The natural world has produced some spectacularly deadly toxins. One of the most deadly is ricin, produced by the castor oil plant. Just one castor bean seed can kill a small child and less than a milligram is enough to kill an adult: Bulgarian émigré journalist Georgi Markov died in 1978 after a tiny pellet containing only 450µg of ricin was fired into his leg from the end of an umbrella on the steps of Waterloo Bridge. Lethal though they may be, ricin and similar toxins have interesting biological properties that Lynne Roberts and colleagues at the University of Warwick believe can be exploited to medical advantage.

Supported by a Wellcome Trust programme grant, Dr Roberts and colleagues are studying ricin and Shiga-like toxin 1 (SLT1), an Escherichia coli protein related to the toxin causing dysentery. Both these proteins disable ribosomes, the large assemblies of protein and RNA that make new proteins in the cell. Actually getting to the ribosomes is no easy task, however. Proteins are taken up by cells in membrane-bound vesicles, from which they need to escape to get into the watery matrix, or cytosol, where the ribosomes are found.

To get at ribosomes, ricin and SLT1 exploit an unusual 'back entrance' into cells. Dr Roberts and Professor Mike Lord are studying this process both to gain a better understanding of how proteins are trafficked around a cell and to develop new ways to deliver useful products into the deep internal workings of the cell.

Reverse transport

Transport of large molecules inside the cell depends on tiny vesicles, which act as cargo vessels moving material between a ramifying network of internal chambers. Ricin or SLT1 molecules taken up by a cell tend to become trapped in membrane-bound (endosomal) compartments near the cell surface and are either recycled back up to the cell surface or destroyed. However, a fraction manage to escape from that pathway and travel to an organelle known as the Golgi body. From there they reach the endoplasmic reticulum (ER), a kind of network of interconnecting corridors.

The entry route for ricin – from the endosomes to the Golgi to the ER – is actually part of the classical secretory pathway operating in reverse. "Most people are familiar with proteins moving along this route in the opposite direction, from the cytosol to the ER for transport towards the cell surface and secretion," says Dr Roberts. "I think everybody now also accepts that ricin and SLT1 undergo retrograde transport along the same pathway to reach the ER. What we don't know is precisely what is happening at the molecular level."

A key issue is how ricin convinces the ER to expel it into the cytosol. "The ER is the only place in the cell from where these particular toxins can get out into the cytosol to act," explains Dr Roberts. "There's normally a degradation process tightly coupled to this activity: the ER spits out abnormal proteins, which are then digested by degradation machinery nearby in the cytosol. But these toxins manage to escape that fate."

As well as looking at the molecular 'tags' that guide toxin movement from the Golgi body to the ER, the group wants to know what happens once the toxins reach the ER. "The strange thing is that although these proteins are robust and properly folded, they somehow manage to masquerade as abnormal proteins. They manage to exploit the so-called ER quality control system to get themselves expelled into the cytosol where they can act on the ribosomes."

Once in the cytosol, by yet another 'trick' the toxins manage to avoid degradation by the digestion machinery. Dr Roberts thinks this could be due to the fact that they have very few lysine amino acids on their surface. The presence of lysine typically leads to the addition of a protein tag known as ubiquitin, which earmarks tagged proteins for degradation and disposal.

To pass out of the ER into the cytosol, the toxins need first to be partially unfolded, so that they can travel through the ER membrane rather like a thread through a needle. Once in the cytosol, however, they need to refold to act. "There is some evidence that the ribosomes are actually helping the toxins to refold, even though in so doing, they are enabling the toxin to destroy them," says Dr Roberts. "This is the ultimate sleight of hand by the toxin. We call a protein that helps another protein to fold a chaperone, so in this instance we would like to refer to ribosomes as suicidal chaperones because by the very act of refolding the toxin they are committing suicide."

