Importance of tetrahedral intermediate formation in the catalytic mechanism of the serine proteases chymotrypsin and subtilisin

Article abstract of DOI:10.1021/bi300688k

Two new inhibitors in which the terminal α-carboxyl groups of Z-Ala-Ala-Phe-COOH and Z-Ala-Pro-Phe-COOH have been replaced with a proton to give Z-Ala-Ala-Phe-H and Z-Ala-Pro-Phe-H, respectively, have been synthesized. Using these inhibitors, we estimate that for α-chymotrypsin and subtilisin Carlsberg the terminal carboxylate group decreases the level of inhibitor binding 3-4-fold while a glyoxal group increases the level of binding by 500-2000-fold. We show that at pH 7.2 the effective molarities of the catalytic hydroxyl group of the active site serine are 41000-229000 and 101000-159000 for α-chymotrypsin and subtilisin Carlsberg, respectively. It is estimated that oxyanion stabilization and the increased effective molarity of the catalytic serine hydroxyl group can account for the catalytic efficiency of the reaction. We argue that substrate binding induces the formation of a strong hydrogen bond or low-barrier hydrogen bond between histidine-57 and aspartate-102 that increases the pK_a of the active site histidine, allowing it to be an effective general base catalyst for the formation of the tetrahedral intermediate and increasing the effective molarity of the catalytic hydroxyl group of serine-195. A catalytic mechanism for acyl intermediate formation in the serine proteases is proposed.

Full text of DOI:10.1021/bi300688k

Article
pubs.acs.org/biochemistry
Importance of Tetrahedral Intermediate Formation in the Catalytic
Mechanism of the Serine Proteases Chymotrypsin and Subtilisin
Teodolinda Petrillo, Catrina A. O’Donohoe, Nicole Howe, and J. Paul G. Malthouse*
UCD School of Biomolecular and Biomedical Science, UCD Centre for Synthesis and Chemical Biology, SEC Strategic Research
Cluster, Conway Institute, University College Dublin, Belfield, Dublin 4, Ireland
ABSTRACT: Two new inhibitors in which the terminal α-
carboxyl groups of Z-Ala-Ala-Phe-COOH and Z-Ala-Pro-Phe-
COOH have been replaced with a proton to give Z-Ala-Ala-
Phe-H and Z-Ala-Pro-Phe-H, respectively, have been synthe-
sized. Using these inhibitors, we estimate that for α-
chymotrypsin and subtilisin Carlsberg the terminal carboxylate
group decreases the level of inhibitor binding 3−4-fold while a
glyoxal group increases the level of binding by 500−2000-fold.
We show that at pH 7.2 the effective molarities of the catalytic hydroxyl group of the active site serine are 41000−229000 and
101000−159000 for α-chymotrypsin and subtilisin Carlsberg, respectively. It is estimated that oxyanion stabilization and the
increased effective molarity of the catalytic serine hydroxyl group can account for the catalytic efficiency of the reaction. We argue
that substrate binding induces the formation of a strong hydrogen bond or low-barrier hydrogen bond between histidine-57 and
aspartate-102 that increases the pKa of the active site histidine, allowing it to be an effective general base catalyst for the formation
of the tetrahedral intermediate and increasing the effective molarity of the catalytic hydroxyl group of serine-195. A catalytic
mechanism for acyl intermediate formation in the serine proteases is proposed.
he catalytic residues serine-195 and histidine-57 of
chymotrypsin were first identified by chemical modifica-
inhibitors to provide insights into how the serine proteases
promote tetrahedral intermediate formation.
T
tion studies.1−3 The third residue of the catalytic triad,
aspartate-102, was identified by X-ray crystallographic studies.4
Kinetic experiments using a highly reactive p-nitrophenol
substrate showed a rapid release of the p-nitrophenol product
that was stoichiometric with enzyme, which was followed by a
slow turnover the substrate.5 This provided the first evidence of
catalysis proceeding via an acyl intermediate. The minimal
kinetic scheme for catalysis involves formation of a Michaelis−
Menten complex (ES), an acyl intermediate (ES′), and the first
product (P1), and hydrololysis of the acyl intermediate to give
the free enzyme (E) and the second product (P2). The
formation and breakdown of the acyl intermediate are thought
to proceed via a tetrahedral intermediate. It is generally
accepted that the rate-limiting step in catalysis is either the rate
of breakdown or the rate of formation of the tetrahedral
intermediate involved in acyl intermediate formation or
breakdown. Therefore, the catalytic efficiency of the serine
proteases is expected to depend on their ability to catalyze
tetrahedral intermediate formation and breakdown. One way
they can do this is by transition state stabilization6 of the
tetrahedral intermediate. Therefore, to understand catalysis, we
should study the transition state stabilization of the tetrahedral
intermediate. However, it is not feasible to study such
tetrahedral intermediates because they do not accumulate
during catalysis with peptide substrates.7 Inhibitors that are able
