Decoding Autism Now
Biology of Autism — Molecular Origins
01

The Core Hypothesis

A mechanistically distinct subgroup of ASD with a single convergent biological origin

Immune-derived autism — a subgroup, not all of ASD

Autism spectrum disorder does not have a single cause. A mechanistically distinct subgroup — immune-derived autism (IDA) — arises from a converging multi-step biological cascade that originates in gut pH dysregulation, progresses through immune activation and metabolic disruption, and converges on the silencing of somatostatin-14 (SST-14) interneurons in the cortex, hippocampus, and hypothalamus.

SST-14 interneuron silencing disrupts the coordinated release of three neuropeptides — oxytocin, vasoactive intestinal peptide (VIP), and secretin — producing the social, sensory, sleep, gastrointestinal, cognitive, and motor features of the ASD phenotype cluster.

This model does not claim to explain all autism. Primarily genetic subtypes — Fragile X, Rett syndrome, tuberous sclerosis, and defined single-gene mutations — involve distinct mechanisms. The claim is more specific: to provide a mechanistically coherent, biomarker-definable account of the immune-derived subgroup with sufficient precision to design testable interventions.

Immune-derived autism (IDA) SST-14 interneurons Convergent node Oxytocin / VIP / Secretin Biomarker-definable subgroup
The cascade from start to finish
02

The Cascade at a Glance

Seven interconnected mechanisms operating simultaneously — not in sequence

This page covers the founding conditions — the initiating insults at the top of the cascade. Each subsequent mechanism is described in detail in its own page in this suite. The overview below shows where founding conditions fit in the full chain.

Immune-derived autism — cascade overview
1

Founding conditions — prenatal, perinatal, environmental, toxic, and ongoing insults elevate gut pH and disable parietal cell hydrochloric acid (HCl) production. This page.

2

Pepsin inactivation and opioid peptide accumulation — elevated gut pH deactivates pepsin; casein and gluten proteins are incompletely digested; casomorphin and gliadorphin accumulate and initiate two parallel downstream arms. Pepsin & Opioid Peptides →

3A

CCK overactivation / gut SST-28 dysregulation — opioid receptor activation drives chronic CCK-A overactivation; gut SST-28 becomes tonically overexpressed, suppressing secretin, VIP, and gastric acid — deepening pH dysregulation. Pepsin & Opioid Peptides →

3B

CD26 blockade / adenosine accumulation / methylation failure — opioid peptides block the CD26 ADA binding site; adenosine accumulates; methionine synthase is rate-limited; S-adenosylmethionine (SAMe) production falls; neurotransmitter regulation, immune switching, DNA methylation, and cellular energy production all fail simultaneously. CD26, Adenosine & Methylation →

4

Lipopolysaccharide (LPS) translocation and systemic immune activation — compromised gut barrier allows bacterial LPS — the cell wall toxin of gram-negative bacteria — into systemic circulation; persistent toll-like receptor 4 (TLR4) and nuclear factor kappa B (NF-κB) activation generates the cytokine environment that drives the two arms of SST-14 silencing. Cascade Explained →

5

SST-14 interneuron silencing — indoleamine 2,3-dioxygenase 1 (IDO1) excitotoxic arm and NF-κB transcriptional suppression arm converge simultaneously on SST-14 interneurons; the biological latch engages. Cascade Explained →

6

Neuropeptide cascade disruption — oxytocin release becomes blunted and uncoupled from social context; VIP G-protein signaling fails across four biological systems; secretin faces compound failure from above and below. Cascade Explained →

7

Observable phenotype — social motivational deficit, modality-nonspecific sensory processing abnormalities, sleep architecture disruption, GI dysmotility, motor coordination difficulties, cognitive rigidity. Not independent comorbidities — simultaneous downstream expressions of one upstream mechanism.

Where the cascade begins
03

Prenatal and Perinatal Founding Conditions

Insults that establish gut pH vulnerability before — or at the moment of — birth

The founding conditions described here are not single sufficient causes. No child develops immune-derived autism from one exposure alone. They are vulnerability-establishing events that narrow the biological margin within which subsequent insults must operate to initiate the cascade. A child carrying several of them requires a smaller additional burden to cross the pH dysregulation threshold; a child carrying none may tolerate exposures that would tip a more vulnerable child.

Gestational
Prenatal hormonal or immune disruption

The enteric nervous system differentiates between weeks 5 and 12 of gestation. Exogenous hormones, maternal stress hormones, maternal infection, or transplacental autoantibodies can alter the development of acid-producing parietal cells before birth — producing a gut that enters postnatal life with structurally reduced HCl production capacity.

