research selective cytotoxicity to her2-positive tumor ... · nist(bid)belongs...

2284

Published OnlineFirst April 14, 2010; DOI: 10.1158/1078-0432.CCR-09-2367

Cancer Therapy: Preclinical Clinical
Cancer
Research

Selective Cytotoxicity to HER2-Positive Tumor Cells by aRecombinant e23sFv-TD-tBID Protein Containing aFurin Cleavage Sequence

Fang Wang1, Jing Ren4, Xiu-Chun Qiu5, Li-Feng Wang1, Qing Zhu6, Ying-Qi Zhang3, Yi Huan4,Yan-Ling Meng2, Li-Bo Yao1, Si-Yi Chen7, Yan-Ming Xu1, and An-Gang Yang2

Abstract

Authors'DepartmenImmunologHospital, aMilitary MeAffiliated HShaanxi, CImmunologMedicine, U

Note: SuppResearch O

CorresponFourth Milit86-29-8477or Yan-MinBiochemistXi'an, Shaa

doi: 10.115

©2010 Am

Clin Canc

Do

Purpose: The HER2 antigen is a recognized target on breast cancer cells for immunotherapy. Toovercome the immunogenicity and systemic toxicity of traditional immunotoxins, a novel humanimmunoproapoptotic molecule was developed and its antitumor activity was investigated.Experimental Design: Recombinant e23sFv-TD-tBID, consisting of a single-chain anti-HER2 antibody

fragment linked to a human active truncated Bid by a 10–amino acid residue furin cleavage sequence, wasbacterially expressed. Purified e23sFv-TD-tBID was tested for binding, internalization, and cytotoxic ac-tivity in cell and for tumor localization and antitumor activity in athymic nude mice bearing establishedhuman tumor xenografts.Results: e23sFv-TD-tBID selectively binds to HER2-positive cells and induces apoptotic cell death

in vitro and in vivo. An investigation of its mechanism of action has revealed that e23sFv-TD-tBID wasinternalized on binding to the surface of HER2-positive tumor cells, proteolytically cleaved and trans-ported directly to cytosol. The antitumor activity of e23sFv-TD-tBID was shown in a dose-dependentmanner when injected i.p. into immunodeficient mice bearing human breast carcinomas. Moreover, thisimmunoproapoptotic protein, either given as a single dose or in combination with chemotherapy agents,significantly inhibited tumor growth without any observed toxic side effects on mice. Magnetic resonanceimaging further showed the specific targeting and good penetration of tumors by e23sFv-TD-tBID in vivo.The therapeutic value of e23sFv-TD-tBID to human was shown by its cytotoxic effects on primary patient-derived breast tumor cells but not on endothelial cells.Conclusion: These data suggest that recombinant e23sFv-TD-tBID has therapeutic potential for

HER2-positive tumors and warrant further testing for clinical applications. Clin Cancer Res; 16(8); 2284–94.

©2010 AACR.

The HER2/neu proto-oncogene encodes a 185-kDa trans-membrane glycoprotein (p185HER2) belonging to a subfami-ly of growth factor receptors with intrinsic tyrosine kinaseactivity (1, 2). Overexpression of theHER2/neu gene is found

Affiliations: 1State Key Laboratory of Cancer Biology,t of Biochemistry and Molecular Biology, 2Department ofy, 3Biotechnology Center, 4Department of Radiology, Xijingnd 5Department of Orthopaedics, Tangdu Hospital, Fourthdical University; 6Department of Medical Oncology, The Firstospital, Xi'an Jiaotong University College of Medicine, Xi'an,hina; and 7Department of Molecular Microbiology andy, Norris Comprehensive Cancer Center, Keck School ofniversity of Southern California, Los Angeles, California

lementary data for this article are available at Clinical Cancernline (http://clincancerres.aacrjournals.org/).

ding Authors: An-Gang Yang, Department of Immunology,ary Medical University, Xi'an, Shaanxi 710032, China. Phone:4528; Fax: 86-29-83253816; E-mail: [email protected] Xu, State Key Laboratory of Cancer Biology, Department ofry and Molecular Biology, Fourth Military Medical University,nxi 710032, China. E-mail: [email protected].

8/1078-0432.CCR-09-2367

erican Association for Cancer Research.

er Res; 16(8) April 15, 2010

Researcon May clincancerres.aacrjournals.org wnloaded from

in 25% to 30% of primary human breast cancers and is asso-ciated with a poor clinical outcome (3). Increased HER2 ex-pression is also observed in colon, prostate, lung, and gastriccancers (4–7). Because of preferential expression in tumorcells and extracellular accessibility, HER2 was recognized asan attractive target for antibody-directed therapy. The human-ized HER2 antibody trastuzumab (Herceptin) was the firstHER2-targeted therapeutics approved by the Food and DrugAdministration for the treatment of metastatic breast cancer.Immunotoxins are targeted fusion toxins designed for

cancer therapy, which are composed of the cytotoxic do-main of natural toxins chemically or genetically linked totarget cell-selective peptide ligands such as monoclonalantibodies (mAb), antibody fragments, growth factors, orcytokines. Immunotoxins combine the potency of the tox-in with the specificity of the attached ligand to kill cancercells. The most potent immunotoxins are made from bac-terial or plant toxins. To date, recombinant immunotoxinshave been successfully used in clinical trials for hematolog-ic malignancies (8–11). However, the clinical use of immu-notoxins for treatment of solid tumors is severely limited

h. 29, 2021. © 2010 American Association for Cancer

http://clincancerres.aacrjournals.org/

Translational relevance

Conventional anticancer therapies, such as chemo-therapy and radiation therapy, are characterized bythe lack of tumor cell specificity. Immunotoxins aspotent targeting cytotoxic agents generally consist ofan antibody and a toxin moiety, which efficiently de-stroys the cells following internalization. Although nu-merous immunotoxins have gone into clinical trials,their therapeutic utility has not been fully realizedbecause of toxin immunogenicity and nonspecifictoxicity. To circumvent these problems, we engineerimmunotoxins containing human-derived effector pro-teins. We construct a novel human immunoproapopto-tic molecule, e23sFv-TD-tBID, and explore its ability tomediate tumor-specific apoptosis in vitro and in vivo.This immunoproapoptotic protein, given not only asa single agent but also in combination with chemother-apy, could inhibit tumor growth strongly and prolonglongevity in mice with no toxic side effects. Our find-ings suggest that recombinant e23sFv-TD-tBID hastherapeutic potential for HER2-positive tumors andwarrants further testing for clinical applications.

