Enzyme catalytic promiscuity: The nonheme Fe2+ center of β-diketone-cleaving dioxygenase Dke1 promotes hydrolysis of activated esters

Article abstract of DOI:10.1002/cbic.200900688

(Chemical Equation Presented) Trying new partners: The nonheme Fe ²⁺ center of β-diketone-cleaving dioxygenase in native (His62, His64, His104) and mutated (His62, His64, Glu104) form catalyzed hydrolysis of 4-nitrophenyl esters of short-chain

Full text of DOI:10.1002/cbic.200900688

DOI: 10.1002/cbic.200900688
Enzyme Catalytic Promiscuity: The Nonheme Fe2+ Center of b-Diketone-
Cleaving Dioxygenase Dke1 Promotes Hydrolysis of Activated Esters
Stefan Leitgeb and Bernd Nidetzky*[a]
Natural enzymes are often described as highly efficient and
finely tuned catalysts for specific chemical transformations of
biological relevance. Some of these enzymes, however, are pro-
miscuous in a sense that they catalyze secondary reactions,
chemically distinct from the reaction promoted in the conver-
sion of the physiological substrate.[1] Catalytic promiscuity has
drawn mechanistic attention as it necessitates that different
reaction coordinates be accommodated by a single enzyme
active site.[1a] Furthermore, it presents an often untapped
source of potentially useful enzymatic functionality.[2] Active-
site redesign, by using minimal structural modifications of the
Figure 1. Metal cofactor coordination by the 3-His center of Dke1. The pic-
original catalytic center, has been exploited successfully with
ture was derived from the crystal structure of a Zn2+ form of Dke1 (PDB ID:
some enzymes to revive latent alternative reactivity present in
3BAL) that is inactive as dioxygenase[4,5] but promotes ester hydrolysis (this
work).
the native form.[1–3]
b-Diketone-cleaving enzyme Dke1 is a nonheme Fe2+-de-
pendent dioxygenase from Acinetobacter johnsonii that con-
verts 2,4-pentanedione and O2 into methyglyoxal and acetate
(Scheme 1).[4] The metal cofactor of Dke1 is coordinated by the
triad His62, His64, and His104 (Figure 1) and is absolutely re-
function of a Fe2+/Zn2+ esterase. The type of catalytic promis-
cuity (oxygenase!esterase) seen for Dke1 is novel, and with
the exception of methionine aminopeptidase,[6] for which a
role for Fe2+ as cofactor has been proposed, hydrolytic activity
of a catalytic Fe2+ site appears to have limited precedence in
reported enzymes (see later).
When a purified preparation of Dke1 (5 mm Fe2+ sites) was
incubated in the presence of 4-nitrophenylpropionate (pNPP;
3 mm in 20 mm Tris-Cl buffer, pH 7.5; 258C) formation of 4-ni-
trophenol occurred at a spectrophotometric rate (405 nm) that
was 40-fold faster than the rate of the uncatalyzed hydrolysis
of the ester substrate under otherwise identical conditions
lacking the enzyme. Apo-Dke1, from which Fe2+ had been re-
quired for oxygenase activity.[5] Other metal ions, like Zn2+
,
Ni2+ or Cu2+, that compete with Fe2+ for binding to the 3-His
site are completely inactive in the enzymatic reaction with
O2.[5] We show here that the catalytic center of Dke1 in its
native Fe2+ form catalyses the hydrolysis of 4-nitrophenylesters
of short-chain alkanoic acids, which is an accidental type[2a] of
catalytic promiscuity. A non-native Zn2+ form of the enzyme
was likewise active as esterase—about ten-times more so than
the Fe2+ enzyme. Substitution of His104 by Glu, which de-
stroys the oxygenase activity,[5] was fully compatible with the
Scheme 1. Proposed catalytic mechanism of O2-dependent Cꢀ-C bond cleavage in 2,4-pentanedione (R1=R2=CH3) and related b-diketones by Dke1. A b-keto-
enolate coordinated to the nonheme Fe2+ center is thought to be the reactive form of the substrate. The type of cleavage shown takes place in the case that
R2 is more electron-withdrawing than R1.[4b]
moved by exhaustive dialysis in the presence of the metal
chelator FereneS, was completely inactive towards pNPP. Dke1
preparations showing a variable Fe2+ content between 0.03
and 0.70 (molmetal per molenzyme subunit of 16 kDa) dis-
played specific activities that increased linearly with increasing
Fe2+ occupancy of the active site (Figure S1 in the Supporting
[a] Dr. S. Leitgeb, Prof. Dr. B. Nidetzky
Institute of Biotechnology and Biochemical Engineering
Graz University of Technology, Petersgasse 12, 8010 Graz (Austria)
Fax: (+43)316-873-8400
E-mail: bernd.nidetzky@tugraz.at
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.200900688.
