`
`II!ALTH Sei£NCB
`
`CIENCE
`
`(
`IVf.1'1tTY ;"lr. \'I I.., " 'IJ ';.' fl ~
`$3.00
`
`25 MARCH 1988
`VoL. 239 • PAGES ~~ R I17q ,~
`
`BIOEPIS EX. 1068
`Page 1
`
`
`
`SciENCE
`
`25 MARCH 1988
`VOLUME 239
`NUMBER-484-7
`
`Amerlclln Auocllltlon tor the Advancement ot Science
`Science serves its readers as a forum for the presentation
`and discussion of important issues related to the advance(cid:173)
`ment of science, including the presentation of minority or con(cid:173)
`flicting points of view, rather than by publishing only material
`on which a consensus has been reached. Accordingly, all ar(cid:173)
`ticles published in Science---<ncluding editorials, news and
`comment, and book reviews-are signed and reflect the indi(cid:173)
`vidual views of the authors and not official points of view .
`adopted by the AAAS or the institutions with which the au(cid:173)
`thors are affiliated.
`
`Publleher: Alvin W. Trivelpiece
`
`Editor: Daniel E. Koshland, Jr.
`
`Daputy Editors: Philip H. Abelson (Engineering and Applied
`Sciences); John I. Brauman (Physical Sciences)
`
`EDITORIAL STAFF
`Managing Editor: Patricia A. Morgan
`Alalllant Managing Editor: Nancy J. Hartnagel
`Senior Editors: Eleanore Butz, Ruth Kulstad
`Auoclllt8 Editors: Martha Coleman, R. Brooks Hanson,
`Barbara Jasny, Katrina L. Kelner, Edith Meyers, Phillip D.
`Szuromi, David F. Voss
`~Attars Editor: Christine Gilbert
`Book Ravlewl: Katherine Livingston, editor; Deborah F.
`Washburn
`Thle Week In Sc/4mce: Ruth Levy Guyer
`Contributing Editor: Lawrence I. Grossman
`Chlat Production Editor: Ellen E. Murphy
`Editing Departrnant: Lois Schmitt, head; Mary McDaniel,
`Patricia L. Moe, Barbara E. Patterson
`Copy Dalk: Joi S. Granger, Beverly Shields, Anna Victoreen,
`Barbara Wittig
`Production Managar: Karen Schools
`Alalllant Production Manager: James Landry
`Grsphlca lnd Production: Holly Bishop, James J. Olivarri
`CoVII'I Editor: Grayce Finger
`ManUICI'Ipt Syllllme Anllyll: William Carter
`
`NEWS STAFF
`Nlwl Editor: Barbara J. Culliton
`Nlwl and Comrnant: Colin Norman, deputy editor; William
`Booth, Mark H. Crawford , Constance Holden, Eliot Marshall,
`Marjorie Sun, John Walsh
`~rch Nlwl: Roger Lewin, deputy editor; Deborah M.
`Barnes, Richard A. Kerr, Jean L. Marx, Leslie Roberts, M.
`Mitchell Waldrop
`Europe~~n Correlpondant: David Dickson
`
`BUSINESS STAFF
`BullnHI Stat! Ma~~~g~~r: Deborah Rivera-Wienhold
`Claaltled Adverllelng Supervleor: Karen Morgenstern
`Membership Recrultrnant: Gwendolyn Huddle
`Member lnd Subecrlptlon Recorde: Ann Ragland
`Guldot to Blollchnology Produc:ll and lnllrurnanll:
`Shauna S. Roberts
`
`ADVERTISING REPRESENTAnVES
`DI!'Kior: Earl J. Scherago
`Trllllc Managar: Donna Rivera
`Trllllc Managar (Recrultrnant): Gwen Canter
`Advertlelng Sales Ma~~~g~~r: Richard L. Charles
`Employment Sales Managar: Edward C. Keller
`Marketing Managar: Herbert L. Burklund
`S.lel: New York, NY 10036: J. Kevin Henebry, 1515 Broad(cid:173)
`way (212-730-1050); Scotch Plains, NJ 07076: C. Richard
`Callis, 12 Unami Lane (201 -889-4873); Chicago, IL 60611 ·
`Jack Ryan, Room 2107, 919 N. Michigan Ave. (312-337-
`4973); San Jose, CA 95112: Bob Brindley, 310 S. 16 St. (408-
`998-4690); Dorset, VT 05251 : Fred W. Dieffenbach, Kent Hill
`Rd. (802-867-5581); Damascus, MD 20872: Rick Sommer,
`24808 Shrubbery Hill Ct. (301 -972-9270); U.K., Europe: Nick
`Jones, +44(0647)52918; Telex 42513; FAX (0392) 31645.
