Human Angiogenin Fused to Human CD30 Ligand (Ang-CD30L) Exhibits Specific Cytotoxicity Against CD30-Positive Lymphoma1

Home , CD30

[CANCER RESEARCH 61, 8737–8742, December 15, 2001] Human Angiogenin Fused to Human CD30 Ligand (Ang-CD30L) Exhibits Specific Cytotoxicity against CD30-positive Lymphoma1

Michael Huhn, Stephanie Sasse, Mehmet K. Tur, Ba¨rbel Matthey, Timo Schinko¨the, Susanna M. Rybak, Stefan Barth,2 and Andreas Engert Fraunhofer IME, Department of Pharmaceutical Product Development, 52074 Aachen, Germany [M. K. T., M. H., S. B.]; Laboratory of Immunotherapy, Department I of Internal Medicine, University Hospital Cologne, 50931 Cologne, Germany [S. S., B. M., T. S., A. E.]; Developmental Therapeutics Program, Division of Cancer Treatment and Diagnosis, National Cancer Institute-Frederick Cancer Research and Development Center, Frederick, MD 21702 [S. M. R.]

ABSTRACT first and then to administer ITs to kill residual H-RS cells. One of the most promising target antigens for immunotherapy of malignant lym- A number of different immunotoxins composed of cell-specific target- phoma such as Hodgkin’s lymphoma or anaplastic large cell lym- ing structures coupled to plant or bacterial toxins have increasingly been phoma is the CD30 receptor. This antigen was originally discovered evaluated for immunotherapy. Because these foreign proteins are highly immunogenic in humans, we have developed a new CD30 ligand-based on cultured H-RS cells using the moab Ki-1 (1). The gene encoding fusion toxin (Ang-CD30L) using the human RNase angiogenin. The com- the CD30 receptor molecule (2) is located on chromosome 1p36. The pletely human fusion gene was inserted into a pET-based expression naturally occurring CD30 ligand has also been identified and cloned plasmid. Transformed Escherichia coli BL21(DE3) were grown under (3). The CD30/CD30 ligand system triggers cytolytic cell death in osmotic stress conditions in the presence of compatible solutes. After malignant lymphoma cell lines and induces proliferation and cytokine ␤ isopropyl -D-thiogalactoside induction, the Mr 37,000 His10-tagged Ang- production in T cells or neutrophils (4). Moabs against CD30 have CD30L was directed into the periplasmic space and functionally purified been explored as vehicles for cytostatic drugs (5) or plant toxins (6). by a combination of metal ion affinity followed by enterokinase cleavage ITs constructed with anti-CD30 moabs chemically linked to catalyt- of the His -Tag and molecular size chromatography. The characteristics 10 ically active toxins demonstrated specificity and potent antitumor of the recombinant protein were assessed by ELISA, flow cytometry, and ؉ activity against Hodgkin’s lymphoma cells in vitro and in mouse toxicity assays showing specific activity against CD30 Hodgkin-derived cells. Specific binding activity of Ang-CD30L was verified by competition models (7, 8). A total of 12 patients with refractory relapsed with anti-CD30 monoclonal antibody Ki-4 and commercially available Hodgkin’s lymphoma were treated with an IT constructed by conju- CD30L-CD8 chimeric protein. Ang-CD30L showed RNase activity in gating the anti-CD30 moab BerH2 to Saporin-6 (Ber-H2-S6; Ref. 9). vitro. The human recombinant immunotoxin showed significant toxicity Rapid regression of tumor masses ranging from 50% to Ͼ 75% toward several CD30-positive cell lines (HDLM-2, L1236, KM-H2, and (lasting 2–4 months) were observed in ϳ50% of patients underlining ؍ L540Cy) and exhibited highest cytotoxicity against L540 cells (IC50 8 the validity of CD30 as a target antigen in HD (10, 11). The major ng/ml) as determined by cell proliferation assays. CD30 specificity was obstacles observed in this and other trials are unspecific toxicities, confirmed by competitive toxicity assays. This is the first report on the mainly related to the vascular leak syndrome, and the immunogenicity specific cytotoxicity of a recombinant completely human fusion toxin with of the foreign proteins resulting in only a limited number of applica- possibly largely reduced immunogenicity for the treatment of CD30- tions (12, 13). positive malignancies. By using phage display technology, we generated several Hodgkin’s lymphoma-specific scFvs, which were then genetically INTRODUCTION fused to ETA’ (14). The rITs recovered from periplasmic space of E. coli grown under osmotic stress conditions in the presence of Hodgkin’s lymphoma is one of the best suited malignancies for compatible solutes were highly functional in terms of specific in vitro targeted immunotherapy for several reasons: (a) the lymphocyte ac- 3 binding and in vivo cytotoxic activity. tivation marker CD30 is expressed in high copy numbers on H-RS Humanized or human antibody fragments have been used to reduce cells; (b) the number of malignant cells that needs to be killed is small the immune response against xenogeneic proteins and indeed attenu- because the majority of cells in Hodgkin’s lymphoma are nonmalig- ated immunogenicity in patients (15). We reported recently on a nant reactive cells; (c) Hodgkin tumors are usually well vascularized, recombinant fusion protein consisting of recombinant human CD30 suggesting sufficient access of an immunotherapeutic agent like an IT ligand fused to ETA’ and showed the CD30-specific cytotoxicity of to the target cells; and (d) although Hodgkin’s lymphoma is known to this monomeric protein (16). Because human RNases are present in respond well to chemotherapy, residual tumor cells remaining after extracellular fluids, human plasma, and tissues, they might be less first-line treatment have been demonstrated to correlate with the immunogenic when used as toxic component of ITs. Human Ang, a probability of a later relapse. Because the selective elimination of human plasma protein with 65% homology to RNase A (17), was residual H-RS cells might enhance the number of patients being cured, documented as a potent inhibitor of protein synthesis in cell-free it seems feasible to eradicate bulky disease by conventional therapy extracts and when injected into Xenopus oocytes (18). The Mr 14,000 single chain polypeptide is not cytotoxic toward a wide range of Received 10/30/00; accepted 10/16/01. cultured cells but exhibits specific cytotoxic activity when fused to a The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with ligand, which is internalized on binding to the target cell (19, 20). 18 U.S.C. Section 1734 solely to indicate this fact. In this paper, we report our results with the human CD30L coding 1 Supported in part by the Deutsche Forschungsgemeinschaft Grant SFB502 (to A. E. region genetically fused to the human Ang gene. We demonstrate that and S. B.). 2 To whom requests for reprints should be addressed, at Fraunhofer IME, Department this new and completely human rIT Ang-CD30L exhibits specific and of Pharmaceutical Product Development, Worringer Weg 1, 52074 Aachen, Germany. effective destruction of CD30ϩ lymphoma cells. Phone: 49-241-80-28399; Fax: 49-241-871062; E-mail: [email protected]. 3 The abbreviations used are: H-RS, Hodgkin/Reed-Sternberg; IT, immunotoxin; HD, Hodgkin’s Disease; ETA’, deletion mutant of Pseudomonas exotoxin; MOPS, 3-[N- MATERIALS AND METHODS morpholino]propanesulfonic acid; Ang, angiogenin; IPTG, isopropyl ␤-D-thiogalactoside; moab, monoclonal antibody; EtBr, ethidium bromide; XTT, 2,3-bis[2-methoxy-4-nitro- 5-sulfophenyl]-2H-tetrazolium-5-carboxanilide inner salt; rIT, recombinant immunotoxin; Bacterial Strains, Oligonucleotides, and Plasmids. E. coli XL1-blue ϩ scFv, single-chain variable fragment; SEC, size exclusion chromatography. {supE44 hsdR17 recA1 endA1 gyr A46 thi relA1 lacF’[pro AB lacIq 8737

