Recent Komodo postings of Dr. Bryan Fry

[venomdoc]

Smy_749, key here is that it is nothing exceptional in regards to types or levels. A good lab could document as many in the mouth of family dogs or lions.

Bacteria-as-a-weapon is evolutionarily implausible since it would act too slow to be under any sort of selection pressure. If water buffalo had never been introduced to the islands, then this enchanting fairy tale about komodos selectively blooming bacteria in their mouth never would have been dreamed up in the malaria-fever restless nights of Auffenberg. He proposed the idea based upon a fundamental misreading of the biological reality rather than recognising the artificial situation for what it was. Introduced water buffalo trapped on small islands with only stagnant water holes has as much to do with reality as catapulting cows to great white sharks.

[venomdoc]

The next step in our research is to culture the watering holes. We will be sampling widely, not only on Rinca and Komodo islands but also in the native range of the water buffalo as well as in northern Australia where they have also been introduced. The below images illustrate just how putrid the water is and the sort of non-lethal wounds that the water buffalo are getting, that facilitates the entry of the bacteria from the disgusting water,

[MDT]

Brian, that last pic looks almost like a prolapse of (?)rectum/uterus/etc., rather than abscess or the like. Can you comment on what is going on in that pic?

Thanks!

[venomdoc]

Indeed, it is an excellent example of the sorts of wounds that occur when komodos attack. They do a unique 'grip and rip' mode, where by they bite and then use the strong forelegs and neck to pull straight back, with each tooth thus cutting in succession, resulting in four parallel deep wounds. In this particular event, it sliced him right in the butt. There are other cuts in the legs. He ended up dying slowly over the next week from septicemia. However, in the case of natural sized prey items, this can result in severed major arteries. At one of our field sites (Rinca Island) a young boy was killed this way, when his femoral artery was cut all the way through. He rapidly bled out.

[MDT]

Nice
I think I'd rather go by hemorrhage rather than by sepsis.....

I appreciate the info!

[Mikoh4792]

Originally Posted by venomdoc

Indeed, it is an excellent example of the sorts of wounds that occur when komodos attack. They do a unique 'grip and rip' mode, where by they bite and then use the strong forelegs and neck to pull straight back, with each tooth thus cutting in succession, resulting in four parallel deep wounds. In this particular event, it sliced him right in the butt. There are other cuts in the legs. He ended up dying slowly over the next week from septicemia. However, in the case of natural sized prey items, this can result in severed major arteries. At one of our field sites (Rinca Island) a young boy was killed this way, when his femoral artery was cut all the way through. He rapidly bled out.

Do you have any details on how the accident took place? Did it chase the boy down? Was the boy walking along and happened to step on it's tail..etc?

[Gatorhunter1231]

Hello dr. Fry,
Congrats on all scientific adventures. I have had time to read your links to see if they give a break down of the enzymes, proteins, etc (AT&T works me like crazy). I have been aware of the anti-colagation agents related to varanus bites. Actually have have been on the receiving end several times.
What kind of potency are we talking about? In normal guy terms. I have been bite by v. Salvadorii many times and up to 5ft with anything major outside of large amounts of bleeding from teeth slicing through my skin (and gloves) like a hot knife through butter. Never had any skin deteration or anything that someone would picture to go with an envenomation.
I have had rabbits survive attacks from an 8ft v.salvadorii for a day and appeared they could live until infection. (I found rabbit and put out of misery). Major lacerations and major blood loss. I would venom could have easily pushed the rabbit over the age with the rabbit already at a weaken state but I had to put it down. A 40lb salvadorii vs 10 rabbit should be a lethal venom delievery when a 200lb Komodo's venom can drop a 500lb plus water buffalo. Sorry for any misspellings I quickly typed it down on my I-phone. Can't get scientific till Tuesday.

[venomdoc]

Key here, as I mentioned in my earlier post, is that the natural sized prey for Varanus komodoensis is 40-50kg. Water buffalo are well outside this range. So the bite effects (with mechanical damage as the primary effect and the venom there to supplement it) are mirrored in this. Please read my earlier post. As for your comments about what someone would picture for envenomation, in that local tissue damage must go with this, this is a view obviously coloured by the effects of viperid snakes which contain large amounts of proteolytic enzymes. In contrast, elapid snakes typically lack such enzymes. So there is no 'general bite effect' picture to be found even within the snakes. Venomous lizards typically do not produce local tissue destruction as they lack such large metalloprotease enzymes. Local pain and swelling may be present due to kallikrein enzymes though.

[murrindindi]

Hi Bryan, I don`t want to take away from the current discussion on V. komodoensis, but I`d like to ask if you have an opinion on whether the Mosasaurs and their closest relatives were likely to have been venomous, that would obviously make them the largest venomous vertebrates ever to have existed, rather than V. priscus?
Thanks in anticipation of a difinitive answer (rather than "it`s possible")!

