New comprehensive biochemistry vol 16 hydrolytic enzymes

In a series of brilliant investigations, they isolated this substance from culture filtrates of a strain ofStreptomyces,establishedits structure, and demonstrated its ability to inhibit

All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted

in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher, Elsevier Science Publishers B.V.(Biomedical Division), P.O Box 1527,

1000 BM Amsterdam (The Netherlands).

No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein Because of the rapid advances in the medical sciences, the Publisher recommends that independent verification of diagnoses and drug dosages should be made Special regulation for readers in the U.S.A.:

This publication has been registered with the Copyright Clearance Center Inc (Ccq, Salem, Massachusetts.

Information can be obtained from the CCC about conditions under which the photocopying of parts of this publication may be made in the U.S.A All other copyright questions, including photocopying outside of the U.S.A., should be referred to the publisher.

ISBN for the series: 0-444-80303-3

ISBN for the volume: 0-444-80886-8

Library of Congress Cataloging-in-Publication Data

(Revised for volume 16)

New comprehensive biochemistry.

Sole distributors for the U.S.A and Canada: Elsevier Science Publishing Company, Inc.

52 Vanderbilt Avenue New York, NY 10017 (U.S.A.)

Vol 16 published: Amsterdam; New York: Elsevier

Science Publishers B.V (Biomedical Division); New York,

N.Y U.S.A.: Sole distributors for the U.S.A and

Canada, Elsevier Science Pub Co.

Includes bibliographical references and indexes.

Contents: v \ Membrane structure / editors, J.B.

Finean and R.H Michell v 3 Stereochemistry /

editor, Ch Tamm v 4 Phospholipids / editors, J.N.

Hawthorne ana G.B Ansell [etc.] v 16 Hydrolytic

enzymes / editors, A Neuberger and K Brocklehurst.

\ Biological chemistry Collected works.

I Neuberger, Albert II Deenen, Laurens L.M van.

[DNLM \ Membranes Anatomy and histology WI NE372F v.l

/ QS 532.5.M534]

QD415.N48 574.19'2 81-3090

ISBN 0-444-80303-3 (Elsevier/North-Holland: set)

Printed in The Netherlands

Hydrolytic Enzymes

Editors

A NEUBERGER and K BROCKLEHURST

The Lister Institute of Preventive Medicine, Charing Cross Hospital Medical School, St Dunstan's Rd., London W6 8RP(U.K.), and Department ofBiochemistry, Medical College

of St Bartholomew's Hospital, University of London, Charterhouse Square, London

ECl M 6BQ (U.K.)

1987 ELSEVIER AMSTERDAM· NEW YORK· OXFORD

A volume dedicated to Hydrolytic Enzymes was perceived as useful for two reasons

In the first place, a number of these enzymes are not dealt with fully in those volumes

of this series in which systems and events are discussed principally in a particularmetabolic or physiological context Secondly, it seems appropriate to bring togetherdiscussion of some of the enzymes that became the focus of attention in the 1960s whenour understanding of the function of protein molecules was revolutionised by theapplication of X-ray crystallography At that time, an account of the structure ofmyoglobin was rapidly followed by reports of the structures of lysozyme, carboxypep-tidase, ribonuclease, chymotrypsin and papain, which permitted for the first time theresults of mechanistic study by kinetic and protein chemical methods to be thoughtabout within a realistic structural framework Six of the eight chapters are devoted tovarious aspects of proteolysis This emphasis is not inappropriate in view of the manyadvances in the search for chemical understanding of biological phenomena that wereachieved during the study of proteolytic enzymes They were among the fIrst enzymes

to be highly purified and crystallised and much of our understanding of molecularaspects of catalytic mechanism and specillcity is founded on the study of these enzymes.Four of the chapters deal with the different types of proteinase that are differentiated

on the basis of Hartley's idea of classifying proteinases by catalytic mechanism ratherthan by origin, specillcity or physiological function These chapters are complemented

by one on proteinase inhibitors and by a short review of intracellular proteolysis Thelatter includes a brief discussion of ATP-dependent proteolysis by ubiquitin, which will

be extended in a subsequent volume dealing with protein metabolism The final twochapters deal respectively with pancreatic ribonuclease A, the best characterised of theendoribonucleases, and with the phosphomonoesterases A particular regret, in view ofthe central importance oflysozyme in the development of studies on structure, specificityand mechanism, is the unavoidable omission of a chapter on glycosidases Itis hopedthat this omission will be rectified in a subsequent volume

The major development of the 1960s in providing three-dimensional structures ofenzymes at atomic resolution is being augmented in the 1980s by the application ofDNA technology to provide designed structural variation in individual amino acidresidues by site-directed mutagenesis This approach should go some way towardsobviating the largest single problem that has held back mechanistic study of enzymecatalysis, namely the inability to vary systematically the structure of both or all of thereactant molecules

We wish to record our thanks to the authors both for their excellent contributions andfor their helpful cooperation in the editorial process

London

December 1987

A Neuberger

K Brocklehurst

(a) Historical background

The aspartyl proteinases represent one of the four known main classes of enzymes thatact at interior peptide bonds of proteins and oligopeptides (endopeptidases); the otherclasses are denoted serine proteinases, cysteine proteinases and metalloproteinases.Because of their optimal action at pH 1.5-5,the aspartyl proteinases were previouslynamed acid proteinases With the recognition that particular carboxyl groups in theseenzymes are essential for catalysis, the term carboxyl proteinase was then used Theidentification of these groups as belonging to aspartyl residues in several members ofthis class has led to the currently-preferred terminology The term 'aspartyl proteinase'(aspartic proteinase and aspartate proteinase have also been used) is more appropriatethan 'acid proteinase' because some enzymes now known to belong to this class actoptimally on their substrates near pH 7

Few enzymes occupy a more important place in the history of biochemistry than theone found in 1834 by Johann Nepomuk Eberle (1798-1834) in extracts of gastricmucosa Two years later, Theodor Schwann (1810-1882) characterized this 'ferment',

named it pepsin, and established its physiological role in the mammalian digestion of

food proteins [1] During the succeeding 60 years, pepsin was considered to be the

prototype of the 'unorganized ferments' (KUhne named them enzymes in 1876) as

distinct from the 'organized ferments' responsible for such processes as the fermentation

of sugar by yeast [2] Many efforts were made to purify pepsin; the work of ErnstWilhelm von Brncke (1819-1892) and of Comelis Adrianus Pekelharing (1848-1922)

is especially noteworthy The high point came in 1930, when John Howard Northrop(b 1891) described the crystallization of pig pepsin [3] Although this achievementfollowed the crystallization of urease by James Sumner, it was Northrop's massiveevidence for the protein nature of pepsin that led to the rejection of the view, advocated

by Richard Willstatter during the 1920's, that enzymes are small catalytic moleculesadsorbed on inactive protein carriers [4].

Another important discovery made in this field before 1900 was the observation byJohn Newport Langley (1852-1925) that a slightly alkaline extract of gastric mucosacontains a material (pepsinogen) which is converted to pepsin on acidification of the

extract [5] The crystallization of pig pepsinogen in 1938 by Roger Moss Herriott(b 1908)made possible the incisive study of its conversion to pepsin [6] The work ofNorthrop and Herriott thus marks the beginning of the modem study of pepsin as aprotein and as a catalytic agent, and has influenced the investigation of the otherenzymes now considered to be aspartyl proteinases

(b) Occurrence and nomenclature

In all vertebrates the gastric juice contains one or more pepsins that arise from secretedpepsinogens; the latter are produced mainly in the chief cells (zymogen cells) of thefundus (corpus) [7,8] The secretion of the pepsinogens is under vagal control bothdirectly on the oxyntic glands of the fundus and indirectly through the release of peptidehormones (gastrins) from the pyloric glands [9]

Multiple forms of pepsinogen, and of the pepsins derived from them have been found

in many vertebrates (e.g., man, monkey, pig, beef, rat, chicken, dogfish) graphic separation of the components has shown that the predominant pepsin A(usually denoted pepsin) of adult mammals is accompanied by pepsin C (the currently-preferred name isgastricsin), as well as by the minor components denoted pepsin B and

Chromato-pepsin D [10] Some investigators have given the individual chromatographic ponents Roman numerals, while others have numbered the gastric proteinases and theirzymogens in the order of decreasing electrophoretic mobility at pH 5.0 or 8.5respectively [11] Immunochemical methods have also been applied to the differen-tiation and numbering of human gastric proteinases [12] In the gastric juice offetal andnewborn mammals the major pepsin-like enzyme ischymosin (derived from the zymogen

com-prochymosin); this was its original name [13], but for many years it was calledrennin

because it is the chief enzymic component of rennet, the calf-stomach (abomasus)extract used in the manufacture of cheese [14] Regrettably, the nomenclature of themammalian gastric proteinases has been in a confused state because of differences inthe terminology used by various groups of investigators; for a helpful clarification, seeFoltmann [15]

Among the aspartyl proteinases from vertebrates is the kidney enzyme renin (also

present in submaxillary tissue), whose important physiological function is the formationfrom plasma angiotensinogen of the decapeptide angiotensin I, which is in tum cleaved

by a 'converting enzyme' to the highly active pressor octapeptide angiotensin II [16,17]

Itshould be noted that although renin is now known to be an aspartyl proteinase, it isnot an acid proteinase, since the pH optimum for its action is 6-8 Another aspartylproteinase is the lysosomal enzymecathepsin D present in many animal tissues (spleen

[18], liver [19], uterus [20,21], thyroid [22], skeletal muscle [23], anterior pituitary [24],brain [25], seminal tissue [26], erythrocytes [27], lymphoid tissue [28]) A cathepsin D

and its inactive precursor from monkey lung appears to resemble gastricsin and itszymogen [29] Among the aspartyl proteinases in plants are those present in Lotus seed[30] and in the insectivorous plants Nepenthes and Drosera [31).

There has been considerable interest in microbial acid proteinases in part because of

a search for suitable rennet substrates The enzymes subjected to the most intensive

study have been penicillopepsin (from Penicillium janthinellum) [32], Rhizopus-pepsin (from Rhizopus chinensis) [33], and the acid proteinases from Endothia parasitica [34] and Mucor miehei[35].Well-characterized microbial acid proteinases have also been

isolated, and in some cases crystallized, from strains of Acrocylindricum sp [36], Aspergillus saitoi [37], Candida albicans [38], Cladosporium sp [39], Fusarium monili- forme [40], Monascus kaoling [41], Mucor pusillus [42], Paecilomyces varioti [43], Penicillium duponti [44], Rhodotorula glutinis [45], Russula decolorans [46], Tramestes sanguinea [47], Trichoderma viride [48], and both Saccharomyces cerevisiae and

S.carlsbergensis(yeast proteinase A) [49] No evidence is available for the existence of

zymogens for the above microbial proteinases The enzymes from Endothia parasitica, Mucor miehei and Mucor pusillus have been used in cheese manufacture.

These various aspartyl proteinases have in common the property of cleaving proteins(e.g., denatured hemoglobin, serum albumin) and suitable oligopeptides at pH 1.5-5.5

A widely-used diagnostic test is their inhibition by the naturally-occurring peptidepepstatin and by active-site-directed diazo compounds; these properties will bediscussed later in this chapter Although future study of other enzymes may show them

to belong to the aspartyl proteinase family, in what follows primary attention willbegiven to those known to exhibit these properties Among the acid proteinases that do

not appear to be inhibited by pepstatin or diazo compounds is the one from Scytalidium Iignicolumn[50]

Apart from the confusion in the naming of the mammalian gastric proteinases,mentioned previously, the nomenclature of the acid proteinases has not been aided bythe Commission on Enzymes of the International Union of Biochemistry [51].Proteolytic enzymes belonging to different classes have been given the same name anddistinguished from each other only by the addition of a different capital letter and theassignment of different numbers Moreover, the Commission has retained the long-outworn distinction between hydrolases and transferases for enzymes that act onpeptide, ester and glycosidic bonds presented in the first edition ofthe treatise of Dixonand Webb [52]

(c) Purification

In his work on crystalline pig pepsin, Northrop noted that the preparations differedconsiderably in homogeneity, as indicated by measurement of their solubility behavior.Part of the inhomogeneity was a consequence of the presence of peptide material,formed by autodigestion Subsequently, Steinhardt performed a careful study of thesolubility properties of crystalline pig pepsin, and gave clear evidence of its inhomo-geneity as a protein [53].After the introduction of ion-exchange chromatography forthe fractionation of proteins, Ryle and his associates established the presence of the

minor components mentioned previously [10] Additional factors that may contribute

to heterogeneity are the presence of multiple gene products (made evident by amino acidsubstitutions) and different degrees of phosphorylation [54] or glycosylation [55]

At present, the preferred method for the preparation of apparently homogeneous pigpepsin A is rapid activation of crystalline pepsinogen (shown to be homogeneous byseveral criteria [56]), and passage ofthe mixture first through sulfoethyl Sephadex C-25

to remove peptides and then through Sephadex G-25 to remove salts This pepsinpreparation is homogeneous on hydroxylapatite or DEAE-cellulose, which may also beused to effect the purification of commercial preparations of crystalline pig pepsin [57].Similar chromatographic procedures have been used for other aspartyl proteinases.Some aspartyl proteinases have been purified by means of affmity chromatography

on columns of Sepharose 4B or agarose to which a substrate analogue (e.g.,L-Phe-D-Phe) [58] or pepstatin [59] has been attached by means of aminohexoylbridges; hemoglobin-agarose columns have also been employed [60] The purification

of cathepsin0 from various animal tissues has been attended with difficulty, as isindicated by the heterogeneity of the enzyme preparations that have been described[61-64]

Crystallization has been usually effected by means of ammonium sulfate or acetone.Aside from pig pepsin A, the following aspartyl proteinases have been obtained incrystalline form: the pepsins from beef [65] and salmon [66], calf chymosin [67],penicillopepsin [32], Rhizopus-pepsin [33], and the proteinases from Endothia parasitica [34], Aspergillus saitoi[37], Mucor pusi//us [42], Paeci/omycetes varioti[43],

Penicillium duponti [44], Rhodotoru/a glutinis [45], and Trametes sanguinea [47] Thecrystallization of chicken liver cathepsin 0 has been reported [68]

Various synthetic peptides have been used in the assay of the aspartyl proteinases

Among these substrates are compounds of the type A-Phe(N02)-Y-B· (where Y=Phe,Leu etc.); the rate of cleavage of the Phe(N02)-Y bond may be followed spectrophoto-metrically at 310 nm [71] Older methods have involved the use of substrates of the typeA-Phe-Y-OH (where Y = Tyr(I2)' Phe etc.), and measurement of the rate of hydrolysis

of the Phe-Y bond by the ninhydrin method [72]; this procedure has been automated[73]

