Abstract
Autism spectrum disorder (ASD) is a human neurological and behavioural disease, seen in 1 on 88 births in the United States. In this review, it will be studied if ASD is a human disease or also a non-human primate disease by comparing the human brain and the brain of the last common ancestor (LCA), which shows that the human brain is enlarged as a whole, and varies in size in several regions compared to the LCA. Also, it is seen that the human brain has very much in common with the brains of great apes. Furthermore, the ability of language is a unique feature in the human brain and is thereby not seen in any other primate. For studying ASD, animal models are used. These animals are modificated to show autistic-like behaviour, such as social impairments. It is shown that for different types of research, different types of animal models are necessary. Also, the reliability of these animals is still questioned, but seem to be approved.
Introduction
The word ‘autism’ was first used in 1908 by psychiatrist Eugen Bleuler. At this time it describes a subset of schizophrenic patients. The name comes from the Greek “autos” meaning “self”, because patients were especially withdrawn and self-absorbed (Virginia Sole-Smith, The History of Autism, Parents).
Autism spectrum disorder (ASD) is a neurodevelopmental disorder. Neurodevelopmental disorders are characterized by a dysfunctional brain development. Also patients attend to have behavioural, cognitive and/or physical abnormalities (Hsiao et al., 2013). Autism in particular is characterized by abnormal repetitive and stereotypic behaviour and also a restricted interest in early childhood compared with a deficiency of the social communication (Ito et al., 2017).
Nowadays, ASD is diagnosed on the presence and intensity of the autistic-like behaviour and on their deficiencies in language and social interaction (Hsiao et al., 2013).
ASD has an incidence of 1 on 88 births in the United States in 2008 (Autism and Developmental Disabilities Monitoring Network Surveillance Year 2008
Principal Investigators and CDC, 2012, cited in Hsiao et al., 2013) and therapies for treating symptoms of autism are limited.
The increase of the human brain size, which comprises mostly the growth of the neocortex (Finlay & Darlington, 1995), is central in the evolution of modern human cognition; other modifications in brain development, structure and function appear to be significant (Sherwood et al., 2008).
Although autism has its own characteristics, there are several diseases with similar characteristics as autism. An example of this syndrome is Rett syndrome, which is a severe developmental disorder, which shows autistic phenotypes (Amir et al., 1999, cited in Liu et al., 2016). In 90% of the patients with Rett syndrome, a mutation in the MECP2 gene is found. The Methyl-CpG binding protein 2 (MeCP2) has crucial roles in transcriptional regulation and microRNA processing.
When the MECP2 gene is duplicated, it will cause MECP2 duplication syndrome, which shares core symptoms with ASD (Ramocki et al., 2009, cited in Liu et al., 2016). Liu et al. (2016) report that lentivirus-based transgenic monkeys (Macaca fascicularis), which express human MeCP2 in the brain, present autism-like behaviours and show germline transmission of the transgene.
By studying the underlying (molecular) mechanisms underlying the development and functions of the brain, the origin of autism can be discovered. For this, rodents, such as mice and rats, have been widely used. They also have been used to discover the pathophysiology and treatments for multiple brain diseases (Kawasaki, 2017).
By studying humans (Homo sapiens), it is seen that humans are very well developed and intellectual, compared to other animals. Yet, despite our distinctiveness, humans show more than 99% similarity in nonsynonymous DNA sequence with chimpanzees (Wildman et al., 2003, cited in Sherwood et al., 2008).
Studying underlying molecular mechanisms in the brain of mice does not completely correspond with the important unique brain structures that humans have. The brain of primates and carnivores correspond in a better way with that of humans, because, the anatomical structure is more similar to each other, due to the developed brain structure, which not found in the brains of mice (Kawasaki, 2017). The challenge to understand how the unique feature of modern human behaviour is mapped onto evolutionary changes in neural structure is still ongoing (Sherwood et al., 2008).
