Dehydrative Annulation Strategy for the Construction of Octahydroindolizine Framework: A Diastereoselective Synthesis of (6 R,8a S)-Octahydroindolizin-6-ol

Article abstract of DOI:10.1080/00397911.2015.1093146

A dehydrative annulation strategy involving an intramolecular ring closure under a Mitsunobu-Type reaction condition has been used for the construction of octahydroindolizine framework successfully. This strategy that was reported to be unsuccessful when

Full text of DOI:10.1080/00397911.2015.1093146

Synthetic Communications
An International Journal for Rapid Communication of Synthetic Organic
Chemistry
ISSN: 0039-7911 (Print) 1532-2432 (Online) Journal homepage: http://www.tandfonline.com/loi/lsyc20
Dehydrative Annulation Strategy for the
Construction of Octahydroindolizine Framework:
A Diastereoselective Synthesis of (6R, 8aS)-
Octahydroindolizin-6-ol
Subbarao Jammula, Venkateswara Rao Anna, Sandip R. Khobare, U. K. Syam
Kumar, Bagineni Prasad & Manojit Pal
To cite this article: Subbarao Jammula, Venkateswara Rao Anna, Sandip R. Khobare, U. K.
Syam Kumar, Bagineni Prasad & Manojit Pal (2015): Dehydrative Annulation Strategy for the
Construction of Octahydroindolizine Framework: A Diastereoselective Synthesis of (6R, 8aS)-
Octahydroindolizin-6-ol, Synthetic Communications, DOI: 10.1080/00397911.2015.1093146
To link to this article: http://dx.doi.org/10.1080/00397911.2015.1093146
Accepted author version posted online: 06
Oct 2015.
View supplementary material
Submit your article to this journal
View related articles
Article views: 17
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=lsyc20
Download by: [University of Cambridge]
Date: 27 October 2015, At: 06:52

Dehydrative Annulation Strategy for the Construction of Octahydroindolizine
Framework: A Diastereoselective Synthesis of (6R, 8aS)-Octahydroindolizin-6-ol
Subbarao Jammula1,2, Venkateswara Rao Anna2, Sandip R. Khobare1, U. K. Syam
Kumar1, Bagineni Prasad3, Manojit Pal3
1Custom Pharmaceutical Services, Dr. Reddy’s Laboratories Limited, Hyderabad, India,
2Department of Chemistry, Koneru Lakshmaiah University (KLU), Guntur, Andhra
Pradesh, India, 3Dr. Reddy’s Institute of Life Sciences, University of Hyderabad Campus,
Hyderabad, India
Address correspondence to Manojit Pal, Dr. Reddy’s Institute of Life Sciences,
University of Hyderabad Campus, Gachibowli, Hyderabad 500 046, Andhra Pradesh,
India. Email: manojitpal@rediffmail.com
Abstract
A dehydrative annulation strategy involving an intramolecular ring closure under a
Mitsunobu type reaction condition has been used for the construction of
octahydroindolizine framework successfully. This strategy that was reported to be
unsuccessful when applied to a similar other system allowed us to perform a
diastereoselective synthesis of (6R, 8aS)-octahydroindolizin-6-ol [a precursor of (-)-8a-
epidesacetoxyslaframine] starting from commercially available chiral (S)-epichlorohydrin
via a piperidine intermediate i.e. (3R, 6S)-6-(3-hydroxypropyl)piperidin-3-ol. The
methodology has potential to afford a library of optically pure small molecules of
pharmacological importance based on the related indolizine framework.
1

