A .V''
`
`;
`.
`! » > > . . . .
`:4;S'i!ll.ii';
`,^J. M
`Sf |1
`^r'j;
`. 1 IS
`' ; Si #•••:>•: 3M\
`
`' -A:^.
`
`.
`
`. . . .
`
`^ v i V
`J.!!
`
`KS I
`• i s I
`i
`1 P
`• Drugs of the future.
`*v. 24, no. 7 (Juiy 1999)
`Genera! Collection
`,,W1 DR897
`Received: 10-01-"1999
`
`p-
`
`f >"
`" •'"1 - »*>; .
`
`ZM
`
`<
`
`- &
`r i}
`"•* JfUr »
`V <• j-.
`T - - ^
`
`Ar"
`
`» l
`
`»
`
`i ^
`
`| f
`
`•2 ^
`
`^
`:s ;*
`v C '
`V + 1 ;
`-rre
`*•
`
`<
`
`*
`
`' %
`
`.i
`?fw' 'J&rt
`* i - .
`S3' >
`1 i
`
`1
`
`f
`
`I
`
`I-
`J,
`Jj
`1 * *' "*f •
`i
`•• £.\2f>U
`^ ' r . ' ®
`*
`* jr'
`
`- • r-
`cr.
`t.
`e^l
`1%^
`jwr
`•
`tr
`i«
`f
`i-j '-'"rr
`|
`1 ">->*' ^ ^"J.fr'-S'r
`WWmfi
`^
`•>
`^
`T
`^ .J
`
`i
`
`IT
`
`* "
`
`-»
`
`-a.
`
`*&
`
`^
`
`. . . .
`Js >. ^
`' ' V " ' ' '
`.-JL
`V ^|'v,r
`S I
`V61 u me" 24,^ N u mbeiv7fJ u ly ^1999
`r-v;
`
`1
`
`'*'1
`•«««w r
`
`jJv*1'
`
`t
`/ - ^
`\
`i
`
`s
`
`Is:!
`
`m
`
`M
`m
`pp • u
`
`v ^
`
`'. ' ^ " "
`DRUGS OF THE FUTURE
`
`?!« v»
`iff a *
`
`a
`
`-
`
`• A , 1 " '
`
`%
`
`S'
`
`I'I ' i' n
`
`i
`
`Os
`
`O ro
`o
`
`1
`i
`s
`
`£ 1999 VOLUME 24 ISSUE ?
`°
`SISAC
`o
`ro
`o
`
`8? (19
`
`: f; 1 - S
`
`OJ
`o
`
`a
`
`- i
`ift l ' < -
`
`<*v
`
`*
`
`1
`
`v
`
`~_-m
`
`Wl DR897
`r-rrng^ 5032533/01
`OOT
`
`IHIIJIW
`
`ryytt'-v* iji"-
`
`T
`
`V/v"
`•» w *'
`
`i . , • £ ' ^ ; ; xy^cjiVV'
`?
`
`
`
`^ ,-i» J-,'"
`
`«•
`
`^
`
`' •*
`«&
`V j w
`CP'
`
`•t *j
`I
`
`lM§jfS,
`I &•*' *A
`^ ^
`^ ^
`
`I- *
`
`NATIONAL
`LIBRARY OF
`MEDICINE
`
`OWMt IMtiruilt •! tiuilH
`
`. . . ,
`
`PROPERTY OF THE
`NATIONAL
`LIBRARY OF
`MEDICINE
`I
`
`x
`
`B ' "
`
`^ ' " X»
`^
`X T
`_
`f s . j * - i
`"'- 1 i,* 'J •>
`
`v.
`:f%v?v •-•
`^"w
`,/ r '* ;S
`
`\< -
`^ » -
`«i
`
`^
`w**
`
`t,
`
`- 4 \ ;
`,
`»' ri
`
`•
`
`4 f
`
`'m.
`*% V *V >•
`
`^ I1 V
`
`a
`
`«i
`
`•%
`
`V-'
`tv '''*
`t»f
`>
`
`"J
`
`'' ^
`
`,
`
`Cr
`
`tf
`
`Th - S't ftK/f S 9 I4?t.
`
`/
`
`i
`i
`
`a
`^ I
`
`'yc"
`i
`
`.
`
`MYLAN - EXHIBIT 1007
`
`

`
`i
`
`iru
`of the
`1
`
`Contents
`
`Volume 24, Number 7,
`
`1999
`
`Monographs
`irritable bowel syndrome.
