Synthesis of novel bicyclic nitroxides using partial favorskii rearrangement

Article abstract of DOI:10.1055/s-2008-1072715

3-Bromo-2,2,6,6-tetramethyl-4-oxopiperidine-1-oxyl was reacted with several C-nucleophiles to give novel bicyclic pyrrolidine nitroxides through partial Favorskii rearrangement. Further reduction with sodium borohydride gave spin probes with free hydroxyl groups and under harsh reduction conditions allowed the Favorskii rearrangement to proceed to completion. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2008-1072715

LETTER
1155
Synthesis of Novel Bicyclic Nitroxides Using Partial Favorskii Rearrangement
S
ynthesis of
N
ove
n
l
Bicyclic
N
i
d
troxides rej Babič,*a Slavko Pečara,b
a
Faculty of Pharmacy, University of Ljubljana, Aškerčeva 7, 1000 Ljubljana, Slovenia
Fax +386(1)4258031; E-mail: andrej.babic@ffa.uni-lj.si
Jožef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
b
Received 24 January 2008
thors have demonstrated that a similar rearrangement can
also proceed in the presence of C-nucleophiles.13–18 The
classical mechanism of Favorskii rearrangement involves
a cyclopropanone intermediate which reacts with another
Abstract: 3-Bromo-2,2,6,6-tetramethyl-4-oxopiperidine-1-oxyl
was reacted with several C-nucleophiles to give novel bicyclic pyr-
rolidine nitroxides through partial Favorskii rearrangement. Further
reduction with sodium borohydride gave spin probes with free hy-
droxyl groups and under harsh reduction conditions allowed the Fa- nucleophile/base molecule to yield a carboxylic acid or its
vorskii rearrangement to proceed to completion.
derivative. The reaction is subjected to the influence of
solvent, acidity of substrate, structure of cyclopropanone
intermediate, carboanion stability, and base structure.19,20
We describe the reaction of 3-Br-TEMPONE with select-
ed C-nucleophiles and present a convenient route to new
nitroxyl spin probes.
Key words: bicyclic compounds, rearrangements, nitroxide free
radicals, spin probes, ring opening
Favorskii rearrangement has been known for over 100
years and it is still often used in synthetic organic chemis-
try; for example, in the synthesis of A-norsteroids,1
amides from a-haloketimines,2 and cyclic analogues of
proline.3 It was important in the field of nitroxide chemis-
try from the very beginning, since it allows for the conve-
nient preparation of pyrroline and pyrrolidine nitroxides4–6
which are more resistant to acidic media and redox condi-
tions in vitro and in vivo.6,7 Recent advances in low-fre-
quency electron paramagnetic resonance imaging have
created a need for spin probes that can provide specific
data in real time.8–10 Nitroxides are very well suited for
this task and novel synthetic approaches that could pro-
vide molecular tools with tailored properties for spin
probing are welcome.11 Existing synthetic approaches to
pyrroline nitroxides are multistep processes which mostly
start from 3,5-dibromo-2,2,6,6-tetramethyl-4-oxopiperi-
dine.4,6 The rearrangement of 3-bromo-2,2,6,6-tetrameth-
yl-4-oxopiperidin-1-oxyl (3-Br-TEMPONE, 1) with O-
nucleophiles gives the corresponding esters of 2,2,5,5-tet-
ramethyl-pyrrolidin-1-oxyl-3-carboxylic acid (2) in one
reaction step, as depicted in Scheme 1. This approach is a
good alternative to tedious multistep syntheses.12
O
O
R
Br
O
N
O•
1
N
O•
2
Scheme 1 Favorskii rearrangement with an O-nucleophile
Compound 1, the key reagent for the Favorskii rearrange-
ment, was synthesized from triacetoneamine according to
known procedures in five reaction steps as outlined in
Scheme 2.4,5,12 Scheme 3 presents the conversion of 3-Br-
TEMPONE (1) to 6a–h under the conditions of Favorskii
rearrangement. However, contrary to N- and O-nucleo-
philes the reaction evidently stops on the tetrahydric inter-
mediate stage, which is a tertiary alcohol when C-
nucleophiles are used. The O- and N-nucleophiles convert
to semiketals and semiazaketals, respectively, which are
unstable under normal reaction conditions and give proper
Favorskii products which, in our case, would be pyrroli-
dine nitroxides.