Graphical Abstract:
KEYWORDS: octahydroindolizine, (S)-epichlorohydrin, Mitsunobu reaction,
diastereoselectivity
INTRODUCTION
The development of synthetic tools to construct optically pure heterocycles of
biological importance is of great interest in both organic and medicinal chemistry as
this effort provides quick access to the relevant natural and unnatural analogs for
various studies including pharmacological activities.
The octahydroindolizine (or 1-azabicyclo[4.3.0]nonane) framework is widespread in
many alkaloids[1],[2],[3] such as coniceine (isolated from the skin secretions of
neotropical amphibians), slaframine (a mycotoxin that generally causes salivation in
most animals and is usually produced by the fungus Rhizoctonia leguminicola),
castanospermine (a potent glycosidase inhibitor) and a number of poisonous-frog
alkaloids typified by pumiliotoxin B1. Thus construction of octahydroindolizine ring
in a stereoselective fashion has attracted enormous attention.[4-11] Being a precursor of
(-)-8a-epidesacetoxyslaframine, the (6-R)-octahydroindolizin-6-ol (A, Fig.1) attracted
our particular attention as this compound could provide us access to a library of
2

indolizine based small molecules in addition to (-)-8a-epidesacetoxyslaframine. This
and our interest in piperidine
derivatives[12] prompted us to report a new approach to the octahydroindolizine
skeleton of compound A starting from enantiomerically pure (S)-epichlorohydrin
using dehydrative annulation strategy as a key step. Notably, previous synthesis of
(±)-A involved intramolecular nitrone cycloaddition strategy, using 2-
pyrrolecarbaldehyde as starting material followed by hydrogenation.[11] However,
formation of a mixture of isomeric cycloadducts during the intramolecular nitrone
cycloaddition step could potentially be a major concern for this strategy to be used as
a practical route to A. Moreover, this methodology do not provide an optically pure
A. Additionally, the dehydrative annulation strategy (the Mitsunobu-like protocol)
was not successful when applied to a similar system earlier by Casiraghi et al.[10]
RESULTS AND DISCUSSIONS
A retro-synthetic analysis of A (Fig. 1) suggested that this (6R)-octahydroindolizin-6-
ol could be obtained from a chiral precursor i.e. (3R,6S)-6-(3-
hydroxypropyl)piperidin-3-ol (B) via intra-molecular displacement of the activated
primary hydoxy group by the piperidine nitrogen. The alcohol B could be obtained
from the chiral γ-hydroxy-ketone (C) by de-benzylation followed by a stereo-
controlled reductive amination process, a strategy similar to that used earlier.[13]
Based on this retro-synthetic analysis we initiated the actual synthesis of compound
A. Initially, the synthesis of its precursor i.e. the piperidine intermediate (B) was
3

undertaken (Scheme 1). Since the synthesis of amide 6 has been reported earlier[13]
hence we followed the same method for its preparation. Thus, the reaction of (S)-
epichlorohydrin and dibenzylamine (1) afforded (S)-N,N-dibenzyl-1-(oxiran-2-
yl)methanamine (3) (via 2) which on treatment with the in situ generated sodium salt
of diethyl malonate afforded the lactone 4 via opening of the oxirane ring of 3.
Subsequent Krapcho[14] type decarboxylation of 4 in the same pot afforded (R)-5-
[(dibenzylamino)-methyl]-dihydrofuran-2(3H)-one (5) with specific optical rotation
(SOR) +11.85°. The Bodroux type reaction of 5 with N,O-dimethylhydroxylamine
afforded Weinreb amide (6).[15] The amide 6 on reaction with excess of [(3-
chloropropoxy)methyl]benzene in the presence of magnesium produced a γ-
hydroxyketone, the hydroxyl group of which was further protected using TBDMS
chloride to give the corresponding compound 7 in 50% yield. Next we performed a
one-pot three step cascade reaction of 7 involving sequential (i) debenzylation, (ii)
intramolecular cyclization and (iii) subsequent reduction of the imine bond formed in
situ in the
same pot. This step was carried out using Pd/C catalyst under hydrogen atmosphere.
The product obtained was treated with 2N HCl to remove the TBDMS group and to
afford the HCl salt of a diastereomeric mixture (i.e. B-1.HCl and B-2.HCl, Scheme 1)
of the piperidine derivative (B) via a 2,3,4,5-tetrahydropyridine intermediate (8)
generated in situ. The HCl salts were washed with DCM to remove the minor
impurities present. The crystalline (3S,6R)-6-(3-hydroxypropyl) piperidin-3-ol
hydrochloride salt on treatment with base regenerated pure B. Notably, the
4

