`
`BIOEPIS EX. 1125
`
`Page 1
`
`BIOEPIS EX. 1125
`Page 1
`
`
`
`
`
`BIOEPIS EX. 1125
`
`Page 2
`
`BIOEPIS EX. 1125
`Page 2
`
`
`
`J. Mol . Biol. (1983) 167, 661- 692
`
`Structure of a Novel Bence-Jones Protein (Rhe)
`Fragment at 1·6 A Resolution
`
`W. FuREY JR, B. C. W ANG, C. S. Yoo AND M. SAX
`
`Biocrystallography Laborat01·y
`Box 12055, V.A. Medical Center, P ittsburgh, Pa 15240 , U.S .A .
`and Department of C1·ystallography
`University of P ittsburgh , Pittsburgh, Pa 15260, U.S.A.
`
`(Received 13 October 1982, and in revised form 3 F ebruary 1983)
`
`The crystal structu re of Rhe, a A.-type Bence-J ones protein fragm ent, has been
`solved a nd refin ed to a resolution of 1·6 A. A model fragm ent consisting of the
`complete variable domain and the first three residues of the constant domain
`yields a crystallog raphic residua l Rr value of 0·1 49. The protein exists as a dimer
`both in solution and in the crysta ls. Although the " immunoglobulin fold " is
`generally preserved in the structure, there a re significant differences in both the
`monomer conformation and in the mode of association of monomers into dimers,
`when compa red to other known Bence-Jones proteins or Fab fragments. The
`variations in conformation within monomers a re particu la rly significant as they
`involve non-hype rvariable resid ues, which prev io usly were believed to be part of a
`"struct ura ll y inva ri a nt" fra mework co mmon to a ll
`immunoglobu lin variable
`domains. The novel mode of dimerization is equally importa nt, as it can result in
`combi ning site shapes and sizes unobtainable with the conventiona l mode of
`dimerization. A co mpa rison of the structure with other variable domain dimers
`reveals furth er that the variations within monomers and between doma ins in the
`dimer a re coupled . Some possible fun ctional implications revealed by this coupling
`are greater variability , indu ced fittin g of the combining site to better accommodate
`antigen ic determinants , and a mecha nism for relay ing binding inform ation from
`one end of the variable doma in dimer to the other.
`In addition
`to providing
`the most accurate atomic parameters for an
`immunoglobulin domain yet obtained ,
`the high resolution and exten ive
`refinement resul ted in identification of several t ightly bound water molecules in
`key stru ctural positions. These water molecules may be regarded as in tegra l
`components of the protein. Other wa ter molecules appear to be req uired to
`stabilize t he novel conformation.
`
`1. Introduction
`Rhe is a ,\-type Bence-Jones protein fragment consisting of a complete variable
`domain and the first three residues of a constant domain . The protein exists solely
`as a non-covalent dim er both in solution and in the crystalline state. The crystals
`are orthorh ombic, with space group P2 12 12 and cell constants a= 54·63(2) A,
`b = 52·22(3) A and c = 42·62(3) A. The unit cell contains four monomers (114
`661
`
`0022- 2836/83/ 190661 - 32 S03.00/0
`
`© 1983 Academic Press Inc. (London) Ltd .
`
`BIOEPIS EX. 1125
`Page 3
`
`
`
`(iii~
`
`residues each) and approximately f) I 0; 0 so h·ent by volume. The crystallization
`eo nditions have been repo rted (Wang & Sax , 197+) , as has a preliminary analy i
`(e~:-carbon level) of the stru eture at 3·0 A resolution (Wang et al .. L 979 ). In the 3 A
`analys is. se,·era l aspects of the structure we re described , the most significant
`being the d iscove r,v of a mode of association of ,·ariable doma ins into dimel'!!
`diffe rent from that obse n ·ed pre,•iously in eithe r Bence--Jo nes proteins or Fab
`fragments. Thi · novel mode of dimerization is correlated with a change in
`conformation within monomer , and does not represent merely a rotation of one
`dom a in relat ive to t he other. Furthe rm ore, the segment within each mono mer
`that cha nged conform ation consists solely of non-hyperva ri ab le
`residue .
`Prev iously , these residues (sequence number·s 36 to 50) were believed to be part of
`a " stru ctura ll y inva ri a nt" fra mework comm on to all immunoglobulins.
`In order to determine factors responsible for this novel conform ation, it is
`n ecessar~' to extend the resolu tion beyond the 3 A level and to complete the
`chemical seq uencing of the protein. Detail of the phase extension and refinement
`to 1·9 A resolu tion were reported ( F'urey et al. , 1979) , but no attempt was made to
`interp ret th e
`tru cture, sin ce this was merely an interim step toward a detailed
`analysis at 1·6 A resolu tion . The current high-resolution a nalys i confi rms the 3 A
`stru cture interpretation and indicates that, a lthough the conformational change
`invol\'es non-hypervari able residue , it is ,·ery likely ca used by interactions with
`residues in the first and second hyperva riab le regions. The analysis also suggests a
`role for solvent in stabilizing the novel protein conform ation . An alternative
`inte rpretation of the data suggests a plausible trigger mechanism for relay ing
`information from the combining site end of variable domain dimers to the switch
`regions connecti ng them to constant doma ins.
