Search

Radiolabeled single-chain Fv (sFv) molecules display highly specific tumor retention in the severe combined immunodeficient (SCID) mouse model; however, the absolute quantity of sFv retained in the tumors is diminished by the rapid renal elimination resulting from the small size of the sFv molecules (Mr 27,000) and by dissociation of the monovalent sFv from tumor-associated antigen. We previously reported significant improvement in tumor retention without a loss of targeting specificity on converting monovalent sFv into divalent [(sFv')2] dimers, linked by a disulfide bond between COOH-terminal cysteinyl peptides engineered into the sFv'. However, our data for enhanced dimer localization in tumors could not distinguish between the contributions of enhanced avidity and increased systemic retention associated with the larger size of 54 kDa [(sFv')2] dimers relative to 27-kDa sFv. In this investigation, we have compared tumor targeting of divalent anti-c-erbB-2/HER2/neu 741F8-1 (sFv')2 homodimers with monovalent 741F8/26-10 (sFv')2 heterodimers (Mr 54,000) and 741F8 sFv monomers (741F8 sFv has binding specificity for erbB-2/HER2/neu and 26-10 sFv specificity for digoxin and related cardiac glycosides). These studies allowed us to distinguish the dominant effect of valency over molecular weight in accounting for the superior tumor retention of 741F8-1 (sFv')2 homodimers. Each of the radioiodinated species was administered i.v. to SCID mice bearing SK-OV-3 human tumor xenografts and tumor localization at 24 hours post i.v. injection was determined for 125I-741F8-1 (sFv')2 (3.57 %ID/g), 125I-741F8/26-10 (sFv')2 (1.13 %ID/g), and 125I-741F8-1 sFv (1.25 %ID/g). These findings substantiate that the improved tumor retention of (sFv')2 homodimers over sFv monomers results from the availability of dual binding sites rather than from the slower systemic clearance of homodimers.

Over-expression of the epidermal growth factor receptor (EGFR) is a hallmark of numerous solid tumors, thus providing a means of selectively targeting therapeutic agents. Heparin-binding epidermal growth factor (HBEGF) binds to EGFRs with high affinity and to heparan sulfate proteoglycans, resulting in increased mitogenic potential compared to other EGF family members. We have investigated the feasibility of using HBEGF to selectively deliver a cytotoxic protein into EGFR-expressing tumor cells. Recombinant fusion proteins consisting of mature human HBEGF fused to the plant ribosome-inactivating protein saporin (SAP) were expressed in Escherichia coli. Purified HBEGF-SAP chimeras inhibited protein synthesis in a cell-free assay and competed with EGF for binding to receptors on intact cells. A construct with a 22-amino-acid flexible linker (L22) between the HBEGF and SAP moieties exhibited an affinity for the EGFR that was comparable to that of HBEGF. The sensitivity to HBEGF-L22-SAP was determined for a variety of human tumor cell lines, including the 60 cell lines comprising the National Cancer Institute Anticancer Drug Screen. HBEGF-L22-SAP was cytotoxic in vitro to a variety of EGFR-bearing cell lines and inhibited growth of EGFR-over-expressing human breast carcinoma cells in vivo. In contrast, the fusion protein had no effect on small-cell lung carcinoma cells, which are EGFR-deficient. Our results demonstrate that fusion proteins composed of HBEGF and SAP exhibit targeting specificity and cytotoxicity that may be of therapeutic value in treating a variety of EGFR-bearing malignancies.

The development of specific antibody probes for characterizing the expression of the family of 4 fibroblast growth factor receptor (FGFR) types has been difficult because of their close homology to each other and high degree of evolutionary conservation. Of the existing anti-FGFR monoclonal antibodies (MAbs), there are few that are useful for staining paraffin-embedded tissues. We have raised MAbs against human FGFR1 and FGFR2 in both rats and mice using bacterial recombinant receptor fusion proteins as immunogens. We used peptide epitope mapping to characterize the immune sera and the selected MAbs. Immunized animals were selected that displayed the broadest reactivity against epitopes unique to the immunizing receptor type. We produced FGFR1 specific MAbs that bind epitopes in immunoglobulin domain I (Ig-I) and FGFR2 specific MAbs that bind epitopes in Ig-I, Ig-II, and the acid box. The specificity of the antibodies was demonstrated by ELISA and immunoblot analysis of purified recombinant FGFR1 and FGFR2 extracellular domains produced both in E. coli and in eucaryotic cells. Based on the lack of epitope homology, these MAbs would not be expected to cross-react with FGFR3 or FGFR4. We isolated MAbs that bound to paraffin embedded tissue and immunoblots of recombinant receptor. These epitope-defined MAbs can distinguish between members of the FGF receptor family and should be useful as tools for assessing FGF receptor expression in a variety of normal and diseased tissues.

Objective: The goals of this clinical trial of intraventricular 454A12-rRA therapy were to identify dose-limiting toxicities, to evaluate the pharmacokinetics of single-dose intraventricular 454A12-rRA, and to detect antitumor activity., Methods: We performed a pilot study of intraventricular therapy with the immunotoxin 454A12-rRA in eight patients with leptomeningeal spread of systemic neoplasia. The immunotoxin 454A12-rRA is a conjugate of a monoclonal antibody against the human transferrin receptor and recombinant ricin A chain, the enzymatically active subunit of the protein toxin ricin. Patients were treated with single doses of 454A12-rRA ranging from 1.2 to 1200 micrograms., Results: The early phase half-life of 454A12-rRA in ventricular cerebrospinal fluid (CSF) averaged 44 +/- 21 minutes, and the late phase half-life averaged 237 +/- 86 minutes. The clearance of the immunotoxin was faster than the clearance of coinjected technetium-99m-diethylenetriamine penta-acetic acid, averaging approximately 2.4-fold greater. No 454A12-rRA degradation was detected by Western blot analysis of ventricular CSF for a period of 24 hours, and bioactivity was retained in CSF paralleling the concentration of immunotoxin. No acute or chronic drug toxicity was identified in patients who received less than or equal to 38 micrograms of 454A12-rRA by intraventricular injection. Doses more than or equal to 120 micrograms caused a CSF inflammatory response that was associated with transient headache, vomiting, and altered mental status. This acute syndrome was responsive to steroids and CSF drainage. No systemic toxicity was detected. In four of the eight patients, a greater than 50% reduction of tumor cell counts in the lumbar CSF occurred within 5 to 7 days after the intraventricular dose of 454A12-rRA; however, no patient had their CSF cleared of tumor, and clinical or magnetic resonance imaging evidence of tumor progression was demonstrated in seven of the eight patients after treatment., Conclusion: Tumoricidal concentrations of the immunotoxin 454A12-rRA can be attained safely in the CSF of patients with leptomeningeal tumor spread.

