The venom hunters

Spiders have been killing insects for around 400 million years – they inject their poisonous venom using sharp fangs. The chemicals in venom range from salts and small organic compounds to large neurotoxins. Most importantly, they contain a trove of peptides that can interfere with a huge number of molecular targets, including lectins, protease inhibitors, sodium channels, glutamate transporters, voltage-gated ion channels, and ligand-gated ion channels. Consequently, the toxins have varied effect. Some rapidly incapacitate invertebrate, but not vertebrate, prey, making them ideal pesticides. Others inflict pain for defensive purposes. And significantly, of the 100,000 species so far characterised, only five are potentially lethal to humans. ‘The chances that a toxin that kills insects will hurt us are remote,’ King says. However, he points out they screen all peptides early to make sure.

‘The toxins that we are most interested in target voltage-gated calcium channels in insects. There aren’t any insecticides out there that target that particular ion channel and therefore there should be no pre-existing resistance to it,’ says King. Despite the vast array of chemical insecticides available, King explains only a very small number of molecular targets are exploited. This has helped drive insecticide resistance. So the venom peptides could be an Achilles heel for pests. ‘The [calcium] channels are essential to insects. Knock them out and the insects die.’ Even small quantities – less than 0.25nmol/g – of the funnel web toxin that King studied back in 1997 were capable of paralysing 50% of test insects, causing loss of coordinated movement, followed by paralysis and death.

Why calcium channel blockers have not been hit, says King, is unclear, though one challenge lies in expressing the channel for high throughput screens. Humans have 10 calcium channels and they can be expressed in a variety of cell types and used for screening molecules that may target that channel. Not so the insect channel. ‘The agrochemical companies have found it difficult to set up screens against the insect calcium channel and that makes it really hard to look for molecules that target this channel,’ King explains. Indeed, difficulty screening for insecticides makes them a harder nut to crack than human drugs, he says.

Another reason is selectivity. ‘When you spray an insecticide onto a field you are worried about its effect on non-target insects, pollinators, on dogs, fish, birds, everything, so you really need a high level of selectivity and that is hard to achieve,’ he explains.

In 2013, King’s group made some progress with their discovery of a peptide toxin from a tarantula that was orally active in an agricultural pest. The spider, the Australian featherleg (Selenotypus plumipes), was chosen because of its enormous size and venom yield. Significantly, insects were killed by ingesting the toxin but not through contact (PLOS One, doi: 10.1371/journal.pone.0073136). This was an important finding because any substance poisonous to insects through touch would be indiscriminate, killing pests and pollinators in equal measure. This spider toxin would only kill pests actively consuming crops. ‘This significantly reduces the likelihood of inadvertent effects on beneficial insect predators and pollinators such as bees,’ notes King.

Spider peptide toxins, such as that of the featherleg, have the added advantage of being highly stable because of the formation of a cysteine knot motif. The knot consists of a ring formed by two disulphide bridges and the intervening sections of peptide backbone, with a third disulphide bond piercing the ring to form a knot. This arrangement makes them resistant to attack by protein-destroying enzymes in its victims. Consequently, the peptides could be sprayed onto crops and would persist. The toxin could also be genetically engineered into crop plants, providing a more targeted and thus greener approach than current insecticide regimes.

However, there is a potential hurdle to killing insects with spider venoms. In general, the venoms are usually around 90-fold less lethal when eaten by insects than when fang injected. This reduced lethality is blamed on the slow rate of absorption of peptides across the insect gut. King and other venom researchers are working on ways to overcome this natural gut barrier.

Molecular biologist John Gatehouse at Durham University, UK, decided to create a ‘fusion protein’ whereby a spider toxin was fastened to a carrier protein to smuggle it across the gut. He used lectin from the snowdrop plant, which he joined to a spider peptide, Hv1a, obtained from the venom of the Australian funnel web spider (Hadronyche versuta) (PLOS One, doi:10.1371/journal.pone.0039389). The peptide targets insect voltage-gated calcium channels and is good at taking out the Colorado beetle, the potato pest.

‘We would like to go back to transgenic plants with this,’ says Gatehouse, ‘but there is too much antipathy toward transgenic plants at the moment.’ He suspects the toxin will end up as a niche product, applied as a spray.

Meanwhile, King, in collaboration with Raymond Leger at the University of Maryland, US, is genetically engineering spider-venom peptides into pathogenic fungi for control of insects. Leger is a pioneer in the field. He has already shown that Metarhizum fungi genetically engineered to express a scorpion venom peptide, AaIT  increases fungal toxicity 22-fold against tobacco horn worm and nine-fold against yellow fever mosquitoes (Aedes aegypti) (Nature Biotechnology, doi:10.1038/nbt1357).

