PER\OOlCAl ROOM
`
`CSL EXHIBIT 1054
`CSL v. Shire
`
`Page 1 of 17
`
`

`

`Journal of Molecular Biology
`
`Editor-in-Chief
`P. Wright
`Depart.mfl nt of Molecular .Biology, Research Institute of Scripps Clinic·
`10661i N. T<Jrrey Pines Road , La J olla, CA 92037, U$.A .
`
`Assistant Editor
`J. Karn
`MRC Laboratory of Molecular Biology
`Hills Road , Cambridge CB2 2Q H, U.K.
`
`Founding Editor
`Sir J ohn Kendrew
`
`Consulting Editor
`Sydney Brenner
`
`Editors
`f'. Chambon. Laboratoir-e tie Genetique Molecula ire des Eucaryotes du CNRS, In:;titut de Chimie Biologique,
`Facultk de Medecine, II Rue Humann . 67085 Stra.~bourg Cedex. France.
`M. Gotle/fman. I nstitute of Cancer Research, College of Physiciuns & Sur·geons of Columbia University.
`701 W. I 68th Street, New York, NY 10032, U.S.A.
`P. von H ippeJ., I nstitute of Molecular· Biology, University of Oregon, Eugene, OR 97403- 1229, U.S.A .
`R. H1tber, Max-P ianck-l nstitut fi.ir Biochemic, 8033 Martinsried bei Miinchen, Uermany.
`A. Klug, MRC Laboratory of Moleculnr Biology, H ills Road , Cambridge CB2 2QH , U.K.
`
`Associate Editors
`C. R. Can/or , Human Genome Center, Donner Laboratory, Lawrenc·e Berkeley Laboratory, Univer·sity of California .
`Berkeley, CA 94720, U.S.A.
`N.-H. Ch-ua. The Rockefeller University , 1230 York Avenue. New York , N Y 1002 1, U.S.A.
`F. E. Cohen. Department of Pha.rmaceutieal Chemistry, School of Pharmacy. niversity of Ca:IiTornia. Ran Francis{'O,
`CA 94143-0446, U.S.A.
`JJ. J . Dt>Ro~tier. Rosenstiel Basic Med ical Sciences L"tesearch Center, Brandeis University, Waltham , MA 02254, L .S.A.
`A. R . Persht, University Chemical Laboratory. Cambridge U niversity, Lensfield Road . Cambridge CB2 l EW, U.K .
`W. A.. Hendrickson, Department of Biochemistr-y & Molecula r Biophysics. College of Physicians & Surgeons of
`Columbia University. 630 West 168th Street, New York. l'rY 10032. U.S. A.
`1.8 . Holland, Institute d.e Genetique et Microbiologic, Biitiment 409, Universitk de Paris Xl. 9J40!l Orsay Cedex 05,
`France.
`I:J. Honig, Department of Biochemistry & Molecular Biophysics, College of l'hysi('ians & :Smgeons of Columbia
`Uni versity, 630 West l68th Street. New York, NY 10032, U.S.A.
`V. Luzzali. Cent re de Genetique Moleculaire, CentrP Na.tional de Ia. Recherche Soientifique. 9 1 Gif-8ur-Yvette. France.
`J . L . Ma.ndel, La boratoire de Geruh,ique Moleculuire des Eucaryotes du CNRS, Institut- de Chimie Biologique.
`Faculte de Medecine, ll Rlle Humann. 67085 Strasbourg Ceclex. France.
`B. MattlumNJ, Institute of Molecula.r Biology, University of Oregon , Eugene. OR 9740:3- 1229, U.S.A.
`J. H. Miller, Department of Microbiology. University ofCtolifornia, 405 H ilgard Avenue, Los Angeles, CA 90024. U.S.A.
`M. F. Moody, School of Pharmacy, University of London . 29/39 Brunswick Squ1~re. London WCJN l AX, U.K.
`T . Richrnond, Institut fUr Molekularbiologie und Biophysik. Eidgenus8i8che Technische Hochschule, Honggerberg.
`CH 8093 Zurich . Switzel'land .
`R. Schleif. Biology Department. J ohns Hopkins University. Charles & 34th Streets, Baltimore, MD 2 12 18, U.l:i.A.
`N. £. Sternberg, Centr·al Resea.rch & DPvelopment Department, E. L du P ont Nemour·s & Comp any. Wilmington .
`DE 19898, U.S. A.
`K . R . Y a.ma:m.oto. Departrnt'>nt of Biochemistry and Biophysic!>,
`San Francisco. CA !)4.143-0448, U.S.A.
`M . ranagida, Department of Biophysics. Faculty of Science, Kyoto L' niversity, Rakyo-Ku . Kyoto 606, .lltpnn .
`
`l:ichool of Medicine, University of California.
`
`Editorial Office
`U. Htwris, ,'loumal of Molecular Biology , IOd St Edwards Passage, Cambridge ('82 3PJ . U.K.
`
`JOURNAL OF MOLECULAR .BIO!,.OG Y: I SSN 0022- 2836. Volumes 2.11 - 2 16, 1990, published twice a mont h on t he
`5th and 20th by Academic Press at 24- 28 Oval Road, L ondon NWJ 7DX, England. Annual subscription price including
`postage: £768 U.K . and U.S.$14f>4 overseaa. Personal su bscription rat.e: £233 U.K. and U.S. $350 overseas. Subscription
`orders should be sent. to Academic Press Limited. Foots Cra.y, Sidcup, K ent DA14 5H P , U.K. (Tel: 08 1- 300 3322). Send
`notices of changes of address to t he publisher at least 6-8 weeks in advance, including both old a.nd n(:>w addresses.
