Skip to main content

So Now Zebrafish Are The Answer

Genome-wide association studies (GWAS) have revealed new facts that might affect the treatment for multiple sclerosis (MS). A study published in Nature1 revealed that a mutation in a single gene could result in patients being unresponsive to a specific treatment. By studying humans, in the form of GWAS, researchers discovered that a mutation affects tumour necrosis factor receptor 1 (TNFR1). This means that some drugs that are used to treat autoimmune disorders such as rheumatoid arthritis will not be effective in certain MS patients. Callaway writes: “Yet despite achieving some encouraging results in animal models of MS, drugs that block the activity of TNF tend not work in patients with MS. In fact, they usually make symptoms worse, and they may even have caused the disease in people predisposed to it . . .” The Nature article may explain why this is the case. Callaway continues: “Fugger hopes that his team’s study will also help to dispel the notion that genome-wide association studies will never offer much that can be used in patient care . . . If doctors had known that TNF-blocking drugs mimic the effects of a major risk factor for the disease, they might have designed their clinical trials differently, he says. ‘The idea is you can use GWAS studies to decide which drugs should be used and which should not be used,’ Fugger says.” Once again, we see human-based medical research resulting in data that can be used for humans and that explains why the animal models were wrong.

Studying humans also led to the discovery that the gene PHF21A is responsible for development of the face and skull and that mutations can lead to cognitive impairment. The researchers studied patients with Potocki-Shaffer syndrome to make this discovery. The press release states: “The scientists confirmed PHF21A's role by suppressing it in zebrafish, which developed head and brain abnormalities similar to those in patients. ‘With less PHF21A, brain cells died, so this gene must play a big role in neuron survival,’ said Kim, lead and corresponding author of the study published in The American Journal of Human Genetics. They reconfirmed the role by giving the gene back to the malformed fish – studied for their adeptness at regeneration – which then became essentially normal. They also documented the gene's presence in the craniofacial area of normal mice.”

I found two aspects of the above interesting. First, animal modelers really believe that in order for something to be proven it must be shown true in animals. This, despite diseases and drug reactions that have never been found to have any counterpart in any species studied and despite the fact that qualifying as a predictive modality means getting the answer correct more than just one time. (For more on proof of concept, see here and here.) Furthermore, in the case of genes, profound dissimilarities have been discovered among species. A gene may cause X in a strain of mouse but Y in another and perhaps is not needed in humans at all. If that were not bad enough, anyone familiar with convergent evolution should know that the same trait can appear by entirely different mechanisms / genes among species.

Second, the zebrafish is getting more attention and being touted as the next model species. First chimpanzees were supposed to be ideal and when they failed, genetically modified rodents, and now we have zebrafish. So we have gone from primates to mammals to fish. One would think maybe the reverse would have worked had the notion any validity to begin with. A recent article was titled: Zebrafish Provide Insights Into Causes and Treatment of Human Diseases. These diseases include

  • Inflammatory Bowel Disease
  • Doxorubicin-Induced Heart Failure
  • Spinal Muscular Atrophy 
  • Acute T-cell Lymphoblastic Leukemia and Lymphoma (T-ALL/T-LBL)

In addition, another article was titled: Zebrafish reveal promising process for healing spinal cord injury. Similar headlines could be found for the animal model du jour over the decades. The reason there have been so many is that none predicted human response to drugs and disease, the raison d'être for the model as well as how animal model use in general is sold to the taxpayers. Think animal models are sold to society on the basis of anything else? When was the last time you saw an animal modeler on TV saying he had to wire up monkey brains for the sake of adding facts to the universe? But claim that it will cure blindness and you have an NIH grant.

In the old days, animals were used to discover and or demonstrate various basic facts about life. While there were ethical issues even back then, the fact remains that the fundamentals of life could have been, and some were, discovered using animals. Of course, some of those same fundamentals could have been discovered, and indeed were discovered using yeast, but today the questions we need answered are very different. Today we want personalized medicine (as opposed to discovering the function of the pancreas) and the only way to get personalized medicine is by studying humans.