Antitumor Activity of e23sFv-TD-tBID


by the poor penetration (12), nonspecific cytotoxicity (13),and immunogenicity (14) of the toxin component. Strate-gies to overcome these limitations are being pursued, in-cluding the use of human or mammalian toxins such asRNases (15) or proapoptotic proteins (16). Based on thepreviously constructed immunotoxin e23sFv-PEA40 (17),we developed a set of recombinant immunoproapoptoticmolecules containing anti-HER2 single-chain variable anti-body fragment (scFv), the translocation domain (domainII) of Pseudomonas exotoxin A, and different proapoptoticmolecules such as active caspase-3, caspase-6, and gran-zyme B of human origin for targeted gene therapy ofHER2-overexpressing tumors (18–21).The removal of malignant cells via the apoptotic pathway

can minimize tissue damage or systemic responses. Mito-chondria play an important role in apoptosis triggered bymany stimuli (22, 23). BH3-interacting domain death ago-nist (Bid) belongs to the proapoptotic Bcl-2 protein family,mainly involved in the regulation of apoptosis at the mito-chondrial level. The proteolytic cleavage of Bid to the trun-cated form tBID has been proposed as a prerequisite forits proapoptotic activity. tBID translocates to the mito-chondria and causes the release of apoptotic factors suchas cytochrome c, Smac/DIABLO, or AIF from the mito-chondrial intermembrane space, which can execute caspase-dependent or caspase-independent cell death, respectively(24, 25). Bid interconnects death receptors to the mitochon-drial amplification loop of the intrinsic pathway; therefore,introducing Bid to induce tumor cell apoptosis may be apromising strategy for treatment of cancer.Here, we report a novel human antitumor immuno-

proapoptotic molecule, e23sFv-TD-tBID, consisting of

www.aacrjournals.org

Researcon May clincancerres.aacrjournals.org Downloaded from

anti-HER2 scFv (e23sFv) and active truncated tBID. A10-residue furin cleavage sequence from diphtheria toxin(DT) translocation domain was incorporated between theantibody fragment and tBID. The fusion protein contain-ing this sequence was previously shown to result inmore potent cell-killing activity than those with otherfurin-sensitive sequences (26). We produced this novelimmunoproapoptotic protein in Escherichia coli anddescribed the antitumor activity in vitro and in vivo. Ourfindings indicate that e23sFv-TD-tBID may be a valuableagent for immunotherapy of HER2-positive carcinomas.

Materials and Methods

Construction of the recombinant cDNA encoding e23sFv-TD-tBID. A cDNA encoding e23sFv-TD-tBID was engineeredby two successive PCRs. The PCR product of BIDΔ1-60with a furin cleavage sequence generated by using primersA (5′-CATGCATGCGACGTCCAGCTGACCCAGTCT-3′)and B (5′-TTTGCGGCCGCGAAAGCCGG‐CAATAGAGT-GAGGAGATCTGTGGGCGGCAACCGCAGCAGC-3′) wascloned downstream of e23sFv sequence in pCMV (27).Primers A and C (5′-TTTTCTAGAGTCGACTTAGTC-CATCCCATTTCTGGC-3′) were then used to amplify thewhole coding sequence. The assembled gene was clonedin pQE30 (Qiagen) and confirmed by sequencing.Expression, purification, and refolding of recombinant

proteins. e23sFv-TD-tBID protein was expressed in E. coliM15 cells by 1 mmol/L isopropyl-1-thio-β-galactopyranoside(IPTG) induction 4 h at 37°C. The isolated inclusion bod-ies were resuspended in solubilizing buffer B with 8mol/Lurea and incubated at 4°C for 1 h. After centrifugation andfiltration, the supernatant was applied to the Ni2+ chelateaffinity column at a flow rate of 0.5 mL/min. Nonspecifi-cally bound proteins were removed by extensive washingwith buffer B, 25 and 100 mmol/L imidazole. Bound pro-teins were eluted with 250 mmol/L imidazole in buffer Band refolded by dialyzing against buffers with decreasingdenaturant in 0.1 mol/L NaCl and 20 mmol/L Tris-HCl(pH 7.5): first 6 mol/L urea, then 4 mol/L urea, then2 mol/L urea and 1 mmol/L EDTA, and finally PBS (pH7.4). Dialysis was done at 4°C for 16 h with three buf-fer changes. After each step, samples were centrifuged andsupernatants were subjected to the next round of dialysis.The refolded proteins were quantitated by bicinchoninicacid assays (Pierce).Western blot analysis. Proteins were separated by

SDS-PAGE and transferred to polyvinylidene difluoridemembranes (Amersham Biosciences). Membranes wereincubated with primary antibodies that recognize His(1:5,000; Qiagen), Bid (1:1,000; Cell Signaling Technolo-gy), and HER2 (1:500; Lab Vision and NeoMarkers) over-night at 4°C in PBS-Tween 20. After incubation withhorseradish peroxidase–conjugated secondary antibody(1:2,000; ZhongShanJinQiao), Western blots were visual-ized using an enhanced chemiluminescence kit (Pierce).Cell culture. The human breast cancer cell lines SK-BR-3,

MDA-MB-453, and MCF7 and ovarian cancer cell SKOV-3

Clin Cancer Res; 16(8) April 15, 2010 2285



Wang et al.

2286


(obtained from the American Type Culture Collection) andgastric cancer cell SGC7901 (raised in our laboratory) werecharacterized previously as HER2 positive (2, 18, 26). Cellsweremaintained in RPMI 1640 or DMEM (Invitrogen) with10% fetal bovine serum.Isolation of primary tumor cells from pleural effusion.

