Antitransferrin receptor antibody-RNase fusion protein expressed in the mammary gland of transgenic mice1

Journal of Immunological Methods 231 Ž1999. 159–167 www.elsevier.nlrlocaterjim

Antitransferrin receptor antibody-RNase fusion protein expressed in the mammary gland of transgenic mice 1 Dianne L. Newton a , Daniel Pollock b, Paul DiTullio b,2 , Yann Echelard b, Merri Harvey b, Brian Wilburn b, Jennifer Williams b, Hennie R. Hoogenboom c , Jef C.M. Raus d , Harry M. Meade b, Susanna M. Rybak e,) a

Intramural Research Support Program, SAIC Frederick, National Cancer Institute-Frederick Cancer Research and DeÕelopment Center, Frederick, MD 21702, USA b Genzyme Transgenic, Framingham, MA 01701, USA c CESAME, Department of Pathology, Maastricht UniÕersity, 6202 AZ Maastricht, Netherlands d D.L. Willems-Institute, B-3590 Diepenbeek, Belgium e Pharmacology Experimental Therapeutics Section, Laboratory of Drug DiscoÕery Research and DeÕelopment, DeÕelopmental Therapeutics Program, DiÕision of Cancer Treatment and Diagnosis, National Cancer Institute-Frederick Cancer Research and DeÕelopment Center, Building 567, Room 152, Frederick, MD 21702-1201, USA

Abstract Antibodies fused to human enzymes offer an alternative to specifically targeting tumors with antibodies linked to plant or bacterial toxins. Since large amounts of these reagents can be administered without eliciting non-specific toxicities, efficient methods of production are needed. The goal of this work was to express a complex immunoenzyme fusion protein Žimmunotoxin. in the mammary gland of transgenic mice. A chimeric mouserhuman antibody directed against the human transferrin receptor ŽE6. was fused at its CH2 domain to the gene for a human angiogenic ribonuclease, angiogenin ŽAng.. It was expressed in the mammary gland of mice and secreted into mouse milk. Expression levels in milk were approximately 0.8 grl. The chimeric protein retained antibody binding activity and protein synthesis inhibitory activity equivalent to that of free Ang. It was specifically cytotoxic to human tumor cells in vitro. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Transgenic mice; Ribonuclease; Immunotoxin; Antitransferrin receptor antibody

AbbreÕiations: Ang, human angiogenin; E6, anti-transferrin receptor IgG monoclonal antibody; RNases, ribonucleases; H chain, heavy chain; L chain, light chain; CH2Ang, angiogenin fused to the CH2 domain of the E6 heavy chain; IC 50 , the concentration of fusion protein which inhibits protein synthesis by 50% ) Corresponding author. Tel.: q1-301-846-5215; fax: q1-301-846-7022; e-mail: [email protected] 1 The publisher or recipient acknowledges the right of the U.S. Government to retain a nonexclusive, royalty-free license in and to any copyright covering the article. 2 Present affiliation: Midas Biologicals, N. Grafton, MA 01536, USA. 0022-1759r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 2 2 - 1 7 5 9 Ž 9 9 . 0 0 1 5 4 - 4

