Efficient catalytic promiscuity for chemically distinct reactions

Full text of DOI:10.1002/anie.200805843

Communications
DOI: 10.1002/anie.200805843
Catalytic Promiscuity
Efficient Catalytic Promiscuity for Chemically Distinct Reactions**
Ann C. Babtie, Subhajit Bandyopadhyay, Luis F. Olguin, and Florian Hollfelder*
The extraordinary rate enhancements achieved by enzymes,
2
6
[1]
with values up to 10 reported for (k /K )/k , are usually
cat
m
2
associated with exquisite specificity for one transition state in
[
2]
textbooks. In this context, it is remarkable that some
enzymes are able to combine efficient catalysis of their
native function with the ability to accelerate chemically
distinct reactions involving formation and cleavage of differ-
ent bonds. This phenomenon is known as catalytic promiscu-
[
3–6]
ity.
It is proposed to play a role in the divergent evolution
of enzymes by providing a head start in activity and a possible
[
6–9]
selective advantage to a duplicated gene.
catalytic proficiencies
In most cases,
for the promiscuous activities are
much lower than that for the parent reaction, not exceeding
[10]
1
3
15
(
k /K )/k values of 10 –10 and typically much less than
cat
m
6,11]
2
[
Scheme 1. PAS hydrolyzes a) phosphate diesters 1, phosphate mono-
ester 3, and b) sulfate monoester 2.
that.
Herein we report a promiscuous phosphodiesterase
activity for Pseudomonas aeruginosa arylsulfatase (PAS) with
1
8
a remarkably high proficiency of 10 . Several experimental
observations suggest that PAS acts as a sulfatase in vivo: it
À1
Menten parameters kcat = 0.55 s and K = 2.2 mm, and multi-
m
[12]
hydrolyzes several aromatic sulfate esters, its expression is
induced under sulfate starvation conditions and repressed in
ple turnovers per active site (Figure 1). The kinetic param-
eters for all reactions catalyzed by PAS are shown in Table 1.
Several experimental methods were used to confirm
phosphate diester hydrolysis as a genuine activity of PAS,
rather than the result of a contaminating enzyme. During
purification of PAS, the diesterase activity co-elutes with the
native sulfatase activity and previously observed promiscuous
[
12,13]
the presence of inorganic sulfate,
part of a gene cluster that also encodes a sulfate ester
and its gene (atsA) is
[13]
transport system. A promiscuous phosphate monoesterase
[14]
activity has been reported for this enzyme, but this reaction
is catalyzed much less efficiently than sulfate ester hydrolysis.
The promiscuous diesterase activity described herein sur-
passes all other reported promiscuous activities, and is
comparable to the proficiency of the native sulfatase activity
[
14]
phosphate monoesterase activity
from three different
columns with a constant ratio between the three activities
1
8 [14]
(
4 ꢀ 10 ) of PAS, thus challenging the idea that efficient
catalysis requires specialization.
PAS expressed in E. coli was purified in three steps as
[
14]
described previously, and appears as a single band on a
SDS-PAGE gel. The purified enzyme hydrolyzes phosphate
diesters 1 (Scheme 1) to produce 4-nitrophenolate; the yellow
color thus enables the progress of the reaction to be followed
spectrophotometrically. For the reaction of diester 1b at
pH 8.0, we observed saturation kinetics with Michaelis–
[
+]
[
*] A. C. Babtie, Dr. S. Bandyopadhyay, Dr. L. F. Olguin,
Dr. F. Hollfelder
Department of Biochemistry, University of Cambridge
Cambridge CB2 1GA (UK)
Fax: (+44)1223-766-002
E-mail: fh111@cam.ac.uk
Homepage: http://www.bioc.cam.ac.uk/~fh111
Figure 1. Promiscuous phosphodiesterase activity of PAS. A typical
+
[
] Current address: Indian Institute of Science Education & Research
IISER), HC—Sector III, Salt Lake City, Kolkata 700106 (India)
Michaelis–Menten plot of the initial rate n for phosphate diester 1b
0
(
([PAS]=15 nm; [1b]=1.1–165 mm). Inset: Time course of the reactions
[
**] This work was supported by the EPSRC, the EU network ENDIRPRO,
and a CASE studentship of the BBSRC and GlaxoSmithKline to A.B.
