Download Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites

Current Topics in Medicinal Chemistry 2004, 4, 403-429
403
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites:
Conformational Studies in Solution, Homology Models and Comparison
with Other Zinc Metallopeptidases
Georgios A. Spyroulias*,1, Athanassios S. Galanis1, George Pairas1, Evy Manessi-Zoupa2 and Paul
Cordopatis*,1
Departments of Pharmacy1 and Chemistry2, University of Patras, GR-26504, Patras, GREECE
Abstract: Angiotensin-I Converting Enzyme (ACE) is a Zinc Metallopeptidase of which the three-dimensional stucture
was unknown until recently, when the Xray structure of testis isoform (C-terminal domain of somatic) was determined.
ACE plays an important role in the regulation of blood pressure due to its action in the frame of the Renin-Angiotensin
System. Efforts for the specific inhibition of the catalytic function of this enzyme have been made on the basis of the Xray structures of other enzymes with analogous efficacy in the hydrolytic cleavage of peptide substrate terminal
fragments. Angiotensin-I Converting Enzyme bears the sequence and topology characteristics of the well-known
gluzincins, a sub-family of zincins metallopeptidases and these similarities are exploited in order to reveal common
structural elements among these enzymes. 3D homology models are also built using the X-ray structure of Thermolysin as
template and peptide models that represent the amino acid sequence of the ACE’s two catalytic, zinc-containing sites are
designed and synthesized. Conformational analysis of the zinc-free and zinc-bound peptides through high resolution 1H
NMR Spectroscopy provides new insights into the solution structure of ACE catalytic centers. Structural properties of
these peptides could provide valuable information towards the design and preparation of new potent ACE inhibitors.
1. INTRODUCTION
1.1. Angiotensin-I Converting Enzyme and Hypertension
Angiotensin-I Converting Enzyme (ACE), isolated in the
mid 50’s, is a Zinc Metallopeptidase and one of the major
components of the so-called Renin-Angiotensin System
(RAS) [1-3]. Renin is responsible for the liberation of
Angiotensin I (AI) in blood, after renin’s catalytic action on
the angiotensinogen. ACE possesses a crucial role in the
regulation of blood pressure since it catalyzes the cleavage of
the C-terminal His-Leu dipeptide of the rather inactive
decapeptide Angiotensin I (AI), in the vasopressor
octapeptide Angiotensin II (AII) [4-6] (Fig. (1)). However,
ACE impact is not only focused on the generation of
Angiotensin II but also extended to inactivation of the
vasodilator peptides Bradykinin (BK) and Kallidin [7-9].
ACE is encountered in two distinct forms in humans, the
somatic and the testis form. These differ from the structural
point of view, mainly in size and number of catalytic sites
[10,11]. According to its function, ACE is classified among
the peptidyl dipeptidases of zinc metallopeptidases superfamily due to its ability to remove C-terminal dipeptide [5]
from substrates. ACE can also exhibit activity of an
endopeptidase against substrates such as Substance P,
Cholecystokinin and Luliberin (LHRH), peptides with
amidated C-end [9,12,13]. As far as the ACE role in blood
pressure is concerned, the inhibition of ACE enzymatic
activity against AI was considered as one of the major
challenges against hypertensive disease and congestive heart
failure [14]. Therapy, today and after extensive research for
*Address correspondence to these authors at Department of Pharmacy,
University of Patras, GR-265 04, Greece; Tel: +30 2610 997 721; Fax: +30
2610 997 714; e-mail: [email protected] and [email protected]
1568-0266/04 $45.00+.00
the last 30 years has been achieved through inhibitors based
on the pioneering work of Ferreira S.H. [15] and Ondetti
M.A. [16,17]. These researchers showed that the venom of a
Brazilian pit viper contained a factor that greatly enhanced
the smooth-muscle-relaxing action of the nonapeptide BK
which also inhibiting ACE. All ACE inhibitors were
prepared in the absence of the ACE’s three-dimensional
structure and bear two main characteristics: (i) designed
on the basis of venom peptide extracting structural
information for the enzyme catalytic site from the crystal
structure of Carboxypeptidase A (CPA) [18], and (ii) high
biological activity strongly coupled with enhanced zinc
binding ability [19].
1.2. ACE Inhibitors and their Impact in Medicine and
Pharmacology
ACE inhibitors are considered among the most potent
antihypertensive drugs and apart their major action, exhibit
beneficial lateral effects in the prevention of cardiovascular
disease in various classes of hypertensive patients.
Additionally ACE inhibitors have been proven more
effective than other hypertensive substances in reducing
proteinuria and retarding the progression of renal damage in
patients with various types of nephropathy. These features
are probably among the reasons that two (International
Society of Hypertension-World Health Organization, ISHWHO; Canadian Society of Hypertension) of the three health
organisations (the third is the British Hypertension Society)
affiliated with hypertension, recommend ACE inhibitors in
the first-line of antihypertensive drug treatment, after the
results of the first trials of this type of inhibitors became
available at 1996.
An alternative treatment of hypertension is focused on
the blockade of AII receptors, AT1R and AT2R (94% amino
© 2004 Bentham Science Publishers Ltd.
404
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
Spyroulias et al.
Fig. (1). Schematic representation of the Renin-Angiotensin system and the role of the two enzymes, Renin and Angiotensin Converting
Enzyme (ACE) on regulation of blood pressure through the generation and release of Angiotensin II vassopressor peptide.
acid sequence identity), with appropriate antagonists. The
class of AT1R which are currently under continuous
development and trials, exhibit at the moment only one main
the advantage over ACE inhibitors, which is the absence of
cough as a side effect. Interestingly, only ISH-WHO has
recommended AT1R antagonists as first line antihypertensive
drugs, and this is probably due to the absence of any
published long-term trial results. For all these reasons, the
design and preparation of new potent ACE inhibitors still
remains one of the main challenges in the intersection of the
fields of chemistry, pharmacology and medicine.
1.3. Zinc Catalytic Sites and their Characteristics
Zinc sites in metalloenzymes and related biomolecules
are classified according to their ligands and coordination
geometry into three types of zinc binding sites [20,21]: (i)
the catalytic, (ii) the cocatalytic and (iii) the structural.
Hydrolases like ACE possess a catalytic zinc site which
usually coordinates with nitrogen, oxygen and sulphur
donors of His, Glu, Asp and Cys residues while His is most
frequently encountered in the coordination sphere of zinc
metal ion. Water is also a zinc ligand in catalytic sites and is
activated for ionisation, polarisation, or displacement by the
identity and arrangement of ligands coordinated with zinc
[22]. The zinc coordination number for this kind of sites
has been found to be four or five and the donor atoms of
residues define a distorted-tetrahedral or trigonal-bipyramidal
coordination geometry. Ionisation and/or polarisation of the
activated H 2O is assisted by the base form of an active siteresidue or in some cases by a “second-shell” residue
that yields hydroxide ions at neutral pH while water
displacement results in Lewis acid catalysis on the part of the
catalytic zinc metal.
The structure of the zinc catalytic sites comprises: (i) the
zinc-bound residues, (ii) the characteristic sequence of the
amino acids around those ligated to the metal, (iii) the short
or long amino acid “spacers” among the three (in some cases
four) protein ligands, and (iv) the conformational features of
the “spacers”. Zinc sites are also characterised by secondary
interactions with neighbouring amino acids that position in
space and conformational features strongly depend on the
overall protein folding and three-dimensional structure.
These features are critical for the structure-function
relationship of this class of metalloenzymes and dictate their
classification into various families.
The characteristic amino acid sequences, which contain
the potential three zinc ligands in the zinc metallopeptidase
family, comprise the binding motif sequences that are a
diagnostic tool in enzyme classification. The first two of the
protein ligands are found in the first three-, four- or fiveresidue binding motif, while the third is found in a second
characteristic motif. These residue-ligands are generally
separated either by short amino acid “spacers” among the
first, second and third ligands, or by a short spacer between
the two first ligands and by a large spacer between the
second and third, or fourth ligand, should one exist. The
magnitude of a short spacer could vary from one to three
amino acids between the first two ligands. On the other hand,
the long spacers usually found in various metallopeptidase
subfamilies could vary from 5 to over 100 amino acids. The
length of the spacer between the two first ligands belonging
to the same binding motif often characterizes the secondary
structure of this protein fragment. For example, a threeresidue spacer is characteristic of a α-helix conformation
while a one-residue spacer indicates a β sheet conformation.
1.4. Focus of this Article
This article aims to provide new structural insights into
ACE, an enzyme whose role in hypertension has stimulated
over the last 35 years extensive and continuous effort
towards designing its potential inhibitors, even without the
most important tool in the hands of biochemists,
enzymologists and drug designers: the ACE threedimensional structure. What follows is an attempt to review
the latest progress in structural biology of zinc
metallopeptidases and extract structural information through:
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
(i) an investigation of ACE sequence and structure
similarities/differences among representative members of
metallopeptidase family, (ii) homologous modeling based on
known enzyme structure, and (iii) a solution conformational
analysis through 1H NMR spectroscopy on 36-residue
synthetic peptides whose sequences represent the two ACE
catalytic sites and their zinc-binding properties.
2. ZINC METALLOPEPTIDASES
ISTICS AND CLASSIFICATION
CHARACTER-
The super-family of zinc-containing bio-macromolecules
has been enormously extended over the last years, with the
same occurring in the frame of the zinc metalloproteases/
peptidases family [23,24] (Fig. (2)). An attempt to classify
previously unknown biopolymers with distinct motifs and
zinc-binding affinity at the beginning of the 90’s led
researchers to identify diagnostic motifs within the
polypeptide amino acid sequences through sequence and
topology comparison [45]. Using this methodology, not only
sequence but also structural similarities have been identified
among various zinc metalloproteases and thus enzyme
groups into certain categories [46-52]. In 1992 Jiang and
Bond [48] compare the sequences around the diagnostic,
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
405
zinc-binding motif, HEXXH (X = any amino acid residue) in
order to classify the known zinc metalloproteases into five
families, where each one has its prototype: (i) Thermolysin
(TLN) [27,53-55] (ii) Astacin (AST) [34], (iii) Serratia [5658], (iv) Matrixin [59,60] and (v) Reprolysin/Adamalysin
[24,49,52,61]. Determination of the zinc metalloproteases
3D crystal structure sheds new light on the structural features
of these enzymes and new terms, such as zincins, metzincins,
aspzincins, gluzincins and inverzincins, have been used in
order to discriminate between them [24,51,52] (see Fig. (2)).
In addition to ACE conformational features extracted through
theoretical and experimental studies, an update of the sequence
and topology characteristics of zinc metalloproteases/
peptidases through the structural analysis of known 3D
structures is presented.
2.1. Zincins, Metzincins and Gluzincins: Where ACE
belongs?
Overall, the zinc metallopeptidases can be divided into
two categories according to the sequence of the first zinc
binding motif: (i) the zinc enzymes with the characteristic
HEXXH motif, where the two histidines are the potential
protein ligands, and (ii) the zinc enzymes without HEXXH
Fig. (2). Classification of Zinc Metalloproteases according to MEROPS Protease Database (July 2002; http://www.merops.co.uk/merops/
index.htm). Data concerning protein source and zinc ligands, amino acid spacer between the binding motifs, accession number and codes for
sequence and coordinates are also given. The X-ray structures of the active site for some representative metalloenzymes are presented. The
zinc protein ligands are labelled. The number of amino acids (aa) refers to the spacer length between the protein zinc ligands. X stands for
any amino acid while, in bold letters, His, Glu and Met residues are noted.
406
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
motif. The presence of HEXXH defines the so-called zincins
family [49]. The amino acid sequence and topology close to
the second binding motif or the motif sequence itself divides
the zincins family into metzincins [49,52] and gluzincins [24]
subfamilies (Fig. (2)). According to Bode et al. [49] the
crystal structure determination of AST (pdb: 1ast [34]) and
Adamalysin II (pdb: 1iag [61]; see also Fig. (3)) revealed
that these two proteinases exhibit low sequence similarity
but significant topological equivalence. Their main feature is
the almost identical zinc environment with a methioninecontaining turn and, according to these findings, Bode has
suggested that these enzymes could be termed as metzincins.
The third zinc ligand in the vast majority of metzincins is a
histidine residue, while in some members of this family a
tyrosyl residue has been identified as the fourth coordinated
protein ligand [49,52]. However, the X-ray crystal structures
of Snapalysin (M7 family, pdb : 1kuh [31], Fig. (3)) and
Peptidyl-Lys Metalloendopeptidase (M35 family, pdb: 1g12
[62], Fig. (3)) revealed that an aspartate residue is the third
protein ligand, although these enzymes belong to the MA(M)
clan and have been termed aspzincins [62].
The term gluzincins was introduced in the mid 90’s by
Hooper N. M. for the zincins whose third zinc ligand is a
glutamic acid [24] found in the consensus binding motif
Spyroulias et al.
sequence EXXXX. TLN is considered the prototype member
of this category of zinc metallopeptidases, whose first X-ray
crystal structure was solved almost 30 years ago [27,53-55]
(see also Fig. (2), (3)). The last residue of the EXXXX
sequence for TLN (pdb: 1lnd) [27]) and Neprilysin (or
Neutral Endopeptidase; NEP) (pdb: 1dmt [28]) crystal
structures has been found to be an aspartatic acid which is
also found in the same position of the proposed second
binding motif of ACE (EAIGD). According to the sequence
identity of the key-residues in the two potential binding
motif sequences, ACE is classified among the gluzincins
superfamily and specifically in the M2 family of MA(E) clan
of zinc metalloproteases (Table (1)).
In 1994, the gluzincin term, according to Hooper,
spanned six families of the metalloprotease MA clan, M1,
M2, M3, M4, M13 and M27 with representative members
Aminopeptidase A (EC 3.4.11.7), ACE (EC 3.4.15.1),
Thimet Oligopeptidase (EC 3.4.24.15), TLN (EC 3.4.24.27),
NEP (EC 3.4.24.11), and Botulinum neurotoxin A (EC 3.4.
24.69) respectively. Until mid 2002, the gluzincin term
encompasses eighteen metallopeptidase families, according
to the Protease Database (http: // www . merops . co.uk/merops/
merops.htm), while there are 9 3D X-ray structures out of 99
assigned peptidase sequences classified in gluzincins,
Fig. (3). 3D X-ray structures of the zinc-containing active sites of the two superfamilies, metzincins MA(M) (left) and gluzincins MA(E)
(right), of zinc metalloproteases, available until July 2002. Representative models, from each clan, which possesses at least one characteristic
enzyme with available 3D structure, are presented and pdb codes are given. Prototype clan members with no 3D structure determined so far
are also listed. Active site helices are shown in blue/gold ribbon representation while non-active helices are depicted in grey/light grey.
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
Table 1.
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
407
Amino Acid Composition of Zinc Binding Motif Sequences for Representative Members of All Known Families of MA(E)
Clan, Gluzincins, Zinc Metallopeptidases. Horizontal Black Shaded Boxes Indicate the Family of the MA(E) Clan. Darkgray Columns Indicate the Metal Binding Ligand which have Either been Identified Through the Available X-ray
Structure or Proposed in Analogy to other Enzymes. PDB Code, when Available, is Given (Last Column).
