Electronic substituent effects on the cleavage specificity of a non-heme Fe2+-dependent β-diketone dioxygenase and their mechanistic implications

Article abstract of DOI:10.1021/ja0460918

Acinetobacter johnsonii acetylacetone dioxygenase (Dke1) is a non-heme Fe(II)-dependent dioxygenase that cleaves C-C bonds in various β-dicarbonyl compounds capable of undergoing enolization to a cis-β-keto enol structure. Results from ¹⁸O labeling experiments and quantitative structure-reactivity relationship analysis of electronic substituent effects on the substrate cleavage specificity of Dke1 are used to distinguish between two principle chemical mechanisms of reaction: one involving a 1,2-dioxetane intermediate and another proceeding via Criegee rearrangement. Oxygenative cleavage of asymmetrically substituted β-dicarbonyl substrates occurs at the bond adjacent to the most electron-deficient carbonyl carbon. Replacement of the acetyl group in 1-phenyl-1,3-butanedione by a trifluoro-acetyl group leads to a complete reversal of cleavage frequency from 83% to only 8% fission of the bond next to the benzoyl moiety. The structure-activity correlation for Dke1 strongly suggests that enzymatic bond cleavage takes place via nucleophilic attack to generate a dioxetane, which then decomposes into the carboxylate and α-keto-aldehyde products. Copyright

Full text of DOI:10.1021/ja0460918

Published on Web 09/10/2004
Electronic Substituent Effects on the Cleavage Specificity of a Non-Heme
Fe2+-Dependent â-Diketone Dioxygenase and Their Mechanistic Implications
Grit D. Straganz,*,† Hannes Hofer,† Walter Steiner,† and Bernd Nidetzky*,†,‡
Institute of Biotechnology and Biochemical Engineering and Research Centre Applied Biocatalysis,
Graz UniVersity of Technology, Petersgasse 12, A-8010 Graz, Austria
Received July 1, 2004; E-mail: grit.straganz@TUGraz.at; bernd.nidetzky@TUGraz.at
Scheme 1. Three Routes of C-C Bond Cleavage by Dke1a
The enzymatic fission of C-C bonds by the incorporation of
one atom of molecular oxygen into each site of bond cleavage is a
key step of specialized pathways through which microorganisms
assimilate xenobiotic and environmentally toxic compounds. It is
an intriguing type of reaction that has no immediate counterpart in
organic chemistry and is thus of a fundamental interest. In nature,
it is generally performed by C-C bond-cleaving dioxygenases that
require a non-heme metal cofactor for activity (NHMCDs). Among
the NHMCDs, intradiol and extradiol-cleaving catechol dioxygen-
ases have been studied in great detail.1 By contrast, mechanistic
studies of the cleavage of aliphatic C-C bonds are sparse,2 limiting
the current view of the chemistry employed by NHMCDs to the
conversion of aromatic substrates.
Acinetobacter johnsonii acetylacetone dioxygenase (Dke1)3 is a
representative Fe(II)-dependent NHMCD that acts on aliphatic
substrates. It catalyzes oxidative degradation of 2,4-pentanedione
(Scheme 1) and related â-dicarbonyl compounds capable of
undergoing enolization to a cis-â-keto-enol structure. Here, we
report a quantitative structure-activity relationship analysis of
electronic substituent effects on the cleavage specificity of Dke1,
which is a novel approach to explore the mechanisms of NHMCDs.
The results, together with evidence from 18O labeling experiments,
support a nucleophilic mechanism of C-C bond cleavage involving
a dioxetane intermediate (see Scheme 1c). These findings are
relevant in consideration of the mechanistic proposals for extradiol-
a Paths a and b show rearrangement of a common peroxide intermediate
and intradiol-cleaving catechol dioxygenases that invoke Criegee
(IV) via acyl group migration (solid arrow) and hydrogen migration (dashed
rearrangement (Scheme 1a).1,4
arrow), respectively, followed by attack of Fe2+-bound OH- (which may
partially equilibrate with solvent) to yield products. Path c shows nucleo-
philic attack of the peroxidate to generate a dioxetane that decomposes to
the products. Mechanisms a-c will differ in substituent effects on the
cleavage pattern. *For 1-phenyl-1,3-butanedione, cleavage is expected
adjacent to the benzoyl moiety in all three routes. **Labeling in (VIIb)
may be partial.
The Dke1-catalyzed conversion of acetylacetone (2,4-pentanedi-
one) was carried out under an 18O2 atmosphere, and the incorpora-
tion of 18O label into the products was monitored by solid-phase
microextraction GC-MS (see the Supporting Information for
experimental setup and detailed results). We determined 97 ( 2%
and 70 ( 5% incorporation of one 18O into acetate and methyl-
glyoxal, respectively, and found no double labeling of either
product. Incomplete 18O incorporation of methylglyoxal reflects the
kinetic competition between the enzymatic reaction and the loss
of label to water solvent due to aldehyde group hydratation (see
Supporting Information). The observed pattern of 18O labeling is
fully consistent with a dioxygenase reaction in which one atom
from molecular oxygen is incorporated into each site of C-C bond
cleavage. Therefore, the plausible mechanism of Dke1 involves a
C-3 peroxide (IV),5 and this central intermediate may undergo
conversion to products via three different routes, as outlined in
Scheme 1: Criegee rearrangement (a), hydrogen migration (b), and
formation and subsequent cleavage of a dioxetane intermediate (c).