Delivery
Cesarean delivery and microbial seeding failure

Vaginal delivery seeds the infant gut with maternal lactobacillus and bifidobacterium species that establish acid-producing commensal populations. C-section delivery bypasses this entirely. Hospital-acquired streptococcal species — which also block the CD26 receptor through streptokinase — can establish dysbiotic colonization from day one of life.

Neonatal
Failure to establish nursing

Breast milk oligosaccharides selectively feed acid-producing commensals. Colostrum shapes early gut immune tolerance. The sucking reflex activates motilin-driven motility. Low muscle tone preventing nursing, premature birth, or early formula dependence removes all three pH-regulating inputs simultaneously.

Maternal nutrition
Oral contraceptive nutritional depletion

Combined OCPs deplete B6, B12, folate, zinc, and magnesium through increased urinary excretion and altered absorption. Zinc depletion is specifically consequential — zinc is the essential cofactor for carbonic anhydrase, which generates the hydrogen ions parietal cells use to produce HCl. A woman conceiving shortly after stopping OCPs may begin pregnancy with depleted reserves across all five nutrients.

Immune autoantibody
Folate receptor alpha antibodies

FRAA block methylfolate transport into the brain independently of dietary folate status. This creates a dual methylation vulnerability when combined with cascade-driven adenosine accumulation in later steps: the methionine synthase cycle is starved of its methyl donor from folate independently of its rate-limitation by adenosine. The Frye et al. folinic acid trial — the single published ASD intervention with consistent positive results — targets precisely this mechanism.

Genetic
Variants affecting acid production and methylation

Proton pump gene polymorphisms, carbonic anhydrase variants, and zinc transporter variants reduce parietal cell HCl output from birth. methylenetetrahydrofolate reductase (MTHFR) C677T — documented at elevated frequency in ASD populations — reduces methylfolate conversion by approximately 70% in homozygous carriers, compounding the adenosine-driven methylation bottleneck of later cascade steps with a genetically determined methyl donor deficit.

maternal immune activation (MIA) / prenatal immune disruption C-section microbiome failure Nursing / motilin OCP zinc depletion Folate receptor antibodies MTHFR C677T Carbonic anhydrase / zinc
Environmental and toxic initiating conditions
04

Environmental and Toxic Founding Conditions

Exposures that disable gut ecology, block the CD26 receptor, or disrupt adenylyl cyclase signaling

Glyphosate and microbiome depletion

Glyphosate — the active ingredient in broad-spectrum herbicides — disrupts the shikimate pathway in bacteria, selectively depleting the commensal species (Lactobacillus, Bifidobacterium, Enterococcus faecalis) that establish the acid-producing gut ecology required for normal pH maintenance, while sparing many LPS-producing gram-negative pathogenic species. Maternal dietary glyphosate exposure depletes the microbiome seed delivered to the infant at birth, establishing dysbiotic colonization that favors the inflammatory inputs driving subsequent cascade steps.

Mercury and the CD26 receptor

Mercury binds directly to the CD26 receptor site on lymphocyte immune cells, blocking adenosine deaminase from docking and initiating the same adenosine accumulation that casomorphin and gliadorphin produce through the same receptor in later cascade steps. Pre-1990s dental amalgam fillings release mercury vapor continuously. Mercury crosses the placenta and reaches the developing fetal CD26 receptor system during gut and immune development — a prenatal CD26-blocking exposure that predates any dietary or infectious insult. Thimerosal, environmental mercury, and dietary fish exposure represent additional routes; effects from multiple concurrent sources are additive.

Mercury's relevance to the cascade is not through neurotoxicity alone — it is specifically through the CD26/adenosine/methylation arm described on CD26, Adenosine & Methylation. This is the mechanistic link that connects mercury exposure to methylation failure rather than treating them as separate phenomena.

Organophosphate pesticides and vagal tone suppression

Organophosphate pesticides inhibit acetylcholinesterase, the enzyme that degrades acetylcholine after vagal nerve terminal release. The vagus nerve drives parietal cell HCl production by releasing acetylcholine onto parietal cell muscarinic receptors. Chronic organophosphate exposure produces muscarinic receptor downregulation through persistent overstimulation — paradoxically reducing parietal cell acid responsiveness over time and contributing to the pH elevation that initiates downstream cascade mechanisms.