The availability of engineered antibody species has catalyzed new developments in radioimmunotargeting. This chapter summarized recent studies of single-chain Fv (sFv) proteins, which are minimal antibody binding sites engineered as single polypeptide chains. The single-chain Fv can be as small as 26 kDa monomers or may be engineered as larger fusion proteins designed to self-associate into dimeric or multimeric species. They typically exhibit rapid clearance that results in high targeting specificity within a matter of hours. We have compared different modes of administration to allow further manipulation of their biodistribution and targeting properties. Results of the present study comparing intravenous (i.v.) and intraperitoneal (i.p.) administration show comparable long-term retention in circulation, but the i.v. route showed an initially high peak blood level while i.p. injection did not. As with a single sFv dose, repeated bolus injections of sFv attained high target-to-background ratios, whereas continuous sFv infusion reached a steady state level of free sFv in blood and kidney that exceeded that in tumor xenografts. We observed improved localization of radioiodinated sFv in tumor xenografts if the radioiodine label resisted dehalogenation from the protein, which was accomplished, for example, through conjugation of a para-131I-benzoyl group to Iysyl epsilon-amino groups of the protein. Modification of the sFv by genetic incorporation of a cysteinyl peptide (to form sFv') provided a chelation site for radiometals that simplified incorporation of 99mTc with the opportunity for improved diagnostic imaging in cancer and other diseases. Therapeutic applications of sFv radioimmunotargeting could rely on sFv' complexed to 186Re or 188Re. Engineering sFv of sFv' with increased antigen-binding affinity and appropriately manipulating their mode of administration should promote sustained tumor retention conducive to clinically useful therapeutic indices.

In order to produce sufficient quantities of fibroblast growth factor-saporin (rFGF-2-SAP) mitotoxin for preclinical evaluation in models of diseases such as cancer and restenosis, we have undertaken the large-scale expression, purification, and characterization of the recombinant molecule. The fusion gene encoding rFGF-2-SAP was cloned into the inducible pET 11a expression vector and transformed into Escherichia coli strain BL21 (DE3). The transformants were grown using a fed-batch fermentation until the A600 reached 85. At this stage, induction of the expression of the fusion protein led to the production of approximately 2.2 mg/liter per A600 unit. The soluble mitotoxin was purified to homogeneity from cell lysates via expanded bed adsorption chromatography followed by cation-exchange, heparin-affinity, and size-exclusion chromatography. Purified rFGF-2-SAP contained less than 0.5 EU/mg of endotoxin, as determined by gel clot analyses. The highly purified rFGF-2-SAP retained the toxin's ability to inhibit protein synthesis as measured in a cell-free system and was cytotoxic to a number of normal and neoplastic cell lines bearing FGF receptors. Binding studies establish that the fusion protein exerts its effects via the FGF high-affinity receptor.

Basic fibroblast growth factor (FGF-2) and saporin were chemically conjugated using the crosslinker, N-succinimidyl-3(2-pyridyldithio)-propionate. When purified, the conjugate was found to be heterogeneous as analyzed by SDS/PAGE, size-exclusion HPLC and reverse-phase HPLC. Therefore, we sought to synthesize a molecule that would be homogeneous and thus easier to characterize and evaluate its efficacy and toxicity for pharmaceutical drug development. A homogeneous chemical conjugate was successfully synthesized by using a mutant FGF-2 with Cys96 replaced by Ser ([S96]FGF-2) and a recombinant saporin mutant containing a single Cys at the -1 position (C-SAP). The latter was expressed in Escherichia coli and isolated to 99% purity by expanded-bed adsorption chromatography followed by cation-exchange chromatography. The cysteine in C-SAP was activated by Ellman's reagent and then reacted with the only available cysteine (position 78) in [S96]FGF-2 to produce the homogeneous conjugate, designated as FGF2-C-SAP. The purified FGF2-C-SAP was more than 98% pure as judged by HPLC. In vitro biological assays indicated that FGF2-C-SAP was a potent inhibitor of protein synthesis in a cell-free system and was cytotoxic to FGF-receptor-bearing cells.

Unlabelled: The goal of this study was to determine if the stabilization of the radioiodine-protein bond by the N-succinimidyl p-iodobenzoate (PIB) method improved the degree and specificity of tumor localization of 125I-741F8-1 (sFv')2, an anti-c-erbB-2 sFv dimer, in an immunodeficient murine model., Methods: Gamma camera images were acquired 21 hr after intravenous administration of 131I-741F8-1 (sFv')2 labeled by the p-iodobenzoate or chloramine T methods. The stability of the radioiodine-protein bond also was assessed in plasma samples after intravenous injection of 125I-741F8-1 (sFv')2 labeled by either the chloramine T or p-iodobenzoate methods., Results: By 6 hr postinjection, 97% of the activity associated with the 125I-741F8-1 (sFv')2 labeled by the p-iodobenzoate method was protein bound compared with 61% after labeling with the chloramine-T method. These observations indicate that increasing the stability of the conjugation between the radioiodine and the sFv molecule can significantly increase the degree and specificity of tumor targeting. Significantly greater tumor retention (p < 0.005) and lower blood (p < 0.001), spleen (p < 0.001) and stomach (p < 0.005) retention were observed in biodistribution studies when the p-iodobenzoate conjugate was used. This resulted in superior tumor-to-organ ratios for all tissue samples studied., Conclusion: These observations may have clinical relevance for the use of radiolabeled sFv as imaging agents.

Purpose: To investigate the ability of fibroblast growth factor (FGF) 2-saporin to prevent lens regrowth in the rabbit., Methods: Chemically conjugated and genetically fused FGF2-saporin (made in Escherichia coli) were used. Extracapsular extraction of the lens was performed on the rabbit, and the cytotoxin either was injected directly into the capsule bag or was administered by FGF2-saporin-coated, heparin surface-modified (HSM) polymethylmethacrylate intraocular lenses. The potential of the conjugate was checked by slit lamp evaluation of capsular opacification and by measuring crystallin synthesis. Toxin diffusion and sites of toxin binding were assessed by immunohistochemistry. Possible toxicity was determined by histologic analysis of ocular tissues., Results: FGF2-saporin effectively inhibited lens regrowth when it was injected directly into the capsular bag. However, high concentration of the toxin induced transient corneal edema and loss of pigment in the iris. Intraocular lenses coated with FGF2-saporin reduced lens regrowth and crystallin synthesis without any detectable clinical side effect. After implantation, FGF2-saporin was shown to have bound to the capsules and, to a lesser extent, to the iris; no histologic damage was found on ocular tissues as a result of implantation of drug-loaded HSM intraocular lenses., Conclusions: Chemically conjugated (FGF2-SAP) and genetically fused FGF2-saporin (rFGF2-SAP) bound to HSM intraocular lenses can prevent lens regrowth in the rabbit.