`
`2. Experimental Procedures
`The X -ray data collection and reduction process has been desc ribed in detail (Furey et
`al .. 1979), so onl y aspects refl ecting the quality of the data need be stated here. Rhe
`crystals diffract X -rays extremely well , so it was possible to collect all data to 1·6 A
`resolution from I crystal with on ly a 15% decrease in standard refl ection in tensities. In
`add ition , the modemte size of the unit cell enabled collection of diffractometer data by the
`0-20 scan method commonly used in small molecule crystallog raphy. The overa ll quality of
`to 1·6 A resolution are
`the data set is refl ected in the fact that over 75% of a ll refl ection
`considered " ob erved''. even by the rather stringent 1/a(l ) > 3 criterion. The percentage of
`obse n ·ed reflections as a fun ction of resoluti on is given in Table I. 1'\ote that the data wer:e
`collected at room temperature with a relatively low power (0·5 kW ) X-ray source. hence rt
`should be possible to collect even higher resolution data if low temperature techniques are
`applied a nd a more powerful (1·5 kW) X -ray tube is used .
`Extending t he proced ure desc ribed in the 1·9 A paper (Furey et al. , 1979) , refin ement was
`resumed after in cluding the add itional X -ray data to 1·6 A resolution. The 1·9 A refined .co·
`ordinates (R r = 0·28, R F = (L'\IFo\-\Pc\ll/L'JF0 1) served as ini t ial parameters . Since chemrcal
`seq uen ce information was not avail able. electron density maps (eith er 2.P0 - F < or F. - !c)
`we re exa mined freq uently. Residues to be examined we re always deleted from the phaS L~g
`process before map calcul ation. Occa ionally. errors in am ino acid seq uence or atornrc
`positions were indicated by the maps. The errors were corrected by non -in teractive
`com puter graphi cs (Furey et al. , 1979 ), optical compa rator tech ni q ues (Richards, 1968) and ,
`in th e Ia t stages, in teractive co mp uter g raphi cs. In a ll case after seq uence changes were
`implemented , restrained reciprocal space refinement (H endri ckson & K onnert. 1978) was
`
`BIOEPIS EX. 1125
`Page 4
`
`
`
`STR.L' ('Tl' RE OF BE:\C'E - .10:\ES PHOTEl:\ Rh c
`
`663
`
`TABU: 1
`Obser-Ped reflections as a j11nction of resolution
`
`dmin
`(A)
`
`3·0
`N
`1·9
`1·6
`
`Total
`
`2667
`509 1
`10.073
`16.665
`
`Obsern•d if I > 3 a( / )
`
`:\o.
`obsen·ed
`
`0
`0
`(in m nge)
`
`0
`0
`(curnulati ,·e)
`
`25R2
`.J.78 1
`8975
`12,842
`
`96·8
`90·7
`8.J.·l
`58·7
`
`96·8
`939
`89·1
`77-1
`
`computation was performed with the space group genera l a rray pmces or
`(Furey el al .. 1982) of the Hendrickson- K onnert program .
`next step in the refin ement process was to introd uce indi,·idua l i otropic the rm a l
`for each of t he non -hydrogen atoms. a nd to in clude solvent molecules in the model.
`the obvio us presence of water molec ules in some of the early ma ps, solvent at oms
`not inco rpora ted into the model until late in the refine ment process (Rr = 0·23) to
`mi ta king erroneously seq uenced side-cha in atoms for water molecules. Wate r
`positi ons we re obtained from difference e lectron den ity map by scanning through
`largest peaks and determining the shortest distance between each peak and a ll atoms in
`current model. If the shortest distance was between 2·3 and 3·.J. A and the model atom
`ved was capa ble of formin g hydrogen bonds , a water oxygen wa added at the peak
`. The new model was then subj ected to se,·eral cycles of least -sq uares refinement.
`re was iterated several times and. a fter discarding water molecules that moved
`their o rig ina l locations. resul ted in the inclusion of 186 water oxygen atoms (I 02
`occupied) . The mean electron density at the water sites is 1·26 e/ A 3
`. The
`811;1mate~d error in the electron density fun ction is -0·20 e/ A 3 (Cru ickshank , 1949), hence
`the water molecu les shou ld be reasonably well -d ete rmined . Therma l and occupancy factors
`for the water molec ules ranged from 5 to 68 A2 a nd 0·30 to 1·00, respectively. H ydrogen
`atom co ntributi ons fo r hydrogen a toms in the protein we re in cluded in a ll structure factor
`calculation , but their positions were recom puted every few cycles rather than refin ed .
