`p,...; mz
`o;o .-~
`C..:l: H<
`orn :x>
`rur u
`rH Vl,
`pn
`<""'
`P. l
`,
`IHP.
`r m(l)
`.... p.
`CD Vl Z
`<""'
`nn ...... o
`:x> t m.-.
`o z m
`...J<'"'G"'l
`-DI.I1 mO
`N CD Vl
`0
`-D
`\>1
`
`'>;:
`0
`0
`0
`0
`0
`N
`0'
`N
`..0
`0
`...J
`~i z
`p.
`
`3:
`p.
`-<
`..0
`0
`
`0
`
`OF CALIFORNIA,
`SAN DIEGO - LIBRARIES
`Received .-;or,: 12···29-89
`'+, 1869-
`v. J.·-
`Nov.
`
`RESERVE COPY
`
`RETURN TO
`CIRCULATION DESK
`
`PFIZER EX. 1549
`Page 1
`
`
`
`nature
`
`14 December 1989
`Vol. 3421ssue no. 6251
`
`~ Sulphur volcanoes grew on the
`desiccated lake floor of Poas crater.
`Costa Rica. earlier this year. This 2-
`metre-high cone
`later collapsed
`revealing a bubbling pool of liquid
`sulphur. Photo by David Stevenson.
`See page 790.
`
`THIS WEEK
`Setting the baseline
`Ttle greenhouse effect, Its exis(cid:173)
`tence, extent and Implicat ions.
`nas been the subject of hundreds
`of model and theoretical studies.
`Yet the physical basis of the
`the infrared radiation
`effect -
`energy trapped by atmosphefic
`gases and clouds - has not pie(cid:173)
`viously been measured
`from
`observational data. Ravel and
`Ramanathan (page 758) have
`now filled that gap, presenting an
`effective method
`for directly
`monitoring future changes in the
`greenhouse effect. See also
`News and Views. page 736.
`Channel vision
`The light-sensitivity of vertebrate
`photoreceptor cells depends on
`the closure of a cation channel
`which Is gated directly by cyclic
`GMP. the Internal messenger of
`visual transduction. Now. the
`cloning and sequencing of the
`complementary DNA
`for this
`channel from bovine retinal rod
`photoreceptors shows
`that it
`could belong to a new family
`or ligand-gated ion channels.
`Page 762.
`Regal matters
`Worker honeybees efficiently
`destroy worker-laid male eggs if a
`queen Is present. This ·worker
`policing' ensures that the colony's
`male reproductlves are sons of
`the queen. pages 796 and 741.
`Positional cues
`The Hox-5 homoeogenes are
`activated in the developing limo
`bud of the chick In a temporal
`order that corresponds to their
`spatial ordering on the chromo(cid:173)
`some. and In a series of partially
`overlapping spatial domains
`such that the region In which all
`five genes are expressed coin(cid:173)
`cides with the zone of polarizing
`activity
`that determines
`the
`antero-posterior limb axis. Pages
`767 and 734.
`
`Insulin mechanism
`Changes In our understanding of
`how Insulin regulates glucose up(cid:173)
`take are likely to follow the dis(cid:173)
`covery that the endothelial cells
`that line the capillaries of the
`insulin-regulated tissues. muscle
`and fat express high levels of the
`insulin-regulated glucose trans(cid:173)
`porter. Page 798.
`Beneath the rings
`Infrared Imaging has revealed
`several new
`features
`in the
`atmosphere of Saturn -
`includ(cid:173)
`Ing a hotspot at the north pole
`and an asymmetric structure at
`the equator. Page 777.
`Yeast shows the way
`The processes of differentiation
`by Induction of different gene
`regulatory proteins under the
`Influence of cell-<:ell signals.
`and of the establishment of stem
`cell llneages, fundamental to
`metazoan development. can be
`Illustrated In the single-celled
`yeast. Recent work Illuminates
`the molecular basis of some of
`these processes. Review Article,
`page 749. See also page 830.
`Hairpin comer
`Oligonucleotide encoding tela(cid:173)
`merle DNA sequences form stable
`dimers In solution, possibly by the
`formation of hydrogen bonds be(cid:173)
`tween two Intramolecular hairpin
`loops to form antlparallel quad·
`ruplexes containing cyl\c guanine
`base tetrads. Pages 825 and 737 _
`Alvin's finds
`Samples of sediment taken by the
`research submersible Alvin at the
`Guaymas Basin hot vents (Gulf
`of California), contain a group
`of methanogenic archaebacteria
`growing, quite unexpectedly. at
`a temperature of at least UO"C
`(page 833). From the same area,
`dense layErs of large filamentous
`sulphur-oxidizing bacteria of the
`genus Beggiatoa 11ave been
`found at a depth of 2010 m (page
`834).
`
`f'/alt<r_. (ISSN 0028-0K;\6) is published ""ckly on Thursd•y. e.«eptthc l•st week in December.