`Rt>l'ond da::;s postag(:> rate paid at J amaic·a. NY I 14:31. U.R.A.
`Air fr-eight and mailing in the U.S.A. by Publications Expediting lnc., 2UO Mt:Jatha m Avenue, Blmont, NY ll()(l:3. L .H. A.
`l'.~.A . POf'TMAf'T RRH: sPnd <·hange of addresses to ,JOURNAL OF MOLEC ULAR BIOLOOY . r/n Pu hlicat,iom•
`Rxpediting. £n(! .. 21)() t\1(:>acha nt Avenue, Elmont. NY 11 00:3, l:.S.A.
`Printed in U. K.
`
`Page 2 of 17
`
`

`

`Journal of Molecular Biology
`
`Volume 214, Number 3
`
`Contents
`
`Communicatio ns
`Str·ategy for Analysing t he Co-operativity of
`Intramolecular InteractioHS in P eptides a nd
`Proteins
`Novel GA BAA R ecep tor a Subun it is Expressed
`only in Cerebellar Granule Cells
`Crystals of cx-Momorcharin. A New
`Ribosome-iru-wtivating Protein
`
`Calci um Binding Sites in T omato Bushy i-lt11nt
`Virus Visualized by Laue Crystallogra.phy
`
`The Longest, Regu lar Poly peptide 3 10 H elix at
`Atomic Resolution
`
`Crystallization and Preliminary X-ray St.udy of
`AaR IT2 , an I nsect-sp ecific T ox in fl'Orn the
`Scorpion Androcton:us australis Hector
`New Crystalline Forms of Neuroaminidase of
`Type B Human InRuenza Virus
`
`Crystallization and P r·eliminary X-ray St.udies of
`o-Serine Dehyd ratase from Escherichia coli
`
`Articles
`Expression a.ncl Function of the U1J8 W G11ne of
`Bacteriophage T4
`Levanase Operon of Bacillus 81tblilis I ncludes a
`Fructose-specific Phosphotransferase System
`R.egulating the Expression of the Operon
`Establishment of de Nouo DNA Methylation
`Patterns. T ranscript.ion F actor Binding and
`Deoxycyticline Methylation at CpG and Non-CpG
`Sequences in an Integrated Adenovirus P romoter
`Deletion of Sites for Initiation of DNA ynthesis
`in the Origin of Broad R ost-ra.nge P lasmid R.11 62
`Myosin Step Size. Estimation F rom Slow Sliding
`Movement of Actin Over L ow Densities of H eavy
`Memmyosin
`
`Determining Local Conformationa l Variations in
`DNA . Nuclear Magnetic Resona.nce Structures of
`the DNA Duplexes d(CGCCTA AT CG) and
`d(CGTCACGCGC) Generated Using
`Back -calculation of t he Nuclear Overhauser Effect
`Spectra. a Distance Geometry Algorithm a nd
`Constrained Molecular Dynamics
`Three-dimensional Structure of the Mammalian
`Cyt.oplasrruc Ribosome
`
`A. Horovitz and A. R. F ersht
`
`(il3- 6l7
`
`K . Kato
`
`Z. F eng, W. W. Li, H. W. Yeung,
`S.-z. Chen, Y.-p. Wang, X.-y. Lin,
`Y.-c. Dong and J.-h. Wang
`J. W. Campbell, I . J. Clifton,
`T. J. Greenhough, J. Hajdu,
`S. C . Harrison, R. C. Liddington and
`A. K . Shrive
`V. Pavone, B. Di Blasio, A. Santini,
`E . Benedetti, C . Pedone, C . Toniolo
`and M . Crisma
`C. Abergel, E . Loret, H. Rochat and
`J. C. Fontecilla-Camps
`
`Y. Lin, M . Luo, W . G . Laver,
`G. M. Air, C. D. Smith and
`R. G. Webster
`G . Obmolova, A. Tepliakov,
`E . Harutyunyan, G . Wahler and
`K. D . Schnackerz
`
`L. K . Derr a nd K . N. Kreuzer
`
`I . Martin-Verstraete, M. Debarbouille,
`A. Klier and G . Rapoport
`
`619- 624
`
`{)25-626
`
`627- 632
`
`633- 635
`
`6:37- 638
`
`639- 640
`
`641- 642
`
`{)43- 656
`
`657- 671
`
`M . Toth, U . Milller and W. DoerOer
`
`673-683
`
`H . Z hou and R. J. Meyer
`
`T. Q. P. U yeda, S. J. Kron and
`J . A .. Spudich
`
`W. J. Metzler, C. Wang,
`D . B. Kitchen, R. M . Levy and
`A. Pardi
`
`685- 697
`
`699- 710
`
`7 11- 736
`
`A. Verschoor and J. F rank
`
`737- 749
`
`Page 3 of 17
`
`

`

`Two-domain .. tructure of the Native and R~active
`Centre Cleaved Forms of CT lnhini tor of Human
`Complement by Neutron Scattering
`
`Three-dirnen siom~l tructtu·e of Human
`l 113Cd1)Metallothionein-2 in olution Determjned
`ny Nuclear Magnetic Resonance Spectroscopy
`
`Amide Proton Exchange in Ruman
`Metallothionein-2 Measured by N uclear Magnetic
`Resonance Spectroscopy
`Phyuobi liprotein Methylation. Effect of the y-N(cid:173)
`MeLhylasparagi ne Residue on Energy Transfer iu
`Phycocyan in and the Phycobilisome
`
`Author Index
`
`S. J. Perkins, K . F. Smith,
`S. Amatayakul, D . Ashford,
`T. W. Rademacher, R. A . Dwek,
`P. J. Lachmann and R. A. Harrison
`B. A. Messerle, A . Schaffer, M. Vasak,
`J. H. R. Kagi and K . Wuthrich
`
`B. A . Messerle, M . Bos, A. Schaffer,
`M. VaSA_k, J. H. R. Kagi and
`K . Wuthrich
`R. V. Swanson and A. N. Glazer
`
`751 - 76:3
`
`765-779
`
`781- 786
`
`787- 796
`
`797- 798
`
`Cooer lllwstration: Conform ation of an RNA Pseudoknot. Pu gli~i el al.. .J . Mol . Biol. 214, 437-453.
