Pterygoplichthys u pclose

I was interviewed for the October edition of Practical Fishkeeping magazine “Do catfishes need to eat wood?”.  You can link to an image here, although you can’t read the actual article unless you purchase a subscription.

They asked me some questions:

Which catfish groups have been observed eating wood?

What happens to these when they are fed wood?

Can they digest it?

Are they really eaters of wood, or do they consume it while grazing other foods?

Some manufacturers are adding lignin to catfish foods.  Is there any point in doing so?

What should wood-eating loricariids be fed?

I would like to thank Matt Clarke for contacting me for the interview.  It is nice that someone read my work.  And, PFK is a first-rate aquarium mag and the interview spread looks great.

You can read my answers (as I wrote them) here.  If you read the PFK article, you will see that some editorial changes were made to my answers making them more definitive than I originally worded them.  And, I NEVER called myself a “world expert” in anything.  That was their doing. 

I would like to respond to some of the criticisms that have popped up on pleco enthusiast websites.  Of course, if one is interested in more of the complete picture of my research on wood-eating catfish, I ask that you read my articles (go to my pubs page), not the “bites” presented in an interview.

The first point I would like to make is that I DO NOT like the title of the interview.  "Do catfishes need to eat wood?"  I would answer that question differently than those that they asked.  If you read my articles, you will see that the fish do indeed ingest wood in the wild, and this is very much part of their natural niche.  So, I would rather ask the question: Can wood-eating catfishes digest wood?  I think that the answer to the question "Do they NEED to EAT wood" may actually be "yes", because that is what they do in nature.  What they get nutritionally from the wood is a different question altogether (see below).  

Second, I am not an aquarist.  I do have three aquaria at home—an 80-gallon tank with wild-caught fish from Florida (including Florida Flag Fish, Mosquito Fish, Platys, Bluegill Sunfish, a minnow species I haven’t keyed out yet, a Pt. disjunctivus captured near Orlando, and a P. nigrolineatus purchased from a wholesaler); a 20-gallon tank (which houses a Green Terror Cichlid and a Hypostomus sp.); and a 10-gallon tank (which houses a Long-eared Sunfish that I captured in Alabama)—but I am not a hardcore aquarium hobbyist.  Rather, I am a biologist that studies wild fish in an evolutionary context and attempts to figure out how fish eat and digest what they do in the wild, not how to best keep them in captivity.  Obviously, I had to house fish in aquaria for feeding trials during my graduate career, and I did this in aquarium laboratories at CSU Fullerton (for the study of digestion in prickleback fishes) and at University of Florida (for the study of digestion in loricariid catfishes).  But, this does not make me an expert “aquarist”.  For example, I say nothing about feeding Panaque carrots in the PFK article, but I know my P. nigrolineatus at home loves carrots.  Given what I found in terms of their digestive physiology, it is pretty clear why.  Do Panaque eat carrots in nature?  I think the answer to that is obvious (no), but they have been observed to chew on roots that are submerged on river banks.  If you want to know the best wood to give Panaque in captivity, ask an aquarium enthusiast who has been observing Panaque in captivity for many decades. 

Third, I went into this project on wood-eating catfishes hoping to find new and novel microbes that could produce ethanol from lignocellulose.  In fact, when I first proposed working on wood-eating catfishes to my committee, I stated that I could see working on these fish for the next decade or more of my career if my original hypotheses panned out.  However, before one can even start to think about microbes playing a role in digestion, steps need to be taken to examine the digestive strategy of the fish to show a reliance on endosymbionts.  Please read the wonderful book by Karasov and Martinez del Rio “Physiological Ecology: How Animals Process Energy, Nutrients, and Toxins”.  By adopting the theoretical considerations presented in that book (and all of the literature cited therein), I, with the support of my committee, reasoned that there were four main areas that had to be addressed before one could make the leap to microbial digestion in these fish: 

1) Intake.  How much do these fish eat and how quickly do they pass it through their guts? Herbivorous animals reliant upon microbially-mediated digestion have long retention of food in the gut (e.g., 24 hours in termites).  

