Mechanisms underlying anxiety in Rett Syndrome: Translational insights from preclinical findings

Article type: Review Article

Keywords: Rett syndrome, Anxiety, Epigenetics, Translational research

Authors: Bethan Impey, Tracey A. Newman, David S Baldwin

License: © 2022 The Authors CC BY 4.0 This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Article links: DOI: 10.1016/j.nsa.2022.100109  |  PubMed: 40656225  |  PMC: PMC12244188

Relevance: Moderate: mentioned 3+ times in text

Full text: PDF (3.0 MB)

Nomenclature:

MECP2
is the gene in humans

Mecp2
is the murine gene

MeCP2
is the protein in mouse and humans

Introduction

Rett syndrome is the second most common cause of severe learning disability in females (occurring in approximately 1:10,000 females). Babies born with Rett syndrome initially develop apparently normally, however, at the age of 6–18 months, development falters and subsequently regresses (Box 1). Rett syndrome has multiple comorbid conditions affecting overlapping physiological systems. Despite this heterogeneity, anxiety is a common feature amongst the comorbidities of Rett syndrome (ref. Gold et al., 2017). The mechanisms underlying anxiety are the focus of this review as further study may lead to new insights into the pathophysiology of anxiety.

The challenge of identifying anxiety in individuals with severe intellectual disability has been recognised. Two validated questionnaires have been developed for identifying presumed anxiety-related behaviours in Rett Syndrome: the Rett Syndrome Behaviour Questionnaire (RSBQ) and the Anxiety, Depression and Mood Scale (ADAMS) (ref. Esbensen et al., 2003; ref. Mount et al., 2002). These questionnaires rely on carer observation. The RBSQ was validated by comparing results in individuals with Rett syndrome and those with other severe intellectual disability; control for physical impairment was also included (ref. Mount et al., 2002). Anxiety symptoms in Rett syndrome appear similar to generalised anxiety and social anxiety as they include worsening of hyperventilation, inconsolable crying, trembling in the absence of fearful situations and withdrawal from social contact (ref. Buchanan et al., 2019).

While the diagnosis of Rett syndrome is made using clinical features, it is usually caused by mutations on chromosome Xq28 in the gene encoding methyl-CpG-binding protein 2 (MECP2). MeCP2 is found in most tissues, but is most abundant in the central nervous system (CNS) (ref. Shahbazian et al., 2002). A key role of MeCP2 is to bind to methylated DNA and it can both activate and repress gene expression, therefore acting as an epigenetic regulator. The effects of MeCP2 are cell-type and region specific and the level of MeCP2 within a cell is critical to the function of the cell. MeCP2 expression increases in the early months of life and the timing of expression in particular brain regions correlates with the maturation of the CNS (a time when sensory-driven neural activity shapes the CNS circuitry) (ref. Zimmermann et al., 2015). In Rett syndrome there is mosaicism (only the maternal or paternal X chromosome will be expressed in any one cell in the body) of MECP2 gene expression and this leads to variability in phenotype (ref. Gold et al., 2017). Further variations in expression of the condition are likely influenced by the numerous mutations in MECP2; to date 518 likely pathogenic mutations have been identified.

A review of autonomic dysregulation in individuals with Rett syndrome has highlighted the complexity of treatment (ref. Singh and Santosh, 2018). Treatments for anxiety disorders in neurotypical individuals are often suboptimal. This is further complicated in neurodivergent individuals in whom the mechanisms and treatments for anxiety are poorly understood. Symptoms suggestive of anxiety are found more commonly in Rett syndrome compared to individuals with other severe intellectual disability (ref. Mount et al., 2002) suggesting the mechanisms of anxiety may be entwined with the disorder. Greater understanding of anxiety mechanisms in Rett syndrome could facilitate development of more specific treatments and improve our mechanistic understanding of anxiety disorders in general.

A number of preclinical models of Rett syndrome have been developed in mice; many of these show similar features to patients with Rett syndrome, including anxiety-like behaviour, irregular breathing, motor impairment and seizures (ref. Katz et al., 2012). This review focuses on the preclinical evidence base. It highlights the heterogeneity in the mutations being studied, the study designs and the need for standardisation of behavioural assays. Despite these limitations, these models provide a system for the study of the cellular and molecular mechanisms of MeCP2 dysfunction and associated alterations in anxiety-like behaviour. The discussion draws together findings from pre-clinical studies with clinical observations and suggests future directions for Rett syndrome translational research, and postulates that there may be broader implications for other disorders of MeCP2 dysfunction where anxiety is a feature.