160

D.L. Newton et al.r Journal of Immunological Methods 231 (1999) 159–167

1. Introduction

2. Materials and methods

Members of the pancreatic RNase A superfamily exhibit a variety of physiological actions, e.g., antitumoral, antibacterial, antiviral, antihelminthic and angiogenic activities Žreviewed in Youle et al., 1993., escalating current scientific interest in these proteins ŽSchein, 1997.. To overcome problems of internalization and routing, as well as to impart tumor cell specificity, noncytotoxic RNases have been linked to tumor associated cell binding ligands to selectively kill tumor cells ŽRybak et al., 1991; Newton et al., 1992, 1994, 1996; Rybak et al., 1992; Jinno et al., 1996; Zewe et al., 1997; Deonarain and Epenetos, 1998.. Because these proteins are human and are not cytotoxic, it is thought that they will elicit fewer toxic side effects and immunogenicity than the plant ŽThrush et al., 1996. and bacterially derived toxins ŽPastan, 1997. currently used in the construction of immunotoxins. Addressing those problems would allow the administration of large as well as repeated doses of the RNase-based immunotoxins ŽRybak et al., 1995.. Moreover, recombinant antibody-enzyme chimeras are also being investigated for thrombolytic therapies Žreviewed in Haber, 1994; Collen, 1997. and partnered with suitable prodrugs for indirect tumor targeting ŽBosslet et al., 1992; Goshorn et al., 1993; Rodrigues et al., 1995; Siemers et al., 1997., demonstrating the breadth of their potential importance. With the use of genetic control elements from different milk protein genes, human proteins have been successfully expressed in the milk of transgenic animals Žreviewed in Maga and Murray, 1995; Colman, 1996; Clark, 1997.. Although antibodies can be correctly assembled and expressed in mouse milk ŽLimonta et al., 1995; Castilla et al., 1998; Sola et al., 1998., we report the first expression, purification and characterization of a humanized immunotoxin in the murine mammary gland. A mouserhuman chimeric antibody against the human transferrin receptor ŽHoogenboom et al., 1990, 1991. was fused to DNA for the human angiogenic RNase angiogenin ŽAng. ŽFett et al., 1985; Strydom et al., 1985.. The expression and biological activities are compared to the same construct expressed in myeloma cells ŽRybak et al., 1992..

2.1. Transgenic mice Transgenic mice were generated following standard published procedures ŽDiTullio et al., 1992; Roberts et al., 1992; Gutierrez et al., 1996.. Founder mice were bred to produce lactating females and the milk collected and diluted with an equal volume of phosphate buffered saline as previously described. Milk was stored at y708C. 2.2. Fractionation of milk Milk containing E6 IgG antibody was applied to a Protein A Sepharose column and eluted with 0.1 M glycine, pH 3.0, into tubes containing 1 M Tris base to adjust pH to neutrality. Milk containing the fusion protein ŽCH2Ang. was made 0.2 M EDTA and incubated on ice for 20 min before centrifugation for 10 min at 48C in an eppifuge. The skim milk was removed carefully from the fat layer and centrifuged again before purification by size exclusion high performance liquid chromatography on a TSK 3000 column ŽToso Haas, Montgomeryville, PA. equilibrated and eluted with 0.1 M phosphate buffer, pH 7.4. The flow rate was 0.5 mlrmin and 1 min fractions were collected. The majority of material reacting with an antibody against Ang eluted in the void volume of the column. This material was pooled and arginine powder was added to a final concentration of 1 M. After an overnight incubation at 48C, the sample was re-chromatographed on the TSK 3000 column as described above. CH2Ang containing milk required a second treatment with 1 M arginine and re-chromatography on the sizing column. 2.3. Protein determination Protein was determined using the following extinction coefficients: E6 IgG antibody, E1%r280 nm s 14.0; CH2Ang, E1%r280 nm s 10.0. 2.4. Protein synthesis assay Cells were plated at 2500 cells per well in 96-well microtiter plates in Dulbecco’s minimum essential

D.L. Newton et al.r Journal of Immunological Methods 231 (1999) 159–167

161

medium supplemented with 10% fetal bovine serum. Additions were made in a total volume of 10 ml, and the plates were incubated at 378C for 3 days before 0.1 mCi of w14 Cx-leucine was added for 2–4 h. Cells were harvested onto glass fiber filters using a PHD cell harvester, washed with water, dried with ethanol and counted. The results are expressed as percent of w14 Cx-leucine incorporation in mock-treated wells. 3. Results 3.1. Expression of an antihuman transferrin receptor antibody and antibody-Ang fusion protein in the milk of transgenic mice The DNA constructs used to produce the transgenic mice are illustrated in Scheme 1 and Fig. 1A. DNA encoding the entire heavy chain of the E6 antibody, a chimeric antibody against the human transferrin receptor ŽHoogenboom et al., 1990., was fused between exons 2 and 7 of a modified goat b-casein gene ŽFig. 1A, I. that is expressed at high levels in the milk of lactating transgenic mice ŽRoberts et al., 1992.. A second transgene encoding an antibody-enzyme fusion was prepared by linking the gene for the human RNase, Ang to the CH2 domain of the antibody ŽScheme 1 and Fig. 1A, II.. Those genes as well as the gene encoding the light chain of the same antibody ŽFig. 1A, III. were all cloned separately, and the appropriate pairs Žheavy ŽH. and light ŽL. chains; CH2Ang and L chain. were purified free of procaryotic DNA and co-injected into mouse embryos that were reimplanted using standard methods ŽRoberts et al., 1992.. Transgenic mice were identified by PCR and Southern blot analysis of DNA obtained from tails of the resulting progeny. Founder mice were bred to produce lactating transgenic females. Milk was collected, diluted with PBS and analyzed for the presence of the antibody chains and Ang. Polyclonal antibodies raised against human Ang only reacted with a band of the expected Mr Ž43 kDa; antibody H chain, 29 kDa; Ang, 14 kDa. in the fusion protein ŽFig. 1B, left panel.. However, anti-IgG antisera strongly reacted with both the H and L chains of the chimeric E6 antibody ŽFig. 1B, right panel.. Whereas the L chain of the antibody fusion protein was clearly observed with anti-IgG antisera, the truncated H chain of CH2Ang

