Photoinduced Vitamin B12-Catalysis for Deprotection of (Allyloxy)arenes

Article abstract of DOI:10.1021/acs.orglett.7b01012

Vitamin B₁₂ is a natural cobalt complex that, while reduced to the "supernucleophilic" Co(I) form, can easily react with electrophiles via an S_N2 mechanism. It is also shown to react via an S_N2′ mechanism with allylic compounds allowing for photochemical deprotection of (allyloxy)arenes. A sustainable alternative to commonly used noble metal-catalyzed deprotection reactions is presented.

Full text of DOI:10.1021/acs.orglett.7b01012

Letter
pubs.acs.org/OrgLett
Photoinduced Vitamin B12-Catalysis for Deprotection of
(Allyloxy)arenes
Maciej Giedyk,‡ Joanna Turkowska,‡ Sandra Lepak, Marcin Marculewicz, Keith o Proinsias,
́
and Dorota Gryko*
Institute of Organic Chemistry Polish Academy of Science, Kasprzaka 44/52, 01-224 Warsaw, Poland
S
* Supporting Information
ABSTRACT: Vitamin B12 is a natural cobalt complex that, while reduced to the
“supernucleophilic” Co(I) form, can easily react with electrophiles via an SN2
mechanism. It is also shown to react via an SN2′ mechanism with allylic compounds
allowing for photochemical deprotection of (allyloxy)arenes. A sustainable
alternative to commonly used noble metal-catalyzed deprotection reactions is
presented.
ature builds and decomposes complex molecules with the
help of enzymes and light as the ultimate energy source.
However, despite enzymes’ fundamental benefits, their
industrial application can be limited by bioavailability, instability
toward certain reaction conditions, or narrow substrate
tolerance. A complementary approach involves bioinspired
methods and in this line vitamin B12 (1, cobalamin) (Figure 1),
central Co(III) cation, which can be reduced to radical Co(II)-
or “supernucleophilic” Co(I)-species 2. Most vitamin B12-
catalyzed reactions involve Co(I)-form 2, as it readily reacts
with electrophiles affording alkylcobalamins, usually via a
typical SN2 mechanism.1a,5 We envisaged that vitamin B12
“supernucleophilic” Co(I)-form 2 should also react with
electrophiles in the SN2′ manner provided that a suitable
partner is present (Scheme 1).
N
Scheme 1. A Hypothesized Nucleophilic Conjugated
Substitution of the Allyl Group
Nucleophilic conjugated substitution is common for allylic
compounds bearing a good living group. Hence, O-allylphenols
should react with reduced vitamin B12 2 generating
allylcobalamin 3 and phenolate that upon protonation furnishes
the desired phenol.2 This concept can have broad ramifications
if applied to the protection−deprotection strategy for the
hydroxyl group, as phenols are common motifs in natural
products and biologically active compounds.6 Currently, the
deprotection of (allyloxy)arenes involves the use of strong
bases, acids, Pd, or other noble metals,7 which renders
development of environmentally benign alternatives highly
desirable.
In order to test our hypothesis, 2-(allyloxy)naphthalene (4a)
was reacted with a catalytic amount of vitamin B12 1a in the
presence of the Zn/NH4Cl reducing system (Scheme 2). The
reaction worked very well, giving 2-naphthol (5a) in 87% yield.
If such a result could be obtained so effortlessly, then
progression into a much greener alternative is a must. Could
Figure 1. Structure of vitamin B12 (1, cobalamin, Cbl).
which in natural systems plays the role of cofactor for adenosyl-
and methylcobalamin-dependent enzymes, creating numerous
opportunities. Though underdeveloped, vitamin B12 catalysis is
particularly appealing, due to the following: (a) a remarkable
stability of this highly functionalized cobalt complex with no
risk of metal leakage; (b) complete nontoxicity and easy
consumption by nature; (c) commercial availability.1
To date, vitamin B12 1 has been exploited in dehalogenation
of polychlorinated compounds, functional group migration,
dimerization of benzyl halides, cyclopropanation and cyclo-
propane ring opening, alkylation of olefins, and synthesis of
esters and amides from (trichloromethyl)benzene, just to name
a few.2−4 These processes rely on the redox properties of the
Received: April 4, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b01012
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 2. Model Deprotection Reaction
Scheme 4. Scope and Limitation Studies
native cyanocobalamin 1a be activated by light e.g. photo-
chemically reduced to Co(I) 2 and as such used in our
deallylation reaction?
