Vitamin B12 catalyzed radical cyclizations of arylalkenes

Article abstract of DOI:10.1055/s-2005-923588

The use of vitamin B₁₂ for synthetic organic transformations has been extensively studied. Herein, we report the intramolecular cyclization reaction of a series of arylalkene substrates catalyzed by vitamin B ₁₂. These reactions proc

Full text of DOI:10.1055/s-2005-923588

LETTER
211
Vitamin B12 Catalyzed Radical Cyclizations of Arylalkenes
Vitamin
B
2-Catahlyzed
C
ycliz
r
ation
R
e
i
actionss M. McGinley, Heather A. Relyea, Wilfred A. van der Donk*
1
Department of Chemistry, University of Illinois, Urbana, IL 61801, USA
Fax +1(217)2448024; E-mail: vddonk@uiuc.edu
Received 29 July 2005
lyzed reaction is selective for mono- or 1,1-disubstituted
arylalkenes,2 and no products from activation of the
trisubstituted arylalkene were detected. Hence the phenyl
substituted alkene functions only as a radical trap. With
Abstract: The use of vitamin B12 for synthetic organic transforma-
tions has been extensively studied. Herein, we report the intra-
molecular cyclization reaction of a series of arylalkene substrates
catalyzed by vitamin B12. These reactions proceed in good yields in
environmentally benign solvents and do not require the use of a pyrroles 1c and 1d, a mixture of products was observed
toxic heavy metal catalyst. Variation of the reaction pH can predict-
ably alter the product distribution.
containing both the saturated and unsaturated cyclized
pyrroles 2 and 3. Pyrrole 1e yielded primarily the cyclized
Key words: vitamin B12, radical cyclization, arylalkenes
saturated product 2e, but dimerization products were also
observed. This observation reflects the lower reactivity of
the unsubstituted alkene as a radical trap compared to the
alkyl and aryl substituted alkenes in substrates 1a–d.
Upon cyclization compound 1e would afford a less favor-
able primary radical intermediate causing the bimolecular
dimerization process to become competitive. The 2-vi-
nylpyrrole substrates 1f and 1g (R1 = H) were also reac-
tive under these conditions, but dimerization products
were almost exclusively recovered with less than 10% cy-
clized products observed. Hence, the initiating arylalkene
must be 1,1-disubstituted for the cyclization reaction to be
successful. No products were detected containing an
exocyclic alkene 4.
Vitamin B12 and its derivatives have received consider-
able attention for their use in synthetic organic chemistry,1
including several recent new applications.2,3 Vitamin B12
contains a cobalt atom surrounded by an equatorial corrin
ring system. The chiral environment of the corrin ring has
been utilized for the development of stereoselective B12
catalyzed reactions.4 Covalently linked to the corrin is a
dimethylbenzimidazole that reversibly occupies one of
the axial coordination sites.1 Herein we report an exten-
sion of our previously reported work on a novel reaction
that utilizes vitamin B12 as a catalyst and Ti(III) citrate as
a reducing agent. In that earlier study, the coupling of sty-
rene derivatives or benzylic halides was reported with un-
usual regioselectivity providing a,a-dimers (Equation 1).2
Mechanistic studies suggested that the reaction takes
place through the intermediacy of benzylic radicals. This
letter expands the scope of this reaction to intramolecular
cyclization processes. In addition, the reaction mecha-
nism was explored uncovering an interesting reversal of
product distribution by altering the reaction pH.
Alternative intramolecular traps of the initially formed
radical were also evaluated. Recently, several examples
have been reported of oximes as effective radical accep-
tors.5,6 Therefore, oxime 5 was subjected to the vitamin
B12 catalyzed conditions affording N-hydroxylamine 6 in
moderate yield and good purity (Equation 2).
N
N
vitamin B12, Ti(III) citrate
EtOH, Tris buffer, pH = 8
63%
5
Me
R1
Me
R1
HN
cat. B12
N
OBn
R1
R2
OBn
6
R2 R2
Ti(III) citrate
Equation 2
Equation 1
The range of potential substrates was further explored
through the use of a-alkyl styrenes 7. Ethers 7a–c all re-
acted in good yields affording the cyclized reduced prod-
ucts with modest diastereoselectivities (Table 2). Similar
to the observations with the vinylpyrroles, ether 7d con-
taining a dimethylalkenyl group yielded a mixture of the
cyclized saturated and unsaturated products 8 and 9, re-
spectively. Compounds 10 containing an exocyclic alkene
were not observed.