`
`Origination by BPCC W hitefria,rs 1~/.d , 11'unb,·idg'' Well.~.
`and 'P'rinte.<l and bound ·in Gt·e<tl Britain by B PC<.: W heaton8 Lid, ExPler
`This journal i.s printed on acid·free ]JaJJPr
`
`Page 4 of 17
`
`

`

`.J. Mol. Bioi. (19HO) 214. 75 1- 763
`
`Two-domain Structure of the Native and
`Reactive Centre Cleaved Forms of CI Inhibitor of
`Human Complement by Neutron Scattering
`
`Stephen J. Perkinst, Kathryn F . Smith
`Depa,rtments of Biochemistry and Chemistry
`and Protein and .NJ olec1.tla1· B1;ology
`Royal Free H ospital School of J.l1 edicine
`R owland Hill Street, London N W3 2PF, U .K.
`
`Supavadee Amatayakul, David Ashford, Thomas W. Rademacher
`Raymond A. Dwek
`
`Glycobiology Unit. Department of B iochem:ist1·y
`University of O:>.;jm·d, South Parks Road, Oxford OXJ 3QU. U. K .
`
`Peter J. Lachmann and Richard A. Harrison
`
`MRC Molecular I mmunopathology Unit, MRC Centre
`Hills Road. Cambridge CB 2 2QH , U .K.
`
`( Received 14 N ovembe1· 1989; ar.cepled 20 April 1990)
`
`The <..:T in hibitor component of huma n complement is a mem ber of the serpin superfamily,
`and controls Cl activation. Carbohydrate analyses showed that there are seven 0-linked
`oligosaccharides in CT inhibitor . Together with six N-l i"nked complex-type oligosaccharides,
`t~ e carbohydrate con tent is t-herefore 26 % by weight, and t-he molecular weight (M,) is
`calculated as 71,100. Neutron scattering gives an M, of 76,000 ( ± 4000) and a match point of
`41·8 to 42·3 % 2H 20 , in agreement with t his carbohydrate and am ino acid composition .
`Ouinier plots to determine the radius of gyration R0 were biphasic. Neutron contrast
`variation of CT inhibitor in H 20 - 2H 20 mixt ures gave an overall radius of gyration R0 at
`infinite contrast of 4·85 nm , from analyses at low Q, and a cross-sectional Ra of 1·43 nm. The
`reactive centre cleaved form of CI inhibitor has the same i'Vl., and structure as the native
`molecule. The length of CI inhibitor . 16 to 19 nm, is far greater than that of t he putative
`serpin domain . This is attributed to a.n elongated structure for the carbohydrate-rich
`11 3-residue N-termina.l domain. The radial i11homogeneity of scattering density, a, is large
`at 59 x I0 - 5 from the R0 data and 28 x 10- 5 from t he cross-sectional analysis, and this is
`accounted for by the high oligosaccharide content of CT inhibitor. The scattering data were
`modelled using small spheres. A two-domain structure of length 18 nm based on two distinct
`scatter·ing densities accounted for all the contrast variation data. One domain is based on
`the crystal str-uctttre of a 1 antit1·ypsin (7 nm x 3 nm x 3 nm). The other· con esponds to an
`extended heavily glycosylated N-terminal domain of length 15 nm, whose long axis is close
`to th~ longest ax-is of t he ser·pin domain. Calculation of t he sedimentation coefficient s~o. w
`for C1 inhibitor using t he hydrodynamic sphere approach showed t hat a two-domain head(cid:173)
`and-tail structure with an M, of 71,000 and longest axis of 16 to 19 nm successfull y
`reproduced the s~o. w of 3·7 S. Possible roles of the N-termina l domain in t he functi on of CT
`inhibitor are discussed.
`
`t Author to whom con espondence should be addressed, in the Department of Biochemistry and Chemistry. Royal
`Free Hospital School of Medicine, Rowland Hill Street, London NW3 2PF. U.K.
`751
`
`© 1!)90 Academic Press Limited
`
`11022- 2836/90/ 15075 1- 13 S03.0<1
`
`
`Page 5 of 17
`
`

`

`752
`
`S. J . Perkins et a l.
