Mechanism-based inactivation of coenzyme B12-dependent 2-methyleneglutarate mutase by (Z)-glutaconate and buta-1,3-diene-2,3- dicarboxylate

Article abstract of DOI:10.1002/ejic.200600405

In the presence of holo 2-methyleneglutarate mutase, buta-1,3-diene-2,3- dicarboxylate and (Z)-glutaconate [(Z)-pent-2-ene-1,5-dicarboxylate], but not (E)-glutaconate, each induced homolysis of the Co-C bond of coenzyme B ₁₂ to afford cob(II)alamin and the 5′-deoxyadenosyl radical. The latter probably added to the double bond in (Z)-glutaconate and one of the double bonds in buta-1,3-diene-2,3-dicarboxylate to afford a corresponding "radical adduct". The formation of new radicals and cob(II)alamin was diagnosed by UV/Visible and EPR spectroscopy. (Z)-Glutaconate rapidly inactivated the mutase with formation of aquocobalamin, which was possibly derived by electron transfer from cob(II)alamin to the radical adduct. In contrast, buta-1,3-diene-2,3-dicarboxylate was a much slower inactivator. In this case, the spectroscopic data revealed a relatively stable complex of the radical adduct with cob(II)alamin in the active site of the enzyme. Wiley-VCH Verlag GmbH & Co. KGaA, 2006.

Full text of DOI:10.1002/ejic.200600405

FULL PAPER
DOI: 10.1002/ejic.200600405
Mechanism-Based Inactivation of Coenzyme
B -Dependent 2-Methyleneglutarate Mutase by (Z)-Glutaconate and
12
Buta-1,3-diene-2,3-dicarboxylate
[a]
[a]
[a]
[a]
[b]
Wolfgang Buckel,* Antonio J. Pierik, Sandra Plett, Ashraf Alhapel, Diana Suarez,
[
b]
[b]
Shang-min Tu, and Bernard T. Golding*
Keywords: 2-Methyleneglutarate / Mutase / Coenzyme B12 / Inhibition / Buta-1,3-diene-2,3-dicarboxylic acid /
Glutaconic acid
In the presence of holo 2-methyleneglutarate mutase, buta-
,3-diene-2,3-dicarboxylate and (Z)-glutaconate [(Z)-pent-2-
and EPR spectroscopy. (Z)-Glutaconate rapidly inactivated
the mutase with formation of aquocobalamin, which was
possibly derived by electron transfer from cob(II)alamin to
the radical adduct. In contrast, buta-1,3-diene-2,3-dicarbox-
ylate was a much slower inactivator. In this case, the spectro-
scopic data revealed a relatively stable complex of the radical
adduct with cob(II)alamin in the active site of the enzyme.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
1
ene-1,5-dicarboxylate], but not (E)-glutaconate, each in-
duced homolysis of the Co–C bond of coenzyme B12 to afford
cob(II)alamin and the 5Ј-deoxyadenosyl radical. The latter
probably added to the double bond in (Z)-glutaconate and
one of the double bonds in buta-1,3-diene-2,3-dicarboxylate
to afford a corresponding “radical adduct”. The formation of
new radicals and cob(II)alamin was diagnosed by UV/Visible
Introduction
methylene radical abstracts the 4-Re-hydrogen atom from
2-methyleneglutarate to yield the substrate-derived 2-meth-
Coenzyme B -dependent enzymes comprise eliminases
1
2
ylene-4-ylglutarate radical (1b) and 5Ј-deoxyadenosine.