The applicability of the reaction with respect to intra-
molecular cyclizations was initially investigated using a
series of N-substituted pyrroles 1 (Table 1). These sub-
strates were reacted with catalytic vitamin B12 and Ti(III)
citrate at pH 8.0. After 24 hours, the reactions were
stopped and the products isolated by extraction with hex-
anes. For compounds 1a and 1b, cyclization proceeded
smoothly affording the 2a,b in good yield. Only the 2-pro-
penyl group is activated in this process as the B12-cata-
While the results in Table 1 and Table 2 show that the
cyclized products can be obtained in moderate to good
yields, the reaction would be substantially more useful for
synthetic purposes if the distribution of the reaction
SYNLETT 2006, No. 2, pp 0211–0214
Advanced online publication: 23.12.2005
DOI: 10.1055/s-2005-923588; Art ID: S06205ST
© Georg Thieme Verlag Stuttgart · New York
0
1.
0
2.
2
0
0
6

212
C. M. McGinley et al.
LETTER
Table 1 Initial Screen of Substrates for Intramolecular Vitamin B12 Catalyzed Cyclizations at pH 8.0a
R1
R2
R1
CH2
N
N
R3
R3
vitamin B12
Ti(III) citrate
N
R1
R2
2
3
EtOH, Tris buffer
pH 8.0
R3
R1
N
1
R2
R3
4
Entry
1a
R1
R2
R3
H
Yield of 2 (%)
Yield of 3 (%)
Yield of Dimer (%)
Me
Me
Me
Me
Me
H
Ph
Ph
Me
Me
H
70
80
60
28
72
<2
6
0
0
0
0
1b
1c
Ph
H
40
42
0
0
1d
1e
Me
H
0
16
>72
84
1f
H
Me
H
0
1g
H
H
0
a Listed values are based on NMR integration and GCMS analysis of isolated crude reaction products.
products could be more predictably controlled. Further- The vitamin B12 pre-catalyst is first reduced by Ti(III)
more, it would be highly desirable if the unsaturated citrate to the catalytically active cob(I)alamin, which dis-
products such as 3 and 9 would be the major or exclusive plays a characteristic purple color with a UV maximum
product since they possess an alkene functionality that near 380 nm. The reduced metal catalyst activates the sub-
provides a convenient handle for further manipulation. In strate in a process that is not completely understood2 to
order to achieve this, we inspected the proposed2 catalytic generate the stabilized radical I. Most likely the activation
cycle for the reaction shown in Scheme 1.
involves an inner sphere process that may involve an
Table 2 Vitamin B12 Catalyzed Cyclization of a-Methylstyrene Derivativesa
vitamin B12
R1
CH2
Ti(III) citrate
pH 8.0
R1
R2
R2
R2
O
O
O
Ph
MeCN
Ph
Ph
8
9
7
R1
R2
O
Ph
10
Entry
7a
R1
R2
H
Yield of 8 (%)
Yield of 9 (%)b
Yield of 10 (%)
dr of 8c
3.6:1
3.2:1
4.1:1
1.5:1
H
78
68
80
36
–
0
0
0
0
7b
Ph
Ph
Me
Ph
H
–
7c
–
7d
Me
18
a Yields based on NMR analysis of crude isolated products.
b Product 9 cannot be formed from substrates 7a–c.
c The stereochemistry of the major isomer was assigned by NMR spectroscopy (HMQC, COSY, NOE).
Synlett 2006, No. 2, 211–214 © Thieme Stuttgart · New York

LETTER
Vitamin B12 Catalyzed Cyclizations Reactions
213
Table 3 Vitamin B12 Catalyzed Cyclization of Pyrroles 1 at pH 6.0a
organocobalamin followed by homolytic cleavage of the
Co–C bond to produce I.2 Radical I subsequently reacts
intramolecularly with the alkene to generate radical II.