`725 Dexloxiglumide. CCK1 (CCKA) receptor antagonist, Treatment of
`L. Revel, F. Makovec, J. Castaner
`729 Prucalopride. 5-HT4 agonist, Treatment of irritable bowel syndrome. X. Rabasseda, J. Castaner
`Ceccarelli
`735 Raxofelast. Treatment of diabetic complications, Antioxidant. S. Cuzzocrea, S.
`740 Robalzotan Tartrate Hydrate. Antidepressant, 5-HT1A antagonist. L.A. Sorbera, P. Leeson, J. Castaner
`747 Weige. Treatment of erectile dysfunction. X-W. Wang
`
`Review Articles
`Indole-containing thiazolidine-2,4-diones as
`751
`B.B. Lohray, V. Bhushan
`759 P2Y2 receptor agonists: structure, activity and therapeutic utility
`B.R. Yerxa, F.L Johnson
`771 Progress in the design of inhibitors of coagulation factor Xa
`W.R. Ewing, H.W. Pauls, A. P. Spada
`
`novel
`
`hypolipidemic agents
`and
`euglycemic
`
`1-23
`
`Information Update
`1263W94
`790
`790 Acreozast
`791 Azelnidipine
`791 Basiliximab
`792 Bucindolol Hydrochloride
`792 C-1311
`793 Celecoxib
`796 Ceterizine Hydrochloride
`797 Daptomycin
`798 DMP-754
`799 Ebastine
`800 Entacapone
`Fosphenytoin Sodium
`801
`Frovatriptan
`801
`Idoxifene
`804
`Idoxuridine
`806
`Imidapril Hydrochloride
`806
`Ipsapirone
`806
`807 KE-298
`Levofloxacin
`808
`
`808
`808
`809
`809
`810
`810
`810
`811
`815
`815
`820
`821
`821
`823
`825
`825
`827
`827
`829
`830
`830
`
`Metazosin
`MKC-442
`Moguisteine
`Nifedipine
`Oltipraz
`Pentostatin
`Phentolamine Mesilate
`Raloxifene Hydrochloride
`Ramoplanin
`Repaglinide
`Risedronate Sodium
`Ritonavir
`Rivastigmine
`Rizatriptan Benzoate
`Saredutant
`Thrombopoietin
`Tolteridone
`Triptolide
`Valaciclovir
`Vatanidipine Hydrochloride
`Zileuton
`
`This materisI was copied
`at the NLM and may be
`
`

`
`fi M
`
`Drugs of the Future is Published monthly. A Pharmacostructurai index appears annually in the December issue.
`Cumulative chemical
`formula and general indices are published
`
`separately every year.
`
`Subscription Rates
`Calendar year subscription price for Volume 24, 1999 is US $ 1015. Airmail additional (optional): America $ 70, Asia $
`120, Europe $ 40. Back volumes are available in book form, offering an Encyclopedia of drugs in various stages of
`
`research, continuously updated as new information becomes available. '
`
`Subscription Information/Sample Copies
`Remittances and orders for subscriptions, notices of change of address, claims for missing numbers and requests for
`sample copies should be sent to the Subscription Service Department, Prous Science, S.A., P.O. Box 540, 08080
`Barcelona, Spain. Street address: Provenza
`388,
`08025
`Barcelona.
`Tel. (34-93)
`459-2220.
`service® prous.es. Please indicate
`
`the code number from mailing
`labels
`in
`all
`
`correspondence relating to your subscrip­
`tion.
`
`Electronic Databases
`information from the print edition of Drugs of the Future is also available in electronic format: on
`Continuously updated
`the Prous Science Ensemble® database; on the CIPSLINE® diskette-based databases (for use with MDL Information
`Dialog
`Systems, Inc.'s
`
`
`for ISIS™/Base software PC® and Macintosh® computers);
`and
`online
`via
`file PRFU.
`
`file
`
`For more information on these databases, contact: Marketing Dept., Prous Science, S.A., P.O. Box 540, 08080 Barcelona,
`Spain. Tel +34 93 459-2220; Fax +34 93 458-1535; e-mail: service@prous.es or Prous Science,
`Inc., 12th Floor, East
`Tower, Centre Square, 1500 Market St., Philadelphia, PA 19102, U.S.A. Tel: (215) 246-3441; Fax: (215) 246-3496; e-mail:
`http://www.prous.com.
`proussci@aol.com or connect
`to
`
`Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by Prous
`
`
`is fee paid directly to Copyright Clearance Center
`
`(CCC), 222 Rosewood Dr.,
`Science, S.A., 0377-8282/99, provided that the appropriate
`Danvers, MA 01923, USA, 978-750-8400. For
`those
`
`that organizations have been granted a
`
`photocopy
`license
`by CCC,
`
`tem of payment has been arranged. Prior to photocopying items
`for educational classroom use,
`please
`contact
`CCC.