Cob(II)alamin coupled to an organic radical, possibly 2-
methylene-4-ylglutarate (1b), can be observed by EPR spec-
and mutases, both of which catalyse molecular rearrange-
ments, whereby a functional group undergoes a 1,2-shift en-
abling conversion of substrate to product. In the eliminases,
the migrating group is amino or hydroxy, whereas in the
mutases it is either amino or a carbon-based group. The
[5]
troscopy. It has been proposed that the substrate radical
either fragments into acrylate and the 2-acrylate radical
·
–
(
CH =C HCO ), which recombine to the product-related
2 2
mechanisms of these reactions have been extensively dis-
(R)-3-methylene-itaconate radical (2b), or the latter radical
cussed and reviewed.[
1–3]
For the mutases, experimental and
is derived via an intermediate 1,2-dicarboxycyclopro-
theoretical data have recently led to the proposal that in
the mutases cob(II)alamin assists the group migration by
stabilising intermediate methylene radicals.[3]
[5]
pylmethyl radical (3) (see Scheme 1). Recently, Newcomb
and Miranda have proposed a variant (“polar/heterolytic”)
of fragmentation–recombination in which the intermediates
2-Methyleneglutarate mutase from the strict anaerobic
–
–
are the 2-carbanion of acrylate (CH =C CO ) and the car-
2
2
bacterium Eubacterium barkeri, formerly called Clostridium
barkeri, catalyses the reversible carbon-skeleton rearrange-
ment of 2-methyleneglutarate (1a) to (R)-3-methylitaconate
· [6]
boxyl radical of acrylate (CH =CHCO ). The reaction
2
2
cycle is completed by a H-atom transfer from 5Ј-deoxy-
adenosine to the 3-methylene-itaconate radical (2b) yielding
the product 2a. Finally, the coenzyme is regenerated by
combining cob(II)alamin and the 5Ј-deoxyadenosyl radical,
thereby reforming the Co–C σ-bond. In this paper we re-
port the use of stable analogues of the proposed intermedi-
ate radicals with π bonds at those carbon atoms that during
(2a, 2-methylene-3-methylsuccinate) in the pathway of nico-
[
4]
tinate fermentation. Binding of the substrate 1a to the
enzyme induces homolytic cleavage of the Co–C σ bond in
the prosthetic group coenzyme B12 to afford cob(II)alamin
and the 5Ј-deoxyadenosyl radical. This highly reactive
2
catalysis are converted into the sp centres of radicals. In
[
[
a] Laboratorium für Mikrobiologie, Fachbereich Biologie, Phil- principle, these compounds could either competitively in-
ipps-Universität,
hibit the enzyme or the 5Ј-deoxyadenosyl radical might add
at a double bond to form an adduct radical and cob(II)-
alamin, thus leading to inactive enzyme. We synthesised
3
5032 Marburg, Germany
E-mail: buckel@staff.uni-marburg.de
b] School of Natural Sciences – Chemistry, Bedson Building, New-
castle University,
(Z)-glutaconic acid [4a, (Z)-pent-2-ene-1,5-dicarboxylic
Newcastle upon Tyne, NE1 7RU, UK
E-mail: b.t.golding@ncl.ac.uk
acid] as a potential analogue of the 2-methylene-4-ylglutar-
3622
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 3622–3626

Mechanism-Based Inactivators of 2-Methyleneglutarate Mutase
FULL PAPER
Scheme 1. Proposed reaction pathways for 2-methyleneglutarate mutase. (Path a) Addition-elimination mechanism with the participation
of an intermediate 1-methylene-1,2-cyclopropanedicarboxylate radical 3, for which the (1S,2R)-isomer is shown {AdoCH = 5Ј-deoxyad-
enosyl; AdoCH = 5Ј-deoxyadenosine}. (Path b) Fragmentation–recombination mechanism for 2-methyleneglutarate mutase involving an
acrylate molecule and a 2-acrylate radical.
2
3
ate radical (1b), as well as buta-1,3-diene-2,3-dicarboxylic nickel(0)-induced coupling to afford di-tert-butyl buta-1,3-
acid (5) as an analogue of the (R)-3-methyleneitaconate diene-2,3-dicarboxylate (11), which was deprotected with
radical (2b). The commercially available (E)-glutaconic acid trifluoroacetic acid to give buta-1,3-diene-2,3-dicarboxylic
(
4b) was also studied. Interestingly, compounds 4a and 5 acid (5).
were inactivators of 2-methyleneglutarate mutase, whereas
b had no effect on the enzyme.