Scheme 1. The chimeric antitransferrin receptor antibody used in the studies described herein was originally fused to human tumor necrosis factor ŽHoogenboom et al., 1991. and then to a human ribonuclease, Ang ŽRybak et al., 1992.. The Ang gene was fused X behind the first three amino acid residues of the 5 region of the CH2 domain of the antibody, thus leaving the hinge region unaffected and dimerization of the heavy chain possible. The goal was to create humanized immunotoxin-like proteins that might elicit less immunogenic side effects when administered to patients. The in vitro mammalian cell expression systems yielded very little material for functional studies, especially when the antibody was fused to the human RNase, Ang. Ang is a member of the RNase A superfamily. All the members of this superfamily are small Ž12–14 kDa., basic ribonucleolytic enzymes found in the pancreas as well as other organs, fluids and tissues of mammals and amphibians. Though these RNases can cleave RNA, physiological actions, e.g., eliciting angiogenesis, host defense actions and antiviral effects have been described for various RNase members. Because RNases might be part of a natural defense system, they have been used to create chemical conjugates and recombinant fusion proteins with a variety of antibodies. Since those studies indicate that RNase-based therapeutics may have potential for the treatment of cancer and AIDS, the original RNase work with the chimeric antibody against the human transferrin receptor was re-explored using newly developed technology for the production of complex proteins in the milk of transgenic animals. The molecular details of the genetic constructs used in these studies are shown above. The Roman numerals correspond to those shown in Fig. 1 panel A and expand on the DNAs cloned between exons 2 and 7 of the goat b-casein gene.

was barely detectable. A truncated CH2TNF fusion protein was recognized by anti-IgG antisera ŽHoogenboom et al., 1991., suggesting that the fusion of Ang to the CH2 domain hindered binding of the antisera to the H chain.

162

D.L. Newton et al.r Journal of Immunological Methods 231 (1999) 159–167

The chimeric IgG antibody was purified by chromatography on Protein A Sepharose. As shown in Fig. 1C, lanes 1 and 2, Western analysis of the final purified product by gel electrophoresis under reduc-

ing conditions showed the presence of L Ž28 kDa. and H chain proteins Žapproximately 55 kDa.. Western analysis under non-reducing conditions ŽFig. 1C, lane 3. demonstrated that the transgenic antibody

D.L. Newton et al.r Journal of Immunological Methods 231 (1999) 159–167

existed as a mixture of IgG and FŽabX . 2 forms. A small amount of free H chain Ž55 kDa. was also seen. Milk containing the CH2Ang fusion protein was similarly collected and diluted with PBS. Protein A Sepharose failed to bind the Ang fusion protein. Analogous results were obtained when the same CH2 antibody fragment previously was fused to TNF and it was postulated that this was due to the deletion of the Protein A binding site believed to be near the CH2–CH3 junction ŽHoogenboom et al., 1991.. The nature of the transgenic antibody-Ang fusion protein was determined by Western blotting. After reduction of the interchain disulfide bonds, the H chain Ang fusion Ž43, kDa. and L chain Ž28 kDa. were dissociated ŽFig. 1C, lane 4.. Western analysis with an anti-IgG antibody under non-reducing conditions ŽFig. 1C, lane 5. demonstrated that the transgenic antibody-enzyme fusion protein existed as a mixture of FŽabX . 2 and FabX forms Ž142 and 71 kDa, respectively.. Identical results were obtained when the analysis was performed with anti-Ang antisera Žnot shown.. Taken together the latter results demonstrate that the L chain is associated with the H chain-Ang fusion. 3.2. Biological characterization of antibody-Ang fusion protein Ang is a potent inhibitor of the translational capacity of the rabbit reticulocyte lysate by a mechanism that depends upon its ribonucleolytic activity ŽSt. Clair et al., 1987.. Fig. 2 shows that the addition of Ang or CH2Ang to the lysate caused the inhibition of protein synthesis as measured by the incorporation of w35 Sxmethionine into acid-precipitable protein. The IC 50 s Ž40 nM. of unfused Ang or CH2Ang were indistinguishable in this assay, indicating that the conformation of the active site residues was not

