Tailoring in Vitro Selection for a Picomolar Affinity HumanAntibody Directed against Vascular Endothelial Growth FactorReceptor 2 for Enhanced Neutralizing Activity*

Received for publication, July 17, 2003, and in revised form, August 8, 2003Published, JBC Papers in Press, August 12, 2003, DOI 10.1074/jbc.M307742200

Dan Lu‡, Juqun Shen‡, Marie D. Vil‡, Haifan Zhang‡, Xenia Jimenez§, Peter Bohlen¶,Larry Witte§, and Zhenping Zhu‡�

From the Departments of ‡Antibody Technology, §Molecular and Cell Biology, and ¶Research, ImClone SystemsIncorporated, New York, New York 10014

Vascular endothelial growth factor (VEGF) and its re-ceptors have been implicated in promoting solid tumorgrowth and metastasis via stimulating tumor-associatedangiogenesis. We previously identified several fully hu-man neutralizing anti-VEGF receptor 2 (or kinase in-serting domain-containing receptor (KDR)) antibodiesfrom a large antibody phage display library. These an-tibodies bind specifically to KDR, block VEGF/KDR in-teraction, and inhibit VEGF-induced proliferation of hu-man endothelial cells and migration of KDR� leukemiacells. Three of these antibodies, interestingly, share anidentical heavy chain variable (VH) sequence. In this re-port, we constructed a new library comprising the singleVH paired with the variable light chain (VL) repertoireobtained from the original naı̈ve human library. Initial invitro selection revealed that the single VH could pair witha number of different VL while retaining its specificity forKDR. However, a consensus VH/VL pair, clone 1121, wasidentified after three or four rounds of selection by tailor-ing the stringency of the panning conditions. Clone 1121showed a >30-fold higher binding affinity to KDR (Kd, 100pM) because of improvement on both association and dis-sociation constants and blocked VEGF/KDR interactionwith an IC50 of �1 nM, compared with that of 3–4 nM for theparent Fab fragments. Further, clone 1121 was more po-tent in inhibiting VEGF-stimulated KDR phosphorylationin endothelial cells. A binding epitope mapping study onclone 1121 and one of the parent clones, 2C6, demon-strated that both antibodies interacted with the third im-munoglobulin domain within the extracellular region ofKDR. Several peptide phage display libraries were uti-lized to further examine the fine binding specificities ofthe two antibodies. All of the 2C6-binding peptides arecysteine-constrained, whereas clone 1121 binds to bothcysteine-constrained and linear peptides. It is noteworthythat most of the 2C6-binding peptides also cross-reactwith clone 1121, but none of the clone 1121-specific pep-tides binds to 2C6, indicating that clone 1121 retainedpart of the original binding epitope(s) of 2C6 while gain-ing new binding specificity. Taken together, our observa-tion suggests that clone 1121 may have great clinical po-tential in anti-angiogenesis therapy. It furtherunderscores the efforts to identify antibodies of highaffinity for enhanced biological activities.

Vascular endothelial growth factor (VEGF)1 and its recep-tors (VEGFR) have been implicated in promoting solid tumorgrowth and metastasis via stimulating tumor-associated angio-genesis (1–3). VEGF, the primary endothelial-specific mitogen,exerts its effects via two high affinity tyrosine kinase receptor,VEGFR1 (or fms-like tyrosine kinase, Flt-1), and VEGFR2 (orkinase insert domain-containing receptor, KDR) (1–3). Amongthese two receptors, KDR appears to be the major transducer ofVEGF signals in endothelial cells that result in cell prolifera-tion, migration, differentiation, tube formation, increase of vas-cular permeability, and maintenance of vascular integrity (3–5). Inhibition of KDR-mediated signal transduction, therefore,represents an excellent approach for anti-angiogenic interven-tion. In fact, inhibition of angiogenesis and tumor inhibitionhas been achieved by using agents that either interrupt VEGF/KDR interaction and/or block the KDR signal transductionpathway (4–6), including antibodies to VEGF (7–9) or KDR(10–12), anti-VEGF immunotoxins (13), ribozyme (14), solublereceptors (15), and small molecule tyrosine kinase inhibitors(16, 17).

Monoclonal antibodies (mAb), because of their high specific-ity toward a given target, represent a unique class of noveltherapeutics as angiogenesis inhibitors (18, 19). With the re-cent progresses in molecular engineering techniques and theavailability of human antibody transgenic mice and humanantibody phage display libraries, chimeric, humanized, andfully human mAb with desired specificities and properties, e.g.low immunogenecity and high affinity, can be readily produced(20, 21). We previously produced a chimeric and several fullyhuman anti-KDR antibodies and demonstrated that these an-tibodies were capable of blocking KDR/VEGF interaction andinhibiting VEGF-stimulated receptor activation and mitogene-sis of human endothelial cells (10, 22, 23). In addition, theseantibodies inhibited VEGF-induced proliferation of human leu-kemia cells in vitro and prolonged survival of nonobese diabet-ic-severe combined immunodeficient mice inoculated with hu-man leukemia cells (11, 24, 25). We observed that three of theneutralizing human anti-KDR antibodies identified from aphage display library share an identical heavy chain variable(VH) sequence (23). In this report, we constructed a new library

* The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby marked“advertisement” in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

� To whom correspondence should be addressed: Dept. of AntibodyTechnology, ImClone Systems Incorporated, 180 Varick St., New York,NY 10014. Tel.: 646-638-5190; Fax: 212-645-2054; E-mail: Zhenping@imclone.com.

1 The abbreviations used are: VEGF, vascular endothelial growthfactor; VEGFR, vascular endothelial growth factor receptor; AP, alka-line phosphatase; CDR, complementarity determining region; ECD,extracellular domain; HRP, horseradish peroxidase; HUVEC, humanumbilical veil endothelial cells; KDR, kinase insert domain-containingreceptor (or VEGFR2); mAb, monoclonal antibody; PAE, porcine aorticendothelial cells; VH, variable domain of antibody heavy chain; VL,variable domain of antibody light chain; PEG, polyethylene glycol; PBS,phosphate-buffered saline; ELISA, enzyme-linked immunosorbentassay.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 278, No. 44, Issue of October 31, pp. 43496–43507, 2003© 2003 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org43496

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

comprising the single VH paired with the light chain variable(VL) repertoire obtained from the original phage library and,together with a tailored stringent in vitro selection process,identified a consensus VH/VL pair with �30-fold affinity im-provement. The high affinity variant was more potent in block-ing KDR/VEGF interaction and in inhibiting VEGF-stimulatedreceptor activation. Epitope mapping studies using peptidephage display libraries revealed that the high affinity variantgained novel epitope(s) on KDR while retaining part of theoriginal epitope(s) to which the parent antibodies interact.

EXPERIMENTAL PROCEDURES

Cell Lines, Proteins, and Phage Display Libraries—Primary culturedHUVEC were maintained in EBM-2 medium (Clonetics, Walkersville,MD) at 37 °C and 5% CO2 as described previously (10). Porcine aorticendothelial cells transfected with KDR (PAE-KDR) was generated atImClone Systems Incorporated (New York, NY) and maintained inDulbecco’s modified Eagle’s medium containing 10% fetal calf serum.The soluble KDR-alkaline phosphatase fusion protein (KDR-AP), itsextracellular Ig domain (ECD) deletion variant-AP fusions, andVEGF165 proteins were produced as previously described (26). IMC-1C11 and MAB664, two mAb directed against KDR, were generated atImClone Systems Incorporated as previously described (10, 12, 27). Thehuman anti-KDR antibodies, clones 2C6, 1H4, and 2H2, were isolatedfrom a large human Fab phage display library (containing 3.7 � 1010

clones) (28) by several rounds of selection against immobilized recom-binant KDR protein as previously described (23). All of the peptidephage display libraries and the Escherichia coli strain K91 were kindgifts from Dr. G. P. Smith of University of Missouri-Columbia (fordetails see www.biosci.missouri.edu/smithgp/PhageDisplayWebsite/PhageDisplayWebsiteIndex.html).

