Aldolase activity of serum albumins

Article abstract of DOI:10.1039/c0ob01219j

Bovine and human serum albumins catalyze the aldol reaction of aromatic aldehyedes and acetone, with saturation kinetics and moderate and opposite enantioselectivity. The reaction occurs at the binding site in domain IIa, and is inhibited by warfarin. Kin

Full text of DOI:10.1039/c0ob01219j

View Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 4417
www.rsc.org/obc
COMMUNICATION
Aldolase activity of serum albumins†
Fabio Benedetti, Federico Berti* and Silvia Bidoggia
Received 20th December 2010, Accepted 11th April 2011
DOI: 10.1039/c0ob01219j
Bovine and human serum albumins catalyze the aldol reaction
of aromatic aldehyedes and acetone, with saturation kinetics
and moderate and opposite enantioselectivity. The reaction
occurs at the binding site in domain IIa, and is inhibited
by warfarin. Kinetic data are consistent with an enamine
mechanism. The activity is conserved in a 103 aminoacid
peptide derived from the albumin sequence.
The aldol reaction is one of the most important methods of
organic synthesis, allowing formation of a carbon–carbon bond
1
with generation of one or two stereocenters. Stereocontrol can
1
2
be achieved by asymmetric induction, by chiral auxiliaries, and
3
by asymmetric catalysis. The latter approach is very attractive
and great effort has been dedicated to the discovery of new
3
,4
catalysts, including small organic molecules (organocatalysts),
2,3
1,5
chiral metal complexes, enzyme mimics as catalytic antibodies
6
and polymers.
Scheme 1 Aldol reactions of aldehydes 1 with acetone.
5
Covalent catalysis is used by class I aldolases, and also by
designed enzymes and catalytic antibodies. A common feature
found in the catalytic site of these proteins is the presence of
7
8
addition of acetone to aromatic aldehydes (Scheme 1) is indeed
accelerated by albumin with an enzyme-like mechanism.
Formation of aldol 2a by reaction of acetone with aldehyde 1a
is accelerated by both HSA and BSA, with saturation kinetics
(Fig. 1; Table 1). The kinetic parameters are very similar for
the two proteins, the catalyzed process being three orders of
magnitude faster than the uncatalyzed one, while the Michaelis–
Menten constant is in the millimolar range in both cases. The
reactions exhibited multiple turnover, thus revealing true catalysis,
and allowing full conversion of 1a to aldol 2a (partial dehydration
of 2a to the a,b-unsaturated ketone was observed at over 80%
conversion).
a low pK
a
, nucleophilic lysine, surrounded by a hydrophobic
environment, which catalyzes the aldol/retroaldol reaction by an
enamine/Schiff base mechanism.
A low pK lysine is also present in the IIa binding site of human
5,7,8
a
(
HSA: Lys-199) and bovine (BSA: Lys-222) albumin. This residue
9
is involved in the covalent binding of substrates, and is responsible
for albumin’s ability to behave as an enzyme-like catalyst in b-
eliminations, in the decomposition of Meisenheimer adducts,
and in the Kemp elimination. Notwithstanding the presence of
an active lysine in a hydrophobic binding site, which might present
albumin with a chance to behave as an aldolase catalyst, BSA was
reported by Barbas not to catalyze the aldol addition of acetone to
10
11
12
Albumin-catalyzed reactions show an opposite enantioprefer-
ence: in the BSA-catalyzed process, the Si face of aldehyde 2a is
preferred, thus giving the S aldol with 40% ee. On the contrary,
the HSA-catalyzed reaction leads to the R product with 60% ee.