163

Fig. 2. Translation of mRNA in the presence of Ang or CH2Ang. Ang or the fusion protein was added to a lysate mixture containing BMV mRNA and w35 Sxmethionine. Protein synthesis was determined by measuring the incorporation of label into newly synthesized proteins as described in ŽNewton et al., 1996.. Data from 2–3 experiments were pooled and plotted "S.E.M. The results are expressed as a percentage of a mock treated control reaction. IC 50 is the concentration of Ang or the Ang fusion protein required to cause a 50% inhibition of protein synthesis and was determined from the dose–response curves. Solid circles, Ang; open circles, CH2Ang.

affected by fusing Ang in this orientation ŽNH 2terminus. to the CH2 antibody domain. The antibody portion of the fusion protein was characterized by competition binding experiments ŽTable 1.. Binding of milk-derived E6 antibody ŽIgG. to the human transferrin receptor was tested and compared to that of the same antibody originally purified from hybridoma cells ŽHeyligen et al., 1985.. The ability of both antibodies to displace the w125 Ixlabeled parental antibody was identical Ž50%

Fig. 1. ŽA. Structure of the transgenic expression vectors for E6 and CH2Ang. The following DNAs were fused between exons 2 and 7 of a modified goat b-casein gene ŽDiTullio et al., 1992. for expression in the mammary gland of mice: the heavy chain of the anti-human X transferrin receptor monoclonal antibody, E6 ŽI.; the same heavy chain fused at the CH2 domain to the 5 end of the gene encoding Ang as previously described ŽRybak et al., 1992. ŽII.; the light chain of the E6 antibody ŽIII.. Open boxes, H chain; crossed hatched boxes, L chain; striped boxes, Ang. ŽB. Western analysis using anti-Ang or anti-IgG antibodies under reducing conditions of milk collected from lactating females producing either the E6 IgG antibody or CH2Ang fusion protein. 15 ml of milk diluted with an equal volume of PBS was applied to the gel. ŽC. Western analysis of purified E6 antibody or CH2Ang under reducing or non-reducing conditions. The blots were analyzed with the indicated antibodies. 0.3 mg E6, lanes 1 and 2; 4 mg E6, lane 3; 0.7 and 0.2 mg CH2Ang lanes 4 and 5, respectively.

D.L. Newton et al.r Journal of Immunological Methods 231 (1999) 159–167

164

Table 1 Binding of E6 and Ang fusion proteins to human transferrin receptor a Construct

Source

Binding b EC 50 ŽnM.

Fold difference c

E6 E6 CH2Ang

hybridoma milk milk

0.8 0.8 140

1 1 175

a Competition of E6 or Ang fusion protein with w125 IxE6 IgG for binding to the transferrin receptor. The binding analyses were conducted on K562 cells as described in Newton et al. Ž1996.. Data from 2–3 experiments were pooled. The S.E.M. was 10% or less. b EC 50 is the concentration of protein necessary to displace 50% of the w125 IxE6 IgG monoclonal antibody from the human transferrin receptor. c Fold differences in binding from E6 assessed at 50% displacement.

displacement by either antibody was 0.8 nM.. The CH2Ang fusion protein was 175 fold less active than the E6 intact antibody Ž140 nM CH2Ang vs. 0.8 nM E6..