Purpose: To test the ability of two preparations of FGF2-saporin, either FGF2 chemically conjugated to saporin (FGF2-SAP) or genetically engineered FGF2-saporin (rFGF2-SAP) to inhibit the growth of bovine epithelial lens (BEL) cells in vitro when in solution and when immobilized on heparin surface-modified (HSM) polymethylmethacrylate (PMMA) intraocular lenses (IOLs)., Method: Bovine epithelial lens cells were incubated with various concentrations FGF2-saporin for as long as 4 days. The number of surviving cells was determined by counting the number of nuclei. Because FGF2 binds to heparin, FGF2-saporin was incubated with HSM PMMA IOLs; excess toxin was washed off, and the BEL cells were grown on the FGF2-saporin-treated IOLs (HSM and non-HSM) for 4 days. Cell density was determined by image analysis., Results: Both FGF2-SAP and rFGF2-SAP were highly cytotoxic (nM range), with rFGF2-SAP 10 times less active than FGF2-SAP. FGF2-saporin bound to the surface of HSM IOLs and eluted by 2M NaCl retained its activity. Toxin bound to HSM IOLs killed more than 90% of the BEL cells placed on the IOL surface within 4 days. The ability of FGF2-saporin to prevent the growth of cells on the IOL surface was strictly dependent on the presence of heparin on the IOL., Conclusions: FGF2-saporin is bound to HSM PMMA IOLs and prevents the growth of epithelial cells on the surface of the lens.

We describe a method to facilitate radioimaging with technetium-99m (99mTc) by genetic incorporation of a 99mTc chelation site in recombinant single-chain Fv (sFv) antibody proteins. This method relies on fusion of the sFv C terminus with a Gly4Cys peptide that specifically coordinates 99mTc. By using analogues of the 26-10 anti-digoxin sFv as our primary model, we find that addition of the chelate peptide, to form 26-10-1 sFv', does not alter the antigen-binding affinity of sFv. We have demonstrated nearly quantitative chelation of 0.5-50 mCi of 99mTc per mg of 26-10-1 sFv' (1 Ci = 37 GBq). These 99mTc-labeled sFv' complexes are highly stable to challenge with saline buffers, plasma, or diethylenetriaminepentaacetic acid. We find that the 99mTc-labeled 741F8-1 sFv', specific for the c-erbB-2 tumor-associated antigen, is effective in imaging human ovarian carcinoma in a scid mouse tumor xenograft model. This fusion chelate methodology should be applicable to diagnostic imaging with 99mTc and radioimmunotherapy with 186Re or 188Re, and its use could extend beyond the sFv' to other engineered antibodies, recombinant proteins, and synthetic peptides.

Background: Antibody-based reagents have failed to live up to their anticipated role as highly specific targeting agents for cancer therapy. Targeting with human single-chain Fv (sFv) molecules may overcome some of the limitations of murine IgG, but are difficult to produce with conventional hybridoma technology. Alternatively, phage display of antibody gene repertoires can be used to produce human sFv., Objectives: To isolate and characterize human single chain Fvs which bind to c-erbB-2, an oncogene product overexpressed by 30-50% of breast carcinomas and other adenocarcinomas., Study Design: A non-immune human single-chain Fv phage antibody library was selected on human c-erbB extracellular domain and sFv characterized with respect to affinity, binding kinetics, and in vivo pharmacokinetics in tumor-bearing scid mice., Results: A human single-chain Fv (C6.5) was isolated which binds specifically to c-erbB-2. C6.5 is entirely human in sequence, expresses at high level as native protein in E. coli, and is easily purified in high yield in two steps. C6.5 binds to immobilized c-erbB-2 extracellular domain with a Kd of 1.6 x 10(-8) M and to c-erbB-2 on SK-OV-3 cells with a Kd of 2.0 x 10(-8) M, an affinity that is similar to sFv produced against the same antigen from hybridomas. Biodistribution studies demonstrate 1.47% injected dose/g tumor 24 h after injection of 125I-C6.5 into scid mice bearing SK-OV-3 tumors. Tumor:normal organ ratios range from 8.9:1 for kidney to 283:1 for muscle., Conclusions: These results are the first in vivo biodistribution studies using an sFv isolated from a non-immune human repertoire and confirm the specificity of sFv produced in this manner. The use of phage display to produce C6.5 mutants with higher affinity and slower k(off) would permit rigorous evaluation of the role of antibody affinity and binding kinetics in tumor targeting, and could result in the production of a therapeutically useful targeting protein for radioimmunotherapy and other applications.

Single-chain Fv molecules in monovalent (sFv) and divalent [(sFv')2] forms exhibit highly specific tumor targeting in mice as a result of their small size and rapid systemic clearance. As a consequence, there is a rapid reversal of the sFv blood/tumor gradient, resulting in diminished retention of sFv species in tumors. In this report we investigate two distinct strategies, dose escalation and repetitive intravenous (i.v.) dosing, aiming to increase the absolute selective retention of radiolabeled anti-c-erbB-2 125I-741F8 (sFv')2 in c-erbB-2-overexpressing SK-OV-3 tumors in mice with severe combined immunodeficiency (SCID). A dose-escalation strategy was applied to single i.v. injections of 125I-741F8 (sFv')2. Doses from 50 micrograms to 1000 micrograms were administered without a significant decrease in tumor targeting or specificity. High doses resulted in large increases in the absolute retention of 125I-741F8 (sFv')2. For example, raising the administered dose from 50 micrograms to 1000 micrograms increased the tumor retention 24 h after injection from 0.46 microgram/g to 9.5 micrograms/g, and resulted in a net increase of greater than 9 micrograms/g. Over the same dose range, the liver retention rose from 0.06 microgram/g to 1 microgram/g, and resulted in a net increase of less than 1 microgram/g. The retention of 9.5 micrograms/g in tumor 24 h following the 1000-micrograms dose of (sFv')2 was comparable to that seen 24 h after a 50-micrograms dose of 125I-741F8 IgG, indicating that the use of large doses of (sFv')2 may partially offset their rapid clearance. When two doses were administered by i.v. injection 24 h apart, the specificity of delivery to tumor observed after the first dose was maintained following the second injection. Tumor retention of 125I-741F8 (sFv')2 was 0.32 microgram/g at 24 h and 0.22 micrograms/g at 48 h following a single injection of 20 micrograms, while 0.04 microgram/ml and 0.03 microgram/ml were retained in blood at the same assay times. After a second 20-micrograms injection at the 24-h assay time, tumor retention increased to 0.49 micrograms/g, and blood retention was 0.06 microgram/ml, at the 48-h point. These results suggest that multiple high-dose administrations of radiolabeled 741F8 (sFv')2 may lead to the selective tumor localization of therapeutic radiation doses.

