Over the past decade there has been considerable interest in the assembly of conjugate toxin molecules whose action is directed toward specific target cells. A wide variety of cell surface-directed ligands (e.g., mAbs, lectins, polypeptide hormones, growth factors) have been coupled through disulfide linkage to nontoxic fragments of plant and/or microbial toxins (reviewed in 1, 2) . In particular, mAbs have been used to direct the cytotoxic action of ricin, both the A chain and intact toxin, and fragments of diphtheria toxin to specific antigenbearing cells . This class of conjugate toxins are widely known as immunotoxins, and have been the subject of several recent reviews (3-6).In contrast to chemical cross-linking to form conjugate toxins, our approach to the development of targeted toxins has involved genetic assembly of chimeric toxin genes that are expressed from recombinant strains of Escherichia coli K12 (6, 7). We have selected targeting ligands whose cell surface receptors are known to undergo obligatory receptor-mediated endocytosis in the design of chimeric toxins that are composed of portions of diphtheria toxin linked to peptide hormones and/or growth factors. We have used rDNA methodologies to construct fusion genes in which that portion of the diphtheria tox gene which encodes the receptor-binding domain has been genetically replaced with the cDNA encoding the ligand of choice . Murphy et al. (6) and Williams et al. (7) have described the genetic construction of fusion toxins in which a-melanocytestimulating hormone (a-MSH)' and IL-2 have served as the receptor-binding component of the hybrid . Since both peptides have been shown to bind to surface receptors and are internalized by receptor-mediated endocytosis (8-10), we reasoned that both a-MSH toxin and IL-2 toxin should be selectively toxic for a-MSH and high affinity IL-2-R-bearing cells, respectively . Indeed, the action

No abstract

The receptor binding requirements for entry of the NAD+ ADP-ribosyltransferase component of DAB486-IL 2 into target cells were examined. Experiments utilizing cell lines bearing either high-affinity or individual subunits of the interleukin 2 receptor (IL 2R) as well as human peripheral blood mononuclear cells with natural killer activity demonstrate that the high-affinity receptor facilitates delivery of fragment A from DAB486-IL 2 to the cytosol approximately 1000 times more efficiently than either the intermediate-(p75) or low-affinity (p55) forms of the IL 2R. We show that elongation factor 2 (EF-2) in these cells is not quantitatively or qualitatively altered indicating that the relative resistance to intoxication displayed by IL 2R variant cell lines cannot be attributed to an altered intracellular target of the hybrid toxin. We also demonstrate that an alteration in the binding of DAB486-IL 2 to the p75 subunit of the IL 2R may account for the selective cytotoxicity of DAB486-IL 2 for cells bearing the heterodimeric high-affinity IL 2R.

A continuous cell line was established from the blood of a patient (HH) with an aggressive cutaneous T-cell leukemia/lymphoma who lacked antibodies to human T lymphotrophic virus, type I. The immunophenotype of the cultured cells was CD2+, CD3+, CD4+, CD5+, CD8-, DR+ and CD25- (Tac, IL-2 receptor alpha chain). Southern-blot hybridization analysis of T-cell-receptor beta chain DNA demonstrated the same rearrangement in freshly isolated blood cells and cultured cells, indicating that the cell line was derived from the patient's malignant clone. Since cultured T-cells grew in complete medium without added IL-2, we investigated whether HH cells could be producing and responding to IL-2 in an autocrine fashion. However, no IL-2 was detectable in supernatant from the cell line, while antibodies to IL-2, or to the IL-2 receptor alpha or beta chains did not inhibit cell growth. In addition, no mRNA message for IL-2 was detectable in these cells. The results appear to exclude an autocrine IL-2-dependent mechanism of cell growth for this T-cell line. Although cultured HH cells lacked detectable IL-2 receptor alpha chain, they did show increased proliferation to exogenous IL-2. Binding studies with 125I-IL-2 demonstrated an intermediate affinity receptor for IL-2, KD = 1.7 nM, with 6400 binding sites per cell, suggesting the presence of an IL-2 receptor beta chain. Consistent with these findings 125I-IL-2 cross-linking studies demonstrated a single receptor calculated to be 75 kDa. Also, the beta chain of the IL-2 receptor was detected by immunofluorescence using specific monoclonal antibodies (MAbs). Nanomolar concentrations of an IL-2-diphtheria toxin fusion protein inhibited cellular protein synthesis, an effect abrogated by native IL-2. These findings indicate that the IL-2 receptor beta-chain was functional. This novel mature T-cell line may be useful in studies of IL-2 receptor regulation and in analysis of the mechanism of T-cell leukemogenesis.

The interleukin 2-diphtheria toxin-related fusion protein (IL-2-toxin) rapidly inhibits protein synthesis in IL-2 receptor (IL-2R)-bearing phytohemagglutinin-activated T cells but transiently stimulates DNA synthesis. At 7 hr after interaction with IL-2R+ phytohemagglutinin-activated T cells, IL-2-toxin-treated cells bear augmented steady-state levels of c-myc, interferon y, and IL-2R mRNA; these effects are indistinguishable from those produced by recombinant IL-2. Amplification of IL-2 sequences by the polymerase chain reaction reveals an increased level of IL-2 mRNA in cell cultures treated with recombinant IL-2, IL-2-toxin, and cycloheximide. These results suggest that IL-2-toxin can affect de novo IL-2 gene transcription/mRNA stabilization through independent mechanisms exerted by both the IL-2R binding domain and ADP-ribosyltransferase activity of the fusion protein. After 20 hr of culture, IL-2R mRNA was markedly decreased in both IL-2-toxin-and cycloheximide-treated phytohemagglutinin-activated T cells. Although interaction of IL-2-toxin with IL-2R+ T cells initially mimics the stimulatory effects of IL-2 upon c-myc, interferon y, IL-2R, and IL-2 gene expression, the consequences of inhibition of protein synthesis mediated by the ADP-ribosyltransferase activity of the toxin dominate after 7 hr and are indistinguishable from those effects mediated by cycloheximide.A fusion gene encoding the interleukin 2-diphtheria toxin fusion protein (IL-2-toxin) was constructed from a truncated diphtheria toxin gene by replacing DNA sequences coding for the toxin receptor binding domain with sequences coding for amino acids 2-133 of human IL-2 (1). The mature form of IL-2-toxin has a deduced molecular mass of 68 kDa, and the fusion protein retains at least some functional attributes of both its diphtheria toxin and IL-2 components. The fusion protein (i) binds with high affinity to the multimeric IL-2 receptor (IL-2R), (ii) is internalized by IL-2R-mediated endocytosis, and (iii) mediates the ADP-ribosylation of elongation factor 2 in the target-cell cytosol (2).Binding of IL-2 to its high-affinity receptor rapidly activates intracellular processes that lead to gene activation and de novo gene transcription, DNA synthesis, and cell cycle progression (3)(4)(5)(6). Even though IL-2-toxin rapidly inhibits protein synthesis in IL-2R-bearing phytohemagglutinin (PHA)-activated T cells, we now note that this chimeric toxin also transiently stimulates DNA synthesis (e.g., thymidine incorporation).In the present study we demonstrate that binding of the IL-2-toxin to the high-affinity IL-2R initially produces effects in the PHA-activated target cell that are indistinguishable from those mediated by IL-2 itself. Subsequently, the fusion protein inhibits cellular protein synthesis. Additional indirect effects of IL-2-toxin upon T-cell gene activation, produced by the inhibition of protein synthesis, are characterized herein. MATERIALS AND METHODSCell Cultures. Human peripheral blood mononuclear cells were isolated by means...