2) Selective retention.  Do these fish selectively retain small, or at least more digestible particles along their guts, thereby allowing microbial symbionts to digest refractory polysaccharides? Herbivorous and xylivorous animals reliant upon microbially-mediated digestion show selective retention of particles somewhere along their gut, usually in some sort of “chamber” separated from the rest of the gut by sphincters (e.g., the rumen of a cow or the hindgut chamber of the herbivorous fish Kyphosus sydneyanus), to engender microbial digestion (which is a slow process.  Microbes need time).

3) Digestive enzyme activity patterns.  Is there a pattern to digestive enzyme activities along the guts of the fish?  Non-ruminant herbivores (including fish) that are reliant upon microbial digestion of refractory polysaccharides in their guts have elevated activities of refractory polysaccharide degrading enzymes in their hindguts (see Skea et al. 2005; 2007). 

4) Redox potentials and short chain fatty acid (SCFA) concentrations.  Is the gut of the fish anaerobic, and if so, are there high levels of fermentation occurring somewhere along the gut?  Again, non-ruminant herbivores reliant upon endosymbionts have elevated concentrations of SCFAs in their hindguts, which are anaerobic.

From a mechanistic standpoint, these areas of digestion are consistent among wood-eating and herbivorous animals that are reliant upon microbial digestion in their GI tracts.  Unless these four areas can each be shown to be conducive to microbial digestion, then the animal probably doesn’t rely upon microbial endosymbionts to meet their daily energetic needs.  It is a consistent problem that people assume that ANY plant eating animal is reliant upon endosymbionts for digestion, when actually, relatively few are.  Most herbivorous fish (and indeed, many herbivorous animals) studied thus far make a living on soluble sugars and polysaccharides in the cell contents of plant cells.  Very few species of fish (e.g., Kyphosus sydneyanus, Odax pullus) have been shown to heavily utilize microbial (endosymbiont) digestion, and it occurs in the hindguts of those that do.  Thus, feeding trials examining digestibility are only the “icing on the cake” in terms of showing whether an animal can digest refractory polysaccharides with the help of microbial endosymbionts.

Fourth, the word "endosymbiont" needs to be defined.  In this case, "endosymbionts" [the word "endo" implying "inside" and the word "symbionts" implying that a mutualistic relationship exists between two types of organisms—the host, which in this case is the fish, and the symbionts (the microbes), which provide a service to the host (cellulose digestion), and in return, get a stable environment inside the gut of the host and a constant supply of food] explicitly refer to resident microorganisms (i.e., bacteria, fungi, and potentially protists) residing in an animals digestive tract that are directly aiding in the process of digestion of plant material; i.e., refractory polysaccharides, like cellulose, for which the animal lacks the enzymes to degrade.  This is different from what Swift et al. (1979) called the "external rumen", or the decomposition occurring in nature that can be taken advantage of by detritivorous animalsthose consuming decaying plant and animal matter.  Thus, the microbial digestion in this latter example is taking place outside of the animal's GI tract, and thus, cannot directly be called a "symbiotic" relationship.  There is a big difference between using microbes in your gut to digest cellulose (like a cow) and consuming rotting plant material that has all sorts of goodies on it (e.g., disaccharides, proteins).     

Fifth, the difference between the words “eat” and “digest” must be defined.  “Eat” means to consume.  That is it.  The word “eat” does not imply that whatever was consumed can be digested and utilized in a nutritional manner by an animal.  The word “digest” implies that what was eaten can be broken down (i.e., digested) by hydrolytic enzymes in the GI tract of an animal and absorbed for use in a nutritional context.  To use these two words in a real life example: Parrotfishes eat coral.  There is no denying that the parrotfishes consume a considerable amount of coral (and by this I mean the hard, calcium carbonate reef); gut content analyses show that the fish eat coral.  However, parrotfishes do not only eat coral, and they certainly do not digest coral in their guts.  There simply isn’t any nutritional value to carbonates.  Many parrotfishes are actually detritivores, taking bites of reef to access the nutritional detritus that has collected in the interstitial spaces of the reef (Crossman et al. 2005).  Much in the same way, the wood-eating catfishes “eat” a considerable amount of wood (~70% of their intake).  However, they poorly “digest” the bulk wood (the cellulose and lignin of the wood proper), but efficiently digest detritus and wood degradation products (e.g., disaccharides).  I hope this analogy helps.  Wood is the reef in the Amazon, providing critical habitat to so many fish species.  Wood-eating catfishes, therefore, are the parrotfish of the Amazon, not the beavers of the fish world (as I commonly end my talks).  I just want to make it clear that I never said that Panaque do not “eat” wood.  They certainly do.  The wood-eating catfishes just poorly “digest” wood itself.                  