Organophosphate flame retardants (OPFRs) — including TDCIPP and TCPP — saturated polyurethane foam in infant sleep surfaces, car seats, and nursing pillows under flammability regulations in force from 1975 to approximately 2015. These compounds migrate from foam into household dust and are detected in cord blood at concentrations confirming prenatal exposure through maternal body burden.

Viral disruption of adenylyl cyclase signaling

Several viruses directly target adenylyl cyclase signaling. Bordetella pertussis toxin ADP-ribosylates the inhibitory G-protein Gαi, locking it in its inactive form and disrupting the contextual regulation of cAMP that SST-14 phasic signaling requires. Herpesviruses including cytomegalovirus and Epstein-Barr virus produce proteins that interact with G-protein regulatory components. A significant viral infection during a critical developmental window may independently contribute to SST-14 transcriptional suppression — through the same cAMP/cAMP response element-binding protein (CREB) pathway disrupted by adenosine accumulation in Arm 2B — before the inflammatory cascade has fully established itself through the gut-immune pathway.

Glyphosate / microbiome depletion Mercury / CD26 blockade Organophosphates / vagal tone OPFR flame retardants Pertussis toxin / Gαi CMV / EBV adenylyl cyclase
Ongoing drivers that perpetuate pH dysregulation
05

Ongoing Drivers

Conditions that maintain elevated gut pH once the cascade has initiated — making resolution unlikely without addressing multiple factors simultaneously

The founding conditions described in the preceding sections initiate the cascade. The ongoing drivers described here are what make it self-perpetuating. A child whose cascade was initiated by prenatal conditions and a disrupted birth microbiome may be maintained in the dysregulated state by any combination of the drivers below — independently of the original initiating conditions, which may no longer be active.

  • H. pylori infection — alkalinises the stomach by producing urease, converting urea to ammonia and directly elevating gastric pH. It damages parietal cells through inflammatory mechanisms, creating a self-perpetuating high-pH environment. H. pylori can be transmitted in early childhood and establishes silently.
  • Recurrent streptococcal infections — streptococcus produces streptokinase, which blocks the CD26 receptor by occupying the adenosine deaminase binding site — the same mechanism by which opioid peptides and mercury block CD26. Each streptococcal episode reintroduces this blocker, extending the adenosine accumulation and methylation suppression cycle. This is why recurrent strep is not just a frequent childhood illness in these children — it is a direct driver of the cascade mechanism.
  • Chronic sympathetic dominance and reduced vagal tone — in chronic stress states (persistent sensory overload, immune activation, gut discomfort, anxiety), sustained sympathetic nervous system dominance suppresses vagal tone. The parietal cell receives insufficient acetylcholine stimulation for adequate HCl production. Contrary to the widely held belief that stress causes excess stomach acid, chronic stress characteristically reduces acid production through vagal withdrawal.
  • Acetaminophen use — in young children with compromised sulphation capacity from the transsulfuration bottleneck of the methylation cascade, acetaminophen metabolism depletes glutathione and compounds oxidative stress at the methylation failure step, reducing the metabolic resilience available to SST-14 interneurons under quinolinic acid excitotoxic pressure.
  • Proton pump inhibitor and H2 blocker therapy — medications prescribed for reflux directly suppress parietal cell acid production. Reflux in the context of gut dysbiosis and impaired motility frequently reflects low motility and fermentation pressure rather than acid overproduction. Acid suppression therapy in this context may relieve the surface symptom while deepening the underlying pH dysregulation — the opposite of what the cascade requires.

The self-reinforcing loop: Once CD26 is blocked by any mechanism — opioid peptides, streptokinase, or mercury — adenosine accumulates, the methionine synthase cycle slows, and the cellular energy available to drive parietal cell activity falls. This reduces gastric acid production further, which generates more opioid peptide fragments, which further block CD26. The cascade maintains itself independently of whether the original triggering conditions remain present. This is why addressing a single driver produces temporary improvement followed by relapse — the loop must be interrupted at multiple points simultaneously.

H. pylori / urease Streptococcal streptokinase Vagal tone / sympathetic dominance Acetaminophen / glutathione PPI therapy / acid suppression Self-reinforcing CD26 loop
Why most children exposed to these conditions do not develop IDA
06

The Threshold Model

Why the same exposures produce different outcomes — and what tips the balance

Most exposed children do not develop immune-derived autism

The founding conditions described on this page — C-section delivery, antibiotic exposure, formula feeding, acetaminophen use, glyphosate dietary exposure — are common across the general population. Most children who experience them do not develop immune-derived autism. Three factors explain this threshold relationship.

