure to afford spec-
troscopically pure 4a (2.023 g, 93%) as a yellow solid that can be
used in the next step without further purification. Purification by
chromatography on silica gel (eluent: CH2Cl2/MeOH (1% NH4OH)
98:2) afforded analytically pure 4a (1.7 g, 78%) as a white solid.
Compound 14: m.p. 78.6–81.88C; [a]D23 =À10 (c=0.5, CHCl3);
1H NMR (CDCl3, 400 MHz): d=7.35–7.24 (m, 5H, HPh), 4.03–3.98 (m,
2H, 3-H+4-H), 3.70 (A part of an AB system, J=12.9 Hz, 1H,
CHHPh), 3.66 (B part of an AB system, J=12.9 Hz, 1H, CHHPh), 3.26
(dd, J=10.1, 6.7 Hz, 1H) and 2.82 (broad d, J=10.4 Hz, 1H) (2-Ha
and 5-Ha), 2.68 (dd, J=10.4, 5.1 Hz, 1H) and 2.34 (dd, J=10.1,
5.4 Hz, 1H) (2-Hb and 5-Hb), 1.19 (s, 9H, CH3 ꢁ3) ppm; 13C NMR
(CDCl3, 50 MHz): d=137.3 (s; CPh), 128.9 (d; 2C, CHPh), 128.3 (d;
2C, CHPh), 127.4 (d; CHPh), 78.6 (d), 78.2 (d; C-3 and C-4), 74.1 (s;
CMe3), 60.6 (t), 60.4 (t; C-2 and C-5), 60.1 (t; CH2Ph), 28.4 (q; 3C,
CH3) ppm; IR (CDCl3): v=2978, 2801, 1392, 1366, 1192, 1090 cmÀ1
;
Compound 4a: mp=75.4–77.28C; [a]D27 =À166 (c=0.7, CHCl3);
1H NMR (CDCl3, 400 MHz): d=6.78–6.76 (m, 1H, 5-H), 4.59–4.56 (m,
1H, 4-H), 4.20–4.12 (m, 2H, 2-Ha +3-H), 3.71–3.64 (m, 1H, 2-Hb),
1.21 (s, 9H, CH3 ꢁ3), 1.19 (s, 9H, CH3 ꢁ3) ppm; 13C NMR (CDCl3,
50 MHz): d=134.9 (d; C-5), 79.0 (d; C-4), 74.8 (s; CMe3), 74.7 (s;
CMe3), 74.2 (d; C-3), 68.3 (t; C-2), 28.2 (q; 6C, CH3) ppm; IR (CDCl3):
v=2978, 2935, 1588, 1392, 1367, 1260, 1187, 1086 cmÀ1; MS (EI):
m/z (%)=229 ([M+], 1), 173 (16), 117 (39), 100 (18), 88 (31), 70 (43),
57 (100), 41 (42); elemental analysis (%) calcd for C12H23NO3
(229.3)=C 62.85, H 10.11, N 6.11, found: C 63.14, H 10.06, N 6.31;
Rf =0.32 (CH2Cl2/MeOH (1% NH4OH) 98:2).
MS (EI): m/z (%)=249 ([M+],6), 192 (27), 91 (100), 57(19); elemental
analysis (%) calcd for C15H23NO2 (249.3)=C 72.25, H 9.30, N 5.62;
found: C 71.99, H 9.18, N 5.54; Rf =0.23 (CH2Cl2/MeOH 95:5).
(3S,4S)-1-Benzyl-3,4-di-tert-butoxypyrrolidine
(ent-12)
and
(3S,4S)-1-benzyl-4-tert-butoxy-3-pyrrolidinol (ent-14): By follow-
ing the same procedure as for 12 and 14, the enantiomeric pyrroli-
dines ent-12 and ent-14 were prepared starting from the l-tartaric
acid derivative ent-13.
Compound ent-12: [a]D25 =+83 (c=0.7, CHCl3); elemental analysis
(%) calcd for C19H31NO2 (305.4)=C 74.71, H 10.23, N 4.59; found:
C 74.41, H 10.41, N 4.97. Spectroscopic properties are identical to
those of 12.
(3S,4S)-3,4-Di-tert-butoxy-3,4-dihydro-2H-pyrrole 1-oxide (ent-
4a): By following the same procedure as for 4a, the enantiomeric
nitrone ent-4a was prepared starting from ent-15.