`
`lntormlllon tor contributors appears on page XI of the 25
`March 1988 issue. Editorial correspondence, including re(cid:173)
`quests for permission to reprint and reprint orders, should be
`sent to 1333 H Street, NW, Washington, DC 20005. Tele(cid:173)
`phone: 202-326-6500.
`
`Advertising correspondence should be sent to Tenth Floor,
`1515 Broadway, NY 10036. Telephone 212-730-1050 or WU
`Telex 968082 SCHERAGO.
`
`25 MARCH 1988
`
`Women in Science
`
`T he threat of a serious shortage of scientific personnel looms in the years ahead. Many
`
`predictions are, of course, notoriously unreliable. If a shortage is a realistic scenario,
`however, it is important to find ways to employ underrepresented groups more
`equitably-for reasons of national interest as well as of equality.
`Women are one conspicuously underrepresented group in the higher echelons of
`academia and industry. Records of their transit through the system may help provide clues
`to appropriate remedial actions. Some trends in the data are promising. For example, in the
`1930s women received 7% of the Ph.D. degrees in mathematics and the physical sciences,
`15% of those in the life sciences, and 16% in the social sciences. But by the early 1980s those
`percentages had all doubled. Recently, however, the figures appear to have leveled off.
`Tracing the progress of women through the system shows that the percentages roughly
`parallel those of men for total percentages in science through high school, college, and
`entrance into graduate school. The serious differential in participation occurs at the
`postdoctoral level. For example, 93,000 men and 94,000 women undergraduates were
`majoring in the biological sciences in 1984; the respective graduate enrollments were
`22,000 and 17,000. At the next level, however, women are poorly represented on faculties
`and on average receive lower salaries than do men in comparable positions. One survey
`showed that although women had 10% of the doctoral degrees in chemistry, they had only
`4% of the faculty jobs. At no stage in the educational process is there an indication that the
`attrition is caused by lack of academic performance.
`Attempts to understand the attrition have so far been unsuccessful, but some theories
`seem better than others. In the past, certainly, prejudice from the "old boys" was
`widespread, and it has only been partly eradicated. Moreover, the perception of this
`historical prejudice can be a subtle deterrent in today's more enlightened, but imperfect,
`world. The lack of role models can be a source of insecurity, a point made eloquently by
`Sheila Widnall in her AAAS presidential address. That situation may change as more women
`take important roles in our society, and particularly in science. But the insecurity may be a
`decisive factor during the period between graduate school and tenure, an interval of intense
`competitive pressure. Those who have pedagogical or administrative roles need to be
`sensitive to the stress of the pressured student or the untenured assistant professor. The
`support of a steady friend with encouragement to stay the course and an occasional
`congratulations for work well done can be crucial in developing the self-confidence that is
`essential for a research investigator.
`Words are important, but actions are more so. Important contributions would be
`programs to make it easier for women during childbearing years to continue their
`professional involvements. Several universities have introduced "stop the clock" programs
`that allow women who are raising children to have tenure decisions postponed. Other
`programs, such as half-time appointments, "extend the clock" on grants, or on-site and
`subsidized day care are particularly appropriate (see also Carl Djerassi, Letters, 1 Jan., p.
`10). Women not only bear the children, they are the prime organizers of their upbringing,
`and in these years they need a special form of encouragement. Since equality of responsibil(cid:173)
`ity is not yet here, not only are the demands on women faculty members greater, but they are
`more subject to criticism. A man who does less teaching because he serves on editorial
`boards is excused as normal, whereas a woman who asks to do less teaching to help raise a
`child is viewed as a burden. Today there is less prejudice at the time of promotion, but
`obstacles confronted before tenure decisions are sufficient to discourage a significant portion
`of talented women scientists.
`Although the problems for ethnic groups are not the same as those that women face,
`they have some of the same characteristics. There are relatively few role models, and the need
`for encouragement of pioneers in potentially hostile territory is real.
`As the country expands into an ever-increasing technological base, the need for women
`and minorities in both academia and industry increases proportionally. It may cost some
`money, some effort, and some understanding, but the voyage to full equality can be even
`more exciting and worthwhile than the voyage into space.-DANIEL E. KosHLAND, JR.
`
`EDITORIAL 1473
`
`BIOEPIS EX. 1068
`Page 2
`
`
`
`34. B. R . Cullen, ibid. 46, 973 (1986) .
`35. P. E. Pellett, K. G. Kousoulas, L. P. Pereira, B.
`Roizman,J. Virol. 53, 243 (1985).
`36. J. D . Dignam, R. M. Lebovitz, R. G. Roeder,
`NudeicAcids Res. 11, 1475 (1983) .
