Article

X-ray, neutron and NMR studies of the catalytic mechanism of aspartic proteinases

School of Biological Sciences, University of Southampton, Bassett Crescent East, Southampton, SO16 7PX, England.
European Biophysics Journal (Impact Factor: 2.22). 10/2006; 35(7):559-66. DOI: 10.1007/s00249-006-0065-7
Source: PubMed
ABSTRACT
Current proposals for the catalytic mechanism of aspartic proteinases are largely based on X-ray structures of bound oligopeptide inhibitors possessing non-hydrolysable analogues of the scissile peptide bond. Until recent years, the positions of protons on the catalytic aspartates and the ligand in these complexes had not been determined with certainty due to the inadequate resolution of these analyses. There has been much interest in locating the catalytic protons at the active site of aspartic proteinases since this has major implications for detailed understanding of the mechanism of action and the design of improved transition state mimics for therapeutic applications. In this review we discuss the results of studies which have shed light on the locations of protons at the catalytic centre. The first direct determination of the proton positions stemmed from neutron diffraction data collected from crystals of the fungal aspartic proteinase endothiapepsin bound to a transition state analogue (H261). The neutron structure of the complex at a resolution of 2.1 A provided evidence that Asp 215 is protonated and that Asp 32 is the negatively charged residue in the transition state complex. Atomic resolution X-ray studies of inhibitor complexes have corroborated this finding. A similar study of the native enzyme established that it, unexpectedly, has a dipeptide bound at the catalytic site which is consistent with classical reports of inhibition by short peptides and the ability of pepsins to catalyse transpeptidation reactions. Studies by NMR have confirmed the findings of low-barrier and single-well hydrogen bonds in the complexes with transition state analogues.

Full-text

Available from: Raj Gill
Abstract Current proposals for the catalytic mecha-
nism of aspartic proteinases are largely based on X-ray
structures of bound oligopeptide inhibitors possessing
non-hydrolysable analogues of the scissile peptide
bond. Until recent years, the positions of protons on
the catalytic aspartates and the ligand in these com-
plexes had not been determined with certainty due to
the inadequate resolution of these analyses. There has
been much interest in locating the catalytic protons at
the active site of aspartic proteinases since this has
major implications for detailed understanding of the
mechanism of action and the design of improved
transition state mimics for therapeutic applications. In
this review we discuss the results of studies which have
shed light on the locations of protons at the catalytic
centre. The first direct determination of the proton
positions stemmed from neutron diffraction data col-
lected from crystals of the fungal aspartic proteinase
endothiapepsin bound to a transition state analogue
(H261). The neutron structure of the complex at a
resolution of 2.1 A
˚
provided evidence that Asp 215 is
protonated and that Asp 32 is the negatively charged
residue in the transition state complex. Atomic reso-
lution X-ray studies of inhibitor complexes have cor-
roborated this finding. A similar study of the native
enzyme established that it, unexpectedly, has a dipep-
tide bound at the catalytic site which is consistent with
classical reports of inhibition by short peptides and the
ability of pepsins to catalyse transpeptidation reac-
tions. Studies by NMR have confirmed the findings of
low-barrier and single-well hydrogen bonds in the
complexes with transition state analogues.
Keywords Aspartic proteinase Æ Neutron diffraction Æ
Atomic resolution X-ray Æ Catalytic mechanism Æ
Low-barrier hydrogen bond
Introduction
The aspartic proteinases are a family of enzymes in-
volved in a number of important biological processes
(Cooper 2002; Dunn 2002). In animals the enzyme
renin has a hypertensive action through its role in the
rennin–angiotensin system. The retroviral aspartic
proteinases, such as the HIV proteinase, are essential
for maturation of the virus particle and inhibitors have
a proven therapeutic record in the treatment of AIDS.
The enzyme b-secretase has been implicated in amy-
loidosis and the stomach enzyme pepsin is known to be
involved in various gastric disorders. All enzymes
in this class are characteristically inhibited by the
microbial peptide pepstatin A which contains the
unusual amino acid statine. Statine is an analogue
of
L-leucine, differing from this amino acid by the
P. T. Erskine Æ S. Mall Æ R. Gill Æ S. P. Wood Æ
J. B. Cooper (&)
School of Biological Sciences, University of Southampton,
Bassett Crescent East, Southampton SO16 7PX, England
e-mail: J.B.Cooper@soton.ac.uk
D. A. A. Myles
Center for Structural Molecular Biology,
Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge,
TN 37831-6100, USA
L. Coates
Neutron Diffraction Group, Bioscience Division,
Mailstop M888, Los Alamos National Laboratory,
Los Alamos, NM 87545, USA
Eur Biophys J (2006) 35:559–566
DOI 10.1007/s00249-006-0065-7
123
REVIEW
X-ray, neutron and NMR studies of the catalytic mechanism
of aspartic proteinases
Leighton Coates Æ Peter T. Erskine Æ Sanjay Mall Æ
Raj Gill Æ Steve P. Wood Æ Dean A. A. Myles Æ
Jonathan B. Cooper
Received: 28 January 2006 / Revised: 24 March 2006 / Accepted: 4 April 2006 / Published online: 4 May 2006
EBSA 2006
Page 1
insertion of a –CHOH–CH
2
group between the Ca
and the main chain carbonyl group.
