Mutants of Immunotoxin Anti-Tac(dsFv) - American Chemical Society

Mutants of Immunotoxin Anti-Tac(dsFv)-PE38 with Variable. Number of Lysine Residues as Candidates for Site-Specific Chemical. Modification. 1. Propert...
0 downloads 0 Views 630KB Size
480

Bioconjugate Chem. 2003, 14, 480−487

Mutants of Immunotoxin Anti-Tac(dsFv)-PE38 with Variable Number of Lysine Residues as Candidates for Site-Specific Chemical Modification. 1. Properties of Mutant Molecules Masanori Onda, James J. Vincent, Byungkook Lee, and Ira Pastan* Laboratory of Molecular Biology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, 37 Convent Drive, Room 5106, Bethesda, Maryland 20892-4264 Received August 8, 2002; Revised Manuscript Received December 15, 2002

Chemical modification of proteins with substances such as poly(ethylene glycol) can add useful properties to proteins. Currently PEGylation is done in a random manner utilizing amino residues dispersed throughout a protein. For proteins such as immunotoxins, which have several different functional domains, random modification leads to inactivation. To determine if we could produce an immunotoxin with a diminished number of lysine residues so that chemical modification could be restricted to certain regions of the protein, we chose the recombinant immunotoxin anti-Tac(dsFv)PE38 that has 13 lysine residues in the Fv portion and 3 in the toxin. We prepared a series of mutants with 0-12 lysines in the Fv and 0 or 3 in the toxin. Almost all of these molecules retain full biological activity. Our data indicate that replacement of lysine residues can be achieve without loss of biological potency. These molecules are a useful starting point to carry out site-specific PEGylation experiments.

INTRODUCTION

Recombinant immunotoxins are chimeric proteins in which a truncated toxin is fused to the Fv domain of an antibody. We have produced many different recombinant immunotoxins in which the Fv portion of an antibody against a tumor related antigen is fused to a 38 kDa mutant form of Pseudomonas exotoxin A (PE)1 that has a deletion of its cell binding domain (Bera et al., 1998; Brinkmann et al., 1991; Chowdhury et al., 1998; Mansfield et al., 1997; Onda et al., 2001b; Pastan, 1997; Reiter et al., 1994). Five of these immunotoxins, anti-Tac(Fv)PE38 (LMB2), B3(Fv)-PE38 (LMB7), B3(dsFv)-PE38 (LMB9), RFB4(dsFv)-PE38 (BL22), and e23(dsFv)-PE38 (erb38), have been evaluated in phase I trials in patients with cancer (Kreitman et al., 2000, 2001; Pai-Scherf et al., 1999). Anti-Tac(Fv)-PE38, also called LMB2, contains the Fv fragment of the anti-Tac MAb that binds to the IL-2 receptor R subunit. Treatment of patients with LMB2 has produced major clinical responses in several different hematologic malignancies (Kreitman et al., 1999, 2000). LMB2 was administered to 35 patients with CD25 positive tumors who had failed standard therapies. One patient had a complete remission, lasting 2 years, and seven others had partial responses of variable durations. The toxic side effects of recombinant immunotoxins are of two types. One type results from specific targeting of normal cells that have the same antigen as the cancer cells. The second type is nonspecific and usually is characterized by damage to liver cells in mice and a fall in serum albumin, weight gain, and evidence of renal or liver toxicity in humans (Kreitman and Pastan, 1995; * To whom correspondence should be addressed. Tel: (301) 496-4797. Fax: (301) 402-1344. E-mail: [email protected]. 1 Abbreviations: ASA, accessible surface area; CDR, complimentarity determining regions; DTE, dithioerythritol; PE, Pseudomonas exotoxin A; PEG, poly(ethylene glycol); TGF, transforming growth factor; TNF, tumor necrosis factor.

Onda et al., 2000). We recently described a new strategy designed to decrease the side effects of LMB2 in which mutations were introduced into the framework region of the Fv to lower its isoelectric point (Onda et al., 1999, 2001a). These mutant immunotoxins are less toxic to mice. We also found that a tumor necrosis factor (TNF) binding protein and/or a cyclooxygenase inhibitor were useful in reducing liver toxicity because TNFR was identified as a major cause of liver toxicity (Onda et al., 2000). Administration of either of these two molecules allowed us to use higher doses of immunotoxin to treat animals with tumors. However, liver damage was still the dose limiting toxicity. Also, this approach is not designed to decrease immunogenicity of the immunotoxin. PEG-conjugated proteins are frequently more effective than their corresponding unmodified parent molecules as therapeutic agents. Many pharmaceutical proteins have been PEGylated and shown to have an improvement in their properties (Chaffee et al., 1992; Katre et al., 1987; Kitamura et al., 1990; Pyatak et al., 1980; Wang et al., 1993). These improved clinical properties include better physical and thermal stability, protection against susceptibility to enzymatic degradation, increased solubility, longer in vivo circulating half-life, decreased clearance, enhanced potency, reduced immunogenicity and antigenicity, and reduced animal toxicity (Bailon and Berthold, 1998). We previously reported that a PEGylated chimeric toxin composed of transforming growth factor (TGF)-R and PE showed an improvement in blood residency time and a decrease in its immunogenicity (Wang et al., 1993). However, PEGylation was accompanied by a significant loss of cytotoxic activity of the TGFR-PE fusion protein. Decreased activity was attributed to a decrease in binding to the TGFR receptor and a consequent decreased ability of the toxin to kill the cell. Every chemical modification or conjugation process involves the formation of a covalent bond. Reactive groups that couple with amine-containing molecules are

