Quinine synthesis studies: A radical-ionic annulation via Mn-mediated addition to chiral N-acylhydrazones

Article abstract of DOI:10.1021/ol7017938

A radical-ionic annulation approach to functionalized perhydroisoquinolines involving Mn-mediated coupling of alkyl iodides and chiral N-acylhydrazones was achieved using only 1.25 equiv of the alkyl iodide. Application of this reaction to alkene-containi

Full text of DOI:10.1021/ol7017938

ORGANIC
LETTERS
2007
Vol. 9, No. 21
4243-4246
Quinine Synthesis Studies:
A Radical Ionic Annulation via
−
Mn-Mediated Addition to Chiral
N-Acylhydrazones
Chandra Sekhar Korapala, Jun Qin,† and Gregory K. Friestad*
Department of Chemistry, UniVersity of Iowa, Iowa City, Iowa 52242
gregory-friestad@uiowa.edu
Received July 26, 2007
ABSTRACT
A radical−ionic annulation approach to functionalized perhydroisoquinolines involving Mn-mediated coupling of alkyl iodides and chiral
N-acylhydrazones was achieved using only 1.25 equiv of the alkyl iodide. Application of this reaction to alkene-containing substrates en route
to quinine offered modest yields, decreasing on scaleup. Control experiments revealed that the alkene interfered with the coupling reaction.
A revised approach involving prior oxidation of the alkene offered 93% yield in the Mn-mediated coupling, with the adduct obtained as a single
diastereomer.
The long-established antimalarial properties of quinine have
fueled great interest in synthetic efforts toward quinine since
the very early days of natural product synthesis.1 A surge of
media attention accompanied Woodward’s formal synthesis
of quinine in 1944,2 but even with practical developments
led by Uskokovic, Taylor, and Gates in the 1970s3 for access
to quinine and quinidine, there was still no total synthesis
amines.7 Toward this end, we have introduced an efficient
with complete configurational control until 2001. Quinine
and highly stereoselective free-radical coupling of alkyl
iodides and chiral N-acylhydrazones.8,9 In general, radical
additions to imino compounds are attractive complements
to polar methods for reasons of functional group compat-
finally succumbed to the first asymmetric total synthesis as
reported by Stork in 2001,4 and syntheses by Jacobsen5 and
Kobayashi6 appeared in 2004.
Quinine attracted our interest in the course of our program
to develop new C-C bond construction approaches to chiral
(4) Stork, G.; Niu, D.; Fujimoto, A.; Koft, E. R.; Balkovec, J. M.; Tata,
J. R.; Dake, G. R. J. Am. Chem. Soc. 2001, 123, 3239-3242.
(5) Raheem, I. T.; Goodman, S. N.; Jacobsen, E. N. J. Am. Chem. Soc.
2004, 126, 706-707.
(6) Igarashi, J.; Katsukawa, M.; Wang, Y.-G.; Acharya, H. P.; Kobayashi,
Y. Tetrahedron Lett. 2004, 45, 3783-3786.
† Current address: Schering-Plough Research Institute, Kenilworth, NJ.
(1) Kaufman, T. S.; Ru´veda, E. A. Angew. Chem., Int. Ed. 2005, 44,
854-885.
(2) Woodward, R. B.; Doering, W. E. J. Am. Chem. Soc. 1945, 67, 860-
874.
(3) (a) Uskokovic, M.; Gutzwiller, J.; Henderson, T. J. Am. Chem. Soc.
1970, 92, 203-204. Gutzwiller, J.; Uskokovic, M. J. Am. Chem. Soc. 1970,
92, 204-205. (b) Gates, M.; Sugavanam, B.; Schreiber, W. L. J. Am. Chem.
Soc. 1970, 92, 205-207. (c) Taylor, E. C.; Martin, S. F. J. Am. Chem. Soc.
1974, 96, 8095-8102.
(7) Review: Friestad, G. K. Eur. J. Org. Chem. 2005, 3157-3172.
(8) (a) Friestad, G. K.; Qin, J. J. Am. Chem. Soc. 2000, 122, 8329-
8330. (b) Shen, Y.; Friestad, G. K. J. Org. Chem. 2002, 67, 6236-6239.
(c) Friestad, G. K.; Draghici, C.; Soukri, M.; Qin, J. J. Org. Chem. 2005,
70, 6330-6338.
10.1021/ol7017938 CCC: $37.00
© 2007 American Chemical Society
Published on Web 09/12/2007