`Copyright ©1999, Prous Science,
`S.A.
`Publishers.
`Reproduction
`in
`whole
`or
`part
`publisher.
`
`a
`
`is
`
`Federation of Periodical Press.
`
`International
`FIPP Member of the
`
`APP Publication not subject to mandatory control of circulation because advertising
`
`pages
`
`
`
`
`
`less are 10% of the
`
`
`
`
`
`
`
`than total number pages. of
`
`reasonable
`a
`The Editor has made
`ISSN: 0377-8282 - Dep6sito Legal: B-154-1.976
`
`and complete does not assume any
`
`supply
`to
`effort
`
`
`•
`Printed in Spain. • Romargraf, S.A. - Hospitalet de
`Llobregat
`
`for
`liability
`(Barcelona)
`
`
`
`accurate errors
`
`or
`
`
`
`information, omissions.
`
`This material was copied
`at tfie NLM and may be
`
`

`
`Drugs of the Future 1999, 24(7): 771-787
`SCIENCE
`Copyright © 1999 PROUS
`CCC: 0377-8282/99
`
`inhibitors
`Progress in the design of
`of coagulation factor Xa
`
`Review Article
`
`William R. Ewing', Henry W. Pauls and
`Alfred P. Spada
`Rhone-Poulenc Rorer, Department of Medicinal Chemistry,
`500 Areola Road, Collegeville, PA 19426, USA.
`'Correspondence
`
`B-
`
`CONTENTS
`
`Introduction
`Background
`Factor Xa structure and function
`Affinity labels
`Transition state analogs
`Peptide inhibitors
`inhibitors . . . .
`Bis-benzamidine
`Benzamidine inhibitors
`inhibitors . . . .
`Nonbenzamidine
`Conclusions
`References
`
`Introduction
`
`771
`771
`774
`775
`775
`776
`777
`779
`782
`783
`783
`
`>
`
`p*
`
`Myocardial infarction, stroke, deep vein thrombosis
`and pulmonary embolism accounted for approximately 2
`million deaths in the United States in 1996 (1). The for­
`related
`is causally
`thrombus
`mation of an occlusive
`pathology of these conditions. As such, antithrombotic
`therapy is a crucial component in both acute intervention
`procedures and chronic prevention strategies for treat­
`ment and management of these diseases.
`requires
`therapy often
`Effective antithrombotic
`istering a combination of antiplatelet and anticoagulant
`agents. Aspirin and heparin represent the current main­
`of coronary
`treatment
`for the
`stay of combination therapy
`syndromes (2). The shortcomings of each have driven
`industry
`the pharmaceutical
`intense efforts within
`tify and develop more effective and safer antiplatelet and
`anticoagulant agents. Significant advances have been
`made in the development of more potent and effective
`antiplatelet agents (3, 4); however, the clinical need for
`improved anticoagulant agents is arguably greater.
`Current treatment options are limited to unfractionated
`heparin (UFH), low molecular weight heparins (LMWHs)
`and warfarin. The challenge remains to achieve consis­
`tent, predictable and clinically effective levels of anticoag­
`ulation while minimizing the risk of bleeding complica­
`tions.
`Recent approaches to identify anticoagulant agents
`focused
`with improved safety and efficacy have
`
`to
`
`on
`
`oping specific inhibitors of enzymes within the coagula­
`extensively
`the most
`remains
`tion cascade (5). Thrombin
`of experience
`targets. Two decades
`investigated of these
`in the design of thrombin inhibitors has led to the devel­
`opment of highly potent and selective inhibitors, several
`of which have been investigated in large-scale clinical
`studies (6-10). In general, the results of these studies
`have fallen short of expectations, demonstrating no clear
`of coronary
`treatment
`the
`in
`advantage over heparin
`have
`results
`dromes (11). More encouraging clinical
`obtained in the treatment of venous thromboembolism
`(12).
`Factor Xa inhibitors, the subject of this review, repre­
`approach
`rapidly evolving
`sent a more recent and
`the development of anticoagulant agents. Theoretically,
`direct inhibition of factor Xa activity should provide a
`limiting
`potentially
`potent anticoagulant devoid of the
`inhibitors
`effects observed with thrombin
`(5, 13).
`
`toward
`
`side
`
`been
`
`the
`to
`Background
`
`factor
`
`To develop superior anticoagulant strategies requires
`an understanding of the biochemical and biophysical
`admin­
`coagulation.