`At this time, partial seq uence information became a\·ai lable (W. Brown . personal
`communication ) for compa rison with the X -ray model. There were II disc repancie , all of
`which were reso lved in the model by adding , deleting o r changing the atomic type of, at
`most, 2 atoms/ residue. The chemical sequence is now known exce pt for distinctions within
`the pairs , Gin , Glu a nd Asn, Asp . Since these residues a re iso teric and each can form
`hydrogen bonds, they cann ot a lway be distingu i heel at 1·6 A resolution by X -ray da ta.
`Therefore acid/a mide distinctions in the model we re based mostly on the seq uences of other
`A-type variab le dom a in s (Kabat el al., 1977). In fa ,·orable case . however, ob vious
`hydrogen bonding partners and/or the chemi cal composition in short peptides ena bled a
`choice to be made. The a min o acid seq uence in the final model is given in Table 2. When
`ignifi cant discrepancies we re
`compared with the seq uence deduced from th e 3 A map .
`revealed , a lthough the original inte rpretation was rea onable (60% correct) .
`The final model conta ins 1019 non -hyd rogen atoms, 833 from protein (all full y occ upied ),
`and the 186 water oxygen atoms. The model fi ts into the electron density extre mely well ,
`~ indicated in Fig. I . The mean isotropic thermal factor i 16·5 A 2 when all atoms are
`•~eluded , a nd 12·5 A2 fo r protein atoms on ly. F'or solvent atom , the mean Bt is 34·2 A2 A
`histogram indi cating the distribution of thermal facto rs for the protein is given in Fig. 2.
`The final crystallog raphi c residual Rf. value is 0·149 for the 12.763 obse rved refl ections with
`d spacings ra nging from I 0 to 1·6 A. Th e R factor is pl otted as a fun ction of resolu tion in
`. t B (isotropic thermal factor) = 87T
`2u2
`, where u is the root-rn ea n·square amplitudP o f atomic
`VIbration .
`
`BIOEPIS EX. 1125
`Page 5
`
`
`
`664
`
`W. FUREY E'fAL.
`
`TABLE 2
`Amino acid sequence of Bence-Janes protein Rhe (V1)
`
`p
`
`8
`p
`
`Res idue
`2
`I
`-!
`II
`12
`10
`5
`17
`16
`15
`9
`6
`13
`I+
`18
`~0.
`19
`20
`s A
`s G
`B v
`p G Q R
`T Q
`L
`Code E
`\' T
`T
`I
`No.
`2 1 22
`24
`23
`29
`28
`27
`25
`26
`:3 1 32 33 34 35 36 37
`30
`38 39 40
`Code s c T G
`s A
`I w y Q Q
`s v
`T D
`I G
`~ N
`\'
`49 50 5 1 52 53 54 55 56 57 58 59 60
`48
`46 47
`No .
`+I
`42
`43
`44
`-!5
`s
`c: v s
`Code p G K
`p
`p
`y
`y
`A
`I
`K
`L
`L
`L
`N D
`L
`No.
`61 62 63
`71
`74 75 76 77
`65 66 67
`6+
`68 69
`72 73
`70
`78 79 80
`s A
`s K
`s G T B A
`s
`s G
`F
`Code D
`A
`L
`R
`I
`L E
`8+ &'J
`No.
`8 1 82
`83
`91
`86 87
`88 89 90
`92 93 9+
`95 96 97
`98 99 100
`Code s E D
`c A
`A w N D
`s L D
`y
`y
`p G
`A D
`E
`E
`,-
`:'\o . 101 102 103 10+ 105 106 107 108 109 11 0 Ill 11 2 113 II+
`p K
`Code F G G G T K
`T
`L G Q
`L
`
`One-lette r amino acid code: A , alanine: C, cysteine; D , aspartate: E , glu tamate: F , phenyla lanine:
`G. g ly cine; I , isole uc ine: K . lysine: L, leucine: :'\ , asparagin e: P . pro line; Q, glutamine: R, arginine; S,
`serine; T. threonine ; V. valine: W. tryptophan: Y , tyr-osine: 0 , water.
`
`Fig. 3. Throughout the refinement , the protein model was restrained with respect to
`stereochemistry , resu lting in a _final model with root-mean-square deviations from " ideal"
`bond distances of 0·024 A. The root-mean-square dev iation from planarity for the
`appropriate atomic groups (including pe ptide links) is 0·020 A. The last refinement cycle
`resulted in a root-mean-square sh ift in atom ic positions of 0·006 A with a corresponding
`estimated standard deviation of 0·01 I A. The latter is most certain ly an underestimate of
`the true error in atomic positions. which we estimate to be ~0·04 to 0·08 A by the method
`of Luzzati (1952). Mean electron densities for the main-chain atoms, N" , Cc.:, (' and 0 are
`2·89 , 2·24, 2·44 and 3·17 e/ A 3
`, respectively .