to react with chymotrypsin to form transition state analogues
mimicking the catalytic tetrahedral intermediate offer a practical
method for studying tetrahedral intermediate stabilization in
the serine proteases. In this study, we use peptide glyoxal
Specific substrate-derived peptide glyoxals have been shown
to be extremely potent competitive inhibitors of the serine
proteases chymotrypsin8−10 and subtilisin.11,12 Using 18O and
2H isotope shifts, it has been shown that the active site serine
hydroxyl group forms a hemiketal with the glyoxal keto
group.13 Therefore, the glyoxal keto carbon should be in the
same position as the peptide carbon of the hydrolyzed peptide
bond of an analogous substrate. Also, the hemiketal that is
formed when the catalytic serine hydroxyl group reacts with the
keto carbon of the glyoxal should be a good analogue of the
tetrahedral intermediate formed during catalysis. The resonance
stabilization of the peptide bond is lost upon formation of
catalytic tetrahedral intermediates from peptide substrates. This
makes tetrahedral intermediate formation energetically unfav-
orable, so the amount of tetrahedral intermediate formed
during catalysis will be extremely small.14 Therefore, it is not
possible study the tetrahedral intermediates formed during
catalysis by techniques such as nuclear magnetic resonance
(NMR).15,16 However, the keto carbons of glyoxal inhibitors
readily form tetrahedral adducts with both chymotrypsin8,9 and
subtilisin.11,12 Oxyanion formation can be followed by 13C
NMR,17,18 and this has made it possible to quantify oxyanion
stabilization in both chymotrypsin− and subtilisin−glyoxal
inhibitor complexes.8,9,11,12 However, it was not possible to
Received: May 25, 2012
Revised: June 24, 2012
Published: July 3, 2012
© 2012 American Chemical Society
6164
dx.doi.org/10.1021/bi300688k | Biochemistry 2012, 51, 6164−6170

Biochemistry
Article
quantify the role of hemiketal formation in glyoxal inhibitor
binding. This is because in addition to an inhibitor glyoxal
group forming a hemiketal with the active site serine hydroxyl
group its peptide chain can also bind in secondary subsites S1−
S4. To estimate how these two processes contribute to inhibitor
binding, we have studied the binding of two derivatives of the
N-protected tripeptides Z-Ala-Ala-Phe-COOH and Z-Ala-Pro-
Phe-COOH (Z = benzyloxycarbonyl). The glyoxal derivatives
were formed by converting the peptide carboxyl group to a
glyoxal group, and the Z-Ala-Ala-Phe-H and Z-Ala-Pro-Phe-H
derivatives were formed by replacing the peptide carboxyl
group with a hydrogen atom. This has also allowed us to
determine how the C-terminal α-carboxylate groups of the N-
protected tripeptides contribute to binding. The mechanistic
significance of these results is discussed.
was determined. Stock solutions of substrate and inhibitor were
dissolved in dimethyl sulfoxide. The same procedure was used
with subtilisin except that the substrate suc-Ala-Ala-Pro-Phe-
pNA was used.
Ki values were estimated when [S0] ≪ KM. Therefore, the
equation for competitive inhibition {d[P]/dt = (kcat[E][S])/
[[S] + KM(1 + [I]/Ki)]} is reduced to d[P]/dt = (kcat/
KM)[E][S]Ki/([I] + Ki). Ki values were estimated by using a
nonlinear least-squares regression program.18
RESULTS
■
Role of the α-Carboxylate Group in Peptide Binding
by α-Chymotrypsin and Subtilisin Carlsberg. The Ki
values for inhibition of α-chymotrypsin by Z-Ala-Ala-Phe-
COOH (Chart 1B) and Z-Ala-Pro-Phe-H (Chart 1C) were
a
EXPERIMENTAL PROCEDURES
Materials. All materials were obtained from Sigma-Aldrich
Chemical Co. (Gillingham, Dorset, U.K.).
Chart 1. Structures of Inhibitors
■
Inhibitor Synthesis. The peptide-derived glyoxal inhibitors
Z-Ala-Pro-Phe-COCHO and Z-Ala-Ala-Phe-COCHO were
synthesized as described by Djurdjevic-Pahl et al.8 and
Cosgrove et al.19 Z-Ala-Pro-Phe-H and Z-Ala-Ala-Phe-H were
synthesized by coupling 2-phenethylamine to either Z-Ala-Ala
or Z-Ala-Pro as required, using 1-ethyl-3-[3-(dimethylamino)-
propyl]carbodiimide hydrochloride (EDCI·HCl) as a coupling
reagent.20,21
NMR Spectra of Z-Ala-Ala-Phe-H and Z-Ala-Pro-Phe-
H. 13C NMR analysis of Z-Ala-Ala-Phe-H gave the folowing
data (75.475 MHz, DMSO-d6): δ 18.11 (1C, CHCH3), 18.49
(1C, CHCH3), 35.01 (1C, C6H5CH2), 40.14 (1C,
C6H5CH2CH2), 48.09 (1C, CHCH3), 50.06 (1C, CHCH3),
65.37 (1C, O-CH2Ph), 126.08−128.68 (10C, CHCH),
137.02−139.33 (2C, CHC), 155.74 (1C, O-CO-NH),
171.92−172.00 (2C, CONH). 13C NMR analysis of Z-Ala-Pro-
Phe-H gave the following data (trans form, 83.5%) ([2H6]-
DMSO): δ 16.85 (1C, CHCH3), 24.41 (1C, CHCH2CH2CH2),
29.23 (1C, CHCH2CH2CH2), 35.13 (1C, C6H5CH2CH2),
46.57 (1C, CHCH2CH2CH2), 48.15 (1C, CHCH3), 40.20 (1C,
C6H5CH2CH2), 59.65 (1C, CHCH2CH2CH2), 65.34 (1C,
C6H5CH2), 126.02−128.68 (10C, CHCH), 137.07−139.45
(2C, CHC), 155.69 (1C, O-CO-NH), 170.84 and 171.43
(2C, CO-NH). The amount of the cis form present (16.5%)
was determined by quantifying the signals from the trans β
(29.2 ppm) and γ (24.4 ppm) proline carbons and the cis β
(31.31 ppm) and cis γ (21.62 ppm) carbons.
