Synthesis of the indolo[2,3-a]quinolizidine ring through the addition of 2-siloxyfurans to imines and intrinsic reaction coordinate calculations

Article abstract of DOI:10.1055/s-0031-1289631

A concise asymmetric diastereoselective strategy for the synthesis of indolo[2,3-a]quinolizidine derivative 1 was developed using diastereoselective addition of 2-siloxyfurans 4 to imine 3 through chiral auxiliary induction. The addition of an ionic liquid as additive in the reaction favored the anti configuration in the major adduct. The stereochemical outcome of the anti/syn (threo/erythro) selectivity was rationalized based on transition state and IRC calculations at DFT (B3LYP) and MP2 theories. MP2 calculations was shown to be the method of choice in these systems, which orbital desymmetrizations were observed in the anti transition state of the addition of 4 to 3 and secondary orbital interactions allowed us to rationalize the production of the major anti-adduct 6. Furthermore, the work also suggested that 2-(triisopropylsiloxy) furan (4a) was the nucleophile of choice in this kind of Mannich reaction. Moreover, the strategy features the use of the Mitsunobu reaction to insert an amino group with the correct configuration into amine 2, key intermediate to achieve 1. The synthetic route can also be applied in the total synthesis of promising aza-β-carboline compounds. Georg Thieme Verlag Stuttgart. New York.

Full text of DOI:10.1055/s-0031-1289631

144
PAPER
Synthesis of the Indolo[2,3-a]quinolizidine Ring through the Addition of
2-Siloxyfurans to Imines and Intrinsic Reaction Coordinate Calculations
Yaneris Mirabal-Gallardo,a Maria D. P. C. Soriano,a Julio Caballero,b Jans Alzate-Morales,b Mario J. Simirgiotis,c
Leonardo S. Santos*a
a
Laboratory of Asymmetric Synthesis, Chemistry Institute of Natural Resources, Talca University, P.O. Box 747, Talca, Chile
Fax +56(71)200448; E-mail: lssantos@utalca.cl
Centre for Bioinformatics and Molecular Simulations, School of Bioinformatics, Talca University, P.O. Box 747, Talca, Chile
Laboratory of Natural Products, Department of Chemistry, Faculty of Basic Sciences, University of Antofagasta,
b
c
P.O. Box 170, Antofagasta, Chile
Received 9 September 2011; revised 18 October 2011
derivatives,3 we decided to extend these studies to dihy-
Abstract: A concise asymmetric diastereoselective strategy for the
dro-b-carboline iminium precursor such as 3 (Scheme 1).
synthesis of indolo[2,3-a]quinolizidine derivative 1 was developed
using diastereoselective addition of 2-siloxyfurans 4 to imine 3
through chiral auxiliary induction. The addition of an ionic liquid as
additive in the reaction favored the anti configuration in the major
adduct. The stereochemical outcome of the anti/syn (threo/erythro)
N
N
H
H
selectivity was rationalized based on transition state and IRC calcu-
lations at DFT (B3LYP) and MP2 theories. MP2 calculations was
shown to be the method of choice in these systems, which orbital
desymmetrizations were observed in the anti transition state of the
addition of 4 to 3 and secondary orbital interactions allowed us to
rationalize the production of the major anti-adduct 6. Furthermore,
the work also suggested that 2-(triisopropylsiloxy)furan (4a) was
the nucleophile of choice in this kind of Mannich reaction. More-
over, the strategy features the use of the Mitsunobu reaction to insert
an amino group with the correct configuration into amine 2, key in-
termediate to achieve 1. The synthetic route can also be applied in
the total synthesis of promising aza-b-carboline compounds.
H
N
3
+
N
N
N
H
N
H
HO
OSiR3
O
O
2
1
4
Scheme 1 Retrosynthetic analysis for the construction of indo-
lo[2,3-a]quinolizidine ring system 2 obtained from the addition of 4
to 3 and its azaeburnane analogue 1
As part of our efforts in the field of biologically relevant
carbolines, we turned our attention towards an alternative
synthetic route for chiral indolo[2,3-a]quinolizidine ring
systems, via addition of 2-siloxyfuran 4 to key iminium
precursor 3. Recently, we reported enantioselective total
syntheses of hydrastine alkaloids, (–)-quinolactacin B an-
tibiotic, and PDE5 inhibitors.4 Structurally, 15 comprises
a carboline framework containing a common quinoline
core.6 The retrosynthetic analysis for the basic framework
of 1 is depicted in Scheme 1 and features the diastereose-
lective addition of 2-siloxyfuran 4 to the iminium ion de-
rived from 3 as the key step. Although demonstrated to be
a useful synthetic method, these asymmetric additions re-
main to be fully explored in the area of total syntheses of
alkaloid natural products.
Key words: azaeburnane, b-carbolines, chiral auxiliary, Mitsunobu
reaction, intrinsic reaction coordinate calculations
Chirality in b-carboline molecules plays an important role
in medicine, and activity is governed by the asymmetry in
molecules. However, even after modern developments in
synthetic organic chemistry, there are still few asymmet-
ric methodologies available to prepare these chiral com-
pounds. One methodology is the Pictet–Spengler
cyclocondensation.1 In this context, we have pursued effi-
cient synthetic routes for novel chemotypes when selec-
tivity is required. Over the last few years, 2-
(trialkylsiloxy)furans2 have been used as versatile re-
agents for the construction of several enantiomerically
pure compounds of biological interest.3 We have investi-
gated the nucleophilic addition of carbon nucleophiles to
cyclic acyliminium ions and found that the N-acylimini-
um ring size plays an important role in the stereochemical
outcome of the reaction.4 As studies involving the inter-
molecular nucleophilic addition of 2-(trialkylsil-
oxy)furans 4 to cyclic N-acyliminium ions are so far
restricted to pyrrolidine and piperidine N-acyliminium ion
rings, isoquinolines and mainly to N-benzyloxycarbonyl
We first explored the reaction of different 2-siloxyfurans
4a–d, which are readily available from furfuraldehyde,7
and iminium precursor 3. Imine 3 was obtained by reflux-
ing ethyl formate (17 h) with commercial available
tryptamine affording the corresponding formamide in
98% yield. Bischler–Napieralski cyclization of the form-
amide with phosphoryl chloride in acetonitrile at 0 °C for
four hours gave 3 in 80% yield. Having prepared imine 3,
the next stage was set for model studies involving the ad-
dition of 2-siloxyfurans 4a–d to the iminium ion interme-
diates obtained from 3 and 5a,b (Scheme 2, Table 1).