Most eukaryotic aspartic proteinases are monomeric
and consist of a single chain of around 330 amino acids
which forms two similar domains with the active site
located in between (Davies 1990). In contrast, retro-
viral aspartic proteinases are dimeric, consisting of two
identical subunits, each roughly equivalent to one do-
main of a eukaryotic aspartic proteinase. Accordingly,
the amino acid sequences of the eukaryotic enzymes
have signs of an internal repeat relating the two halves
of the molecule, their identity being greatest in the
vicinity of the active site which involves two conserved
Asp-Thr-Gly sequences (Tang et al. 1978; Pearl and
Blundell 1984). Hence the eukaryotic aspartic pro-
teinases may have evolved divergently from a primitive
dimeric enzyme (resembling the retroviral proteinase)
by gene duplication and fusion. In all aspartic pro-
teinases, the base of the active site cleft is made of
b-strands which contain the catalytic aspartate residues
(32 and 215). The side chains of the aspartates are held
coplanar and within hydrogen bonding distance by an
intricate arrangement of H-bonds involving surround-
ing main chain and conserved side chain groups. A
solvent molecule is found between both aspartate
carboxyls in all native aspartic proteinase crystal
structures and is presumed to be a water molecule. In
numerous chemical studies, the failure to trap cova-
lently bound substrate indicated that the reaction in-
volves an intermediate which binds non-covalently to
the enzyme (Hofmann et al. 1984). NMR studies using
an inhibitor with a ketone analogue of the scissile bond
suggested that it binds to the enzyme in a hydrated
gem-diol form (–C(OH)
2
–) (Rich et al. 1982). Thus
current proposals for the catalytic mechanism (such as
that shown in Fig. 1) invoke nucleophilic attack of the
active site water molecule on the scissile bond carbonyl
generating a tetrahedral gem-diol intermediate
(Veerapandian et al. 1992).
The best synthetic inhibitors of aspartic protein-
ases are those in which one or both of the hydroxyl
groups of the putative transition state are mimicked
(intermediate 2 in Fig. 1). One hydroxyl binds by
hydrogen bonds to both of the catalytic aspartates in
the same position as the solvent molecule in the
native enzyme and most of the transition state ana-
logues (e.g. statine) mimic this group alone. In con-
trast, fluoroketone analogues (–CO–CF
2
–) mimic
both hydroxyls of the putative intermediate since
they readily hydrate to the gem-diol form (–C(OH)
2
CF
2
–). In current mechanisms, the active site water
molecule becomes partly displaced upon substrate
Fig. 1 The catalytic
mechanism for aspartic
proteinases proposed by
Veerapandian et al. (1992).
This mechanism is based on
the X-ray structure of a
difluoroketone (gem-diol)
inhibitor bound to
endothiapepsin. A water
molecule tightly bound to the
aspartates in the native
enzyme is proposed to
nucleophilically attack the
scissile bond carbonyl. The
resulting tetrahedral
intermediate (2) is stabilised
by hydrogen bonds to the
negatively charged carboxyl
of aspartate 32. Fission of the
scissile C–N bond is
accompanied by transfer of a
proton to the leaving amino
group either from Asp 215
(with nitrogen inversion) or
from bulk solvent. Dashed
lines indicate hydrogen bonds
560 Eur Biophys J (2006) 35:559–566
123
Page 2
binding and polarised by one of the aspartate car-
boxyls, as suggested by Suguna et al. (1987). The
water may then nucleophilically attack the scissile
bond carbonyl group to form the tetrahedral inter-
mediate. Structural comparison of numerous inhibitor
complexes (e.g. Bailey and Cooper 1994) largely
confirmed earlier predictions by Pearl (1987) that the
mechanism involved strain imposed by tight binding
of the substrate at sub-sites adjacent to the catalytic
centre.
Early proposals for the mechanism based on
crystal structures generally invoked proton transfer
between the two inner carboxyl oxygen atoms of the
catalytic diad (Polgar 1987; Pearl 1987). However
inspection of the molecular geometry of the catalytic
carboxyls in numerous high resolution structures
suggests that a direct hydrogen bond interaction be-
tween these two groups would have poor geometry.