4
Synthesis of Inhibitors
The synthesis of (Z)-glutaconic acid (4a, Scheme 2) was
based on the procedure described in ref.[ (Scheme 2). Thus,
laevulinic acid (6) was cleanly dibrominated to 3,5-dibro-
molaevulinic acid (7), which underwent a Favorski re-
arrangement on treatment with potassium hydrogen car-
bonate to give exclusively (Z)-glutaconic acid with no trace
7]
1
of the (E)-isomer 4b according to H NMR analysis.
Scheme 3. Synthesis of buta-1,3-diene-2,3-dicarboxylic acid (5)
reagents and conditions: (i) Br , DCM, Ͻ10 °C; (ii) Et N, diethyl
ether, 0 °C to room temperature; (iii) (cycloocta-1,5-diene) Ni,
Ph P, diethyl ether, 25 °C, 4.5 h; (iv) TFA, room temperature].
[
2
3
2
3
Enzyme Inhibition Studies
Scheme 2. Synthesis of (Z)-glutaconic acid (4a) [reagents: (i) Br
2
,
aq. 48% HBr; (ii) aq. KHCO ].
3
The activity of 2-methyleneglutarate mutase was deter-
mined in a coupled assay using 3-methylitaconate ∆-iso-
[
5]
The synthesis of buta-1,3-diene-2,3-dicarboxylic acid (5, merase as an auxiliary enzyme. The formation of 2,3-di-
Scheme 3) was performed via the corresponding di-tert-bu- methylmaleate was followed spectrophotometrically at λ =
tyl ester, the preparation of which was closely modelled on 256 nm. At first all three compounds were checked for their
[
8]
that of dimethyl buta-1,3-diene-2,3-dicarboxylate. Thus, ability to inhibit or inactivate the isomerase. Whereas the
tert-butyl acrylate (8) was treated with bromine to give tert- glutaconates (4a, 4b) did not affect the enzyme at all, 1 m
butyl 2,3-dibromopropionate (9), which was dehydrobromi- buta-1,3-diene-2,3-dicarboxylate (5) caused complete inacti-
nated to tert-butyl 2-bromoacrylate (10) using triethylamine vation within 3 min under the conditions of the assay.
as base. The tert-butyl 2-bromoacrylate was subjected to Possibly a cysteine residue at the active site of the isomerase
Eur. J. Inorg. Chem. 2006, 3622–3626
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3623

W. Buckel, B. T. Golding et al.
FULL PAPER
added to one of the conjugated double bonds of this sub- 4a to the holo-enzyme. The peak at 535 nm shifted to
strate analogue. This interesting result was not pursued fur- 521 nm and a new peak arose at 495 nm. The same concen-
ther for the time being.
tration of the (E) isomer 4b modified the spectrum in a
Because both glutaconates did not inhibit the isomerase, similar manner but to a lesser extent, and is ascribed to 2%
their action on the mutase could be studied directly in the 4a as an impurity. The new peaks revealed the formation of
coupled assay. It was found that 0.2 m (Z)-glutaconate aquocobalamin, the oxidation product of cob(II)alamin.