The stereoselectivity is moderate if compared to the 97% displayed
by antibody 38C2 in the same reaction,‡ but significantly higher
than the 8% ee obtained with Tanaka’s catalytic peptides in the
13
3
-(4-acetamidophenyl)-propanal. However, based on albumin’s
9a,14
known specificity for aromatic molecules,
we reasoned that the
lack of catalytic activity observed by Barbas might be due to poor
recognition of the aliphatic aldehyde by the protein and decided
to further investigate the reaction. We now report that the aldol
16
reaction of acetone and 1e. The switch in enantioselectivity going
from BSA to HSA might reflect the different position of the lysine
residue (222 in BSA and 199 in HSA) within the otherwise identical
sequence of the IIa binding site, thus suggesting an active role for
this aminoacid in the catalytic mechanism.
Department of Chemical and Pharmaceutical Sciences, University of Trieste,
Via Giorgieri 1, 34127 Trieste, (Italy). E-mail: fberti@units.it; Fax: (+39)
0
40-558-2402
†
Electronic supplementary information (ESI) available: HPLC and kinetic
data, stereoselectivity, Hammett Plot, sequence of IIa103 peptide and
details of its preparation and purification, DFT geometries of transition
states. See DOI: 10.1039/c0ob01219j
BSA showed no aldolase activity after reaction with fluorescein
11c
isothiocyanate. This reagent selectively modifies Lys-222, thus
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4417–4420 | 4417

View Online
◦
Table 1 Kinetic parameters (at 37 C) for the uncatalyzed and albumin-catalyzed aldol reactions
a
3
-1
b
3
-1
b
unc [M-1]
c
Entry
Substrate/albumin
k
unc 10 [min ]
k
cat 10 [min ]
k
cat/kunc
K
M
[mM]
k
cat/K
M
k
ee %
5
5
4
4
4
3
1
2
3
4
6
7
1a/HSA
1a/BSA
1b/BSA
1c/BSA
1d/BSA
1e/BSA
0.00804
10.0
10.4
6.72
23.6
427.1
672.0
1250
1290
206
92
263
280
2.1
3.3
4.3
2.7
22.3
66.1
4.1 ¥ 10
2.4 ¥ 10
4.8 ¥ 10
3.4 ¥ 10
1.2 ¥ 10
4.2 ¥ 10
60 (R)
40 (S)
0.0326
0.258
1.62
2.39
a
b
c
3
observed pseudo first order values. apparent values in 10% aqueous acetone. at 70% conversion by Eu(hfc) NMR; abs. configuration assigned by
15
comparison with that of 2a obtained by antibody 38c2 catalysis.
Fig. 2 Dixon plot for the warfarin-inhibited aldol reaction of aldehyde
a and acetone. The concentrations of 1a are 0.5 mM (᭹) and 1 mM (᭺).
1
Fig. 1 Michaelis–Menten and Lineweaver–Burk (insert) plots for the
BSA (᭹) and HSA (᭺) catalyzed aldol reactions of aldehyde 1a and
acetone.
The intermediate value of r (+2.08) obtained for our uncatalyzed
process at pH 7.5 is consistent with a reaction occurring between
neutral species with moderate charge transfer to the aldehyde in the
transition state and a similar conclusion can be drawn for the BSA-
catalyzed reaction (r +2.08). An enamine-mediated mechanism,
strongly supporting the hypothesis that this residue plays a key role
in the catalytic mechanism. In order to further confirm this, the
reaction was studied in the presence of (±)-warfarin, a well-known
involving a low pK
24h6A from a +2 value of r and also seems entirely compatible
with our results. We have also found that incorporation of O by
a
lysine, has been inferred for aldolase antibody
17
21
ligand of the IIa site. BSA’s aldolase activity is competitively
inhibited by the drug, with a Ki of 2.94 mM (Fig. 2). This
experiment not only confirms that the small molecule binding
site of subdomain IIa is the aldolase site, but also provides clear
evidence in favour of a specific catalytic process.
1
8
1
8
acetone (10% in H
2
O) is 15 times faster in the presence of 200 mM
BSA. The rate of exchange is comparable to that obtained with
22
aldolases on ketone substrates, which has been postulated to
proceed via the same enamine intermediate of the aldol reaction.