Thus, the scope of the addition of 4a to imine 3 mediated
by chiral auxiliaries 5a and 5b with and without additives
under different conditions was investigated (Table 1). The
SYNTHESIS 2012, 44, 144–150
Advanced online publication: 01.12.2011
DOI: 10.1055/s-0031-1289631; Art ID: M88411SS
x
x
.x
x
.2
0
1
1
© Georg Thieme Verlag Stuttgart · New York

PAPER
Indolo[2,3-a]quinolizidines by Addition of 2-Siloxyfurans to Imines
145
4a, R = i-Pr
Table 1 Addition of 2-Siloxyfurans 4a–d to Iminium Ions Derived
from Imine 3 and Chiral Auxiliaries 5a,b
4b, R = Me
4c, R = t-BuMe2
4d, R = Bu
OSiR3
O
O
Entry Solvent
5
4
Additive
–
Ratioa 6/7a Yieldb
+
N
(%)
N
(Table 1)
Cl
OR*
H
3
5a, R* = 8-phenylmenthyl
1
2
3
4
5
6
7
8
9
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
5a 4a
4.1:1,
6a/7a
55c
65
72
58 c
68
70
60c
65
66
65c
72
70
74
66
63
5b, R* = trans-2-phenyl-cyclohexyl
H
H
5a 4a CsF
4.8:1,
6a/7a
N
H
N
N
H
CO2R*
N
H
CO2R*
H
+
5a 4a [bmim][BF4] 5.9:1,
6a/7a
O
O
O
O
5b 4a
–
3.5:1,
6b/7b
6a,b, threo
7a,b, erythro
Scheme 2 Model study for the addition of 4 to iminium ion interme-
diate obtained from 3 and chiral auxiliaries the chloroformate of (–)-
8-phenylmenthol (5a) or trans-2-phenylcyclohexanol (5b)
5b 4a CsF
4.9:1,
6b/7b
5b 4a [bmim][BF4] 5.0:1,
6b/7b
N-acyl derivatives synthesized by quaternization of imine
3 with 5a by in situ formation of the N-acyliminium ion
and subsequent addition of nucleophile 4a in the absence
of additives afforded anti-isomer 6a in diastereomeric ra-
tios ranging from 4.1–5.4:1 (anti/syn, entries 1 and 7).
Chiral auxiliary 5b resulted in a lower diastereomeric ra-
tio (anti/syn, 6b/7b, 3.5–3.8:1, entries 4 and 10) for the
addition of 4a to 3 in the absence of additives. Further-
more, without additives, the reactions required longer re-
action times and resulted in moderated yields (55–65%)
depending on the nature of the solvents employed
(Table 1).
Different salts, such as metal fluorides (e.g., CsF,8a,b KF,8c
CuF28d), lithium alkoxides,8e and imidazolium hydrox-
ide,8f have been successfully used to catalyze nonstereo-
selective additions of trimethylsilyl cyanide to iminium
ions. As the activation of R3SiNu by a Lewis base is pri-
marily due to the affinity of silicon for fluorine or oxygen,
THF–CH2Cl2 5a 4a
–
5.4:1,
6a/7a
THF–CH2Cl2 5a 4a CsF
7.3:1,
6a:7a
THF–CH2Cl2 5a 4a [bmim][BF4] 8:1,
6a/7a
10 THF–CH2Cl2 5b 4a
–
3.8:1,
6b/7b
11 THF–CH2Cl2 5b 4a CsF
3.6:1,
6b/7b
12 THF–CH2Cl2 5b 4a [bmim][BF4] 4.6:1,
6b/7b
13 THF–CH2Cl2 5a 4b [bmim][BF4] 4.3:1,
6a/7a
facilitating the formation of a reactive pentacoordinate or 14 THF–CH2Cl2 5a 4c [bmim][BF4] 7:1,
6a/7a
hexacoordinate silicon intermediate, we employed cesium
fluoride and 1-butyl-3-methylimidazolium tetrafluorobo-
rate ([bmim][BF4]) in an attempt to improve the nucleo-
philicity and stereospecificity of 4. Thus, the role of the
addition of cesium fluoride was studied as an additive (en-
tries 2, 5, 8, and 11), and diastereoselections were ob-
tained favoring the anti isomer (anti/syn 3.6–7.3:1) in
shorter reaction times. Best results were obtained when 5a
was employed as a chiral auxiliary. Next, we tested the ad-
dition of [bmim][BF4] (entries 3, 6, 9, and 12) as an addi-
tive.9 Ionic liquid [bmim][BF4] showed selectivity
improvement when 5a and tetrahydrofuran–dichloro-
methane were employed (entry 9). Finally, to study the
influence of the R group (Me, t-BuMe2, Bu) on the diaste-
reoselectivity, 3, 5a, and [bmim][BF4] were reacted with
siloxyfurans 4b–d in tetrahydrofuran–dichloromethane.