`
`3. The Molecular Structure
`(a) The monomeric subunit
`The Rhe monomer consists of nine strands, as is the case for all immunoglob ulin
`variable domains of known structure. A stereoscopic drawing of the ex-carbon
`structure is given in Figure 4. If we neglect the N-terminal residue, which is
`highly disordered and cannot be located accurately, the structure begins with a~
`turn in volving residues 2 to 5. The nine strands contain residues 6 to 12 , 17 to 24,
`33 to 39 , 44 to 50 , 54 to 57 , 60 to 67 , 72 to 79 , 84 to 92 and 97 to 11 4. All strands
`are in an extend ed conformation ; however, there is one short helix of roughly 1·5
`turns (residues 25 to 32) connecting strands 2 and 3.
`As seen in Figure 4, the nine strands are connected by ten turns, resulting in a
`pair of antiparallel ,8-sheets. The first , designated sheet A, contains strands 2, 6
`and 7, while the second (sheet B) contains strands 3, 4 , 8 and 9. The two sheets are
`also connected by a disulphide bridge between Cys22 (strand 2) and Cys89
`(strand 8), and by a salt-bridge between Arg62 and Asp83. Only strands 1 and 5
`are isolated from the sheets, although segments of strands 3 and 4 eventuallY
`break away fmm sheet B . This latter feature has important consequences with
`regard to the mode of domain- domain association upon dimerization. In prev ious
`
`BIOEPIS EX. 1125
`Page 6
`
`
`
`STR L. CTL' HE OF BE:'\CE - ,10 :\ ES PROTE I:'\ Hh l'
`
`665
`
`I c l
`
`F 1<:. I .
`
`BIOEPIS EX. 1125
`Page 7
`
`
`
`(j(j(i
`
`W . FL' REY
`
`( d )
`
`FIG. I . ~te reosco pi c pmjeet ions of some typ ic-al resid ues sup€' rim posed on th e 1·6 A electron density
`map illust rati ng t he fi t . T he electron density contours begin a t 0·6 e/ A 3 a nd increase in steps of 0·6 e/ A3
`•
`Note t ha t a ll rings, even p rolines. ha ve " holes' ' in the ir centers. Also note the electron density for a
`wa t er mo lecule near the 0 H group in (d ). Coefficients for t he elect ron densit.v synthesis were
`P0 exp (icf>cl with the ill ustrated atoms omit ted from th e phasing.
`
`descri pti ons of the immunoglobulin fold (A mzel & P oljak, 1979 ; Davies et al.,
`1975), both t hree and four-strand structures a re men t ioned , but with definition
`oppos ite to those g iven a bove. Us ually, strands 1, 2, 6 and 7 are referred to as the
`four-strand sheet wi t h strands 3, 8 and 9 forming t he t hree-strand sheet. As wiJI
`be shown later, t he hyd rogen-bond ing pattern in Rh e suggests t hat strand 1 is
`much more closely related to sheet B, and t hat it should not be considered a
`member of t he " four-strand " sheet (A). Surp risingly , t he corresponding hyd rogen-
`
`.. .. .. .. .. .. .. .. .. .. .. .. ..
`
`Ut
`
`ttl
`c
`
`0 "' ::2:
`
`. ..
`
`t:t t
`
`I
`
`t
`l
`t tt tt
`
`I
`
`t UU
`
`I
`
`I
`
`I
`
`I
`
`l t't l
`t tt t
`t t#t
`
`fUt
`
`. '() ..
`..
`...
`...
`...
`"•
`I ::.:;::lilt
`: :::~ ;
`~) ::: ttll t ,;,:::::. ::::::(::;; : ::::::: I I It lilt ::~ :: .~:::::Et [~::;::::::
`:
`::; ::: :::l: I :::::::::;:;::;~:::: I: : ::::::;::: : ;:: : ~:::;:::::: ::::::~=~~:::::::::::~
`
`fttt; tt ttt;
`
`ttt"t:tft t t tttttt r t:ttt I
`
`I ttt•·tt Ill
`
`lltttt t ,t. t t**** ' ****** f t
`
`f ttftttt'tf** '
`
`t• t l. I .tt't l * 't H ***ttH •ttttt::t l t«f*** l ******* ... tH l I .ttt.tt •******** I ******E**f ************ I I [ *ftt:*l****l t.tt: l .tt* .t.H f
`t U I 't: I U- •r :tt t• t.t:•tt$':t.ttt't tH f f".t:UU-I **U *'ttt:tt.t:•••***"«********** I ****** f **f ***********;t l I ( l:****f U:* tE .ttU t*t ,. ;;f
`u .t 1 t • t:tH"tt:t t f*Htt .. ·•••••• r •n:******* *••n·•••••••tuttH****** 1 ******E**E**************E***••t •*' H *****"* , .....