Furthermore there are shorter hydrogen bond con-
tacts between the aspartate carboxyls and the active
site water molecule suggesting that the active site
protons are involved in these interactions rather than
in the interaction between the two aspartate car-
boxyls. In the ground state of the enzyme-substrate
complex (shown as intermediate 1 in Fig. 1)itis
assumed that the protons on the active site water
molecule are oriented towards the two aspartates so
that both of the water oxygen lone pairs are pointed
towards the substrate for nucleophilic attack. Thus
in more recent proposals, which are based on the
bound structures of difluoroketone inhibitors (e.g.
Veerapandian et al. 1992; James et al. 1992; Silva
et al. 1996), there are no direct proton transfers
between the two catalytic carboxyl groups and in-
stead all proton transfers are mediated by the
nucleophilic water molecule. These mechanistic pro-
posals were consistent in general terms but lacked
direct experimental evidence for the protonation
states of the active site groups. Since, the active site
hydrogen atoms cannot be located by X-ray analysis
of proteins even at high resolution, their putative
positions have to be inferred from the local geometry
of surrounding polar atoms. One of the key features
of the mechanism proposed by Veerapandian et al.
(1992) (shown in Fig. 1) is the stabilisation of the
transition state by a negative charge localised on Asp
32. The assignment of a negative charge to this res-
idue was made solely on the basis that its hydrogen
bonding capacity is satisfied to a greater extent than
that of Asp 215 in complex with a gem-diol inhibitor.
However, the protonation states of the catalytic as-
partates had not been determined with any certainty
at that stage.
Neutron diffraction studies
Crystals of endothiapepsin bound to the hydroxyeth-
ylene inhibitor (H261) of sufficient size for neutron data
collection were obtained and were partially deuterated
by vapour diffusion (Cooper and Myles 2000). Neutron
Laue data were collected using the LADI detector at
ILL (Grenoble) on large crystals of the H261 inhibitor
complex allowing the structure to be refined at a reso-
lution of 2.1 A
˚
(Coates et al. 2001). This represents one
of the largest protein structures to be refined in
molecular detail using neutron data. The final neutron
dataset had a merging R-factor of 7.5% and a com-
pleteness of 84.5%. The final neutron refinement
R-factor was 23.5% and the R-free was 27.4%, which
are slightly high in comparison with the values expected
for a refined X-ray structure at comparable resolution.
However, they are consistent with R-values obtained in
other analyses which have used neutron Laue data. The
neutron structure superimposes very well with the
previously solved X-ray structure of this complex
(Veerapandian et al. 1990); the rms Ca deviation is only
0.2 A
˚
.
Whilst partial deuteration of the crystal was under-
taken primarily to reduce incoherent scatter from bulk
solvent, the neutron data show that the endothiapepsin
molecule itself has become extensively deuterated by
the vapour diffusion protocol. The parts of the mole-
cule most protected from H-D exchange are the buried
b-strand regions. The majority of polar side chains
have exchanged and, in general, the loops and helical
regions have exchanged to a greater extent than the
b-sheet regions. Many aspartic proteinases have pI
values which are unusually low, e.g. pepsin has a pI
value close to 2.0 (Fruton 1976). It has been suggested
that these low pI values are due to buried carboxylate
groups which remain deprotonated even at very low
pH by interacting with conserved neutral polar resi-
dues. Evidence for this was provided by the neutron
data since there are a number of buried carboxylates in
endothiapepsin which the neutron data show conclu-
sively to be deprotonated (Coates et al. 2001).
The bound inhibitor (H261) possesses a hydroxy-
ethylene analogue (–CHOH–CH
2
–) in place of the
scissile peptide bond. The analogue –OH group mimics
one hydroxyl of the putative tetrahedral intermediate
and replaces the water molecule found at the catalytic
centre of the native enzyme. A neutron difference
Fourier for the structure, without hydrogens or deu-
teriums modelled at the catalytic centre, shows positive
difference density for deuterium atoms on the inhibitor
hydroxyl and the outer carboxyl oxygen of Asp 215
(Fig. 2). To test possible interpretations, the structure
Eur Biophys J (2006) 35:559–566 561
123
Page 3
was refined using different tautomeric models of the
deuterium positions and the resulting maps and deu-
terium occupancies inspected. These results again
strongly suggested that Asp 215 is deuterated (or
protonated in vivo) in the transition state complex and
provided the first direct experimental evidence in fa-
vour of the mechanism proposed by Veerapandian
et al. (1992).
Atomic resolution X-ray analysis
Atomic resolution X-ray data have been collected from
crystals of native endothiapepsin allowing the structure
to be refined at 0.9 A
˚
resolution giving a final R-factor
of 12.1% and free R-value of 14.7% (Erskine et al.