Sequence1
MA(E) - M2
ACEN
H390
ACEC
390
H
EMG
H394
EMG
394
H
PDB
REF
---23---
E418
AIGD
---23---
418
AIGD
1O8A
[148-9]
E318
GHTV
1hs6
[25]
E503
VPSQ
1i1i
[26]
E415
IAST
E166
AISD
1lnd
[27]
E239
GFAD
E429
GIAE
E446
GGAE
E646
NIAD
1dmt
[28]
E 1518
FFAK2
E261
LRTF
3bta
[29]
E269
AFAM
E 305
SQSR3
1k9x
[150]
E768
FFAE
1j7n
[30]
E 458
GWSD4
E 476
EIIY5
E
MA(E) – M1
LTA4 Hydrolase
295
H
EIS
299
H
---18--M3A
Neurolysin
H474
EFG
H478
ETG
476
---24--M3B
Oligopeptidase F
387
H
H
---23--M4
Thermolysin
142
H
ELT
146
H
---19--M5
Mycolysin
201
H
EAG
205
H
---33--M9A
Microbial collagenase
H401
Collagenase colA (H)
414
EYV
H405
EYT
418
---23--M9B
H
H
---27--M13
Neprilysin
583
H
EIT
587
H
---58--M26
IgA1 protease
H1494
EMT
H1498
---19--M27
Botulinum neurotoxin
222
H
ELI
226
H
---34--M30
Hyicolysin
246
H
EYQ
250
H
---44--M32
Carboxypeptidase Pfu
276
H
EMG
280
H
---24--M34
Anthrax Lethal factor
H719
EFG
H723
EYT
432
---44--M36
Fungalysin
429
H
H
---25--M41
FtsH endopeptidase
414
H
E_G
418
H
---57--M47
408
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
Spyroulias et al.
Table 1. (Contd….)
Sequence1
PRSM1 metallopeptidase
MA(E)
H204
ELG
H208
ELG
339
--------
PDB
----
----6
E415
FQAD
E 222
FHAD4
E 267
VWNN7
E 282
GQTQ8
E
XXXX
REF
M48_
CAAX prenyl protease
335
H
H
---75--M48_
HtpX endopeptidase
H139
E_S
H143
---78--M60
Enhacin
248
H
ELG
252
H
---14--M61
248
252
Glycyl aminopeptidase
H
ELG
H
Consensus Sequence
H
EXX
H
---29---
1
Residue numbering is according to the sequence record deposited at Swiss-Prot (http://www.expasy.ch/) when no 3D structure is available. With peptidases for which
coordinates have been deposited at Protein Data Bank (http://www.rcsb.org/pdb/) the residue numbering follows the PDB record. Numbers between the 2nd and 3rd protein ligand
indicate the magnitude of amino acid spacer.
2
Tentative assignment – E 1518 is the 1 st conserved glutamate in all known sequences of M26 family. The 2nd conserved E is found after 90 intervening residues in a pentapeptide
that possesses the sequence EGNSI.
3
Tentative assignment – E305 is found in the consensus sequence of HESQSX (where X = R or L) of M32 family. This E has been assigned as 3rd zinc ligand according to the
conserved residues of the motif and to the preceding histidyl residue, which is also found in the same position in ACE sequences.
4
Tentative assignment – E 458 has been assigned as 3rd zinc ligand since the 4th residue of the consensus sequence EXXXD is an Aspartic acid as has also been found in the same
position in M4, M5, M13, M48A and ACE sequences.
5
Tentative assignment – E476 is the only conserved glutamate in all known sequences of M41 family.
6
No assignment is attempted since only one sequence for M47 family is available. First E residue is found after 19 intervening residues in the sequence EWPGG.
7
Tentative assignment – E 267 is the 1 st conserved glutamate in all known sequences of M60 family in the consensus sequence EXWZN (where X = I or V and Z= N or T). The 2nd
conserved E is found after 50 intervening residues in the non-consensus motif sequence ERNIA.
8
Tentative assignment – E282 is the only conserved glutamate in all known sequences of M61 family in the consensus sequence EGXTZ (where X = T, F or Q and Z= S or Q).
yielding representative 3D models for six of the families
(M1, M3, M4, M13, M27 and M34; see Fig. (3)).
The catalytic, zinc-containing, sites extracted from the
available X-ray structures of representative members of zinc
metalloproteases metzincins, MA(M), and gluzincins, MA(E)
clans are illustrated in Fig. (3).
2.2. Sequence and Structure Characteristics: Binding
Motifs and Active Sites of Zincins, Metzincins and
Gluzincins
Enzymes from each of the above mentioned
metallopeptidase clans and families are grouped below
according to three main characteristics concerning the
characteristic sequence of: (i) the first zinc binding motif, (ii)
the second zinc binding motif and (iii) the amino acid spacer
between the two binding motifs and its magnitude.
The amino acid composition of their binding motif
sequences and the number of residues which constitute the
spacer for gluzincin sequences for which both binding motifs
have been distinguished and characterized is illustrated at
Table (1) while analogous data for metzincins are presented
at Table (2).
According to Hooper N. M. [24] the consensus sequence
of gluzincins is HEXXH-spacer-EXXXX and the magnitude
of the spacer varies between 18 and >80 amino acids. Except
for the three zinc ligands and the glutamic acid that is present
in all zincins’ first binding motif there is no other conserved
amino acid in the two binding motifs. However, among the
18 gluzincins families (plus three subfamilies), in 7 families
the fourth residue of the second, the “glutamate” binding
motif, is an aspartic acid, as also happens in the ACE
sequence. Of these families, only two, MA(E)-M4 and -M13,
possess sequences for which X-ray crystal structures are
available (Fig. (3)). Representative members of these two
families are TLN and NEP respectively, exhibiting the
HEXTH and EXIXD consensus zinc binding motif sequences,
which exhibit significant similarities with the HEMGH and
EAIGD sequences of ACE. Additionally, TLN’s active site
possesses a 19 amino acid spacer that is comparable to that
of the 23 amino acid spacer attributed to ACE’s active site
polypeptide, while NEP possesses a spacer consisting of 58
amino acids.
Despite the low overall sequence identity between
metzincins and gluzincins peptidases, such as for AST and
TLN respectively, significant topological similarities in their
active site have been implied by sequence comparison
[24,29]. Crystal structure determination of metzincins
representatives in the early 90’s, such as Adamalysin [pdb:
1iag, 61] and AST [pdb: 1ast, 34], revealed that the first
binding consensus zincins sequence HEXXH constitutes a
helix, which has also been found in the X-ray structure of
TLN [49,27,53-55]. Crystal structures of metzincins and
gluzincins peptidases solved over the last decade verify the
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
Table 2.
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
409
Amino Acid Composition of Zinc Binding Motif Sequences for Representative Members of MA(M) Clan Families,
Metzincins, Zinc Metallopeptidases which Posses at Least One Member with known Xray Crystal Structure. Horizontal
Black Shaded Boxes Indicate the Family of the MA(M) Clan. Dark-gray Columns Indicate the Metal Binding Ligand
According to Xray Structure. The Light-gray Column Indicates the Three-residue Sequence of the Characteristic ‘Metturn’ while Tyrosyl Residue in bold Characters Indicates the Assigned Fourth Protein Zinc Ligand. PDB Code, when
Available, is Given (Last Column).
Sequence1
MA(M) – M7
H83
Snapalysin
ETG
H87
----5---
PDB
REF
D93
HYSG
--5-
M103SG
1kuh
[31]
H334
IKMR
--6-
M345AP
1lml
[32]
H228
STDI
--3-
M236YP
1fbl
[33]
H186
PGDY
-23-
M214SY
1srp
[58]
H102
EHTR
-40-
M147HY
1ast
[34]
H152
DGKD
--9-
M166RP
1iag
[61]
D130
YAYG
----
----3
1g12
[62]
H/D
XXXX
M8
Leishmanolysin
264
H
EMA
268
H
---67--M10A
2
Collagenase 1
218
H
ELG
222
H
----5--M10B
Serralysin2
H176
EIG
H180
----5--M12A
Astacin
2
92
H
EAM
96
H
----5--M12B
Adamalysin
142
H
ELG
146
H
----5--M35
1
2
3
Peptidyl-Lys
endopeptidase
H117
ESS
H121
Consensus Seq.
H
EXX
H
----9---
MXX
Residue numbering is according to the record with the structure coordinates (PDB record) which have been deposited at Protein Data Bank (http://www.rcsb.org/pdb/). Numbers
between the 2nd and 3rd protein ligand, as well as before the “Met-turn” Methionine, indicate the magnitude of amino acid spacer.
Tyrosine in the Met-turn sequence is noted with bold characters and indicates the 4th zinc ligand.
No Met-turn motif has been identified for this peptidase either in identified sequence or in X-ray structure. Last residue in sequence (P81054; Swiss-Prot :
http://www.expasy.ch/) and in X-ray structure is Ser167.
assumption that the HEXXH first binding sequence for all
zincins is part of a helix, the so-called ‘active site helix’. The
second binding motif of metzincins and gluzincins varies
structurally as implied by sequence comparison between
various peptidases of the two MA(M) and MA(E) clans. The
third protein zinc ligand is found in a helical fragment for all
gluzincins structures solved, but this is not the case with
metzincins peptidases. The latter possess a histidine or
aspartate residue as a zinc third ligand, which has not arisen
from a second helix but is located in an open coil segment.
No matter what the amino acid composition, the length or the
conformation of the spacer between the two binding motifs
in gluzincins (Fig. (3)), the EXXXX sequence always
comprises a second ‘active site helix’. Consequently, the
active site environment in gluzincins is characterised by two
‘active site helices’, which constitute the main component of
the zinc catalytic pocket. Additionally, the major structural
feature that characterises the conformation of the 18-58
amino acids spacer is the α-helix, especially when the
magnitude of this spacer is longer than 20 residues.
Summarising the information drawn from the analysis of
zincins structure available so far are one should note that:
(i) Topological equivalence may occur even without
sequence equivalence, (ii) conformation of first and second
binding zinc motif sequence for both metzincins and
gluzincins is remarkably similar among members of each
family even when sequence identity of active site is low (1525%) and intervening residues differ in nature and length.
2.3. Functional and Structural Relationship of ACE with
other Zinc Metallopeptidases
Structural and catalytic properties of ACE have been
discussed in the literature in comparison either with a zincin,
the gluzincin metallopeptidase TLN, or with a non-zincin
metallopeptidase the CPA [63-69]. ACE is a dipeptidyl
carboxypeptidase while CPA is an exopeptidase [63,64] and
TLN is an endopeptidase [66,69].
Carboxypeptidase A (see Fig. (1)) has a HXXE as a first
binding motif sequence and HXX as a second. CPA first
binding motif is a four-residue motif in contrast to the fiveresidue zincins motif and possesses the glutamate and the
histidine as second and third protein ligand. In contrast, the
first zincins motif sequence bears the two histidines and the
glutamate follows as the third ligand sited at the second
binding motif. None of the CPA binding motifs comprises a
helical fragment while the amino acid spacer between them
410
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
varies from 108 to 135 amino acids among the various
carboxypeptidases sequences.
2.3.1. Carboxypeptidase A.
Research on the inhibition of ACE’s catalytic action
started in the early 1970s [15-17], and has been based on
function similarities of ACE with other enzymes with known
3D structures [70-72]. One of these enzymes was CPA (M14
family of MC clan), the structure of which was determined
by X-ray crystallography (pdb: 1yme, Fig. (4)) [18,36] and
its active site and catalytic mechanism has been extensively
investigated. [63,73]. CPA possesses a zinc-binding motif
with the characteristic HXXE sequence, and is able to cleave
a single amino acid from the carboxy-terminal end of a
peptide substrate, in contrast with ACE, which hydrolyses
the COOH-terminal dipeptides. Ondetti and Cushman, who
were recently honored for their pioneering work and
continuous effort in the design and study of potent and
specific ACE inhibitors [19], proposed a late 1970’s
Spyroulias et al.
hypothetical model of ACE based on the active center of
CPA. (see Fig. (4C)) [72]. They also suggested that the
catalytic mechanism of ACE is similar to that of CPA. Based
on these studies several research groups designed and
synthesized numerous potential ACE inhibitors, arriving at
the first pharmacologically promising antihypertensive
compounds that bind at ACE’s active sites [19].
Analysis of 3D structure of CPA catalytic cavity [73]
revealed that a positively charged residue (noted by “+” in
Fig (4C)), Arg145, extends its side-chain NH3 groups
towards the metal site and has the role of forming ionic
bonds with the negatively charged carboxyl group of the
substrate’s C-terminal (Scheme ( 1A)). As an analogy to that,
a positively charged residue is believed to exist at ACE
active site and participate in the catalytic mechanism of the
hydrolytic dipeptide cleavage [72]. Also, in CPA, the zinc
ion, which polarizes the carbonyl group of the scissile
peptide bond, has been found in a substrate’s single-residue
far from Arg145, which interact with the substrate’s COO-
Fig. (4). (A ) X-ray crystal structure of CPA (pdb: 1yme; top) and TLN (pdb: 1lnd; bottom). (B) Conformation of structural and catalytical
important residues of the zinc active site of CPA (B, top) and TLN (B, bottom). Side-chains of residues are presented in ball and stick. (C)
Cartoon representation of active site of CPA based on its X-ray structure and a CPA-based hypothetical model of ACE active site where
various known inhibitors are accommodated. Sub-sites noted as S1, S2, S1’ and S2’ are cavities or areas in an enzyme’s active site where
amino acid groups interact with adjacent side-chains of the substrate’s (peptide or inhibitors) groups in a molecular recognition and complexformation procedure.
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
411
Scheme 1. Proposed catalytic mechanisms for the (A) Carboxypeptidase A and (B) Thermolysin catalyzed cleavage of peptides.
terminal group. On the other hand, at the ACE catalytic site,
zinc metal ion has to be found in a position at a distance of a
dipeptide-residue, which itself must be far from the
enzyme’s positively charged residue. Furthermore, Arg127,
Glu270 orient their side-chains towards a zinc-catalytic site
(Fig. (4B) & Scheme (1A)) and are implicated in CPA
catalytic mechanism. They have the task of polarizing the
scissile carbonyl bond, while the Tyr248 OH group is
hydrogen bonded to the terminal carboxylate of the substrate
[63] (Scheme (1A)).
2.3.2. Thermolysin
Thermolysin is a thermostable extracellular gluzincin
endopeptidase which belongs to the M4 family of MA(E)
zinc metallopeptidase clan and has been isolated from
Bacillus thermoproteolyticus. TLN catalyses the hydrolytic
cleavage of the peptide bond specifically on the imino side
of large hydrophobic residues, and especially leucine,
isoleucine and phenylalanine [74] His142, His146 and
Glu166 together with a solvent water are found coordinated
with the zinc ion in a distorted tetrahedral coordination [27].