Fig. 3. In vitro toxicity of CH2Ang to SF539 and MDA-MB231md r1 cells as assessed by protein synthesis inhibition. Cytotoxicity assays were performed by measuring the incorporation of w14 Cxleucine into cell proteins as described in Section 2. The assays were conducted in the presence of serum and changed to leucine- and serum-free medium prior to pulsing with w14 Cxleucine. IC 50 is the concentration of the Ang fusion proteins required to cause a 50% inhibition of protein synthesis after 3 days and was determined directly from the dose response curves. The S.E.M. is shown when it is larger than the symbol. Solid symbols, SF539 human glioma cells; open symbols, MDA-MB-231md r1 human breast cancer cells.

Table 2 Cytotoxicity of CH2Ang a Cell line

CH2Ang, IC 50 ŽnM.

SF539 HS578T MDA-MB-231md r1 MALME ACHN

15 70 45 40 30

a Cytotoxicity assays were performed by measuring the incorporation of w14 Cxleucine proteins as described in Section 2. The IC 50 values, the concentration of CH2Ang that caused a 50% inhibition of protein synthesis after 3 days incubation, were determined from semilogarithmic dose–response curves. The S.E.M. was 10% or less.

The cytotoxic effects of the Ang fusion protein on human tumor cells was assessed by measuring w14 C xleucine incorporation into newly synthesized proteins. Typical dose–response curves are depicted in Fig. 3. CH2Ang inhibited the protein synthesis of SF539 human glioma cells and MDA-MB-231mdr1 breast cancer cells with IC 50 s of 15 and 45 nM, respectively. Cytotoxicity on other human tumor cell lines is compared in Table 2. The IC 50 s ranged from 15 to 70 nM. Cytotoxicity was specific to the fusion protein since no activity was observed on an antigen negative cell line Žmouse NIH3T3 cells, data not shown. and a five fold molar excess of the unfused chimeric antibody reversed cytotoxicity by approximately 50%. Whereas CH2Ang inhibited protein synthesis to 99% of mock treated cells, protein synthesis was decreased to 45% of mock treated cells in the presence of a five fold molar excess of antibody. Since neither the unfused antibody ŽRybak et al., 1992. or free Ang ŽNewton et al., 1996. are cytotoxic, the two domains in the fusion proteins must be covalently joined to elicit cytotoxicity.

4. Discussion Ang was isolated from tumor cell conditioned medium by following angiogenic activity in the chicken embryo chorioallantoic membrane and rabbit corneal assays ŽFett et al., 1985.. Its homology to ribonuclease and distinctive nucleolytic activity ŽShapiro et al., 1986. coupled to its angiogenic activity yield unique biological properties that may pro-

D.L. Newton et al.r Journal of Immunological Methods 231 (1999) 159–167

mote enhanced tumor cell killing when Ang is targeted to tumor cells with cell specific targeting agents. Angiogenic activity is maintained when Ang is expressed as a fusion protein ŽNewton et al., 1996.. Ang also binds a cell surface proteoglycan on human colon carcinoma cells ŽSoncin et al., 1994.. Accordingly, localization to tumor sites by the antibody could be increased by the tumor cell binding properties of Ang while increased angiogenesis could conceivably aid tumor penetration by increasing tumor vascularization ŽNewton et al., 1996.. Moreover, antagonists of Ang prevent tumor growth ŽOlson et al., 1995; Piccoli et al., 1998.. Thus, Ang activities are pleiotropic; their manifestation is governed by the cellular milieu to which Ang is exposed, e.g., targeting the cytosolic protein synthesis machinery causes cytotoxicity ŽSt. Clair et al., 1987; Rybak et al., 1991. while endocytosis and translocation of Ang to the nucleus in endothelial cells has been reported to elicit angiogenesis ŽMoroianu and Riordan, 1994.. These biological properties of Ang afford unique opportunities to design both cytostatic Žantiangiogenic. and cytotoxic Žantitumor cell. therapeutic strategies by antagonizing or specifically targeting this protein, respectively. The realization of human enzyme-based multi-domain targeted therapeutics for cancer ŽRybak et al., 1991, 1992; Newton et al., 1992, 1994, 1996; Jinno et al., 1996; Zewe et al., 1997; Deonarain and Epenetos, 1998. and cardiovascular disease ŽHaber, 1994; Collen, 1997. depends on developing expression systems capable of producing these reagents for preclinical characterization and eventual clinical use. Expression of a two chain antibody-Ang fusion protein in the milk of transgenic mice was accomplished and presented in this study. It was not obvious that Ang could be successfully expressed as a fusion protein in transgenic mice because a similar fusion protein was expressed only at very low levels from cultured myeloma cells presumably due to retrograde transport during secretion leading to the selection of low producers ŽRybak et al., 1992.. Remarkably, in the natural environment of the mammary gland the efficiency of expression was increased 160,000 times over the cell culture system Ž0.8 grl vs. 5 mgrl in milk and myeloma cells, respectively.. Thus, it was possible to purify sufficient amounts of the Ang fusion protein for biological characterization. One of