502
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2010, 11, 502 – 505

Information). Fe2+ in solution (33–167 mm) did not promote hy-
drolysis of NPP. The nonheme Fe2+ center of Dke1 is therefore
essential for the esterase activity of the enzyme.
these site-directed substitutions[5] could explain the absence of
Fe2+ esterase activity in the four Dke1 variants, it is interesting
that the corresponding Zn2+ enzymes were also inactive de-
spite the fact that, according to literature,[5] binding of Zn2+
has been hardly affected in the His62 and His64 mutants as
compared to wild-type Dke1. The presence of esterase activity
in a His104!Glu enzyme (H104E) that was nonreactive as b-
diketone-cleaving dioxygenase is therefore striking. Both Fe2+
and Zn2+ forms of H104E were active, as shown in Table 1.
However, binding of Fe2+ appeared to have been strongly im-
paired as a result of the substitu-
Variants of Dke1 in which the native Fe2+ had been substi-
tuted by another divalent metal were either totally inactive
toward pNPP (Cu2+, Mn2+, Ni2+) or showed activity (Zn2+
)
comparable to that seen for the Fe2+ enzyme. However, when
applied in solution, none of the metals used for replacement
of Fe2+ caused rate acceleration in hydrolysis of pNPP. Table 1
summarizes the kinetic parameters of Fe2+ and Zn2+ forms of
tion His104 by Glu, arguably ex-
Table 1. Kinetic parameters for hydrolysis of ester substrates by Fe2+ and Zn2+ forms of wild-type Dke1 and
the metal-site variant H104E at pH 7.5.
plaining the relatively low kcat for
the Fe2+ as compared to the
Zn2+ dependent catalytic reac-
tion with this mutant.
Wild-type enzyme
Zn2+
H104E
Parameter
kcat [sꢀ1
Fe2+
Fe2+
0.3ꢁ10ꢀ2
0.7ꢁ10ꢀ2
0.1ꢁ10ꢀ2
4.3ꢁ102
1.8ꢁ103
94
7.4
3.6
Zn2+
4.5ꢁ10ꢀ2
3.1ꢁ10ꢀ2
0.7ꢁ10ꢀ2
9.2ꢁ102
2.1ꢁ103
1.7ꢁ102
There are two principle mech-
anisms by which Fe2+/Zn2+ cen-
ters of the native 3-His form of
Dke1 or the mutant 2-His–1-Glu
form thereof might provide cata-
lytic assistance to 4-nitrophenyl-
ester hydrolysis (Scheme 2). Un-
catalyzed hydrolysis is thought
to occur through simultaneous
attack of hydroxide and hydroni-
um to form a gem-diol inter-
mediate, the elimination of
which yields the products
(Scheme 2C).[8] We measured the
pH dependence of the conver-
sion of pNPP by the different
Fe2+/Zn2+ enzyme preparations,
considering that pH-rate profiles
]
pNPP
pNPA
pNPB
pNPP
pNPA
pNPB
pNPP
pNPA
pNPB
pNPP
pNPA
pNPB
pNPP
pNPA
pNPB
3.0ꢁ10ꢀ2
2.7ꢁ10ꢀ2
1.3ꢁ10ꢀ2
0.6ꢁ10ꢀ2
61
2.6ꢁ102
17
4.5ꢁ102
0.5ꢁ10ꢀ2
0.7ꢁ10ꢀ2
4.2ꢁ102
1.0ꢁ103
2.5ꢁ102
KM [mm]
kcat/KM [mꢀ1 sꢀ1
]
71
5.1
29
48
14
41
49
3.6ꢁ102
12
kuncat [sꢀ1
]
3.6ꢁ10ꢀ4
1.0ꢁ10ꢀ3
n.d.[c]
kcat/KM kuncat [mꢀ1
]
2.0ꢁ105
5.0ꢁ103
n.d.[c]
1.2ꢁ106
4.8ꢁ104
n.d.[c]
2.0ꢁ104
3.5ꢁ103
n.d.[c]
1.3ꢁ105
1.4ꢁ104
n.d.[c]
[b]
[a] Relative s.d. on kcat and KM was ꢁ15 and ꢁ20%, respectively; [b] kuncat was determined at a constant sub-
strate concentration of 4.2 mm; [c] not determined: uncatalyzed rate was too low to be determined precisely.
pNPP: 4-nitrophenylpropionate; pNPA: 4-nitrophenylacetate; pNPB: 4-nitrophenlybutyrate.