Glu143 and Asp170 are positioned in the so-called “second
coordination shell” of the zinc cation and both belong to the
consensus HEXXH and EXIXD sequences of the first and
second binding zinc motifs, respectively (Fig (4B)). Glu143
is highly possibly hydrogen-bonded with its side-chain
carboxylate to the coordinated H2O, while Asp170, which is
conserved in various gluzincins including ACE, is considered
a structurally and/or functionally important residue (Scheme
(1B)). The Asp170 structural role arises from the fact that its
charged side chain forms a salt link with the imidazole ring
of His142, the first protein zinc-ligand [11,27,53]. According
to extensive structural studies of enzyme-inhibitor
complexes, the catalytic role of the outer shell residues
His231, Tyr157 and Asp226 in the enzyme’s function has
been also suggested, Fig. (4B) [75-80].
The peptide substrate forms a Michaelis complex
(Scheme (1B); 2 nd step) with an enzyme’s active site and its
carbonyl oxygen is accommodated among His231, Tyr157
and the coordinated H2O, which however becomes slightly
displaced from its original position after peptide binding.
Coordinated H 2O is then sited closer to Glu143, in a position
favorable for its polarization, which is also assisted by zinc
cation. Glu143 and metal ion therefore enhance the
nucleophilicity of the H2O and promote its attack on the
carbonyl carbon. Glu143 accept a proton which in turn is
donated to the substrate’s peptide bond nitrogen, forming a
gem-diolate intermediate (Scheme (1B); 3rd step). At this
stage the zinc metal is in a five-coordination state and the
peptide carbon possesses a tetrahedral geometry. The
geometry of the metal site and the intermediate has been
identified through transition-state analogue inhibitors [7580]. The formation of the intermediate product is believed to
be assisted by a hydrogen bond network where the sidechains of His231, Tyr157 and the carbonyl oxygen of
substrate are involved. The role of Asp266 is focused on the
stabilization of the positive charge required for catalysis
through a salt-bridge between its negative carboxylate sidechain and the protonated imidazole of His231 [74,80].
Finally, the peptide C-N bond is cleaved and the protonated
product is released, while Glu143 is proposed to abstract the
second water proton, shuttling it to the amine nitrogen
(Scheme (1B); 4th step).
2.3.3. Angiotensin-I Converting Enzyme and Inhibitors
The discovery of two active sites in somatic ACE has
provoked many assays to establish functional or structural
412
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
difference in these two highly homologous enzyme regions.
Studies of an ACE somatic isoform fragment that contains
only the N-terminal active site has been performed with the
aim of identifying the functional differences or the selective
inhibition between the two catalytic centers of ACE. For
example, whereas the most important peptide substrates such
as BK and AI [4,9, 81,82] are hydrolyzed at both sites, while
some others, like LHRH or Angiotensin1-7, are cleaved
preferentially only by N-domain active site [83-86].
Additionally, it has been reported that the phosphinic
compound RXP407 is an N-selective inhibitor of ACE, able
to differentiate between the two active sites [87,88], in
similarity with some other peptide inhibitors (BBPs) which
have been reported recently to selectively inhibit C-domain
active sites. Furthermore, it has been reported that ACE
catalytic activity requires monovalent anions such as Cl- for
maximal activity and that the extent of activation is substrate
and chloride dependent [89-91]. Data suggest that the Cdomain active site is more sensitive to chloride concentration
for the hydrolysis of some substrates [81,92].
The catalytic action of ACE has long been discussed in
concert with that of CPA and/or TLN [74,80], whose 3D
crystal structure is available. In the absence of an ACE
structure, speculations are made on the basis of sequence
homology, or topology similarities with the above mentioned
enzymes and structure-function correlation studies of
mutated enzymes. Between ACE and CPA/TLN, there is
undoubtedly low sequence identity while the known
structures of CPA and TLN differ considerably. Moreover,
ACE binding motif and spacer resemble those of TLN while
both differ remarkably from those of CPA. ACE binding
motif sequences consist of the residues HEMGH.….EAIGD
and between them 23 intervening residues have been
identified in analogy to the sequences of TLN,
HELTH…..EAISD separated by 19 residues. However, a
profound analysis of the TLN and CPA zinc environment in
their 3D structures suggest that there are some common
elements. For example CPA Glu270 is relative to TLN
Glu143 (see Fig. (4A) and (4B)), and in analogy to
Glu362/960 of the two ACE active sites considered vital for
enzyme activity [93]. On the other hand the two residues
suggested as the proton donors in the catalytic mechanisms
of CPA and TLN, Tyr248 and His231 respectively, are not
found in a comparable position in their structures. This
structural variation has promoted the aspect that there was no
absolute requirement for the involvement of a histidine or a
tyrosine in zinc neutral proteases hydrolytic action.
Nevertheless, recently a histidyl residue, His1089 has been
suggested to possess a similar role to that of TLN His231.
Additionally, sequence comparison of the amino acid spacer
that separates the second and the third zinc ligand in TLN
and ACE reveals three and two tyrosines (368/965, 369 and
372/970; somatic form numbering) for the two ACE catalytic
sites analogous to Tyr157 in TLN.
Since until recently ACE structure remained obscure, the
binding affinity of potential inhibitors to ACE and structure
activity relationships have not been totally illuminated.
Extensive studies have been performed towards a detailed
understanding its active centers specificity. Data indicate that
the majority of ACE inhibitors exhibit specificity on
Spyroulias et al.
interaction with mainly three subsites or pockets at the ACE
active sites, named S1, S1' and S2 (Fig. (4C)). Substrates
containing phenyl group (R3) that interacts with the ACE S1
subsite have proved to be significant for the binding properties
of inhibitors [94-97] like antihypertensive compounds
enalapril, lisinopril, trandolapril, fosinopril etc. (Fig. (4C)).
Furthermore, S1' subsite has been found to have a weaker
binding affinity toward the peptide substrate, while it may
not be large enough to accommodate peptide substrates with
extended, long, side-chains [98,99]. Additionally, most
substrates that specifically inhibit ACE active sites bear
proline or proline-like analogues at their carboxy terminus,
which interact with the S2' binding pocket [72,100]
(captopril, enalapril, lisinopril etc.; see Fig. (4C)). Another
feature that might be essential in a potential enzyme inhibitor
is its free C-terminal carboxylate group [101]. According to
the zinc-binding group, inhibitors or potential inhibitors can
be classified in three categories: (i) those with a sulfydryl
group, such as captopril [72,100,102], (ii) those with a
carboxyl group, bound to zinc ion as enalapril [102],
lisinopril [97, 103] etc and (iii) inhibitors with a phosphonate
group, like fosinopril [104]. The structures of the above
mentioned compounds accommodated at the proposed
hypothetical ACE catalytic cavity are presented in Fig. (4).
2.4. Structural Insights to ACE Active Sites Through
Site-directed Mutagenesis
Attempts to extract conformational characteristics essential
for an enzyme’s function have been also focused on the
substitution of residues, which in analogy to other peptidases
are believed to be actively involved in ACE’s catalytic
mechanism. Such residues should be sited close to the zinc
site (Table (1)) and their role elucidated through site directed
mutagenesis and structure-function relationship studies.
When the two histidines of the first potential ACE zincbinding motif have been mutually substituted by other amino
acids, it is revealed that enzyme [11,92,103,105,106]
completely abolishes its activity. These data indicate that
these two histidyl residues are essential for ACE catalytic
activity. Moreover, these amino acids are believed to be the
two first protein zinc-ligands. Further site-directed
mutagenesis experiments yield a GluàAsp substitution,
where the replaced glutamate residue is that found in the
possible two-histidyl binding motif with the sequence
HEMGH, which has resulted in suppression of enzyme
catalytic activity [92]. Aspartate residue retains the negative
charge in the sequence, which, however, has been displaced
of approximately 1.4 Å. This glutamate is involved in a basic
attack of the substrate peptide bond and its role is of
crucial importance to the catalytic efficacy of ACE. This
strongly supports the role postulated for Glu143 of the
HELTH in TLN [86].
As far as the nature and properties of the second ACE
possible binding motif are concerned, the glutamate residue
in EAIGD sequence is conserved in many other zinc
metalloproteases which highly possibly plays the role of the
third protein zinc ligand. This glutamate characterizes the
gluzincins and is, indeed, coordinated with the zinc metal ion
according to the X-ray structures solved for other members
of this family. When an aspartic residue or a valine replaces
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
this glutamic acid, ACE catalytic activity has been decreased
by more than two orders of magnitude, or has been
completely extinguished, respectively [11]. Finally, mutation
studies performed by the same research group with aspartic
acid in EAIGD motif as target residue, which is replaced by
a glutamic, or an alanine residue, indicate a specific
functional role of its carboxylate side-chain. Asp170 role has
been proposed to as the salt link formation with the
imidazole ring of the first potential zinc coordinated histidine
[11], which is observed in TLN X-ray structure [27,53]. A
similar role in ACE function is expected for Asp393/991
(somatic isoform numbering).
Site directed mutagenesis studies have been also
performed in order to probe the nature of a basic amino acid,
adjacent to an enzyme’s active site, which is believed to bind
a chloride ion. Concentration of chloride ions has been
reported among the crucial parameters that adjust ACE
catalytic activity [90,107] and Arg1098 [108] has been
reported critical for the chloride dependence of ACE
catalytic activity and binds the anion. However, this is in
contrast with the results obtained through earlier chemical
modification studies on lysyl residues [109], which had
proposed active involvement in chloride and other
monovalent anion binding [109,110].
Furthermore, substitution of His1089 of human somatic
form, by an Ala or Leu, has prompted the researcher to
propose that this histidyl residue stabilizes the transition state
complex through hydrogen bonding with the tetrahedral
intermediate product [111]. A similar role has been also
attributed to His231 in the catalytic mechanism of TLN
[74,80]. Finally, mutation studies of human and rabbit testis
ACE isoform have demonstrated a functional role of Tyr200
and Tyr236, respectively [112-113] in analogy to the Tyr198
in CPA.
3. 3D HOMOLOGY MODELS OF ACE ACTIVE SITES
Sequence alignment [114] of ACE and other gluzincin
active site fragments for which an X-ray crystal structure has
been determined is performed and presented in Fig. (5A).
The structure of TLN active site (pdb: 1lnd [27]) has been
chosen as a template in order to generate 3D homology
models of the 36-residue peptide that represent the
ACEN[His361-Ala396] and ACEC[His959-Ala994] zinc
active sites, as shown in Fig. (5B) [115]. TLN zinc site
fulfils the following criteria: (i) high sequence identity with
ACE binding motifs, (ii) similar amino acid spacer
magnitude, (iii) topological and conformational similarities
resulting from theoretical prediction of ACE secondary
structure at NPSA server (Fig. (5C)). Despite the low
primary structure similarity (<25%), the above mentioned
features indicate remarkable topological analogy.
3D homology ACE peptide models indicate that the
backbone folds in two helical fragments, one at each
terminus, where the two binding motifs are sited. Therefore,
the two histidyl side-chains of the HEMGH motif are
separated by a α-helix turn and their side-chains are parallel.
Therefore, they adopt the desired geometry in order to
donate the imidazole nitrogens to the zinc coordination
sphere. The oxygen atom of the third zinc protein ligand that
of glutamate is also found in favourable geometry for zinc
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
413
coordination. The aspartate residue of EAISD motif orients
its side-chain parallel to that of glutamate and towards the
first histidine imidazole ring.
The two antiparallel terminal helices in ACE homology
models are in great agreement with the “two active helices
site” model adopted according to the two helices observed in
the catalytic cavities of all the 9 X-ray gluzincin crystal
structures available hitherto. These features appear
independently of the active site amino acid composition and
magnitude. Additionally, theoretical prediction of secondary
structure of ACE structure in Network Protein Sequence
Analysis [116] through GOR protocols [117-119] reveals
two major helices, one towards the N-terminus and the other
towards the C-terminus, both of which include the HEMGH
and EIAGD bonding motifs.
Having analysed all the above data we wish to make an
effort to elucidate experimentally the structural features of
ACE active sites. This attempt is performed through conformational analysis in solution, using NMR spectroscopy, of the
synthetic 36-residue ACE model peptides. The results of
NMR data analysis for apo- and zinc-bound forms of
ACEN[His361-Ala396] peptide and NMR solution structure
of the free peptide are presented below.
4. SYNTHESIS OF THE TWO ACE Zn-SITES – WHAT
COULD BE THE BENEFIT?
X-ray crystallography and NMR spectroscopy are the
two experimental methods, which could yield a highresolution three-dimensional structure of biopolymers.
However, both of these techniques have to overcome some
limitations. X-ray crystallography has the drawback of
single-crystal and heavy-atom derivative preparation. On the
other hand, NMR bypasses the time-consuming stage of
crystallization but is restricted by the size of the molecule.
Larger NMR structures of biomolecules or biomolecular
complexes reaches the limit of 50-60 kDa while sporadic are
the examples that complete sequence specific resonance
assignment has been performed for proteins with more than
400 amino acid. But even in that case, NMR studies become
feasible only after the development and the successful
applications of molecular biology and biotechnology
techniques which permits the selective labeling and
deuteration in concert with 15N and 13C labeling. Until
recently, ACE structure had not been elucidated, neither in
solid-state, probably due to unsuccessful attempts to obtain
high-quality crystals, nor in solution since its molecular
weight exceeds 80 and 140 kDa, for testis and somatic
isoform respectively. On the other hand, solid-phase
synthesis of peptides and polypeptides is able to produce
peptides or polypeptides with any given sequence-bearing
protein or even non-protein amino acids in satisfactorily high
yields and purity [120]. Additionally, synthetic peptides of
metal-binding sequences are widely used in order to probe
the structure, the metal-binding affinity and transport
pathways of heavy metals to target proteins, since their
structure closely resembles the native active site of a
metallobiomolecule [121-123].
Zinc metal is essential for ACE function and substrate
binding and for this reason our approach to investigate the
conformational features of ACE catalytic cavity begins with
414
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
Spyroulias et al.
Fig. (5). (A) Sequence alignment of the polypeptide fragments comprising the zinc active sites of ACE and representative gluzinicins of
which 3D X-ray structures are available. (B) TLN zinc-containing active site polypeptide (32 residue X-ray, pdb: 1lnd) and 3D homology
models of ACEC/ACEN 36-residue peptides. Side chains of zinc ligands and F/Y, M/L, A/S and E/R variable ACEC/ACEN residues are
shown in ball and stick representation for all models while the four different residues in ACEC and ACE N sequences are also illustrated. (C)
Prediction of secondary structure at NPSA server through GOR protocols.
the solid-phase synthesis of 36-residue peptide whose amino
acid composition and sequence represent the ACE active site
fragment. ACE zinc-binding sequence comprises a 29residue fragment which contains the three proposed protein
ligands; His361/959, His365/963 and Glu389/987 for the
two zinc-sites, found at HEMGH and EAIGD (somatic
isoform numbering). 36-residue synthetic peptides represent
the ACE N[His361-Ala396] and ACEC[His959-Ala994] zinc-
binding sequences (ACEN and ACEC refer to zinc sites
towards N- and C- terminus respectively; Fig. (5C)).
