Dead genes for taste in carnivores

March 15, 2012 • 5:09 am

Ceiling Cat knows that we hardly need more evidence for evolution, but of course it just keeps pouring in, for Evolution is True. The latest piece—and it’s a nice one—is a report of “dead” genes (i.e., existing but nonfunctional genes) for taste receptors in some carnivores.

In WEIT, I mentioned how the presence of dead genes in some species, which have been inactivated by mutations and don’t make a protein product, can be explained only if those genes were active in an ancestor.  Such genes include olfactory receptors in aquatic mammals (who don’t need them, since they can’t smell in air) and the final gene in the pathway for synthesizing vitamin C, which is dead in humans (we get enough vitamin C in our diet) but functional in our mammalian relatives.  No creationist “theory,” save that of a Cosmic Trickster, can explain why a designer would put nonfunctional genes into the genomes of plants and animals.

The latest discovery is given in a paper published in PNAS (reference below) by Peihua Jiang and colleagues from the Monell Chemical Senses Center in Philadelphia and the Anthropological Institute and Museum at the University of Zurich. They sequenced genes for taste-receptor proteins in 12 mammal species and, combining this data from earlier studies, shows that several species of carnivores have inactivated taste genes, and presumably can’t experience the taste sensations which those genes ultimately produce. And the particular genes that are inactive are ones that aren’t much use given the species’ diets.

Scientists have found that mammals can perceive five taste qualities:

  • sweet
  • bitter
  • umami, a “savory” sensation involving the detection of certain amino acids and ribonucleotides
  • salty
  • sour

Each of these sensations has its own taste receptors in the “taste buds,” and the sweet, bitter and umami sensations are detected through the conjunction of foodstuffs with special receptor proteins in the buds.  (Salty and sour detection works differently, through the direct action of food ions on the nervous system.)  The researchers worked only with sweet, bitter, and umami genes, since those have well-defined functions.  The genes for these have complicated names, but we’ll refer to them as S-, B- and U-proteins respectively.

The researchers sequenced some or all of these three genes in 12 species of mammals: the aardwolf, the Canadian otter, the spectacled bear, the raccoon, the red wolf, the sea lion, the fur seal, the Pacific harbor seal, the small-clawed otter, the spotted hyena, the banded linsang, and the fossa.  Here are the three you might not know:

The aardwolf, Proteles cristata, which eats insects
The fossa (Cryptoprocta ferox), a catlike carnivore, related to civets, that's found in Madagascar. It eats lemurs, rodents, and other animals.
The banded linsang (Prionodon linsang), a carnivorous civet from southeast Asia

We have known for a while that housecats and other felids lack functional sweet receptors, so when you feed your kitteh ice cream, it’s going for the dairy taste, not the sweetness! The authors predicted that mutations that inactivated sweet receptors might be seen in carnivores other than cats, especially sea lions and dolphins, which have “an atrophied taste system, exemplified by few taste buds present in their lingual epithelium.”  Those two groups also swallow their food whole, which the authors postulate would “minimize opportunities and needs for taste input.”  If you are evolved to go after anything that looks like a fish, then you don’t need to know how it tastes.

The general result is that taste receptor proteins are inactivated if the animal either doesn’t encounter that taste in their diet (and hence don’t need the receptors to ensure that they’re eating the right thing), or they eat by swallowing and not chewing chunks of meat, so there’s not a real need to taste the food:

  • Sweet receptor genes that produce S-proteins are inactivated in the linsang, the fossa, the spotted hyena, the sea lion and fur seal, the Pacific harbor seal, and the asian small-clawed otter.  All of these are either carnivores or piscivores (fish-eaters), and hence don’t encounter many sugars in their diets.
  • There were all kinds of mutations inactivating the S-genes of these species: stop codons, insertions, nonsense codons, and deletions.  And in each lineage the gene was inactivated independently since the disruptive mutations were different in different species. (The only exception was the S-genes in sea lions and fur seals, which are sister species who shared inactivating mutations that presumably occurred in their common pinniped ancestor.  Since these genes don’t have much use in a meat- or fish-eater, mutations inactivating them presumably paid no selective penalty, or could have even been advantageous if they produced an “expensive” protein that was no longer needed.  Earlier work also showed that the bottlenose dolphin also lacked functional S-genes.