10.1021/bc020069r CCC: $25.00 © 2003 American Chemical Society Published on Web 02/27/2003

Immunotoxin Mutants

by far the most common functional groups present on cross-linking or modification reagents. An amine-coupling method is most often used to couple protein or peptide molecules to each other as well as to other macromolecules. The primary coupling chemical reactions for modification of amines is proceeded by acylation or alkylation. Most of these reactions are rapid and occur in high yield to give stable amide or secondary amine bonds (Hermanson, 1996). In most cases, PEGylation of proteins is nonspecific and may target all of the lysine residues in the protein, some of which may be in or near the active site of the protein. To overcome this drawback, we previously performed site-specific PEGylation of an immunotoxin in the peptide connector that attaches the Fv to the toxin using site directed mutagenesis to insert a cysteine residue in the connector (Tsutsumi et al., 2000). The PEGylated molecule retained full in vitro specific cytotoxicity. Stability, plasma half-life, antitumor activity, immunogenicity, and animal toxicity were greatly improved. Thus, thiol chemistry is useful for the attachment of PEG to a cysteine residue at a specific site. Modification of thiol residues has certain difficulties. It is difficult to produce large amounts of recombinant immunotoxins with one free thiol and very difficult if two or more are present because of refolding problems. To overcome these problems, we have chosen to carry out modification of lysine residues by making immunotoxins with a limited number of lysine residues. In the current study, we have used site directed mutagenesis to change the lysine residues of a recombinant immunotoxin to other residues to make site-specific chemical modification possible. A series of molecules containing 0-13 lysines in the Fv and 0 or 3 lysines in the toxin was prepared. Molecular modeling was used to locate exposed amino acids that are remote from the complimentarity determining regions (CDRs) of the Fv and could be mutated to lysine residues. We find that all of these mutant molecules retain specific binding and cytotoxic activities. Our studies indicate that controlling the number and location of lysine residues in a protein can be accomplished successfully and could be a useful approach to allow specific chemical modification using lysine chemistries. EXPERIMENTAL PROCEDURES

Molecular Models and Calculation of Accessible Surface Area (ASA) of Lysine Residues. Molecular models of anti-Tac(dsFv) were created using SWISSMODEL (Peitsch, 1996). These models were used to compute the ASA of lysines. The algorithm of Shrake and Rupley (1973), as implemented in GRASP (Nicholls et al., 1991), was used for all ASA calculations. Mutagenesis of Anti-Tac(dsFv)-PE38. Mutagenesis of M1(dsFv)-PE38 was done by Kunkel’s method with some modifications (Kunkel, 1985). CJ236 cells were transformed with pOND9-1 and pOND9-2 (Onda et al., 1999, 2001a). The transformants were grown in 2× YT medium containing 100 µg/mL ampicillin at 37 °C. At an OD600 of 0.36, the cells were infected with the helper phage M13 at a multiplicity of infection of 5. After incubation at 37°C/110 rpm for 1 h, the culture was maintained at 37°C/300 rpm for another 6 h. The bacterial cells then were precipitated by centrifugation, and the phage from the supernatant was precipitated with poly(ethylene glycol). The single-stranded, uracil-containing DNA from the purified phage was extracted with phenol and chloroform and precipitated with sodium

