The Neuroscience of Autism

Autism Spectrum Disorders (ASDs) are classified as a wide range of developmental disorders that present early in life, consisting of a range of developmental disorders, including Asperger syndrome, Childhood Disintegrative Disorder, Rett syndrome, and, most commonly, autistic disorder (1). The phenotypes associated with ASDs are broad as well, including problems in communication, social interactions, and cognitive functioning, usually resulting in mental retardation (2). The incidence of these disorders has increased to approximately one child in every 166 born (3), which is a significant increase from studies performed in earlier years, resulting in a higher percentage of the population afflicted with this range of disorders. Several research groups have identified genetic roots to ASDs as well, meaning that it is passed on from one generation to subsequent ones to varying degrees (4). There is further evidence that ASDs result from an imbalance between the ratio of excitatory and inhibitory signals in sensory, mnemonic, and social systems (5). Because of the ever-increasing number of persons afflicted with this class of disorders, it is certainly a field of interest. However, the specific mechanism responsible for ASDs is not yet established, there is no animal model that is representative of ASDs, and there is no known cure or therapy to alleviate or reduce the negative effects associated with these disorders.

The primary focus of this paper is to investigate the work of Tabuchi et al, entitled “A Neuroligin-3 Mutation Implicated in Autism Increases Inhibitory Synaptic Transmission in Mice”. The authors of this paper claim to have engineered a mutant mouse model of a phenotype similar to that of ASD patients, which could have further research implications in determining the cause for ASDs, as well as testing potential therapeutic treatments for this class of disorders. The model that Tabuchi et al utilized is the Arg451→Cys451 (R451C) mutation within neuroligin-3. This specific mutation has been found in a subset of human individuals afflicted with ASDs, and is one of the few genetic links that have been established for autistic behaviors. By knocking in the R451C mutation into the neuroligin-3 molecules of mice and comparing these mutant mice to control groups (wild-type mice), the authors investigated whether or not this specific mutation contributes to a phenotype in the mice similar to that seen in patients with ASDs.

Neuroligin-3 is a postsynaptic cell adhesion molecule, which is responsible for mediating synaptic signals. Neuroligin-1 and -2 have previously been investigated in mice, and result in changes to excitatory or inhibitory synaptic behavior (6), while neuroligin-3 has not been investigated in mouse models. Tabuchi et al, upon examining these neuroligin-3 mutant mice, found that this substitution significantly increases the levels of inhibitory synapse markers when compared to controls, as can be seen in their Figure 1. No change was found, however, in the levels of excitatory synapse markers, which is a significant result, demonstrating the potential imbalance that has been hypothesized to be responsible for ASDs. Whole cell recordings were performed on the mutant mice in the somatosensory cortex, as shown in their Figure 3, and they found that the mutant mice exhibited increased spontaneous inhibitory synaptic transmission when compared to KO mice or controls. They also found an increase in inhibitory synaptic strength in the presence of the R451C neuroligin-3 mutation. These results are all in agreement with the increase in inhibitory synapse markers from their Figure 1. This is important because it provides further evidence to the excitatory / inhibitory synaptic potential imbalance that has previously been suggested as the mechanism for ASDs.

Furthermore, Tabuchi et al compared the behavior and learning of the mutant mice to control mice. The mutant mice showed a lower propensity to interact with other mice, indicating decreased social behavior as compared to control mice. In order to test spatial learning, they employed the Morris water maze under both initial and reversal training. The mutant KI mice showed no difference in finding the visible platform, but performed significantly better than controls in finding the hidden platform. This indicates an increased spatial learning over wild type mice, which is a curious result. Most ASDs present with mental retardation and decreased cognitive functioning, but not these neuroligin-3 R451C mice. This could be due to the fact that some persons with ASDs exhibit the rare symptom of enhanced ability, such as the autistic savant, and this mutation tapped into that rare ability. It will be intriguing to see, as time progresses, whether any more research is performed on the enhancement of spatial learning associated with neuroligin mutation. Overall, Tabuchi et al have demonstrated that the R451C mutation acts as a gain of function substitution, resulting in increased inhibitory synaptic transmission in addition to behavior that is characteristic to ASDs.

It is interesting that the authors this paper attempts to develop a model for ASDs utilizing a mutation that is only found in a small percentage of individuals with ASDs. The R451C mutation was only found in two families that Tabuchi et al cited, whereas there have been hundreds of individuals with ASDs investigated. One author indicates that this Neuroligin-3 mutation has only found in 0.8% of individuals with ASDs that have been investigated prior to 2009 (7). This in itself is evidence that there is still much to be uncovered about this class of disorders, and raises the question: If the other classes of ASDs are not associated with a genetic mutation, then what is their cause? It is also a concern that if potential therapies were based off of this rare neuroligin-3 mutation, then the testing of potential therapies on this model may not be accurate beyond the scope of this specific mutation. For example, therapies that are found to be curative to these mutant knock-in mice may not be curative to the vast numbers of individuals with ASDs. However, it is a start, and can provide vital information to researchers who are trying to further understand the mechanisms involved with ASDs so that they can develop therapies to treat these debilitating disorders. Further testing should be performed to determine if there are any other mutations that are present in a greater percentage of those inflicted with ASDs. Also, it is important to note that the ASDs are very broad in their presentation in the population, and there could be many different potential causes that have not been investigated. The work of Tabuchi et al could very well provide the necessary tools needed to understand a subclass of ASDs, which is why its results are so compelling to the field of autism research.