Here’s a phylogeny of 18 species, with the red diamonds showing those species having inactive sweet-receptor proteins:

  • The authors did taste tests on the otter and the spectacled bear to see if they could taste sugar: these involve giving the animals dishes of water containing sugar as well as control dishes containing either plain water or non-sweet chemicals.  The otter had no preference, as one might expect since it has no sweet-receptor proteins, while the bear showed a strong preference for sugars (the bear is largely herbivorous and noms fruit, berries, cacti, and other plants with sugars).
  • In sea lions and bottlenose dolphins, the umami receptors are also “dead,” and in the bottlenose dolphin the bitter gene also seems to be permanently broken. Those animals don’t chew their food, but rip off chunks and swallow them.

These results not only strengthen the already-airtight case for evolution and common ancestry, but show that whether taste genes are active or inactive depends on the animal’s diet, as one might expect. It also shows that gene loss in different lineages has occurred independently. I look forward to the Discovery Institute explaining this one (no worries—they’ll have a cockamamie explanation soon).

An interesting speculation, floated by my colleague Andrew Berry, is that the mutations inactivating taste receptors and olfactory receptors in humans (yes, we have some) have spread not because inactivated genes are selectively advantageous, but are actually “neutral” (i.e., have no effect on reproductive fitness). If that’s the case, then taste and smell genes could be polymorphic: some individuals might have active copies, other inactive ones. (“Neutral mutations” can hang around for long periods of time in a population.) This might explain why different people can taste and smell different things, and why some people, like wine- and perfume-testers, have extraordinarily keen abilities to detect scents and flavors.  We know that there is polymorphism for genes affecting, for example, our ability to taste PTC, and for “how” we taste stuff like cilantro (I hate the stuff, as it tastes like soap to me; others love it).

As a sideline, the authors note that the sweet-receptor genes have also been inactivated in chickens, vampire bats, and “tongueless Western clawed frogs.” Taste tests show that neither chickens nor vampire bats can perceive sweetness (no sugar in blood!) but I found this sentence endearing:

“It is yet to be established how Western clawed frogs respond to sweeteners.”


Jiang, P., J. Josue, X. Li, D. Glaser, W. Li, J. G. Brand, R. F. Margolskee, D. R. Reed, and G. K. Beauchamp. 2012. Major taste loss in carnivorous mammals. Proc. Nat. Acad. Sci. USA  online early edition,

55 thoughts on “Dead genes for taste in carnivores

  1. “… cilantro (I hate the stuff, as it tastes like soap to me; others love it).”

    My wife hates it too! Over the past few years in the UK, restaurants seem to have become obsessed with putting it in everything… (Confusingly, they refer to it as coriander, which I always thing of as the spice [ground seeds] rather than the leaves.)

    But quite apart from this culinary tidbit, a fascinating article.


      1. Stupid Brits! Fancy calling it something which is almost exactly the same as the genus name – the Spanish name is obviously so much less confusing. 😉

    1. They tested only vampire bats. I said that in the initial post, but have clarified it further in the second sentence of the conclusions. Thanks.

  2. Blood doesn’t have sugar? That contradicts what I learned (and have taught students) in physiology class. The glucose concentration in blood might be too low to taste.

    1. Yes, of course you’re right. What I meant is that vampire bats drink only blood, that it’s not sweet, and they don’t need to discriminate among foods because they are adapted to eat only a single thing.

      1. Thanks for the clarification, Dr. Coyne. Apologies for being pedantic, I see the point you were making.

  3. My personal experience with cilantro is not whether it tastes like soap or not but that I don’t mind that it tastes like soap. My wife hates the stuff with a passion except when she has a cold and can’t smell it.

    1. That is an interesting observation. Disposition is one thing, attitude is another. I wonder if that is true for animals too?

  4. I believe some mammals, including humans, also have taste receptors for lipids. It is thought that the functioning of these receptors help explain why many fat substitutes produced by food companies are not as satisfying to most people as the real thing. I’m not sure if anyone has looked for functional polymorphisms.

    1. I think you’re thinking of CD36, a known lipid-binding protein which is suspected of also being a taste receptor for fat. This was reported just a month or so ago.

    2. The ability to taste fat is the latest thing in taste science, two of my friends are doing PhDs on the subject as we speak.

  5. I love the stuff on certain foods only, such as Asian or Mexican (eg. guacamole and ceviche). The soft leafy tops are called cilantro. The stalks and roots are coriander, and the seeds are coriander seeds.

    I cook a lot and would love to cook for Jerry some time, as I’ve noticed he’s even fond of Tree Ears!