Most people seem to be caught up on the feeding trial that I performed for the article: Inside the guts of wood-eating catfishes: can they digest wood? J Comp Phys B 179: 1011-1023 [pdf] 

There are some central themes to the criticisms that I would like to address.   

The main point is that water oak wood does not represent wood the fish would encounter in the Amazon.  This is entirely true, and I acknowledge that in the above referenced article on page 11 and in the PFK article where I call for more research on the effects of tannins on these fish.  That being said, some comments went as far as to suggest that I should have used monocot “wood” for my feeding trials based on an observation of Panaque collected on a fallen palm tree in Brazil.  

This observation largely depends on where one collects Panaque.  I went on an extensive collecting expedition with seven ichthyological experts (you can e-mail me and ask with who I collected fish) in the Río Marañon in northern Peru, where wood-eating catfishes are most diverse and abundant.  We NEVER saw the fish feeding on palm or bamboo, another monocot tree.  In fact, as I say in the above referenced article (Page 6, under the subheading “fiber digestibility and gut transit”), the fish appeared to feed non-discriminately on wood from any fallen tree. 

All decaying trees had grazing scars on them, except for the monocots, which were not very abundant.  Furthermore, the wood the fish were feeding on was heavily degraded to the point that one could break off pieces of wood with their bare hand.  Even hardwoods can become soft over time if they are saturated in water and are heavily degraded.  Thus, except for the tannins, which may be a significant problem (not sure yet), my choice of equally degraded water oak wood (which I could break, when wet, with my fingers) was not out of line.  Degraded wood is loaded with microbes and microbial degradation products (e.g., disaccharides).   

If the argument devolves to the point that the wood-eating catfish require a certain type of wood, that would demand that the wood-eating catfishes be specialists on a specific wood type.  This simply isn't the case in nature and this constitutes "hand waving" to explain why a particular belief is not upheld by empirical study.  If the fish are digesting wood via an endosymbiotic community, they should be able to show growth on many different types of wood in captivity, which has never been shown.  Termites, for example, will show growth and reproductive output on many types of wood, although some wood types (e.g., pine) are not as nutritious as others (Morales-Ramos and Rojas 2003).  Nevertheless, the termites were able to digest the pine with the aid of their endosymbionts and grow and reproduce (just at a lower rate). Thus far there is one anecdotal account (i.e., a personal observation) of growth on a wood-only diet by Panaque maccus in captivity.   Empirical evidence of this account has never been published, nor has it been replicated.  I understand that the real problem people have with this previous observation (and my own digestibility study showing the opposite) is that fishes obtained via the aquarium trade were used.  Hence, there may have been "endosymbiont loss" during shipment of the fish from South America.  However, the digestive enzyme, fermentation, and electron microscopy data from wild-caught fish also do not support the presence of a large, cellulose-digesting endosymbiotic community residing in the fishes' guts.  What I would find believable is that wood used in captive feeding trials thus far have not replicated the critical microbial biomass found in wood Panaque consume in nature, and also have lacked the presence of amorphous detritus (i.e., detritus that is not of "wood origin" and is higher in protein) to provide that extra needed protein boost.  But this argument only supports my findings.... the fish are reliant upon microbial degradation in the environment, not in their guts.       

There is no such thing as "monocot wood".  Monocot plants, like palms, do not show secondary growth like dicot plants, which produce true wood.  Because palm does not contain true wood it may be easier for microbes to degrade palm in a riverine setting, and the stringy structure of palm would certainly act as a trap for non-wood detritus.  So, just because someone sees Panaque on a palm doesn’t mean that the fish prefer it, and, because palm doesn’t contain true-wood that would mean that the “wood-eating catfishes” wouldn’t be eating “wood” when consuming palm.  Panaque are known to like "hearts of palm" in captivity, but palm hearts are apical meristem tissue, which are comparatively low in fiber and richer in soluble polysaccharides, proteins, and lipids.