`
`l. Introduction
`The comple men t system comprises a fa mily of'
`plasma a nd cell-surface g lycop roteins Lha.t, ope rate
`as a casc·ade a nd play a ma jo r r ole in immune
`defe nce mechanisms. The classical p athway of
`compleme nt a<'t.ivatio n is initiated by t.he first com(cid:173)
`po nent. C I., whi ch
`reacts largely with
`immune
`complexes of l g~I or TgC immunoglobulins bound to
`foreign ma teria l
`(Coope r,
`1985: Reid ,
`1986;
`~c hum ake r el at .. 1987). The unrestrained a ctivatio n
`of Cl would cause exeesi':ive eons umpt io n of C2 a.nd
`C4 . tim::; C l is regula t.ecl by CT inhibitor (H a rpel,
`1976; f;im & R e boul. 198 1; Coope r 1985: D av is ,
`1988). This protea se inhibi tor , presen t a t about
`0·20 mg/ ml in plas ma, forms stable l: I complexes
`\\'ith Lhe CTr a nd CTs !'mbeompo ne nts of CT to
`preven t. fu r·t.her comple ment activa tio n . Complex
`fo rm a tio n
`is characterized by
`the
`in·ever·sible
`d eavarft- o f t.he peptide b ond at A rg 444-Thr445 in
`e
`-
`t.he reactive centre of C l inhibi tor (Salvesen et at. ,
`1985). C'onfm·ma.Lional cha nges after cleav age lead
`lo a mu t'h more st.a ble protein str ucture (Bmch el
`al. , 1988; P e mberton PI al. , 1989 ). While CT inhibitor
`is t he only k11o wn inhibitor ofC'Tr and CTs, it. is a lso
`a cri t ica.l. regula.tory component o f t he t·oa.gulat.ion ,
`fib ri noly tiC' a nd kinin-rt>leasing systems, and its
`absence o r
`ina<'liva.tio n
`by
`" non-productive"
`cleavage o f the Arg444-Thr445 peptide b ond is
`importa nt in certain disPase st a tes (D avis. 198R) .
`('omposit io na l studies o f' CT inhibito r· a re reqt1ired
`to undersLand the meeha nis m of eon t ro l of Cl
`a.ctiv atio n in semm . cT inhibitor· is a member of t he
`serpin superfamily (serine prOte.itse inhibitor ) of
`ser·um inhibitor·s (Dav is el nl .. 1986 : Boek et a!.,
`I ~tit)). Its M, has been estimaLecl as 98,000 to
`11 6,000 by gel e lectr·ophoresis a nd ult racent ri(cid:173)
`fu gatio n a na lyses (P en sky et a.l. , 196 1: H a upt et a.l. ,
`1970; R e houl et al. , 1977; Nilsson & \Nima n , 1982;
`H a rrison . 1983). tl owe ve r, the sequence sho ws t hai
`the m a ture pi'Otein M, is only 52.300 and it has been
`suggested that t his implies a carbohy d ra te content
`as high as 49 % (BocJ, et al ., 1986). T his is not
`s upported
`by previous carbo hydra.te a.na.lyses,
`which found a t.ota.l of33 % Lo 35 % by mass (H a upt
`et n.l. , I 970; H a n·iRon , 1983). The structur·e of t he
`N-linked and 0-linked oligosaccha rides have been
`determined , bu t not t,he ir tot.a.l a mo unts (Rt.recker et
`a.l .. 1985). Accordingly, fur·t her carbohy clra.te ana ly (cid:173)
`seR were p erformed in order to cl£•fine t he t•omposi(cid:173)
`t io n of CT inhibitor.
`RLructura l dala a re required a lso to understand
`the fun ct.io n of CT inhihit.or. :'3edimen tat.io n d a t.a for
`CT inhibitor f(ive sgo, w values mostly at 3·7 S
`(~chul tze el. al.. 19()2; H a upt et al., 1970; R eboul et
`al., 1977 ; Chesne PI rd., 1982) fro m which hig hly
`e longated stl'llcturE>s h~wc been d eduf'ed (Odermatt
`et al. , 1981; P erkins, l !l85). Electron microscopy
`suggested that CT inhibitor ha~ an e long ated two(cid:173)
`d o main " hea d -and -La il " sLructtu·e (Odermatt et 111 ..
`198 1 ). a ncl mitTO<:alorimPt.ry e xper·iments supported
`t his t wo-d oma in s tructure (Le nniek et nl., 1985).
`T he er·yst.al st ruet ure of t he ho mo logous serpin
`
`a 1-an t it ry psin can be u sed to model t he C'-termina l
`365 a mino acid residues (Lo ber ma nn et al.. 198-i;
`Bock el al. , 1986 ; H a rrison , 1989 ). whi l'h probably
`corresp onds
`to t he " head " of CT inhi bitor . The
`··tail"
`is
`t he refore
`t he
`heavily
`g lyeosy la Lecl
`N-termina l 113 a mino acid
`r·esid ue::;. with a
`presumed length of a ny where between 26 nm a nd
`53 nm . The structure of Lhe N-tennina l d oma in is
`poorly
`tmde rsLond : indeed . the fun et.iona l con sP(cid:173)
`quences of this do main ar·e wholly unknown (Davis.
`1988).
`Nen t r·on scattering is an effective multipa rameter
`tool fo r structura l studies of g lyeoprot eins (Pe r·kins
`et al. , 1985; S mit h et rtf., 1990). Contr·ast. varia t ion in
`H1 0 - 2H 10 mixt ures will iden t ify t.he oYe rall struc(cid:173)
`t.nre under condit.ions close to physiological,
`in
`wh ich t he
`in terna l a rra ngement of protein a nd
`carbohy dra te can be a.llowed fM ( Pe r·kins , 1988a, b).
`The conformational t ra nsition between t he native
`and reactive cen t re cleavf:'d pro te ins can be moni (cid:173)
`torecl. The I wo-d oma in model for CT inhibito r can
`be compared qua n t it atively wit h t he erysta.l struc(cid:173)
`t ure of a 1-a nt it rypsin (Lobermann el al., 198-i_!
`~mith et al .. 1990), whe re molecula r mode ls fo r Cl
`inh ibi tor can be de ve loped . P ossihle roles for l he
`heavily g lyeosylat.ed N-t e rminal d oma in for the
`fun ction of c T inhibito r a rf' discussed .