165

the consequences of this work is that the importance of the orientation of Ang in a fusion protein is demonstrated for the first time. In single chain antibody Ang fusion proteins Ang was fused at the C-terminus to the N-terminus of the antibody ŽNewton et al., 1996.. Subsequently, it became known that the last three amino acid residues of the C-terminal region of Ang constitute an active center subsite ŽRusso et al., 1996.. Whereas Ang in the CH2 fusion protein and free Ang were equipotent in the rabbit reticulocyte lysate assay, Ang in a single chain fusion protein was two fold less effective than free Ang to inhibit protein synthesis in the lysate assay ŽNewton et al., 1996.. This is the first demonstration, in general, that antibody-enzyme fusion proteins can be expressed at high levels in the mammary gland. In particular, the demonstration that antibody-Ang fusions can be expressed in the mammary gland has implications for the development of transgenic mouse models for breast cancer. Promoters from other milk specific genes have been used to cause the expression of transgenes during lactation, initiating the onset of neoplasias ŽAmundadottir et al., 1996.. Since the results of the present study show that a milk specific promoter can induce expression of an active immunotoxin, double transgenic strains could be developed to test whether the expression of an Ang fusion protein targeted against the engineered neoplasia could prevent or alter the progression of the disease. These results are especially relevant to Ang since murine counterparts are available ŽBond et al., 1993.. In summary, these results demonstrate for the first time that complex heterologous fusion proteins can be expressed in the mammary gland of mice in larger amounts and with superior biological properties than mammalian cell culture ŽRybak et al., 1992. and E. coli expression systems ŽNewton et al., 1996.. The results impact both the possibility of producing these fusion proteins as therapeutics as well as the possibility of creating new animal models for breast cancer.

Acknowledgements The technical support of Dale Ruby is gratefully acknowledged. We very much appreciate our excel-

166

D.L. Newton et al.r Journal of Immunological Methods 231 (1999) 159–167

lent administrative help and thank Ms. Beverly A. Bales, Robin L. Reese, and Jamie M. Tammariello. We are grateful for the interest and support of Dr. Edward A. Sausville. This project has been funded in whole or in part with Federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. NO1-CO-56000. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.

References Amundadottir, L.T., Merlino, G., Dickson, R.B., 1996. Transgenic mouse models of breast cancer. Breast Can. Res. Treat. 39, 119. Bond, M.D., Strydom, D.J., Vallee, B.L., 1993. Characterization and sequencing of rabbit, pig and mouse angiogenins: discernment of functionally important residues and regions. Biochim. Biophys. Acta 1162, 177. Bosslet, K., Czech, J., Lorenz, P., Sedlacek, H.H., Schuermann, M., Seemann, G., 1992. Molecular and functional characterization of a fusion protein suited for tumor specific prodrug activation. Br. J. Cancer 65, 234. Castilla, J., Pintado, B., Sola, I., Sanchez-Morgado, J.M., Enjuanes, L., 1998. Engineering passive immunity in transgenic mice secreting virus-neutralizing antibodies in milk. Nat. Biotechnol. 16, 349. Clark, A., 1997. Gene expression in the mammary glands of transgenic animals. Biochem. Soc. Symp. 63, 133. Collen, D., 1997. Thrombolytic therapy. Thromb. Haemostasis 78, 742. Colman, A., 1996. Production of proteins in the milk of transgenic livestock: problems, solutions and successes. Am. J. Clin. Nutr. 63, 639S. Deonarain, M.P., Epenetos, A.A., 1998. Design, characterization and antitumor cytotoxicity of a panel of recombinant, mammalian ribonuclease-based immunotoxins. Br. J. Cancer 77, 537. DiTullio, P., Cheng, S.H., Marshall, J., Gregory, R.J., Ebert, K.M., Meade, H.M., Smith, A.E., 1992. Production of cystic fibrosis transmembrane conductance regulator in the milk of transgenic mice. BiorTechnology 10, 74. Fett, J.W., Strydom, D.J., Lobb, R.R., Alderman, E.M., Bethune, J.L., Riordan, J.F., Vallee, B.L., 1985. Isolation and characterization of angiogenin, an angiogenic protein from human carcinoma cells. Biochemistry 24, 5480. Goshorn, S.C., Svensson, H.P., Kerr, D.E., Somerville, J.E., Senter, P.D., Fell, H.P., 1993. Genetic construction, expression

