feature

Striking Beauties:
Venomous Snakes

Joe Slowinski

A snake handler at a small roadside zoo in Arizona is momentarily distracted as he handles a large western diamondback rattlesnake. The snake seizes the opportunity to plunge its two fangs into the man's forearm. Within seconds the venom molecules begin to destroy tissue around the wound. The forearm quickly swells and becomes black and purple; the pain is intense. But these are only the first, localized symptoms. As the blood spreads the toxins around the body, the venom molecules attack the circulatory system: destroying red blood cells, punching holes in the blood vessels, and interfering with blood clotting. Twelve hours after the bite, large, blood-filled blisters appear all over the arm. As blood continues to leak out through perforated vessels, the blood pressure drops precipitously.

Fortunately, the patient arrives at a hospital and is administered antivenom for rattlesnakes. The antivenom molecules quickly neutralize the venom, and the patient's blood pressure becomes stabilized. The patient will live, but his arm will be permanently scarred and weakened.

Snake venoms are probably the most complicated animal venoms in the world. They are toxic cocktails of a dozen or more distinct proteins-both enzymes and non-enzymes-each coded for by a different gene. Because of variation in these genes, the venom of a single snake species can differ geographically, and even between individuals. Moreover, the particular venom combination can also vary with age in the same snake. This complexity and variability is a headache for physicians treating snakebite victims, because the symptoms are often unpredictable and can differ from patient to patient.

But it provides an exciting challenge to those who study snake venoms. In the past two decades, scientists have made significant advances in understanding the pharmacology and toxicology of snake venoms. Much more recently, they have been examining their evolution, and some very exciting results are emerging.

Most venomous snakes occur in one of two families: the Elapidae or the Viperidae. The almost 300 species of elapids include such snakes as cobras, coral snakes, taipans, and sea snakes, while the 223 species of viperids include such familiar snakes as vipers, rattlesnakes, copperheads, and water moccasins.

Elapid venoms tend to kill via proteins called neurotoxins, which disable muscle contraction and bring about paralysis. Death from elapid bites generally results from asphyxiation because the diaphragm can no longer contract.

Viperid venoms, on the other hand, contain an abundance of protein-degrading enzymes called proteases, which produce the symptoms like those encountered by the unlucky rattlesnake handler: strong local swelling and necrosis, and blood loss from cardiovascular damage complicated by coagulopathy, or disruption of the blood clotting system. Death from viperid bites is generally caused by a collapse of blood pressure. But the distinction is not always so straightforward. Elapid bites can also produce symptoms typical of viperid bites, and many viperid bites result in neurotoxic symptons.

Reliable figures do not exist, but elapids and viperids together may kill between 10,000 to 50,000 people a year worldwide. The vast majority of these deaths occur in the tropics, where, for reasons not understood, venomous snakes tend to have more toxic venoms, and where people come into contact with snakes more frequently. In the United States, where snake bites generally cause fewer than five deaths a year, many bites happen to people who handle snakes. The large-scale killing of venomous snakes under the pretense of saving human lives is indefensible, and many of the bites that occur each year befall the misguided people participating in rattlesnake roundups, annual events that occur across the United States to rid natural areas of venomous snakes.

Which snake is the world's deadliest?" As a herpetologist who specializes in venomous snakes, I hear this question all the time. Unfortunately, there is no easy answer. First, we must define "deadliest." It could be the snake with the longest fangs, largest venom glands, or the most aggressive behavior toward humans (see sidebar). But, I suspect that what most people mean is: Which snake is the most toxic? In other words, which species has the strongest venom?

Venom toxicities are generally measured in terms of mouse LD-50s, which is the dose required to kill one-half of a test group of mice which have been given the venom. LD-50s have been measured for many species of venomous snakes; the results are posted and regularly updated on the International Venom and Toxin Database (http://www.uq.edu.au/~ddbfry/ind ex.html). So we now know which snake is the most toxic in the world: the small-scaled taipan (Oxyuranus microlepidota), a medium-sized elapid that lives in Australia. It takes only 500-billionths of a gram of this snake's venom to produce a 50 percent chance of killing the average mouse. When you consider that the average amount of venom in the venom glands of a small-scaled taipan equals 88,000 mouse LD-50 doses, it becomes dramatically clear how dangerous this particular snake is. Of course, mice are not simply scaled-down humans, and it is impossible to extrapolate this LD-50 in a precise way to people. Nonetheless, it is fair to say that untreated human envenomations from this species are almost always fatal.