`tu • • 1 ••• r ••u· •••• ••• • ·•*****E***** ' •"'*** tH ***** •••••**'******** 1 ****** E**E**************E" *****E****E******!••••
`!!: :! : :!!~! ::::: !! :::: !!:!!:~;!:!:::;::::::;;:;:!:!!!!!!!!:::::: : ::::!!~!:~:::::::::::;::~!!!!!~:!!!~::::::u•••
`I L VI TCH'I'!!~SG 11'1 001-<V I i Gl I GS f'l In I ;1S tJ SV I WYnn~f·Gt-.tiF'I\lll ~YNI1t L I 'SGVS tlr<FSf'ISt-.Sr; ; !:i AS l A t sr.u:Sfii[AIIYYC~AWN[15L I11 f·i;rGGG I t>.L TV~ GO
`co
`
`F iG. 2. H istogram of mean the r·m a l farto rs (8) for each residue. (*)Main cha in : (!)side cha in ; E.
`main and side-chains eq ual.
`
`1 0
`
`t OO
`
`JIO
`
`.. "
`.. "
`.. "
`,, ,,
`"
`I tt t l
`:::::::i
`: ~==~=~~
`~::::~::: .. [t:: ::!::::r.
`
`$'$ l 1 I
`
`t t• f
`
`!f: l
`
`I l ft t f tt l
`
`1
`
`t"
`
`f U t F:
`
`BIOEPIS EX. 1125
`Page 8
`
`
`
`HTR UC' T ll RE O F HE:\ <' E- ci O :\E S PR O T E I!\ R he
`
`667
`
`0 ·10 0 ·20
`
`0 ·30
`
`0·50
`0 -40
`2 sm.8 / >.
`
`0 ·60
`
`0 ·70
`
`F' iG . 3. Pl ot of R fact or t>rrs us resolu t ion .
`
`ng diagrams for ot her immunoglobulin doma ins (Amzel & P oljak , 1979)
`with t hat for Rh e, yet t he sheet designations a re d ifferent.
`In terms of t he vari a bili ty of a min o acid seq uence, residues 23 to 35 rep resent
`first hy pervari a ble region , while the second a nd third hypervari a ble regions
`
`F1 n. ~ . Rhe monome r. (\-carbon skeleto n.
`
`BIOEPIS EX. 1125
`Page 9
`
`
`
`668
`
`W . F U I{EY 8'l' r1L.
`
`corres pond to residues 51 to 57 a nd 90 to 100, res pectiv ely. All other residue
`positions usually show much less ,·a riation in seq uence for ..\-type Y L t domains.
`
`(b ) JVlain chain torsion angles and t11rns
`In Figure 5 th e d istribution of r/J , if; a ngles is shown in the usual Ramachandran
`to Ramakrishnan &
`plot , with
`"allowed"'
`regions ma rked
`according
`.Ramachandran (1965). The most str iking feature of th e plot is the small number
`of non -glyc ine
`residues outside a llowed
`regions.
`indicating a very stable
`conformation. There a re on ly three violations, with exten uating circumstance
`exp laining the occ urrence of eac h. For example. th e abnormal if; ,-al ue for Asp28 is
`read ily expla ined wh en neighboring
`residues are also considered. F rom
`F igure 6(a), it is clear that any change in if; wou ld weaken hydrogen-bonding
`interactions with the main chain ~H groups of Re r3l and Asn32. A change in ~
`here wou ld a lso affect th e direction of the NH bond of Ile29. Since this group i
`already ideall y oriented for hydrogen bond format ion to the side-chai n carboxyl
`gro up of Asp28 , any change co uld only destabi lize the system. There are even
`more compell ing reasons for the unusual angles in Asn52. From Figure 6(b), it is
`clear that any change in rP wou ld weaken the hy drogen-bond interactions between
`NH of Asn52 and 0 of Val3+, and between main chain a toms 0 of Tyr5l and NH
`of Asp53. Likewise, any change in if; would weaken t he hy drogen bond between
`main chain 0 of Asn52 and water oxygen 0 11 8. A change in if; would also
`weaken th e NH-0 hydrogen bond already mentioned between Asp53 and Tyr5l.
`
`1 80 ~~~==~~~--~r----------------,
`
`G
`
`120
`
`60
`
`-60
`
`-120
`
`-180
`
`,:·········
`
`<\
`'
`
`* D28
`
`G
`,. ............................................. ,
`- 120
`-60
`
`G
`
`0
`
`60
`
`120
`
`180
`
`"' ( 0)
`F1< :. 5. The Ra marhandra n plot for Rhe . G indi rates g lyrine residues.
`t AbbreYiation used: V L• va ri ab le domain of lig ht cha in .
`
`BIOEPIS EX. 1125
`Page 10
`
`
`
`ST R l ' ("ITRE OF B E l'\ CE - J OKE fi PR OTEIK Rh e
`
`669
`
`(a)
`
`~ 1 17
`
`~ I
`
`'
`
`e0 U 8_
`
`(b)
`
`FI G. 6 . E1w ironments around (a) Asp28 and (b) Asn52.