After obtaining informed consent, primary breast cancercells were collected from pleural effusion of a patientdiagnosed with medullary carcinoma with liver, bone,and pleura metastases, HER2(+++), ER(−), PR(−). Cellsin pleural fluid diluted with RPMI 1640 were centri-fuged, and the pellet was washed twice with PBS. Cellswere then collected from the interface after centrifugingover a Ficoll density gradient (28, 29). The freshly iso-lated primary breast cancer cells were resuspended inRPMI 1640 with 10% fetal bovine serum, 50 μg/mLpenicillin, and 100 μg/mL streptomycin and identifiedas HER2 positive.Cellular ELISA. HER2-positive SK-BR-3 cells and

HER2-negative human umbilical vascular endothelial cells(HUVEC) were seeded in a 96-well plate (1 × 105 perwell). After blocking with PBS/6% bovine serum albumin,the plate was incubated with 1 μg/mL purified e23sFv-TD-tBID in ELISA buffer (PBS/3% bovine serum albumin) for90 min. The cells were then washed twice and incubatedwith anti-His mAb and horseradish peroxidase anti-mouseIgG for 1 h each. After applying 3,3′,5,5′-tetramethylben-zidine (Sigma-Aldrich), binding was determined from450-nm absorbance values and reported as the mean ofat least three determinations.

Clin Cancer Res; 16(8) April 15, 2010


Flow cytometric analyses. Suspensions of 1 × 105 cellswere washed with PBS buffer containing 0.2% bovine se-rum albumin and 0.05% sodium azide and incubated in100 μL of 1 μg/mL purified e23sFv-TD-tBID for 45 minon ice. After washing, cells were incubated with 1:3,000anti-His mAb for 25 min on ice and 1:100 FITC-linkedsecondary antibody for 20 min at 4°C. The cells wereanalyzed by flow cytometry (BD Bioscience).Internalization of e23sFv-TD-tBID. Cells grown on

coverslips to 60% confluence were incubated withe23sFv-TD-tBID (100 μg/mL) for 2 h at 37°C or 4°C. Cellsthen were washed, fixed, and permeabilized. Intracellulare23sFv-TD-tBID was detected with anti-His mAb, followedby Cy3-anti-mouse IgG. Cells were imaged by confocalmicroscopy (FluoView FV1000, Olympus).In vitro cytotoxicity of e23sFv-TD-tBID. Cells were seed-

ed in 96-well plates at 5,000 per well. After 24 h, serialdilutions of purified protein in 0.2% human serum albu-min were added to the cells and incubated at 37°C foranother 48 h. The effect of e23sFv-TD-tBID on cell growthwas determined by using MTT assays. Cell survival wasexpressed as percentage of viable cells in the presence ofthe protein compared with control cells.Assessment of apoptosis by monitoring exposure of phos-

phatidylserine. An early event in apoptosis is exposure ofphosphatidylserine on the outer cellular membrane. Thisfeature was analyzed by flow cytometry using an AnnexinV-FITC/propidium iodide (PI) kit (BD Pharmingen).Assessment of apoptosis by monitoring loss of mitochondrial

membrane potential. Mitochondrial membrane potential

h. 29, 2021. © 2010 Ame

Fig. 1. The recombinantimmunoproapoptotic moleculee23sFv-TD-tBID. A, schematicrepresentation of e23sFv-TD-tBID codingregion in pQE30. The anti-HER2single-chain antibody fragment e23sFv isfused to the truncated Bid by the 10-residuefurin cleavage site from DT. NH2 terminuscontains the hexahistidine tag. B, bacterialexpression of e23sFv-TD-tBID wasanalyzed by SDS-PAGE. Lane 1, solublefraction of e23sFv-TD-tBID M15 celllysates; lane 2, inclusion body ofe23sFv-TD-tBID. C, SDS-PAGE analysisof purified fusion protein. D, Western blotanalysis of the renatured protein usinganti-His and anti-Bid antibodies.

Clinical Cancer Research

rican Association for Cancer




(ΔΨ) was analyzed by flow cytometry using the cell-permeant green fluorescent lipophilic dyeDiOC6 (MolecularProbes), which is actively taken up by intact mitochondriaof living cells only. After treatment, cells were harvested bycentrifugation (1,000 × g, 5 min), incubated for 30 min at37°C with fresh medium containing 0.1 μmol/L DiOC6,washed once with PBS, and analyzed by flow cytometry.In vivo antitumor activity. Female BALB/c athymic mice

(4-6 wk old) were inoculated s.c. with 5 × 106 humanbreast cancer SK-BR-3 cells in the right mammary fatpad. Eight days later, when tumors reached ∼50 mm3,six animals were randomly assigned to each treatmentgroup and treated i.p. with therapeutic agents. The dosageof these agents tested was as follows: e23sFv-TD-tBID(5 mg/kg, days 0, 2, 4, 6, and 8), Herceptin (10 mg/kg,days 0, 2, 4, 6, and 8), the taxane drug paclitaxel(15 mg/kg, days 1-3), the anthracycline antibiotic doxo-rubicin (5 mg/kg, day 1), and PBS as control. During thetreatment, tumor volumes (V) were measured and calculat-ed using the formula of rational ellipsoid: V = A × B2/2,



where A is the axial diameter and B is the rotationaldiameter. The mice were weighed, and tumor sizes weremeasured two to three times per week. Survival curve plotsand Kaplan-Meier analysis were done using Prism 5(GraphPad Software).Contrast-enhanced magnetic resonance imaging of tumor-

bearing mice. The contrast agent Gold Magnetic nanopar-ticle conjugated with e23sFv-TD-tBID was prepared byLifeGen Co. Magnetic resonance imaging (MRI) was doneon a 3.0T superconducting unit (Siemens magnetomtrio#tim3.0T) using loop coils. T2-weighted SE coronalimages were obtained with the following parameters:repetition time, 4,000 ms; echo time, 90 ms; slice thick-ness, 3 mm; acquisition matrix, 448 × 448; field of view,150 mm; and number of signal averaging, 6.