Bioconjugate Chem., Vol. 14, No. 2, 2003 481

chloride and ethanol. This single-stranded DNA codes for the sense strand of M1(dsFv)-PE38. The following primers were used for mutagenesis: VL 18, 5′-GGCACTGCAGGTTATGGTGACTTGCTCCCCTGGAGATGCAGACAT-3′; VL39, 5′-CCATAGCTGGGGAGAAGTGCCAGGCCTCTGCTGGAACCAGTGCATGTA-3′; VL 45, 5′-GGATGTGGTATAAATCCATAGCTGGGGAGAAGTGCCTGGCTT-3′; VL 77, 5′-GGCAGCATCTTCAGCCTCCATATTCGAAATTGTGAGAGAGTAATCGGT-3′; VL 103-107, 5′-GTTAGCAGCCGAATTCTATTCGAGTTCCAGCTCGGTCCCGCAACCGAACGTGAG-3′; VH 13, 5′-CATCTTCACTGAGGCCCCAGGTTCTGCTAGCTCAGCCCCAGA-3′; VH 19-22-23, 5′-AGTAAAGGTGTAGCCAGAAGCTGCGCAGGACAGCTGCACTGAGGCCCCAGGTTCTGC-3′; VH23, 5′-AAAGGTGTAGCCAGAAGCTGCGCAGGACATCTTCACTGA-3′; VH 38-40, 5′-AATCCATTCCAGACACTGTTCAGGTGTCTGTCTTACCCAGTGCATCCTGTAG-3′; VH 62-64, 5′GGAGGAATCGTCTGCAGTTAACGTGGCCTTGTCTCTGAATCTCTGATTGTATTCAGTATA-3′; VH6264-66-67, 5′-GGAGGAATCGTCTGCAGTTAACGTGAATCTGTCTCTGAATCTCTGATTGTATTCAGTATA-3′; VH 73, 5′-GCTCAGTTGCATGTAGGCAGTACTGGAGGAATCGTCTGCAGTCAA-3′; PE604K, 5′-CAGGTCCTCGCGCGGCGGTTGGCCGGGGCCACCTTTCCCACCCTGGCTGGCGTAGTCCGGCAG-3′. The primers were phosphorylated using polynucleotide kinase and T4 DNA ligase buffer from Boehringer Mannheim (Indianapolis, IN). These phosphorylated primers were used to introduce mutations in the uracil template of pOND9-1 and pOND9-2 using the Bio-Rad (Richmond, CA) Muta-Gene kit. The product at the end of the mutagenesis reaction was used to directly transform DH5R competent cells. Mutations in the clones were confirmed by automated DNA sequencing. Production of Recombinant Immunotoxin. The two components of the recombinant immunotoxins were expressed in Escherichia coli BL21 (λDE3) and accumulated in inclusion bodies as previously described (Onda et al., 1999). Inclusion bodies were solubilized in Guanidine hydrochloride solution, reduced with dithioerythritol (DTE) and refolded by dilution in a refolding buffer containing arginine to prevent aggregation, and oxidized and reduced glutathione to facilitate redox shuffling. Active monomeric protein was purified from the refolding solution by ion exchange and size exclusion chromatography (Onda et al., 1999). Protein concentrations were determined by a Bradford assay (Coomassie Plus; Pierce, Rockford, IL). Cytotoxicity Assays. The specific cytotoxicity of each immunotoxin was assessed by protein synthesis inhibition assays (inhibition of incorporation of tritium-labeled leucine into cellular protein) in 96-well plates as previously described (Onda et al., 1999). The activity of the molecule is defined by the IC50, the toxin concentration that reduced incorporation of radioactivity by 50% compared with cells that were not treated with toxin. Binding Assays. Binding of anti-Tac(dsFv)-PE38 and its derivatives to the human adult T cell leukemia cell line, HUT102, was analyzed in a displacement assay. Human anti-Tac(HAT)-IgG (Queen et al., 1989) was modified by Bolton-Hunter reagent (Amersham Pharmacia Biotech, Piscataway, NJ) and purified by gel filtration on a PD-10 column (Amersham Pharmacia Biotech). HUT102 cells were plated at 2 × 105 cells/well in 96-well plates. Cells were washed with binding buffer (DMEM containing 0.2% BSA). An initial experiment showed that the binding of 125I-HAT-IgG (specific activity 0.97 mCi/mg) to HUT102 cells reached equilibrium

482 Bioconjugate Chem., Vol. 14, No. 2, 2003

Onda et al.

Figure 1. Structural model of anti-Tac(dsFv). The structure was generated by homology modeling. The antigen binding site is located at the top center in this model and is colored green. Lysine residues of anti-Tac(dsFv) are colored red in the ribbon structure and have white labels indicating Kabat numbering. Table 1. Accessible Surface Area (ASA) of Lysine Residues in Modeled Anti-Tac Fv position

ASA (Å2)

VH K13 VH K19 VH K23 VH K38 VH K62 VH K64 VH K66 VH K73 VL K18 VL K39 VL K45 VL K103 VL K107

160.6 127.6 98.7 0.9 69.9 83.5 44.3 125.2 153.9 59.6 112.2 120.7 144.3

by 1.5 h. In the competition studies, various concentrations of unlabeled MAbs or immunotoxins were added to these cells in the presence of a fixed concentration of labeled HAT-IgG (1.2 nM). Bound radioactivity was measured in an automated gamma counter. Statistical Analysis. Values are expressed as mean ( SD. For comparison between the two experimental groups Student’s t test was used. p < 0.05 was considered statistically significant. RESULTS