`37. Supported by the National Cancer Instirute (grants
`CA08494 and CA19264), the National Instirute for
`Allergy and Infectious Diseases (grant AI124009) ,
`
`the United States Public Health Service, and the
`American Cancer Society (MV2W). N.M., T .M.K.,
`and D.S. were predoctoral trainees ofUSPHS train(cid:173)
`ing grants AI07099, CA09241, and GM7281, re·
`spectively.
`
`23 October 1987; accepted 31 December 1987
`
`Reshaping Human Antibodies: Grafting an
`Antilysozyme Activity
`
`MARTINE VERHOEYEN, CESAR MILSTEIN, GREG WINTER*
`
`The production of therapeutic human monoclonal antibodies by hybridoma technolo(cid:173)
`gy has proved difficult, and this has prompted the "humanizing" of mouse monoclonal
`antibodies by recombinant DNA techniques. It was shown previously that the binding
`site for a small hapten could be grafted from the heavy-chain variable domain of a
`mouse antibody to that of a human myeloma protein by transplanting the hypervaria(cid:173)
`ble loops. It is now shown that a large binding site for a protein antigen (lysozyme) can
`also be transplanted from mouse to human heavy chain. The success of such
`constructions may be facilitated by an induced-fit mechanism.
`
`contrast, the three-dimensional structure of
`the complex of lysozyme and the mouse
`antibody Dl.3 has been solved (6), and
`about 690 A2 of the solvent-accessible sur(cid:173)
`face of the antibody is buried on complex
`formation. Both V H and light-chain variable
`(V L) domains make extensive contacts to
`lysozyme, but most of the hydrogen-bond(cid:173)
`ing contacts are made to the CDRs of the
`heavy chain. Thus, of 12 hydrogen bond
`interactions proposed (6), 9 are made to the
`heavy chain. We have replaced the hypervaria(cid:173)
`ble loops of the human NEW heavy chain (5)
`with those from the Dl.3 heavy chain.
`The variable domains of the mouse anti(cid:173)
`body to lysozyme were cloned and se-
`
`a
`H
`
`lg pro
`
`e
`E EH
`
`•
`
`SB
`
`F OR PASSIVE IMMUNITY OR ANTIBODY
`
`therapy in humans, monoclonal anti(cid:173)
`bodies designed to eliminate toxins,
`viral and bacterial pathogens, or other cells
`would ideally be of human origin (1). Un(cid:173)
`fortunately it has proved difficult to make
`human monoclonal antibodies by hybrido(cid:173)
`ma technology (2) . Chimeric antibodies
`with mouse variable and human constant
`domains have been constructed by linking
`together the genes encoding each domain
`(3, 4), and expressing the recombinant anti(cid:173)
`bodies in myeloma cells. However, the
`mouse variable region may itself be seen as
`foreign (1). We have therefore attempted to
`insett the antigen-binding site of a mouse
`antibody, rather than the whole variable
`region, directly into a human antibody. In
`previous work, the three heavy-chain hyper(cid:173)
`variable regions [or complementarity-deter(cid:173)
`mining regions (CDRs)] from a mouse anti(cid:173)
`body to a hapten were transplanted onto the
`framework regions of the heavy-chain vari(cid:173)
`able (V H) domain of a human myeloma
`protein. In combination with the mouse
`light chain, the reshaped heavy chain bound
`tightly to hapten (5). Although it seems
`likely that both heavy and light chains make
`contacts to the hapten, the relative contribu(cid:173)
`tion of each chain is unknown. Nor is it clear
`whether the small hydrophobic hapten
`NP ( 4-hydroxy- 3-nitrophenacetylarninocap(cid:173)
`roate) simply binds to a hydrophobic pocket
`at the base of the hypervariable loops. By
`
`Medical Research Council Laboratory of Molecular Bioi·
`ogy, Hills Road, Cambridge CB2 2QH, England.
`
`*To whom correspondence should be addressed.
`
`153+
`
`quenced as described (7). To reshape the
`NEW heavy chain, we started from a syn(cid:173)
`thetic gene in an Ml3 vector (Fig. lb)
`containing the framework regions of human
`NEW and the CDRs from mouse antibody
`Bl-8 (5). Long oligonucleotides with multi(cid:173)
`ple mismatches with the template ( 8) were
`used to replace each of the hypervariable
`loops in turn by site-directed mutagenesis:
`the central mismatched portion of the prim(cid:173)
`er encoded each CDR of the Dl.3 heavy
`chain, and the 5' and 3' ends of the primer
`were complementary to the flanking frame(cid:173)
`work regions. Thus after three rounds of
`mutagenesis, the reshaped gene (HuV~.