Methods

A systematic review was carried out using the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines (ref. Moher et al., 2009). Records were searched through PubMed, MEDLINE and Embase. The following search terms were used [Rett syndrome] AND [anxiety]. Searching was completed on February 3, 2022. Fig. 1 illustrates the selection process. All included studies made statements that animal experiments complied with ethical guidelines, except for ref. McGill et al., 2006, ref. Orefice et al., 2016 and ref. Wither et al., 2013 (ref. McGill et al., 2006; ref. Orefice et al., 2016; ref. Wither et al., 2013), where statements were not explicitly made. All included studies were published in peer-reviewed journals.

Results

Overview

Anxiety is postulated to develop when there are disruptions in the interconnected macro- and micro-circuits that process stimuli detected from the external world. As such, the first stage in the anxiety macrocircuit involves the sensory systems. Sensory signals are interpreted as potentially threatening through the interconnected activity (via neuronal projections between regions) of the amygdala, bed nucleus of the stria terminalis, ventral hippocampus and prefrontal cortex. Evaluation, where the interpreted signals are regulated, then occurs. This involves the medial prefrontal cortex, ventral tegmental area, nucleus accumbens, ventral pallidum, and hypothalamus. Following coordinated processing of a threat stimulus, a response is initiated (ref. Calhoon and Tye, 2015). Microcircuits function within each region of the anxiety macrocircuit and involve short-range connections between neuronal subtypes and associated interactions with glial cells, the blood vessel endothelium and surrounding specialised extracellular matrix (perineuronal net). The account of the preclinical studies has been structured to highlight the effect of MeCP2 loss in anxiety macrocircuit regions and within cellular subtypes on anxiety-like behaviour and/or physiological response (illustrated within Fig. 2) and subsequently summarises potential underlying molecular mechanisms. At present, there is limited literature examining macro- and micro-circuitry connections, which is briefly described in section 3.6. However, an effect of MeCP2 loss in CNS regions and cell subtypes on anxiety behaviour may suggest an impact on the anxiety circuitry, albeit with significant limitation on interpretation.

The tests of anxiety-like behaviours are described within Box 2. Generation of the mouse models relies on mutation of the Mecp2 alleles. In keeping with the variability of mutations identified in people with Rett syndrome, the mouse models involve varying degrees of severity of mutation – null alleles, large mutations and point mutations. For clarity, male Rett syndrome mice are referred to as ‘Mecp2 null’ and female Rett syndrome mice are referred to as ‘Mecp2^+/−‘. Most earlier Rett syndrome studies used male mice, although this is changing with the recognition of the importance of studying both sexes and the obvious importance of using female mice given that in humans, females are affected. The reason that male mice had been predominantly used is because Mecp2 mutation causes a less severe phenotype in mice than humans and the male mice show a more pronounced phenotype than females. In addition, females have a more variable phenotype than males, due to mosaicism (ref. Ribeiro and MacDonald, 2020). Table 1 documents studies that correlate molecular changes with anxiety behaviour; included studies are arranged alphabetically on the basis of the first author(s). This table details experimental conditions and enables changes in anxiety behaviour to be identified within and between studies, as well as highlighting whether male and/or female mice were used. Outcomes of the studies within the table are summarised in the text. The sections documenting MeCP2 loss in macrocircuit regions and cell subtypes have been structured to summarise where this was associated with a change/no change/mixed effects on anxiety behaviour. The description ‘mixed effects’ was determined by there being inconsistent changes between the different behavioural tests within a study. The impacts of MeCP2 loss within anxiety macrocircuit regions and cell subtypes are summarised within Fig. 2. Details of the mutation used in each study have been documented within Table 2.

Circuitry regions and cellular subtypes where loss of MeCP2 altered anxiety behaviour

Macrocircuitry regions

Loss of MeCP2 in the peripheral sensory ganglia has been associated with increased anxiety behaviour. Mice with this mutation had enhanced sensitivity to tactile stimuli when compared to controls. It is important to note that mice would have a greater anxiety response to touch than humans. However, the authors comment that other studies have shown that humans who have had sensory deprivation in early life have increased anxiety (ref. Orefice et al., 2016). Deletion of Mecp2 in the amygdala was associated with an increase in anxiety behaviour in male mice (ref. Adachi et al., 2009; ref. Gemelli et al., 2006). There is evidence for deletion of Mecp2 in the nucleus accumbens and prefrontal cortex being associated with increased anxiety (Mecp2 was deleted from multiple anatomic regions, therefore the relevance of individual regions is not clear) (ref. Gemelli et al., 2006).

Cellular subtypes

Selective reintroduction of Mecp2 in astrocytes of Mecp2 null mice was associated with anxiety behaviour being more typical of behaviours of control mice (although interpretation is complex as the null mice showed mixed anxiety behavioural changes compared to control) (ref. Lioy et al., 2011).