`
`2. Materials and Methods
`(a) Pre.pnral ion.~ of the native and r PacliN' cenlrP dPfLrPd
`form.s of(' r inhib·ilor
`r T inhibit.m· was pr·epa red
`from human plas ma
`according to t he method of Ha rrison & Lachmann ( 1986)
`a nd citlw r used fresh or stor·ed frozen at - 70°C unt il
`requirt>d . Tf frozen. samplt>s wer·e p reJ.la rt>d for· tocat t1•ring
`:;tud ies by gel fi lt ration o n ~epharose 6 B in 12 mM -sodiurn
`p hosphate. 200 mlii· NaCI. I mM-EDTA . pH 7·0. ThE' peak
`fra.ction(s) a t abou t 10 nlfl/ ml were then dia lysed against
`t he same b uffe r, filte rE-d throug h a 0·2 Jllll filter (GE.'Ima n)
`into ster·ile g la.ss o r plastie via ls. a nci held a t 4 oc unt il
`requir·E.>d fo r a na lysis. Two forms o f CT inhibitor wt>re u~ed
`in a na lysis. Tht> na tive form showed a ~< ingle band on
`SDSfpolyac·r·yla mide gel electropho r"t'si!; (Lae mrnli , 1970),
`a nd was fully aetive against C is a nd plasrnirr. ThE' spli t
`fo rm o f CT inhib itor wns genemted s ubsequen t to isolation
`by the action of a n unident.ified protease(s). Cel e lectro (cid:173)
`phoresis
`(l..aemnlli . 1970;) a nd high-pressure
`liquid
`ch roma.togrttphy a.na lyses o f t his s howed it to haxe been
`split a t 2 s it.e>;, unt> a t or close to tlw rt>.aetive centre
`exposed loop, and t be other close to t he a mino terrniuus of
`t he protein , gt>nc rating fragmen ts indisting uis ha ble fi·om
`those gt>ne rat.f'd by Pseudomcrna.~ aer uf!inn~a elastase
`(PE>mhe rton PI al., 1989 ). The sites o f clenvage wer·e
`confirmE'd by a rnino-tennina l sequeneE.' a na lysis (pel·(cid:173)
`fnrrn ed by Dr L. Packma n at tlw l'rott'in Sequt>ncing
`Facility. Depa r t men t of Biochemistry , U niverRity of
`Cambridge) o f t he clt>a vt>u ma teria l bot h before a nd after
`dia lysis agains t scattering buffers. This ind ica ted t hat
`clea vage had oc-eurred bet ween Ser44 1 a nd Va l442 in t he
`rt>activt> cent,r•e a nd. in appi'Oximatt>ly eq uimola r a mounts.
`eit her betwt•m ~l eta ! a nd Leu32 or bet.ween Leu32 a nd
`Phe:33 in t h(' a mino-termina l regio n. Whitt> t he C-termina l
`peptid~. oonla ining the l't>lwtivt> {'f'nt.re residue:;, was fully
`retained in the clialysed clf'aved molecule, pa rt ia l (up to
`:30 %) loss nf t.he a mino-termina l 3 1/ 32 resitlut>s may have
`
`Page 6 of 17
`
`

`

`Solution S tructure of C' T 1 nh:ibitor
`
`753
`
`occurred, although t he nat.ure of the residues at t he
`amino terminus of the protein (N-P-N(CHO)-A-T-S-S-R-8)
`make thi$ diffiC'ult to quantify precisely. Samples we re
`reanalysed by gel electrophor esis subseq uent to solution
`scattering studies. a nd JH) differences before or a fte•· t he
`scattering experiment~ were detected. Tn addition, native
`CT inhibi to1· after the scattering expe riments had an una l(cid:173)
`tered specific activity a.gainst active-site titrated pla-smin.
`
`(h ) Carbohydralf analy"i" of C 7 inhibitor
`The methods fo r
`t he
`re ll'ase o f ~-linked oligo(cid:173)
`saet·ha•·ides b.v hyd razino lysis a nd tht>ir subseque nt analy(cid:173)
`sis htwe been described (As hford el al ., 1987). Enzyme
`digestion of oligosacc·ha rides with
`neura rninida><e
`(Arlhrobarler ureajacif-n.~) and fJ-... V-ac·etylhexosarninidase
`(J ack bean) ha ve a lso been descri bed (P a rekh PIn/ .. 19~7).
`0-linked o ligo;;accharides wPre relea:;cd and r·educed by
`alkaline sodium borohydride treatme nt using a modifi ra(cid:173)
`t.ion o f t he method desC'rihed b.\' Mizuochi el al. (1980).
`T y pically . 2·5 111~ of f'T inhibitor was t reated with 2150111
`of 0·6 M-8odium [3 H ]borohydride (360 mCi/ mmo l, New
`1-':ngland Nucleat·) in 0·05 M-sodiurn hydroxide for 20 h at
`
`50 °(.; with shaking. Tlw mixt ure was r·ooled to ~ 0 (' and
`12 Jll of acetir a nhyd ride was added. Afte1· lO min , a n
`addit iona l l 2Jtl of aeetic a nhyd ride was added a nd the
`1'1'-<tction a llowed to proeeed at room tempe rat ur•p for a
`furt her 10 min . This addition and
`incubation was
`1·epeated t.wit·e more. 'l'lw reaC'tion mixturE" Will'; adjust.f'd
`to appmxima te ly p H 6 hy dropwise a ddi t ion of I i\HWf'tir·
`to a
`l ml eolum n of Dowex
`a.e id. a nd then a pplied
`AUfiO x 12 ( H + form , Bio-RRd). The c·olumn was e lutt•d
`with 5 column volumes of dist.illcd water. The cl uatt' wa:s
`filtert>d t hroug h a 0·5Jtrn T efi un tiJ ter and evaporated tu
`dryness. Borat!' was remO\' Pd by repeated evaporation
`(5 x) with methanol (300 Jl] ). The sample was t lw n
`s ubjected to d e~ce ndin g paper chromatog raphy for +8 h
`with butun- 1-olfetha nolfwater (+ : I : I . by vol).
