Autism spectrum disorders include patients who display a triad of impaired social interactions, deficits in communication, and restricted/repetitive interests and behaviors. However, the syndrome is clinically heterogeneous, likely a result of the diverse origins of autism spectrum disorder.
Presently, there are no clear biomarkers or effective treatments identified for autism spectrum disorder (ASD). Some causes of autism can be traced to single gene disorders, such as Fragile X syndrome, tuberous sclerosis and others; but distinct mutations, genetic syndromes, and de novo copy number variants only account for a minority of ASD cases.1
Diagnosis of ASD has steadily increased for 4 decades, and now 1 in 88 children are diagnosed with ASD;2 this rapid rise is unlikely due solely to inherited genetic change, and leads us to examine additional causes of autism that interplay with pre-existing genetic factors. ASD “susceptibility genes” confer vulnerability to developing an ASD in response to a second hit of environmental factors and stressors, in the absence of which a person might never develop a clinical ASD.
Environmental factors and stressors leading to ASD are thought to be mostly prenatal and perinatal, with many deriving from either immune dysfunction in the mother or the fetus/neonate. The resultant dysfunction may influence the developmental neuronal circuitry, myelination of the brain, and the neuronal and somatic environments.
Autoimmunity, susceptibility genes, placental dysfunction, maternal/neonatal infection, maternal/neonatal stress, and environmental toxins can all or each cause an increase in inflammation that is measurable by an increase in, among other responses, mast cell activation and corresponding changes in interleukin-6 (IL-6), tumor necrosis factor (TNF), granulocyte macrophage–colony stimulating factor (GM-CSF), extracellular mitochondrial DNA and other pro- and anti-inflammatory factors, sometimes evidenced by allergy-like responses and disruption of the blood–brain barrier causing somatic and brain inflammation.
These factors affect pathways involved in cell growth, proliferation and survival, and in messenger RNA (mRNA) transcription/translation. They are the focus of intense clinical study;3,4 in addition, biomarkers within specific brain regions are being identified that demonstrate chronic inflammatory responses, increased mitochondrial superoxide production and protein, and DNA destruction due to oxidation.5
Immune dysfunction may now be directly implicated in some of the neurologic deficits and developmental, plastic, and structural changes that cause the core behavioral and cognitive deficits of ASD.6 The immune system may become a therapeutic target for ASD symptom relief.
Autism, Neurobiology, and Inflammation
ASD is a heterogeneous group of neurodevelopmental disorders with core behavioral features, but it is also a highly heritable disease. Also, there are environmental elements, such as prenatal infection with rubella, anticonvulsant or antiemetic use in pregnancy, and perinatal hypoxia.
Postnatal infections including encephalitis and metabolic abnormalities also have been implicated in the development of ASD.7 It is thought that there needs to be “two hits” to develop ASD: a genetic vulnerability, followed by a second insult that may lead to the aberrant patterning and connection found in the ASD central nervous system (CNS).8
Brain specimens examined from autism patients show evidence of chronic inflammation and altered gene pathways associated with active immune signaling and immune function.9 Many genes found to be aberrant in autistic patients have direct roles in immune function, and it is thought that neuroimmune dysfunction contributes to or may cause certain forms of autism. Cytokines and other immune activators that affect neuronal pathways during critical periods of brain development may cause some of the core dysfunctional symptoms found in autism (see Figure).7
Figure. Model of genetic and environmental causes of autism. Cytokines and other immune activators that affect neuronal pathways during critical periods of brain development may cause some of the core dysfunctional symptoms found in autism.Image courtesy of Laura N. Antar, MD, PhD. Reprinted with permission.