Favorskii rearrangement of a-haloketones is usually car-
ried out with N- or O-nucleophiles, although several au-
The partial Favorskii reaction was carried out by slowly
adding appropriate sodium salts, prepared in the usual
O
O
O
O
Br
i
ii, iii
iv, v
N
O•
N+
N
O•
N
H
3
H
Cl–
OH
4
5
1
Scheme 2 Synthesis of 3-Br-TEMPONE. Reaction conditions: (i) H2O2, Na2WO4, Na2EDTA, H2O, r.t.; (ii) H2, Pd/C, MeOH, r.t.; (iii) HCl(aq)
,
(iv) Br2, CHCl3, r.t.; (v) NaNO2, CHCl3, r.t.
SYNLETT 2008, No. 8, pp 1155–1158
x
x.
x
x.
2
0
0
8
Advanced online publication: 16.04.2008
DOI: 10.1055/s-2008-1072715; Art ID: G02608ST
© Georg Thieme Verlag Stuttgart · New York

1156
A. Babič, S. Pečar
LETTER
Table 1 Partial Favorskii Rearrangement Products 6a–h, Reaction Conditions, and Yields
Compd
6a
R1
R2
R3
H
Solvent
Et2O
THF
THF
THF
THF
THF
THF
THF
Na salt (equiv) Temp (°C)
Time (h)
24
Yield (%)a
COMe
COMe
COMe
COMe
CO2Me
COPh
CN
CO2Et
CO2Me
CO2t-Bu
COMe
CO2Me
CO2Et
CO2Me
CO2Et
2.0
1.4
1.4
2.5
1.4
1.4
4.0
1.05
34
50
50
30
50
50
50
50
65
81
71
31
75
81
37
60
6b
H
1.5
6c
H
1.5
6d
H
24
6e
H
1.5
6f
H
1.5
6g
H
1.5
6h
CO2Et
Me
1.5
a Isolated yields.
practical for spin-probe synthesis.22 Even though the rear-
rangement mechanistically requires minimal excess of
base/nucleophile, it was advantageous to use up to 1.4
equivalents of sodium salts. The yields of 6d and 6g23
could not be improved, even with bigger excess of base,
indicating an important influence of the nucleophile/base
on the outcome of the reaction. It is noteworthy to add that
the bicyclic structure was unambiguously confirmed by
indirect 1H NMR and 13C NMR spectroscopy of reduced
form of 6e,24 which clearly showed cyclopropane protons
and carbon atoms, respectively. Since 6e was a nitroxide
free radical and was therefore paramagnetic it had to be
reduced to its hydroxylamine derivative prior to all NMR
experiments.25 The reaction products, conditions, and
yields are summarized in Table 1.
R1
R2
HO
O
R3
Br
i
N
N
O•
6a–h
O•
1
Scheme 3 Partial Favorskii rearrangement with C-nucleophiles. Re-
action conditions: (i) R1R2R3-C–Na+, THF, reflux.
manner,21 to a stirred solution of 1 in dry solvent under
conditions listed in Table 1. Diethyl ether and THF were
found to be suitable solvents for these reactions and, due
to the higher boiling point of THF, reactions in this sol-
vent were faster. Furthermore, the higher solubility of so-
dium salts in THF also contributed to increased reaction
rate. It is noteworthy, however, that 6a could also be ob-
tained in good yield in diethyl ether. The optimal reaction
temperature proved to be 50 °C and reaction times were
no longer than 1.5 h, which makes the reaction highly
Scheme 4 outlines the further reaction steps used to obtain
new spin probes and Table 2 presents reaction conditions
and yields. Compounds 6f and 6c were partially reduced
at 0 °C in methanol using NaBH4, giving esters 8a and
8b,26 respectively, as four diastereomers. At low tempera-
R4
O
HO
O
R5
O
N
O•
6f
6b,f
i
iii
R4
R4
HO
HO
HO
6c,f
OH
OH
HO
ii
R4
HO
HO
N
N
O
R5
O•
O•
9a,b
7
O
N
O•
8a,b
Scheme 4 Sodium borohydride reduction products. Reaction conditions: (i) NaBH4, MeOH, 0–25 °C; (ii) NaBH4, MeOH, 0 °C; (iii) NaBH4,
THF–MeOH, 40 °C.