Enzyme Solutions. α-Chymotrypsin and subtilisin Carls-
berg were obtained from Sigma as crystallized and lyophilized
powders. The amounts of fully active chymotrypsin (83%),
subtilisin Carlsberg (45−65%), and subtilisin BPN (73%) were
determined as described by Finucane et al.22 and O’Connell et
al.23
Inhibition of Chymotrypsin and Subtilisin Carlsberg
by Z-Ala-Ala-Phe-COOH and Z-Ala-Ala-H. The inhibition
of the chymotrypsin-catalyzed hydrolysis of suc-Phe-pNA or
suc-Ala-Ala-Pro-Phe-pNA was studied at 25 °C in 3 mL
cuvettes containing 0.1 M buffers [pH 7.1−7.3 (potassium
phosphate), 3.2 (sodium formate), and 10.6 (potassium
carbonate)] and 3.3% (v/v) dimethyl sulfoxide. The initial
rate of hydrolysis of suc-Phe-pNA (pH 7.1−7.3 and 10.6) or
suc-Ala-Ala-Pro-Phe-pNA (pH 3.2) was followed by the
measuring the release of p-nitroaniline (E410 = 8800 M−1cm−1
24) over a 5−15 min period. The pH of the reaction mixture
a
R is either ZAP (Z-Ala-Pro-) or ZAA (Z-Ala-Ala-).
determined by determining how they inhibited the chymo-
trypsin-catalyzed hydrolysis of suc-Phe-pNA when the substrate
concentration was nonsaturating (Figure 1). The Ki values with
subtilisin were measured in the same way except that suc-Ala-
Ala-Pro-Phe-pNA was used as a substrate.
Replacing the α-carboxyl group of Z-Ala-Ala-Phe-COOH
and Z-Ala-Pro-Phe-COOH with a hydrogen atom to give Z-
Ala-Ala-Phe-H and Z-Ala-Pro-Phe-H, respectively (Chart 1C),
led to 3−4-fold tighter binding at pH 7 (Table 1). Therefore,
the presence of an α-carboxylate group at pH 7 decreases the
effectiveness of peptide binding. This result is expected because
for catalytic efficiency we would expect that the enzyme will
bind the peptide products of catalysis less effectively than the
parent peptide of the substrate. Therefore, when α-chymo-
trypsin and subtilisin catalyze the hydrolysis of the peptide
substrates, the α-carboxylate of the peptide product will help
promote its dissociation from the enzyme. This shows that the
6165
dx.doi.org/10.1021/bi300688k | Biochemistry 2012, 51, 6164−6170

Biochemistry
Article
replaced with a hydrogen atom to give Z-Ala-Ala-Phe-H and
Z-Ala-Pro-Phe-H, respectively (Chart 1C), the Ki values of the
inhbitors were ∼3 orders of magnitude larger than those of the
corresponding glyoxal inhibitor (Table 1).
NMR studies have shown that at pH ∼7 the active site
catalytic serine residues of both chymotrypsin8,9,13 and
subtilisin11,12 form negatively charged hemiketals with the
glyoxal inhibitors Z-Ala-Ala-Phe-COCHO and Z-Ala-Pro-Phe-
COCHO (Chart 2). With subtilisin Carlsberg and Z-Ala-Ala-
Chart 2. Negatively Charged Hemiketal Formed When
Glyoxal Inhibitors React with the Serine Proteases
Phe-COCHO, an equal amount of the fully hydrated form of
the structure in Chart 2 is also formed.12 Therefore, while the
negatively charged carboxylate group decreases binding
efficiency, the formation of a negatively charged hemiketal
with a glyoxal inhibitor facilitates binding. This tight binding
demonstrates that hemiketal formation is energeticaly favored
and confirms that the negatively charged hemiketal is a good
transition state analogue of the catalytic tetrahedral inter-
mediate.
Scheme 1 represents the minimal scheme that we have used
to analyze hemiketal formation when glyoxal inhibitors interact
with chymotrypsin or subtilisin. In this model, KH1 is the
hydration constant for the glyoxal inhibitor (KH1 = hydrate/
keto group = [GH]/[G]), Ks is the dissociation constant [Ks =
([E][G])/[EG]] of the noncovalent enzyme−glyoxal inhibitor
complex (EG), and KHK(obs) is the observed equilibrium
constant for hemiketal formation [KHK(obs) = [hemiketal]/
[EG]]. The observed binding constant Ki(obs) = [E]([G] +
[GH])/([EG] + [hemiketal]).
Figure 1. Inhibition of the α-chymotrypsin-catalyzed hydrolysis of suc-
Phe-pNA by Z-Ala-Pro-Phe-H (ZAPF-H) and Z-Ala-Pro-Phe-COOH
(ZAPF-COOH). All samples contained 3.3% dimethyl sulfoxide and
0.1 M potassium phosphate buffer ( pH 7.2). (A) For inhibition by
ZAPF-COOH (10 data points), the α-chymotrypsin and suc-Phe-pNA
concentrations were 5.2 and 54 μM, respectively. The solid line was
calculated using the equation d[P]/dt = [(kcat/KM)[E][S]Ki]/([I] +
Ki) and the fitted values of 14.3 0.4 M−1 s−1 and 233 24 μM for
kcat/KM and Ki, respectively. (B) For inhibition by ZAPF-H (9 data
points), the α-chymotrypsin and suc-Phe-pNA concentrations were 5
and 54 μM, respectively. The solid line was calculated as described for
panel A using the fitted values of 17.9 0.5 M−1 s−1 and 84.2 6.4
μM for kcat/KM and Ki, respectively. All errors are the standard errors
obtained on fitting the experimental data.
α-carboxylate group does not make a positive contribution to
binding in α-chymotrypsin and subtilisin.