All of the world's ten most toxic snakes are elapids, and half of these are sea snakes. One must go pretty far down the list to encounter some of the venomous snakes familiar to Americans. The coral snake (Micrurus fulvius) of the southeastern United States has a venom 52 times less toxic than the small-scaled taipan; the timber rattlesnake (Crotalus horridus) of the eastern United States has a venom 124 times less toxic. One generalization that emerges is that elapids are more toxic than viperids. The neurotoxic proteins they have evolved are extremely effective at bringing about death with minimal disturbance to the body. Viperid venoms, on the other hand, are messy -tearing apart tissues and literally melting cells. So an elapid venom is more likely to kill you, but will leave a far prettier corpse.

But elapids and viperids are not the entire story. A third snake family, the Colubridae, also has venomous species. This family contains approximately 64 percent of the more than 2,500 snake species and includes such familiar and harmless ones as garter snakes, gopher snakes, racers, and rat snakes. The night snake and lyre snake of southern California are two of the venomous colubrids, but neither poses a threat to people.

Although it had long been understood that some colubrids are venomous, the potential threat posed to human life by these species was not really taken seriously until two of the most prominent herpetologists of the twentieth century found out the hard way that some colubrids can inflict fatal bites. Karl Schmidt of the Field Museum of Natural History in Chicago died after suffering a bite from an African boomslang in 1957. And in 1975 a bite from an African twig snake killed German herpetologist Robert Mertens. Neither man sought medical help at the time, largely because the lack of severe symptoms early on led each to underestimate the gravity of his condition.

These unfortunate incidents drove home the point in the herpetological community that at least some colubrids should be treated with extreme caution. But which ones, other than boomslangs and twig snakes? We don't really know, although we think that at least 25 percent of colubrid species are venomous. But we have little idea how dangerous they are. Unlike the front-fanged elapids and viperids, the venomous colubrids are rear-fanged snakes, which must hang on and chew when they bite to inject any venom. Further, many venomous colubrids are docile and reluctant to bite, even when picked up.

I suspect that several other colubrid species will turn out to be lethally toxic to humans. Several years ago, a beautiful Asian snake, the red-necked keelback (Rhabdophis subminiata), was commonly sold into the pet trade as harmless, until one fatal and several near-fatal bites occurred to people handling them. In fatal bites by the toxic colubrids, the venom acts in the following insidious way. Initially, the venom proteins promote massive blood clotting, which leads to headaches, dizziness, and vomiting. This subsides after a day or so, and the victim begins to feel better. But the victim is actually dying. The person's clotting protein has been used up, and can no longer plug leaks in the circulatory system. An autopsy of Karl Schmidt revealed that, even as he prepared to go back to work, he had been hemorrhaging from small leaks all over his body. People tend to categorize snakes as either venomous or harmless, but, as colubrids demonstrate, the reality is a continuum, from completely harmless, to mildly venomous, to deadly.

Althought we have learned quite a bit over the past two decades about the biochemistry, pharmacology, and toxicity of snake venoms, we still know very little about their evolution. Where do snake venom proteins come from, and what drives their evolution? This is all terra incognita. But researchers exploring the molecular evolution of snake venoms have begun to make progress. First, we know that snake venoms evolved primarily to subdue or kill prey. The defensive role for venoms probably had very little to do with their evolution.

One piece of evidence that supports this is that many bites to humans are dry, where no venom is injected. Venomous snakes possess some ability to meter their release of venom, to size up the situation and deliver an appropriate amount. Large prey receive more venom than smaller prey, and often an attacking snake considers it unnecessary to inject venom into a potential predator (such as a human) that it is trying to repel, rather than kill. Venom proteins, after all, take energy to make. But one group of snakes, those familiar elapids known as cobras, has evolved secondarily the ability to spit venom, as a defense, at the eyes of a perceived predator.