Compound ent-14: [a]D27 =+8 (c=0.4, CHCl3); elemental analysis
(%) calcd for C15H23NO2 (249.3)=C 72.25, H 9.30, N 5.62; found:
C 72.54, H 9.57, N 5.85. Spectroscopic properties are identical to
those of 14.
25
Compound ent-4a: [a]D23 =+151 (c=0.8, CHCl3), {lit. [a]D
=
+153.7 (c=0.68, CCl4)[16]}; elemental analysis (%) calcd for
C12H23NO3 (229.3)=C 62.85, H 10.11, N 6.11; found: C 62.59,
H 10.49, N 6.02. Spectroscopic properties are identical to those of
4a.
(3R,4R)-3,4-Di-tert-butoxypyrrolidine
(15):
AcOH
(4.3 mL,
75 mmol) was added to a mixture of pyrrolidine 12 (2.3 g,
7.5 mmol) and 10% Pd/C (0.493 g, 0.46 mmol) in MeOH (48 mL) at
08C. The reaction mixture was stirred in a H2 atmosphere (1 atm)
at RT for 3 hours, then filtered through a short pad of Celite and
(1R,2R,7S,8aR)- and (1R,2R,7R,8aS)-1,2-Di-tert-butoxyoctahydro-
7-indolizinol (23 and 24): A mixture of nitrone 4a (229 mg,
1 mmol) and alkene 17 (905 mg, 4 mmol) was stirred at RT for
230
www.chempluschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 77, 224 – 233

Indolizidine Alkaloids
3 days. The reaction mixture was suspended in Et2O (10 mL),
cooled at 08C, treated with glacial AcOH (2 mL), and activated Zn
dust (981 mg, 15 mmol) and then stirred at RT for 24 hours. The
excess Zn dust was filtered off and the solution was treated with
base to give pH 9 with a 3m aqueous NaOH solution at 08C. The
two layers were separated and the aqueous one was extracted
with CH2Cl2 (3ꢁ20 mL). The combined organic layers were washed
with brine, dried over Na2SO4, filtered, and concentrated under re-
duced pressure to afford a diastereomeric mixture of indolizidines.
Purification by column chromatography on silica gel (eluent:
CH2Cl2/MeOH (1% NH4OH) first 97:3 and then 95:5) afforded indoli-
zidine 23 (170 mg, 60%) as a white solid and a mixture of 24 and
a third diastereomer (40 mg, 14%) in 6:1 ratio. Further separation
afforded a sample of pure 24. The excess dipolarophile 17 could
be recovered in the early chromatography fractions (644 mg, 95%)
and reused.
(1R,2R,7S,8aR)-Octahydroindolizine-1,2,7-triol (3): A solution of
23 (55 mg, 0.193 mmol) in TFA (0.83 mL) was stirred at RT for
24 hours. After this time the reaction mixture was concentrated
under reduced pressure. The crude material was dissolved in
MeOH and Ambersep 900-OH was added at 08C. The mixture was
slowly stirred at 08C for 10 minutes and then shaken at RT for
80 minutes on a flat shaker at 180 rpm. After this time the reaction
mixture was filtered through cotton wool and concentrated under
reduced pressure. Purification of the crude product by chromatog-
raphy on silica gel (eluent: CH2Cl2/MeOH/NH4OH 35:19:1) afforded
3 (29 mg, 87%) as a white solid.
Compound 3: m.p. 180–1828C (decomp); [a]D26 =À3 (c=0.47,
MeOH), {lit. ent-3: [a]D22 =+2.1 (c=0.36, MeOH);[6] [a]D21 =+1.8 (c=
1
0.81, MeOH)[7c]}; H NMR (D2O, 400 MHz): d=4.11 (ddd, J=7.5, 3.9,
1.8 Hz, 1H; 2-H), 3.70 (dd, J=8.5, 3.9 Hz, 1H; 1-H), 3.65 (pseudo tt,
J=11.1, 4.6 Hz, 1H; 7-H), 2.97 (ddd, J=11.2, 4.5, 2.5 Hz, 1H; 5-Ha),
2.84 (dd, J=11.2, 1.8 Hz, 1H; 3-Ha), 2.65 (dd, J=11.2, 7.5 Hz, 1H; 3-
Hb), 2.24–2.04 (m, 3H, 5-Hb +8-Ha +8a-H), 1.97–1.89 (m, 1H; 6-Ha),
1.47 (pseudo ddt, J=11.3, 4.5, 12.6 Hz, 1H; 6-Hb), 1.28 (pseudo q,
J=11.4 Hz, 1H; 8-Hb) ppm; 13C NMR (D2O, 50 MHz): d=82.8 (d; C-
1), 77.0 (d; C-2), 68.7 (d; C-7), 67.3 (d; C-8a), 59.8 (t; C-3), 50.1 (t; C-
5), 36.5 (t; C-8), 33.1 (t; C-6) ppm; HRMS (ESI): calcd for C8H16NO3
[M+H]+: 174.11247; found: 174.11242; Rf =0.5 (CH2Cl2/MeOH/
NH4OH 35:19:1).