F.H. is an ERC Starting Investigator. We thank Tony Kirby and
members of the Hollfelder group for comments on the manuscript.
of PAS with phosphate diester 1b (&, [PAS]=10 nm, [1b]=2 mm) with
more than 200 reaction cycles per active site. For comparison, the
reaction of sulfate monoester 2 (*, [PAS]=0.5 nm, [2]=2.3 mm) and
the corresponding uncatalyzed reactions (&,*) are shown. Condi-
tions: 258C, 0.1m Tris-HCl (pH 8.0), 0.5 mgmL BSA, 0.5m NaCl.
Tris=tris(hydroxymethyl)aminomethane.
À1
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200805843.
3
692
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 3692 –3694

Angewandte
Chemie
Table 1: Kinetic parameters for the PAS-catalyzed hydrolysis of phos-
[
a]
phodiesters 1, sulfate monoester 2, and phosphate monoester 3.
À1
À1 À1
[b]
Substrate Km [mm]
kcat [s
]
kcat/Km [m
s
]
(kcat/K )/k
m 2
1
5
1
1
1
2
3
a
b
c
617Æ61
2.2Æ0.3
25Æ2
0.073Æ0.004 119Æ7
6.5ꢀ10
1.3ꢀ10
5
18
0.55Æ0.02
0.19Æ0.01
(2.5Æ0.3)ꢀ10
(7.6Æ0.5)ꢀ10
(4.9Æ0.8)ꢀ10
3
7
Figure 2. Transition states during phosphodiester (associative; left)
and sulfate monoester hydrolysis (dissociative; right).
[
c]
18
13
0.29Æ0.03 14.2Æ0.6
4.3ꢀ10
1.6ꢀ10
[
c]
29.1Æ2.0
0.023Æ0.001 790Æ58
À1
[
a] Reaction conditions: 258C, 0.1m Tris-HCl (pH 8.0), 0.5 mgmL
[
15]
15
bovine serum albumin (BSA), 0.5m NaCl (for 1a–c). [b] For hydrolytic
lyzes phosphate diesters ((k /K )/k = 8 ꢀ 10 ) with lower
cat m 2
reactions, the second-order rate constant k for the uncatalyzed reaction
sulfate and phosphate monoesterase activities (proficiencies
less than 10 ).
2
À1
[11,14]
is taken as k /55m. Values for k /s at 258C, pH 8.0:
1.0ꢀ
11 [20,21]
uncat
uncat
À10
À12
À11
À9
1
0
(1a), 1.1ꢀ10
(1b), 6.2ꢀ10
(2), 2.7ꢀ10 (3). [c] Data for
It is remarkable that PAS can achieve large and com-
parable rate enhancements for two reactions with distinct
features. Both catalyzed reactions are hydrolytic, but the
intrinsic chemistry, which is defined by bond-making and
bond-breaking, involves mechanistic alternatives. In solution,
the hydrolysis of phosphodiesters 1 occurs via a transition
state that is more associative in nature, with little negative
[
14]
sulfate (2) and phosphate (3) monoesters are included for compar-
ison.
(
Supporting Information, Figure S1). A phosphate diester
substrate acts as a competitive inhibitor of the native sulfatase
activity of PAS (Supporting Information, Figures S2 and S3),
indicating that both reactions occur in the same active site.
Furthermore, a mutant, C51S, lacking the wild-type nucleo-
phile (see Scheme 2) has a reduced kcat value for the reactions
[22–24]
charge development on the leaving group.
By contrast,
sulfate and phosphate monoesters proceed through dissocia-
tive transition states, with significant negative charge on the
[
14]
[22,25,26]
with both phosphate diester 1b and sulfate monoester 2 (by
leaving group (Figure 2).