diastereomeric mixture of B was found to contain (3R,6S)-6-(3-
hydroxypropyl)piperidin-3-ol (B-1) as a major isomer and (3R,6R)-6-(3-
hydroxypropyl)piperidin-3-ol (B-2) as a minor isomer in a ratio of B-1 / B-2 ~ 10:1
(yield of pure B-1 and B-2 was 30.9% and 3.37% respectively with respect to 7.
While the HCl salt of B-1/B-2 mixture was isolated as a crystalline solid in ~ 70%
yield, the loss of material occurred during separation and purification of B-1 and B-
2). The observed diastereoselectivity during the one-pot conversion of 7 to B can be
explained by using a transition state involving the intermediate 8 generated in situ as
shown in Fig 2 and 3. While the hydrogentation of the imine bond formed in situ may
takes place in two different pathways
depending on approach of hydrogen molecule towards the –C=N- of 8, the steric bulk
of the hydroxypropyl chain seemed to have played a key role in the observed
diastereoselectivity. Thus the approach of hydrogen molecule towards the –C=N-
from the side opposite to the bulky hydroxypropyl chain is expected to be more
favorable (Fig. 2) over that involved the approach from the same side of the
hydroxypropyl chain (Fig. 3). The former pathway affords the cis isomer whereas the
other one affords the trans isomer. It is therefore not surprising that the cis isomer i.e.
B-1 was obtained as a major product in this step. Nevertheless, both the diasteromers
B-1 and B-2 were separated via a semi preparative HPLC method by using
phenomenex (Lux amylose 21.2* 250 mm) 10u column {mobile phase I: II = 80: 20
where I is n-hexane (0.1%, DEA) and II is IPA (0.1%, DEA) with a flow rate of 4
mL/min}. The 1HNMR spectra obtained for the cis isomer B-1 showed a signal at δ
5

4.23 (s, 1H) due to the tertiary proton of –CH(OH) group. It is worthy to mention that
the same proton of a structurally very similar compound reported in the
literature[13],[16,17] appeared at δ 4.22 (s, 1H)[13] or 4.08 (s,1H)[16] for the cis isomer and
at δ 3.89 (m, 1H)[13] or 3.83 (m, 1 H)[17] for the trans isomer. Indeed, the appearance
of this proton at δ 3.89 (s, 1H) in case of B-2 clearly suggests that B-1 is a cis isomer.
Having prepared (3R, 6S)-6-(3-hydroxypropyl)piperidin-3-ol (B-1) as a precursor of
A we then attempted to construct the final bicyclic skeleton of A. A literature search
revealed that the construction of a similar bicyclic skeleton was performed earlier
using a tetrahydroxypiperidine derivative.[10] While, the cyclization i.e. intramolecular
displacement of the activated primary -OH group by the piperidine nitrogen was
successful when carried out in the presence of PPh3 / CCl4 / Et3N, the use of a
Mitsunobu-like protocol in the presence of PPh3 / DEAD (diethyl azodicarboxylate)
was reported to be unsuccessful. Although the exact reason for this observation was
not clear the interference caused by other hydroxyl groups present on the side chain
could be a possible reason. Based on this we anticipated that the Mitsunobu-like
protocol might work in our case as B-1 contains less number of hydroxyl groups
compared to that used earlier.[10] To our satisfaction, when B-1 was treated with PPh3
and DEAD,[18,19] the reaction proceeded well via an intramolecular dehydrative
annulation process to give the target product i.e. (6R, 8aS)-octahydroindolizin-6-ol
(A) in 72% yield (Scheme 2).
6