Results

Recombinant e23sFv-TD-tBID protein was expressed andpurified from E. coli. The chimeric protein (Fig. 1A) is

Fig. 2. Binding and internalization of purified recombinant e23sFv-TD-tBID. A, binding of e23sFv-TD-tBID to HER2-negative HUVEC and HER2-positiveSK-BR-3 cells in a cellular ELISA. Data are representative from two independent experiments done in triplicate each. B, flow cytometric analysis of purifiede23sFv-TD-tBID (open histograms) or his-NusA protein (negative control, filled histograms) binding to HER2-positive cell lines of different origins orHER2-negative HUVEC indicated that the hexahistidine tag at the NH2 terminus did not affect binding specificity. C, HER2-specific binding activity ofe23sFv-TD-tBID detected by competitive flow cytometric analysis with FITC-labeled anti-His antibody. SK-BR-3 cells were incubated with purified protein(green curve) or his-NusA (filled histograms). For competition, cells were incubated with 10 μg/mL anti-HER2 mAb and 1 μg/mL e23sFv-TD-tBID (red curve).D, internalization of e23sFv-TD-tBID in SK-BR-3 cells as visualized by confocal microscopy. Cells were incubated with e23sFv-TD-tBID at 4°C (a) orat 37°C (b). The punctate staining indicates internalized fusion proteins. Magnification, ×600.




Wang et al.

2288


composed of anti-HER2 e23sFv and active tBID separatedby a 10-residue sequence containing a furin cleavage site(A187GNRVRRSVG196) from DT translocation domain.The recombinant cDNA was cloned in a T5 promoter-based E. coli expression vector (pQE30) equipped witha hexahistidine affinity tag for purification. The re-combinant plasmid was transformed into E. coli, and theIPTG-induced target proteins were detected by SDS-PAGE.As shown in Fig. 1B, the observed molecular weightof e23sFv-TD-tBID was ∼43 kDa, consistent with theexpected size. These fusion proteins, mainly existing in in-clusion bodies (Fig. 1B, lane 2), were solubilized with8 mol/L urea and purified by Ni2+ chelate affinity chroma-tography under denaturing conditions (Fig. 1C). The pro-teins were then refolded by dialysis against graduallydecreased urea concentrations (from 4 to 2 mol/L) andfinally in PBS (pH 7.4). The purity of the target proteinreached >90%, and it was further analyzed by Western blotwith anti-His and anti-Bid antibodies (Fig. 1D).Purified e23sFv-TD-tBID protein could bind and subse-

quently be internalized into HER2-positive tumor cells. Bind-ing of purified e23sFv-TD-tBID protein to HER2-positivecells was investigated by cellular ELISA and flow cytome-try. As shown in Fig. 2A, purified e23sFv-TD-tBID proteinbound to HER2-overexpressing breast cancer cell SK-BR-3but not to HER2-negative HUVECs. To further analyzethe binding specificity of e23sFv-TD-tBID proteins, flowcytometric and competitive flow cytometric analyses weredone. The data showed that e23sFv-TD-tBID proteincould bind to the surfaces of all five HER2-positive cells



(SK-BR-3, MDA-MB-453, MCF7, SKOV-3, and SGC7901).Although we detected various levels of e23sFv-TD-tBIDbinding to the different cells due to the differences inHER2 expression, no binding to HER2-negative HUVECswas detected (Fig. 2B), confirming the specificity of the re-combinant e23sFv-TD-tBID for HER2. Furthermore, bind-ing of the e23sFv-TD-tBID protein was highest with theHER2-overexpressing SK-BR-3 cell line. This interactionwas completely blocked by addition of 10 μg/mL anti-HER2 mAb (Fig. 2C), indicating that the attachment oftBID to anti-HER2 scFv did not alter the binding specificityof e23sFv.To exert its cytotoxicity, internalization of e23sFv-TD-

tBID is required. When SK-BR-3 cells were incubated withe23sFv-TD-tBID at 4°C for 2 hours, the protein was largelyfound on the cell surface; however, at the permissive tem-perature of 37°C, the fusion protein was internalized intocytoplasmic vesicle structures (Fig. 2D), which were iden-tified as early endosomes by their pattern of distribution(26). These observations indicated that the fusion proteinsbound the cell surface and were internalized.Purified e23sFv-TD-tBID protein recognized breast cancer

cells from patients and preserved the ability to kill HER2-positive tumor cells after incubation in human serum. Bind-ing of e23sFv-TD-tBID to freshly isolated breast cancercells was investigated by immunofluorescence microscopicanalysis of paraformaldehyde-fixed primary cells afterconfirming the expression level of HER2 (SupplementaryFig. S1; Fig. 3A). With his-NusA protein serving as anegative control, specific staining with e23sFv-TD-tBID

h. 29, 2021. © 2010 Ame

Fig. 3. Binding ability of e23sFv-TD-tBID tobreast cancer cells from patients. Primarybreast cancer cells were freshly isolatedfrom the pleural effusion of one patient withmetastatic breast cancer and wereanalyzed. A, flow cytometric analysis ofHER2 expression level. B, confocalmicroscopic analysis of e23sFv-TD-tBIDbinding on human primary breast tumorcells was done. Bound e23sFv-TD-tBIDwas detected with anti-His mAb followed byanti-mouse Cy3-conjugated antibody.Magnification, ×600. a, his-NusA proteinwas the negative control.


