Cell Line Infections, or, my dog has no bones

[ I wrote this back in 2002]

Host cell lines are an entirely new class of disease-causing organism. They are strange and unusually disgusting, and if you read this you may end up showering in Lysol for the next few weeks. Consider yourself warned.

A host cell line is a microorganism that was until fairly recently a part of some higher organism – roughly speaking, a contagious cancer. We know of one good example, transmissible venereal tumor, also known as canine venereal sarcoma or Sticker’s sarcoma, a contagious neoplasm of dogs. It is not contagious in the same sense as liver or cervical cancer, which are (usually) consequences of viral infections. In those cases, it is the virus that is infectious; here it is the cancer itself. Viable cells become engrafted onto mucous membranes and grow in the new host animal. Transmission is usually sexual, but licking or inhaling sometimes causes oral or nasal tumors. Chromosomal and genetic studies indicate that all cases of TVT share a common origin – all share a particular pattern of chromosomal rearrangement and carry characteristic insertions.
TVT is a real disease that affects a significant number of dogs in the tropics and subtropics. It is not always fatal; in most cases, especially in adult dogs, it regresses after a few months. This kind of spontaneous remission is very rare in conventional forms of cancer.
TVT is also very, very strange. In fact it is astonishing. It sounds logical enough, and one can see how it might happen. Cancer cells are probably selected for low immunological visibility, and cancer cells that originated on one dog’s mucous membrane might occasionally be transmitted by contact, escape notice by the other dog’s immune system, and grow. Once this happens, natural selection would gradually optimize transmission.
But although the process is not so mysterious, the implications are vast. First, we have here the most abrupt evolutionary transition known, a jump from mammal to one-celled organism in a single step. For this transmissible tumor, also called canine venereal sarcoma, is in every sense an independent infectious organism, just as much so as Salmonella typhi or Plasmodium falciparum. It reproduces and metabolizes. Although descended from dogs, it is genetically different from dogs, with a different chromosome number (57 rather than 78), due to extensive chromosomal rearrangements. Since it cannot, as far as we know, exchange genetic material with dogs, it is a new species.

Although its phenotype differs considerably from dogs (no brain, no bones, no eyes, no fur, asexual) classification by descent clearly implies that it is a canid and mammal – certainly the most unusual mammal ever discovered. And classification by descent is the right way to go. If you try to classify animals by similarity of surface appearance, you’ll get into trouble. Recent molecular work shows that elephant shrews, elephants, manatees and aardvarks descend from a common African ancestor. An elephant shrew looks a lot like a shrew or rat, but it is more closely related to an elephant. In the same way, the organism causing canine venereal sarcoma is more closely related to a wolf than a fox is, even though you need a microscope to examine it.
Are there other cell line infections? There is at least one, a contagious leukemia in hamsters. HeLa, a long established experimental human cell line, comes pretty close. It originated as a cervical adenocarcinoma in Henrietta Lacks in 1951, and is wildly successful in the limited niche of tissue culture. It has repeatedly outcompeted and replaced other cell lines via cryptic contamination, ruining a great deal of research in the process. At one point, most of the supposedly different human cell lines used were actually HeLa. Leigh Van Valen and Virginia Maiorana have pointed out that HeLa should logically be considered a separate species. and they’re completely correct. I think that TVT, as the cause of a common natural disease, is even more interesting, since its existence in nature suggests that there may be many such cell-line species. The existence of a whole new category of disease organism, of the same level of generality as bacteria or viruses or protozoa, would be of the first importance.
Why do I think that there are probably many other host cell line infections? First, TVT exists in dogs, and dogs are really not all that unusual, except for their close association with humans. That close association probably does not make dogs especially susceptible to this kind of infection, but it greatly increases our chance of noticing it. We know much more about the diseases of domesticated animals than we do about diseases of wild species. The transmission modes for TVT are not particularly unusual and should work in many species. For all we know, many species have one or more cell line infections.

Infectious cell lines may not be very noticeable – it seems possible that they might pass for host cells. Certainly people have not been looking for such things. In addition, diseases caused by infectious cell lines would not have to look or act anything like cancer. TVT itself hints at this; even if you think of TVT as a cancer, forgetting its infectiousness, it’s pretty unusual, mainly because of that high rate of spontaneous remission.