The amino acid sequence of the above ACEN and ACEC
peptides was built “step by step” on the acid-sensitive 2chlorotrityl chloride resin (substitution 0.6 mmol/g) applying
the Fmoc strategy [124-126]. Final purification was achieved
by semipreparative HPLC on a RP C-18 support (Phase Sep
C-18 S10 ODS2) eluted with a linear gradient 20% to 60%
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
acetonitrile (0.1% TFA) over 30 min at a 2 ml/min flow rate
[127]. The final products were determined to be at least 96%
pure by analytical HPLC. The overall yield was 48% for
both 36-residue constructs (Scheme 2).
These peptides share 89% sequence identity and are the
main fragment of the ACE catalytic sites region (see Table
(3)). Active site reconstitution has been performed through
addition of ZnCl2 in a peptide solution and metal binding
properties are monitored in solution.
5.
CONFORMATIONAL
STUDY
CATALYTIC SITE IN SOLUTION
OF
ACEN
5.1. 1H NMR Spectroscopy of ACEN[His361-Ala396] 36residue Zinc-Free Peptide
5.1.1. NOE and Secondary Structure
The complete spin-system of 35 out of 36 residues and
sequential assignment has been accomplished through the
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
415
combined analysis of the TOSCY [128,129] (Fig. (6A-B))
and NOESY [130,131] spectra. His1 is the only amino acid
where backbone and aliphatic chain protons have not been
identified. However, the characteristic cross peak between
the non-exchangeable protons of His1 imidazole ring has
been assigned. No HN proton resonance was identified for
Glu2. Assignment and chemical shifts are given in Table (4).
Fig. (6C) shows the short and medium range NOE
observed for the backbone and CβH protons in the ACEN
NOESY maps. There are three regions with diagnostic
connectivities for helix conformation, such as HN-HN(i,i +
2), Hα-HN(i,i + 2), Hα-HN(i,i + 3) and Hα-Hβ(i,i + 3), that
were observed. Met3-Tyr12, Val17-Gly22 and Gly26-Ala36
define these regions in both peptides. On the other hand HαHN(i,i + 4) NOE are observed only in the first of the abovementioned regions. This is due to signal overlapping which
does not allow unambiguous assignment of such type of
NOE for the Val17-Ala36 region.
Scheme 2. Flow chart for solid phase, step-by-step, synthesis and purification procedure for the ACE 36-residue constructs (HOBt, 1hydroxybenzotriazole; DIC, N,N’-diisopropylcarbodiimide; Pip, pideridine; EDT, 1,2-ethanedithiol, TFA, trifluoroacetic acid).
416
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
Table 3.
Spyroulias et al.
Amino Acid Composition of the Active Sites’ 44 Residue Polypeptide Containing One or Two Zinc Binding Motifs of
ANGIOTENSIN-I Converting Enzyme Known Sequences. Dark-gray Columns Indicate the Metal Binding Ligand
According to the SWISS-PROT Record.
Sequence
1st B. M.1
ACE_HUMAN3
VAH
H390
EMG
“
---
H988
ACE_MOUSE3
TVH
H395
“
993
IA3
ACE_CHICK
H
288
TVH
“
H
886
---
ACE_RABIT
3,4
H
395
TVH
H
992
Spacer
2nd B. M.1
H394
IQYYLQYKDLPVSLRRGANPGFH
E418
AIGD
VLALS
---
H992
---FM-------A--E-------
E1016
----
-----
EMG
H399
VQYYMQYKDLHVSLRRGANPGFH
E423
--EMG
--EMG
997
H
292
H
890
H
399
H
996
I--F------P-TF-E------VQYYLQYKDQPVSFRGGANPGFH
---F---M-Q-I---D------VQYYLQYKDQPVSLRR - ANPGFH
AA2
AIGD
VLALS
E
1021
----
IM---
E
316
AIGD
VLSLS
E
913
----
-MA--
E
422
AIGD
VLALS
1020
----
-----
“
V--
H
---
H
I--FM----L--A--EG------
E
ACE_RAT3
TVH
H396
EMG
H400
VQYYLQYKDLHVSLRRGANPGFH
E424
AIGD
VLALS
“
IA-
H994
---
H998
I--FM-----P-TF---------
E1022
5
ACE_DROME
ACE_HAEIE
6
ACET_HUMAN
ACET_MOUSE
367
TVH
H
367
---
H
414
VAH
H
413
I--
ACET_RABIT
VV-
Cons. Seq.
XXH
419
-MEMG
---
H
371
H
418
H
417
H
423
IQYFLQYQHQPFVYRTGANPGFH
----------------------IQYFMQYKDLPVALREGANPGFH
------------TF---------
1312
1193
1310
1310
----
-----
E
395
AVGD
VLSLS
615
E
395
----
-----
615
E
442
AIGD
VLALS
732
E
441
----
IM---
732
447
----
-----
737
AXGD
XXXLS
H
---
H
-----------------------
E
H
EXG
H
XQYXXQYXXXXXXXRXGANPGFH
E
1
Amino acids belong to 1 and 2 binding motif. Numbering is according to the sequence record deposited at Swiss-Prot (http://www.expasy.ch/).
2
Number of amino acids found in the sequence.
3
Somatic ACE isoforms possess two zinc-containing active sites, while testis isoform contains only one.
4
A residue is missing at the position of conserved Gly residue in all the other ACE sequences and noted by -
5
st
H
ELG
371
1306
nd
Drosophila melanogaster (Fruit fly).
5.1.2. 3JHNHα Coupling Constants
Spin-spin coupling constant values could also be of great
importance for the diagnosis of the secondary polypeptide
structure. The observed J depends on the average of the Js of
each available conformation multiplied by its statistical
weight, and α- and 310- helical structures exhibit experimental
J values in the range of 4.8 to 5.6 [132], somewhat larger
than the ideal values of 3.9 and 4.2 Hz respectively. On the
other hand, extended structures such as the parallel or antiparallel β sheet give rise to J values larger than 8.0 Hz. Most
(14 out of 18 3JHNHα values) of the J coupling constants
measured in ACE peptides exhibits values below 6.2 Hz (see
Fig. (6C)) [133,134].
5.1.3. Chemical Shift Index
The strong relationship between the backbone
conformation and chemical shift Hα values serves as a strong
indicator for the assignment of secondary structure in any
polypeptide sequence [135,136]. Chemical shift difference
analysis between the observed Hα shift values and the
corresponding random coil values [137] is presented in Fig.
(6D) and provides strong evidence for the conformational
preference of the majority of amino acids towards the helical
configuration. There are three distinct regions where the
∆δHα values are negative suggesting helical character (Ile6Lys13, Val17-Gly22 and Gly26-Val34) and three shorter
regions where the ∆δHα values are positive suggesting
extended conformation (Glu2-His5, Asp14-Pro16 and
Ala23-Pro25). The most negative ∆δHα value is measured in
both peptides for Val17 (~ 0.4 ppm) while Asn24 exhibits
the larger positive ∆δHα value (~ 0.3 ppm) (Fig. (6D)).
An overall evaluation of the observed NOE, 3JHNHα and
∆δHα values, implies that in three fragments the skeleton of
the 36-residue ACE N active site model peptide adopts a welldefined α-helical structure even in the absence of zinc metal
[138]. The longer helices are anticipated for their C- and Ntermini, and a smaller one for the intermediate region. The
two prolines, at positions 17 and 25, act as helix-stop
residues between the three helices. The second binding motif
EAIGD seems to be part of the C-terminal helix while no
safe conclusion could be extracted for the first binding motif,
HEMGH, which comprise the N-terminal pentapeptide.
Taking into account the flexibility of any terminal peptide
fragment and the fact that resonances for all protons for His1
and HN proton of Glu2 have not been observed the
conformation of this pentapeptide should be the average over
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
417
Fig. (6). 1H-1H 2D TOCSY 600-MHz NMR of ACEN (A & B) and ACEC (F & G) fingerprint region of H_-HN protons (H2O/TFE-d2
34%/66% v/v, at pH=5.0, T= 298 K). The number of the amino acid in the ACE sequences, to which the H_-HN and side-chain proton
connectivities belong, is noted. (C) Sequential connectivities for ACEN are presented together with the predicted secondary structure
elements. 3JHNH_ coupling constants are illustrated by arrows (↓ for values in the range 4.2 –5.6, ↑ for values equal or above 8.0 Hz) or by
filled circles (values in the range of 6.0-7.0 Hz). (D) ∆δH_ chemical shift difference (in ppm) from random coil values according to CSI. The
NMR data were acquired on a Bruker AVANCE 600 spectrometer. (E) Number of meaningful NOE constraints per residue for ACEN.
White, grey, and dark grey bars respectively represent intraresidue, sequential, and medium-range connectivities. Long-range connectivities
have not been observed. Schematic presentation of the predicted secondary structures according to the sequential connectivities is shown at
the bottom of panel (C) and according to CSI at the top of panel (D).
various conformers. However, the positive ∆δHα values are
relatively large for Glu2 and His5 (>0.10 ppm) and close to
the random coil values for Met3 and Gly4. The J values,
feasibly measured for Met3 and His5, found ≤ 6.2 Hz for
ACEC, the same holding for Met3. These data are
controversial with ∆δHα values, but are in great agreement
with the NOE sequential connectivities illustrated at Fig.
(6C). According to the observed NOE, almost all the helixdiagnostic connectivities of Hα-HN(i,i + 2), Hα-HN(i,i + 3),
Hα-Hβ(i,i + 3), and Hα-HN(i,i + 4) type, involving the
Met3-Gly4-His5 amino acids, have been observed in
NOESY maps. These data suggest that these three residues
of the N-terminal binding motif are in α-helical structure and
possibly comprise the initial turn of the first 10-12-residue
helix of ACEN/C.
5.2. High Resolution NMR Solution Structures of
ACEN[His361-Ala396] 36-residue Zinc-Free Peptide
DYANA [139,140] structure calculations have been
performed using 22.0 NOE-derived distance constraints per
residue (>15 meaningful) together with 20 H-bond distance
constraints (two distance limits for each H-bond) and 18 ϕ
constraints. The resulting DYANA family of 30 structures
has rmsd values (calculated for residues 3-33) of 0.57 ± 0.24
Å and 1.25 ± 0.21 Å respectively, for backbone and heavy
atoms. The target function lies in the range 0.39-0.45 Å2
(0.43 ± 0.018 Å2). The final REM [141,142] family exhibits
pairwise rmsd values for the 30 structures 0.57 ± 0.24 Å,
1.27 ± 0.21 Å and to the mean structure 0.40 ± 0.18 Å, 0.89
± 0.11 Å for backbone and heavy atoms respectively (Fig.
(6E)). Restraint violations and structural and energetic
statistics for the ACEN[His361-Ala396] 36-residue Zinc-Free
peptide are reported at Table (5).
The ACEN polypeptide chain (Fig. (7)) is characterized
by the high content of helical structure, which is distributed
in three fragments; the α-helical N- and C- terminal together
with the 310-helix observed in the center of the intermediate
fragment between the two other helices. Two turns of the
peptide skeleton, the first after the end of the N-terminal
helix in the region of Asp14-Pro16 and the second after the
intermediate helix in the region of Asn24-Pro25 creates a Ushaped cavity in the middle of the peptide sequence. This Uturn structure is comprised of the residues Lys13-Pro25 and
brings the two terminal helices to a distance, which varies
418
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
Table 4.
Spyroulias et al.
Chemical Shifts (ppm) of the Protons of the Residues in the [1-36]ACEN at 298K (H2O/TFE-d2 34%/66% v/v, pH=4.9).
Residue
HN
1
His
2
Glu
3
Met
4
Hα
Hβ
other
Ηδ2 8.263; Ηε1 7.298
4.457
2.184, 2.040
Ηγ 2.389
8.692
4.539
2.150, 2.094
Ηγ 2.618
Gly
8.385
3.994
5
His
8.214
4.743
3.370, 3.323
Ηδ2 8.457; Ηε1 7.246
6
Ile
8.032
4.071
2.039
Ηγ 1.628, 1.313; γCΗ3 1.010; δCΗ3 0.963
7
Gln
8.511
4.029
2.115, 2.081
Ηγ 2.392 ; δΝΗ2 6.554, 7.141
8
Tyr
7.785
4.268
3.088
Ηδ 6.948; Ηε 6.754
9
Tyr
7.823
4.311
3.253, 3.158
Ηδ 7.177; Ηε 6.881
10
Leu
8.202
4.052
1.945, 1.874
Ηγ 1.526; δCΗ3 0.930
11
Gln
7.780
4.123
1.985, 1.890
Ηγ 2.191, 2.138 ; δΝΗ2 6.389, 6.975
12
Tyr
7.850
4.426
3.135, 2.819
Ηδ 7.015; Ηε 6.775
13
Lys
7.830
4.082
1.795, 1.751
Ηγ 1.398; Ηδ 1.624; εCΗ3 3.032
14
Asp
7.915
4.809
2.875, 2.657
15
Leu
7.778
4.532
1.779
Ηγ 1.597; δCΗ3 0.970, 0.943
16
Pro
4.450
2.316, 2.105
Ηγ 2.041, 1.967; Ηδ 3.862, 3.653
17
Val
7.688
3.815
2.179
γCΗ3 1.053, 0.999
18
Ser
7.999
4.240
3.980, 3.924
19
Leu
7.796
4.345
1.733
Ηγ 1.650; δCΗ3 0.936, 0.891
20
Arg
7.754
4.207
1.929
Ηγ 1.748, 1.628 ; Ηδ 3.169; Ηε 7.063
21
Arg
8.016
4.278
1.895, 1.835
Ηγ 1.744, 1.647 ; Ηδ 3.174; Ηε 7.153
22
Gly
7.953
3.959
23
Ala
7.780
4.380
1.416
24
Asn
7.956
5.020
2.931, 2.762
δΝΗ2 6.613, 7.443
25
Pro
4.444
2.387, 2.319
Ηγ 2.054, 1.990; Ηδ 3.900, 3.804
26
Gly
8.331
3.931
27
Phe
7.816
4.512
3.153, 3.117
Ηδ 7.180; Ηε 7.299; Ηζ 7.249
28
His
8.053
4.458
3.306, 3.223
Ηδ2 8.493; Ηε1 7.303
29
Glu
8.244
4.228
2.123, 2.063
Ηγ 2.394
30
Ala
8.097
4.323
1.457
31
Ile
7.722
4.091
1.849
32
Gly
7.952
3.954, 3.863
33
Asp
7.960
4.728
2.810
34
Val
7.759
4.099
2.226
γCΗ3 0.990
35
Leu
7.795
4.400
1.700
Ηγ 1.629; δCΗ3 0.918, 0.880
36
Ala
7.476
4.179
1.402
Ηγ 1.453, 1.196; γCΗ3 0.883; δCΗ3 0.830
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
Table 5.
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
Statistical Analysis for the REM a and <REM> a Structures of [1-36]ACEN.