`by M;tcmtllan Mai!Jilincs ltd (4 little E3S<~ Street, London WO R JLF). Annual •ubscnpuon
`for USA and C:ln•d• USS295 (instiMionaUcorpor>te), USS I2S (indtvldual making pmonal
`p~ymtnt). USA and C•no.d1an ordento:Naturt, Subscriptfon 01.'pt, PO Do• l733, Riverton, NJ
`fl!!OTI-1l:l3, USA. Olher order; to Namu. Brunei Road, Ba.ing!lo~~· Honl! RG21 2XS. UK.
`Seco"ddasspostuge pi'idat New York, NY 10012 and :.dduuma.l ma1hng offic~s. AuthOfi.~Uon
`to photocopy material f<'r,ince~nal or pcnonal use .. or inter~:d ot person~t.l use of specific da~nl5,
`is- &r~nu:d by Nut11r~ to hbrane~ and other~ regtstcred with the Copynght Oearance Center
`!CCC) Trunsactionat Reporting Service. provided the base rce uf St.OO a copy pha SO. lOe~ pa,ge
`IS paid direct to CCC~ '21 CQngrc~ S1ree1. Salem. MA 01970, USA, ldenuficauon «<!e for·
`"'""'"' 0028·0836189 SLOO + 50.10. US Po.,mOitcr send addrm change.> to: Nawrr, 65
`Blet.c:ker Slrt:et, New York, NY 10012. Publbhe<l i.n J~tpan byNatureJapttn.K.K, , Shin:~fit$uke
`Bldg. J6 lchigaya Tal!1a<hi, ShinJuku·ku, Tokyo Jh2, JApan.© 1989 M•cm•llan Magal1neo ltd,
`
`721
`
`729
`
`731
`
`733
`
`NATURE SAYS
`Europe must define Its objectives • An outstanding
`broadcasting system is endangered
`NATURE REPORTS
`India's new stance • Human genome matters • After the
`'quake • Early reptile on the move • Drifl·nets cause
`conflict • West German reactor • Support for East
`Germany • A viatlon milestone • Mono Lake reprieved •
`French universities • MRC • AIDS epidemic •IBM awards 723
`CORRESPONDENCE
`Ranking universities • Reviewing poUcy • Carbon tax •
`Journal costs • Etcetera
`COMMENTARY
`Keeping up with the Russians
`G W Fisher, P C Grew & B Yardley
`NEWS AND VIEWS
`Is the salami sliced too thin? John M addox
`Vertebrate limb development: A pattern emerges
`Julian L ewis & Paul Martin
`Diamond films: Low-pressure nucleation routes
`M ichael Pinneo
`Greenhouse effect: Gauging water-vapour feedback
`Robert D Cess
`DNA structure: T he turn ofthequadruplex?
`M axim Frank-K amenetskii
`Palaeontology: Four legs to stand on for Devonian
`vertebrates
`H enry Gee; Per E rik Ahlberg
`Scanning atomic microscopy: Variations on
`an original theme
`CFQuate
`Mid-ocean ridges: Propagating rifts exposed
`Ken C Macdonald
`Sociallnsects: Who are tbe drone police?
`Jon Seger
`Daedalus: A clean press
`
`734
`
`735
`
`736
`
`737
`
`738
`
`739
`
`740
`
`741
`742
`
`SCIENTIFIC CORRESPONDENCE
`Sampling the lithospher e N W Rogers, C J Hawkesworth,
`D S Ormerod & P D Kempton: Reply-
`W F McDonough, K P Jochum, H Palme & B Spittel
`Cleaning up after Chernobyl K F Baver stock
`Early warning for L TP R Anwyl
`BOOK REVIEWS
`Origins and Evolution of Planetary and Sate.llite
`Atmospheres S K Atreya, I B Pollack & M S Mal/hews eds
`William B McKinnon
`Debating A r chaeology by L R Biriford Warwick Bray •
`Diverse Divers: Physiology and Behavior
`by G L Kooyman P J B utler
`Boninites and R elated Rocks A J Crawford ed,
`Carbonatites; Genesis and Evolution K Bell ed
`HSYoderJr
`Speciation and its Consequences D Olle & J A Endler ed.s
`M ark Ridley
`REVIEW ARTICLE
`A regulatory hierarch y for cell specialization in yeast
`1 Herskowitz
`
`743
`
`744
`
`74S
`
`746
`
`747
`
`746
`
`749~
`
`v
`
`PFIZER EX. 1549
`Page 2
`
`
`
`ARTICLES
`
`:zs Vll~f. C o recton~ts Bf'ID LariO(Otms (lon!man. New Yoo... l9S1l
`26 '•tal. F J & M~le< I A. Spec Pubis &eo! Soc S. Alr. ll, 247-26IH1984),
`'ll> K. G 1 P,<Joi 21. 629- 650 (1980)
`27
`~ 'o&lar<l, P c &. ~Joln•r. P (Jeology l•n lM , ess)
`29 ""ui S & fullar6 V J Geo!Dgla do BfaSII•E~teta da ll'llver!ldode dl! SAo P...,o. 1983).
`:JO •!VI<ara. A J el aJ Ctt!Sfll/ EI'OWI/On Of SOWJern A/ric• ISj>rloger, New Yor11. 199.2),
`.-:ro~ C. (ed ) ~ntdftJttc Geosc1f!f1Ce hJm-veiSily or WiSCO(lSln Press:. MadiSOI'\ 1982)
`;Jt
`
`32. Ktng, l- C. Soorn 11/ricM Sreneq 2nd edn (Oliver aoc1 Boyll. Edtrburgn, 1951).