(4a) already inactivated the mutase by 60% within 1 min;
after 5 min the inactivation was complete. Furthermore,
whereas 1 m (Z) isomer 4a completely inactivated the mu-
tase within 1 min, 1 m (E) isomer 4b exhibited no effect
during this time. Only with 4 m 4b was there a slight devi-
1
ation from the control observed after the first minute. H
NMR spectroscopy revealed that the commercial 4b used
in this study contained ca. 2% 4a. This was deduced from
the integrated signal at δ = 6.55 ppm (β-H of 4a) as com-
pared to 7.03 ppm (4b). Hence the observed inactivation is
completely specific for the (Z) isomer 4a. In order to show
that inactivation of 2-methyleneglutarate mutase only oc-
curred with the holo-enzyme, coenzyme B -free apo-2-
1
2
methyleneglutarate mutase was incubated for 10 min with
a, which was then removed by gel filtration through Se-
4
phadex G25. Following the addition of coenzyme B12 to the
recovered apo-enzyme, the reconstituted holo-enzyme was
found to be fully active towards 2-methyleneglutarate.
The effect of 5 on the enzyme was assessed with 15 µ Figure 1. Visible spectra of 20 µ reconstituted holo-2-methyl-
eneglutarate mutase (A) in the presence of 25 m 2-methyleneglut-
arate (B) or 10 m buta-1,3-diene-2,3-dicarboxylate (5, C) or
apo-2-methyleneglutarate mutase (subunit m = 67 kDa),
100 µ coenzyme B , 2 m dithiothreitol and 1 m 5 in
12
1
0 m (Z)-glutaconate (4a, D). The addition of the dicarboxylate
100 m potassium phosphate (pH 7.4) which were incu-
caused 10% dilution. The spectra were run 6 min after the ad-
bated for 30 min at 25 °C. Afterwards, the inhibitor was re- ditions.
moved by gel filtration and the activity of the recovered
enzyme was measured in the coupled assay. The results
showed that incubation of the holo-enzyme with 5 de-
creased the activity of the enzyme by 45± 2% as compared
to that of a control without 5. We also studied the mutase
under steady-state conditions and found that in the pres-
ence of excess of coenzyme B12 and 10 m 2-methyleneglut-
arate the apo-mutase lost 80% of its activity after 20 min
(50% after 10 min) or about 6 million turnovers; within the
next 20 min the decrease was much slower, leading to 15%
residual activity. In the absence of either coenzyme B12 or
substrate the enzyme remained stable during this time.
UV/Visible spectroscopy of the reconstituted (holo) 2-
methyleneglutarate mutase, from which the excess of coen-
zyme B12 had been removed by gel filtration, exhibited a
broad peak at 535 nm and a shoulder at around 570 nm
(Figure 1). Upon addition of 25 m 2-methyleneglutarate
the absorbance of both features decreased, and the peak
shifted by about 5 nm to 530 nm. A new peak arose at
4
70 nm indicating the formation of cob(II)alamin. These
Figure 2. EPR spectra of 270 µ 2-methyleneglutarate mutase incu-
spectral changes were already observed after 1 min; incu-
bation for a further 5 min revealed only a slight increase in
the intensities of the new peaks. Addition of 10 m 5 to
holo-mutase resulted in the same changes but with a two- or buta-1,3-diene-2,3-dicarboxylate (5, E). EPR conditions: tem-
fold increase in intensity. An almost identical spectrum was perature, 40 K for A, 77 K for B–E; microwave power 2 mW (A),
bated in the presence of 1.6 m coenzyme B12 and 40 m of 2-
2
methyleneglutarate (A) or acrylate (B) or [2,3,3- H
3
]acrylate (B,
sharper trace) or (E)-glutaconate (4b, C) or (Z)-glutaconate (4a, D)
1.3 mW (B), 10 mW for C–E; microwave frequency 9463± 1 MHz
obtained after 30 min incubation of the apoenzyme with
m 5 and an excess of coenzyme B , followed by removal
(
1
A) or 9435± 2 MHz (B–E); modulation amplitude 0.5 mT (A),
.25 mT (B–E). Spectra were normalised to identical amplitudes,
except for (E)-glutaconate. Thus, the spectra of the glutaconates
ferent spectral changes were observed by addition of 10 m can be directly compared.