⌬ r lacZ M15 Tn10(tet )]} was used for propagation of plasmids and E. coli Enterokinase Cleavage. To remove the NH2-terminal His10-Tag, highly Ϫ Ϫ Ϫ BL21[DE3; F ompT hsSB(rB MB ) gal dcm DE3] as host for synthesis of specific cleavage of rIT (behind asp-asp-asp-asp-lys) by recombinant enteroki- rITs. Synthetic oligonucleotides were synthesized by Eurogentec (Seraing, nase (1 unit/50 ␮g rIT) was performed in duplicate using the enterokinase Belgium). Plasmids were prepared by the alkaline lysis method and purified cleavage kit (Novagen, Abingdon, United Kingdom) as described by the using plasmid kits from Qiagen (Hilden, Germany). Restriction fragments or manufacturer. Functional rIT was finally purified using SEC on the BioLogic PCR products were separated by horizontal agarose gel electrophoresis and workstation by separation in PBS (pH 7.4). Purified protein was analyzed by extracted with Qiaex II (Qiagen). Cloning into plasmid vectors was performed SDS-PAGE and quantified by densitometry (GS-700 Imaging Densitometer; by standard methods. Bio-Rad) after Coomassie staining in comparison with BSA standards and Plasmid Construction. CD30L cDNA gene was released from plasmid verified by Bradford assays (Bio-Rad). pDC202 (21) by PCR using the oligonucleotides CD30LBack (CAC-TTG- SDS-PAGE and Western Blotting. SDS-PAGE and Western blotting GAT-CAG-TCA-ATT-TTC-CGT-CGT-CCG-gaa-ttc-CAG-AGG-ACG-GAC- were performed as described (31). Ang-CD30L was detected by anti-Ang- TCC-ATT-CCC-AAC-TCA-CCT; underlined: EcoRI consensus; double un- biotin moab (Sigma Chemical Co.). Bound antibody was stained with an derlined: 3Ј-Ang region) and CD30LFor (cgg-cgg-ggt-acc-TTA-GTC-TGA- alkaline-phosphatase-conjugated antimouse-IgG moab (Sigma Chemical Co., ATT-ACT-GTA-TAA-GAA-GAT-GGA-CAA; underlined: KpnI consensus). Deisenhofen, Germany) and a solution of Tris-HCl (pH 8.0) and 0.2 mg/ml The Ang coding region was similarly amplified using the oligonucleotides naphtol-AS-Bi-phosphate (Sigma Chemical Co.) plus 1 mg/ml Fast-Red AngBack (tat-tat-aag-ctt-CAG-GAT-AAC-TCC-AGG-TAC-ACA-CAC- (Serva, Heidelberg, Germany). TTC-CTG; underlined: HindIII consensus) and AngFor (AGG-TGA-GTT- Sandwich-ELISA. The binding activity of Ang-CD30L was determined by GGG-AAT-GGA-GTC-CGT-CCT-CTG-gaa-ttc-CGG-ACG-ACG-GAA-AAT- CD30 receptor Sandwich-ELISA as documented previously (DAKO CD30 TGA-CTG-ATC-CAA-GTG; underlined: EcoRI consensus; double kit). The Microwell Strips was precoated with mouse moab to human CD30 underlined: 5Ј-CD30L region). PCR was performed in 50-␮l reaction mixtures receptor. After washing, 30 units/ml of CD30 (in Tris-buffered saline contain-

containing 20 mM Tris-HCl (pH 8.3), 10 mM KCl, 2 mM MgCl2,6mM ing 0.5% BSA and 0.05% Tween 20) were incubated for2hatroom