rican Association for Cancer




was detected on primary breast cancer cells (Fig. 3B),suggesting that the hexahistidine tag at the NH2 terminusof e23sFv-TD-tBID did not affect its binding properties.The stability of immunoproapoptotic protein under

physiologic conditions is required for their potentialas therapeutic agents. To estimate the stability ofe23sFv-TD-tBID, we incubated purified e23sFv-TD-tBIDproteins in human serum at 37°C for varying time pe-riods before testing their integrity as a functional pro-tein determined by cytotoxicity assays on SK-BR-3 cells.The e23sFv-TD-tBID proteins retained nearly full cyto-toxic activity after incubation in human serum for upto 48 hours (Supplementary Fig. S2), suggesting thatthey were folded properly and very stable under thoseconditions.Purified e23sFv-TD-tBID protein could induce HER2-

positive tumor cell death, which was dependent on furinactivity and could be blocked by endosome/lysosome inhib-itor bafilomycin A1. To characterize the cytotoxic activityof the recombinant e23sFv-TD-tBID in vitro, we evaluatedthe proliferation of several HER2-positive cells after incu-



bation with escalating doses of e23sFv-TD-tBID. Growthinhibition of HER2-positive cells was documented byMTT assays. Cell-killing effects were observed against a va-riety of HER2-positive cell lines, with a calculated IC50 of∼25 ng/mL on SK-BR-3 cells, 48 ng/mL on MDA-MB-453cells, 35 ng/mL on SKOV-3 cells, and 40 ng/mL onSGC7901 cells (Fig. 4A). Furthermore, e23sFv-TD-tBIDdid not affect the growth of HER2-negative HUVECsat concentration up to 1 μg/mL (Fig. 4A) or even up to10 μg/mL (data not shown). These results suggest a corre-lation between surface expression levels of HER2 and sen-sitivity of cells to e23sFv-TD-tBID.The effects of e23sFv-TD-tBID on the induction of apo-

ptosis of bothHER2-positive breast cancer cell lines and pri-mary breast cancer cells were examined by flow cytometry.Treatment of HER2-positive and HER2-negative cells with1 μg/mL e23sFv-TD-tBID for 12 hours resulted in pro-nounced induction of apoptosis as measured by AnnexinV-FITC and PI staining (Fig. 4B). e23sFv-TD-tBID–mediated apoptosis was characterized by the reduction ofmitochondrial transmembrane potential (Fig. 4C). This

Fig. 4. In vitro cytotoxicity of e23sFv-TD-tBID. A, dose-response curves of HUVEC, MCF7, SK-BR-3, MDA-MB-453, SKOV-3, and SGC7901 cell lines treatedfor 48 h with e23sFv-TD-tBID. B, apoptosis induction was assessed by Annexin V/PI staining. Data are representative of three independent experiments.C, for analysis ofΔΨm, cells were incubated with DiOC6 for 30min and subjected to flow cytometric analysis. D, effects of e23sFv-TD-tBID on SK-BR-3 cells inthe presence or absence of inhibitors of vacuolar-type H+-ATPase (baf), furin protease (dec-RVKR-cmk), and trans-Golgi network (BFA).




Wang et al.

2290


suggests that the observed cytotoxic effect of this fusionprotein is mediated through an apoptotic mechanism.Immunotoxins bind specifically to the cell surface target,

and the pathway of internal translocation to an intra-cellular compartment for cytotoxic activity is different fordifferent toxins (30–32). Here, we investigated thepathway of e23sFv-TD-tBID from the extracellular spaceto the cytosol. SK-BR-3 cells were treated with e23sFv-TD-tBID (100 ng/mL) with or without the specific endosome/lysosome inhibitor bafilomycin A1 (baf; 10 nmol/L), thefurin inhibitor dec-RVKR-cmk (25 μmol/L), or the proteintransport inhibitor brefeldin A (BFA; 10 nmol/L). After48 hours, baf and dec-RVKR-cmk inhibited the cell-killingactivity of e23sFv-TD-tBID, but BFA did not influence itsactivity (Fig. 4D). Because inactivation of endosome/lysosome or furin blocks its entry to the cytosol and its



cytotoxic activity, we conclude that internalized e23sFv-TD-tBID, after cleavage by furin in the endosome, wasdirectly translocated to the cytosol.Purified e23sFv-TD-tBID proteins inhibited HER2-positive

tumor growth without toxic side effects. The in vivo therapeu-tic effects of e23sFv-TD-tBID were tested in nude micebearing human breast cancer xenografts. SK-BR-3 cells(5 × 106) were implanted into the fat pad of nude mice.Treatment was started on day 8 when the tumors reached∼50 mm3. Animals were given three single i.p. injectionswith 1, 5, or 10 mg/kg every other day for a total of fivedoses. The control groups received PBS. Data shownin Fig. 5A showed that the tumor growth was obviouslyinhibited in all the three groups after e23sFv-TD-tBIDtreatment in a significant dose-dependent way as thetumor growth in 5 or 10 mg/kg almost stopped (Fig. 5A).

Fig. 5. Antitumor activity of e23sFv-TD-tBID in vivo. A, antitumor effect of i.p. injections of e23sFv-TD-tBID doses of 1, 5, or 10 mg/kg on SK-BR-3tumors growing in nude mice. Mice were s.c. injected with 5 × 106 SK-BR-3 cells, and treatment was initiated when tumor volume reached ∼50 mm3.Points, mean tumor volume versus time; bars, SD. Arrows showed the day intervals of e23sFv-TD-tBID injection. B, effects of combined treatment withe23sFv-TD-tBID and chemotherapeutic drugs on HER2-overexpressing breast carcinoma xenografts. Mice (n = 6/group) were s.c. injected with 5 × 106

SK-BR-3 cells, and treatment was initiated when tumor volume reached ∼50 mm3. C, Kaplan-Meier survival curve. D, TUNEL assay of apoptotic cells intumor tissues from e23sFv-TD-tBID–treated mice. PBS served as negative control.






clincancerres.aaDownloaded from


Fig. 6. Toxicity and specificity ofe23sFv-TD-tBID on mice. A, toxicity ofe23sFv-TD-tBID on mice monitored by animalweight loss. B, H&E staining of liver, kidney,spleen, lung, and heart tissues from xenograftmice treated with e23sFv-TD-tBID.Magnification, ×200. C, immunohistochemicalstaining was done on cancer, liver, kidney,spleen, lung, and heart tissues, and there is noe23sFv-TD-tBID detected besides cancertissue from e23sFv-TD-tBID–treated group.Magnification, ×200. D, contrast-enhancedthree-dimensional MRI of mice bearinghuman breast tumors treated with 200 μLof MR contrast agent Gold Magneticnanoparticle coupled with (b) or without (a)e23sFv-TD-tBID (protein to chelator ratio of1:2) at preinjection and 6, 12, and 24 h after i.v.administration. Note the extensive uptake inthe tumors.