Compound 23: m.p. 117.5–119.5; [a]D26 =À45 (c=0.6, CHCl3);
1H NMR (CDCl3, 400 MHz): d=3.83 (ddd, J=7.1, 4.0, 1.7 Hz, 1H; 2-
H), 3.67 (dd, J=8.5, 4.0 Hz, 1H; 1-H), 3.58 (pseudo tt, J=10.9,
4.6 Hz, 1H; 7-H), 2.93 (ddd, J=11.2, 4.5, 2.5 Hz, 1H; 5-Ha), 2.89 (dd,
J=10.1, 1.7 Hz, 1H; 3-Ha), 2.42 (dd, J=10.1, 7.1 Hz, 1H; 3-Hb), 2.21–
2.15 (m, 1H; 8-Ha), 1.96 (pseudo dt, J=2.6, 11.8 Hz, 1H; 5-Hb),
1.90–1.77 (m, 2H; 6-Ha +8a-H), 1.61 (pseudo ddt, J=11.2, 4.5,
12.3 Hz, 1H; 6-Hb), 1.27 (pseudo q, J=11.2 Hz, 1H; 8-Hb), 1.20 (s,
9H, CH3 ꢁ3), 1.16 (s, 9H, CH3 ꢁ3) ppm; 13C NMR (CDCl3,100 MHz):
d=83.2 (d; C-1), 77.9 (d; C-2), 73.8 (s; CMe3), 73.7 (s; CMe3), 69.7
(d; C-7), 65.4 (d; C-8a), 61.1 (t; C-3), 50.5 (t; C-5), 37.9 (t; C-8), 34.2
(t; C-6), 29.2 (q; 3C, CH3), 28.7 (q; 3C, CH3) ppm; IR (CDCl3): v=
Biological assays
Cell cultures
3689, 3608, 2978, 2257, 1602, 1391, 1367, 1216, 1190, 1067 cmÀ1
;
Human acute lymphoblastic Tcell line MOLT-3 (Zooprofilactic Insti-
tute, Brescia, Italy); human colorectal adenocarcinoma HT-29 cell
line and human breast cancer MCF-7 cell line (American type Cul-
ture Collection, RockvilleMD); human melanoma cell line M-14,
kindly provided by Prof. Sinibaldi Vallebona, University of Rome
‘Tor Vergata’, were grown, respectively, in suspension culture
(MOLT-3) and as adherent cells. Cell lines were cultured in Roswell
Park Memorial Institute (RPMI) 1640 (Invitrogen, Carlsbad, CA) sup-
plemented with 10% fetal bovine serum (FBS, Invitrogen), 2 mm
glutamine (Hyclone, Cramlington, England, UK), 50 UmLÀ1 penicillin
and 50 UmLÀ1 streptomycin (Hyclone). The neuroblastoma SH-
SY5Y (Zooprofilactic Institute, Brescia, Italy) cells were routinely
grown in Dulbecco’s modified Eagle’s medium (D-MEM) supple-
mented with 10% heat inactivated FBS (GIBCO, Gaithersburg, MD).
All the cell lines were cultured at 378C under humidified 5% CO2
atmosphere, in the presence or absence of compound 3 at the
concentrations of 1, 10, 50, 100, 250, 500 and 1000 mm for
24 hours.
MS (EI): m/z (%)=285 ([M+], 2), 228 (51), 172 (18), 128 (9), 113 (46),
100 (16), 57 (100), 41 (54); elemental analysis (%) calcd for
C16H31NO3 (285.42)=C 67.33, H 10.95, N 4.91; found: C 67.29,
H 11.21, N 4.77; Rf =0.11 (CH2Cl2/MeOH (1% NH4OH) 97:3).