2003). The quality of the map showed conclusively that
one region of the protein involving an -Asp-Gly-
dipeptide had cyclised forming a succinimide. Bio-
chemical studies have shown that -Asp-Gly- sequences
undergo slow succinimide formation when the glycine’s
main chain nitrogen nucleophilically attacks the side
chain carboxyl of the aspartate. Succinimide and sub-
sequently b-aspartic acid formation in proteins is a
characteristic of ageing and it has been suggested that
the accumulation of these unusual residues may be a
triggering factor in amyloidogenesis (Lindner and
Helliger 2001).
The electron density map of the native enzyme at
atomic resolution also demonstrated that a short pep-
tide (probably a Ser-Thr dipeptide) was bound non-
covalently in the active site cleft (Fig. 3) (Erskine et al.
2003). The N-terminal nitrogen of the dipeptide
interacts with the aspartate diad of the enzyme by
hydrogen bonds involving the carboxyl of Asp 215 and
the catalytic water molecule. The two amino acids of
the dipeptide lie in the same positions as the P
1
¢ and P
2
¢
residues of bound inhibitors. In refinement, the occu-
pancy of the Ser-Thr dipeptide converged to a value of
39%, i.e. unlike the tight-binding transition state ana-
logue inhibitors, the dipeptide is not present in all
enzyme molecules in the crystal. This is consistent with
classical findings that the aspartic proteinases can be
inhibited weakly by short peptides and that these en-
zymes can catalyse transpeptidation reactions (Fruton
1976). In this process the ability of the active site cleft
to retain one of the products of the hydrolytic reaction
is clearly important for the subsequent condensation
step involving a second peptide that binds at the active
site. Previously, transpeptidation has been cited
as evidence for a covalently bound intermediate.
However, these results demonstrated that a covalent
Fig. 2 The neutron map at the catalytic centre of the aspartic
proteinase endothiapepsin formed by aspartates 32 and 215.
Green lines indicate the 2F
o
F
c
neutron density at 2.1 A
˚
resolution contoured at 1.2 rms. Positive difference density
(contoured at 2.5 rms is shown with dark blue lines) indicates
that the outer oxygen of Asp 215 is deuterated and that the
inhibitor hydroxyl has a D oriented towards Asp 32 (Coates et al.
2001)(red indicates negative difference density at a contour level
of –2.5 rms). This provided the first experimental evidence for
the mechanism shown in Fig. 1
562 Eur Biophys J (2006) 35:559–566
123
Page 4
intermediate may not be necessary for transpeptidation
to occur.
It is conceivable that the dipeptide could have
originated from autolysis of the enzyme’s N-terminal
Ser-Thr- sequence during crystallisation. The exact
sequence of the dipeptide, as deduced from the elec-
tron density, is not certain but several efforts at
refinement with different amino acids built in to the
density lent faith to the conclusion that it is a Ser-Thr
dipeptide. In addition, the following ‘structurally sim-
ilar’ dipeptides were synthesised to test whether any
possessed inhibitory activity against endothiapepsin:
Ser-Thr, Pro-Thr and Ser-Val (Erskine et al. 2003).
These dipeptides were chosen since they could fit the
electron density for the ligand almost equally well. The
finding that only the Ser-Thr dipeptide was able to
inhibit the enzyme lends further credence to the idea
that this is the ligand observed in the active site al-
though it is conceivable that a mixture of short pep-
tides are bound. A similar finding of a short peptide
bound in the ‘prime’ side of the active site was reported
by Symersky et al. (1997) for the aspartic proteinase
from Candida tropicalis.
As mentioned above, the protonation state of the
catalytic residues has been a long-standing controversy
in the aspartic proteinase field. In the optimal pH
range it is likely that the aspartate diad possesses a
single negative charge. The putative salt-bridge inter-
action made by the active site dipeptide’s amino group
with Asp 215 indicates that this aspartate could be
negatively charged in the ground state of the enzyme.
In general, unrestrained refinement using atomic res-
olution X-ray data provides a powerful means of
defining the protonation states of carboxyl groups in
well-ordered parts of the structure. Neutral carboxyl
groups have a significant difference between their C–
OH and C = O bond lengths (1.21 A
˚
for the C = O
bond and 1.32 A
˚
for the C–OH bond) whereas ionised
carboxylates have identical C–O bond lengths (typi-
cally 1.27 A
˚
) due to resonance. Unrestrained refine-
ment of the carboxyl groups of native endothiapepsin
at 0.9 A
˚
resolution showed that both catalytic aspar-
tates have C–O bond lengths which are almost identi-
cal within the errors of measurement (ESDs = 0.01 A
˚
).
