Skip to Content

Whatever Happened to the Fish and the Frogs?

Eight or ten years ago, there was a good deal of excitement about non-mammalian small animal model systems for compound screening – specifically fish and frogs. More specifically, zebrafish and Xenopus. A number of small companies started up to do this sort of thing, and large companies paid attention as well. A correspondent, though, wrote me the other day noting that few (if any) of these companies seem to have made it. Phylonix and Znomics seem to be inactive, and InDanio, while apparently still with us, has a low profile. (Are there others?)
More generally, it’s worth asking what’s become of the whole idea. I’ve read some interesting papers over the years using these systems in compound screening, mostly uncovering effects developmental pathways. But how has it worked out overall? There are still people working in the field, of course, but have they been able to get any traction in drug discovery? From my perspective, I just know that I seem to hear a lot less about this than I did a few years back.
Is there still a zebrafish case to be made, or one for the (evolutionarily closer) Xenopus? If so, is that case best made for developmental biology/embryology, or for more open-ended high-content phenotypic screening, or for toxicology? Thoughts welcome.

22 comments on “Whatever Happened to the Fish and the Frogs?”

  1. Anonymous says:

    Perlstein Lab (link given above), a new start up interested in orphan diseases, is focusing exclusively on phenotypic screens in model organisms. Too early to say if they have tractable hits and leads.

  2. @Derek says:

    Here is a cute and concise video from Roche about zebrafish models in biological research:

  3. sim_2d says:

    For zebrafish, at least still active, there is a Spanish CRO, biobide.
    From some CRO conferences, I only heard about zebrafish, nothing on frogs.

  4. Sideline Chemist says:

    Teleos Therapeutics in Boston area is doing some interesting work on using zebrafish in behavior-based screening for CNS diseases.
    Most of these companies are very young and it will take years before a compound from their screens make it to clinical trial.

  5. innovorich says:

    The models are good and the scientific validation is growing. The challenge is really inserting into the drug design/med chem process in a way that reduces costs rather than increases them – and this has so far failed. This is really about a dogma and practised way of doing things amongst project leaders; i.e. the reality ended up being that the data looked good but everyone had to check, in larger animal models that they trusted more, therefore defeating the object of the use of the zebrafish/xenopus models as a real filter. And therefore, realising ahead of time that they would do that, they would just not bother next time.

  6. Virgil says:

    On the academic side, I’m seeing more and more C. elegans and Drosophila labs having a hard time getting funded. Sure, there are a few areas such as aging research where short lifespan is useful, but for most other stuff the adage of “simpler and cheaper than mouse” just doesn’t hold water any more. I would guess CRISPR/Cas is only going to accentuate that problem. In the major medical research universities the fly/worm labs are fewer and fewer, but they’re still popular in the smaller colleges and undergrad teaching institutions, primarily because they lend themselves so easily to short projects for students. The model organism labs that seem to be thriving, are the ones with a clear path to translation into mammalian systems. Going after the mammalian orthologs of genes, seeing if the system behaves the same way in mice or in human cell lines.
    Part of the driving force for this (IMO) is a raising of the bar in terms of what is required by biotech/pharma for biting at a project (and who can blame them?) A long time ago, cell data was enough, then they wanted mouse/rat/rabbit, now they want dog/pig/sheep before blinking at an IP portfolio from academia. I don’t see venture capitalists lining up to sign drug screen results from university invertebrate research labs.

  7. Anonymous says:

    “Whatever Happened to the Fish and the Frogs?”
    The frogs ate the fish. Then the French ate the frogs.

  8. anonymouse says:

    The history of Exelixis is informative. Founded twenty years ago with much hoopla and hype on the thesis that model organisms such as Drosophila would reveal novel pathways and targets for drugs. Fast forward 20 years and they ended up going after the same RTKs as everybody else. And not very successfully from the looks of it. Lexicon Pharmaceuticals (nee Lexicon Genetics) with their mouse knockout model target discovery system doesn’t seem to have been any more successful. So what do zebra fish and Xenopus bring to the table that others haven’t already tried?

  9. steve says:

    I know of one example where zebrafish screening has led to a clinically-relevant product. Len Zon at Harvard found that PGE2 acted on zebrafish hematopoietic stem cells and FATE Therapeutics (now in Ph2) is trying to develop that clinically.

  10. phenotypic screener says:

    I think the answer to your question stems from what your view is of the current level of understanding of biological systems. If you think we have it mostly figured out, then it would make sense to study systems most closely related to human biology. If on the other hand you think our level of understanding of biological systems is rather limited, that biology is pretty complicated, then it makes sense to work in experimentally tractable model systems. While the current mindset at NIH study sections and most but not all pharma seems to align with the first model. I think in 40 years we’ll see the second view to be more accurate. Advances in technology like CRISPr will certainly help make mammalian systems more experimentally tractable but that will also be true of invertebrates and their cold-blooded vertebrate relatives. As a scientist the level that you can work at experimentally in worms or flies is far superior in terms of defining an in vivo biological process. For example, in a high quality invertebrate paper the analysis of typically dozens if not hundreds of compound mutations is usually required compared to maybe a handful in the average high quality mouse paper– despite that being a more complex system.
    In terms of phenotypic screening, what is closer to a real biological system? cells in a dish– even primary human cells or an intact organism even distantly related to humans with compensatory pathways, native cell-cell interactions and physiological concentrations of growth factors nutrients etc… The trick in all phenotypic screening for drug discovery whether it be mammalian cells or yeast is picking a phenotype to screen on that is therapeutically relevant for a human disease indication– an obvious concept but not so easy to execute. What the non-mammalian organisms do allow you to do is ask open ended questions, to give yourself “optionality” in Taleb parlance, minimize the cost of experimental failures, before taking the free option to jump down the rabbit hole. In other systems you have fewer options before jumping down that hole.