Compound 24:
[a]D23 =À19
(c=0.3,
CHCl3);
1H NMR
(CDCl3,400 MHz): d=3.98 (pseudo dt, J=1.0, 6.3 Hz, 1H; 2-H), 3.73
(dd, J=5.0, 1.0 Hz, 1H; 1-H), 3.68 (pseudo tt, J=11.0, 4.7 Hz, 1H; 7-
H), 3.40 (dd, J=9.4, 6.8 Hz, 1H; 3-Ha), 3.04 (ddd, J=11.2, 4.2,
2.7 Hz, 1H; 5-Ha), 2.16–2.08 (m, 1H; 8a-H), 2.00 (pseudo dt, J=1.9,
11.8 Hz, 1H; 5-Hb), 1.92 (dd, J=9.4, 5.8 Hz, 1H; 3-Hb), 1.91–1.81 (m,
2H; 6-Ha +8-Ha), 1.63–1.51 (m, 2H; 6-Hb +8-Hb), 1.18 (s, 9H, CH3 ꢁ
3), 1.16 (s, 9H, CH3 ꢁ3) ppm; 13C NMR (CDCl3,100 MHz): d=79.9 (d;
C-1), 79.7 (d; C-2), 73.7 (s; CMe3), 73.4 (s; CMe3), 70.0 (d; C-7), 64.5
(d; C-8a), 63.0 (t; C-3), 50.6 (t; C-5), 35.0 (t; C-8), 34.4 (t; C-6), 28.7
(q; 3C, CH3), 28.6 (q; 3C, CH3) ppm; IR (CDCl3): v=2978, 2934, 1468,
1391, 1366, 1191, 1070 cmÀ1; MS (EI): m/z (%)=285 ([M+], 4), 228
(85), 212 (8), 172 (25), 113 (58), 100 (22), 70 (9), 57 (100), 41(56); el-
emental analysis (%) calcd for C16H31NO3 (285.4)=C 67.33, H 10.95,
N 4.91; found: C 67.19, H 11.13, N 4.94; Rf =0.23 (CH2Cl2/MeOH (1%
NH4OH) 95:5).
The SN38 (a metabolite of irinotecan, a topoisomerase I inhibitor)
at the concentration of 10 mm was used as the positive control in
MTS and apoptosis assays.
Evaluation of apoptosis
(1S,2S,7R,8aS)- and (1S,2S,7S,8aR)-1,2-Di-tert-butoxyoctahydro-7-
indolizinol (ent-23 and ent-24): By following the same procedure
as for 23 and 24, the enantiomeric indolizidines ent-23 and ent-24
were prepared starting from nitrone ent-4a.
Apoptosis was evaluated, after 24 hours of incubation, by flow cy-
tometry analysis of hypodiploid events following treatment of the
cells with detergent and PI staining, by using a method that distin-
guishes nuclei from apoptotic, necrotic, or viable cells, as previous-
ly described.[29] Isolated nuclei were analyzed by using a FACScan
flow cytometry (BD Biosciences, San Josꢂ, CA). Detectors and am-
plifier gains for forward and orthogonal scatter were adequately
selected to simultaneously detect nuclei from viable, apoptotic,
and necrotic cells. Events were gated on forward versus orthogonal
Compound ent-23: [a]D23 =+49 (c=0.4, CHCl3), {lit. [a]D23 =+53.0
(c=1.02, CHCl3);[7c] [a]D23 =+49.7 (c=0.45, CHCl3)[9]}; elemental
analysis (%) calcd for C16H31NO3 (285.4)=C 67.33, H 10.95, N 4.91;
found: C 67.22, H 11.20, N 5.01. Spectroscopic properties are identi-
cal to those of 23.
ChemPlusChem 2012, 77, 224 – 233
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
231

F. M. Cordero, B. Macchi et al.
scatter in such a way that degraded DNA from cell debris or from
doublets was excluded, and nuclei from viable, apoptotic, and ne-
crotic cells were assayed. Data acquisition and analysis were per-
formed by using CellQuestTM software on a minimum of 5000
events for each sample (BD Biosciences, San Josꢂ, CA).