In fact, the preference for the syn-isomer 7a increased (en-
try 13) when 4b (R = Me) was employed, compared with
4a (R = i-Pr). When 4c and 4d were used (entries 14 and
15), the preference for anti-isomer 6a was modestly lower
than 4a (entry 9). The diastereomeric ratios of 6/7 ob-
tained were determined by isolated products and con-
15 THF–CH2Cl2 5a 4d [bmim][BF4] 7.2:1,
6a/7a
a Diastereomeric ratio [dr (%)] calculated based on HPLC.
b Yields of anti/syn isomer mixtures. All reactions were performed at
–78 °C to –20 °C, 12 h.
c Reaction time ca. 48 h.
firmed by HPLC analysis. The relative stereochemistry of
compounds 6 and 7 could be deduced principally by com-
parison of their 13C NMR spectra since it was noted that
the 13C chemical shifts of C5¢ and C9 in the anti isomer
are, respectively, further downfield and higher upfield
than those of the corresponding carbons of the syn isomer.
Hydrogenation reaction of anti-isomer 6a followed by re-
moval of chiral auxiliary with sodium methoxide in meth-
anol gave the 1-hydroxylactam 8 in 80% yield and >98%
ee after spontaneous lactamization (Scheme 3). The enan-
tiomeric excess of 8 was determined by chiral HPLC
(Welk-01 column, hexanes–i-PrOH–i-Pr2NH, 90:10:0.1;
0.8 mL/min, l = 262 nm). The chiral auxiliary 8-phenyl-
© Thieme Stuttgart · New York
Synthesis 2012, 44, 144–150

146
Y. Mirabal-Gallardo et al.
PAPER
solution (AlH3)4c,10 in tetrahydrofuran afforded 2 in 96%
yield {[a]D –32.0 (c 0.7, MeOH)}.11 Next, the inversion of
configuration of the (R)-hydroxy group in 2 to give 9 was
achieved using the Mitsunobu reaction. The N-alkylation
of 1H-indole through an intermolecular reaction when us-
ing classical Mitsunobu reaction conditions, i.e. triphe-
nylphosphine and diethyl azodicarboxylate, was not a
problem.12a Furthermore, the NH moiety of indolyl com-
pounds is weakly acidic, and competitive processes
showed that indolyl N-alkylation proceeded smoothly
compared with other NH groups.
1) H2, Pd/C
EtOAc
H
N
O
anti-6a
N
H
2) NaOMe/MeOH
80%
HO
8
i. Ph3P, DEAD
phthalimide, THF
ii. NH2NH2, H2O–EtOH
AlH3
THF
H
61% (two steps)
N
N
H
or
96%
i. Ph3P, DEAD, HN3
ii. Ni2B, MeOH–H2O
MW (70 W), 11 min
79% (two steps)
HO
2
Thus, reaction of 2 with phthalimide, triphenylphosphine
and diethyl azodicarboxylate in tetrahydrofuran followed
by hydrazine treatment in water–ethanol as solvent af-
forded 9 in overall yields around 61%.12f,13 Then, testing
Mitsunobu reaction of 2 with diethyl azodicarboxylate,
triphenylphosphine, and hydrazoic acid14 gave the corre-
sponding azide, which was reduced (without further puri-
fication) with nickel boride (freshly prepared)15 to afford
9 in 79% yield under microwave-assisted irradiation (70
W) in a sealed vessel at 39 °C.16 Finally, chloroacetic acid
was converted into the respective acyl chloride using thio-
nyl chloride, and 1.01 equivalents reacted with 9 using po-
tassium carbonate in a highly dilute concentration in N,N-
O
Cl
H
H
Cl
N
N
N
N
H
K2CO3
DMF, 12 h
58%
H2N
9
N
H
O
1
Scheme 3
menthol was recovered in 85% yield and continuously
used without loss of induction. Then, reduction of the car-
bonyl group of 8 with freshly prepared aluminum hydride
OSiR3
O
+N
O
N
H
H
O
H
OSiR3
R = i-Pr, ΔHrel = 0.42 kcal/mol
R = Me, ΔHrel = 0.60 kcal/mol
R = t-BuMe2, ΔHrel = 0.54 kcal/mol
R = Bu, ΔHrel = 0.15 kcal/mol
+N
H
H
O
O
+N
O
N
N
H
H
O
H
O
O
H
R = i-Pr, ΔHrel = 0.97 kcal/mol
R = Me, ΔHrel = 1.06 kcal/mol
R = t-BuMe2, ΔHrel = 1.23 kcal/mol
R = Bu, ΔHrel = 5.43 kcal/mol
R3SiO
R = i-Pr, ΔHrel = 0.00 kcal/mol
R = Me, ΔHrel = 1.38 kcal/mol
R = t-BuMe2, ΔHrel = 1.30 kcal/mol
R = Bu, ΔHrel = 4.76 kcal/mol
B
A
C
syn-7a
anti-6a
OSiR3
O +
+N
H
O
N
D
F
O
N
N
H
O
E
H
H
O
O
H
R = i-Pr, ΔHrel = 1.97 kcal/mol
R = Me, ΔHrel = 1.02 kcal/mol
R = t-BuMe2, ΔHrel = 1.08 kcal/mol
R = Bu, ΔHrel = 2.84 kcal/mol
H
OSiR3
R = i-Pr, ΔHrel = 1.21 kcal/mol
R = Me, ΔHrel = 1.28 kcal/mol
R = t-BuMe2, ΔHrel = 2.43 kcal/mol
R = Bu, ΔHrel = 5.00 kcal/mol
+N
O
R3SiO
N
H
O
H
O
H
R = i-Pr, ΔHrel = 0.42 kcal/mol
R = Me, ΔHrel = 0.39 kcal/mol
R = t-BuMe2, ΔHrel = 0.36 kcal/mol
R = Bu, ΔHrel = 2.20 kcal/mol
Figure 1 Calculated energies (B3LYP/6-31G*) for the transition-state models for the production of syn-isomer 7a (erythro) and anti-isomer
6a (threo) adducts
Synthesis 2012, 44, 144–150
© Thieme Stuttgart · New York

PAPER
Indolo[2,3-a]quinolizidines by Addition of 2-Siloxyfurans to Imines
147
dimethylformamide as solvent (10–5 M) afforded com- reach energy convergence. The step size along the reac-
pound 1 in 58% yield.17
tion path, leading to products, was set up to 10. The results
showed that syn product, in each array, was lower in ener-
gy that anti product, indicating that syn isomers are also
favored by thermodynamic control of the reaction when
DFT theory was employed, as depicted in Figure 2.