REM
RMS Violations per Experimental Distance Constraints (Å)
<REM>
b
intraresidue
(128)
0.0213 ± 0.0030
0.0215
sequential
(253)
0.0258 ± 0.0018
0.0311
medium-range
(173)
0.0265 ± 0.0015
0.0237
total
(544)
0.0251 ± 0.0013
0.0269
Average Number of Violations per Structure
intraresidue
7.00 ± 1.84
10.0
sequential
17.67 ± 2.05
17.0
medium-range
14.07 ± 1.39
14.0
total
38.73 ± 2.94
41.0
average no. of NOE violations > than 0.3 Å
0.000 ± 0.00
00.0
largest residual NOE distance violation (Å)
0.203
0.259
0.415 ± 0.03
0.466
2
average distance penalty function (Å )
Statistics of Other Structural Constraints
ϕ constraints from JHNHα
3
(18)
RMS violations per ϕ constraint
0.0627 ± 0.24
average no. of ϕ violations per structure
0.0667 ± 0.25
largest residual ϕ violations
0.0062
-1
average torsion penalty function (kJ mol )
-1
AMBER energy (kJ mol )
0.0008 ± 0.003
0.0004
-1078.44 ± 59.8
-1107.12
a
REM indicates the energy-minimized family of 40 structures and <REM> the mean energy-minimized structure.
b
Numbers in parenthesis indicate the number of meaningful upper distance limits per class.
Fig. (7). Backbone representation of: (A) Family of 30 ACEN REM models and (B) mean REM structure.
419
420
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
from 13.5 to 24.5 Å (between the C_ atoms of Gln7 and
His28, the carbon atom of -COO- and the nitrogen atom of NH3, respectively), one above the other, while their dipoles
form an angle of almost 90°. The helical fragments
comprised by 12 (Glu2-Lys13) and 8 (Phe27-Val34) residues
for the N- and C- termini, respectively, and by 3 residues
(Arg20-Gly22) for the intermediate [138].
The structure of the C-terminal fragment has been
determined with higher resolution than that of the Nterminal, due to the higher number of NOE and angle
constraints. No proton resonances have been identified for
the backbone and side-chain aliphatic protons of His1 and
amide proton of Glu2, probably due to the conformational
averaging and/or exchange with solvents of labile NH
protons. The two prolines, in position 16 and 25 of the
sequence, found in the intervening residues between the
helices probably interrupt the helicity of the backbone as
suggested also by sequential NOE and ∆δHα values. Additionally, next to Pro25 is found Gly26 and these two residues
comprise a notorious residue-pair with helix-destabilizing
ability, since both one characterized by the lowest helix
propensity among the 20 natural amino acids [143,144].
Spyroulias et al.
As far as the structure of the two potential binding motifs
sited at the N- and C- terminus of the ACE model peptide is
concerned, they are found in helix-like or helical fragments
(Fig. (8)). The helical conformation of the C- terminus has
been unambiguously determined and supported by all NMR
experimental data, such as helix-diagnostic sequential
NOE-connectivities (dNN(i,i+2), daN(i,i+2), daN(i,i+3) and
da(i,i+3), see Fig. (6C)), J-coupling constants and chemical
shift analysis (Fig. (6D)). On the other hand helix-like
conformation for the N- terminus resulted for the mean,
energy-minimized, NMR structure supported by some
sequential NOE typical for helix conformation, observed for
the N-terminus starting from Met3 (see Fig. (6C)). Apart
from loss of NOE information, the only contradiction for a
helix structure in this region arises with CSI, and the positive
∆δHα values in the Glu2-His5 (Fig. (6D)). Helix structure has
been calculated for the Ile6-Lys13 residues and all NMR
data is fully consistent with this conformation.
At this point we should note that this N-terminal motif
HEMGH sequence comprise the well-known zincins’ first
zinc-binding motif which is always found in all 3D crystal
structures to be a helix, the so-called “active-site helix”. This
ambiguity in determination of this binding motif conformation
Fig. (8). Colour-coded chemical shift perturbation mapping in identification of conformational changes in free ACEN mean NMR structure.
Shift differentiation is illustrated: (A) between the ACEN and ACEC sequences which differs in 4 residues: Tyr9Phe, Leu10Met, Ser18Ala
and Arg21Glu, (B) when zinc metal is added to the solution of free peptide and (C) the ∆δH_ (CSI) chemical shift differences are indicative of
the zinc-bound peptide’s secondary structure (grey for ∆δH_ > 0; dark grey for ∆δH_ < 0 suggesting _-helix conformation). Chemical shift
differences in bar diagrams are also presented at the bottom. The NMR data were acquired on a Bruker AVANCE 600 spectrometer.
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
strongly depends on the fact that the first histidine of the
motif is the first peptide residue. However, even in this case
the mean NMR solution structure of the free peptides fits
well with the “two active-site helices” model suggested for
gluzincins. Both binding motifs are in helix or helix-like
conformation. The side-chains of the two histidines are
parallel, the same occurring with side-chains of the third
protein zinc-ligand, the glutamate, and the aspartate four
residues after glutamate towards the C-terminus, whose
conformation plays a crucial role in the catalytic activity of
some gluzincins.
5.3. Analysis of the Tyr9Phe, Leu10Met, Ser18Ala and
Arg21Glu Differentiation Between ACEN and ACEC
Through Chemical Shift Perturbation Mapping
The two ACE 36-residue peptides designed and
synthesized differ, as actually happens in native sequence, in
four residues (89% sequence identity), found in positions 9,
10, 18 and 21. The aromatic and aliphatic character of amino
acids in positions 9 and 10 are in general maintained
between the two sequences, due to Tyr/Phe and Leu/Met
substitution. On the other hand, the polar Ser18 which is
considered a “surface” residue is replaced by the non-polar,
“internal”, Ala. Additionally, the basic, positively charged
Arg21 has been replaced by the acidic, negatively charged,
glutamic acid without perturbing the hydrophilic properties
at this point of the sequence. These residue variations
between the two ACE peptides do not seem to provoke any
noticeable conformational change in the secondary structure
of the studied biopolymers, since according to NMR data
(sequential connectivities and ∆δHα values) the helical
character of the fragments where the four residues reside is
not disturbed.
However, comparative analysis of proton resonances
through Chemical Shift Perturbation Mapping [145], provides
∆δHα -HN values between 0.15 and 0.20 ppm. These differences
are mapped in the backbone of the 3D average NMR
structure of ACEN peptide and presented in different colors
at the top and center of Fig. (8A). Chemical shift changes
around Tyr/Phe9 and Leu/Met10 are moderate and slightly
exceed 0.05 ppm, while the relative changes for the peptide
fragment Leu15-Arg/Glu21 fluctuate between 0.07 and 0.17
ppm. Additionally, Asn24 present a chemical shift difference
value of around 0.075 ppm. Differences around 0.05 ppm
have also been estimated for the His28-Ala30 tripeptide.
Plots of the Hα and HN shift differences are also presented
at the bottom of Fig. (8A). Assignment and chemical shifts
for the 36-residue peptide ACEC are given in Table (6).
5.4. 1H NMR Spectroscopy, Zn(II) Binding Properties
and Conformational Features of ACEN[His361-Ala396]
36-residue Peptide
5.4.1. NOE and Secondary Structure – Differences with
Free Peptides
Color-coded chemical shift changes, upon Zn(II) addition,
are presented in Fig. (8B) while chemical shift difference
analysis between the observed Hα shift values and their
corresponding random coil values are presented in Fig. (8C).
Zinc binding properties of ACE peptides are monitored
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
421
through 1H 1D and 2D NMR Spectroscopy and many proton
resonances in the NH region shift considerably and their
half-width is considerably increased when Zn(II) is added
(Fig. (9A) & (9D)). Additionally, the previously degenerated
resonance of the two geminal Gly Hα protons has been split
into two well-defined resonances (Fig. (9B) & (9C). This
resonance different-iation for the previously degenerated Hα
glycine protons suggests a metal-coupled structural change
of the N-terminal motif and indicates a different magnetic
environment for those nuclei in metal-peptides from those in
their free state. NMR data suggest similar elements of
secondary structure between free and zinc-bound peptides
[146].
The Hα values of residue fragment with negative or
positive ∆δHα values in free peptides exhibit the same nature
and magnitude of deviation from their corresponding random
coil values. In the presence of Zn(II) the majority of residues
(23 residues) still exhibit negative deviation (larger than 0.05
ppm) from random coil values (Fig. (8C)). ∆δHα values in
ACEN-Zn peptide define three possibly helical regions : (i) a
7-residue fragment close to the peptide N-terminus (Ile6Lys13), (ii) another 6-residue fragment in the middle of the
sequence (Val17-Arg21), and (iii) a third comprising of 10
residues close to peptide C-terminus (Gly26-Val34). These
data strongly suggest that in the zinc-bound peptide, the
second proposed binding motif (EAIGD) retains its helical
conformation, also observed for the free peptides. As far as
the two histidyl motif is concerned, no definite conclusion
could be reached [146]. In the TLN [27] crystal structure,
that binding motif, which possesses similar sequence to that
of ACE, has been found in helical conformation.
However, helix-diagnostic dαN(i + 3) type connectivity
between Gly4 and Gln7, present in the NOESY spectrum of
free peptides disappears when Zn(II) is added (Fig. (9B)).
Nevertheless, typical helix dαN(i + 4) and dαN(i + 3) connectivities between Ile6-Met10 and Gln7-Met10 respectively, in
free peptides are still present after zinc addition [146]. These
data strongly suggest that the HEMGH motif conformation
might undergo a conformational transition from helix-like or
partial helix to non-helix structure when zinc coordinates
with the donor atoms of the peptide ligands. This could
probably be due to the fact that the first peptide zinc ligand is
also the first residue in the peptide sequence. Upon the lack
of other residue(s), the His1 backbone and side-chain possess
a remarkably high degree of conformational freedom. Thus,
when zinc binds the peptide bonds of the pentapeptide
binding sequence HEMGH, it could be accommodated in a
random, open coil segment. Such conformation justifies the
absence of the typical sequential NOE for a helix and has
resulted after preliminary DYANA structure calculation for
ACEN 36-residue peptide. On the other hand, no changes for
the fragment Ile6-Lys13 are implied by NMR data.
Moreover, the same type as for free peptides sequential NOE
has been detected and structure calculation indicates that the
first HEMGH open coil structure in ACEN follows a helical
fragment which is extended from Ile6 to Tyr12/Lys13.
Another difference between free and zinc-bound peptides,
which has also been identified through sequential connectivities diagram and preliminary structure calculations, is the
absence of the helical fragment in the area Ser18-Arg21 in
422
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
Table 6.
Spyroulias et al.
Chemical Shifts (ppm) of the Protons of the Residues in the [1-36]ACEC at 298K (H2O/TFE-d2 34%/66% v/v, pH=4.9).
Residue
HN
1
His
2
Glu
3
Met
4
Hα
Hβ
other
Ηδ2 8.349 ; Ηε1 7.342
4.467
2.188, 2.050
Ηγ 2.407
8.668
4.557
2.154, 2.095
Ηγ 2.628
Gly
8.362
3.994
5
His
8.252
4.719
3.355, 3.315
Ηδ2 8.485; Ηε1 7.232
6
Ile
8.043
4.063
2.026
Ηγ 1.628, 1.301; γCΗ3 1.006; δCΗ3 0.954
7
Gln
8.443
4.099
2.098, 2.079
Ηγ 2.397 ; δΝΗ2 6.564, 7.160
8
Tyr
7.841
4.292
3.077
Ηδ 6.936; Ηε 6.748
9
Phe
7.950
4.377
3.312, 3.229
Ηδ 7.322; Ηε 7.360; Ηζ 7.344
10
Met
8.283
4.225
2.200, 2.104
Ηγ 2.743, 2.692
11
Gln
7.804
4.149
1.977, 1.903
Ηγ 2.170 ; δΝΗ2 6.384, 6.987
12
Tyr
7.869
4.413
3.129, 2.819
Ηδ 7.010; Ηε 6.751
13
Lys
7.796
4.069
1.790, 1.745
Ηγ 1.398; Ηδ 1.639; εCΗ3 3.040
14
Asp
7.881
4.822
2.906, 2.690
15
Leu
7.772
4.451
1.773
Ηγ 1.646; δCΗ3 0.971, 0.938
16
Pro
4.384
2.394, 2.130
Ηγ 2.048, 1.890; Ηδ 3.874, 3.650
17
Val
7.579
3.690
2.169
γ CΗ3 1.062, 0.981
18
Ala
7.920
4.083
1.452
19
Leu
7.900
4.240
1.766
Ηγ 1.595; δ CΗ3 0.917, 0.884
20
Arg
7.686
4.150
1.950
Ηγ 1.754, 1.633; Ηδ 3.156; Ηε 7.090
21
Glu
8.224
4.294
2.096, 1.985
Ηγ 2.418, 2.387
22
Gly
7.936
3.938
23
Ala
7.766
4.390
1.434
24
Asn
7.888
5.028
3.002, 2.798
δΝΗ2 6.598, 7.498
25
Pro
4.450
2.343
Ηγ 2.048, 1.999; Ηδ 3.908, 3.835
26
Gly
8.341
3.930
27
Phe
7.818
4.488
3.156
Ηδ 7.151; Ηε 7.276; Ηζ 7.234
28
His
8.040
4.415
3.314, 3.255
Ηδ2 8.534; Ηε1 7.343
29
Glu
8.198
4.212
2.125, 2.082
Ηγ 2.418
30
Ala
8.059
4.292
1.455
31
Ile
7.718
4.069
1.826
32
Gly
7.925
3.939, 3.867
33
Asp
7.961
4.728
2.827
34
Val
7.752
4.090
2.225
γ CΗ3 0.999
35
Leu
7.758
4.397
1.707
Ηγ 1.638; δ CΗ3 0.917, 0.878
36
Ala
7.491
4.194
1.409
Ηγ 1.417, 1.172; CΗ3 0.856; δ CΗ3 0.803
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
423
Fig. (9). Downfield regions of 600 MHz 1H 2D NMR NOESY (upper panel) and 1H 1D NMR spectra with characteristic His imidazole ring
and Hα-HN cross-peaks of ACE peptide before (A & B) and after Zn(II) addition (C & D) in solution. Intraresidue Hα-HN cross-peaks of
selected N-terminal residues are illustrated in dark grey and black, while interesidue d_N(i + 1,3,4) cross-peaks are illustrated in grey.
ACEN zinc peptides. The majority of HN-HN and Hα -HN
sequential connectivities identified in free ACEN have not
been detected after zinc addition. However, Hα-Hβ(i + 3)
connectivity, present in free peptide is still detectable
between Asp14 and Val17. Other helix-indicative sequential
connectivities such as Hα-HN(i + 2) between Val17-Leu19,
Arg21-Ala23 and Gly22-Asn24 together with Hα-HN(i + 3)
type NOE between Arg20-Ala23 have also been detected for
the intervening fragment between the two terminus helices
[146]. The limited number of sequential, helix-type
connectivities, could be related to the fact that many of these
protons have been found to resonate in similar field values
possessing almost identical chemical shifts.