One comment compared the Panaque to Pandas…. Perfect.  That is a perfect comparison of an animal with a LONG intestine, rapid gut throughput, and little gastrointestinal fermentation (Dierenfeld et al. 1982; Stevens and Hume 1998).  Thus, Pandas show poor cellulose digestibility (Dierenfeld et al. 1982).  That would be analogous to what I saw with the loricariids.  But, Pandas don’t gnaw on wood like the Panaque do; Pandas tend to eat the leaves and shoots of living bamboo, which have more soluble polysaccharides and sugars in them than degrading wood has.  Also, I am not sure one can call bamboo “soft”.  People commonly use bamboo for weaponry and flooring, so it is hard, too. 

A comment noted that the P. nigrolineatus in my study lost less weight and excreted less nitrogen than Pt. disjunctivus, a detritivorous loricariid, on a wood diet.  As I say in the above referenced article (Page 11), “Individuals of Pt. disjunctivus likely did worse from weight and fecal nitrogen loss perspectives than individuals of P. nigrolineatus because the former lack the spoon-shaped teeth necessary to gouge wood in significant quantities (i.e., they had lower daily intake rates of wood). This is also supported by the observation that it took 6 weeks for Pt. disjunctivus to produce amounts of feces that P. nigrolineatus produced in 4 weeks.”  Of course the Panaque did better—they were eating more because they have the correct teeth for gouging wood.  In a subsequent study using stable isotopes (German and Miles, 2010), I ground up the wood before feeding it to Pt. disjunctivus.  They were able to assimilate soluble components of wood detritus, but not the intact cellulose.  Furthermore, seeing more or less weight loss doesn't suggest much.  Fish don't show weight loss like terrestrial animals because lipids and/or proteins burned for fuel are replaced by water in fish.  Thus, a fish can be totally starving but lose little weight (i.e., wet weight).  For example, as a negative control for my stable isotope study I starved Pt. disjunctivus for 150 days (before you think I am cruel please understand that plecos can endure starvation in the wild for up to six months, and I did this under the watchful eye of veterinarians at UF), and the fish lost (average ± standard deviation) 3.67 ± 1.49% of their body mass.  Amazingly, only two fish died during this experiment, and one of them was a "control" fish consuming algae discs, so I can't be sure on the cause of death. So, just because P. nigrolineatus lost "less weight" than Pt. disjunctivus doesn't mean the former were not functionally starving. 

When looking at the loricariid phylogeny it is parsimonious that the spoon-shaped teeth arose to allow the fish access to a resource (wood detritus) that isn’t accessible to loricariids with villiform teeth.  But, it is poor inference to assume that this means that the wood-eating catfishes are “eating” AND “digesting” wood.  There is fierce resource competition in any environment.  In fact, theory predicts that when competition for surface resources becomes strong enough, then digging for resources becomes a viable strategy (Richards 2002).  Most loricariids that are sympatric with wood-eating loricariids are grazers, which scrape and suck the surface detritus from the river bed and from the surface of wood.  Meanwhile, detritus is collecting in the interstitial spaces of wood, not to mention that the wood itself is organic and is being degraded, and thus represents an additional resource.  With so many loricariids in South America, competition for the surface resources is probably fierce enough to elicit digging as a viable foraging strategy.  Hence, we see spoon-shaped teeth emerge twice (independently) in the loricariid phylogeny.  Nathan Lujan, Kirk Winemiller, and I have a paper in press at Functional Ecology in which we show isotopic and morphological evidence for niche partitioning among wood grazing lorcariids. See the pubs page for this article (when it becomes available).           

