Mitochondrial alterations in adult ADHD: insights from hiPSC-derived neurovascular models of PARK2 copy number variants

Background

Attention-deficit/hyperactivity disorder (ADHD) is a prevalent neurodevelopmental condition affecting 5–8% of children and adolescents worldwide and often persists into adulthood. Despite available treatments, up to 60% of adults experience inadequate symptom relief. ADHD is highly heritable and has been linked to genetic factors, including copy number variants (CNVs) in the mitochondrial quality control gene PARK2. These CNVs might influence mitochondrial and cellular function, particularly in energy-demanding cells such as cortical neurons (CNs) and brain capillary endothelial cells (BCECs).

Aims

To investigate the effects of PARK2 CNVs on mitochondrial dynamics during neuronal maturation and on blood-brain barrier (BBB) function in ADHD using human induced pluripotent stem cell (hiPSC)-derived neuronal and neurovascular models, and to assess responses to oxidative stress and polyunsaturated fatty acid (PUFA) supplementation.

Methods

hiPSC-derived neuronal progenitor cells (NPCs), immature, maturing and mature CNs, and BCECs were generated from ADHD individuals with PARK2 CNVs (n=4), ADHD wildtype (WT) individuals (n=4), and healthy WT controls (n=4). Cellular identity and functionality were assessed by immunofluorescence and multielectrode array electrophysiology. Reactive oxygen species (ROS) were measured using DCFDA at baseline and after tert-butyl hydroperoxide (TBHP)-induced oxidative stress. Mitochondrial biogenesis and adenosine triphosphate (ATP) production were assessed by COX-I/SDH-A ratios and Seahorse flux analysis. Mitochondrial morphology was quantified from MitoTracker-labelled confocal images, and mito–lysosome colocalisation was analysed as a proxy for mitophagy. BBB integrity was assessed in static CN–BCEC co-cultures by transendothelial electrical resistance. Statistical analyses used one-way or two-way ANOVA with Tukey’s post hoc tests; significance was set at p<0.05.

Results

Differentiation capacity and cellular identity of the generated models were preserved across genotypes. No differences in ROS levels were observed in NPCs. In PARK2 CNV neurons, ROS was elevated from the immature stage onwards, with strongest effects in maturing CNs at baseline (p<0.0001) and after TBHP (p=0.006), persisting in mature CNs at baseline (p<0.0001) and after TBHP (p=0.0002). Two-week PUFA supplementation (10 µM EPA + 10 µM DHA) reduced TBHP-induced ROS across genotypes (p<0.0001) and selectively reduced baseline ROS in PARK2 CNV neurons (p≤0.003), abolishing genotype differences. Maturing PARK2 CNV neurons showed increased mitochondrial sphericity at baseline (p=0.0006) and after TBHP (p=0.0014), and reduced mean branch length at baseline (p=0.0046) and after TBHP (p=0.0036), indicating a more fragmented mitochondrial network. Mito–lysosome overlap events were reduced in both ADHD-derived groups compared with healthy controls (WT ADHD: p=0.0178; PARK2 CNV: p=0.0167), without differences between ADHD groups (p=0.9991). Mitochondrial biogenesis and ATP production were unchanged across groups. BBB integrity was unaffected by genotype, and both CNs and BCECs retained their cellular identity in co-culture, validating the model.

Conclusions

PARK2 CNV neuronal models showed increased oxidative stress and altered mitochondrial morphology during maturation, without detectable changes in mitochondrial biogenesis, ATP production or BBB integrity. Reduced mito–lysosome association across ADHD-derived neurons suggests mitochondrial quality-control alterations independent of PARK2 CNV status. These findings support a potential role for mitochondrial dysregulation in ADHD, with PARK2 CNVs contributing to oxidative and structural mitochondrial alterations in a subgroup of patients. PUFA-mediated ROS reduction may represent a genotype-informed strategy warranting further investigation.

References

No references