Toxins with a twist

Dr Roberts and Professor Lord are hoping to exploit the unusual trafficking properties of these toxins to help develop new T-cell vaccines. "Many vaccines elicit a B-cell response which produces antibodies," says Dr Roberts. "Antibodies bind to microbes or free foreign proteins – antigens – which are floating in the serum, but they can't detect abnormal proteins that are made within the cell, like viral proteins or certain tumour antigens."

Cytotoxic T cells (CTL), or 'killer' cells, are required to destroy intracellular pathogens such as viruses. However, inducing a CTL response through vaccination is more problematic than eliciting a B-cell antibody response – largely because of the way antigens introduced into the body are handled.

To stimulate an immune response, fragments of broken down (processed) antigens have to be 'presented' to T cells in association with special proteins – known as major histocompatibility complex (MHC) proteins – found on the surface of particular host cells. There are two types of MHC protein, class I and class II. MHC class I molecules present processed antigen to CTL whereas MHC class II present antigen to a second type of T cell, the T helper cell. One important role of these T helper cells is to activate B cells.

So, introducing an antibody response can be a relatively straightforward process, involving the introduction of foreign antigens into the body. T helper cells are activated once the foreign antigens are taken up by what are known as antigen-presenting cells (APCs). "Collectively, APCs hoover up free antigens and invading pathogens," explains Professor Lord. "Foreign proteins get taken into APCs and broken down in the endosomal compartments where MHC class II molecules are located." Fragments of antigen become bound to MHC class II and are cycled back to the surface of the APC for ‘presentation’ to the T helper cells. Activated helper cells in turn stimulate secretion of protective antibodies by B cells.

Stimulation of a protective CTL response is more challenging, however, because MHC class I molecules are found deep within the ER. Most proteins taken into cells like APCs never get that far, having become trapped in the endosomal system.

Dr Roberts and Professor Lord are looking at exploiting the natural trafficking route of ricin and SLT1 to deliver antigens into the MHC class I pathway. "We thought we could use our toxins to deliver viral peptides into the ER, where the toxin is clipped off, enabling the peptide to join up with the MHC class I molecules waiting in the ER and travel back to the cell surface for presentation," says Dr Roberts.

In collaboration with Professor Vincenzo Cerundolo at the Institute of Molecular Medicine (IMM) in Oxford, Dr Roberts and Professor Lord have already shown, in vitro, that SLT1 and ricin can deliver antigenic flu peptides in association with MHC class I proteins to the cell surface in sufficient quantities to promote a CTL response. This promising lead will eventually be followed up by studies in mice, again using ricin as a vehicle.

At Warwick, ricin will be disarmed to render it harmless, and then attached to viral peptides or proteins that researchers at the IMM know would elicit a good CTL response if delivered to the right part of the cell. "It's a really good multidisciplinary collaboration," says Dr Roberts. "IMM is a centre of excellence with a wide range of CTL resources. They provide the immunology expertise and we have the toxin expertise."

Although still at an early stage, this delivery strategy may provide novel anti-viral and antitumour vaccines in future. Ironically, toxins responsible for countless deaths in the past may yet end up saving many lives instead.

External links

• Toxin Research Group at the University of Warwick: Research interests of Dr Lynne Roberts and Professor Mike Lord
• Georgi Markov: An account of the assassination

Further reading

Sandvig K, van Deurs B (2000). Entry of ricin and Shiga toxin into cells: molecular mechanisms and medical perspectives. EMBO J. 19: 5943-5950

Lord J M, Smith D C, Roberts L M (1999). Toxin Entry: How bacterial proteins get into mammalian cells. Cellular Microbiology 1: 85-91

Lord J M, Smith D C, Roberts L M (1998). Toxin Entry: Retrograde transport through the secretory pathway. J. Cell Biol. 140: 733-736

Lord J M, Smith D C, Roberts L M (1998). Retrograde Transport: Going against the flow. Current Biology 8: R56-R58

Bona C A, Caseres S, Brumeanu T-D (1998). Towards the development of T cell vaccines. Immunology Today 19: 126-132

Guidebook to Protein Toxins and their Use in Cell Biology (1997). Rappuoli R, Montecucco C (Editors). Oxford University Press.