Dke1 for hydrolysis of 4-nitrophenylesters of alkanoic acids dif-
fering in chain length from two to four carbon atoms. The
Zn2+ enzyme was the more efficient catalyst (in kcat/KM terms)
of the two Dke1 preparations. Turnover of substrate (kcat) was
only about two (pNPP and 4-nitrophenylacetate; pNPA) or
three orders of magnitude (4-nitrophenylbutyrate; pNPB)
slower than O2-dependent conversion of 2,4-pentanedione, the
presumed natural reaction of the Fe2+ enzyme.[4a] The observa-
tion that irrespective of the metal cofactor used Dke1 could be
saturated with substrate concentrations in the micromolar
range suggests a mechanism in which the applied 4-nitrophen-
lyesters bind to the Fe2+/Zn2+ active site of the enzyme where
catalytic assistance to their hydrolysis is provided. The rate
acceleration contributed by the two nonheme metal centers
over the respective uncatalyzed reaction (kuncat) at pH 7.5 was
might be mechanistically revealing. Results are shown as
double-log plots in Figure 2. The apparent kcat for wild-type
enzyme (Fe2+, Zn2+) was level at low pH (ꢁ7.0–7.5) and in-
creased as the pH was raised in the range ~7.5–11 (Figure 2A).
Interestingly, the pH dependence of kcat for H104E (Zn2+) was
level at high pH (ꢂ9.5) and decreased below an apparent pK
of ~9 (Figure 2B). Unlike the wild-type enzyme (Fe2+, Zn2+
)
the activity of H104E (Zn2+) did not reach a constant level at
pH values down to 6.0. Hydrolysis of pNPP by H104E (Fe2+
)
was pH dependent in the range 6.5–11.0. These pH dependen-
cies generally support a role for base catalysis in the enzymatic
reactions, and arguably involve participation from a metal–hy-
droxide species (Scheme 2B). However, the pH effects on kcat
for wild-type enzyme and H104E were clearly not uniform. It is,
therefore, conceivable for Dke1 (and would be consistent with
studies of other metalloenzymes like carbonic anhydrase and
small molecule models thereof[9]) that changes in the primary
coordination sphere can result in a slightly altered catalytic
action of the active-site metal. However, a more detailed ex-
amination of a potential ligand field effect in Fe2+/Zn2+ Dke1
was considered to be beyond the scope of this work.
in the range 15–120-fold (Table 1). A catalytic proficiency (kcat/
[7]
KM kuncat
)
of between 2.0ꢁ105 and 1.2ꢁ106 mꢀ1 was calculated
from the data in Table 1.
Mutants of Dke1 in which the native 3-His center of non-
heme Fe2+ had been changed at position His62 (to Glu) and at
position His64 (to Glu, Asp, or Asn) were nonreactive as Fe2+
-
and Zn2+-dependent esterases. While disruption of the catalyt-
ic Fe2+ site that was previously shown to occur as a result of
Despite notable efforts toward design of artificial metalloen-
zymes from nonenzymatic protein scaffolds,[10] there is current-
ChemBioChem 2010, 11, 502 – 505
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
503

diol dioxygenases, both of which
catalyze the oxidative cleavage
of catechol, are interesting ex-
amples in which hydrolysis of
oxidized reaction intermediates
(intradiol: anhydride; extradiol:
lactone) promoted by metal-
bound hydroxide constitutes the
final step of the proposed enzy-
matic mechanism.[12] The extra-
diol dioxygenase MhpB from Es-
cherichia coli was shown to pro-
vide Fe2+-dependent catalysis to
the hydrolysis of a seven-mem-
bered lactone analogue that was
chemically synthesized to mimic
the putative reaction intermedia-
te.[12c] Di-zinc aminopeptidase
provides an interesting case of
reverse (hydrolase!oxidase) cat-
alytic promiscuity for which sub-
stitution of the native Zn2+ by
Cu2+ has yielded a proficient
catechol oxidase dependent on
H2O2.[13] Also relevant, incorpora-
tion of vanadate into hydrolase
phytase has generated a novel
peroxidase.[14] Generally, metal
exchange was in several en-
zymes accompanied by reactivity
change.[3c–e]
Scheme 2. Possible mechanisms for catalytic participation of the Dke1 Fe2+/Zn2+ center in hydrolysis of ester sub-
strates. A) Lewis acid stabilization; B) activation of water for attack on carbon; C) uncatalyzed hydrolysis.