(NH4)2SO4, 0.1% Triton X-100, and 100 mM deoxynucleotide triphosphate, 50 temperature. After washing, samples were incubated for 2 h. Unbound IT was nmol of each primer, and 1 unit plaque-forming unit DNA polymerase; 20–50 removed, wells were thoroughly washed and subsequently incubated with ng of template DNA was used. General cycling conditions: 94°C for 5 min, anti-Ang moab-biotin (R&D Systems) for 1 h. After washing, 100 ␮lof 60°C hold (hot start); 94°C for 1 min, 50°C for 3 min, and 72°C for 6 min Streptavidin-POD (Roche Diagnostics, Mannheim, Germany; 1:5000 in Tris- (30ϫ); and 72°C for 3 min in a Biometra Personal Cycler. Both fragments buffered saline containing 0.5% BSA and 0.05% Tween 20) were added and were joined together by SOE-PCR (22), restricted with HindIII/KpnI, and samples incubated for1hatroom temperature. Bound Ang-CD30L was finally inserted into a HindIII/KpnI-restricted expression vector pBR substi- detected by addition of 100 ␮l O-phenylenediamine-dihydrochloride (Sigma tuting for a modified ricin A-chain gene behind a enterokinase-cleavable Chemical Co.) after incubation for 1 h. Absorbance at 405 nm was measured

His10-Tag (23). The resulting pBMAng-CD30L plasmid was transformed into with an ELISA reader (MWG Biotech, Ebersberg, Germany). Native moabs E. coli BL21(DE3). were used as controls. Cell Culture. All of the cell lines including the CD30ϩ Hodgkin-derived Flow Cytometry Analyses. Cell binding activity of Ang-CD30L expressed cell lines L540, L540cy (24), HDLM-2 (25), KM-H2 (26), L1236 (27), and in E. coli BL21(DE3) was evaluated by flow cytometry as described (31). CD30Ϫ -MyZ (28; kindly provided by B. Do¨rken, Berlin, Germany) were Briefly, cell suspensions containing 5 ϫ 106 L540cy cells/ml were washed in

cultivated in complex medium (RPMI 1640) supplemented with 10% (v/v) PBS containing 0.2% w/v BSA and 0.05% w/v sodium azide (PBS/BSA/N3Ϫ) ␮ ␮ heat-inactivated FCS, 50 g/ml penicillin, 100 g/ml streptomycin, and 2 mM and incubated for 30 min in blocking solution (10% BSA in PBS/N3Ϫ). After ␮ L-glutamine. All of the cells were cultured at 37°Cina5%CO2 in air one wash in PBS/BSA/N3Ϫ, the cells were incubated in 100 l of the Ang- atmosphere, cultivated in complex medium (RPMI 1640), and supplemented CD30L protein extract for2hat4°C. The cells were washed again three times with 10% (v/v) heat-inactivated FCS, 50 ␮g/ml penicillin, 100 ␮g/ml strep- and treated with anti-Ang moab (R&D Systems) for 1 h. After three more tomycin, and 2 mML-glutamine. All of the cells were cultured at 37°Cina5% washes, the cells were incubated with FITC-labeled goat-antimouse immuno-

CO2 atmosphere. globulin for1hat4°C. The cells were subsequently washed three times and Periplasmic Expression and Purification of the Recombinant analyzed on a FACScan (Becton Dickinson, Heidelberg, Germany). For com- Ang-CD30L. Recombinant ITs were expressed under the control of the IPTG- petition, 10 ␮g/ml CD30L-CD8-Biotin (Ancell, Bayport, MN) or moab Ki-4 inducible T7lac promoter in E. coli BL21(DE3) as described recently (29). were used; Ang-CD30L-his bound to L540 cells was detected by moab anti- Briefly, bacteria were grown overnight at 26°C in Terrific Broth (30) contain- penta-His (Qiagen) and FITC-labeled antimouse IgG. ␮ ing 50 g/ml kanamycin and 0.5 mM ZnCl2. The culture was diluted 30-fold RNase Assay. To measure the ribonucleolytic activity of human Ang, a

in 200 ml of the same medium. At an A600 of 2 it was supplemented with 0.5 tRNA assay was performed as described (32). In this study, conditions were M sorbitol, 4% NaCl, and 10 mM glycine betaine, and incubated at 26°C for and optimized as followed. Yeast-tRNA (0.6 ␮g; Sigma Chemical Co.) were additional 30–60 min. Expression of rIT was induced by the addition of 2 mM incubated with enterokinase-cleaved and molecular size-purified Ang-CD30L IPTG at 26°C. Later (15 h), cells were harvested by centrifugation at 3,700 ϫ g in 30 mM Tris-HCl (pH 7.5) containing 30 mM NaCl in a total volume of 40 for 10 min at 4°C. For all of the next steps, tubes were kept on ice. The ␮l. After 20 h at 4°C, the reaction was terminated by addition of sample buffer bacterial pellet was centrifuged and its wet weight determined. Cells were consisting of 48% formaldehyde, 48% glycerol supplemented with 0.25% frozen at Ϫ196°C. After thawing, the cells were resuspended in 75 mM bromphenol blue and 2% 1 M sodium phosphate (pH 7.5). The samples were Tris-HCl (pH 8), 300 mM NaCl, 1 capsule of protease inhibitors/50 ml loaded onto a 1% agarose gel containing 20 mM MOPS, 5 mM NaOAc, 1 mM (Complet, Roche Diagnostics, Mannheim, Germany), 5 mM DTT, 10 mM EDTA, 65 mM formaldehyde (pH 7; Qiagen) and electrophoresed in gel EDTA, and 10% (v/v) glycerol, and sonicated six 6 times for 30 s at 200 W. running buffer (20 mM MOPS,), 5 mM NaOAc, 1 mM EDTA, 65 mM formal- The periplasmic fraction was recovered after centrifugation at 21,000 ϫ g for dehyde, and 0.1% EtBr (pH 7). 30 min at 4°C and transferred to 75 mM Tris-HCl (pH 8), 1 M NaCl, and 10% Colorimetric Cell Proliferation Assay. The cytotoxic effect of Ang- glycerol using Hitrap desalting columns (Pharmacia, Freiburg, Germany). rIT CD30L of the target cells was determined by measurement of metabolization was partially purified by immobilized metal-ion affinity chromatography that of XTT to a water soluble orange formazan dye as published previously (29). contained nickel-nitriloacetic acid chelating Sepharose (Qiagen) on a BioLogic Various dilutions of the recombinant toxin were distributed in 100-␮l aliquots workstation (Bio-Rad, Mu¨nchen, Germany). Bound protein was eluted with in 96-well plates. Target cells (2–4 ϫ 104) in 100-␮l aliquots of complete 250 mM imidazole in 75 mM Tris-HCl (pH 8), 1 M NaCl, and 10% glycerol. medium were added, and the plates were incubated for 48 h at 37°C. Subse- Fractions containing the Ang-CD30L were pooled, concentrated by ultrafil- quently, the cell cultures were pulsed with 100-␮l fresh culture medium tration, and finally purified using SEC with Bio-Prep SE-100/17 columns supplemented with XTT/phenazine methosulfate (final concentrations of 0.3 (Bio-Rad) on the BioLogic workstation. Protein was eluted with PBS (pH 7.4) mg and 0.383 ng, respectively) for 4 h. The spectrophotometrical absorbances and 1 M NaCl, analyzed by SDS-PAGE, quantified by densitometry (GS-700 of the samples were measured at 450 and 650 nm (reference wavelength) with Imaging Densitometer; Bio-Rad) after Coomassie staining in comparison with an ELISA reader (MWG Biotech). The concentration required to achieve an

