a human fab-based immunoconjugate speciﬁc for the lmp1 … · therapeutic discovery a human...

Therapeutic Discovery

A Human Fab-Based Immunoconjugate Specific for theLMP1 Extracellular Domain Inhibits NasopharyngealCarcinoma Growth In Vitro and In Vivo

Renjie Chen1, Dawei Zhang1, Yuan Mao2, Jin Zhu1,3, Hao Ming4, Juan Wen1, Jun Ma5, Qing Cao1,Hong Lin1, Qi Tang1, Jie Liang1, and Zhenqing Feng1

AbstractNasopharyngeal carcinoma (NPC) is a major cause of cancer-related death in Southeast Asia and China.

Metastasis and relapse are the primary cause ofmorbidity andmortality inNPC. Recent evidence suggests that

the Epstein–Barr virus latent membrane protein 1 (LMP1) is exclusively expressed in most NPC and is a

potential target for biotherapy. In this study, we successfully prepared a novel human antibody Fab

(HLEAFab) against LMP1 extracellular domain, which was subsequently conjugated with mitomycin C

(MMC), thus forming an immunoconjugate (HLEAFab-MMC). The effects ofHLEAFab-MMConproliferation

and apoptosis inNPC cell linesHNE2/LMP1 and the inhibition rate of growth ofNPC xenografts in nudemice

were examined. The inhibition rate of HNE2/LMP1 cell proliferation was the highest for HLEAFab-MMC

(76%) compared with MMC (31%) and HLEAFab (22%) at a concentration of 200 nmol/L and showed dose-

dependent fashion. The apoptosis rate of HNE2/LMP1 cell lines was 13.88% inHLEAFab-MMC group, 3.04%

in MMC group, 2.78% in HLEAFab group, and 2.10% in negative control group at the same concentration,

respectively. In vivo, the inhibition rate of growth of NPC xenografts in nude mice was 55.1% in HLEAFab-

MMC group, 26.5% in MMC group, and 5.64% in HLEAFab group. In summary, our findings show that

HLEAFab-MMC is a unique immunoconjugate with the potential as a novel therapeutic agent in the treatment

of LMP1-expressing NPC. Mol Cancer Ther; 11(3); 594–603. �2011 AACR.

IntroductionEpstein–Barr virus (EBV) is a herpes virus prevalently

associated with a variety of malignant diseases, such asnasopharyngeal carcinoma (NPC), Burkitt’s lymphoma,Hodgkin’s lymphoma, and some gastric carcinomas (1).An association of EBV with NPC has been consistentlyestablished by numerous studies. Various prints of EBVhave been found in the tissues of NPC and antibodiesagainst EBV present in serum of NPC patients. In NPC,

most tumor cells are latently infected and express 3 latentmembrane proteins (LMP): LMP1, LMP2A, and LMP2B,as well as EBV nuclear antigen 1 (EBNA1). Among these,LMP1 and EBNA1 have been detected inNPC biopsies byWestern blotting analysis. LMP1 is considered to be themajor EBV oncoprotein (2). Therefore, both EBNA1 andLMP1 are important tumor-specific markers for NPC.Furthermore, LMP1-positive NPCs are reported to bemore progressive than LMP1-negative NPCs and showan increasing tendency of lymph node metastasis (3).Structurally, LMP1 is an integral membrane protein con-sisting of a short cytoplasmic N-terminus of 20 aminoacids, a transmembrane domain with 6 membrane-span-ning segments that anchor LMP1 in a patchy distributionalong the plasma membrane, and a long cytoplasmicC-terminus of 200 amino acids (4–6).

In this study, we chose the extracellular domains of theoncogenic LMP1 as a target of therapy.We had screened afully humannaive Fabphage library after panning againstextramembrane domains (EMD) polypeptide, which con-sists of 3 EMDs proximal 16 amino acid, by panning onliving cells and immobilized LMP1 (7), followed by char-acterizing its binding specificity and affinity to nativeLMP1 as well as the NPC cell line overexpressing LMP1.Using this strategy, we showed that this human anti-LMP1 EMDs antibody Fab fragment (HLEAFab) can bindto the extracellular domains of LMP1.We then conjugated

Authors' Affiliations: 1Nanjing Medical University, The Key Laboratory ofCancer Biomarkers, Prevention & Treatment Cancer Center, The KeyLaboratory of Antibody Technique of Ministry of Health and The SecondAffiliated Hospital, Department of Otolaryngology Head and Neck Surgery,Nanjing; 2Department of Otolaryngology Head and Neck Surgery, JiangsuProvincial Official Hospital; 3Huadong Medical Institute of Biotechniques,Nanjing; 4Department of Otolaryngology Head and Neck Surgery, XuzhouCenter Hospital, Xuzhou; and 5Department of Otolaryngology Head andNeck Surgery, Yiji Shang Hospital of Wannan Medical College, Anhui,China

Note: R. Chen and D. Zhang contributed equally to this work.

Corresponding Author: Zhenqing Feng, Nanjing Medical University, TheKey Laboratory of Cancer Biomarkers, Prevention & Treatment CancerCenter, TheKeyLaboratory ofAntibodyTechniqueofMinistry ofHealth andThe Second Affiliated Hospital, Department of Otolaryngology Head andNeck Surgery, 140HanzhongRoad, Nanjing 210029. China. Phone: 86-25-868-63100; Fax: 86-25-866-63193; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-11-0725

�2011 American Association for Cancer Research.

MolecularCancer

Therapeutics

Mol Cancer Ther; 11(3) March 2012594

on June 30, 2021. © 2012 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst December 14, 2011; DOI: 10.1158/1535-7163.MCT-11-0725

http://mct.aacrjournals.org/

this HLEAFab with mitomycin C (MMC), a chemothera-peutic drug, to generate a potential immunoconjugateagent. This study is the first to clone, express, and func-tionally validate the HLEAFab targeting the LMP1. Theimmunoconjugate, which consists of a Fab antibody frag-ment derived from a fully human Fab phage library andMMC, killed LMP1-positive NPC cell lines in vitro andsuppressed NPC growth in a nude murine transplanta-tion model.