‘The cool thing about these fungi is that many of them have a really restrictive host range. There is one used in Australia for locust control and it kills only particular species of grasshoppers,’ King enthuses. The downside of the pathogenic fungi without the peptide toxin is they work slowly. But if you engineer toxin genes into the fungi, you keep the selectivity and ramp up speed.

King is currently exploring the possibility of using genetically engineered fungi to express spider toxins against mosquitoes; toxic spores could be laced onto sugar feeding stations to attach themselves and grow on male and females insects, while also hitting them with spider toxin. ‘We are particularly interested in this in terms of mosquito control, where there is malaria or dengue, because we’re showing that it not only kills mosquitoes but also stops them feeding very quickly, which is key for mosquito-borne diseases.’ 

Viruses could provide another option for smuggling toxins across the insect gut. A recent study – led by Bryony Bonning at Iowa State University, US – used the coat protein of a plant virus to smuggle a toxin into the circulation of an aphid. Aphids that fed on this protein showed signs of neurotoxin-induced paralysis (Nature Biotechnology, doi:10.1038/nbt.2753).

The agrochemical companies are beginning to take an interest in this research. The US insecticide development company, Vestaron, already has a pipeline of novel biopesticides derived from spider toxins. And King is optimistic that others will follow. The very same toxins kill insects each day when delivered by spiders; the challenge now is for scientists to come up with novel ways of getting these same toxins into insects which consume our crops.

Glenn King believes venoms are a huge potential resource for new drugs. In 2013, he reported on a  painkiller from a centipede’s venom. The peptide inhibited an ion channel, the voltage-gated sodium channel NaV1.7, in mammals and outperformed morphine in reducing pain in rodent models (Proceedings of the National Academy of Sciences, doi: 10.1073/pnas.1306285110).

But despite this success, there are only a few venom-inspired drugs on the market. One is the painkiller Prialt, which a synthetic version of a small peptide obtained from the cone snail (Conus magus). Toxicologist Dietrich Mabs, a veteran researcher on venoms in Germany, says the word ‘toxins’ had often proved a turn off for industry; the pharma industry, he says, fears side effects.

Biochemist Christopher Shaw of Queens University Belfast, UK, lays the blame for a lack of venom-inspired drugs squarely on the pharmaceutical industry. ‘Small biologically active peptides are the vocabulary of cellular communication,’ he says, ‘and they must be the single most important molecules to study with respect to drug development because they control almost all the behaviour of cells.’ Slight changes in amino acid sequences, he says, often sharply alter their biological activity, something too often overlooked in the past.

Added to this, Shaw points to the range of potential targets of the venom peptides. ‘It never ceases to amaze me the enormous targeting specificity of these venom components, which act upon very discrete molecular targets, many of which are disease associated,’ he says. His molecular therapeutics group has worked on venom peptides from snakes, spiders and scorpions, but his big focus right now is amphibians. The secretions of frogs and toads contain a broad spectrum of biologically active compounds. 

Shaw’s group recently identified peptides from leaf frogs that affect smooth muscles in mammals as part of a basic research programme fishing for novel drug targets, and ones from a scorpion that showed antimicrobial and anticancer activities. He also discovered a new class of antimicrobial peptides in frog secretions (Amino Acids, doi. org/10.1007/s00726-013-1655-1). 

‘This is the new chemical space for drug discovery,’ says Shaw. ‘We have gone through the natural plant products and we have gone through the combinatorial chemical libraries – which were a bit of a disaster – but we are now in the third chemical space: smart venom peptides.’ He says pharmaceutical companies need to embrace biological active peptides. ‘

Toxin expert Kini Manjunatha, of the National University of Singapore, has spent over two decades studying snake venoms. Generally, there are two kinds – most are neurotoxins, but others affect the cardiovascular system. Manjunatha has isolated a number of novel toxins, including one derived from king cobra venom that was a more potent painkiller than morphine. The neurotoxin, hannalgesin, is now under development by French biotech Theralpha. ‘It is orally active, it does not cause addiction and does not show withdrawal symptoms, nor does it affect motor function,’ says Manjunatha. 

Other companies are beginning to take an interest in venom-derived drugs. Kineta in Seattle, US, for example, has a pipeline of peptide drugs; its lead compound, ShK-186, comes from a sun anemone and is being investigated for the treatment of multiple sclerosis, skin inflammation and other autoimmune diseases. The company also has a small peptide derived from the venom of cone snail being investigated as a treatment for severe, chronic pain.

Given how much easier it is to sequence genes and produce small peptides using molecular approaches, Manjunatha and Shaw believe that venoms and their toxins offer real bite in terms of potential health gains from new therapeutics.