BSA standards and verified by Bradford assays (Bio-Rad). IC50 of protein synthesis relative to untreated control cultures was calculated 8738

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2001 American Association for Cancer Research. HUMAN IT FOR CD30-POSITIVE LYMPHOMA TREATMENT by nonlinear regression using SPSS. All of the measurements were done in triplicate. Competition experiments were performed on L540Cy cells in the presence or absence of 10 ␮g/ml purified Ki-4 moab as described (14) or 10 ␮g/ml CD30L-CD8 chimeric protein (Ancell) as competitors for a dilution series of Ang-CD30L.

RESULTS Construction and Expression of Ang-CD30L. PCR-amplified and assembled Ang-CD30L cDNA was inserted into the expression vector pBR, thus replacing the ricin A-chain gene (Fig. 1, a–c). The pBR expression vector is a derivative of the pET27b plasmid and contains an IPTG-inducible lac operator, a pelB signal peptide followed by an enterokinase-cleavable His10-Tag (23). Successful cloning was verified by sequence analyses. After cultivation of transformed E. coli BL21(DE3) under osmotic stress conditions in the presence of compatible solutes, recombinant foreign protein was functionally directed into the periplasmic space.

The Mr 34,000 Ang-CD30L was purified by combinations of immobilized metal-ion affinity chromatography followed by enterokinase cleavage and SEC (Fig. 1d) to a final minimum concentration of about 1 mg/4 g cell paste from 1 liter of bacterial shaking cultures (Fig. 2). Binding Properties of Ang-CD30L. Binding properties of the enterokinase-cleaved Ang-CD30L against CD30ϩ target cell lines was measured by Sandwich-ELISA using captured soluble human Fig. 2. Functional activity of purified rITs after molecular size chromatography. a, CD30 receptor (Fig. 2a) and flow cytometry. Binding properties of the elution profile monitored at 280 nm (OOO) combined with binding activity of the eluted fractions documented by Sandwich-ELISA (⅐⅐⅐⅐). Captured recombinant human recombinant protein against the native target antigen was documented CD30 receptor was incubated with various concentrations of Ang-CD30L. Specifi- by flow cytometry using an anti-Ang moab. Ang-CD30L bound to the cally bound IT was detected after incubation with polyclonal goat anti-Ang-biotin CD30ϩ Hodgkin-derived cell line L540Cy but not to the CD30Ϫ followed by Streptavidin-POD. Converted phosphatase substrate (O-phenylenediamine-dihydrochloride) was measured as absorbance at 405 nm. b, cell-binding activity HD-MyZ (Fig. 2b). CD30-specificity was documented by competitive of Ang-CD30L evaluated by flow cytometry analysis. L540cy cells were incubated ϩ flow cytometric experiments. Binding of Ang-CD30L against CD30 with PBS (gray filling), His-tagged Ang-CD30L (black filling), enterokinase-cleaved Ang-CD30L (black line), or moab Ki-4 (doted line) for 60 min at 4°C. Cells were L540 cells was inhibited by 10 ␮g/ml chimeric CD30L-CD8-Biotin stained with mouse-anti-Ang moab and with goat-antimouse FITC-conjugated anti- (Fig. 3d)or10␮g/ml moab Ki-4 (Fig. 3e) confirming this moab as a body. Immunofluorescence (FL1 channel) was measured by flow cytometry using a competitor for CD30L as described recently (33). FACScan (Becton Dickinson).

Fig. 1. Cloning and expression of Ang-CD30L. a, schematic structure of the Ang-CD30L insert in the E. coli expression plasmid. The expression module is composed of the signal peptide of the pectate lyase of Erwinia carotovora (pelB), the IPTG inducible lac operator, a synthetic and enterokinase-cleavable His10 cluster (His-Tag), the human Ang gene, and the CD30L coding region. b, plasmid pBRAng-CD30L for the expression of the rIT in E. coli. The plasmid contains the expression module, a kanamycin resistance gene (kanr), an E. coli origin of replication (pBR322 origin), an M13 origin of replication (f1 origin), and the lactose repressor gene (lacI). The toxic moiety (Ang) is genetically fused to the binding structure (CD30L). c, Ang-CD30L open reading frame after SOE-PCR and restriction with HindIII/KpnI (Lane 1, Ang-CD30L (ϳ980 bp); M, 100 bp ladder. d, Ang-CD30L protein after enterokinase cleavage and SEC as documented after SDS-PAGE and Comassie staining; M, Prestained Marker (Bio-Rad); 1, Ang-CD30L (34 kDa). 8739