`F or q uantification of the number of 0-linked cha ins on
`t'T inhibito r. lac·to~e was added prior to the a lka line.>
`borohydride t rratmcn t. After isolation of the r·ed ueed,
`rad io-labe lled o ligosaeeharides from paper chromato·
`t hey were
`t r·eated with neuraminidase and
`graph~r ,
`subjected to hig h-vo ltage e lectrophoresis
`in j)yridint>/
`acetic acidjwater (3: I : 378, by vol.) buffer, pH 5·4 at
`80 V em - •
`fo r 45 min. The neutral oligosaccharides
`re maining at the o rig in WE're e luted wit h water ancl
`applied to a Bio-Gel P-+ high -resolutio n gel filtmt inn
`syste m . Radioaet.ive fractions e lut ing from I to 5 glucose
`units. by compa rison with
`t h e
`internal
`isomalto(cid:173)
`oligosaccharide standards. we re pooled together. LaC'titol
`was separated from redur-ed 0-glycans by hig h-voltage
`electrophoresis in borate buffer (Ashfor·d et at.. 1987). The
`radioactive areas were e luted from t hE" electrophoreto(cid:173)
`g ram and quant ified by liquid scintillation coun t ing.
`A portion of t he Bio-Gel P-4 pool was a lso s ubjeC'ted to
`lor
`the ]Jroc.:edure
`reducing-terminal monosaeeharide
`determination (Ashford et al., 1987). The radioac:tive a r·cas
`of the result ing electrophorc>togram C'orresponding
`to
`glucitol and N-aretylgalactosam initol wer-e quantified hy
`integmtion of t he radioa ctive peaks detec-t ed by the linear
`a na lyt<<:>r. t'oneetions in t ht> q unn t ifiC'ation wc>re madf' for
`traP!' contamina nt,<;; in the lactit.ol and for background .
`
`(~') Ne11fron du/11 f'ollerlion mul anoly.SP8
`
`Neut r·on du.ta W t't't> r·ul iPded 0 11 Tn><t.rumcnt Dl7 at tlw
`Tns t.il'ut-Laue-T.angt'v in . Gr't'nohiP. Cuinif'J'
`radius o f
`jtyrat,inn RG d ata Wl'I'P bas(•rl on a sample' to clPtf'c·tor
`
`d istance of 3·46 m a nd neut.ron wavelt>ngths of 1·38fi to
`to a Q range
`1·395 nm or 1·600 nm, corresponding
`(Q = ~tr s in 0/A. where 20 is the sc-attering angle) of 0·0:33
`to 0·4-86 nm - •. Data a t. larger Q were obtained with a
`sample to d etector distance of l ·~O m. wavt-lengths of
`1·001 to 1·004 nm . and main beam to detector angle:> of0°
`and 19·89° to g ive Q ranges of 0· 1::! to 1·60 nm - t a nd 0·8 to
`3·6 nm- •. Instrument Dll with a sample to detector
`distan ce of I 0·5 m and wavelength of I ·00 nm was used to
`obtain a lowm· Q ranue of 0·0019 to 0·220 nm - •. Sa mplrs
`were dialysed at 6°C with stirring into 12 mlJ ·sodium
`phosphate, 200 mM -NaCI, I mM- EDTA (JJH 7·0 in 0 %,
`80 % o r 100 % 2H 2 0 solutions wit.h 4 changc>s over· :!6 to
`48 h ). All neut ron data were recorded at zo•c. Concen(cid:173)
`trations were measm·ed using a n absorption coefficient.
`A : ·~",.280 of 3·6 (Hanison , 1983; Salvesen el a/., 1985).
`Data reduetion was based on standard Grenoble software
`(Ghosh . 198 1 ), and the final analyses in London were
`based on SCTPL (Pe rkins & S im, 1986). Statistical analy(cid:173)
`ses were pe1formed using MTNTTAB (ve•·sion 6.1) on a
`microcomputer (Ryan et al. , 1985).
`At small Q, the CuiniPr equation gives t he radi us of
`gyration Rc a.nd the fonvard scattering at zero st·<tttel'ing
`a ng le 1(0) (Guinier & F o urnpt .. 1955):
`
`In / (Q) = In f (O)- R~Q 2f 3.
`
`In a given solute-solvent con t rast, Rc measures t he
`d egree of e longation of a g lyc-oprotein . If t his struetu•·e is
`sufficiently e longated , the Rc of the cross-sectiona l struC'(cid:173)
`ture Rxs and t he cross-ser·tiona.l intensity at zero angle
`f f( Q) X Q]Q-0 are obtained from :
`In (i(Q) x Q] =In (J (Q) x Q]Q_,0 -RfcsCif2.