1
1
2
of the unbound small molecules through gel filtration. Dif-
3624
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 3622–3626

Mechanism-Based Inactivators of 2-Methyleneglutarate Mutase
FULL PAPER
EPR spectra were obtained by mixing 270 µ apo-meth- spectra, however, which were taken after 1 and 6 min (Fig-
yleneglutarate mutase (final concentration; 18 mg protein ure 1), only aquocob(III)alamin rather than cob(II)alamin
–
1
mL ) in 50 m Tris/HCl (pH 7.5) and 75 m NaCl with could be seen. This could be due to an electron transfer
.6 m coenzyme B , and 40 m substrate or inactivator, from cob(II)alamin to the initially derived adduct radical
1
1
2
followed by freezing in liquid pentane/N within 30 s. The from 4a and the 5Ј-deoxyadenosyl radical.
2
results are shown in Figure 2. The spectra revealed that 2-
methyleneglutarate, as well as the inactivators 4a and 5, led
to the formation of stable enzyme-bound organic radicals,
which interact with cob(II)alamin.[ In the case of 4a the
concentration dependence of the EPR signal was studied.
Saturation was observed with an apparent Ki of
5]
0
.9± 0.1 m. The determination of the nature of the radi-
2
13
cals (see below) has to await the syntheses of H- and C-
labelled inactivators. Interestingly, acrylate also induced the
formation of radicals even though it was recently shown
not to act as an inhibitor.[ This result was confirmed with
perdeuterated acrylate, which sharpened the signal.
5]
The adduct radical is possibly 13, arising from addition
of the 5Ј-deoxyadenosyl radical to C-3 of 4a. Similar elec-
tron transfers have been observed from cob(II)alamin to
Discussion
[10]
substrate-derived radicals in β-lysine 5,6-aminomutase
In this paper we describe two mechanism-based inacti-
vators of 2-methyleneglutarate mutase. Buta-1,3-diene-2,3-
dicarboxylate (5) can be regarded as a close analogue of the
product (R)-3-methylitaconate (2a): just the methyl group
is converted to a second methylene group. Unfortunately we
were not able to perform steady-state kinetics with 5, be-
cause it reacted rapidly and irreversibly with the auxiliary
enzyme, the isomerase. However, incubation of 5 with the
mutase resulted in partial inactivation, formation of oxy-
gen-insensitive cob(II)alamin and induction of an organic
radical interacting with cob(II)alamin. Interestingly the
[
11]
and in variants of methylmalonyl-CoA mutase.
Experimental Section
Enzymology: The preparation and assay of 2-methyleneglutarate
mutase and 3-methylitaconate ∆-isomerase, as well as the method
[
5]
of recording the EPR spectra, have been described previously. In
the standard assay, 2-methyleneglutarate mutase activity (0.05–0.1
unit) was measured with 10 m 2-methyleneglutarate as substrate
in 100 m potassium phosphate (pH 7.4, 25 °C) using 3-methylita-
conate isomerase (10–30 units/mL) as auxiliary enzyme. The 2,3-
same features were observed with the substrate 2-methyl- dimethylmaleate formed was measured by its UV absorbance (ε
2
56
–
1
–1
eneglutarate (1a) itself; only the EPR spectrum was some- = 0.66 m cm ). For the inactivation studies (E)- and (Z)-gluta-
what different. Hence 5 behaved superficially like 2-methyl- conate and buta-1,3-diene-2,3-dicarboxylate were dissolved in
aqueous KOH to give 0.1  solutions (pH ca. 7). In order to re-
eneglutarate. However, whereas the 5Ј-deoxyadenosyl radi-
move small molecules from the enzymes, PD-10-desalting columns
cal abstracts a hydrogen atom from C-4 of 2-methyleneglut-
arate, with 5 the 5Ј-deoxyadenosyl radical most likely added
at one of its methylene groups leading to an allyl radical
containing Sephadex® G-25 Medium from GE Healthcare (Amer-
sham Biosciences) were used. Concentrations (m or µ) given in
this paper are the final concentration for a particular component
in the reaction mixture.