Whilst this suggests that both are ionised, this is un-
likely due to the proximity of both groups. Instead it is
Fig. 3 The 0.91 A
˚
resolution X-ray map of endothiapepsin
showing the dipeptide bound to the ‘native’ enzyme at the
catalytic centre (Erskine et al. 2003). The 2F
o
F
c
density (shown
in green) is contoured at 1.0 rms. The occupancy of this dipeptide
refined to 39%. The dipeptide probably originates from autolysis
of the enzyme’s N-terminal Ser-Thr-dipeptide during crystalli-
sation
Eur Biophys J (2006) 35:559–566 563
123
Page 5
likely that both are equally likely to be charged in the
ground state i.e. Asp 32 and 215 will each be negatively
charged (and able to make a salt-bridge) in only 50%
of the active sites in the crystal at any one time due to
proton transfer. This correlates with the finding that
the occupancy of the dipeptide is less than 50%. Thus,
atomic resolution studies of the native endothiapepsin
enzyme suggest that the distribution of charge at the
catalytic centre in the absence of substrate is sym-
metric—an effect which was first predicted earlier
(Pearl and Blundell 1984) but has not been substanti-
ated until now. However, an asymmetric charge dis-
tribution may arise in the presence of substrate.
To analyse this, X-ray data have been collected to
atomic resolution on a total of seven peptide-mimetic
inhibitors bound to endothiapepsin (Coates et al. 2002a,
b; Erskine et al. 2003). These represent a wide range of
transition state analogues including the hydroxyethyl-
ene analogue (–CHOH–CH
2
–) as well as the statine
(–CHOH–CH
2
–CO–), reduced bond (–CH
2
–NH–),
norstatine (–CHOH–CO–), phosphinate (–PO(OH)–
CH
2
–) and the difluoroketone (–C(OH)
2
–CF
2
–)
analogues. In one structure there is excellent electron
density for a hydrogen atom on the inhibitor hydroxyl
orientated toward the inner oxygen of Asp 32, as shown
in Fig. 4. This X-ray structure represents the most or-
dered of the inhibitor complexes analysed in terms of
average B-factor—something which may stem from
favourable freezing conditions for this particular crystal.
Significantly, the occurrence of a proton at this position
is consistent with the neutron data (Coates et al. 2001)
and unrestrained refinement of the carboxyl bond
lengths (Coates et al. 2002a).
The atomic resolution X-ray data showed that the
transition state analogues of these inhibitors bind by
making several hydrogen bonds with the catalytic
aspartates, some of which have donor–acceptor dis-
tances as short as 2.5 A
˚
and, in the case of the
phosphinate analogue, as short as 2.4 A
˚
. Whilst dis-
tances as short as these had been observed in earlier
analyses at around 2.0 A
˚
resolution, the errors in the
bond lengths of atomic resolution structures are con-
siderably lower (of the order of 0.01 A
˚
). This confirms
that these short hydrogen bonds are a real feature of
Fig. 4 The electron density at 0.98 A
˚
resolution for a hydrogen
atom on a transition state analogue (paler contours indicate
positive difference density) (Coates et al. 2002a). The putative
hydrogen atom on the inhibitor hydroxyl is oriented towards Asp
32. This is consistent with the mechanism shown in Fig. 1
564 Eur Biophys J (2006) 35:559–566
123
Page 6
the enzyme-inhibitor complex and not errors in the
structure analysis. Hydrogen bonds as short as this are
referred to as low barrier hydrogen bonds (LBHB)
since the proximity of the donor and acceptor atoms
reduces the energy barrier which normally prevents
transfer of the hydrogen atom from the donor to the
acceptor group (Cleland et al. 1998). The locations of
the LBHB’s observed in the endothiapepsin com-
plexes are shown in Fig. 5 where it can be seen that
the inhibitor hydroxyl forms an LBHB with the inner
oxygen of Asp 32 and another with the outer oxygen
of Asp 215. This is consistent with the deuterium
positions found in the neutron analysis (Coates et al.
2001) and other atomic resolution X-ray structures of
aspartic proteinases (e.g. Fujimoto et al. 2004).