Binding of Glyoxal Inhibitors. In contrast to the results
described above, when the glyoxal groups of Z-Ala-Ala-Phe-
COCHO or Z-Ala-Pro-Phe-COCHO (Chart 1A) were
Therefore
Table 1. Binding of Inhibitors to Chymotrypsin and Subtilisin Carlsberg
a
Ki(obs) (μM)
enzyme
pH
ZAAF-H
ZAAF-COOH
ZAAF-COCHO
KiZAAF‑H/KiZAAF‑COOH
KiZAAF‑H/KiZAAF‑COCHO
b
α-chymotrypsin
7.2
7.1
166
9
532 50
419 32
Ki(obs) (μM)
0.365
0.31
0.23
455
subtilisin Carlsberg
97.6 12.6
0.0876 0.016
1110
a
enzyme
pH
ZAPF-H
ZAPF-COOH
ZAPF-COCHO
KiZAPF‑H/KiZAPF‑COOH
KiZAPF‑H/KiZAPF‑COCHO
b
α-chymotrypsin
7.2
7.2
77.4 6.4
241
ND
7
0.0335
0.32
2310
1600
c
c
subtilisin Carlsberg
1600 600
1.00 0.03
ND
a
b
c
Errors are the standard deviations of three determinations. From ref 9. Not determined.
6166
dx.doi.org/10.1021/bi300688k | Biochemistry 2012, 51, 6164−6170

Biochemistry
Article
([glyoxal][H2O])]. Therefore, the KHK(obs)/KH1 ratio gives the
molarity of water required to be as effective as the enzyme’s
serine hydroxyl group in hemiketal formation. At pH 7.2 with
Z-Ala-Ala-Phe-COCHO, the effective concentration of the
active site serine hydroxyl group of chymotrypsin relative to a
water hydroxyl group is 1170/0.0284 = 41200 M. For subtilisin
Carlsberg at pH 7.1, the effective concentration is even larger
(2870/0.0284 = 101000 M). These correspond to stabilizations
of 26.3 and 28.6 kJ/mol for Z-Ala-Ala-Phe-COCHO with
chymotrypsin and subtilisin Carlsberg, respectively (Table 2).
With Z-Ala-Pro-Phe-COCHO at pH 7.2, the effective
concentration of the active site serine hydroxyl group relative
to a water hydroxyl group is larger for chymotrypsin (229000
M) and subtilisin Carlsberg (159000 M). These correspond to
stabilizations of 30.6 and 29.7 kJ/mol for Z-Ala-Pro-Phe-
COCHO with chymotrypsin and subtilisin Carlsberg, respec-
tively (Table 2). For subtlisin BPN, the effective concentration
of the active site serine hydroxyl group is ∼20 times lower than
that for subtilisin Carlsberg (Table 2).
Scheme 1. Minimal Scheme for Hemiketal Formation by the
Serine Proteases
Ki(obs) = Ks[1 + KH1(obs)]/[1 + KHK(obs)
]
(1)
(2)
KHK(obs) = {Ks[1 + KH1(obs)] − Ki(obs)}/Ki(obs)
The hydration constant [KH1(obs)] of the keto carbon of the
glyoxal group of Z-Ala-Ala-Phe-COCHO was estimated by
quantitative 13C NMR to be 1.58 0.05. It was assumed that
the Ki(obs) value for Z-Ala-Ala-Phe-H was a good approximation
for Ks, the disassociation constant for the noncovalent
enzyme−glyoxal inhibitor complex (EG) in Scheme 1.
Effect of pH on Hemiketal Formation. From pH 7.2 to
3.2, there is an ∼17-fold decrease in the level of binding of Z-
Ala-Pro-Phe-COCHO and a similar ∼5-fold decrease in the
level of binding of Z-Ala-Pro-Phe-H (Table 3). This
Table 3. Effect of pH on Hemiketal Formation in Glyoxal
Therefore, using the values of Ki(obs) for Z-Ala-Ala-Phe-
COCHO and Z-Ala-Ala-Phe-H (Table 2), we calculated the
KHK(obs) values for the glyoxal inhibitors (Table 2). For
chymotrypsin and subtilisin, the KHK(obs) values were 1170 and
2870, respectively, for ZAAF-COCHO at pH 7 (Table 2).
The hydration constant [KH1(obs)] of the keto carbon of the
Inhibitor Complexes with Chymotrypsin
a
Ki(obs) (μM)
pH
ZAPF-H
ZAPF-COCHO
KHK(obs)
b
3.2
7.2
381 41
77.4 6.4
242 16
0.581
1494
5267
b
0.0335
glyoxal group of Z-Ala-Pro-Phe-COCHO is 1.28
0.05.13
b
10.6
0.201
2744
a
b
Therefore, we could estimate KHK(obs) for Z-Ala-Pro-Phe-
COCHO inhibiting both chymotrypsin and subtilisin. In this
case, even larger values of KHK(obs) were obtained for
chymotrypsin and subtilisin Carlsberg (Table 2). However,
the hydration constant [KH1(obs)] is not a true thermodynamic
constant and must be divided by the water concentration to
give the correct thermodynamic association constant (KH1). For
the hydrates of Z-Ala-Pro-Phe-COCHO and Z-Ala-Ala-Phe-
COCHO, the values of KH1 were 0.0230 and 0.0284 M−1,
respectively (Table 2).
Effective Molarity of the Catalytic Serine in α-
Chymotrypsin and Subtilisin. Hemiketal formation within
the enzyme−inhibitor complex is a unimolecular process
[KHK(obs) = [hemiketal]/[EG]], while the hydration of the
unbound glyoxal inhibitor is a bimolecular process involving the
reaction of water with the glyoxal group [KH1 = [hemiketal]/
Errors are the standard deviations of three determinations. From ref
9.
demonstrates that protonation of the hemiketal oxyanion
weakens binding by only a small amount (∼3-fold). The fact
that the Ki values for both Z-Ala-Pro-Phe-COCHO and Z-Ala-
Pro-Phe-H increase from pH 7.2 to 10.6 by similar factors of
∼6 and ∼3, respectively (Table 3), supports the earlier
suggestion9 that this is due to a conformational change
resulting from the ionization of the ion pair between
isoleucine-16 and aspartate-194 in chymotrypsin.