Research. on May 29, 2021. © 2010 American Association for Cancercrjournals.org


Wang et al.

2292


The taxane drug paclitaxel and the anthracycline anti-biotic doxorubicin are two of the most active chemother-apeutic agents for breast cancer treatment. Treatment withe23sFv-TD-tBID, Herceptin, paclitaxel, or doxorubicin sig-nificantly slowed the tumor growth compared with thecontrol group (P < 0.01). Doxorubicin or paclitaxel com-bined with e23sFv-TD-tBID significantly reduced tumorvolumes compared with either agent alone (P < 0.05), in-dicating that combination therapy with e23sFv-TD-tBIDenhanced the antitumor activity of chemotherapy(Fig. 5B). Kaplan-Meier survival analysis showed that micetreated with e23sFv-TD-tBID, Herceptin, or chemothera-peutics lived significantly longer than untreated mice(Fig. 5C). The average survival time of mice treated withchemotherapeutic agents was shorter than e23sFv-TD-tBID–treated or Herceptin-treated groups, likely due tothe higher systemic toxicity of chemotherapeutic agents.Apoptotic cells in tumor tissues were identified by the ter-minal deoxynucleotidyl transferase–mediated dUTP nickend labeling (TUNEL) assay (Fig. 5D). The data in Fig. 5Band C also showed that treatment with 5 mg/kg of e23sFv-TD-tBID could get the same effect as Herceptin in 10mg/kg.The animal toxicity at each of these dose levels was

evaluated by body weight of the mice, and no significantdifferences were seen (Fig. 6A). Compared with untreatedcontrols, the level of alanine aminotransferase/aspartateaminotransferase activity in the plasma of treated micedid not change significantly (data not shown). Consistentwith these findings, pathologic changes neither in liver norin any other tissue were detected with histologic analysisafter e23sFv-TD-tBID treatment (Fig. 6B). By immunohis-tochemical staining, no cross-reactivity was seen in liver,kidney, spleen, lung, and heart tissues (Fig. 6C). However,as shown in Supplementary Fig. S3, e23sFv-TD-tBID couldstill recognize mouse HER2 on mouse mammary tumorEMT6 cells by immunofluorescence staining.To measure tumor deposition, mice bearing human

breast tumors were injected i.v. with e23sFv-TD-tBID conju-gated to Gold Magnetic nanoparticles. Their localization atdifferent time points is shown in Fig. 6D. The low signal in-tensity in the tumor tissue was shown by three-dimensionalMRI within 6 hours after injection. The signal intensitygradually increased 24 hours later. Accumulation ofe23sFv-TD-tBID was greatest in tumors with relatively littlenormal tissue distribution (background levels). These re-sults showed that e23sFv-TD-tBID can quickly accumulateand efficiently penetrate into tumor masses. We concludethat e23sFv-TD-tBID can specifically target the HER2-overexpressing tumors with potent antitumor activity butwithout damaging normal tissues.

Discussion

In this study, human immunoproapoptotic moleculeswere designed to circumvent some of the problems en-countered with earlier immunotoxins, such as dose-limiting toxicities presenting as vascular leak syndrome,thrombocytopenia, hepatic damage, and immunogenicity



of powerful bacterial or plant toxins. HER2 has long beenrecognized and considered as a promising target forimmunotherapeutic applications, including mAbs (33),vaccines (34), and immunotoxins (35, 36).Bid is the molecular linker bridging various peripheral

death pathways to the central mitochondria pathway. Bidcleavage and activation can be exerted by different pro-teases, such as caspase-8, granzyme B, calpain, or cathepsinB, suggesting Bid as a key factor in many paradigms of celldeath (37). Truncated Bid is much more potent than full-length Bid in activating Bax or Bak, leading to mitochondri-al function loss and the release of multiple proapoptoticfactors. The 15-kDa tBID is smaller than other proapoptoticmolecules (e.g., caspase-3, caspase-6, granzyme B, and AIF)and may facilitate better tumor penetration, improvepharmacokinetics, and reduce immunogenicity. Such aprotein would be preferred in the construction of animmunoproapoptotic agent.Our long-term goal is to develop humanized immuno-

toxins for cancer therapy. The fusion protein we describehere is an improved form of larger immunoproapoptoticmolecules previously constructed with exogenous translo-cation domains. e23sFv-TD-tBID consists of an anti-HER2scFv and the truncated form of Bid. The 10–amino acid DTtranslocation domain was chosen to link e23sFv and tBIDbecause the proteolytically cleavable spacer was shown toenhance cell-killing activity (38, 39). We used a prokaryot-ic expression system to produce biologically active e23sFv-TD-tBID, which retained the binding specificity for HER2of the parental scFv and was rapidly internalized intoHER2-positive target cells.The intracellular mechanism of e23sFv-TD-tBID was fur-

ther conducted. The data show that cytotoxicity of e23sFv-TD-tBID was significantly diminished by the endosome/lysosome inhibitor and furin inhibitor but not by thetrans-Golgi network inhibitor, indicating that e23sFv-TD-tBID accessed the cytosol depending on endosome/lyso-some processing and furin activity and not through thetrans-Golgi network (40). Based on the functional studies,we conclude that e23sFv-TD-tBID kills target cells by acomplex series of steps that are initiated by binding to cellsurface, followed by endocytosis, proteolytic processing,and translocation into the cytosol. In cytosol, active tBIDanchors the mitochondrial membrane, leading to irrevers-ible mitochondrial damage and the release of cytochromec to induce apoptotic cell death. Here, we report for thefirst time the mechanism of the antitumor action of thishuman immunoproapoptotic protein and contribute tothe rational design of immunoconjugates with improvedproperties.Of therapeutic interest, the immunoproapoptotic pro-