The stereochemical outcome of the above reaction did not
come as a surprise to us as previous results with 2-siloxy-
furans led us to predict the preferential formation of the
anti isomer.4 However, in an attempt to predict the best 2-
siloxyfuran (R = i-Pr, Me, BuMe2, Bu) in this kind of re- Martin and co-workers have found a similar result for the
action, theoretical calculations of the transition-state geo- transition-state calculations (RHF/3-21G*) in the addition
metries associated with the addition of 2-siloxyfurans of 2-methoxyfuran to a five-membered N-acyl-N-(meth-
4a–d to the iminium ion derived from 3 at DFT level oxycarbonyl)iminium ion.19 Generally, it is observed that
(B3LYP/6-31G*) and MP2 were carried out. Theoretical the anti adduct is formed in the addition of 2-siloxyfurans
calculations of the transition-state geometries associated to iminium ions as the major species.2,4 However, for sev-
with the addition of 2-siloxyfuran 4a (R = i-Pr) to the imi- eral years, we have not been able to rationalize the rever-
nium ion derived from 3 at DFT level (B3LYP/6-31G*) sal of the stereochemical outcome observed when 2-
showed that arrays which lead to formation of syn product siloxyfurans are employed in the addition to iminium
(A, B, and C) are lower in energy than arrays leading to ions, the unexpected preference for the syn adduct by ab
formation of anti isomer (D, E, and F). Specifically, array initio and DFT calculation models.4a,7,19 With the aim to
A showed (relative energy: 0.0 kcal/mol), expected to be elucidate the disagreement in the experimental and theo-
the lowest energy transition state, displaying an synclinal retical data when 4a was used, we performed single point
approach of the p systems of the nucleophile and iminium energy calculations at MP2 level, using 6-31G* basis set,
ion was indeed lower in energy than array C (relative en- on last points in previous IRC curves to the lowest energy
ergy: 0.97 kcal/mol), as depicted in Figure 1.
transition state described by arrays A (R = i-Pr, syn) and
D (R = i-Pr, anti). MP2 results afforded the picture for the
stereochemical outcome in this Mannich reaction, which
the anti transition state presented energy 0.29 kcal/mol
lower than the syn transition state, as depicted in Figure 3.
Additionally, we analyzed the LUMO orbitals for both
transition states in order to interpret the gain in stabiliza-
tion energy of the anti adduct. We figure that in the anti
transition state a secondary orbital interaction of indolic
nitrogen electrons lone pair to the transient p*allyl siloxy-
furan system, during the production of the C–C bond,
should have a higher contribution than the transition state
stabilized by the pCO shown in array A, Figure 3. This re-
sult suggests that, at least, the MP2 level of calculation is
The same analysis can be performed for the anti energies
of transition states (arrays D and F). For remaining 2-sil-
oxyfurans 4b,c,d the arrays B and E were lower in energy,
both leading to syn and anti products, respectively. Tran-
sition-state calculations suggest that formation of syn in-
termediates is generally faster, which means it is
kinetically controlled. In order to check if thermodynamic
control favors the formation of the anti isomer, we per-
formed intrinsic reaction coordinate (IRC) calculations18
on each array of 2-siloxyfuran 4a to the iminium ion de-
rived from 3 (Figure 2). IRC calculations were performed
at the same level used to characterize the transition-state
structures, and they required between 30 and 40 points to
Figure 2 DFT energy profile along the optimized reaction coordinate going from transition state (TS) to products (P). The different curves
correspond to possible arrays A–F adopted by 2-siloxyfuran 4a during the attack to the iminium ion derived from 3 to form syn-7a and anti-6a
products.
© Thieme Stuttgart · New York
Synthesis 2012, 44, 144–150

148
Y. Mirabal-Gallardo et al.
PAPER
Figure 3 Proposed secondary orbital interaction stabilization for the transition state of the addition of 4a to iminium ion derivative from 3
obtained at MP2/6-31G* level of calculation. Note that calculated energies (MP2/6-31G*) for the transition-state models for the production of
anti-isomer 6a and syn-isomer 7a present now a difference of 0.29 kcal/mol favoring the anti adduct.
N-Formyltryptamine
necessary to obtain a right representation of relative ener-
gy stabilities of isomer products in this kind of Mannich
reaction. Moreover, the inclusion of exchange and corre-
lation energy as well as the desymmetrization of molecu-
lar orbitals (Figure 3), through MP2 method, allowed us
to explain the gain in energy stabilization of the anti-iso-
mer product.
Tryptamine (0.48 g, 3.0 mmol) and HCO2Et (5.0 mL) were refluxed
for 17 h. The solvent was evaporated under reduced pressure, and
resulting the crude oil was diluted with CH2Cl2 (5.0 mL) and
washed with 1 M HCl (5.0 mL), aq sat. K2CO3 soln, and brine. The
organic phase was evaporated under reduced pressure, and the re-
sulting oil dried under high vacuum for 3–4 h to give pure N-formyl-
tryptamine as a colorless oil; yield: 0.553 g (89%).
1H NMR (400 MHz, CDCl3): d = 3.01 (t, J = 8.8 Hz, 2 H), 3.67 (t,
J = 8.8 Hz, 2 H), 7.16 (t, J = 7.8 Hz, 1 H), 7.22 (t, J = 8.1 Hz, 1 H),
7.38 (d, J = 8.1 Hz, 1 H), 7.61 (d, J = 7.8 Hz, 1 H), 8.36 (s, 1 H).
In conclusion, we explored an alternative and efficient
route for the synthesis of azaeburnane systems. The em-
ployed method features the use of 2-siloxyfurans 4a–d to
imines mediated by chiral auxiliaries to introduce the
chirality in 1. The spectra data of our synthetic com-
pounds, as well as the final products, were in accordance
with those previously described.
