The short version
Dopamine is made, moved, packaged, released, recycled and cleared in a sequence of steps. Each step depends on particular proteins, and each protein is built from a gene. The site's working model treats this as a chain: the overall signal can be limited by its weakest step. Each step's biology is well established (Tier A). Whether a weak step explains ADHD in any given person is an open question (Tier D).
The usual mix-up
"Low dopamine" is a slogan, not a mechanism. Dopamine signalling can be reduced by making too little, packaging too little, releasing too little, clearing too fast, or receiving poorly. Those are different problems with different genes, and they do not all respond to the same medicine.
The chain, step by step
| # | Step | Main proteins (gene) | What happens when it fails in a known single-gene condition | ADHD link |
|---|---|---|---|---|
| 1 | Raw material. Phenylalanine and tyrosine come from food. Phenylalanine is converted to tyrosine. | PAH | Phenylketonuria (PKU) | None established |
| 2 | Crossing into the brain. Amino acids use transporters across the blood-brain barrier. | LAT1 (SLC7A5) | - | Open research question (Tier D) |
| 3 | Making L-DOPA. Tyrosine is converted to L-DOPA. This is the rate-limiting step of dopamine synthesis. It needs a helper molecule, BH4. | TH; BH4 is made via GCH1 (rate-limiting for BH4) and other enzymes | Dopa-responsive dystonia family; TH deficiency [1, 2] | None established |
| 4 | L-DOPA to dopamine. | AADC (DDC) | AADC deficiency | None established |
| 5 | Packaging. Dopamine is loaded into vesicles so it can be stored and released in controlled amounts. | VMAT2 (SLC18A2) | Brain dopamine-serotonin vesicular transport disease [3] | One small platelet study, Tier C [4] |
| 6 | Release and reception. Vesicles release dopamine; receptors on the next cell respond. | Receptors D1 to D5 (DRD1 to DRD5) | - | Candidate-gene signals for DRD4 and DRD5 with heterogeneity, pre-GWAS [5] |
| 7 | Clean-up. Reuptake back into the cell, or breakdown. | DAT (SLC6A3); MAO; COMT | - | Candidate-gene signal for DAT1 with heterogeneity [5] |
Two points that often get muddled
- What powers VMAT2. VMAT2 moves dopamine into a vesicle using a proton gradient that the vesicle's own pump (a V-ATPase, which uses ATP) builds across its membrane. It is not driven directly by the mitochondrial proton gradient. Energy supply matters indirectly, through ATP. (Tier A)
- What the "rate-limiting" step is. In dopamine synthesis it is tyrosine hydroxylase, step 3 (Tier A, textbook biochemistry). Its helper molecule BH4 has its own rate-limiting step, catalysed by the GCH1 enzyme [2]. "Rate-limiting" describes synthesis, not the whole system. Packaging, release and clean-up set their own limits.
What medicines do in this picture
- Stimulants act at steps 5 to 7: they raise the amount of dopamine and noradrenaline available between cells. Methylphenidate mainly blocks the transporters that clear it. Amphetamines are taken up by the transporter and promote release. (Tier A)
- Levodopa medicines replace step 3. They still need step 4 (AADC) and step 5 (packaging) to work.
- Why that matters for diagnosis. A good levodopa response suggests that the steps after L-DOPA are working. Worsening on levodopa has been reported in a true VMAT2 loss-of-function condition, where direct dopamine agonists helped instead [3]. Specialists use response patterns as clues, always alongside examination and tests.
The working model: component-level bottlenecking
The site's throughline is that ADHD is best understood at the level of components. Think of a motherboard rather than a single clock. One weak component can limit a system that looks fine elsewhere, and different people can have different weak components. The model draws on established biochemistry (flux through a pathway is limited by its slowest steps). As an account of ADHD it is Tier D: a hypothesis that predicts that different people will have different bottlenecks, and that has not been tested person-by-person.
Limits
- The table is a teaching model. Real dopamine signalling has feedback loops, and the same symptom can come from different steps.
- A gene in the table is not an ADHD gene unless the ADHD column says so, and even then only at the stated tier.
- Single-gene conditions in the table are rare and mostly present in childhood. They are included because they show which steps are clinically important, not because they explain typical ADHD.
- No claim here is about any individual's biology.
Sources
Bibliographic details checked against PubMed where an ID is given.
- Weissbach A et al. Relationship of genotype, phenotype, and treatment in dopa-responsive dystonia: MDSGene review. Mov Disord 2022;37(2):237-252. DOI · PMID 34908184
- Novelli M et al. Autosomal recessive guanosine triphosphate cyclohydrolase I deficiency: redefining the phenotypic spectrum and outcomes. Mov Disord Clin Pract 2024;11(9):1072-1084. DOI · PMID 39001623
- Rilstone JJ, Alkhater RA, Minassian BA. Brain dopamine-serotonin vesicular transport disease and its treatment. N Engl J Med 2013;368(6):543-550. DOI · PMID 23363473
- Toren P et al. Eur Neuropsychopharmacol 2005;15(2):159-162. DOI · PMID 15695060
- Gizer IR, Ficks C, Waldman ID. Hum Genet 2009;126(1):51-90. DOI · PMID 19506906
Standard pharmacology and biochemistry textbooks cover steps 1 to 7 (Tier A). AADC deficiency, PKU, LAT1 and the V-ATPase/VMAT2 relationship are stated from standard references and have not been individually re-checked for this page.