Identification of Mutation Sites in the Fv. To identify residues that could be mutated without affecting protein stability, we have calculated the frequencies at which various amino acids occur at each position of the Fv using the Kabat database, reasoning that replacement of a lysine with an amino acid that also occurs frequently at that position would not decrease protein stability. M1(dsFv)-PE38 is a mutant of anti-Tac(Fv)-PE38 in which the pI has been lowered from 10.2 to 6.82 by selective mutations of surface residues (Onda et al., 1999). It is less toxic to mice than anti-Tac(Fv)-PE38. Therefore, we decided to use this molecule as a starting point to perform the modifications. M1(dsFv)-PE38 has16 lysine residues (K): 13 in the Fv and 3 in PE38. We replaced exposed lysines in a stepwise manner. The location of lysine residues in selected mutants is shown in Figures 1 and 2. Because residues on the surface of antibodies tend to be hydrophilic (Chothia et al., 1998), we chose to replace the selected residues with hydrophilic residues using the frequency table shown in Figure 3. For example, K103 in VL is an exposed hydrophilic residue (Table 1).

Because a lysine residue is highly conserved at this position, the frequency table did not suggest a good replacement candidate. In this case, we chose glutamic acid (E), which is hydrophilic. This choice also lowered the pI, which is likely to diminish nonspecific toxicity (Onda et al., 1999). This molecule is named M3(dsFv)PE38. For the same reasons, we replaced the lysine at position 107 in VL with glutamic acid, producing mutant M4. We then combined the mutations of M3 and M4 into one molecule, which is called M5 (Figure 2). In a similar fashion, M6, M7, and M8 were made by the mutations VL K18Q (glutamine), VL K45Q, and VL R77N (asparagine), respectively. These mutations also reduced the pI of each molecule. Combining the mutations in M6, M7 and M8 produced M9. K13 was the most exposed residue in VH. M10 was made by replacing K13 in VH with E even though E is not common at this position. The change was made to lower the pI. M11 was produced by replacing K73 in VH with aspartic acid (D). K73 in VH was also replaced with N, which is more common than D at this position, to produce M12. All the mutants produced by this approach had full cytotoxic activity (Table 2). M13 contained the mutations present in M10 and M11. M14 contained the mutations present in M3, M4, M6, M7, and M8. We also combined the mutations in M6, M7, M8, M10, and M11 to make M15. M16 contained the mutations in M3, M4, M6, M7, M8, M10, and M11. In M16, the number of lysine residues in the Fv is reduced to 7, and there is no loss of cytotoxic activity (Onda et al., 2001a). To produce M18, K23 in the VH of M1 was replaced with alanine (A), the next most common amino acid at that position. Because M18 and M12 were fully active, we combined their mutations to produce M19. PE38QQR is a 38 kDa toxin molecule in which lysines 590 and 606 have been replaced with glutamines and lysine 613 with arginine (Debinski and Pastan, 1994). Although PE38QQR does not contain lysine residues, it is still fully active. We combined PE38QQR with M16 Fv, which has seven lysine residues in the Fv, to produce M16(dsFv)-PE38QQR. This molecule retained full cytotoxic activity. Because the pI of M16 Fv is very low (pI ) 4.42), we did not select negatively charged residues to replace K. Instead, we replaced K with Q or R to retain hydrophilicity. K62 and K64 in VH are in the CDR2 of VH. Because these residues could potentially react with antigen, we replaced them with arginines to maintain the positive charge in the M1 background. Both mutants were fully active (Table 2). We then made these mutations in M16(dsFv)-PE38QQR to make M27, which has five lysines and is fully active. To make M28, we replaced K19 with Q, K23 with A and methionine (M) 20 with leucine (L). Leucine is much more common than methionine at position 20 in the VH, and we speculated this mutation would help stability. M28 has three lysines and is fully active. We then proceeded stepwise to make M30 with two lysines, M31 with one lysine, and M32 with no lysine residues. All were fully active. When we replaced K38 in VH with R, we replaced R40 with T to help the stability of the Fv. Of 669 mouse Fv VH sequences with K at position 38, 62.5% have R at position 40. Of the 727 sequences with R at position 38, none has R at position 40. However, when R occurs at position 38, T occurs 20.9% of the time at position 40. Crystal structures of Fvs containing K38 and R40 show K38 is buried and R40 is partially buried. The side chains of these two residues are in close proximity and are parallel to each other. Replacing K38 with R would cause unfavorable interactions. Therefore, we replaced R40 with T. Also, when we

Immunotoxin Mutants

Bioconjugate Chem., Vol. 14, No. 2, 2003 483

Figure 2. Structural models of anti-Tac(dsFv) mutants. Lysine labels indicate which lysine residues are present in each mutant.