12. A similar proposal has been made for the inactivation of
[
9]
glutamate mutase induced by 2-methyleneglutarate. The
partial inactivation of 2-methyleneglutarate mutase by 5
suggests that the proposed radical addition may be revers-
ible. The substrate analogue 5 may fit sufficiently well into
the enzyme active site that the adduct produced by the in-
teraction of 5 with the 5Ј-deoxyadenosyl radical protects
cob(II)alamin from interaction with oxygen.
3
,5-Dibromolaevulinic Acid (7): Bromine (13.7 g, 86 mmol) was
added dropwise to a mixture of laevulinic acid (5.0 g, 43 mmol)
and 48% aqueous HBr (4.3 mL, 38 mmol) at 0 °C over 30 min. The
mixture was stirred at 0 °C for 4 h. The excess of bromine was
removed using a nitrogen flow. The resulting solution was stirred
overnight at 0 °C to give a solid, which was taken up in ethyl ace-
tate (40 mL) and washed with saturated aqueous sodium metabi-
4
The inactivator 4a is not so closely related to the sub- sulfite (3×25 mL). After drying (MgSO ), the solvent was removed
strate 2-methyleneglutarate because it lacks the exo-methyl- to afford crude product as a pale yellow solid. Recrystallisation
from ethyl acetate/petroleum ether afforded the title compound as
a white solid (7.83 g, 66% yield). M.p. 110–112 °C (lit. m.p. 111–
ene group. However, it apparently fits well into the active
[7]
site, possibly because the distance between the two carbox-
ylates is similar to that in the enzyme-bound conformation
of (R)-3-methylitaconate. This could be the explanation for
the inability of the (E) isomer 4b to inactivate the enzyme.
The (Z) isomer reacts rapidly with the mutase as shown by
activity measurements and EPR spectroscopy. Already 60 s
1
1
1
13). H NMR (300 MHz, CDCl
7.6 Hz, 1 H, CHHCO H), 3.29 (dd, J = 8.1 and 17.4 Hz, 1 H,
H), 4.07 (d, J = 13.2 Hz, 1 H, CHHBr), 4.28 (d, J =
2.9 Hz, 1 H, CHHBr), 4.94 (br. t, J = 6.0 Hz, 1 H, CHBr) ppm.
3
): δ = 2.97 (dd, J = 5.6 and
2
CHHCO
2
1
13
C NMR (75 MHz, CDCl
), 174.5 (C-1), 194.4 (C-4) ppm. C
3
): δ = 30.9 (C-2), 38.4 (C-5), 41.1 (C-
Br : calcd. C 21.93, H
3
5
H
6
O
3
2
after mixing under anoxic conditions an organic radical in- 2.21, found C 22.02, H 2.27. IR (KBr disc): ν˜ = 3000 (m), 1724 (s),
–
1
teracting with cob(II)alamin was detected. In the visible 1700 (s), 611 (m) cm .
Eur. J. Inorg. Chem. 2006, 3622–3626
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3625

W. Buckel, B. T. Golding et al.
FULL PAPER
(
(
2
Z)-Glutaconic Acid (4a): KHCO
210 mL) was added to 3,5-dibromolaevulinic acid (5.5 g,
1 mmol). The mixture was vigorously stirred for 19 h. When con-
3
(17.0 g, 169 mmol) in water
taining fractions (R
collected and the solvent was removed to afford the title compound
as a colourless oil (0.48 g, 47% yield). H NMR (300 MHz,
f
= 0.36 on silica, same solvent system) were
1
stant titration values against methyl orange were obtained, the re-
action was quenched by acidification (aq. HCl) and extracted with
diethyl ether (10×30 mL). After drying the combined organic ex-
CDCl
CHH), 6.06 (d, J = 1.7 Hz, 2 H, CHH) ppm. C NMR (75 MHz,
CDCl ): δ = 28.4 [C(CH ], 81.3 [C(CH ], 116.6 (CH ), 126.0 (C-
2, C-3), 165.3 (C-1) ppm.