Materials and MethodsPhage library, helper phage, and bacterial strainsA human naive Fab phage library was constructed as

previously described (8). Before the first-round panning,the library was titrated and 2 � 1013 phage clones werecollected for panning. The VCSM13 helper phage and theE. coli strain XL1-Blue and another E. coli strain, Top 10 F’,for expressing the Fab, were saved and provided by TheKey Laboratory of Cancer Biomarkers, Prevention &Treatment Cancer Center and The Key Laboratory ofAntibody Technique of Ministry of Health, NanjingMed-icalUniversity,Nanjing,China. Both strainswere tested topreclude any wild-type phage contamination.

Cell lines and purified EMDs of LMP1Two cell lines were used for biopanning and Fab char-

acterization as well as for in vitro bioassays: HNE2 is aEBV-LMP1–negative human NPC cell line, HNE2/LMP1is a cell line constantly expressing LMP1 after the intro-duction of full-length LMP1 cDNA into HNE2 cells, pur-chased from Xiangya Central Experiment Laboratory,Hunan, China and were authenticated in this laboratory.The initial characterization of HNE2 and HNE2/LMP1cells has previously been described (9). On the basis ofPCR and Western blotting analyses for LMP1, HNE2/LMP1 cells are LMP1 positive, whereas HNE2 cells areEBV DNA negative. Both cell lines were cultured in aRPMI-1640 medium (GIBCO; Invitrogen) with 10% FBS,antibiotics. All of the cell lines are routinely screened forMycoplasma species (GenProbe Detection Kit; Fisher). Allcells weremaintained at 37�C in a humidified atmosphereof 95% air and 5% CO2. For the experiments, HNE2 andHNE2/LMP1 were used between passages 3 to 7. TheEMDs of LMP1 and GST fusion protein, which wasprepared by our research team previously (7), were affin-ity purified from Glutathione Sepharose 4B (GE productcode:27-4574-01). The eluted LMP1 was verified byWest-ern blotting and quantified by BCA Protein Assay (Bio-Rad).

Biopanning methodsTo isolate phage that specifically bind to LMP1 on the

cell surface, subtractive panning was carried out on bothHNE2 cell line and HNE2/LMP1 cell line, as well as onimmobilized LMP1 extracellular domains. Seven roundsof panning were carried out. The first, second, third, andsixth rounds of panning were done on the cells, and one

million cells were used in each round. The cells werestarved in serum-freemedium for 12 hours, washed twiceusing pH 7.4 PBS, detached from the flask with celldissociation buffer (catalog no. 13151-014; Invitrogen) at37�C for 10 minutes, and washed twice in PBS. LMP1-negative HNE2 cells were resuspended and mixed with1013 phagemids in 1 mL of 1% bovine serum albumin(BSA)-PBS, shaken at 37�C for 30minutes, and centrifugedat 5,000 � g for 3 minutes. The supernatant was collectedandmixedwithHNE2/LMP1 and then incubated at 37�Cfor 1 hour. The cells were washed with RPMI-1640 andtreated with 3 mL Trypsin-EDTA (Gibco, 25300.054) at37�C for 5 minutes. The phages were used to infect E. coli.XL1-Blue for amplification and next-round panning. Toincrease binding specificity, as well as affinity to LMP1,the fourth, fifth, and seventh rounds of panning werecarried out on immobilized LMP1, and the amount ofcoated LMP1 was gradient decreased in each of the 3rounds (400 ng for fourth round, 200 ng for fifth round,and 100 ng for seventh round).

ELISA screening of LMP1–binding positive phageclones

Single phage clones from the E. coli XL1-Blue infectedby the seventh round of eluted phage were picked up andgrown in 1 mL super broth (SB) medium containing 50mg/mL carbenicillin and 1% glucose, with shaking at37�C until the exponential phase. VCSM13 helper phage(1 � 109) was added to each vial. The culture was shakenovernight with 70 mg/mL kanamycin. Fifty microliters ofsupernatant from each vial was added to each well of 96-well EIA plates coated with 100 ng GST-LMP1 fusionprotein that had been preblocked with 5% milk blockingbuffer (5% milk, 0.5% Tween-PBS). Purified GST proteinwas used as negative control. After incubation at roomtemperature for 1 hour, the plate was washed using 0.5%Tween-PBS; 50 mL of 1:4,000 diluted horseradish peroxi-dase (HRP)-conjugated anti-M13 antibody (Amershamcode number: 27-942-01) in blocking buffer was added toeach well and incubated for another hour and then theplatewaswashedagain. Fiftymicroliters ofHRPsubstratesolution (Prod# 34021; Pierce) were added and developed30minutes before stopping by addition of 1mol/LH2SO4,and the absorbance value at 450 nm was read by Multis-kan Spectrum Microplate (Themo Instruments Inc.). AllELISA assays were repeated at least 3 times.

Expression and purification of a soluble HLEAFabfragment

The recombinant HLEAFab was expressed in the E. coliTop 10 F’ strain. Briefly, the overnight culture of a singleclone was reinoculated at 1:100 in SB medium with 50mg/mL carbenicillin until the OD600 reached 1.0, inducedby 1 mmol/L IPTG in the presence of 4% sucrose at 25�Cand harvested 24 hours later. Western blotting was doneon both bacteria lysate and sonicated supernatant todetermine whether the HLEAFab was expressed afterIPTG induction, with a purified human Fab as positive

Fab-Based Immunoconjugate Specific for LMP1

www.aacrjournals.org Mol Cancer Ther; 11(3) March 2012 595




control. The soluble HLEAFab was purified from theperiplasm of the bacteria using Protein L (catalog no.L00239; GenScript Corporation) affinity purification.

Cell ELISANPC cell line (HNE2 and HNE2/LMP1) were cultured

in complete medium in 96-well flat bottom plates (Corn-ing) to form a subconfluent monolayer and further incu-bated overnight in RPMI-1640 medium containing 10%FBS. The cells were washed 3 times for 5 minutes eachwith PBS, pH 7.4 and cells were fixed by treatment with4% paraformaldehyde for 10 minutes at 4�C. The endog-enous peroxidase activity was blocked with 3% H2O2 indistilled water for 7 minutes at room temperature fol-lowed by 3 � 5 minutes washes in PBS, pH 7.4. Nonspe-cific binding was blocked with 3% BSA for 1 hour at roomtemperature. Cells were washed once in PBS and incu-bated with purified HLEAFab diluted in 1% BSA for 1hour at room temperature. The unrelated antibody Fabfragmentwas used as the negative control antibody. After3 � 10 minutes washes in PBS, HRP-conjugated anti-human IgG(Fab)’ (catalog no. A0293; Sigma) was addedfor 1 hour at room temperature. The cells were washedagain in PBS, following by TMB substrate solution, andincubated for 20 minutes at room temperature. The colorreaction was stopped by adding 1 mol/L H2SO4 and theintensity was read at 450 nm in a Multiskan SpectrumMicroplate (Thermo Labsystems; refs. 10–12).

