Discussion

We found that adolescents with ADHD had reduced circulating α-klotho compared with healthy peers, independent of demographics, adiposity, clinical symptom burden and current medication, which may suggest klotho as a trait marker of ADHD. In addition, klotho levels were negatively associated with the percentage of non-perseverative errors on WCST, indicating a correlation between lower klotho levels and worse executive function. Following Bonferroni correction for 4 comparisons, the significance threshold was adjusted from α = 0.05 to 0.0125; the association between klotho levels and percentage non-perseverative errors remained significant (P = 0.002 < 0.0125), indicating a robust relationship between klotho and executive function.

Studies of klotho have primarily been conducted among middle-aged and older adults, due to its known age-suppressive qualities. ^{Reference de Vries, Staff, Noble, Muetzel, Vernooij and White3,Reference Alex, Buss, Davis, Campos, Donald and Fair4} In the current decade, the neurodevelopmental role of klotho has gained scientific and clinical attention, which may reflect evidence linking neurodevelopmental and neurodegenerative processes. ^{Reference de Vries, Staff, Noble, Muetzel, Vernooij and White3,Reference Alex, Buss, Davis, Campos, Donald and Fair4} Increasing evidence suggests that klotho levels are associated with those of psychological stress and cognitive function in young people. ^{Reference de Vries, Staff, Noble, Muetzel, Vernooij and White3,Reference Birdi, Tomo, Sharma, Yadav, Charan and Sharma6,Reference Beckner, Conkright, Eagle, Martin, Sinnott and LaGoy14} Using in situ hybridisation, Clinton et al found that levels of klotho messenger RNA expression among developing and adult rat brains in regions such as the cortex, hippocampus and amygdala increased during the first week of life, decreased during the second week and then gradually increased again to reach adult levels by postnatal day 21. ^{Reference Clinton, Glover, Maltare, Laszczyk, Mehi and Simmons15} Beckner et al assessed klotho levels and cognitive function in 54 young service members who underwent a 5-day simulated military training protocol that included daily physical exertion and 48 h of restricted sleep and caloric intake, which led to a 7.2% reduction in klotho levels and a 5.6% decrease in working memory. ^{Reference Beckner, Conkright, Eagle, Martin, Sinnott and LaGoy14} A klotho genotyping neuroimaging study of 1387 children and adolescents aged 3–21 years found that individuals carrying KL-VS heterozygotes had better cognitive function, including executive function and attention, than those without, particularly before the age of 11 years. ^{Reference de Vries, Staff, Noble, Muetzel, Vernooij and White3} Gupta et al found that klotho acutely increased synaptic plasticity in the hippocampus in male mice, in a histological pattern similar to that in individuals undergoing cognitive stimulation, suggesting that klotho had enhanced cognition in young mice and countered cognitive deficits in older mice. ^{Reference de Vries, Staff, Noble, Muetzel, Vernooij and White3} The findings of de Vries et al and Gupta et al imply that insufficient klotho levels may lead to impaired attention and executive function, an endophenotype of ADHD, during critical neurodevelopmental periods. ^{Reference de Vries, Staff, Noble, Muetzel, Vernooij and White3,Reference Gupta, Moreno, Wang, Leon, Chen and Hahn16}

Our study results suggested that reduced klotho levels may serve as a trait marker of ADHD and related executive dysfunction, independent of clinical symptoms and ADHD medication use. Siahanidou et al measured klotho levels in 50 healthy neonates – 25 preterm (with a mean gestational age of 33.7 weeks) and 25 full-term – on days 14 and 28 of life, and found that these were significantly higher in full-term compared with preterm infants at both time points. ^{Reference Siahanidou, Garatzioti, Lazaropoulou, Kourlaba, Papassotiriou and Kino17} A preclinical study of postnatal day 10 mice reported that hypoxic injury decreased hippocampal klotho levels at post-injury day 4. ^{Reference Herrmann, Kochanek, Vagni, Janesko-Feldman, Stezoski and Gorse18} Preterm and neonatal hypoxia are believed to be risk factors for ADHD. ^{Reference Franz, Bolat, Bolat, Matijasevich, Santos and Silveira19,Reference Bala, Bala, Pell and Fleming20} Furthermore, Lian et al found that klotho reduced oxidative stress and mitochondrial injury caused by sevoflurane in the hippocampus of newborn rats. ^{Reference Lian, Lu, Zhao, Zou, Lu and Zhou21} Xie et al reported that neonatal exposure to sevoflurane may lead to impulse control deficits in mice by disrupting excitatory neurons in the medial prefrontal cortex. ^{Reference Xie, Liu, Hu, Wang, Zhu and Jiang22} Prather et al linked klotho levels with chronic psychological stress, indicating that high-stress individuals exhibited lower klotho levels compared with low-stress. ^{Reference Prather, Epel, Arenander, Broestl, Garay and Wang23} Schnorr et al identified an inflammatory biotype of ADHD that presented with increase in both chemokine signalling and the NF-κB pathway, and was associated with chronic stress as measured by the perceived stress scale. ^{Reference Schnorr, Siegl, Luckhardt, Wenz, Friedrichsen and El Jomaa8} Other studies have shown a significant association between high levels of inattention and hyperactivity/impulsivity and a high level of perceived stress. ^{Reference Salla, Galera, Guichard, Tzourio and Michel24,Reference Oster, Ramklint, Meyer and Isaksson25} These findings indirectly suggest a potential association between klotho and ADHD, indicating that early childhood adversity and psychological stress, combined with insufficient klotho levels, may increase vulnerability to ADHD. However, the findings of our cross-sectional study are insufficient to confirm a causal association among klotho levels, ADHD and executive dysfunction. ^{Reference Siahanidou, Garatzioti, Lazaropoulou, Kourlaba, Papassotiriou and Kino17} Longitudinal studies would be required to validate our findings.

Our study has several limitations. First, we assessed executive function using only the association between WCST performance and klotho levels. Additional research would be required to evaluate associations between klotho levels and other aspects of cognitive function, such as inhibitory control and working memory. Second, although previous studies have shown a high correlation between ADHD symptoms and perceived psychological stress, we did not assess perceived psychological stress among those adolescents with ADHD who participated in our study, and we compared klotho levels between groups only after adjusting for clinical symptoms. Further research would be necessary to clarify an association among klotho levels, perceived psychological stress and ADHD symptoms. Third, given the Barkley model of ADHD as an executive function deficit disorder, ^{Reference Barkley26,Reference Castellanos and Tannock27} it is difficult to determine whether lower klotho levels are specific to ADHD per se or, rather, reflect executive dysfunction more broadly. The potential transdiagnostic role of reduced klotho levels in executive dysfunction warrants further investigation. Fourth, we focused on soluble α-klotho because it is the predominant circulating form, with validated serum ELISAs and the clearest prior links to cognition/executive function. By contrast, β- and γ-klotho act mainly as membrane co-receptors for endocrine fibroblast growth factors, have low/uncertain circulating levels and lack widely standardised serum assays and reference ranges. Further research would be warranted to clarify the associations of β- and γ-klotho with ADHD and executive function. Fifth, whether the observed reduction in klotho among adolescents with ADHD may generalise to adults with ADHD remains to be determined.

In conclusion, adolescents with ADHD exhibited decreased klotho levels compared with healthy adolescents following adjustments for demographic characteristics, clinical symptoms and ADHD medication use. In addition, lower klotho levels were associated with higher executive dysfunction, indicating that klotho may serve as a trait marker of ADHD. The exact neuromechanisms underlying the association among klotho, ADHD and executive function require further investigation.