* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Structural Features of Angiotensin-I Converting Enzyme Catalytic Sites
Point mutation wikipedia , lookup
Ancestral sequence reconstruction wikipedia , lookup
Ligand binding assay wikipedia , lookup
Genetic code wikipedia , lookup
Deoxyribozyme wikipedia , lookup
Evolution of metal ions in biological systems wikipedia , lookup
Zinc finger nuclease wikipedia , lookup
Amino acid synthesis wikipedia , lookup
Peptide synthesis wikipedia , lookup
Enzyme inhibitor wikipedia , lookup
Structural alignment wikipedia , lookup
Proteolysis wikipedia , lookup
Catalytic triad wikipedia , lookup
Nuclear magnetic resonance spectroscopy of proteins wikipedia , lookup
Biosynthesis wikipedia , lookup
Homology modeling wikipedia , lookup
Biochemistry wikipedia , lookup
Ribosomally synthesized and post-translationally modified peptides wikipedia , lookup
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.