How the Effective Molarity of the Catalytic Serine and
Oxyanion Stabilization Contribute to Catalysis. During
catalysis by chymotrypsin, formation of the acyl intermediate is
expected to proceed via a tetrahedral intermediate (Scheme 2)
Table 2. Effective Molarities of the Catalytic Serine of the Serine Proteases upon Formation of Hemiketals with Glyoxal
Inhibitors
a
Ki(obs) (μM)
enzyme
pH
ZAAF-H
ZAAF-COCHO
KHK(obs)
KH1 (M−1
)
effective molarity (M)
ΔG at 25 °C (kJ/mol)
b
α-chymotrypsin
7.2
7.1
166
9
0.365
1170
2870
0.0284
0.0284
41200
−26.3
−28.6
subtilisin Carlsberg
97.6 12.6
0.0876 0.016
101000
a
Ki(obs) (μM)
enzyme
pH
ZAPF-H
ZAPF-COCHO
KHK(obs)
KH1 (M−1
)
effective molarity (M)
ΔG at 25 °C (kJ/mol)
b
α-chymotrypsin
subtilisin Carlsberg
subtilisin BPN
7.2
7.2
7.0
77.4 6.4
0.0335
5270
3650
192
0.023
0.023
0.023
229000
159000
8390
−30.6
−29.7
−22.4
1600 600
1.00 0.03
d
c
451 47
5.32
a
b
c
d
Errors are the standard deviations of three determinations. From ref 9. From ref 11. pH 7.3.
6167
dx.doi.org/10.1021/bi300688k | Biochemistry 2012, 51, 6164−6170

Biochemistry
Article
its conjugate acid are bound with similar affinities. Both the
oxyanion and its conjugate acid should be hydrogen bonded
within the oxyanion hole.30,31 Tighter binding of the negatively
charged oxyanion relative to its neutral conjugate acid is
expected because hydrogen bonds involving charged groups are
stronger than those involving uncharged groups.32,33 However,
it has been shown that in glyoxal inhibitor complexes the
oxyanion pKa is reduced by ∼5 pKa units.8,9,11,12 It is clear that
this large decrease in pKa cannot be attributed to better binding
of the oxyanion but must reflect destabilization of its conjugate
acid.
Scheme 2. Formation of the Acyl Intermediate by the Serine
Proteases
We have shown that the binding of glyoxal inhibitors to
chymotrypsin9 and subtilisin12 increases the pKa of the active
site histidine to a value of >11. This led to the suggestion9,12
that this increase in pKa would allow the positively charged
imidazolium ring of the active site histidine to lower the
hemiketal oxyanion pKa and also lower the pKa of the active site
serine hydroxyl, increasing its nucleophilicity. The hydroxide
ion is solvated in water and so not very reactive.27 However, in
nonaqueous solvents, its reactivity may be increased by as much
as 14 orders of magnitude.34 Therefore, in the enzyme−
inhibitor complex, the entropic advantage of fixing the reactive
groups within the Michaelis complex35 combined with
desolvation, a lower pKa for the active site serine hydroxyl,
and general base catalysis by the imidazole group of histidine-
57 can easily explain the effective concentration of ∼100000 M
for the active site serine hydroxyl group relative to the hydroxyl
group of water.
analogous to the hemiketal formed when Z-Ala-Ala-Phe-
COCHO or Z-Ala-Pro-Phe-COCHO reacts with chymotrypsin.
It is therefore reasonable to assume that both tetrahedral
species will be formed and stabilized in the same way with the
active site histidine acting as a general base catalyst for
tetrahedral intermediate formation and a general acid catalyst
for tetrahedral intermediate breakdown (Scheme 2). Therefore,
we would expect that the enzyme-catalyzed formation of the
tetrahedral intermediate would be ∼100000 times more
effective than the uncatalyzed reaction in water (Table 2).
It has been determined that the rate of hydrolysis of a
peptide bond in water at 25 °C and pH 7 is 3 × 10−9 s−1.25 In
the enzyme-catalyzed reaction, the greater effective molarity of
the serine hydroxyl group should increase this rate by at least
100000 to a value of 3 × 10−5 s−1. This rate is still considerably
slower than the kcat values usually observed with chymotrypsin
and similar substrates such as Ac-Ala-Pro-Phe-NH2,26 for which
kcat is 2.3 s−1 at pH 8. Therefore, kcat would be expected to be
∼1 s−1 at pH 7 as kcat depends on a pKa of ∼7.
Blow4 first identified the catalytic triad of serine-195,
histidine-57, and aspartate-102, with aspartate-102 forming a
hydrogen bond to Nδ1 of histidine-57. His canonical forms are
still be being incorrectly quoted as suggesting a neutral
aspartate and histidine even though this possibility was not
considered likely by the authors.36 Recently, it has been
proposed that the catalytic aspartate is protonated in
Michaelis−Menten complexes.37 However, subsequent studies
contradict this as they claim to show that aspartate-102 has a
low pKa of ≤1.5 that they propose is compatible with the
reaction-driven ring flip mechanism but not mechanisms that
require a strong hydrogen bond between histidine-57 and
However, we have also shown that in chymotrypsin8,9 and
subtilisin11,12 the pKa for the formation of an oxyanion by the
glyoxal hemiketal is reduced by ∼5 pKa units. If the catalytic
rate is directly proportional to the amount of oxyanion formed,
then this would increase the catalytic rate from 3 × 10−5 to 3
s−1. This is very similar to the expected value of ∼1 s−1.