`LYS) encoded the framework regions of
`NEW with the hypervariable regions of
`Dl.3 (Figs. lc and 2). This was assembled
`with the heavy-chain constant region of
`human immunoglobulin G2 (HulgG2) (9)
`to give the plasmid pSVgpt-HuVHLYS(cid:173)
`HuigG2. The plasmid was transfected by
`electroporation (10) into the myeloma line
`J558L (11), which secretes a mouse A light
`chain. Transfectants resistant to mycophen(cid:173)
`olic acid were screened for secretion of
`immunoglobulin by ~el electrophoresis of
`supernatants from [3 S]methionine-labeled
`cells. The sea-eted product (Hu V HLYS(cid:173)
`HuigG2, A) was purified on protein A(cid:173)
`Sepharose; the A light chain was exchanged
`for the Dl.3 K light chain in vitro; and the
`HuVHLYS-HuigG2, K antibody was puri(cid:173)
`fied (12). In parallel experiments as control,
`the mouse Dl.3 variable region (MoVH(cid:173)
`LYS) was attached to the heavy-chain con(cid:173)
`stant region of mouse IgGl (MolgGl) in
`a pSVgpt vector
`(pSVgpt-MoVHLYS(cid:173)
`MolgGl), and antibody was expressed and
`reassociated as above (Fig. ld).
`Fluorescence quench was used to measure
`
`SCIENCE, VOL. 239
`
`j+ . W - W
`
`L
`
`HuVHNP
`
`Fig. 1. Vectors for heavy-chain expression: (a)
`Mouse VNP gene (4) and designated here as
`MoVHNP, (b) reshaped HuVNP gene (5) desig(cid:173)
`nated here as HuVHNP, (c) HuVHLYS gene
`Hb
`with the heavy-chain constant region gene of
`I
`human IgG2, (d) MoVHLYS gene with the
`lg pro CDRs 1
`c
`heavy-chain constant region gene of mouse IgGl ,
`H
`L
`HuVHLYS
`B B CH1
`CH2 CH3
`B
`and (e) the backbone of the pSV gpt vector with
`
`1 : - ' ~,.-t·~~·[][• :::rl!• :rrl•=I1!• I : :• ::J---'' ~
`immunoglobulin heavy-chain enhancer (Igh enh)
`lg pro CDRs 1
`2
`3
`HulgG2
`d
`(4). The VH domains are denoted in black, stip-
`pled, or open boxes to signify sequences encoding
`~•:-;:.,. ... ~ .... IITITI!illM!illoviiiT"iiiTLYill!sDI:I----Il~~ ~ CH1
`CH2 CH3
`-:olgG1 mouseantibodytoNP(Bl-8), mouseantibody to
`•
`lgpro
`•
`lysozyme (Dl.3), and human myeloma protein
`(NEW), respectively. The constant domains are
`denoted in black or open boxes to signify se(cid:173)
`quences encoding mouse or human constant do(cid:173)
`mains. Ig pro, heavy-chain promoter; L, leader
`exon; H, Hind III; B, Bam HI; S, Sac I; E, Eco
`Rl restriction sites. The reshaped NEW heavy chain was expressed from a vector derived from the
`pSV gpt-Hu V HNP vector (b) . Thus the Hu V HNP gene, cloned initially in M 13mp9 as a Hind III-Bam
`HI fragment, contains heavy-chain promoter, the CDRs of the Bl-8 antibody, and the framework
`regions of human NEW (5). The CDRs of Bl-8 were then replaced with those of Dl.3 using long
`mutagenic oligonucleotides (see Fig. 2). The Hind III-Bam HI fragment, now carrying the HuV HLYS
`gene (c) was excised from the Ml3 vector and cloned into the pSV vector (e) along with a Bam HI
`fragment encoding the heavy-chain constant region of human IgG2 (9). The construction of the
`pSVgpt-MoVHLYS-MoigGl vector to express recombinant Dl.3 is summarized in (d).
`
`BIOEPIS EX. 1068
`Page 3
`
`
`
`~~~~~~~ .
`
`ATGCAAATCCTCTGAATCTACATGGTAAATAiAGGTTTGTCTATAC
`
`RNA starts
`....._.