Comparing to normal age-based values, in young children with ASD, it is seen that the global brain developmental abnormalities appear in the archicortex, cerebellum, brainstem, and other subcortical structures. It appears with region-specific intensity of neuropathology (Wang et al., 2009; Wegiel et al., 2015; Sacco R and Gabriele S Persico, 2015, cited in Varghese et al., 2017). The increase in brain size in children with ASD, may be due to an increased number of neurons, or to no change in the amount of neurons, but due to an increased neuropil, which is a dense network of interwoven nerve fibers and their branches and synapses, together with glial filaments (Varghese et al., 2017).
Study show that the brain consists of various regions. The fusiform gyrus (FG) is important in the processing of faces, which makes it a crucial part of our capacity to interact appropriately in social situations (Varghese et al., 2017). Most studies state that in ASD patients, there is a hypoactivation of the FG (Bolte et al., 2006; Kanwisher et al., 1999; Pierce et al., 2001a; Pierce et al., 2001b; Piggot et al., 2004, cites in Varghese et al., 2017).
Earlier research has shown that anxiety-associated behaviour is seen in autism patients and in mouse models with MeCP2 overexpression as well (Ramocki et al., 2009; Samaco et al., 2012 cited in Liu et al., 2016). Also, the impairment of social interaction is a sign of MECP2-associated disorders as well as an autistic characteristic (Liu et al., 2016).
In nature, we see a broad spectrum of communication methods, which may seem unique to its possessors. One of them is the language of human spiecies (Hedeager, 2012). It is long been acquainted that language is a unique behavioural feature in Homo sapiens (humans) (Locke and Bogin, 2006). The human brain shows a strong population-level dominance for language functions in the left hemisphere. This is especially seen among right-handed individuals. Some of the cortical areas, which are associated with language function in humans, also display asymmetries in extant great apes. This suggests that the asymmetries were present in the last common ancestor (LCA) (Sherwood et al., 2008). In addition to this, the population-level left hemisphere dominant orangutans, gorillas, chimpanzees, bonobos and humans share asymmetry of the planum temporale, which is a surface feature of the cerebral cortex in the region of Wenicke’s area (Gannon et al. 1998; Hopkins et al. 1998). These neuroanatomical asymmetries suggests that due to computational demand to process sequential information, some aspects of functional processing were already lateralized in the brain of the LCA prior to the evolution of language (Ringo et al. 1994). Probably, hemispheric dominance for the process of communication signals was advanced at an earlier stage in primate evolution. Altogether, these findings point out that the neural machinery for the process of acoustic signals, which are contained in species-specific communication, was present and lateralized in the LCA. This provided a framework, from which language evolved (Sherwood et al., 2008).
These findings lead to the central question: Can autism be a disease of other primates next to humans?
In order to examine this question, this review describes the differences between the human brain and the brains of non-human primates, focussing on autism related features, over the last decades. The cellular and neuroanatomical features will both be exposed. In addition to this, the importance of language in the human evolution will be reported. Furthermore, this review will discuss the used animal models and their reliability that are used in research on ASD.
Human brain varies in different regions from the LCA, but shares several features with great apes
To answer the main question, the differences between the human brain compared to other primates will be described. This paragraph will focus on the cellular and anatomical aspects of the brain of both humans and primates. To discover neuroanatomical differences between humans and other primates, a variation of regions in the brain will be discussed. In this paragraph, also the cellular differences between humans and primates will be presented. Furthermore, neuroanatomical differences in cortical regions between patients with ASD and their controls will be reported in this paragraph. Are there anatomical differences that make it possible to explain the development of autism in humans?
