Melatonin inhibits attention-deficit/hyperactivity disorder caused by atopic dermatitis-induced psychological stress in an NC/Nga atopic-like mouse model

Atopic dermatitis (AD) is a chronic inflammatory skin disease with the hallmark characteristics of pruritus, psychological stress, and sleep disturbance, all possibly associated with an increased risk of attention-deficit/hyperactivity disorder (ADHD). However, the etiology of the possible association between AD and ADHD is still not well understood. 2,4-dinitrochlorobenzene or corticosterone was used to evaluate the atopic symptom and its psychologic stress in the atopic mice model. Melatonin, corticotropin-releasing hormone, corticotropin-releasing hormone receptor, urocortin, proopiomelanocortin, adrenocorticotropic hormone, corticosterone, cAMP, cAMP response element-binding protein, dopamine and noradrenaline were analyzed spectrophotometrically, and the expression of dopamine beta-hydroxylase and tyrosine hydroxylase were measured by Western blotting or immunohistochemistry. AD-related psychological stress caused an increase in the levels of dopamine beta-hydroxylase and tyrosine hydroxylase, degradation of melatonin, hyper-activity of the hypothalamic-pituitary-adrenal axis, and dysregulation of dopamine and noradrenaline levels (ADHD phenomena) in the locus coeruleus, prefrontal cortex, and striatum of the AD mouse brain. Notably, melatonin administration inhibited the development of ADHD phenomena and their-related response in the mouse model. This study demonstrated that AD-related psychological stress increased catecholamine dysfunction and accelerated the development of psychiatric comorbidities, such as ADHD.

In vitro and in vivo models. Immortalized human SH-SY5Y cell culture and the 2,4-dinitrochlorobenzene (DNCB) or CORT treatment animal models were established according to previously published methods 25,41,42 . The institutional animal care committee of the Korea Institute of Oriental Medicine (KIOM) approved the experimental protocols KIOM-16-015 and KIOM- . The experiments were performed according to the guidelines of the Animal Care and Use Committee at KIOM 25 . The animals were sacrificed between 11:00 a.m. and 14:30 p.m., 7 weeks after sensitization with DNCB or CORT ( Fig. 1 shows the study timeline).
Brain tissue preparation and immunofluorescence analysis. Brain tissue preparation was performed as previously described 25 . Briefly, the mice were decapitated; the skull was then removed and the brain was dissected around selected regions including the locus coeruleus (LC), prefrontal cortex (PC), and striatum (ST) using a brain matrix for kit-based analyses. All tissues were frozen in situ by immediate direct immersion in liquid nitrogen in order to prevent decomposition. Immunofluorescence analysis was then performed as described in a previous study 25 . Measurement of stress-related factors. Melatonin Statistical analysis. All statistical parameters were calculated using Graphpad Prism 5.0 software (Graphpad Software, San Diego, CA, USA). Values are expressed as means ± standard error of the mean (S.E.M.). Statistical comparisons between the different treatments were performed using a one-way ANOVA with Tukey's multiple comparison post hoc test. p values of <0.05 were considered to be statistically significant.

Effects of CORT and melatonin on DβH levels in SH-SY5Y cells.
We measured DβH levels using ELISA kits after exposing SH-SY5Y cells to CORT or melatonin. CORT exposure significantly increased DβH levels ( Fig. 2A

Effects of melatonin on DNCB-induced HPA axis activity biomarkers.
To determine whether melatonin affected stress hormone responses, we measured CRH and CRHR levels. Treatment with DNCB significantly increased CRH levels (Fig. 4A-C, and Supplementary Table 3), while treatment with 20 mg/kg melatonin reduced the DNCB-induced CRH level increase (Fig. 4A-C, and Supplementary Table 3). Also, treatment with DNCB significantly increased CRHR levels (Fig. 3D-F, and Supplementary Table 3), while treatment with 20 mg/ kg melatonin reduced the DNCB-induced CRHR increase ( Fig. 4D-F, and Supplementary Table 3).    Table 5).

Effects of melatonin on DNCB-induced dopamine metabolic enzyme levels.
To determine whether melatonin affects dopamine responses, we measured TH and DβH levels. Treatment with DNCB and DNCB + melatonin did not change the levels of TH compared with the control group ( Fig. 7A-C, and Supplementary Table 6). However, treatment with DNCB significantly increased the levels of DβH (Fig. 7E-G, and Supplementary Table 6), while treatment with 20 mg/kg melatonin reduced the DNCB-induced DβH increase ( Fig. 7D-F, and Supplementary Table 6).  Table 7). Furthermore, treatment with DNCB significantly increased the levels of norepinephrine ( Fig. 8D-F, and Supplementary Table 7), while treatment with 20 mg/kg melatonin reduced this DNCB-induced norepinephrine increase ( Fig. 8D-F, and Supplementary Table 7).  Table 8).