2 Molecular properties

(a) Physical-chemical properties

By means of the sedimentation-equilibrium method, values of 32700 ± 1200 and

40400 ±1600 were obtained for the molecular weight of pig pepsin A and pepsinogen Arespectively [74] These values may be compared to 34644 and 39637 calculated fromthe amino acid sequences Other methods (for example, sedimentation-velocity-diffusion, light scattering, osmotic pressure) gave values for pepsin ranging from 32000

to 35000 Estimates of the molecular weight of other gastric proteinases and ofmicrobialacid proteinases (in some cases determined by means of sodium dodecyl sulfate-agargel electrophoresis) have given values ranging from about 31000 to about 40000; forthe zymogens the values range between 36000 and 43000 [75]

The aspartyl proteinases are acidic proteins, as a consequence of the preponderance

of dicarboxylic acid residues as compared to the basic amino acid residues In the case

of pig pepsin, the paucity of lysine and arginine residues is especially marked Earlystudies by Tiselius and Herriott suggested that the isoe1ectric point of pig pepsin liesbelow 1, since the protein still migrated as an anion at this pH value This conclusion

is probably incorrect, as is suggested by more recent studies in which the isoelectricfocussing technique was employed [76-78] However, in view of the heterogeneity ofthe preparations, and the extended time required in this method, no defmite isoelectricpoint can be assigned, except to infer that pepsin A has a pI between 2 and 3 In thecase of other aspartyl proteinases where the balance between acidic and basic aminoacids is less extreme, the isoelectric points are between 3 and 5 In contrast to theextremely low isoelectric point of pig pepsin, that of pig pepsinogen is about 3.7; thisdifference is consistent with the cationic character of the peptide removed from thezymogen upon its conversion to pepsin (see Section 3(a))

• Abbreviations (in alphabetical order) used in this chapter and not defined in the Recommendations of the IUPAC-IUB Joint Commission on Nomenclature on the Nomenclature and Symbolism for Amino Acids and Peptides [Eur J Biochem (1984) 138,9-37]: DAN, diazoacetyl-nt-norleucine methyl ester; Dns, 5-dimethylamino-I-naphthalenesulfonyl; EPNP, 1,2-epoxy-3-(p-nitrophenoxy)propane; Mns, 6-(N-

methylanilino)-2-naphthalenesulfonyl; Nle, L-norleucyl; Nva, L-norvalyl; OP4P, 3-(4-pyridyl)propyl-I-oxy; Phe(N02 ) ,p-nitro-L-phenylalanyl; Pia, p-phenyl-L-Iactyl;Pol, L-phenylalaninol; TPDM, p-toluenesulfonyl- t-phenylalanyldiazomethane; TNS, 2-p-toluidinylnaphthalene-6-sulfonate; Tyr(I2 ) , 3,5-diiodo-L-tyrosyl The abbreviated designation of amino acid residues denotes the L form, except where otherwise indicated.

Highly-purified preparations of several aspartyl proteinases exhibit the presence ofmultiple components on isoelectric focussing Among them are beef spleen cathepsin D[61], and Rhizopus pepsin [79].

(b) Amino acid composition and sequence

Many investigators have studied the amino acid sequence of segments of pig pepsin A;the complete sequence proposed by Sepuvelda et al [80], has been widely accepted(Fig 1) The amino acid composition implied by this sequence (total, 327 residues)differs from that reported (total, 321 residues) by Rajagopalan, Stein and Moore [57].There are several notable features in the sequence: (1) the paucity of strongly basicside-chain cationic groups(l Lys, 2 Arg), all located within the 20-amino acid carboxyl-terminal section of the protein; (2) the overwhelming predominance of side-chaincarboxyl groups; (3) the presence of a relatively large number of residues ofhydroxyamino acids, of proline and of aromatic amino acids; (4) the presence of threedisulfide bridges in relatively small loops Earlier work had shown that pig pepsin Acontains one phosphoryl group per molecule [81], and that this group is absent in bothpig pepsin D and its zymogen [82]; in the sequence shown it is Ser-68 that isphosphorylated The sequence presented in Fig 1 is for the major component of theenzyme preparation analyzed by Sepuvelda et al [80] Some of the pepsin moleculeshad an additional H-Ala-Leu- unit at the amino terminus or a deletion of Ile-230 or thereplacement of Ser-255 by a glutamine residue

In the conversion of pig pepsinogen A to pepsin A, the 44-residue amino-terminalaction of the zymogen is removed The amino acid sequence of this fragment has beenreported [83] to be:

The amino acid composition of pepsinogen A and pepsin A from other mammalianspecies (man, monkey, beef, rat) resembles those from the pig, and extensive sequencehomology is evident [84,85] In all cases studied, the gastric proteinases have 6 half-cystine residues to form three disulfide bridges In chicken pepsin (and pepsinogen),however, in addition to these bridges there is a cysteinyl residue [55] It is alsonoteworthy that chicken pepsin has a larger proportion of strongly basic amino acids(8 Lys+ 4 Arg) and an isoionic point near pH 4; the small net negative charge of thisenzyme may account for its stability at pH values above 6, where pig pepsin is rapidlydenatured [86]

A complete amino acid sequence has been reported for calf chymosin, and itszymogen prochymosin [87] The latter resembles pig pepsinogen A in having 365 amino

Fig I Amino acid sequence of pig pepsin A (from ref 80).

acid residues, and 42 residues are removed from the amino terminus upon activation

to chymosin Like other mammalian gastric proteinases, chymosin has 3 disulfide bridges, but in contrast to pig pepsin A (and like chicken pepsin) chymosin has a relatively large proportion of strongly basic amino acids (9 Lys + 6 Arg).

In contrast to the aspartyl proteinases from the mammalian gastric mucosa, which are single-polypeptide-chain proteins, the enzyme renin (from mouse submaxillary glands) has two chains linked by one disulfide bridge [88] Similarly, cathepsin D (from

pig spleen) is composed of two chains (total, 339 residues) whose sequences have beenreported [89] On the other hand, the microbial aspartyl proteinases for which sequencedata are available all appear to be single-polypeptide chain proteins Like the pepsins,they are composed of about 330 amino acid residues, but differ from pig pepsin A inhaving many more (9-15) lysine residues per molecule, and their isoelectric points arenear 4-5 Also, most of them are lower in their content of half-eystine residues, andsome (penicillopepsin,Russula decoloransproteinase) appear to have none [46] For avaluable discussion of the problem of homology in the amino acid sequences of thegastric proteinases and the microbial enzymes, see Foltmann and Pedersen [84].

It should be added that several of the aspartyl proteinases have been shown to beglycoproteins Thus chicken pepsin contains 3 mannose and 7 glucosamine units perprotein molecule [55] and in pig spleen cathepsin D two asparagine residues areglycosylated [89] Moreover, among the microbial enzymes those from Penicillium duponti, Candida albicans andMonascus kaolianghave been shown to be glycoproteins

(c) Chemical modification

Early studies on the kinetics of the action of pig pepsin on synthetic peptide substratessuggested the presence in the enzyme of two catalytically important prototropic groupswithpK avalues near 1 and 4 [90] A reasonable interpretation of this finding was that

a carboxylate group and a carboxyl group are involved in the catalytic mechanism.Support for this view has come from studies on the esterification of pig pepsin by means

of diazo compounds Diazomethane [91] and diphenyldiazomethane [92] were shown

to inactivate pig pepsin near pH 5 but nearly complete loss of activity required theesterification of up to five groups per pepsin molecule Subsequent work, however, led

to finding of diazoketones of the type RCOCHN2or diazoacetamido compounds of thetype N2CHCONHR which can inactivate pig pepsin by the introduction of a singlesubstituent group per protein molecule

The first of the diazoketones shown to be a specific reagent for pepsin A wasL-I-diazo-4-phenyl-3-tosylamidobutanone-2 (Tos-Phe-CHN2 ) which rapidly inacti-vates the enzyme at pH 5.4 [93] (Fig 2) The rate of the reaction is greatly increased

Fig 2 Active-site directed inhibitors of aspartyl proteinases.

by the addition of cupric salts With 14C-labeled reagent (prepared from alanine), the rate of the loss of pig pepsin A activity toward protein and peptide substratewas shown to be the same as the rate ofincorporation ofthe tosyl-L-phenylalanyl group;complete inactivation was achieved upon the introduction of one such group permolecule of pepsin No incorporation was observed with pig pepsinogen A, whosepotential enzymic activity was unaffected by treatment with the diazo compound, norwas the label incorporated to a significant extent into alkali-denatured pepsin The 0

14C-L-phenyl-isomer of the reagent reacted with pepsin much more slowly than the L compound, andthe reactivity of the corresponding Tos-Gly-CHN2 was found to be intermediatebetween the L and 0 forms of Tos-Phe-CHN2 •The tosyl group may be replaced byothers such as benzyloxycarbonyl or by chromophoric groups such as 2,4-dinitrophenyl

or l-dimethylaminonaphthalene-5-sulfonyl (dansyl) to introduce them into pepsin and

to serve in fluorescence studies as resonance energy acceptors from tryptophan residues

in the protein [94]

The first of the diazoacetamido compounds to be described as a specific inhibitor ofpig pepsin was diazoacetyl-m.-norleucine methyl ester, chosen to permit the determi-nation of the norleucine content ofthe modified protein in the amino acid analyzer [95].The rate of the reaction at pH 5 is accelerated by cupric ion; in its absence theinactivation is slower and the incorporation is not stoichiometric In contrast to thesite-specific diazoketones, the diazoacetyl compounds do not exhibit stereospecificitysince the L and 0 forms ofN2CHCO-Phe-OEtreact at the same rate [93] The coppercatalysis of the reactions of the diazoacetamido compounds involves the intermediateformation of a metal-complexed carbene [96]; it is not surprising therefore that theesterification of carboxyl groups of a protein is accompanied by the oxidation oftryptophan, methionine, tyrosine and cysteine [97]

The above diazo compounds have played a large role in the identification of Asp-2I5

of pig pepsin A as a catalytically-active group of the enzyme [98]; thiswillbe discussedlater in this chapter Despite the less exacting specificity of the diazoacetylamidocompounds, as compared to that ofthe diazoketones, diazo-nt-norleucine methyl ester(DAN) has become a standard test material for the identification of aspartylproteinases Itshould be noted that whereas with pig pepsin complete inactivation isachieved upon the incorporation of one molecule of DAN per protein molecule, thecomplete inactivation of penicillopepsin and of Rhizopus-pepsin requires the intro-duction of 1.3-2 molecules of reagent [32,79] In addition to diazomethane anddiphenyldiazomethane, mentioned above, other diazo compounds have been shown toinactivate pepsin; it is not clear in all cases whether the reaction was stoichiometric andspecific for Asp-2I5 [99,100] Partial inactivation was also effected by means ofp-bromophenacyl bromide, and evidence was presented for the view that an aspartylresidue had been esterified [10 I ]

Another esterification method applied to the study of the carboxyl groups of pepsinhas been the use of suitable epoxides In particular, I,2-epoxy-3-(4-nitrophenoxy)pro-pane (EPNP) was reported to inactivate pig pepsin A at pH 4.6 with the apparentintroduction of two molecules of substituent per molecule ofprotein [ 102] With chickenpepsin, the incorporation of 3-4 molecules of the inhibitor was reported [103] In the

case of pig pepsin A, one of the sites of reaction was shown to be an aspartyl residuelater identified as Asp-32, and the other appears to have been the one modified byspecific diazo compounds [104,105] Together with the results obtained with the latterreagents, the work on the epoxy compounds was in agreement with the concept thattwo protein carboxyl groups are important in the catalytic action of pepsin; studies oncalf chymosin gave similar results [106] This view received further support fromexperiments on the modification of pepsin by means of 14C-labeled trimethyloxoniumfluoroborate [107] Among the other reagents that inactivate pig pepsin A with theapparent esterification of carboxyl groups is bis(p-chloroethyl)sulfide (mustard gas)[ 108].

In addition to the chemical modification of the carboxyl groups ofpepsin, much workhas been done on the role of the tyrosyl residues Indeed, the first studies on the chemicalmodification of crystalline pig pepsin were on the effect of acetylation by means ofketene, and showed that increased acetylation of the phenolic hydroxyl groups wasaccompanied by progressively greater inhibition of proteinase activity [109] Subse-quent work, using the more selective acetylating agent acetylimidazole at pH 5.5, led tothe significant finding that whereas the proteinase activity toward hemoglobin isinhibited by the acetylation of tyrosyl residues in pepsin, the rate of cleavage of smallsynthetic substrates is enhanced; these effects are reversed upon deacetylation of themodified protein by treatment with hydroxylamine [73,110] Similar observations havebeen made in studies on the carbamylation of pig pepsin with potassium cyanate [Ill],and upon nitration with tetranitromethane [112]

Iodination of pig pepsin led to the formation of 3-iodotyrosyl and 3,5-diiodotyrosylresidues and partial inactivation ofthe enzyme [113] However, in contrast to the effect

of acetylation with acetylimidazole, the iodination of pepsin leads to the parallel loss

of proteinase, peptidase and esterase activity [73] These findings led to the view thatone or more tyrosine residues form part of the extended binding site of pig pepsin Morerecent studies have identified as sites of iodination Tyr-9 and Tyr-174 [114] On theother hand, it has also been reported that modification of Tyr-189 by means ofp-nitrophenyldiazonium hydrochloride abolishes the peptidase and reduces theproteinase activity of pig pepsin, and it was suggested that this residue may also beinvolved in substrate binding [115]

As regards the chemical modification of tryptophyl residues of pig pepsin, treatmentwith 2-hydroxy-5-nitrobenzyl bromide at pH 3.5 leads to the incorporation of twosubstituent molecules per molecule of pepsin with the loss of only about 25%of theproteinase and peptidase activity [116] More extensive inactivation was observed withN-bromosuccinimide [117]

Phenylglyoxalhas been found to inhibit partially the proteinase and peptidase activity

of pig pepsin [118] This reagent has been used to modify specifically arginyl residues

in proteins Another is butane-2,3-dione, which has been reported to inhibit pig pepsinwith the modification of Arg-316 [119] Further studies with this reagent have shownthat it inactivates several aspartyl proteinases at pH 6 in the presence of light, but not

in the dark, and that the inactivation is a consequence of a photosensitized modification

of tryptophan and tyrosine residues [120] Partial inactivation of pig pepsin has also

been observed after treatment with tx-bromo-4-amino-3-nitroacetophenone at pH

values below 3; this leads to the alkylation of Met-290 [121]