DISCUSSION In this study we report the construction of a novel, completely human recombinant fusion toxin targeting CD30 receptor. For this purpose we fused the CD30L coding region to a modified human Ang. The major findings to emerge from our study are: (a) functional Ang-CD30L was directly isolated from the periplasmic space of E. coli cultured under osmotic stress conditions in the presence of compatible solutes by a combination of immobilized metal affinity

and molecular size chromatography; (b) a free NH2-terminal region of the fusion protein was generated after enterokinase cleavage of the ϩ His10-Tag; (c) Ang-CD30L bound to CD30 cells as documented by flow cytometry and CD30 receptor ELISA and binding was competed by the addition of moab Ki-4 or chimeric CD30L-CD8; and (d) the recombinant toxin exhibited specific cytotoxicity toward CD30 receptor-expressing Hodgkin-derived cell lines HDLM-2, KM-H2, L1236, L540, and L540Cy, but not against CD30-negative HD-MyZ cells. Different concentrations of CD30 receptor on the surface of the various Hodgkin-derived cell lines tested (34), might be the reason for

varying IC50s. CD30 specificity was again demonstrated by competition experiments.

Table 1 Growth inhibition of Hodgkin-derived cell lines after incubation with Ang- CD30L as documented by cell viability assays ϩ Ϫ L540Cy, L540, L1236, HDLM-2, KM-H2 (all CD30 ), or HD-MyZ (CD30 ) were treated with enterokinase-cleaved rIT. L540 was also treated with rIT-HIS and their ability to metabolize the tetrazolium salt, XTT, to a water-soluble formazan salt was measured as absorbance at 450 and 650 nm (reference wavelength) according to Jost et al. (42). Measurements were performed in triplicate. Results are presented as percentage of untreated control cells. The IC50 was calculated by nonlinear regression using SPSS.

Cell line IC50 (ng/ml) L540 ϩ Ang-CD30L 8 (ϩ/Ϫ 0,5) L540Cy ϩ Ang-CD30L 13,5 (ϩ/Ϫ 1,5) KM-H2 ϩ Ang-CD30L 11,5 (ϩ/Ϫ 2) L1236 ϩ Ang-CD30L 11,5 (ϩ/Ϫ 1,1) HDLM-2 ϩ Ang-CD30L 9,8 (ϩ/Ϫ 0,2) HD-MyZ ϩ Ang-CD30L — L540 ϩ His-Ang-CD30L — Fig. 3. CD30-specific binding activity of Ang-CD30L documented by competitive flow cytometric analyses. L540 cells were incubated with: a, Ang-CD30L; b, CD30L- CD8-biotin; c, moab Ki-4; d, Ang-CD30L competed with CD30L-CD8-biotin; and e, CD30L-CD8-biotin competed with moab Ki-4. For competition with 10 ␮g/ml CD30L- CD8-biotin (Ancell) or moab Ki-4 were used, Ang-CD30L-his bound to L540 cells was detected by moab anti-penta-His (Qiagen, Du¨sseldorf, Germany) and FITC-labeled antimouse IgG. Cells were stained with PBS as negative controls (gray curves and solid line in e).

RNase Assay. Human Ang and RNase A (Sigma Chemical Co.) were used as positive controls to establish the RNase assay. The ribonucleolytic activity of enterokinase-cleaved Ang-CD30L is shown in Fig. 4. The major product generated by endonucleolytic cleavage of tRNA was visualized after 1% denaturing agarose gel electrophoresis and EtBr staining. After 20 h of incubation at 4°C, enterokinase- cleaved Ang-CD30L showed ribonucleolytic activity against yeast tRNA (0.6 ␮g) at an absolute concentration of 150 ng. Cytotoxic Activity in Vitro. To characterize the cytotoxic activity of the rIT in vitro, we measured the proliferation of different target cells on incubation with different amounts or rIT. As shown in Fig. 5, Ang-CD30L released by enterokinase-cleavage was cytotoxic to ϩ ␮ CD30 L540 with a calculated median IC50 of 8 ng/ml. The fusion Fig. 4. Ribonucleolytic activity of Ang. Yeast-tRNA (0.6 g; Sigma Chemical Co.) toxin also showed effective killing of all CD30-positive cell lines were incubated with enterokinase-cleaved and molecular size-purified Ang-CD30L in 30 mM Tris-HCl (pH 7.5), containing 30 mM NaCl in a total volume of 40 ␮l. After 20 h at tested in parallel, including HDLM-2, KM-H2, L1236, L540, and 4°C, the reaction was terminated by addition of sample buffer consisting of 48% form- Ϫ L540Cy (Table 1). The CD30 Hodgkin-derived cell line HD-MyZ aldehyde and 48% glycerol supplemented with 0.25% bromphenol blue and 2% 1 M was not affected at recombinant protein concentrations of up to 10 sodium phosphate (pH 7.5). The samples were loaded onto a 1% agarose gel containing 20 mM MOPS, 5 mM NaOAc, 1 mM EDTA, and 65 mM formaldehyde (pH 7; Qiagen), and ␮g/ml. His-tagged Ang-CD30L also did not exhibit any cytotoxic electrophoresed in gel running buffer (20 mM MOPS), 5 mM NaOAc, 1 mM EDTA, 65 mM effect against L540 cells. In addition, the cytotoxicity against CD30ϩ formaldehyde, and 0.1% EtBr (pH 7). Lanes A–C, enterokinase-cleaved and molecular size-purified Ang-CD30L (150ng, 75 ng, and 37 ng); Lanes D, E, I, and J, negative cells was specifically inhibited by an excess of purified moab Ki-4 or controls; Lanes K–M, human Ang (75 ng, 35 ng, and 18 ng); Lanes F–H, RNase A (75 ng, CD30L-CD8 chimeric protein. 35 ng, and 18 ng). 8740

Fig. 5. Growth inhibition of Hodgkin-derived cell lines after incubation with Ang-CD30L as documented by cell viability assays. a, L540 (CD30ϩ)or HD-MyZ (CD30Ϫ) were treated with enterokinase- cleaved rIT, and their ability to metabolize the XTT to a water-soluble formazan salt (formed by mito- chondrial dehydrogenase activity) was measured as absorbance at 450 and 650 nm (reference wavelength) in a cell viability assay (43). Measurements were performed in triplicate. Results are presented as percentage of untreated control cells. b, cytotoxic activity of Ang-CD30L (⅐⅐⅐⅐) on L540 and its competition with 10 ␮g/ml CD30L-CD8 (OO ⅐⅐ OO)or moab Ki-4 (OOO) HD-MyZ (----) cells were used as negative controls. Measurements were again performed in triplicate.