Genetic background determines the baseline. Variants in carbonic anhydrase, proton pump genes, MTHFR, AHCY, and CD26 receptor efficiency each reduce the margin within which environmental insults must operate to produce cascade initiation. A child carrying several of these variants requires a smaller environmental burden to cross the threshold than one carrying none.

Co-occurring insults across multiple steps are required. C-section delivery alone is insufficient. C-section delivery combined with antibiotic treatment, formula feeding that eliminates breast milk oligosaccharide protection, and early acetaminophen exposure represents a qualitatively different biological burden. The epidemiological associations are weakest when individual factors are examined in isolation and strongest when exposure combinations are modelled.

Timing within critical developmental windows determines impact. The enteric nervous system differentiates between weeks 5 and 12 of gestation; the first months of postnatal life are critical for microbiome establishment. The same exposure at six months postnatally produces substantially less cascade burden than at six days. The majority of children exposed to individual founding conditions without concurrent vulnerabilities at the right developmental moment do not cross the biological threshold required for sustained cascade propagation.

Cascade burden and the clinical threshold

Threshold → SST-14 silencing
Low burden — subclinical → Threshold crossed → Immune-derived autism expressed

This threshold model also explains regressive autism — where a child with apparently normal early development loses skills between 18 and 36 months. Compensatory mechanisms may maintain adequate SST-14 interneuron output while cascade burden accumulates below threshold. A second biological challenge — febrile illness, streptococcal infection, antibiotic-induced microbiome disruption — crosses the threshold at which SST-14 interneurons can no longer maintain coordinating output. The regression is the moment the threshold was crossed, not the moment pathology began.

What the biomarkers reveal
07

The Biomarker Signature

Laboratory findings that identify immune-derived autism and distinguish it from other ASD subtypes

Immune-derived autism is distinguishable from genetic and structural ASD subtypes by a specific biomarker panel. These markers are not merely associated with ASD — each one corresponds to a specific mechanistic step in the cascade described on this page and its companion pages. A child carrying this biomarker combination has a measurable, mechanistically coherent upstream driver.

Biomarker Direction Cascade mechanism
Kynurenine:tryptophan (K:T) ratio ↑ Elevated IDO1 activation — active tryptophan diversion away from serotonin; confirms excitotoxic arm is operative
Plasma quinolinic acid (QUIN) ↑ Elevated Kynurenine pathway overactivation; direct N-methyl-D-aspartate (NMDA) excitotoxic pressure on SST-14 interneurons
Plasma homocysteine ↑ Elevated Methylation cycle stall from adenosine rate-limiting of methionine synthase; the CD26 arm fingerprint
Lactate:pyruvate (L:P) ratio ↑ Elevated (>25) Mitochondrial respiratory chain dysfunction — State 2 SST-14 metabolic exhaustion indicator
Neural autoantibody panel ↑ Positive titres Adaptive immune activation against SST-14 interneuron surface proteins — State 1 functional silencing
Cytokine panel (IL-6, TNF-α, IL-1β) ↑ Two or more elevated Active NF-κB-driven inflammatory environment; upstream IDO1 and transcriptional suppression drivers
Red blood cell glutathione ↓ Depleted Transsulfuration bottleneck from methylation failure; reduced antioxidant defense at SST-14 interneurons
Plasma amino acids (Phe, Tyr, Trp) ↓ Low despite adequate diet Pepsin inactivation — proline-bonded essential amino acids not released from dietary protein
dipeptidyl peptidase IV (DPP-IV) / CD26 activity ↓ Reduced CD26 receptor blockade by opioid peptides, streptokinase, or mercury — adenosine arm operative

Two or more of the above criteria met constitutes a strong signal for the immune-derived autism subgroup. The combination of K:T ratio elevation + cytokine elevation + homocysteine elevation + autoantibody positivity is the core enrolment criterion for the biomarker-stratified trial described in the IMIG Clinical Trial Initiative. Full guidance on testing is on the Testing Strategy and Test Reference pages.

Where to go from here

Where to Go Next

Each page in the Foundation series goes deeper into one mechanism

i

Framework note: This page describes the immune-derived autism (IDA) subgroup — a mechanistically coherent, biomarker-definable subgroup of ASD. The model does not claim to explain all autism. Primarily genetic subtypes — Fragile X, Rett syndrome, tuberous sclerosis, and defined single-gene mutations — involve distinct mechanisms. The cascade described here applies to a specific, identifiable subgroup. Individual mechanistic steps are supported by published primary literature; the integrated cascade as a complete unified model awaits prospective validation. This is a research framework, not a clinical protocol. All intervention decisions require qualified clinical oversight.