`~ RNA starts
`CACAAACAGAAAAACATGAGATCACAGTTCTCTCTACAGTTACTGAGCACACAGGACCTC
`
`t V A T A T )
`E
`( M G W $
`C
`ACCATGGGATGGAGCTGTATCATCCTCTI'CTTGGTAGCAACAGCTACAGGTAAGGGGCTC
`
`~j 1al I
`
`ACAGTAGCAGGCTTGAGGTCTGGACATATATATGGGTGACAATGACATCCACT'M'GCCTT
`
`10
`5
`1
`sianal
`( G V H S ) Q V Q L Q E S G P G L V R
`T=CCACAGGTGTCCACTCCCAGGTCCAACTGCAGGAGAGCGGTCCAGGTCTTGTGAGA
`
`CORl
`30
`25
`20
`15
`S ~
`0 T L S LTC TV S G S T F
`S
`P
`CCTAGCCAGACCCTGAGCCTGACCTGCACCGTGTCTGGCAGCACCTTCAGC =ATGGT
`
`A
`
`primer II
`
`Fig. 2. Sequence of the reshaped
`V H domain. The reshaped domain
`is
`based on
`the
`HuVHLYS
`Hu V HNP gene (4), with the frame(cid:173)
`work regions of human NEW alter(cid:173)
`nating with the hypervariable re(cid:173)
`gions of mouse Dl.3 (7) . Three
`oligonucleotides I, II, and ill were
`used to replace each of the B l-8
`CDRs with those from Dl.3. Each
`oligonucleotide has 12 nucleotides
`at the 5' end and 12 nucleotides at
`the 3' end which are complemen(cid:173)
`4~ E w I G 1'@ yrlRa G 1
`4~ p G R G
`tary to the NEW framework re- ~ w v R
`0
`giOnS, whereas the Central portion
`GTAAAC TGGGTGAGACAGC~~~CGAGGTCTTGAGTGGATTGGA ATGATTTGGGGT
`encodes CDRs l, 2, or 3 of the
`55 coR2
`6o
`65
`10
`Dl.3 antibody. The primers are
`complementary to the marked por-
`tiOnS Of the coding Strand Of the
`reshaped domain. Three rounds of
`mutagenesis were used, transfect-
`ing intO the Escherichia co/i Strain
`BMH7l-l8mutL, plating on a
`lawn of BMH7l-l8, and probing
`infected colonies with the mutagen-
`ic primers as in (8). ln washing the
`filters, it WaS necessary tO gradually
`GTC TGT GTT TCC ATC ACC CCA AAT CAT TCC AAT CCA CTC 3', III 5 ' GCC TTG
`increase the wash temperature from
`ACC CCA GTA GTC AAG CCT ATA ATC TCT CTC TCT TGC ACA ATA 3'
`65° to 75•c to distinguish the mutant from wild-type clones. The yield of mutants at each step was
`about 5%.
`
`1 ~TJl,.JCA&,JCTlTJTTb.cbc~akT2c1 A~~GA~~~~A~c
`c
`82
`85
`80
`75
`8
`T A V
`S V T A A 0
`S
`L R L
`S
`S K N O F
`T
`ACCAGCAAGAACCAGTTCAGCCTGAGACTCAGCAGCGTGACAGCCGCCGACACCGCGGTC
`9~ , c A R
`w G 1o~ G
`5
`1
`1
`0
`TATTATTGTGCAAGA GAGAGAGATTATAGGCTTGACTAC TGGGGTCAAGGCAGCCTC
`no
`primer III
`v T v s s
`BamHI
`GTCACAGTCTCCTCAGGT . • ••• • 3 ,
`
`9£ R 9)"-\ R' o~
`
`,
`
`L
`
`~~·~i~~r~r3.~r~1 \~'CAc;GT~~ ~Tg:{' r:;: ;:;c,;.g\~\'{;.\TT ATA
`
`Fig. 3. Competition of reshaped and engineered
`mouse antibody for lysozyme. The wells of a
`rnicrotiter plate were coated with lysozyme at a
`concentration of 100 j.Lg/ml in phosphate-buff(cid:173)
`ered saline (PBS), and remaining binding sites on
`the plastic were blocked with l% bovine serum
`albumin (in PBS). The binding of 1251-labeled
`mouse antibody Dl.3 (specific activity, > 0.6 j.LCi
`per micrograms of Dl.3) to lysozyme was com(cid:173)
`peted with increasing concentrations of (• ) engi(cid:173)
`neered antibody (MoVHLYS-MolgGl , K) or (D)
`reshaped antibody (HuVHLYS-HulgG2, K), and
`the data presented are typical of two independent
`experiments. Antibody Dl.3 and reassembled
`antibody Dl.3 competed with the labeled Dl.3 as
`effectively as did the engineered antibody Dl.3.
`Radioiodination was carried out by the dliora(cid:173)
`mine T method (22).
`
`~ 80
`"' c: :;;
`c: :c
`~
`~
`Qj
`a:
`
`40
`
`0
`
`0.1
`
`Competing antibody (llglml)
`
`10
`
`20
`
`the binding of lysozyme to these antibodies.
`The binding to both antibodies was tight,
`with 2 mol of lysozyme molecules binding l
`mol of antibody. Our results indicate that
`the affinity is better than 5 nM, although
`previous work had indicated a value of 20 to
`40 nM ( 6, 13). The antibodies were then
`compared in a competition assay for lyso(cid:173)
`zyme; 1251-labeled mouse antibody Dl.3
`competing with mouse antibody or with the
`reshaped antibody for binding to lysozyme.