When looking at the neuroanatomical differences between humans and their last common ancestor (LCA) it is seen that there are features that are distinctive to the LCA. It is seen that, in humans, the primary visual cortex (V1) is reduced in overall size and histological modifications of layer IVA, compared to the V1 of the LCA. It can be assumed that the expansion of the posterior parietal cortex perhaps have contributed to the reduction of the V1 in humans. Furthermore, in humans a relative enlargement of the temporal lobe is visible. An enlargement in total volume and white matter can be seen. When the Von Economo Neurons (VENs) are studied, which are located in layer V of the anterior cingulate and frontoinsular cortex, it is seen that humans show larger and more clustered VENs than great apes. Also, VENs are non-existent in non-hominid primates. Studying the frontal cortex in the LCA shows that there is a relative enlargement, particularly in the dorsolateral prefrontal cortex, compared to humans. Also, there is a relative enlargement of the lateral hemispheres seen in the cerebellum of the LCA. In humans, asymmetry is seen in the right frontal petalia ans in the left occipal petalia, where in the LCA, there is a left dominant asymmetry of sulci in the frontal-orbital cortex and a left dominant asymmetry in the planum temporale. In addition to these differences, humans show a greater prefrontal cortex (PFC) gyrifraction than the LCA. Also, the LCA displays a longer primary motor cortex and primary somatosensory cortex than humans. Overall, it can be seen that the human brain is enlarged compared to the LCA (Nimchinsky et al., 1999; Sherwood et al., 2008).
It is been estimated that the human brain has around 100 billion neurons and 10 times more glial cells (Kandel et al., 2000; Noctor et al., 2007; Allen and barres, 2009; Ullian et al., 2001; Doetsch, 2003; Nishiyama et al., 2005, cited in Herculano-Houzel, 2009). Isotropic fractionation was used to investigate the cellular scaling rules that apply to brain allometry in different mammalian orders. By comparing the estimation of absolute numbers of neuronal and non-neuronal cells in the brains of different mammalian species to individuals, cellular composition of the human and primate brain could be made (Herculano-Houzel et al., 2007, cited in Herculano-Houzel et al., 2009). These rules lead to the estimation that a common primate brain, with a brain weight of around 1.5kg, should have 93 billion neurons and 112 billion non-neuronal cells, which consist of glial cells. This is about half of the brain cells. This means that the common primate brain should have a cerebral cortex of an estimated 1.4 kg, which contains 24 billion neurons. Furthermore, it contains a cerebellum that weighs 120 g, with 70 billion neurons.
When the same scale rules are applied to humans, it is seen that the human brain, with a weight of around 1.5kg, consists of 86 bullion neurons and 85 billion non-neuronal cells. This leads to the human cerebral cortex with an average weight of 1233 g, which contains 16 billion neurons.
When specifically looking at the cellular composition of great apes, it is seen that in the cerebellum of orang-utans and one gorilla that there is a match between the predicted sized of the cerebellum and cerebral cortex and their actual sizes. This proposes that these brains are built to similar scaling rules that apply to humans and other primates (Herculano-Houzel and Kaas, in preparation, cited in Herculano-Houzel et al., 2009).
To discover neuroanatomical differences between patients with ASD and their controls, different cortical regions, which are consistently altered in ASD, will be reported in this paragraph.
Fusiform gyrus
Different studies report that several layers of the FG (I to VI) show various neuropathological changes in patients with ASD, compared to an age-matched control subject. This is investigated by perfoming patches of the brain layers in subjects with autism and controls. It is seen that the subjects with ASD have a significantly lower neuron density in layer III, a significantly lower total numbers of neurons in layers III, V, and VI. Also, in layers V and VI, a significantly smaller mean perikambryal volume is shown. All these results are compared to controls (Varghese et al., 2017).
Prefrontal cortex
In subjects with ASD, the prefrontal cortex (PFC), which main role is cognitive control, is enlarged. It coordinates memory, planning and executive activity of different brain areas with behavioural outcomes (Fuster et al., 2001; Miller and Cohen, 2001, cited in Varghese et al., 2017). Young children (age 2-5 years) have abnormal overgrowth in the PFC (Carper and Courchesne, 2005; Carper et al., 2002; Hazlett et al., 2011; Shumann et al., 2010; Sparks et al., 2002, cited in Varghese et al., 2017). This is shown due to the fact that a decrease in neurons that are expressing layer- or cell-specific markers are identified in the dorsolateral PFC of subjects with ASD (Stoner et al., 2014, cited in Varghese et al., 2017). It is suggested that there is no global downregulation of expression in glia-specific markers, because these remain unchanged. In addition to this, it is suggested that subjects with ASD have cell- and region-specific abnormalities (Morgan et al., 2012, cited in Varghese et al., 2017).