Circuitry regions and cellular subtypes where loss of MeCP2 had mixed effects on anxiety behaviour

Macrocircuitry regions

Lower MeCP2 expression in the hippocampus in Mecp2^+/− mice was associated with lesser anxiety behaviour. However, expression of MeCP2 was not measured in anxiety-associated brain regions other than the hippocampus, making interpretation of these results difficult (ref. Wither et al., 2013). In contrast when Mecp2 was removed from the forebrain (inclusive of hippocampus), there was an increase in anxiety behaviour (ref. Gemelli et al., 2006). Sim-1 is a gene regulatory element required for the development of neurons in the paraventricular, supraoptic, and posterior hypothalamic nuclei, and in the nucleus of the lateral olfactory tract of the amygdala. Deletion of Mecp2 in Sim1-expressing neurons led to a decrease in MeCP2 expression in these regions. This did not result in a reliable difference in ‘anxiety’ behaviour compared to controls (ref. Fyffe et al., 2008).

Cellular subtypes

Altered MeCP2 function in cholinergic (ref. Zhou et al., 2017; ref. Zhang et al., 2016), glutamatergic (ref. Meng et al., 2016; ref. Goffin et al., 2014) and GABAergic (ref. Chao et al., 2010; ref. Ure et al., 2016) neurons resulted in a variable pattern of changes in anxiety behaviour. Mice with specific removal of Mecp2 from tyrosine hydroxylase-expressing dopaminergic (DA)/noradrenergic (NA) neurons showed no difference in anxiety behaviour compared to controls (ref. Samaco et al., 2009). Rescue of Mecp2 in catecholaminergic neurons normalised anxiety behaviour in male and female Mecp2-deficient mice (ref. Lang et al., 2012). However, there was no significant effect on anxiety behaviour when Mecp2 was rescued in the dorsal striatum of Mecp2 deficient male and female mice (ref. Su et al., 2015). Overall, this implies that if MeCP2 function in DA/NA neurons does affect anxiety behaviour, that this occurs via a specific subtype of catecholaminergic neuron and/or particular brain region.

In adult mice, microglial cells develop within the bone marrow and migrate to the brain (in addition to slow proliferation of microglia within the CNS). Mecp2^+/− mice were irradiated to kill resident microglia and then given a bone marrow transplant from wildtype mice. Behavioural testing indicated that transplantation (intended to give rise to normal microglia) resulted in anxiety behaviour reducing to wildtype levels (ref. Derecki et al., 2012). A later study used Mecp2 null mice and replicated these methods. Anxiety behaviour was not explored, but other symptoms were not reversed through introduction of wildtype microglia, suggesting that anxiety behaviour may also not have been altered (ref. Wang et al., 2015).

Circuitry regions and cellular subtypes where loss of MeCP2 did not alter anxiety behaviour

Cellular subtypes

Specific loss of MeCP2 in GABAergic interneurons – parvalbumin, somatostatin, vasoactive intestinal peptide (separate or combined) – was not associated with a change in anxiety behaviour (ref. Ito-Ishida et al., 2015; ref. He et al., 2014; ref. Mossner et al., 2020).

Mice in which Mecp2 was deleted in PET1-expressing serotonergic neurons showed no difference in anxiety behaviour compared to controls. (ref. Samaco et al., 2009). Mecp2 null mice were found to have lower 5-HT7R (serotonin receptor) density in the cortex and hippocampus compared to wildtype. Intraperitoneal injection of LP-211 (a 5-HT7R selective agonist) in Mecp2 null mice was associated with increased cortical 5-HT7R density, but had no clear effect on anxiety behaviour (ref. De Filippis et al., 2014). Serotonin is a key neurotransmitter in anxiety disorders, so this apparent lack of effect on anxiety behaviour is of note. It has been shown that MeCP2 is involved in enhancing transcription of 5-HT1AR (ref. Philippe et al., 2018) and so it is possible that either MeCP2 does not modulate 5-HT7 production or that a greater concentration of 5-HT has less impact if there is lower receptor expression. In addition, studies have shown that administration of fluoxetine (a selective serotonin reuptake inhibitor, SSRI) resulted in raised MeCP2 levels in GABAergic neurons (ref. Cassel et al., 2006) and increased MeCP2 expression in the prefrontal cortex but not the CA3 region of the hippocampus (ref. Villani et al., 2021), highlighting that MeCP2 function within serotonergic neurons may be different to the effects of serotonin itself.

Cannabidiolic acid administration did not reduce anxiety in Mecp2 null mice (ref. Vigli et al., 2021). This contrasts with an earlier study that reported cannabidivardin improved symptoms (anxiety behaviour not assessed) in Mecp2 null mice (ref. Vigli et al., 2018), so it is possible that cannabinoid receptors may be relevant to anxiety in Rett syndrome.