`The reshaped antibody competes, albeit less
`effectively (about tenfold less)
`than the
`mouse antibody (Fig. 3). Nevertheless, t4e
`grafting of hypervariable
`regions from
`mouse to human framework regions is suffi(cid:173)
`cient to transfer the lysozyme-binding site,
`an extensive surface of interaction, despite
`the 31 of 8 7 residues that differ between the
`heavy-chain framework regions.
`The results confirm that the hypervariable
`
`regions (mainly loops) fashion the antigen(cid:173)
`binding site and that the more conserved
`regions (mainly ~-sheet) form a structural
`framework (14). However, the result is re(cid:173)
`markable given the several assumptions un(cid:173)
`derlying the building of an antigen-binding
`site on a new framework region-namely,
`( i) the V H and V L domains pack together in
`the same way (15) , (ii) the two ~-sheets
`within each variable domain pack together
`in the same way (16), (iii) the antigen
`usually binds to the hypervariable regions,
`and (iv) the hypervariable regions are sup(cid:173)
`ported by the framework regions in a similar
`way (17) . In particular, the hypervariable
`loops are not stand-alone structures, but
`make extensive contacts to the framework
`and to other hypervariable regions. Since
`the crystallographic structures of both par(cid:173)
`ent antibodies are known, we could confirm
`that in the reshaped antibody, the Dl.3
`
`25 MARCH 1988
`
`CDRs could be supported by the NEW
`framework regions with the same (or simi(cid:173)
`lar) contacts as are used with the Dl.3
`framework regions. However, we also iden(cid:173)
`tified some potential problems. For exam(cid:173)
`ple, in CDRl of Dl.3, Tyr-32 m*es a van
`der Waals contact with the Pjle-27 in FRl;
`in NEW this phenylalanine is replaced by
`serine, and this contact is therefore lost
`when the Dl.3 CDRs are mounted on
`NEW framework regions. Furthermore, in
`antibody Dl.3 the loop including the end of
`FRl bulges out from the surface, whereas in
`NEW (and other solved strucrures), it is
`pinned back to the surface. Despite these
`problems, the reshaped antibody is able to
`bind lysozyme.
`It is conceivable that the several contacts
`made by the FRl/CDRl loop to lysozyme
`make only a small contribution to the overall
`energetics of binding. Problems in this re(cid:173)
`gion would then have only a slight effect on
`the affinity. It is also possible that the shape
`of the binding site could adjust to the
`binding of antigen (18) and therefore that
`small errors in assembling the CDRs on a
`new framework might be "self-correcting"
`when the antigen binds. The energetic price
`for such putative self-correcting induced fit
`would be low, given the small loss in bind(cid:173)
`ing affinity suggested by the competition
`data of Fig. 3. It is not possible to deduce
`the exact change in affinity from the compe(cid:173)
`tition data, as the competition may only
`reflect the relative on-rates. However, even
`if the difference in affinities were tenfold,
`this would correspond to no more than the
`loss of a single hydrogen bond interaction
`(19) . The shape of the antigenic epitope
`could also adjust to the binding of antibody
`(20), although no structural change in lyso(cid:173)
`zyme was detected in its complex with the
`mouse Dl.3 antibody (6). Furthermore, the
`whole surface of interaction could reorien(cid:173)
`tate slightly, perhaps by rocking on side
`chains (21 ), to make a set of new contacts.
`This is an attractive model for large surfaces
`making multiple interactions. Thus self-cor(cid:173)
`rection could ocqJ.r at the level of the anti(cid:173)
`body, the antigen, or the complex.
`In conclusion, the structure of antibodies
`may incorporate features that assist the
`grafting of hypervariable regions. The criti(cid:173)
`cal features are that framework contacts are
`largely conserved between V H and V L do(cid:173)
`mains, between the sheets within a domain,
`and to the CDR 'loops. In addition, the
`shape of the antigen binding site or the
`antigenic determinant (or both) might be
`able to adjust to each other in an induced-fit
`mechanism.
`REFERENCES AND NOTES
`l. R. A. Miller, A. R. Oseroff, P. T. Stratte, R. Levy,
`Blood 62, 988 (1983).
`
`REPORTS 1535
`
`BIOEPIS EX. 1068
`Page 4
`
`
`
`2. D. A. Carson and B. D. Freimark, Adv. Immunol.
`38, 275 (1986).
`3. G. L. Boulianne, N. Hozwni, M. J. Shulman,
`Nature (London) 312, 643 (1984); S. L. Morrison,
`M . J. Johnson, L. A. Herzenberg, V. T . Oi, Proc.
`Nat/. Acad. Sci. U.SA. 81, 6851 (1984).
`4. M. S. Neuberger et al., Nature (London) 314, 268
`(1985).
`5. P. T . Jones, P. H. Dear, J. Foote, M. S. Neuberger,
`G. Winter, ibid. 321, 522 (1986).