Discussion
The data presented herein demonstrate that the atopy-induced stress response significantly increased the presence of signaling molecules involved in ADHD. This response is due to factors such as dopamine and noradrenalin imbalances (via the up-regulation of DβH), which exacerbate HPA dysfunction and suppress the melatonin feedback system. Atopic stress led to: (1) induction of CRH-related and suppressing melatonin signaling in the (2) increased levels of DβH that did not alter TH levels; and (3) decreased dopamine levels and increased noradrenalin levels. Melatonin reversed these effects on dopamine and noradrenalin via stimulation of the normal HPA axis. To the best of our knowledge, in the first study to demonstrate that AD-caused psychologic stress increases catecholamine dysfunction and accelerated the development of psychiatric comorbidities, such as ADHD via dysregulation of the HPA/melatonin signaling pathways. ADHD, characterized by inattention, hyperactivity/impulsivity, or both, is one of the most common psychiatric disorders of childhood 43 . Approximately one third of medication-free children with ADHD experience chronic sleep-onset insomnia [43][44][45] . Although the safety and efficacy of melatonin treatment for sleep-onset insomnia in children without ADHD have been well-documented, melatonin efficacy has not been studied in medication-free children with ADHD and sleep-onset insomnia; this patient group is of special interest for several reasons 46,47 . First, medication-free children with both disorders exhibit a delayed evening increase in endogenous melatonin levels, and this phase delay predicts strengthening of the sleep phase, which normalizes the effect of exogenous melatonin in children without ADHD 46 . Second, because treating sleep-related disorders other than insomnia improves daytime function in children with ADHD, treating insomnia may have important consequences for ADHD treatment strategies [48][49][50] . The night time circadian rise in melatonin levels correlates with a nighttime circadian drop in glucocorticoids 51,52 . Any chronic, late-night stressor (e.g., shift-work 53 ) can result in excessively high nighttime cortisol levels, which may impair the normal morning circadian increase in corticosteroid levels 54 . It is known that circadian rhythms are highly related to the HPA axis, which is a key neuroendocrine mediator of physiological responses to psychological stressors [55][56][57] . In a previous study, we found Values are means ± standard error of the mean. *P < 0.05, **P < 0.01, **P < 0.01 and ***P < 0.001 compared with the control group; # P < 0.05, ## P < 0.01, and ### P < 0.001 compared with the 2% DNCB-alone group. DNCB, dinitrochlorobenzene; CRH, corticotropin releasing hormone; HPA, hypothalamic-pituitary-adrenal; UCN; POMC, pro-opiomelanocortin; ACTH, adrenocorticotropic hormone; CORT, corticosterone; ELISA, enzymelinked immunosorbent assay.
Scientific REPORts | (2018) 8:14981 | DOI:10.1038/s41598-018-33317-x that hyper-activation of HPA axis induced by atopic chronic psychological stress was associated with neurotoxicity and cognitive impairment, and it also blunted neuroendocrine responses to stress 25 . In this and our previous paper, we focused on the idea that neuroendocrine contribution to the responses in skin to stress is promoted, in part, by local synthesis of all elements of the HPA axis. Skin has the ability to synthesize glucocorticoids from cholesterol or steroid intermediates of systemic origin 58,59 . By interacting with glucocorticoid receptors, they regulate skin immune functions as well as functions and phenotype of the epidermal, dermal and adnexal compartments 58,59 . Most of the biochemical (enzyme and transporter activities) and regulatory principles of cutaneous glucocorticosteroidogenesis (neuropeptides mediated activation of cAMP and protein kinase A dependent pathways) are similar to those operating in classical steroidogenic organs [58][59][60][61] .
According to a recent report, stimulation of cutaneous corticosteroidogenesis can occur via this skin homologue of the HPA axis which is dependent on the functional activating CRHR and processing POMC 60,61 . Specifically, the stimulating cortisol synthesis by IL1 is intriguing because IL1 serves as a local signal of skin inflammatory injury 60,61 . It is possible that IL1 may stimulate corticosteroidogenesis indirectly, through up-regulating CRH or POMC peptides or by itself, which has been reported to occur in the adrenals 60,61 . CRH, in addition to indirect stimulation may directly stimulate local corticosteroidogenesis because it increases cAMP 62,63 . A similar direct action of CRH on adrenocortical region has been reported 62,63 . An additional regulating-mechanism involves the activation or inactivation of glucocorticosteroids by locally expressed cortisone reductase, such as 11β-hydroxysteroid dehydrogenase type1 and 2 [62][63][64] .