Fluorescence-activated cell sorting analysisHNE2 and HNE2/LMP1 cells were blocked using 1%

BSA-PBS at 4�C for 30 minutes, then incubated with100 mg/mL HLEAFab at 4�C for 45 minutes, and stained

using 1:50 diluted fluorescein isothiocyanate (FITC)-labeled human anti-Fab IgG (catalog no. F5512; Sigma)at 4�C for 20 minutes. The unrelated antibody Fab frag-ment was used as the negative control antibody. Fluores-cence intensity was analyzed with CellQuest software(Becton Dickinson Bioscience). The cells incubated onlywith secondary antibody were analyzed as controls.

Fluorescence stainingHNE2/LMP1 and HNE2 cells were grown in 6-well

strips of cell culture plates for staining.At 80%confluence,cells were fixed using equal volumes of acetone andmethanol, blocked by 5% milk PBS at 37�C for 1 hour,and coincubated with 50 mg/mL HLEAFab at 37�C for 1hour before 1:50 FITC-labeled human anti-Fab IgG (cat-alog no. F5512; Sigma) in the dark. As mentioned above,the fluorescence staining experimentwas done using cellsfrom different passage numbers and repeated at least 3times.

Conjugation of MMC to HLEAFabThe principle used to conjugate smallmolecule antican-

cer drug MMC (Zhejiang Haizheng Pharmaceutical Co.,Ltd.) to HLEAFab involved the activation of the aminogroup with N-succinimidyl 3-(2-pyridyldithio) propio-nate (SPDP; Pierce), followed by disulfide exchange withiminothiolane-modified HLEAFab (Fig. 1; ref. 13). Foractivation, MMC was reacted with SPDP at 4�C for 10hours and stored in small aliquots at �80�C until furtheruse. HLEAFab in PBS was reduced by mixing with2-iminothiolane for 1 hour under nitrogen. The reducedantibody was purified by size-exclusion chromatographyusing Desalting Column (Pierce) preequilibrated with

O

O

O O

O

NN

NN

O

O

O

O

N OH

O SS

S

S

S N–

N–

S

S NH

NH

NH

NH

HS

NH

OC

OC OC

N

O

N

MMCSPDP

A B

CNHS

MMC-SPDP activation HLEAFab-MMC immunoconjugate

HLEAFab reduction

HLEAFab

HLEAFab

2-Iminothiolane

HLEAFabH2N

H2N

H2N

OCH3

OCH3 OCH3

NH

CH2OCONH2

CH2OCONH2 CH2OCONH2

H3C

H3C

H2N

H3C

Figure 1. Conjugation of anti-LMP1 antibody Fab andMMC.MMCwas first conjugatedwith the SPDP linker (A), and Fabwas reducedwith 2-iminothiolane (B).The final immunoconjugate construct is shown in C.

Chen et al.

Mol Cancer Ther; 11(3) March 2012 Molecular Cancer Therapeutics596




PBS. The conjugation of activated MMC with thiolatedHLEAFab (MW:� 50 kDa)was done at room temperaturefor 1 hour under nitrogen using a 1:1 molar ratio. Excessfree MMC–SPDP were removed using Desalting Column(Pierce) preequilibrated with PBS. HLEAFab-MMC pro-ducts were filter sterilized using 0.2-mmSuporMembrane(Pall). The dosage of the HLEAFab-MMC conjugate forcytotoxicity and in vivo experiments was calculatedaccording to the total amount of MMC added. The con-jugations of MMC with the unrelated Fab fragment weredone the same way as described above.

MTT assay for antiproliferation of HNE2/LMP1 cellsby HLEAFab-MMCMTT assay (14) was done on HNE2/LMP1 and HNE2

cells on a 96-well plate. A total of 5 � 103 cells weredispensed into each well and cultured for 12 hours.Following removal of cultured media, fresh mediumcontaining MMC, naked antibody, or immunoconjugates(25–200 nmol/L)was added to eachwell for an additional48-hour incubation. Subsequently, the culture mediumwas removed, and 100 mL of MTT (1 mg/mL in RPMI-1640) was added to each well. After 5 hours of incubationwith MTT at 37�C in 5% CO2 incubator, the supernatantwas removed, and 150 mL of dimethylsulfoxide wasadded to each well followed by shaking at 150 rpm for5 minutes. Absorbance at 490 nm was determined spec-trophotometrically. The cells without any antibody treat-ment served as the negative control. MTT assay wasrepeated at least 3 times.

Induction of apoptosis of HNE2/LMP1 cells byHLEAFab-MMCTo investigate the apoptotic effects of HLEAFab-MMC

on NPC cancer cells, Apo-Direct Kit (BD Pharmingen556381) was used for detection and quantitation of apo-ptosis at a single cell level. Briefly, HNE2/LMP1 andHNE2 cells were cultured in RPMI-1640medium at 37�Cin a 5% CO2 incubator for 12 hours until all cells wereattached to microwells. HLEAFab-MMC (200 nmol/L)was added to the attached cells for coincubation of 48hours. As the negative control, MMC, HLEAFab, theunrelated Fab, the immunoconjugate of MMC with theunrelated Fab, of the same concentration or PBS wasused for the same incubation period. At the end ofincubation, the attached cellswere removed fromculturewells by Cell Dissociation Buffer (Invitrogen). Apoptosisof treated HNE2/LMP1 cells was determined quantita-tively by terminal deoxynucleotidyl transferase–medi-ated dUTP nick end labeling (TUNEL) assay, with theinstructions provided by Apo-Direct Kit (BD pharmin-gen 556381). Morphologic changes of HNE2 and HNE2/LMP1 cells were observed after treatment with HLEA-Fab-MMC under fluorescence microscope by livingcells\apoptotic cells\necrotic cells Identification Kit(Nanjing KeyGen Biotechnology Development Co. Ltd).As mentioned above, the TUNEL assay was repeated atleast 3 times.