Itshould be recalled that chicken pepsin contains a single sulfhydryl group per proteinmolecule Alkylation of this group or its conversion to a mixed disulfide decreased theproteinase activity of the enzyme only slightly, and markedly enhanced the rate of itsaction on a synthetic peptide substrate [122] Itshould also be noted that acetylation

of the amino groups of pig pepsin, or their deamination with nitrous acid, does notappear to affect the proteinase activity of the enzyme [109,123] The absence ofhistidine, arginine or methionine from some of the microbial aspartyl proteinasessuggests that these amino acids do not playa significant role in the general mechanism

of action of this class of enzymes

(~ Prorem-ligandmrerocMn

If chemical modification by means of active-site-directed reagents can shed light on the

nature of the catalytically-active groups in an enzyme, the study of the non-covalentinteraction of an enzyme with small ligand molecules, especially strong inhibitors, canilluminate features of the active site that are important in the binding of substrates toform productive complexes In the case of the aspartyl proteinases one such inhibitor,

discovered by Umezawa and his associates, and which theynamedpepstatin,has played

a leading role in the study of these enzymes [124] In a series of brilliant investigations,

they isolated this substance from culture filtrates of a strain ofStreptomyces,establishedits structure, and demonstrated its ability to inhibit the action of pepsin, gastric sin, renin,cathepsin D and several microbial proteinases on protein and peptide substrates Thebinding of pepstatin to pig pepsin is stoichiometric and the dissociation constant of theenzyme-inhibitor complex, when measured with a peptide substrate of the typeA-Phe(N02)-Phe-B,is approximately 10- 10M Upon chemical modification of pepsinwith either diazo or epoxide active-site-directed reagents, the apparent dissociationconstant increases to about 10-6M, and with either alkali-denatured pepsin or nativepepsinogen the value is about 10-5M [125]

As shown in Fig 3, pepstatin is an isovaleryl derivative of a pentapeptide composed

of 2 residues of L-valine, 1 of L-alanine, and 2 of a new hydroxyamino acid 3S-hydroxy-6-methylheptanoic acid), for which the name statinehas been proposed.Although esterification of the carboxyl group of pepstatin does not affect its inhibitoryability,this is markedly reduced by acetylation of the two hydroxyl groups Replacement

(4S-amino-of the N-terminal isovaleryl group by an acetyl group, as in the material (SPI) isolatedfrom aStreptomyces[126], does not alter significantly the inhibitory property Reduction

of the chain length of pepstatin markedly increases the apparent dissociation constant;for example, the value for Ac-Val-statine is about 10-6M [127] A ketone analogue ofpepstatin, in which the 3S-hydroxyl group had been oxidized (isovaleryl-Val-statone-Ala-isoamylamide) is a strong inhibitor, with an apparent dissociation constant of about10-9M [128] The above estimates of the dissociation constants of complexes of pigpepsin with various pepstatin derivatives were estimated from the apparentK,values

in kinetic measurements with synthetic oligopeptides as substrates More directmeasurements have been made with radioactive pepstatin (labeled with 14C or 3H) ormodified by the addition of 125I-Iabeled iodotyrosine methyl ester to the carboxyl-terminus; the value of the apparent dissociation constant(K o ) of the complex of pig

pepsin with the last-named derivative gave a value of 10-IIM, near that for the complexwith unmodified labeled pepstatin [129] Upon the binding of pepstatin, pig pepsinundergoes conformational changes detectable by[IH]NMR spectroscopy [130].Although other known aspartyl proteinases are also inhibited by pepstatin, at aconcentration of inhibitor sufficient to abolish completely the activity of pig pepsin orhuman pepsin, human gastric sin is inhibited to 50%, while calfchymosin and Rhizopuspepsin are only inhibited slightly (ca 10%) [131] These differences indicate thatalthough the various aspartyl proteinases may be alike in their requirement for twoaspartyl carboxyl groups in the bond-breaking phase of enzymic catalysis, theseenzymes appear to differ greatly in the binding of peptide ligands (including peptidesubstrates) at their active sites

An outstanding structural feature of the pepstatin molecule is the preponderance ofhydrophobic groups, and its tight binding at the active site of pig pepsin supports theview that in this enzyme the extended active site is strongly hydrophobic in nature Thishad become evident from studies on the interaction of pig pepsin with peptides bearingfluorescent probe groups Earlier studies by means of a refined gel-filtration technique[132] had shown that for peptides of the type A-Phe-Phe-B the principal bindingenergy is provided by the interaction of the Phe-Phe unit with the active site of pigpepsin, and that this binding is largely hydrophobic in character [133] For fluorescencestudies, peptides of this type bearing a 6-(N-methylanilino)-2-naphthalenesulfonyl group(mansyl, Mns) in the A or B group were used This probe group [134] offers markedadvantages over the more widely employed I-dimethylaminonaphthalene-5-sulfonyl(dansyl, Dns) or 2-p-toluidinylnaphthalene-6-sulfonyl (TNS) group [135,136], seeFig 4 These groups have been used as probes for hydrophobic interaction, becausetheir fluorescence is enhanced and their emission maximum is shifted to a shorter

wavelength when they are transferred from an aqueous environment to a solvent of lowpolarity or when they are bound to proteins The mansyl group is much more sensitive

to changes in the polarity of the solvent than the dansyl group, and its introduction intopeptides is attended by fewer difficulties than in the case of TNS

With substances that are completely resistant to pepsin action (e.g., Mns-NH2 ,Mns-Gly-Gly-Y) or with substrates that are cleaved very slowly (Mns-Phe-Phe-Y), it

is possible to perform steady-state fluorescence measurements to determine the fraction

of the ligand that is bound to pig pepsin when successively larger amounts of enzymeare added to a constant amount of ligand If it is assumed that binding involves a singlesite that interacts more strongly with the ligand than do other sites, a Scatchard plotgives an estimate of the value ofK o In the case of Mns-Phe-Phe-OP4P,K o was found

to be 0.07 mM at pH 2.35 and 25°C [137]; this may be compared to the determined value ofK m = 0.095±0.015 mM under the same conditions [138] Theavailable evidence indicates that the fluorescent group of Mns-Phe-Phe-OP4P is drawninto the active site of pig pepsin by virtue of the interaction of the Phe-Phe unit withcomplementary active site groups and that, in addition, pepsin has a weaker bindinglocus (or loci) for the mansyl group, distinct from the extended active site of the enzyme.This conclusion is based on the effect of the addition of pepstatin The active site ofpepsin has relatively little intrinsic affmity for the mansyl group, as judged by the factthat the increase in fluorescence with Mns-Gly-Gly-OP4P is small and is not altered

kinetically-by the addition of pepstatin, but with Mns-Phe-Phe-OP4P the large increase influorescence is reduced by pepstatin to the value observed with Mns-Gly-Gly-OP4P orMns-NH2 [139]

Itis noteworthy that the fluorescence of both Mns-NH2and Mns-Phe-Phe-OP4P isgreatly enhanced upon interaction with pig pepsin that had been inactivated bystoichiometric reaction with tosyl-L-phenylalanyl diazomethane This fluorescenceenhancement is not diminished by pepstatin, indicating that the introduction of theTos-Phe-CH2 group had not only blocked the active site but had also altered theconformation of the protein in such a manner as to increase the ability of the secondarybinding locus to accept the mansyl group [137] Analogous binding of 2-p-toluidinyl-naphthalene-6-sulfonate to a secondary binding site has been reported for chymosin;this binding is enhanced in the presence of the pepstatin-analogue SPI [140].Other studies on the interaction of pig pepsin with peptide inhibitors bearing afluorescent probe group have involved dansyl derivatives of fragments of the amino-terminal sequence of pig pepsinogen [141] As in the earlier studies with dansyl andmansyl peptides [137,142] an enhancement of fluorescence and energy transfer fromprotein tryptophan were observed In addition to the naphthalene sulfonyl derivatives,acridine compounds have also been tested as fluorescent probes Whereas thefluorescence of 9-acetylaminoacridine at pH 5 is not changed significantly in thepresence of pig pepsin, that of 9-(H-Phe-Phe-amino)acridine is markedly altered; thiseffect is abolished by the addition of pepstatin [143] The interaction of the acridine dyeacriflavine with pig pepsin has also been studied [144]

(e) Denaturation

It has long been known that at temperatures near 25°C pig pepsin is irreversiblyinactivated at pH values above 6.0, and this alkali lability has been attributed to thescission of hydrogen bonds [145] In the pH range 2-6, pig pepsin appears to bestabilized largely by hydrophobic interactions, since its enzymic activity and opticalrotation are unaffected by heating to 60° or by treatment with 4 M urea or 3 Mguanidinium chloride [146] However, within the stability range a conformationaltransition has been detected by means of spectroscopic measurements; this transitionhas been attributed to the presence in the protein of two globular units held together

by relatively flexible segments of the polypeptide chain [147]

In contrast to pig pepsin, pig pepsinogen can undergo reversible denaturation afterbeing heated to about 60°C at pH 7 or after treatment withalkali (up to pH 11) Thisprocess has been considered to be biphasic, a deprotonation step preceding one asso-ciated with marked changes in optical rotation or viscosity [148] Ithas also beenconcluded from such studies that the amino-terminal portion of pig pepsinogen, wherethe strongly basic amino acids are clustered, participates in the stabilization of thezymogen through electrostatic interactions with carboxylate groups of the protein [149].Although subsequent fluorescence studies [150] have given support to the idea of atwo-step process in the reversible denaturation of pig pepsinogen, calorimetric measure-ments [151] have suggested that the process is more complex in nature

Chicken pepsin exhibits greater stability than does pig pepsin at pH values up toabout 8, and this may be correlated with the smaller net negative charge of the chickenenzyme [55] Other aspartyl proteinases, such as pig kidney renin [152] or variousmicrobial enzymes [35,51] are also more stable; as noted previously in this chapter, likechicken pepsin, they have a higher lysine content than does pig pepsin

(f) Three-dimensional structure

Although valuable inferences have been drawn from studies involving chemical cation and optical (e.g circular dichroism) methods regarding the active sites andconformations of the aspartyl proteinases, the most incisive information has come after

modifi-1975 from the description of the crystal structures of several of these enzymes X-rayphotographs of crystals of pig pepsin were obtained in 1934 by Bernal and Crowfoot[153], but the new techniques of X-ray crystallography have been applied only recently.The three-dimensional structure of the followingaspartyl proteinases has been presented

at better than 2.8Aresolution: pig pepsin [154], penicillopepsin [155], ticaproteinase [156-159] and Rhizopus pepsin [160] Progress has also been made inthe determination of the crystal structure of chymosin [161,162] and renin [163].Striking similarities in the general structure of the aspartyl proteinases have emergedfrom these studies They all appear to be composed of two globular domains that arepredominantly in the form of P-structures, with little helical content, and the active site

Endothiaparasi-is located in an extended cleft (25-30A)at the junction of the two lobes Each domaincontributes one aspartyl residue, corresponding to Asp-32 and Asp-215 of pig pepsin,

to the catalytic center of the extended binding site Sequence studies had shownconsiderable homology for various aspartyl proteinases in amino acid sequence of theamino acid residues attached to those corresponding to Asp-32 (Asp-Thr-Gly-Ser) and

to Asp-215 (Asp-Thr-Gly-Thr) Moreover, the X-ray analyses have given evidence of

an intramolecular twofold symmetry axis that relates two topologically similar domainsand the active site residues.Ithas been suggested that present-day aspartyl proteinaseshave evolved by gene duplication of an ancestral protein of about 150 residues having

a fold similar to that of one lobe of pig pepsin [164] This view has received supportfrom the determination of the nucleotide sequence of the human renin gene [164a]

In addition, crystal structures have been described for complexes of aspartylproteinases with inhibitors considered to be substrate analogues Thus, upon thebinding of isovaleryl-Val-Val-statone-OEt to penicillopepsin, a significant confor-mational change was observed in the active site region [165], although no similar changewas found in model studies for the binding of pepstatin to Rhizopus pepsin [158] Inthe case where a substrate analogue (H-Phe-Tyr(12)-OH)was used, X-ray data werepresented in favor of the view that the Tyr(12 )residue is bound in a strongly hydrophobicregion of the active site of pig pepsin [166], in accordance with earlier evidence on theprimary specificity of pepsin, to be discussed later in this chapter Also, in the fmalsection, consideration will be given to contributions of crystal-structure determinations

to the problem of the mechanism of the catalytic action of the aspartyl proteinases

3 Action on protein substrates

(a) Activation of zymogens

Of special importance is the activation of the zymogens, which represent the form inwhich the extracellular proteinases of vertebrates are secreted In the case of pigpepsinogen, the activation occurs by removal of the 44-amino acid amino-terminal unit,with the scission of the Leu-44-Ile-45 bond, and may proceed by one of twomechanisms, depending on the pH In his pioneer work, Herriott [167] demonstratedthat near pH 4.5 the reaction is autocatalytic, and pig pepsin catalyzes its ownproduction At more acid pH (ca 2) however, the activation is unimolecular and therate is independent of pepsinogen concentration [168-171] In this process, the firstpeptide bond to be cleaved is Leu-16-Ile-17 to form an active 'pseudopepsin' and thecleavage of other bonds (including Leu-44-Leu-45) follows It appears likelythat uponacidification a portion of the amino-terminal portion of the zymogen is loosened fromthe active site of pig pepsin, as a consequence of the protonation of carboxyl groups,and the resulting conformational change elicits enzymic activity [172] Limited pro-teolysis then follows, the site of cleavage varying with such factors as the amino acidsequence of the zymogen and the specificityof the individual enzyme It should be noted,however, that the activation of calf prochymosin at pH 2 occurs predominantly as asecond-order intermolecular process with the sole cleavage of Phe-27-Leu-28;ifthe

activation is conducted near pH 5, the reaction is still autocatalytic but results in achymosin with an N-tenninal glycine [173].