Dke1 belongs to the cupin
protein superfamily, a large and
widely spread group of b-barrel
proteins in which a mononuclear
metal center is both a conserved
feature of the structure and a
rich source of functional diversi-
ty.[15] The metal binding residues
of Dke1 (Figure 1) are contribut-
2+
Figure 2. The pH profiles of kcat (sꢀ1) for hydrolysis of pNPP by: A) wild-type Dke1, and B) H104E; : Fe enzymes;
!
*
2+
: Zn enzymes. Note that all initial rates used in determination of kcat were carefully corrected for the corre-
sponding rates of uncatalyzed hydrolysis.
ly limited evidence for catalytic promiscuity in natural enzymes
displaying activity dependent on a metal cofactor. Carbonic an-
hydrase, which like Dke1 utilizes a 3-His motif for coordination
of its native Zn2+ cofactor, shows Zn2+/Co2+ dependent ester-
ase activity.[11] The catalytic efficiency of carbonic anhydrase for
ed from two highly conserved core motifs that provide the sig-
nature for the superfamily. Metal exchange, as shown for Fe2+
and Zn2+ forms of Dke1, could be one way of introducing new
reactivity patterns in existing enzymes. Cupins have evolved di-
verse enzymatic functions spanning primary E.C. classes 1, 3,[16]
4, and 5. Design of biocatalysts with novel or improved func-
tionalities could focus on interconversion of homologous en-
zymes or the exploitation of accidental catalytic promiscuity,
like the one observed in Dke1.
hydrolysis of pNPP (516mꢀ1 ꢀ1; pH 8.5) can be compared with
s
that of wild-type Dke1 in the Zn2+ form.[11e,f] Rational design
and directed-evolution strategies have been successfully ap-
plied to enhance the activity and the substrate scope of car-
bonic anhydrase as an ester hydrolase,[11d] and show that these
secondary activities of metalloenzymes present a potential op-
portunity for biocatalytic synthesis. Although recent work on
methionine aminopeptidase reveals that Fe2+ sites in proteins
are capable of promoting a hydrolysis reaction,[7] the promiscu-
ous function of an oxygenase Fe2+ catalytic center as an ester-
Experimental Section
Experimental details (chemicals and reagents; site-directed muta-
genesis, enzyme production and purification; metal exchange stud-
ies; enzymatic assays; steady-state kinetic analysis) are provided in
the Supporting Information.
ase active site has not been described before. However, Fe3+
dependent intradiol dioxygenases and Fe2+-dependent extra-
-
504
www.chembiochem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2010, 11, 502 – 505

B. Nidetzky, ChemBioChem 2006, 7, 1536–1548; d) S. Leitgeb, B. Nidetz-
ky, Biochem. Soc. Trans. 2008, 36, 1180–1186.
Acknowledgements
[5] S. Leitgeb, G. D. Straganz, B. Nidetzky, Biochem. J. 2009, 418, 403–411.
[6] a) S. Chai, W.-L. Wang, Q.-Z. Ye, J. Biol. Chem. 2008, 283, 26879–26885;
b) M. Leopoldini, N. Russo, M. Toscano, J. Am. Chem. Soc. 2007, 129,
7776–7784.
Financial support from Graz University of Technology is gratefully
acknowledged.
[7] On definition and use of enzyme catalytic proficiency: a) X. Zhang, K. N.
Houk, Acc. Chem. Res. 2005, 38, 379–385; b) B. G. Miller, R. Wolfenden,
Annu. Rev. Biochem. 2002, 71, 847–885.
Keywords: enzyme
metalloenzymes · nonheme iron · zinc
catalysis
·
esterase
activity
·
[8] X. Zhang, J. DeChancie, H. Gunaydin, A. B. Chowdry, F. R. Clemente,
A. J. T. Smith, T. M. Handel, K. N. Houk, J. Org. Chem. 2008, 73, 889–899.
[9] V. M. Krishnamurthy, G. K. Kaufman, A. R. Urbach, I. Gitlin, K. L. Gudiksen,
D. B. Weibel, G. M. Whitesides, Chem. Rev. 2008, 108, 946–1051.
[10] a) C. Letondor, T. R. Ward, ChemBioChem 2006, 7, 1845–1852; b) M.
Creus, T. R. Ward, Org. Biomol. Chem. 2007, 5, 1835–1844.