We have also carried out a series of DFT calculations on the
reaction of aldehydes 1b–e with acetone’s enol 3 (Scheme 2) as
a model for the uncatalyzed reaction and N-methylethenamine
BSA proved active on the whole set of p-substituted benzalde-
hydes 1b–e (Table 1), although worked less efficiently than with
aldehyde 1a, with kcat/kunc ratios in the 100–300 range. Similar
values of K
M
were found for aldehydes 1a–c, while K increases
M
23
by one order of magnitude when the aldehyde bears a polar group
4 as a model of the catalyzed process. Both reactions prefer
a concerted mechanism, via cyclic transition states. Formation
of the C–C bond is significantly more advanced in the enamine
transition state, with proton transfer lagging behind (see ESI†).
A good agreement is found between experimental constants for
the albumin-catalyzed reaction and calculated constants for the
model enamine reaction, thus further supporting the proposed
enamine mechanism for the catalyzed process (Table 2). On the
contrary, the relative rate constants calculated for the enol reaction
do not reproduce well the experimental values for the uncatalyzed
reaction. Probably the model is an over-simplification of the actual
in the para position (1d, 1e). This is consistent with the topology
9a
of the IIa binding site, which is sock-shaped, with the bottom of
both its subsites layered with hydrophobic residues.
In a Hammet plot for kcat and kunc (see ESI†), both series of data
correlate well with s , with similar slopes of +2.08 (uncatalyzed)
p
and +1.95 (catalyzed). In the addition of the acetone enolate to
benzaldehydes, a higher r value of around +3 can be estimated
18
for base-catalysis, while the acid-catalyzed adol addition shows
19
a moderately negative r value. A concerted reaction such as the
20
borane reduction occurs with a moderately positive r of +0.66.
4
418 | Org. Biomol. Chem., 2011, 9, 4417–4420
This journal is © The Royal Society of Chemistry 2011

View Online
Table 2 Experimental relative kinetic constants for the uncatalyzed
and BSA-catalyzed aldol reactions of aldehydes 1b–e with acetone and
calculated values for the reaction of the same aldehydes with acetone’s
enol and N-methylethenamine
Table 3 Initial rates of formation of 2a in the presence of proteins
a
9
-1 -1
v
obs 10 [mol l s ]
No protein
MBP
IIa103-MBP
HSA
BSA
0.325
0.303
8.73
16.5
9.9
a
b
a
k
rel,unc exp
k
rel,enol calc
k
rel
,
cat exp
krel,enamine calc
1
1
1
1
b
c
d
e
1
7.9
50
73
1
1
4.8
11.8
1
1
3.9
26
88
3.5
64
100
a
◦
2
mM 1a and 200 mM protein in 10% acetone; 37 C.
a
◦
B3LYP/6-311++G(3d,3p)/CPCM; in water at 37 C; from the activation
b
barriers after ZPE correction. BSA.
HSA’s residues 191–294 which contains all the residues of the IIa
binding site (see ESI†). This fragment corresponds approximately
to half the domain II and contains all the disulfide-forming
cysteines. IIa103 has been expressed in E. Coli, in fusion with
27
the maltose binding protein (MBP). Formation of aldol 2a is
accelerated by IIa103–MBP, its effect being comparable to that of
native albumins (Table 3), while MBP alone is devoid of catalytic
activity.
In addition to providing the first example of a functional
albumin subdomain, this result enforces the evidence that the
observed aldolase activity is not due to impurities in commercial
albumin preparations. This is always a concern with albumins, but
can be ruled out in this case because: a) albumins originating
from two different species show similar activity, but opposite
stereoselectivity; b) the catalytic activity is inhibited by warfarin,
a typical albumin ligand, and irreversibly suppressed by modifi-
cation of the active site lysine with fluorescein isothiocyanate; c)
catalytic activity is maintained in the artificial protein IIa103-
MBP obtained from bacteria. In conclusion we have shown
that the aldol addition can also be included in the assortment
of reactions catalyzed by serum albumins. Both bovine and
human proteins catalyze the reaction with a specific enamine
mechanism taking place inside the IIa binding site, and with
opposite stereoselectivity. The finding that aldolase activity is
preserved in the albumin-derived peptide IIa103 opens the way
to the possibility of selecting new aldolase peptides with enhanced
efficiency and stereoselectivity from mutated libraries.