Treatment of HNE2/LMP1 xenografts withHLEAFab-MMC

Female 4-week-old BALB/c nude mice with a bodyweight of approximately 20 g were purchased fromShanghai SLAC Laboratory Animal Co. Ltd. and keptunder specific pathogen-free conditions. For HNE2/LMP1 xenograft model, mice injected with 2 � 106/mLHNE2/LMP1 cells subcutaneously were killed aftertumor diameter of each of the 4 nude mice was about 1cm; the rapidly growing tumors were excised, cut asep-tically into 2 mm3 pieces, and inoculated subcutaneouslyinto the right flank of experimental mice. After inocula-tion, tumor sizes were measured every 5 days with Ver-nier calipers, volumes were calculated by the formula:(smaller diameter)2 � larger diameter � 0.5, and tumorswere allowed to grow for 14 days before treatment (15). Totest HLEAFab-MMC on tumor growth inhibition, themice with HNE2/LMP1 xenograft were given HLEA-Fab-MMC location injected (16, 17) at dose levels rangingfrom HLEAFab (8 mg/kg, 1.6 � 10�5 mol/kg) in HLEA-Fabgroup,HLEAFab-MMC(8mg/kg, 1.6� 10�5mol/kg)inHLEAFab-MMCgroup,MMC (0.053mg/kg, 1.6� 10�5mol/kg) in MMC group on days 1, 4, and 7 includingthe unrelated Fab, the immunoconjugate ofMMCwith theunrelated Fab, and normal saline for negative control. Themice were followed for observation of xenograft bodyweight changes. At the end of the treatment period, themice were killed, and the tumors were excised to deter-mine tumor weight.

ResultsSelection of specific LMP1-binding phage andexpression and purification of soluble HLEAFab

After 7 rounds of panning, 45 single phage clones wererandomly picked up and amplified to test for specificbinding toLMP1EMDsbyphageELISA.A ratio of sampleOD450 versus the blank of greater than 2.5 was set as thestandard for selecting positive clones. As shown in Fig.2A, no. 36 clone with strongest binding to LMP1 EMDswere analyzed by DNA sequencing andwere all identical(Table 1). Soluble expression of HLEAFab was inducedovernight at low temperature (25�C). The expressedHLEAFab was found mainly in the periplasmic space ofE. coli (Fig. 2B). SDS-PAGE and Coomassie blue stainingshowed equal expression of heavy and light chains. Thepurity was above 95% after Protein L affinity purification(Fig. 2C).

The HLEAFab reacts with the extracellular domainsof LMP1 in native conformation

Further analysis of HLEAFab reactivity was carried outby comparing the binding ability of the antibody frag-ments to LMP1-positive (HNE2/LMP1) and LMP1-neg-ative (HNE2) NPC cells using cell surface ELISA. Shownin Fig. 3A, the purified HLEAFab recognized LMP1 pres-ent in HNE2/LMP1 cells. We further evaluated whetherthe HLEAFab binds to extracellular domains of LMP1






using a flow cytometric assay. Both HNE2/LMP1 andHNE2 cells were treatedwith orwithoutHLEAFab beforeaddition of fluorescence-labeled secondary conjugate.

The result in Fig. 3B shows that the population ofHLEAFab-treated HNE2/LMP1 cells was clearly sepa-rated from nontreated cells by fluorescent intensity,whereas no difference was noted between HLEAFab-treated and nontreated HNE2 cells, suggesting that theHLEAFab binds to LMP1-positive cell membranes only.Furthermore, to provide morphologic evidence thatthis HLEAFab binds to the cell membrane, a cell singlestaining experiment was conducted, and again, thisHLEAFab stained only LMP1-positive NPC HNE2/LMP1 cells and not LMP1-negative NPC HNE2 cells(Fig. 3C).

Cytotoxicity and proapoptotic activity ofimmunoconjugates HLEAFab-MMC againstLMP1-positive NPC HNE2/LMP1 cells in vitro

After confirming the binding capability of thisHLEAFab, we conjugated it with MMC to investigate itsantitumor effect in vitro.Mass spectrophotometer analysisshowed that the molar ratio of HLEAFab to MMC wasabout 1:1 (average molecular weight of the conjugateincreased 342 Da in comparison with HLEAFab, and themolecularweight ofMMC is 334.3). Thebinding affinity ofMMC-conjugated HLEAFab to LMP1 did not decreasesignificantly as compared with nonconjugated HLEAFab(data not shown), and the stability of MMC-conjugatedHLEAFab is reasonable, as it was exposed at room tem-perature for days during conjugation reaction. To deter-mine the efficacy of HLEAFab-MMC to specificallyinhibit the growth ofHNE2/LMP1 cells,MTTassayswerecarried out. Viable HNE2/LMP1 or HNE2 cells were

Figure2. A,ELISA result of 45 individual phage clones randomly pickedupfrom the eluted phage pool after the 7th round of biopanning. The last 2columns on the right were negative control phages that served as blanksB, Western blotting characterization of HLEAFab fragment expressedin E. coli. Lane 1 was Top10F' cell lysate control; lane 2 was 25 mL of celllysates after sonication; lane 3 was 10 mL of cell sedimentation aftersonication; lane 4 was 25 mL of culture medium after cells culture. Mouseantihuman Fab HRP conjugate was used at 1:1,000 dilution. C, purifiedHLEAFab fragment was separated on a 10% SDS-PAGE gel andstained with Coomassie blue. The double bands were heavy chain Fd(top) and light chain k (bottom), and light chain k were more expressed.