The right way to understand this is by considering the evolutionary pressures on an infectious cell line, which are very different from those that shape ordinary, non-transmissible cancers. In conventional cancers, the cells that grow the fastest win out – if you call being the first to kill the host winning. The amount of possible evolutionary change in a nontransmissible cancer is limited, since all such change must occur within one host’s lifetime.
Infectious organisms, including transmissible cell lines, often do best by adopting more moderate strategies. ‘Grow like mad’ is often not the best approach; it is likely to harm or kill the host, reducing opportunities for future transmission. Infectious organisms usually adopt strategies that involve slower-than-maximal reproduction, allowing extended host survival. Infectious organisms that depend on host mobility for transmission, rather than some vector, are especially likely to allow prolonged host survival. Thus we see that some infectious diseases (like the rhinoviruses that cause the common cold) are mild and never kill the host, some persist indefinitely while doing little harm to the host (Herpes I), others, like HIV, kill quite slowly. Diseases caused by long-established cell line infections might look like any of these, but would probably not kill quickly. Diseases caused by long-established cell lines should look more like other transmissible diseases (tuberculosis, say) than cancer.
We can make some predictions about the likely characteristics of cell line infections. First, infectious cell lines would grow fairly slowly. Viruses and bacteria have doubling times as low as 20 minutes, but cells of higher organisms take considerably longer. They would probably not be very hardy. Some bacteria and viruses can survive for long periods in the exterior environment – anthrax spores can remain viable for decades – but it is hard to see how an infectious cell line could do the same, unless it has existed for a long time and diverged very far from its ancestral cancer.
Even though infectious cell lines would probably not be very durable, there are still a number of possible transmission routes. From the example of TVT, we already know that sexual and oral transmission are both possible. From what we know of venereal disease, sexual transmission usually implies vertical transmission as well. Respiratory transmission is probably possible, since aerosols can transmit HeLa. If the host cell line has existed for a long time, long enough for substantial adaptation, we might see vectorborne transmission by arthropods. Aquatic organisms, especially filter feeders like clams, might manage to transmit cell line infections short distances through water (and there is a contagious leukemia of clams that looks very suspicious).
Cell line infections would probably not be susceptible to antibiotics; since antibiotics take advantages of the differences between host and parasite, things like the bacterial cell wall or the bacterial ribosome. Here those differences would be very small, and it would be difficult to find chemotherapeutic agents that hit the host cell line infections lots harder than the host. This could be an important epidemiological clue.
Cell line infections may be capable of very complex manipulations, since they start out with the ability to make every hormone and signal chemical. In other words, they start out knowing how to push your buttons. They might be able to trick the immune system or change your behavior in ways that maximize transmission.
If cell line infections are reasonably common, it might be that immunological rejection of foreign tissue, which so complicates organ transplants, is a necessary function rather than a side-effect.

The most important practical point is that those few people who are actually looking for causes of diseases should consider this possibility. There are diseases that look as if they might be infectious where no causative organisms has ever been found – diseases like sarcoidosis. They might be caused by some disease that started out as your second cousin Frank. It might be older than that – we might find surviving Neanderthal cell line diseases.

Gregory Cochran

P.S. There are some other organisms and diseases that have some similarities to cell line infections. A number of parasitic organisms have lost so many nonessential functions that it is almost impossible to tell what they once were by any methods other than molecular biology. For example, the organism that causes ‘whirling disease’ in trout is a degenerate jellyfish. And there are known examples in humans where cells from someone else can exist for long periods and, at least occasionally, cause trouble. Many women who have had sons will have a few XY leukocytes, even decades later. We routinely irradiate blood transfusions, because when we didn’t, in rare cases totipotent leukocyte cell lines (usually from close relatives, immunologically similar) proliferated in the transfusee and killed him.

[ 2015] We now know of another example, DFTD, (Devil facial tumour disease), that afflicts Tasmanian devils, threatening them with extinction. And there’s a paper just out in Science confirming that contagious clam leukemia is also a cell line infection.  Told you so.
This entry was posted in Uncategorized and tagged . Bookmark the permalink.