Frontoinsular and cingulate cortex
Von Economo neurons (VENs), present in the frontoinsular (FI) and anterior cingulate cortices (ACC), are regularly affected in patients with ASD (Allman et al., 2002; Fajardo et al., 2008; Nimchinsky et al., 1999); Nimchinsky et al., 1995, cited in Varghese et al., 2017).
The VENs are a particular population of defined neurons. A core characteristic in a child with ASD is the resistance to change and routing. The anterior midcingulate cortex (aMCC) is involved in this, because of its role in decision-making during uncertain situations. It is seen that young children with ASD show a positive correlation between the numbers of pyramidal neurons and VENs (Uppal et al., 2014, cited in Varghese et al., 2017).
Hippocampus
Subjects of ASD show decreases in neuronal size, increased cell packing density and presence of less complex dendritic arboziation in the hippocampus. These are perhaps an indication of disorganised neuronal maturation (Bailey et al., 1998; Kemper and Bauman, 1993; Raymond et al., 1996, cited in Varghese et al., 2017).
This leads to the conclusion, that where the human brain differs from the LCA and other primates in a variation of regions, looking at neuroanatomical features. Also, the human brain is very much enlarged, compared to other primates. Nevertheless, this does not mean that brain size plays a main part in defining the uniqueness of the human brain. In addition to this, it is reported that, when looking at the cellular features, the estimated brain size of primates corresponds with those of humans. This is when the cellular composition is estimated by the rule described earlier.
From this, we can conclude that despite the fact that humans have certainly acquired many novel cognitive and neural specializations in the course of evolution, a broad spectrum of features are shared exclusively with our fellow great apes.
Language is a unique feature of humans
Humans are characterised by their unique ability to communicate by language. It is thought that syntax is the element of human language that most clearly separates humans from other animals. Syntax is the arrangement of words and phrases to create a sentence, which is well formed in a language (Wiener, 1984). How is language evolved in humans, compared to other primates?
Language is evolved anatomically different in humans, compared to other primates, such as chimpanzees. Kellog (1968), cited in Wiener (1984) shows in an experiment that it is not yet possible to teach a chimpanzee more than a few barely comprehensible words. Chimpanzees show anatomical disability to produce language, as we know in humans. By investigating the precise muscular control of the chimpanzee tongues, using only verbal communication of the American Sign Language (ASL) and maximal exposure this language by human beings (Gardner R. and Gardner B, 1969). It was shown that separation of the larynx and glottis is human specific. Lacking this control is due to the lips and larynxes (Noback, 1982, cited in Wiener, 1984). The separation of the larynx and glottis is only seen in humans, and therefore not present in other primates. This separation is an effect of evolution of upright posture in Hominids. It allows humans to produce a greater variety of speech sounds compared to other primates (Fink and Frederickson, 1978; Sutton, 1979, cited in Wiener, 1984).
It is seen that monkeys who are reared in social isolation are producing species-typical call types soon after birth but they do not invent new vocal sounds (Arbib et al., 2008).
As there is limited access to information about this subject, because not many studies are performed on language, it is too indistinguishable to get a clear view of the concept of language in humans. Although, it is seen that language is a unique feature in humans.
Different animal models for studying autism and its various characteristics
Animal models are widely used to study autistic characteristics, such as abnormal behaviour, seen in humans with ASD. Animal models are used to study ASD only show autistic symptoms. It is therefore questionable to use these animal models to study this disease.
There are a wide variety of animal models that are used to study ASD. Different types of animal models should be used for different type of research.