`33 Klfli, L. C. The Motpllology olt/le Eartl! (Oliver ana Boyd, EdtnbU<gh, 196~)
`3A Partr~dt;e, T_ C. !. ~ R. R, S Mr. geo/, J 90, 179- 208 119871.
`35 Cotlt!n.L .. Snellt~ N J. Oelhal J, & Va.I,J. R. The GoocivonoloB)' ono f'loluUon of Mr<"a (Cla<etldOo.
`0Afotd, 1984\,
`36. ~=~: R. w_ 6tJrlnt Con(. Fosslon Tra:I<OB<IrJII Ab•~- Voi.Ur>lverslt6 de fr""'-'llt!.Q>mte,Besa~.
`
`C-onformations of immunoglobulin
`hypervariable regions
`Cyrus Chothia*t, Arthur M. Lesk·~, Anna Tramontano*, Michael Levitf§,
`Sandra J. Smith-Gilll, Gillian Air~, Steven Sheriff"'··, Eduardo A. Padlan#,
`David Davies#, William R. Tuliptt, Peter M. Colmantt, Silvia Spinelli**,
`Pedro M. Alzariu & Roberto J. Poljak**
`
`• MRC laboratory of Molecular Biology, Hills Road, Cambridge CB2 2QH. UK
`t Eoropean Molecular Biology laboratory, Meyerhofstrasse 1, Postfach 1022.09, 0-6900 Heidelberg. FRG
`~Department of Cell Biology, Stanford University Medical School. Stanford. California 94305, USA
`t Cl1ristopher l11gold laboratory. University College london, 20 Gordon Street, london WClH OAJ, UK
`11 NatioJlal Institute of Ca11cer and # National Institute of Diabetes and Digestive and Kidney Diseases. National Institutes of Health. Bethesda,
`Maryland 20892. USA
`Department of Microbiology, Universtty of Alabama, Birmingham, Alabama 35294. USA
`tt CSiRO Divlston of Biotechnology, 343 Royal Parade. Parlwille 3052, Australia
`t :f Unite d'lmmunologie Structurale, Departement d'lmmunologie, lnstitut Pasteur, 25 rue du Dr Roux, 75724 Paris. France
`
`On the basis of comparative studies of known
`antibody structures and sequences It has been
`argued that there is a small repertoire of main(cid:173)
`chain conformations for at least five of the six
`hypervariable regions of antibodies, and that the
`particular conformation adopted is determined by
`a few key conserved residues. These hypotheses
`are now supported by reasonably successful pre(cid:173)
`dictions of the structures of most hypervariable
`regions of various antibodies. as revealed by
`comparison with their subsequently determined
`structures.
`
`relationships between the amino-acid sequences of
`T HE
`im111unoglobulins and the structures of their antigen-binding
`sites are important for understanding the molecular mechanisms
`of the generatio n and maturation of the immune response and
`for designing engineered a ntibodies. Antigen-binding sites are
`formed by s ix loops of polypeptide, the hypervariable regions;
`three rrom the variable domain of the light chain (YL) a nd ttuee
`from the variable domain of the heavy chain ( VH ), denoted LJ.
`L~, L3, and HI , H2, H3, respectively (Fig. I a ). Within the
`d•nnains, the loops are conne1..ted to a 13-sheet framework whose
`2
`structure is conserved 1
`• The specifi city and affin ity of the bind(cid:173)
`•
`ing sites are governed by the structures of the six hypervariable
`rcgions3·•.
`fwo models can be proposed for the relationship between
`the amino-acid sequence and structure of the binding-site loops.
`In o ne model, different sequences produce diffe rent conforma(cid:173)
`ti11ns for both the main cha in and side chains of the loo ps.
`Because hy pe rvariable regions have different sequences in
`d1flerent antibodies, this model implies that each region adopts
`a different conformation in different antibodies. In the other
`
`• Present address: The Squibb l]lstitute for Medical Research, PO Box 4000,
`Pnnceton, New Jersey 08543-4000, USA.
`
`NATURE · VOL 342 · 21/ 28 DECEMBER 1989
`
`model, antibodies have only a few main-cha in conformations
`or 'canonical structures' for each hy pervariable region. Most
`sequence variations would only modify the surface provided by
`the side chains on a canonical main-chain structure. Sequence
`changes at a few specific sets of positions would switc h the main
`chain to a different cano nical conformation.
`
`Canonical structure model
`Experimental evidence indicates that the cano nical structure
`model describes the relationship between amino-acid sequence
`9
`and structure for a t least five of the six hypervariable regions5
`"
`.
`Kabat et al. 5 fou nd conserved residues at sites within certain
`sets of hypervariable regio ns a nd suggested that they had a
`structural role. Padlan and Davies6
`, and more recently de Ia
`Paz el a/. 7
`, showed that some of the hypervariable regions
`in the immunoglobulins of known structure have the same
`main-chain conformation in spite of several differe nces in
`sequence.