DISCUSSION
■
It is accepted that for chemical reactions in aqueous solutions
reactions proceed via neutral tetrahedral intermediates.27
However, it is clear from studies with both reversi-
ble8,9,11,12,15,28 and irreversible inhibitors17,18,23,29 that the
serine proteases preferentially stabilize tetrahedral adducts
with a negatively charged oxyanion.
aspartate-102.38 However, H NMR has been used to observe
1
the Nδ1 proton shared between histidine-57 and aspartate-
102.39,40 Bachovchin41 used 15N enrichment of the Nδ1 atom
of histidine-57 to confirm that the NMR signal at 14−18 ppm is
due to the Nδ1 proton of histidine-57. Pioneering studies of
peptidyl trifluoromethyl ketone complexes of chymotrypsin
showed the presence of a 1H NMR signal at 18.7 ppm that was
Protonation of the hemiacetal oxyanion and hemiketal
oxyanion had only a small effect (an 5−17-fold increase in
Ki) on inhibitor binding (Table 3). Therefore, the oxyanion and
Scheme 3. Microscopic Ionization States of Histidine-57 and Aspartate-102 in the Serine Proteases
6168
dx.doi.org/10.1021/bi300688k | Biochemistry 2012, 51, 6164−6170

Biochemistry
Article
Funding
attributed to a hydrogen strongly hydrogen bonded between
histidine-57 and aspartate-102.28 It was proposed that such
proton chemical shifts of ∼16−20 ppm show the presence of
strong hydrogen bonds called low-barrier hydrogen bonds.42
This has been confirmed by D/H fractionation factors and
activation energies43 and also by deuterium44 and tritium45
isotope shifts in peptidyl trifluoromethyl ketone complexes of
chymotrypsin. Therefore, a chemical shift of 16−18 ppm for
Nδ1 of histidine-57 shows that histidine-57 is fully protonated
and confirms that a strong hydrogen bond exists between
aspartate-102 and Nδ1 of histidine-57.
This work was supported by a grant from the Programme for
Research in Third-Level Institutions (PRTLI-4). Grant
055637/Z/98 from the Wellcome Trust was used to purchase
the NMR spectrometer used in these studies, and funding from
Science Foundation Ireland was used to upgrade the NMR
machine. We thank University College Dublin for research
demonstratorships for C.A.O. and N.H.
Notes
The authors declare no competing financial interest.
We have shown that in glyoxal inhibitor complexes proton
signals at 18.7 and 17.4 ppm can be observed at pH 3.5 and
10.9, respectively, showing that the pKa of the active site
histidine has been increased to a value of >11 (structure B in
Scheme 3) on binding the glyoxal inhibitor.9,12 The fact that
the signal at 18.7 ppm is present at low pH when the oxyanion
is present as its conjugate acid shows that the increase in pKa
must be due to the interaction of aspatate-102 with histidine-57
and not with the oxyanion. However, it has been argued that
because aspartate-102 has a low pKa a strong hydrogen bond
cannot exist between histidine-57 and aspartate-102.38 How-
ever, as a strong hydrogen bond is formed between the
negatively charged carboxylate group and the Nδ1 proton of
the positively charged histidine, then it will be difficult to
protonate the aspartate as this will result in disruption of the
low-barrier hydrogen bond. Consequently, the aspartate is
expected to have a low pKa of <1.5 (structure B in Scheme 3) as
recently reported.38 Therefore, when inhibitors or substrates
are bound to chymotrypsin, the zwitterionic species (structure
B in Scheme 3) is the predominant species formed from pH 1.5
to 11.The strong hydrogen bond between the aspartate and the
histidine shows that reaction-driven ring flip mechanism is not
possible. The fact that the low-barrier hydrogen bond is present
even when the oxyanion has been protonated suggests that the
binding of inhibitors and substrates must increase the pKa of
histidine-57. This, as our work shows, can increase the effective
molarity of the active site hydroxyl group by more than 5 orders
of magnitude (Table 2). The catalytic serine is expected to have
a pKa of ∼15, so the increase in the pKa of the active site
histidine will allow it to enhance the nucleophilicity of the
serine hydroxyl group so that it can react at the substrate
peptide carbonyl to form a tetrahedral intermediate. The
strengthening of the hydrogen bond between histidine-57 and
aspartate-102 may result from a decrease in the dielectric
constant due to the substrate excluding water from the active
site17 and or by the induction of steric compression between
histidine-57 and aspartate-102.46 Therefore, when substrate
binds, water is excluded from the active site and the pKa of the
histidine is increased, allowing it to abstract the proton from
the serine hydroxyl, which then reacts with the peptide
carbonyl of the substrate to form a tetrahedral intermediate
(Scheme 2). A small reorientation of the imidazole ring47 will
then allow histidine-57 to act as a general acid catalyst
protonating the nitrogen of the leaving group (Scheme 2),
which is expected to have a pKa of ∼10.17,48 The formation of
the acyl intermediate will then complete the first half of the
catalytic cycle.
REFERENCES
■
(1) Ong, E. B., Shaw, E., and Schoellmann, G. (1965) The
identification of the histidine residue at the active centre of
chymotrypsin. J. Biol. Chem. 240, 694−698.
(2) Schaffer, N. K., May, S. C., Jr., and Summerson, W. H. (1953)
Serine phosphoric acid from diisopropylphosphoryl chymotrypsin. J.
Biol. Chem. 202, 67−76.
(3) Schoellmann, G., and Shaw, E. (1963) Direct evidence for the
presence of histidine in the active centre of chymotrypsin. Biochemistry
2, 252−255.
(4) Blow, D. M., Birktoft, J. J., and Hartley, B. S. (1969) Role of a
buried acid group in the mechanism of action of chymotrypsin. Nature
221, 337−340.