Although this work provides legitimate evidence supporting their model, it is important to note another study that took place shortly after this work that claims a different perspective. In 2008, Chadman et al published a paper suggesting that the neuroligin-3 R451C mutation in a mouse model shows little to no behavioral characteristics similar to autism spectrum disorders (8). Furthermore, they advise that the results of Tabuchi et al are less than compelling, and that the Neuroligin-3 R451C KI mouse model is, in actuality, not one that is representative of ASDs. Chadman et al produced a R451C KI mouse in similar manner to Tabuchi. They subjected the mouse to numerous behavioral experiments as well as learning ones, and found no significant difference between the mutant mouse and littermate controls. They did admit, however, that there were minor differences between the mutant and control mice, just not ones that demonstrate the phenotype of ASDs.

This is further confounded by the publishing of another article by Radyushkin et al in June of 2009, who claim to have induced autistic behavior in neuroligin-3 deficient mice (9). Radyushkin et al found that knocking out neuroligin-3 in mice leads to behavior similar to ASDs, including reduced vocalization and lack of social interactions. They also found, interestingly, that there was a decreased olfactory sense available to these neuroligin-3 deficient mice. Similar olfactory loss has been found in humans with ASDs, giving further evidence that the neuroligin-3 mutant mouse could be a representative model of autistic phenotype. Although the mutation described by Radyushkin was not the same R451C mutation, they found similar behavior mechanisms to Tabuchi et al for the Morris Water Maze behavior. They found that the Neuroligin-3 KO mice showed normal performance for spatial learning and memory, and even showed some increase in spatial learning ability in reverse Morris Water Maze training when compared to controls. This is interesting, especially because it mirrors the increased spatial learning found in the R451C KI mice. Radyushkin’s work also supported Tabuchi’s findings in relation to socialability. Both the Neuroligin-3 R451C mutants and the Neuroligin-3 KO mice showed deficient sociability, and less time spent interacting socially with littermates. Also interesting is the lack of the increased inhibitory synaptic transmission in the Neuroligin-3 KO mice which was found in the R451C mutant ones. This could indicate that there is something more complex at play than just a single point mutation, or something more than simply an imbalance between inhibitory and excitatory synaptic transmission that contributes to the phenotype of ASDs. More likely, it is a combination of several different events working in conjunction to cause ASDs.

Tabuchi et al found that there was a 90% decrease in neuroligin-3 in the forebrain, as measured by immunoblotting and shown in their Figure 1. It could be the case that this downregulation was enough that it allowed for the behavioral patterns seen in their mutant mice to appear as if the neuroligin-3 was absent, as the behavior was similar in nature to the neuroligin-3 KO mice of Radyushkin’s group. This thought is pure speculation, however, and further studies could be performed to give more evidence of this. The work of Radyushkin et al, nonetheless, provides further evidence that neuroligin-3 can be modified in mice to produce a mouse with characteristics of ASDs, contrary to that which was demonstrated by Chadman et al.

The methods described by Tabuchi et al could provide potential therapeutic applications in the analysis and treatment of ASDs. Having access to a model that is a close homolog to the developmental disorder of ASDs, especially one as easy to manipulate and investigate as a mouse model, could yield a tool for further research in examining exactly what the mechanism is behind autistic behaviors, as well as the investigation of ways to reverse that mechanism. Tabuchi et al solidified the notion that that the inhibitory / excitatory synaptic transmission imbalance is a contributing factor to ASDs, and further testing can be performed based on this concept. For example, one potential examination that could be performed is the investigation of a way to rectify the inhibitory / excitatory synaptic potential imbalance, bringing the animal into balance, and possibly ameliorating the negative phenotypical effects associated with ASDs. Another question this raises is: Could this specific R451C mutation’s effects be reversed if the inhibitory synaptic transmission were to be downregulated to account for its unnatural increase?

Tabuchi et al have certainly brought to light some interesting and exciting new research in examining this broad class of disorders that are increasing in number. The potential for new research, the investigation of controversial results, and the unusual increase in spatial learning all would be potential topics of interest for future research.

References
1. E. DiCicco-Bloom et al., J. Neurosci. 26, 6897 (June 28, 2006, 2006).
2. K. Tabuchi et al., Science 318, 71 (Oct, 2007).
3. J. E. Carr, L. A. LeBlanc, Primary Care 34, 343 (Jun, 2007).
4. R. Muhle, S. V. Trentacoste, I. Rapin, Pediatrics 113, e472 (May 1, 2004, 2004).
5. J. L. R. Rubenstein, M. M. Merzenich, Genes Brain Behav. 2, 255 (Oct, 2003).
6. A. A. Chubykin et al., J. Biol. Chem. 280, 22365 (Jun, 2005).
7. C. Lintas, A. M. Persico, Journal of Medical Genetics 46, 1 (January 2009, 2009).
8. K. K. Chadman et al., Autism Res. 1, 147 (Jun, 2008).
9. K. Radyushkin et al., Genes Brain Behav. 8, 416 (Jun, 2009).