Table 1. HLEAFab variable region DNA and amino acid sequence

VL (immunoglobulin kappa light chain VLJregion)

VH (immunoglobulin heavy chainvariable region)

DNA sequence GAGCTCCAGATGACCCAGTCTCCATCCTCCCTGTCTGCATCTGTAGGAGACAGAG TCGCCATCACTTGCCGGGCAAGTCA GAGCATTAGCAGCCATTTAAATTGGTATCAGCAGAAACCAG GGAAAGCCCCTAAGCTCCTGATCTATGCTGCATCCAGTTTGCAAAGTGG GGTCCCATCAAGGTTCAGTGGCA GTGGATCTGGGACAGATTTCACTCTCACCATCAGCAGTCTGCAACCTG AAGATTTTGCAACTTACTACTGTCA ACAGAGTTACAGTACCCCGTACACTTTTGGCCAGGGGACCAAGCTGGAGATCAAA

GAGGTGCAGCTGGTAGAGTCTGGTGGAGGCTTGGTCCAGCCTGGGGGGTCCCTGAGACTCTCCTGTGCAGCCTCTGGATTCACCTTCAGTAGCTATTG GATGAGCTGGGTCCGCCAGGCTCCAGGGAAGGGGCTGGAGTGGGTGGCC AACATAAAGCAAGATGGAAGTGAGAA AT ACTATGTGGACTCTGTGAAGGGC CGATTCACCATCTCCAGAGACAACG CCAAGAACTCACTGTATCTGCGAATGAACAGCCTGAGAGCCGAGGACACGG CTG TGTATTACTGTGCGAGAGCCTGGGGGTATAGCAGCTACTACTTTG ACTACTGGGGCCAGGGCACCCTG GTCACCGTCTCCCCT

Amino acid sequence ELQMTQSPSSLSASVGDRVAITCRASQSISSHLNWYQ QKPGKAPKLLIYAASSLQSGVPSRFSGS GSGTDFTLTISSLQP EDFATYYCQQSYSTPYTF GQGTKLEIK

EVQLVESGGGLVQPGGSLRLSCAAS GFTFSSYWMSWVRQAPGKGLEWV ANIKQDGSEKYYVDS VKGRFTISRDNA KNSLYLRMNSLRAEDTAVYYCARAWG YSSYYFDYWGQGTLVTVSP

Chen et al.





treated with different concentrations of HLEAFab-MMC,HLEAFab, MMC, unrelated Fab, or unrelated Fab-MMC.The results showed that HLEAFab-MMCwas effective ininhibiting 76% cell proliferation at 200 nmol/L. However,MMC and HLEAFab yielded only 31% and 22% inhibi-tion, respectively. In fact, the cell inhibition rate decreasedas the concentration of HLEAFab-MMC declined, denot-ing a dose-dependent response. These data indicated thatthe cell cytotoxicity seen with HLEAFab-MMC is medi-ated by LMP1 specifically in a dose-dependent manner(Fig. 4A). MMC, naked antibody, or immunoconjugatesexhibited very modest or no cytotoxicity at the highestdose tested (200 nmol/L) inHNE2 cells (Fig. 4B). By usingTUNEL assay, inductions of apoptosis following treat-ments of cancer cells with HLEAFab-MMC were quanti-tatively determined. As shown in Fig. 4C, significantincreases in apoptosis of treated HNE2/LMP1 cells weredetected upon the treatment with HLEAFab-MMC (200nmol/L), as compared with the negative control andHNE2 cells. The highest apoptosis rate of HNE2/LMP1cell lines was observed in HLEAFab-MMC group(13.88%) compared with the MMC group (3.04%), HLEA-Fab group (2.78%), unrelated Fab (2.34%) or unrelatedFab-MMC (3.2%), or negative control (2.10%) at the sameconcentration. Characteristic morphologic changes of cellapoptosis were observed by fluorescence microscopyafter HLEAFab-MMC acted on NPC HNE2/LMP1 cell

lines. Normal cells were round, uniform, cells werestained green, the size and the shape were single (Fig.4D1); necrotic cells have oval shaped cells stained orange,the size and shape were single (Fig. 4D2); early apoptoticcells were green, cell shape was irregular, like a crescent(Fig. 4D3); late apoptotic cell nuclei were orange, thenucleus broke into pieces, showing bud-like cellular pro-cesses (Fig. 4D4).

The HLEAFab-MMC immunoconjugate suppressestumor development in a HNE2/LMP1 xenograftmodel

Mice models bearing HNE2/LMP1 xenografts wereestablished and were confirmed for their LMP1-overex-pressing by immunohistochemistry (IHC) method (Fig.5A). As illustrated in Fig. 5B, HNE2/LMP1 xenograftsmice treated with HLEAFab-MMC (1.6 � 10�5 mol/kg,MMC equivalent at 0.053 mg/kg HLEAFab equivalent at8 mg/kg) achieved significant regressions from 20 to40 dayswhen comparedwith controls (P < 0.05). AlthoughMMC (0.053 mg/kg) injection was potent to inhibit tumorgrowth, the efficacy was much lower than that of HLEA-Fab-MMC (MMC: 0.053mg/kg),which suggested an extraadvantage of LMP1-mediated specific tumor targeting. Asshown in Fig. 5C, the HLEAFab-MMC immunoconjugateresulted in an average 55.1% inhibition of HNE2/LMP1tumor growth in comparison with their controls (P < 0.05).

Figure 3. The HLEAFab binds tothe LMP1 extracellular domain innative confirmation. A, LMP1-positive cells, HNE2/LMP1; LMP1-negative cells, HNE2; control, humanhepatoma BEL-7402 cells; and theunrelated antibody Fab fragmentwas used as the negative controlantibody. B, flow cytometry analysis:the fluorescent intensity wassignificantly different betweenLMP1-positive HNE2/LMP1 cellsand LMP1-negative HNE2 cells.Background staining (black line) wasobtained by the unrelated Fab. C,fluorescence microscope analysis ofHNE2/LMP1 cells and HNE2 cells. 2,HLEAFab-stained HNE2/LMP1 cellswith mouse antihuman Fab FITCconjugate; 3 and 4, cell morphologyunder light microscopy; and 1 and 3,HNE2 cell negative control (�200,Olympus digital camera; C).






However, the MMC alone produced 26.5% inhibition,HLEAFabalone5.64%, unrelatedFab 9.63%, andunrelatedFab-MMC 24.4%. So the HLEAFab-MMC immunoconju-gate remarkably enhanced the tumor growth inhibition.During the entire treatment period, the antineoplasticantibiotic MMC, the antibody, and immunoconjugatesadministration were nontoxic according observation ofweight, survival, andanimal activities aswell aspathologicexamination results (Fig. 5D).No other complications suchas anaphylaxis and skin necrosis were detected.