  6. This article made me realise that I’m unclear on what it means to sequence a gene. Could someone offer a brief explanation or point me to some reading matter which could help a non-scientist better understand?

    Another question: I can see the point of an animal losing the ability to taste things it no longer eats (Use it or lose it on a whole nother level!?), but what if an animal needs to taste or smell things it shouldn’t eat?

    1. BC,

      DNA sequencing means determine the entire sequence of nucleotides in a stretch of DNA. One uses the DNA of interest to make new DNA that incorporates fluorescent markers for each of the nucleotides and then reads the sequence off with lasers in fancy expensive machines

      I think our senses are tuned to potentially harmfull things. One of the reasons we tend not to like very bitter foods is that bitter foods often contain compounds called alkyloids. Plants put toxic alkyloids in their tissues to discourage animals from eating them. They also try to trick animal senses: hot peppers arent actually hot but capsaicin seems hot because it binds to heat receptors.

      Evolutionary Psychology is a favorite target of IDers and many scientists since it borders on quasi-science. I tend to agree with the latter but one of the best supported studies of EP concerns taste preferences of pregnant women. Pregnant women may have their taste preferences ‘tuned’ during pregnancy to discourage them from eating foods that may be fine for adults but might risk birth defects in the fetus. A disclaimer: I read about this years ago so it might no longer be valid


    2. Thanks, Rod. I knew it had to do with getting a picture of the sequence in a stretch of DNA but I also somehow had it in my head that it had something to do with determining the link between a sequence and a particular structure or capacity in a grown organism. People seem to talk about those things in the same breath a lot so I guess that’s what has me a bit mixed up. Is there special term for studies that make the links between particular sequences and what they produce?

      Here’s another question, and I’m sorry if it’s another clueless one: What is the difference between a dead gene and junk DNA?

      1. BS

        Once one has the sequence its very easy to get the sequence of the protein by using the rules of translation. Once one has the sequence of the protein one can figure out the function of its parts or the whole by comparisions to known proteins and/or by doing lab experiments. In the paper JAC discussed, its very easy, once one has the sequence of the taste receptor gene to see how changes in the sequence could destroy the function of the protein.

        Dead genes are a subclass of junk DNA. In my opinion ‘junk DNA’ is still a valid term despite what IDers claim. While its true that some junk DNA has funtion ( and so it isnt really junk) its very clear the these functions are secondary gains-of-function in a very small fraction.
        Most ‘junk’ is repetitive DNA which increases or decreased in the genome as a reult of the sloppiness of processes involved in egg and sperm formation, or transposable elements or jumping genes = these are sequences that just multiply on their own – like viruses without a coat. A small number wind up being beneficial; having fucntion. A greater number cause disease and the vast majority have no effect.
        Another disclaimer – this is my take on it. Experts in the field have some disagreement over those last points

    3. “Once one has the sequence of the protein one can figure out the function of its parts or the whole by comparisions to known proteins and/or by doing lab experiments.”
      I’m going to have to read up about how exactly that is done. It really amazes me.

      Interesting what you say about junk DNA. I wonder if it gets called that because it is of no use to the individual or the species in it’s current state, or if some consideration is given to it’s potential use in the evolutionary process.

      Thanks for your explanations. I feel quite daunted by all my own questions! I wish I had more time to read!

  7. Thanks for that paper! We do the bitter tasting polymorphism excercise in our 1st year bio lab – that paper will make great reading for the write-up.
    There are also polymorphisms associated with olfaction. I know this from experience. In the 80s I drove across country with a friend were I discovered that I’m completely oblivious to certain odors that he perceived strongly (I wont elaborate).

    As for creationists, this wont even cause them to break their stride. YECers will just say the mutations happened after The Fall. IDers will say ‘sure, random mutations inactivated those receptors, but it takes a Designer to make the tongue and taste-buds and receptor pathways in the first place’


    1. Yes, when I brought up the Vitamin C (GULO?) issue with my creationist friend, he speculated that it may have been the designer’s plan to create the cravings for certain foods and thus allow them to experience the pleasures of eating those foods.

