PLA2 Substrate Specificity
A R T I C L E S
recently, it has been shown that a selection of only a few key
residues or even truncated residues often can provide a satisfying
model for the real active site.30-32 In line with this approach,
the current study was carried out using a model system, where
the enzyme part consisted of the two most crucial residues
(His47 and Asp48) along with the required calcium ion. The
histidine is believed to deprotonate the attacking water molecule,
whereas the aspartate is responsible for binding the calcium ion.
To limit the computational demand, the phospholipid headgroup
was modeled by a simple methoxy group, and the fatty acid
chains were both truncated to propyl. The system illustrated in
Figure 7 consist of 71 atoms (including Ca2+), which allowed
treatment of the entire system with density functional theory
(B3LYP33-35 /LACVP36) including a SCRF-PCM37,38 solvation
model for water as described in the Supporting Information.
To determine the barrier for the deprotonation of the water
molecule by histidine and subsequent addition to the carbonyl
group, a series of energy minimizations were carried out. In
each calculation, the distance between the carbon atom in the
carbonyl and the oxygen in the water molecule was kept fixed
to either 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 2.2, 2.4, 2.6, or 2.8 Å. For
the shortest distance, the constraint was removed, and for both
substrates, a stable tetrahedral intermediate was obtained. This
illustrates that the model of the active site is capable of
stabilizing this intermediate for the thio-ester (which was
unstable in the absence of the model enzyme), thereby inducing
an associative hydrolysis pathway for this substrate. When
imposing a distance constraint of 2.8 Å, the relative energies
for the two substrates become markedly different with the
O-ester being more than 30 kJ/mol lower in energy as indicated
in Figure 9, which displays the energy profile as the oxygen
atom in the water molecule approaches the carbon atom in the
carbonyl group of the ester.
To identify the inherent conformational preferences for the
two esters, initial conformational searches were carried out using
the OPLS-2005 force-field. The lowest energy conformations
of the two esters were virtually identical, except for the longer
C-S bond (1.8 Å vs 1.6 Å for C-O), as expected. Since the
MD calculations (Vide infra) revealed that both substrates fit
perfectly into the binding pocket and do not impair the incoming,
nucleophilic water molecule, the differences in rate of hydrolysis
must instead be a more direct result of the substitution of oxygen
by sulfur. One example of the importance of this electronic
difference was obtained when the anionic tetrahedral intermedi-
ates were optimized using DFT using an implicit solvation
model for water. Here, it was only possible to obtain a stable
energy minimum for the free O-ester, whereas the free S-ester
spontaneously dissociated upon energy minimization. This is a
strong indication that there is a difference in the inherit
hydrolysis mechanism that each of the two esters follows under
normal, basic aqueous conditions.
The size and flexibility of the investigated systems prevent a
stringent location of the transition state; however, as judged from
the energy profiles, it is around 1.6-1.7 Å for the O-ester, and
at a slightly longer distance for the S-ester at approximately
1.8 Å. In Figure 10, examples of the high-energy structures are
shown.
The water molecule is situated in an intermediate position
with the oxygen atom interacting with the carbonyl carbon of
the respective esters, while one of the hydrogen atoms is
hydrogen-bonded to the histidine. The calculated structures
suggest that the crucial nucleophilic addition involves a
simultaneous deprotonation of the water molecule by the
histidine.22 For both substrates, the tetrahedral intermediate is
a stable energy minimum (Figure 11), and at this point in the
reaction pathway, the difference in energy is 14 kJ/mol.
2.5. DFT Study Using a Model Active Site. As a reference
state, we have chosen the empty active site, the isolated
substrate, and an isolated water molecule. For both substrates
the binding of the substrate in the binding pocket (along with
the water molecule) is exothermic (∼-120-125 kJ/mol) when
using the SCRF-PCM solvation model for water.37,38 However,
the two substrates adopt significantly different conformations
(Figure 8), which result in a much longer distance between the
carbonyl and the water molecule in the S-ester compared to
the O-ester (S-ester: C-O 4.8 Å vs O-ester: 3.4 Å).
For both substrates, the tetrahedral intermediate has a higher
energy than the initial complex of the reaction partners (O-ester:
+48 kJ/mol, S-ester: +65 kJ/mol). We also inspected the
elimination-addition pathway for both substrates by performing
a series of calculations with fixed, increased C-O/S distances
starting from the initial esters. As expected, this is very
unfavorable for the O-ester, but also for the S-ester, this pathway
cannot compete with the enzyme-assisted formation of the
tetrahedral intermediate (see Supporting Information).
After formation of the tetrahedral intermediate, a proton shift
to the ether-oxygen atom of the alcohol part of the ester should
occur. The fidelity of this step was tested by manually moving
the proton from the hydroxyl to the ether oxygen in the
tetrahedral intermediate. Upon optimization, both substrates
spontaneously separated to the acid and alcohol/thiol, as
expected (Figure 12).
(25) Ramos, M. J.; Fernandes, P. A. Acc. Chem. Res. 2008, 41, 689–698.
(26) Claeyssens, F.; Harvey, J. N.; Manby, F. R.; Mata, R. A.; Mulholland,
A. J.; Ranaghan, K. E.; Schutz, M.; Thiel, S.; Thiel, W.; Werner, H. J.
Angew. Chem., Int. Ed. 2006, 45, 6856–6859.
(27) Friesner, R. A.; Guallar, V. Annu. ReV. Phys. Chem. 2005, 56, 389–
427.
(28) Warshel, A. Annu. ReV. Biophys. Biol. 2003, 32, 425–443.
(29) Gao, J. L.; Truhlar, D. G. Annu. ReV. Phys. Chem. 2002, 53, 467–
505.
When comparing the fully energy minimized structures, the
overall reaction was found to be 12 kJ/mol more favorable for
the O-ester compared to the S-ester (-40 and -28 kJ/mol,
respectively). The energy minimized structures again reveal
differences between the O-ester and the S-ester, namely in the
distance between the carbonyl group and the oxygen atom in
the leaving alcohol. As observed earlier for the coordination of
the water molecule, this distance is significantly longer for the
S-ester (4.30 Å) than for the O-ester (3.32 Å), which again
underlines that there are subtle but not negligible differences
between the properties of these two substrates. In the absence
of the enzyme, the O-ester would still follow this mechanism
(addition-elimination through tetrahedral intermediate), whereas
(30) Siegbahn, P. E. M.; Blomberg, M. R. A. Chem. ReV. 2000, 100, 421–
437.
(31) Siegbahn, P. E. M.; Borowski, T. Acc. Chem. Res. 2006, 39, 729–
738.
(32) de Visser, S. P. J. Am. Chem. Soc. 2006, 128, 9813–9824.
(33) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
(34) Becke, A. D. J. Chem. Phys. 1993, 98, 1372–1377.
(35) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. ReV. B 1988, 37, 785–789.
(36) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299–310.
(37) Marten, B.; Kim, K.; Cortis, C.; Friesner, R. A.; Murphy, R. B.;
Ringnalda, M. N.; Sitkoff, D.; Honig, B. J. Phys. Chem. 1996, 100,
11775–11788.
(38) Tannor, D. J.; Marten, B.; Murphy, R.; Friesner, R. A.; Sitkoff, D.;
Nicholls, A.; Ringnalda, M.; Goddard, W. A.; Honig, B. J. Am. Chem.
Soc. 1994, 116, 11875–11882.
9
J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009 12199