DiscussionIn this study, we successfully prepared the immunocon-

jugate HLEAFab-MMC against LMP1-positive NPC. This

involved the following procedures: (i) The HNE2/LMP1cell line, which has high LMP1 expression, and the LMP1-negative HNE2 cell line were selected for subtractive pan-ning to increase the chances of capturing the phage thatbinds tonativeLMP1. (ii)Wewere able to generate ahumanHLEAFab fragment that binds to the LMP1-positive NPCcell lines. (iii) We showed that the major immunogenicregion of the LMP1 extracellular domain can be used asan immunoconjugate target to kill LMP1-positive NPC celllines in vitro. (iv) An antibody-based immunoconjugateHLEAFab-MMC killed LMP1-positive NPC cells in vitroand significantly slowed tumor growth in a mouse LMP1-positiveNPC transplantationmodel, althoughHLEAFaborMMCalonewas insufficient to generate comparable effects.

Figure 4. Inhibitory effect of MMC, naked antibody or immunoconjugates on growth of NPC HNE2/LMP1 or HNE2 cell lines. A, MTT assay to revealdose-dependent antiproliferative effects of 25 to 200 nmol/L ofHLEAFab-MMC (red;P

In our study, the repeated panningwith living cells andcoated LMP1 EMD protein in microtiter plates ensuredthe enrichment of specific LMP1 EMD–binding phage.After 7 rounds of panning, 1 of 45 randomly selectedphage clones showed evidently positive results (Fig. 2).Moreover, the DNA sequencing results (Table 1) wereidentical to the phagemids that gave the strongest bindingsignal by ELISA, indicating that the phage with highspecificbinding toLMP1were selectedout from thepanel.The ability of the expressed HLEAFab to specifically bindto native LMP1 on the HNE2/LMP1 cell surface wasfurther confirmed by Cell ELISA, fluorescence-activatedcell sorting, and fluorescence staining analyses (Fig. 3).Antibody-based targeting immunotherapy is a well-

established treatment for various tumor types, such asErbB2-positive breast cancer (18), Hodgkin lymphoma,acute myeloid leukemia, colon cancer, lung cancer, mel-anoma, and some of head and neck squamous cell carci-nomas (19–24). Immunoconjugate therapy has not yetbeen applied to NPC, although some antigens have beenfound since many years ago to expresses at high levels onthe surface of these tumor cells (25, 26). This obstacle wasovercome after we discovered that the LMP1 EMD islikely to be as a therapeutic target of the vast majority ofNPC.

The LMP-specific monoclonal antibodies (mAb) CS1-4,S12, and other mAbs are known to recognize intracellularparts of LMP1 (27), and LMP antibody–positive serastained our LMP-expressing cells in indirect immunoflu-orescence assays only when these cells were first permea-bilized. These antibodies and sera, however, were unableto stain these cells without prior permeabilization, sug-gesting that these antibodies recognized only intracellularparts of the protein (28). Therefore these intrabodies arenot fit for the preparation of antibody-based immunocon-jugate unless they are internalized by the target cells (29).Accordingly, we selected LMP1 EMD for a specific targetposition as an extracellular domain antibody is an idealcarrier of antibody-based immunoconjugate.

An antibody per se can be a drug for immunotherapy.Indeed, several neutralizing antibodies have beenapproved by the U.S. Food and Drug Administration forclinical treatment of a variety of diseases. Besides, anti-body can also be used as a carrier to deliver a broad rangeof therapeutics selectively to cancer cells by recognizing aspecific tumormarker on the cell surface. The therapeuticsmight include nucleic acids, chemotherapy compounds,toxins, radionuclides, antibody, cytokine or antibody–ligand fusion proteins, and enzyme prodrugs. Such deliv-ery may reduce or eliminate side effects caused by the

Figure 5. Growth inhibition on tumorxenografts by HLEAFab-MMC.A, positive expression on LMP1tumor xenografts was confirmed byIHC method (�200, Olympus digitalcamera). B, themicewere grouped (8mouse per group) and administeredin the tumor local with 1.6 � 10�5mol/kg HLEAFab, HLEAFab-MMC,MMC, unrelated Fab, unrelated Fab-MMC, or control (N.S., normal saline)with indicated doses on days 1, 4,and 7. Tumor sizewasmeasured onetime per 5 days and tumor volumewas calculated according to theequation: tumor size ¼ width2 �length/2. �, P < 0.05 compared withcontrol. C, at the end of experiment,the mice bearing tumors weresacrificed and tumors were removedand weighed. �, P < 0.05 comparedwith control. D, measurement ofbody weight throughout theexperiments.






drugs in normal cells. For the purposes of immunochem-otherapy, an antibody capable of receptor binding isessential because, in that process, the antibody can helpaccumulate the conjugated molecules in the cancer cells(20, 30). LMP1 has been reported to be overexpressed inmost solid tumors at various levels, making it an idealtarget for antibody-directed drug delivery (31, 32). In thisstudy,we selectedMMC tobe conjugated to an anti-LMP1EMD HLEAFab and made an immunochemotherapeuticagent. Its biologic efficacy was tested in in vitro antitumoractivity via cell proliferation inhibition and apoptosisassays (Fig. 4). The results showed that HLEAFab-MMCwas effective in inhibiting 76% HNE2/LMP1 cell prolif-eration at 200 nmol/L, in contrast to the inhibitory rate of31% by MMC and 22% by HLEAFab. The highest apo-ptosis rate of HNE2/LMP1 cell lines was observed inHLEAFab-MMC group (13.88%) compared with theMMC group (3.04%), HLEAFab group (2.78%), or nega-tive control (2.10%) at the same concentration. These datasuggest that theHLEAFab-MMCconjugate ismore potentfor treating LMP1-positive cancers and has less sideeffects relative to those seen with MMC alone.