Treatment studies that may provide further insights into these mixed effects on anxiety behaviour

Electrophysiological analysis of isolated cortical cell membranes of Mecp2 null mice showed reduced GABA and glutamate currents relative to those from Mecp2 null mice that had been administered mirtazapine prior to analysis. However, administration of mirtazapine and desipramine to Mecp2 null mice did not result in a clear change in anxiety behaviour (ref. Bittolo et al., 2016). Mecp2^+/− mice initially showed a reduction in anxiety compared to wildtype and anxiety apparently increased with mirtazapine. However, this was subsequently attributed to increased whisker sensitivity altering the behaviour of the mice in the closed arms of the assay. Albeit whisker (sensory) sensitivity may be relevant to anxiety, in line with earlier findings (ref. Orefice et al., 2016)). There was a reduction in expression of parvalbumin positive neurons (fast-firing GABAergic interneurons) in the barrel cortex and basolateral amygdala in Mecp2^+/− mice; expression was increased in the barrel cortex but not the amygdala following mirtazapine treatment, suggesting a possible mechanism for the increased sensitivity (ref. Flores Gutiérrez et al., 2020).

The mechanism of action of deep-brain magnetic stimulation (DMS) is largely unknown, though it may modulate the balance of excitatory and inhibitory circuits. Treatment with DMS for 5 months showed some reduction in anxiety behaviour of Mecp2 null mice to wildtype levels (ref. Zhang et al., 2014). Deep brain stimulation (DBS) also appeared to reduce anxiety behaviour in Mecp2^+/− mice. However, the same effect was found with sham DBS, indicating handling and exposure as the reason for decreased anxiety (ref. Hao et al., 2015).

Molecular changes that potentially associate with the anxiety behavioural changes/physiologic response

MeCP2 effects on circuitry connections and molecular changes that are cell type and region specific

Loss of MeCP2 in the peripheral sensory ganglia may cause loss of the β3 unit of the GABA_A receptor in these regions, resulting in increased peripheral nerve conduction, which may act as a mechanism for the increased anxiety behaviour (ref. Orefice et al., 2016). Excitatory projections from the cortex and thalamus to the lateral nucleus of the amygdala (LA) are involved in fear acquisition. MeCP2 specifically affected the functioning of the cortico-LA pathway (not the thalamo-LA pathway) through effects on stabilisation of AMPA receptors and synaptic elimination and maturation, highlighting a potential mechanism of anxiety in Rett syndrome, since there is an overlap between fear and anxiety circuitry (ref. Gambino et al., 2010). (ref. Gambino et al., 2010).

Loss of MeCP2 in cholinergic neurons in the basal forebrain led to decreased expression of choline acetyltransferase (ChAT). This was associated with reduced excitability of the cholinergic neurons, leading to downstream increased excitability of hippocampal pyramidal neurons. According to anxiety behavioural changes, it appears that these effects of MeCP2 were pathway-specific (ref. Zhang et al., 2016). By turn, choline supplementation significantly increased ChAT activity in Mecp2 null and wildtype mice, and reduced Mecp2 null anxiety levels to those of wildtype mice (ref. Ricceri et al., 2011). Choline supplementation in mice with specific loss of MeCP2 in excitatory neurons of the forebrain was associated with decreased anxiety and increased neurite length, number of dendritic branches and dendritic spine density, in the basal forebrain suggesting further molecular changes resulting from MeCP2 function in cholinergic neurons (ref. Chin et al., 2019).

Tyrosine hydroxylase (Th) and tryptophan hydroxylase 2 (Tph2) are rate-limiting enzymes in the production of DA/NA and 5-HT, respectively. MeCP2 may be necessary for their production as MeCP2 was found to bind to the Th and Tph2 promoters and there were reduced RNA levels for Th and Tph2 in Mecp2 null mice relative to wildtypes. Again this appears to be region-specific, since the expression of Th was lower in the locus coeruleus, midbrain containing substantia nigra and ventral tegmental area, but not the medulla, in Mecp2 null mice relative to wildtypes and Tph2 expression was lower in the hindbrain containing raphe nucleus B1-3, but not in the hindbrain region containing raphe nucleus B4-9, in Mecp2 null mice relative to wildtypes. These changes in enzyme expression were associated with reduced levels of monoamines (ref. Samaco et al., 2009).

Influence on cell signalling pathways

IGF-1 and BDNF both act via the phosphoinositide 3-kinase (PI3K)/Protein kinase B (Akt) pathway and the extracellular signal-regulated kinase (ERK) pathways. Lower serum IGF-1 (ref. Mellios et al., 2014) and lower hippocampal BDNF expression (ref. Kondo et al., 2016) were found in Mecp2^+/− mice compared to wildtype mice. Increasing the serum IGF-1 levels (via clenbuterol treatment, a beta adrenergic agonist (ref. Mellios et al., 2014)) and environmental enrichment (ref. Kondo et al., 2016) resulted in anxiety behaviour returning to wildtype levels. Intraperitoneal rhIGF-1 treatment resulted in an increase of anxiety behaviour of MeCP2 null mice to wildtype levels (ref. Castro et al., 2014). Overall, these results indicate that modulation of IGF-1 and BDNF expression by MeCP2 may contribute to anxiety behaviour.