In the case of beef pepsinogen, the initial site of cleavage is Leu-17- He-18, and thecomplete conversion to beef pepsin involves removal of the 45 amino-terminal residues

of the zymogen at Leu-45-Val-46 [174], whereas with chicken pepsinogen the processappears to resemble that for calf prochymosin with initial scission of a Phe-26-Leu-27bond [175] As will be evident from the subsequent discussion of the specificity of theaspartyl proteinases, these fmdings are in general accord with the known preference ofthe gastric proteinases for bonds linking hydrophobic amino acid residues Itwillbe alsoevident, however, that pig pepsin and calf chymosin differ greatly in their secondaryspecificity,thus accounting in part for the site of cleavage of prochymosin at pH valuesnear 5 It is also noteworthy that with rat pepsinogen, the activation involves removal

of the 46 amino-terminal residues through the cleavage of Tyr-46-Ser-47 [85] Theappearance of an amino-terminal serine in rat pepsin suggests that this enzymeresembles beef gastricsin, for which this amino acid has been identified as the principalN-terminus [176] Evidence has been presented for the existence of proenzymes of thecathepsin 0 of some animal tissues [29] and of renin [17]

Pig pepsin A is inhibited by the peptides formed from the amino terminus of itspepsinogen, and recent work has shown that the amino-terminal lysine-rich 16-residueunit is especially inhibitory [177] This is not a general phenomenon for all gastricproteinases, however; calf chymosin does not appear to be inhibited by the activationpeptides [173]

As noted above, the conversion of pig pepsinogen A to pepsin is accompanied by achange in conformation of the protein molecule This is evident from fluorescencestudies on the binding of fluorescent probe groups [137] Thus, upon activation ofpepsinogen at pH 2.35 in the presence of Mns-NH2the mansyl fluorescence decreases,indicating that the secondary binding site becomes less accessible to the probe; this site(separate from the active site) responds in a similar manner to 6-p-toluidinyl-2-naphthalene sulfonate upon pepsinogen activation [178] On the other hand, the mansylfluorescence of Mns-Phe-Phe-OP4P greatly increases during pepsin formation, as aconsequence of the ability of the Phe-Phe unit to draw the mansyl group into thenewly-formed active site Both changes in mansyl fluorescence are abolished by theaddition of pepstatin [137]

Another specific limited proteolysis leading to the formation of a proteinase is theaction of many microbial aspartyl proteinases on beef trypsinogen In the autocatalyticconversion of trypsinogen to trypsin, the essential step is the cleavage of the Lys-6-He-7bond in the zymogen, with the release ofH-Val-(AspkLys-OH [179] This cleavage

is effected by many microbial enzymes, e.g., penicillopepsin, Rhizopus pepsin, and theacid proteinases from Mucor meihei, Rhodotorula glutinis, Aspergillus saitoi, amongothers; in some cases the hexapeptide is hydrolyzed to give smaller fragments [51].These enzymes differ therefore from pig pepsin in their preference for a lysyl residue,rather than a hydrophobic amino acid residue, as the donor of the carbonyl to thesensitive bond in their substrates

(b) Cleavage ofprotein substrates

As was indicated previously, the considerable interest in the mode of action of chymosinderives from its importance in the manufacture of cheese The principal casein fractionaffected by chymosin in the primary phase of the milk-clotting process is the glyco-protein fraction (ca 15%) designated kappa-casein The selective enzymic cleavage at

pH 7 of the Phe-105-Met-106 bond leads to the removal of a soluble glycopeptide thusdestroying the ability of the kappa-casein to stabilize the casein micelle and, in thepresence of calcium ions, the micelles coagulate [180] Apart from this selectiveproteolysis, chymosin can cleave peptide bonds (e.g., Phe-Phe) in other proteins,including casein fractions other than kappa-casein [181] As will be evident from thelater discussion of the specificity of chymosin, secondary interactions at a distance fromthe site of cleavage playa significant role As for the microbial aspartyl proteinases thatcan serve as rennet substitutes, although the milk-clotting mechanism is the same as thatfor chymosin, the site of the initial cleavage of kappa-casein does not appear to havebeen identified Still another example of selective limited cleavage of a protein by anaspartyl proteinase is provided by renin, in its attack at pH 7 on the Leu-lO-Leu-llbond of the plasma renin substrate (angiotensinogen) It has also been reported thatcathepsin D from pig anterior pituitary cleaves the Leu-77- Phe-78 bond of P-lipotropinand the same bond in p-endorphin to form y-endorphin [182], and that chicken livercathepsin D selectively removes the carboxyl-terminal propeptide from procollagen[183]

Turning to the less selective cleavage of protein substrates by aspartyl proteinases,brief mention may be made first of the extensively-studied autolysis of pig pepsin Acareful study has shown that the cleavage of the protein is more rapid than the loss ofproteinase activity, but claims [184-186] for the isolation of enzymically-active low-molecular-weight dialyzable fragments could not be confirmed [187]

In its action on some general protein substrates (e.g., hemoglobin and serumalbumin), pig pepsin acts optimally near pH 2, but after these proteins are denaturedthe pH optimum shifts to about 3.5 [188] Also, for those proteins that can undergoreversible denaturation under a given set of conditions, the more rapid cleavage of thedenatured form will pull the equilibrium toward denaturation [189] For example, in thecase of serum albumin, which undergoes an expansion of the molecule at pH 4 (theso-called N-F transition), pig pepsin preferentially attacks the expanded protein atflexiblesegments that link smaller globular portions ofthe protein [ 190,191] In addition

to the factor of accessibility, if an accessible peptide segment of a protein substrate fitsinto a complementary active-site region, the favorable binding may result in more rapidhydrolysis of a sensitive peptide bond than when that bond is in a small polypeptidewith the same sequence Conversely, even if a potentially sensitive bond is on the surface

of a native protein, but is located in a relatively rigid peptide segment that does not fitinto the extended active site of the proteinase, the hydrolysis of that bond may not occuruntil the protein has been denatured For these reasons, inferences about the relativespecificity of enzymes such as pig pepsin from data on the nature of the peptide bondscleaved in intact proteins are fraught with uncertainty [192]

The use of pig pepsin as a reagent in the partial cleavage of proteins showed that manykinds of peptide bonds had been broken and led to the conclusion that pepsin is aproteinase of broad side-chain specificity [193] Although, in general, the sensitivebonds were present in dipeptidyl units containing at least one hydrophobic amino acidresidue such as Phe, Tyr, Leu or Met [194], a number of exceptions were noted.Examples are the scission of a Ser-Thrbond in the a-chain of human hemoglobin A and

of Gly-Lys, Asp-Pro and Ala-Gly bonds in the f1-chain of human hemoglobin A [195]

In considering fmdings of this kind, it should be recalled that in most sequence studiesthe primary objective has not been the delineation of the specificity of the enzyme inquestion, that relatively large amounts of enzyme (often of uncertain homogeneity) andprolonged incubation periods were employed, and that the ionic strength of the solutionmay affect the rate of cleavage of individual peptide bonds

The long-chain peptide most widely used for the examination of the specificity ofproteinases has been the 30-amino-acid oxidized p-chain of beef insulin Early workshowed that this substrate is attacked by pig pepsin A primarily at Leu-ll-Val-12;Glu-13-Ala-14, Ala-14-Leu-15, Tyr-16-Leu-17, Phe-24-Phe-25 and Phe-25-Tyr-26[196,197] Clearly, where two adjacent peptide bonds are both potential sites of attack,the more rapid cleavage of one of themwill render the other more resistant to peptidecleavage because of the appearance of an a-carboxyl or an a-amino group in immediateadjacence to the less susceptible bond Other aspartyl proteinases whose action on theoxidized p-chain of insulin has been examined include beef pepsin A [197], piggastricsin [198], calf chymosin [199], beef spleen cathepsin D [18], penicillopepsin[200], Rhizopus pepsin [201], and the proteinases fromMucor meihei[202,203] and

Endothia parasitica[204] Although there are many similarities, notably in the cleavage

of Leu-ll-Val-12, Phe-24-Phe-25, and Phe-25-Tyr-26, these bonds are cleaved atdifferent rates by the various enzymes, and several enzymes appear to act effectivelyatbonds relatively resistant to pig pepsin A Among such bonds are His-IO-Leu-ll(gastricsin, Rhizopus pepsin), Leu-17-Val-18 (chymosin), Gly-20-Glu-21 (penicil-lopepsin,Endothia parasitica) and Asn-3-Gln-4(Endothia parasitica).

These fmdings are in general agreement with the results of systematic studies on thecleavage of synthetic peptide substrates, to be discussed in the next section of thischapter All the aspartyl proteinases that have been examined carefully appear to sharewith pig pepsin A a preference for a hydrophobic amino acid residue as the donor ofthe imino group of the sensitive bond, but there are notable exceptions, some of whichmay be a consequence of the favorable effect of secondary peptide-enzyme interactions

so as to promote the cleavage of a bond resistant to pig pepsin A An attempthas been made to assess the specificity of pig pepsin A by taking published reports of

1020 cases of cleavage in proteins and naturally occurring peptides and subjecting thedata to probability analysis [205] The limitations of this approach are obvious fromthe fact that, in the calculations, each amino acid residue was considered to be indepen-dent of the others in the substrate, and was assumed to fit into a defmed 'sub-site' Theconcept, introduced into the field of proteolytic enzymes by Schechter and Berger [206],has been widely adopted because of its heuristic appeal, but is an oversimplification ofthe process whereby an oligopeptide substrate and the active site of a proteinase interactcooperatively to produce a productive enzyme-substrate complex [207]

4 Action on synthetic substrates *

(a) Primary specificity

The discovery in 1938that Z-Glu-Tyr-OH is hydrolyzed by pig pepsin A at the Glu-Tyrbond [208] provided strong evidence for the peptide theory of protein structure at a timewhen the theory was under serious challenge [209] However, this substrate and otherslike it (e.g., Z-Met-Tyr-OH [210)) are cleaved very slowly and are sparingly soluble at

pH values below 5 The demonstration that Ac-Phe-Phe-OH or Ac-Phe-Tyr-OH is abetter substrate indicated a preference for peptide bonds linking two aromatic L-aminoacids [211] The best of the substrates of this type is Ac-Phe-Tyr(I2)-OH(II in Fig 5)which is cleaved by pig pepsin at pH 2 and 37°C with kc a t = 0.2 sec - I and

Km =0.08 mM [53] The pH optimum for the hydrolysis of the acyl dipeptides is near2; substitution of the carboxyl group (as in Ac-Phe-Tyr-OMe) leads to a shift in pHoptimum to the region 3-4.5 [212] The latter type of substrate is insufficiently soluble

in aqueous butTers to permit reliable kinetic studies, and organic solvents must be added,but such solvents inhibit the action of pepsin on small oligopeptide substrates [213].For this reason, a second group of synthetic peptide substrates bearing a cationicgroup have been used; their protonation in the pH region 1-5 increases the solubility

of the compounds in water One type has the general structure Z-His-X-Y-OMe (orOEt) where X and Y are L-amino acid residues forming the bond cleaved by the enzyme[214] Systematic variation of the nature of X and Y showed that, of the substratestested, the most sensitive ones were those in which X=Phe and Y=Trp, Tyr or Phe[215]; for Z-His-Phe-Phe-OMe (III in Fig 5), kc a t =0.17 sec-I and Km =0.33 mM

at pH 4 and 37°C The results of this study strengthened the conclusion that thepreferred substrates of pig pepsin are those in which the sensitive peptide bond isflanked by two aromatic L-amino acid residues It should be noted, however, thatreplacement of either L-phenylalanyl residue in Z-His-Phe-Phe-OMe by a L-phenylgly-cyl residue or a D-phenylalanyl residue rendered the bond resistant to pepsin action[216] The favorable etTect of an aromatic and planar substituent at the p-earbon of the

X and Y residues was emphasized by the fmding that when X =p-eyclohexyl-L-alanylthe value of kc a tis much lower than that found for the corresponding substrate in which

X or Y =Phe, and is similar to that for substrates in which the X- or V-position isoccupied by an aliphatic amino acid residue larger than Ala (Nva, Nle, Leu, Met) [215]

Itwas also shown that the replacement of Phe in the X-position by Val or He renderedthe X-V bond more resistant to the action of pig pepsin than when X= Gly, indicating

• The kinetic parameters mentioned in this chapter are defined by the equation

v=kc a t[E],[S]o/(K m+ [S]o) for the process:

E + S ~ ES ; ES'( + PI) 4 E + P2

k_t

where v= initial velocity, [E], = total enzyme concentrations, [S]o = initial substrate concentration,

k =k2k3/(k2+k3 ) andK = [(L, +k2)/k,] [k3/(k2+k3 ) ] Other symbols used areK. =K o = L tikI'

Fig 5 Symnetic substrates tor aspartyl proteinases.

Bis-p-nitrophenyl suiti te (VI)

that when the X-position is occupied by a residue that is branched at the p-carbon, one

of the catalytic groups of pepsin may be prevented from attacking the carbonyl group

of the sensitive bond The available data on the primary specificity of pig pepsin A inits action on small oligopeptide substrates may be summarized therefore in terms of anapparent requirement for a hydrophobic (preferably aromatic) L-amino acid as thedonor of the NH group to the sensitive bond, and a strong preference for another suchamino acid (but not Val or lIe) as the donor of the CO group

The fmding that the replacement of the L-phenylalanyl residue in the X-position ofZ-His-Phe-Phe-OMe by a p-nitro-L-phenylalanyl residue did not alter the kineticparameters significantly permitted the development of a spectrophotometric method forfollowing the hydrolysis of the Phe(N02)-Phebond [71] Itshould be noted that thismethod measures the rate of formation of the acidic product (Z-His-Phe(N02)-OH),

in contrast to colonmetric (ninhydrin) aud fluorimetric (fluorescamine) methods thatmeasure the release of the amine product (H-Phe-OMe)

In addition to the cationic substrates bearing an imidazolium group, an extensiveseries of pyridylpropyl esters of suitable peptide has been examined; an example isZ-Phe-Phe-OP4P (IV in Fig 5) [217] In place of the pyridinium group, themorpholinium or N-methyl pyridinium group has also been used [218,219]

The development of the cationic substrates of the type Z-His-Phe(N02)-Phe-OMepermitted the unequivocal demonstration of the esterase activity of pig pepsin A [220].Replacement of the phenylalanyl residue by a p-phenyl-L-Iactyl (PIa) residue gave adepsipeptide (V in Fig 5) that is cleaved at the Phe(N02)-Plabond more rapidly than

is the comparable peptide substrate.Itshould be noted that the esterase activity of pigpepsin differs in principle from the action of serine proteinases like chymotrypsin onsubstrates of the type Ac-Tyr-OMe or cysteine proteinases such as papain onZ-Leu-OEt For these enzymes, the primary specificity resides solely in the nature ofthe amino acid residue that donates the carbonyl group to the sensitive bond In the case

of pig pepsin, however, the PIa unit is needed to meet the primary specificity requirementfor a p-substituted L-hydroxy acid, and it cannot be replaced by its o-enantiomer or by

a methoxy or ethoxy group It is noteworthy that the amide bond of Ac-Pla-Phe-OH

is resistant to pepsin action under conditions where Ac-Phe-Phe-OH is readilyhydrolyzed [221] Also, Z-His-Phe-Pol is not cleaved by pig pepsin at the Phe-Pol bond[222] Both Ac-Pla-Phe-OH and Z-His-Phe-Pol are competitive inhibitors of pepsin,withK, values near the K m values for the corresponding substrates.Itwould appear,therefore, that in peptide or depsipeptide substrates of pig pepsin an imino group isrequired on the carbonyl side of the sensitive bond, and a carbonyl group is required

on the amino side of the bond

Pig pepsin also catalyzes the hydrolysis of several organic sulfites such as fite at pH 2 [223] The same enzymic site is involved as in the cleavage ofpeptides, sincethe reaction is inhibited by compounds such as Z-Phe-Tyr-OH and abolished bytreatment of the enzyme with diazoacetyl-nt-norleucine methyl ester Bis-p-nitrophenylsulfite (VI in Fig 5) is cleaved very rapidly, with kca t = 143 sec - 1 and

diphenylsul-K m = 0.08 mM at pH 2 and 25.0C [224] Because ofthe asymmetry about the pyramidalsulfur atom, the synthetic sulfite esters represent racemic mixtures, and the fast resolu-tion of such compounds has been effected by means of pig pepsin in the case of phenyltetrahydrofurfuryl sulfite [225]