`Chothia and Lesk& identified th~ residues that thro ugh pack(cid:173)
`ing, hydrogen bonding, or the ability to assume unusua l values
`of the torsion angl~s </:>,
`tJr o r w, are primarily responsible for
`the ma in-chain conformations of the hypervariable regions in
`the structures then known-the Fab fragme nts of N EW (ref.
`10), Mc PC603 (ref. 11), KOL ( ref. 12) a nd J5J9 ( ref. 13) and
`the VL domains of REI (ref. 14) a nd RH E ( ref. IS). The
`conformations are determined by the inte ractions of a few
`residues at spel:ific sites in t he hypervariable regions and, for
`certain loops, in the framework regions. H ypervariable regions
`that have the same conformatio ns in different immunoglobulins
`have the same or very similar residues at these s ites ( Fig. t and
`Table t ). Examination o f the a mino-acid sequence of the anti(cid:173)
`body 01.3 s howed that its hypervariable regions are the same
`size as those in known structures and contain the same or similar
`residues a t the s ites responsible for known conformations•. On
`the basis o f these observations the atomic structure of the YL-YH
`dimer of Dl.3 was predicted before its experimental determina(cid:173)
`tion. Comparison ofthis predicted structure with the preliminary
`crystal structure showed that the conformatio ns of fou r of the
`hypervariable regions had been predicted correctly; the confor(cid:173)
`matio n of L3 was significantly different from that predicted, and
`
`877
`
`PFIZER EX. 1549
`Page 3
`
`
`
`ARTICLES
`
`L1 RegJonst
`
`Canomcal
`Structure
`
`TABlE 1 Sequences and conformations of v, and VH hypervariable regions of known structure
`
`Protein
`
`26
`
`27
`
`28
`
`29
`
`30
`
`31
`
`a
`
`<l
`
`e
`
`32
`
`2
`
`25
`
`33
`
`71
`
`J539
`HyHEl-5
`NQ10
`
`REI
`D1.3
`HyHEl-10
`NC41
`
`McPC603
`
`4- 4·20
`
`2
`
`3
`
`4
`
`s
`s
`s
`Q
`G
`Q
`Q
`
`v
`v
`v
`
`I
`I
`v
`
`s
`N
`R
`
`H
`G
`s
`
`K
`N
`N
`T
`
`N
`
`H
`
`s
`y
`y
`v
`y
`N
`A
`
`y
`
`L
`M
`M
`
`y
`y
`y
`
`y
`y
`F
`y
`
`A
`A
`A
`
`A
`A
`A
`A
`s
`s
`
`v
`
`G
`
`N
`
`N
`
`E
`
`G
`
`K
`
`N
`
`N
`
`T
`
`s
`s
`s
`s
`s
`s
`s
`s
`s
`
`s
`s
`s
`D
`N
`s
`D
`s
`s
`
`Q
`
`v
`
`s
`s
`
`Total no. of sequences known for l1 regions: human, 95: mouse. 299.
`1
`2
`3
`4
`Canonical suucture
`60
`5
`5
`HUman sequences that fit (%)
`25
`20
`10
`Mouse sequences that fit (%)
`
`15
`
`L2 Regions
`
`canonical
`Structure
`
`Protein
`
`REJ
`MoPC603
`J539
`D1.3
`Hyl-£l-5
`Hyl-£1.-10
`NC41
`NQlO
`4-4-20
`
`50
`
`51
`
`5:<
`
`48
`
`64
`
`E
`G
`E
`y
`D
`y
`w
`D
`K
`
`A
`A
`I
`T
`T
`A
`A
`T
`v
`
`s
`s
`s
`T
`s
`s
`s
`s
`s
`
`G
`G
`G
`G
`G
`G
`G
`G
`G
`
`Total no. of sequences known for l2 regions: human, 69t mouse. 183.
`1
`Canonical structure
`Human sequences that fit (%)
`95
`95
`Mouse sequences \hat fit (%)
`
`l3 Regions
`
`Canomcal
`Sllucture
`
`2
`3
`
`Protein
`
`REI
`McPC603
`01.3
`HyHEl-10
`NC41
`4-4-20
`NQ10
`
`J539
`
`HyHEL-5
`
`91
`
`92
`
`93
`
`94
`
`95
`
`96
`
`y
`D
`F
`s
`H
`s
`w
`w
`w
`
`Q
`H
`w
`N
`y
`T
`s
`T
`G
`
`s
`s
`s
`s
`s
`H
`s
`y
`
`R
`
`l
`y
`T
`w
`p
`v
`N
`p
`N
`
`p
`p
`p
`p
`p
`p
`p
`
`l
`p
`
`y
`l
`R
`y
`w
`w
`L
`
`90
`
`Q
`N
`H
`Q
`Q
`Q
`Q
`Q
`Q
`
`Total no. of sequences known for L3 regions: human. 52; mouse. 152.
`2
`1
`3
`Canon1cal structure
`90
`Human sequences that lit (%)
`2
`80
`Mouse sequences that fit (%)
`1
`
`10
`
`HI had a very different fold from that predicted9
`. ( We report
`below that the refined conformation of 01.3 corresponds more
`closely to the predicted structure.)