Two large population-based studies examined perinatal responses that could be related to later ASD diagnosis. ASD was found in children with mothers with autoimmune diseases such as rheumatoid arthritis and celiac disease, as well as mothers with diabetes type 1, thyroid disorders, psoriasis, asthma, and allergies.9 Similarly, autistic patients have an increased innate and adaptive immune response, which suggests that there is localized inflammation and potential autoimmunity in the ASD patients themselves.8
Immunogenetics and Neurobiology
Although no single gene has been found to be the cause of ASD, researchers have analyzed gene-interaction networks, examining ASD-implicated genes that are highly expressed in normal brains during development. From this analysis, a subset of interactome (clusters of molecules that interact) networks revealed that immune signaling molecules in both neurons and glia were likely important in ASD signaling pathways, including NF-kappaB, mitogen-activated protein kinase (MAPK), TNF, transforming growth factor (TGF)-beta, Myc and Jnk. The analysis examined gene ontology, canonical pathways and interactome networks for every gene suggested in the autism database at the National Institutes of Mental Health (NIMH).10
Although the list is long and there are many other genes implicated in ASD, there are other neuro-immune genes implicated in ASD, such as: human leukocyte antigen (HLA)-DR4; the IL-4 receptor; the complement C4B null allele; macrophage migration inhibitory factor; MET tyrosine receptors; protein phosphatase and tensin homolog (PTEN); reelin, serine; and threonine kinase C gene (PRKCB1).11
Neuro-Immune Signaling Pathways
Gene expression and gene polymorphism research has uncovered pathways involved in both neuronal development and the immune system.9 Two examples are examined in more depth: the MET-HGF interaction and the PI3K-Akt-mTOR pathway. The hepatocyte growth factor (HGF) and tyrosine kinase metastasis receptor (MET) pathway is important for normal neuronal organization, is abnormal in patients with ASD, and abnormality doubles the risk of diagnosis.12 Increased levels of HGF are found in the cerebral spinal fluid (CSF) of ASD patients.
Meanwhile, decreased levels of MET, a receptor with tyrosine kinase activity that binds HGF, are found in ASD patient tissue.13 MET is located on the 7q31 locus that is associated by linkage studies with ASD. HGF is a neurotrophic factor for motor and sensory neurons in the parasympathetic nervous system, and in the CNS affects neuronal migration, dendritic development and morphology, and neurite outgrowth.
Additionally, MET is expressed in dendritic cells and in activated monocytes of the immune system. HGF-stimulated monocytes secrete chemo-attractants and have immunosuppressive effects; therefore, they may be involved in the inflammatory response.9 These findings point to the many roles, both immunologic and neurophysiologic, played by the interaction of this ligand-receptor pair, which is dysregulated in ASD.
Another example of neuronal and immunologic systems converging on one metabolic pathway highly associated with ASD occurs in the PI3K-Akt-mTOR system (see Figure 2) In this system, neuromodulators such as neurotransmitters, neurotrophins and growth factors, act on both Ras and PI3K, which in turn repress TSC1/TSC2 through ERK1/2 or Akt, respectively.
Figure 2. The PI3K-Akt-mTOR pathway and Ras-MEK pathway are sites where neuronal and immunologic systems converge. Altered protein function in these sites is corrolated with autism spectrum disorder.Adapted from Oostra BA, Willemsen R. FMR1: a gene with three faces. Biochim Biophys Acta. 2009;1790(6):467–477.
Additionally, immunomodulators such as cytokines, antigen receptors, and growth factors and co-stimulatory molecules act on Ras and PI3K as well. TSC1/2 then serves to inhibit Rheb, which activates the mTOR1 molecule, which then stimulates several important systemic functions:14 cell growth and metabolism (which affects T cells through activation and anergy), cell proliferation, cell survival (apoptosis is intimately bound with the immune system) and mRNA transcription and translation.
Within this complex pathway, there are several single gene disorders that are associated with ASD. These include the FMR1 gene, which is silenced in Fragile X syndrome and acts as a translational repressor in the mTOR system. Also included are the TS1/2 genes, which may be mutated in tuberous sclerosis and that act as an mTOR repressor; and the neurofibromatosis type 1 gene NF1, which acts on Ras GAP. The PI3K pathway is a central focus of much research in autism.
It may be that damage in one of these genes, or others in the pathway, provide the “susceptibility” while changes in ligand concentration and availability due to environmental stressors provide the “second hit.”
Immunologic factors have been directly associated with some of the core symptoms of children with ASD. Diminished immunoglobulin levels have been found in ASD, including overall low immunoglobulin (Ig) G and low IgA. Patients with low gamma globulin (for example, patients with DiGeorge syndrome) have been found to have autistic traits that somewhat remit with intravenous IgG infusion.16
Ashwood and colleagues15 found elevated levels of chemokines in the brain and CSF of children with ASD compared with typically developing counterparts, and they also found that increased chemokine concentrations were associated with greater aberrant behavior scores, development, and adaptive function.