Scheme 1
ibility, but applications to total synthesis have lagged because
trisubstituted piperidine precursor A, which in turn would
derive from oxidative cleavage of the C6-C7 bond of an
isoquinoline such as B (Scheme 1). We envisaged an efficient
access to the isoquinoline ring system through Mn-mediated
radical addition to construct either of the two C-C bonds at
the C3 stereogenic carbon (disconnections a and b). The
stereoselective radical addition would be followed by closed-
shell nucleophilic displacement of a leaving group X by the
imino nitrogen to complete the piperidine heterocycle.
Some important aspects of this strategy warrant specific
mention. First, the Mn-mediated coupling enables interchange
of the iodide and hydrazone functionality, such that either
component could serve as radical precursor or acceptor. Thus,
the choice of path a or b is flexible, to be defined at the
benchtop according to optimal yields or selectivities. Second,
the strategy reveals a common precursor C, for which C2
symmetry could be exploited in an efficient access to both
sets of coupling components. Finally, if the chiral auxiliary
X* is the dominant control element (as was expected from
all available precedents), then the enantiomeric iodide (e.g.,
ent-C) could generate the C8-C3/4 stereochemical relation-
ship of quinidine.
methods exhibiting versatility with respect to both radical
and acceptor are limited.10 The emergence of Mn-mediated
photolytic initiation of these intermolecular radical additions
has addressed this problem to some extent, enabling ap-
plication to precursors bearing electrophilic functionality in
either of the coupling components.11 Quinine presented an
ideal challenge to the applicability of these Mn-mediated
coupling reactions to synthetic problems in a multifunctional
molecular setting.
Our approach to quinine focuses on strategic application
of our Mn-mediated hybrid radical-ionic annulation, a
radical-polar crossover reaction,12 which had previously
been described for preparation of simple pyrrolidines and
piperidines.11a,b Employing Stork’s disconnection of the
azabicyclo[2.2.2]octane ring system suggested a 2,4,5-
(9) For other applications of chiral N-acylhydrazones, see the following.
(a) Allylsilane additions: Friestad, G. K.; Ding, H. Angew. Chem., Int. Ed.
2001, 40, 4491-4493. Friestad, G. K.; Korapala, C. S.; Ding, H. J. Org.
Chem. 2006, 71, 281-289. (b) Radical additions: Ferna´ndez, M.; Alonso,
R. Org. Lett. 2003, 5, 2461-2464. (c) Allylindium additions: Cook, G.
R.; Maity, B. C.; Kargbo, R. Org. Lett. 2004, 6, 1749-1752. (d) Mannich-
type reactions: Jacobsen, M. F.; Ionita, L.; Skrydstrup, T. J. Org. Chem.
2004, 69, 4792-4796. (e) Strecker reactions: Ding, H.; Friestad, G. K.
Heterocycles 2006, 70, 185-199. (f) Hydride additions: Qin, J.; Friestad,
G. K. Tetrahedron 2003, 59, 6393-6402.
(10) Reviews: (a) Friestad, G. K. Tetrahedron 2001, 57, 5461-5496.
(b) Bertrand, M.; Feray, L.; Gastaldi, S. C. R. Acad. Sci. Paris, Chim. 2002,
5, 623-638. (c) Miyabe, H.; Ueda, M.; Naito, T. Synlett 2004, 1140-
1157. For selected recent developments in intermolecular radical addition
to CdN bonds, see: (d) Yamada, K.; Yamamoto, Y.; Maekawa, M.;
Akindele, T.; Umeki, H.; Tomioka, K. Org. Lett. 2006, 8, 87-89. (e) Ueda,
M.; Miyabe, H.; Sugino, H.; Miyata, O.; Naito, T. Angew. Chem., Int. Ed.
2005, 44, 6190-6193. (f) McNabb, S. B.; Ueda, M.; Naito, T. Org. Lett.
2004, 6, 1911-1914. (g) Miyabe, H.; Yamaoka, Y.; Takemoto, Y. J. Org.
Chem. 2005, 70, 3324-3327. (h) Clerici, A.; Cannella, R.; Pastori, N.;
Panzeri, W.; Porta, O. Tetrahedron 2006, 62, 5986-5994. (i) Risberg, E.;
Fischer, A.; Somfai, P. Tetrahedron 2005, 61, 8443-8450. (j) Ferna´ndez,
M.; Alonso, R. Org. Lett. 2003, 5, 2461-2464.
Testing these hypotheses began with preparation of an
iodide encompassing the structural features of C. For this
purpose, the C2-symmetric diester 1 (ROH ) (-)-menthol)
was acquired using the known enantioselective Diels-Alder
reaction of dimenthyl fumarate13 (Scheme 2). Reductive
removal of the (-)-menthol and monosilylation of the
Scheme 2a
(11) (a) Friestad, G. K.; Qin, J. J. Am. Chem. Soc. 2001, 123, 9922-
9923. (b) Friestad, G. K.; Qin, J.; Suh, Y.; Marie´, J.-C. J. Org. Chem. 2006,
71, 7016-7027. (c) Friestad, G. K.; Deveau, A. M.; Marie´, J.-C. Org. Lett.
2004, 6, 3249-3252.
(12) (a) Review: Murphy, J. A. In Radicals in Organic Synthesis;
Renaud, P., Sibi, M. P., Eds.; Wiley-VCH: Weinheim, 2001; Vol 1, pp
298-316. For leading references, see: (b) Callaghan, O.; Lampard, C.;
Kennedy, A. R.; Murphy, J. A. J. Chem. Soc., Perkin Trans. 1 1999, 995-
1001. (c) Jahn, U.; Muller, M.; Aussieker, S. J. Am. Chem. Soc. 2000, 122,
5212-5213. (d) Rivkin, A.; Nagashima, T.; Curran, D. P. Org. Lett. 2003,
5, 419-422. (e) Denes, F.; Chemla, F.; Normant, J. F. Angew. Chem., Int.
Ed. 2003, 42, 4043-4046. (f) Tojino, M.; Uenoyama, Y.; Fukuyama, T.;
Ryu, I. Chem. Commun. 2004, 2482-2483. (g) Bazin, S.; Feray, L.;
Vanthuyne, N.; Bertrand, M. P. Tetrahedron 2005, 61, 4261-4274. (h)
Ueda, M.; Miyabe, H.; Sugino, H.; Miyata, O.; Naito, T. Angew. Chem.,
Int. Ed. 2005, 44, 6190-6193.
a ROH ) (-)-menthol; the antipodes employed (+)-menthol.
4244
Org. Lett., Vol. 9, No. 21, 2007