The production of several single-chain Fv (sFv) antibody proteins was examined by three modes of mammalian cell expression. Our primary model was the 741F8 anti-c-erbB-2 sFv, assembled as either the VH-VL or VL-VH, and expressed alone, with C-terminal cysteine for dimerization, or as fusion proteins with carboxyl-terminal effector domains, including interleukin-2, the B domain of staphylococcal protein A, the S-peptide of ribonuclease S, or hexa-histidine metal chelate peptide. Constructs were expressed and secreted transiently in 293 cells and stably in CHO or Sp2/0 cell lines, the latter yielding up to 10 mg per liter. Single-chain constructs of MOPC 315 myeloma and 26-10 monoclonal antibodies were also expressed, as were hybrids comprising unrelated VH and VL regions. Our results suggest that mammalian expression is a practical and valuable complement to the bacterial expression of single-chain antibodies.

Purpose: To determine the effect of a rabbit transferrin conjugated to recombinant ricin A chain (Tfr-rRA) and the carboxylic ionophore monensin on proliferating and density-arrested human retinal pigment epithelial cells and rabbit dermal fibroblasts., Methods: Cells were seeded on 24-well plates at 20,000 cells/cm2 and exposed to Tfr-rRA (0.1-10,000 ng/ml) with or without monensin (0.01 microM), and with or without human transferrin (65.7 mg/l) for 5 minutes to 7 days. Cells were studied morphologically and counted at 1, 2, 4, and 7 days., Results: Tfr-rRA (10-10,000 ng/ml) killed proliferating human retinal pigment epithelial cells and rabbit dermal fibroblasts in a dose-dependent manner (p < or = 0.01) up to a maximum of 86% and 93%, respectively. In contrast, Tfr-rRA had minimal effect on density-arrested human retinal pigment epithelial cells and rabbit dermal fibroblasts. The cytotoxicity of Tfr-rRA was inhibited by the addition of human transferrin (65.7 mg/l), an effect that was partially overcome by longer treatment with Tfr-rRA. Monensin (0.01 microM) increased the cytotoxicity of Tfr-rRA by 4.8-fold over Tfr-rRA alone, shortened the onset of cell kill with Tfr-rRA from 48 to 24 hours (P = 0.04), and partially reversed the neutralizing effect of human transferrin., Conclusions: The results indicate that Tfr-rRA effectively inhibited the proliferation of human retinal pigment epithelial cells and rabbit dermal fibroblasts in vitro. The inhibitory effect could be modified by the addition of human transferrin or monensin. Thus, this ricin A chain conjugate may interrupt the proliferation of cells necessary in the pathogenesis of proliferative vitreoretinopathy.

We examined the effects of the immunotoxin 260F9 Mab-recombinant ricin A (developed against human breast carcinoma) on proliferating and confluent human corneal epithelium (HCE) cells in vitro. HCE cells derived from explants of discarded human donor corneoscleral rims were established as proliferating and confluent cell cultures, and were exposed continuously for 7 days to immunotoxin. Final cell counts at day 7, and thymidine uptake measured at days 1 and 7 postexposure, showed > 95% suppression of proliferating cells at an immunotoxin concentration of 10 ng/ml, with confluent HCE cells relatively unaffected. This immunotoxin may prove useful in treatment of proliferative ocular epithelial diseases such as epithelial downgrowth or squamous cell carcinoma of the ocular surface.

The in vivo properties of monovalent and divalent single-chain Fv (sFv)-based molecules with the specificity of the anti-c-erbB-2 monoclonal antibody 741F8 were examined in scid mice bearing SK-OV-3 tumor xenografts. 741F8 sFv monomers exhibited rapid, biphasic clearance from blood, while a slightly slower clearance was observed with the divalent 741F8 (sFv')2 comprising a pair of 741F8 sFv' with a C-terminal Gly4Cys joined by a disulfide bond. Following i.v. injection, the 741F8 sFv monomer was selectively retained in c-erbB-2-overexpressing SK-OV-3 tumor, with excellent tumor:normal organ ratios uniformly exceeding 10:1 by 24 h. The specificity of this effect was demonstrated by the lack of retention of the anti-digoxin 26-10 sFv monomer, as evaluated by biodistribution studies, gamma camera imaging, and cryomacroautoradiography studies. The specificity index (741F8 sFv retention/26-10 sFv retention) of 741F8 monomer binding, measured by the percentage of injected dose per g of tissue, was 13.2:1 for tumor, and 0.8 to 2.1 for all tested normal organs, with specificity indices for tumor:organ ratios ranging from 7.0 (kidneys) to 16.7 (intestines). Comparing divalent 741F8 (sFv')2 with the 26-10 (sFv')2, similar patterns emerged, with specificity indices for retention in tumor of 16.9 for the Gly4Cys-linked (sFv')2. These data demonstrate that, following their i.v. administration, both monovalent and divalent forms of 741F8 sFv are specifically retained by SK-OV-3 tumors. This antigen-specific binding, in conjunction with the 26-10 sFv controls, precludes the possibility that passive diffusion and pooling in the tumor interstitium contributes significantly to long-term tumor localization. 741F8 (sFv')2 species with peptide spacers exhibited divalent binding and increased retention in tumors as compared with 741F8 sFv monomers. Since the blood retention of the (sFv')2 is slightly more prolonged than that of the monomer, it was necessary to demonstrate that the increased tumor localization of the peptide-linked (sFv')2 was due to its divalent nature. The significantly greater localization of the divalent bismalimidohexane-linked 741F8 (sFv')2 as compared with a monovalent 741F8 Fab fragment of approximately the same size suggests that the increased avidity of the (sFv')2 is a factor in its improved tumor retention. This is the first report of successful specific in vivo targeting of tumors by divalent forms of sFv molecules. The improved retention of specific divalent (sFv')2 by tumors may have important consequences for targeted diagnostic or therapeutic strategies.