PAS displays greater selec-
4
3
[16]
more than 10 and 10 respectively; Table 2).
tivity for substrate structure than for the transition-state
characteristics of the background reactions: we observed a
3
1
0 -fold variation in k /K for diesters 1a–c, with an
cat
m
apparent preference for diesters with two aromatic rings,
yet the values for diester 1b and sulfate 2 differ by just 100-
fold. Substrate charge appears to have a large influence on
reactivity; the increase in charge on phosphate monoester 3
relative to sulfate 2 is penalized by a 6 ꢀ 10 -fold decrease in
k /K , despite both reactions occurring via a dissociative
cat m
[
a]
Table 2: Kinetic parameters for the C51S mutant of PAS.
À1
À1 À1
Substrate
Km [mm]
kcat [s
]
kcat/Km [m
s
]
À5
1
2
b
3.7Æ0.3
(<5.5Æ0.1)ꢀ10
<15Æ1
[
b]
À3
4
4
0.25Æ0.06
(5.4Æ0.2)ꢀ10
(2.1Æ0.5)ꢀ10
À1
[
a] Reaction conditions: 258C, 0.1m Tris-HCl (pH 8.0), 0.5 mgmL BSA,
0
.5m NaCl (for 1b). [b] Data for sulfate monoester 2 included for
transition state in solution. The nature of the transition states
for PAS-catalyzed reactions remain to be determined.
[
14]
comparison.
[
27,28]
[22,29]
[30]
AP,
many other phosphatases,
and a sulfatase use
similar mechanisms to those in solution, but an example of a
promiscuous enzyme that hydrolyzes phosphate and phos-
phonate monoesters with transition states that are closer to
one another than those of the non-enzymatic reaction was
Catalytic promiscuity is a familiar feature in PAS, which
also hydrolyzes phosphate monoesters, such as 3, but much
[
14]
less efficiently than sulfate monoesters, such as 2.
The
[
31]
catalytic proficiencies (k /K )/k of the enzymatic reactions
recently reported.
cat
m
2
1
3
18
for substrate 3 is 1.6 ꢀ 10 and 4.3 ꢀ 10 for 2 (Table 1).
However, the phosphodiesterase activity (1b) of PAS has an
Efficient hydrolysis of phosphate diesters by PAS may
result from active-site groups that could carry out similar roles
during catalysis of both native and promiscuous reactions.
1
8
exceptionally high catalytic proficiency of 1.3 ꢀ 10 . Promis-
cuous cyclic phosphodiesterase activity has been previously
Specifically, the proposed double displacement mechanism
[
17]
[32]
observed for a sulfatase, but with lower efficiency.
This
for sulfate hydrolysis
involves Lewis acid catalysis by a
value surpasses the rate accelerations observed for the native
reactions of many enzymes, and to our knowledge is greater
than the proficiency of any previously reported promiscuous
calcium ion, a reactive nucleophile, and efficient substrate
binding and general acid catalysis by several cationic active
site groups (Supporting Information, Figure S4). The
decreased activity of the C51S mutant confirms the key role
of the formylglycine (FGly) nucleophile, which is formed by
[
6,11]
enzyme activity.
In comparison, two structurally and
evolutionarily related enzymes that belong to the same
superfamily as PAS catalyze the same three hydrolytic
reactions but show greater discrimination between the
native and promiscuous activities. E. coli alkaline phospha-
tase (AP) is a native phosphate monoesterase that also
[33]
post-translational modification of cysteine 51,
in both
reactions. This nucleophile can be regenerated by hemiacetal
cleavage, enabling multiple turnovers per active site
(Scheme 2 and inset in Figure 1).