and characterization of a single chain anti-carcinoma antibody fused to b-lactamase. Cancer Res. 53, 2123. Gutierrez, A., Meade, H.M., DiTullio, P., Pollock, D., Harvey, M., Jimenez-Flores, R., Anderson, G., Murray, J., Medrano, J., 1996. Expression of a bovine kappa-CN cDNA in the mammary gland of transgenic mice utilizing a genomic milk protein gene as an expression cassette. Transgenic Res. 5, 271. Haber, E., 1994. Antibody-plasminogen activator conjugates and recombinant proteins. Methods Mol. Genet. 5, 111. Heyligen, H., Thijs, C., Weber, W., Bosmans, E., Raus, J., 1985. Monoclonal antibodies detecting human T cell activation antigens: development of monoclonal antibodies and expression of antigens on activated T cells and leukemic cells. Fed. Proc. 44, 787. Hoogenboom, H.R., Raus, J.C.M., Volckaert, G., 1990. Cloning and expression of a chimeric antibody directed against the human transferrin receptor. J. Immunol. 144, 3211. Hoogenboom, H.R., Volckaert, G., Raus, J.C.M., 1991. Construction and expression of antibody-tumor necrosis factor fusion proteins. Mol. Immunol. 28, 1027. Jinno, H., Ueda, M., Ozawa, S., Kikuchi, K., Ikeda, T., Enomoto, K., Kitajima, M., 1996. Epidermal growth factor receptor-dependent cytotoxic effect by an EGF-ribonuclease conjugate on human cancer cell lines: a trial for less immunogenic chimeric toxin. Can. Chemother. Pharmacol. 38, 303. Limonta, J., Pedraza, A., Rodriguez, A., Greyre, F.M., Barral, A.M., Castro, F.O., Lleonart, R., Gracia, C.A., Gavilondo, J.V., DelaFuente, J., 1995. Production of active anti-CD6 mouserhuman chimeric antibodies in the milk of transgenic mice. Immunotechnology 1, 107. Maga, E., Murray, J., 1995. Mammary gland expression of transgenes and the potential for altering the properties of milk. BiorTechnology 13, 1452. Moroianu, J., Riordan, J.F., 1994. Nuclear translocation of angiogenin in proliferating endothelial cells is essential to its angiogenic activity. Proc. Natl. Acad. Sci. U. S. A. 91, 1677. Newton, D.L., Ilercil, O., Laske, D.W., Oldfield, E., Rybak, S.M., Youle, R.J., 1992. Cytotoxic ribonuclease chimeras: targeted tumoricidal activity in vitro and in vivo. J. Biol. Chem. 267, 19572. Newton, D.L., Nicholls, P.J., Rybak, S.M., Youle, R.J., 1994. Expression and characterization of recombinant human eosinophil-derived neurotoxin and eosinophil-derived neurotoxin-anti-transferrin receptor sFv. J. Biol. Chem. 269, 26739. Newton, D.L., Xue, Y., Olson, K.A., Fett, J.W., Rybak, S.M., 1996. Angiogenin single-chain immunofusions: influence of peptide linkers and spacers between fusion protein domains. Biochemistry 35, 545. Olson, K.A., Fett, J.W., French, T.C., Key, M.E., Vallee, B.L., 1995. Angiogenin antagonists prevent tumor growth in vivo. Proc. Natl. Acad. Sci. U. S. A. 92, 442. Pastan, I., 1997. Targeted therapy of cancer with recombinant immunotoxins. Biochim. Biophys. Acta 1333, C1. Piccoli, R., Olson, K.A., Vallee, B.L., Fett, J.W., 1998. Chimeric anti-angiogenin antibody cAb 26-2F inhibits the formation of human breast cancer xenografts in athymic mice. Proc. Natl. Acad. Sci. U. S. A. 95, 4579.