We have successfully constructed highly potent and selective anti-CD19 PAP immunotoxins using each of the three crosslinking agents, SPDP, LC-SPDP, or SMPT, to generate an intermolecular bridge between the B43 MoAb and PAP toxin moieties. These immunotoxins were selectively immunoreactive with and cytotoxic against CD19+ B-lineage ALL cells. In this report, we compared (a) in vivo chemical, immunological, and biological stability, (b) in vivo immunogenicity, and (c) in vivo anti-leukemic activity of various B43-PAP immunotoxin constructs. Our data recommend the use of SPDP and SMPT rather than LC-SPDP for generation of B43(anti-CD19)-PAP immunotoxins as clinical anti-leukemic agents. To our knowledge, this is the first comparative analysis of the in vivo pharmacokinetic features, immunogenicity, and anti-leukemic activity of anti-CD19 PAP immunotoxins that were prepared with different heterobifunctional crosslinking agents.

The effect of monensin on the cytotoxic effect of antitransferrin receptor immunotoxin (IT) was determined on cultured, human retinal pigment epithelial (hRPE) cells. Human RPE cells were treated with 0.1-10,000 ng/ml IT with and without 0.01-0.1 microM monensin, a lysosomotropic reagent that can influence IT activity. Monensin (0.01 microM) shortened the onset of cell kill with IT (10,000 ng/ml) from 48 to 24 hours (p = 0.0016). Although 0.01 microM monensin alone was not cytotoxic to hRPE cells, a single 7-day treatment with monensin caused up to a 4.1-fold increase in antiproliferative potency of IT on proliferating hRPE cells (p < or = 0.0001). Enhancement was obtained with only a 1-hour exposure to 0.1 microM monensin (p = 0.0001). In contrast, IT (0.1-10,000 ng/ml) combined with monensin (0.01 microM) had minimal effect on density-arrested cells. IT with or without monensin did not inhibit proliferation of Rhesus monkey RPE cells. Our results indicate that monensin enhances the selective cytotoxic effect of IT on proliferating hRPE cells in culture.

The ability to target proliferating cells is important for agents used to modulate wound healing by decreasing the growth of fibroblasts. Proliferating cells are known to express increased numbers of transferrin receptors and have increased receptor turnover. 454A12 Mab-rRA, an immunotoxin containing anti-human transferrin receptor monoclonal antibody conjugated to recombinant ricin A chain, was shown to inhibit the proliferation of human subconjunctival fibroblasts in vitro. A dose-related reduction of cell counts was observed in proliferating cells. More than 90% inhibition was achieved with an immunotoxin concentration of 10 ng/ml per 20,000 cells plated. In contrast, confluent fibroblasts were markedly less sensitive to the immunotoxin at equivalent concentrations. Comparative experiments demonstrated that 5-fluorouracil has less specificity for proliferating cells, with significant death of confluent fibroblasts at high drug concentrations.

Autologous bone marrow transplantation (ABMT) may aid in the management of breast cancer, but is currently limited to patients without bone marrow metastases. In earlier studies, 5 logs of malignant clonogenic breast cancer cells could be eliminated from human bone marrow using a combination of chemoseparation with 4-hydroperoxycyclophosphamide (4-HC) and immunoseparation with monoclonal antibodies and magnetic microspheres. In this report the authors compare chemoimmunoseparation to treatment with immunotoxins for elimination of tumor cells from human bone marrow and for the preservation of normal precursors. Breast cancer cells from each of five cell lines were mixed with a tenfold excess of irradiated human bone marrow cells. Treatment with a combination of five immunotoxins reduced clonogenic tumor cell growth by 1.8 to 5.5 logs depending upon the cell line. With two of the five cell lines, clonogenic tumor cells were eliminated quantitatively. Using the CAMA-1 breast cancer cell line, treatment with multiple immunotoxins was compared with chemoimmunoseparation with 4-HC, a panel of five unconjugated monoclonal antibodies and magnetic microspheres. Chemoimmunoseparation eliminated 3.5 to 5.4 logs of malignant cells, while preserving 21% of Colony-forming unit-granulocyte-macrophage (CFU-GM) and 37% of burst-forming unit-erythrocyte (BFU-E). No clonogenic breast cancer cells could be detected. Immunotoxin treatment eliminated 2.2 to 5.4 logs of clonogenic breast cancer cells, but had no effect on the bone marrow precursors. In seven of ten experiments, however, clonogenic breast cancer cells remained after immunotoxin treatment. Consequently, treatment with 4-HC, multiple murine monoclonal antibodies and magnetic microspheres provided more consistent elimination of tumor cells than separation with immunotoxins, but was significantly more toxic for marrow precursors.

An immunotoxin conjugate of a murine monoclonal antibody against human transferrin receptors and the A chain of ricin was examined for its potential to inhibit selectively the growth of retinal pigment epithelial (RPE) cells which grow in an uncontrolled manner in proliferative vitreoretinopathy. The probable efficacy of such an agent in vivo stems from the observation that actively proliferating cells possess many more transferrin receptors than normal quiescent cells. The authors showed in vitro that the immunoconjugate (454A12 MAB-rRA) inhibits protein synthesis in actively dividing RPE cells but has a smaller or no effect on protein synthesis by confluent, nondividing RPE cells. The effect was specific in that neither the free ricin A chain (rRA) nor the monoclonal antibody (454A12 MAB) alone has any inhibitory effect. Furthermore, the antibody competes with the immunotoxin and suppresses the latter's toxicity. This immunotoxin has applications for therapy in conditions in which the pathologic proliferation of RPE cells occurs.

We studied an immunotoxin consisting of recombinant ricin A chain (rRA) conjugated to 454A12 MoAb, a monoclonal antibody which recognizes an epitope on the human transferrin receptor, and compared the ability of 454A12 MoAb-rRA immunotoxin to inhibit the growth of erythroid burst-forming units (BFU-e) and myeloid colony-forming units (CFU-c) with unconjugated 454A12 MoAb. A significant reduction in BFU-e colony growth was observed at 0.001 microgram/ml of 454A12 MoAb-rRA versus 0.1 microgram/ml of unconjugated 454A12 MoAb (p = 0.005). Comparison of the effects of 454A12 MoAb-rRA and 454A12 MoAb on myeloid colony development gave markedly different results. Unconjugated antibody had no effect on CFU-c colony growth; in contrast, 0.01 microgram/ml of 454A12 MoAb-rRA reduced the number of colonies from 139 per 1 X 10(5) to 75 per 1 X 10(5) cells plated (p = 0.0005). No myeloid progenitor colonies developed at 0.1 microgram/ml of immunotoxin. These observations suggest that 454A12 MoAb-rRA inhibits growth by a potent, ricin A chain-mediated toxic effect on any proliferating cells expressing transferrin receptors, whereas the 454A12 MoAb exerts a selective inhibitory effect primarily on erythroid progenitors by perturbing the transferrin cycle. While growth factor receptors expressed on hematopoietic cells represent promising targets for immunotoxin therapy, our data indicate that an immunotoxin could inhibit cellular proliferation by a different mechanism than the corresponding unconjugated MoAb. Depending on the antibody used, these differences may be important in trials using immunotoxins for in vivo treatment or in vitro purging of malignant hematopoietic cells.