Generation of the Light Chain-shuffled Phage Display Library—Theoriginal human Fab phage display library (28), from which the anti-KDR antibodies 2C6, 1H4, and 2H2 were identified, was used as thesource of the VL repertoire in the shuffled library. The phagemidpreparation from the original library was first digested with SfiI andNotI followed by electrophoresis on an agarose gel to separate the VHgene fragments from the antibody light chain-containing backbone vec-tor to delete the entire VH repertoire. The gene encoding the VHdomain of clone 2C6/1H4/2H2 was generated by digestion of 2C6-codingphagemid with the same restriction enzymes and purified as describedabove. The 2C6 VH-coding gene was then ligated into the purifiedbackbone vector to create the VL-shuffled Fab repertoire. E. coli TG1cells were transformed with the ligation mixtures via electroporation(Electro Cell Manipulator 600, BTX Electroporation System, San Diego,CA). The transformed TG1 cells were growth with shaking in 2YTmedium at 37 °C for 60 min and plated onto several 150-mm 2YT agarplates containing 100 �g/ml ampicillin and 2% glucose (2YTAG plates).After incubation overnight at 30 °C, all of the colonies grown on theplates were scraped into 5 ml of 2YTAG medium, mixed with 1.2 ml of50% glycerol (final concentration, 10%), aliquoted, and stored at �70 °Cas the library stock.

Selection of the VL-shuffled Library against KDR—The library stock(100 �l) was grown to log phase in 20 ml of 2YTAG medium, rescuedwith M13K07 helper phage, and amplified overnight in 2YTAK medium(2YT containing 100 �g/ml of ampicillin and 50 �g/ml of kanamycin) at30 °C. The phage preparation was precipitated in 4% PEG, 0.5 M NaCl,resuspended in 1 ml of 3% fat-free milk/PBS containing 500 �g/ml of APprotein and incubated at 37 °C for 1 h to block nonspecific binding. Forfirst round selection, KDR-AP-coated Maxisorp Star tubes (Nunc,Rosklide, Denmark) were first blocked with 3% milk/PBS at 37 °C for1 h and then incubated with the phage preparation at room tempera-ture for 1 h. The tubes were washed 15 times with PBST (PBS contain-ing 0.1% Tween 20) followed by 15 washes with PBS. The bound phagewas eluted at room temperature for 10 min with 1 ml of a freshlyprepared solution of 100 mM triethylamine (Sigma). The eluted phagewere incubated with 10 ml of mid-log phase TG1 cells at 37 °C for 30min stationary and 30 min shaking. The infected TG1 cells were pel-leted, plated onto several large 2YTAG plates, and incubated overnightat 30 °C. All of the colonies grown on the plates were scraped into 3–5ml of 2YTA medium, mixed with glycerol (final concentration: 10%),aliquoted, and stored at �70 °C. For subsequent rounds of selection,100 �l of the phage stock was added to 20 ml of 2YTAG medium andgrown to mid-log phase. The culture was rescued with M13K07 helperphage and amplified overnight in 2YTAK medium, and phage-contain-ing supernatant was used directly (without PEG precipitation) for se-

lection followed the procedure described above, with conditions of var-ious stringency as described in Table I.

Phage-based KDR Binding and Blocking ELISA—Individual TG1clones recovered after each round of selection were randomly picked andgrown at 37 °C in 96-well plates and rescued with M13K07 helperphage as described above. The amplified phage preparation was blockedwith 1⁄6 volume of 18% milk/PBS at room temperature for 1 h and addedto Maxi-sorp 96-well microtiter plates (Nunc) coated with KDR-AP orAP (1 �g/ml � 100 �l at 4 °C overnight). After incubation at roomtemperature for 1 h, the plates were washed three times with PBST andincubated with a mouse anti-M13 phage-HRP conjugate (AmershamBiosciences). The plates were washed five times, TMB peroxidase sub-strate (KPL, Gaithersburg, MD) was added, and the absorbance at 450nm was read using a microplate reader (Molecular Device, Sunnyvale,CA).

For KDR/VEGF blocking ELISA, 100 �l of the blocked phage weremixed with 100 ng of KDR-AP and incubated at room temperature for1 h, after which the mixture were transferred to 96-well plates coatedwith VEGF (200 ng/well) at incubated for additional 2 h. The plateswere washed five times with PBST, and the substrate for AP (p-nitro-phenyl phosphate; Sigma) was added, followed by reading the absorb-ance at 405 nm to quantify the bound KDR-AP molecules (10)

DNA Pattern Analysis and Nucleotide Sequencing—The diversity ofthe VL-shuffled library and the anti-KDR Fab clones recovered aftereach round of selection was analyzed by restriction enzyme digestionpatterns (i.e. DNA fingerprints) using a frequent cutting enzyme, BstNI(10, 23). DNA sequences of representative clones from each digestionpattern were determined by dideoxynucleotide sequencing. Classifica-tion and alignments of the VH and the VL genes were performed usingNCBI IgBlast (www.ncbi.nlm.nih.gov/igblast) and V base (www.mrc-cpe.cam.ac.uk).

Expression and Purification of the Soluble Fab Fragments and theFull-length IgG—Phagemids of the individual selected clones were usedto transform a nonsuppressor E. coli host HB2151. Expression of theFab fragments in HB2151 and purification of the soluble Fab proteinsfrom the periplasmic extracts of the E. coli were carried out as previ-ously described (10, 23). The IgG1 forms of clones 2C6, 1H4, and 1121(IMC-2C6, IMC-1H4, and IMC-1121, respectively) were constructedand produced in NS0 cells under serum-free conditions following pro-cedures previously described (11).

Quantitative KDR Binding and Blocking Assay—In the direct bind-ing assay, various amounts of anti-KDR antibodies were added toKDR-coated 96-well Maxi-sorp microtiter plates and incubated at roomtemperature for 1 h, after which the plates were washed three timeswith PBST. The plates were then incubated at room temperature for 1 hwith 100 �l of a rabbit anti-human Fab antibody-HRP conjugate (Jack-son ImmunoResearch Laboratory Inc., West Grove, PA). The plateswere washed and developed, and the absorbance at 450 nm was readfollowing the procedure described above for the phage ELISA. In thecompetitive KDR/VEGF blocking assay, various amounts of anti-KDRantibodies were mixed with a fixed amount of KDR-AP (100 ng) andincubated at room temperature for 1 h. The mixture were then trans-ferred to 96-well microtiter plates precoated with VEGF165 (200 ng/well) and incubated at room temperature for an additional 2 h, afterwhich the plates were washed five times, and the substrate for AP wasadded, followed by reading the absorbance at 405 nm to quantify thebound KDR-AP molecules (10). IC50, i.e. the antibody concentrationrequired for 50% inhibition of KDR binding to VEGF, was thencalculated.

Antibody Affinity Determination by BIAcore Analysis—The bindingkinetics of various antibodies to KDR was measured using a BIAcore3000 biosensor and evaluated using the program BIA Evaluation 2.0(Biacore, Inc., Uppsala, Sweden). The affinity constant, Kd, was calcu-lated from the ratio of dissociation rate (koff)/association rate (kon) (10).