`
`he only othe r abn orm al conform ation occurs for ~e r2 . Sin ce th e t orsion a ngles
`for er2 are likely t o be ad ve rsely affected by the disorder in residue l , a nd in a ny
`event are just slight ly outs ide of a n allowed region, no furth e r comment is
`required .
`. In Table 3 t he ten sharp turns in Rh e a re classified according to t he sc heme
`llltroduced by Venk a t acha la m (1968) , a nd extended by Lewis et al. (1973 ). The
`fact that all three turns of t ype II have Gly in position 3 is in excellent ag ree ment
`\Vith th e findin gs of Ve nk atacha la m (1968) a nd Cra wford et al. (1973 ). They claim
`that for a 1-4 hydrogen bond to remain intact in a type II bend , position 3 must
`be glycine. In a more recent study on t he freq uency of occ urrence of each a mino
`in 29 proteins, Chou & F as ma n (1977 ) found
`acid in vari ous positions of fl-turn
`that the most frequ entl y occurrin g a mino ac id residues a re Pro, Gly, Asn , Asp and
`Ser. The stru ct ure of Rhe is consistent wi t h th eir findings in th a t of the 40
`residues inv olved in t he ten turns, 23 (57% ) conform to th ose fi ve amino acid
`
`BIOEPIS EX. 1125
`Page 11
`
`
`
`670
`
`w . FU REY ETA/,.
`
`T.-\BLE 3
`{3-T urn pammeters
`
`Residue no.
`
`:-;eq uence
`
`r/>2 (0)
`
`.p2 n
`
`cf>3 (")
`
`</;3 (0)
`
`d C" I- Ccx4
`(Al
`
`rl 0 1- :\4
`(A )
`
`2- 5
`13- 16
`40-43
`51 - M
`56- 59
`60--63
`68- 71
`8Q-83
`93- 96
`94- 97
`
`H\ 'LT
`TPGQ
`\ ' PGK
`YNDL
`PHGV
`SDRF
`HGTS
`EHED
`XDSL
`DSLD
`
`- 76
`-50
`- 55
`70
`-57
`-72
`5R
`-53
`-65
`-81
`
`-47
`130
`138
`- 46
`142
`3
`-117
`- 37
`-25
`-36
`
`- 104
`95
`79
`- 143
`79
`-72
`- 104
`-64
`-8 1
`-86
`
`146
`- II
`I I
`12
`-6
`-I
`15
`- 14
`-36
`-2 1
`
`6·27
`5·93
`5·7 1
`5·86
`6·04
`6·10
`5·60
`5·49
`5·34
`5·00
`
`5·39
`3·13
`3·14
`4·10
`3·38
`3· 19
`3·20
`2·91
`3·17
`3·72
`
`Class
`
`I(NI)
`II
`11
`IV
`II
`III
`II'
`Ill
`III
`I
`
`Nl, non -idea l. i.e. one angle off by more tha n 50°.
`
`types. There is only one a bnormal tum (ty pe IV) , which is the result of the
`unusual torsion angles for Asn52 disc ussed abO\'e. In this case, th e usual 1-4
`hydrogen bond is replaced by a l-3 hydrogen bond. The only " non -ideal " entry
`(residues 2 to 5) is probably still an a rtifact of the disordered N-terminal residue.
`
`(c) Secondary str-ucture
`The second a ry stru cture of Rh e is almost exclusively {3-sheet, and therefore Rhe
`is classified as a {3-protein according to the scheme of Levitt & Chothia (1976).
`Sheet A (strands 2, 6 and 7) , along with strands 1 and 5, form s one half of a {3-
`barrel. Th ese five strands make up appmximately one-half of the molecule and
`the corresponding main chain atoms are shown in Figu re 7(a). From the Figure, it
`is obvious that the sheet is quite regular, with no {3-bu lges of the types reported
`by Ri chardson et al. (1978) . There are few hydrogen bonds from th e inner strands
`to strands l or 5. Indeed . most of the carbonyl and NH groups in strands 1 and 5
`are nearly perpendicular to the plane of sheet A. This is particula rly th e case for
`strand 1, with Thr5 and Thrl3 being exceptions. The hydrogen bond to Th r13 is a
`consequence of the {3-turn in volving residues 13 to 16. In stra nd 5, even if t he 1H
`and carbonyl groups were ideally oriented , hydrogen bonds would not exist as this
`strand is considerably below the plane of sheet A. Note that strand 5 consists
`almost exclu sively of members of the second hyper variab le region , hence
`interactions between th ese residues and strand 6 are likely to vary considerably
`within VL domains from different som ces.
`Analysis of sheet B (strands 3, 4, 8 and 9) is cons iderably more complicated.