Another comment implied that a long gut—which loricariids have—equates to long transit time of food in the gut and thus, a reliance on gastrointestinal fermentation by endosymbiotic microbes.  This is absolutely incorrect.  I have been studying fish digestion for 10 years now.  One conclusion that I have come to (and I am not alone) is that a long gut actually equates to rapid gut transit.  This is predicted theoretically (Sibly and Calow 1986; Horn and Messer 1992) and has been observed in other animals, like rats eating a low-quality food.  High intake equates to rapid gut transit and a lengthening of the gut, not longer retention time.  Hence, detritivores (e.g., mullets, gizzard shad, minnows, some chaetodontids, and loricariids), with high intake, have some of the longest guts measured, and transit times less than six hours (Horn 1989; Hood et al. 2005; German 2009).  Furthermore, they have little GI fermentation occurring in their guts.  Some of the fishes with the highest rates and levels of GI fermentation (e.g., Odax pullus; Mountfort et al. 2002) do not have long guts (~2X their body length), but do have relatively long retention of food in their guts (12-20 hours; Clements and Rees 1998).  So please, stop perpetuating the rumor that a long gut implies long retention time, it simply isn’t true.

A comment was made that fast egestion may allow the Panaque to “seed” other fish (perhaps younger fish) with microbes to aid in the digestion of wood.  Although corprophagy is common in ruminants and in insects for the transference of intestinal microbiota to other individuals of the same species (i.e., proctodeal trophollaxis), fast egestion does not imply that is occurring. 

A comment was made that other xylivorous animals (an example would have been nice) have to eat and then move from place to place, so they hold their food in their guts longer in comparison to Panaque, which sit on their food, thus allowing them to eat all day and have fast egestion.  Termites (and most xylivorous animals; look it up) live with their food, too, so that is a bad example. Fast egestion with no selective particle retention equates to little microbial (i.e., endosymbiont) digestion. 

I never said the catfishes efficiently absorb “fermentation products”.  Disaccharides—which the fish efficiently digest—are NOT fermentation products, like the SCFAs acetate, propionate and butyrate, the concentrations of which are low in the intestines of the wood-eating catfishes.  Disaccharides are produced by extracellular digestive enzymes secreted by microbes onto wood that hydrolyze larger poly- and oligosaccharides under environmental conditions (aerobic or anaerobic).  Fermentation implies the partial oxidation of a monosaccharide (e.g., glucose, mannose) under anaerobic conditions, which result in reduced byproducts like SCFAs.  SCFAs are transported across the gut wall by different transporters (an anion exchanger; Titus and Ahearn 1988) than glucose (or other monomers) would be.

To address it again.....there was a concern that I used fish from the aquarium trade for my study.  This is only true for the P. nigrolineatus for the digestibility study.  All other fish I studied were wild-caught.  So, all of the gut size, electron microscopy image, digestive enzyme data, etc., were from wild fish that I caught in Peru or Florida.  Because all of that data says the same thing (these fish are not reliant upon endosymbionts to digest wood), the digestibility data support my other findings.  Even if you remove the digestibility data, the story doesn’t change and is just as well supported by the physiological and isotopic data from wild-caught fish.  Furthermore, what would microbes cultured from wild-caught fish tell you, especially in light of all of the digestive physiology data?  Unless one provides physiological evidence that the fish utilize microbial pathways to meet their energetic needs, cultured microbes can be a) ingested with food and transient; b) residing in the gut but performing other functions (e.g., vitamin synthesis, detoxification of plant secondary compounds).  

A comment was made that wild-caught fish were found with guts full of wood.  This is true, but gut content analyses performed with the naked eye are not quantitative.  I quantitatively determined the proportions of different gut content items in the guts of wild-caught fish with the aid of microscopes.  Wood is not the only thing these fish eat, as they also had detritus, algae, and diatoms in their guts.  Wood is the most obvious item in their guts and makes up the largest proportion of their gut contents, but it is not the only thing they consume.  See my article referenced above.