4,9-Dihydro-3H-b-carboline (3)
N-Formyltryptamine (0.553 g, 2.94 mmol) was dissolved in MeCN
(3.50 mL), and the temperature was maintained at 0 °C. Distilled
POCl3 (0.43 mL, 4.60 mmol) in MeCN (0.70 mL) was added over
35–40 min. The mixture was stirred at 0 °C for 3 h and then 5% HCl
(20 mL) was added. The mixture was washed with Et2O (3 × 6.0
mL). To the aqueous phase at 0 °C under vigorous stirring was add-
ed aq KOH soln [3.15 g in H2O (6.0 mL)] dropwise; a white precip-
itate formed that could not be filtered off. The mixture was extracted
with CH2Cl2 (4 × 15.0 mL) and the combined organic phases were
dried (Na2SO4), filtered, and evaporated under reduced pressure to
give imine 3 as an orange solid; yield: 0.400 g (80%).
Commercial available chemical reagents were used without further
purification. HN3, DEAD, HCO2Et, POCl3, CsF, [bmim][BF4],
NH2NH2, Ni2B, SOCl2, K2CO3, tryptamine, formamide, and chloro-
acetic acid were purchased from Sigma-Aldrich. (1R,2S,5R)-(+)-5-
Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl chloroacetate was
purchased from Aldrich and hydrolyzed to (–)-8-phenylmenthol.
Ph3P was purchased from Merck. Anhyd MeCN and THF, CH2Cl2,
MeOH, and DMF were prepared by distillation under a N2 atmo-
sphere over Na/benzophenone, CaH2, Na/P2O5, and CaH2/molecu-
lar sieves, respectively, and were used for reactions. Solvents for
extraction and column chromatography were distilled prior to use.
NaN3 was handled with care by wearing safety glasses, facemask,
gloves, and reactions were performed in a fume hood. All micro-
wave reactions were performed in CEM Discover LabMate equip-
ment in a closed vessel (built-in IR sensor) with cooling system.
Infrared spectra were recorded as film on KBr cells. Melting points
were measured with an Electrothermal apparatus and are uncorrect-
ed. 1H and 13C NMR spectra were recorded on Bruker WH Avance
300, and Varian Unity 400 MHz spectrometers using TMS as the in-
ternal standard. Mass spectra were recorded on a QTof Micro
(Waters-Micromass). Column chromatography was performed
using silica gel 100–200 mesh. TLC analyses were performed with
silica gel plates using I2, KMnO4, and UV lamp for visualization.
1H NMR (400 MHz, CDCl3): d = 2.91 (t, J = 8.8 Hz, 2 H), 3.95 (t,
J = 8.8 Hz, 2 H), 7.16 (t, J = 8.0 Hz, 1 H), 7.29 (t, J = 8.0 Hz, 1 H),
7.38 (d, J = 8.0 Hz, 1 H), 7.60 (d, J = 8.0 Hz, 1 H), 8.54 (s, 1 H).
(1R,2S,5R)-5-Methyl-2-(1-methyl-1-phenylethyl)cyclohexyl
anti-1-(2,5-Dihydro-5-oxofuran-2-yl)-4,9-dihydro-3H-b-carbo-
line-2-carboxylate (6a)
To a soln of imine 3 (306 mg, 1.80 mmol) in anhyd CH2Cl2–THF
(1:1, 10.0 mL) and (–)-8-phenylmenthyl chloroformate (5a, 530
mg, 1.80 mmol) at –78 °C was added [bmim][BF4] (0.020 mL, 1.80
mmol) and the mixture was stirred for 30 min under an argon atmo-
sphere, followed by slow addition of 2-(triisopropylsiloxy)furan
(4a, 460 mg, 1.80 mmol) in CH2Cl2–THF (1:1, 1.00 mL). After 12
h, sat. aq NH4Cl (10.0 mL) was added and the layers were separated.
The aqueous phase was extracted with CH2Cl2 (2 × 50 mL), and the
combined organic phases were dried (MgSO4) and concentrated un-
der reduced pressure. The crude diastereomeric mixture (8:1) was
submitted to flash column chromatography purification (hexane–
EtOAc, 2:1) affording pure anti-isomer 6a (543 mg, 1.06 mmol,
Synthesis 2012, 44, 144–150
© Thieme Stuttgart · New York

PAPER
Indolo[2,3-a]quinolizidines by Addition of 2-Siloxyfurans to Imines
149
59%) and syn-isomer 7a (64.0 mg, 0.126 mmol) contaminated with
6a.
7.10–7.11 (m, 2 H), 7.30 (d, J = 7.9 Hz, 1 H), 7.46 (d, J = 7.9 Hz, 1
H), 8.12 (s, 1 H).
[a]D –64 (c 1.0, CHCl3).
13C NMR (75 MHz, CD3OD): d = 21.4, 28.3, 34.0, 47.2, 48.7, 53.1,
56.0, 56.9, 66.7, 111.1, 112.2, 127.1, 128.3, 145.3, 147.0.
FTIR (film): 3088, 3060, 2956, 2921, 2869, 2850, 1760, 1696,
1682, 1457, 1427 cm–1.
HRMS (ESI): m/z [M + H]+ calcd for C15H19N2O: 243.1497; found:
243.1500.
1H NMR (400 MHz, DMSO-d6, 360 K): d = 0.83–0.90 (m, 1 H),
0.90 (t, J = 7.4 Hz, 3 H), 1.10–1.42 (m, 9 H), 1.47–1.67 (m, 2 H),
1.75–1.86 (m, 2 H), 1.99–2.06 (m, 1 H), 2.45–2.50 (m, 1 H), 2.45–
2.73 (m, 4 H), [3.97–4.11/4.66–4.75 (m, m, 1 H)], [4.95–5.10/5.25–
5.28 (m, m, 1 H)], 5.61 (br d, J = 6.3 Hz, 1 H,), 6.99 (d, J = 6.3 Hz,
1 H), 6.95 (t, J = 7.5 Hz, 1 H), 6.04 (t, J = 7.4 Hz, 1 H), 7.23–7.38
(m, 7 H).