These various types of synthetic substrates have been used in studies on the primaryspecificity of aspartyl proteinases other than pig pepsin A, and significant differenceswere found For example chicken pepsin does not appear to hydrolyze Ac-Phe-Tyrtlj)[226], and although human gastric sin hydrolyzes the Tyr-Ala bond of Z-Tyr-Ala-OHeffectively, this substrate is relatively resistant to pig pepsin A [227] This finding is ofinterest in relation to the presence of an amino-terminal serine in human gastricsin[228] Beef gastric sin is much less active toward Ac-Phe-Tyrflj) or Z-His-Phe-Leu-OMe than is pig pepsin A, but can cleave the Phe(N02)-Nle bond of H-Leu-Ser-Phe(N02)-Nle-Ala-Leu-OMeat a comparable rate [176,198] The last-named peptidederivative has been proposed as a reference substrate for chymosin, which cleaves itselectivelyat the Phe(N02)-Nlebond [229] As was noted previously, the milk-clottingmechanism is initiated by the hydrolysis of a Phe-Met bond in kappa-casein, andsynthetic peptides containing this bond have also been prepared for test with calfchymosin [230,231] These studies have shown that the rate of cleavage of thePhe(N02)-Nleor Phe-Met bond depends greatly on the chain-length and amino acidsequence of the substrate For example, removal of the amino-terminal leucine from

H-Leu-Ser-Phe-Met-Ala-Ile-Pro-Pro-OH reduces thekcatlKmvalue for the cleavage ofthe Phe-Met bond by a factor of about 500 [231] The difference in specificity betweencalf chymosin and pig pepsin is clearly evident from the finding that one of the bestavailable synthetic substrates for pepsin, H-Phe-Gly-His-Phe(N02}-Phe-Ala-Phe-OMe, is hydrolyzed by chymosin at less than 1%of the rate for pepsin [232] A similardifference was found in the case of beef spleen cathepsin D in its action on thePhe(N02)-Phebond of this peptide, although the replacement of the carboxyl-terminalAla-Phe-OMe by Val-Leu-OMe increased the rate of cleavage by cathepsin D 25-fold[61].It has been shown that cathepsin D can release angiotensin I from renin substratesthrough cleavage of the Leu-Leu bond hydrolyzed by pig kidney renin itself [233].Synthetic peptides containing this bond are readily cleaved by reninifthe Leu-Leu unit

is part of the sequence H-Pro-Phe-His-Leu-Leu-Val-Tyr-Ser-OH, but shorter peptidesare relatively resistant [234] Consequently, although the primary specificity of theseseveral mammalian aspartyl proteinases appears to be directed toward the cleavage ofbonds linking hydrophobic amino acid residues, of far greater importance indetermining the rate of enzymic hydrolysis are the secondary enzyme-substrate inter-actions involving substrate residues not involved in the sensitive bond Theseconsiderations have had practical utility in the design of renin inhibitors of possibletherapeutic value in the treatment of hypertension [235,235a]

Extensive studies have been conducted on the kinetics of the hydrolysis of syntheticsubstrates for microbial aspartyl proteinases [236] Although most of those testedreadily hydrolyze Phe-Phe and Phe-Leu bonds in good pepsin substrates (e.g.,Z-His-Phe(N02)-Phe-Ala-Ala-OMe, Z-Phe-Leu-Ala-Ala-OH) [232], the microbialacid proteinases (e.g., Rhizopus pepsin, penicillopepsin) that can activate trypsinogenalso act effectively at Lys-Leu or Lys-Phe(N02 ) bonds [237-239] Some specificitydifferences were observed for the acid proteinases from a strain of Scytalidium lignicolumn,which are not inhibited by diazoacetyl-nt-norleucine methyl ester [50]

(b) Secondary specificity

In contrast to serine proteinases such as pancreatic trypsin, whose primary specificityfor Lys-Y and Arg-Y bonds is the dominant feature of its action, pig pepsin and otheraspartyl proteinases are more sensitive to the effect of secondary enzyme-substrateinteractions involving structural elements of the substrate distinct from the amino acidresidue joining the sensitive peptide bond [192] Systematic kinetic studies with anextensive series of synthetic substrates of the type A-Phe-Phe-B (or A-Phe(N02 ) -Phe-B) have shown that variation of either the A or B portion of the substrate mayproduce large changes in the value ofkcatlKm Itshould be noted that in the interaction

of oligopeptide substrates with a proteinase, the possibility of nonproductive bindingmust be considered For this reason, comparison ofkcat/Kmvalues is desirable becausethis parameter is independent of the effect of such nonproductive binding [240,241].The experimental data for the action of pig pepsin A on various A-Phe-Phe-B sub-strates have been discussed in several review articles [75,242,243] For example, the

kcatlKmvalue for the hydrolysis ofZ-Ala-Ala-Phe-Phe-OP4P is 2000 times greater than

that for the cleavage of Z-Phe-Phe-OP4P (see Table 1), and thekcat/Kmvalue for thehydrolysis of Z-His-Phe(N02)-Phe-Val-Leu-OMe is 2500 times greater than that ofZ-His-Phe(N02)-Phe-OMe[244,245] Of special significance is the finding that whilethe values of Km (shown to correspond to the dissociation constant Ks ) vary withinroughly a single order of magnitude, the values of kcat range over 3-4 orders ofmagnitude These data have permitted an approach to several problems of themechanism of peptide action, to be discussed in thefinalsection of this chapter Also,the results led to the estimate of about 25Afor the size of the extended active site ofpig pepsin [246], a value close to that later found by determination ofits crystal structure[154].

The Gibbs-energy change in the process:

RCOO - (1 M)+ +NH3R'(1 M) =RCONHR' (1 M) + H20(liq)

is near zero at pH 4 [247,248], but in dilute solution the equilibrium is far in the direction

of hydrolysis To effect the condensation reaction, high concentrations of the and +NH3R' components are required, or the synthetic product must be removed fromsolution [249] Interest in the abilityof pepsin to catalyze the synthesis of peptide bonds

RCOO-in condensation reactions stems from early work on the 'plasteRCOO-in' reaction [250] Thisreaction was observed when pepsin was allowed to act on egg albumin at pH 1.6, thedigest adjusted to pH 4 and concentrated, and more pepsin was added; the gradualformation of a precipitate was considered to be a consequence of peptide bond

formation.Itwas soon shown, however, that this plastein material represents a complexmixture of small peptides of unknown structure [251,252] Although the nature of theplastein reaction remained obscure, later work demonstrated that at pH 5 and highsubstrate concentration, pig pepsin can catalyze the conversion of oligopeptides topolymeric products [253,254] For example, H-Tyr-Leu-Gly-Glu-Leu-OH is anexcellent monomeric substrate; its amino-terminal residue may be replaced by Phe, butnot by n-Phe, and its carboxyl-terminal residue may be replaced by Phe, but not byD-Phe, lIe, Val or Ala Clearly, the preference for aromatic or hydrophobic aliphaticL-residues on both sides of the newly-formed peptide bond, evident from systematicstudies on the primary specificity of pepsin in the hydrolysis of synthetic peptidesubstrates, also applies to pepsin-catalyzed condensation reactions In more recentstudies [255-257], various condensation reactions leading to the formation of well-defined peptide derivatives were shown to be catalyzed efficiently by pig pepsin A Thekinetics of such condensation reactions have been investigated for the pepsin-catalyzedsynthesis of oligopeptides of the type A-Phe-Leu-B from A-Phe-OH and H-Leu-B[258] Variation of the A group led to large changes in the initial rates of the conden-sation reaction, and the effect of such changes was found to be similar to that previouslyobserved for the secondary specificity of pig pepsin in the hydrolysis of oligopeptidesubstrates Also, replacement of the Phe and Leu residues of A-Phe-OH and H-Leu-B

by other amino acid residues gave relative rates of synthesis in accord with the knownprimary specificity of the hydrolytic action of pig pepsin

(d) Transpeptidation reactions

The suggestion that proteinases may be efficient catalysts of transfer reactions not only

to water, but also to other acceptor molecules, has received extensive experimentalsupport [259] Such transfer reactions, termed transpeptidation or transamidationreactions, may be of two types: (l) acyl transfer, in which the RCO portion of a substrateRCO-X is transferred to an acceptor (YNH2 or YOH) to yield RCO-NHY orRCO-OY; and (2) amino transfer in which the NHR' portion of a substrate X-NHR'

is transferred to an acceptor (YCOOH) to yield YCO-NHR'

In the case of pig pepsin, it was found that near pH 4.5, with a relatively poorsubstrate such as Z-Glu-Tyr-OH or Z-Tyr-Tyr-OH, significant amounts ofH-Tyr-Tyr-

OH are formed [260,261] Similar transpeptidation was also found with substrates ofthe type H-(GlY)n-Tyr-Tyr-OH (n= 1-3) [262] These observations suggested theintermediate formation of an amino-enzyme (ECO-Tyr-OH) and its reaction with thesubstrate (X-Tyr-OH) to form X-Tyr-Tyr-OH, which is then cleaved at the X-Tyr bond.Although some additional evidence was offered in favor of this explanation [263,264],doubt was cast on its validity by subsequent work [265,266]

In addition to amino-transfer reactions, pig pepsin also acts at pH 3.4 on substrates

of the type H-Leu-Y (e.g., H-Leu-Trp-Met-Arg-OH) with the formation

ofH-Leu-Leu-OH and H-Leu-Leu-Leu-ofH-Leu-Leu-OH [267] This was taken as evidence for the existence of anintermediate covalent acyl enzyme (H-Leu-E) which reacts with more of the substrate

to generate the transpeptidation products by an acyl transfer mechanism In a study with

H-[I4C]Leu-Tyr-[3H]Leu-OH, it was shown that pig pepsin can produce bothH-[ 14C]Leu-[14C]Leu-OH (by acyl transfer) and H-[ 3H]Leu-[ 3H]Leu-OH (by aminotransfer) [268] Such acyl- and amino-transfer reactions are also catalyzed by penicil-lopepsin [269,270].

Transpeptidation reactions have also been considered in relation to the finding thatsome peptide derivatives (e.g., Z-Leu-Met-OH) markedly increase the rate of cleavage

of poor pepsin substrates such as H-Leu-Tyr-NH2and enhance the formation of thetranspeptidation product H-Leu-Leu-Leu-OH [271] Although acyl enzyme inter-mediates were initially assumed to be involved, subsequent work showed that the keystep is the pepsin-catalyzed condensation reaction to form a new substrate (e.g.,Zvl.eu-Met-Leu-Tyr-Nrlj) which is then cleaved at a different bond [272-274]

(a) Binding of substrate at active site

A considerable body of data has been gathered to show that in the action of pig pepsin A

on peptide substrates such as Ac-Phe-Phe-Olf, Z-His-Phe-Phe-OMe, or Gly-Phe-Phe-OP4P, the value ofK m determined under conditions where [S]o~ [E],approximates the values ofK. (orK D ) = k_l/kl for the process [243]:

As was noted previously in this chapter, the extensive kinetic data on the cleavage

of comparable synthetic peptide substrates by pig pepsin have shown that structuralalterations leading to striking changes in catalytic efficiency (kc a t ) are often notaccompanied by significant changes in binding affmity Thus, for substrates of the type

A-Phe-Phe-OP4P, in which A= Z-Gly-Gly, Z-Gly-Ala, and Z-Gly-Pro, at pH 3.5 and37°C, K m varied between 0.1 and 0.4 mM, near the value of 0.25 mM for the disso-ciation constant of the complex of pepsin with the Phe-Phe unit, whereas the kc a tvalueswere 72, 410 and 0.06 sec -1 respectively [245] Similar results were obtained withsubstrates of the type Phe-Gly-His-Phe(N02)-Phe-Bin which B = OMe, Ala-OMe andAla-Ala-OMe; theK mvalues feU between 0.2 and 0.4 mM but the kc a tvalues were 0.1,3.3 and 28 sec -1respectively [244] These results are consistent with the view that thesecondary interactions of the A and B groups with the extended active site (corre-sponding to a peptide segment of about 7 amino residues) may affect catalysis by theutilization of the potential binding energy in the enzyme-substrate interaction to lowerthe Gibbs energy of activation in the catalytic process [275] This could arise byconformational change in the substrate or the catalytic site of the enzyme, or both,leading to strain or distortion at the sensitive bond The conformational change in thesubstrate may produce a transition state for which the active site has greater affmitythanfor either the free substrate or the products If the active site is not a rigid structure, butcan undergo conformational change in response to its interaction with the substrate, aportion of the potential binding energy could be used to achieve a transition state inwhich the active site is strained or distorted, and catalysis would be favored by thetendency of the enzyme to return to its native state [242] Such complementary confor-mational changes in the substrate and the enzyme may therefore be involved in thecontribution of the entropy loss in the formation of the enzyme-substrate complex tothe energy required to reach the transition state [276].