The fate of this radical is dependent on the reaction condi-
tions. It can either be reduced to yield product III, or
cob(II)alamin can abstract a b-hydrogen atom to give the
unsaturated product IV. This would also produce hydri-
docobalamin, which would be rapidly deprotonated (pKa
ca. 1)7,8 regenerating the active cob(I)alamin catalyst.
Several studies have provided strong support for such a
homolytic mechanism for generation of alkenes as op-
posed to a b-hydride elimination mechanism from an
alkylcobalamin species.9–11
N
N
N
vitamin B12
+
R
R
Ti(III) citrate pH 6.0
R
2
3
Substrate
R
Yield of 2 (%)
Yield of 3 (%)
1c
H
Trace
5
66
70
1d
Me
a Yields based on NMR analysis of crude isolated products.
N
vitamin B12
Ti(III) citrate
procedure was repeated using pyrrole 1d and at pH 6.0
the unsaturated product 3d was again the predominant
product (Table 3).
N
I
The product distribution in the cyclization of ethers 7a
and 7d could also be biased toward formation of the cy-
clized unsaturated products 9a and 9d (Table 4). Hence
the lowering of the reaction pH provides a general solu-
tion for the production of these types of products. One
limitation was encountered with substrates such as 1b and
7b, which cannot form products 3b or 9b. It was hoped
that lowering of the pH might provide access to products
4b and 10b instead. However, reaction of 1b did not pro-
duce the unsaturated product 4b under all conditions in-
vestigated. The intermediate radical for this type of
phenyl-substituted alkene substrate does not contain a
readily accessible b-hydrogen atom since it resides at a
tertiary carbon, rather than on a primary carbon. Evident-
ly, this hydrogen atom is not reactive towards
cob(II)alamin.
N
H
hydrogen
abstraction
from solvent
β-hydrogen
abstraction from II
by cob(II)alamin
II
N
N
H
H
III
IV
Scheme 1 Proposed catalytic mechanism for the B12-catalyzed
cyclization reaction
According to Scheme 1, the formation of the unsaturated
product IV is dependent on b-hydrogen atom abstraction
by cob(II)alamin. However, the concentration of the latter
at any time is quite low since it is rapidly reduced to
cob(I)alamin by Ti(III) citrate. Thus, it was envisioned
that by slowing this reduction step and using solvents
without readily available hydrogen atoms, the product ra-
tio might be changed to favor product IV. The reduction
potential of Ti(III) citrate has been shown to decrease in a
pH-dependent manner.12,13 A smaller thermodynamic
driving force for reduction of Co(II) to Co(I) at lower pH
could result in a kinetically slower reduction process,
which might make the hydrogen atom abstraction from II
more competitive.
Table 4 Vitamin B12 Catalyzed Cyclization of Ethers 7 at pH 6.0
R
R
R
vitamin B12
O
O
O
Ti(III) citrate
pH 6.0
8
9
Substrate
R
Yield of 8 (%) Yield of 9 (%)
dr
7a
7d
H
Trace
Trace
64
82
4.0:1
3.4:1
Me
In summary, we present a novel cyclization reaction for
substituted arylalkenes that is mild, environmentally
friendly, and can be tuned with respect to the product dis-
tribution by adjusting the pH.14 In comparison with other
strategies, the regiochemistry of the activation of the aryl-
alkene is well-defined, resulting in the formation of a qua-
ternary center, which is not readily achieved by other
means. Efforts to improve the stereoselectivity of the
reaction are underway.
To test this hypothesis, the cyclization of pyrrole 1c was
first performed at pH 8.0 in tert-butanol to minimize for-
mation of product III but this resulted only in a very small
change (ca. 5%) in the product distribution. The reaction
was then repeated at pH 6.0. After 24 hours, essentially
exclusive formation of the unsaturated product 3c was
observed in good yield (Table 3). On the basis of these
results the increased formation of 3c at the expense of 2c
is attributed to more effective hydrogen atom abstraction
from intermediate II by cob(II)alamin at pH 6.0. The
Synlett 2006, No. 2, 211–214 © Thieme Stuttgart · New York