13C NMR (100 MHz, CDCl3): d = 13.6, 17.5 (2 ×), 21.8, 27.2, 28.7,
31.4, 34.7, 39.2, 49.7, 50.9, 60.4, 75.3, 84.2, 110.8, 118.0, 118.2,
119.4, 121.7, 124.3, 124.8, 125.5, 126.6, 128.0, 128.4, 134.8, 135.8,
151.7, 153.4, 155.6, 155.3, 173.3.
(1S,12bR)-1,2,3,4,6,7,12,12b-Octahydroindolo[2,3-a]quinolizin-
1-amine (9)
DEAD (0.30 mL, 1.88 mmol) was added dropwise to soln of 2
(115.7 mg, 0.48 mmol), 2.7 M HN3 in toluene (0.78 mL, 2.18
mmol), Ph3P (498 mg, 1.90 mmol), and THF (7.0 mL) at 0 °C over
15 min using a syringe pump. After an additional 30 min, hexanes
(9.0 mL) were added, and Ph3PO was filtered off. The crude mixture
was dissolved in MeOH (6.0 mL), and then Ni2B (114 mg, 1.47
mmol) and 1 M HCl (1.5 mL) were added. The reaction was sub-
jected to microwave-assisted irradiation at 70 W using CEM Dis-
cover LabMate equipment in a closed vessel with the temperature
monitored by a built-in infrared sensor for 11 min at 39 °C with
cooling. The resulting mixture was filtered, and then the MeOH was
evaporated under vacuum. Sat. NaHCO3 soln (15 mL) was added to
the mixture, which was extracted with CH2Cl2 (3 × 15 mL). The
combined organic layers were dried (Na2SO4) and rotaevaporated
under reduced pressure to afford pure 9 as a white solid; yield: 91.4
mg (79%); mp 133.0–135.0 °C.
HRMS (ESI): m/z [M + H]+ calcd for C32H37N2O4: 513.2753; found:
513.2750.
(1R,12bR)-1-Hydroxy-2,3,6,7,12,12b-hexahydroindolo[2,3-
a]quinolizin-4(1H)-one (8)
To a soln of anti-isomer 6a (1.20 mmol) in EtOAc (12.0 mL) was
added 10% Pd/C (38 mg) and the mixture was stirred under H2 (1
atm) overnight. The mixture was then filtered through Celite, and
the pad was rinsed with EtOAc–MeOH (4:1, 200 mL). The organic
layer was concentrated under reduced pressure. Then, a soln of 1.1
M NaOMe in MeOH (5.2 mL) was added to the crude at 0 °C, and
the mixture was stirred at r.t. After 2 h, 2 M HCl–MeOH soln (10.0
mL) was carefully added. The organic layer was concentrated under
reduced pressure, purified by flash chromatography (silica gel) pro-
viding 8 as a colorless oil (246 mg, 0.96 mmol, 80%). The enantio-
meric excess of 8 was determined by HPLC analysis using a Welk-
01 column [eluent: hexanes–i-PrOH–i-Pr2NH (90:10:0.1); flow
rate: 0.8 mL/min; l = 262 nm]: tR (major) = 7.5 min, tR (minor) = 9.1
min, 99:1 ratio (>98% ee).
[a]D +3.7 (c 1.0, MeOH).
FTIR (film): 3345, 3269, 2920, 2805, 2740, 1464 cm–1.
1H NMR (400 MHz, CDCl3): d = 1.22–1.27 (m, 1 H), 1.67–1.91 (m,
4 H), 2.05–2.10 (m, 1 H), 2.35 (dt, J = 11.4, 3.0 Hz, 1 H), 2.53–2.75
(m, 2 H), 2.82 (dt, J = 10.3, 4.0 Hz, 1 H), 2.89–3.14 (m, 4 H), 7.06
(t, J = 7.0 Hz, 1 H), 7.12 (t, J = 7.0 Hz, 1 H), 7.32 (d, J = 7.7 Hz, 1
H), 7.46 (d, J = 7.7 Hz, 1 H), 10.54 (s, 1 H).
13C NMR (100 MHz, CDCl3): d = 22.2, 25.1, 39.2, 53.2, 54.2, 55.8,
66.0, 107.8, 111.5, 118.3, 119.0, 121.1, 127.4, 135.9, 136.0.
HRMS (ESI): m/z [M + H]+ calcd for C15H20N3: 242.1657; found:
242.1661.
[a]D –102.0 (c 1.0, MeOH).
FTIR (film): 3440, 2910, 1622 cm–1.
trans-(+)-14-Oxo[1,12]iminoethano-1,2,3,4,6,7,12,12b-octahy-
droindolo[2,3-a]quinolizine (1)
1H NMR (300 MHz, CDCl3): d = 1.74–1.77 (m, 1 H), 1.88–1.91 (m,
1 H), 2.42–2.47 (m, 1 H), 2.59–2.62 (m, 1 H), 2.68–3.00 (m, 4 H),
3.34 (br s, 1 H, OH), 4.11 (d, J = 10.0 Hz, 1 H), 4.75 (m, 1 H), 7.10
(dt, J = 7.2, 1.2 Hz, 1 H), 7.15 (dt, J = 7.2, 1.2 Hz, 1 H), 7.29 (d,
J = 7.9 Hz, 1 H), 7.47 (d, J = 7.9 Hz, 1 H), 9.65 (s, 1 H).
13C NMR (75 MHz, CDCl3): d = 21.0, 31.5, 32.8, 40.2, 54.1, 66.6,
109.0, 111.0, 118.5, 119.7, 122.1, 126.8, 133.4, 136.0, 169.0.