Scheme
2
Model aldol reactions of acetone enol
3
and
N-methylethenamine 4 via cyclic transition states.
mechanism for the spontaneous aldol reaction in water, which
might involve discrete water molecules.
24
The proposed enamine mechanism may also account for the
switch in enantioselectivity from BSA to HSA. The binding sites
of the two proteins are almost superimposable, the only difference
being BSA’s Lys222 and Arg199 which exchange positions in
25
HSA. If we assume that the orientation of the substrate 1a is
dictated by the binding site topology, so that the large aromatic
group occupies the major hydrophobic subsite, then the active site
lysine would be facing opposite sides of the carbonyl in the two
proteins. Accordingly, the enamine should attack the Si face of the
aldehyde in BSA and the Re face in HSA (Fig. 3).
Notes and references
‡
With commercial preparations of antibody 38C2 the reaction was
considerably slower than with albumins and ee was measured at 10%
conversion.
1
2
Modern Aldol Reactions, R. Mahrwald, ed., Wiley-VCH, Berlin, 2004.
L. M. Geary and P. G. Hultin, Tetrahedron: Asymmetry, 2009, 20, 131–
1
73.
3
4
B. M. Trost and C. S. Brindle, Chem. Soc. Rev., 2010, 39, 1600–1632.
S. G. Zlotin, A. S. Kucherenko and I. P. Beletskaya, Russ. Chem. Rev.,
2
009, 78, 737–784.
T. D. Machajewski and C.-H. Wong, Angew. Chem., Int. Ed., 2000, 39,
352–1374.
5
1
Fig. 3 Origin of the stereoselectivity switch in the enamine-mediated
addition of acetone to aldehyde 1a.
6 (a) J. Matsui, I. A. Nicholls, I. Karube and K. Mosbach, J. Org.
Chem., 1996, 61, 5414–5417; (b) D. Carboni, K. Flavin, A. Servant,
V. Gouverneur and M. Resmini, Chem.–Eur. J., 2008, 14, 7059–7065.
7
J. K. Lassila, D. Baker and D. Herschlag, Proc. Natl. Acad. Sci. U. S. A.,
010, 107, 4937–4942.
Finally, we investigated the possibility of obtaining fully func-
tional aldolase peptides starting from the sequence of serum
albumins. Single domains of HSA can be expressed in a functional
2
8 X. Zhu, F. Tanaka, R. A. Lerner, C. F. Barbas III and I. A. Wilson,
J. Am. Chem. Soc., 2009, 131, 18206–18207 and ref. therein.
9 (a) J. Ghuman, P. A. Zunszain, I. Petitpas, A. A. Bhattacharya, M.
Otagiri and S. Curry, J. Mol. Biol., 2005, 353, 38–52; (b) N. Diaz, D.
Su a´ rez, T. L. Sordo and K. M. Merz, Jr., J. Am. Chem. Soc., 2001, 123,
7574–7583.
26
form, but we reasoned that shorter peptides might also be stable,
provided they contain essential structural features. We have thus
identified a 103 aminoacid sequence (IIa103) corresponding to
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4417–4420 | 4419

View Online
1
1
0 (a) G. Klein and J.-L. Reymond, Bioorg. Med. Chem. Lett., 1998, 8,
16 (a) F. Tanaka and C. F. Barbas III, Chem. Commun., 2001, 769–770;
(b) F. Tanaka, R. Fuller and C. F. Barbas III, Biochemistry, 2005, 44,
7583–7592.