One reason even monozygotic twins do not have the same susceptibilities to disease and demonstrate different responses to drugs lies in differences in the intrauterine environment. Studies in both humans and animals have revealed that epigenetic changes can occur in utero. A nice example of what comparative research can reveal. Dr. Jeffrey Craig of the Murdoch Children’s Research Institute (MCRI) in Australia stated: “This must be due to events that happened to one twin and not the other.” His paper on differences in epigenetics between human monozygotic twins was published in Genome Res, July 16, 2012. If monozygotic twins do not respond the same to drugs and disease, society should not be funding animal models just because researchers claim they can. Society should not fund the search for perpetual motion machines, clinical trials for homeopathy, or the investigation of UFOs reported by people whose blood alcohol level was the only thing in the stratosphere that evening. Nonsense costs money and lives, and society has better thing to spend money one. Moreover, basing treatments on animal models harms humans.

Along the same lines, the drug acadesine has been shown ineffective for cardio-protection in a large clinical trial.2 This follows large clinical trials for three other drugs, also shown cardio-protective in animal models that likewise failed in humans.3-5 These clinical trials were based largely on animal studies and were very expensive to carry out. Granted, small clinical trials appeared to show that all of the drugs might be effective but small trials have a poor track record hence the need for large trials. Had these drugs been tested on humans early in the development process, the lack of efficacy might have been noted. However, even if all of the drugs had needed large trials to be shown inefficacious, at the very least the reliance on animal models did nothing to prevent the cost and risks entailed by exposing large numbers of patients to the drugs. Moreover, this fact must be placed into the context of animal models that have allowed drugs in clinical trials that harmed patients in addition to the fact that animal models have derailed, or would have derailed, drugs that were efficacious, or would have been efficacious, in humans. For example, had Tamoxifen been tested on a certain rat strains it would have been shown to cause liver tumors and would not have been developed.6

But many seem to be blind to the dangers from animal models. Broussalis et al. wrote in Drug Discovery Today 2012: “Despite successful neuroprotective therapy in animal models of stroke, the translation of neuroprotective benefits from bench to bedside has not yet been successful.”7 True, but Broussalis et al. then asked: “Why neuroprotection is not yet established?” and suggested the following reasons.

  • Time window
  • Differences in outcome measurement
  • Differences in evaluation
  • Differences in comorbidities
  • Diversity of stroke types
  • Differences in physiological variables. Animals are tightly controlled in laboratory parameters, blood pressure and temperature in addition to other metabolic factors in the acute phase in contrast to humans.
  • Differences in-between gray and white matter damage.

Note that nowhere in their explanations did Broussalis et al. suggest that animals are evolved complex systems that are differently complex from humans and hence it might not be possible for animal models to be predictive modalities for humans. Patients have been seriously injured from drugs that were neuroprotective in rodents. There are better ways to develop drugs and better ways to find drug targets and all of them are human-based.

Until the scientific community as a whole stands up to the few that bring in the most money to universities and insists on integrity in research, animal models will continue to be sold as the be-all and end-all in terms of finding cures for disease. Universities, animal vendors, animal modelers, vivisection activists, and others who are disingenuous or merely misinformed will rally in support of the cause.

Don’t look for the scientific community to grow a spine any time soon.


1.         Gregory, A.P., et al. TNF receptor 1 genetic risk mirrors outcome of anti-TNF therapy in multiple sclerosis. Natureadvance online publication(2012).

2.         Newman MF, F.T.W.J.A. & et al. Effect of adenosine-regulating agent acadesine on morbidity and mortality associated with coronary artery bypass grafting: The red-cabg randomized controlled trial. JAMA: The Journal of the American Medical Association308, 157-164 (2012).

3.         Mentzer, R.M., Jr., et al. Sodium-hydrogen exchange inhibition by cariporide to reduce the risk of ischemic cardiac events in patients undergoing coronary artery bypass grafting: results of the EXPEDITION study. The Annals of thoracic surgery85, 1261-1270 (2008).

4.         Alexander, J.H., et al. Efficacy and safety of pyridoxal 5'-phosphate (MC-1) in high-risk patients undergoing coronary artery bypass graft surgery: the MEND-CABG II randomized clinical trial. JAMA : the journal of the American Medical Association299, 1777-1787 (2008).

5.         Verrier, E.D., et al. Terminal complement blockade with pexelizumab during coronary artery bypass graft surgery requiring cardiopulmonary bypass: a randomized trial. JAMA : the journal of the American Medical Association291, 2319-2327 (2004).

6.         White, I.N. Tamoxifen: is it safe? Comparison of activation and detoxication mechanisms in rodents and in humans. Curr Drug Metab4, 223-239 (2003).

7.         Broussalis, E., et al. Foundation review: Current therapies in ischemic stroke. Part B. Future candidates in stroke therapy and experimental studies. Drug Discovery Today17, 671-684 (2012).


Popular Video