214
C. M. McGinley et al.
LETTER
with hexane (3 × 50 mL). The combined organic layers were
washed with H2O (2 × 75 mL), dried with MgSO4, filtered
and concentrated, yielding 15.9 mg of total product (1:14
mixture of 2d and 3d, 75%) as a colorless oil. 1H NMR (400
MHz, CDCl3): d = 1.07 (s, 3 H), 1.41 (s, 3 H), 1.75 (s, 3 H),
3.12 (t, J = 7.9 Hz, 1 H), 4.01 (m, 2 H), 4.79 (m, 1 H), 4.95
(pentet, J = 1.5 Hz, 1 H), 5.73 (dd, J = 3.6, 1.3 Hz, 1 H), 6.19
(t, J = 2.9 Hz, 1 H), 6.53 (dd, J = 2.6, 1.3 Hz, 1 H). 13C NMR
(100.6 MHz, CDCl3): d = 23.0 (CH3), 23.8 (CH3), 28.6
(CH3), 41.4 (Cq), 48.7 (CH2), 60.5 (CH), 96.7 (CH), 111.5
(CH), 113.1 (CH), 113.6 (CH2), 143.0 (Cq). HRMS (EI):
m/z calcd for C12H17N [M+]: 175.1361; found: 175.1360.
A flask was charged with 7b (29.4 mg, 0.090 mmol) and
vitamin B12 (12.4 mg, 0.009 mmol) and purged with Ar.
Degassed MeCN (20 mL) was added followed by aqueous
Ti(III) citrate pH 8.0 (7 mL). The purple solution was stirred
in the dark for 23 h, then poured into H2O (50 mL) and
extracted with hexane (3 × 50 mL). The combined organic
layers were washed with H2O (2 × 75 mL), dried with
MgSO4, filtered and concentrated yielding 20.0 mg of
product 8b (68%) as a colorless oil. Both diastereomers were
separated by HPLC and the stereochemistry of the major
isomer was determined using NOE after assignment of the
protons by HMQC and COSY spectroscopy. 1H NMR (500
MHz, CDCl3, major diastereomer): d = 1.47 (s, 3 H), 3.63
(m, 2 H), 3.77 (d, J = 8.8 Hz, 1 H), 3.87 (d, J = 7.7 Hz, 1 H),
3.98 (dt, J = 1.8, 7.4 Hz, 1 H), 4.22 (d, J = 8.8 Hz, 1 H),
6.84–6.90 (m, 3 H), 7.05–7.35 (m, 12 H). Minor
Acknowledgment
This research was supported by the Petroleum Research Fund
administered by the American Chemical Society (39921-AC3).
References and Notes
(1) Brown, K. L. Chem. Rev. 2005, 105, 2075.
(2) Shey, J.; McGinley, C. M.; McCauley, K. M.; Dearth, A.;
Young, B.; van der Donk, W. A. J. Org. Chem. 2002, 67,
837.
(3) (a) Forbes, C. L.; Franck, R. W. J. Org. Chem. 1999, 64,
1424. (b) Kleban, M.; Kautz, U.; Greul, J.; Hilgers, P.;
Kugler, R.; Dong, H.-Q.; Jager, V. Synthesis 2000, 1027.
(c) Huang, L.; Chen, Y.; Gao, G.-Y.; Zhang, X. P. J. Org.
Chem. 2003, 68, 8179. (d) Chen, Y.; Zhang, X. P. J. Org.
Chem. 2004, 69, 2431. (e) Shimakoshi, H.; Tokunaga, M.;
Kuroiwa, K.; Kimizuka, N.; Hisaeda, Y. Chem. Commun.
2004, 50.
(4) (a) Zhang, Z. D.; Scheffold, R. Helv. Chim. Acta 1993, 76,
2602. (b) Su, H.; Walder, L.; Zhang, Z. D.; Scheffold, R.
Helv. Chim. Acta 1988, 71, 1073. (c) Fischli, A. Helv.
Chim. Acta 1982, 65, 2697.
(5) Keck, G. E.; McHardy, S. F.; Murry, J. A. J. Org. Chem.
1999, 64, 4465.
(6) Keck, G. E.; Wager, T. T.; Rodriquez, J. F. D. J. Am. Chem.
Soc. 1999, 121, 5176.
(7) Lexa, D.; Savéant, J. J. Chem. Soc., Chem. Commun. 1975,
872.
(8) Waddington, M. D.; Finke, R. G. J. Am. Chem. Soc. 1993,
115, 4629.