The connection is that chemokines are found to regulate neuronal cell migration and proliferation as well as differentiation and intercommunication between neurons and microglia. It is not clear that the elevated chemokines have a direct impact on neuronal structure in this study; however, the data are suggestive enough to consider looking at plasma chemokine levels in children and noting if there are alterations in those with ASD.16
Inflammation and the Brain
In addition to genetic anomalies, dysfunction of certain cell types (over- or under-activation) has also been found in many patients with ASD, both adult and child. In the CNS, neuroglia such as astrocytes and microglia, and also perivascular macrophage and endothelial cells, have been shown to play important roles in neuronal health and function.
The microglia and astroglia play important roles in axonal guidance, synaptic plasticity, and the general organization of the cortex. Neuroglia protect neurons from over-excitation, and produce growth factors that nourish them. Astroglia, secondary to activation, can produce factors that modulate the inflammatory response such as pro-inflammatory cytokines, metalloproteinases, and chemokines.17
On the other hand, neural-immune cross-talk can be protective. Cytokines have been found to play important roles in normal neural development, and also in brain inflammation and disease. Anti-inflammatory cytokines can be neuroprotective, whereas proinflammatory cytokines can cause neuron and neural pathway destruction. Evidence is emerging that the balance between protective and destructive roles of cytokines in the brain predicts their beneficial or deleterious effect on the brain and cognition.17 There is evidence from postmortem studies that there is activation of astroglia and microglia in cerebellum, and ongoing neuroinflammation in the brains of people with ASD.18
Role of T Cells
In addition to genetic and cellular abnormalities, particular branches of the immune system are hypothesized to be overactive in ASD, activating these CNS cells. It is thought that ASD patients have overly robust innate and adaptive immune responses through the Th1 pathway, and questionably through the Th2 pathway. The Th1 and Th2 pathways respond to different immune responses. Th1 cells secrete proinflammatory cytokines like IFN-gamma, TNF-beta, and IL-2. Their job is to phagocytose and destroy intracellular microbes.
Th2 cells help to mount an extracellular response. They secrete cytokines such as IL-4 (which helps stimulate antibody production), IL-5 (which stimulates eosinophilic responses), IL-6 (both pro- and anti-inflammatory actions; and mediate IL-1 and TNF-alpha), and IL-13 (allergic inflammation), that are protective against extracellular parasites. There is also evidence for decreased lymphocyte numbers, and a diminished T-cell mitogen response.17
Neural-immune crosstalk may be detrimental; however, cytokines also play an important part in healthy brains, such as stem cell renewal, determining cell fates, differentiation of neurons, and even synaptic plasticity itself.4 Therefore, the degree of inflammation and immunomodulation that shifts the balance from beneficial to detrimental effects on cognition becomes the important question for research.
“Susceptibility genes” and environmental factors may play a role in abnormal inflammation in ASD; the necessary interplay between the two may explain why some patients are affected with ASD, whereas their siblings with similar genetics are not.
Pre-term birth and low birth weight, which can be due to a wide variety of factors, have been found to be associated with increased risk for ASD. However, environmental factors such as smoking early in pregnancy, maternal birth outside Europe or North America, being delivered by cesarean section, low Apgar scores, being small for gestational age, and congenital malformations,3 all of which can lead to premature birth or low birth weight, might be implicated more than the length of gestation or birth weight themselves.
Other factors also attributable to pre-term birth and low birth weight include intra-uterine inflammation that can lead to fetal brain injury, and obesity, which is also an independent factor for ASD.
Maternal stress, which stimulates corticotropin-releasing hormone (CRH), can induce release of vascular endothelial growth factor (VEGF) from mast cells.19 Mast cell activation may also be due to allergies, infection, and could lead to release of proinflammatory neurotoxicity and ASD pathology. Mast cell activation has been attributed to mastocytosis and monoclonal mast cell activation disorder in addition to secondary causes, including allergies, inflammation, cancer, urticaria, anaphylaxis, and mast cell activation syndrome.
Mast cells have been found to participate in diseases such as asthma, dermatitis, food intolerances, irritable bowel disease, interstitial cystitis, migraines, multiple sclerosis, and allergic rhinitis.14 Various cytokines and immune modulating factors serve as modulators of key pathways noted to be aberrant in patients susceptible to ASD.20
Auto-Antibodies and Maternal Anti-Brain Antibodies
Rates of autoimmunity and allergies are higher in some children with autism and also their parents. A large number of ASD children have auto-antibodies to specific brain proteins, such as anti-myelin antibodies, and anti-endothelial brain capillary antibodies. These can be detected in peripheral blood, and they react with CNS components.