Synlett 2008, No. 8, 1155–1158 © Thieme Stuttgart · New York

LETTER
Synthesis of Novel Bicyclic Nitroxides
1157
Table 2 NaBH4 Reduction Products, Conditions, and Yields
Compd
7
Starting compound R4
R5
–
Solvent
Temp (°C)
Time (h)
Yield (%)a
6f
6f
6c
6f
6b
Ph
MeOH
0–25
12
3
73
73
93
21
34
8a
Ph
Et
t-Bu
–
MeOH
0
0
8b
Me
Ph
MeOH
3
9a
THF–MeOH
THF–MeOH
40
40
3
9b
Me
–
3
a Isolated yields.
(4) Rozantzev, E. G. Free Nitroxyl Radicals; Plenum Press:
New York, 1970, 204.
(5) Volodarsky, L. B.; Reznikov, V. A.; Ovacharenko, V. I.
Synthetic Chemistry of Stable Nitroxides, 40; CRC Press:
Boca Raton, 1994, 89.
ture the reaction proceeded stereoselectively with the ratio
between the two enantiomeric pairs of 4.8 and 6.2 for 8a
and 8b, respectively, determined by HPLC. Carbonyl
group as well as the ester moiety of 6f, however, could
also be fully reduced to bicyclic triol 727 by adding an ex-
cess of NaBH4 several times, allowing the reaction tem-
perature to raise and prolonging the reaction time which
also resulted in the drop of the ratio between enantiomeric
pairs to 3.0. The products of the reduction reactions 7, 8a,
and 8b were also indirect proofs that Favorskii rearrange-
ment indeed proceeded only partially.
Interestingly, 9a28 was first observed as a side product,
present in very small amounts, in the synthesis of 7. Mod-
ification of reaction conditions lead to a slight increase in
yield of 9a and 9b. The harsh reduction conditions, with
the addition of methanol producing sodium methylate in
situ, opened the cyclopropane ring and therefore allowed
the final step of the Favorskii rearrangement to take place.
This gave a pyrrolidine type of spin labels, 9a and 9b as a
mixture of several diastereomers.
(6) Hankovszky, O. H.; Hideg, K.; Bódi, I.; Frank, L. J. Med.
Chem. 1986, 29, 1138.
(7) Rozantzev, E. G.; Krinitzkaya, L. A. Tetrahedron 1965, 21,
491.
(8) Elas, M.; Williams, B. B.; Parasca, A.; Mailer, C.; Pelizzari,
C. A.; Lewis, M. A.; River, J. N.; Karczmar, G. S.; Barth, E.
D.; Halpern, H. J. Magn. Reson. Med. 2003, 49, 682.
(9) Shen, J.; Liu, S.; Miyake, M.; Liu, W.; Pritchard, A.; Kao, J.
P. Y.; Rosen, G. M.; Tong, Y.; Liu, K. J. Magn. Reson. Med.
2006, 55, 1433.
(10) Kao, J. P. Y.; Barth, E. D.; Burks, S. R.; Smithback, P.;
Mailer, C.; Ahn, K.-H.; Halpern, H. J.; Rosen, G. M. Magn.
Reson. Med. 2007, 58, 850.
(11) Berliner, L. J. Spin Labeling; Academic Press: New York,
1976, 274.
(12) Marc, G.; Pečar, S. Synth. Commun. 1995, 25, 1015.
(13) Sosnovsky, G.; Cai, Z. J. Org. Chem. 1995, 60, 3414.
(14) Sakai, T.; Amano, E.; Kawabata, A. A.; Takeda, A. J. Org.
Chem. 1980, 45, 43.
In summary, we present a convenient route to new bicy-
clic nitroxides, using a partial Favorskii rearrangement of
3-bromo-TEMPONE with C-nucleophiles. The cyclopro-
pane ring of bicyclic nitroxides remains intact under mild
reduction conditions while, at elevated temperatures, it
was possible to obtain classical products of Favorskii re-
arrangement. This gave new nitroxides with multiple
functional groups applicable for spin labeling or used
alone as spin probes.
(15) Sakai, T.; Katayama, T.; Takeda, A. J. Org. Chem. 1981, 46,
2924.
(16) Sakai, T.; Tabata, H.; Takeda, A. J. Org. Chem. 1983, 48,
4618.
(17) Barbee, T. R.; Hedeel, G.; Heeg, M. J.; Albizati, K. F. J. Org.
Chem. 1991, 56, 6773.
(18) De Angelis, F.; Feroci, M.; Inesi, A. Bull. Soc. Chim. Fr.
1993, 130, 712.
(19) Rappe, C.; Knutsson, L.; Turro, N. J.; Gagosian, R. B. J. Am.
Chem. Soc. 1970, 92, 2032.