      1. His rationale makes no sense. The vit C knockout doesn’t create cravings, it just means if you don’t get enough vit C you get scurvy and die. Like most creationist arguments the arguments are incompatible with any reasonable designer and reduces to an ad hoc- god works in mysterious ways- type of argument.
        Our taste receptors are so we can have the pleasure of eating? Presumably the creator gave vampire bats their receptors so they’d have the ‘pleasure’ of drinking blood

        1. We grew up in the same religious sub-culture, so I identify with that compulsion to do intellectual battle for Jesus, throwing anything you’ve got, however unreasonable, at threatening evidence. God works in mysterious ways…who can know the mind of god?…the great cosmic conflict of good and evil…the fall of man and influence of the devil (vampire bats may fit well here)…and we’ll all understand by and by – very difficult to break out of that way of thinking.

    2. Rod, I have little sympathy for the ID people and I’m loathe to disagree with you but those vampire bats – I simply cannot imagine them drinking blood for any other reason than the pleasure or satisfaction of it. The only alternative to the bat being driven by desire and the satisfaction/pleasure of fulfilment is for the bat to have reasons related to the benefits of blood … and we all know how poor a motivation reasons can be.

      ID people are like the Flat Earth Soc, no matter what you say, they will construe it to support Design. Even if it’s sloppy, consists of a lot of junk, they can simply say it shows God is more like Van Gough than Da Vinci.

      In their favour, I’d say I don’t take them to be merely using Design to defend the existence of a Designer but also to be defending that sense of wonder we have for the universe. I think they have a hard time seeing how they could still have that wonder were the universe not deliberate. One is ordinarily more impressed by stuff that is well made (design) than by stuff that gets found on the beach (chance). I’m still trying to bend my own brain around that. What do you think?

      1. BC,

        Well, yes I do think vampire bats get some degree of pleasure in drinking blood – in proportion to the complexity of their emotions. What I think is ridiculous is explaining that as God creating that pleasure.

        Its great that you’re so interested in this stuff. I dont think its really possible to learn in it bits and pieces like this though. I’d suggest that either you start reading textbooks on biology or you find good online lectures. There are some great ones at Want me to make a few recs?

    3. Indeed, FWIW, SDAs believe everything was vegetarian prior to the fall, and so they would specifically expect to find such degeneration of initial genetic information to be associated with the descent of carnivores.

      1. Yep, you’re right…that’s the religious subculture I grew up in. No evolution education in SDA schools either, so I’ve had a lot of catching up to do. Lucky to have this and other sites.

  8. I can proudly and smugly say that I have seen the first two of those mammals in the wild; aardwolves many times. And I love cilantro/coriander.

  9. Wow, i love the photos in this post. And youre right, the sentence about the frogs and sweeteners is great. 🙂

    Gee, my comment is not very cerebral.

  10. Extremely interesting post, thanks. The ‘Cosmic Trickster’ may sound like a joke to us, but didn’t Phillip Johnson suggest (apparently with a straight face) that such things as peacock’s tails were created by Yahweh for his own amusement?

  11. Quite odd about linsangs since other Viverrids like palm civets and bintirongs have so much fruit in their diet. I also seem to remember a more crude taste experiment involving red pandas and aspartame.

    By the way, I would classify a linsangs as a Prionodontidae (closer to the cats) rather than a civet (I assume you mean Viverrid) and the fossa I’m pretty sure is now in the mongoose classification.

  12. Very interesting post…thanks! I’m wondering if/how a tongueless frog can taste anything. Probably showing my ignorance.

  13. The banded linsang is in its own family, the Prionodontidae, and not a civet. Prionodon linsang and Prionodon pardicolor are the only two species. They are both from south-east Asia. Prionodontidae is the sistergroup of the Felidae, the cats.

    This was the result of molecular phylogenetic work, by Gaubert & Veron in 2003. Actually, Gaubert & Veron relate that Horsfield,the official describer of the banded linsang in 1821, thought the animal near to the cats on morphological grounds. However, the two Asiacatic linsangs generally ended up in the general Feliform family Viverridae. Linsangs might very well be thought of as proto-cats.

    The African linsang is however a viverrid.

  14. Mrs DiscoveredJoys loves Brussels Sprouts. I find they taste horrid, horrid, horrid. I understand that the explanation for this is our differing sensitivity to particular sulphur compounds.

    However ‘red’ Brussels Sprouts were available this Christmas and were claimed to be ‘sweet’. I found them acceptable in small quantities.

    Isn’t science (and horticulture) wonderful!

  15. This being the third article/blog I’ve read on this topic, a distantly-related question came to mind, considering the comment about dolphins not needing to taste their food.