The question of the flexibility of active sites is one of the central unsolved problems

of enzymic catalysis In the case ofthe aspartyl proteinases, evidence for such flexibilityhas been adduced from the changes in the crystal structure of penicillopepsin uponbinding a pepstatin analogue [165] On the other hand, no significant dislocation ofactive site groups was observed in the crystal structure of Rhizopus pepsin upon thebinding of pepstatin [158] The latter negative fmding led the authors to suggest thatthe variations of kc a t /K m values for pepsin substrates may be attributed to nonproduc-tive binding rather than to conformational changes in the enzyme [158]; this criticism

is inappropriate since, when Michaelis-Menten kinetics are obeyed, kcat/Kmis

indepen-dent of nonproductive binding Further studies on the relation of the results of crystalstructure determinations of complexes of enzymes with substrate analogues to theobservations made with enzymes in the dissolved state are needed In this connection

it may be noted that a change in the circular dichroism spectrum of penicillopepsin wasobserved upon its interaction of H-Leu-Gly-Leu-OH; this peptide appears to interactwith the enzyme at a secondary binding site [277]

As was noted previously in this chapter, some of the microbial proteinases canactivate trypsinogen by the cleavage of a Lys-Ile bond, and can hydrolyze syntheticsubstrates at such bonds Studies on penicillopepsin and Rhizopus pepsin haveindicated that, in addition to hydrophobic interactions, binding of the peptide substrateAc-Ala-Ala-Lys-Phe(N02)-Ala-Ala-NH2 also involves carboxylate groups of theseenzymes [278] Clearly, with this substrate, the two Ala-Ala units contribute to thesusceptibility of the Lys-Phe(N02 ) bond through secondary enzyme-substrate inter-

actions Similar effects of secondary interactions have been found in the search forspecific inhibitors of other aspartyl proteinases For example, beef spleen cathepsin D

is inhibited by pyroGlu-D-Phe-Pro-Phe-Phe-Val-D-Trp-OH with aK,of 10-8M, nearthat found with pepstatin [279J This peptide inhibits pig pepsin (K j= 4 x 10-7M).Also, a specific inhibitor of human renin, H-Pro-His-Pro-Phe-His-Phe-Phe-Val-Tyr-Lys-OH, is effective in vivo [280J

The association of pig pepsin with oligopeptide substrates is a very rapid process.Attempts to determine its rate at 250C through stopped-flow kinetic measurements ofthe increase in tluorescence of a substrate such as Dns-Ala-Ala-Phe-Phe-OP4P werenot successful [138], and it can only be stated that the estimated second-order rate ofassociation is greater than 106M -1sec - 1.However, the possibility is not excluded thatwith such oligopeptide substrates, there may be stepwise binding with an initial'nucleation' step involving the Phe-Phe segment, followed by a cooperative process inwhich the remaining segments of the oligopeptide are drawn into the site If such aprocess of mutual conformational adjustment of both the substrate and the active siteoccurs in discrete successive steps, the overall activation energy in the associationprocess may be lower than in the interaction of a substrate with a rigid active site [281]

(b) Transition state and bond cleavage

Since in the kinetics of pepsin action on peptide and depsipeptide substratesK m hasbeen shown to approximate the dissociation constant of the first detectableenzyme-substrate complex, it is reasonable to conclude that the rate-limiting step inthe cleavage of the sensitive bond is the formation of the transition-state complex Theinvolvement of at least two enzymic carboxyl groups, one as carboxylate and the other

in its undissociated form, in this step was inferred from pH-dependence data, fromstudies on the deuterium-isotope effect on the kinetics of hydrolysis of H-Gly-(Glyh-Phe(N02)-Phe-OMeby pepsin [282], and on the pepsin-eatalyzed exchange ofAc-Phe-

OH with H2180 [283,284] The involvement of an undissociated carboxyl group as anacid catalyst would be consistent with the similarity in the rates of hydrolysis of the esterbond of Z-Phe(N02)-Pla-OMe and of the amide bond of the corresponding peptide[71J Also, studies on the hydrolysis of dialkylmaleamic acids provided an attractivemodel for pepsin action [285] According to this model, the attack at the amide bond

is initiated by a neighboring carboxyl group with the formation of tetrahedral mediate, and an additional carboxyl group (in its carboxylate form) promotes theinterconversion of the neutral and dipolar forms of this intermediate (Fig 6) As applied

inter-to the mechanism of pepsin catalysis, the model suggests a tetrahedral intermediate

leading to the formation of a covalent acyl enzyme (an acid anhydride) and the amineproduct Thus far, however, efforts to detect burst reactions in the hydrolysis of goodsubstrates of pig pepsin or of penicillopepsin have been unsuccessful [286] and, asmentioned previously in this chapter, the existence of covalent acyl enzyme or aminoenzyme intermediates inferred from transpeptidation studies with poor substrates isunlikely.

The demonstration through sequence and chemical modification studies and structure determinations that it is Asp-32 which provides the carboxylate group left openthe question whether this group acts as nucleophile in attacking the carboxyl carbondirectly, or whether it acts as a general base in abstracting a proton from a watermolecule as shown in Fig 7 Strong evidence for the latter alternative has come fromthe fmding that when pig pepsin A acts on H-Leu-Tyr-NH2in the presence of H2 180,the isotope is incorporated into the H-Leu-Leu-OH formed by transpeptidation [287].This important result excludes the intermediate formation of a covalent acyl enzyme ofthe kind established for serine proteinases such as chymotrypsin or cysteine proteinasessuch as papain In the action of these enzymes, two tetrahedral intermediates areenvisaged: one involved in the formation of the acyl enzyme and the other involved inits deacylation It has been suggested that the transition-state stabilization of suchintermediates may account for much of the catalytic efficiencyof the proteinases [288]

crystal-In support of this view is the demonstration that peptide aldehydes inhibit someproteinases; for example Ac-Phe-glycinal is a strong inhibitor of papain, by virtue of itsreaction with the active site sulfhydryl group to form a thiohemiacetal that resemblesthe presumed tetrahedral intermediate in papain catalysis [289,290] It has beensuggested that pepstatin is a strong inhibitor of aspartyl proteinases by virtue of thepresence of the statyl residue, whose tetrahedral carbinol group is viewed as beinganalogous to the transition state in the catalytic mechanism of these enzymes [131] It

has been found that the [13C]NMR spectrum of a ketone analogue of pepstatin,enriched with 13C in its C-3 position, undergoes a marked shift when bound to pigpepsin [291] This fmding is consistent with the formation of a tetrahedral C-3 carbon

as a consequence of the addition of an oxygen nucleophile to the carbonyl group.Moreover, NMR studies on the binding of a ketomethylene isostere of a pepsin sub-strate isovaleryl-Val-[4_13C]_(4-oxo-5S)-amino-7-methyl-octanoyl-Ala-isoamylamide,support the view that the formation of the transition state in pepsin-catalyzed reactionsinvolves acid-base catalysis, rather than nucleophilic attack of an enzymic carboxylategroup at the carbonyl carbon of the substrate [292] Although the formation of acidanhydride intermediates in the cleavage of peptide substrates by pig pepsin appears to

be excluded, the possibility exists that such covalent acyl enzymes play a role in theenzymic hydrolysis of sulfite esters [293]

Much is still unclear about the nature of the transition state in the catalytic action

of the aspartyl proteinases, but the available evidence is more consistent with anacid-base mechanism than with one involving direct nucleophilic attack by an enzymiccarboxylate group at the carboxyl carbon of the sensitive bond Support for this viewhas come from the determination of the crystal structure ofpenicillopepsin [294], andalthough the details of the mechanism proposed on the basis of this structure mayrequire revision, it is clear that as suggested before [243], the aspartyl proteinasesrepresent a group of enzymes in which intermediates in the catalytic process are held

at the extended active site by noncovalent interaction

(c) Release ofproducts

A necessary consequence of the conclusion that noncovalent interactions are important

in the intermediate steps in catalysis by aspartyl proteinases is that there may be thepossibility of the ordered release of products, so as to explain the occurrence oftranspeptidation reactions of the acyl- or amino-transfer type However, attempts todetect by direct experimental means the sequential release of the products of a pepsin-catalyzed reaction have thus far been unsuccessful For example, stopped-flowfluorescence measurements with substrates of the type A-Phe-Phe-OP4P, where the Agroup bears an amino-terminal mansyl or dansyl group, have been conducted underconditions of [E],~ [S]o in order to determine the rate of conversion of the initialenzyme-substrate complex to the equilibrium established between the acidic product

and its complex with pepsin Thus, with Dns-Gly-Gly-Phe-Phe-OP4P, K; =0.1 mM,and the value of K D for Dns-Gly-Phe-OH is 0.3 mM The resulting decrease influorescence follows strict first-order kinetics, and a plot of l/k o b sagainst l/[E]t gavevalues ofk 2that agreed with the kc a tvalues obtained under conditions of [S]o~ [E],for the release of H-Phe-OP4P [138].Itmay be inferred, therefore, that in the hydrolysis

of good substrates both products leave the active site at the same time, within the limits

of experimental measurement

The occurrence of transpeptidation reactions with relatively poor substrates, and notwith substrates that cleaved rapidly, has therefore been explained by assuming that theinteractions of the split products with the extended site of pepsin may be coupled, so

that the nature of one product influences the rate of departure of the other productthrough the effect it has on the conformational state of the active site, and the twoproducts may therefore leave in a manner that resembles the formation of an acylenzyme or an amino enzyme [243] Given a sufficient lifetime of the enzyme-productcomplex, entry into the extended active site of a new acidic or amino component wouldthen lead, through condensation reactions, to the transpeptidation products Thishypothesis depends on a feature of the specificity of the aspartyl proteinases notapplicable to serine- or cysteine-proteinases, namely the requirement for a hydrophobict-amino acid (or hydroxy acid) as the donor of the NH (or oxygen) to the sensitive bond.One way to test this hypothesis is to study the kinetics of the condensation reactionscatalyzed by pepsin, by examining the effect of variations in the structure of the aminecomponent (e.g., H-Phe-Y) on the rate of entry of the acidic component (e.g.,Z-Phe-OH) into the process leading to the product (e.g., Z-Phe-Phe-Y), but the sparingsolubility of the reactants near pH 5 made this approach impracticable [258].Itshould

be noted, however, that this kind of experiment has been possible with themetalloproteinase thermolysin, whose primary specificity and mechanism of actionresemble the aspartyl proteinases, except for the replacement of the carboxyl group of

an aspartyl residue by a zinc ion Since thermolysin acts optimally near pH 7.5 it waspossible to show that the rate of entry of H-Phe-Gly-OMe into the condensationreaction with Z-Phe-OH is 100 times greater than with H-Phe-OMe as the aminecomponent [295]

Acknowledgement

The preparation of this chapter and the research of our laboratory reported therein wereaided by a grant from the National Institutes of Health (GM-18172)

References

1 Schwarm, T (1836) Muliers Arch Anat Physiol., 90-138.

2 Hickel, E (1975) Naturwiss Rundsch 28, 14-18.

3 Northrop, J.H (1930) J Gen Physiol 13, 739-780.

4 Fruton, J.S (1977) Trends Biochem Sci 2, 210-211.

5 Langley, J.N and Edkins, B (1886) J Physiol (London) 7, 371-415.

6 Herriott, R.M (1938) J Gen Physiol 21, 501-540.

7 Hirschowitz, B.I (1957) Physiol Rev 37,475-511.

8 Taylor, W.H (1962) Physiol Rev 42, 519-533.

9 Johnson, L.R (1977) Physiol Rev 39, 135-158.

10 Ryle, A.P (1970) Methods Enzymol 19,319-336.

11 Etherington, DJ and Taylor, W.H (1969) Biochem J 113,663-668.

12 ZOller, M., Matzku, S and Rapp, W (1976) Biochim Biophys, Acta 427,708-718.

13 Deschamps, J.B (1840) J Pharm 26, 412-420.

14 Foltmann, B (1966) C R Trav Lab Carlsberg 35, 143-231.

15 Foltmann, B (1981) Essays Biochem 17,52-84.

16 Peach, MJ (1977) Physiol Rev 57, 313-370.

17 Inagami, T., Chang, U., Dykes, C.W., Takii, Y., Kisaragi, M and Misono, KS (1983) Proc Fed Am Soc Exp BioI 42, 2729-2734.

18 Press, E.M., Porter, RR and Cebra, J (1960) Biochern J 74, 501-514.

19 Figueiredo, A.F.S., Takii, Y, Tsuji, H., Kato, K and Inagami, T (1983) Biochemistry 22,5476-5481.

20 Woessner, J.F., Jr (1977) in: Acid Proteases (Tang, J., Ed.), pp 313-327 Plenum Press, New York.

21 Barth, R and Afting, E.G (1984) Biochern J 219, 899-904.

22 Smith, G.D., Murray, M.A., Nichol, L.W and Trikojus, V.M (1969) Biochim Biophys, Acta 171, 288-298.

23 Fukushima, K., Gnoh, G.H and Shinano, S (1971) Agr, BioI Chern 35, 1495-1502.

24 Orlowski, M (1983) Mol Cell Biochern 52, 49-74.

25 Marks, N., Grynbaurn, A and Lajtha, A (1973) Science (Washington, D.C.) 181,949-951.

26 Ruenwongsa, P and Chulavatnatol, M (1975) J BioI Chern 250, 7574-7578.

27 Reichelt, D., Jacobsohn, E and Haschen, R.I (1974) Biochim Biophys Acta 341, 15-26.

28 Bowers, W.E., Panagides, J and Yago, N (1977) in: Acid Proteases (Tang, J., Ed.), pp.301-312, Plenum Press, New York.

29 Moriyama, A., Kageyama, T and Takahashi, K (1983) Eur J Biochern 132,687-692.

30 Shinano, S and Fukushima, K (1971) Agr BioI Chern 35, 1488-1494.

31 Arnagase, C (1972) J Biochern (Tokyo) 72, 73-81.

32 Sodek, J and Hofmann, T (1970) Can J Biochern 48, 1014-1017.

33 Fukumoto, J., Tsuru, D and Yamamoto, T (1976) Agr BioI Chern 31, 710-717.

34 Hagemeyer, K, Fawwai, I and Whitaker, J.R (1968) J Dairy Sci 51, 1916-1930.

35 Ottesen, M and Rickert, W (1970) C R Trav Lab Carlsberg 37, 301-325.

36 Uchino, F., Kurono, Y and Doi, S (1976) Agr BioI Chern 31,428-434.

37 Yoshida, F (1956) Bull Agr Chern Soc Japan 20 Suppl., 252-256.

38 Rnchel, R (1981) Biochim Biophys Acta 659 99-113.

39 Oda, K., Kamada, M and Murao, S (1972) Agr BioI Chern 36, 1103-1108.

40 Kolaczewski, M.K., Wieczorek, M and Polanowski, A (1983) Eur J Biochern 132, 557-561.

41 Hwang, J and Hseu, T H (1980) Biochim Biophys Acta 614, 607-612.

42 Arirna, K., Yu, J and Iwasaki, S (1970) Methods Enzymol 19,446-459.

43 Sawada, J (1963) Agr BioI Chern 27, 677-683.

44 Erni, S., Myers, D.V and Iacobucci, G.A (1976) Biochemistry 15,842-848.

45 Kamada, M., Oda, K and Murao, S (1972) Agr BioI Chern 36, 1095-110l.

46 Rudenskaya, G.N., Gaida, A.V and Stepanov, V.M (1979) Biokhimiya (Moscow) 45,561-568

47 Tornoda, K and Shimazono, H (1964) Agr BioI Chern 28, 770-778.

48 Gaida, A.V., Ostermann, A.L., Rudenskaya, G.N and Stepanov, V.M (1981) Biokhimiya (Moscow)

46, 181-189.