KDR Phosphorylation Assay—The assay was carried out on bothHUVEC and PAE-KDR cells following a protocol previously described(10, 22). Briefly, the serum-starved cells were incubated with variousamounts of antibodies at room temperature for 30 min, followed bystimulation with 20 ng/ml of VEGF165 for additional 15 min. The cellswere lysed, and KDR protein was precipitated from the cell lysates witha polyclonal anti-KDR antibody (ImClone Systems) followed by proteinA-Sepharose beads (Santa Cruz Biotechnology, Santa Cruz, CA). Theproteins were resolved with SDS-PAGE and subjected to immunoblot-ting analysis using an anti-phosphotyrosine antibody-HRP conjugate(Santa Cruz Biotechnology, Santa Cruz, CA). The signals were detectedusing enhanced chemoilluminescence (Amersham Biosciences).

Antibody Epitope Mapping Using KDR Ig Domain Deletion Vari-ants—Full-length KDR (KDR(Ig1–7)) and KDR Ig domain deletion vari-

Antibody Affinity Maturation by Tailoring Selection 43497

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

ants, including KDR(Ig1–3) and KDR (Ig3), the deletion variants thatcontain KDR ECD N-terminal Ig domains 1–3 and 3 only, respectively,were first immobilized onto a 96-well plate (Nunc). The plate was thenincubated with various anti-KDR antibodies at room temperature for1 h, followed by incubation with a mouse anti-human Fc-HRP conjugate(for IMC-1C11, IMC-2C6, and IMC-1121) or a rat anti-mouse Fc-HRPconjugate (for MAB664 only). The plate was washed and developed asdescribed above.

Antibody Epitope Mapping Using Peptide Phage Display Librar-ies—A total of eight randomized pVIII phage-displayed peptide librar-ies were used to dissect out the fine binding specificities of IMC-2C6 andIMC-1121, including a linear 15-mer library and 7 cysteine-constrained

15-mer libraries with the configuration of -XnCX0–6CXn- (n � 3–7). Twoparallel selections were carried out using either the single 15-merlinear peptide library or the mixture of all of the 7 cysteine-constrainedpeptide libraries. Briefly, 96-well ELISA plates were coated with 600 ngof IMC-2C6 or IMC-1121 in 150 �l of PBS overnight and blocked with3% (w/v) skim milk/PBS at 37 °C for 1 h. The peptide-phage librarypreparation (�1–2 � 1011 transforming units in 100 �l of PBS) weremixed with 50 �l of 9% milk/PBS, added into the antibody-coated wells,and incubated at room temperature for 1 h. A nonspecific human IgGwas added in the solution (1 mg/ml) to absorb IgG Fc-binding phages.The plate was washed 15 times with PBS/Tween 20 (0.1%), and thebound phages were eluted by incubation with 150 �l of HCl with bovine

TABLE ISelection and identification of anti-KDR antibodies from the VL-shuffled library under various conditions

FIG. 1. Deduced amino acid sequences of the single VH and the various VL domains of the anti-KDR clones recovered after threeto four rounds of selection under different conditions described in Table I. All of the CDRs are underlined according to the definition ofKabat et al. (29).

Antibody Affinity Maturation by Tailoring Selection43498

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

serum albumin (0.1 M HCl, 1 mg/ml bovine serum albumin, adjusted topH 2.2 with glycine) at room temperature for 10 min, followed byneutralization with 25 �l of 1 M Tris-HCl (pH 9.0). Recovered phageswere then rescued by infecting 75 �l of mid-log phase E. coli K91 cellsat room temperature for 15 min, followed by incubation at 37 °C for 45min. The infected cells were amplified in 4 ml of 2YT/tet (tetracycline at40 �g/ml) medium at 37 °C overnight, followed by PEG precipitationand resuspension in PBS for the subsequent round of selection. For thesecond and third round selections, 300 and 150 ng, respectively, ofantibodies were used to coat the microwells of the 96-well plate.

Identification of Antibody-binding Peptide Phages—ELISA was usedto identify peptide phage clones that bind to IMC-2C6 or IMC-1121after three rounds of selection. Individual phage clones were inoculatedin 200 �l of 2YT/tet/isopropyl-�-D-thiogalactopyranoside (1 mM) in 96-well plates and grown at 37 °C overnight. After centrifugation at 2,000rpm for 10 min, the culture supernatants were mixed with 1⁄3 volume of9% milk/PBS and incubated at room temperature for 60 min. Aliquotsof 50 �l of the mixture were transferred to 96-well plates coated with ananti-M13 phage-antibody (1 �g/ml) and incubated at room temperaturefor 60 min to capture the phage particles on the plate. IMC-2C6 orIMC-1121 antibody were added to the plate at 1 �g/ml, and the amountsof bound antibodies were detected by an anti-human Fc antibody-HRPconjugate as described. A total of two to three dozen positive clones wererandomly picked and subjected to DNA sequencing using the primer5�-CCTGTTGACAATTAATCATCGGCTC-3�.

Dose-dependent Antibody Binding and Blocking Assays—To producephage preparations for quantitative binding and blocking assays, indi-vidual selected clones were inoculated into 5 ml of 2YT/tet and incu-bated at 37 °C with shaking for 5 h. The cell cultures were then trans-ferred into 20 ml of 2YT/tet/isopropyl-�-D-thiogalactopyranoside in250-ml flasks and cultured overnight. Phage were purified twice byPEG precipitation, and the phage pellets were resuspended in PBS atfinal concentration of �1013 phages/ml. The binding assay was carriedout following the procedure described above, starting at 1:10 phagedilution (in 3% milk/PBS) and followed by 2� titration up to 1:1280. Inthe blocking assay, various amounts of phages were mixed with 10 �l ofKDR-AP (2.5 �g/ml) first and then added to 96-well plates coated withIMC-2C6 or IMC-1121. After incubated at room temperature for 1 h, theamounts of bound KDR-AP were detected by the addition of AP sub-strate as previously described (10, 22).

RESULTS

Construction of the VL-shuffled Library—In constructing theVL-shuffled library, the single VH gene segment shared by thethree anti-KDR antibodies, 2C6, 1H4, and 2H2, was subclonedinto the large Fab phage display library replacing the originaldiverse VH repertoire. After electroporating into TG1 cells, alibrary of 1.5 � 108 plaque-forming units was obtained. In thislibrary, the single VH gene is randomly paired with, in theory,1.5 � 108 different VL genes, thus creating the same numbersof independent Fab binding specificities. DNA fingerprintingand nucleotide sequencing of several dozen randomly pickedcolonies revealed no identical patterns and nucleotide se-

quences between different clones and between these clones andthe three parent clones, indicating an excellent diversity of theshuffled library.

Selection of High Affinity Anti-KDR Clones—Two differentselection processes were carried out in parallel using conditionsdescribed in Table I. A total of four rounds of selection wereperformed: the first two in Maxi-sorp star tubes and the lasttwo in 96-well plates. After initial phage binding to immobi-lized KDR, the tubes or wells were washed 15 times with PBST(PBS containing 0.1% Tween 20) followed by another 15 timeswith PBS. Approximately 59% of the phage clones recoveredafter the first round selection were specific KDR binders. In thesubsequent selections, conditions of varying stringency wereapplied. First, to select anti-KDR clones with fast associationrate (or on rate, kon), the number of input phages were reducedin each subsequent selections, and the duration of initial bind-ing processes between the phages and the immobilized antigenwas shortened from 1 h in the first round to 0.5 h in the secondand the third rounds to 0.2 h in the fourth round. Further, aprocess aiming to select clones with improved dissociation rate(or off rate, koff) was also applied; after the initial binding, theKDR-bound phages were subjected to extensive period ofwashes, from 0.5 h in the second round, 3 h in the third roundto 24 h in the fourth round selection, with PBS or variousamounts of KDR (from 100 to 2000 nM) in solution as a com-petitor (see Table I for details).