Analysis by NMR
A distinguishing property of LBHBs is a large down-
field shift detected by
1
H-NMR typically between 16
and 21 ppm. Thus, to confirm the existence of the short
interactions suggested by the atomic resolution X-ray
data, 1D
1
H-NMR solution spectra of native endo-
thiapepsin and a number of complexes were recorded
(Coates et al. 2002a). Whilst the NMR spectrum of free
endothiapepsin (or the free inhibitor) showed no peaks
outside the normal region for protein signals 0–
11 ppm, the complexes gave several peaks between
15.5 and 18.5 ppm (Fig. 6). In the spectra of endo-
thiapepsin complexed with the phosphinic acid ana-
logue inhibitor, there is a peak with an even larger
Fig. 5 The hydrogen bond
distances for three atomic
resolution structures of
endothiapepsin bound to
hydroxyl-containing
transition state analogues
(Coates et al. 2002a). The
presence of hydrogen bonds
as short as 2.5–2.6 A
˚
suggests
that low barrier hydrogen
bonds are involved
Fig. 6 The proton NMR
spectra of native
endothiapepsin and that of
complexes with four different
transition state analogues
(Coates et al. 2002a). The
peaks around 16 and 18 ppm
observed in the complexes are
indicative of low-barrier and
single-well hydrogen bonds,
respectively. There are no
peaks in this region in the
native enzyme spectrum or in
the spectra of the inhibitors
themselves (data not shown).
The sharp peak and dip at
around 13.6 ppm in some
spectra are an artifact
Eur Biophys J (2006) 35:559–566 565
123
Page 7
downfield shift of 18.5 ppm. Most likely, this corre-
sponds to a very short hydrogen bond (2.41 A
˚
) ob-
served in the X-ray structure between the outer oxygen
of Asp 32 and an oxygen of the phosphinate group.
Such extremely short hydrogen bonds are referred to
as single-well hydrogen bonds.
Concluding remarks
Since LBHB formation facilitates rapid proton trans-
fer, the role of these interactions in enzyme catalysis is
the subject of much discussion in the enzymology field
(Cleland et al. 1998). An early proposal for the
involvement of LBHBs in aspartic proteinase catalysis
based on deuterium isotope effects (Northrop 2001)
invoked a somewhat different pattern of LBHB for-
mation at the catalytic centre involving a proton shared
equally between the two inner aspartate oxygens.
Overall, recent work using a range of techniques (de-
scribed above) has provided the first structural evi-
dence that low-barrier hydrogen bonds may be
significant for aspartic proteinase catalysis and has
helped to pin-point where the LBHB interactions are
at the catalytic centre, thus giving a sounder footing for
evaluating their role in the mechanism.
The neutron and atomic resolution work described
in this review strongly suggest that Asp 32 is negatively
charged in the transition state complex of aspartic
proteinase catalysis. The negatively charged carboxyl
of this aspartate residue forms hydrogen bonds with
both of the hydroxyls of the neutral gem-diol inter-
mediate in a symmetric manner (Fig. 1,intermediate2).
In contrast with serine proteinases where the enzyme
provides an oxyanion binding hole or pocket, the
oxyanion of the aspartic proteinase mechanism appears
to be part of enzyme itself, rather than part of the
transition state intermediate.
References
Bailey D, Cooper JB (1994) A structural comparison of 21
inhibitor complexes of the aspartic proteinase from Endo-
thia parasitica. Protein Sci 3:2129–2143
Cleland WW, Frey PA, Gerlt JA (1998) The low barrier hydrogen
bond in enzymatic catalysis. J Biol Chem 273:25529–25532
Coates L, Erskine PT, Wood SP, Myles DAA, Cooper JB (2001)
A neutron laue diffraction study of endothiapepsin:
implications for the aspartic proteinase. Biochemistry
40:13149–13157
Coates L, Erskine PT, Crump MP, Wood SP, Cooper JB (2002a)
Five atomic resolution structures of endothiapepsin inhibi-
tor complexes: implications for the aspartic proteinase
mechanism. J Molec Biol 318:1405–1415
Coates L, Erskine PT, Mall S, Williams PA, Gill R, Wood SP,
Cooper JB (2002b) The structure of endothiapepsin com-
plexed with a gem-diol inhibitor PD-135,040 at 1.37 A
˚
res-
olution. Acta Crystallogr D59:978–981
Cooper JB (2002) Aspartic proteinases in disease, a structural
perspective. Curr Drug Targets 3:155–174
Cooper JB, Myles DAA (2000) A preliminary neutron Laue
diffraction study of the aspartic proteinase endothiapepsin.