It is unknown if these antibodies are primary or secondary pathology to an earlier neuronal insult. Antibody reactivity to some proteins in the cerebellum is associated with an ASD/autism diagnosis. These antibodies are associated with poorer cognition and adaptive function.21 However, although brain-reactive antibodies may segregate with different behavioral profiles, circulating brain-reactive antibodies may be common in typical development as well as ASD populations.9
Some mothers with ASD children have been found to have atypical prenatal immune responses; maternal infection during pregnancy has been implicated in autism. Anti-fetal brain antibodies are found in about 12% of mothers with ASD children. The damage is presumably due to the antibodies binding their neuronal antigen and interfering with subsequent neurodevelopment.
Some of the core features of ASD, such as stereotyped behaviors, impaired learning, and hyperactivity, can be replicated in animal models of autism. Macaque monkeys injected during development with IgG from human mothers with ASD children developed atypical behaviors, whereas macaques injected with IgG from mothers of normotypic children did not.14
One of the first challenges in determining a therapeutic approach to neuro-inflammation in ASD is to distinguish if the neuro-inflammation is a primary or secondary process to a neurobiological problem. Whether we are treating symptomatology or disease, elevated cytokine levels could be targets for future therapy, as control of the level, no matter the source, could be therapeutic. Such therapies could regulate anti-inflammatory cytokines (eg, IL-10) to reverse the process of inflammation. Similarly, stem cell therapy could target cytokine regulation. Other related challenges, such as analysis of blood or CSF sufficient to identify CNS markers of immune abnormality, are more practical.14
Excitotoxicity due to glutamatergic overactivity is also implicated in ASD and in patients with seizure disorders (50% to 60% of ASD patients). Epileptologists have found that cytokine changes may be responsible for epileptiform change. Valproic acid has helped to quell several problems, including seizures, in patients with ASD. Also, NMDA receptor antagonists have been considered for glutamatergic excitotoxicity; D-cyclosporine (an antibiotic that binds NMDA receptors) and memantine have improved symptoms in patients with ASD.9
Between 50% and 70% of children with ASD use some kind of complementary or alternative medicine therapy for symptom relief. These include probiotics, antibiotics, antifungals, dietary supplements, and antioxidants (like N-acetyl cytsteine).
However, there are only limited data examining the benefits of these and immunomodulators such as steroids; intravenous IgG (to block endogenous auto-antibodies and minimize T-cell activation of B cells that produce abnormal antibodies) and vitamin D supplementation. Treatment with immuno-suppressants such as prednisone and metronidazol (for patients complaining of comorbid autoimmune problems); and minocycline (for cytokine abnormalities) have been tried. Most studies are limited by size, by lack of placebo controls, or lack of randomization.14
ASD is a neurobehavioral developmental disorder that occurs early in life and may be triggered by interaction of several different risk factors, such as genetic susceptibility, neurotoxic exposures, infectious (maternal and fetal) exposure, metabolic deficits, all of which may interact early on to adversely affect neurogenesis, cortex and synapse formation, and neuroimmunity.
Neuroimmunity affects neuronal function, and certainly excitotoxicity can lead to apoptosis and trigger an immune response, which further affects neuronal function. Therefore, neuroinflammation may initiate or maintain CNS abnormalities during neurodevelopment. There are direct correlations emerging between autistic symptomatology and immunocytochemical abnormality.
The organism with aberrant neurodevelopment is subject to significant environmental consequences both within the CNS and within the world at large, as patients with autism are more likely to receive medications such as atypical antipsychotics, stimulants, antidepressant/anxiolytic therapies, nontraditional therapies, and special education environments, all of which continue to sculpt the impressionable and developing brain.
- Abrahams BS, Geschwind DH. Advances in autism genetics: on the threshold of a new neurobiology. Nat Rev Genet. 2008;9(5):341–355 doi:10.1038/nrg2346 [CrossRef]
- Autism and Developmental Disabilities Monitoring Network Surveillance Year 2008 Principal InvestigatorsCenters for Disease Control and Prevention. Prevalence of autism spectrum disorders — Autism and Developmental Disabilities Monitoring Network, 14 sites, United States, 2008. MMWR Surveill Summ. 2012; 61(3):1–19.