`An examination of the library of the known immunoglobulin
`sequences shows that many immunoglobulins have hyper(cid:173)
`variable regions that are the same size as those in the known
`structures and contain the same or closely related residues at
`the sites responsible for the known conformations8
`. These
`observations indicate that for at least five of the hypervariable
`regions there is only a small repertoire of canonical main-chain
`conformations and that the conformation actually present can
`often be predicted from the sequence by the presence of specific
`residues.
`The accuracy of the canonical structure model
`for
`immunoglobulin binding sites depends on (I ) the correct deter-
`
`878
`
`mination of the sets of residues responsible for the observed
`conformations and (2) changes in the identity of residues at
`other sites not significantly affecting the conformations of the
`canonical structures. The model can be tested, refined and
`extended by using it to predict the atomic structures of binding
`sites in immunoglobulins before their structures have been deter(cid:173)
`mined by X-ray crystallography.
`We have now tested the canonical structure model by using
`it to predict the structures of four immunoglobulins before their
`structures had been experimentally determined. These
`immunoglobulins are HyHEL-5 (ref. 16), HyH EL-10 (ref. 17),
`NC41 (ref. 18) and NQIO (S.S., P.M.A. and R.J.P., manuscript
`in preparation). The analysis of the amino-acid sequences of
`the.se immunoglobulins indicated that 19 of their 24 hypen'ari(cid:173)
`able regions s hould have conformations close to known canoni-
`
`NATURE · VOL 342 · 21/ 28 DECEMBER 1989
`
`PFIZER EX. 1549
`Page 4
`
`
`
`I' ""g""'s
`L ·~r..cal
`, r retute
`
`Protetn
`
`26
`
`27
`
`28
`
`29
`
`30
`
`31
`
`32
`
`34
`
`94
`
`ARTICLES
`
`Mc:PC603
`KOl
`.1539
`013
`~~-5
`NC4l
`NQ10
`4-4-20
`NEW
`~HEL-10
`
`G
`G
`G
`G
`G
`G
`G
`G
`
`G
`G
`
`T
`I
`0
`s
`T
`T
`T
`T
`T
`s
`i ~f no. of sequenc.es known for Hl regoons human, 50; mouse. 321
`1
`t +nomcal structUff:!
`H •man sequer100s that ht 1%)
`50
`M~use sequences that fit ('lbl
`80
`
`'
`
`F
`r
`y
`~
`
`F
`s
`0
`
`s
`s
`s
`T
`s
`T
`s
`s
`s
`T
`
`0
`s
`K
`G
`0
`N
`s
`0
`
`N
`0
`
`F
`y
`y
`y
`y
`v
`f
`y
`
`0
`0
`
`L
`F
`f
`F
`F
`t
`
`M
`M
`M
`v
`I
`M
`M
`M
`y
`w
`
`R
`R
`R
`R
`R
`R
`R
`G
`R
`N
`
`1 rl Reg1ons§
`
`~.arr011~
`tructure
`
`2
`
`Protein
`
`52 a
`
`0
`
`53
`
`54
`
`55
`
`71
`
`NEW
`013
`f+tl'£l·10
`
`Hyt£1.-5
`I\C41
`
`KOl
`.1539
`M)lO
`
`McPC603
`4-4·20
`
`H
`0
`s
`
`s
`T
`
`(•)
`G
`s
`s
`(•)
`K
`N
`
`G
`G
`G
`(o)
`G
`G
`
`s
`G
`s
`
`y
`y
`
`G
`v
`
`G
`1\
`
`0
`0
`G
`
`N
`v
`
`p
`T
`
`0
`p
`s
`
`N
`N
`
`t<
`K
`
`G
`p
`
`A
`L
`
`R
`R
`R
`
`R
`R
`
`oJtal "" ot sequer1ees ~nown tor H2 regions human, 54, mouse, 248
`,&>Onocaf StiUCtU<O
`1
`2
`3
`4
`><uman sequences tMt f•t t'lt.)
`15
`1
`40
`15
`Moose sequences that f•t (91.1
`15
`40
`5
`20
`
`t>e tes•d!Jes I> sled 11ere (single-letter code) <~re !hoSe that form the hypervar~able regions and those In the frameworil teglons that ate 1mportat1t tor the observed conformations
`OJ •hese reg•ons~ The hypervat~able reg.ons are taken as those outs>!le the frameworll J3-sl'lt!et8 E.cept for H2. they are similar to, out not •oont>cal With the teglons that show
`h•t~ seqUence vor~auors and whiCh Kabat er al.'" use to define hypervariable reg•ons. The sequences are grouped so that those that nave the same malo-chain confO<matlon
`01 ranorncoJ structure are adjacent. The canonical structure numbers used below refer to the confotmatlons shown In fig 1. The resld\Jes In the hypervariable and framewon;
`r<~•ons that a<e mtl<nly •espons1ble for tl'lt!se confotmauons• are lndtcated by ar aste11~ 11'1!! class•flcatlon and sequence requ,.ements of the H2 con!ormatloos have been
`••·v•sed In the light ol worl< oosenbed here and elsewhete'" For eacl1 hyperV8ltable regiOn tho numtlcr of human and mouse sequenoes listed by Kabat e1 8110 are giVen Wo
`a1 "' g.-e the percenta~e of these sequences t11at are the same slle as the ~nown canoniCal structures and have the same res•aves at the positions mat!(ea by an aster~Si<
`Canonical structure ·4 IS Illustrated In Fig 4 AlthOugh the siZe of the known U structures vanes oetween 6 ana 13 res•d~s. they nave Closely related folds w1th residUes
`:l6 l9 and 32 p.JCI<ed aga1nst the lrdmeworr. In the same corformation8 The remammg residues form a turn or loop or the surface lf•gs 1 and 4). The ends of the long loops
`h ''~ some fle\!OIIIty There are anotlwr 75% ot the human sequences and 20% of I he mouse sequences that have one more residue than structure 2. 01 one fewer than structure
`4 otld whose sequences satisfy me reQuwements listed abOve, It IS e~pectea thai these a.ffer orly '" the conformations of the t1ps of tile surface looos.