`6. A. G. Amit, R. A. Mariuzza, S. E. V. Phillips, R. J.
`Poljak, Science 233, 747 (1986).
`7. M . E. Verhoeyen, C. Berek, G. Winter, in prepara(cid:173)
`tion.
`8. P. Carter, H . Bedouelle, G. Winter, Nucleic Acids
`Res. 13, 4431 (1985).
`9. J. Ellison and L. Hood, Proc. Nat/. A cad. Sci. U.S A.
`79, 1984 (1982).
`10. H. Potter, L. Weir, P. Leder, ibid. 81, 7161 (1984).
`11. V. T . Oi, S. L. Morrison, L. A. Herzenberg, P.
`Berg, ibid. 80, 825 (1983).
`12. Clones secreting antibody were grown in 1-liter
`roller bottles, and antibody was purified on protein
`A-Sepharose. The mouse ~ light chain of antibodies
`secreted from the J 558L myeloma was exchanged
`for the D 1.3 K light chain in vitro. For the D 1.3
`antibody and the two recombinant antibodies (i) 1
`
`to 2 mg of antibody (in 1 ml) was clialyzed over(cid:173)
`night at 4•c against O.SM tris, pH 8.0, (ii) inter(cid:173)
`chain clisulfides were reduced with O.lM 2-mercap(cid:173)
`toethanol for 1 hour at room temperature, (iii) the
`free sulfhydryls were alkylated with 0.15M iodoace(cid:173)
`tarnide for 15 minutes at room temperature, (iv) the
`heavy and light chains were fractionated on a Du(cid:173)
`pont Zorbax G250 column in 5M guanicline hydro(cid:173)
`chloride and 20 mM soclium phosphate, pH 8.0.
`The D 1.3 K light chain was refractionated on high(cid:173)
`performance liquid chromatography (HPLC) and
`an aliquot was checked on analytical HPLC to
`ensure no contamination with the D 1.3 heavy chain,
`(v) appropriate heavy chains were mixed with equal
`amounts of D 1.3 K light chain and clialyzed against
`O.lM trisCI, pH 7.4, at 4•c for 2 days, and (vi)
`reassembled antibody was purified on a protein A(cid:173)
`Sepharose column. The reassembled mouse anti(cid:173)
`body (MolgG 1) eluted at pH 6, and the reshaped
`antibody (HulgG2) at pH 4. From 1 mg of anti(cid:173)
`body, the yield of reassembled antibody was less
`than 20%.
`13. M. Harper, F. Lema, G. Boulot, R. J. Poljak, Mol.
`Immunol. 24, 97 (1987).
`14. E. A. Kabat, Adv. Protein Chem. 32, 1 (1978).
`15. C. Chothia, J. Novomy, R. Bruccoleri, M. Karplus,
`]. Mol. Bioi. 186, 651 (1985).
`
`16. A. M. Lesk and C. J. Chothia, ibid. 160, 325
`(1982).
`17. C. Chothia and A. M. Lesk, ibid. 196, 901 (1987).
`18. A. B. Edmundson and K. R. Ely, in Synthetic
`Peptides as Antigens, R. Porter and J. Whelan, Eds.
`(Wiley, New York, 1986), p. 107; E. D. Getzoff et
`al., Science 235, 1191 (1987).
`19. A. R. Fersht et a!:, Nature (London) 314, 235
`(1985).
`20. P. M. Colman et al. , ibid. 326, 358 (1987).
`21. C. Chothia, A. M. Lesk, G. G. Dodson, D . C.
`Hodgkin, ibid. 302, 500 (1983).
`22. L. Hudso n and F. C. Hay, Praaical Immunology
`(Blackwell, Oxford, 1980), p. 240.
`23. We thank R. Poljak, A. Amit, S. Phillips, and R. A.
`Marriuzza for the crystallographic coorclinates of
`D 1.3 lysozyme complex; R. A. Marriuzza, J. Foote,
`C. Chothia, and A. Lesk for cliscussions and com(cid:173)
`ments on the paper; J. Foote for use of the fluores(cid:173)
`cence titration method and associated computer
`programs in advance of publication; and J. Jarvis, P.
`T. Jones, and R. Pannell for technical assistance.
`M.V. is a senior research assistant of the Belgian
`National Fund for Scientific Research and a fellow
`of the Commission of the European Communities.
`
`29 September 1987; accepted 8 February 1988
`
`Peroxisomal Membrane Ghosts in Zellweger
`Syndrome-Aberrant Organelle Assembly
`M. J. SANTOS, T. lMANAKA., H. SHIO, G. M. SMALL, P. B. LAZAROW
`
`Peroxisomes are apparently missing in Zellweger syndrome; nevertheless, some of the
`integral membrane proteins of the organelle are present. Their distribution was
`studied by immunofluorescence microscopy. In control fibroblasts, peroxisomes
`appeared as small dots. In Zellweger fibroblasts, the peroxisomal membrane proteins
`were located in unusual empty membrane structures of larger size. These results
`suggest that the primary defect in this disease may be in the mechanism for import of
`matrix proteins.