Scheme 1 shows that the completeness of incorporation of 18O
into acetate is of a mechanistic relevance. Partial exchange of label
with solvent is not consistent with a dioxetane intermediate but
may accompany Criegee rearrangement or hydrogen migration.6
To confirm the results of 18O incorporation, we carried out a reverse
labeling experiment in H218O as solvent and using unlabeled O2.
No 18O was incorporated into acetate from the C-C bond cleavage
(see Supporting Information for detailed results). This is in line
with all three mechanistic routes in Scheme 1.
To differentiate between the intermediacy of a dioxetane or a
rearrangement product, we investigated the cleavage pattern for the
Dke1-catalyzed degradation of a homologous series of asymmetric
â-carbonyl substrates in which the electronic properties of R1 and
R2 (Scheme 1) were different and systematically varied. Considering
that the conversion of (I) is quasi-irreversible, the observed product
pattern immediately mirrors the ratio of cleavage velocities at the
two candidate C-C bonds and is not attenuated by substituent
effects on the thermodynamic equilibrium constant of the reaction.
The broad substrate tolerance of Dke1 allowed us to analyze
electronic substituent effects on cleavage specificity without adding
a strong steric bias. Introduction of electron-withdrawing fluorine
† Institute of Biotechnology and Biochemical Engineering.
‡ Research Centre Applied Biocatalysis.
9
12202
J. AM. CHEM. SOC. 2004, 126, 12202-12203
10.1021/ja0460918 CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
What, however, cannot be discounted on the basis of the available
evidence, is a rearrangement that involves migration of a hydrogen
and subsequent cleavage of the resulting tricarbonyl intermediate
in the active site (Figure 1b). By analogy, rearrangement of C-3
substituted substrates such as 3-methyl- (MPD) and 3-phenyl-
pentane-2,4-dione (PPD) might take place via migration of a methyl
and phenyl group, respectively, in which case the expected
formation of the methyl- and phenyl-ester of acetate would be
diagnostic. Alternatively, if exchange of an iron-bound alkoxide
with bulk water occurred (see Scheme 1b in which OR- would
replace OH-), methanol and phenol would be expected as products.
Within limits of detection (g0.1 mol % of PPD consumed), the
Dke1-catalyzed conversion of PPD did not yield an ester or alcohol
product, essentially eliminating the possibility that migration of a
phenyl cation occurs. Therefore, this supports the notion that
cleavage of C-C bonds by Dke1 does not occur via migration of
the C-3 substituent. Consequently, we propose that C-C bond
cleavage in Dke1 proceeds via nucleophilic attack of the carbonyl
group by the peroxidate moiety leading to a dioxetane intermediate,
which then undergoes cleavage to acetate and methylglyoxal
(Scheme 1c).10
Figure 1. HPLC chromatogram of the products of Dke1-catalyzed
conversion of 4,4,4-trifluoro-1-phenyl-1,3-butanedione (blue solid line) and
1-phenyl-1,3-butanedione (red dashed line). Inset: Correlation of logarithmic
cleavage ratios of methylgyloxal (c2) and acetate (c1) or, in the case of
benzoylic substrates (R1 ) C6H6), phenylglyoxal (c2) and benzoate (c1),
which are formed by cleavage adjacent to R2 and R1, respectively, and the
corresponding ∆σ* values.7 Substrates investigated were: 1-phenyl-1,3-
butanedione (PB), 4,4-difluoro-1-phenyl-1,3-butanedione (DFPB), 4,4,4-
trifluoro-1-phenyl-1,3-butanedione (TFPB), 2,4-pentanedione (PD), 1,1-
difluoro-2,4-pentanedione (DFPD), and 1,1,1-trifluoro-2,4-pentanedione
(TFPD).
Supporting Information Available: Experimental setup and results
of the 18O isotope incorporation experiments, detailed results of the
substrate cleavage pattern, a pictorial representation of the calculated
HOMO for acetylacetonate, and a stereoelectronic analysis of a Criegee
intermediate for the reaction with DFPD. This material is available
free of charge via the Internet at http://pubs.acs.org.
atoms into R2 should strongly enforce the susceptibility of the
adjacent carbonyl group for nucleophilic attack by the distal oxygen
of the C-3 peroxide anion, thus favoring bond cleavage next to the
substituted acetyl group in the case of a dioxetane intermediate. In
the Criegee mechanism, by contrast, those moieties that can best
stabilize a positive charge migrate preferentially, implying a low
migratory aptitude for a fluorinated acetyl group compared to the
nonfluorinated counterpart. The selected acetylacetone derivatives
were incubated in a concentration range of 0.2-3.0 mM with 20
µM purified Dke1 at 25 °C until all substrate was converted (2-4
h). The cleavage ratio was obtained from the product concentrations
measured by HPLC (Figure 1).