(5) Hartley, B. S., and Kilby, B. A. (1954) The Reaction of p-
nitrophenyl Esters with Chymotrypsin and Insulin. Biochem. J. 56,
288−297.
(6) Wolfenden, R. (1976) Transition state analog inhibitors and
enzyme catalysis. Annu. Rev. Biophys. Bioeng. 5, 271−306.
(7) Fastrez, J., and Fersht, A. R. (1973) Mechanism of
Chymotrypsin: Structure, Reactivity, and Nonproductive Binding
Relationships. Biochemistry 12, 1067−1074.
(8) Djurdjevic-Pahl, A., Hewage, C., and Malthouse, J. P. G. (2002) A
13C-NMR study of the inhibition of δ-chymotrypsin by a tripeptide-
glyoxal inhibitor. Biochem. J. 362, 339−347.
(9) Spink, E., Cosgrove, S., Rogers, L., Hewage, C., and Malthouse, J.
P. G. (2007) 13C and H NMR studies of ionizations and hydrogen
1
bonding in chymotrypsin-glyoxal inhibitor complexes. J. Biol. Chem.
282, 7852−7861.
(10) Walker, B., McCarthy, N., Healy, A., Ye, T., and McKervey, M.
A. (1993) Peptide glyoxals: A novel class of inhibitor for serine and
cysteine proteinases. Biochem. J. 293, 321−323.
(11) Djurdjevic-Pahl, A., Hewage, C., and Malthouse, J. P. G. (2005)
Ionisations within a subtilisin-glyoxal inhibitor complex. Biochim.
Biophys. Acta 1749, 33−41.
(12) Howe, N., Rogers, L., Hewage, C., and Malthouse, J. P. G.
(2009) Oxyanion and tetrahedral intermediate stabilization by
subtilisin: Detection of a new tetrahedral adduct. Biochim. Biophys.
Acta 1794, 1251−1258.
(13) Spink, E., Hewage, C., and Malthouse, J. P. G. (2007)
Determination of the structure of tetrahedral transition state analogues
bound at the active site of chymotrypsin using 18O and H isotope
2
shifts in the 13C NMR spectra of glyoxal inhibitors. Biochemistry 46,
12868−12874.
(14) Fastrez, J. (1983) On the stability of tetrahedral intermediates
within the active sites of serine and cysteine proteases. Eur. J. Biochem.
135, 339−341.
(15) Malthouse, J. P. G. (2007) 13C- and 1H-NMR studies of
oxyanion and tetrahedral intermediate stabilization by the serine
proteinases: Optimizing inhibitor warhead specificity and potency by
studying the inhibition of the serine proteinases by peptide-derived
chloromethane and glyoxal inhibitors. Biochem. Soc. Trans. 35, 566−
570.
AUTHOR INFORMATION
■
Corresponding Author
(16) Malthouse, J. P. G. (1986) 13C NMR of Enzymes. Prog. Nucl.
Magn. Reson. Spectrosc. 18, 1−60.
*Telephone: +353 1 7166872. Fax: +353 1 7166893. E-mail: j.
paul.g.malthouse@ucd.ie.
6169
dx.doi.org/10.1021/bi300688k | Biochemistry 2012, 51, 6164−6170

Biochemistry
Article
(17) Finucane, M. D., and Malthouse, J. P. G. (1992) A study of the
stabilization of tetrahedral adducts by trypsin and δ-chymotrypsin.
Biochem. J. 286, 889−900.
(18) Malthouse, J. P. G., Primrose, W. U., Mackenzie, N. E., and
Scott, A. I. (1985) 13C NMR Study of the Ionizations within a
Trypsin-Chloromethyl Ketone Inhibitor Complex. Biochemistry 24,
3478−3487.
(19) Cosgrove, S., Rogers, L., Hewage, C., and Malthouse, J. P. G.
(2007) An NMR study of the inhibition of pepsin by glyoxal
inhibitors: Mechanism of tetrahedral intermediate stabilization by the
aspartyl proteinases. Biochemistry 46, 11205−11215.
in vitro evolved substrate-like inhibitor: A refined mechanism of serine
protease action. J. Biol. Chem. 286, 3587−3596.
(38) Everill, P., Sudmeier, J. L., and Bachovchin, W. W. (2012) Direct
NMR observation and pKa determination of the Asp102 side chain in a
serine protease. J. Am. Chem. Soc. 134, 2348−2354.
(39) Robillard, G., and Shulman, R. G. (1972) High Resolution
Nuclear Magnetic Resonance Study of the Histidine-Aspartate
Hydrogen Bond in Chymotrypsin and Chymotrypsinogen. J. Mol.
Biol. 71, 507−511.
(40) Robillard, G., and Shulman, R. G. (1974) High Resolution
Nuclear Magnetic Resonance Studies of the Active Site of
Chymotrypsin. J. Mol. Biol. 86, 519−540.
(41) Bachovchin, W. W. (1985) Confirmation of the assignment of
the low-field proton resonance of serine proteases by using specifically
nitrogen-15 labeled enzyme. Proc. Natl. Acad. Sci. U.S.A. 82, 7948−
7951.
(42) Frey, P. A., Whitt, S. A., and Tobin, J. B. (1994) A Low-Barrier
Hydrogen Bond in the Catalytic Triad of Serine Proteases. Science 264,
1927−1930.
(43) Lin, J., Westler, W. M., Cleland, W. W., Markley, J. L., and Frey,
P. A. (1998) Fractionation factors and activation energies for exchange
of the low barrier hydrogen bonding proton in peptidyl trifluor-
omethyl ketone complexes of chymotrypsin. Proc. Natl. Acad. Sci.
U.S.A. 95, 14664−14668.