17 I. Fitos, J. Visy and J. Kardos, Chirality, 2002, 14, 442–448.
18 E. Coombs and D. P. Evans, J. Chem. Soc., 1940, 1295–1300.
19 D. S. Noyce and W. A. Pryor, J. Am. Chem. Soc., 1959, 81, 618–
620.
1
113–1116; (b) F. Badalassi, D. Wahler, G. Klein, P. Crotti and J.-L.
Reymond, Angew. Chem., Int. Ed., 2000, 39, 4067–4070.
1 (a) R. P. Taylor, V. Chau, C. Bryner and S. Berga, J. Am. Chem. Soc.,
1
975, 97, 1934–1943; (b) R. P. Taylor, V. Chau, C. Bryner and S. Berga,
J. Am. Chem. Soc., 1975, 97, 1943–1948; (c) R. P. Taylor, J. Am. Chem.
Soc., 1976, 98, 2684–2686; (d) R. P. Taylor and A. Silver, J. Am. Chem.
Soc., 1976, 98, 4650–4651.
20 T. Kudo, T. Hyashida, S. Ikadake and H. Yamataka, J. Org. Chem.,
1
2 (a) F. Hollfelder, A. J. Kirby and D. S. Tawfik, Nature, 1996, 383, 60–63;
2005, 70, 5157–5163.
(
1
b) K. Kikuchi, S. N. Thorn and D. Hilvert, J. Am. Chem. Soc., 1996,
21 H. Shulman, C. Makarov, A. K. Ogava, F. Romesberg and E. Keinan,
J. Am. Chem. Soc., 2000, 122, 10743–10753.
18, 8184–8185; (c) F. Hollfelder, A. J. Kirby, D. S. Tawfik, K. Kikuchi
and D. Hilvert, J. Am. Chem. Soc., 2000, 122, 1022–1029; (d) L. C.
James and D. S. Tawfik, Protein Sci., 2001, 10, 2600–2607; (e) Y. Hu,
K. N. Houk, K. Kikuchi, K. Hotta and D. Hilvert, J. Am. Chem. Soc.,
22 (a) P. Model, L. Ponticorvo and D. Rittemberg, Biochemistry, 1968, 7,
1339–1347; (b) E. J. Heron and R. M. Caprioli, Biochemistry, 1974, 13,
4371–4375.
2
004, 126, 8197–8205; (f) G. Boucher, S. Robin, V. Fargeas, T. Dintinger,
M. Math e´ -Allainmat, J. Lebreton and C. Tellier, ChemBioChem, 2005,
, 807–810.
23 X. Zhang and K. N. Houk, J. Org. Chem., 2005, 70, 9712–
9716.
6
24 T. J. Pickerson and K. D. Janda, J. Am. Chem. Soc., 2002, 124, 3220–
3221.
1
3 C. F. Barbas III, A. Heine, G. Zhong, T. Hoffmann, S. Gramatikova,
R. Bj o¨ rnstedt, B. List, J. Anderson, E. A. Stura, I. A. Wilson and R. A.
Lerner, Science, 1997, 278, 2085–2092.
25 For the crystal structure of the warfarin-HSA complex see ref. 9a, for
a homology model of BSA see ref. 14.
1
1
4 F. Berti, S. Bincoletto, I. Donati, G. Fontanive, M. Fregonese and F.
Benedetti, Org. Biomol. Chem., 2011, 9, 1987–1999.
5 T. Hoffmann, G. Zhong, D. Shabat, J. Anderson, S. Gramatikova, R.
Lerner, C. F. Barbas and III, J. Am. Chem. Soc., 1998, 120, 2768–2779.
26 M. Dockal, D. C. Carter and F. R u¨ cker, J. Biol. Chem., 1999, 274,
29303–29310.
27 E. Azzoni, P. Curto, I. Luisi, R. Marzari, D. Sblattero, unpublished
results.
4
420 | Org. Biomol. Chem., 2011, 9, 4417–4420
This journal is © The Royal Society of Chemistry 2011