Recently, Hisaeda’s group has succeeded in photochemical
reduction of heptamethyl cobyrinate (Cby) and applying it to
dehalogenation reactions,8 but photochemical reduction of
vitamin B12 1a still remains challenging.
In our initial experiments TiO2 was used as the photoredox
mediator, UV irradiation (254 nm) as the light source, and
MeOH as both the solvent and sacrificial electron donor. In the
presence of vitamin B12 1a the deprotection of (allyloxy)-
naphthalene (4a) afforded desired product 2-naphthol (5a) in
59% yield. The reaction proved highly chemoselective; only O-
allyl-ethers were cleaved while O-benzyl- and O-Me remained
intact.
In the next step, the reaction conditions were optimized with
respect to solvent, catalyst, crystalline type of TiO2, and its
loadings. Among all polar solvents tested, only simple alcohols
afforded desired compound 5a, and the highest yield was
obtained for the reaction in MeOH (see Supporting
Information (SI)). Reactions in the presence of pure mineral
TiO2 forms: rutile or anatase gave the same result as a
polymorphic nanopowder mixture; therefore, the latter was
used in subsequent experiments (see SI). The optimal loading
of vitamin B12 1a was found to be 6 mol % (see SI), although
even 2 mol % gave comparable results.
a
Reaction conditions: Method A: (i) (allyloxy)arene (4, 0.5 mmol),
TiO2 (0.5 equiv), (CN)Cbl (1a, 6 mol %), MeOH (2 mL), 20 h, UV
light (254 nm); (ii) HClaq in MeOH (0.05 M, 10 mL), 50 °C, the
reaction was monitored by TLC until completion; Method B:
(allyloxy)arene (4, 0.5 mmol), Zn (6 equiv), NH4Cl (3.4 equiv),
b
(CN)Cbl (1a, 6 mol %), MeOH (2 mL), 20 h, 60 °C. Reaction
conducted on a larger scale: 0.5 g (2.7 mmol). In a 2.7:1 mixture of
c
Moreover, a detailed analysis of the reaction mixture revealed
the presence of a side product (6a), which could be
quantitatively transformed into desired naphthol 5a by treating
it with HClaq in MeOH, at a concentration as low as 0.05 M, for
16 h at 50 °C (Scheme 3).
product 5f and debrominated phenol 5c (76% combined yield).
aliphatic alcohols can be deprotected utilizing our methodology
in a very good yield.
In a subsequent study, the scale-up of our catalytic protocols
was explored. The photochemical reaction of 2-(allyloxy)-
naphthalene (4a) on a scale of 0.5 g (2.7 mmol) gave desired
product 5a in 73% yield, compared to the 81% yield in the test
scale. Similarly, the reaction using the Zn/NH4Cl system
proceeded smoothly with only a 3% difference in the yield
compared to the model reaction.
Scheme 3. Isomerization of 2-Alliloxynaphthalene (4a)
The use of protecting groups, for which deprotection could
be achieved by photochemical means, is an attractive
approach.9 Thus, we considered our newly developed method-
ology as a promising tool in the synthesis of highly
functionalized molecules bearing the phenol unit. This strategy
is especially important in peptide synthesis; therefore, tyrosine
was chosen to examine orthogonality with other protecting
groups.