Cultured human retinal pigment epithelial cells were exposed to an immunotoxin composed of a monoclonal antibody, 454A12, directed against transferrin receptors conjugated to a toxin, recombinant ricin A chain. Exposure of proliferating human retinal pigment epithelial cells to the immunotoxin (0.1 to 10,000 ng/mL) caused a statistically significant (P less than .0001) decrease in the number of cells. This inhibitory effect was induced after an exposure to the immunotoxin as short as 5 minutes and was maximal after 24 hours of exposure. The diminution in cell number was dose dependent over the range from 0.1 to 100 ng/mL. Monoclonal antibody alone, recombinant ricin A chain alone, or an irrelevant immunotoxin, MOP21C monoclonal antibody-recombinant ricin A, did not diminish the number of cells. There was a marked decrease in DNA synthesis measured by nuclear tritiated thymidine incorporation that accompanied the immunotoxin-mediated decrease in cell number. Viable cells remaining after exposure to the immunotoxin (0.1 to 10,000 ng/mL) were morphologically abnormal; typically the cells had elongated spindle-shaped processes and had lost their normal cuboidal appearance. In contrast, cell number was not decreased in confluent human retinal pigment epithelial cells after treatment with maximal doses of immunotoxin. Morphologic changes similar to those seen in proliferating cells were observed in confluent cells exposed to more than 100 ng/mL of immunotoxin. The effect of the immunotoxin was species specific because large doses of immunotoxin did not reduce the number of viable cells in proliferating or confluent pig retinal pigment epithelial cells or cause observable morphologic changes in this cell type. Our results indicate that the immunotoxin selectively inhibited proliferating retinal pigment epithelial cells by receptor-mediated internalization of the antitransferrin receptor monoclonal antibody-recombinant ricin A chain conjugate.

Substantial heterogeneity has been observed in the expression of individual antigens within tumor cell populations. Immunotoxins which bind to different cell surface antigens might exert additive or synergistic cytotoxicity when used in combination to eliminate all clonogenic cells within a tumor. Immunotoxins have been prepared by conjugating recombinantly derived toxin A chain to different monoclonal reagents which recognize cell surface determinants of Mr 42,000 (317G5), 55,000 (260F9), and 200,000 (741F8). Each immunotoxin was evaluated for binding, internalization, and cytotoxicity with four breast cancer cell lines. Each of the three immunotoxins bound to the SKBr3 cell line and exerted antitumor activity in a limiting dilution clonogenic assay. Simultaneous treatment with two immunotoxins produced additive antitumor activity with each of the possible combinations. Additive binding could be demonstrated by immunofluorescent techniques, however, with only one of three combinations. With two of the three combinations, subpopulations of tumor cells could be identified which lacked one or the other antigenic determinant but not both. Consequently, log-additive antitumor activity was produced by immunotoxins in combination, and heterogeneity of antigenic targets may have contributed to the combined cytotoxicity in some but not all cases.

Autologous bone marrow transplantation may contribute to the treatment of several types of lymphoreticular malignancies. Recent studies have suggested that a combination of chemoseparation and immunoseparation may be more effective than either modality alone in eliminating malignant cells from human bone marrow. In this report an immunotoxin has been prepared by conjugating pokeweed antiviral protein (PAP) to the 3A1 murine monoclonal antibody that recognizes a 40 kD (CD7) determinant expressed by most T cell acute lymphoblastic leukemias and a majority of normal mature peripheral T cells. When HSB-2 T lymphoma cells were mixed with normal human bone marrow and incubated with 3A1-PAP and 100 microM chloroquine, approximately 3 logs of clonogenic T cells could be eliminated from a 20-fold excess of bone marrow. Treatment of cell mixtures with 2'deoxycoformycin (2'-dCF) and deoxyadenosine (dAdo) eliminated 2 logs of clonogenic tumor cells. The use of 3A1-PAP and chloroquine with dCF/dAdo was more effective than either single modality, eliminating up to 6 logs of HSB-2 tumor cells in optimal experiments. Anti-tumor activity of the combined treatment extended to T leukemia cells taken directly from patients. Although 3A1-PAP reduced CFU-GM by only 13% and BFU-E by 36%, the addition of 2'-dCF and dAdo was more toxic for normal marrow precursors, further reducing CFU-GM, GEMM and BFU-E as well as preventing recovery of CFU-GM in long-term bone marrow culture.(ABSTRACT TRUNCATED AT 250 WORDS)

Pokeweed antiviral proteins enzymatically inactivate the 60 S subunit of eucaryotic ribosomes in cell-free preparations. Three different species of the enzyme can be isolated from spring leaves, summer leaves, and seeds of pokeweed. Sequence analyses of the NH2-terminal residues show that pokeweed antiviral protein, isolated from spring leaves and seeds, are homologous and differ in 11 of the 28 residues compared. Ricin contains a polypeptide (ricin A chain) that has functional similarities to pokeweed antiviral protein, yet the sequences of the pokeweed proteins show little similarity with ricin A chain. Ricin B chain is responsible for helping ricin A chain across the plasma membrane; since pokeweed antiviral has no counterpart to ricin B chain, it is not nearly as cytotoxic as ricin. However, when pokeweed antiviral protein was covalently coupled to ricin B chain, a cytotoxic species was formed. Pokeweed antiviral protein fails to interact noncovalently with the ricin B chain to produce a cytotoxic species equivalent in function to ricin.

Immunotoxins directed against human transferrin receptor have been evaluated in a nude mouse model of human malignant mesothelioma. Immunotoxins were constructed by linking ricin A chain to murine monoclonal antibodies reactive with the human transferrin receptor. A chain was obtained either by isolation from the parent toxin or by recombinant DNA techniques. These immunotoxins acted as potent in vitro cytotoxins against human malignant mesothelioma cells (H-MESO-1) (ID50, 2 X 10(-9) M). Cytotoxic potency and kinetics of cell kill were potentiated in vitro by the carboxylic ionophore monensin. For in vivo trials, nude mice were injected i.p. with 6-9 X 10(6) human malignant mesothelioma cells 24 h prior to the start of i.p. immunotoxin treatments. The survival of tumor-bearing mice was extended by 149-404%, representing a probable cell kill of 2-4 logs. Specificity of this antitransferrin receptor immunotoxin response was confirmed by the ineffectiveness of irrelevant control immunotoxins and blockade of specific immunotoxin action by excess free antibody. Monensin showed limited in vivo potentiation of immunotoxin effect, but a derivative formed by esterification of monensin with linoleic acid gave improved survival times over treatment with immunotoxin alone. Immunotoxins constructed with ricin A chain have significant tumoricidal activity in this model of regional antitumor therapy. These results may have direct relevance for treatment of i.p. malignancy in clinical settings.