teins we prepared effectively killed HER2-positive cell linesand primary patient-derived breast cancer cells without af-fecting the endothelial cell growth. e23sFv-TD-tBID inhib-ited proliferation of HER2-overexpressing SK-BR-3 cells atan IC50 of 25 ng/mL. The antitumor activity of e23sFv-TD-tBID was also shown in a dose-dependent manner in vivowithout any obvious drug-related toxic reactions at doses






up to 10 mg/kg. Furthermore, in combination with che-motherapy, e23sFv-TD-tBID therapy of solid tumors mayachieve better response, potentially reducing the requireddrug dosage and lowering systemic toxicity. Of criticalimportance to the eventual clinical development ofe23sFv-TD-tBID will be examination of the maximumtolerated dose toxicity profile and efficacy studies of thisfusion protein in other well-characterized human HER2-positive xenograft models.Based on previous findings (41, 42) that suggested a

correlation between MRI signal changes and drug distribu-tion, we initiated the development of tumor-specific con-trast agents to evaluate the distribution of the therapeuticagents. MRI confirmed good tumor-targeting propertiesand penetration of e23sFv-TD-tBID. Because of its smallmolecular weight, e23sFv-TD-tBID targeted tumors ra-pidly, and quickly cleared from the blood, leading to alower exposure overall to normal tissues.In conclusion, we have produced and characterized a

novel immunoproapoptotic protein, e23sFv-TD-tBID.The potent and selective cytotoxicity in vitro, favorabletumor localization properties, as well as the powerfulantitumor effect of e23sFv-TD-tBID in vivo suggest itspotential in clinical application in the treatment of



HER2-positive carcinomas. Because e23sFv-TD-tBID isfully humanized, stable, and selectively cytotoxic to targetcells and can be produced in large amounts, it deservesfurther development for cancer therapy.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank Dr. Weihua Wang for the constructive discussions and YunxinCao for technical assistance in flow cytometer.

Grant Support

National High Technology Research and Development Program of Chi-na grant 2007AA021104 (A-G. Yang), Program for Changjiang Scholarsand Innovative Research Team in University grant IRT0459 (A-G. Yang),National Natural Science Foundation of China grant 30530810 (A-G.Yang), and National Natural Science Foundation of China grant30870497 (Y-M. Xu).

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 toindicate this fact.

Received 08/30/2009; revised 01/29/2010; accepted 02/16/2010;published OnlineFirst 04/06/2010.

References
1. Schechter AL, Hung MC, Vaidyanathan L, et al. The neu gene: an
erbB-homologous gene distinct from and unlinked to the geneencoding the EGF receptor. Science 1985;229:976–8.

2. Yarden Y, Sliwkowski MX. Untangling the ErbB signalling network.Nat Rev Mol Cell Biol 2001;2:127–37.

3. Slamon DJ, Godolphin W, Jones LA, et al. Studies of the Her-2/neuproto-oncogene in human breast and ovarian cancer. Science 1989;244:707–12.

4. Kapitanovic S, Radosevic S, Kapitanovic M, et al. The expression ofp185(HER-2/neu) correlates with the stage of disease and survival incolorectal cancer. Gastroenterology 1997;112:1103–13.

5. Osman I, Mikhail M, Shuch B, et al. Serum levels of shed Her2/neuprotein in men with prostate cancer correlate with disease progres-sion. J Urol 2005;174:2174–7.

6. Turken O, Kunter E, Cermik H, et al. Prevalence and prognosticvalue of c-erbB2 expression in non-small cell lung cancer (NSCLC).Neoplasma 2003;50:257–61.

7. Oshima CT, Lanzoni VP, Iriya K, Forones NM. C-erbB-2 oncoproteinin gastric carcinoma: correlation with clinical stage and prognosis. IntJ Biol Markers 2001;16:250–4.

8. Pastan I, Hassan R, FitzGerald DJ, Kreitman RJ. Immunotoxintherapy of cancer. Nat Rev Cancer 2006;6:559–65.

9. Kreitman RJ. Immunotoxins for targeted cancer therapy. AAPS J2006;8:E532–51.

10. Kreitman RJ, Stetler-Stevenson M, Margulies I, et al. Phase II trial ofrecombinant immunotoxin RFB4(dsFv)-PE38 (BL22) in patients withhairy cell leukemia. J Clin Oncol 2009;27:2983–90.

11. Pasqualucci L, Flenghi L, Terenzi A, et al. Immunotoxin therapy ofhematological malignancies. Haematologica 1995;80:546–56.

12. Messmer D, Kipps TJ. Treatment of solid tumors with immunotoxins.Breast Cancer Res 2005;7:184–6.

13. Posey JA, Khazaeli MB, Bookman MA, et al. A phase I trial of thesingle-chain immunotoxin SGN-10 (BR96 sFv-PE40) in patients withadvanced solid tumors. Clin Cancer Res 2002;8:3092–9.

14. Benbrook DM. An ELISA method for detection of human antibodiesto an immunotoxin. J Pharmacol Toxicol Methods 2002;47:169–75.

15. De Lorenzo C, Nigro A, Piccoli R, D'Alessio G. A new RNase-based

immunoconjugate selectively cytotoxic for ErbB2-overexpressingcells. FEBS Lett 2002;516:208–12.

16. Aqeilan R, Yarkoni S, Lorberboum-Galski H. Interleukin 2-Bax: anovel prototype of human chimeric proteins for targeted therapy.FEBS Lett 1999;457:271–6.

17. Chen SY, Yang AG, Chen JD, et al. Potent antitumour activity of anew class of tumour-specific killer cells. Nature 1997;385:78–80.

18. Jia LT, Zhang LH, Yu CJ, et al. Specific tumouricidal activity of asecreted pro-apoptotic protein consisting of HER2 antibody andconstitutively active caspase-3. Cancer Res 2003;63:3257–62.

19. Xu YM, Wang LF, Jia LT, et al. A caspase-6 and anti-human epider-mal growth factor receptor-2 (HER2) antibody chimeric moleculesuppresses the growth of HER2-overexpressing tumours. J Immunol2004;173:61–7.