40 Responses to Cell Line Infections, or, my dog has no bones

  1. marcel proust says:

    Although its phenotype differs considerably from dogs (no brain, no bones, no eyes, no fur, asexual) classification by descent clearly implies that it is a canid and mammal – certainly the most unusual mammal ever discovered. And classification by descent is the right way to go. If you try to classify animals by similarity of surface appearance, you’ll get into trouble. Recent molecular work shows that elephant shrews, elephants, manatees and aardvarks descend from a common African ancestor. An elephant shrew looks a lot like a shrew or rat, but it is more closely related to an elephant. In the same way, the organism causing canine venereal sarcoma is more closely related to a wolf than a fox is, even though you need a microscope to examine it.

    I don’t know if you are just being provocative. If not, I offer the following mildly educated opinion and hope that you can show me where I have gone astray.

    Mammals are grouped together in evolutionary trees (clades?), but isn’t it the case that they also share certain functional features: lungs, sexual reproduction, hair or at least vestigial follicles, etc., etc.? Mammals are descended from some kind of reptile – synapsids – and while mammals are the only extant synapsids (according to Wikipedia), that does not make us reptiles. Similarly, TVT may be very closely related to one species of mammal, but the utility of classifying (and that ultimately I think is the value of any classification scheme) it either as a canid or a mammal is beyond me. At some point, functionality changes sufficiently that the descendant organisms are better classified as something else.

    Birds are sometimes colloquially called dinosaurs, but it is not clear to me that this is anything more than a figure of speech to emphasize their descent.
    Is it useful to consider the source of the whirling disease a jellyfish anymore, since (and I am speculating here) functionally it shares nothing with jellyfish?

  2. MawBTSS says:

    Fascinating – I only wish I understood it better.

    Is it easy (or possible) for these cell line infections to cross the species line – ie, infect humans? Thinking mainly of monkeys and livestock, since they’ve given us so many conventional diseases.

    To the casual observer they’d look like a regular sarcoma, wouldn’t they?

  3. Anonymous says:

    Mind sticking some extra breaks between paragraphs?

    I’d like to send this to some friends and have them land at something readable

  4. Elliot says:

    This is easily the most interesting thing I’ve read in a month. The possibilities open for a human cell line to effect the humans it ends up in is intriguing.

    Also, funny coincidence: my dog has no nose.

  5. Ursiform says:

    If only those devils had listened!

  6. Ursiform says:

    I suppose I should have said “if only those poor devils had listened?” …

  7. teageegeepea says:

    There was a comment I thought I made on another thread (and thus off topic) but apparently got eaten or something. You’ve written about Greg Clark’s work on intergenerational mobility, so I wondered what you thought of Gary Solon’s critique.

  8. Eugine_Nier says:

    Well, a lot of infectious deceases were first identified as infectious long before we were able to identify the agents.

    Also, since CLI’s can’t recombine with other CLI’s, I’d imagine in the long run, they’d be at a disadvantage with respect to things like bacteria.

  9. ursiform says:

    If you buy the premise of this about to be released book, maybe TVT choose to evolve:

    Intended Evolution
    Dr. Dongxun Zhang, Bob Zhang
    Greenleaf Book Group
    River Grove Books
    Pub Date May 5 2015

    In Intended Evolution, authors Dongxun and Bob Zhang introduce a different perspective on the theory of evolution: Life is not only selected by nature but intentionally interacts with it, learning how to better its future. They explain that applying this idea to generally accepted principles of biology can have startling results in your ability to affect your own health—and even your evolution.

    According to the theory of intended evolution, organisms gather information through sensory experience and use that knowledge to effect change in themselves and their environments. The authors propose that organisms use this saved information to make choices projected to enhance their survival. It is through experience, choices, and action, within a given environment, that life changes itself from moment to moment and determines what changes are needed for future generations.

    Because of humans’ unique ability to understand how our own evolution functions, we can effect changes within ourselves to influence and enhance our health and fitness, even to lengthen our lifespan.

  10. minosp says:

    It’s not so easy to establish a tumor in an animal using a cell line. You’ve got to either have a severely immunodeficient host or cells that are extremely closely related to the host. This is why there’s a market for immunodeficient animals that cost the researcher 5-10x what a standard strain does. And how many hundreds of gallons of HeLa-contaminated medium must have absentminded researchers splashed onto themselves over the past 50-years? Not that long ago, people were using mouth-pipetting when doing cell culture too.

    • gcochran9 says:

      Southam injected people with HeLa on a number of occasions, but most got over it.