3 3 3
): δ = 1.42 [s, 18 H, C(CH ) ], 5.60 (d, J = 1.7 Hz, 2 H,
1
3
3
3
)
3
3
)
3
2
2 4
tracts (Na SO ) the solvent was removed. The resulting yellow solid
was recrystallised from acetonitrile (decolourisation by charcoal)
to give the title compound as a white solid (0.21 g, 8% yield). M.p.
Buta-1,3-diene-2,3-dicarboxylic Acid (5): Trifluoroacetic acid (TFA)
1.0 mL) was added to di-tert-butyl buta-1,3-diene-2,3-dicarboxyl-
(
[
7]
[12]
136–136.5). 1H NMR
1
(
30–132 °C (lit. m.p. 136–136.5, lit.
300 MHz, CD OD): δ = 3.71 (dd, J = 3.0 and 7.0 Hz, 2 H,
H), 5.93 (dt, J = 2.0 and 11.5 Hz, 1 H, CH), 6.52 (dt, J =
ate (0.16 g, 0.63 mmol) in DCM (2 mL). The resulting solution was
stirred at room temperature for 20 min. The solvent was removed
and the title compound was obtained as a white solid (0.08 g, 90%
yield). Recrystallisation from acetonitrile gave an analytically pure
3
2 2
CH CO
13
7
.0 and 11.6 Hz, 1 H, CH) ppm. C NMR [δ
δ = 35.2 (C-4), 123.5 (C-3), 142.2 (C-2), 169.5 (C-1), 174.7 (C-5)
ppm. C : calcd. C 46.16, H 4.65; found C 45.99, H 4.62. IR
KBr disc): ν˜ = 3066 (m), 1686 (s), 1649 (s) cm .
C 3
(75 MHz, CD OD)]:
sample. M.p. 184–186 °C (lit.[ m.p. 186–188, lit. 192–192.5). H
13]
[8]
1
5
6 4
H O
NMR (300 MHz, CD
.22 (d, J = 1.5 Hz, 2 H, CHH) ppm. 13C NMR (75 MHz,
CD OD): δ = 128.1 (CH ), 140.1 (C-2, C-3), 167.1 (C-1) ppm. IR
3
OD): δ = 5.85 (d, J = 1.5 Hz, 2 H, CHH),
–1
(
6
tert-Butyl 2,3-Dibromopropionate (9): Bromine (2.9  ) in DCM
10 mL) was added dropwise to tert-butyl acrylate (2.0 g,
5.6 mmol) in DCM (20 mL), cooled below 10 °C over 20 min. The
3
2
–
1
(
1
(KBr disc): ν˜ = 3066 (m), 1675 (s), 1614 (s) cm .
mixture was stirred at ca. 5 °C for 2 h at which stage an orange
Acknowledgments
colour persisted. The solvent was removed to afford the title com-
1
pound as a colourless oil (4.41 g, 90% yield). H NMR (300 MHz,
We are grateful to the European Commission, Deutsche For-
schungsgemeinschaft, and Fonds der Chemischen Industrie for
funding this work.
CDCl
H, CHHBr), 3.81 (dd, J = 9.8 and 11.3 Hz, 1 H, CHHBr), 4.27
dd, J = 4.4 and 11.3 Hz, 1 H, CHBr
) ppm. 13C NMR (75 MHz,
CDCl ): 28.1 [C(CH ], 41.0 (C-3), 43.3 (C-2), 83.9
C(CH ], 166.8 (C-1) ppm.
