Introduction
Autism and Asperger’s disorder (a better prognostic variety) that used to be classified under pervasive developmental disorders (PDD) are currently grouped under autistic spectrum disorders (ASD). These disorders in the past were labeled “schizophrenic reaction of childhood type”! Regardless of the terminology, ASD like schizophrenia are neurodevelopmental disorders and that is why in the past was classified under childhood psychoses or schizophrenic reaction of childhood. These neurodevelopmental disorders, while share some common features in underlying pathogenesis and clinical manifestations, they have their own specific differences at both pathophysiologic and symptomatic levels. These disorders share an early insult to the developing brain, through the prenatal maternal infections that trigger maternal and fetal immune reactions, damaging the fetal neurodevelopment. Although the insult such as infection could be non-specific, the severity, acuteness, latency of the invasion and the time window of neurodevelopment, determines which disorder to be manifested. Moreover the pathophysiology of these neurodevelopmental disorders seems to be a multi-steps process. In the first step there is the microbial attack at a specific time of neurodevelopment that triggers the second step of maternal and fetal immune reaction response that would be damaging the fetal brain development. In the third step, the microbial invasion causes genetic mutations to pass on its impact to the next generation through genetic inheritance.
First: The Prenatal Maternal Infections
The first time in the literature, Desmond and colleagues in 1967 reported a link between congenital rubella and autism, but without discussing the criteria of autism and the detail of behavioral symptoms. In 1971 Stella Chess, well known for the studies on children temperament with her husband, Alexander Thomas, reported the same link between congenital rubella and autism. Six years later, Chess in a follow up report of these children reported significant improvement in the most autistic symptoms, even improvement in their intelligence. Soon other congenital infections such as cytomegalovirus were linked with autistic symptoms as early as 6 months of age, and the infection hypothesis of autism was established as early as late 70’s and early 80’s. The infection theory of autism has also raised some concerns about the impact of common vaccines to children, specially MMR (Measles-mumps-rubella) vaccine. Although some biological assays have lent support to the association between measles virus or MMR and autism, some epidemiologic studies have shown no association between MMR and autism. This suspicion and concern led the American Academy of Pediatrics in 2000 to convene a conference including parents, practitioners, and scientists presenting for and against the link between MMR vaccine and ASD (Autistic Spectrum Disorders). Despite admitting to the possibility of such link, the conference concluded that there are not sufficient evidence to support such suspicion.
Atladóttir and colleagues in their large and longitudinal study on all children born in Denmark from 1980, through 2005, have reported no association between any maternal infection and diagnosis of ASDs in the child when looking at the total period of pregnancy, but high admission rates to hospital due to maternal viral infection in the first trimester and maternal bacterial infection in the second trimester associated with diagnosis of ASDs in the offspring. In the postmortem brains of autistic patients, several traces of infections with different viruses have been reported. Animal models of autism by injecting viral particles have also supported the infectious hypothesis of autism. Autistic like behaviors in the offsprings of rodents have been shown after the maternal injections of influenza virus, Borna virus, cytomegalovirus, and synthetic viral particles. Higher titers of autoantibodies to brain proteins/antigens have also been reported in autistic subjects. Also cytokines, such as interleukin (IL)-6, tumor necrosis factor (TNF) have also been found in the brain and blood of autistic patients. Some children with autism have also shown a significant reduction in stereotyped behaviors when treated with antimicrobials effective against intestinal clostridia.