25
The compound A was isolated as a viscous liquid that showed optical rotation [α]D
=
-27.94° (c 1.0, CHCl3). This compound was characterized by NMR, IR and HRMS
data and was compared with that reported earlier for (±)-A2h (see Table 1 in the
supplementary content). It should be noted that the NMR data of optically pure A has
not been reported in the literature. Moreover, the reported data of (±)-A was recorded
using CDCl3 as a solvent and a 300 MHz instrument whereas D2O as well as DMSO-
d6 along with 400 MHz instrument was used in the present case. However, in spite of
1
differences observed between the HNMR data generated by us and that reported
earlier, the 13C NMR data of optically pure A and (±)-A showed good correlations.
Since the two stereocenters of B-1 did not participate in the reaction leading to A
hence their configuration was expected to remain same in the product too. Indeed, the
position and nature of 1HNMR signal of tertiary proton of –CH(OH) group that
appeared at δ 4.23 (s, 1H) in case of B-1 was not changed significantly in case of
product A and appeared at δ 4.03 (s, br, 1H) or 4.06 (s, 1H) when recorded in D2O or
DMSO-d6, respectively (see Table 1 in the supplementary content).
CONCLUSIONS
In conclusion, a dehydrative annulation strategy was used for the construction of
octahydroindolizine framework successfully. This strategy involved an intramolecular
ring closure under a Mitsunobu type reaction conditions that was reported to be
unsuccessful when applied to a similar system earlier. The strategy allowed us to
perform a diastereoselective synthesis of (6R, 8aS)-octahydroindolizin-6-ol starting
from commercially available chiral (S)-epichlorohydrin that afforded the piperidine
7

precursor i.e. (3R,6S)-6-(3-hydroxypropyl)piperidin-3-ol as a major product after
several steps. The last step being a sequential combination of debenzylation,
intramolecular cyclization, and subsequent reduction of the imine bond formed in situ
was performed in a single pot. The diastereoselectivity of the imine bond reduction
appeared to be governed by the steric bulk of the hydroxypropyl chain. Overall, the
present synthesis of (6R,8aS)-octahydroindolizin-6-ol could provide a useful
alternative to the previously reported method involving intramolecular nitrone
cycloaddition strategy. The present method could also provide access to the indolizine
based natural and unnatural products and may find wide usage in both organic and
medicinal chemistry.
EXPERIMENTAL
Preparation Of (6R, 8as)-Octahydroindolizin-6-Ol (A)
To a solution of (3R, 6S)-6-(3-hydroxypropyl) piperidin-3-ol (B-1, 150 mg, 0.0009
mmol) in dry THF (4 mL) was added triphenyl phosphine (472 mg, 0.0018 mmol, 2
eq) and diethyl azodicarboxylate (313 mg, 0.0018 mmol, 2 eq) under argon at 0-5 °C.
The reaction mass was aged for 4 h at 0-5°C and quenched with (5 mL) 20 wt %
aqueous NaHCO3. The reaction mixture was diluted with DCM (15 mL), washed with
DM water (2x5 mL) followed by brine solution and dried over anhydrous Na2SO4,
and concentrated under reduced pressure. The residue was purified by column
chromatography using (10:2) DCM/MeOH to afford 96 mg of title compound (A) as a
viscous liquid; [α]D25 = -27.94° (c 1.0, CHCl3); Yield: 72%; IR (Neat): 3019.3,
1
1217.6, 1020.1, 928.6, 775.3, 669.3, 627.0; H NMR (400 MHz, D2O) δ: 4.03 (s, br,
8