`blood
`regulating
`mechanisms activating and
`requires maintaining
`The process of normal hemostasis
`the dynamic processes of pro-
`delicate balance between
`and anticoagulant activities in circulating blood. Both are
`highly integrated cascade
`finely tuned and
`governed by a
`iden­
`vas­
`to
`that amplify the response
`of enzymatic processes
`cular injury preventing hemorrhage and initiating the
`the
`repair processes. An abbreviated scheme highlighting
`Figure
`in
`key aspects of these processes is presented
`Two convergent procoagulant pathways have
`evolved, each capable of being activated in response to
`different stimuli. The intrinsic and extrinsic coagulation
`pathways consist of a cascade composed of serine pro­
`stimulus
`initial
`the
`teases which effectively amplify
`vide a strong and rapid signal to initiate the coagulation
`process (14). These pathways ultimately converge upon
`to
`conversion
`its
`the formation of factor X and
`the prothrombinase complex.
`devel-
`
`This material wascopied
`at the NLM and may
`Subject US Copy right Laws
`
`be
`
`J
`
`

`
`772
`
`Factor Xa inhibitors
`
`Coagulation Palhways
`
`Regulatory Palhways
`
`Intrinsic pathway:
`P'XIla
`
`h
`
`FVIIIa/FIXa
`Caw , PL
`
`FIX
`
`Extrinsic pathway:
`Vascular insult
`
`H
`
`FVIIa/TF
`Ca++, PL
`
`FV
`
`FVIU
`
`FX
`
`1
`Ca",PL
`F Xa 1'Va
`
`prothroinbin
`
`FVa
`
`FVIIla
`
`aPC, protein S,TM
`
`r
`
`FVIII
`
`FV
`
`^ Prolcin C,
`thrombonKxIulin
`
`Tlirombin
`
`Fibrin
`
`Fibrinogen
`
`Plaiolet activation
`
`FXIII
`
`FXlIIa
`
`Stabilized clot
`
`of enzymatic processes procoagulant and anticoagulant activities
`
`
`of the
`Fig. 1. Schematic diagram
`PL = phospholipid, IF = tissue factor, TM =
`thrombomodulin,
`aPC
`=
`activated
`
`in
`
`circulating
`protein
`
`C.
`
`blood.
`
`a
`
`The intrinsic pathway is stimulated by the contact of
`flowing blood with foreign surfaces or upon vascular
`injury and exposure of subendothelial matrix collagen
`(15). Three circulating factors, high molecular weight
`kininogen, prekallikrein and factor XII, bind together on
`the surface to yield the catalytically active serine pro­
`tease, factor XI la. This branch of the coagulation cascade
`ultimately produces
`factor IXa. Factor IXa combines with
`
`factor Villa, phospholipid, Ca2+ and factor X in the intrin­
`sic Xase complex
`to
`afford
`
`factor Xa.
`The extrinsic pathway is initiated following vascular
`injury and the resulting exposure of tissue
`factor (TF) onto
`the surface of endothelial cells and macrophages
`(16, 17). Tissue factor binds to factor VII and catalyzes its
`
`conversion to factor Vila. This complex, in the presence
`of Ca2+ and phospholipid, converts
`factor X to factor Xa.
`Factors XI and IX can also be activated by tissue
`factor/factor Vila, thus providing a link with the intrinsic
`pathway.
`Factor Xa, generated by either pathway, combines
`with the nonenzymatic cofactor Va and Ca2+ on the phos­
`
`pholipid surface of platelets or endothelial cells to form
`the prothrombinase complex. The
`catalytic
`activity
`tor Xa in this complex is increased 300,000-fold relative to
`its activity in solution. Factor Xa in the prothrombinase
`complex converts prothrombin
`(factor
`II) by
`olysis to release catalytically active thrombin (factor I la)
`(18).
`control
`in
`central position
`Thrombin holds a
`lation processes. It catalyzes the cleavage of fibrinogen
`to fibrin, thus initiating the process of clot formation and
`
`to
`activates factor XIII which cross-links fibrin monomers
`to
`stabilize the developing clot. Thrombin binds avidly
`rin and remains catalytically active within the growing
`thrombus (19). Thrombin also catalyzes
`the formation of
`cofactors Va and Villa
`and
`thus
`provides
`its continued generation and the pro­
`back mechanism
`for
`gression of clot development
`(20).
`In
`addition
`actions on the coagulation cascade, thrombin is also a
`potent agonist of platelet activation. The
`direct activation
`of platelets is an
`important
`first
`response
`to
`
`mal hemostasis, allowing
`the
`
`generation a homeostat-of
`ic plug following injury (20). Prolonged activation and
`recruitment of platelets will
`further
`
`potentiate the develop­
`ment of clot formation.