Ricin A chain and pokeweed antiviral protein (PAP), two enzymes that inhibit the action of eukaryotic ribosomes, were coupled by cleavable, N-succinimidyl-3-(2-pyridyldithio)propionate, and noncleavable m-maleimidobenzoyl-N-hydroxysuccinimide ester, cross-linking reagents to monoclonal antibodies directed against Thy 1.1 antigen. Leukemia cells that contained Thy 1.1 antigen were selectively killed compared to Thy 1.2-containing cells. The composition of the conjugates was determined by radioimmunoassay, and most of the immunotoxins contained about equal molar quantities of antibody and ribosomal inhibitor. Ricin A chain linked to antibody by a noncleavable m-maleimidobenzoyl-N-hydroxysuccinimide ester cross-link was not cytotoxic, but PAP coupled to the same antibody was. Both immunotoxins linked by a cleavable disulfide bond were cytotoxic. Disulfide-linked F(ab')2-PAP was cytotoxic, but it was about 45 times less efficient than disulfide-linked IgG-PAP. There was only a 3.2-fold difference in their ability to inhibit ribosomes in vitro. The relative difference between in vitro action and cytotoxicity could be accounted for by differences in the affinity of the immunotoxins to the cell surface. Neither PAP or ricin A chain disulfide linked to a monoclonal antibody against Mr 15,000 envelope protein, a murine leukemia virus coat protein, were not cytotoxic, although both conjugates bound to the cell surface. Because of its stability, ease of purification, and lack of an analogue of ricin B chain, PAP may be more useful than ricin A for immunotoxin synthesis.

The therapeutic efficacy of whole ricin, or recombinant ricin A chain, coupled to a monoclonal antibody that reacts with the idiotype of the surface IgM expressed on guinea pig L2C lymphoblasts, was assessed. In vitro studies were done to characterize the immunotoxins (IT) and to demonstrate their specificity before use in vivo. The concentration of whole ricin IT (M6-Ricin) that inhibited protein synthesis by 50% (IC50) in L2C cells was 1.4 X 10(-9) M, in a 5-hr assay, in the presence of lactose to block non-antibody-directed toxicity. M6-Ricin did not inhibit protein synthesis in two control guinea pig cell lines that did not express the idiotype, nor did a whole ricin IT prepared with an isotype-matched monoclonal antibody of irrelevant specificity inhibit protein synthesis in L2C cells. Two recombinant ricin A chain IT, which differed from one another by a factor of 2 to 3 in the number of A chains conjugated per antibody molecule, were less effective in vitro than M6-Ricin (IC50 of greater than 5 X 10(-8) M). For in vivo experiments, the IT were given by the i.p. route 24 hr after the i.p. inoculation of 1 X 10(5) L2C cells. The highest doses of M6-Ricin and M6-Ricin A chain IT tested, 30 micrograms/kg and 3000 micrograms/kg, respectively, were within fourfold to fivefold of their maximum tolerated doses; no deaths or ill effects due to ricin toxicity were noted. These doses increased the median survival time of L2C-bearing guinea pigs to 31 to 34 days, compared with 15 days for untreated animals. This magnitude of increase in survival indicates that 99.999% (5 logs) of injected tumor cells were eliminated, thus accounting for the 12% long-term survival rate obtained. Median survival times for guinea pigs treated with 30 micrograms/kg of the A chain IT were 18 and 21 days for the two conjugates tested, and the median survival for guinea pigs treated with 3000 micrograms/kg of unconjugated antibody was 18 days. Our data demonstrate that recombinant A chain IT are active in vivo and that the B chain of ricin can potentiate IT activity in vivo. Although the potency differs by 100-fold, the therapeutic index of the intact ricin IT is similar to that of the ricin A chain IT.

Neuronal microtubules in a variety of nerve cell types are unaffected by high hydrostatic pressures over a range of 1400-10,000 pounds/inch(2) and periods of 10-45 min. Similarly, purified tubulin polymerized to form microtubules in vitro were not depolymerized by the same range of pressures. The depolymerization of microtubules in several types of non-neuronal cells, which has been reported, may have been over-generalized with regard to the direct action of pressure on microtubule stability.

mAb directed toward the idiotype of the 38C13 murine B cell lymphoma can be used to treat and cure a high percentage of mice challenged previously with an otherwise lethal dose of tumor cells. Tumors developing in animals despite antibody therapy were examined by immunofluorescence and found to demonstrate either loss of surface Ig, or expression of an altered idiotype that no longer bound the antibody used for treatment. Further immunofluorescence analysis of the variant tumors revealed individual patterns of cross-reactivity with anti-38C13 idiotype mAb other than that used for therapy. The variant tumor cells were fused to myeloma cells and hybrids were isolated which secreted large quantities of the altered idiotype proteins. Polyclonal antibodies and mAb prepared against the mutant proteins demonstrated cross-reactivity with the original 38C13 protein and its other variants. But the variants and wild type cells could be distinguished from each other by their patterns of reactivity with the panels of anti-idiotype antibodies. Differences in apparent m.w. were demonstrated in the L chains of each of the mutant proteins. Southern blot analysis of the H chain locus of these mutants established that they were all clonally related; however, the L chain loci were grossly different. Thus, rare cells with alteration in their Ig L chain genes and expressed proteins can give rise to idiotype variants in this B cell tumor.

A radioimmunoassay for ricin and ricin A- and B-chains was developed. Amounts as low as 100 pg of A-chain and 500 pg of B-chain could easily be quantitated. We showed, however, that the free chains were more reactive in the radioimmunoassay than the equivalent quantity of the individual chains when combined in intact ricin. The usefulness of the assay was demonstrated by determining the concentration of contaminating A- or B-chains in preparations of the separate polypeptides purified by DEAE Bio-Gel A chromatography and by chromatofocusing.

The binding of 4-methylumbelliferyl galactose and 4-methylumbelliferyl N-acetylgalactosamine to ricin, Ricinus communis agglutinin, and ricin B chain was studied by fluorescence polarization and equilibrium dialysis. The binding of [3H]galactose to ricin was also studied by equilibrium dialysis. The results were consistent with binding of 2 mol of ligand to ricin (which contains 1 A chain and 1 B chain) and ricin B chain and 4 mol of ligand to the agglutinin (which contains 2 A chains and 2 B chains). There was no evidence for interaction between the 2 sites on any of the B chains.