Second: The Maternal and Fetal Immune Reactions
Higher titers of autoantibodies to brain proteins/antigens have been reported in autistic subjects. Also cytokines, such as interleukin (IL)-6, tumor necrosis factor (TNF) have been found in the brain and blood of autistic patients. These inflammatory or autoimmune reactions of the body cause damages to the frontal cortex of the brain through glial activation (microgliosis > astrocytosis) and apoptosis or neuron cell deaths that are prominent throughout the brain and persists for several weeks in concert with increased levels of proinflammatory cytokine mRNAs and progressive hippocampal and cerebellar damage. The level of measles antibody, but not mumps or rubella antibodies, has also been reported to be significantly higher in autistic children as compared with normal children or siblings of autistic children. The human immune reaction to infections or the immunopathogenesis of autism has been associated with the presence of pro-inflammatory cytokines and autoimmunity involving MHC (Major Histocompatibility Complex) region and HLA (Human Leukocyte Antigen) to infectious agents such as measles, herpesvirus-6, etc. The cell-mediated immunity in autism has been shown to be impaired as evidenced by low numbers of CD4 cells and a concomitant T-cell polarity with imbalance of different T-cells. Impaired humoral immunity on the other hand is evidenced by decreased IgA (Immunoglobulin) causing poor gut protection. Studies showing elevated brain specific antibodies in autism also support an autoimmune mechanism. Viruses may initiate the process but the subsequent activation of cytokines and other pro-inflammatory agents are the damaging factor causing autism. Inflammatory mediators in autism as discussed above involve activation of astrocytes, microglial cells, immune suppression by inhibiting T-cell proliferation and maturation and downregulation MHC, cytokine alteration of TNF-alpha, secretion of autoantibodies (IgG isotype) against neuron-axon filament protein (NAFP) and glial fibrillary protein, etc.
It has been suggested that even most heterogeneous symptoms of ASD have a common immunopathological pathway associated with dysregulation of glutamatergic neurotransmission in the brain with enhancement of excitatory receptor function by pro-inflammatory immune cytokines as the underlying mechanism. Maternal immune activation (MIA) to infections in second and third trimesters have been proposed in the etiology of neuro-developmental disorders such as schizophrenia and autism, with no specific differentiation between these disorders. The footprint of an early prenatal infection has been evidenced by a longer, ongoing hyper-responsive inflammatory-like state in many young as well as adult autism subjects, years after the first impact in the cerebral spinal fluid, blood and postmortem brains of autistic subject! This impact is not only limited to immune-pathogenesis but lead to strong and common gene expression changes in the embryonic brain. Most notably, there is an acute and transient upregulation of the α, β and γ crystallin gene family. Furthermore, levels of crystallin gene expression are correlated with the severity of MIA as assessed by placental weight.
Many of the proteins encoded by the major histocompatibility complex (MHC) play a vital role in the formation, refinement, maintenance, and plasticity of the brain. Manipulations of levels of MHC molecules by prenatal infections have been shown to significantly alter brain connectivity and function, evidenced in ASD (Autism Spectrum Disorders). Elevation of cerebrospinal fluid levels of tumor necrosis factor-alpha has been shown to be significantly higher than controls in autistic patients. Active neuroinflammatory process in the cerebral cortex, white matter, and notably in cerebellum have been also shown in autistic patients. Immunocytochemical studies have also demonstrated marked activation of microglia and astroglia, and macrophage chemoattractant protein (MCP)-1 and tumor growth factor-beta1, derived from neuroglia, as the most prevalent cytokines in brain tissues of autistic subjects. Elevated different cytokines have also been shown in many studies of autistic subjects, as early as age 2, as signs of neuro-inflammation of their brains but with no link to any specific one. In addition, increasing cytokine levels are associated with more impaired communication and aberrant behaviors in autistic children. Perhaps the most recent evidence of neuro-inflammation in autism as a result of an infection invasion, has been the evidence of increased serum levels of high mobility group box 1 protein in patients with autistic disorders. High-mobility group box 1 protein (HMGB1) is a nuclear protein that is released passively when cells such as neurons die after an insult such as microbial invasions. HMGB1 behaves as a trigger of inflammation, attracting inflammatory cells, and of tissue repair, recruiting stem cells and promoting their proliferation. Moreover, HMGB1 activates dendritic cells (DCs) and promotes their functional maturation and their response to lymph node chemokines. Activated leukocytes actively secrete HMGB1 in the microenvironment.