In the first study, transgenic monkeys that express human MeCP2 in the brain were created. To test if the transgenic monkeys show autism-like behaviours they perform a couple of tests to examine stress responses, interaction time when paired with other transgenic monkeys and cognitive functions of the transgenic monkeys. The stress response was measured by the threat-related anxiety and defensive test. The interaction time was measured with social interaction tests and the Wisconsin general test apparatus measured cognitive functions (Liu et al., 2016).
For examining the stress response, wild-type (WT) monkeys and transgenic (TG) monkeys are placed in a cage for 5 days while recording the locomotion every day for 20 minutes (n=8 for each group). The monkeys are aged 12-18 months.
The results of this test show that the stress responses, which are represented as alterations in locomotions, are higher for all the 8 TG monkeys compared to the WT monkeys.
For examining the interaction time, Liu et al., (2016) studied the average time WT monkeys spent with other WT monkeys or TG monkeys at age 18 months during an observation period of 60 minutes for 5 days (n=6, TG; n=9 WT).
This experiment shows that the average time a TG monkey sat with a WT monkey was significantly lower than the time a WT monkey sat with another WT monkey (Liu et al., 2016).
To examine the cognitive functions of the transgenic monkeys, the researchers performed tests by using the Wisconsin general test apparatus (WGTA). A black/white test was performed, shown in Figure 1. In this test the TG monkeys showed a considerable difference in time course, where the average time course of the WT and TG groups were similar (Liu et al., 2016). This shows that the TG monkeys differ in cognitive capabilities.
Figure 1. Black/white test
When animal models are used for experiments, there are four main criteria:
There must be a similarity in:
– The inductive conditions, which are obtained most of the time in animal models via manipulation; Methods of manipulation are for example: changing the environment or manipulate in a pharmacological manner;
– Behaviour;
– Underlying neurobiological mechanisms;
– Treatment response, which is based on a shared pharmacological identity.
Clinical aspects seen in mouse models can be linked to human behaviour with the disorder. When social interaction is tested, ‘autistic mice’ show a decrease in huddle, groom, barber, play behaviour, social exploration, sexual activity and aggression. When cognitive and communication impairments are examined in mice, they show a distress calls, mating calls and submissive calls. Stereotyped behaviours and other behaviours of autism can be studied in mice by the mice showing repeated motor activities, self-injurious behaviours and other self-involved behaviours (Tordjman et al., 2006).
There are different animal models to study autism. For the examination of social interaction, rat embryos were exposed to borna virus. After this, they showed a loss of cerebellar neurons associated with social interaction impairments. This reduced play behaviour in the treated animals (Hornig et al., 1999; Pletnikov et al., 1999, cited in Tordjman et al., 2006).
For the study of social interaction and stereotypies, the GS guinea pig is used. When Lev-Ram et al. (1993) used GS guinea pigs derived from albino Peruvian long hair littermates, they showed cerebrocortical and cerebellar defects, which naturally occurred. This lead to a significant decrease in interaction frequency with each other, compared to controls.
These GS guinea pigs showing autistic-like behaviours, which are associated with cerebellar abnormalities, make it an interesting model in the study of ASD (Bauman and Kemper, 1985; Bauman et al., 1985; Gaffney et al., 1987, cited in Tordjman et al., 2006). These are examples of animal models used in social and a combination of social and stereotypies.
Altogether, Liu et al. (2016) shows an increase in the frequency of repetitive circular locomotion, an increase in anxiety and reduced social interaction in the transgenic monkeys. Also, the monkeys show a relatively weak cognitive phenotype. Liu et al. (2016) state that with these results, genetically engineered macaque monkeys are efficient enough to study brain disorders.
Also, there are specific criteria for using an animal model to study autistic symptoms. Although, every study needs its own animal model, such as social interaction and stereotypies should be studied with GS guinea pigs, where rat embryos are valuable for studying only social interaction.