Possible changes to metabolic activity of CNS cells

Alterations in access to food in Mecp2 null mice suggested an improvement in anxiety. Mice fed a ketogenic diet or standard diet with caloric restriction had reduced anxiety compared to those given standard diet. However, the mice on restricted diets were more active as they persistently searched for food, wildtype mice on a similar restricted diet also showed a decrease in anxiety behaviour. This shows little evidence for the effects of cellular metabolic activity on anxiety behaviour or evidence that MeCP2 might affect metabolic activity (ref. Mantis et al., 2009).

Altered HPA axis function

mRNA for FKBP5 (a negative feedback regulator of the glucocorticoid receptor) and serum and glucocorticoid-inducible kinase (Sgk) levels were upregulated in whole-brain samples from Mecp2 null mice compared to wildtype, however there was no difference in levels in the basolateral amygdala, suggesting MeCP2’s modulation of their expression is region-specific (ref. Adachi et al., 2009; ref. Nuber et al., 2005). Mecp2 null mice in environmentally enriched housing showed increased mRNA expression of Sgk1in the hypothalamus and cortex that was associated with anxiety returning to wildtype levels, identifying a possible role for Sgk1 regulation in anxiety. (ref. Kerr et al., 2010).

MeCP2 regulation of corticotropin releasing hormone (CRH) expression differs between brain regions and between male and female mice: there was elevated CRH in the paraventricular nucleus of the hypothalamus, central amygdala and the bed nucleus of the stria terminalis (ref. McGill et al., 2006) of Mecp2 null compared to wildtype mice, whilst Mecp2^+/− mice had lower expression of CRH mRNA in the hypothalamus than wildtype mice (ref. Kondo et al., 2016). CRH mRNA expression in the basolateral amygdala was not significantly different in mice without MeCP2 in the amygdala, compared to controls (ref. Adachi et al., 2009). The HPA axis was normal in Mecp2 null mice and so increased CRH expression (in certain regions) was identified as a key source of HPA axis dysfunction in Mecp2 null mice (ref. McGill et al., 2006). However, CRH1 antagonist administration significantly reduced anxiety behaviour of wildtype, but not Mecp2 null mice, whilst under restraint stress. Wildtype and Mecp2 null mice showed an irregular breathing pattern when restrained using a small restraint chamber, indicating a possible anxiety response. CRHR1 antagonist administration resulted in a closer to normal breathing pattern in the wildtype mice in the small chamber but notably affected breathing in Mecp2 null mice in both chambers (ref. Ren et al., 2012).

Physiological responses to stress, assessed through serum corticosterone levels were equivalent in Mecp2 null and wildtype mice at baseline and following 5–15 ​min of restraint stress (ref. McGill et al., 2006; ref. Nuber et al., 2005). However, this response altered after 30–60 ​min of restraint, at which point higher serum corticosterone was measured in Mecp2 null mice (ref. McGill et al., 2006; ref. Ren et al., 2012). Mecp2 null mice showed a greater increase in anxiety behaviour when moved from a large to a small chamber (restraint stress) compared to wildtype mice (ref. Ren et al., 2012). Environmental enrichment is associated with a reduction in corticosterone levels in Mecp2^+/− mice and anxiety behaviour being similar to wildtype (ref. Kondo et al., 2016).

Environmental changes overlap with/modify the effects of genetic loss of MeCP2

Environmental enrichment normalised some anxiety behaviours in Mecp2 null mice. Behavioural testing was performed in two mouse strains: 129/SvJ and mixed 129/SvJ and C57BL/6J background. 129/SvJ mice are known to perform poorly in the elevated plus maze (ref. Kerr et al., 2010). Environmental enrichment normalised anxiety in Mecp2^+/− mice (ref. Kondo et al., 2016). Environmentally enriched Mecp2 null mice had significantly lower mRNA expression for synaptophysin and post-synaptic density-95 protein in the hypothalamus, and lower mRNA expression for syntaxin 1A and synaptotagmin in the cortex, when compared to standard housed Mecp2 null mice (ref. Kerr et al., 2010).

Maternally-derived cortisol was shown not to affect anxiety behaviour of Mecp2 null and wildtype pups, or their response to restraint stress. Despite this, changes were found in mineralocorticoid receptor (MR) mRNA expression, which was increased in both genotypes in mice with mothers who drank cortisol solution; this effect was more marked in the Mecp2 null mice. Glucocorticoid receptor (GR) mRNA expression was increased in Mecp2 null mice with mothers who drank cortisol solution compared to Mecp2 null mice without cortisol supplementation. (ref. De Filippis et al., 2013).