`With the exception of th e short helical region a nd a fe w turns linking to sheet A,
`these four strands complete the monomeric subunit of Rhe. The main chain atoms
`for this half of the molec ule are shown in Figure 7(b). The most obvious features
`a re the regular {3-type hydrogen bonds between the two central strands, and the
`" wide" {3-bulge (Richardson et al., 1978) to the left involving residues Glyl03 ,
`Glyl04 and Ty r88.
`The ri ghtmost trand (strand 4) in clud es a " classic"' {3-bulge in volving residues
`
`BIOEPIS EX. 1125
`Page 12
`
`
`
`\•''"· 7 . (•) St•~d• 5, 6 , 7, 2 ""'
`
`(a l
`
`the other half. (c) Helical region .
`
`( b )
`
`( c )
`
`( ''""' left 00 <ight) w m p<i•log ,.,e. h• lf of the fthe moo om" . (b) '""'"" 0. 8. 3 "od ' (f••m left "' ''• " ) '""''"'"'"'
`
`BIOEPIS EX. 1125
`Page 13
`
`
`
`672
`
`W. FUREY ETA 1 ...
`
`Leu-t8 , lle49 and Trp36. In similar mgions of Bence-Jones proteins Rei and
`McP C603 , a classic bulge was detected by Ri chardson et al. (1978) , in volving
`residues Le u47 , lle48 and Trp35. There is another bulge furth er down strands 3
`and 4 after they turn away from sheet B. The residues in volved a re Ala4-t, Pro45
`and Gln39, but it is not clear whether this bulge should be classified as wide or
`classic. Note in Fig ure 7(b) that as strands 3 and 4 proceed downward , they also
`move out, ending up nearly perpend icul a r to sheet B. It is this feature that
`ultimately determines the mode of association of variable domain mon omers into
`a dim er.
`In terms of th e variability of amino ac ids, nearly all r·esidues shown in the ~
`sheet portion of Figure 7(b) are generall y considered conserved. The loop shown in
`the upper left containing residues 90 to 100, however, is made up exclusively of
`the third hyper variable region. The only other feature with
`residues from
`considerable seconda ry structure is the short helical region (residues 25 to 32)
`shown in Figure 7(c). Note that the region is made up exclusively of members of
`th e first hypervariable region . Although th is segment is obviously helical, the
`distan ces between ex-carbons i and i + 3 are considerably large r than usually found
`in helices. For the five co nsecutive tetrapeptides beginning with Ser25. the mean
`for helical
`Ccxi-Ccxi+ 3 distan ce
`is 5·80A. The mean Co:i- C!Xi+ 3 distances
`tetrapeptides in myoglobin , lysozyme and pancreatic trypsin inhibitor are 5·15,
`5·33 and 5·40 A, respectively (Chou & Fasman , 1977) . The extension of th e helix is
`apparently req uired to accommodate the unusual angles for Asp28 discussed
`ea rlier. This is evident from an examination of the main chain torsion angles for
`residues 26 to 31 , the inner residues of the helix . If Asp28 is excluded , the mean r/1,
`1/J values for the fiv e residues are -73° and -28°, respectively, which are close to
`the normal helical values of 4> ""-60° and 1/J """-30°. Therefore, if a tetrapeptide
`did not include Asp28, one would expect to observe normal helical Co:i
`i+3
`distan ces.
`The complete Rh e monomer is generated by placing the two large segments
`shown in Figure 7 back to back connecting th em with th e appropriate ,8-turns.
`The short helical segment co nnects strand 2 to strand 3, and therefore also
`connects sheet A to sheet B . The interaction between sheets is furth er sta bilized
`by a disulphide bridge between Cys22 (strand 2, sheet A) and C'ys89 (strand 8,
`sheet B) , by hydrophobic interactions in th e co re of the protein , and by a salt(cid:173)
`bridge between Arg62 (strand 6, sheet A) and Asp83 (strand 8, sheet B ). Strand 1
`also in teracts with strand 9 (sheet B ), formin g a seam between the two large
`reg ions of ,8-structure. This is ev ident in Figure 8, where th e co mplete main chain
`hydrogen bonding pattern is shown. Apparently , this is the reason why strand 1
`interacts only weakly with sheet A, and also explains th e carbonyl and NH
`orientations in strand 1 mentioned above. The skewed hydrogen bonds in Figure 8
`result from the fact that these two strands a re parallel. Of the nine strands in
`Rhe, this is the only occurrence of ,8-type interactions between parallel strands :
`all other interacting strands run in opposite directions. Note the kink in strand 1
`caused by the Pro-Pro seq uence for residues 7 and 8. In Bence-~Tones protein Rei
`(Huber & Steigema nn, 1974), re idue 8 is also Pm , but was found to be in the cis
`conformation. All eight pmline residues in Rhe are in the trans conformation.