The antitumor effect of this immunoconjugate was alsoassessed in vivo (Fig. 5). As illustrated in Fig. 5, 8 micemodels with xenograft tumor were treated with HLEA-Fab-MMC (MMC equivalent 0.053 mg/kg, HLEAFabequivalent at 8 mg/kg) and achieved tumor inhibitionand maintained tumor inhibitory effect for more thananother 20 days. Although MMC injection was potentto inhibit tumor growth, the efficacy was much lowerthan that of HLEAFab-MMC, which suggested the greatadvantage of HLEAFab-mediated specific tumor target-ing. To circumvent the relatively significant errors madeby directly measuring tumor size, tumors were removedand weighed after the mice were sacrificed at the end ofexperiment (day 40). The average tumorweight of HLEA-Fab-MMC treatment group was much smaller than thosefrom other groups. The tumors shrank to 55.1% comparedwith the control group (N.S. injected). In contrast, theMMC-alone treated tumors only shrank to 26.5%. Theseresults clearly showed that HLEAFab-MMC can specifi-cally inhibit tumor growth in vivo. HLEAFab-MMC sig-nificantly inhibits BALB/c nude mice transplantationtumor growth of human HNE2/LMP1 NPC in vivo.

Actually, different type of NPCs and different cells inthe same tumor tissue are not always the same level of

LMP1 expressing. Only 20% to 65%NPCs express LMP1that can be detected by Western blot or immunohisto-chemical analysis and another (25, 33). Although theresults presented here seem promising, the differentlevel of LMP1 expressing in EBV-related carcinomas,may lead to the different biologic efficacy of this Fab-based immunoconjugate. To strengthen the antitumorbiologic efficacy of HLEAFab-MMC to the low LMP1-expressing NPC, our next strategy is that the antibodywill conjugate liposomes or polymer nanoparticlesencapsulating more MMC to elevate potential efficiencyand decrease clearance rate. The conjugation can kill notonly the cancer cells of low-expressing LMP1 by tar-geted efficacy but also the cancer cells of LMP1 negativethrough a bystander effect. This phenomenon was alsoreported by Kovtun and his colleagues (34). On thewhole, the evaluation of LMP1 expression in indivi-duals is required before this therapeutic strategy isapplied in NPC treatment.

In summary, our findings indicate that the LMP1 EMDis a specific marker of humanNPC that can be recognizedby a HLEAFab-based immunoconjugate (HLEAFab-MMC). The immunoconjugate, the first targeting agentbased on a fully human antibody Fab fragment describedso far, is effective both in vitro and in vivo againstNPCcellsand might represent a promising chemotherapeutics thatmerits further preclinical studies.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

AcknowledgmentsThe authors thank Dr. J. Yongjun for helpful discussions, AbMax

Biotechnology Co., Ltd. for their excellent technical assistance, and C.Xuan for helpful image processing as well as Mr. Patrick Karle and Prof.Gordon James Leitch from America for polishing our English.

Grant SupportThe project was supported by Natural Science Foundation of

Jiangsu Province (BK2008481), Technology Support Program of Jiangsu(BE2009152), a grant from State Major Basic Research Development Pro-gram of China (973 Project, 2010CB933902) and Natural Science Founda-tion of Nanjing Medical University (NY2005D2D13).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received September 29, 2011; revised December 2, 2011; acceptedDecember 7, 2011; published OnlineFirst December 14, 2011.

References1. De-Th�e G, Geser A, Day NE, Tukei PM, Williams EH, Beri DP, et al.

Epidemiological evidence for causal relationship between Epstein-Barr virus and Burkitt's lymphoma from Ugandan prospective study.Nature 1978;274:756–61.

2. Morris MA, Dawson CW, Young LS. Role of the Epstein-Barr virus-encoded latent membrane protein-1, LMP1, in the pathogenesis ofnasopharyngeal carcinoma. Future Oncol 2009;5:811–25.

3. Hu LF, Chen F, Zhen QF, Zhang YW, Luo Y, Zheng X, et al. Differencesin the growth pattern and clinical course of EBV-LMP1 expressing and

nonexpressing nasopharyngeal carcinomas. Eur J Cancer 1995;31A:658–60.

4. Smee RI, Meagher NS, Broadley K, Ho T, Williams JR, Bridger GP.Recurrent nasopharyngeal carcinoma: current managementapproaches. Am J Clin Oncol 2010;33:469–73.

5. Chen ML, Tsai CN, Liang CL, Shu CH, Huang CR, Sulitzeanu D, et al.Cloning and characterization of the latentmembrane protein (LMP) of aspecific Epstein-Barr virus variant derived from the nasopharyngealcarcinoma in the Taiwanese population. Oncogene 1992;7:2131–40.

Chen et al.





6. Tarrand JJ, Keating MJ, Tsimberidou AM, O'Brien S, LaSala RP, HanXY, et al. Epstein-Barr virus latent membrane protein 1 mRNA isexpressed in a significant proportion of patients with chronic lympho-cytic leukemia. Cancer 2010:116;880–7.

7. Zhang DW, Zhu J, Chen RJ, Feng ZQ. Expression and purification ofGST-LMP-1 extracellular domain fusion protein of EB virus. Acta UnivMed Nanjing 2008;28:420–3.

8. Jiao YJ, Zhao P, Zhu J, Grabinski T, Feng Z,Guan X, et al. Constructionof human naive Fab library and characterization of anti-met Fabfragment generated from the library. Mol Biotechnol 2005;31:41–54.

9. Ding L, Li LL, Yang J, Tao YG, Ye M, Shi Y, et al. Epstein-Barr virusencoded latent membrane protein 1 modulates nuclear translocationof telomerase reverse transcriptase protein by activating nuclearfactor-kappaB p65 in human nasopharyngeal carcinoma cells. Int JBiochem Cell Biol 2005;37:1881–9.

10. Dantas-Barbosa C, Faria FP, Brigido MM, Maranh~ao AQ. Isolation ofosteosarcoma-associated human antibodies from a combinatorialFab phage display library. J Biomed Biotechnol 2009;2009:157531.

11. Sagane K, Ishihama Y, Sugimoto H. LGI1 and LGI4 bind to ADAM22,ADAM23 and ADAM11. Int J Biol Sci 2008;4:387–96.

12. Saini D, Kala M, Jain V, Sinha S. Targeting the active site of theplacental isozyme of alkaline phosphatase by phage-displayed scFvantibodies selected by a specific uncompetitive inhibitor. BMC Bio-technol 2005;5:33.

