Expression, purification, and refolding of a novel immunotoxin containing humanized single-chain fragment variable antibody against CTLA4 and the N-terminal fragment of human perforin

Immunotoxins might be potential in treatment of cancer for their ability to kill selected cell populations. We constructed a novel immunotoxin hS83P34 by fusing N-terminal 34 amino acid fragment of human perforin to the C-terminus of humanized single-chain fragment variable antibody against CTLA4. The fusion protein was inductively expressed as inclusion bodies at a high level about 30% of total bacterial proteins. After washing with buffer containing 2 M urea, the purity of inclusion body was about 71%. The washed inclusion bodies were solubilized in 8 M urea and further purified to homogeneity (approximately 92% purity) by cation-exchange chromatography and Ni-agarose affinity chromatography under denaturing condition. The inclusion body refolding conditions were optimized following Pro-Matrix Protein Refolding Guide. After refolded in Tris buffer (pH 8.0) containing 1M urea, 0.8 M l-arginine, and 2 mM GSH:0.2 mM GSSG or 2 mM GSH:0.4 mM GSSG for 18h at 4 degrees C, over 90% proteins were recovered from inclusion bodies. In vitro dose-dependent cytotoxicity assay demonstrates that hS83P34 is only toxic to CTLA4-positive cells. IC(50) of hS83P34 for leukemic cells Raji and 6T-CEM are about 0.85 and 1.3 microM individually. Whereas, CTLA4-negative endothelial cell ECV-304 is resistant to hS83P34.     

Cytotoxic properties of a ricin A chain immunotoxin recognising the cluster-5A antigen associated with human small-cell lung cancer

Summary The cytotoxic properties of a ricin A chain immunotoxin made with the mouse monoclonal antibody SWA20, recognising a family of sialoglycoprotein antigens selectively expressed by human small-cell lung cancer (SCLC), were examined using a panel of tumour cell lines in tissue culture. SWA20—ricin-A-chain was selectively toxic to the SW2, NCI-H69 and GLC-8 SCLC cell lines, inhibiting the incorporation of [³H]leucine by 50% at a concentration of 0.2–2 nM, but had no selective activity against the NCI-H23 and NCI-H125 lung adenocarcinoma or the control CEM T-lymphoblastoid cell lines. The SWA20 immunotoxin intoxicated the SW2 cell line rapidly, inhibiting [³H]leucine incorporation by 50% within 2 h compared with 0.5 h for ricin. Analysis of the effects of SWA20—ricin-A-chain on the growth of SW2 cells using a limiting-dilution clonogenic assay revealed that the immunotoxin could eliminate 95% of clonogenic malignant cells. Although SWA20—ricin-A-chain was found to be rapidly active against the majority of tumour cells, its action was limited by the presence of insensitive cells expressing low levels of the target antigen.     

T lymphocyte killing by a xanthine-oxidase-containing immunotoxin

Summary We report on the preparation of an immunotoxin consisting of xanthine oxidase, a free-radicalproducing enzyme, covalently linked to an anti-CD3 monoclonal antibody. The immunotoxin retained both enzymic and immunological properties and its toxicity to target cells (a) was greater than that of the free enzyme, (b) was proportional to the enzyme concentration, and (c) was reduced either in the absence of hypoxanthine or by an excess of free anti-CD3 monoclonal antibody. The cytotoxicity and selectivity of the hypoxanthine/conjugated xanthine oxidase system were potentiated by the addition of chelated iron and by washing away the unbound immunotoxin prior to the addition of substrate. The same system was not toxic to bone marrow progenitor cells. A possible use of this immunotoxin for the ex vivo purging of organs to be transplanted from T lymphocytes, to avoid the graft-versus-host reaction, is suggested.     

Immunotoxins to the HER-2 oncogene product: Functional and ultrastructural analysis of their cytotoxic activity

Two immunotoxins were prepared using monoclonal antibodies (mAb) directed towards two distinct epitopes of the gp185^HER-2 extracellular domain, and the type I ribosome inactivating protein (RIP) plant toxin saporin 6. Cell protein synthesis inhibition assay reveals that the immunotoxins display a potent and specific cytotoxicity that is characterized by a slow rate, since the time required to inhibit incorporation of radiolabeled leucine completely ranges from 36 h to 60 h depending on the target cell line and the immunotoxin. Because this feature may hamper the immunotherapeutic use of these conjugates we analysed this further by studying the early phases of internalization of immunotoxins by immunoelectron microscopy. The results of this study have demonstrated that the distribution pattern of the immunotoxins and of the unconjugated mAb over the cell surface overlaps. Similarly the mAb and immunotoxins are internalized into the cell by two different pathways: via clathrin-coated pits or via smaller uncoated pits and vesicles. A higher degree of internalization is achieved when the two immunotoxins are used in combination. Unlike the slow kinetics of cell intoxication the process of immunotoxin endocytosis is characterized by a rapid rate of internalization (above 40% at 5 min in the SK-BR-3 cell line). Although these findings provide no clue to explain the mechanisms of the slow rate of cytotoxicity of the two immunotoxins their rapid internalization indicates that these reagents can be exploited in immunotherapeutic approches to gp185^HER-2-expressing malignancies.     