49 Dreyer, T., Halkjer, B., Svendsen, I and Ottesen, M (1985) in: Aspartic Proteinases and their Inhibitors (Kostka, V., Ed.), de Gruyter, Berlin.

50 Oda, K, Murao, S., Oka, T and Morihara, K (1976) Agr BioI Chern 40, 859-866.

51 Sodek, J and Hofmann, T (1971) Methods Enzymol 19,372-396.

52 Dixon, M and Webb, E.C (1958) Enzymes, Academic Press, London.

53 Steinhardt, J (1939) J BioI Chern 129,135-170.

54 Meitner, P.A and Kassell, B (1971) Biochern J 121,249-256.

55 Bohak, Z (1969) J BioI Chern 244,4638-4648.

56 Van Vunakis, H and Herriott, RM (1957) Biochim Biophys Acta 23, 600-608.

57 Rajagopaian, T.G., Moore, S and Stein, W.H (1966) J BioI Chern 241,4940-4950.

58 Chernaya, M.M., Adli, K., Lavrenova, G.I and Stepanov, V.M (1976) Biokhimiya (Moscow) 41, 732-739.

59 Esurni, H., Sato, S., Sugirnura, T and Furihata, C (1978) FEBS Lett 86,33-36.

60 Smith, R and Turk, V (1974) Eur J Biochern 48, 245-254.

61 Ferguson, J.D., Andrews, J.R, Voynick, I.M and Fruton, J.S (1973) J BioI Chern 248, 6701-6708.

62 Cunningham, M and Tang, J (1976) J BioI Chern 251, 4528-4536.

63 Yamamoto, K., Katsuda, N., Himano, M and Kato, K (1979) Eur J Biochern 95, 459-467.

64 Huang, J.S., Huang, S.S and Tang, J (1979) J BioI Chern 254, 11405-11417.

65 Northrop, J.H (1933) J Gen Physiol 16,615-623.

66 Norris, E.R and Elam, D.W (1940) J Biol, Chern 134,443-454.

67 Bunn,csi Camerman, N., Tsai, L.T., Moews, P.e and Baumber, M.E (1970) Philos Trans R Soc London Ser B 257, 153-158.

68 Kazakova, O.V and Orekhovich, V.N (1976) Biochern Biophys Res Commun 72 747-752.

69 Anson, M.L (1940) J Gen Physiol 23, 695-704.

70 Kunitz, M (1938) J Gen Physiol 21, 601-620.

71 Inouye, K and Fruton, J.S (1967) Biochemistry 6, 1765-1777.

72 Jackson, W.T., Schlamowitz, M and Shaw, A (1965) Biochemistry 4, 1537-1543.

73 Hollands, T.R and Fruton, J.S (1968) Biochemistry 7,2045-2053.

74 Williams, R.C and Rajagopalan, T.G (1966) J Bioi Chern 241,4951-4954.

75 Fruton, J.S (1971) in: The Enzymes (Boyer, P.D., Ed.), Vol 3, pp 119-164, Academic Press, New York.

76 Vesterberg, O (1973) Acta Chern Scand 27, 2415-2420.

77 Jonsson, M (1972) Acta Chern Scand 26, 3435-3440.

78 Strongin, A.Y., Levin, E.D and Stepanov, V.M (1973) Biokhimiya (Moscow) 38, 1098-1100.

79 Graham, J.E.S., Sodek, J and Hofmann, T (1973) Can J Biochern 51, 789-796.

80 Sepuve1da, P., Marciniszyn, J., Liu, D and Tang, J (1975) J BioI Chern 250, 5082-5088.

81 Perlrnann, G.E (1952) J Am Chern Soc 74, 6308-6309.

82 Lee, D and Ryle, A.P (1967) Biochern J 104,735-748.

83 Pedersen, V.B and Foltrnann, B (1973) FEBS Lett 35,255-256.

84 Foltrnann, B and Pedersen, V.B (1977) in: Acid Proteases (Tang, J., Ed.), pp 3-22, Plenum Press, New York.

85 Arai, K.M., Muto, N., Tani, S and Akahane, K (1984) Biochim Biophys Acta 788, 256-261.

86 Green, M.L and Llewellin, J.M (1973) Biochern J 133, 105-115.

87 Foltrnann, B., Pedersen, V.B., Jacobsen, H., Kauffman, D and Wybrandt, G (1977) Proc Natl, Acad Sci (U.S.A.) 74, 2321-2324.

88 Misono, K.S., Chang, U and Inagami, T (1982) Proc Nat! Acad Sci (U.S.A.) 79, 4858-4862.

89 Shewale, J.G and Tang, J (1984) Proc Natl Acad Sci (U.S.A.) 81, 3703-3707.

90 Fruton, J.S (1970) Adv Enzymol 33,401-443.

91 Herriott, RM (1947) Adv Prot Chern 3, 169-225.

92 Delpierre, G.R and Fruton, J.S (1965) Proc Natl Acad Sci (U.S.A.) 54, 1161-1167.

93 Delpierre, G.R and Fruton, J.S (1966) Proc Natl Acad Sci (U.S.A.) 56, 1817-1822.

94 Badley, R.A and Teale, F.W.J (1970) Biochern J 116,341-348.

95 Rajagopalan, T.G., Stein, W.H and Moore, S (1966) J BioI Chern 241,4295-4297.

96 Lundblad, RL and Stein, W.H (1969) J BioI Chern 244, 154-160.

97 Widmer, F and Visnawatha, T (1980) Carlsberg Res Commun 45, 149-159.

98 Bayliss, RS., Knowles, J.R and Wybrandt, G.B (1969) Biochern J 113,377-386.

99 Erlangen, B.F., Vratsanos, S.M., Wassermann, N and Cooper, A.G (1967) Biochern Biophys Res Cornrnun 28,203-208.

100 Kozlov, L.V., Ginodman, L.M and Orekhovich, V.N (1967) Biokhimiya (Moscow) 32, 1011-1019.

101 Gross, E and Morell, J.L (1966) J BioI Chern 241, 3638-3639.

102 Tang, J (1971) J BioI Chern 246, 4510-4517.

103 Llewellin, J.M and Green, M.L (1975) Biochern J 151,319-326.

104 Hartsuck, J.A and Tang, J (1972) J Bioi Chern 247, 2575-2580.

105 Chen, K.e.S and Tang, J (1972) J BioI Chern 247, 2566-2574.

106 Chang, W and Takahashi, K (1973) J Biochern (Tokyo) 74, 231-237.

107 Paterson, A.K and Knowles, J.R (1972) Eur J Biochern 31, 510-517.

108 Herriott, RM., Anson, M.L and Northrop, J.H (1946) J Gen Physiol 30, 185-210.

109 Herriott, RM (1935) J Gen Physiol 19,283-299.

110 Perlrnann, G.E (1966) J BioI Chern 241, 153-157.

III Rirnon, S and Perirnann, G (1968) J BioI Chern 243, 3566-3572.

112 Kozlov, L.V., Kogan, G.A and Zavada, L.L (1969) Biokhimiya (Moscow) 34, 1257-1265.

113 Herriott, RM (1937) J Gen Physiol 20, 335-352.

114 Mains, G., Burchell, RH and Hofinann, T (1973) Biochern Biophys Res Commun 54,275-282.

115 Zavada, L.L., Kozlov, L.V., Volkova, L.I and Antonov, V.K (1977) Bioorg Khim 3, 1671-1681.

116 Dopheide, T.A.A and Jones, V.M (1968) J BioI Chern 243, 3906-391 I.

117 Lokshina, L.A.,Orekhovich, V.N and Pandakova, V.N (1962) Dold Akad.Nauk SSSR 142,471-473.

118 Kitson, T.M and Knowles, J.R (1971) FEBS Lett 16, 337-338.

119 Huang, W.Y and Tang, J (1972) J BioI Chern 247, 2704-2710.

120 Gripon, J.C and Hofinann, T (1981) Biochern J 193,55-65.

121 Tarasova, N.I., Lavrenova, G.I and Stepanov, V.M (1980) Biochern J 187, 345-352.

122 Schechter, Y., Rubinstein, M., Becker, R and Bohak, Z (1975) Eur J Biochern 58, 123-131.

123 Philpot, J.S.L and Small, P.A (1938) Biochern J 32, 542-551.

124 Urnezawa, H (1972) Enzyme Inhibitors of Microbial Origin, University of Tokyo Press, Tokyo >

125 Kunirnoto, S., Aoyagi, T., Nishizawa, R, Komai, T., Takeuchi, T and Urnezawa, H (1974)1 Antibiot 27,413-418.

126 Fukurnura, M., Satoi, S., Kuwana, N and Murao, S (1971) Agr Bioi Chern 35, 1310-1312.

127 Marciniszyn, J., Hartsuck, J.A and Tang, J (1976) J BioI Chern 251, 7088-7094.

128 Rich, D.H., Boporai, A.S and Bernatowicz, M.S (1982) Biochern Biophys Res Commun 104, 1127-1133.

129 Workman, R.I and Burkitt, D.W (1979) Arch Biochern Biophys 194, 157-164.

130 Schmidt, P.G., Bernatowicz, M.S and Rich, D.H (1982) Biochemistry 21, 6710-6716.

131 Marciniszyn, J., Hartsuck, J.A and Tang, J (1977) in: Acid Proteases (Tang, J., Ed.), pp 199-210 Plenum Press, New York.

132 Fairclough, G.F., Jr and Fruton, J.S (1966) Biochemistry 5, 673-683.

133 Raju, E.V., Humphreys, RE and Fruton, J.S (1972) Biochemistry II, 3533-3536.

134 Cory, RR, Becker, R.R, Rosenbluth, R and Isenberg, I (1968) J Am Chern Soc 90, 1643-1647.

135 Edelman, G.M and McClure, W.O (1968) Acc Chern Res 1,65-70.

136 Stryer, L (1968) Science (Washington, D.C.) 162,526-533.

137 Sachdev, G.P., Brownstein, A.D and Fruton, J.S (1973) J Bioi Chern 248, 6292-6299.

138 Sachdev, G.P and Fruton, J.S (1975) Proc Natl Acad Sci (U.S.A.) 72, 3424-3427.

139 Sachdev, G.P., Brownstein, A.D and Fruton, J.S (1975) J BioI Chern 250, 501-507.

140 Nakatani, H., Tsuchiya, M., Morita, T., Ohki, S and Hiromi, K (1980) Biochim Biophys Acta 614, 144-150.

141 Dunn, B.M., Pham,c.,Ranay, L., Abayasekara, D., Gillispie, W and Hsu, A (1981) Biochemistry 20, 7206-721 I.

142 Sachdev, G.P., Johnston, M.A and Fruton, J.S (1972) Biochemistry II, 1080-1086.

143 Sachdev, G.P., Brownstein, A.D and Fruton, J.S (1974) J BioI Chern 249, 420-427.

144 Antonov, V.K and Dyakov, V.L (1975) Bioorg Khim I, 1324-1331.

145 Edelhoch, H (1958) J Am Chern Soc 80, 6640-6655.

146 Perirnann, G.E (1967) Adv Chern Ser 63,268-281.

147 Kozlov, L.V., Meshcheryakova, E.A., Zavada, L.L., Efrernov, E.S and Rashkovetskii, L.G (1979) Biokhimiya (Moscow) 44, 338-349.

148 Edelhoch, H., Frattali, V and Steiner, R.F (1965) J BioI Chern 240, 122-127.

149 Gounaris, A.D and Perlmann, G.E (1967) J BioI Chern 242, 2739-2745.

150 McPhie, P (1982) Biochemistry 21, 5509-5515.

151 Privalov, P.L., Mateo, P.L., Khechinashvili, N.N., Stepanov, V.M and Revina, L.P (1982)J Mol Bioi 152,445-464.

152 Murakami, K and Inagami, T (1975) Biochern Biophys Res Commun 62, 757-763.

153 Bernal, J and Crowfoot, D (1934) Nature (London) 133, 795-796.

154 Andreeva, N.S., Zdanov, A.S., Gustchina, A.S and Fedorov, A.A (1984) J BioI Chern 259, 11353-11365.

155 James, M.N.G and Sielecki, A.R (1983) J Mol BioI 163,299-361.

156 Subramanian, E., Swan, I.DA, Liu, M., Davies, RD., Jenkins, J.A., Tickle, I.J and Blundell, T.L (1977) Proc Natl Acad Sci (U.S.A.) 74, 556-559.

157 Wong,cn.,Lee, TJ., Lee, T.Y., Liu, T.H and Yang, I.H (1979) Biochemistry 18, 1638-1640.

158 Bott, R, Subramanian, E and Davies, D.R (1982) Biochemistry 21, 6956-6962.

159 Pearl, L and Blundell, T (1984) FEBS Lett 174,96-101.

160 Subramanian, E., Liu, M., Swan, D.A and Davies, D.R (1977) in: Acid Proteases (Tang, J., Ed.),

pp 33-41, Plenum Press, New York.

161 Jenkins, J., Tickle, I., Sewell,T., Ungaretti, L., Wollmer, A and Blundell,T.L (1977)in: Acid Proteases (Tang, J., Ed.), pp 43-60, Plenum Press, New York.

162 Safro, M.G and Andreeva, N.S (1979) Dold Akad Nauk SSSR 247, 758-761.

163 Blundell, T.L., Sibanda, B.L and Pearl, L (1983) Nature (London) 304, 273-275.

164 Tang, J., James, M.N.G., Hsu, LN., Jenkins, J.A and Blundell, T.L (1978) Nature (London) 271, 618-621.

164a Miyazaki, H., Fukamizu, A., Hirose, S., Hayashi, T., Hori, H., Ohkubo, H., Nakanishi, S and Murakami, K (1984) Proc Nat! Acad Sci (U.S.A.) 81, 5999-6003.