resulting C2-symmetric diol furnished desymmetrized (+)-
2, which was readily converted to the corresponding iodide
(+)-3a. The silyl ether was cleaved by acidic methanolysis
to afford the corresponding alcohol (+)-3b. Both enantiomers
of 1-3 were prepared in similar fashion.
Table 1. Mn-Mediated Addition Studiesa
Iodide 3a was prepared as a radical precursor; to test the
alternative C-C bond construction (path b of Scheme 1) it
was also necessary to interchange the functionality from
iodide to the homologous N-acylhydrazone. This was readily
accomplished via a simple five-step sequence (Scheme 3).
Scheme 3
a Conditions: InCl3 (2.2 equiv) and hydrazone (ca. 0.2 mmol) in CH2Cl2
(0.02 M), 40 min at rt, then R-I and Mn2(CO)10 (1.2 equiv), irradiation
(300 nm, Rayonet) for 10-20 h; Et3N (5 equiv) then silica gel flash
chromatography. b Solvent ) 10:1 PhH/MeCN, DBU in place of Et3N.c Ca.
25% yield, but inseparable impurities prevented characterization of 12b.
d 0.65 mmol scale. e dr 1:1 (both trans).
Cyanide homologation and conversion of OTBS to Cl was
followed by partial reduction of the nitrile to the correspond-
ing aldehyde and condensation with N-aminooxazolidinone
(R)-7. This furnished radical acceptor 8a in an overall 56%
yield. The related acceptor 8b was prepared in three steps
from 3a with an overall yield of 74%. Benzoate and acetate
analogues were synthesized from 3a by a related dithiane-
based homologation route.14
With a set of radical precursors 3a,b and acceptors 8a-d
at hand, screening their Mn-mediated addition reactions with
hydrazone 911a and chloroiodomethane, respectively, was
examined (Table 1). In the presence of InCl3 (employed as
a chelating Lewis acid11b), the desired adducts were obtained,
generally as single diastereomers.15 Unfortunately, the yields
in these couplings never exceeded ca. 50% (entries 1-6),
and a slight scaleup decreased the yield from 49% to 31%
(entries 3 and 7). Even more troubling was the poor mass
balance; in contrast to our prior experience with these Mn-
mediated coupling reactions, no starting material was re-
coverable from lower yielding runs. Moreover, no tractable
side products were found, which obscured any obvious
corrections to the conditions.
The question remained of whether the remote alkene
functionality of 3a,b could interfere in some unanticipated
manner.16 This was easily tested by experiment. Thus, a
saturated analogue (()-3 was prepared and subjected to the
coupling. The racemic diacid (()-trans-cyclohexane-1,2-
dicarboxylic acid was transformed to saturated iodide (()-
3c via a sequence analogous to Scheme 2. Now absent the
alkene functionality, the radical addition of (()-3c to 9 was
significantly improved, furnishing 67% yield of 13 without
any optimization (Table 1, entry 8).17 This control experiment
confirmed that the alkene interfered with the Mn-mediated
coupling.18
With this clear guidance for improving the key C-C bond
construction, the original C6-C7 bond cleavage (e.g., by
alkene ozonolysis) was revised to a plan entailing oxidation
of the C6-C7 alkene to a diol prior to the radical addition,
(13) Furuta, K.; Iwanaga, K.; Yamamoto, H. Tetrahedron Lett. 1986,
27, 4507-4710.
(14) See the Supporting Information.
(16) Cyclization of radicals derived from 3a,b onto the alkene functional-
ity may be proposed, but significant ring strain would be expected in the
relevant bicyclic 4-exo or 5-endo transition states.
(15) (a) In Table 1, only entry 1 showed evidence of a minor diastereomer
(dr 94:6 by 1H NMR); others were obtained with dr >95:5. (b) Configura-
tions of 10, 11, 12a-d, and 13 were assigned on the basis of precedent
(refs 7-9 and 11), supported by NOE analysis of 18.
(17) The addition of racemic (()-3c (1.3 equiv) to 9 afforded only two
of the four possible diastereomeric adducts.
(18) At this time, there is insufficient evidence to draw conclusions
regarding potential mechanisms of interference by alkene functionality.
Org. Lett., Vol. 9, No. 21, 2007
4245