`Thrombin also plays a key role in the initiation of
`inhibitory pathways to downregulate the coagulation
`process. Upon release from the prothrombinase complex,
`thrombin can bind to thrombomodulin, a glycoprotein pre­
`sent on the surface of endothelial cells
`
`(22, 23). Thrombin
`has high affinity
`
`for thrombomodulin which alters its sub­
`strate specificity from fibrinogen and the procoagulant
`C.
`Cleavage
`factors V, VIII and XIII to protein
`of
`fac­
`protein C. This
`bound to thromomodulin affords
`activated
`complex, in the presence of the nonenzymatic cofactor
`protein S, cleaves the procoagulant cofactors Va and
`limited prote­
`Villa, thus providing a mechanism ot inactivating the
`intrinsic coagulation pathway. In addition, active protein C
`activates the fibrinolytic system by stimulating the
`release
`of coagu­
`of tissue plasminogen activator from endothelial cell,
`which provides a physiologically important link between
`anticoagulant and
`fibrinolytic pathways
`(24).
`
`fib­
`
`maintain
`
`of
`
`This materia I was copied
`3t ttie NLM and may be
`Subject US Copyright Laws
`
`

`
`Drugs Put 1999, 24(7)
`
`773
`
`Tissue factor inhibitor controls the regulation of the
`Low molecular weight heparins
`(LMWHs) offer distinct
`extrinsic pathway. This protein
`is an active-site inhibitor of
`advantages when compared with UFH (36, 37). They pro­
`factor Xa that binds with
`tissue
`factor/factor Vila
`to
`form
`a
`vide predictable and well-controlled anticoagulant
`quaternary inhibitor complex (25, 26). Tissue factor
`response with fixed-dose administration. The bioavailabil­
`inhibitor is synthesized by endothelial cells of which
`ity and half-life of LMWHs following s.c. administration
`approximately 80% remains associated with
`the
`endothe­
`are generally good, allowing for once- or twice-daily dos­
`lial surface (25).
`ing (38). Combined, these attributes enable LMWHs
`to be
`Antithrombin III (ATIII) is an important circulating
`used safely outside of the hospital setting and avoid the
`endogenous inhibitor of coagulation that acts in addition
`need for continuous patient monitoring. Additionally, one
`to the two dynamic regulatory mechanisms discussed
`LMWH, enoxaparin, has demonstrated superior efficacy
`above. ATIII inhibits plasma
`thrombin,
`factor
`Xa and, to a
`when compared with heparin
`in several clinical trials (39,
`lesser extent, IXa activities. ATIII binds irreversibly to
`40). It is interesting to note
`that
`not
`all
`LMWHs
`these serine proteases to form 1:1 inhibitory complexes.
`same level of efficacy when compared with standard
`The rate of this inhibition, which is relatively slow other­
`heparin (37, 41). This may be the result of different
`wise, is accelerated 4000-fold upon binding with the gly-
`anti-Xa/anti-lla ratios, ranging from 1.9:1 to 5.0:1,
`cosaminoglycan, heparin (27). Heparin, which is synthe­
`depending upon
`the specific product
`preparation
`(37,
`sized in the liver, is an integral cell surface component
`42). As with heparin, however, LMWHs do
`not
`inhibit
`clot-
`which is exposed to flowing blood on the surface of
`bound thrombin or platelet bound Xa nor are oral formu­
`endothelial cells. Therefore, under conditions of normal
`lations available (36).
`hemostasis, ATIII is bound and activated by heparin on
`Warfarin is the only orally active anticoagulant cur­
`the endothelial cell surface as well as within the suben-
`rently available. It inhibits the vitamin K-dependent con­
`dothelium, sequestering and inactivating circulating pro-
`version of glutamic acid to gamma-carboxyglutamic acid
`coagulant enzymes
`thrombin and
`factor Xa (28).
`residues of the procoagulant factors
`II, VII, IX and X and
`In clinical practice, the use of unfractionated heparin
`anticoagulant proteins C and S. This posttranslational
`(UFH) has been a part of standard anticoagulant therapy
`modification, which occurs
`in
`liver, is essential for the
`the
`for several decades (29). The ATlll-heparin complex
`Ca2+ and membrane binding properties of these proteins
`effectively neutralizes soluble thrombin but, owing to the
`without which catalytically effective complexes cannot be
`size of the formed complex, it is incapable of inhibiting
`formed. The onset of action of warfarin is slow and often
`thrombin activity once bound to fibrin in the growing
`accompanied by a paradoxical increase in coagulation
`thrombus (19). This physiochemical
`limitation
`significant­
`activity (43). This reflects
`the shorter half-lives
`of the
`ly restricts the ability of heparin to maximally inhibit the
`coagulant proteins C (6 h) and S
`(30
`h) as
`compared
`actions of thrombin. It is estimated that approximately
`the half-lives of factors II (60 h), VII (5 h),
`IX (24
`and X
`h)
`40% of thrombin generated within the developing throm­
`(40 h) (44). As a direct consequence of this mechanism
`bus remains bound
`to the
`clot and is therefore resistant
`to
`of action, it may take 72-96 h to achieve peak anticoagu­
`inhibition by heparin-ATIII (30). In addition, thrombin
`lant activity. Warfarin possesses a high oral bioavailabili­
`bound to soluble fibrin
`fragments produced
`during
`throm­
`ty (~ 100%) with a long terminal half-life
`1 week). As a
`(~
`bolysis is also protected from inhibition but still remains
`result, the anticoagulant activity of warfarin may continue
`capable of activating platelets and coagulation (31, 32).