One of the major problems identified in clinical trials with chem- recently (16). Additionally, the concentration needed to achieve 50% ically linked ITs is the development of neutralizing antibodies against inhibition of cell growth for Ang-CD30L is in a very similar range to both the murine IgG and the toxic moiety resulting in a limited the Pseudomonas exotoxin-based anti-CD30 IT Ki-4(scFv)-ETA’ (6 number of applications in about 40–60% of the patients (35, 36). This ng/ml; Refs. 29, 42). Cross-linking of the CD30 target antigen does problem might, at least in part, be overcome by using recombinant not seem to be necessary for the monomeric IT to be internalized. This DNA technology to construct smaller and less immunogenic ITs. Very phenomenon correlates with our data from recombinant CD30L-ETA’ recently, the first clinical trial with such a construct, anti-Tac(Fv)- and anti-CD30 scFvs fused to ETA’ as binding structures. Both PE38 (LMB-2), in patients with CD25ϩ hematological malignancies binding activity and cytotoxic efficiency of the recombinant, mono- has been published suggesting a reduced antibody response (37, 38). valent fusion toxin might be additionally increased by creating poly- To additionally diminish this immunogenicity, humanized antibody valent modifications thereby enhancing internalization efficacy by fragments or cytotoxic activities derived from human resources (39) receptor mediated endocytosis. might be used. Human CD30 ligand has been described recently as a In summary, we demonstrated the construction of a recombinant, type II membrane protein. Its COOH-terminal, extracellular sub- tag-free fusion toxin, which specifically binds to the CD30 receptor. domain shares sequence homologies with corresponding regions of The use of modified recombinant human CD30L as a binding struc- tumor necrosis factor ␣, tumor necrosis factor ␤, and CD40L (40). ture for recombinant immunotherapeutics against Hodgkin’s lym- Thus, we have used the putative human soluble CD30L portion as phoma and in combination with Ang will be additionally investigated binding subdomain for the construction of a Pseudomonas exotoxin- for in vivo application in humans. based fusion protein (16). Although ETA’ was genetically fused to the COOH-terminus of the CD30L, which normally remains free when ACKNOWLEDGMENTS displayed on the surface of eukaryotic cells, the protein showed CD30-specific binding and cytotoxic activity. Because the NH ter- We thank Dr. Hans-Jo¨rg Gruss for providing CD30L. We also thank Silke 2 Drillich for cloning assistance and Gisela Scho¨n for performing the toxicity minus is crucial to the cytotoxicity of RNases (39), CD30L was in this assays. study fused to the COOH-terminus of human Ang, thus inserting the extracellular spacer region of CD30L followed by the COOH-terminal REFERENCES receptor binding domain (41). The periplasmically expressed, non- glycosylated recombinant fusion protein used in this experimental 1. Schwab, U., Stein, H., Gerdes, J., Lemke, H., Kirchner, H., Schaadt, M., and Diehl, V. Production of a monoclonal antibody specific for Hodgkin and Sternberg-Reed setup exhibited specific cytotoxicity after removal of the NH2- cells of Hodgkin’s disease and a subset of normal lymphoid cells. Nature (Lond.), 299: 65–67, 1982. terminal His10-Tag. The calculated cytotoxic activity of Tag-free ϭ 2. Durkop, H., Latza, U., Hummel, M., Eitelbach, F., Seed, B., and Stein, H. Molecular Ang-CD30L against L540 cells was three times higher (IC50 8 cloning and expression of a new member of the nerve growth factor receptor family ϭ ng/ml) compared with CD30L-ETA’ (IC50 24 ng/ml) published that is characteristic for Hodgkin’s disease., Cell, 68: 421–427, 1992. 8741