1H), 3.63-3.50 (m, 2H), 3.40-3.18 (m, 2H), 2.85 (s, br, 1H), 2.23-2.06 (m, 5H), 1.74-
1.53 (m, 3H);1H NMR (400 MHz, DMSO-d6) δ: 4.06 (s, 1H), 3.39-3.42 (m, 3H), 2.96-
3.14 (m, 3H), 1.58-2.18 (m, 8H); 13C NMR (100 MHz, DMSO) δ: 65.9, 62.0, 56.4,
52.4, 26.9, 26.4, 22.9, 20.0; HRMS Calcd m/z for C8H14NO+H: 142.1232 found:
142.1236.
ACKNOWLEDGEMENTS
The authors thank Dr. Vilas Dahanukar, Dr. H. Rammohan of Dr. Reddy’s
Laboratories Limited for useful discussions and the Analytical Department of Dr.
Reddy’s Laboratories Limited, for spectra.
SUPPLEMENTAL MATERIAL
Full experimental detail, spectral data, 1H and 13C NMR spectra, and HPLC traces for this
article can be accessed on the publisher’s website.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
Michael, J. P. Nat. Prod. Rep. 2005, 22, 603-626.
Michael, J. P. Nat. Prod. Rep. 1999, 16, 675-696.
Franklin, A. S.; Overman, L. E. Chem. Rev. 1996, 96, 505-522.
Groaning, M. D.; Meyers, A. I. Chem. Commun. 2000, 1027-1028.
Arai, Y.; Kontani, T.; Koizumi, T., Chem. Lett., 1991, 2135-2138.
Nukui, S.; Sodeoka, M.; Shibasaki, M., Tetrahedron Lett. 1993, 34, 4965-4968.
Davies, S. B.; McKervey, M. A. Tetrahedron Lett., 1999, 40, 1229-1232.
9

[8]
[9]
Sanchez-Sancho, F.; Herradon, B.; Tetrahedron Asymm. 1998, 9, 1951-1965.
Takahata, H.; Kubota, M.; Takahashi, S.; Momose, T. Tetrahedron
Asymm. 1996, 7, 3047-3054.
[10] Casiraghi, G.; Rassu, G.; Spanu, P.; Pinna, L.; Ulgheri; F. J. Org. Chem. 1993, 58,
3397-3400.
[11] Arnone, A.; Broggini, G.; Passarella, D.; Terraneo, A.; Zecchi, G. J. Org. Chem.
1998, 63, 9279-9284.
[12] Kodimuthali, A.; Prasunamba, P. L.; Pal, M. Beilstein J. Org. Chem. 2010, 6,
No. 71, doi:10.3762/bjoc.6.71
[13] Khobare, S. R.; Gajare, V. S.; Jammula, S.; Kumar, U. K. S.; Murthy, Y. L. N.
Tetrahedron Lett. 2013, 54, 2909-2913.
[14] Hiromi, U.; Ryo, K.; Yuuki, A.; Miki, H.; Yu, K. Org. Lett. 2011, 23, 6268-6271.
[15] Michael, W.; Ronald, B.; Nobuyoshi, Y.; George, M.; Ulf-H, D.; Edward, G.
Tetrahedron Lett. 1995, 36, 5461-5466.
[16] For cis isomer, see: Bates, R. W.; Sivarajan, K.; Straub, B. F. J. Org. Chem. 2011,
76, 6844-6848.
[17] For trans isomer, see: Lovick, H. M.; Michael, F. E. J. Am. Chem. Soc. 2010, 132,
1249-1251.
[18] Mitsunobu, O. Synthesis, 1981, 1-28.
[19] Mitsunobu, O.; Yamada, M. Bull. Chem. SOC. Jpn. 1967, 40, 2380-2382.
10

Figure 1. Retro-synthetic analysis for the construction of octahydroindolizine framework
of A
11

Figure 2. Approach of hydrogen molecule towards the –C=N- from the side opposite to
the bulky hydroxypropyl chain.
12

Figure 3. Approach of hydrogen molecule towards the –C=N- from the same side of
the bulky hydroxypropyl chain.
13

Scheme 1. Synthesis of piperidine intermediate (B)
14

Scheme 2. Synthesis of (6R, 8aS)-octahydroindolizin-6-ol (A) via the dehydrative
annulation strategy.
15