We also know that venomous snakes have co-opted many proteins for their venoms that are already used elsewhere in the body. For example, many venom enzymes have a digestive function, breaking down various molecules in the prey such as lipids, nucleic acids, and proteins. These same enzymes exist elsewhere in the snake's body, where they serve roles in normal metabolism. Are the venom and body enzymes coded for by the same genes? No. The pattern now emerging suggests that early in the evolution of these snakes, genes for enzymes already serving important functions in the body were duplicated to create identical copies elsewhere in the genome. These new gene copies became expressed in the venom glands. In this way, the genes could be tinkered with to perfect them for their new role of killing envenomated animals. But this doesn't tell the whole story. Some of the proteins in snake venoms, especially some elapid neurotoxins, are not similar to any other known proteins. Clearly, more work needs to be done.

A critical first step in understanding the evolution of snake venoms is to determine the sequence of the particular nucleotides forming the gene, a process that has begun only recently for snake venom proteins. The few viperid sequences that have been published reveal a startling pattern: The coding regions of these genes evolve at a very high rate.

Genes of all non-bacterial and non-viral organisms are constructed in common fashion, with segments that code for the final protein interspersed with segments that code for nothing. The coding segments are called exons, the non-coding segments introns. Typically, in all non-bacterial and non-viral organisms, introns evolve much faster than exons because mutations to introns have little functional significance. On the other hand, mutations to exons that actually alter the protein itself will generally be deleterious and thus eliminated through natural selection. But researchers sequencing viperid venom genes found just the opposite-that the exons were evolving at a much higher rate. They also found that duplication of viperid venom genes is common, and that any given species of viperid is likely to have several copies of the gene for a particular venom protein. How can we explain such surprising results?

The hypothesis that myself and others have formulated considers venomous snakes to be locked in an evolutionary "arms race" with their prey. Prey are constantly evolving resistance to snake venoms, rendering the venoms ineffective. This places enormous pressure on the snakes to circumvent the resistance by changing their venoms. The pressure is so great that just about any change that alters the amino acid sequence of the venom protein will be favored by natural selection. But then, of course, the prey re-evolves resistance, and so on. This explains why the exons evolve faster than the introns. The frequent gene duplications provide additional venom proteins, expanding the snake's arsenal and increasing the chance that the prey will be killed. To quote evolutionary biologist Leigh Van Valen, quoting Lewis Carroll, "Here, you see, it takes all the running you can do to keep in the same place."

Three intriguing lines of recent research seem to provide corroborative evidence for the arms race hypothesis. First, researchers at the University of Wales examined the venom of the Malayan pitviper (Calloselasma rhodostoma), a species widespread in southeast Asia. They found that, like all venomous snakes, this species shows tremendous geographic variation in its venom proteins. They then compared the pattern of venom variation to geographic variation in diet and found a correlation: shifts in diet corresponded to shifts in the venom proteins, clear evidence that the rapid rate of change in the Malayan pitviper's venom proteins is in response to their diet.

The second piece of corroborating evidence comes from recent studies revealing that many prey species have proteins in their blood and tissues that confer some degree of resistance to venoms. This is backed up by anecdotal reports of small mammals in certain parts of the world that are virtually immune to the bites of local venomous snakes.

Finally, a really neat development is that researchers studying other venomous animals are reporting similar results. For example, scientists working on cone snails, venomous predatory gastropods, have found a pattern of gene duplications, followed by rapid evolution of exons. This, too, is almost certainly due to an evolutionary arms race.

Elapids have not yet been examined for these phenomena. But here at the Academy's Osher Foundation Laboratory for Molecular Systematics, Robin Lawson and I have begun to sequence venom protein genes in elapids in an effort to find out whether or not they share the viperid pattern of gene duplications and rapid exon evolution. I predict that we will find the same pattern and suspect that this pattern will ultimately be found to characterize most, if not all, of the world's venomous animals.

I hope to have shown that venomous snakes and their venoms are fascinating. Of course, snakebite victims may not share my appreciation. But as with that unlucky snake handler in Arizona or crazy museum curators, the fact is that many bites in this country occur to people who pick snakes up. The best way to avoid being bitten by a venomous snake is simply to leave it alone.

Joe Slowinski is Assistant Curator of Herpetology at the California Academy of Sciences.