`for several days following the termination of treatment.
`Heparin therapy is associated with a number of
`Warfarin has a narrow therapeutic margin with many
`well-recognized
`risks and disadvantages, several of
`known drug-drug interactions which
`which are highlighted below (29). Firstly, the anticoagu­
`requires
`that
`be treated individually and continually monitored for
`lant response to heparin varies widely among patients. As
`coagulant activity (45).
`a result, treatment with heparin requires close monitoring
`to maintain effective and safe levels of anticoagulant
`Numerous preclinical and clinical studies have
`activity. The lack of predictability and the need
`demonstrated that direct low molecular weight thrombin
`for contin­
`uous monitoring prevents the use of UFH outside of the
`inhibitors are capable of effectively blocking clot-bound
`hospital setting. Secondly, it is estimated that approxi­
`thrombin activity (46). Therefore, superior efficacy
`should
`mately 3% of patients treated with UFH will develop a
`be expected from direct acting inhibitors relative to
`severe and potentially
`heparins (11). The results of clinical trials with these
`life-threatening
`form of
`heparin-induced thrombocytopenia (33). Lastly, a pro-
`agents, however, have generally been disappointing and
`thrombotic "rebound" phenomenon has been observed
`portend several potential limitations associated with this
`following the termination of heparin treatment, which
`approach (11, 47). Two large-scale studies conducted
`results in recurrent episodes of unstable angina
`with hirudin, GUSTO-lla and TIMI-9A, were terminated
`(34).
`In
`a
`separate study, it was demonstrated that this reactivation
`due to excessive cranial bleeding (48, 49). Both studies
`phenomenon is the result of increased thrombin activity
`were reinitiated (GUSTO-llb and TIMI-9B) with lower
`following termination of heparin
`infusion
`The precise
`(35).
`doses of hirudin and heparin (50, 51). Unfortunately,
`mechanism for the rebound phenomenon is
`not clear but
`hirudin did not demonstrate a statistical improvement
`it has been suggested that continued thrombin genera­
`over standard heparin therapy in either of these studies at
`tion, along with other procoagulant factors, during the
`their respective 30-day combined primary endpoints.
`course of treatment may play a role (35).
`Clinical data obtained with inogatran and argatroban
`
`patients
`
`>
`
`>
`
`>
`
`>
`
`>
`
`!>•
`
`This material was co-pied
`at the NLM and may be
`Subject US Copyright Laws
`
`

`
`774
`
`Factor Xa inhibitors
`
`Table I: Potential advantages
`factor Xa inhibitors.
`
`of
`Inhibition of the source of thrombin generation rather than its catalytic activity
`No direct effect upon
`thrombin-activated
`
`platelet aggregation, thus minimizing
`No direct effect on thrombin-mediated generation
`of aPC-via
`flla/TM
`Minimum risk of thrombotic reactivation
`rebound
`and
`associated
`
`risk
`
`bleeding
`complex
`ischemic
`events
`
`have been similarly disappointing (52-54). In addition,
`same level of maturity as has thrombin inhibition.
`cessation of treatment with either agent was associated
`Although academic interest in the inhibition of factor Xa
`coronary
`with thrombotic rebound which
`led
`to
`increased
`has a long history, drug
`companies
`had
`not
`events (52, 55). At the present time, hirudin
`is approved
`challenge in force until the early part of this decade.
`
`
`of in Europe for the prevention deep vein thrombosis
`fol­
`Given their similarities, some
`of
`the
`design
`lowing hip and knee replacement surgery (56).
`in the development of thrombin inhibitors were adapted
`Argatroban is approved in
`Japan
`
`for peripheral vascular
`and incorporated into the design of factor Xa inhibitors.
`disease and acute cerebral infarction (54).