Statistical analysis
Each experiment was performed in duplicate or triplicate. All data
were presented as meansÆSD. Data analysis was performed by
using the SPSS statistical software system (version 12.0 for Win-
dows, Chicago, IL). Comparison of means of apoptosis in response
to the different drugs was carried out by using the Bonferroni’s
post-hoc multiple comparison ANOVA test.
Western blot analysis
For Western blot analysis, a total number of 5ꢁ106 cells from all
cell lines, treated with 100 mm of compound 3 for 6 hours and
24 hours, were solubilised at 48C in lysis buffer (50 mm Tris-HCL at
pH 7.4, 1 mm EDTA, 1 mm EGTA at pH 7.4, 1% Triton-X, 150 mm
NaCl, 0.25% sodium deoxycholate, 1% NP-40 and, freshly added,
1 mm PMSF, 5 mm DTT, 1 mgmLÀ1 leupeptin, 1 mgmLÀ1 pepstatin,
2 mgmLÀ1 aprotinin, 1 mm Na3VO4, 20 mm Na3F, all from Sigma)
and centrifuged at 10,000 g for 20 minutes. An amount of protein
obtained from 5ꢁ105 cells was loaded onto a 10% SDS–polyacryl-
amide gel, subjected to electrophoresis, and transferred to nitrocel-
lulose membrane (Bio-Rad laboratories, Hercules, CA). After block-
ing the membrane in 10% non-fat dried milk and 3% BSA in TTBS
(20 mm Tris-HCL at pH 8.0, 0.9% NaCl, 0.03% Tween 20, all from
Sigma), the blots were incubated overnight at 48C with diluted pri-
mary antibody and, subsequently, washed and then incubated
with anti-mouse (Bio-Rad laboratories), anti-rabbit, or anti-goat IgG
chain specific conjugated to peroxidase. (Calbiochem, Merck Biosci-
ences, Darmstadt, Germany). Binding of antibodies was detected
by chemoluminescence staining using the ECL detection kit (Amer-
sham Bioscences). The following antibodies were used: rabbit poly-
clonal antibodies against human caspase 3 (1:3000, BD Bioscence
Pharmingen), mouse monoclonal antibodies against human cas-
pase 8 (1:4000, BioSource Int., Camarillo, CA), and human b-actin
(1: 8000 Novus Biologicals, Inc Littleton, CO).
Acknowledgements
We thank the Ministry of University and Research (MIUR Rome-
Italy) for financial support (PRIN2008BRXNTY).
Keywords: apoptosis
·
cycloaddition
·
iminosugars
·
indolizidine alkaloids · protecting groups
[1] For reviews on iminosugars bioactivity, see: a) G. Horne, F. X. Wilson, J.
Tinsley, D. H. Williams, R. Storer, Drug Discovery Today 2011, 16, 107–
118; b) P. Merino, I. Delso, E. Marca, T. Tejero, R. Matute, Curr. Chem. Biol.
2009, 3, 253–271; c) D. D’Alonzo, A. Guaragna, G. Palumbo, Curr. Med.
Chem. 2009, 16, 473–505; d) B. G. Winchester, Tetrahedron: Asymmetry
2009, 20, 645–651; e) E. Borges de Melo, A. da Silveira Gomes, I. Carval-
ho, Tetrahedron 2006, 62, 10277–10302; f) T. M. Wrodnigg, F. K. Spreng-
er, Mini-Rev. Med. Chem. 2004, 4, 437–459; g) N. Asano, Curr. Top. Med.
Chem. 2003, 3, 471–484; h) N. Asano, Glycobiology 2003, 13, 93R–
104R; i) A. A. Watson, G. W. J. Fleet, N. Asano, R. J. Molyneux, R. J. Nash,
Phytochemistry 2001, 56, 265–295; j) N. Asano, R. J. Nash, R. J. Moly-
neux, G. W. J. Fleet, Tetrahedron: Asymmetry 2000, 11, 1645–1680.
[2] P. Compain, O. R. Martin in Iminosugars: From Synthesis to Therapeutic
Applications, Wiley-VCH, Weinheim, 2007.
[3] A. Brandi, S. Cicchi, F. M. Cordero, R. Frignoli, A. Goti, S. Picasso, P. Vogel,
J. Org. Chem. 1995, 60, 6806–6812.
[4] F. Cardona, A. Goti, A. Brandi, Eur. J. Org. Chem. 2007, 1551–1565.