(9) Baldwin, D. A.; Betterton, E. A.; Chemaly, S. M.; Pratt, J.
M. J. Chem. Soc., Dalton Trans. 1985, 1613.
(10) Ng, F. T. T.; Rempel, G. L.; Mancuso, C.; Halpern, J.
Organometallics 1990, 9, 2762.
diastereomer: d = 1.57 (s, 3 H), 3.16 (d, J = 11.4 Hz, 1 H),
3.38 (m, 1 H), 3.52 (dd, J = 10.2, 9.7 Hz, 1 H), 3.85 (dd,
J = 8.1, 8.3 Hz, 1 H), 3.91 (d, J = 8.6 Hz, 1 H), 4.21 (d,
J = 8.7 Hz, 1 H), 7.05–7.35 (m, 15 H). 13C NMR (125.6
MHz, CDCl3, major diastereomer): d = 18.7 (CH3), 47.9
(CH2), 55.0 (CH), 77.0 (CH2), 85.0 (CH2), 125.7 (CH), 126.0
(CH), 126.7 (CH), 127.5 (CH), 128.0 (CH), 128.1 (CH),
128.9 (CH), 144.4 (Cq). HRMS (EI): m/z calcd for C24H24O
[M+]: 328.1827; found: 328.1826
(11) Garr, C. D.; Finke, R. G. J. Am. Chem. Soc. 1992, 114,
10440.
A flask was charged with 7d (18.8 mg, 0.093 mmol) and
vitamin B12 (13.2 mg, 0.01 mmol) and purged with Ar.
Degassed MeCN (20 mL) was added, followed by aqueous
Ti(III) citrate pH 6.0 (8 mL). The dark red solution was
stirred in the dark for 24 h, and then the reaction was poured
into H2O (50 mL) and extracted with hexane (3 × 60 mL).
The combined organic layers were washed with H2O (2 × 75
mL), dried with MgSO4, filtered and concentrated yielding
15.4 mg of product 9d (82%) as a colorless oil. 1H NMR
(500 MHz, CDCl3): d = 1.37 (s, 3 H), 1.48 (s, 3 H), 3.11 (t,
J = 8.1 Hz, 1 H), 3.82 (d, J = 8.6 Hz, 1 H), 4.05 (t, J = 8.8
Hz, 1 H), 4.18 (d, J = 8.7 Hz, 1 H), 4.15 (t, J = 8.2 Hz, 1 H),
4.70 (s, 1 H), 4.89 (s, 1 H), 7.25–7.35 (m, 3 H), 7.48–7.50
(m, 2 H). 13C NMR (125.6 MHz, CDCl3): d = 19.8 (CH3),
24.2 (CH3), 57.5 (CH), 71.4 (CH2), 82.7 (CH2), 112.4 (CH2),
126.5 (CH), 127.4 (CH), 128.1 (CH), 128.6 (CH), 146.5
(Cq). HRMS (EI): m/z calcd for C14H18O [M+]: 202.1358;
found: 202.1357.
(12) Holliger, C.; Pierik, A. J.; Reijerse, E. J.; Hagen, W. R. J.
Am. Chem. Soc. 1993, 115, 5651.
(13) Gaertner, P.; Weiss, D. S.; Harms, U.; Thauer, R. K. Eur. J.
Biochem. 1994, 226, 465.
(14) Solvents used for cyclization reactions were deoxygenated
under reduced pressure (10 min) and purged with N2; this
was repeated three times before storage under argon. When
t-BuOH was used as a solvent, it was distilled over CaH2. All
cyclizations were conducted on a 0.1 mmol scale, and
initiated in the dark before covering the flask with aluminum
foil. Representative procedures are provided for a select set
of substrates.
A flask was charged with 1d (20.3 mg, 0.116 mmol) and
vitamin B12 (12.8 mg, 0.009 mmol) and purged with Ar.
Degassed t-BuOH (15 mL) was added, followed by a Ti(III)
citrate solution, pH 6.0 (8 mL) prepared as previously
described.2 The purple solution was stirred in the dark for 24
h. The reaction was poured into H2O (50 mL) and extracted
Synlett 2006, No. 2, 211–214 © Thieme Stuttgart · New York