(20) House, H. O.; Frank, G. A. J. Org. Chem. 1965, 30, 2948.
(21) Becker, H. G. O. Organikum; Wiley-VCH: Weinheim,
2001, 608.
Acknowledgment
(22) Typical Procedure for the Synthesis of Products 6a–h
3-Bromo-2,2,6,6-tetramethyl-1-oxyl-piperidine-4-one (1,
400 mg, 1.61 mmol) was dissolved in anhyd THF (20 mL).
The sodium salt of the appropriate C-nucleophile was
dissolved in anhyd THF (5 mL) and added dropwise to the
stirred solution of 1 over a period of 5 min under argon. The
reaction mixture was refluxed at 50 °C for 1.5 h. After that
time the solvent was evaporated under reduced pressure and
the yellow solid dissolved in cold citric acid (10 mL,
pH 3.5). The aqueous phase was immediately extracted with
EtOAc. The organic phase was washed with brine, dried
with Na2SO4, and the solvent evaporated under reduced
pressure. The crude product was purified using circular
chromatography (Et2O–hexane, 3:1). A bright yellow solid
was obtained.
Financial support by the Ministry of Education, Science, and Sport
of Slovenia is gratefully acknowledged. We would also like to thank
Jožef Stefan Institute for MS and EPR spectra, the Faculty of Che-
mistry and Chemical Technology, University of Ljubljana for
microanalysis and Dr. R. Pain for critical reading of the manuscript.
References and Notes
(1) Kende, A. S. The Favorskii Rearrangement of Haloketones,
In Organic Reactions, Vol. 2; John Wiley and Sons: New
York, 1960, 261.
(2) De Kimpe, N.; Verhe, R.; Buyck, L.; Moens, L.; Schamp, N.
Tetrahedron Lett. 1981, 22, 1837.
(3) Nagasawa, H. T.; Elberling, J. A. Tetrahedron Lett. 1969,
44, 5393.
Synlett 2008, No. 8, 1155–1158 © Thieme Stuttgart · New York

1158
A. Babič, S. Pečar
LETTER
(23) Methyl 2-Cyano-2-{6-hydroxy-3-oxyl-2,2,4,4-
tetramethyl-3-azabicyclo[3.1.0]hex-6-yl}acetate (6g)
Yield 37%; yellow solid; mp 120–124 °C. IR (KBr): 3213,
2984, 2257, 1738, 1442, 1306, 1220, 1016, 928, 712 cm–1.
MS (EI): m/z = 267 [M]+. 1H NMR (300 MHz, CDCl3): d =
1.28 (s, 2 H, CH), 1.32 (s, 12 H, CH3), 3.30 (s, 1 H, CH), 3.84
(s, 3 H, COOCH3) ppm. EPR: aN = 1.55 mT. Anal. Calcd for
C13H19N2O4 (%): C, 58.41; H, 7.19; N, 10.48. Found: C,
58.17; H, 7.33; N, 10.71.
(27) Preparation of 2-{6-Hydroxy-3-oxyl-2,2,4,4-tetramethyl-
3-azabicyclo[3.1.0]hex-6-yl}-1-phenyl-propane-1,3-diol (7)
Ethyl 2-{6-hydroxy-3-oxyl-2,2,4,4-tetramethyl-3-
azabicyclo[3.1.0]hex-6-yl}-3-oxo-3-phenylpropanoate (6f,
170 mg, 0.47 mmol) was dissolved in MeOH (10 mL) and
cooled to 0 °C. While being stirred, NaBH4 (42 mg, 1.06
mmol) was added. The same amount of NaBH4 (42 mg, 0.53
mmol) was again added after 1 h and 2 h. After the third
addition the reaction mixture was allowed to warm to
ambient temperature. Then, 12 h later, citric acid was added
and the pH adjusted to 6. The solvents were evaporated
under reduced pressure and citric acid (10%) was added to
adjust the pH to 3. The water phase was extracted with
EtOAc. The organic phase was washed with brine, dried
with Na2SO4, and the solvent evaporated under reduced
pressure. The crude product was purified using circular
chromatography (CH2Cl2–MeOH, 15:1). A bright yellow
solid was obtained; yield 73%; mp 147–157 °C. IR (KBr):
3421, 2978, 1649, 1457, 1287, 1174, 1032, 705 cm–1. MS
(EI): m/z = 320 [M]+. HRMS (EI): m/z calcd for C18H26N1O4:
320.186184 [MH+]; found: 320.187330.