- Angelidou A, Asadi S, Alysandratos KD, Karagkouni A, Kourembanas S, Theoharides TC. Perinatal stress, brain inflammation and risk of autism — review and proposal. BMC Pediatrics. 2012; 12:89 doi:10.1186/1471-2431-12-89 [CrossRef]
- Asadi S, Theoharides T. Corticotropin-releasing hormone and extracellular mitochondria augment IgE-stimulated human mast-cell vascular endothelial growth factor release, which is inhibited by luteolin. J Neuroinflam. 2012;9(85) 1–6. doi:10.1186/1742-2094-9-85 [CrossRef]
- Rose S, Melnyk S, Pavliv O, et al. Evidence of oxidative damage and inflammation associated with low glutathione redox status in the autism brain. Transl Psychiatry. 2012;10 (2):e134. doi:10.1038/tp.2012.61 [CrossRef]
- Onore C, Careaga M, Ashwood P. The role of immune dysfunction in the pathophysiology of autism. Brain Behav Immun. 2012;26:383–392. doi:10.1016/j.bbi.2011.08.007 [CrossRef]
- Ashwood P, Wills S, Van de Water J. The immune response in autism: a new frontier for autism research. J Leukocyte Biol. 2006;80:1–15. doi:10.1189/jlb.1205707 [CrossRef]
- Li X, Chauhan A, Sheikh AM, et al. Elevated immune response in the brain of autistic patients. J Neuroimmunol. 2009;15:297(1–2):111–116. doi:10.1016/j.jneuroim.2008.12.002 [CrossRef]
- Careaga M, Van de Water J, Ashwood P. Immune dysfunction in autism: a pathway to treatment. Neruotherapeutics. 2010;7:283–292. doi:10.1016/j.nurt.2010.05.003 [CrossRef]
- Ziats MN, Rennert OM. Expression profiling of autism candidate genes during human brain development implicates central immune signaling pathways. PLOS One. 2011;6(9):e24691. doi:10.1371/journal.pone.0024691 [CrossRef]
- Ashwood P, Krakowiak P, Hertz-Picciotto I, et al. Altered T cell responses in children with autism. Brain Behav Immun. 2011;25:840–849. doi:10.1016/j.bbi.2010.09.002 [CrossRef]
- Campbel D, Sutcliff J, Ebert P, et al. A genetic variant that disrupts MET transcription is associated with autism. Proc Natl Acad Sci U S A. 2006;102:16834–16839. doi:10.1073/pnas.0605296103 [CrossRef]
- Pardo C, Eberhart C. The neurobiology of autism. Brain Pathol. 2007;17:434–447. doi:10.1111/j.1750-3639.2007.00102.x [CrossRef]
- Chez M, Guido-Estrada N. Immune therapy in autism: historical experience and future directions with immunomodulary therapy. Neurotherapeutics. 2010;7:293–301. doi:10.1016/j.nurt.2010.05.008 [CrossRef]
- Ashwood P, Krakowiak P, Hertz-Picciotto I, et al. Associations of impaired behviors with elevated plasma chemokines in autism spectrum disorders. J Neuroimmunol. 2011;232:196–199. doi:10.1016/j.jneuroim.2010.10.025 [CrossRef]
- Pardo C, Vargas D, Zimmerman AW. Immunity, neuroglia and neuro-inflammation in autism. Int Rev Psychiatry. 2005;17(6)485–495. doi:10.1080/02646830500381930 [CrossRef]
- McAllister A, van de Water J. Breaking boundaries in neural-immune interactions. Neuron. 2009;64:9–12. doi:10.1016/j.neuron.2009.09.038 [CrossRef]
- Ashwood P, Corbett B, Kantor A, et al. In search of cellular immunophenotypes in the blood of children with autism. PLoS One. 2011;6(5): e19299. doi:10.1371/journal.pone.0019299 [CrossRef] . doi:10.1371/journal.pone.0019299 [CrossRef]
- Angelidou A, Alysandratos K, Aadi S, et al. Brief report: “allergic symptoms” in children with autism spectrum disorders. More than meets the Eye?J Autism Dev Disord. 2011;41:1579–1585. doi:10.1007/s10803-010-1171-z [CrossRef]
- Goines P, Haapanen L, Boyce R, et al. Autoantibodies to cerebellum in children with autism associate with behavior. Brain Behav Immun. 2011;25(3):514–523. doi:10.1016/j.bbi.2010.11.017 [CrossRef]
- Rossi C, Van de Water J, Rogers SJ, Amaral DG. Detection of plasma autoantibodies to brian tissue in young children with and without autism spectrum disorders. Brain Behav Immun. 2011; 25:1123–1135. doi:10.1016/j.bbi.2011.02.011 [CrossRef]