The substantial rate accelerations show that PAS is a
highly efficient catalyst for both sulfate monoester and
[
18,19]
hydrolyzes sulfate monoesters and phosphate diesters,
but the proficiencies for the promiscuous reactions are at least
6
1
0 -fold lower than for the native phosphatase activity ((k /
phosphate diester hydrolysis. This efficiency is also reflected
cat
1
7
À17
K )/k = 7 ꢀ 10 ). Xanthomonas axonopodis nucleotide phos-
m
in the tight transition state binding (K = 1.3 ꢀ 10 m and
2
tx
À17
phodiesterase/pyrophosphatase (NPP) preferentially hydro-
4.2 ꢀ 10 m for substrates 1b and 2, respectively). This
Angew. Chem. Int. Ed. 2009, 48, 3692 –3694
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3693

Communications
[
[
10] Catalytic proficiency, (kcat/K )/k , is a ratio of the second-order
m 2
rate constants for the enzyme-catalyzed reaction and the
uncatalyzed reaction in solution; it is a measure of transition-
state stabilization by an enzyme, in this case comparing the
reaction of the FGly nucleophile with the reaction of water and
the sulfate or phosphate ester.
11] S. Jonas, F. Hollfelder in The Protein Engineering Handbook,
Vol. 1 (Eds.: S. Lutz, U. Bornscheuer), Wiley, Chichester, 2008,
pp. 47 – 72.
[
12] S. Beil, H. Kehrli, P. James, W. Staudenmann, A. M. Cook, T.
Leisinger, M. A. Kertesz, Eur. J. Biochem. 1995, 229, 385.
13] J. Hummerjohann, S. Laudenbach, J. Retey, T. Leisinger, M. A.
Kertesz, J. Bacteriol. 2000, 182, 2055.
[
Scheme 2. Role of the key nucleophilic residue FGly51 for the multiple
turnover of the phosphodiesterase activity of PAS. The intermediate is
[14] L. F. Olguin, S. E. Askew, A. C. OꢁDonoghue, F. Hollfelder, J.
Am. Chem. Soc. 2008, 130, 16547.
broken down by base catalysis, leading to CÀO bond cleavage. This
[15] BSA was added to stabilize PAS at low concentrations, as
[
14]
mechanism is available regardless of whether a phosphate or sulfate
intermediate is generated after reaction with the FGly nucleophile,
resulting in a common intermediate breakdown step for the three
reactions catalyzed by PAS. The metal ion and the residues acting as
general acid and base catalysts are not shown for clarity (see
Supporting Information, Figure S4).
described previously. Although it does not itself catalyze the
reaction, it may slightly increase the K : in the absence of BSA
m
m
the K for 1b was approximately 20 mm, but rapid loss of enzyme
activity prevented accurate determination of kinetic parameters
under these conditions.
[16] The stated diesterase activity of the C51S mutant represents an
upper limit owing to the possibility of small amounts of a
contaminating phosphodiesterase in the enzyme preparation.
Low levels of phosphodiesterase contamination of a comparable
activity were seen for a C51A mutant preparation. However, the
large decrease in activity on mutation of the catalytic nucleo-
phile confirms that PAS is indeed responsible for the observed
combination of considerable catalytic proficiency and a lack
of specificity is remarkable. Presumably there has been little
selective pressure against the secondary activity for this
enzyme during evolution, which may be an advantage for the
organism. The recycling of catalytic functionality for three
reactions indicates that, despite different transition-state
structures, substrate charges, and sizes, evolutionary crossover
between such reactions might be achieved with relative ease.
This observation also suggests that the ability to catalyze a
second reaction can not only occur in early stages of
phosphate diester hydrolysis. The presence of
a possible
contamination would not affect the analysis of the diesterase
activity of wild-type PAS, as the activity is negligible (less than 1/
10000 of the kcat/K
17] Ox liver sulfatase hydrolyzes cAMP with kcat/Km values 10 –10
value of wild-type PAS).
m
3
4
[
[
35]
fold lower than the measured sulfatase activity
and a
1
6
proficiency (k /K )/k = 1 ꢀ 10 calculated by OꢁBrien and
cat
m
2
[6]
Herschlag using an estimated value for the uncatalyzed
[
34]
functional adaptation with low rate accelerations, but also
in catalysts of high proficiency. A catalyst with multiple head-
start activities (in this case with k /K values differing only
[36]
hydrolysis of cAMP. However, this value relies on a rather
long extrapolation.
[18] P. J. OꢁBrien, D. Herschlag, J. Am. Chem. Soc. 1998, 120, 12369.
cat
m
4
[19] P. J. OꢁBrien, D. Herschlag, Biochemistry 2001, 40, 5691.