After each round selection, 96 phage clones were pickedrandomly and assayed for KDR binding activity. The KDR-positive rates in the PBS-washed group were 34.4, 21.1, and81.1% after the second, third, and fourth round selections,respectively, which are comparable with that of 32.2, 22.2, and63.3%, respectively, in the KDR-washed group (Table I). Ten totwenty clones recovered from each round selection were ana-lyzed by DNA sequencing. In the PBS-washed group, 12 of 20clones (60%) sequenced after the third round selection wereclone 2C6, one of the parent clones. A total of seven newsequences were identified, with one sequence, designated asclone 1121, repeated once (Table I). Five different sequenceswere recovered after the fourth round selection; only one of tenclones (10%) sequenced was 2C6. Six of ten sequences (60%)were clone 1121, and one clone (4B4) was also seen after thethird round selection (clone 3B10). In the KDR-washed group,only three different sequences were recovered after the thirdround selection; twelve of twenty clones (60%) were 2C6, andseven of twenty (35%) were clone 1121. All of the clones (10/10)sequenced after the fourth round selection were clone 1121(Table I). In total, ten new anti-KDR VH/VL pairs were iden-

TABLE IIClassification of the VL and the VH genes of the various anti-KDR antibodies

V-gene Human V-gene familyaNo. of mutations from germlinea, b

Closest human-germline genea

Nucleotides Amino acids

Clone 2C6 V�III 13 8 Vg/38KClone 2H2 V�I 6 4 DPK6/Vb�/Vb�Clone 1H4 V�II 29 18 DPL10/V1-7Clone 1121 V�I 26 14 DPK5/Vb/Vb�Clone 3A1 V�I 21 14 DPL8/V1-13Clone 3B5 V�I 20 11 DPL8/V1-13Clone 3B10 V�II 44 22 DPL11/V1-4Clone 3D2 V�I 14 10 DPK5/Vb/Vb�Clone 3D5 V�I 26 13 DPL8/V1-13Clone 3F2 V�I 21 12 DPL8/V1-13Clone 3F7 V�I 32 16 DPL8/V1-13Clone 4A3 V�I 24 12 HK137Clone 4C7 V�I 12 9 DPK6/Vb�/Vb�VH VH3 1 1 DP-77

a V-gene families and the closest human germlines were deduced from NCBI IgBlast at www.ncbi.nlm.nih.gov/igblast and V-Base at www.mrc-cpe.cam.ac.uk.

b Germline gene used for comparison is the closest human germline gene to the individual variable domain sequence.

Antibody Affinity Maturation by Tailoring Selection 43499

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

tified from the VL-shuffled library after three or four rounds ofselection (Fig. 1 and Table I).

Sequence alignments using NCBI IgBlast and V Base of thesingle VH domain showed that the gene has the highest homol-ogy with the germline DP77 segment of the human VH3 family,with only one amino acid substitution from the germline se-quence (Table II). Interestingly, alignments of the VL domainsrevealed both V� and V� gene families; two of the originalclones, 2C6 and 2H2, along with four newly identified clonesincluding 1121 belong to V�, whereas the other clones, includ-ing the original clone 1H4 and six new clones, are from the V�family. It is noteworthy that all the six V� genes have the samecanonical structure for L1, L2, and L3 (2-1-1), whereas the sixnewly selected V� genes share identical length of both CDR1and CDR2 (14 and 7 amino acids, respectively) (30, 31). Taken

together, these observations underscore the structure require-ment on VL genes for pairing with the fixed VH segment tomaintain KDR binding specificity.

KDR binding and blocking activities of all ten positive cloneswere examined using phage-based ELISA. Compared with theparent clone 2C6, six new clones demonstrated moderate tostrong enhancement in KDR-binding activity, with clone 1121being the best binder (Fig. 2A). However, only three clones,1121, 3A1, and 4C7, were better blockers to KDR/VEGF inter-action than clone 2C6 (Fig. 2B). Clone 1121 was thereforeselected as the leading candidate for further studies.

KDR Binding and KDR/VEGF Blocking by the Anti-KDRAntibodies—Soluble Fab fragments of all the three parent anti-KDR clones, 2C6, 1H4, and 2H2, and the consensus clone 1121were produced and compared quantitatively in their antigen

FIG. 2. Binding to KDR and block-ing KDR/VEGF interaction by vari-ous anti-KDR phage clones recoveredafter three to four rounds of selec-tion. A total of ten different KDR-bindingVH/VL pairs were recovered and assayedby phage-based ELISA. Clone 3B10 isidentical to clone 4B4. A, phage superna-tant obtained from overnight culture in96-well plates were added to microwellscoated with KDR-AP and incubated atroom temperature for 1 h, after which theplates were incubated with a mouse anti-M13 phage antibody-HRP conjugate. Theplates were washed, peroxidase substratewas added, and A450 nm was read. B, inhi-bition of binding of KDR to immobilizedVEGF by the anti-KDR phages. Thephage supernatants were incubated witha fixed amount of KDR-AP at room tem-perature for 1 h, after which the mixtureswere transferred to 96-well plated coatedwith VEGF and incubated for an addi-tional 2 h. The amount of KDR-AP thatbound to the immobilized VEGF wasquantified by incubation of the plateswith AP substrate and reading of A405 nm.

Antibody Affinity Maturation by Tailoring Selection43500

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

binding efficiency and potency in blocking KDR/VEGF interac-tion. Consistent with the phage binding result (Fig. 2A), clone1121 Fab binds much more efficiently to KDR than all threeparent Fab (Fig. 3A). It is also a much stronger blocker toKDR/VEGF interaction (Fig. 3B), with an IC50 value of �1 nM,compared with that of 3–5 nM for 2C6, 1H4, and 2H2 Fab.Three Fab fragments, 2C6, 1H4, and 1121, were converted intothe full-length bivalent IgG1 format. IMC-1121 again is thestrongest KDR-binder and the most potent blocker of KDR/VEGF interaction (Fig. 4). The IC50 values for blocking KDRbinding to VEGF were 0.8, 1.2, and 6.2 nM, for IMC-1121,IMC-2C6, and IMC-1H4, respectively (Fig. 4B).

The binding kinetics of various anti-KDR antibodies weredetermined by surface plasmon resonance on a BIAcore instru-ment (Table III). As seen previously, the three parent Fabfragments bind to immobilized KDR with a similar overallaffinity (Kd) of 3.1–4.2 nM, although each individual Fabshowed different kon and koff rates; clone 2C6 has the fastest kon

and the fastest koff, whereas clone 1H4 possesses the slowestkon and also the slowest koff among the three fragments. Clone1121 Fab possesses an affinity of 100 pM, representing �30-foldimprovement over the parent clones, because of improvement

on both kon and, more remarkably, koff. As a bivalent IgG,IMC-1121 showed an overall affinity of 50 pM, representing a�4-fold improvement over IMC-2C6, and 12-fold improvementover IMC-1H4 (Table III).