5.4.2. Metal-induced Chemical Shift Perturbation & Zincbinding Implication
Chemical shift perturbation mapping [145] provides
valuable insight for Zn(II) induced structural changes when
the metal is added to a peptide solution. The largest changes
were identified for the N- and C- terminal decapeptide
containing the HEMGH and EAIGD sequences where
potential zinc ligands are located. The proton chemical shift
differences for Glu29 are the largest throughout the peptide
Gln7-Ala36 region, suggesting its coordination to zinc. The
His5 protons resonance exhibits a remarkably large
difference while the largest one is identified for Met3.
Unfortunately, no proton resonance was identified for His1
and for the amide proton of Glu2.
According to the data presented above, it is highly
possible that zinc coordination has a dual effect on the
peptides’ N-terminal free-to-bound structure differentiation:
(i) it provokes a structural transition from free peptide
conformation, helix or helix-like, to a well-defined but nonhelical structure of zinc-bound peptide, and (ii) diminishes
the conformational flexibility of this fragment leading to an
ordered structure where residues adopt the desired geometry
prior zinc coordination. Other significant chemical shift
differences are identified in the region covered by Leu15Gly22, which constitute the intermediate fragment of 23residue spacer between the two proposed binding motifs.
This HN and Hα proton shift value differentiation is
probably due to the fact that zinc coordination forces the two
peptide termini to approach one another in order the donor
atoms of the potential peptide ligands sited at N- and C- ends
to achieve a four-ligand distorted-tetrahedral coordination
geometry. This kind of geometry is that most frequently
encountered in other members of the zinc metallopeptidase
family and bend polypeptide fragments have been identified
424
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
Spyroulias et al.
in other gluzincs, such as TLN. Such a model is consistent
with a bend peptide skeleton of a U-like shape, as has been
observed in the case of TLN X-ray crystal structure [146].
TLN and the two helices approach each other. The
NMR structure of the ACEN free peptide reveals that
the potential zinc ligands could bind the metal without
severely disturbing the elements of its tertiary and
secondary structure. The conformation of the amino
acid spacer (310 helix fragment) and the relative
orientation of the two helices could be substantially
differentiated upon zinc binding to the polypeptide,
and the peptide structure could further resemble that
of the models and TLN.
6. CONCLUSIONS
6.1. Common and Variable Structural Features Between
Free ACE Peptides, 3D Homology ACE Models and
Catalytic Sites of Known Gluzincin Structures
Having gone through theoretical and experimental studies
in order to elucidate the solution structure of the ACE zinccontaining active sites various valuable structural
information for these biologically interesting catalytic
centres has been extracted:
i.
Despite the low sequence similarities among known
sequences of gluzincins, all the available X-ray crystal
structures possess a characteristic “two helix activesite”, no matter what the length of the inter-helical
amino acid spacer (Fig. (3)).
ii.
Theoretical study of the secondary structure of the 36residue active site model ACE peptides predicts
helical conformation for the N- and C- terminus,
where the two zinc-binding motif sequences are sited
(HEMGH and EAIGD, respectively, Fig. (5C)).
iii.
3D homology models of these 36-residue peptides
possess two helices, one at each peptide terminus,
suggesting that the peptide bonds could be
accommodated in conformation with high helical
degree (Fig. (5B)). These helices are found to form an
angle similar to that of TLN helices (Fig. (10)).
iv.
Experimental study of the free peptides in solution
through high resolution 1H NMR spectroscopy,
followed by structure calculation using NMR-derived
structural constraints (NOEs, 3JHNHα) in concert with
CSI suggest helix conformation for the N- and Cterminus (Fig. (6C-E), (7) and (8)).
v.
Minor conformational differences between free and
zinc-bound peptides could be identified for the
secondary structure of the first binding motif
sequence, constituting the first pentapeptide fragment.
The conformational freedom of the first peptide zincligand, His1, could be responsible for this structural
variation, for which however the helical structure has
also been predicted (Fig. (5C)).
vi.
Remarkable backbone conformational similarities
could be identified even between the free model
peptides studied in solution and the structure of
TLN’s active site in solid state. Among them the two
helices in homologous regions of the polypeptide
chains such as the binding motifs (the N- and CACEN peptide termini) are of immense interest (Fig.
(10), upper panel).
vii.
The NMR solution structure of the free ACE N peptide,
the 3D homology ACE-Zn models and the TLN active
site structure exhibit striking similarities. Data suggest
that even in the absence of metal ion the synthesised
polypeptide folds in a tertiary structure where the
helical content is comparable to that of the model and
viii.
The conformation of the 23-residue spacer could
either retain or perturb the 310 helix conformation of
the Arg20-Gly22 fragment observed in free peptide
NMR structure. The small negative value exhibited by
Leu19 (see Fig. (8C) bottom) could be considered
within the error of the CSI method and if that happens
would indicate a helix-stop signal for this peptide
fragment. Nevertheless, a non-helix, coil-like structure
is consistent with TLN active site and helix structure
could be also consistent with the Neurolysin structure,
which possess a spacer similar to ACE 24-residue
with a small helical fragment (see also Fig. (3)).
ix.
One of the main structure-property related differences
between the TLN and ACE active centre is the lack of
any positively-charged residue in the TLN
polypeptide fragment, which comprise its zinccontaining catalytic site (Fig. (10), lower panel).
Electrostatic potentials not only in the surface but also
in the substrate channel and cavity are crucial for the
molecular recognition and complex formation process
between enzyme and substrate.
6.2. Outlook and Perspectives
According to the above data, free and zinc peptides
exhibit conformational features highly similar to the zinccontaining active sites of various known zinc metallopeptidases that belong to the same super-family as ACE, the
gluzincins. The structural calculation based on NMR-derived
constraints for the two ACE-Zn peptides is under way and
the models of both zinc-containing ACE catalytic sites will
become available in the form of 36-residue peptides. These
models could be used in enzyme-substrate complex
simulation/docking studies in an effort to understand the
factors governing the association of the two molecules.
The structural information acquired for ACE active sites
through NMR spectroscopy and structure calculation indicate
that the multidimensional process based on design, solidphase synthesis and conformational analysis of ACE metalbinding sequences seems to be able to provide a suitable
maquette of the actual ACE catalytic sites. Valuable information could be acquired and would be exploited in structurebased design of biologically interesting substances against
hypertension, until the three-dimensional structure of the
entire Angiotensin-I Converting Enzyme becomes available.
Note added in proof
While this paper was in press, the crystal structures of human
testicular [148] and Drosophila [149] ACE were reported.
Additionally, shortly before the release of ACE model, the
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
425
Fig. (10). Comparison between the ribbon representation of the 3D structures of zinc active sites of (A) Thermolysin, (B), ACE C (C) ACEN
3D homology models, and (D) NMR structure is presented in upper panel. The helix angles calculated in terms of helix dipoles are also
given. The distribution of the electrostatic potential on the surface of the models is also presented for the peptides in lower panel. Figures
were generated with the program MOLMOL [147].
Xray structure of another gluzincin metalloprotease, that of
Carboxypeptidase Pyrococcus Furiosus (M32 clan), had
been determined [150].
ACKNOWLEDGEMENTS.
University of Patras for a K. Karatheodoris Research Grant
(P.C., G.A.S.) and General Secretariat of Research and
Technology of Greece-Pened 99 Program (G.P., E.M.-Z.)
are acknowledged for financial support. We also thank
EC’s Access to Research Infrastructures Action of the
Improving Human Potential Program (PARABIO, Contract
No. HPRI-CT-1999-00009) for further support (G.A.S., P.C.
and A.S.G.).
REFERENCES
[1]
[2]
Inagami, T. The renin–angiotensin system. Essays Biochem. 1994,
28, 147–164.
Ondetti, M. A.; Cushman, D. W. Enzymes of the renin-angiotensin
system and their inhibitors. Annu. Rev. Biochem. 1982, 51, 283-308.
426
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
Roks, A.; Buikema, H.; Pinto, Y. M.; van Gilst, W. H. The reninangiotensin system and vascular function. The role of angiotensin
II, angiotensin-converting enzyme, and alternative conversion of
angiotensin I. Heart Vessels 1997, 12, 119-124.
Skeggs, L. T.; Kahn, J. R.; Shumway, N. P. Preparation and
function of the hypertension converting enzyme. J. Exp. Med. 1956,
103, 295-299.
Soffer, R. L. Angiotensin-converting enzyme and the regulation of
vasoactive peptides. Annu. Rev. Biochem. 1976;45, 73-94.
Ehlers, M. R.; Riordan, J. F. Angiotensin-converting enzyme: new
concepts concerning its biological role. Biochemistry 1989, 28,
5311-5318.
Dorer, F. E.; Kahn, J. R.; Lentz, K. E.; Levine, M.; Skeggs, L. T.
Hydrolysis of bradykinin by angiotensin-converting enzyme. Circ.
Res. 1974, 24, 824–827.
Kokkonen, J. O.; Lindstedt, K. A.; Kuoppala, A.; Kovanen, P. T.
Kinin-degrading pathways in the human heart. Trends Cardiovasc.
Med. 2000, 10, 42-45.
Yang, H. Y. Y.; Erdös, E. G.; Levin, V. A dipeptidyl
carboxypeptidase that converts angiotensin I and inactivates
bradykinin. Biochem. Biophys. Acta 1970, 214, 374-376.
Soubrier, F.; Alhenc-Gelas, F.; Hubert, C.; Allegrini, J.; John, M.;
Tregear, G.; Corvol, P. Two putative active centers in human
angiotensin I-converting enzyme revealed by molecular cloning.
Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 9386-9390.
Williams, T. A.; Corvol, P.; Soubrier, F. Identification of two active
site residues in human angiotensin I-converting enzyme. J. Biol.
Chem. 1994, 269, 29430-29434.
Skidgel, R. A.; Erdös, E. G. Novel activity of human angiotensin I
converting enzyme: release of the NH2- and COOH-terminal
tripeptides from the luteinizing hormone-releasing hormone. Proc.
Natl. Acad. Sci. U.S.A. 1985, 82, 1025-1029.
Ehlers, M. R.; Riordan, J. F. Angiotensin-converting enzyme: zincand inhibitor-binding stoichiometries of the somatic and testis
isozymes. Biochemistry 1991, 30, 7118-7126.
Gavras, H. Angiotensin converting enzyme inhibition and its impact
on cardiovascular disease. Circulation 1990, 81, 381-388.
Ferreira, S. H.; Bartelt, D. C.; Greene, L. J. Isolation of bradykininpotentiating peptides from Bothrops jararaca venom . Biochemistry
1970, 9, 2583-2593.
Ondetti, M. A.; Williams, N. J.; Sabo, E. F.; Pluscec, J.; Weaver, E.
R.; Kocy, O. Angiotensin-converting enzyme inhibitors from the
venom of Bothrops jararaca. Isolation, elucidation of structure, and
synthesis. Biochemistry 1971, 10, 4033-4039.
Cheung, H. S.; Cushman, D. W. Inhibition of homogenous
angiotensin-converting enzyme of rabbit lung by synthetic venom
peptides of Bothrops jararaca. Biochim. Biophys. Acta 1973, 293,
451-463.
Steitz, T. A.; Ludwig, M. L.; Quiocho, F. A.; Lipscomb, W. N. The
structure of carboxypeptidase A. V. Studies of enzyme-substrate
and enzyme-inhibitor complexes at 6 Å resolution. J. Biol. Chem.
1967, 242, 4662-4668.
Cushman, D. W.; Ondetti, M. A. Design of angiotensin converting
enzyme inhibitors. Nat. Med. 1999, 5, 1110-1113.
Vallee, B. L.; Auld, D. S. Zinc Coordination, Function, and
Structure of Zinc Enzymes and Other Proteins. Biochemistry 1990,
29, 5647-5659.
Vallee, B. L.; Auld, D. S. New Perspective on Zinc Biochemistry:
Cocatalytic Sites in Multi-Zinc Enzymes. Biochemistry 1993, 32,
6493-6500.
Vallee, B. L; Auld, D. S. Active-site zinc ligands and activated H2O
of zinc enzymes. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 220-224.
Auld, D. S. Zinc sites in metalloenzymes and related proteins. In
Handbook on Metallopreoteins; Bertini, I.; Sigel, A.; Sigel, H., Eds,
Marcel Dekker, 2001, pp. 881-959.
Hooper, N. M. Families of zinc metalloproteases. FEBS Lett. 1994,
354, 1-6.
Thunnissen, M. M. G. M.; Nordlund, P. N.; Haeggstrom, J. Z.
Crystal Structure of Human Leukotriene A4 Hydrolase, a
Bifunctional Enzyme in Inflammation. Nat. Struct. Biol. 2001, 8,
131-135.
Brown, C. K.; Madauss, K.; Lian, W.; Beck, M. R.; Tolbert, W. D.;
Rodgers, D. W. Structure of Neurolysin Reveals a Deep Channel
that Limits Substrate Access. Proc. Nat. Acad. Sci.U.S.A. 2001, 98,
3127-3132.
Spyroulias et al.
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
Holland, D. R.; Hausrath, A. C.; Juers, D.; Matthews, B. W.
Structural analysis of zinc substitutions in the active site of
thermolysin. Protein Sci. 1995, 4, 1955-1965.
Oefner, C.; D'Arcy, A.; Hennig, M.; Winkler, F. K.; Dale, G. E.
Structure of Human Neutral Endopeptidase (Neprilysin) Complexed
with Phosphoramidon. J. Mol. Biol. 2000, 296, 341-349
Lacy, D. B.; Tepp, W.; Cohen, A. C.; Dasgupta, B. R.; Stevens, R.
C. Crystal Structure of Botulinum Neurotoxin Type a and
Implications for Toxicity. Nat. Struct. Biol. 1998, 5, 898-902.
Pannifer, A. D.; Wong, T. Y.; Schwarzenbacher, R.; Renatus, M.;
Petosa, C.; Bienkowska, J.; Lacy, D. B.; Collier, R. J.; Park, S.;
Leppla, S. H.; Hanna, P.; Liddington, R. C. Crystal Structure of the
Anthrax Lethal Factor. Nature 2001, 414, 229-233
Kurisu, G.; Kinoshita, T.; Sugimoto, A.; Nagara, A.; Kai, Y.; Kasai,
N.; Harada, S. Structure of the zinc endoprotease from
Streptomyces caespitosus. J. Biochem. 1997, 121, 304-308.
Schlagenhauf, E.; Etges, R.; Metcalf, P. The crystal structure of the
Leishmania major surface proteinase leishmanolysin (gp63).
Structure 1998, 6, 1035-1046.
Li, J.; Brick, P.; O'Hare, M. C.; Skarzynski, T.; Lloyd, L. F.; Curry,
V. A.; Clark, I. M.; Bigg, H. F.; Hazleman, B. L.; Cawston, T. E.;
Blow, D. M. Structure of full-length porcine synovial collagenase
reveals a C-terminal domain containing a calcium-linked, fourbladed beta-propeller. Structure 1995, 3, 541-549.
Bode, W.; Gomis-Ruth, F. X.; Huber, R.; Zwilling, R.; Stocker, W.