`-The Hl hypervarlable rc&tOns "''th canoriiCal structure 1 haVe very som11ar con'ormauons the r.nu differences •n the coordtnates or the• ma.,>-<ham atoms aoe 0.3-0 8 A
`T • H1 rel/()ns In NEW and ~~-10 Ol\ly partly s~Usfy the sequence reqUirements for structure 1 and nave a Olstorted versiOn of Its oonformauon.
`4 The H2 regoon here comp11sas res.oues 52a-55 fhe regiOn w1th hrgh sequenCe vartatoon 1S 50-65 (ref 261 In the known structures the millr\oCilain CO<Iformat""' of 50_52
`"'d 56 63 ao flOt 01Her s~&nofcantly" IF1g 101 (• ). The restaues at pos1tiors 55 ot 54 1n tt>e taroll'Cal structures 2. 3 an<1 4 have reSioues wtlh poSitiVe •atues r01 0 and .,
`• ~usually but flO! 1n aH cases Gjy Asn or Asp IS found at these 51\es for a sequence to match !hat of canorncal strUC!U<e 2. 3 or 4 the presence of these reSidUes at Sites
`51 01 55 ts reou,.ed
`
`CJ I s1ructures. We then comp:ued the predicted structures of
`tl1cse hypervariable regions with the subsequently determined
`structures. Another immunoglobulin 5tructurc, 4-4-20 (ref. 19)
`h,ts recently been reponed. We did not have the opponunity to
`ptedict the structure of4-4-20 before its experimental determinl·
`tion, and we discuss here only how its hypervariable regions
`h.rve the conformations e xpected from the known canonic1l
`structures. Also, we repon that the refined conformation of D 1.3
`(n.:f. 20) corresponds more closely to the predicted structure.
`
`Model building procedure
`The main-chain conformations of the hypervariable regions in
`the VI( nnd VH domain~ of known structure are shown in Fig.
`I The re~idues responsibk for these conformafions are listed
`
`M TURE · VOL 342 · 21/ 28 DECEMBER 1989
`
`in Table I. Each hypcrvariable region in the immunoglobulins
`o f unknown structure was examined to determine ( I ) whether
`it ha~ the ~amc ~i.~:c: a~ any homologous hypervariable region of
`known structure and (2) whether its sequence contains the set
`of residues res ponsible for a known conformation. Except for
`LJ in Hy H EL-5, all the light-chain regions correspond to a
`known canonical structure, t\s do a li the HI regions and the H2
`region in H y H EL- 10 (Table 1). The conformation predicted for
`the H2 regions in NC4 1 and Hy H EL-5 was based on the analysis
`of the H2 region in the preliminary st ructure of J539 (ref. 8).
`In all three of these antibodies the H2 region is a four-residue
`turn with Gly at the fourth position and the predicted conforma(cid:173)
`tio n is that almost always found for such turns: ', ( Below we
`present a mo re accurate analysis of H2 regions.) For HJ regions
`in Hy H EL-5, HyHEL-10, NC4 1 and NQIO, no predi~:tion of
`
`879
`
`PFIZER EX. 1549
`Page 5
`
`
`
`ARTICLES
`
`conformation could be made on the basis of the known canonical
`structures.
`The sequences of the VL and YH domains were compared to
`see which of the known framework structures have sequences
`close to those of the unknown structures. From the comparisons
`
`Antibody
`
`Antigen binding site
`
`b
`
`VK
`
`of the hypervariable and framework regions, one known VL and
`one known VH structure were taken as the starting points(cid:173)
`parents-for the model of the predicted structure. If the confor(cid:173)
`mation predicted for a hypervariable region was not present in
`the parent, but was present in another known structure, the
`hypervariable region in the parent was replaced by that in the
`other structure. Side chains in the parent that were different
`23 and the
`from those in the unknown structure were replaced 22
`•
`resulting model subjected to a very limited energy refinement24
`.