Behavioural testing of Mecp2^+/− and wildtype mice separated from their mothers (intended to induce early life stress) indicated that Mecp2^+/− mice were less anxious and that maternal separation did not alter this. However, when exposed to a more anxiogenic paradigm – elevated plus maze under bright-light conditions – maternal separation resulted in both wildtype and Mecp2^+/− mice being less anxious than their non-separated counterparts. Although maternal separation is usually used to induce early life stress, the authors noted that Mecp2^+/− mothers (mothers of both wildtype and Mecp2^+/− offspring in this study) tend to be neglectful and cannibalistic and C57BL/6 mice (the background in this study) also tend to be neglectful. Therefore, separation may have enabled the mice to become more resilient than the mice who stayed with the mothers. (ref. Abellán-Álvaro et al., 2021).

Discussion

Summary of findings

Many of the included studies illustrated effects on anxiety behaviour through use of mouse models with specific loss/reintroduction of Mecp2 from specific brain regions or cell subtypes. Loss/reintroduction of Mecp2 to some anxiety macrocircuit regions was associated with a change in anxiety behaviour, but for the majority of studies with loss/reintroduction of Mecp2 from particular cell subtypes, there were mixed effects on anxiety behaviour. Fig. 2, panel A summarises the brain regions identified as contributing/not contributing to anxiety in models of Rett syndrome and maps them onto the regions currently recognised to have a role in anxiety processing (ref. Calhoon and Tye, 2015). There is evidence (ref. Orefice et al., 2016) that the anxiety response could be modulated as early as the sensory detection stage. Fig. 2, panel B summarises the cell types identified as contributing/not contributing to anxiety in models of Rett syndrome. The mixed results found within some studies (ref. Zhou et al., 2017; ref. Zhang et al., 2016; ref. Chao et al., 2010; ref. Ure et al., 2016; ref. Samaco et al., 2009; ref. Lang et al., 2012; ref. Su et al., 2015) may reflect: variability in the validity of the anxiety behavioural tests (discussed in detail in section 4.2); the specificity of MeCP2 function within particular cell types; the fact that removal of Mecp2 from a cell subtype throughout the brain may mask the effects of that cell in a specific microcircuit; or may arise from the complexity of anxiety microcircuits (making a change to one neuronal or neurotransmitter subtype may not cause a definitive change to the overall circuit). If this last aspect were the case, there would be implications for anxiolytic drug treatments, which at present tend to target single or only few neurotransmitters. Inflammation is increasingly regarded as a contributor to mental disorders. Although the relevance of MeCP2 function in microglia is unclear (ref. Derecki et al., 2012; ref. Wang et al., 2015), there is evidence for an altered systemic inflammatory profile and for altered microglial activation (which can lead to increased release of proinflammatory molecules) in Rett syndrome (ref. Jin et al., 2017) suggesting possible routes for neuroinflammation contributing to anxiety symptoms.

Loss of MeCP2 in certain brain regions has been associated with reduced neurotransmitter (ref. Orefice et al., 2016; ref. Zhang et al., 2016; ref. Samaco et al., 2009; ref. Ricceri et al., 2011) and growth factor expression (ref. Kondo et al., 2016; ref. Mellios et al., 2014; ref. Castro et al., 2014), loss of neurotransmitter receptor subunits (ref. Orefice et al., 2016), reduced stabilisation of AMPA receptors (ref. Gambino et al., 2010), and alterations to synapses (ref. Gambino et al., 2010).

There is evidence for MeCP2 loss having an impact on regulation of the HPA axis, and mechanisms are beginning to be understood (ref. McGill et al., 2006; ref. Adachi et al., 2009; ref. Nuber et al., 2005; ref. Kerr et al., 2010; ref. Ren et al., 2012). There is evidence for increased FKBP5 mRNA expression in whole brains of null mice and CRH expression may be altered, although this appears to vary between brain regions and between male and female mice (ref. McGill et al., 2006; ref. Kondo et al., 2016; ref. Nuber et al., 2005). Mecp2 null mice show elevated serum corticosterone levels at a delayed timepoint following restraint stress (ref. McGill et al., 2006; ref. Ren et al., 2012).

Validity of the pre-clinical studies

Mecp2 mouse models have reasonable construct validity, as the majority of humans with Rett syndrome have MECP2 mutations. However, most mouse models are null for Mecp2; this is representative of large deletions in humans, which account for only 10% of cases. Although mouse models with conditional deletion of the Mecp2 gene are useful in understanding function of MeCP2 in individual cell subtypes, they do not recapitulate Rett syndrome, since they do not represent the global loss of MeCP2 and therefore have limited translational value. In models where there is conditional deletion of the Mecp2 gene in the targeted region, it is notable that they do not always result in complete loss of MeCP2 expression in that region (ref. Adachi et al., 2009; ref. Gemelli et al., 2006).