Inhibition of VEGF-stimulated KDR Phosphorylation—Thebiological activity of IMC-1121 and IMC-2C6 was evaluated ina VEGF-stimulated KDR phosphorylation assay using bothHUVEC and PAE-KDR cells. VEGF stimulation resulted insignificant KDR phosphorylation in both HUVEC and PAE-KDR cells (Fig. 5). Neither antibody alone had any effects onKDR phosphorylation. In the presence of VEGF, both antibod-ies neutralized KDR phosphorylation in a dose-dependentmanner. Consistent with its potent blocking activity, IMC-1121was a much more stronger (�5–25-fold) inhibitor in both cellassays than IMC-2C6 (Fig. 5).

Binding Epitope Mapping for IMC-2C6 and IMC-1121—Wepreviously showed that all of the neutralizing anti-KDR anti-bodies we produced, via both the hybridoma technique and thephage display libraries, block KDR/VEGF interaction by bind-ing to epitope(s) that are locates within KDR ECD Ig domain1–3 (e.g. IMC-1C11) or domain 3 alone (e.g. MAB664) (26, 27).Both IMC-2C6 and IMC-1121 bind equally well to full-length

FIG. 3. Binding to KDR and block-ing KDR/VEGF interaction by anti-KDR Fab fragments. A, dose-dependentbinding of the anti-KDR Fab to immobi-lized KDR. Various amounts of Fab wereadded to 96-well plates coated with KDR(1.0 �g/ml) and incubated at room tem-perature for 1 h, after which the plateswere incubated with a mouse anti-humanantibody-HRP conjugate. The plates werewashed, peroxidase substrate was added,and A450 nm was read. B, inhibition ofbinding of KDR to immobilized VEGF bythe anti-KDR Fab. Various amounts ofFab were incubated with a fixed amountof KDR-AP in solution at room tempera-ture for 1 h, after which the mixtureswere transferred to 96-well plated coatedwith VEGF and incubated for an addi-tional 2 h. The amount of KDR-AP thatbound to the immobilized VEGF wasquantified by incubation of the plateswith AP substrate and reading of A405 nm.The data points are the means � S.D. oftriplicate determinations and are repre-sentative of at least three separateexperiments.

Antibody Affinity Maturation by Tailoring Selection 43501

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

KDR ECD, KDR(Ig1–3), and KDR(Ig3), thus locating theirbinding epitope(s) within the single Ig domain 3 of KDR ECD(Fig. 6).

BIAcore analysis was employed to examine whether IMC-2C6 and IMC-1121 competes with each other or with otherneutralizing anti-KDR antibodies for binding to KDR ECD. Inthis assay, an antibody was first injected onto a KDR-coatedchip at high concentrations to saturate all of the receptor im-mobilized on the chip, which was followed by injection of a

second antibody. An increase in binding density (as measuredby reference units) to the KDR-coated chip upon the injection ofthe second antibody indicates no competition for binding be-tween the two antibodies, i.e. the two antibodies are most likelybind to different, nonoverlapping epitope(s) on KDR, as seen inthe case between MAB664 and IMC-1C11 (Table IV; also seeRef. 26). Binding of IMC-2C6 to the KDR-coated chip blockedfurther binding by IMC-1121 and vice versa, suggesting thetwo antibodies interact with either the same or overlappingepitope(s) within KDR ECD Ig domain 3. Both IMC-2C6 andIMC-1121 also blocked subsequent binding by IMC-1C11 butnot by MAB664. Because IMC-1C11 requires KDR ECD Igdomain 1–3 for binding, this observation suggests that IMC-1C11 may share an overlapping epitope(s) with both IMC-2C6and IMC-1121 within Ig domain 3. It is noteworthy that theoverall KDR binding reference units increased when IMC-1C11was followed by either IMC-2C6 or IMC-1121, but there was nosuch increase when IMC-2C6 or IMC-1121 was followed byIMC-1C11. This phenomenon is most likely a reflection thatboth IMC-2C6 and IMC-1121 are stronger KDR binder andthat they replace KDR chip-bound IMC-1C11 when injectedover the chip at high concentration.

We then used a panel of peptide phage display libraries to

TABLE IIIBinding kinetics of the anti-KDR antibodies to immobilized receptor

Antibody kon koff Kd

104M

�1 s�1 10�4 s�1 nM

2C6 Fab 17.1 � 5.7a 5.5 � 0.76 3.6 � 1.71H4 Fab 5.6 � 0.59 1.5 � 0.22 3.1 � 0.862H2 Fab 12.4 � 2.9 4.9 � 0.18 4.2 � 0.621121 Fab 29.6 � 7.3 0.30 � 0.06 0.10 � 0.02

IMC-2C6 IgG 21.2 � 8.1 0.43 � 0.03 0.20 � 0.01IMC-1H4 IgG 10.0 � 1.78 0.61 � 0.26 0.60 � 0.14IMC-1121 IgG 47.9 � 2.4 0.25 � 004 0.05 � 0.01

a All of the numbers are determined by BIAcore analysis and repre-sent the mean � S.E. from at least three separate determinations.

FIG. 4. Binding to KDR and block-ing KDR/VEGF interaction by full-length anti-KDR antibodies (IgG1).See the legend of Fig. 3 for details. Thedata points are the means � S.D. of trip-licate determinations and are represen-tative of at least three separateexperiments.

Antibody Affinity Maturation by Tailoring Selection43502

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

further map out the fine binding specificities of IMC-2C6 andIMC-1121. After three rounds of selection, �50% of IMC-2C6-selected phage clones and 81% of IMC-1121-selected phageclones demonstrated positive binding to their respective anti-body. Approximately 65% of IMC-2C6-positive phage clonesalso cross-reacted with IMC-1121, but none of the IMC-1121-positive clones bound to IMC-2C6. A total of 19 IMC-2C6-selected clones and 28 IMC-1121-selected clones were pickedand sequenced; 11 different peptides were identified form theIMC-2C6 clones and 15 from the IMC-1121 clones (Table V). Of

note, all IMC-2C6-selected clones, including the three clones(clones 2D5, 1A3, and 1A5) identified from the randomized15-mer linear peptide library, are cysteine-constrained peptides,whereas IMC-1121 binders include both linear and cysteine-constrained peptides. Eight of the 11 IMC-2C6-selected peptideshave the same XCX5CX motif, with the consensus motif ECP(H/P/S)X2RC(P/Q). IMC-1121-selected peptides were structurallymore diverse; of 15 sequences obtained there were five linearpeptides, four peptides with XCX5CX, three with XCX4CX, twowith XCX9CX, and one with XCX3CX motif (Table V).

FIG. 5. Epitope mapping for variousanti-KDR antibodies. The anti-KDRantibodies were added to the microwellsof a 96-well plate coated with KDR(Ig1–7)(full-length KDR ECD), KDR(Ig1–3) (var-iant that contains only the first three N-terminal Ig domains of KDR ECD), orKDR(Ig1) (variant that contains only thefirst N-terminal Ig domain of KDR ECD)and incubated at room temperature for1 h. The plate was washed and then incu-bated with a mouse anti-human Fc anti-body-HRP conjugate for additional 1 h,after which the plate was developed asdescribed above. Antibody binding to thetwo Ig deletion variants are shown as rel-ative (as percentages) to their binding tothe full-length KDR ECD.

FIG. 6. Inhibition of VEGF-stimu-lated KDR phosphorylation in bothHUVEC and PAE-KDR cells by full-length anti-KDR antibodies. Serum-starved cells were incubated with variousamounts of antibodies at room tempera-ture for 30 min, followed by stimulationwith 20 ng/ml of VEGF165 for additional15 min. The cells were lysed, and KDRprotein was precipitated from the cell ly-sates with a polyclonal anti-KDR antibodyfollowed by protein A-Sepharose beads.The proteins were resolved with SDS-PAGE and subjected to immunoblottinganalysis using an anti-phosphotyrosineantibody. The signals were detected usingenhanced chemoilluminescence.