[venomdoc]

Unfortunately when talking about paleontology, there is really no such thing as a definitive answer. While the Mosasaurs certainly appear to be varanoid, the relative placement is unclear. If they are in fact within the Anguimorpha, then they would indeed share a common venomous ancestor with the other anguimorph lizards. However, the envenomation strategy of these lizards would not be favored in a purely aquatic lifestyle since any venom would wash off from the open cuts or skin. This is in contrast to venom delivered via a puncture wound in the two independent lineages of marine snake (sea kraits and sea snakes) as well as cone snails or venomous fish. So Mosasaurs indeed emerged from within the venomous lineages, I suspect any venom system would be relicital, with powerful crocodilian-like jaws favoured instead for predation.

[venomdoc]

The last sentence should have read:

So if Mosasaurs indeed emerged from within the venomous lineages, I suspect any venom system would be relicital, with powerful crocodilian-like jaws favoured instead for predation.

[Rob_H]

Thanks Bryan. Follows is a lengthy reply summing up my perspective in more detail (although at the risk of repeating myself).

These seem to be the points you’re arguing:

1) Komodos use venom to kill their prey

2) Predation facilitated by sepsis cannot occur through inoculations from Komodo bites.

3) Predation facilitated by sepsis comes instead via the pathogens in the water-holes the buffalo live in following a lacerating Komodo bite

4) Pathogenic bacteria found in the mouths of Komodo’s are obtained from the environment where natural predation occurs, so they cannot be important in their foraging ecology.

5) sepsis-facilitated predation, from whatever source, is caused by human-introductions, so has not had enough time for selection to operate on, and is therefore not important.

6) Sepsis works too slowly to function as the primary means of disabling prey, as most animals have likely run off before the Komodo can catch them.

I’ll quickly address these in turn.

1) You’ve shown mechanistically that this is very possible from your great lab work. But no one has yet shown whether venom is important functionally. If you’re referring to Tim Jessop’s published predation observations from his PLoS One 2010 paper, this only involved 17 predation observations, none that involved buffalo, and with no analysis of whether venom or sepsis was killing the animals. Since no field study has been done on the functional importance of venom in Komodos in a natural setting, it is therefore, by definition, still speculation.

2) The current evidence for this is that oral swabs taken from Komodos in a natural setting contain bacterial flora that can cause sepsis; these flora do not spontaneously colonise except in the environmental context where natural predation occurs (acknowledged by Auffenberg); observations of Komodos predating animals with sepsis following a bite are common; no study to date has investigated whether animals such as buffalo will die if bitten by a Komodo without bathing in nasty water, which is necessary to confirm/dismiss this claim.

Again, I and many other’s would conclude that totally dismissing the role of septic bites in Komodo predation requires field data to be convincing. There is currently more evidence for it than not.

3) This is very convincing, and I doubt anyone would argue that this doesn’t contribute to how Komodo’s kill prey items such as pigs and buffalo that frequently wallow.


4) Maybe I don’t understand this one, but there are tonnes of examples of evolutionarily important adaptations that must be obtained from the environment (poisons, pheromones, scent mark compounds, digestive aids). Maybe the point of this argument, and your recent paper, has escaped me…. If you’re arguing that Komodos don’t therefore have any evolutionary adaptations to facilitate colonisation by these bacteria then I don’t think this is the best study to have addressed that.


5) This is simply a misnomer. There are many, many published examples of species responding to selection in response to human introductions (see my previous example, and a good review is: Strauss et al (2006) Evolutionary responses of natives to introduced species: what do introductions tell us about natural communities? Ecology Letters). Some of the best examples of this are in fact lizards (Anolis, Podarcis etc), where measurable responses to selection can be found in less than ten generations. Either way, an evolved response does not detract from whether a behaviour or trait offers a fitness advantage now.

6) I see exactly where you’re coming from: obviously it must be unlikely to evolve since there are few (any?) examples apart from Komodos, but again…. No field data! I don’t need to tell you that we’re talking about an animal with a highly developed olfactory system that is famed for tracking carcasses and prey from long distances…. If there is one species that is likely to be able to use a method of predation that takes days to kill large prey items, then Varanids are probably right up there.



As far as the current evidence is concerned, there is a very strong story that Komodos and many other Varanids are venomous, scant to no scientifically presented evidence that any species use this venom functionally when killing their prey, and absolutely no evidence, in my mind, that Komodos do NOT use bacteria as a ‘venom’ when killing prey.

I have no agenda in regards to whichever hypothesis is accepted, but I do believe people need to be particularly rigorous when discrediting previous hypotheses and the scientists that proposed them.