[5] B. Macchi, A. Minutolo, S. Grelli, F. Cardona, F. M. Cordero, A. Mastino, A.
Brandi, Glycobiology 2010, 20, 500–506.
[6] A. Goti, F. Cardona, A. Brandi, S. Picasso, P. Vogel, Tetrahedron: Asymme-
try 1996, 7, 1659–1674.
[7] a) A. Goti, F. Cardona, A. Brandi, Synlett 1996, 7, 761–763; b) A. E.
McCaig, K. P. Meldrum, R. H. Wightman, Tetrahedron 1998, 54, 9429–
9446; c) F. Cardona, A. Goti, S. Picasso, P. Vogel, A. Brandi, J. Carbohydr.
Chem. 2000, 19, 585–601.
Comparative analysis of the bands was performed by quantitative
densitometry BY using the Tina software (version 2.10, Raytest,
Straubenhardt, Germany). Specific densitometry values for the
bands were calculated by normalization to the appropriate b-actin
values.
MTS assay
[8] a) F. M. Cordero, P. Bonanno, S. Neudeck, C. Vurchio, A. Brandi, Adv.
Synth. Cat. 2009, 351, 1155–1161; b) F. M. Cordero, P. Bonanno, M.
Chioccioli, P. Gratteri, I. Robina, A. Moreno Vargas, J. Brandi, Tetrahedron
2011, 67, 9555–9564.
[9] D. Socha, M. Jurczak, M. Chmielewski, Carbohydr. Res. 2001, 336, 315–
318.
Cell metabolic activity, measured by reduction of MTS to formazan,
was evaluated by using a colorimetric commercial kit, MTS (Cell
Titer 96 Aqueous One Solution, Promega). The assay was per-
formed by seeding 1ꢁ104 of MOLT-3, HT-29, SH-SY5Y, MCF-7, and
M-14 in 100 mL in the presence or absence of compound 3 at
seven different concentrations in complete medium RPMI supple-
mented with 5% FBS for MOLT-3, HT-29, MCF-7, or M-14 and D-
MEM supplemented with 5% FBS for SH-SY5Y neuroblastoma cells.
Twenty microliters of “CellTiter 96 Aqueous One Solution Reagent”
was added directly to culture wells at the end of the culture, incu-
bated for 1–4 hours and then the absorbance was read at 490 nm.
[10] J. Revuelta, S. Cicchi, A. Goti, A. Brandi, Synlett 2007, 485–504, and ref-
erences therein.
[11] For some recent examples of 1,3-DC, see: a) A. Benz, F. Cardona, A. Goti,
Synlett 2011, 231–234; b) S. Bigotti, L. Malpezzi, M. Molteni, A. Mele, W.
Panzeri, M. Zanda, Tetrahedron Lett. 2009, 50, 2540–2542; c) F. M. Cor-
dero, M. Salvati, C. Vurchio, A. Brandi, A. de Meijere, J. Org. Chem. 2009,
74, 4225–4231; d) J. Jakowiecki, R. Loska, M. Makosza, J. Org. Chem.
2008, 73, 5436–5441.
[12] For some examples of nucleophilic additions, see: a) K. P. Kaliappan, P.
Das, Synlett 2008, 841–844; b) F. Cardona, G. Moreno, F. Guarna, P.
Vogel, C. Schuetz, P. Merino, A. Goti, J. Org. Chem. 2005, 70, 6552–
6555; c) R. Pulz, S. Cicchi, A. Brandi, H.-U. Reissig, Eur. J. Org. Chem.
2003, 1153–1156; d) P. Merino, T. Tejero, J. Revuelta, P. Romero, S.
Cicchi, V. Mannucci, A. Brandi, A. Goti, Tetrahedron: Asymmetry 2003, 14,
367–380; e) M. Lombardo, C. Trombini, Tetrahedron 2000, 56, 323–326.
[13] Reductive coupling with carbonyl compounds: S.-F. Wu, Y.-P. Ruan, X.
Zheng, P.-Q. Huang, Tetrahedron 2010, 66, 1653–1660.
Calculation of inhibitory concentration
Cumulative results from at least three different determinations
were used to calculate the drug concentration required to inhibit
mitochondrial enzyme activity by 50% (MAIC50 value), as evaluated
by the MTS assay in all cell lines. The MAIC50 values were calculated
according to the best-fit curve, y value versus log x, where y is the
value of the examined function and x is the drug concentration.