First, model compounds 7 and 9 bearing variously protected
carboxyl or amino groups respectively were tested under
developed conditions (Scheme 5). In both methods methyl
ester 7a gave product 8a in a good yield while tert-butyl-2-(4-
(allyloxy)phenyl)acetate (7b) proved to be more compatible
with the Zn/NH4Cl Method B system (88% yield). Expectedly,
reducing conditions were incompatible with benzyl ester 8c,
which cleaved rapidly. In the case of an amino group, O-allyl
tyramine derivatives protected with either Ac- (9a) or Boc-
(9b) groups were subjected to our reaction. The photocatalytic
method gave products 10a and 10b in moderate yields, but the
use of Zn/NH4Cl allowed for an increase in the yield up to 69%
At this point, we had two sets of conditions: the first using
the optimized photochemical Method A and the second
involving our initial discovery using Zn/NH4Cl (Method B).
The scope and limitations of both methods were examined
(Scheme 4). Under light irradiation (254 nm), phenols with
electron-withdrawing and -donating substituents reacted
equally well. For example, methoxy-derivative 5g was obtained
in 83% yield and p-hydroxybenzaldehyde (5k) in 73% yield
with no formyl group reduction observed. Gratifyingly,
photochemical Method A assured the formation of an even
more synthetically important product, p-vanillin 5n. Some
(allyloxy)arenes were problematic; e.g., in the deprotection of
(allyloxy)-3-bromo-benzene (4f) the unwanted debromination
process predominated. This problem was diminished when
catalyst 1a was reduced with the Zn/NH4Cl system, resulting in
desired product 5f being favored. Method B was also found to
be superior for salicylic acid derivative 4m, giving desired
product 5m in 70% yield. To our delight, also O-allyl-protected
B
DOI: 10.1021/acs.orglett.7b01012
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
heptamethyl cobyrinate possessing Co in the +2 oxidation state
(Table 1, entry 5), contrary to reactions in which stronger
reducing agents were employed (Table 1, entries 1 and 7). This
observation corroborates the participation of Co(I)-form 2 as
an active species. The reaction in deuterated methanol afforded
products with deuterium exclusively incorporated at the
terminal carbon of the double bond as revealed by NMR
(see SI).
Scheme 5. Deprotection of Functionalized Phenols
Moreover, it is known that the relative rate of cleavage versus
isomerization depends on the substitution on the allyl moiety.7c
On that account, we performed experiments with substituted
allyloxy derivatives 13 and 15 (Scheme 6).
Scheme 6. Deprotection of Substituted
(Allyloxy)naphthalenes
Expectedly, crotyl derivative 13 was inert in both methods. 2-
Methylprop-2-enyl ether (15) furnished predominantly trisub-
stituted vinyl derivative 16 under photochemical conditions,
but in the presence of Zn/NH4Cl deprotection predominated.
It can be rationalized by the accessibility of an acidic proton and
the effect of a stronger reducing agent in the latter case. It has
been firmly established that such conditions promote hydrogen
evolution10 thus lowering the concentration of Co(III)−H
species. As a result, the contribution of the isomerization
pathway is diminished.
The proposed mechanism for the vitamin B12-catalyzed
deallylation is presented in Scheme 7. Based on the
experimental evidences, we assume that, first, vitamin B12 1a
is reduced either photochemically or chemically to the
catalytically active Co(I)-form 2. Then, the nucleophilic
a
Reaction conditions: Methods A and B as described in Scheme 4.
Transesterification occurred; product 8a was isolated. No acid
b
c
treatment was performed.
for substrate 9a and 90% for N-Boc-protected derivative 9b.
Finally, we examined methyl esters of N-Ac- (11a) and N-Boc-
protected (11b) L-tyrosine. In both cases chemical reduction
proved superior, allowing for isolation of compounds 12a and
12b in 75% and 67% yield, respectively. The photochemical
method also gave desired compounds in moderate yields.