TABLE IVEpitope mapping for the various anti-KDR antibodies using BIA core analysis

First antibody injectionSecond antibody injection

IMC-1C11 MAB664 IMC-2C6 IMC-1121

IMC-1C11 112/115 115/334 117/190 120/234MAB 664 214/348 189/218 218/372 227/414IMC-2C6 174/163 160/337 167/160 187/200IMC-1211 210/203 195/382 217/213 214/217

The numbers represent the number of reference units bound to the KDR-coated chip upon the injection of the first antibody followed by thenumber of reference units bound to the chip after the injection of the second antibody and are the representatives of three separate experiments.

Antibody Affinity Maturation by Tailoring Selection 43503

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

Based on their frequency of occurrence and binding effi-ciency, four peptide phage clones were selected for each anti-body for further analysis. The phage clones were amplified andassayed quantitatively for their efficiency in binding to bothIMC-2C6 and IMC-1121 and their ability in blocking the anti-bodies from binding to their cognate antigen, KDR. As shown inFig. 7 (top panels), all four IMC-1121-selected phage clonesbound specifically to IMC-1121 without any cross-reactivity toIMC-2C6. On the other hand, all the four IMC-2C6-selectedphage clones reacted with both IMC-2C6 and IMC-1121. West-ern blot analysis confirmed the same reactivity pattern of thesephage clones (not shown). In another assay, the peptide phageswere used to compete with soluble KDR for binding to immo-bilized antibodies. Consistent with the antibody binding re-sults, all eight peptide phages were able to compete with KDRfor binding to IMC-1121, but only IMC-2C6-selected phagesblocked KDR/IMC-2C6 interaction (Fig. 7, bottom panels). Acontrol cysteine-constrained phage clone, 9C10 (YCDPRCM-RESKYVIS), did not bind to IMC-2C6 and IMC-1121, nor did itblock the antibodies from binding to KDR.

DISCUSSION

We previously identified four fully human anti-KDR antibod-ies from a large human naı̈ve antibody library and demon-strated that these antibodies were able to block KDR/VEGFinteraction and neutralize VEGF-induced angiogenic activity(23). Three of these antibodies share, interestingly, an identicalVH sequence, suggesting that this single VH domain may pairwith different VL domains while preserving KDR binding/blocking activities. In this report, we successfully improved thebinding affinity of these anti-KDR antibodies by �30-foldthrough VL shuffling in association with a tailored selectionprocess. Several observations are noteworthy in our study.First, by following a tailored stringent selection procedure, wewere able to identify a consensus VL/VH pair, clone 1121, thatshowed a picomolar binding affinity because of significant im-provement on both the association (kon) and dissociation (koff)rates. Second, by shuffling against a large VL repertoire, weidentified ten different VL sequences (in addition to the origi-nal three VL sequences) that could pair with the single VH

TABLE VDeduced amino acid sequences of anti-KDR antibody-selected peptides

a Clone 1A11 is identical to clone 2D5 in (A). ND, not determined.

Antibody Affinity Maturation by Tailoring Selection43504

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

domain while preserving KDR binding specificity. Consistentwith our previous study, these different VL/VH pairs bind KDRand block KDR/VEGF interaction with varying efficiency. It isinteresting to note that both the V� and the V� families areutilized almost equally among the anti-KDR VL/VH pairs, andof the three best KDR/VEGF blocking clones, two (clone 1121and 4C7) belong to V�, and one (clone 3A1) belongs to V�.Third, our epitope mapping studies revealed that althoughclone 1121 competes with 2C6 for binding to certain overlap-ping epitope(s) on KDR ECD Ig domain 3, the new variant alsogained novel binding specificity that is distinct from 2C6.Taken together, these observations suggest that in all of theanti-KDR VL/VH pairs, the single VH domain likely plays adominate role in determining KDR binding specificity, whereasthe VL domains, on the other hand, may contribute to both thebinding strength and the fine epitope specificity via influencingthe overall folding/presentation of the antigen-binding surfaceof the antibodies. It is pertinent to note that VL sharing hasalso been reported in a earlier study among a panel of phage-derived human antibodies directed against several totally un-related antigens (32), again indicating the dominating role ofVH domains in the binding specificities of these antibodies.

A number of approaches have been developed and utilizedsuccessfully to improve antibody binding affinity. These ap-proaches usually start with the construction of mutagenized

antibody libraries and followed by an array of selections toidentify variants with improved affinity. A variety of methods,including randomized mutations (33–35), designed mutationssuch as “hot spot” and “codon-based” mutations (guided bystructural knowledge and/or computer modeling) (36–38) and“CDR walking” (39) have been used to construct the mu-tagenized libraries. These libraries were then displayed on thesurface of filamentous phage (40, 41), bacterial (42–44), yeast(45, 46), or ribosome (47–49) and selected on relevant targetsunder conditions specifically designed for the enrichment ofvariants with desired properties (i.e. high affinity) (for review,see Ref. 50). A common practice in these previous studies isthat most of them were focused on selection of variant(s) withan improved koff rate. Selection for improvement on kon rate, onthe other hand, was largely neglected. Because the overallbinding affinity (Kd) of an antibody equals the ratio of koff overkon, we reasoned that it should be of significant benefit if wecould develop a strategy that would enable us to simulta-neously select variants with improvement on not only koff butalso kon.

The three original neutralizing anti-KDR Fab, clones 2C6,2H2, and 1H4, possess similar overall binding affinity (Kd);however, their individual binding kinetics, i.e. kon and koff, aresignificantly different. For example, clone 2C6 Fab has thefastest kon and also the fastest koff rates, whereas clone 1H4

FIG. 7. Quantitative binding to anti-KDR antibodies and blocking KDR/antibody interaction with selected peptide phage clones.Amplified phages were purified by PEG precipitation and resuspended in PBS at final concentration of �1013 phages/ml. In the binding assay,various amounts of phages were added to 96-well plates coated with an anti-M13 phage antibody to capture the peptide phage particles, followedby incubation with IMC-2C6 or IMC-1121 at room temperature for 1 h, after which the plates were incubated with a mouse anti-human Fcantibody-HRP conjugate. The plates were washed, peroxidase substrate was added, and A450 nm was read. In the blocking assay, the phages weremixed with 25 ng of KDR-AP first and then added to 96-well plates coated with IMC-2C6 or IMC-1121. After incubation at room temperature for1 h, the amounts of bound KDR-AP were detected by the addition of AP substrate and reading of A405 nm. Note that all four IMC-2C6-selected phageclones (solid lines) also bind to IMC-1121 and block KDR/IMC-1121 interaction, whereas none of the IMC-1121-selected clones (dotted lines) bindto IMC-2C6 or block KDR/IMC-2C6 interaction. 9C10, the control phage clone, does not bind to IMC-2C6 and IMC-1121 nor block the antibodiesfrom binding to KDR.