20. Zhao J, Zhang LH, Jia LT, et al. Secreted antibody/granzyme Bfusion protein stimulates selective killing of HER2-overexpressingtumour cells. J Biol Chem 2004;279:21343–8.

21. Yu CJ, Jia LT, Meng YL, et al. Selective pro-apoptotic activity of asecreted recombinant antibody/AIF fusion protein in carcinomasoverexpressing HER2. Gene Ther 2006;13:313–20.

22. Desagher S, Martinou JC. Mitochondria as the central control pointof apoptosis. Trends Cell Biol 2000;10:369–77.

23. Igney FH, Krammer PH. Death and anti-death: tumor resistance toapoptosis. Nat Rev Cancer 2002;2:277–88.

24. Gross A, Yin XM, Wang K, et al. Caspase cleaved BID targetsmitochondria and is required for cytochrome c release, whileBCL-XL prevents this release but not tumor necrosis factor-R1/Fasdeath. J Biol Chem 1999;274:1156–63.

25. Li H, Zhu H, Xu CJ, Yuan J. Cleavage of BID by caspase 8 mediatesthe mitochondrial damage in the Fas pathway of apoptosis. Cell1998;94:491–501.

26. Wang T, Zhao J, Ren JL, et al. Recombinant immunoproapoptoticproteins with furin site can translocate and kill HER2-positive cancercells. Cancer Res 2007;67:11830–9.

27. Qiu XC, Xu YM, Wang F, et al. Single-chain antibody/activated BIDchimeric protein effectively suppresses HER2-positive tumor growth.Mol Cancer Ther 2008;7:1890–9.




Wang et al.

2294


28. Ethier SP, Mahacek ML, Gullick WJ, Frank TS, Weber BL. Differentialisolation of normal luminal mammary epithelial cells and breastcancer cells from primary and metastatic sites using selective media.Cancer Res 1993;53:627–35.

29. Grimshaw MJ, Cooper L, Papazisis K, et al. Mammosphere culture ofmetastatic breast cancer cells enriches for tumorigenic breast cancercells. Breast Cancer Res 2008;10:R52.

30. Thrush GR, Lark LR, Clinchy BC, Vitetta ES. Immunotoxins: anupdate. Annu Rev Immunol 1996;14:49–71.

31. Pastan I, Chaudhary V, FitzGerald DJ. Recombinant toxins as noveltherapeutic agents. Annu Rev Biochem 1992;61:331–54.

32. Ogata M, Fryling CM, Pastan I, FitzGerald DJ. Cell-mediatedcleavage of Pseudomonas exotoxin between Arg279 and Gly280generates the enzymatically active fragment which translocates tothe cytosol. J Biol Chem 1992;267:25396–401.

33. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al. Trastuzumabafter adjuvant chemotherapy in HER2-positive breast cancer. N EnglJ Med 2005;353:1659–72.

34. Baxevanis CN, Sotiriadou NN, Gritzapis AD, et al. ImmunogenicHER2/neu peptides as tumor vaccines. Cancer Immunol Immunother2006;55:85–95.

35. Batra JK, Kasprzyk PG, Bird RE, Pastan I, King CR. Recombinant



anti-erbB2 immunotoxins containing Pseudomonas exotoxin. ProcNatl Acad Sci U S A 1992;89:5867–71.

36. Azemar M, Schmidt M, Arlt F, et al. Recombinant antibody toxinsspecific for ErbB2 and EGF receptor inhibit the in vitro growth ofhuman head and neck cancer cells and cause rapid tumor regressionin vivo. Int J Cancer 2000;86:269–75.

37. Yin XM. Bid, a BH3-only multi-functional molecule, is at the crossroad of life and death. Gene 2006;369:7–19.

38. O'Hare M, Brown AN, Hussain K, et al. Cytotoxicity of a recombinantricin-A-chain fusion protein containing a proteolytically-cleavablespacer sequence. FEBS Lett 1990;273:200–4.

39. Goyal A, Batra JK. Inclusion of a furin-sensitive spacer enhancesthe cytotoxicity of ribotoxin restrictocin containing recombinantsingle-chain immunotoxins. Biochem J 2000;345:247–54.

40. Lord JM, Roberts LM. Toxin entry: retrograde transport through thesecretory pathways. J Cell Biol 1998;140:733–6.

41. Lidar Z, Mardor Y, Jonas T, et al. Convection-enhanced delivery ofpaclitaxel for the treatment of recurrent malignant glioma: a phase I/IIclinical study. J Neurosurg 2004;100:472–9.

42. Mardor Y, Roth Y, Lidar Z, et al. Monitoring response to convection-enhanced Taxol delivery in brain tumor patients using diffusion-weighted magnetic resonance imaging. Cancer Res 2001;61:4971–3.




2010;16:2284-2294. Published OnlineFirst April 14, 2010.Clin Cancer Res Fang Wang, Jing Ren, Xiu-Chun Qiu, et al. Cleavage SequenceRecombinant e23sFv-TD-tBID Protein Containing a Furin Selective Cytotoxicity to HER2-Positive Tumor Cells by a

Updated version

10.1158/1078-0432.CCR-09-2367doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2010/04/06/1078-0432.CCR-09-2367.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/16/8/2284.full#ref-list-1

This article cites 42 articles, 17 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/16/8/2284.full#related-urls

This article has been cited by 1 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/16/8/2284To request permission to re-use all or part of this article, use this link

Research. on May 29, 2021. © 2010 American Association for Cancerclincancerres.aacrjournals.org Downloaded from


http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-09-2367

http://clincancerres.aacrjournals.org/content/suppl/2010/04/06/1078-0432.CCR-09-2367.DC1

http://clincancerres.aacrjournals.org/content/16/8/2284.full#ref-list-1

http://clincancerres.aacrjournals.org/content/16/8/2284.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/16/8/2284


research selective cytotoxicity to her2-positive tumor ... · nist(bid)belongs...

Documents