      The cell line infections in Syrian hamsters and Tasmanian devils both flourished in hosts with limited genetic variation. Canine venereal sarcoma doesn’t need that genetic similarity, apparently: it can spread to foxes and coyotes..

  11. Although its phenotype differs considerably from dogs (no brain, no bones, no eyes, no fur, asexual) classification by descent clearly implies that it is a canid and mammal – certainly the most unusual mammal ever discovered. And classification by descent is the right way to go. If you try to classify animals by similarity of surface appearance, you’ll get into trouble. Recent molecular work shows that elephant shrews, elephants, manatees and aardvarks descend from a common African ancestor. An elephant shrew looks a lot like a shrew or rat, but it is more closely related to an elephant. In the same way, the organism causing canine venereal sarcoma is more closely related to a wolf than a fox is, even though you need a microscope to examine it.

    This makes me think that what it is that we mean by the word ‘dog’ isn’t really quite the same as what canind means. For non-scientific purposes, something could conceivably be a ‘dog’ no matter what it was descended from, as long as did what it is that dogs do. Classifying animals by function or appearance does get you into trouble if you are looking for common descent, but if you don’t care about that the functional definition will suffice.
    Much the same, HeLa, while descended from a person, isn’t what anyone would reasonably call a person, while nonetheless being part of the genus Homo. Common descent isn’t good enough here. Like the Bene Gesserit in Dune who differentiated between humans and animals who happened to be homo sapiens

  12. Mike Johnson says:

    I’ve always thought it’d be handy to have a single quantitative measure for the information-theoretic ‘distance’ between two genomes.

    My attempt at one: a “logarithmic evolution distance” (LED) metric, a measure of how many generations it would take (given a certain arbitrary per-generation budget of mutations/indels/deletions etc) to change the genome of organism X into the genome for organism Y. Everything would be done computationally in software, of course. Kind of a rough-and-ready “what’s the shortest computational route to mutate X into Y” thing, without really bothering with phenotype stuff.

    ‘LED’ between a chimp and a baboon? Maybe 1.8 (~8,000 generations). LED between a chimp and a hippo? Maybe 6.6 (60,000,000 generations). LED between e. coli and a starfish? Maybe 25 or so. (Don’t take these numbers seriously, I made em up.)

    It’d be a terribly lossy abstraction– and I want to heavily stress that this metric would be useful in a relative sense, not a literal one, at least initially– but damnit, it’d be a single, not-terribly-unrealistic number you could use to compare ANY two genomes at a glance. Or gene pools, though that’d be more complicated.

  13. Cattle Guard says:

    So, a very poor choice for biological warfare because of the slow growth and tendency to evolve towards low lethality. Still, it might make an interesting sci-fi plot about a mad scientist who wants to destroy humanity by turning himself into a cancer.

  14. reiner Tor says:

    Although its phenotype differs considerably from dogs (no brain, no bones, no eyes, no fur, asexual) classification by descent clearly implies that it is a canid and mammal – certainly the most unusual mammal ever discovered. And classification by descent is the right way to go.

    No doubt they should be classified together with mammals, but I wouldn’t call these things mammals at all. ‘Mammal’ is a traditional word with a traditional definition, which in light of recent DNA research was thought to be a monophyletic group until the discovery of these things. Just as when it was discovered that essentially reptiles are a paraphyletic group they didn’t redefine ‘reptile’ to include humans and penguins, but instead they created a new monophyletic taxonomic group called amniotes, similarly we should create a new monophyletic group including these organisms and acknowledge that mammals are paraphyletic.

    So I’d propose that these transmissible cell lines are not mammals, while on the other hand, because of their existence, mammals are now discovered to be a paraphyletic group. Calling them mammals only causes confusion, just as calling orangutans and lions and ostriches ‘reptiles’ would cause confusion.

    • tautology says:

      Orangutans and lions are not reptiles. Reptilia, which is synominous with Sauropsida, is the most inclusive clade containing sparrows, lizards but not humans. So Ostriches are reptiles but lions are Synapsids, amniotes but not reptiles.

      • reiner Tor says:

        This is what I wrote. Instead of redefining Reptilia to include birds and mammals, we accepted the fact that Reptilia is a paraphyletic group, and instead created a new, correct monophyletic group, amniotes. The same thing should be done for these cell lines, they shouldn’t be classified mammals, instead some new category should be created to include both mammals and these cell lines, and then we’d have to accept that mammals are a paraphyletic group. At least that was my $0.02.