Antibody Affinity Maturation by Tailoring Selection 43505

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

Fab has the slowest kon and koff rates (Table III). These obser-vations indicate that VL domains have significant influence onthe binding kinetics of these anti-KDR antibodies, which sharean identical VH domain. In this report, we therefore con-structed a large library by shuffling the single VH domainshared by the three anti-KDR antibodies against the entire VLrepertoire from the original naı̈ve human library in an attemptto identify the ideal VH/VL pair(s) with improved binding ki-netics (i.e. faster kon and slower koff). The VL-shuffled librarywas selected on immobilized KDR protein under conditions ofincreasing stringency that were designed for both kon and koff

selection; the on rate selection was achieved by reducing thenumbers of input phages and by shortening the initial bindingperiod, whereas the off rate selection was accomplished byextending the washing process (up to 24 h) in the presence of ahigh concentration of the soluble receptor protein. This combi-national selection resulted in the identification of a consensusVL/VH pair (clone 1121) with a Kd of 100 pM, representing�30-fold improvement over the parent Fab fragments. It ispertinent to note that the enhanced affinity of clone 1121 wasa result of improvement on both kon and koff rates.

It is generally accepted that to achieve effective tumor bind-ing and antitumor activity, the affinity (Kd) of a mAb needs tobe 10�8 M or better (51). Although some investigators presumethat high affinity antibody is desirable (52), there is consider-able debate whether improving mAb affinity beyond 10�9 M

would result in a further increase in its tumor localization andbiological activity (53). Recently, Adams et al. (54) employedseveral anti-HER2 antibodies of varying affinities derived fromthe same parent antibody by side-directed mutations andshowed a positive correlation between the amount of antibodylocalized in tumor and antibody affinity up to 10�9 M. Furtherincrease in affinity beyond 10�9 M did not, however, result inhigher tumor localization of the antibody (55). Histologicalstudy on the tumor specimens demonstrated that antibodieswith moderate affinity (between 10�8 and 10�9 M) penetrateddeeper in to tumors and percolated more homogeneously intotumor mass than antibodies with high affinity (10�10 M),which were trapped around the perivascular tumor cells withmuch less localization in the distant tumor tissues (55). Theimpaired tumor penetrating capacity of antibodies with veryhigh affinity may negatively affect their anti-tumor activitywhen the antibodies are used as targeting devices to delivercytotoxic agents to tumor cells within a large tumor mass (53,55). This argument may not be relevant in the context ofanti-KDR therapy, because the target antigen (i.e. KDR) isexpressed on the surface of vessel-lining endothelial cells,which are in direct contact with the systemically administeredantibody (i.e. no tumor penetration is required for antigenaccess). Antibodies with higher affinity, such as IMC-1121,should be able to bind more tightly to their target for a pro-longed period of time, thus efficiently preventing the growthfactors, such as VEGF, from interacting to their receptors. Inthis setting, high affinity antibodies are therefore likely to bemore effective agents than those with low affinity, hence thepreferable choice for anti-angiogenesis therapy. Here we dem-onstrated that the high affinity antibody, IMC-1121, is indeedmore potent in inhibiting KDR/VEGF interaction and in neu-tralizing VEGF-stimulated KDR activation than its parent an-tibodies. Taken together with our previous observation thatIMC-1121 was more potent than IMC-2C6 in inhibiting KDR-leukemia growth in an animal model (11), these results lendstrong support to IMC-1121 for potential clinical evaluation asan anti-angiogenic/antitumor agent.

REFERENCES

1. Ferrara, N. (1999) J. Mol. Med. 77, 527–5432. Klagsbrum, M., and D’Amore, P. A. (1996) Cytokine Growth Factor Rev. 7,

259–2703. Neufeld, G., Cohen, T., Gengrinovitch, S., and Poltorak, Z. (1999) FASEB J. 13,

9–224. Margolin, K. (2002) Curr. Oncol. Rep 4, 20–285. Zhu, Z., Bohlen, P., and Witte, L. (2002) Curr. Cancer Drug Targets 2, 135–1566. Zogakis, T. G., and Libutti, S. K. (2001) Exp. Opin. Biol. Ther. 1, 253–2757. Kim, K. J., Li, B., Winer, J., Armanini, M., Gillett, N., Phillips, H. S., and

Ferrara, N. (1993) Nature 362, 841–8448. Kanai, T., Konno, H., Tanaka, T., Baba, M., Matsumoto, K., Nakamura, S.,

Yukita, A., Asano, M., Suzuki, H., and Baba, S. (1998) Int. J. Cancer 77,933–936

9. Margolin, K., Gordon, M. S., Holmgren, E., Gaudreault, J., Novotny, W., Fyfe,G., Adelman, D., Stalter, S., and Breed, J. (2001) J. Clin. Oncol. 19,851–856

10. Zhu, Z., Rockwell, P., Lu, D., Kotanides, H., Pytowski, B., Hicklin, D. J.,Bohlen, P., and Witte, L. (1998) Cancer Res. 58, 3209–3214

11. Zhu, Z., Hattori, K., Zhang, H., Jimenez, X., Ludwig, D. L., Dias, S., Kussie, P.,Koo, H., Kim, H. J., Lu, D., Liu, M., Tejada, R., Friedrich, M., Bohlen, P.,Witte, L., and Rafii, S. (2003) Leukemia 17, 604–611

12. Prewett, M., Huber, J., Li, Y., Santiago, A., O’Connor, W., King, K., Overhol-ser, J., Hooper, A., Pytowski, B., Witte, L., Bohlen, P., and Hicklin, D. J.(1999) Cancer Res. 59, 5209–5218

13. Olson, T. A., Mohanraj, D., Roy, S., and Ramakrishnan, S. (1997) Int. J. Cancer73, 865–870

14. Pavco, P. A., Bouhana, K. S., Gallegos, A. M., Agrawal, A., Blanchard, K. S.,Grimm, S. L., Jensen, K. L., Andrews, L. E., Wincott, F. E., Pitot, P. A.,Tressler, R. J., Cushman, C., Reynolds, M. A., and Parry, T. J. (2000) Clin.Cancer Res. 6, 2094–2103

15. Holash, J., Davis, S., Papadopoulos, N., Croll, S. D., Ho, L., Russell, M.,Boland, P., Leidich, R., Hylton, D., Burova, E., Ioffe, E., Huang, T., Rad-ziejewski, C., Bailey, K., Fandl, J. P., Daly, T., Wiegand, S. J., Yancopoulos,G. D., and Rudge, J. S. (2002) Proc. Natl. Acad. Sci. U. S. A. 99,11393–11398

16. Fong, T. A., Shawver, L. K., Sun, L., Tang, C., App, H., Powell, J. T., Kim,Y. H., Schreck, R., Wang, X., Risau, W., Ullrich, A., Hirth, P. K., andMcMahon, G. (1999) Cancer Res. 59, 99–106

17. Wood, J. M., Bold, G., Buchdunger, E., Cozens, R., Ferrari, S., Frei, J., Hof-mann, F., Mestan, J., Mett, H., O’Reilly, T., Persohn, E., Rosel, J., Schnell,C., Stover, D., Theuer, A., Towbin, H., Wenger, F., Woods-Cook, K.; Menrad,A.; Siemeister, G., Schirner, M., Thierauch, K. H., Schneider, M. R., Drevs,J., Martiny-Baron, G., and Totzke, F. (2000) Cancer Res. 60, 2178–2189

18. Glennie, M. J., Johnson, P. W. M. (2000) Immunol. Today 21, 403–41019. Zhu, Z., Hicklin, D. J., Bohlen, B., Waskal, H., and Witte, L. (2001) in Recent

Research Developments in Cancer (Pandalai, S. G., ed) Vol. 3, pp. 369–384,Transworld Research Network, Trivandrum, India

20. Little, M., Kipriyanov, S. M., Gall, F. L., and Moldenhauer, G. (2000) Immunol.Today 21, 364–370

21. Hoogenboom, H. R., and Chames, P. (2000) Immunol. Today 21, 371–37822. Zhu, Z., Lu, D., Kotanides, H., Santiago, A., Jimenez, X., Simcox, T., Hicklin,

D. J., Bohlen, P., and Witte, L. (1999) Cancer Lett. 136, 203–21323. Lu, D., Jimenez, X., Zhang, H., Bohlen, P., Witte, L., and Zhu, Z. (2002) Int. J.