Hence my suggestion that the most important component you now need is field data (no matter how difficult it is to collect) to disentangle the role of bite-inflicted trauma, venom, and sepsis in the foraging ecology of these animals to advocate the role of venom, and discredit Auffenberg’s suggestion for the role of pathogenic bacteria.

Cheers,
Rob

[venomdoc]

1) Key is that water buffalo always escape. They flee quite some distance from the attacker. Thus if and when the buffalo dies, the original dragon is very unlikely to be the beneficiary. Thus there would be no selection pressure operating upon such a predation strategy as nature does not run a charity

2) Actually, the evidence does not support this at all and indeed the Pasteurella multocida section was the weakest part of the Montgomery study. Pasteurella multocida is part of the normal oral flora of dogs, cats, and other animals and is only associated with infection when outside its usual ecologic niche. Montgomery et al.33 proposed P. multocida as the cause of Komodo-associated prey sepsis and mortality, even though it was reported in only 5% (2/39) of their dragons. Although infections from this in bite wounds of mammals may be on occasion pathogenic, they are associated with sepsis infrequently. At best this species can be called an opportunistic pathogen that given time in the right environment in a wound would be serious and perhaps lead to eventual death but are in no way pathogens in the sense we regard Strep pyogenes, Staph aureus, gangrene-causing bacteria

The rest of the gram-negative organisms that they isolated are commonly found in soil and water. They also found 29 gram-positive species, 25 from wild and eight from captive dragons; most were staphylococci and streptococci that are typically found on animal skin, as well as Bacillus spp., that are commonly found in soil and on vegetation. They considered 54 of the 57 species to be ‘‘potentially pathogenic’’; however, these members of the normal microbiome are in fact of low virulence and unlikely to be the cause of rapid fatal infection when present in a wound.

So the extreme variability of bacteria present in the wild dragons argues against a use since there is obviously no selection pressure at work, combined with the fact that even in the wild dragon study the types of bacteria present were quite ordinary. The difference between our study and theirs was that I involved clinical microbiologists in mine, thus there was no over-reaching as to the nature of the bacteria present.

3) it is an artificial situation so it has zero relevance to the evolution of predatory behaviour

4) If komodos were in fact evolutionarily selected to facilitate the growth of bacteria in their mouth, then there would be a consistent trend as to the types and levels of bacteria present. Instead, the wild and captive studies both showed simply the normal types of transient oral flora expected to be found in the mouth of a lion or the family dog. There was the expected levels of extreme variability in species present. If they were indeed cultivating it, then there would be evidence of evolutionary selection but this was conspicuously lacking.

5) there is no fitness advantage to be gained if the biting animal does not obtain the prey. It is not like there is a komodo dragon charity at work with water buffalo bodies dropping all over the island. It is actually a quite rare event.

[venomdoc]

6) regarding field observations, of all the extensive tracking and filming of komodos, there has never been a single documented case of the same komodo biting a water buffalo and following it until it dies.

[MDT]

Originally Posted by venomdoc

2) Actually, the evidence does not support this at all and indeed the Pasteurella multocida section was the weakest part of the Montgomery study. Pasteurella multocida is part of the normal oral flora of dogs, cats, and other animals and is only associated with infection when outside its usual ecologic niche. Montgomery et al.33 proposed P. multocida as the cause of Komodo-associated prey sepsis and mortality, even though it was reported in only 5% (2/39) of their dragons. Although infections from this in bite wounds of mammals may be on occasion pathogenic, they are associated with sepsis infrequently. At best this species can be called an opportunistic pathogen that given time in the right environment in a wound would be serious and perhaps lead to eventual death but are in no way pathogens in the sense we regard Strep pyogenes, Staph aureus, gangrene-causing bacteria

The rest of the gram-negative organisms that they isolated are commonly found in soil and water. They also found 29 gram-positive species, 25 from wild and eight from captive dragons; most were staphylococci and streptococci that are typically found on animal skin, as well as Bacillus spp., that are commonly found in soil and on vegetation. They considered 54 of the 57 species to be ‘‘potentially pathogenic’’; however, these members of the normal microbiome are in fact of low virulence and unlikely to be the cause of rapid fatal infection when present in a wound.

So the extreme variability of bacteria present in the wild dragons argues against a use since there is obviously no selection pressure at work, combined with the fact that even in the wild dragon study the types of bacteria present were quite ordinary. The difference between our study and theirs was that I involved clinical microbiologists in mine, thus there was no over-reaching as to the nature of the bacteria present.

Exactly....I have treated multiple canine/feline bites (never a varanid ) and to my knowledge, none of them have resulted in sepsis. Pastuerella is the predominant org in a feline bite. But there are also numerous others as well (both aerobic and anaerobic). Bites happen frequently and sepsis does not.

It does strike me as an opportunistic situation and not the primary causation for the death of the prey.