To a solution of 9 (50.0 mg, 0.207 mmol) in DMF (3.0 mL) under
vigorous stirring was added chloroacetyl chloride (0.016 mL, 0.209
mmol) at r.t. followed by K2CO3 (0.057 mg). After 12 h of reaction,
H2O (10.0 mL) was added, and intensive stirring was continued for
an additional 10 min. The mixture was kept at 5 °C, and after 6 h a
colorless precipitate was obtained, collected by filtration, and
washed twice with H2O (2 × 0.1 mL). Drying under high vacuum
afforded a solid, which was dissolved in hot MeOH (6.0 mL) and
fumaric acid (11.4 mg, 0.10 mmol) was added. After 20 min, color-
less crystals of pure 1 were obtained and filtered off; yield: 33.7 mg
(58%); mp 255–258 °C.
HRMS (ESI): m/z [M + H]+ calcd for C15H17N2O2: 257.1290; found:
257.1294.
(1R,12bR)-1,2,3,4,6,7,12,12b-Octahydroindolo[2,3-a]quinoliz-
in-1-ol (2)
To a soln of 8 (65.5 mg, 0.256 mmol) in anhyd THF (4.3 mL) was
added 1.55 M AlH3 in THF (0.330 mL, 0.512 mmol) at r.t. After 10
min, the reaction was quenched with sat. aq Na2SO4 soln (1.0 mL)
and filtered. The solids were washed with CH2Cl2 (50 mL), and the
filtrate was dried (Na2SO4), evaporated, and concentrated under
vacuum. Purification of the residue by column chromatography
(10% CHCl3–MeOH) afforded a colorless oil (59.4 mg, 0.2458
mmol, 96%).
[a]D +8.1 (c 1, pyridine).
FTIR (film): 1695 cm–1.
1H NMR (400 MHz, pyridine-d5): d = 1.61–1.82 (m, 3 H), 2.09–
2.25 (m, 2 H), 2.49–2.52 (m, 1 H), 2.67 (dd, J = 14.2, 5.3 Hz, 1 H),
2.85 (dd, J = 12.3, 3.2 Hz, 1 H), 2.94–3.12 (m, 2 H), 3.13 (d, J =
10.7 Hz, 1 H), 3.78 (ddd, J = 10.7, 10.0, 5.0 Hz, 1 H), 4.74 (d, J =
2.2 Hz, 1 H), 5.14 (d, J = 2.2 Hz, 1 H), 7.24 (dt, J = 8.0, 2.0 Hz, 1
H), 7.32 (dt, J = 8.0, 2.0 Hz, 1 H), 7.50 (dd, J = 8.0, 2.0 Hz, 1 H),
7.58 (dd, J = 8.0, 2.0 Hz, 1 H).
13C NMR (100 MHz, pyridine-d5): d = 22.1, 24.5, 31.2, 50.1, 51.2,
54.2, 54.8, 66.7, 107.3, 110.5, 118.8, 120.2, 122.2, 128.5, 134.9,
135.8, 169.9.
[a]D –32.0 (c 0.7, MeOH).
FTIR (film): 3505, 2070, 1425 cm–1.
1H NMR (300 MHz, CD3OD): d = 1.54–1.72 (m, 3 H), 1.92–2.00
(m, 2 H), 2.46 (br t, J = 7.9, 1.1 Hz, 1 H), 2.67–2.73 (m, 2 H), 2.90–
3.00 (m, 2 H), 3.04 (br s, 1 H, OH), 3.55 (br s, 1 H), 4.19 (br s, 1 H),
© Thieme Stuttgart · New York
Synthesis 2012, 44, 144–150

150
Y. Mirabal-Gallardo et al.
PAPER
HRMS (ESI): m/z [M + H]+ calcd for C17H20N3O: 282.1606; found:
282.1610.
Furtmuller, R.; Jursky, F.; Sigel, E.; Sieghart, W.; Dodd, R.
H. Bioorg. Med. Chem. Lett. 2000, 10, 2579. (c) Kim, S. S.;
Lee, J. T.; Lee, S. H. Bull. Korean Chem. Soc. 2005, 26,
993. (d) Onaka, M.; Higuchi, K.; Sugita, K.; Izumi, Y.
Chem. Lett. 1989, 1393. (e) Wilkinson, H. S.; Grover, P. T.;
Vandenbossche, C. P.; Bakale, R. P.; Bhongale, N. N.; Wald,
S. A.; Senanayake, C. H. Org. Lett. 2001, 3, 553.
(f) Yamaguchi, K.; Imago, T.; Ogasawara, Y.; Kasai, J.;
Kotani, M.; Mizuno, N. Adv. Synth. Catal. 2006, 348, 1516.
(9) (a) Duran-Lara, E. F.; Shankaraiah, N.; Geraldo, D.; Santos,
L. S. J. Braz. Chem. Soc. 2009, 20, 813. (b) Gozzo, F. C.;
Santos, L. S.; Augusti, R.; Consorti, C. S.; Dupont, J.;
Eberlin, M. N. Chem. Eur. J. 2004, 10, 6187.
Acknowledgment
L.S.S. thanks FONDECYT (Project 1110022). Y.M.G. thanks
Talca University for financial support. M.D.P.C.S. thanks PBCT
(PSD-50) for financial support. J.A.M. and J.C. thank Centro de
Bioinformática y Simulación Molecular (CBSM) from the Univer-
sity of Talca for computer resources provided for this work.
References
(10) Yoon, N. M.; Brown, H. C. J. Am. Chem. Soc. 1968, 90,
2927.
(1) For asymmetric Pictet–Spengler reactions, see: (a)Raheem,
I. T.; Thiara, P. S.; Peterson, E. A.; Jacobsen, E. N. J. Am.
Chem. Soc. 2007, 129, 13404. (b) Cox, E. D.; Cook, J. M.
Chem. Rev. 1995, 95, 1797. (c) Chrzanowska, M.;
Rozwadowska, M. D. Chem. Rev. 2004, 104, 3341.
(d) Allin, S. M.; Gaskell, S. N.; Elsegood, M. R. J.; Martin,
W. P. Tetrahedron Lett. 2007, 48, 5669. (e) Wanner, M. J.;
van der Haas, R. N. S.; de Cuba, K. R.; van Maarseveen, J.