[20] J. K. Lassila, D. Herschlag, Biochemistry 2008, 47, 12853.
[21] J. G. Zalatan, T. D. Fenn, A. T. Brunger, D. Herschlag, Bio-
chemistry 2006, 45, 9788.
2
00-fold for two and 6 ꢀ 10 -fold for three reactions) may
provide a good starting point for future evolutionary adapta-
tion: versatile catalysts, such as PAS, could be useful cellular
tools as generalist scavengers that can respond quickly to a
range of environmental influences.
[22] W. W. Cleland, A. C. Hengge, Chem. Rev. 2006, 106, 3252.
[23] A. J. Kirby, M. Younas, J. Chem. Soc. B 1970, 1165.
[24] A. J. Kirby, M. Younas, J. Chem. Soc. B 1970, 510.
Received: December 1, 2008
Published online: April 16, 2009
[25] S. J. Benkovic, P. A. Benkovic, J. Am. Chem. Soc. 1966, 88, 5504.
[26] A. J. Kirby, W. P. Jencks, J. Am. Chem. Soc. 1965, 87, 3209.
[
[
[
27] F. Hollfelder, D. Herschlag, Biochemistry 1995, 34, 12255.
28] J. G. Zalatan, D. Herschlag, J. Am. Chem. Soc. 2006, 128, 1293.
29] Y. L. Zhang, F. Hollfelder, S. J. Gordon, L. Chen, Y. F. Keng, L.
Wu, D. Herschlag, Z. Y. Zhang, Biochemistry 1999, 38, 12111.
30] E. Chapman, M. C. Bryan, C. H. Wong, Proc. Natl. Acad. Sci.
USA 2003, 100, 910.
Keywords: catalytic promiscuity · enzyme catalysis · hydrolases ·
phosphatases · sulfatases
.
[
[
31] C. McWhirter, E. A. Lund, E. A. Tanifum, G. Feng, Q. I. Sheikh,
A. C. Hengge, N. H. Williams, J. Am. Chem. Soc. 2008, 130,
[
1] R. Wolfenden, Chem. Rev. 2006, 106, 3379.
[
2] A. Fersht, Structure and Mechanism in Protein Science, W. H.
Freeman, New York, 1999.
13673.
[
32] I. Boltes, H. Czapinska, A. Kahnert, R. von Bꢂlow, T. Dierks, B.
[
3] U. T. Bornscheuer, R. J. Kazlauskas, Angew. Chem. 2004, 116,
Schmidt, K. von Figura, M. A. Kertesz, I. Uson, Structure 2001,
6156; Angew. Chem. Int. Ed. 2004, 43, 6032.
9, 483.
[
[
4] K. Hult, P. Berglund, Trends Biotechnol. 2007, 25, 231.
5] O. Khersonsky, C. Roodveldt, D. S. Tawfik, Curr. Opin. Chem.
Biol. 2006, 10, 498.
[
[
[
[
33] T. Dierks, C. Miech, J. Hummerjohann, B. Schmidt, M. A.
Kertesz, K. von Figura, J. Biol. Chem. 1998, 273, 25560.
34] A. Aharoni, L. Gaidukov, O. Khersonsky, Q. G. S. Mc, C.
Roodveldt, D. S. Tawfik, Nat. Genet. 2005, 37, 73.
[6] P. J. OꢁBrien, D. Herschlag, Chem. Biol. 1999, 6, R91.
[7] R. A. Jensen, Annu. Rev. Microbiol. 1976, 30, 409.
[8] S. Bershtein, D. S. Tawfik, Mol. Biol. Evol. 2008, 25, 2311.
[9] M. E. Glasner, J. A. Gerlt, P. C. Babbitt, Curr. Opin. Chem. Biol.
35] T. Uchida, F. Egami, A. B. Roy, Biochim. Biophys. Acta
Enzymol. 1981, 657, 356.
36] J. Chin, X. Zou, Can. J. Chem. 1987, 65, 1882.
2006, 10, 492.
3
694 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 3692 –3694