Discussion
In this review, the differences between humans and primates at neurological and cellular level are described, it is seen that humans have a very enlarged brain compared to other primates of the same weight. Also different regions of the brain are enlarged or increased in the human brain, compared to the LCA. At cellular level it is seen that a broad spectrum of features are shared exclusively with our fellow great apes by estimating the cellular composition.
When looking at the role of language in humans, it is seen that this is a unique feature evolved in Homo sapiens.
Furthermore, this review describes the use of different types of animals and its reliability. In the use of transgenic monkeys, overexpressing human MeCP2, an increased frequency of repetitive circular locomotion, an increase in anxiety and reduced social interaction is observed. Also it is reported that for different types of research, different animal models are necessary, but that they are, in fact, very reliable.
By these findings, we can say that the human brain differs in various ways from the brain of non-human primates. Also, the role of language, where disabled communication is a core characteristic in autism, is unique in humans.
This leads to the conclusion that autism is a human disease, and therefore not seen in any non-human primate.
In this review it is said that when studying the macaque monkeys, some sort of autistic features are visible. Yet, a core characteristic of ASD is the delay in language development, which is unique in humans. Although other animals/primates have a form of communication, it is nowhere near that of humans. This is why I believe that ASD is a human disease, although other primates can show several symptoms of ASD, which correspond to ASD in humans.
Although, despite the fact that humans have certainly acquired many novel cognitive and neural specializations in the course of evolution, a broad spectrum of features are shared exclusively with our fellow great apes.
Another criticism is that the human brain differs from others, because of its enlarged size. Sherwood et al. (2006) also state that this is a unique feature. On the contrary, our brain is not the biggest brain seen in species, what implies that our superior cognitive abilities cannot be explained by just the brain size (Herculano-Houzel, 2009). A thing we can conclude is that humans have a smaller body but a larger brain than great apes. In addition to this, Jerison (1973) and Marino (1998), cited in Herculano-Houzel (2009), report that the human brain distinguishes itself because the brain is 5-7 times too large for its body size, when looking at the relationship between body and brain size in other primates, which includes great apes. This leads to a conclusion that brain size is not a reliable indicator of cognitive function of the number of neurons across orders.
Language is a word without a universally accepted definition and criteria for its use does not exist. Before we can study language, we need to define very specifically what it means (Hedeager, 2012).
The experiment of Liu et al. (2016) it they used transgenic monkeys to where human MeCP2 in the brain was expressed. MeCP2 was known for its presence in patients suffering from Rett syndrome, a syndrome with very similar characteristics as ASD. This experiment can be criticized, because the monkeys are actually modificated to show Rett syndrome, not ASD. This makes the experiment doubtful.
While studying the behaviour of the WT and the TG monkeys, the researchers recorded them for 5 days, 20 minutes a day. The reliability of this assay can be questioned. By filming only 20 minutes a day for 5 days, the results may be biased. The results can be even more biased when they were recording the monkeys at the same time for 5 days. Perhaps they could have filmed more than 20 minutes a day at different time intervals.
Furthermore, we can say that the four criteria used to define an animal model for the study of autism, are widely interpretable. Every researcher can interpret this in its own way, what makes it difficult to draw a line in which animal model is the best.
In further research the gene expression of genes stronger associated with ASD could be studied by using genome-wide-association-studies (GWAS). An example of this can be by studying a mutation present in autism and Asperger syndrome, SHANK3, which is also known as ProSAP2. SHANK3 regulates the structural organization of dendritic spines. It also plays a role in the binding of neuroligins. The genes that encode for neuroligins are mutated in autism and Asperger. This makes that research into this could clarify more on the symptoms of ASD, instead of using a gene present in Rett syndrome (Durand et al., 2006.
Also, in addition to the research where the expression of human MeCP2 in transgenic monkeys is tested, it is possible elaborate on the differences between MeCP2 in humans compared to MeCP2 in monkeys.
In summary, additional research necessary to study autism in the human, and non-human primate brain.