    My understanding is that Great White Sharks can tell the difference between humans and seals. They use a torpedo-like attack on seals, but generally not on humans. One hypothesis was that they recognize that humans are not seals (by sight or smell), and bite off a leg or arm to “taste” whether humans are appetizing or not.

    So the question is, do GWS have much in the way of taste-receptors? Or is taste important to them at all (considering all the other inedible junk they’ll swallow)? If it isn’t, what are the other reasons for the difference between the attack methods, considering the size and swimming profile of humans and seals aren’t that different?

    1. From reading around I understand that sharks primarily use their sense of smell at long range & the sensing of electric charges (prey muscle contractions) at short range with visual cues at ranges in between. They can also sense changes in water pressure.

      I am guessing that an attacking shark cannot tell the difference between a human, seal & surfboard until he is already on the final attack phase with jaws open & eyes shielded. Perhaps he veers off at the very last moment if the target isn’t generating a ‘normal’ electric field?

      Just guessing you understand. It would be interesting to hear from an expert.


      1. In general, sharks merely “test bite” humans and generally reject us. Alas, the test bites of many sharks are enough to cause serious damage. If they really wanted to kill us they could do so easily.

        I had an interesting encounter with a shark about a half mile from San Luis Pass in Galveston. Midday in February, sunny but murky as an early spring front had just blown in. Water was chilly, but probably warmer than the California coast in early June.

        Felt a hard bump followed by a brush that felt like fine sandpaper. Saw a large dark shape moving away, about 5-6 feet long. It was either a large lemon shark or a medium sized bull shark. I didn’t leave the water immediately, but about ten minutes later I saw huge schools of croaker, a few black drum and a couple of scout dolphins. I figured a feeding frenzy was about to begin and that for each dolphin there were many sharks. I got out of the water fast.

  16. “The researchers worked only with sweet, butter, and umami genes, since those have well-defined functions.”

    In my case, the butter gene does have well-defined functions. Apparently in your case it overrides your finger control.

  17. Yet another question: How do the scientists who study this stuff distinguish between an animal having absolutely no capacity to taste something and an animal simply not having learned it yet? I’m sure we are all familiar with learning to identify tastes and smells we might previously have been unaware of. To put my question differently, how do they know it’s a dead gene / totally lost capacity and not just a case of poor awareness on the part of the person or animal being tested? Sorry if that’s a stupid question.

    1. I would guess it has to do with whether one can make recordings (patch-clamp?) from the specific receptors. If the receptors don’t fire when they contact a particular substance, it may not be perceivable by gustation.

      1. Patch clamping is for ion channels. However, many receptors activate channels (including olfactory receptors), so this could potentially be used in those cases. Some taste receptors are channels themselves, and some activate channels through second messengers.

    2. Taste and odorant receptors are encoded by genes. If the gene becomes mutated, the effect on the protein can be devastating. For example, the protein can become too short (premature stop codon) or can have a region in which amino acids are changed either singly (non-synonymous mutation) or in large number (e.g., an indel (insertion or deletion) causing a frame shift mutation). Nearly three quarters of human odorant receptors have become pseudogenes through mutation.

  18. I’d just like to point out that no sugar in blood would be an abnormal (even lethal) condition. There may not be much, but the sugars are there.

    1. If there are no sugars in your blood your liver will generate ketones instead. This is an issue if you are diabetic but not otherwise. In normal people this will happen if you don’t eat any carbs, it is the goal of both the Dukan diet, the Atkins diet and many others.

  19. It seems like nature is a sloppy programmer that often just REMs out lines of code rather than removing them from the program like you would expect an intelligent designer would do.

  20. I’ve read before about cats not being able to taste sweetness, but I did not realize it was because of gene inactivation, so this was very interesting.

    But why the heck does my cat like to lick strawberries? Could it be because of some tart/sour taste she can perceive that I don’t because it is overwhelmed by the sweetness?

    My crazy cat has no interest in meat, but in addition to strawberries also likes tomatoes and carrots, and celery makes her bonkers the same way catnip does.

    1. I was wondering the same thing too. I had a cat who loved watermelon and even cantaloupe. I wonder what chemical/taste she was responding too. Hobie even liked cherry lollipops. Cats are such strange creatures.

    2. I had a cat that ate bread! He was a Burmese who always think with their stomachs. I had to buy a bread bin or he would chew a hole in the plastic and eat the corner of the bread. He also ate pancakes but I could understand that on the basis of the butter they were cooked in.

Leave a Reply