In this and our previous paper, we confirmed that melatonin control anti-inflammation by regulating the HPA axis via the skin. Melatonin plays an important role in the regulating circadian timing and neuro-immunologic function 54,65 . Recent studies have linked decreases in melatonin output to insomnia in aged patients and sleep disturbances in patients with Alzheimer's disease 25,[66][67][68][69] . In AD, a severity score ≥48.7 predicts elevated levels of immunoglobulin, sensitive allergen, and poor sleep efficiency, possibly because of reduced nocturnal melatonin secretion, pruritus, and associated scratching 25,69 . One study reported disrupted melatonin secretion in eczema, possibly due to partial action of the sympathetic nervous reduction that regulates secretion of melatonin 17,25 . Recent studies have shown that melatonin is synthesized in numerous extrapineal sites 54,65,70 . Additionally, melatonin is regulated by rapid metabolism in the liver and peripheral organs including the skin 54 . Several researchers have proposed that melatonin and its metabolites affect skin functions and structures through actions mediated by intracutaneously expressed cell-surface and putative nuclear receptors 54,70 . Melatonin exerts both receptor-dependent and receptor-independent protective effects against oxidative stress and can attenuate environmental skin stressor-induced damage 71 . The effects of the common environmental skin stressors are modulated by melatonin via a complex intra-cutaneous melatonergic anti-oxidative system, with ultraviolet radiation-enhanced melatonin metabolism generating bioactive melatonin metabolites such as acetyl-N-formyl -5-methoxykynurenamine 70,71 . These properties suggest that melatonin is an important endogenous effector of intracutaneous stress responses. In fact, there is a shortage of information concerning the regulation of HPA axis-melatonin-circadian rhythms on the glucocorticosteroidogenesis signaling system in skin [61][62][63] . However, previously, we found that NC/Nga mice exposed to atopic stress exhibited substantial reductions in hypothalamic melatonin membrane receptor expression and in hypothalamic and intracutaneous melatonin expression 25 . Additionally, historically, oriental medicine has suggested a similar connection with the thought that biological rhythm is an inherent connotation of "harmony between human and nature" 72 . Our previous study did not show whether intracutaneous melatonin affects the brain, but many other studies have suggested that it has an important effect. Further research is certainly needed, but our observation of decreased melatonin levels in the LC, PC, and ST in DNCB-exposed NC/Nga mice supports this hypothesis.
In the present study, we tried to investigate effects of stress in the LC. Some anti-depressants as well as the ADHD medication atomoxetine, are believed to act on LC neurons 73,74 . The LC is responsible for mediating several sympathetic effects of stress; it is activated by stress and responds by increasing norepinephrine secretion, which in turn alters cognitive function (through the PC), increases motivation (through nucleus accumbens), and activates the HPA axis 75 . After HPA axis stimulation, norepinephrine stimulates the secretion of CRH from the hypothalamus, which induces ACTH release from the anterior pituitary and subsequent cortisol synthesis in the adrenal glands 55,76,77 . Norepinephrine released from the LC will inhibit its own production, and CRH will inhibit its own production while causing the LC to increase norepinephrine production 73 . Thus, to determine whether mechanisms of the dopamine and noradrenaline imbalance caused by atopic stress were regulated by melatonin, we measured ADHD signaling patterns in an NC/Nga atopic mouse model. We showed that DNCB significantly increased levels of HPA axis-related response substances such as CRH, CRHR, UCN, POMC, ACTH, and CORT, while treatment with melatonin significantly reduced these levels in DNCB-treated mice. Further, corticosteroid-mediated psychological stress responses utilize various signal transduction systems [78][79][80] . Studies have shown that in hippocampal neurons or slices, ERK1/2 respond to stressful stimuli through the transcription factor CREB (cyclic response element binding protein), which activates c-Fos via CRE sites in promoter regions 79 . Additionally, stress increases phosphorylating CREB 80 . Moreover, the PI3K-cAMP-CREB pathway activity is elevated in the noradrenergic neurons of the LC, which is the main source of noradrenaline in the brain 37 . PI3Kc gene KO mice show increased CREB activation via elevation of cAMP levels in the LC and alters the dopamine/ noradrenaline balance in the PC and ST 7,37 . These changes facilitate the development of core ADHD-related phenotypes, including hyperactivity and attention deficit, as well as secondary features such as memory and social impairments 7,37 . Overexpression of CREB in the LC of normal animals produces similar behavioral changes, and down-regulation of CREB activity in the LC of mutant mice reverses the phenotype 7,37 . In the present study, we observed significantly increased cAMP and pCREB expression in the LC; subchronic melatonin reduced cAMP and pCREB expression in DNCB-treated mice.