165 James, M.N.G., Sielecki, A., Salituro, F., Rich, D.H and Hofmann, T (1982) Proc Natl, Acad Sci U.S.A 79, 6137-6141.

166 Andreeva, N.S., Gushchina, A.E and Fedorov, A.A (1981) Dold Akad Nauk SSSR 259,1261-1264.

167 Herriott, RM (1939) J Gen Physiol 22,65-78.

168 Bustin, M and Conway-Jacobs, A (1971) J BioI Chem 246, 615-620.

169 McPhie, P (1974) Biochem Biophys Res Commun 56, 789-792.

170 Dykes, e.W and Kay, J (1976) Biochem J 153, 141-144.

171 Christensen, KA., Pedersen, V.B and Foltmann, B (1977) FEBS Lett 76,214-218.

172 Glick, D.M., Valier, MJ., Rowlands,cc,Evans, J.e and Kay, J (1982) Biochemistry 21, 3746-3750.

173 Pedersen, V.B., Christensen, K.A and Foltmann, B (1979) Eur J Biochem 94, 573-580.

174 Harboe, M., Andersen, P.M., Foltmann, B., Kay, J and Kassell, B (1974) J BioI Chem 249, 4487-4494.

175 Keilova, H., Kostka, V and Kay, J (1977) Biochem J 167, 855-858.

176 Martin, P., Trieu-Cuot, P., Collin, J.C and Ribadeau Dumas, B (1982) Eur J Biochem 122,31-39.

177 Dunn, B.M., Deyrup, C., Moesching, W.G., Gilbert, WA, Nolan, RJ and Trach, M.L (1978)J BioI Chem 253, 7269-7275.

178 Wang, J.L and Edelman, G.M (1971) J BioI Chem 246, 1185-1191.

179 Davie, E.W and Neurath, H (1955) J BioI Chem 212, 515-529.

180 Armstrong, C.E., Mackinlay, A.G., Hill, RJ and Wake, RG (1967) Biochim Biophys Acta 140, 123-131.

181 Hill, R.D., Lahav, E and Givol, D (1974) J Dairy Res 41, 147-153.

182 Graf, L., Kenessey, A., Patthy, A., Grynbaum, A., Marks, N and Lajtha, A (1979) Arch Biochem Biophys 193, 101-109.

183 Helseth, D.L and Veis, A (1984) Proc Natl Acad Sci U.S.A 81, 3302-3306.

184 Kraut, H and Tria, E (1973) Biochem Z 290, 277-288.

185 Perlmann, G.E (1954) Nature (London) 173,406.

186 Funatsu, M and Tokuyoso, K (1959) J Biochem (Tokyo) 46, 1441-1451.

187 Determann, H., Jaworek, D., Kotitschke, R and Walch, A (1969) Hoppe-Seyler's Z Physiol Chem.

350, 379-388.

188 Schlamowitz, M and Peterson, L.U (1959) J BioI Chem 234, 3137-3145.

189 Linderstrem-Lang, K (1939) Proc R Soc London B127, 17-18.

190 Weber, G and Young, L.B (1964) J BioI Chem 239, 1415-1431.

191 Wilson, W.D and Foster, J.F (1971) Biochemistry 10, 1772-1780.

192 Fruton, J.S (1975) in: Proteases and Biological Control (Reich, E., Rifkin, D.B and Shaw, E., Eds.),

pp 33-50, Cold Spring Harbor Laboratory, New York.

193 Hill, R.L (1965) Adv Protein Chem 20, 37-107.

194 Tang, J (1963) Nature (London) 199, 1094-1095.

195 Konigsberg, W., Goldstein, J and Hill, RJ (1963) J BioI Chem 238, 2028-2033.

196 Sanger, F and Tuppy, H (1951) Biochem J 49, 481-490.

197 Pedersen, D.D (1977) Acta Chern Scand B31, 149-156.

198 Ryle, A.P., Leclerc, J and Falla, F (1968) Biochem J 110,4P.

199 Bang-Jensen, B., Foltmann, B and Rombauts, W (1964) e R Trav Lab Carlsberg 34, 326-345.

200 Mains, G., Takahashi, M., Sodek, J and Hofmann, T (1971) Can J Biochem 49, 1134-1149.

201 Tsuru, D., Hattori, T., Yamamoto, T and Fukumoto, J (1969) Agr BioI Chern 33, 1419-1426.

202 Rickert, W (1970) e R Trav Lab Carlsberg 38, 1-17.

203 Sternberg, M (1972) Biochim Biophys Acta 285, 383-392.

204 Williams, D.e., Whitaker, J.R and Caldwell, P.V (1972) Arch Biochem Biophys 149, 52-61.

205 Powers, J.e., Harley, A.D and Myers, D.V (1977) in: Acid Proteases (Tang, J., Ed.), pp 141-157, Plenum Press, New York.

206 Schechter, I and Berger, A (1966) Biochem Biophys Res Commun 27, 157-162.

207 Lowbridge, J and Fruton, J.S (1974) J BioI Chern 249,6754-6761.

208 Fruton, J.S and Bergmann, M (1938) Science (Washington, D.e.) 87, 557.

209 Fruton, J.S (1979) Ann N Y Acad, Sci 325, 1-18.

210 Dekker, CA, Taylor, S.P and Fruton, J.S (1949) J BioI Chern 180, 155-173.

211 Baker, L.E (1951) J BioI Chern 193,809-819.

212 Clement, G.E., Rooney, J., Zakheim, D and Eastman, J (1970) J Am Chern Soc 92, 186-189.

213 Tang, J (1965) J BioI Chern 240, 3810-3815.

214 Inouye, K., Voynick, I.M., Delpierre, G.R and Fruton, J.S (1966) Biochemistry 5, 2473-2483.

215 Trout, G.E and Fruton, J.S (1969) Biochemistry 8, 4183-4190.

216 Fruton, J.S (1973) in: Peptides 1972 (Hanson, H and Jakubke, H.D., Eds.), pp.355-362, North-Holland, Amsterdam.

217 Sachdev, G.P and Fruton, J.S (1969) Biochemistry 8, 4231-4238.

218 Tikhodeeva, A.G., Rumsh, L.D and Antonov, V.K (1975) Bioorg Khim 1,993-994.

219 Deyrup, e and Dunn, B.M (1983) Anal Biochem 129, 502-512.

220 Inouye, K and Fruton, J.S (1967) J Am Chern Soc 89, 187-188.

221 Rumsh, L.D., Tikhodeeva, A.G and Antonov, V.K (1974) Biokhimiya (Moscow) 39, 899-902.

222 Inouye, K and Fruton, J.S (1968) Biochemistry 7, 1611-1615.

223 Reid, TW and Fahrney, e (1967) J Am Chern Soc 89, 3941-3943.

224 May, S.W and Kaiser, E.T (1969) J Am Chern Soc 91, 6491-6495.

225 Reid, TW., Stein, T.P and Fahrney, D (1967) J Am Chern Soc 89,7125-7126.

226 Levchuk, T.P and Orekhovich, V.N (1963) Biokhimiya (Moscow) 28, 1004-1010.

227 Huang, W.Y and Tang, J (1969) J BioI Chern 244, 1085-1091.

228 Sepuvelda, P., Jackson, K.W and Tang, J (1975) Biochem Biophys Res Commun 63, 1106-1112.

229 Raymond, M.N., Bricas, E., Salesse, R., Garnier, J., Gamot, P and Ribadeau Dumas, B (1973)J Dairy Sci 56,419-422.

230 Raymond, M.N., Gamier, J., Bricas, E., Cilianu, S., Blasnic, M., Chaix, A and Lefrancier, P (1972) Biochimie 54, 145-154.

231 Visser, S., van Rooijen, PJ., Schattenkerk, e and Kerling, K.E.T (1976) Biochim Biophys Acta 438, 265-272.

232 Voynick, I.M and Fruton, J.S (1971) Proc, Natl Acad Sci (U.S.A.) 68, 257-259.

233 Dorer, F.E., Lentz, K.E., Kahn, J.R., Levine, M and Skeggs, L.T (1978)J BioI.Chern 253, 3140-3142.

234 Skeggs, L.T., Lentz, K.E., Kahn, J.R and Hochstrasser, H (1968) J Exp Med 128, 13-34.

235 Haber, E and Burton, J (1979) Proc Fed Am Soc Exp BioI 38,2768-2773.

235a Boger, J., Lohr, N.S., UIm, E.H., Poe, M., Blaine, E.H., Fanelli, G.M., Lin, T.Y., Payne, L.S., Schorn, T.W., LaMont, B.I., Vassil, T.e., Stabilito, 1.1., Veber, D.F., Rich, D.H and Bopari, A.S (1983) Nature (London) 303, 81-84.

236 Morihara, K (1974) Adv Enzymol 41, 179-243.

237 Morihara, K and Oka, T (1973) Arch Biochem Biophys 157,561-572.

238 Oka, T and Morihara, K (1974) Arch Biochem Biophys 165,65-71.

239 Hofmann, T and Hodges, R.S (1982) Biochem J 203, 603-610.

240 Hein, G.E and Niemann, e (1962) J Am Chern Soc 84,4495-4503.

241 Bender, M.L and Kezdy, J (1965) Annu Rev Biochem 34,49-76.

242 Fruton, J.S (1974) Ace Chem Res 7,241-246.

243 Fruton, J.S (1976) Adv Enzymol 44, 1-36.

244 Medzihradszky, K., Voynick, I.M., Medzihradszky-Schweiger, H and Fruton, J.S (1970) Biochemistry

9, 1154-1162.

245 Sachdev, G.P and Fruton, J.S (1970) Biochemistry 9, 4465-4470.

246 Sampath-Kumar, P.S and Fruton, J.S (1974) Proc Natl Acad Sci (U.S.A.) 71, 1070-1072.

247 Dobry, A., Fruton, J.S and Sturtevant, J.M (1952) J BioI Chem 195, 149-154.

248 Kozlov, L.V., Ginodman, L.M., Orekhovich, V.N and Valueva, T.A (1982) Biokhimiya (Moscow) 31, 315-321.

249 Fruton, J.S (1982) Adv Enzymol 53, 239-306.

250 Wasteneys, H and Borsook, H (1930) Physiol Rev 10, 110-145.

251 Folley, SJ (1932) Biochem J 26, 99-105.

252 Flosdorf, E.W (1941) Science (Washington, D.C.) 93, 157.

253 Determann, H., Bonhard, K., KOhler, R and Wieland, T (1963) Helv Chim Acta 45,2489-2509.

254 Determann, H., Heuer, J and Jaworek, D (1965) Justus Liebig's Ann Chem 690, 189-196.

255 Isowa, Y., Ohmori, M., Ichikawa, T., Kurita, H., Sato, M and Mori, K (1977) Bull Chem Soc Japan

50, 2762-2765.

256 PeUigrini, A and Luisi, P.L (1978) Biopolymers 17,2573-2580.

257 Morihara, K and Oka, T (1981) J Biochem (Tokyo) 89, 385-395.

258 Bozler, H., Wayne, S.I and Fruton, J.S (1982) Int J Pept Protein Res 20, 102-109.

259 Fruton, J.S (1957) Harvey Lect 51,64-87.

260 Neumann, H., Levin, Y., Berger, A and Katchalski, E (1959) Biochem J 73,33-41.

261 Fruton, J.S., Fujii, S and Knappenberger, M.H (1961) Proc Nat! Acad Sci U.S.A 47, 759-765.

262 Terada, S., Yoshida, S and Izumiya, N (1971) J Biochem (Tokyo) 70, 133-142.

263 Kitson, T.M and Knowles, J.R (1971) Biochem J 122,249-256.

264 Antonov, V.K., Rumsh, L.D and Tikhodeeva, A.G (1974) FEBS Lett 46, 29-33.

265 Silver, M.S., Stoddard, M and Kelleher, M.H (1976) J Am Chem Soc 98,6684-6690.

266 Richman, P.G and Fruton, J.S (1976) Proc Nat! Acad Sci (U.S.A.) 73, 3915-3917.

267 Wang, T.T and Hofmann, T (1976) Biochem J 153,691-699.

268 Newmark, A.K and Knowles, J.R (1975) J Am Chem Soc 97, 3557-3559.

269 Takahashi, M and Hofmann, T (1975) Biochem J 147, 549-563.

270 Wang, T.T and Hofmann, T (1977) Can J Biochem 55,286-294.

271 Wang, T.T and Hofmann, T (1976) Biochem J 153,701-712.

272 Silver, M.S and James, S.L.T (1981) Biochemistry 20, 3177-3189.

273 Zinchenko, A.A., Rumsh, L.D., Ginodman, L.M and Antonov, V.K (1981) Bioorg Khim 7, 1195-1200.

274 Silver, M.S and James, S.L.T (1983) Biochim Biophys Acta 743, 13-22.

275 Hollands, T.R, Voynick, I.M and Fruton, J.S (1969) Biochemistry 8, 575-585.

276 Jencks, W.P (1975) Adv Enzymol 43, 219-410.

277 Wang, T.T., Dorrington, K.J and Hofmann, T (1974) Biochem Biophys Res Commun 57, 865-869.

278 Hofmann, T., Hodges, RS and James, M.N.G (1984) Biochemistry 23, 635-643.

279 Lin, T.Y and Williams, H.R (1979) J BioI Chem 254, 11875-11883.

280 Cody, RJ., Burton, J., Evin, G., Poulsen, K., Herd, J.A and Haber, E (1980) Biochem Biophys Res Commun 97, 230-235.

281 Burgen, A.S.V., Roberts, G.C.K and Feeney, J (1975) Nature (London) 253, 753-755.

282 Hollands, T.R and Fruton, J.S (1969) Proc Nat! Acad Sci (U.S.A.) 62, 1116-1120.

283 Silver, M.S., Stoddard, M and Stein, T.P (1970) J Am Chem Soc 92, 2883-2890.

284 Kozlov, L.V (1974) Biokhimiya (Moscow) 39, 512-516.

285 Kirby, A.J., McDonald, RS and Smith, C.R (1974) J Chern Soc Perkin Trans 2, 1495-1504.

286 Dunn, B.M., Hofmann, T and Fink, A.L (1984) Biochemistry 23, 5241-5256.

287 Antonov, V.K., Ginodman, L.M., Rumsh, L.D., Kapitannikov, Y.V., Barshevskaya, T.N., Yavashev, L.P., Gurova, A.G and Volkova, L.I (1981) Eur J Biochem 117, 195-200.

288 Wolfenden, R (1972) Ace Chern Res 5, 10-18.

289 Westerik, J.O and Wolfenden, R (1972) J BioI Chern 247, 8195-8197.

290 Mattis, JA, Henes, J.B and Fruton, J.S (1977) J Bioi Chern 252,6776-6782.

291 Rich, D.H., Bematowicz, M.S and Schmidt, P.G (1982) J Am Chern Soc 104,3535-3536.

292 Holladay, M.W., Salituro, F.G., Schmidt, P.G and Rich, D.H (1985) Biochern Biophys Res Cornrnun 129,597-602.

293 Kaiser, E.T and Nakagawa, Y (1977) in: Acid Proteases (Tang),J., Ed.), pp 159-177, Plenum Press, New York.

294 Jones, M.N.G., Hsu, LN and Delbaere, L.T.J (1977) Nature (London) 267, 808-813.

295 Fruton, J.S (1984) Carlsberg Lab Cornrnun 49, 231-239.