In order to establish the mechanism for the deprotection of
(allyloxy)arenes, several experiments were conducted. First, we
proved that each reaction’s component (vitamin B12, TiO2, and
light, or Zn, NH4Cl) is required for the reaction to take place
(Table 1, entries 1−3, 7, and 8). Second, isolated (vinyloxy)-
Scheme 7. Plausible Mechanism for Vitamin B12 Catalyzed
Deallylation
Table 1. Mechanistic Studies
b
entry
substrate
catalyst
hν or t
reductant
5a (%)
1
2
3
4a
4a
4a
4a
6a
4a
4a
1a
−
UV
TiO2
81
UV
TiO2
10
1a
1c
1a
1a
−
−
UV
TiO2
0
a
5
HCO2Na
TiO2
traces
0
6
7
8
UV
60 °C
60 °C
Zn, NH4Cl
Zn, NH4Cl
87
traces
a
Cby(II) 1c and HCO2Na were used to ensure the presence of the
b
Co(II) form in the reaction mixture. Isolated yields.
arene 6a did not decompose to 2-naphtol (5a) when subjected
to standard reaction conditions (entry 6). Hence, this
compound is not an intermediate but rather a byproduct.
Third, the addition of TEMPO, a radical scavenger, did not
allow for detection of carbon radical intermediates, and as a
consequence, a radical mechanism was excluded. Moreover, the
desired phenol 5a did not form in the reaction catalyzed by
C
DOI: 10.1021/acs.orglett.7b01012
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
conjugated substitution yields phenol II and vinylcobalamin 3
possessing a Co−C bond that is easily cleaved under
photochemical or thermal conditions. On the other hand, the
isomerization reaction is believed to involve cobalamin hydride
17, resulting from the reaction of supernucleophilic Co(I)-form
2 with MeOH. Such observations remain in good agreement
with those of Norton, who reported the facile isomerization of
double bonds mediated by cobalt hydride,11 and of Hisaeda and
Shimakoshi, who recently proposed a mechanism of hydrogen
evolution in which vitamin B12 derived cobyrinic acid hydride is
involved.12 The alkylcobalamin III formed undergoes dehy-
drocobaltation generating isomerized product IV and hydrido-
cobalamin (Co(III)-H) 17 that upon reduction regenerates the
catalyst or reacts with the next molecule of (allyloxy)arene I.
In summary, native vitamin B12 1 can be reduced
photochemically to its catalytically active Co(I)-form 2 and as
such catalyzes the cleavage of the C−O bond in allyl-aryl ethers.
The photochemically induced reaction tolerates starting
materials bearing various functional groups, including: −CN,
−CO2Me, −OMe, and even very reactive −CHO. The reaction
seems also promising for deprotection of aliphatic alcohols.
Additionally, for less reactive substrates, the Zn/NH4Cl
reducing system can be used. Newly developed methodology
proved chemoselective; only O-allyl-ethers are cleaved while O-
benzyl- and O-Me remained intact. Compared to known
processes our methods do possess advantages, such as mild
photochemical conditions, the exclusion of harsh acid or bases,
and elimination of precious metal catalysts, e.g. Pd.
REFERENCES
■
(1) For vitamin B12 bioactivity and structure, see reviews:
(a) Krautler, B.; Golding, B. T.; Arigoni, D. Vitamin B12 and B12-
̈
Proteins; Wiley-VCH: 2008. (b) Banerjee, R. Chemistry and
́
Biochemistry of B12; John Wiley&Sons: 1999. (c) o Proinsias, K.;
Giedyk, M.; Gryko, D. Chem. Soc. Rev. 2013, 42, 6605−6619.
(d) Brown, K. L. Chem. Rev. 2005, 105, 2075−2150. (e) Zelder, F.
Chem. Commun. 2015, 51, 14004−14017. (f) Gruber, K.; Puffer, B.;
Krautler, B. Chem. Soc. Rev. 2011, 40, 4346−4363.
̈
(2) Reviews: (a) Giedyk, M.; Goliszewska, K.; Gryko, D. Chem. Soc.
Rev. 2015, 44, 3391−3404. (b) Hisaeda, Y.; Nishioka, T.; Inoue, Y.;
Asada, K.; Hayashi, T. Coord. Chem. Rev. 2000, 198, 21−25.
(c) Scheffold, R.; Abrecht, S.; Orlinski, R.; Ruf, H.-R.; Stamouli, P.;
Tinembart, O.; Walder, L.; Weymuth, C.; Orflnski, R.; Bern, U.; Bern,
C. Pure Appl. Chem. 1987, 59, 363−372. (d) Hisaeda, Y.; Shimakoshi,
H. Handbook of Porphyrin Science; World Scientific Publishing Co.:
2010.
