Handle those catfish with care!


Do you have a venomous fish swimming round in your aquarium? Chances are, that if you keep catfish, you do. Jeremy J. Wright, who has been studying catfish venom, explains...

Approximately how many catfishes are known to be venomous to some degree and which family is thought to have the highest number of venomous species?

The results of my survey last year, indicated that anywhere between 1250-1625 species are likely to be venomous. 

Although I was able to examine material from nearly all of the currently recognised catfish families, this still represented only about one quarter of the known genera. I’m continuing to examine new material as it becomes available, which should allow the generation of more precise estimates in the future. 

As it stands, it would appear that bagrids, callichthyids, and mochokids are all in the running for having the highest levels of venomous species diversity. All of these families contain many undescribed species, most of which are also likely to be venomous, so the winner is likely to change over time and will depend upon the rate of description of these species by taxonomists.

What type of venom is produced and what are its effects?

The toxic compounds contained within the venom are proteins of varying molecular weights. In humans, the most common result of envenomation is severe pain and redness or swelling at the site of the wound, which can persist for several hours.

Secondary infections and complications arising from remaining foreign material (broken bits of fin spine) at the wound site are also common.

In other fishes, catfish venoms can produce several additional symptoms, including dramatic loss of colour, muscle spasms, haemorrhage, loss of equilibrium, and in some cases, rapid mortality.

Are all catfish venoms the same?

No. While they do all share some similarity in protein composition, the exact identity of the toxic proteins and the effects that they produce varies between, and sometimes within, different groups.

Where are the venom glands located and how is the venom delivered?

The venom glands are associated with the pectoral and dorsal-fin (in the species that have them) spines, though the cellular morphology, size, and orientation of the glands vary from group to group.

When the spines penetrate a potential predator’s (or occasionally an aquarist’s!) tissue, the glandular cells surrounding the spine are torn, releasing venom into the wound.

How do the venom glands of Doradids differ from other siluroid families? The distinctive venom delivery apparatus of a doradid catfish.

The venom glands of the doradid species that I’ve examined are restricted to the spaces between the serrations found on the fin spines of these species, as opposed to being found along the length of the spines in other groups.

Microscopically, the glands are separated into discrete chambers by integumentary tissue, with each chamber containing a cluster of glandular cells. The glands of other species have no integumentary subdivisions. 

Finally, doradid venom glands are often visible without magnification, having a pronounced granular appearance.

The above picture shows the location of the venom glands in the Ripsaw catfish, Oxydoras niger.

How have catfish venom glands evolved and what is their purpose?

Their current purpose would appear to be predator deterrence. How they came about is a matter of some uncertainty, but for some time, it was thought that they were derived from toxin secreting epidermal cells. 

However, when I started looking at the effects produced by the venoms and the qualities of the proteins that they contained, I noticed some striking similarities to other epidermal proteins that have been shown to accelerate wound healing. 

The development of high concentration of these cells in an area such as the fin spines, which are likely to be damaged in an attack by a predator, and evolutionary derivation of venom glands from these types of cells makes a lot of intuitive sense. 

Genetic and structural characterisations of venom proteins will go a long way towards clarifying this issue, and are something that I’m actively pursuing.

Is it possible that Corydoras and other Callichthyids may have evolved them independently?

Not only possible, but very likely! Callichthyids represent an independent evolution of venom glands within the Loricarioidei, as they are the only group in which I found evidence of venom glands, though I examined multiple representatives of all of the other loricarioid families as well. 

The results of a fairly recent (2006) molecular phylogenetic study indicate that the doradids, with their unique venom gland morphology, may also represent an example of an independent evolutionary development of these structures.

Do you have any ideas as to why some species have them, while others don't?

In malapterurids, amphiliids, cetopsids, and some silurids, just to name a few examples, bony spines have been secondarily lost. In the absence an effective delivery device for venoms, it wouldn’t make much sense to continue to put energy into their production. 

In some cases, such as the North American flathead catfish (Pylodictus olivaris), individuals reach such large sizes that outside of humans (and possibly alligators in some areas), no predatory species would pose a significant threat. If predators are not much of an issue, it doesn’t make sense to continue investing energy into an anti-predatory trait. 

In other cases, like aspredinids and erethistids, it’s much harder to develop possible explanations for why these relatively small-bodied catfishes with well-developed bony spines would lack venom glands. If anyone out there has any ideas, I’d love to hear them!

Taxonomic distributions and estimates of venomous catfish diversity

Siluriformes - Catfishes, 1250-1625 species in survey

Taxon                                               Number presumed venomous

Akysidae - Asian stream catfishes       48

Amblycipitidae - Torrent catfishes       26-28

Anchariidae - Madagascan catfishes    4-6

Ariidae - Sea catfishes                       67-134

Bagridae - Bagrid catfishes                176-198

Callichthyidae - Armored catfishes      182-194

Chacidae - Angler catfishes                3

Clariidae - Labyrinth catfishes            79-114

Claroteidae - Claroteid catfishes         56-84

Cranoglanididae - Armorhead catfishes  3

Doradidae - Thorny catfishes              91-160

Ictaluridae - North American catfishes 57-64

Mochokidae - Squeakers                    166-189

Pangasiidae - Shark catfishes             27-30

Pimelodidae - Antennae catfishes       41-79

Plotosidae - Eeltail catfishes               17-37

Pseudopimelodidae - Bumblebee cats 21-31

Schilbidae - Glass catfishes                48-62

Siluridae - Sheath catfishes                74-83

Table from Wright JJ, 2009 Diversity, phylogenetic distribution, and origins of venomous catfishes. BMC Evolutionary Biology 9:282