replaced K66 in VH with R, we replaced A67 with F, using similar logic. We restored the lysine residue at position 13 in the VH distant from the CDRs. Presumably K13 might be PEGylated without affecting its ability to bind to CD25 on the cell surface. This molecule is M33(dsFv)-PE38QQR. Finally, we placed a lysine residue at position 604 of PE38 to make M32(dsFv)-PE38QQR(604K). This would allow us to PEGylate the carboxyl region of PE38, which is reported to have many B cell epitopes (Roscoe et al., 1994, 1997). Expression and Purification of Disulfide BondStabilized Fv Recombinant Immunotoxins. Mutations were confirmed by DNA sequencing. Recombinant proteins were expressed in Escherichia coli, BL21 (λDE3), where they all accumulated in inclusion bodies. The immunotoxins were purified by our standard protocol, which consists of ion exchange and size exclusion chromatography using renatured inclusion body protein (Onda et al., 1999). Each immunotoxin eluted as a monomer upon TSK gel-filtration chromatography, and each migrated as a single major band of about 63 kDa in SDS-PAGE (Figure 4).

Cytotoxicity Studies. The data in Table 2 show the activity of each of the mutant molecules tested on ATAC4 cells, which contain the CD25 antigen. All but two of the mutant immunotoxins had similar cytotoxic activity on target cells. For example, the IC50s of M1(dsFv)-PE38 and M16(dsFv)-PE38 are 0.05 ( 0.01 and 0.04 ( 0.01 ng/mL, respectively. The IC50s of M13(dsFv)-PE38 and M15(dsFv)-PE38 are 0.12 and 0.13 ng/mL, respectively. We have not investigated the reason for the decreased activity of these two molecules. Replacement of the lysine residues in VH CDR2 with arginine did not cause a loss of cytotoxic activity in immunotoxins M27, M28, M30, M31, M32, and M33. Typical cytotoxicity curves from which the IC50s were calculated are shown in Figure 5A. Figure 5B shows the activity of each of the mutant molecules tested on OHS cells, which do not express CD25. Each of the mutant molecules tested had no cytotoxic activity on these antigen negative cells, whereas immunotoxin TP3(dsFv)-PE38, which targets an antigen on these cells, is active (Onda et al., 2001b). Binding Activity of Immunotoxins. Figure 6 shows a binding assay in which several different mutant im-

484 Bioconjugate Chem., Vol. 14, No. 2, 2003

Onda et al.

Figure 3. Sequences of Fvs aligned with frequency table. Top, the first sequence is VL of anti-Tac(dsFv). Numbering is according to Kabat. The remaining sequences are the anti-Tac Fv mutants constructed in our laboratory. Below is a table of amino acids sorted into bins according to their frequency in the Kabat database for each position in the framework region. Bottom, corresponding alignments for VHs. Residues shaded in gray indicate mutations made from anti-Tac(dsFv).

munotoxins were analyzed. All the immunotoxins displaced 125I-labeled human anti-Tac antibody in a similar manner with 50% displacement at 2 nM. This shows that the mutations did not affect binding to CD25. As expected, the bivalent antibody competed somewhat better than the monovalent immunotoxins with 50% displacement at 1 nM. Thus, substituting lysine residues with

other amino acids does not affect binding of the immunotoxin to CD25 positive cells. DISCUSSION

In this work, we describe the design, production, and cytotoxic properties of mutants of recombinant immunotoxin derived from anti-Tac(dsFv)-PE38. These mutants

Immunotoxin Mutants

Bioconjugate Chem., Vol. 14, No. 2, 2003 485

Table 2. Mutations and Biological Activity of Anti-Tac Immunotoxins

anti-Tac immunotoxin Ml(dsFv)-PE38 M3(dsFv)-PE38 M4(dsFv)-PE38 M5(dsFv)-PE38 M6(dsFv)-PE38 M7(dsFv)-PE38 M9(dsFv)-PE38 M10(dsFv)-PE38 M11(dsFv)-PE38 M12(dsFv)-PE38 M13(dsFv)-PE38 M14(dsFv)-PE38 M15(dsFv)-PE38 M16(dsFv)-PE38 M16(dsFv)-PE38QQR M18(dsFv)-PE38 M19(dsFv)-PE38 M27(dsFv)-PE38QQR M28(dsFv)-PE38QQR M30(dsFv)-PE38QQR M31(dsFv)-PE38QQR M32(dsFv)-PE38QQR M32(dsFv)-PE38QQR(604K) M33(dsFv)-PE38QQR

residues mutated VL: VL: VL: VL: VL: VL: VH: VH: VH: VH: VL: VL: VL: VL: VH: VL: VL: VH: VL: VH: VL: VH: VL: VH: VL: VH: VL: VH: PE: VL: VH:

