MECP2 Duplication Syndrome: Evidence of Enhanced Oxidative Stress. A Comparison with Rett Syndrome

Rett syndrome (RTT) and MECP2 duplication syndrome (MDS) are neurodevelopmental disorders caused by alterations in the methyl-CpG binding protein 2 (MECP2) gene expression. A relationship between MECP2 loss-of-function mutations and oxidative stress has been previously documented in RTT patients and murine models. To date, no data on oxidative stress have been reported for the MECP2 gain-of-function mutations in patients with MDS. In the present work, the pro-oxidant status and oxidative fatty acid damage in MDS was investigated (subjects n = 6) and compared to RTT (subjects n = 24) and healthy condition (subjects n = 12). Patients with MECP2 gain-of-function mutations showed increased oxidative stress marker levels (plasma non-protein bound iron, intraerythrocyte non-protein bound iron, F2-isoprostanes, and F4-neuroprostanes), as compared to healthy controls (P ≤ 0.05). Such increases were similar to those observed in RTT patients except for higher plasma F2-isoprostanes levels (P < 0.0196). Moreover, plasma levels of F2-isoprostanes were significantly correlated (P = 0.0098) with the size of the amplified region. The present work shows unique data in patients affected by MDS. For the first time MECP2 gain-of-function mutations are indicated to be linked to an oxidative damage and related clinical symptoms overlapping with those of MECP2 loss-of-function mutations. A finely tuned balance of MECP2 expression appears to be critical to oxidative stress homeostasis, thus shedding light on the relevance of the redox balance in the central nervous system integrity.


Introduction
Methylated-CpG binding protein 2 (MeCP2) is a nuclear protein encoded by the X-linked MECP2 gene (OMIM Ã 300005). MeCP2 can be defined as a multifunctional protein, due to its involvement in chromatin architecture, regulation of RNA splicing, and its role both as transcriptional repressor or activator [1]. MECP2 appears to be universally expressed in all cell types with few exceptions, including microglia and rod photoreceptors [2]. Loss-of-function mutations in MECP2 is the main cause of Rett syndrome (RTT), which is a neurodevelopmental disease with severe cognitive impairment occurring at a ratio of approximately 1:10,000 girls [3,4].
However, RTT is not the only known pathological condition related to MECP2 mutations, as a wide series of conditions, collectively termed as MECP2-related disorders [5], has been reported. These disorders include asymptomatic female carriers, boys with MECP2 mutations typically causing a RTT phenotype in girls, and rare individuals with mutations in MECP2 showing other neurodevelopmental disorders [5]. Interestingly, intellectual disability (ID) is a common feature between RTT as well as MECP2-related disorders. Gain-of-function mutations in relation to MECP2 also lead to a severe neurodevelopmental disorder, named MECP2 duplication syndrome (MDS) or MECP2 triplication syndrome [6][7][8]. The phenotypes include major cognitive and motor deficits, stunted motor development, early onset hypotonia, epilepsy, and progressive spasticity, clinical features which are overlapping with some of those seen in RTT [6,7].
Although the prevalence of MDS is unknown, it has been estimated that MDS could be responsible for 1 to 2% of all X-linked ID cases [9]. MDS is well documented in males, with 150 affected individuals reported in the literature, while it rarely occurs in females [9], as female carriers of MECP2/Xq28 duplications are almost always asymptomatic due to extremely skewed X-inactivation [9]. This is in contrast to females with classical RTT harboring a loss-offunction mutations of the MECP2 [10,11].
Interestingly, a genotype-phenotype correlation in relation to the size of the Xq28 duplicated region has emerged. The minimal region of duplication that is sufficient to cause the core MDS phenotype involves the MECP2 and IRAK1 [12][13][14]. Evaluating previously data and data from mouse models, Ramocki and colleagues posit that MECP2 is the primary dosage-sensitive gene responsible for the neurological phenotypes in the Xq28 duplications [7].
Overall, MeCP2 appears to play a key role in the brain as a regulator of synaptic and neuronal plasticity as well as an etiological role in the development of RTT and MDS [15].
Oxidative stress (OS) is a nonspecific pathological condition that has frequently been associated with neurological disorders, including several diseases linked to cognitive impairment [16]. Compelling evidence between MECP2 loss-of-function and aberrant redox homeostasis has been shown by our and other research groups [17,18]. Specifically, a cause-effect relationship between oxidative brain damage and Mecp2 loss-of-function has been reported by our group in several murine models of the disease [19].
Since both a critical role for MECP2 in central nervous system integrity [20], and an involvement of OS in brain synaptic plasticity have been reported [21][22][23], MDS can represent a clinical model for testing the hypothesis that a finely tuned balance of MECP2 expression is critical to the control of redox homeostasis. We hypothesize that overexpression of MECP2 would lead to a status of enhanced OS.
The work was carried out in accordance to the rules expressed in the Declaration of Helsinki, and written informed consent was obtained by the parents of the enrolled subjects. This study was approved by the competent institutions: Ethics Committee of Western Zealand, Denmark (for the patient admitted at the Danish Epilepsy Centre), and Ethics Committee of the Tuscan Region, Azienda Ospedaliera Universitaria Senese, Siena, Italy (for the subjects recruited at the Child Neuropsychiatry Unit).

