Translocation of the Toxin

The next step in the entry process is endocytic uptake of the bound toxin. The uptake is comparatively slow and appears to occur from clathrin-coated pits (Morris ef al., 1985). Once in endosomes, the toxin is exposed to acidic conditions a pH below 5.3 is necessary for the penetration to the cytosol (Sandvig and Olsnes, 1981). If acidification of the endosomes is prevented by exposure of the cells to NH4CI or monensin, the cells are protected from poisoning. However, the protection can be overcome if the cells are exposed to buffer with acidic pH (Sandvig and Olsnes, 1981). Under these conditions toxin bound at the cell surface penetrates directly through the surface membrane (Moskaug etal., 1988). 

---

Werner ED, Brodsky JL, McCracken AA (1996) Proteasome-dependent endoplasmic reticulum-associated protein degradation an unconventional route to a familiar fate. Proc Natl Acad Sd USA 93 13797-13801 Wesche J, Rapak A, Olsnes S (1999) Dependence of ridn toxicity on translocation of the toxin A-chain fi om the endoplasmic reticulum to the cytosol. J Biol Chem 274 34443-34449... 

Pertussis toxin consists of two components, the enzymatically active A protomer consisting of a single polypeptide, and the penta-meric B oligomer (Tamura et ai, 1982). The B component is involved in binding to the surface of eukaryotic target cells, and presumably in translocation of the toxin across the plasma membrane. Once inside the cell, the enzymatically active A component needs to undergo activation that depends on reduced gluthathione this effect can be... 

Lanzrein M, Sand O, Olsnes S (1996) GPI-anchored diphtheria toxin receptor allows membrane translocation of the toxin without ion channel activity. EMBOJ... 

Wesche, J., Rapak, A. and Olsnes, S. (1999) Dependence of ricin toxicity on translocation of the toxin A-chain... 

Figure 1 The mode of action for bacterial AB-type exotoxins. AB-toxins are enzymes that modify specific substrate molecules in the cytosol of eukaryotic cells. Besides the enzyme domain (A-domain), AB-toxins have a binding/translocation domain (B-domain) that specifically interacts with a cell-surface receptor and facilitates internalization of the toxin into cellular transport vesicles, such as endosomes. In many cases, the B-domain mediates translocation of the A-domain into the cytosol by pore formation in cellular membranes. By following receptor-mediated endocytosis, AB-type toxins exploit normal vesicle traffic pathways into cells. One type of toxin escapes from early acidified endosomes (EE) into the cytosol, thus they are referred to as short-trip-toxins . In contrast, the long-trip-toxins take a retrograde route from early endosomes (EE) through late endosomes (LE), trans-Golgi network (TGN), and Golgi apparatus into the endoplasmic reticulum (ER) from where the A-domains translocate into the cytosol to modify specific substrates.

C. botulinum toxins belong to the AB group of toxins, which also includes diphtheria toxin, pseudomonas exotoxin A, anthrax toxin, Shiga(like) toxin, cholera toxin, pertussis toxin, and plant toxins, e.g., ricin. Moiety A has an enzymatic activity and usually modified cellular-target entering cytosol. Moiety B consists of one or more components and binds the toxin to surface receptors, and is responsible for translocation of the A component into cells. AB toxins are produced in a non-active form and are activated by a split between two cysteine residues within a region (Falnes and Sandvig, 2000). 

The modified elongation factor reacts normally with GTP, but the complex so formed is unable to participate in translocation. A concentration of the toxin in the cytoplasm of 10 8 M is sufficient to promote the fatal reaction. The reaction with diphthamide parallels that of cholera toxin (Box 11-A). 

The enzymatic specificity of diphtheria toxin deserves special comment. The toxin ADP-ribosylates EF-2 in all eukaryotic cells in vitro whether or not they are sensitive to the toxin in vivo, but it does not modify any other protein, including the bacterial counterpart of EF-2. This narrow enzymatic specificity has called attention to an unusual posttranslational derivative of histidine, diphthamide, that occurs in EF-2 at the site of ADP-ribosylation (see fig. 1). Although the unique occurrence of diphthamide in EF-2 explains the specificity of the toxin, it raises questions about the functional significance of this modification in translocation. Interestingly, some mutants of eukaryotic cells selected for toxin resistance lack one of several enzymes necessary for the posttranslational synthesis of diphthamide in EF-2 that is necessary for toxin recognition, but these cells seem perfectly competent in protein synthesis. Thus, the raison d etre of diphthamide, as well as the biological origin of the toxin that modifies it, remains a mystery. 

To further address the signaling pathways involved in opioid-induced cardioprotection, Schultz et al. [55] determined the involvement of a Gi/o protein in mediating delta -induced cardioprotection produced by the selective nonpeptide delta opioid agonist, TAN-67. Pretreatment with pertussis toxin for 48 h prior to TAN-67 administration completely blocked its cardioprotective elfect as well as that to IPC, suggesting that a Gi/o protein is intimately involved in the cardioprotection produced by these two interventions. Subsequently, Miki et al. [56] found that morphine produced a cardioprotective elfect in isolated rabbit hearts which was blocked by pretreatment with chelerythrine, a protein kinase C (PKC) inhibitor at a concentration that had no elfect on infarct size in the absence of morphine. More recently, Fryer et al. [57] extended these findings to the intact rat heart and showed that the protective elfect of TAN-67 to reduce infarct size was blocked by chelerythrine and GF 109203X, two selective PKC inhibitors with different binding sites, and that TAN-67 produced a selective translocation of the PKC-delta isoform to the mitochondria. 