Acta Crystallogr D56:246–248
Davies DA (1990) The structure and function of the aspartic
proteinases. Ann Rev Biophys Biophys Chem 19:189–215
Dunn BM (2002) Structure and mechanism of the pepsin-like
family of aspartic peptidases. Chem Rev 102:4431–4458
Erskine PT, Coates L, Mall S, Gill R S, Wood SP, Myles DAA,
Cooper JB (2003) Atomic resolution analysis of the catalytic
site of an aspartic proteinase and an unexpected mode of
binding by short peptides. Protein Sci 12:1741–1749
Fruton JS (1976) Mechanism of catalytic action of pepsin and
related acid proteinases. Adv Enzymol 44:1–36
Fujimoto Z, Fujii Y, Kaneko S, Kobayashi H, Mizuno H (2004)
Crystal structure of aspartic proteinase from Irpex lacteus in
complex with inhibitor pepstatin. J Mol Biol 341:1227–1235
Hofmann T, Dunn BM, Fink AL (1984) Cryoenzymology of
penicillopepsin. Mechanism of action of aspartyl proteinas-
es. Biochemistry (USA) 23:5247–5256
James MNG, Sielecki AR, Hayakawa K, Gelb MH (1992)
Crystallographic analysis of transition-state mimics bound to
penicillopepsin -difluorostatine-containing and difluorost-
atone-containing peptides. Biochemistry 32:3872–3886
Lindner H, Helliger W (2001) Age-dependent deamidation of
asparagine residues in proteins. Exp Geront 36:1551–1563
Northrop DB (2001) Follow the protons: a low-barrier hydrogen
bond unifies the mechanisms of the aspartic proteases. Acc
Chem Res 34:790–797
Pearl L (1987) The catalytic mechanism of aspartic proteinases.
FEBS Lett 214:8–12
Pearl LH, Blundell TL (1984) The active-site of aspartic pro-
teinases. FEBS Lett 174:96–101
Polgar L (1987) The mechanism of aspartic proteases involved
‘push–pull’ catalysis. FEBS Lett 219:1–4
Rich DH, Bernatowicz MS, Schmidt PG (1982) Direct C-13
NMR evidence for a tetrahedral intermediate in the binding
of a pepstatin analog to porcine pepsin. J Am Chem Soc
104:3535–3536
Silva AM, Cachau RE, Sham HL, Erickson JW (1996) Inhibition
and catalytic mechanism of HIV-1 aspartic protease. J Mol
Biol 255:321–340
Suguna K, Padlan EA, Smith CW, Carlson WD, Davies DR
(1987) Binding of a reduced peptide inhibitor to the aspartic
proteinase from Rhizopus-chinensis—implications for a
mechanism of action. Proc Natl Acad Sci USA 84:7009–7013
Symersky J, Monod H, Foundling SI (1997) High-resolution
structure of the extracellular aspartic proteinase from
Candida tropicalis yeast. Biochemistry 36:12700–12710
Tang J, James MNG, Hsu IN, Jenkins JA, Blundell TL (1978)
Structural evidence for gene duplication in evolution of acid
proteases. Nature 271:618–621
Veerapandian B, Cooper J, S
ˇ
ali A, Blundell TL (1990) 3-
Dimensional structure of endothiapepsin complexed with a
transition-state isostere inhibitor of renin at 1.6 A
˚
resolu-
tion. J Mol Biol 216:1017–1029
Veerapandian B, Cooper JB, S
ˇ
ali A, Blundell TL, Rosati RL,
Dominy BW, Damon DB, Hoover DJ (1992) Direct
observation by X-ray-analysis of the tetrahedral intermedi-
ate of aspartic proteinases. Protein Sci 1:322–328
566 Eur Biophys J (2006) 35:559–566
123
Page 8
  • Source
    • "That is, in the crystal structure, the two Asp dyad protonation states shown schematically in (Fig. 6) are equally populated. Overall, the work described above has provided the first structural evidence that low-barrier hydrogen bonds may be significant in the catalytically driven reaction pathway in aspartic proteases and has helped to locate the LBHB interactions in the catalytic centre , thus giving a sounder footing for evaluating their role in the mechanism [30]. "
    [Show abstract] [Hide abstract] ABSTRACT: Aspartic proteases (AP) are a family of important hydrolytic enzymes in medicinal chemistry, since many of its members have become therapeutical targets for a wide variety of diseases from AIDS to Alzheimer. The enzymatic activity of these proteins is driven by the Asp dyad, a pair of active site residues Asp residues that participate in the catalysis based hydrolysis of peptides. Hence, the protonation state of these and other acidic residues present in these enzymes determines the catalytic rate and the affinity for an inhibitor at a given pH. In the present work we have reviewed the effect of the protonation states of the titratable residues in AP's both on catalysis and inhibition in this family of enzymes. The first section focuses on the details of the catalytic reaction mechanism picture brought about by a large number of kinetic, crystallographic and computational chemistry analysis. The results indicate that although the mechanism is similar in both retroviral and eukaryotic enzymes, there are some clear differences. For instance, while in the former family branch the binding of the substrate induces a mono-ionic charge state for the Asp dyad, this charge state seems to be already present in the unbound state of the eukaryotic enzymes. In this section we have explored as well the possible existence of low barrier hydrogen bonds (LBHB's) in the enzymatic path. Catalytic rate enhancement in APs could be explained in part by the lowering of the barrier for proton transfer in a hydrogen bond from donor to acceptor, which is a typical feature of LBHB's. Review of the published work indicates that the experimental support for this type of bonds is rather scarce and it may be more probable in the first stages of the hydrolytic mechanism in retroviral proteases. The second section deals with the effect of active site protonation state on inhibitor binding. The design of highly potent AP inhibitors, that could be the basis for drug leads require a deep knowledge of the protonation state of the active site residues induced by their presence. This vital issue has been tackled by experimental techniques like NMR, X-ray crystallography, calorimetric and binding kinetic techniques. Recently, we have developed a protocol that combines monitoring the pH effect on binding affinities by SPR methods and rationalization of the results by molecular mechanics based calculations. We have used this combined method on BACE-1 and HIV-1 PR, two important therapeutic targets. Our calculations are able to reproduce the inhibitor binding trends to either enzyme upon a pH increase. The results indicate that inhibitors that differ in the Asp dyad binding fragments will present different binding affinity trends upon a pH increase. Our calculations have enabled us to predict the protonation states at different pH values that underlie the above mentioned trends. We have found out that these results have many implications not only for in silico hit screening campaigns aimed at finding high affinity binders, but also (in the case of BACE-1) for the discovery of cell active compounds.