3. Gruss, H. J., and Dower, S. K. The TNF ligand superfamily and its relevance for 22. Horton, R. M., Cai, Z. L., Ho, S. N., and Pease, L. R. Gene splicing by overlap human diseases. Cytokines Cell. Mol. Ther., 1: 75–105, 1995. extension: tailor-made genes using the polymerase chain reaction. Biotechniques, 8: 4. Wiley, S., Goodwin, R., and Smith, C. Reverse signaling via CD30 ligand. J. Immu- 528–535, 1990. nol., 157: 3635–3639, 1996. 23. Matthey, B., Engert, A., and Barth, S. Ki-4(scFv)-Ricin, a new recombinant immu- 5. Sahin, U., Hartmann, F., Senter, P., Pohl, C., Engert, A., Diehl, V., and Pfreundschuh, notoxin targeting CD30 isolated from E.coli periplasmic space exhibits specific M. Specific activation of the prodrug mitomycin phosphate by a bispecific anti-CD30/ cytotoxic activity against Hodgkin cells in vitro. submitted, 2000. anti-alkaline phosphatase monoclonal antibody. Cancer Res., 50: 6944–6948, 1990. 24. Kapp, U., Wolf, J., von Kalle, C., Tawadros, S., Rottgen, A., Engert, A., Fonatsch, C., 6. Terenzi, A., Bolognesi, A., Pasqualucci, L., Flenghi, L., Pileri, S., Stein, H., Kadin, Stein, H., and Diehl, V. Preliminary report: growth of Hodgkin’s lymphoma derived M., Bigerna, B., Polito, L., Tazzari, P. L., Martelli, M. F., Stirpe, F., and Falini, B. cells in immune compromised mice. Ann. Oncol. 3(Suppl 4): 21–23, 1992. Anti-CD30 (BERϭH2) immunotoxins containing the type-1 ribosome-inactivating 25. Drexler, H. G., Gaedicke, G., Lok, M. S., Diehl, V., and Minowada, J. Hodgkin’s proteins momordin and PAP-S (pokeweed antiviral protein from seeds) display disease derived cell lines HDLM-2 and L-428: comparison of morphology, immu- powerful antitumour activity against CD30ϩ tumour cells in vitro and in SCID mice. nological and isoenzyme profiles. Leuk. Res., 10: 487–500, 1986. Br. J. Haematol., 92: 872–879, 1996. 26. Kamesaki, H., Fukuhara, S., Tatsumi, E., Uchino, H., Yamabe, H., Miwa, H., 7. Engert, A., Burrows, F., Jung, W., Tazzari, P. L., Stein, H., Pfreundschuh, M., Diehl, Shirakawa, S., Hatanaka, M., and Honjo, T. Cytochemical, immunologic, chromo- V., and Thorpe, P. Evaluation of ricin A chain-containing immunotoxins directed somal, and molecular genetic analysis of a novel cell line derived from Hodgkin’s disease. Blood, 68: 285–292, 1986. against the CD30 antigen as potential reagents for the treatment of Hodgkin’s disease. 27. Wolf, J., Kapp, U., Bohlen, H., Kornacker, M., Schoch, C., Stahl, B., Mucke, S., von Cancer Res., 50: 84–88, 1990. Kalle, C., Fonatsch, C., Schaefer, H. E., Hansmann, M. L., and Diehl, V. Peripheral 8. Schnell, R., Linnartz, C., Katouzi, A. A., Schon, G., Bohlen, H., Horn-Lohrens, O., blood mononuclear cells of a patient with advanced Hodgkin’s lymphoma give rise to Parwaresch, R. M., Lange, H., Diehl, V., Lemke, H., and Engert, A. Development of permanently growing Hodgkin-Reed Sternberg cells. Blood, 87: 3418–3428, 1996. new ricin A-chain immunotoxins with potent anti-tumor effects against human 28. Bargou, R. C., Mapara, M. Y., Zugck, C., Daniel, P. T., Pawlita, M., Dohner, H., and Hodgkin cells in vitro and disseminated Hodgkin tumors in SCID mice using Dorken, B. Characterization of a novel Hodgkin cell line, HD-MyZ, with my- high-affinity monoclonal antibodies directed against the CD30 antigen. Int. J. Cancer, elomonocytic features mimicking Hodgkin’s disease in severe combined immunod- 63: 238–244, 1995. eficient mice. J. Exp. Med., 177: 1257–1268, 1993. 9. Pasqualucci, L., Wasik, M., Teicher, B. A., Flenghi, L., Bolognesi, A., Stirpe, F., 29. Barth, S., Huhn, M., Matthey, B., Klimka, A., Galinski, E. A., and Engert, A. Polito, L., Falini, B., and Kadin, M. E. Antitumor activity of anti-CD30 immunotoxin Compatible-solute-supported periplasmic expression of functional recombinant pro- (Ber-H2/saporin) in vitro and in severe combined immunodeficiency disease mice teins under stress conditions. Appl. Environ. Microbiol., 66: 1572–1579, 2000. xenografted with human CD30ϩ anaplastic large-cell lymphoma. Blood, 85: 2139– 30. Sambrook, J., Fritsch, E., and Maniatis, T. (eds.). Molecular Cloning. A laboratory 2146, 1995. Manual, Second Ed. New York: Cold Spring Harbor Laboratory Press, 1989. 10. Pasqualucci, L., Flenghi, L., Terenzi, A., Bolognesi, A., Stirpe, F., Bigerna, B., and 31. Barth, S., Huhn, M., Wels, W., Diehl, V., and Engert, A. Construction and in vitro Falini, B. Immunotoxin therapy of hematological malignancies. Haematologica, 80: evaluation of RFT5(scFv)-ETA’, a new recombinant single-chain immunotoxin with 546–556, 1995. specific cytotoxicity toward CD25ϩ Hodgkin-derived cell lines. Int. J. Mol. Med., 1: 11. Falini, B., Pileri, S., Pizzolo, G., Durkop, H., Flenghi, L., Stirpe, F., Martelli, M. F., 249–256, 1998. and Stein, H. CD30 (Ki-1) molecule: a new cytokine receptor of the tumor necrosis 32. Denefle, P., Kovarik, S., Guitton, J. D., Cartwright, T., and Mayaux, J. F. Chemical factor receptor superfamily as a tool for diagnosis and immunotherapy. Blood, 85: synthesis of a gene coding for human angiogenin, its expression in Escherichia coli 1–14, 1995. and conversion of the product into its active form. Gene, 56: 61–70, 1987. 