(3) Olefin alkylation: Giedyk, M.; Goliszewska, K.; o
Gryko, D. Chem. Commun. 2016, 52, 1389−1392.
(4) Formation of esters and amides from trichlorinated organic
compounds: (a) Shimakoshi, H.; Hisaeda, Y. Angew. Chem., Int. Ed.
2015, 54, 15439−15443. (b) Shimakoshi, H.; Luo, Z.; Inaba, T.;
Hisaeda, Y. Dalton Trans. 2016, 45, 10173−10180.
(5) (a) Somasundaran, P. Electrochemical Reactions in Microemulsions;
Encyclopedia of Surface and Colloid Science; CRC Press: 2006.
(b) Schrauzer, G. N.; Deutsch, E. J. Am. Chem. Soc. 1969, 91, 3341−
3350. (c) Shey, J.; McGinley, Ch. M.; McCauley, K. M.; Dearth, A. S.;
Young, B. T.; van der Donk, W. A. J. Org. Chem. 2002, 67, 837−846.
(6) Whiting, D. A. Nat. Prod. Rep. 2001, 18, 583−606.
(7) For reviews, see: (a) Weissman, S. A.; Zewge, D. Tetrahedron
2005, 61, 7833−7863. (b) Greene, T. W.; Wuts, P. G. M. Protective
Groups in Organic Synthesis; Wiley, 2006;. (c) Kocienski, P. J. Protecting
Groups; Thieme: 2000.
́
Proinsias, K.;
Mechanistic investigations imply that photochemically
formed vitamin B12 “supernucleophilic” Co(I)-form 2 is an
active form of the catalyst, and as such it reacts with O-
allylphenols in the SN2′ manner. Moreover, our studies strongly
support previous reports on the involvement of a cobalt-
hydride intermediate in the isomerization of the double bond,
though direct evidence has not yet been found.
(8) (a) Shimakoshi, H.; Hisaeda, Y. ChemPlusChem 2017, 82, 18−29.
and ref therein (b) Tahara, K.; Mikuriya, K.; Masuko, T.; Kikuchi, J.;
Hisaeda, Y. Supramol. Chem. 2016, 28, 141−150.
(9) For reviews, see: (a) Rajasekharan Pillai, V. N. Synthesis 1980,
1980, 1−26. (b) Pelliccioli, A. P.; Wirz, J. Photochem. Photobiol. Sci.
̌
́
2002, 1, 441−458. (c) Klan, P.; Solomek, T.; Bochet, Ch. G.; Blanc,
A.; Givens, R.; Rubina, M.; Popik, V.; Kostikov, A.; Wirz, J. Chem. Rev.
2013, 113, 119−191.
ASSOCIATED CONTENT
* Supporting Information
■
S
(10) For reviews, see: (a) Marinescu, S. C.; Winkler, J. R.; Gray, H. B.
Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 15127−15131. (b) Artero, V.;
Chavarot-Kerlidou, M.; Fontecave, M. Angew. Chem., Int. Ed. 2011, 50,
7238−7266.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b01012.
(11) Li, G.; Kuo, J. L.; Han, A.; Abuyuan, J. M.; Young, L. C.;
Norton, J. R.; Palmer, J. H. J. Am. Chem. Soc. 2016, 138, 7698−7704.
(12) Shimakoshi, H.; Hisaeda, Y. ChemPlusChem 2014, 79, 1250−
1253.
1
Experimental procedures, optimization studies, H and
13C NMR spectra for all new compounds (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: dorota.gryko@icho.edu.pl.
ORCID
Dorota Gryko: 0000-0002-5197-4222
Author Contributions
‡These authors contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The National Science Centre, Grants SYMFONIA (DEC
2014/12/W/ST5/00589), SONATA 2013/11/D/ST5/02956;
the Foundation for Polish Science grant (START 31.2016) for
M.G.
D
DOI: 10.1021/acs.orglett.7b01012
Org. Lett. XXXX, XXX, XXX−XXX