Blood sampling
Blood sampling was carried out after the overnight fast. Blood samples were collected in heparinized tubes, and centrifuged at 2,400Xg for 15 min at room temperature. The platelet poor plasma was saved, and the buffy coat was removed by aspiration. Erythrocytes were washed twice with physiological solution, suspended in Ringer solution, pH 7.4 as a 50% (vol/vol) suspension, and then used for the determination of intraerythrocyte non-protein bound iron (IE-NPBI). Plasma was used for free F 2 -isoprostanes (F 2 -IsoPs), F 4 -neuroprostanes (F 4 -Neu-roPs), F 2 -dihomo-isoprostanes (F 2 -dihomo-IsoPs), and plasma non-protein bound iron (p-NPBI). For all isoprostane determinations, butylated hydroxytoluene (BHT) (90 μM) was added to plasma as an antioxidant.

Routine blood tests
Serum concentrations of total cholesterol, high density lipoproteins (HDL)-cholesterol, triglycerides, immunoglobulin class G, class A and class M (IgG, IgA, IgM), were performed by a fully automatic analyzer (Cobas 6000 System; Roche Diagnostics). Total cholesterol, HDL-cholesterol, and triglycerides were assayed by using an enzymatic methods [29][30][31]. Serum fibrinogen concentration was determined by using Thromborel Reagent on BCS XP coagulation analyzer (Siemens Healthcare) [32]. Erythrocyte Sedimentation Rate (ESR) was assayed by using an automated system [33]. Blood cells counts were assayed on Sysmex XT-2100 system [34].

Data analysis
All variables were tested for normal distribution (D'Agostino-Pearson test). Differences between groups were evaluated by either one-way analysis of variance (ANOVA), or the Kruskal-Wallis test, as appropriate. Associations between variables were tested using linear regression analyses (for continuous normally distributed data) or the Spearman rank correlation (for continuous non normally distributed variables). A two-tailed P < 0.05 was considered to indicate statistical significance. The MedCalc ver. 12.0 statistical software package (MedCalc. Software, Mariakerke, Belgium) was used for data analysis.

Results
All the enrolled patients showed a MECP2 duplication/triplication whose position and size was defined according to the GRCh37/hg19 human genome annotation (Fig 1 and Table 1).
The typical features of the MDS, including developmental regression, epilepsy, magnetic resonance imaging abnormalities, hypotonia/spasticity were present in all the examined patients with the single exception of epilepsy, reported in five out of six subjects (Table 1).
All the MDS patients showed relevant differences in four out of five of the investigated OS markers, as compared to the healthy controls (Fig 2). In particular, significantly increased plasma levels of p-NPBI and IE-NPBI, forms of redox active iron, and both F 2 -IsoPs and F 4 -NeuroPs, non-enzymatic oxidized products from polyunsaturated fatty acids (i.e., arachidonic and docosahexaenoic acid, respectively), were detected. No significant differences were evidenced for F 2 -dihomo-IsoPs, biomarkers of adrenic acid oxidation. Both examined forms of iron (i.e., p-NPBI and IE-NPBI) are considered pro-oxidant agents [26], whereas the investigated isoprostanoids are indexes of lipid peroxidation (i.e., F 2 -IsoPs) [26], and brain gray (i.e., F 4 -NeuroPs) or white (i.e., F 2 -dihomo-IsoPs) matter oxidative damage [27,28]. A comparison of the OS status in MDS patients and RTT patients vs the healthy control population evidenced similarities, with the exceptions of significantly higher plasma levels of F 2 -IsoPs and lower plasma levels of F 2 -dihomo-IsoPs in MDS. Indeed, the levels of p-NPBI, IE-NPBI, and F 4 -NeuroPs in MDS were found to be comparable to those observed in RTT (Fig 2). Moreover, the levels of all the examined plasma and intraerythrocyte biomarkers were significantly higher in RTT as compared to the control group (Fig 2).
A significant positive correlation between the Xq28 region duplication/triplication size and plasma levels of F 2 -IsoPs (r = 0.9181; 95% C.I.: 0.4181 to 0.9912; P = 0.0098) was evidenced (Fig 3). Additionally, a correlation trend between the Xq28 region duplication size and the erythrocyte sedimentation rate (ESR) (r = 0.732; P = 0.1598) was observed. On the other hand, no significant correlations were detected between OS markers and age, thus reinforcing the hypothesis of a link between the MECP2 gain-of-function mutations and OS derangement. Significant differences were observed for total cholesterol, ESR, and ALC (post-hoc comparison: RTT > MDS % Control Subjects; MDS % RTT > Control Subjects; MDS % RTT > Control Subjects, respectively) ( Table 2). In contrast, no significant differences were reported for all the other parameters of the typical clinical-chemical pattern ( Table 2). No  significant differences were present for immunoglobulin class G and class M, high density lipoproteins-cholesterol, white blood cells, absolute counts for neutrophils, eosinophils, or basophils. (data not shown).