Cancer 97, 393–39924. Dias, S., Hattori, K., Heissig, B., Zhu, Z., Wu, Y., Witte, L., Hicklin, D. J.,

Tateno, M., Bohlen, P., Moore, M. A., and Rafii, S. (2001) Proc. Natl. Acad.Sci. U. S. A. 98, 10857–10862

25. Dias, S., Hattori, K., Zhu, Z., Heissig, B., Choy, M., Lane, W., Wu, Y., Chad-burn, A., Hyjek, E., Gill, M., Hicklin, D. J., Witte, L., Moore, M. A., andRafii, S. (2000) J. Clin. Invest. 106, 511–521

26. Lu, D., Kussie, P., Pytowski, B., Persaud, C., Bohlen, P., Witte, L., and Zhu, Z.(2000) J. Biol. Chem. 275, 14321–14330

27. Witte, L., Hicklin, D. J., Zhu, Z., Pytowski, B., Kotanides, H., Rockwell, and P.,Bohlen, P. (1998) Cancer Metastasis Rev. 17, 155–161

28. De Haard, H. J., va Neer, N., Reurs, A., Hufton, S. E., Roovers, R. C., Hen-derikx, P., de Bruine, A. P., Arends, J.-W., and Hoogenboom, H. R. (1999)J. Biol. Chem. 274, 18218–18230

29. Kabat, E. A., Wu, T. T., Perry, H. M., Gottesman, K. S., and Foeller, C. (1991)Sequences of Proteins of Immunological Interest, 5th Ed., National Insti-tutes of Health, Bethesda, MD

30. Chothia, C., Lesk, A. M., Tramontano, A., Levitt, M., Smith-Gill, S. J., Air, G.,Sheriff, S., Padlan, E. A., Davies, D., Tulip, W. R., et al. (1989) Nature 342,877–883

31. Chothia, C., Lesk, A. M., Gherardi, E., Tomlinson, I. M., Walter, G., Marks,J. D., Llewelyn, M. B., and Winter, G. (1992) J. Mol. Biol. 227, 799–817

32. Merchant, A. M., Zhu, Z., Yuan, J. Q., Goddard, A., Adams, C. W., Presta, L. G.,and Carter, P. (1998) Nat. Biotechnol. 16, 677–681

33. Low, N. M., Holliger, P. H., and Winter. G. (1996) J. Mol. Biol. 260, 359–36834. Schier, R., McCall, A., Adams, G. P., Marshall, K. W., Merritt, H., Yim, M.,

Crawford, R. S., Weiner, L. M., Marks, C., and Marks, J. D. (1996) J. Mol.Biol. 263, 551–567

35. Coia, G., Hudson, P. J., and Irving, R. A. (2001) J. Immunol. Methods 251,187–193

36. Wu, H., Beuerlein, G., Nie, Y., Smith, H., Lee, B. A., Hensler, M., Huse, W. D.,and Watkins, J. D. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 6037–6042

37. Chen, Y., Wiesmann, C., Fuh, G., Li, B., Christinger, H. W., McKay, P., de Vos,A. M., and Lowman, H. B. (1999) J. Mol. Biol. 293, 865–881

38. Chowdhury, P. S., and Pastan, I. (1999) Nat. Biotechnol. 17, 568–57239. Yang, W. P., Green, K., Pinz-Sweeney, S., Briones, A. T., Burton, D. R., and

Barbas, C. F., III (1995) J. Mol. Biol. 254, 392–40340. Hawkins, R. E., Russell, S. J., and Winter, G. (1992) J. Mol. Biol. 226, 889–89641. Riechmann, L., and Weill, M. (1993) Biochemistry, 32, 8848–8855

Antibody Affinity Maturation by Tailoring Selection43506

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

42. Daugherty, P. S., Chen, G., Olsen, M. J., Iverson, B. L., and Georgiou, G. (1998)Protein Eng. 11, 825–832

43. Daugherty, P. S., Olsen, M. J., Iverson, B. L., and Georgiou, G. (1999) ProteinEng. 12, 613–621

44. Chen, G., Hayhurst, A., Thomas, J. G., Harvey, B. R., Iverson, B. L., andGeorgiou, G. (2001) Nat. Biotechnol. 19, 537–542

45. van Antwerp, J. J., and Wittrup, K. D. (2000) Biotechnol. Prog. 16, 31–3746. Feldhaus, M. J., Siegel, R. W., Opresko, L. K., Coleman, J. R., Feldhaus, J. M.,

Yeung, Y. A., Cochran, J. R., Heinzelman, P., Colby, D., Swers, J., Graff, C.,Wiley, H. S., and Wittrup, K. D. (2003) Nat. Biotechnol. 21, 163–170

47. Hanes, J., Schaffitzel, C., Knappik, A., and Pluckthun, A. (2000) Nat. Biotech-nol. 18, 1287–1292

48. Jermutus, L., Honegger, A., Schwesinger, F., Hanes, J., and Pluckthun, A.(2001) Proc. Natl. Acad. Sci. U. S. A. 98, 75–80

49. Irving, R. A., Coia, G., Roberts, A., Nuttall, S. D., and Hudson, P. J. (2001)J. Immunol. Methods 248, 31–45

50. Amstutz, P., Forrer, P., Zahnd, C., and Pluckthun, A. (2001) Curr. Opin.Biotechnol. 12, 400–405

51. Cannon, J. B., and Hui, H. W. (1990) Targeted Diagn. Ther. 3, 121–13952. Schlom, J., Eggensperger, D., Colcher, D., Molinolo, A., Houchens, D., Miller,

L. S., Hinkle, G., and Siler, K. (1992) Cancer Res. 52, 1067–107253. Juweid, M., Neumann, R., Paik, C., Perez-Bacete, M. J., Sato, J., van Osdol,

W., and Weinstein, J. N. (1992) Cancer Res. 52, 5144–515354. Adams, G. P., Schier, R., Marshall, K., Wolf, E. J., McCall, A. M., Marks, J. D.,

and Weiner, L. M. (1998) Cancer Res. 58, 485–49055. Adams, G. P., Schier, R., McCall, A. M., Simmons, H. H., Horak, E. M.,

Alpaugh, R. K., Marks, J. D., and Weiner, L. M. (2001) Cancer Res. 61,4750–4755

Antibody Affinity Maturation by Tailoring Selection 43507

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from

Witte and Zhenping ZhuDan Lu, Juqun Shen, Marie D. Vil, Haifan Zhang, Xenia Jimenez, Peter Bohlen, Larry

Activityagainst Vascular Endothelial Growth Factor Receptor 2 for Enhanced Neutralizing

Selection for a Picomolar Affinity Human Antibody Directedin VitroTailoring

doi: 10.1074/jbc.M307742200 originally published online August 12, 20032003, 278:43496-43507.J. Biol. Chem. 

  10.1074/jbc.M307742200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/278/44/43496.full.html#ref-list-1

This article cites 54 references, 16 of which can be accessed free at

by guest on Novem

ber 28, 2020http://w

ww

.jbc.org/D

ownloaded from