Structure of astacin and implications for activation of astacins and
zinc-ligation of collagenases. Nature 1992, 358, 164-167.
Mcauley, K. E.; Jia-Xing, Y.; Dodson, E. J.; Lehmbeck, J.;
Ostergaard, P. R.; Wilson, K. S. A Quick Solution: Ab Initio
Structure Determination of a 19 kDa Metalloproteinase Using
Acorn Acta Crystallogr., Sect. D 2001, 57, 1571-1578.
Greenblatt, H. M.; Feinberg, H.; Tucker, P. A.; Shoham, G.
Carboxypeptidase A: native, zinc-removed and mercury-replaced
forms. Acta Crystallogr., Sect. D. 1998, 54, 289-305.
1lbu is the pdb code for the “Refined Structure of a Zn2+ D-AlanylD-Alanine Carboxypeptidase of S. Albus G at 1.8 Angstroms
Resolution” (Wery, J.-P.; Charlier, P.; Dideberg, O.) which refers to
the original enzyme structure with pdb code 0zgp : Dideberg, O.;
Charlier, P.; Dive, G.; Joris, B.; Frere, J. M.; Ghuysen, J. M.
Structure of a Zn2+-containing D-alanyl-D-alanine-cleaving
carboxypeptidase at 2.5 A resolution. Nature 1982 299, 469-470.
Taylor, A. B.; Smith, B. S.; Kitada, S.; Kojima, K.; Miyaura, H.;
Otwinowski, Z.; Ito, A.; Deisenhofer, J. Crystal Structures of
Mitochondrial Processing Peptidase Reveal the Mode for Specific
Cleavage of Import Signal Sequences. Structure 2001, 9, 615-625.
Burley, S. K.; David, P. R.; Taylor, A.; Lipscomb, W. N. Molecular
structure of leucine aminopeptidase at 2.7-A resolution. Proc. Natl.
Acad. Sci. U.S.A. 1990, 87, 6878-6882.
Roderick, S. L.; Matthews, B. W. Structure of the cobalt-dependent
methionine aminopeptidase from Escherichia coli: a new type of
proteolytic enzyme. Biochemistry 1993, 32, 3907-3912.
Rowsell, S.; Pauptit, R. A.; Tucker, A. D.; Melton, R. G.; Blow, D.
M.; Brick, P. Crystal structure of carboxypeptidase G2, a bacterial
enzyme with applications in cancer therapy. Structure 1997, 5, 337347.
Chevrier, B.; Schalk, C.; D'Orchymont, H.; Rondeau, J. M.; Moras,
D.; Tarnus, C. Crystal structure of Aeromonas proteolytica
aminopeptidase: a prototypical member of the co-catalytic zinc
enzyme family. Structure 1994, 2, 283-291.
Pearson, M. A.; Michel, L. O.; Hausinger, R. P.; Karplus, P. A.
Structures of Cys319 variants and acetohydroxamate-inhibited
Klebsiella aerogenes urease. Biochemistry 1997, 36, 8164-8172.
Fritsche, E.; Paschos, A.; Beisel, H.-G.; Bock, A.; Huber, R. Crystal
Structure of the Hydrogenase Maturating Endopeptidase Hybd from
Escherichia Coli. J. Mol. Biol. 1999, 288, 989-998.
Laszlo, P.; Blundell, T. L. Sequences and topology. Curr. Opin.
Struct. Biol. 1995, 5, 357-359
Jongeneel, C. V.; Bouvier, J.; Bairoch, A. A unique signature
identifies a family of zinc-dependent metallopeptidases. FEBS Lett.
1989, 242, 211-214.
Vallee, B. L.; Auld, D. S. Short and long spacer sequences and
other structural features of zinc binding sites in zinc enzymes.
FEBS Lett. 1989, 257, 138-140.
Jiang, W.; Bond, J. S. Families of metalloendopeptidases and their
relationships. FEBS Lett. 1992, 312, 110-114.
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71]
Bode, W.; Gomis-Rüth, F.-X.; Stöckler, W. Astacins, serralysins,
snake venom and matrix metalloproteinases exhibit identical zincbinding environments (HEXXHXXGXXH and Met-turn) and
topologies and should be grouped into a common family, the
`metzincins'. FEBS Lett. 1993, 331, 134-140.
Rawlings, N. D.; Barrett, A. J. Evolutionary families of peptidases.
Biochem. J. 1993, 290, 205-218.
Blundell, T. L. Metalloproteinase superfamilies and drug design.
Nat. Struct. Biol. 1994, 1, 73-75.
Bode, W.; Stöckler, W. Structural features of a superfamily of zincendopeptidases: the metzincins. Curr. Opin.. Struct. Biol. 1995, 5,
383-390.
Colman, P. M; Jansonius, J. N.; Matthews, B. W. The structure of
thermolysin : an electron density map at 2-3 A resolution. J. Mol.
Biol. 1972, 70, 701-24.
Matthews B.W., Weaver L.H., Kester W.R. The conformation of
thermolysin. J. Biol. Chem. 1974, 249, 8030-8044.
Holmes MA, Matthews BW. Structure of thermolysin refined at 1.6
A resolution. J. Mol. Biol. 1982, 160, 623-639.
Baumann U. Crystal structure of the 50 kDa metallo protease from
Serratia marcescens. J. Mol. Biol. 1994, 242, 244-251.
Katsuya, Y.; Hamada, K.; Hata, Y.; Tanaka, N.; Sato, M., Katsube,
Y.; Kakiuchi, K.; Miyata, K. Preliminary X-ray studies on Serratia
protease. J. Biochem. 1985, 98, 1139-1142.
pdb code 1srp is linked to the reference : Hamada, K.; Hiramatsu,
H.; Katsuya, Y.; Hata, Y.; Matsuura, Y.; Katsube, Y. Structural
Analysis of Serratia Protease. Acta Crystallogr., Sect. A 1993, 49,
153. However, this citation seems to be a misprint, as it does not
exist to the corresponding journal. A relative article to serralysin
structure should be the following : Katsuya, Y.; Hamada, K.; Hata,
Y.; Tanaka, N.; Sato, M.; Katsube, Y.; Kakiuchi, K.; Miyata, K.;
Preliminary X-ray studies on Serratia protease. J. Biochem. 1985
98, 1139-1142.
Spurlino, J. C.; Smallwood, A. M.; Carlton, D. D.; Banks, T. M.;
Vavra, K. J.; Johnson, J. S.; Cook, E. R.; Falvo, J.; Wahl, R. C.;
Pulvino, T. A.; Wendoloski, J. J.; Smith, D. L. 1.56 Å structure of
mature truncated human fibroblast collagenase. Proteins 1994, 19,
98-109.
Browner M. F.; Smith W. W.; Castelhano A. L. Matrilysin-inhibitor
complexes : common themes among metalloproteases.
Biochemistry 1995, 34, 6602-6610.
Gomis-Ruth, F. X.; Kress, L. F.; Bode, W. First structure of a snake
venom
metalloproteinase:
a
prototype
for
matrix
metalloproteinases/collagenases. EMBO J. 1993, 12, 4151-4157.
Hori, T.; Kumasaka, T.; Yamamoto, M.; Nonaka, T.; Tanaka, N.;
Hashimoto, Y.; Ueki, T.; Takio, K. Structure of a New 'Aspzincin'
Metalloendopeptidase from Grifola Frondosa: Implications for the
Catalytic Mechanism and Substrate Specificity Based on Several
Different Crystal Forms. Acta Crystallogr., Sect. D 2001, 57, 361368.
Christianson, D. W.; David, P. R.; Lipscomb, W. N. Mechanism of
carboxypeptidase A: hydration of a ketonic substrate analogue.
Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 1512-1515.
Christianson, D. W.; Lipscomb, W. B. Carboxypeptidase-A. Acc.
Chem. Res. 1989, 22, 62-69.
Osterman, A. L.; Grishin, N. V.; Smulevitch, S. V.; Matz, M. V.;
Zagnitko, O. P.; Revina, L. P.; Stepanov, V. M. Primary structure of
carboxypeptidase T: Delineation of functionally relevant features in
Zn-carboxypeptidase family. J. Protein Chem. 1992, 11, 561-570.
Lipscomb, W. N.; Sträter N. Recent Advances in Zinc Enzymology.
Chem. Rev. 1996, 96, 2375-2433.
Makarova, K. S.; Grishin, N. V. The Zn-peptidase superfamily:
functional convergence after evolutionary divergence. J. Mol. Biol.
1999, 292, 11-17.
Greenblatt, H. M.; Feinberg, H.; Tucker, P. A.; Shoham, G.
Carboxypeptidase A: native, zinc-removed and mercury-replaced
forms. Acta Crystallogr., Sect. D 1998, 54, 289-305.
Blundell T. L. Metalloproteinase superfamilies and drug design.
Nat. Struct. Biol. 1992, 2, 73-75.
Cushman, D. W.; Pluscec, J.; Williams, N. J.; Weaver, E. R.; Sabo,
E. F.; Kocy, O.; Cheung, H. S.; Ondetti, M. A. Inhibition of
angiotensin-coverting enzyme by analogs of peptides from
Bothrops jararaca venom. Experientia 1973, 29, 1032-1035.
Elisseeva, Y. E.; Orekhovich, V. N.; Pavlikhina, L. V.; Alexeenko,
L. P. Carboxycathepsin--a key regulatory component of two
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
[72]
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
[88]
[89]
[90]
[91]
427
physiological systems involved in regulation of blood pressure.
Clin. Chim. Acta. 1971, 31, 413-419.
Ondetti, M. A.; Rubin, B.; Cushman, D. W. Design of specific
inhibitors of angiotensin-converting enzyme: new class of orally
active antihypertensive agents. Science 1977, 196, 441-444.
Quiocho, F. A.; Lipscombe, W. N. Carboxypeptidase A: a protein
and an enzyme. Adv. Protein Chem. 1971, 25, 1-78.
Pelmenschikov, V.; Blomberg, M. R. A.; Siegbahn, P. E. M. A
theoretical study of the mechanism for peptide hydrolysis by
thermolysin. J. Biol. Inorg. Chem. 2002, 7, 284-298.
Kester, W. R.; Matthews, B. W. Crystallographic study of the
binding of dipeptide inhibitors to thermolysin: implications for the
mechanism of catalysis. Biochemistry 1977, 16, 2506-2516.
Holmes, M. A.; Matthews, B. W. Binding of hydroxamic acid
inhibitors to crystalline thermolysin suggests a pentacoordinate zinc
intermediate in catalysis. Biochemistry 1981, 20, 6912-6920.
Hangauer, D. G.; Monzingo A. F.; Matthews, B. W. An interactive
computer graphics study of thermolysin-catalyzed peptide cleavage
and inhibition by N-carboxymethyl dipeptides. Biochemistry 1984,
23, 5730-5741.
Holden, H. M.; Tronrud, D. E.; Monzingo, A. F.; Weaver, L. H.;
Matthews, B. W. Slow- and fast-binding inhibitors of thermolysin
display different modes of binding: crystallographic analysis of
extended phosphonamidate transition-state analogues. Biochemistry
1987, 26, 8542-8553.
Tronrud, D. E.; Holden, H. M.; Matthews, B. W. Structures of two
thermolysin-inhibitor complexes that differ by a single hydrogen
bond. Science 1987, 30, 571-574.
Matthews, B. W. Structural Basis of the Action of Thermolysin and
Related Zinc Peptidases. Acc. Chem. Res. 1988, 21, 333-340.
Jaspard, E.; Wei, L.; Alhenc-Gelas, F. Differences in the properties
and enzymatic specificities of the two active sites of angiotensin Iconverting enzyme (kininase II). Studies with bradykinin and other
natural peptides. J. Biol. Chem. 1993, 268, 9496-9503.
Erdös, E. G.; Deddish, P. A.; Marcic, B. M. Potentiation of
Bradykinin Actions by ACE Inhibitors. Trends Endocrinol. Metab.
1999, 10, 223-229.
Skidgel, R. A.; Erdos, E. G. Novel activity of human angiotensin I
converting enzyme: release of the NH2- and COOH-terminal
tripeptides from the luteinizing hormone-releasing hormone. Proc.
Natl. Acad. Sci. U. S. A. 1985, 82, 1025-1029.
Ehlers, M. R.; Riordan, J. F. Angiotensin-converting enzyme: zincand inhibitor-binding stoichiometries of the somatic and testis
isozymes. Biochemistry 1991, 30, 7118-7126.
Casarini, D. E.; Plavinik, F. L.; Zanella, M. T.; Marson, O.;
Krieger, J. E.; Hirata, I.Y.; Stella, R. C. Angiotensin converting
enzymes from human urine of mild hypertensive untreated patients
resemble the N-terminal fragment of human angiotensin Iconverting enzyme. Int. J. Biochem. Cell Biol. 2001, 33, 75-85.
Deddish, P. A.; Marcic, B.; Jackman, H. L.; Wang, H. Z.; Skidgel,
R. A.; Erdös, E. G. N-domain-specific substrate and C-domain
inhibitors of angiotensin-converting enzyme: angiotensin-(1-7) and
keto-ACE. Hypertension 1998, 31, 912-917.
Dive, V.; Cotton, J.; Yiotakis, A.; Michaud, A.; Vassiliou, S.;
Jiracek, J.; Vazeux, G.; Chauvet, M. T.; Cuniasse, P.; Corvol, P.
RXP 407, a phosphinic peptide, is a potent inhibitor of angiotensin I
converting enzyme able to differentiate between its two active sites.
Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 4330-4335.
Junot, C.; Gonzales, M. F.; Ezan, E.; Cotton, J.; Vazeux, G.;
Michaud, A.; Azizi, M.; Vassiliou, S.; Yiotakis, A.; Corvol, P.;
Dive, V. RXP 407, a selective inhibitor of the N-domain of
angiotensin I-converting enzyme, blocks in vivo the degradation of
hemoregulatory peptide acetyl-Ser-Asp-Lys-Pro with no effect on
angiotensin I hydrolysis. J. Pharmacol. Exp. Ther. 2001, 297, 606611.
Corvol, P.; Williams, T. A. In Handbook of Proteolytic Enzymes;
Barrett, A. J.; Rawlings, N. D.; Woessner, J. F., Eds.; Academic
Press: New York, 1998; pp 1066-1076.
Shapiro, R.; Holmquist, B.; Riordan, J. F. Anion activation of
angiotensin converting enzyme: dependence on nature of substrate.
Biochemistry 1983, 22, 3850-3857.
Lamango, N. S.; Sajid, M.; Isaac, R. E. The endopeptidase activity
and the activation by Cl- of angiotensin-converting enzyme is
evolutionarily conserved: purification and properties of an
428
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
[111]
[112]
[113]
[114]
[115]
Fernandez, M.; Liu, X.; Wouters, M. A.; Heyberger, S.; Husain, A.
Angiotensin I-converting enzyme transition state stabilization by
His1089: evidence for a catalytic mechanism distinct from other
gluzincin metalloproteinases. J. Biol. Chem. 2001, 276, 4998-5004.