Only 11/38 of the studies included female mice (all found within Table 1), whereas Rett syndrome almost exclusively affects girls (ref. Katz et al., 2012). This is of particular significance in the study of anxiety disorders, since the prevalence of anxiety disorders in the general population is twice as high in women compared to men (ref. Bandelow and Michaelis, 2015) and sex-dependent effects on anxiety and stress-vulnerability have been found in Mecp2-deficient mice (ref. Philippe et al., 2018; ref. Cosentino et al., 2021).

As shown in Table 2, Mecp2 mouse models have differing genetic backgrounds and have been obtained from different sources. This makes comparison between studies and reproducibility more difficult. Although it has been shown that certain mouse models may show increased anxiety behaviour compared to others, there is no clear pattern amongst the studies reviewed in this paper (ref. Samaco et al., 2013). The effect of environmental enrichment on anxiety in mice with Mecp2 mutations emphasises that standardised animal husbandry practices are important in securing the reproducibility of results (ref. Katz et al., 2012; ref. Kondo et al., 2016; ref. Kerr et al., 2010) and the maternal separation study indicates that maternal temperament may affect anxiety (ref. Abellán-Álvaro et al., 2021).

There is debate as to the validity of mouse anxiety behavioural experiments. The behavioural experiments discussed in this paper best approximate to generalised anxiety disorder and specific phobias. These tests are imperfect, but there is no ‘gold standard’ anxiety behavioural experiment. Tests similar to the elevated plus maze and open field test have been performed in humans and behaviour within these tests is comparable between mouse and human. However, further work is needed to understand if the neurobiological anxiety mechanisms are comparable between mouse and human in these paradigms (ref. Bach, 2022). One key limitation of mouse behavioural experiments in the context of Rett syndrome mouse models is the motor impairment component of the phenotype. As anxiety behavioural experiments all rely on locomotion, this presents a clear a problem in interpretation. Measures such as ultrasonic vocalisation may be a useful alternative that does not rely on motor function (ref. Mun et al., 2015).

Although decreased time in the open arm and increased time in the closed arm of the elevated plus maze are typically interpreted as signs of increased anxiety in mice, it has been shown that female Mecp2^tm1.1Bird mice avoid the closed arms due to whisker hypersensitivity, rather than true preference for the open arm (ref. Flores Gutiérrez et al., 2020). This potentially affects the interpretation of three of the papers involving female mice (ref. Kondo et al., 2016; ref. Meng et al., 2016; ref. Abellán-Álvaro et al., 2021). It would be important to clarify further whether this response is found across different Mecp2 mutations, mouse genders and with loss of MeCP2 from specific cells/anatomical regions. It also suggests that the open field test and light-dark assay may be preferable tests for quantifying anxiety behaviour in Rett syndrome mouse models, compared to the elevated plus maze or elevated zero maze (ref. Lioy et al., 2011; ref. Ure et al., 2016; ref. Flores Gutiérrez et al., 2020).

A diverse number of behavioural tests and variety of behavioural measures have been used in the studies within the review. For example, some studies measured latency to enter the open area of the elevated plus maze and others measured the time spent in the open area. This is a further factor that makes comparison between studies difficult. Physiological measures, such as heart rate may be considered as a more objective and translatable measure of anxiety response and it has been shown that physiological changes occur when mice are in the elevated plus maze (ref. Okonogi et al., 2018).

Anxiety mechanisms identified within human studies

Neurotransmitters and their metabolites

Prior to the identification of MECP2 mutation as the main cause of Rett syndrome, early research into the pathology of the syndrome identified that reduction of levels of biogenic amines and their metabolites was a common feature; however biogenic amines were not correlated with anxiety levels (ref. He et al., 2014; ref. Roux and Villard, 2010). Individuals with the pArg168X mutation (associated with more severe overall symptoms of Rett syndrome) had significantly lower cerebrospinal fluid (CSF) 5-hydroxy indole acetic acid (5-HIAA) levels than individuals with the pArg133Cys (associated with less severe global impairment) mutation or controls (ref. Samaco et al., 2009). RSBQ and ADAMS anxiety scores have significant (or trend towards) inverse correlation with level of global impairment (ref. Barnes et al., 2015; ref. Anderson et al., 2014; ref. Robertson et al., 2006). These findings suggest that lower CSF 5-HIAA levels are associated with lower anxiety levels in Rett syndrome (ref. Samaco et al., 2009; ref. Barnes et al., 2015). A study has not yet been done exploring the efficacy of SSRIs in individuals with Rett syndrome, however as part of the Rett Syndrome Natural History Study, parents have reported that they observed improvement in anxiety symptoms with SSRI treatment (ref. Buchanan et al., 2022).