3 3 3
): δ = 1.43 [s, 9 H, C(CH ) ], 3.57 (dd, J = 4.4 and 9.8 Hz 1
(
2
3
δ
=
3 3
)
[
3 3
)
[1] R. Banerjee, Chemistry and Biochemistry of B12, John Wiley &
Sons Inc., New York, 1999.
tert-Butyl 2-Bromoacrylate (10): Triethylamine (0.77 g, 0.80 mL,
[
[
2] K. L. Brown, Chem. Rev. 2005, 105, 2075–2149.
3] W. Buckel, C. Kratky, B. T. Golding, Chem. Eur. J. 2006, 12,
7
2
.7 mmol) in diethyl ether (5 mL) was added dropwise to tert-butyl
,3-dibromopropionate (3.0 g, 7.0 mmol) in dry diethyl ether
352–362.
(
15 mL), cooled to 0 °C . The mixture was stirred overnight at room
temperature and then filtered. The filtrate was washed with water
3×10 mL), dried (Na SO ) and the solvent was removed. The title
[
4] W. Buckel, G. Bröker, H. Bothe, A. J. Pierik, B. T. Golding, in
Chemistry and Biochemistry of B12, John Wiley & Sons, Inc.,
New York, 1999, p. 757.
[5] A. J. Pierik, D. Ciceri, R. F. Lopez, F. Kroll, G. Bröker, B. Be-
atrix, W. Buckel, B. T. Golding, Biochemistry 2005, 44, 10541–
10551.
(
2
4
compound was obtained as a pale yellow oil for which no further
1
purification was needed (1.72 g, 80% yield). H NMR (300 MHz,
CDCl
.77 (d, J = 1.5 Hz, 1 H, CHH) ppm. C NMR (75 MHz, CDCl
δ = 28.3 [C(CH ], 83.4 [C(CH ], 123.8 (C-3), 129.2 (C-2), 161.1
C-1) ppm.
3 3 3
): δ = 1.45 [s, 9 H, C(CH ) ], 6.13 (d, J = 1.5 Hz, 1 H, CHH),
13
[6] M. Newcomb, N. Miranda, J. Am. Chem. Soc. 2003, 125, 4080–
6
3
):
4086.
3
)
3
3 3
)
[
[
7] C. Rappe, R. Adeström, Acta Chem. Scand. 1965, 19, 383–389.
8] M. F. Semmelhack, P. Helquist, L. D. Jones, L. Keller, L. Men-
delson, L. S. Ryono, J. G. Smith, R. D. Stauffer, J. Am. Chem.
Soc. 1981, 103, 6460–6471.
(
Di-tert-butyl Buta-1,3-diene-2,3-dicarboxylate (11): Bis(1,5-cyclooc-
tadiene)nickel(0) (1.5 g, 5.3 mmol) and triphenylphosphane (2.8 g,
[
9] M. S. Huhta, D. Ciceri, B. T. Golding, E. N. G. Marsh, Bio-
chemistry 2002, 41, 3200–3206.
10.6 mmol) were suspended in dry diethyl ether (20 mL) at 25 °C
under nitrogen in a flame-dried Schlenk tube. Neat tert-butyl 2-
bromoacrylate (1.87 g, 8.9 mmol) was added rapidly via syringe
and the reaction mixture was stirred at 25 °C for 4.5 h. The mixture
was diluted with diethyl ether (20 mL), washed with water
[
10] K. H. Tang, C. H. Chang, P. A. Frey, Biochemistry 2001, 40,
5190–5199.
[
[
[
11] M. D. Vlasie, R. Banerjee, Biochemistry 2004, 43, 8410–8417.
12] R. Malachowski, Ber. Dtsch. Chem. Ges. 1929, 62, 1323–1326.
13] S. G. Deshpande, N. P. Argade, Synthesis 1999, 8, 1306–1308.
Received: May 3, 2006
(
2×25 mL), dried (MgSO
crude product was purified by medium pressure chromatography
silica, elution with 5% ethyl acetate in petrol). The product con-
4
) and the solvent was removed. The
(
Published Online: August 2, 2006
3626
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 3622–3626