`Tlw mat<:hpoint is obtained from a g mph of J I (O)fc7~1
`against % 2 H 20 (7', is sample transmission . l is sample
`t hickness, c is concentration), llcnd from tbe cross-sec-tional
`analy»es usi ng Jf / (Q) x Q]Q-olc'I'. t. Experirnt> ntal matc·h (cid:173)
`point:s can be compared with the amino a.cid sequen ce on
`the basi~ of t he unhyrlratc>d Rhape of a 1-AT. t he use of
`crystallogra phic· volumt>s (unhyd ra.ted ). a nd t ht' 10 °,{1
`no n-excha ngt> of t he main -cha in a mide proto n,.; (Pe rkins.
`198()). The contrast \'ariation analysis of t he arrangenwnt
`of ca.rbohydrate a nd protein is based on t he Htuhnnann
`equation ( fbel & 1-ltuhnnann. 1975):
`H}; = Rb-e+ac x llp - •- Pc x llp- z
`Ris = Rh.c+«xs x llp - •-Pxs x !lp - z,
`
`where RG.c and Rxs.c a re t he r adii o f gyr ation of the
`macromolPcule a nd its cross-sectio n at infinite con t rast,
`aG a nd axs measure t he corrt>spondi rw radial inho mo(cid:173)
`geneity of scattering densit ies, and llp - ? is t he J'flCi proC'al
`solvent- solute !'ontra.st diffe re nce. The te rms in Pc and
`flxs m('asure the displacemc.>nt of the centre of srattering
`density as t he cont rast is varied . From t he term in p, t he
`d istance /l betwc.>en
`the centre:; of 2 tompo nen t!:l of
`distinct SNlttering densities p 1 a nd p 2 nnd volumes 1'1 a nd
`1'2 , respectively. c·an be calculated; however , :3 pammc.>t.er(cid:173)
`weighted least;-squa res fitting of t he.> d ata in Fig . 3(b)
`showed t hat t his was not measurable.
`The neu tr on-sca ttering curve T(Q) in reciprocal space
`(•an be t ransfor·med into real spaee P(r) by use of thE'
`indir-ect t ra nsfonnation p r·oced ure (T'l'P) met hod (Clatter,
`1982). Th i,; r·equired the full sc:attering cmve to 3·6 nm - •.
`It. offe rs a n alternative r·alculation of t he Rc and the
`length L for r.T inhibitor. The 7fi experimental I (Q) du ta
`points (in 3 subdoma ins to C'Orrespond to t he 3 Dl7
`config urat.ions) were bt>st fitted us ing 10 t.o 15 C'ubic
`><p linE"~. whil'h were tmnsiorrned into 10 1 points of t he
`
`Page 7 of 17
`
`

`

`754
`
`,'J. J. P erkins et al.
`
`P(r) fund inn hu><t•d on n nu1ximum lt'ngth of :!0 nm .
`Dt>~<lllf'Arin~ nf t lw rwnt ron t·urvl' wa" pt'rfnrnwrl on thf'
`basis of u Dl i \\tl\f'lt'ngth ~prvnd of 10° 0 full -widt h to
`half-maJ>imum. Tht> -.lit wid th and lt-nj!th corrections werf'
`~•'t tn hr Pqual 1\ll d ha~·rl on a bPam din•rgenre of
`11·011:? r·ad nt l n .. trunwnt l>li.
`
`(.I ) .1/udtllilr!J"f t/11 11/l'ltl'lttre of('[ inhibitor
`(llodf'lling of tlw m•utron·:>l'IHterm~ l'urn'l! \\at! ba.!t'<J
`nn tlw dr.'' \·olunw of the ~ly"op1·otein (Chothia. 1975;
`Pt>r·kms.
`I!) '(i). and U~>t<cl DPinf' ~irnulations baSt>d on
`sphl•n•..; uf
`:! di!ltim·t "'l·alte;ing densities to follow
`dl'!Wr ihed pr-oct>dure" (P Prki rls & Weis;~, 19 3; Per·kins.
`1985; Smith PIal., l!l!)O). ~l nd t>lling of the sedimentation
`l'O(' n! ('it' nt .-r~o. w \l'll'i lm,;!'d on tht' h,vdmted volume. whic·h
`Porresponcl ~ t o t lu• ~'""' nf thP ,·olnrnf'R nf th e dry g lyco(cid:173)
`protl'in oncl I he hydmt io n t~ht<ll (as~:< uming a hydration of
`o-:J g H l O/~ g lyt·oprot:t•in ll lld 1\11 Plcctt'08tricted water
`molet·ult' volumt• of 0 .11:!-!fi nml), and t.his !Pads to a
`,.nJu nw ;; o f 0·721 rnlfg for cT inhibitor
`partial s peritil-
`(T'Prkins . IHR!I). Wlwn h,vdnulyuamil' sph1•1'('~ arP usN!.
`rhiK h.vtlratiull is illl'l't:ll~ed tu O·:J9!!; in o rd t'r to ~·urnvc·n­
`sal!• for thl' void ~Jint·es l!etw~n th£> nun-overlappi ng
`Flpht>rPs (PNkiniS. 198Hli). ( 'alt·ulations wPre performed
`us i n~ tlw pmgram OJ•;:-< I) I A (C:arria de Ia Torre &
`Blonmfirld. I !)77,. h).
`
`3. Results and Discussion
`(a) I 'nrbo!lydrnll' nnn/y.~l'8 of ('J inhibitor
`
`to
`Ca rbo hydrate nnalyl'\es WE'rP carried out
`confirm
`the stnwtu rl.'(s) of the
`:'\-linked oligo(cid:173)
`sncc:'haridci! and lo quantify the numbe r· of O-lin ked
`oligosacchal'idt•s bound to C'T i11hibito r . 'While the
`pro tein sequence ( ijol·k e/ al., 1986) indicate six
`occupied N-lin kcd g lycosylation siles, the situation
`for the 0-linked c·hains is less clea r.