Discussion
Enhanced OS has been claimed to be involved in several pathological processes [35], including RTT, a rare genetic disease in which a deficiency of MeCP2 is demonstrated [36,37]. In particular, we have previously shown that OS in brains of mice with MECP2 loss-of-function mutations takes place and can be rescued, along with neurological signs, by Mecp2 brain specific gene reactivation [19]. However, to date, no hints on the opposite being true exist, i.e., no evidence is available for supporting the paradoxical concept that "too much" MeCP2 could lead to similar biochemical events as observed in conditions of "too little" MeCP2.
In the present study, we demonstrate, for the first time, the occurrence of a redox imbalance in patients with MECP2 overexpression. Moreover, our data, while adding new evidence with regard to the MECP2-OS link, suggest that a fine tuning of the MECP2 dosage may play a key role in regulating redox homeostasis in humans.
Whether mitochondria would be the main, or the only major, intracellular sources for the abnormal redox status in MDS and RTT is still to be ascertained [19,[38][39][40][41]. To this regard, an ultrastructural analysis of primary cultures of skin fibroblasts from RTT patients has shown no major morphological changes in the mitochondria [42]. At the same time, to the best of our knowledge, no studies on mitochondrial function in MDS have been published to date.
Oxidized products from polyunsaturated fatty acids appear to be promising molecules to be investigated in MDS, as they could mirror an ongoing oxidative brain damage. To this regard, in RTT mouse models, the presence of isoprostanes was evidenced in brain concomitant to elevated isoprostane plasma levels [19]. The reason for the strongly increased F 2 -IsoPs plasma levels in patients presenting MECP2 gain-of-function mutations remains to be elucidated. Nevertheless, the evidenced dose-effect relationship between Xq28 duplication size and F 2 -IsoPs production further supports the existence of a close link between MECP2 and redox homeostasis control. A close link between MECP2 gene expression and F 2 -IsoPs formation has been reported in RTT where the MECP2 mutations associated to more severe phenotypes exhibited higher F 2 -IsoPs plasma levels [17]. Interestingly, increased F 4 -NeuroP plasma levels in both MDS and RTT (Fig 2), suggest the involvement of the gray matter oxidative damage in both conditions. Currently, F 4 -NeuroPs are investigated as potential biomarker for neurological disease. In human RTT, F 4 -NeuroPs has been related to neurological symptoms severity, mutation type and clinical presentation [27], whereas their levels were significantly elevated in the brain of RTT symptomatic null mice [19]. The relevance of F 4 -NeuroPs in the neurological disease is also reported for the pathogenesis of Alzheimer's disease [43,44]. Gain-of-function and loss-of-function MECP2 mutations were found to generate distinct F 2 -dihomo-IsoPs patterns, oxidized fatty acid products considered as biomarkers of free radical damage to myelin [28,45]. Although a reduced volume of the brain white matter has been recently reported in patients affected by Xq28 duplication involving the MECP2 gene [46], our data might suggest a differential involvement of the brain white matter damage for RTT and MDS, on the base of the recovered F 2 -dihomo-IsoPs plasma levels.
This peculiar OS markers pattern further reinforces the specificity of the MECP2-OS link, which, far from being an epiphenomenon, appears to be intimately related to the specific MECP2-related disorders [17]. For some of those molecules known to be implicated in the general regulation of the redox homeostasis, an epigenetic modulation by MeCP2 has already been demonstrated [20]. To this regard, a possible role for contiguous genes to MECP2, i.e., IRAK1 (OMIM Ã 300283), cannot be ruled out, as a compensatory upregulation of IRAK1 has been reported in Mecp2 loss-of-function mutations associated with experimental RTT murine models [47,48].
Given that MDS and RTT both lead to severe neurodevelopmental disorders sharing several similar features (Table 3), our data, beyond stressing the critical role of the MECP2 function for the OS status, shed further light on the relevance of redox homeostasis in the central nervous system integrity.
Alike RTT, MDS seems to elicit an inflammatory response. This condition could be linked to either imbalanced gene control or abnormal redox status. Although recurrent infections are a quite common feature of MDS [49], present in the clinical history of five out of our six patients, no signs of ongoing infection were detectable at the time of the blood sampling. However, it is also possible that the elevated ESR values are not directly related to the underlying genetic abnormality. Nevertheless, the inflammatory similarities observed in MDS and RTT well comply with the potential role for MeCP2 in regulating cytokine-dependent T helper cell differentiation [20]. Thus, MECP2 seems to exert a finely tuned regulation of major biological processes.

Conclusion
Taken together, our data indicate that a MECP2 disequilibrium, either due to loss-of-function or gain-of-function mutation, leads to similar phenotypes in terms of OS status, thus adding new evidence on the relationship between OS and MECP2.