  15. Robert says:

    [First comment on this blog, I don’t work in this field but the content is fascinating.]

    Isn’t this type of organism a plausible vector in the “homosexuality = transmitted disease” scenario in humans/hominids?

    • Del says:

      No. The most plausible vector for that would be a virus, since those actually change genes inside cells, while these (like bacteria) reproduce the original infectious organisms directly.

      That said, it’s a ridiculous scenario no matter what you postulate.

  16. JRM says:

    If a human acquired a cell line infection, would things that increase autophagy (eg. fasting) help get rid of the infection?

  17. Karl says:

    Is anything known about the age of infectious cell lines? As cancer as rather common, I’d expect new infectious cell lines to originate fairly often. If so, most will have to die out rather quickly or they would be extremly common. Of course, there will also be some old ones.

    Who do they compete with? With other infectious cell lines? Or also with other pathogens, like bacteria. If the latter, are they competitive with bacteria over long times?

  18. anonymous says:

    Yet another candidate for the gay germ. Especially if the human cell line infection was of opposite gender from the victim. This could not only explain the inverted sexual attraction but also the gay face/gay voice, etc.

  19. austmann says:

    This is why I regularly visit your blog! Life is a bitcht, aint it?

    • austmann says:

      B-I-T-C-H, (gotta learn English soon)

      • MEH 0910 says:

        Hank: “It just doesn’t work that way with biker couples. Lumpy and Pepperoni Sue have a great relationship, and she never rides up front. In fact, the spot behind the driver is called the…er…’bitch seat’.”
        Peggy: “What? So then that makes me a –”
        Hank: “No! It’s a motorcycle term, I don’t even think it’s spelled the same.”

  20. L says:

    TVT sounds a lot like the Tasmanian Devil Tumors. I’ll refer you to this talk if anyone here is curious.


    Off the top of my head, mammals are classified as air breathing animals which has a backbone. Without those features, it is considered something else entirely even if it shares a significant degree of similarity in DNA or even if it branched off directly from that organism. I don’t see the logic in putting TVT into the same morphological classification as it’s host simply because it came from it’s host. Would HeLa cells then be classified as mammals in this way? Sorry if I sound a bit like a simpleton, genetics was never my favorite subject and I’m certainly no cancer biologist.

  21. gothamette says:

    I can’t say I understand this fully, but this is what bothers me about assisted reproductive technologies such as where the defective mtDNA is replaced with healthy – creating a “2nd mother”. I cannot put it into words but isn’t there a chance that bad stuff will be intro’d into the fetus, which will manifest in future generations?

  22. Doug Muir says:

    Greg, insects have been pulling this trick for a while now.

    Parasitoid wasps lay an egg inside their victims, yes? Well, the eggs of some wasps can produce nasty little cells called teratocytes. Teratocytes are embryonic wasp cells, and they have the wasp’s DNA (though some may be polyploid). But they look and act like free-living single-celled organisms. Depending on the species of wasp, they can perform a variety of functions. Some appear to attack the host creature’s immune system;. Others swarm to the host’s fat bodies and start breaking them down to feed the growing embryo. Still others appear to attack the host’s ability to pupate, either by secreting hormones or by attacking and devouring the glands that drive pupation.

    Teratocyes are fairly large as single-celled creatures go — typically 200 to 500 micrometers — and many of them display cilia, villi, and other complex structures. They can divide and reproduce on their own, and can be cultured. In nature, their numbers tend to surge and then crash naturally during the course of embryonic development, presumably because they’re “designed” not to divert too many resources away from the developing wasp larva. The mechanisms that drive this are still unclear (this is an area where research is very much a work in progress) but presumably it’s not unlike the natural processes of apoptosis that all bilaterian embryos go through. That said, if you were to sample the hemolymph of a caterpillar that had a wasp embryo growing inside it, good chance you’d find it swarming with mysterious single-celled “protozoans”… which would genetically, cladistically, be insects.

    (This is not even close to the weirdest thing parasitoid wasps do, btw. Don’t even get me started on polydnaviruses.)


    Doug M.

  23. Pingback: Links for May 2015 - foreXiv

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s