12. Schnell, R., Vitetta, E., Schindler, J., Borchmann, P., Barth, S., Ghetie, V., Hell, K., 33. Franke, A. C., Jung, D., and Ellis, T. M. Characterization of the CD30L binding Drillich, S., Diehl, V., and Engert, A. Treatment of refractory Hodgkin’s lymphoma domain on the human CD30 molecule using anti-CD30 antibodies. Hybridoma, 19: patients with an anti-CD25 ricin A-chain immunotoxin. Leukemia, 14: 129–135, 43–48, 2000. 2000. 34. Gruss, H. J., Da Silva, N., Hu, Z. B., Uphoff, C. C., Goodwin, R. G., and Drexler, 13. Engert, A., Diehl, V., Schnell, R., Radszuhn, A., Hatwig, M. T., Drillich, S., Schon, H. G. Expression and regulation of CD30 ligand and CD30 in human leukemia- G., Bohlen, H., Tesch, H., Hansmann, M. L., Barth, S., Schindler, J., Ghetie, V., Uhr, lymphoma cell lines. Leukemia, 8: 2083–2094, 1994. J., and Vitetta, E. A phase-I study of an anti-CD25 ricin A-chain immunotoxin 35. Grossbard, M. L., Gribben, J. G., Freedman, A. S., Lambert, J. M., Kinsella, J., (RFT5-SMPT-dgA) in patients with refractory Hodgkin’s lymphoma. Blood, 89: Rabinowe, S. N., Eliseo, L., Taylor, J. A., Blattler, W. A., Epstein, C. L., and et al. 403–410, 1997. Adjuvant immunotoxin therapy with anti-B4-blocked ricin after autologous bone 14. Klimka, A., Barth, S., Matthey, B., Roovers, R. C., Lemke, H., Hansen, H., Arends, marrow transplantation for patients with B-cell non-Hodgkin’s lymphoma. Blood, 81: J. W., Diehl, V., Hoogenboom, H. R., and Engert, A. An anti-CD30 single-chain Fv 2263–2271, 1993. selected by phage display and fused to Pseudomonas exotoxin A (Ki-4(scFv)-ETA’) 36. Vitetta, E. S., Thorpe, P. E., and Uhr, J. W. Immunotoxins: magic bullets or is a potent immunotoxin against a Hodgkin-derived cell line. Br. J. Cancer, 80: misguided missiles? Immunol. Today, 14: 252–259, 1993. 1214–1222, 1999. 37. Kreitman, R. J., Wilson, W. H., White, J. D., Stetler-Stevenson, M., Jaffe, E. S., Giardina, S., Waldmann, T. A., and Pastan, I. Phase I trial of recombinant immuno- 15. Khazaeli, M. B., Conry, R. M., and Lo Buglio, A. F. Human immune response to toxin anti-Tac(Fv)-PE38 (LMB-2) in patients with hematologic malignancies. J. Clin. monoclonal antibodies. J. Immunother., 15: 42–52, 1994. Oncol., 18: 1622–1636, 2000. 16. Barth, S., Matthey, B., Huhn, M., Diehl, V., and Engert, A. CD30L-ETA’: a new 38. Kreitman, R. J., Wilson, W. H., Robbins, D., Margulies, I., Stetler-Stevenson, M., recombinant immunotoxin based on the CD30 ligand for possible use against human Waldmann, T. A., and Pastan, I. Responses in refractory hairy cell leukemia to a lymphoma. Cytokines Cell. Mol. Ther., 5: 69–78, 1999. recombinant immunotoxin. Blood, 94: 3340–3348, 1999. 17. Shapiro, R., Riordan, J. F., and Vallee, B. L. The characteristic ribonuclease activity 39. Rybak, S. M., and Newton, D. L. Natural and engineered cytotoxic ribonucleases: of angiogenin. Biochem. Insert., 25: 3527–3532, 1986. therapeutic potential. Exp. Cell Res., 253: 325–335, 1999. 18. Saxena, S. K., Rybak, S. M., Davey, R. T. J., Youle, R. J., and Ackerman, E. J. 40. Smith, C. A., Gruss, H. J., Davis, T., Anderson, D., Farrah, T., Baker, E., Sutherland, Angiogenin is a cytotoxic, tRNA-specific ribonuclease in the RNase A superfamily. G. R., Brannan, C. I., Copeland, N. G., and Jenkins, N. A. CD30 antigen, a marker J. Biol. Chem., 267: 21982–21986, 1992. for Hodgkin’s lymphoma, is a receptor whose ligand defines an emerging family of 19. Yoon, J. M., Han, S. H., Kown, O. B., Kim, S. H., Park, M. H., and Kim, B. K. cytokines with homology to TNF. Cell, 73: 1349–1360, 1993. Cloning and cytotoxicity of fusion proteins of EGF and angiogenin. Life Sci., 64: 41. Gruss, H. J., and Dower, S. K. Tumor necrosis factor ligand superfamily: involvement 1435–1445, 1999. in the pathology of malignant lymphomas. Blood, 85: 3378–3404, 1995. 20. Newton, D., Xue, Y., Olson, K., Fett, J., and Rybak, S. Angiogenin single-chain 42. Barth, S., Huhn, M., Matthey, B., Klimka, A., Tawadros, S., Schnell, R., Diehl, V., immunofusions: influence of peptide linkers and spacers between fusion protein and Engert, A. Ki-4(scFv)-ETA’, a new recombinant anti-CD30 immunotoxin with domains. Biochemistry, 35: 545–553, 1996. highly specific cytotoxic activity against disseminated Hodgkin tumors in SCID mice. 21. Sims, J. E., March, C. J., Cosman, D., Widmer, M. B., MacDonald, H. R., McMahan, Blood, 95: 3909–3914, 2000. C. J., Grubin, C. E., Wignall, J. M., Jackson, J. L., and Call, S. M. cDNA expression 43. Jost, L. M., Kirkwood, J. M., and Whiteside, T. L. Improved short- and long-term cloning of the IL-1 receptor, a member of the immunoglobulin superfamily. Science XTT-based colorimetric cellular cytotoxicity assay for melanoma and other tumor (Wash. DC), 241: 585–589, 1988. cells. J. Immunol. Methods, 147: 153–165, 1992.

8742

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2001 American Association for Cancer Research. Human Angiogenin Fused to Human CD30 Ligand (Ang-CD30L) Exhibits Specific Cytotoxicity against CD30-positive Lymphoma

Michael Huhn, Stephanie Sasse, Mehmet K. Tur, et al.

Cancer Res 2001;61:8737-8742.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/61/24/8737

Cited articles This article cites 37 articles, 18 of which you can access for free at: http://cancerres.aacrjournals.org/content/61/24/8737.full#ref-list-1

Citing articles This article has been cited by 9 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/61/24/8737.full#related-urls

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/61/24/8737. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.