Unlike diphtheria toxin, little is known about the structures required for the translocation of the enzymatic subunit of PT. In diphtheria toxin and Pseudomonas aeruginosa exotoxin A, the B moiety can be clearly subdivided into two distinct domains, one responsible for receptor binding, composed essentially of (3 sheets, and one responsible for translocation of the A subunits, essentially composed of a helices (Allured etal., 1986 Choe etal., 1992). There is no clear translocation domain in PT, and much less is known about the internalization step of PT, compared to diphtheria toxin and exotoxin A. 

It is a general property of internalized toxins and certain viruses that they form channels in membranes. The portion of the toxin molecule associated with channel formation is typically the same as that needed for internalization. This could mean that channel formation is the mechanism that underlies translocation, or that it is an epiphe-nomenon that occurs coincidentally with translocation. In either case,... 

Since the L chain of TeTx and BoNTs is responsible for the cytosolic activity of the CNTs, at least this domain of the toxin molecule must reach the cytosol. Pharmacological and morphologic evidence indicates that the CNTs enter the cell by endocytosis (Black and Dolly, 1986 b) and that TeTx and BoNTs have to pass through a low pH step for neuron intoxication to occur (Williamson and Neale, 1992 Simpson et at., 1994). Acidic pH does not activate the toxin directly via a structural change, since the direct introduction of the L chain in the neutral pH environment of the cytosol is sufficient to block exocytosis (Penner et at., 1986 Anhert-Hilger et al., 1989 b Bittner et al., 1989 a, b Mochida et al., 1989 Weller et al., 1991). Hence, low pH is necessary for the process of L chain membrane translocation from the vesicle lumen into the cytosol, by analogy with the other bacterial protein toxins with a three-domain structure (Montecucco et al., 1994). 

Reduction of the toxin and translocation of the L chain into the cytoplasm To gain access to the cytoplasm, the L chain needs to cross the membrane of the endocytic compartment. For this translocation, the H chain is required, probably by forming a proteinaceous translocation complex in the membrane that exposes (and possibly releases) the L chain to the cytoplasm. In the reductive intracellular environment, the disulfide bond linking the H and L chains is reduced (Kistner and Habermann, 1992). 

Structural homologies between PFTs and other toxins have not been identified. However, the process of membrane permeabilization may be operative in many cases where proteins have to escape from an intracellular compartment. Well known examples are diphtheria toxin, the neurotoxins and anthrax toxin. Specific domains in many intracel-lularly active toxins have in fact been shown to produce pores in artificial lipid bilayers, and membrane permeabilization is thought to form the basis for translocation of the active moieties from the late endo-some to the cytoplasm (reviewed in Montecucco et ai, 1994). The molecular mechanism of this translocation remains obscure. In the... 

Like diphtheria toxin. Pseudomonas aeruginosa exotoxin A requires low pH to act (FitzGerald ef al., 1980). In spite of this, it has not been possible to induce translocation of Pseudomonas toxin across the surface membrane by exposure to low pH. It appears that the toxin must be transported beyond the endosomes, possibly to the trans-Golgi network or even to the endoplasmic reticulum to find conditions required for translocation (Chaudhary et al., 1990). In fact domain III ends with an amino acid sequence that (after removal of a terminal... 

The main limitations of the translocation system described here are that (a) many passenger proteins interfere with the structure of the toxin so that it cannot recognize the toxin receptor, and that (b) the passenger protein may not be able to unfold. We have tested a large number of proteins, and most of them are not translocated due to one of these problems (Klingenberg and Olsnes, 1996). This may also be a problem with smaller peptides, but usually less than with protein passengers. Conceivably, modification of the fusion proteins with polypeptide linkers and by mutations in the passenger protein may overcome the problem, but this will require a considerable amount of work in each case. 

The question whether toxins can be used to translocate nucleic acids into cells is very interesting, but this has not been tested seriously up to now. Toxins are interesting candidates for delivery of anti-sense RNA, ribozymes and genes for transformation and gene therapy. So far there is not convincing evidence that toxins have been effective for such purposes. From what is now known about the translocation of diphtheria toxin, it is likely that the nucleic acid would have to be linked end-to-end to the N-terminus of the A-fragment of this... 

Falnes P0, Olsnes S (1995) Cell-mediated reduction and incomplete membrane translocation of diphtheria toxin mutants with internal disulfides in the A-fragment. J Biol Chem 270 20787-20793. 

Papini E, Sandona D, Rappuoli R, Montecucco C (1988) On the membrane translocation of diphtheria toxin at low pH the toxin induces ion channels on cells. EMBO J 7 3353-3359. 

---