`HyHEL·S, HyHEL-10 and NC41 hypervariable regions
`The atomic structures of Fab fragments of immunoglobulins
`HyHELrS, HyHEL-10 and NC41 in complexes with their protein
`16
`antigens were determined by X-ray crystallography16
`. The
`-
`resolution of the X-ray data used to determine the structures
`and value of the residual (R) after refinement are (complex,
`resolution, R value): HyHEL-5-lysozyme. 2.5 A, 20%; HyHEL-
`10- lysozyme, 3.0 A, 24%; and NC41-neuraminidase, 2.9 A,
`19%. These structures have been determined at medium resol(cid:173)
`ution. The tracing of the polypeptide chain of the hypervariable
`regions is unambiguous. although the orientation of some of
`
`L1 f15." ~
`
`32
`
`26
`
`Leu
`
`3
`
`26
`
`~
`
`9
`v.
`1
`
`26
`
`!:>
`
`lie
`
`2
`
`95
`P10
`
`L3
`
`1
`
`L2
`
`~fl"
`
`lie 4
`
`9j
`
`2
`
`VH
`
`2ti~rwt2
`Phe29~
`
`H1
`
`H2 ·~~
`3:-~58
`1
`
`880
`
`~56
`:t·f-8
`
`3
`
`4
`
`2
`
`FIG. l a. Antigen-binding sites of immuno(cid:173)
`globulins are formed by six loops of PQIY·
`peptide, three from the Vl domain L1, L2
`and L3 and three from the VH domain H1, H2
`and H3 (wavy lines). These loops are attached
`to strands (0) of a conserved 13-sheet. b.
`Canonical structures for the hypervarlabte
`regions of V 1< and VH domains. In each drawing
`the region Is viewed so that the accessible
`surface is at top and the framewoti< region
`below. The main-chain conformation and some
`of the side chains that determine this confor-
`mation are shown. For the definition of the
`hypervariable regions used here, see Table 1.
`The Immunoglobulins In which the different
`canonical structures occur are listed In Table 1.
`
`NATURE · VOL 342 · 21/28 DECEMBER 1989
`
`PFIZER EX. 1549
`Page 6
`
`
`
`ARTICLES
`
`tht peptide groups is uncertain. Most side chains are unequivo(cid:173)
`cally placed.
`hgure 2 shows the predicted and observed structures of each
`of the hypervariable regions, superposed by a least-squares fit
`of their main-chain atoms. Table 2a gives, for each predicted
`, 31HI observed hypervariable region, the r.m.s. difference in posi(cid:173)
`tion of the main-chain atoms.
`l he main-chain conformations of the predicted and observed
`hypervariable regions are very similar (Fig. 2, Table 2a). The
`only exception is the HI region of HyH EL-5. Although the
`oh1erved and predicted conformations of residues 26-29 and
`32 are the same, residues 30 and 31 are in quite different
`po>itions. The recently refined structure of Fab JS39 (T. N. Bhat,
`E.A. P. and D.O., manuscript in preparation) shows that these
`dilterences were inherited as a result of an error in the original
`determination of the 1539 structure used as the parent for this
`region. Rebuilding the predicted model with the refined J539
`structure puts residues JO and J I in the correct position and
`gives an r.m.s. difference between the predicted and observed
`H 1 regions of 0.6 A.
`Given the medium resolution of the structures used to derive
`tht: models and of the experimental structures, the agreement
`of the predicted and oblierved loop conformations is excellent.
`Relative positions of the hypervariable regions
`Figure 3 shows the positions of the hypervariable regions relative
`to each other and to the framework for the predicted and
`ob~erved structures. To produce this figure the predicted and
`ob~erved structures were superposed by a least-squares fit of
`framework residues. In Table 2b, the differences in position of
`Jhc: hypervariable regions are reported.
`Small differences in the relative positions of the hypervariable
`reg1ons in the predicted and observed structures might be expec·
`teJ because of two factors not corrected for in the model build-
`
`y
`ing. First, the predicted structures were built using
`. h f
`.
`h h
`parent
`'d
`d
`omams t at ave some rest ues m 1 e ramework and VL- VH
`interface that are. different from those in the final predicted
`structure. These dtfferences have small but significant effect
`the main-chain structure of the individual domains and th son
`d'
`e way
`h 22s,z6
`k
`they pac
`toget er ·
`. Second, dJerences between th
`dieted and observed structures could arise from the Pfl~~·:tSpref-
`.
`.
`. h h
`.
`l
`h
`18
`0
`~ w
`t c anugen . n ot er proteins I'
`t e assoctallon w11
`d
`h
`I . h
`f
`I
`. , lgan
`b. d'
`tn tng can resu t m t e movement o c ose-packed s
`of polypeptide relative to each other by }-2 A., and thee:~:~~
`loops are able to move somewhat more·'.
`The differences in the positions of the predicted and ob e d
`H2 regions in HyHEL-5 and NC41 (Table 2b) are large~ ~e
`. an
`expected from these factors. The interactions that the H2
`.
`f h
`d
`•
`regions
`make w1th the rest o t e VH omam were therefore examined.