Despite the hardly debatable scientific evidence in the literature of the link between prenatal infections and causation of ASD and other neuro-developmental disorders, through a neuro-inflammatory reaction of the body, there is yet a lack of direct evidence between any specific infection, any specific inflammatory reaction and any specific neuro-developmental disorder. Therefore and thus far it seems that any prenatal infections, viral or bacterial, could cause an inflammatory reaction in defense in the pregnant mother and fetus through cytokines, MHC and HLA systems that disrupt the fetal developing brain and causing any of the neurodevelopmental disorders. Now lets see if there is any specific neuro-developmental time window in causing any of the neuro-developmental disorders specifically. Knowing what specific infection and what specific inflammatory reaction or at least what specific neuro-developmental time window causing what specific neuro-developmental disorder is essential in early prediction and even prevention of any of these disorders, as the current treatment of these disorders have never led to acceptable symptomatic reliefs, lest recovery!
Third: The Sensitive Neurodevelopmental Time Window
Meyer and colleagues argue that there is a “specific gestational windows” associated with a differing vulnerability to infection-mediated disturbances in normal brain development, and it seems that “the earlier the worse” would be the impact. The early gestational infections would not only interfere with fundamental neurodevelopmental events such as cell proliferation and differentiation, but also predispose the developing nervous system to additional failures in subsequent cell migration, target selection, and synapse maturation, eventually leading to multiple brain and behavioral abnormalities in the offsprings. It has been shown that early viral infections in late first trimester and middle-late second trimester in pregnant mice significantly alters levels of serotonin, 5-hydroxyindoleacetic acid, and taurine, modeling disruptions that occur in patients with schizophrenia and autism.
Four: The Genetic damage or vulnerability
As in schizophrenia, genetics does not seem to be directly involved, as the concordance rate for twins is only 60% and nill in dizygotic twins. Moreover the candidate genes in autism, while may share with other neuro-developmental or other psychiatric disorders, seem to be more associated with the phenotypic surface of the disorder, e.g. speech, language, and social aptitudes (region at 7q31-q33, cytogenetic abnormalities at the 15q11-q13 locus, & chromosome 15) than any direct underlying pathophysiology. Lastly like in schizophrenia, the genetic vulnerability in autism seems to be a byproduct of infections footprint in our genome in the form of mutation in our genes such as CNVs (Copy Number Variants) & SNP (Single Nucleotide Polymorphism).
Conclusion
Autism and other autism spectrum disorders (ASD) like any other neuro-developmental disorders such as schizophrenia, seem to be the result of prenatal infections that in turn cause maternal and fetal inflammatory reactions that all disrupt the normal brain development. While we have come so far to know this, we are still at the tip of the discovery iceberg, there is not yet any specificity in the type of infection, inflammatory reactions and the timing of the prenatal damage. If the clinical onset of any of this disorders could lead us to any timing assumptions, then it seems that the earlier the onset of the disorder, e.g. autism would be associated with an earlier infection, e.g. in the first trimester. In the same token, a later onset disorder such as schizophrenia, a later infection, e.g. in the second trimester and if we include disorders such as Bipolar Disorder, it could have been caused by infection insults in the third trimester! But these assumptions, while holding some factual evidence (e.g. 60), are in strong need of validation by the future research aiming specifically to find such timing links. The future research needs to identify if there is any specific associations between the type of prenatal infections or even maternal and fetal inflammatory reactions to any of these disorders, or all are non-specific and only timing of the insults is the crucial determining factor!
For the time being, as far as we are certain of the role of prenatal infections in the causation of autism and other neurodevelopment disorders, the prevention seems to be easier than the treatment! By educating the public, specially the mother about the role of infections during any stage of pregnancy and the risk of these disorders, use of low risk vaccinations, treating or preventing the infections could be a great step towards the prevention of these life-long disorders. Although we may not be able to prevent or treat any of these maternal infections, or the pregnant women may not wish to end their pregnancy for such a possible not proven risk, identifying the maternal and fetal inflammatory reactions that could be the main sharp edge of the insult, causing these disorders, could be a better feasible preventive approach! While we cannot recommend abortions upon presence of prenatal infections, we may be able to do so when destructive pathological maternal and fetal inflammatory reactions have been detected and presented to the pregnant mother and her family!