D.L. Newton et al.r Journal of Immunological Methods 231 (1999) 159–167 Roberts, B., DiTullio, P., Vitale, J., Hehir, K., Gordon, K., 1992. Cloning of the goat b-casein-encoding gene and expression in transgenic mice. Gene 121, 255. Rodrigues, M.L., Presta, L.G., Kotts, C.E., Wirth, C., Mordenti, J., Osaka, G., Wong, W.L.T., Nuijens, A., Blackburn, B., Carter, P., 1995. Development of a humanized disulfide-stabilized anti-p185 HER2 Fv-b-Lactamase fusion protein for activation of a cephalosporin doxorubicin prodrug. Cancer Res. 55, 63. Russo, N., Nobile, V., DiDonato, A., Riordan, J.F., Valee, B.L., 1996. The C-terminal region of human angiogenin has a dual role in enzymatic activity. Proc. Natl. Acad. Sci. U. S. A. 93, 3243. Rybak, S.M., Saxena, S.K., Ackerman, E.J., Youle, R.J., 1991. Cytotoxic potential of ribonuclease and ribonuclease hybrid proteins. J. Biol. Chem. 266, 21202. Rybak, S.M., Hoogenboom, H.R., Meade, H.M., Raus, J.C., Schwartz, D., Youle, R.J., 1992. Humanization of immunotoxins. Proc. Natl. Acad. Sci. U. S. A. 89, 3165. Rybak, S.M., Newton, D.L., Xue, Y., 1995. RNase and RNase immunofusions for cancer therapy. Tumor Targeting 1, 141. Schein, C.H., 1997. From housekeeper to microsurgeon: the diagnostic and therapeutic potential of ribonucleases. Nat. Biotechnol. 15, 529. Shapiro, R., Riordan, J.F., Vallee, B.L., 1986. Characteristic ribonucleolytic activity of human angiogenin. Biochemistry 25, 3527. Siemers, N.O., Kerr, D.E., Yarnold, S., Stebbins, M.R., Vrudhula,

167

V.M., Hellstrom, I., Hellstrom, K.E., Senter, P.D., 1997. Construction, expression and activities of L49-sFv-b-Lactamase, a single-chain antibody fusion protein for anticancer prodrug activation. Bioconjugate Chem. 8, 510. Sola, I., Castilla, J., Pintado, B., Sanchez-Morgado, J.M., Whitelaw, B.A., Clark, A.J., Enjuanes, L., 1998. Transgenic mice secreting coronavirus neutralizing antibodies into the milk. J. Virol. 72, 3762. Soncin, F., Shapiro, R., Fett, J.W., 1994. A cell-surface proteoglycan mediates human adenocarcinoma HT-29 cell adhesion to human angiogenin. J. Biol. Chem. 269, 8999. St. Clair, D.K., Rybak, S.M., Riordan, J.F., Vallee, B.L., 1987. Angiogenin abolishes cell-free protein synthesis by specific ribonucleolytic inactivation of ribosomes. Proc. Natl. Acad. Sci. U. S. A. 84, 8330. Strydom, D.J., Fett, J.W., Lobb, R.R., Alderman, E.M., Bethune, J.L., Riordan, J.F., Vallee, B.L., 1985. Amino acid sequence of human tumor derived angiogenin. Biochemistry 24, 5486. Thrush, G.R., Lark, L.R., Clinchy, B.C., Vitetta, E.S., 1996. Immunotoxins: an update. Annu. Rev. Immunol. 14, 49. Youle, R.J., Newton, D.L., Wu, Y.N., Gadina, M., Rybak, S.M., 1993. Cytotoxic ribonucleases and chimeras in cancer therapy. Crit. Rev. Ther. Drug Carrier Syst. 10, 1. Zewe, M., Rybak, S.M., Dubel, S., Coy, J.F., Welschof, M., Newton, D.L., Little, M., 1997. Cloning and cytotoxicity of a human pancreatic RNase immunofusion. Immunotechnology 3, 127.