`
`BIOEPIS EX. 1125
`Page 14
`
`
`
`STR l l(''f l ' RE OF' BE:\C'E- JO:'\ES PROTE I:\ Rhe
`
`673
`
`F'IG. 8. Schematic diagram show ing main chai n hydrogen bond ·.
`
`unlike in the Re i structure , there is no evidence of disorder in the disulphide
`in Rhe. The
`- S distance is 2·018 A with Cp-8y- Y angles of 103·4° and
`. The Cp-S-S- Cp torsion angle is 87·4°.
`
`(d) Hydrogen bonding
`If we temporari ly restrict our attention to only one monomer, then the pmtein-
`. hydrogen -bonding interactions can be separated into three categories;
`chain- main chain , main chain- side chain , and
`ide chain- side chain.
`• H\•dr,DIYfm bonds were included in Figure 8 if the N- 0 distance was less than
`A, the N- H- 0 angle greater than 120°, and the C-0- H angle greater than
`· Although these criteria are somewhat arbitrary , they are reasonable in t hat
`strong hydrogen bonds will certainl y be detected , all totally unrealistic ones
`will be rejected , yet a considerable margin for error is prov ided for. Most
`bonds are in fact quite normal , as indicated by mean N- 0 distances and
`- H- 0 angles of 2·990 A and 160°, respectively . It should be pointed out that
`although protein stereochemistry wa restrained during the refinement process, no
`restraints were included for any of the hydrogen bond inter-actions. Therefore , the
`hydrogen bond parameters represent unbiased stru ctural observations. There are
`57 main chain- main chain , 21 main chain- side chain , and ll side cha in- side cha in
`hydrogen bonds stabilizing the Rh e monomer. The dominance of main chain(cid:173)
`lllain cha in in teractions is not surprising , considering that two ,8-sheets constitute
`lllost of the molec ule.
`For the main chain- side chain hydmgen bonds, the mean donor- acceptor
`distance decreases to 2·908 A and the tandarcl de,·iation rises to 0·213 A. This is
`rnainly the result of in cluding three hyclmgen bori.ds with oxygen ato ms as both
`the donor and acceptor. If the e three terms are neglected , the mean distance rises
`
`BIOEPIS EX. 1125
`Page 15
`
`
`
`()74
`
`w F REY ET~ /, .
`T AB LE 4
`Side clwin- s1:de chain hydrogen bonds
`
`Donor
`
`Acceptor
`
`TI050G I
`S770G
`S250G
`N930Dl
`D940 D2
`D830Dl
`D830D2
`N520DI
`D280Dl
`1'>930DI
`E8~0E2
`
`Q6NE2
`R17NH2
`T270GI
`T270GI
`N32ND2
`R62NH I
`R62N H2
`K67~Z
`N93ND2
`S950G
`Q l1 2i'\E2
`Mean
`Standard deviation
`
`d D- A D-H- A C- 0 - H
`n
`(0)
`(A)
`11 4
`171
`1 ~0
`99
`102
`123
`121
`103
`119
`123
`96
`119
`22
`
`3·005
`3· 166
`3·209
`3· 148
`3· 11 5
`2·830
`2·994
`2·9 12
`2·858
`2·682
`3 ·25~
`3·016
`0·181
`
`1 7 ~
`152
`155
`164
`178
`174
`169
`172
`169
`172
`156
`167
`9
`
`to 2·961 A and the stand ard deviation drops to 0·174 A. The adjusted statistics
`then are in reasonab le agreement with the results for main chain- main chain
`hydrogen bonds. The mean donor- H- acceptor a ngle for these in teractions is also
`160°.
`in Table 4), five
`Of the 11 side cha in- side cha in hydrogen bonds (listed
`co rrespond
`to interactions exclusively between hy pervariable residues, and
`therefore are likely to be unique to the Rhe structu re. Of the remammg six
`hydrogen bonds, the most im portant seem to be the two invol ved in the salt(cid:173)
`bridge between Arg62 and Asp83 shown in Figure 9. In addition to connecting the
`two ,8-sheets, the residues in the Figure serve to insul ate the non -pola r core
`residues from solvent. Since these residues are not from hy pervariable regions,
`this feature may be common to all V L domains.
`Although the mean donor- acceptor distances a nd donor- H- acceptor angles
`are simila r in all three categories, the " acceptor angles" 0-0 - H show a marked
`tendency to decrease as the number of main cha in atoms in the hydrogen bond
`decreases (146°, 132° and 119° for main chain- main cha in , main chain- side chain
`and side cha in- side chain interactions, respectively). This is probably the result of
`increased fl ex ibility in the side-chains, enabling the more favorable acceptor
`
`FIG. 9. Salt-bridge between Arg62 and Asp83.
`
`BIOEPIS EX. 1125
`Page 16
`
`
`
`::;TRUC'TURE OF BENC'E - JOXE::; PROTEI.~\ Rh e
`
`675
`
`F'J<:. 10. All non-polar res idues shown relati,·e to