It has been shown that a number of peptides linked to the N-terminal end of the diphtheria toxin A-fragment are translocated to the cytosol (Stenmark etal., 1991, Stenmark etal., 1992; Ariansen etal., 1993). It is an interesting question to what extent such peptides can subsequently be presented by class I major histocompatibility antigens. If this is the case, fusion proteins of viral or cancer-related peptides and enzymatically inactive mutants of diphtheria toxin could be used for vaccination purposes to expand desired populations of cytotoxic CD 8+ T-lymphocytes.
The Clostridium botulinum exoenzyme C3 is able to ADP-ribosylate the small G-protein, rho, in a cell free system, but is unable to enter cells (except at very high concentrations) because it lacks a B-moiety. Constructs where exoenzyme C3 was linked to the N-terminal end of diphtheria toxin or to toxin where the N-terminal part of the Afragment had been deleted, were able to translocate the exoenzyme-containing fusion protein into the cytosol and interfere with the organization of actin filaments (Aullo et a/., 1993).
Externally added aFGF (acidic fibroblast growth factor) contained in a fusion protein with diphtheria toxin A-fragment is able to stimulate DNA synthesis in serum-starved cells without a measurable increase in tyrosine phosphorylation (Wiedlocha et ai, 1994; Wied-locha et a/., 1996). The DNA synthesis is only stimulated when the fusion protein is translocated to the cytosol (and subsequently to the nucleus), but not when the translocation is prevented by the presence of heparin or by introduction of disulfide bridges into the toxin A-fragment to prevent unfolding of the protein. A deletion mutant of the growth factor lacking an N-terminal putative nuclear localization sequence was also unable to stimulate DNA synthesis, even though it was efficiently translocated to the cytosol. It is a tempting interpretation that accumulation of the growth factor in the nucleus is required to stimulate DNA synthesis (Imamura et ai, 1990), but the possibility must also be kept in mind that deletion of the N-terminal region could induce a structural perturbation. Thus, the sensitivity to trypsin and pronase of the deletion mutant measured in the presence of heparin was clearly higher than that of full length aFGF. Once translocated into the cells, fusion protein with growth factor containing the nuclear localization signal was still detectable after 48 h. On the other hand, AaFGF disappeared completely from the cells upon incubation for 24 h after removal of AaFGF from the medium. It is not clear if it is degraded in the cells, or if growth factor that is not retained in the nucleus is transported out of the cells using the same mechanism by which aFGF is exported from cells that produce this growth factor.
Dihydrofolate reductase has been used extensively in translocation experiments. A fusion protein with diphtheria toxin A-fragment was shown to be translocated to the cytosol (Klingenberg and Ols-nes, 1996). The translocation was inhibited by methotrexate, which induces a tight folding of the protein. A fusion with a mutated dihydrofolate reductase that does not bind methotrexate tightly was also translocated, and in this case methotrexate was not able to prevent the translocation. This indicates that not only must the toxin Afragment be unfolded, but the passenger protein must also be able to unfold for translocation to occur.
The bacterial RNase, barnase, linked to Pseudomonas aeruginosa exotoxin A or to a non-toxic deletion mutant of the toxin lacking the enzymatic domain, was found to be toxic to cells to a greater extent than either component alone (Prior et a/., 1991; 1992). A C-terminal KDEL sequence was required for toxicity. Cells resistant to the intracellular action of the toxin were also sensitive to the fusion protein. This suggests that the fusion protein is translocated to the cytosol, although direct evidence for this was not provided.
Fusion proteins have been constructed from peptide epitopes from influenza A antigens and the binding and translocation domains of Pseudomonas exotoxin A (Donelly eta/., 1993). When target cells were incubated with these fusion proteins, and subsequently exposed to cytotoxic T lymphocytes (CTLs) specific for the relevant epitopes, a CTL mediate lysis of the target cells was observed. These experiments suggest that the translocation machinery supplied by protein toxins may be useful tools for bringing peptides into cells for presentation via the major histocompatibility class I (MHC I) system. It should be noted that no direct evidence was provided that the translocation occurred by the toxin pathway, and it cannot be excluded that the toxin was only instrumental in accumulating the peptide on the surface of the ceils and in the endocytic pathway.
The N-terminal part (residues 1-254) of the lethal factor (LF) component of anthrax toxin is nontoxic, but its binding to the protective antigen (PA) is retained (Arora and Leppla, 1993). When the enzymatic domains of diphtheria toxin (Arora and Leppla, 1994; Milne et a/., 1995), Pseudomonas exotoxin A (Arora and Leppla, 1994), or shiga toxin (Arora and Leppla, 1994) were fused to this fragment of LF, the resulting fusion proteins were highly toxic to cells in the presence of PA. These results suggest that the anthrax toxin translocation machinery may be useful for bringing foreign peptides and proteins into cells, but more work remains to be done to establish the efficiency of the translocation process.
Was this article helpful?