`However, in contrast to thrombin, small molecule
`The design of direct-acting thrombin inhibitors has
`inhibitor/factor Xa X-ray structures have been published
`been the subject of several recent reviews (8, 10). Many
`on only two occasions, the first one appearing in 1996
`highly potent and selective inhibitors have been
`(73). Thus, structure information has only begun to have
`described. However, until recently, combining these
`an impact on the design of factor Xa inhibitors as com­
`essential features
`into
`inhibitors with strong oral pharma­
`pared to thrombin inhibitors (8). Nonetheless, much
`cokinetic properties has
`remained elusive
`(57).
`progress has been made in recent years and will be dis­
`Sanderson et al. recently reported a series of potent
`cussed from an historical perspective using representa­
`pharmacokinet­
`thrombin inhibitors with very encouraging
`tive examples.
`ic properties (58). This may herald an important advance
`in the development of safe and effective oral anticoagu­
`
`lants to replace warfarin. Although there are many earlier
`reports of orally active thrombin inhibitors, the actual
`weak.
`rather
`pharmacokinetic profiles of
`these agents
`are
`Human factor Xa consists of two chains (62): a
`
`sug­
`data
`This may be particularly
`important
`since
`clinical
`chain which incorporates the catalytic triad (63) and a
`gests that long-term and/or prophylactic anticoagulant
`light chain which contains a chymotrypsin cleavage
`site,
`therapy can provide a significant benefit over current
`Tyr44-Lys45 (64). Structural
`
`information for factor Xa was
`standard treatment (59).
`first obtained by Tulinsky et al. (65) on a
`large molecular
`Direct inhibition of factor Xa activity prothrombi-in
`
`
`fragment of the enzyme generated by autocatalysis dur­
`nase complex blocks the single physiological source of
`ing crystallization studies. The des(N-terminal 1-45) fac­
`thrombin generation. Inhibiting the source of thrombin
`
`tor Xa X-ray structure was solved to 2.2 A resolution and
`generation rather than its catalytic activity offers several
`revealed that the side chain of the C-terminus Arg439
`supe­
`potential mechanistic advantages
`that
`could
`afford
`was inserted into the S1 specificity site of a neighboring
`rior anticoagulant agents. Direct inhibition of factor Xa
`molecule of factor Xa. This ruled out small molecule
`activity should have minimal impact upon normal hemo­
`inhibitor work using classical soaking techniques. It is
`static response/regulation processes. Platelets,
`for exam­
`also worth noting that
`this
`form
`of the
`enzyme
`ple, would remain responsive to
`
`the low levels of catalyt-
`ciated factor Va, which is necessary for large rate
`ically active thrombin and thus, the formation of platelet
`enhancements observed
`(18).
`However,
`in vivo
`hemostatic plugs would not be compromised. As a con­
`ture was useful for modeling studies and in fact several
`sequence, the risk of bleeding complications might be
`groups took advantage of this approach for inhibitor
`minimized. The endogenous pathway to downregulate
`design (see below).
`thrombin production via the thrombin/thrombomodulin
`The catalytic triad residues, Ser195, His57 and
`complex would also remain intact. The risk of provoking
`Asp102, provide the catalytic machinery
`for the
`hydrolysis
`prothrombotic rebound episodes observed with heparin
`of the peptide linkage of the substrate.
`The
`mechanism
`and thrombin inhibitors would be minimized as well
`this process has been extensively studied for serine pro­
`(Table I).
`teinases in general (66, 67). Initial binding between
`The prospects for factor Xa inhibitors in therapy, espe­
`
`enzyme and substrate is
`followed by nucleophilic attack
`cially in contrast to thrombin inhibitors, were examined
`on the amide carbonyl
`to form a tetrahedral
`
`intermediate
`previously in this journal (5). The most recent compre­
`(Scheme 1 A). Collapse of this intermediate gives
`the
`acyl
`hensive review of factor Xa inhibitors focused on the
`enzyme and liberates the amine and subsequent hydrol­
`to
`identifying
`potential of a combinatorial approach
`factor
`ysis yields the acid.
`Interaction with
`the
`nucleophilic
`Xa inhibitors (60). Factor Xa inhibitors have also been
`ponents of the triad, i.e., the serine hydroxyl and the his-
`included in the Annual Reports of Medicinal Chemistry
`since 1995 (61). Notwithstanding the potential advan­
`tidine imidazole, has been a general approach for the
`has
`not
`tages, the field of factor Xa inhibition
`reached
`the
`inhibition of serine proteinases
`(68).