(44) Cassidy, C. S., Lin, J., and Frey, P. A. (2000) The deuterium
isotope effect on the NMR signal of the low-barrier hydrogen bond in
a transition-state analog complex of chymotrypsin. Biochem. Biophys.
Res. Commun. 273, 789−792.
(45) Westler, W. M., Frey, P. A., Lin, J., Wemmer, D. E., Morimoto,
H., Williams, P. G., and Markley, J. L. (2002) Evidence for a strong
hydrogen bond in the catalytic dyad of transition-state analogue
inhibitor complexes of chymotrypsin from proton-triton NMR isotope
shifts. J. Am. Chem. Soc. 124, 4196−4197.
(46) Lin, J., Cassidy, C. S., and Frey, P. A. (1998) Correlations of the
basicity of His 57 with transition state analogue binding, substrate
reactivity, and the strength of the low-barrier hydrogen bond in
chymotrypsin. Biochemistry 37, 11940−11948.
(20) Carpino, L. A. (1993) 1-Hydroxy-7-azabenzotriazole. An
efficient peptide coupling additive. J. Am. Chem. Soc. 115, 4397−4398.
(21) Sheehan, J. C., and Ledis, S. L. (1973) Total Synthesis of a
Moncyclic Peptide Lactone Antibiotic, Etamycin. J. Am. Chem. Soc. 95,
875−879.
(22) Finucane, M. D., Hudson, E. A., and Malthouse, J. P. G. (1989)
A
13C-NMR investigation of the ionizations within an inhibitor-α-
chymotrypsin complex: Evidence that both α-chymotrypsin and
trypsin stabilize a hemiketal oxyanion by similar mechanisms. Biochem.
J. 258, 853−859.
(23) O’Connell, T. P., and Malthouse, J. P. G. (1995) A study of the
stabilization of the oxyanion of tetrahedral adducts by trypsin,
chymotrypsin and subtilisin. Biochem. J. 307, 353−359.
(24) Erlanger, B. F., Kokowsky, N., and Cohen, W. (1961) The
Preparation and Properties of Two New Chromogenic Substrates of
Trypsin. Arch. Biochem. Biophys. 95, 271−278.
(25) Kahne, D., and Clark Still, W. (1988) Hydrolysis of a Peptide
Bond in Neutral Water. J. Am. Chem. Soc. 110, 7529−7534.
(26) Bauer, C. A., Thompson, R. C., and Blout, E. R. (1976) The
active centers of Streptomyces griseus protease 3, α-chymotrypsin, and
elastase: Enzyme-substrate interactions close to the scissile bond.
Biochemistry 15, 1296−1299.
(27) Cox, R. A. (2011) A Greatly Under-Appreciated Fundamental
Principle of Physical Organic Chemistry. J. Mol. Sci. 12, 8316−8332.
(28) Liang, T. C., and Abeles, R. H. (1987) Complex of α-
Chymotrypsin and N-Acetyl-L-leucyl-L-phenylalanyl Trifluoromethyl
Ketone: Structural Studies with NMR Spectroscopy. Biochemistry 26,
7603−7608.
(47) Zhou, Y. Z., and Zhang, Y. K. (2011) Serine protease acylation
proceeds with a subtle re-orientation of the histidine ring at the
tetrahedral intermediate. Chem. Commun. 47, 1577−1579.
(48) Komiyama, M., and Bender, M. L. (1979) Do cleavages of
amides by serine proteases occur through a stepwise pathway involving
tetrahedral intermediates? Proc. Natl. Acad. Sci. U.S.A. 76, 557−560.
(29) O’Sullivan, D. B., O’Connell, T. P., Mahon, M. M., Koenig, A.,
Milne, J. J., Fitzpatrick, T. P., and Malthouse, J. P. G. (1999) 13C-NMR
Study of How the Oxyanion pKa Values of Subtilisin and
Chymotrypsin Tetrahedral Adducts Are Affected by Different Amino
Acid Residues Binding in Enzymes Subsites S1−S4. Biochemistry 38,
6187−6194.
(30) Henderson, R. (1970) Structure of Crystalline α-Chymotrypsin.
IV. The Structure of Indoleacryloyl-α-Chymotrypsin and its Relevance
to the Hydrolytic Mechanism of the Enzyme. J. Mol. Biol. 54, 341−
354.
(31) Kraut, J. (1977) Serine Proteases: Structure and Mechanism of
Catalysis. Annu. Rev. Biochem. 46, 331−358.
(32) Fersht, A. R. (1987) The hydrogen bond in molecular
recognition. Trends Biochem. Sci. 12, 301−304.
(33) Fersht, A. R., Shi, J. P., Knill-Jones, J., Lowe, D. M., Wilkenson,
A. J., Blow, D. M., Brick, P., Carter, P., Waye, M. M. Y., and Winter, G.
(1985) Hydrogen bonding and biological specificity analysed by
protein engineering. Nature 314, 235−238.
(34) Dolman, D., and Stewart, R. (1967) Strongly basic systems. VIII.
H-function for dimethyl sulfoxide-water-tetramethylammonium hy-
droxide. Can. J. Chem. 45, 911−924.
(35) Jencks, W. P. (1981) On the attribution and additivity of
binding energies. Proc. Natl. Acad. Sci. U.S.A. 78, 4046−4050.
(36) Blow, D. M. (1997) The tortuous story of Asp···His···Ser:
Structural analysis of α-chymotrypsin. Trends Biochem. Sci. 22, 405−
408.
(37) Wahlgren, W. Y., Pal, G., Kardos, J., Porrogi, P., Szenthe, B.,
Patthy, A., Graf, L., and Katona, G. (2011) The catalytic aspartate is
protonated in the Michaelis complex formed between trypsin and an
6170
dx.doi.org/10.1021/bi300688k | Biochemistry 2012, 51, 6164−6170