括楼素的纯化及其免疫毒素的制备

利用凝胶过滤及离子交换层析从Trichosanthes kirilowii Maximowicz种子中分离出一种核糖体失活蛋白,称之为括楼素(Trichokirin)。这种毒素在无细胞体系中对蛋白质生物合成具有明显的抑制活性而对完整细胞的毒性很低。与单克隆抗体偶联后,括楼素对靶细胞显示了明显的特异性细胞毒作用。     

Production and introduction of a novel immunotoxin based on engineered RNase A for inducing death to Her1-positive cell lines

The present study was performed to design an immunotoxin consisting of engineered RNase A and scFv of Cetuximab. To accomplish this study goal, at first to evade RNase A from its inhibitors in the cytoplasm, six amino acids of RNase A were substituted, then the physicochemical features of engineered RNase A were assessed. To investigate the interaction between the engineered RNase A and the ribonuclease inhibitor, protein–protein docking was performed. After engineering the RNase A, it was theoretically conjugated with scFv of Cetuximab using a cleavable linker to produce scFv-engineered RNase A. Then, wild-RNase A (14 kD), engineered RNase A (14 kD) and scFv-engineered RNase A (42 kDa) were expressed in the BL21 (DE3) strain of Escherichia coli and purified by Ni-NTA columns. To confirm the expressed proteins, western blot analysis was performed. The functioning of wild-RNase A and engineered RNase A were investigated by RNA fragmentation assay. Finally, to evaluate the cytotoxicity of scFv-engineered RNase A, a dose–response cytotoxicity assay was performed on Her1-positive and Her1-negative cell lines. The results showed that engineered RNase A could maintain its structure and disulfide bonds and evade its inhibitor. Expression and purification were successfully conducted and both enzymes could degrade yeast RNA. The result of cytotoxicity showed that the engineered immunotoxin could induce cell death to Her1-positive cell lines with an IC₅₀ of 50 nM. It appears that scFv-engineered RNase A can be a promising molecule for use.     

Different cytotoxic activity and intracellular fate of an anti-CD5-momordin immunotoxin in normal compared to tumour cells

We investigated the different sensitivity of peripheral blood mononuclear cells (PBMC) and human T cell leukaemias (Jurkat and CEM) to an anti-CD5-momordin immunotoxin. In a short-term assay, the immunotoxin displayed different cytotoxic activity on normal and tumour cells: for leukaemic cell lines an incubation time of 72 h was necessary for the immunotoxin to reach the IC₅₀ of 41–53 pM, compared to the 1 h sufficient for 6 pM immunotoxin to inhibit 50% of PBMC protein synthesis. In a long-term clonogenic assay (15 days), the immunotoxin demonstrated a compareble efficacy of clonogenic cell killing for both cell types. We investigated the immunotoxin internalization pathway by a flow-cytometric method and our data seem to indicate that the molecules meet a different intracellular fate in the two cell populations. It may be assumed that the low cytotoxic activity of immunotoxins on tumour cells, detected in the shortterm assay, is due to infefficient delivery to their cytoplasmatic target, while a longer exposure of the cells to the immunotoxin promotes adequate intracellular distribution.     

Characterization of T-cell repertoire in hairy cell leukemia patients before and after recombinant immunotoxin BL22 therapy