`
`features
`
`heavy
`
`lacks asso­
`
`the
`
`Factor Xa structure and function
`
`the
`
`This matariaI was copied
`at the MLM and may ba
`Subject US Copyri^it Laws
`
`

`
`Drugs Fut 1999, 24(7)
`
`775
`
`Transition state analogs
`
`of the
`the dansyl group fills the S4 pocket and the specificity
`pocket is occupied by
`
`the arginine side chain (65). Given
`the inherent reactivity of affinity labels in general, this
`approach is out of favor as a viable strategy for drug
`design.
`
`the middle
`The p-sheet, which stretches across
`Xa active site, is also found in thrombin and trypsin.
`Proteinaceous and peptide-based substrates and
`inhibitors of factor Xa often bind in an extended confor­
`mation. In this standard binding mode the peptide back­
`bone engages in H-bonding
`
`interactions with the Gly216
`residue of the p-sheet (69). Tick anticoagulant protein
`
`(TAP) appears to be an exception (70).
`trypsin-like
`The specificity or S1 binding pocket of the
`serine proteinases is one of the prime determinants of
`Another strategy for the inhibition of serine proteinas­
`factor Xa, the S1 pocket is a nar­
`substrate specificity. In
`es has been to replace the scissile bond of a truncated
`
`row cleft with an aspartate residue at its base. Except for
`peptidyl substrate with electrophilic functions such as
`small changes, this
`feature
`is also
`found
`in
`related
`serine
`boronic acids, aldehydes or activated ketones (81). The
`proteinases such as trypsin (71) and thrombin (72) and,
`covalent adduct which results upon incubation with
`as such, presents a challenge for the design of specific
`enzyme, although reversible, is tightly bound because it
`inhibitors. The aromatic or S4
`of
`binding
`
`pocket factor Xa
`mimics the transition state formed during catalysis
`is one of its unique structural
`
`features. The floor and walls
`(Scheme 1C). Proteinaceous substrates for factor Xa
`of this open-ended box are defined by two electron-rich
`bound in the prothrombinase complex include prothrom­
`aromatic side chains (Trp215, Tyr99) and Phe174. The
`bin (82) and factor V and several inhibitors such as antis-
`terminus of the box is lined with H-bond acceptors, i.e.,
`
`tasin (83), AcAP-6 (84) and tissue factor pathway inhibitor
`the carbonyl functions of Lys96 and Glu97 which bind a
`(85). The active site sequences of these ligands have
`structural water and an acidic side chain, located at the
`been summarized (86). Except for AcAP-6, all incorporate
`periphery (Glu97). This unique
`juxtaposition of functional­
`a P1 arginine and are thought to inhibit factor Xa in the
`ity, termed the "cation hole", has been
`invoked to explain
`extended substrate conformation. AcAP-6 has a pheny­
`the potency of inhibitors with positively charged moieties
`lalanine P1, suggesting a
`tolerance for aromatic rings in
`at the putative P4 or P3 position (73).
`the S1 subsite. The P4 residue is usually hydrophobic
`Given the apparent difficulty of obtaining small mole­
`except for factor V which contains a
`lysine.
`cule/factor Xa crystal structures, various groups have
`A number of laboratories have taken this approach for
`resorted to using the related serine proteinases trypsin
`the design of factor Xa inhibitors and representative
`(74, 75) and thrombin (76) as surrogates for factor Xa.
`examples are given below. Several of the first transition
`This is possible since the factor Xa inhibitors of interest
`state analog inhibitors of factor Xa described incorporat­
`often retain some
`level of activity against
`these
`enzymes.
`ed arginine at P1 and aromatic residues at P2 and P3.
`Trypsin has been used more frequently in this effort due
`Aldehydes 1 (87) and boronic acids 2 (86) are effective
`to its availability and higher degree of homology to factor
`inhibitors of factor Xa. A series of peptidyl
`inhibitors
`Xa (77). Even so, differences exist
`in
`both
`the
`S1 and
`the
`been described which
`incorporate arginines at P1 and P3
`S4 pockets. The most significant difference in SI is an
`and which utilize ketoamides 3 (88) or ketothiazoles 4
`Ala190 (factor Xa) to Ser190
`(trypsin)
`variation.
`This
`sug­
`(89) as serine traps.
`
`
`the In general, presence of the ser­
`gests that in factor Xa the S1 subsite is somewhat more
`ine trap has a profound impact on the activity of these
`hydrophobic than the trypsin S1 pocket. The differences
`peptide derivatives and underscores the potential for
`in S4 are the most profound
`involving Phe174 and
`Tyr99
`potency enhancements
`in
`
`trypsin-like serine proteinases
`(factor Xa) to Glu174 and Ile99 (trypsin) mutations,
`via catalytic triad
`interactions
`(90).
`respectively. Although the possibility for strong 7t-cation
`Assuming that interaction of the electrophilic function
`interactions (78) are greatly diminished in trypsin