[14] Kinugasa reaction: a) S. Stecko, A. Mames, B. Furman, M. Chmielewski, J.
Org. Chem. 2008, 73, 7402–7404; b) A. Mames, S. Stecko, P. Mikolajczyk,
232
www.chempluschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 77, 224 – 233

Indolizidine Alkaloids
M. Soluch, B. Furman, M. Chmielewski, J. Org. Chem. 2010, 75, 7580–
7587.
[25] a) F. Sꢄnchez-Izquierdo, P. Blanco, F. Busquꢅ, R. Alibꢅs, P. de March, M.
Figueredo, J. Font, T. Parella, Org. Lett. 2007, 9, 1769–1772; b) C. Gella,
ð. Ferrer, R. Alibꢅs, F. Busquꢅ, P. de March, M. Figueredo, J. Font, J. Org.
Chem. 2009, 74, 6365–6367.
[26] a) A. Goti, S. Cicchi, F. M. Cordero, V. Fedi, A. Brandi, Molecules 1998, 4,
1–12; b) A. Brandi, F. Cardona, S. Cicchi, F. M. Cordero, A. Goti, Chem.
Eur. J. 2009, 15, 7808–7821.
[15] Friedel–Crafts-type reactions: J.-K. Su, Y.-M. Jia, P.-X. Rui, J. Xiang, C.-Y.
Yu, R. He, N. Han, X. He, X. Chen, J. Zhu, Synlett 2010, 1609–1616.
[16] S. Cicchi, I. Hçld, A. Brandi, J. Org. Chem. 1993, 58, 5274–5275.
[17] S. Cicchi, M. Marradi, A. Goti, A. Brandi, Tetrahedron Lett. 2001, 42,
6503–6505.
[18] S. Cicchi, M. Corsi, A. Goti, J. Org. Chem. 1999, 64, 7243–7245.
[19] a) R. Ballini, E. Marcantoni, M. Petrini, J. Org. Chem. 1992, 57, 1316–
1318; b) E. Marcantoni, M. Petrini, O. Polimanti, Tetrahedron Lett. 1995,
36, 3561–3562; c) R. W. Murray, K. Iyanar, J. Org. Chem. 1996, 61, 8099–
8102; d) C. Chevrier, A. Defoin, C. Tarnus, Bioorg. Med. Chem. 2007, 15,
4125–4135.
[20] a) G. Soldaini, F. Cardona, A. Goti, Org. Lett. 2007, 9, 473–476; b) M. For-
cato, M. Mba, W. A. Nugent, G. Licini, Eur. J. Org. Chem. 2010, 740–748.
[21] G. Uray, W. Lindner, Tetrahedron 1988, 44, 4357–4362.
[22] A. Barge, E. G. Occhiato, C. Prandi, D. Scarpi, S. Tabasso, P. Venturello,
Synlett 2010, 812–816.
[27] a) J. J. Tufariello, J. J. Tegeler, Tetrahedron Lett. 1976, 17, 4037–4040;
b) J. J. Tufariello, Acc. Chem. Res. 1979, 12, 396–403; c) F. M. Cordero, F.
Machetti, F. De Sarlo, A. Brandi, Gazz. Chim. Ital. 1997, 127, 25–29;
d) F. M. Cordero, C. Faggi, F. De Sarlo, A. Brandi, Eur. J. Org. Chem. 2000,
3595–3600.
[28] M. Padrꢆ, J. A. Castillo, L. Gꢆmez, J. Joglar, P. Clapꢅs, C. deBolꢆs, Glyco-
conjugate J. 2010, 27, 277–285.
[29] C. Matteucci, S. Grelli, E. De Smaele, C. Fontana, A. Mastino, Cytometry
1999, 35, 145-153Matteucci 1999.
ˇ
ˇˇ
´
[23] D. Rejman, P. Kocalka, M. Budesꢃnsky, R. Pohl, I. Rosenberg, Tetrahedron
2007, 63, 1243–1253.
Received: December 2, 2011
[24] R. Ballini, G. Bosica, G. Cioci, D. Fiorini, M. Petrini, Tetrahedron 2003, 59,
3603–3608.
Published online on February 6, 2012
ChemPlusChem 2012, 77, 224 – 233
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
233