In the Age of Autism blog, JB Handley repeatedly referenced a now withdrawn study by Hewitson et al. in Neurotoxicology which looked at vaccinated VS unvaccinated monkeys. This study has been analyzed by this blog and by Orac and I think it goes without saying that it was pretty bad. While there were numerous inconsistencies with this publication the question it raised for me for was, how appropriate is a monkey study in the first place? This question is important because monkeys are some of the most expensive, difficult to use and controversial animals to use in research. If using a different animal model is effective or if using animals to model autism in the first place does not work very well, then a study which uses monkeys in the way described in Hewitson et al. is unethical at best and useless at worst.
It is worth pointing out here that the term animal model can mean slightly different things based on the concept of the experiment. An animal can be bred to be strongly genetically disposed to a certain condition or disease. This is often done through genetic manipulation of the animal which causes a particular gene to be suppressed. For example, people with mutation in the Nlgn4 gene have autistic behavior, and mice bred such that this gene is turned off also display autistic behavior. This animal is referred to as a “knockout”. It is useful for study because it gives you an animal which you can be fairly certain has or will develop a particular condition. Using this animal, therapies can then be tested. Another way one can refer to an animal model is in an experiment where a healthy animal, commonly referred to as a “wild-type” (meaning not genetically altered) is exposed to some experimental process to observe the outcome. The Hewitson study performed this sort of test.
In a search of recent publications to see what sort of model was preferred in current research, it is pretty clear that rat or mouse models are the most common. Llaneza et al reports in Physiology and Behavior , “Robust, easily replicable assays have been developed to quantify social behaviors relevant to ASD (Autistic Spectrum Disorder) in mouse models”. This means that for a researcher trying to evaluate whether an experimental condition has treated or caused ASD, there are tools which exist to clearly verify the outcomes of such experiments. Llaneza also reports on several methods which can be used to create mouse models for ASD, mainly through the creation of knockout species which results in mice that have ASD symptoms. So it appears there are robust tools for performing many ASD experiments using mice, but what about other animals? Llaneza does not comment on the use of primates or any other species for this purpose (Llaneza et al. ).
Primate models however, do exist. Machado and Bachevalier discuss the use of macaque monkeys for the purposes of studying human developmental disorders (Machado, Bachevalier 2003). Their analysis is complex and far too detailed for the purposes of this entry but it is a fascinating glimpse at the development of a species similar to humans. Of relevance to us is a discussion of the best developmental stages for testing, where best here means stages in which there is rapid development. The first of such stages is the two week period of the infamous monkey paper. However this period does not contain the kind of social developments that are characteristic of autism. These behaviors are described as reflexive behaviors, motor and grasp, and aversion to staring; only the last one can be considered even closely related to a social development. By the very next week these animals begin to exhibit social vocalizations, the very kind of thing one might thing is a good test for autistic behaviors, but by this time in Wakefield’s study, the animals were dead. Also of note is that these developments in the macaque are related to the growth of the amygdyla, a structure in the brain related to emotional and social behavior. However, in the human infant, these structures are more developed by birth than in the macaque. Still, the conclusions of Machado and Bachevalier state that the macaque can provide an excellent model for examining social development disorders (Machado, Bachevalier 2003).
Amaral , Bauman and Schumann summarize and discuss the results of various studies in which primates were used to study autism (Amaral, Bauman & Mills Schumann 2003). In this study, the amygdyla was surgically impaired and the behavior of the primates was observed. They conclude the amygdyla is not likely to play a large role in the social implications of autism but rather plays more of a role in mediating feelings of fear. However this finding is not relevant to the discussions here, but rather the methodology. These studies discussed evaluated the results of their animals at 2 months and 6 months of age. Given the discussion of macaque develop by Machado and Bachevalier, this makes sense, as by that time macaque monkeys should have obtained clear social milestones. Again, this is much longer than the time the monkeys were evaluated by Wakefield and his colleagues at thoughtful house.
For the sake of comparison, I read another study concerning autism and rhesus monkeys. This one was particularly odd me to me but that’s aside from the point. It is titled “Stereotypies and hyperactivity in rhesus monkeys exposed to IgG from mothers of children with autism”, by Martin et al. which comes from the M.I.N.D institute at UC Davis (Martin et al. 2008). This study used a total of 15 monkeys, 4 injected with immunoglobulin from mothers of children with Autism, 4 injected with immunoglobulin from mothers of children without autism, and 5 were not injected with anything. The animals were injected prenatally and followed after birth for an entire year. I can’t really comment on the overall design because behavioral studies are really not my thing but at very least they followed the monkeys for a much longer period of time and allowed them access to their mothers, which is important for sociological development (Machado, Bachevalier 2003). However, they also used a small amount of monkeys for this test, which means gleaning a lot of meaning is difficult. The conclusions of the paper were, as taken from the abstract
“Rhesus monkeys gestationally exposed to IgG class antibodies from mothers of children with ASD consistently demonstrated increased whole-body stereotypies across multiple testing paradigms. These monkeys were also hyperactive compared to controls. Treatment with IgG purified from mothers of typically developing children did not induce stereotypical or hyperactive behaviors. These findings support the potential for an autoimmune etiology in a subgroup of patients with neurodevelopmental disorders. This research raises the prospect of prenatal evaluation for neurodevelopmental risk factors and the potential for preventative therapeutics (Martin et al. 2008).”
This study seems a little weak to me, but the point is that they seemed to be more careful in their evaluation of the monkeys used than in the Hewitson study. Also note in the Hewitson study, they cite a reference which indicates taking the monkeys away from their parents was correct, which seems to contradict the reference I found. Hewitson’s reference however is from 1992 while mine is from 2003.
So what are the final conclusions of this analysis? The use of primates for behavioral studies certainly has advantages. They are more similar to human physiologically and socially and would seem to be a better fit for a study involved with a neurological disorder. However there are several caveats. For one, it’s simply easier to do a large sample size study with mice or rats because they are smaller, have a shorter life cycle, and easier to care for. We know that there exist useful models that are well established for this purpose. It also does not seem to me that the protocol for using monkeys in these sorts of studies are as well established or standardized. If you are going to use monkeys, it should be for an experiment that needs monkeys. So in conclusion I would say that yes, it can be appropriate to use monkeys for research involving autism, but that these studies must be carefully designed to take advantage of the usefulness of primates. The study that Hewitson performed seems like it would have been better served to have used mice or rats instead, especially since they only observed the monkeys for a short amount of time. Lastly, I think it is worth pointing out that even though these models may work very well, they are not studies using human and no animal model, primate or rodent can ever do as a good a job.
Amaral, D.G., Bauman, M.D. & Mills Schumann, C. 2003, "Review The amygdala and autism: implications from non-human primate studies", Genes, Brain & Behavior, vol. 2, no. 5, pp. 295-302.
Llaneza, D.C., DeLuke, S.V., Batista, M., Crawley, J.N., Christodulu, K.V. & Frye, C.A. "Communication, interventions, and scientific advances in autism: A commentary", Physiology & Behavior, vol. In Press, Corrected Proof.
Machado, C.J. & Bachevalier, J. 2003, "Non-human primate models of childhood psychopathology: the promise and the limitations ", Journal of child psychology and psychiatry, and allied disciplines, vol. 44, no. 1, pp. 64-87.
Martin, L.A., Ashwood, P., Braunschweig, D., Cabanlit, M., Van de Water, J. & Amaral, D.G. 2008, "Stereotypies and hyperactivity in rhesus monkeys exposed to IgG from mothers of children with autism ", Brain, behavior, and immunity, vol. 22, no. 6, pp. 806-816.