Scheme 4
with periodate cleavage to take place at a later stage. In the
yield in 1 mmol scale, giving 17 as a single diastereomer.20
Efficient addition of the multifunctional alkyl group of 3d
to an imino compound by other existing methods is improb-
able.21
revised plan (Scheme 4), diastereoselective preparation of a
C2-symmetric trans-diol preserved the simplifying desym-
metrization aspect of the strategy. The enantiopure diester
ent-1 was treated with m-CPBA, and the resulting epoxide
was subjected to acidic alcoholysis with benzyl alcohol.
Acid-catalyzed protection of the remaining free hydroxyl
with benzyl trichloroacetimidate furnished bis-benzyl ether
14. Reduction to the C2-symmetric diol, monosilylation to
15, and conversion to iodide 3d followed the previously
established sequence.
In the key Mn-mediated coupling of 3d and 16,19 we were
pleased to find quite remarkable yields. In most inter-
molecular radical additions to imino compounds, large
excesses (10-20 equiv or more) of radical precursors are
required. Clearly this would be a prohibitive stoichiometric
requirement for an iodide such as 3d, prepared through
several synthetic steps. To our great delight, the Mn-mediated
coupling of 3d with only 1.25 equiv of 16 proceeded in 93%
Completion of the hybrid radical-ionic annulation would
ideally occur in situ during the Mn-mediated coupling, but
this goal has not yet been achieved. However, a stepwise
radical-ionic annulation process proved efficient; successive
treatment with TsCl and NaI provided decahydroisoquinoline
18 in a quite satisfactory overall yield (85% for three steps).22
Further studies to elaborate the azabicyclic ring system of
quinine are in progress.
In conclusion, a challenging Mn-mediated coupling of a
multifunctional alkyl iodide and a chiral N-acylhydrazone
was addressed by prior oxidation of a remote alkene
functionality, affording efficient access to a functionalized
perhydroisoquinoline. Notably, the optimized conditions
required only 1.25 equiv of alkyl iodide, a finding which
should enable broader applications of this Mn-mediated
coupling process in complex target synthesis.
(19) Hydrazone 16 was prepared by condensation of glycolaldehyde
dimer with (S)-7, followed by O-silylation.
(20) The indicated configuration of 17 was established by NOE analysis
of 18 and was in accord with all prior examples of Mn-mediated addition
to chiral N-acylhydrazones (refs 7-9 and 11).
(21) (a) For a review of asymmetric addition to CdN bonds, see:
Friestad, G. K.; Mathies, A. K. Tetrahedron 2007, 63, 2541-2569. (b) For
example, diorganozinc additions are powerful methods for alkyl addition
to imino compounds, yet still require 2-6 molar equiv of the diorganozinc
reagent (i.e., 4-12 equiv of the organic fragment).
(22) Related ring closures affording decahydroisoquinolines are uncom-
mon. Chatterjee, A.; Sahu, A.; Saha, M.; Benerji, J. Indian J. Chem. Sect.
B 1997, 36, 121-122. Lohse, C.; Detterbeck, R.; Acklin, P.; Borschberg,
H.-J. HelV. Chim. Acta 2002, 85, 945-962. Hattori, K.; Grossman, R. B.
J. Org. Chem. 2003, 68, 1409-1417.
Acknowledgment. We thank the NIH (R01-GM67187)
for generous support of this work and Dr. M. Soukri for
preparation of 1.
Supporting Information Available: Preparative details
and characterization data for 2-18. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL7017938
4246
Org. Lett., Vol. 9, No. 21, 2007