K103 K107 K103, K107 K18 K45 K18, K45, R77 K13 K73 K73 K13, K73 K18, K45, K103, K107 K18, K45, R77 VH: K13, K73 K18, K45, R77, K103, K107 VH: K13, K73 K18, K45, R77, K103, K107 VH: K13, K73 K23 K103, K107 VH: K23, K73 K18, K45, R77, K103, K107 K13, K62, K64, K73 KI8, K45, R77, K103, K107 K13, K19, M20, K23, K62, K64, K73 K18, K39, K45, R77, K103, K107 K13, K19, M20, K23, K62, K64, K73 K18, K45, R77, K103, K107 K13, K19, M20, K23, K38, R40, K62, K64, K66, A67, K73 K18, K39, K45, R77, K103, K107 K13, K19, M20, K23, K38, R40, K62, K64, K66, A67, K73 K18, K39, K45, R77, K103, K107 K13, K19, M20, K23, K38, R40, K62, K64, K66, A67, K73 604 K18, K39, K45, R77, K103, K107 K19, M20, K23, K38, R40, K62, K64, K66, A67, K73

Figure 4. SDS-PAGE analysis of purified immunotoxins. The purified proteins were run on 4-20% gradient SDS-PAGE under nonreducing conditions (A) and under reducing conditions (B). The gels were stained with Coomasie Blue. Lane 1, M1(dsFv)-PE38; lane 2, M16(dsFv)-PE38QQR; lane 3, M27(dsFv)PE38QQR; lane 4, M32(dsFv)-PE38QQR; lane 5, M32(dsFv)PE38QQR(604K); lane 6, M33(dsFv)-PE38QQR; M, molecular mass standards are (top to bottom) 201, 130, 94, 48.6, 36.4, 29.8, 20.6, and 6.6 kDa, respectively.

number of lysine in Fv

number of lysine in whole ITs

IC50 (ng/mL)

13 12 12 11 12 12 11 12 12 12 11 9 9 7 7 12 11 5

16 15 15 14 15 15 14 15 15 15 14 12 12 10 7 15 14 5

0.05 ( 0.01 0.05 ( 0.01 0.05 ( 0.01 0.05 ( 0.01 0.05 ( 0.01 0.05 ( 0.01 0.05 ( 0.01 0.05 ( 0.01 0.05 ( 0.01 0.05 ( 0.01 0.12 ( 0.03 0.06 ( 0.01 0.13 ( 0.02 0.04 ( 0.01 0.04 ( 0.01 0.05 ( 0.01 0.05 ( 0.01 0.04 ( 0.007

3

3

0.04 ( 0.01

2

2

0.05 ( 0.01

1

1

0.03 ( 0.02

0

0

0.04 ( 0.01

0

1

0.05 ( 0.02

1

1

0.05 ( 0.02

Figure 5. In vitro specific cytotoxicity of anti-Tac(dsFv) immunotoxins and TP-3(dsFv)-PE38 on ATAC4 cells (A) and the human osteosarcoma cell line, OHS (B). Various immunotoxins were diluted with 0.2% BSA in DPBS. ATAC4 cells were seeded at 1.6 × 104 cells/well in 96-well plates 24 h before the addition of immunotoxin, then incubated at 37 °C for 20 h, and assayed by measuring inhibition of incorporation of 3H-Leucine. M1(dsFv)-PE38 (O), M16(dsFv)-PE38QQR (4), M32(dsFv)-PE38QQR (0), M33(dsFv)-PE38QQR (b), M32(dsFv)-PE38QQR(604K) (2), and TP3(dsFv)-PE38 (9).

486 Bioconjugate Chem., Vol. 14, No. 2, 2003

Figure 6. Displacement of 125I-HAT-IgG with anti-Tac(dsFv)immunotoxins and TP-3(dsFv)-PE38 using HUT102 cell line, which expresses CD25 but not TP-3 antigen. Triplicate sample values were averaged and the standard deviation was calculated for each data point. M1(dsFv)-PE38 (O), M16(dsFv)-PE38QQR (4), M32(dsFv)-PE38QQR (0), M33(dsFv)-PE38QQR (9), M32(dsFv)-PE38QQR(604K) (2), TP3(dsFv)-PE38 (b), and HAT-IgG (+).

differ in the number of lysines they contain. Almost all the immunotoxins were fully cytotoxic to ATAC4 cells and retained their affinity for CD25. The aim of this study was to show it was possible to diminish the number of lysine residues without losing biological function. AntiTac(dsFv)-PE38 has a total of 16 lysines. There are three lysines in PE38. The other 13 are in the Fv (Figure 1). Two of these lysines are in CDR2 of VH and could contribute to binding. We used the frequencies of residues at specific locations in the Fv to construct these mutants. For replacement, we chose hydrophilic residues because these are frequently present on the surface of proteins (Debinski and Pastan, 1994). The final molecule is M32(dsFv)-PE38QQR. It has no lysine residues yet has completely retained its cytotoxic activity and antigenbinding activity. Having established that we could produce an active immunotoxin without lysines and that it is fully active, we inserted lysine residues at positions that should minimally interfere with immunotoxin function. One is at VH K13, which is distant from the CDRs of the Fv. The other is at position 604 of PE38, a region previously shown to be highly immunogenic (Roscoe et al., 1994). We applied our strategy to another molecule. We could remove all lysine residues from SS1(dsFv)-PE38 without losing biological activity (Beers et al., unpublished data). We are also trying to apply our strategy to another protein. We believe that our approach has an ability to be generalized. In summary, we show that all lysine residues can be removed from anti-Tac(dsFv)-PE38 and that lysine residues can be introduced at new and potentially useful sites in the immunotoxin without loss of activity. The next paper in this series will describe the properties of immunotoxins modified by site specific PEGylation. ACKNOWLEDGMENT