`Residue 71 and position and conformation of H2
`At the sam_e time as the structures of Hy~EL-5 and Nc41
`became ava1lable, the refinement of the atom1c structure of 1539
`was completed (T. N. Bhat, E.A.P. and D.O., manuscri 1 ·
`preparation). The conformation of H2 in the refined stru~t 10
`!s not like t~at in Hy~EL-5 and NC41 but i~ the same as ~~~~
`tn KOL. Thts was quue unexpected. The ma1n determina 1 f
`0
`the conformation of small turns is usually the position of glyn ·
`' 1 . KOL Gl
`'d
`· ·
`cme
`res1 ues· : 10
`y occurs at pos~t~on 54, and in JS39
`,
`HyHEL-5 and NC41, Gly occurs at posthon 55.
`'
`An examination of the environments of the H2 rcgio s"
`shows that in KOL and JS39 the side chain of framework resi~ue
`Arg 7 1 packs between, and forms hydrogen bonds to HI
`d
`H2. In HyHEL-5, resid~e 71 is Ala: and h~rc: the c;;vity ~"at
`would be created by th1s smaller stde cham is filled by th
`insertion of a residue from the H2 region-Pro 52a. In KO~
`and J539 the side chain at position 52a is on the surface Th
`relative movement of position 52a involves a change in th:
`
`Hy ftEl· S l t
`
`FIG. 2 The predicted (bro~<en line) and ob~erved
`(conll/luous hne) conformations of the llypervari(cid:173)
`abfe regions. The sllucrures have been super_
`posed by a least-squares flt of their rnain.Cilain
`atoms. Residue numbers and the r.m s. difference
`in position of the superposed atoms are given In
`Table 2a Predicted and observeo Side-chain con(cid:173)
`formations are shown for all regions except Hl
`and L3 In NQlO wtlere they Ollseure the mamella1n,
`After. our prediCtiOn of lhe NC41 siiiJcture several
`rev1s1ons were made to the S!!Quence and some
`of the differences can be seen here_
`
`t<y HEl S t<l
`
`~~ ,.,El 10 >~•
`
`"JCO: t W
`
`NO'O HI
`
`<J-N0'0 HZ
`r0r ~ ~
`
`tiClt H7
`
`Hy HEl·S H2
`
`t-ty HEL '0 H2
`
`NATURE · VOL 342 21/28 DECEMBER 1989
`
`881
`
`PFIZER EX. 1549
`Page 7
`
`
`
`ARTICLES
`
`TABlE 2 Differences in strooture of predicted and Observed hypervariable regions
`
`(a) Differences In local co~formatlon (A)
`
`Hypervariallle
`region
`
`L1:26-32
`L2:49- 53
`L3:90-97
`ti1:.26- 32
`ti2:52-56
`
`R.rn.s. difference In atomic positions of rnaln-chaln
`atoms after optimal superposition
`NC41
`Hyt£)..-.10
`
`H'jl-£L-5
`
`NQ10
`
`0.8
`0.9
`
`1.4
`1.l
`
`1.1
`0.8
`0.3
`1.3
`1.0
`
`0.7
`0.4
`0.5
`0.9
`0.7
`
`0.4
`0.9
`0.6
`0.3
`0,4
`
`(b) Differences in position relaU~e to frameworll (A)
`Range of differences In positions of c. atoms after
`superposition of frameworks residues (A)
`New structure
`Hypervarlable
`Original
`region
`prediction
`library
`
`Protein
`
`Hyt£)..-5
`
`Hyt£1.-10
`
`NC41
`
`NQlO
`
`0.8- 2.1
`1.2-2.3
`0.7-2.1
`0.5-2.1
`
`1.5-2.6
`1.0- 2.D
`1.8-3.0
`1..0-2.1
`0.4-1.9
`
`L1
`L2
`H1
`H2
`
`L1
`L2
`L3
`Hl
`H2
`u
`L2
`L3
`H1
`H2
`
`Ll ·
`L2
`L3
`H1
`H2
`
`2.0-3.8
`1.6-1.6
`0.8--4.1
`3.0-7.2
`
`0.7- 1.6
`0.6-.13
`0.8-1.5
`1.3- 3.5
`0.6 -2.9
`
`1.4- 2 .4
`1 .1- 1.8
`.1.6- 2.3
`1.3- 2.0
`2.1--4.4
`
`OA- 2.7
`0 .5- 1.4
`0.6-1.5
`0.6- 1.2.
`0.6- 0.9
`
`Superposition of s are Illustrated In Flg. 2, The differences In the positions of the
`hypervarlable regions In the original predictions and the observed structures of b are
`shown In Fig. 3 . The predictions With a new sllucture library Involve rebuilding the
`fttf£1.·5 model using the refv.ed VI. J539 and the VH NC41 strootures, and rebuilding
`the NC41 model using VL McPC603 and VH Hyi-EL-5: see te<t
`
`conformation of H2. It also tilts the H21oop so that the positions
`of residues at the top of the loop in H yHE L-5 differ in p osition,
`relative to those in KOL and 1539. by -4.5 A. In NC4 1, the
`residue at position 71 is Leu and that at position 52a is Tbr,
`and the shift is not as large as that in flyHEL-5.
`Tbis analysis implies that the conformation and position of
`four residue H2 region s a re determined by the packing against
`the VH framework and by the identity of the residu e at position
`71 in panicular, a