Further, we examined DβH and TH levels. DβH is anenzyme that converts DA into NE and is coreleased with catecholamines 81 . TH is involved in the conversion of phenylalanine to dopamine 81 . As the rate-limiting enzyme in the synthesis of catecholamines, TH plays a key role in the physiology of adrenergic neurons; TH is regularly used as a marker for dopaminergic neurons [81][82][83] . DβH is a genetic marker and may reflect individual susceptibility Figure 8. Effects of melatonin on DNCB-induced expression of the dopamine and noradrenaline contents in the locus coeruleus, prefrontal cortex, and striatum of the brain. The expression of dopamine (A-C) and noradrenaline (D-F) were measured using ELISA kits. Values are means ± standard error of the mean. *P < 0.05 and ***P < 0.001 compared with the control group. DNCB, dinitrochlorobenzene; ELISA, enzymelinked immunosorbent assay.
Scientific REPORts | (2018) 8:14981 | DOI:10.1038/s41598-018-33317-x to developing psychosis in the context of exposure to traumatic events 84 . Generally, adrenal catecholamines are known to mediate many of the physiological consequences of the "fight or flight" response to stress 84,85 . However, the mechanisms by which the long-term responses to repeated stress exposure are mediated are not well understood. According to McMahon et al., both TH and DβH levels are elevated by single and repeated exposure to immobilization stress 84,85 . Surprisingly, we observed no or slight change in TH expression in the LC. However, DβH levels were significantly upregulated in the LC, PC, and ST of DNCB-treated mice. Further, subchronic melatonin reduced DβH expression. We also observed significantly decreased dopamine and increased noradrenaline levels in the LC, PC, and ST of DNCB-treated mice; subchronic melatonin treatment reduced these levels. This pattern demonstrated similar results in CORT injection model. In this study, we suggested that the imbalance resulting from increased release of dopamine and noradrenaline for potential modulation of physiological or immunological responses is the consequence of the upregulated DβH expression and the unchanged TH expression in DNCB exposed to atopic stress. However, future research will be needed to analyze the specific mechanisms system. Thus, AD-related neuropsychological stress caused the relationship of normal glucocorticoid/ melatonin disruption and accelerated dopamine dysregulation.
This study has some limitations which have to be pointed out. We did not assess whether atopic stress induced ADHD behavioral pattern of chronic stress, because the scratching behavior and memory impairments interfered with the rodent behavioral tasks. Several pilot experiments were conducted, but they were unsuccessful. Additionally, melatonin is correlated with the circadian rhythm phase shift of the circadian clock; hence, a change in sleep-wake time is expected. However, the experiments were unsuccessful as itching interfered with Figure 9. Effects of melatonin on CORT-induced expression of the dopamine and noradrenaline contents in the locus coeruleus, prefrontal cortex, and striatum of the brain. The expression of dopamine (A-C) and noradrenaline (D-F) were measured by ELISA kits. Values are means ± standard error of the mean. *P < 0.05 and ***P < 0.001 compared with the control group. CORT, corticosterone; ELISA, enzyme-linked immunosorbent assay.
Scientific REPORts | (2018) 8:14981 | DOI:10.1038/s41598-018-33317-x the sleep-wake time and the sleep-wake time criterion was unclear. Thus, future studies will need to analyze the wake-sleep states, specific response of stress, and their potential behavioral outcomes.

Conclusions
In conclusion, we found that atopic stress accelerated HPA-axis dysfunction, increased norepinephrine, and decreased dopamine (Fig. 10). Atopic stress induced HPA axis-related response dysfunction and ERK-CREB signaling pathway, which reduced melatonin levels in the LC, PC, and ST. Moreover, atopy-induced stress accelerated DβH level increases and dopamine consumption. Further, we discovered that melatonin inhibited reduced dopamine consumption by the inhibition of DβH via regulating HPA in AD models. Therefore, our findings suggest that the CORT-melatonin disequilibrium might contribute to dysfunction of dopamine by causing or enhancing neurodegeneration, which could lead to disorders such as ADHD. Figure 10. Schematic of the mechanism proposed for the effects of melatonin on the hypothalamic-pituitaryadrenal (HPA) axis and attention deficit hyperactivity disorder (ADHD) pathogenesis.