4-Azidophenyl-beta-D-galactopyranoside and 4-diazophenyl-beta-D-galactopyranoside were used to inactivate ricin. Galactose, but not glucose, protected against inactivation as measured by the retention of the ability of ricin to bind to Bio-Gel A, a galactose-containing gel. Nearly complete inhibition of binding to Bio-Gel A, to monosaccharides, or to cell surface receptors could be achieved by reaction of ricin with either label, but neither label impaired the ability of the A chain to inhibit translation in vitro. The diazonium salt-modified ricin still inhibited cellular protein synthesis, but ricin modified by the photoactivated label was 280 times less efficient than ricin in inhibition of cellular protein synthesis.

Studies were performed to evaluate the anti-tumor activity of mafosfamid, a new synthetic derivative of cyclophosphamide. We tested its ability to eliminate lymphoblastic leukemia cells from autologous bone marrow grafts following a 30 min preincubation in a highly sensitive clonogenic assay. Treatment with 50-100 micrograms mafosfamid/ml eliminated more than 4 logs of contaminating clonogenic tumor cells from a 200-fold excess of normal bone marrow. Flow cytometric studies showed differences in cell cycle kinetics between mafosfamid-resistant and mafosfamid-susceptible tumor cell clones. Compared to drug susceptible clonogenic tumor cells, clones that resisted treatment with 100 micrograms mafosfamid/ml exhibited a smaller percentage of cells in S-phase, indicating that mafosfamid is mostly cytotoxic to rapidly cycling tumor cells. The combination of mafosfamid and a target cell selective immunotoxin containing pokeweed anti-viral protein was superior to mafosfamid alone or immunotoxin alone for purging mafosfamid-resistant leukemic cells from human marrow.

Two classes of ricin cDNA clones have been identified and sequenced. The cDNA clone pBL-1 closely matches in nucleotide sequence the ricin genomic clone pAKG previously described by Halling et al., 1985 (Nucl. Acids Res. 13:8019). A second group of cDNA clones, represented by pBL-3, encode a hybrid protein (ricin E), having a ricin-like A chain and N-terminal half of the B chain and an RCA (Ricinus communis agglutinin)-like C-terminal half of the B chain.

Ricin, was nitrated with tetranitromethane and reduced with sodium dithionite. Of the 8.0 nitro groups incorporated, 3.2 were on the A chain and 5.1 were on the B chain. Nitrated ricin1 was somewhat less active than nitrated and reduced ricin1 in inhibiting protein synthesis in vitro, but both were highly inhibitory. However, the modified toxins were less than 1% as active as ricin in inhibiting protein synthesis in cultured cells. Indirect immunofluorescence assays demonstrated tha both modified toxins were specifically bound to the cell surface and could be displaced by galactose.

The sensitivity of three human colon adenocarcinoma cell lines (LoVo, LS174T, and SW1116) and a human pancreatic adenocarcinoma cell line (Hs766T) to a recombinant ricin A chain-antitransferrin receptor immunotoxin was studied. In addition, the carboxylic ionophore monensin was used in conjunction with the immunotoxin to determine the possibility of increased cytotoxicity without loss of specificity. The immunotoxin, 454A12-rRTA, is composed of the monoclonal antibody 454A12 directed against transferrin receptor and of ricin A chain, which was produced by recombinant DNA techniques. In 18 h dose-response cytotoxicity assays, the median inhibitory dose (ID50) against LoVo, LS174T, and SW1116 was found to be 3 X 10(-10), 3.6 X 10(-11), and 3.6 X 10(-10) M, respectively; in the same assay, the ID50 for Hs766T was found to be 4 X 10(-10) M. In the presence of monensin, the ID50 for the adenocarcinoma cell lines was reduced 9-fold, 28-fold, and 5-fold, respectively. In cytotoxic kinetic assays, 50% of control protein inhibition was reached in immunotoxin-treated LS174T cells 12-fold faster in the presence of monensin than in its absence. Immunotoxin-treated LoVo cells reached 50% inhibition of control protein synthesis fivefold faster in the presence of monensin than in its absence. Furthermore, no toxicity of immunotoxin or potentiation by monensin was observed in either a control cell line (Swiss albino mouse 3T6) treated with specific immunotoxin or with a control immunotoxin assay. These results show the in vitro specificity and selectivity of 454A12-rRTA immunotoxin for human gastrointestinal and pancreatic cancer cell lines.

Ricin A chain was radioactively labeled using reductive alkylation, lactoperoxidase catalyzed iodination, and reaction with iodoacetamide or N-ethylmaleimide (NEM). The inhibition of cell-free rat liver protein synthesis by the modified A chains and the ribosome binding characteristics of each of the labeled derivatives was examined. [3H] NEW was found to quantitatively react with the A chain sulfhydryl group normally involved in a disulfide bond with the B chain in intact ricin. Labeling the protein with [3H] NEM had no effect on the in vitro inhibition of protein synthesis by the A chain. [3H] NEM-labeled A chain binds to rat liver ribosomes in a manner which is dependent on the concentrations of NaCl and Mg2+. At optimal Mg2+ concentration (5.5 mM), A chain binding to ribosomes is saturable and fully reversible either by dilution of the reaction mixture or by addition of unlabeled A chain. At 5.5 mM Mg2+, A chain was found to bind to a single site on rat liver ribosomes with a dissociation constant of 6.2 x 10(-8) M. [3H] NEM-labeled A chain did not bind to isolated 40S ribosomal subunits and bound to 60S ribosomal subunits with a 1 : 1 molar stoichiometry and a dissociation constant of 2.2 x 10(-7) M. The relationship between ribosome binding and A chain inhibition of eucaryotic protein synthesis is discussed.

Ricinus communis agglutinin dissociated to lower molecular weight forms when heated in sodium dodecyl sulfate in the absence of reducing agents, while ricin was little affected by such treatment. The data suggest that strong noncovalent bonds hold together two A-B heterodimers in the Ricinus communis agglutinin tetramer. Protease inhibitors such as diisopropylfluorophosphate, phenylmethansefulonyl fluoride, and EDTA, did not prevent the sodium dodecyl sulfate-heat induced dissociation; however, sulfhydryl specific reagents (N-ethylmaleimide, 5,5'-dithiobis (2-nitrobenzoic acid) and p-chloromercuribenzoate) were effective. Titration of the lectins in sodium dodecyl sulfate indicated that ricin contains one sulfhydryl and Ricinus communis agglutinin four sulfhydryl groups, none of which react in the presence of 8 M urea. The sulfhydryl groups that could be titrated in the intact proteins in sodium dodecyl sulfate were on the A chains.