[11] a) Y. Pocker, M. W. Beug, Biochemistry 1972, 11, 698–707; b) A. Thor-
slund, S. Lindskog, Eur. J. Biochem. 1967, 3, 117–123; c) B. Elleby, B. Sjç-
blom, S. Lindskog, Eur. J. Biochem. 1999, 262, 516–521; d) S. M. Gould,
D. S. Tawfik, Biochemistry 2005, 44, 5444–5452; e) G. Hçst, L.-G. Mꢂr-
tensson, B.-H. Jonsson, Biochim. Biophys. Acta 2006, 1764, 1601–1606;
f) G. Hçst, B.-H. Jonsson, Biochim. Biophys. Acta 2008, 1784, 811–815.
[12] a) T. D. H. Bugg, Tetrahedron 2003, 59, 7075–7101; b) T. D. H. Bugg, G.
Lin, Chem. Commun. 2001, 941–952; c) J. Sanvoisin, G. J. Langley,
T. D. H. Bugg, J. Am. Chem. Soc. 1995, 117, 7836–7837.
[13] G. F. Z. da Silva, L.-J. Ming, J. Am. Chem. Soc. 2005, 127, 16380–16381.
[14] F. van de Velde, L. Kçnemann, F. van Rantwijk, R. A. Sheldon, Biotechnol.
Bioeng. 2000, 67, 87–96.
[15] J. M. Dunwell, A. Purvis, S. Khuri, Phytochemistry 2004, 65, 7–17.
[16] Ureidoglycolate hydrolase (E.C. 3.5.3.19) is dependent on Mn2+ and
Zn2+, however, not an esterase; X. E. Wells, E. M. Lees, Arch. Biochem.
Biophys. 1991, 287, 151–159.
[1] a) M. D. Toscano, K. J. Woycechowsky, D. Hilvert, Angew. Chem. 2007,
119, 3274–3300; Angew. Chem. Int. Ed. 2007, 46, 3212–3236; b) I.
Nobeli, A. D. Favia, J. M. Thornton, Nat. Biotechnol. 2009, 27, 157–167;
c) N. Tokuriki, D. S. Tawfik, Science 2009, 324, 203–207;d) C. Li, M. Hass-
ler, T. D. Bugg, ChemBioChem 2008, 9, 71–76; e) L. L. Lairson, A. G.
Watts, W. W. Warkachuk, S. G. Withers, Nat. Chem. Biol. 2006, 2, 724–
728; f) O. Khersonsky, C. Roodveldt, D. S. Tawfik, Curr. Opin. Chem. Biol.
2006, 10, 498–508; g) P. J. O’Brien, D. Herschlag, Chem. Biol. 1999, 6,
91–105.
[2] a) K. Hult, P. Berglund, Trends Biotechnol. 2007, 25, 231–238; b) R. J. Ka-
zlauskas, Curr. Opin. Chem. Biol. 2005, 9, 195–201; c) U. T. Bornscheuer,
R. J. Kazlauskas, Angew. Chem. Int. Ed. 2004, 43, 6032–6040.
[3] a) H. Jochens, K. Stiba, C. Saville, R. Fujii, J. G. Yu, T. Gerassenkov, R. J. Ka-
zlauskas, U. T. Bornscheuer, Angew. Chem. 2009, 121, 3584–3587;
Angew. Chem. Int. Ed. 2009, 48, 3532–3535; b) R. Kourist, S. Bartsch, L.
Fransson, K. Hult, U. T. Bornscheuer, ChemBioChem 2008, 9, 67–69; c) I.
Sanchez-Moreno, L. Iturrate, R. Martin-Hoyos, M. L. Jimeno, M. Mena, A.
Bastida, E. Garcia-Junceda, ChemBioChem 2009, 10, 225–229; d) K.
Okrasa, R. J. Kazlauskas, Chem. Eur. J. 2006, 12, 1587–1596; e) A. Fernan-
dez-Gacio, A. Codina, J. Fastrez, O. Riant, P. Soumillon, ChemBioChem
2006, 7, 1013–1016; f) Q. Jing, K. Okrasa, R. J. Kazlauskas, Chem. Eur. J.
2009, 15, 1370–1376.
[4] a) G. D. Straganz, A. Glieder, L. Brecker, D. W. Ribbons, W. Steiner, Bio-
chem. J. 2003, 369, 573–581; b) G. D. Straganz, H. Hofer, W. Steiner, B.
Nidetzky, J. Am. Chem. Soc. 2004, 126, 12202–12203; c) G. D. Straganz,
Received: November 12, 2009
Published online on January 28, 2010
ChemBioChem 2010, 11, 502 – 505
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
505