We previously reported that hairy cell leukemia (HCL) patients have high percentages of CD56+/CD57+/CD3+ large granular lymphocytes consistent with cytotoxic T-lymphocytes (CTLs), and other investigators have reported skewing of the T-cell repertoire. In previous studies of up to seven HCL patients, many of the 22 established T-cell receptor (TCR) beta variable region (TRBV) families showed mono- or oligoclonal restriction. To determine whether percentages of CTLs are correlated with TRBV clonal excess, we studied 20 HCL patients with flow cytometry, PCR of TCR gamma and TRBV regions, and fractional gel electrophoresis of PCR-amplified TRBV CDR3 domains (CDR3 spectratyping). Increased percentages of CD3+/CD8+/CD57+ CTLs correlated with more mono/oligoclonal and fewer polyclonal TRBV families (r=0.53; P=0.016). Age correlated with number of mono/oligoclonal TRBV families (r=0.51; P=0.022). Time since last purine analog therapy correlated with number of polyclonal TRBV families (r=0.46; P=0.040), but treatment with the anti-CD22 recombinant immunotoxin BL22 was not related to clonal excess. We conclude that abnormalities in the T-cell repertoire in HCL patients may represent deficient immunity, and may be exacerbated by purine analogs. Increased CD3+/CD57+ T-cells may be a useful marker of abnormal TRBV repertoire in HCL patients, and might prove useful in deciding whether patients should receive biologic antibody-based treatment rather than repeated courses of purine analog for relapsed disease.     

Treatment of hepatocellular carcinoma in a mouse xenograft model with an immunotoxin which is engineered to eliminate vascular leak syndrome

Vascular leak syndrome (VLS) is the major dose-limiting toxicity of immunotoxin and interleukin-2 therapy. It has been evidenced that VLS-inducing molecules share a three-amino acid consensus motif, (x)D(y), which may be responsible for initiating VLS. Here we have constructed a recombinant immunotoxin (SMFv-PE38KDEL) by genetically fusing PE38KDEL to a single-chain antibody derived from SM5-1 monoclonal antibody, which has a high specificity for melanoma, hepatocellular carcinoma and breast cancer. In order to eliminate VLS induced by this PE38KDEL-based immunotoxin, a panel of mutants were generated by changing amino acid residues adjacent to its three (x)D(y) motifs in the three-dimensional structure. One of the SMFv-PE38KDEL mutants, denoted as mut1, displayed a similar protein synthesis inhibitory in a reticulocyte lysate translation assay compared to the wild-type SMFv-PE38KDEL (wt). The in vitro cytotoxicity assay indicated that mut1 specifically killed SM5-1 binding protein-positive tumor cells, although its cytotoxicity was slightly less than wt. In contrast, mut1 was shown to be much weaker in inducing VLS in mice than wt. The LD₅₀ values of wt and mut1 in mice were investigated with the result that the LD₅₀ of mut1 was about tenfold higher than that of wt. The in vivo antitumor activity of wt and mut1 were also compared in tumor-bearing nude mice. Both wt and mut1 were effective in inhibiting the tumor growth but mut1 showed improved therapeutic efficacy. These studies suggest mut1 may be a novel PE-based immunotoxin with much less toxicity for clinical use. Hao Wang, Shuichuan Song and Geng Kou contributed equally to this paper.     

Optimization of expression and purification of two biologically active chimeric fusion proteins that consist of human interleukin-13 and Pseudomonas exotoxin in Escherichia coli

We have previously reported that a variety of solid human tumor cell lines express a large number of receptors for interleukin-13 (IL-13). These receptors could be targeted with a chimeric fusion protein consisting of human IL-13 and a truncated form of Pseudomonas exotoxin (PE). We describe here optimization of critical steps involved in high yield expression of two recombinant chimeric fusion proteins for obtaining highly purified and biologically active cytotoxins in Escherichia coli. The chimeric constructs of human IL-13 and two 38 kDa truncated PEs: (i) PE38 and (ii) PE38QQR, (three lysine residues in PE38 at 590, 606, and 613 substituted with two glutamine and one arginine) were used for protein expression in pET prokaryotic expression vector system with kanamycin as a selection antibiotic. Our results suggest that fresh transformation of E. coli and induction by isopropyl-beta-D-thiogalactopyranoside (IPTG) for 6 h resulted in maximum protein expression. To further improve the yield, we used a genetically modified E. coli strain, BL21(DE3)pLysS, which carries a plasmid for lysozyme with a weak promoter that inhibits T7 RNA polymerase and minimizes protein production in the absence of IPTG. Use of this strain eliminated the need for lysozyme digestion of the induced bacteria to release inclusion bodies, which resulted in expression of purer protein as compared to the conventional BL21(DE3) strain. Additional protocol optimizations included 16 h solubilization of inclusion bodies, constitution of refolding buffer, and timing of dialysis. These proteins were finally purified by Q-Sepharose, mono-Q, and gel filtration chromatography. Between 14-22 and 21-28 mg highly purified and biologically active protein was obtained from 1L of BL21 (DE3) and BL21 (DE3) pLysS bacteria culture, respectively. As IL-13R targeting for brain tumor therapy offers an exciting treatment option, optimization of production of IL-13PE will enhance production of clinical grade material for Phase III clinical trials.