H.; Hiemstra, H. Angew. Chem. Int. Ed. 2007, 46, 7485.
(2) For reviews on the use of siloxyfurans, see: (a) Casiraghi,
G.; Rassu, G. Synthesis 1995, 607. (b) Rassu, G.; Zanardi,
F.; Battistini, L.; Casiraghi, G. Chem. Soc. Rev. 2000, 29,
109.
(3) (a) Hermange, P.; Dau, M. E. T. H.; Retailleau, P.; Dodd, R.
H. Org. Lett. 2009, 11, 4044. (b) Martin, S. F.; Corbett, J.
W. Synthesis 1992, 55. (c) Morimoto, Y.; Nishida, K.;
Hayashi, Y.; Shirahama, H. Tetrahedron Lett. 1993, 34,
5773. (d) Pichon, M.; Figadere, B.; Cave, A. Tetrahedron
Lett. 1996, 37, 7963. (e) Morimoto, Y.; Iwahashi, M. Synlett
1995, 1221. (f) Zanardi, F.; Battistini, L.; Rassu, G.; Pinna,
L.; Mor, M.; Culeddu, N.; Casiraghi, G. J. Org. Chem. 1998,
63, 1368. (g) Martin, S. F.; Barr, K. J.; Smith, D. W.; Bur, S.
K. J. Am. Chem. Soc. 1999, 121, 6990. (h) D’Oca, M. G.
M.; Pilli, R. A.; Vencato, I. Tetrahedron Lett. 2000, 41,
9709.
(11) 1H NMR for compound 2 is in accordance with: Wasserman,
H. H.; Kuo, G. H. Tetrahedron 1992, 48, 7071.
(12) (a) Bombrun, A.; Casi, G. Tetrahedron Lett. 2002, 43, 2187.
(b) Frohner, W.; Monse, B.; Braxmeier, T. M.; Casiraghi, L.;
Sahagun, H.; Seneci, P. Org. Lett. 2005, 7, 4573.
(c) Tholander, J.; Bergman, J. Heterocycles 1999, 51, 1275.
(d) Iyer, M. S.; Palomo, M.; Schilling, K. M.; Xie, Y.;
Formanski, L.; Zembower, D. E. J. Chromatogr., A 2002,
944, 263. (e) Morita, N.; Krause, N. Eur. J. Org. Chem.
2006, 4634. (f) Roush, W. R.; Straub, J. A.; Brown, R. J.
J. Org. Chem. 1987, 52, 5127.
(13) 1H and 13C NMR are in accordance with that described in the
literature: (a) Melnyk, P.; Ducrot, P.; Demuynck, L.; Thal,
C. Tetrahedron Lett. 1993, 34, 5085. (b) Lee, Y. S.; Cho, D.
J.; Kim, S. N.; Choi, J. H.; Park, H. J. Org. Chem. 1999, 64,
9727.
(14) Overman reported the Mitsunobu reaction with hydrazoic
acid to install an azide group in similar systems. See: Cohen,
F.; Overman, L. E. J. Am. Chem. Soc. 2006, 128, 2594.
(15) (a) Schlesinger, H. I.; Brown, H. C.; Finholt, A. E.;
Gilbreath, J. R.; Hoekstra, H. R.; Hyde, E. K. J. Am. Chem.
Soc. 1953, 75, 215. (b) Brown, C. A. J. Org. Chem. 1970,
35, 1900. (c) Hoffer, L. J. E.; Schultz, J.; Panson, R. D.;
Anderson, R. B. Inorg. Chem. 1964, 3, 1783. (d) Seltzman,
H. H.; Berrang, B. D. Tetrahedron Lett. 1993, 34, 3083.
(e) Santos, L. S.; Pilli, R. A. Synthesis 2002, 87.
(16) Shankaraiah, N.; Markandeya, N.; Espinoza-Moraga, M.;
Arancibia, C.; Kamal, A.; Santos, L. S. Synthesis 2009,
2163.
(4) (a) Soriano, M. D. P. C.; Shankaraiah, N.; Santos, L. S.
Tetrahedron Lett. 2010, 51, 1770. (b) de Oliveira, M. C. F.;
Santos, L. S.; Pilli, R. A. Tetrahedron Lett. 2001, 42, 6995.
(c) Santos, L. S.; Pilli, R. A. Tetrahedron Lett. 2001, 42,
6999. (d) Shankaraiah, N.; Pilli, R. A.; Santos, L. S.
Tetrahedron Lett. 2008, 49, 5098.
(17) Schmitt, P.; Melnyk, P.; Bourde, O.; Demuynck, L.; Pujol,
(5) Ducrot, P.; Rabhi, C.; Thal, C. Tetrahedron 2000, 56, 2683.
(6) (a) Lopez, I.; Diez, A.; Rubiralta, M. Tetrahedron 1996, 52,
8581. (b) Melnyk, P.; Legrand, B.; Gasche, J.; Ducrot, P.;
Thal, C. Tetrahedron 1995, 51, 1941.
(7) Santos, L. S.; Pilli, R. A. J. Braz. Chem. Soc. 2003, 14, 982.
(8) (a) Kim, S. S.; Rajagopal, G.; Song, D. H. J. Organomet.
Chem. 2004, 689, 1734. (b) Razet, R.; Thomet, U.;
J.-F.; Thal, C. Med. Chem. Res. 1993, 3, 24.
(18) (a) Fukui, K. Acc. Chem. Res. 1981, 14, 363. (b) Gonzalez,
C.; Schlegel, H. B. J. Chem. Phys. 1989, 90, 2154.
(c) Gonzalez, C.; Schlegel, H. B. J. Chem. Phys. 1990, 94,
5523.
(19) Bur, S. K.; Martin, S. F. Org. Lett. 2000, 2, 3445.
Synthesis 2012, 44, 144–150
© Thieme Stuttgart · New York