We especially thank Drs. Satoshi Nagata and Kenneth Santora for helpful discussions. We also thank Anna Mazzuca for expert editorial assistance. LITERATURE CITED (1) Bailon, P., and Berthold, W. (1998) Polyethylene glycolconjugated pharmaceutical proteins. Pharm. Sci. Technol. Tox. 1, 352-356. (2) Bera, T. K., Kennedy, P. E., Berger, E. A., Barbas, C. F. III and Pastan, I. (1998) Specific killing of HIV infected lymphocytes by a recombinant immunotoxins directed against the HIV1 envelope glycoprotein. Mol. Med. 4, 384-391.

Onda et al. (3) Brinkmann, U., Pai, L. H., FitzGerald, D. J., Willingham, M., and Pastan, I. (1991) B3(Fv)-PE38KDEL, a single chain immunotoxins that causes complete regression of a human carcinoma in mice. Proc. Natl. Acad. Sci. U.S.A. 88, 86168620. (4) Chaffee, S., Mary, A., Stiehm, E. R., Girault, D., Fischer, A., and Hershfield, M. S. (1992) IgG antibody response to polyethylene glycol-modified adenosine deaminase in patients with adenosine deaminase deficiency. J. Clin. Invest. 89, 1643-1651. (5) Chothia, C., Gelfand, I., and Kister, A. (1998) Structural determinants in the sequences of immunoglobulin variable domain. J. Mol. Biol. 278, 457-479. (6) Chowdhury, P. S., Viner, J. L., Beers, R., and Pastan, I. (1998) Isolation of a high affinity stable single chain Fv specific for mesothelin from DNA-immunized mice by phage display and construction of a recombinant immunotoxin with anti-tumor activity. Proc. Natl. Acad. Sci. U.S.A. 95, 669674. (7) Debinski, W., and Pastan, I. (1994) An immunotoxin with increased activity and homogeneity produced by reducing the number of lysine residues in recombinant Pseudomonas exotoxin. Bioconjugate Chem. 5, 40-46. (8) Hermanson, G. T. (1996) The chemistry of reactive groups. In Bioconjugate Techniques, pp 137-168, Academic Press, San Diego. (9) Katre, N. V., Knauf, M. J., and Laird, W. J. (1987) Chemical modification of recombinant interleukin 2 by polyethylene glycol increases its potency in the murine Meth A sarcoma model. Proc. Natl. Acad. Sci. U.S.A. 84, 1487-1491. (10) Kitamura, K., Takahashi, T., Takashina, K., Yamaguchi, T., Noguchi, A., Tsurumi, H., Toyokuni, T., and Hakomori, S. (1990) Polyethylene glycol modification of the monoclonal antibody A7 enhances its tumor localization. Biochem. Biophys. Res. Commun. 171, 1387-1394. (11) Kreitman, R. J., and Pastan, I. (1995) Targeting Pseudomonas exotoxin to hematologic malignancies. Semin. Cancer Biol. 6, 297-306. (12) Kreitman, R. J., Wilson, W. H., Robbins, D., Margulies, I., Stetler-Stevenson, M., Waldmann, T. A., and Pastan, I. (1999) Responses in refractory hairy cell leukemia to a recombinant immunotoxins. Blood 94, 3340-3348. (13) Kreitman, R. J., Wilson, W. H., White, J. D., StetlerStevenson, M., Jaffe, E. S., Giardina, S.,Waldmann, T. A., and Pastan, I. (2000) Phase I trial of recombinant immunotoxins anti-Tac(Fv)-PE38 (LMB-2) in patients with hematologic malignancies. J. Clin. Oncol. 18, 1622-1636. (14) Kreitman, R. J., Wilson, W. H., Bergeron, K., Raggio, M., Stetler-Stevenson, M., FitzGerald, D. J., and Pastan, I. (2001) Efficacy of the anti-CD22 recombinant immunotoxins BL22 in chemotherapy-resistant Hairy-cell leukemia. N. Engl. J. Med. 345, 241-247. (15) Kunkel, T. A. (1985) Rapid and efficient site-specific mutagenesis without phenotypic selection. Proc. Natl. Acad. Sci. U.S.A. 82, 488-492. (16) Mansfield, E., Amlot, P., Pastan, I., and FitzGerald, P. (1997) Recombinant RFB4 immunotoxins exhibit potent cytotoxic activity for CD22-bearing cells and tumors. Blood 80, 2020-2026. (17) Nicholls, A., Sharp, K., and Honig, B. (1991) Protein folding and association: insights from the interfacial and thermodynamic properties of hydrocarbons. Proteins: Struct. Funct. Genet. 11, 281-296. (18) Onda, M., Kreitman, R. J., Vasmatzis, G., Lee, B., and Pastan, I. (1999) Reduction of the nonspecific animal toxicity of anti-Tac(Fv)-PE38 by mutations in the framework regions of the Fv which lower the isoelectric point. J. Immunol. 163, 6072-6077. (19) Onda, M., Willingham, M., Wang, Q. C., Kreitman, R. J., Tsutsumi, Y., Nagata, S., and Pastan, I.(2000) Inhibition of TNF-R produced by Kupffer cells protects against the nonspecific liver toxicity of immunotoxins anti-Tac(Fv)-PE38, LMB-2. J. Immunol. 165, 7150-7156. (20) Onda, M., Nagata, S., Tsutsumi, Y., Vincent, J. J., Wang, Q., Kreitman, R. J., Lee, B., and Pastan, I. (2001a) Lowering the isoelectric point of the Fv portion of recombinant immu-