(24) Dimethyl 2-{6-Hydroxy-3-oxyl-2,2,4,4-tetramethyl-3-
azabicyclo[3.1.0]hex-6-yl} Malonate (6e)
Yield 75%; yellow solid; mp 146–150 °C. IR (KBr): 3442,
2981, 1742, 1438, 1253, 1179, 1011, 914, 715, 624 cm–1. MS
(EI): m/z = 300 [M]+. 1H NMR (300 MHz, CDCl3): d = 1.21
(s, 12 H, CH3), 1.30 (s, 2 H, CH), 3.15 (s, 1 H, CH), 3.67 (s,
6 H, COOCH3) ppm. 13C NMR (75 MHz, CDCl3): d = 23.21,
25.36, 31.58, 52.89, 58.09, 61.71, 65.92, 168.32 ppm. EPR:
aN = 1.55 mT. Anal. Calcd for C14H22NO6 (%): C, 55.98; H,
7.38; N, 4.66. Found: C, 55.78; H, 7.57; N, 4.69.
(25) NMR spectra were obtained after reduction with phenyl
hydrazine. The compound was dissolved in perdeuterated
solvent, transferred into a NMR tube, and flushed with
argon. Phenyl hydrazine (2 equiv) was added and the NMR
tube shaken to allow the reduction to proceed to completion
prior to the NMR spectroscopic analysis.
(28) Preparation of 1-(1-Oxyl-2,2,5,5-
tetramethylpyrrolidine-3-yl)-2-hydroxymethyl-3-
phenylpropane-1,3-diol (9a)
Ethyl 2-{6-hydroxy-3-oxyl-2,2,4,4-tetramethyl-3-
(26) Preparation of tert-Butyl 3-Hydroxy-2-{6-hydroxy-3-
oxyl-2,2,4,4-tetramethyl-3-azabicyclo[3.1.0]hex-6-
yl)butanoate} (8)
azabicyclo[3.1.0]hex-6-yl}-3-oxo-3-phenylpropanoate (6f,
300 mg, 0.83 mmol) was dissolved in THF (10 mL), and
NaBH4 (148 mg, 3.72 mmol) was added. The reaction
mixture was refluxed at 40 °C and MeOH (0.5 mL) was
added over 0.5 h. Following the addition of MeOH the
temperature was maintained at 40 °C for another 2.5 h.
Several drops of H2O were then added and the solvents
evaporated under reduced pressure, giving a bright yellow
foam. Citric acid (10%, 10 mL) was added and the water
phase extracted with EtOAc after the evolution of hydrogen
stopped. The organic phase was washed with brine, dried
with Na2SO4, and the solvent evaporated under reduced
pressure. The crude product was purified using circular
chromatography (CH2Cl2–MeOH, 15:1). A bright yellow
solid was obtained; yield 21%; mp 157–163 °C. IR (KBr):
3290, 2976, 1474, 1411, 1367, 1295, 1257, 1050, 763, 706
cm–1. MS (EI): m/z = 322 [M]+. MS–FAB: m/z = 322 [M]+,
324 [M + 2]+. HRMS (EI): m/z calcd for C18H28N1O4:
322.201834 [M]+; found: 322.202650.
tert-Butyl 2-{6-hydroxy-3-oxyl-2,2,4,4-tetramethyl-3-
azabicyclo[3.1.0]hex-6-yl}-3-oxobutanoate (6c) (200 mg,
0.61 mmol) was dissolved in MeOH (10 mL) and cooled to
0 °C. Sodium borohydride (24 mg, 0.61 mmol) was added
with stirring. The temperature was maintained at 0 °C for 2
h, after which the reaction was quenched with citric acid and
the pH adjusted to 6. The solvents were evaporated under
reduced pressure and citric acid (10%) was added to adjust
the pH to 3. The water phase was extracted with EtOAc. The
organic phase was washed with brine, dried with Na2SO4,
and the solvent evaporated under reduced pressure. The
crude product was purified using circular chromatography
(Et2O–hexane, 3:1). A bright yellow solid was obtained;
yield 93%; mp 101–108 °C. IR (KBr): 3460, 2976, 1702,
1477, 1375, 1160, 993, 646 cm–1. MS (EI): m/z = 328 [M]+.
MS–FAB: m/z = 330 [M + 2]+. HRMS (EI): m/z calcd for
C17H30N1O5: 328.212398 [MH]+; found: 328.213250.
Synlett 2008, No. 8, 1155–1158 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.