13. Shao W, Zhao S, Liu Z, Zhang J, Ma S, Sato JD, et al. Inhibition ofhuman tumor xenograft growth in nude mice by a conjugate ofmonoclonal antibody LA22 to epidermal growth factor receptor withanti-tumor antibiotics mitomycin C. Biochem Biophys Res Commun2006;349:816–24.

14. Lee G, Ge B. Growth inhibition of tumor cells in vitro by usingmonoclonal antibodies against gonadotropin-releasing hormonereceptor. Cancer Immunol Immunother 2010;59:1011–9.

15. ShenG,HuangH, ZhangA, Zhao T, HuS, Cheng L, et al. In vivo activityof novel anti-ErbB2 antibody chA21 alone and with paclitaxel ortrastuzumab in breast and ovarian cancer xenograft models. CancerImmunol Immunother 2011;60:339–48.

16. Kitamura K, Takahashi T, Kotani T,Miyagaki T, YamaokaN, TsurumiH,et al. Local administration of monoclonal antibody-drug conjugate: anew strategy to reduce the local recurrence of colorectal cancer.Cancer Res 1992;52:6323–8.

17. Kanellos J, McKenzie IF, Pietersz GA. Intratumour therapy of solidtumours with ricin-antibody conjugates. Immunol Cell Biol 1989;67:89–99.

18. Jarvinen TA, Liu ET. HER-2/neu and topoisomerase IIa in breastcancer. Breast Cancer Res Treat 2003;78:299–311.

19. Liu C, Strobl JS, Bane S, Schilling JK, McCracken M, Chatterjee SK,et al. Design, synthesis, and bioactivities of steroid linked taxol analo-gues as potential targeted drugs for prostate and breast cancer. J NatProd 2004;67:152–9.

20. Schrama D, Reisfeld RA, Becker JC. Antibody targeted drugs ascancer therapeutics. Nat Rev Drug Discov 2006;5:147–59.

21. Liu G, Yao K,Wang B, Chen Y, Zhou F, Guo Y, et al. Immunotherapy ofEpstein-Barr virus associated malignancies using mycobacterialHSP70 and LMP2A356–364 epitope fusion protein. Cell Mol Immunol2009;6:423–31.

22. Long HM, Parsonage G, Fox CP, Lee SP. Immunotherapy for Epstein-Barr virus-associated malignancies. Drug News Perspect 2010;23:221–8.

23. Hannigan A, Wilson JB. Evaluation of LMP1 of Epstein-Barr virus as atherapeutic target by its inhibition. Mol Cancer 2010;9:184.

24. KleinG, Klein E, KashubaE. Interaction of Epstein-Barr virus (EBV)withhuman B-lymphocytes. Biochem Biophys Res Commun 2010;396:67–73.

25. Young LS, Dawson CW, Clark D, Rupani H, Busson P, Tursz T, et al.Epstein-Barr virus gene expression in nasopharyngeal carcinoma.J Gen Virol 1988;69:1051–65.

26. Salomon DS, Brandt R, Ciardiello F, Normanno N. Epidermal growthfactor related peptides and their receptors in humanmalignancies. CritRev Oncol Hematol 1995;19:183–232.

27. Fang CY, Chang YS, Chow KP, Yu JS, Chang HY. Construction andcharacterization of monoclonal antibodies specific to Epstein-Barrvirus latent membrane protein 1. J Immunol Methods 2004;287:21–30.

28. Lee AW, Law SC, Ng SH, Chan DK, Poon YF, Foo W, et al. Retro-spective analysis of nasopharyngeal carcinoma treated during 1976–1985: late complications followingmegavoltage irradiation. Br J Radiol1992;65:918–28.

29. Rowe M, Finke J, Szigeti R, Klein G. Characterization of the sero-logical response in man to the latent membrane protein and the sixnuclear antigens encoded by Epstein-Barr virus. J Gen Virol 1988;69:1217–28.

30. Gattenl€ohner S, J€orissen H, Huhn M, Vincent A, Beeson D, Tzartos S,et al. A human recombinant autoantibody-based immunotoxin specificfor the fetal acetylcholine receptor inhibits rhabdomyosarcomagrowthin vitro and in a murine transplantation model. J Biomed Biotechnol2010;2010:187621.

31. Fa�hraeus R, Fu HL, Ernberg I, Finke J, Rowe M, Klein G, et al.

Expression of Epstein–Barr virus-encoded proteins in nasopharyngealcarcinoma. Int J Cancer 1988;42:329–38.

32. Elliott K. Epstein-Barr virus-increasing evidence of a link to carcinoma.N Engl J Med 1995;333:724–6.

33. Brooks L, Yao QY, Rickinson AB, Young LS. Epstein-Barr virus latentgene transcription in nasopharyngeal carcinomacells: coexpressionofEBNA1, LMP1, and LMP2 transcripts. J Virol 1992;66:2689–97.

34. Kovtun YV, Audette CA, Ye Y, Xie H, Ruberti MF, Phinney SJ, et al.Antibody-drug conjugates designed to eradicate tumors with homo-geneous and heterogeneous expression of the target antigen. CancerRes 2006;66:3214–21.






2012;11:594-603. Published OnlineFirst December 14, 2011.Mol Cancer Ther Renjie Chen, Dawei Zhang, Yuan Mao, et al.

In Vivo and In VitroExtracellular Domain Inhibits Nasopharyngeal Carcinoma Growth A Human Fab-Based Immunoconjugate Specific for the LMP1

Updated version

10.1158/1535-7163.MCT-11-0725doi:

Access the most recent version of this article at:

Cited articles

http://mct.aacrjournals.org/content/11/3/594.full#ref-list-1

This article cites 31 articles, 4 of which you can access for free at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mct.aacrjournals.org/content/11/3/594To request permission to re-use all or part of this article, use this link


http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-11-0725http://mct.aacrjournals.org/content/11/3/594.full#ref-list-1http://mct.aacrjournals.org/cgi/alertsmailto:[email protected]://mct.aacrjournals.org/content/11/3/594http://mct.aacrjournals.org/

a human fab-based immunoconjugate speciﬁc for the lmp1 … · therapeutic discovery a human...

Documents