    Full-text · Article · Nov 2012 · Current pharmaceutical design
  • [Show abstract] [Hide abstract] ABSTRACT: Cells of Candida albicans (C. albicans ) can invade hu- mans and may lead to mucosal and skin infections or to deep-seated mycoses of almost all inner organs, especially in immunocompromised patients. In this context, both the host immune status and the ability of C. albicans to modulate the expression of its virulence factors are relevant aspects that drive the candidal susceptibility or resistance; in this last case, culminat- ing in the establishment of successful infection known as candidiasis. C. albicans possesses a potent arma- mentarium consisting of several virulence molecules that help the fungal cells to escape of the host immune responses. There is no doubt that the secretion of aspartyl-type proteases, designated as Saps, are one of the major virulence attributes produced by C. albicans cells, since these hydrolytic enzymes participate in a wide range of fungal physiological processes as well as in different facets of the fungal-host interactions. For these reasons, Saps clearly hold promise as new potential drug targets. Corroborating this hypothesis, the introduction of new anti-human immunodeficiency virus drugs of the aspartyl protease inhibitor-type (HIV PIs) have emerged as new agents for the inhibition of Saps. The introduction of HIV PIs has revolutionized the treatment of HIV disease, reducing opportunistic infections, especially candidiasis. The attenuation of candidal infections in HIV-infected individuals might not solely have resulted from improved immunologi- cal status, but also as a result of direct inhibition of C. albicans Saps. In this article, we review updates on the beneficial effects of HIV PIs against the human fungal pathogen C. albicans , focusing on the effects of these compounds on Sap activity, growth behavior, morphological architecture, cellular differentiation, fungal adhesion to animal cells and abiotic materials, modulation of virulence factors, experimental candidia- sis infection, and their synergistic actions with classical antifungal agents.
    No preview · Article ·
  • [Show abstract] [Hide abstract] ABSTRACT: Z-Ala-Ala-Phe-glyoxal (where Z is benzyloxycarbonyl) has been shown to be a competitive inhibitor of pepsin with a Ki = 89 +/- 24 nM at pH 2.0 and 25 degrees C. Both the ketone carbon (R13COCHO) and the aldehyde carbon (RCO13CHO) of the glyoxal group of Z-Ala-Ala-Phe-glyoxal have been 13C-enriched. Using 13C NMR, it has been shown that when the inhibitor is bound to pepsin, the glyoxal keto and aldehyde carbons give signals at 98.8 and 90.9 ppm, respectively. This demonstrates that pepsin binds and preferentially stabilizes the fully hydrated form of the glyoxal inhibitor Z-Ala-Ala-Phe-glyoxal. From 13C NMR pH studies with glyoxal inhibitor, we obtain no evidence for its hemiketal or hemiacetal hydroxyl groups ionizing to give oxyanions. We conclude that if an oxyanion is formed its pKa must be >8.0. Using 1H NMR, we observe four hydrogen bonds in free pepsin and in pepsin/Z-Ala-Ala-Phe-glyoxal complexes. In the pepsin/pepstatin complex an additional hydrogen bond is formed. We examine the effect of pH on hydrogen bond formation, but we do not find any evidence for low-barrier hydrogen bond formation in the inhibitor complexes. We conclude that the primary role of hydrogen bonding to catalytic tetrahedral intermediates in the aspartyl proteases is to correctly orientate the tetrahedral intermediate for catalysis.
    No preview · Article · Oct 2007 · Biochemistry
Show more