Chen, Y. N.; Ehlers, M. R.; Riordan, J. F. The functional role of
tyrosine-200 in human testis angiotensin-converting enzyme.
Biochem. Biophys. Res. Commun. 1992, 184, 306-309.
Sen, I.; Kasturi, S.; Abdul Jabbar, M.; Sen, G. C. Mutations in two
specific residues of testicular angiotensin-converting enzyme
change its catalytic properties. J. Biol. Chem. 1993, 268, 2574825754.
Thompson, J. D.; Gibson, T. J.; Plewniak, F.; Jeanmougin, F.;
Higgins, D. G. The CLUSTAL_X windows interface: flexible
strategies for multiple sequence alignment aided by quality analysis
tools. Nucleic Acids Res. 1997, 25, 4876-4882.
Martí-Renom, M. A.; Stuart, A. C.; Fiser, A.; Sánchez, R.; Melo,
F.; Sali, A. Comparative protein structure modeling of genes and
genomes. Annu. Rev. Biophys. Biomol. Struct. 2000, 29, 291-325.
Combet, C.; Blanchet, C.; Geourjon, C.; Deléage, G. NPS@:
Network Protein Sequence Analysis. TIBS 2000, 25, 147-150.
Garnier, J.; Osguthorpe, D. J.; Robson, B. Analysis of the accuracy
and implications of simple methods for predicting the secondary
structure of globular proteins. J. Mol. Biol. 1978, 120, 97-120.
Gibrat, J. F.; Garnier, J.; Robson, B. Further developments of
protein secondary structure prediction using information theory.
New parameters and consideration of residue pairs. J. Mol. Biol.
1987, 198, 425-443.
Garnier, J.; Gibrat, J. F.; Robson, B. GOR method for predicting
protein secondary structure from amino acid sequence. Methods
Enzymol. 1996, 266, 540-553.
Merrifield R. B. Solid Phase Synthesis. I. The synthesis of a
Tetrapeptide. J. Am. Chem. Soc. 1963, 85, 2149-2154.
Veglia, G.; Porcelli, F.; DeSilva, T.; Prantner, A.; Opella, S.J. The
structure of the metal-binding motiff GMTCAAC is similar in an
18-residue linear peptide and the mercury binding protein Merp. J.
Am. Chem. Soc. 2000, 122, 2389-2390.
Opella, S. J.; DeSilva, T. M.; Veglia, G. Structural biology of
metal-binding sequence. Curr. Opin. Chem. Biol. 2002, 6, 217-223.
DeSilva, T. M.; Veglia, T.; Porcelli, F.; Prantner, A. M.; Opella, S.
J. Selectivity in heavy metal-binding to peptides and proteins.
Biopolymers 2002, 64, 189-197.
Carpino, L. A.; Han, G. Y.; 9-Fluorenylmethoxycarbonyl function,
a new base-sensitive amino-protecting group J. Am. Chem. Soc.
1970, 92, 5748-5749.
Barlos, K.; Gatos, D.; Kallitsis, J.; Papaphotiou, G.; Sotiriou, P.;
Wenquing, Y.; Shafer, W. Darstellung geshützter peptid-fragmente
unter einsatz substituierter triphenylmethyl-harze, Tetraedron Lett.
1989, 30, 3943-3946.
Papazacharias, S.; Pairas, G.; Kouimtzoglou, E.; Dermitzaki, E.;
Manessi-Zoupa, E.; Gravanis, A.; Cordopatis, P. Synthesis and
Biological Evaluation of New h/rCRH Peptide Analogs. In Peptides
2000 ; Martinez, J.; Fehrentz, J.-A., Eds, EDK, 2001, pp. 933-934.
King, D. S.; Fields, C. G.; Fields G. B. A cleavage method which
minimizes side reactions following Fmoc solid phase peptide
synthesis. Int. J. Pept. Protein Res. 1990, 36, 255-266
Braunschweiler, L.; Ernst, R. R. Coherence transfer by isotropic
mixing : application to proton correlation spectroscopy. J. Magn.
Reson. 1983, 53, 521-528.
Bax, A.; Davis, D. G. MLEV-17-based two-dimensional
homonuclear magnetization transfer spectroscopy. J. Magn. Reson.
1985, 65, 355-360.
Marion, D.; Wüthrich, K. Application of phase sensitive twodimensional correlated spectroscopy (COSY) for measurements of
1
H-1H spin-spin coupling constants in proteins. Biochem. Biophys.
Res. Commun. 1983, 113, 967-974.
Jeener, J., Meier, B. H., Bachmann, P.; Ernst R.R. Investigation of
exchange processes by two-dimensional NMR spectroscopy. J.
Chem. Phys. 1979, 71, 4546-4553.
Wüthrich, K. In NMR of Proteins and Nucleic Acids; Wiley: New
York, 1986.
Wang, Y.; Nip, A.M.; Wishart D. S. A simple method to
quantitatively measure polypeptide JHNHα coupling constants from
TOCSY or NOESY spectra. J. Biomol. NMR 1997, 10, 373-382.
Smith, L. J.; Bolin, K. A.; Schwalbe, H.; MacArthur, M. W.;
Thorton, J. M.; Dobson, C. M. Analysis of main chain torsion
>
angiotensin-converting enzyme from the housefly, Musca
domestica. Biochem. J. 1996, 314, 639-646.
[92]
Wei, L.; Alhenc-Gelas, F.; Corvol, P.; Clauser, E. The two
Homologous Domais of Human Angiotensin I-converting Enzyme
Are Both Catalically Active. J. Biol. Chem. 1991, 266, 9002-9006.
[93]
Kester, W. R.; Matthews, B. W. Comparison of the structures of
carboxypeptidase A and thermolysin. J. Biol. Chem. 1977, 252,
7704-7710.
[94]
Mayer, M.; Meyer, B. Mapping the active site of angiotensinconverting enzyme by Transferred NOE Spectroscopy. J. Med.
Chem 2000, 43, 2093-2097.
[95]
Bünning, P.; Holmquist, B.; Riordan, J. F. Substrate specificity and
kinetic characteristics of angiotensin converting enzyme.
Biochemistry 1983, 22, 103-110.
[96]
Patchett, A. A.; Cordes, E. H. The design and properties of Ncarboxyalkyldipeptide inhibitors of angiotensin-converting enzyme.
Adv. Enzymol. Relat. Areas Mol. Biol. 1985, 57, 1-84.
[97]
Patchett, A. A.; Harris, E.; Tristram, E. W.; Wyvratt, M. J.; Wu, M.
T.; Taub, D. Peterson, E. R.; Ikeler, T.; ten Broeke, J.; Payne, L. G.;
Ondeyka, D. L.; Thorsett, E. D.; Greenlee, W. J.; Lohr, N. S.;
Hoffsommer, R. D.; Joshua, H.; Ruyle, W. V.; Rothrock, J. W.;
Aster, S. D.; Maycock, A. L.; Robinson, F. M.; Hirschmann, R.;
Sweet, C. S.; Ulm, E. H.; Gross, D. M.; Vassil, T. C.; Stone, C. A.
A new class of angiotensin-converting enzyme inhibitors. Nature
1980, 288, 280-283.
[98]
Moore, V. A.; Irwin, W. J.; Timmins, P.; Chong, S.; Dando, S. A.;
Morrison, R. A. A rapid screening system to determine drug
affinities for the intestinal dipeptide transporter 1 : system
characterisation. Inter. J. Pharm. 2000, 210, 15-27.
[99] Moore, V. A.; Irwin, W. J.; Timmins, P.; Chong, S.; Dando, S. A.;
Morrison, R. A. A rapid screening system to determine drug
affinities for the intestinal dipeptide transporter 2: affinities of ACE
inhibitors. Inter. J. Pharm. 2000, 210, 29-44.
[100] Cushman, D. W.; Cheung, H. S.; Sabo, E. F.; Ondetti, M. A. Design
of potent competitive inhibitors of angiotensin-converting enzyme.
Carboxyalkanoyl and mercaptoalkanoyl amino acids Biochemistry
1977, 16, 5484-5491.
[101] Petrillo, E. W.; Ondetti, M. A. Angiotensin-converting enzyme
inhibitors: medicinal chemistry and biological actions. Med. Res.
Rev. 1982, 2, 1-41.
[102] Cushman, D. W.; Cheung, H. S.; Sabo, E. F.; Ondetti, M. A.
Development and design of specific inhibitors of angiotensinconverting enzyme. Am. J. Cardiol. 1982, 49, 1390-1394.
[103] Wu, M. T.; Douglas, A. W.; Ondeyka, D. L.; Payne, L. G.; Ikeler,
T. J. Joshua, H.; Patchett, A. A.; Synthesis of N2-[(S)-1-carboxy-3phenylpropyl]-L-lysyl-L-proline (lisinopril). J. Pharm. Sci. 1985,
74, 352-354.
[104] Krapcho, J.; Turk, C.; Cushman, D. W.; Powell, J. R.; DeForrest, J.
M.; Spitzmiller, E. R.; Karanewsky, D. S.; Duggan, M.; Rovnsak,
G.; Schwartz, J.; Natarajan, S.; Godfrey, J. D.; Ryono, D. E.;
Neubeck, R.; Atwal, K. S.; Petrillo, E. W. Jr. Angiotensinconverting enzyme inhibitors. Mercaptan, carboxyalkyl dipeptide,
and phosphinic acid inhibitors incorporating 4-substituted prolines.
J. Med. Chem. 1988, 31, 1148-1160.
[105] Michaud, A.; Chauvet, M. T.; Corvol, P. N-domain selectivity of
angiotensin I-converting enzyme as assessed by structure-function
studies of its highly selective substrate, N-acetyl-seryl-aspartyllysyl-proline. Biochem. Pharmacol. 1999, 57, 611-618.
[106] Araujo, M. C.; Melo, R. L.; Cesari, M. H.; Juliano, M. A.; Juliano,
L.; Carmona, A. K. Peptidase specificity characterization of C- and
N-terminal catalytic sites of angiotensin I-converting enzyme.
Biochemistry 2000, 39, 8519-8525.
[107] Ehlers, M. R. W.; Kirsch, R. E. Catalysis of Angiotensin I
Hydrolysis by Human Angiotensin-Converting Enzyme : Effect of
Chloride and pH. Biochemistry, 1988, 27, 5538-5544.
[108] Liu, X.; Fernandez, M.; Wouters, M. A.; Heyberger, S.; Husain, A.
Arg(1098) is critical for the chloride dependence of human
angiotensin I-converting enzyme C-domain catalytic activity. J.
Biol. Chem. 2001, 276, 33518-33525.
[109] Bünning, P.; Holmquist, B.; Riordan, J. F. Functional residues at
the active site of angiotensin converting enzyme. Biochem. Biophys.
Res. Commun. 1978, 83, 1442-1449.
[110] Shapiro, R.; Riordan, J. F. Critical Lysine Residue at the Chloride
Binding Site of Angiotensin Converting Enzyme. Biochemistry
1983, 22, 5315-5321.
Spyroulias et al.
[116]
[117]
[118]
[119]
[120]
[121]
[122]
[123]
[124]
[125]
[126]
[127]
[128]
[129]
[130]
[131]
[132]
[133]
[134]
Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
[135]
[136]
[137]
[138]
[139]
[140]
[141]
angles in proteins: prediction of NMR coupling constants for native
and random coil conformations. J. Mol. Biol. 1996, 255, 494-506.
Wishart, D.; Sykes, B.; Richards, F. Relationship between nuclear
magnetic resonance chemical shift and protein secondary structure.
J. Mol. Biol. 1991, 222, 311-333.
Wishart, D.; Sykes, B.; Richards, F. The chemical shift index: a fast
and simple method for the assignment of protein secondary
structure through NMR spectroscopy. Biochemistry 1992, 31, 16471651.
Merutka, G.; Dyson, H. J.; Wright, P. E. 'Random coil' 1H chemical
shifts obtained as a function of temperature and trifluoroethanol
concentration for the peptide series GGXGG. J. Biomol. NMR
1995, 5, 14-24.
Galanis, A. S.; Spyroulias, G. A.; Pairas, G.; Manessi-Zoupa, E.;
Cordopatis, P. NMR solution structures of apoform ACEN and
ACEC active sites model peptides. In preparation.
Güntert, P.; Braun, W.; Wüthrich, K. Efficient computation of
three-dimensional protein structures in solution from nuclear
magnetic resonance data using the program DIANA and the
supporting programs CALIBA, HABAS and GLOMSA. J. Mol.
Biol. 1991, 217, 517-530.
Güntert, P.; Mumenthaler, C.; Wüthrich, K. Torsion angle dynamics
for NMR structure calculation with the new program DYANA. J.
Mol. Biol. 1997, 273, 283-298.
Pearlman, D. A.; Case, D. A.; Caldwell, J. W.; Ross, W. S.;
Cheatham, T. E.; Ferguson, D. M.; Seibel, G. L.; Singh, U. C.;
Weiner, P. K.; Kollman, P. A. AMBER 5.0. University of
California, San Francisco, 1997.
Current Topics in Medicinal Chemistry, 2004, Vol. 4, No. 4
[142]
[143]
[144]
[145]
[146]
[147]
[148]
[149]
[150]
429
Pearlman, D. A.; Case, D. A.; Caldwell, G. C.; Siebel, G. L.; Singh,
U. C.; Weiner, P.; Kollman, P. A. AMBER 4.0. University of
California, San Francisco, 1991.
Shortle, D.; Clarke, N. Alpha helix propensity of amino acids.
Science 1993, 260, 1637-1640.
Shortle, D.; Clarke, N. Alpha helix propensity of amino acids.
Science 1993, 262, 917-918.
Zuiderweg, E. R. P. Mapping Protein-Protein Interactions in
Solution by NMR Spectroscopy. Biochemistry, 2002, 41, 1-7.
Galanis, A. S.; Spyroulias, G. A.; Pierattelli, R.; Tzakos, A.;
Troganis, A.; Gerothanassis, I. P.; Pairas, G.; Manessi-Zoupa, E.;
Cordopatis, P. Zinc-Binding in Peptide Models of ACE Active Sites
Studied Through 1H NMR & Chemical Shift Perturbation Mapping.
Biopolymers, 2003, 69, 244-252.
Koradi, R.; Billeter, M.; Wüthrich, K. MOLMOL: a program for
display and analysis of macromolecular structures. J. Mol. Graph.
1996, 14, 51-55.
Natesh, R.; Schwager, S. L. U.; Sturrock, E. D.; Acharya, K. R.
Crystal structure of the human angiotensin-converting enzymelisinopril complex. Nature 2003, 421, 551-554
Kim, H. M.; Shin, D. R.; Yoo, O. J.; Lee, H.; Lee, J. O. Crystal
structure of Drosophila angiotensin I-converting enzyme bound to
captopril and lisinopril. FEBS Lett. 2003, 538, 65-70
Arndt, J. W.; Hao, B.; Ramakrishnan, V.; Cheng, T.; Chan, S. I.;
Chan, M. K. Crystal structure of a novel carboxypeptidase from the
hyperthermophilic archaeon Pyrococcus furiosus. Structure 2002,
215-224.