References
(1) (a) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L., Jr. Chem. ReV. 2004,
104, 939-986. (b) Solomon, E. I.; Brunold, T. C.; Davis, M. I.; Kemsley,
J. N.; Lee, S. K.; Lehnert, N.; Neese, F.; Skulan, A. J.; Yang, Y. S.; Zhou,
J. Chem. ReV. 2000, 100, 235-349. (c) Bugg, T. D. H. Tetrahedron 2003,
59, 7075-7101. (d) Groce, S. L.; Lipscomb, J. D. J. Am. Chem. Soc.
2003, 125, 11780-11781.
(2) (a) Dai, Y.; Pochapsky, T. C.; Abeles, R. H. Biochemistry 2001, 40, 6379-
6387. (b) Steiner, R. A.; Kalk, K. H.; Dijkstra, B. W. Proc. Natl. Acad.
Sci. U.S.A. 2002, 99, 16625-16630. (c) Leuenberger, M. G.; Engeloch-
Jarret, C.; Woggon, W. D. Angew. Chem., Int. Ed. 2001, 16, 2613-2617.
(3) Straganz, G. D.; Glieder, A.; Brecker, L.; Ribbons, D. W.; Steiner, W.
We observe a marked preference for enzymatic cleavage of the
bond adjacent to the most electron-deficient carbonyl carbon. The
electronic substituent effect is strong, as emphasized by the full
reversal of the product ratio during enzymatic conversion of
benzoylic substrates upon changing R2 from methyl (0.2) to
trifluoromethyl (12). The pronounced correlation between C-C
bond cleavage and the TAFT factor (σ*)7 of the adjacent substit-
uents strongly supports a mechanism of C-C bond fission via
nucleophilic attack at carbonyl carbon. A Criegee mechanism of
Dke1 governed by electronic migratory aptitudes is not consistent
with the observed correlation in Figure 1. A primary stereoelectronic
effect that has been shown to control Criegee rearrangement in
sterically rigid systems8 will not attenuate the clearcut mechanistic
distinction for Dke1, because conformational restrictions are clearly
lacking in the â-dicarbonyl substrates used here. With the widely
held assumption that the group that migrates does so from a position
antiperiplanar to the oxygen-oxygen bond of the peroxide (IVa),
the stereoelectronic analysis of the 1,1-difluoroacetylacetone-
peroxide does not reveal conflicts in dipole orientation for a
conformer in which the electronically favored CH3CdO group is
properly aligned for migration. Finally, we rule out that the cleavage
pattern reflects a substituent effect on the enol-enol equilibrium
of â-dicarbonyl compounds. Compared to acetylacetone, this
equilibrium changes by a factor of maximally 1.5 in response to
the electronic properties of the substituents investigated.9
Biochem. J. 2003, 369, 573-581.
(4) Bugg, T. D. H.; Linn, G. Chem. Commun. 2001, 11, 941-952.
(5) Me2+-dependent NHMCDs are thought to activate molecular oxygen
through electron transfer from the deprotonated, metal bound substrate.1,2
We perform Hartee-Fock calculations of the electronic state of monoan-
ionic acetylacetone and find that the HOMO is centered at C-3 and oxygen.
Nucleophilic attack by C-3 of the substrate would directly lead to the
peroxide intermediate. Single electron transfer from the substrate via the
metal yields a conjugated radical that is located both on C-3 and the two
oxygen atoms. The resulting activated oxygen species will subsequently
add to C-3 (see (II) and (III), Scheme 1). Therefore, irrespective of the
exact mechanism of oxygen activation in Dke1, the same central peroxide
intermediate will occur and is the point of departure for our mechanistic
considerations.
(6) (a) Sanvoisin, J.; Langley, G. J.; Bugg, T. D. H. J. Am. Chem. Soc. 1995,
117, 7836-7837. (b) Mayer, R. J.; Que, L., Jr. J. Biol. Chem. 1984, 259,
13056-13060.
(7) Hansch, C.; Leo, A.; Hoekman, D. H. In Exploring Qsar: Hydrophobic,
Electronic, & Steric Constants; Heller, S. R., Ed.; American Chemical
Society: Washington, DC, 1995.
(8) (a) Crudden, C. M.; Chen, A. C.; Calhoun, L. A. Angew. Chem., Int. Ed.
2000, 39, 2851-2855. (b) Goodman, R. M.; Kishi, Y. J. Am. Chem. Soc.
1998, 120, 9392-9393.
(9) Geraldes, C. F. G. C.; Barros, M. T.; Maycock, C. D.; Silva, M. I. J.
Mol. Struct. 1990, 238, 335-346.
(10) Considering that conversion of a dioxetane ring is a highly exothermic
reaction, we looked at the occurrence of luminescence during the Dke1-
catalyzed cleavage of pentanedione at 25 °C and pH 7.5. We did not
observe any, even at high enzyme concentrations of 200 µM. However,
we emphasize that the nature of the rate-limiting step in the reaction of
Dke1 is not known and the fraction of enzyme-bound dioxetane intermedi-
ate may be too small to allow detection of luminescence.
JA0460918
9
J. AM. CHEM. SOC. VOL. 126, NO. 39, 2004 12203