Immunotoxin Mutants notoxins leads to decreased nonspecific animal toxicity without affecting antitumor activity. Cancer Res. 61, 50705077. (21) Onda, M., Olafsen, T., Tsutsumi, Y., Bruland, Ø. S., and Pastan, I. (2001b) Cytotoxicity of antiosteosarcoma recombinant immunotoxins composed of TP-3 Fv fragments and a truncated Pseudomonas exotoxin A. J. Immunother. 24, 144150. (22) Pai-Scherf, L., Villa, J., Pearson, D., Watson, T., Liu, E., Willingham, M. C., and Pastan, I.(1999) Hepatotoxicity in cancer patients receiving erb-38 a recombinant immunotoxins that targets the erbB2 receptor. Clin. Cancer Res. 5, 23112315. (23) Pastan, I. (1997) Targeted therapy of cancer with recombinant immunotoxins. Biochim. Biophys. Acta: Rev. Cancer 1333, C1-C6. (24) Peitsch, M. C. (1996) ProMod and Swiss-Model: Internetbased tools for automated comparative protein modeling. Biochem. Soc. Trans. 24, 274-279. (25) Pyatak, P. S., Abuchowski, A., and Davis, F F. (1980) Preparation of a polyethylene glycol:superoxide dismutase adduct, and an examination of its blood circulation life and anti-inflammatory activity. Res. Commun. Chem. Pathol. Pharmacol. 29, 113-127. (26) Queen, C., Schneider, W. P., Selick, H. E., Payne, P. W., Landolfi, N. F., Duncan, J. F.,Avdalovic, N. M., Levitt, M., Junghans, R. P., and Waldmann, T. A. (1989) A humanized antibody that binds to the interleukin 2 receptor. Proc. Natl. Acad. Sci. U.S.A. 86, 10029-10033.

Bioconjugate Chem., Vol. 14, No. 2, 2003 487 (27) Reiter, Y., Brinkmann, U., Jung, S. H., Lee, B., Kasprzyk, P. G., King, C. R., and Pastan, I. (1994) Improved binding and antitumor activity of a recombinant anti-erbB2 immunotoxins by disulfide stabilization of the Fv fragment. J. Biol. Chem. 269, 18327-18331. (28) Roscoe, D. M., Jung, S. H., Benhar, I., Pai, L., Lee, B. K., and Pastan, I. (1994) Primate antibody response to immunotoxins: serological and computer-aided analysis of epitopes on a truncated form of Pseudomonas exotoxin. Infect. Immunol. 62, 5055-5065. (29) Roscoe, D. M., Pai, L. H., and Pastan, I. (1997) Identification of epitopes on a mutant form of Pseudomonas exotoxin using serum from humans treated with Pseudomonas exotoxin containing immunotoxins. Eur. J. Immunol. 27, 1459-1468. (30) Shrake, A., and Rupley, J. (1973) Environment and exposure to solvent of protein atoms. Lysozyme and insulin. J. Mol. Biol. 79, 351-371. (31) Tsutsumi, Y., Onda, M., Nagata, S., Lee, B., Kreitman, R. J., and Pastan, I. (2000) Site-specific chemical modification with polyethylene glycol of recombinant immunotoxins antiTac(Fv)-PE38 (LMB-2) improves antitumor activity and reduces animal toxicity and immunogenicity. Proc. Natl. Acad. Sci. U.S.A. 97, 8548-8553. (32) Wang, Q. C., Pai, L. H., Debinski, W., FitzGerald, D. J., and Pastan, I. (1993) Polyethyleneglycol-modified chimeric toxin composed of transforming growth factor alpha and Pseudomonas exotoxin. Cancer Res. 53, 4588-4594.

BC020069R