  11. If I may, I would add Perlstein Lab to the list of biotechs doing drug discovery with genetic model organisms, in our case baker’s yeast, nematodes, fruit flies and zebrafish. (No offense to frogs, we ultimately aspire to a Noah’s Ark of model orgs).
    @1 is correct that we just got off ground in April of this year. But in that relatively short time we’ve validated worm and fly models of Niemann-Pick C, our first disease target. Whole-organism phenotypic screens have commenced, and we expect to have promising leads in early 2015.
    What’s different about our approach is an exclusive focus on Mendelian (aka rare/orphan) diseases. You could argue that we’re Exelixis rebooted for rare diseases. Where I think Exelixis went astray was by targeting p53 cancer and falling in love with kinase inhibitors. I can’t blame for not going Mendelian because rare diseases were on nobody’s radar screen in the mid-90s (except maybe Genzyme’s).
    The therapeutic thesis propelling PLab is the exquisite “product-market fit” between simple model orgs and highly penetrant monogenic diseases, of which there an estimated 3,500. I won’t belabor @11’s comment about disease in a dish vs disease in an organism in a dish, with which I completely agree.
    Chew on this: the vast majority of Mendelian disease genes are ancient. (,
    In other words, only a small minority of Mendelian are uniquely mammalian, and even fewer are uniquely human.

  12. Slurpy says:

    Not that it falls under industry, but a few of the biologists at my undergrad (linked in name) are still using zebrafish for cell-bio work.

  13. Anonymous says:

    A zebrafish is just a human with less mutations.

  14. Jumbo says:

    Ethan – (#12) – it is a long way from ‘target validation’ with a model organism and having a ‘clinically relevant’ drug. No small organism like zebrafish can accurately model large animal ADME or Tox, for example. While I wish every success, there are a lot of hurdles before proving a drug ‘validated’ in an IND-ready sense. Having said that, I am also aware of numerous (as in up to a dozen, certainly not dozens) of mid and large biotechs and pharmas that either in-house or in collaboration are screening candidates with non-mammalian animal models (eg. Drosophila). All it will take is for a few of these to get to the clinic and demonstrate significant advantages (speed with which they progressed, increased efficacy, something) and then everyone will start beating a path to these methods. I also agree, they will likely remain more useful for diseases with defined genetic underpinning (i.e. diseases with a single gene defect) as opposed to more complex, multigenic/multifactorial genes. Sorry, all you neuroscientists out there!

  15. Anonymous says:

    A couple of good friends worked on zebrafish in relation to developing models or useful screening platforms for a human disease. It never worked out. Trying to get them to express some proteins of interest didn’t work, and treating in various ways with relevant proteins didn’t work either. They’re pretty good at regeneration, so for looking at diseases in which damage accumulates over time they seems to be often a non-optimal model. And while you may be able to train a fish to do a t-maze or the like, it’s less easy than training a mouse or rat and the fish makes many more “mistakes.” They seem like they’d be ideal for developmental biology work, maybe reasonable for some cell biology work–but in some areas I’d say they are neither as easy to work with nor as useful as they may be billed as.

  16. DHChemist says:

    Not my usual field, but I’m aware of a lab (my name has a link to one of their Nature papers) who used Zebrafish for compound screening, with their big result being the discovery of the use of Leflunomide (predominately used in the treatment of Arthritis) as showing potential for the treatment of Melanoma. The fact that it’s already an approved drug obviously offers significant benefits. They’ve since concentrated more on using Xenopus rather than Zebrafish.

  17. Moreso says:

    Another consideration for studying fish and frogs: Reproductive Toxicology. Not that I’m a big fan of Haeckel’s theories, but conducting Repro Tox and Teratogenicity studies on mammalian systems is difficult from technical, regulatory, and ethical standpoints. Having easily examined embryos en masse can be really helpful in Repro Tox. Besides, data reliability in general can be boosted by having large N, consistent treatments, and well-defined evaluation criteria.

  18. Frenchman says:

    Whatever happened to the zebrafish unit at the Novartis Institutes of BioMedical Research (NIBR) in Cambridge? Did they succeed in phenotypic drug discovery along the lines pioneered by Mark Fishman, formerly at MGH, since 2002 President of NIBR?

  19. Frenchman says:

    Whatever happened to the zebrafish unit at the Novartis Institutes of BioMedical Research (NIBR) in Cambridge? Did they succeed in phenotypic drug discovery along the lines pioneered by Mark Fishman, formerly at MGH, since 2002 President of NIBR?

  20. Frenchman says:

    Whatever happened to the zebrafish unit at the Novartis Institutes of BioMedical Research (NIBR) in Cambridge? Did they succeed in phenotypic drug discovery along the lines pioneered by Mark Fishman, formerly at MGH, since 2002 President of NIBR?

  21. Fishlover says:

    Along the lines of Frenchman, there is another story about pharma acquiring zebrafish drug screening technology. In 2009 Evotec in Germany purchased Summit, a British based zebrafish company. The press release mentioned “Strong addition to Evotec’s core business in drug discovery”. Does anybody have news what happened to Summit after the acquisition now five years ago?

Comments are closed.