`the N-linked
`H igh -\·olta~t' (' )ertroJ!hOI't>Ris of
`rt>leased by
`oligosacc-harides of C' l
`inhibitor.
`hydt·azinol.v:~is and rcdutl'd wit.h Nai33H4 . indicated
`that 9(;·6 % of the ol i go~me{\h a rides catTied at least
`one negn t ivl' rha r·gp (Fig. I (a)). T he distribution o f
`charged Rpt>rif's was conRist.ent with a mixtu re of
`mo no-
`and
`cl i-::Jia ly lal~:d hitLnl~:nnary oligo(cid:173)
`saccharidi's (A- 1 and A-2) a nd mo no-, di- and t ri(cid:173)
`sial~· l ah:-d lriantennar·y oligosac·c·haridcs (A'-1 , A'-2
`and .-\ -3). The rcla tt\'e proportiOns of these com(cid:173)
`JKmt'nts wt'n:: A- I, 14·9 ° 0 ; A-2, 58· 1 ° 0 ; A'-1 , 2· 1 °0 :
`A'-2. 5· 1 °0 : A-3, 16·4°0 . E lectrophoresis of the
`charged ~-linked oligostL<·chnrides afte r treatment
`with ne un\minida!l(' s ho wed t hnt all the charged
`components wPrc• c·cmvprted to nt>utral speci P~ and
`remained at tht' o rigin . The refo re the only charged
`moiety was :.iulit• atid. Bio-Gel P-4 gel filtration
`chro m a togro phy
`of
`thE'
`ne ura minidase-treated
`oligosacchal'idcs ga,·e tht> elution pr·ofile sho"n in
`Figure I (b). ThC' ma jo r peak eluted a.l 13·5 gluco~e
`un its and minor t·om po nents eluted at 14·5. 16·5,
`17·3 and HHi glm·osc unit:;. These peaks were pooled
`as s hown in Fig ure I (b). The relati,·e propo rtions of
`Pal'h pool wer·<· as follo wa: I, 2·0 %; II. 6·5 %; I II ,
`IV, 20·fl 0
`0 ; V. 6Hi %. Tht> <•ha.racteristi{l
`9·3 1}0 ;
`elutio n pro fi le poRitio ns 1111 Bio-Grl P -4 (Yamashita
`
`e/ n/. , 1982) indic·atNI that peak a w all a t etra(cid:173)
`antennary oligofl!lcdnuidC', p<·a ks I; anrl r were tri(cid:173)
`an t Pnnn r_y oligo><arrha ridr.
`(pPa k
`rorf'
`fuco(cid:173)
`{J
`sylated), and pt>aks rJ and I' Wt:l't' ui-antennary
`oligosat·rharidC's (peak d <'Ore fucosylated). These
`oligusat•charide a:-.signm<"nts w('r•C' c·onfirml:'d using
`sequent ial exoglycosicln!Se ~((llf' llf'ing (data not
`s ho wn).
`H i~ lt - ,·oltagt• ell'<'trnphorel'iS of a po rt ion of t ht>
`radioa<'tivl' o l i~o:mc·cha rirl<•t; ol>tllined l>~· t rl:latment
`~;orl ium [3 RJhoro(cid:173)
`of C' l
`inhibito r with nlka lirw
`hydrid<· in t hC' pn•sr1wc of lactose'. s howed nC'ga(cid:173)
`Li\' cly chargt>d mate rial '' illt u ~im il a r mo bility to
`sial.vl-ladose and 1.\Uthc-nt ic r\eu~Ac2-3(6)G ai/Jl -
`3Ga iNAc in ndditiun lo l~wiit u l. WhPn the c harged
`mate r1al was l t'Pail·d with ne um m inirlase and t•esub(cid:173)
`jec-ted to C' I E•rt r·ophur·Psi~ . the radioactivit y re mained
`al the nrig in 1\nd wa~ Lhcref(Jrt: ne u t ral, confirm ing
`that o n ly s iu.ly l1tLed t'llf.u·ged s pecies were prPsent
`(d atil not. s hown). 'l'hf• hulk or tlw rndioar-tive o ligo(cid:173)
`saecharill£•s wcrr then 11'\'ntf'<l with neuraminidase.
`l'it'c'l ro phol'l's i:< and
`the
`resulting
`subjreted
`l o
`nc utml oligosa<·chnridrs Wt'rt' ~; uuj el· t{'d to Bio-Gel
`P-~ gel filtra tion . 'l'he <·hrnma.toJ(I'Ilm s howed a
`;;eri<>s of peak:;; Pluting 1\1 3·5 and 2·5 glucose units
`and a minor peak l'lutin~ at ~-4 glucose uni ls (data
`not shown). T he (Wak at 2·5 glucose units corres(cid:173)
`ponds
`to
`la.t•titol nnd
`that at 3·5
`to Ga l/JI ·
`3GaiX Ac01 . I n>:ignifirant a mo unts of X-linkcd oligo(cid:173)
`saccharides were pn'scnt. Aeid h~·drolysis of t ht'
`pooled fnu·t ions from OIIC' to
`tivc g lucose units
`confirmed that t h<' only reducing-terminal mono-
`accharides p rest•nl wer•f' N -aeetylgalactosaminitol
`a nd glucitol. The ·e da ta taken together are con(cid:173)
`· i~;t en t with Lhc· finding of Htrecker P/ at. ( 1985) thrLl
`the 0-linked oligosaccha r·idcs of CT inhibitor ar·t:'
`predominantl