`
`Z ELL WEGER SYNDROME IS A DISEASE
`
`in which an entire organelle, the
`peroxisome, appears to be missing,
`as first recognized by Goldfischer et aJ. (1).
`The peroxisome is nearly ubiquitous in eu(cid:173)
`karyotic cells and functions in fatty acid 13-
`oxidation, plasmalogen biosynthesis, cellular
`respiration (H20rforming), gluconeogene(cid:173)
`sis, . bile acid synthesis, and purine catabo(cid:173)
`lism (2) . This human genetic disorder, char(cid:173)
`acterized by profound neurological impair(cid:173)
`ment, metabolic disturbance, and neonatal
`death, has taught us much about peroxi(cid:173)
`some function (3) and promises to teach us
`more about peroxisome assembly. Some
`peroxisomal proteins are synthesized nor(cid:173)
`mally in Zellweger syndrome (4), but they
`are not assembled into peroxisomes. Many
`of these proteins are rapidly degraded, with
`the result that important soluble matrix en(cid:173)
`(4) and
`zymes (catalyzing
`13-oxidation)
`membrane-bound enzymes (catalyzing the
`initial steps in plasmalogen biosynthesis) (41
`5) are missing or seriously deficient, thus
`
`The Rockefeller University, New York, NY 10021.
`
`causing severe metabolic abnormalities (3).
`On the other hand, some peroxisomal en(cid:173)
`zymes (for example, catalase) are present in
`normal amounts in Zellweger cells but are
`located free in the cytosol (41 61 7).
`These defects correlate with the known
`facts of peroxisome biogenesis. All peroxi(cid:173)
`somal proteins investigated thus far are syn(cid:173)
`thesized on free p()lyribosomes and are as(cid:173)
`sembled posttranslationally into preexisting
`peroxisomes ( 8). If the organelle is missing
`(1), newly made proteins will just diffuse
`through the cytosol, unable to enter a per(cid:173)
`oxisome. Under these circumstances, it is
`not surprising that many are degraded.
`The autosomal recessive genetics of Zell(cid:173)
`weger syndrome indicates that a single mu(cid:173)
`tation is responsible for the defects. One
`explanation would be that the mutation
`prevents the assembly of the peroxisomal
`membrane. In this case peroxisomal integral
`membrane proteins (PxiMPs) (9) should be
`absent, since those that have been studied
`are made on free polyribosomes (10-12)
`and are unlikely to be stable if not integrated
`into a membrane. Two other possibilities
`
`were suggested by the unexpected finding
`that several PxiMPs are present in normal
`amounts in Zellweger liver (13). The peroxi(cid:173)
`somal membranes could be assembled in
`Zellweger syndrome, but the import of ma(cid:173)
`trix proteins is defective. This would result
`in empty (or nearly empty) membrane
`ghosts, which would not be recognizable by
`electron microscopy or cytochemistry.
`Alternatively, the PxiMPs, in the absence of
`peroxisomes, might have violated the rules
`of protein sorting and inserted into the
`wrong intracellular membrane(s) .
`To differentiate among these possibilities,
`we analyzed fibroblasts with a polyspecific
`antiserum against rat liver PxiMPs (polyspe(cid:173)
`cific anti-PxiMPs) (12) (Fig. 1A, lane 1).
`This serum detected three human PxiMPs
`with masses of approximately 140, 69, and
`53 kD in control cells (Fig. 1A, lanes 2 to
`4) . These PxiMPs were also present in Zell(cid:173)
`weger fibroblasts (lane 5), and they cosedi(cid:173)
`mented in equilibrium density centrifuga(cid:173)
`tion of Zellweger fibroblast homogenates.
`However, the PxiMPs were found at an
`abnormally low density of 1.10 glcm3 (in(cid:173)
`stead of the usual fibroblast peroxisome
`density of 1.17 g/cm3
`) [lanes 6 and 7; further
`details in (14)]. Several other cell mem(cid:173)
`branes also sedimented in the low-density
`region of the gradient where the PxlMPs
`were found (14). Therefore, the fraction(cid:173)
`ation data are consistent with the possibility
`that the PxiMPs are present in aberrant
`peroxisomal membrane ghosts, but they do
`not exclude erroneous localization in lyso(cid:173)
`somes, the endoplasmic reticulum, or some
`other low-density organelle. Immunofluo(cid:173)
`rescence studies were carried out to resolve
`this uncertainty.
`
`SCIENCE, VOL. 239
`
`BIOEPIS EX. 1068
`Page 5
`
`