One “respondent” went as far to call me “a cocky grad student who is not even a doctorate certified ichthyologist” and therefore, my data are not trustworthy.  I will acknowledge that I can be arrogant (always trying to keep that in check), but who can say they are never arrogant?  The individual that made this comment has contempt for science and the process of earning a doctorate (I wonder if they have one?).  Are they saying that no graduate student produces good, serious work?  Are you kidding me?  Grad students are the leading authors on groundbreaking work every single day.  I am not saying that my work is groundbreaking (far from it); it is almost as if this individual assumes that I did my dissertation work in a dark corner by myself (imagine Golem).  I worked under a strong committee that featured one of the most well-known fish physiologists (Dr. David H. Evans), a prominent nutritional ecologist (Dr. Karen A. Bjorndal), an avian ecologist with strong digestive physiology credentials (Dr. Douglas J. Levey), a fish and poultry digestive physiologist (Dr. Richard D. Miles), and a well-known ichthyologist and PI of the All Catfish Species Inventory (Dr. Larry M. Page).  Not to mention all of the other grad students and undergrads that helped along the way.  Moreover, all of the catfish articles (German 2009; German and Bittong 2009; German et al. 2010; German and Miles 2010; Lujan et al. 2011) were peer-reviewed.  That means that they were read and critiqued by other nutritional physiologists.  So, if you think I was just a grad student that did crap work for his dissertation, well that implies that I did crap work and slipped it by each of my committee members and the reviewers of the manuscripts.  Please e-mail any one of my committee members and see what they think of my project.  Plus, I did earn an actual doctorate in zoology from the University of Florida.  That does make me a “board certified” biologist.

Literature Cited

Clements KD, Rees D (1998) Preservation of inherrent contractility in isolated gut segments of herbivorous and carnivorous marine fish. Journal of Comparative Physiology B 168:61-72 

Crossman DJ, Choat JH, Clements KD (2005) Nutritional ecology of nominally herbivorous fishes on coral reefs. Marine Ecology Progress Series 296:129-142

Dierenfeld ES, Hintz HF, Robertson JB, Van Soest PJ, Oftedal OT (1982) Utilization of bamboo by the giant panda. Journal of Nutrition 112:636-641 

German DP (2009) Inside the guts of wood-eating catfishes: can they digest wood? Journal of Comparative Physiology B 179: 1011-1023

German DP, Bittong RA (2009) Digestive enzyme activities and gastrointestinal fermentation in wood-eating catfishes. Journal of Comparative Physiology B 179: 1025-1042

Hood JM, Vanni MJ, Flecker AS (2005) Nutrient recycling by two phosphorous-rich grazing catfish: the potential for phosphorous limitation of fish growth. Oecologia 146: 247–257

Horn MH (1989) Biology of Marine Herbivorous Fishes. Oceanography and Marine Biology Annual Review 27:167-272

Horn MH, Messer KS (1992) Fish guts as chemical reactors: a model for the alimentary canals of marine herbivorous fishes. Marine Biology 113:527-535 

Karasov WH, Martínez del Rio C (2007) Physiological ecology: how animals process energy, nutrients, and toxins. Princeton University Press, Princeton, NJ USA

Morales-Ramos J, Rojas M (2003) Nutritional ecology of the Formosan subterranean termite (Isoptera: Rhinotermitidae): growth and survival of incipient colonies feeding on preferred wood species. Journal of Economic Entomology 96:106-116 

Mountfort D, Campbell J, Clements KD (2002) Hindgut fermentation in three species of marine herbivorous fish. Applied and Environmental Microbiology 68:1374-1380 

Richards SA (2002) Temporal partitioning and aggression among foragers: modeling the effects of stochasticity and individual state. Behavioral Ecology 13:427-438 

Sibly RM, Calow P (1986) Physiological Ecology of Animals, an Evolutionary Approach. Blackwell Scientific Publications, Oxford, England 

Skea G, Mountfort D, Clements KD (2005) Gut carbohydrases from the New Zealand marine herbivorous fishes Kyphosus sydneyanus (Kyphosidae), Aplodactylus arctidens (Aplodactylidae), and Odax pullus (Labridae). Comparative Biochemistry and Physiology Part B 140:259-269 

Skea G, Mountfort D, Clements KD (2007) Contrasting digestive strategies in four New Zealand herbivorous fishes as reflected by carbohydrase activity profiles. Comparative Biochemistry
            and Physiology Part B 146:63-70

Stevens CE, Hume ID (1998) Contributions of Microbes in Vertebrate Gastrointestinal Tract to Production and Conservation of Nutrients. Physiological Reviews 78:393-427

Swift MJ, Heal OW, Anderson JM (1979) Decomposition in terrestrial ecosystems. University of California Press, Berkeley, CA, USA

Titus E, Ahearn GA (1988) Short-chain fatty acid transport in the intestine of a herbivorous teleost. Journal of Experimental Biology 135:77-94

Last modified 5 July 2011

My response to the Practical Fishkeeping Magazine Interview