HPA axis

Despite there being a number of pre-clinical studies suggesting a role for MeCP2 affecting HPA axis function, there have been few clinical studies exploring this. Total free urinary cortisol/24 ​h correlates well with free plasma cortisol levels (ref. El-Farhan et al., 2017). A study involving 10 females with Rett syndrome (stage III-IV) compared to age- and sex-matched controls showed no significant difference in total urinary cortisol/24 ​h (ref. Motil et al., 2006). In healthy individuals there is a 90% reduction in evening cortisol compared to morning concentration. An exploratory study of 30 individuals with Rett syndrome identified that in individuals with mutations leading to less severe impairment, the diurnal variation was more pronounced (closer to normal) than individuals with mutations leading to more severe global impairment (ref. Byiers et al., 2020). Further study, including correlation of cortisol with RSBQ, will be needed to understand the impact of MeCP2 on the HPA axis.

Growth factors

Although preclinical studies suggest a role for IGF-1 in anxiety behaviour, clinical trials for IGF-1 have produced a complex picture, discussed by ref. Singh and Santosh (2018). IGF-1 treatment is due to be studied further in a Phase 3 trial (ref. Neul et al., 2022). A post-mortem histology study suggested that BDNF expression is region-specific and that levels were higher in the cerebellum in Rett syndrome patients (ref. Pejhan et al., 2020a) although a study by the same group using ELISA (quantifying protein levels) suggested BDNF levels were lower in all regions (ref. Pejhan et al., 2020b); more detailed study of BDNF and IGF-1 expression in anxiety regions may be helpful for understanding its mechanism in Rett syndrome.

Gut-brain axis

Gastrointestinal symptoms are a common feature of Rett syndrome. Differences in the gut microbiome have been reported between Rett syndrome patients, but these did not correlate with differences in anxiety symptoms (as measured by the Gastrointestinal Health Questionnaire, validated tool for use in Rett syndrome patients), however it is possible that confounding factors affect interpretation of this correlation (ref. Thapa et al., 2021).

Sex-hormone influence

Fifteen out of twenty-one young women with Rett syndrome reported that premenstrual syndrome was a problem, and eight women described anxiety as being one of the premenstrual symptoms, suggesting their mood was modulated by hormonal changes (ref. Hamilton et al., 2012).

Potential environmental influence

Hospital Anxiety and Depression Scales (HADS) (ref. Zigmond and Snaith, 1983) were completed by mothers of individuals with Rett syndrome who were found to have greater anxiety, but not depression, than a British normative sample. Siblings of those with Rett syndrome had similar emotion regulation (according to the Strengths and Difficulties Questionnaire) (ref. Goodman, 1997) to the normal population. An increased severity of the RSBQ score (in the child with Rett syndrome) was correlated with a higher level of maternal stress and anxiety (ref. Cianfaglione et al., 2015). Unlike in preclinical studies, environmental enrichment did not alter anxiety symptoms (ref. Downs et al., 2018).

Summary

When combined with findings from a previous review of autonomic dysregulation in Rett syndrome (ref. Singh and Santosh, 2018), the findings described within the current review indicate that there are multiple potential mechanisms contributing to anxiety in Rett syndrome. As identified from the preclinical studies, further study is needed to understand the precise mechanisms of serotonin in contributing to anxiety in Rett syndrome. It appears that certain MECP2 mutations are associated with greater anxiety symptoms.

Epigenetics

Within rett syndrome

Although Rett syndrome is a genetic disorder, the mutation of MECP2 has epigenetic consequences, which are complex (ref. Marano et al., 2021). This MeCP2 dysfunction can interact with other epigenetic mechanisms; for example there may be differing levels of DNA methylation between individuals with Rett syndrome, adding another layer to the way MeCP2 can function (ref. Marano et al., 2021). Pre-clinical and clinical studies within this paper highlight potential environmental influences (with likely epigenetic effect) which may contribute to anxiety in Rett syndrome (ref. De Filippis et al., 2013; ref. Abellán-Álvaro et al., 2021; ref. Cianfaglione et al., 2015).

Rett syndrome may inform understanding of the epigenetics of anxiety

Epigenetic modulators, including MeCP2, are increasingly recognised as the mechanism by which development (and later life changes) can be affected by environment. Early life stress is a key environmental influence that predisposes individuals to mental health disorders, including anxiety disorders (ref. Gottschalk and Domschke, 2016; ref. Bearer and Mulligan, 2018; ref. Schiele et al., 2020). Preclinical studies have demonstrated different mechanisms linking reduced MeCP2 function with early life stress (ref. Murgatroyd et al., 2009; ref. Lewis et al., 2016) and these are depicted in Fig. 3. This mechanism is of interest since it suggests mechanistic overlap with Rett syndrome and that downstream cellular and molecular anxiety mechanisms of Rett syndrome could also be relevant to those with anxiety and a history of early life stress. Further, MECP2 mutation and epigenetic modification leading to altered MECP2 gene expression is reported in patients with autism spectrum condition (ASC) and schizophrenia (ref. Qiu, 2018; ref. Ciernia and LaSalle, 2016; ref. Richetto and Meyer, 2021). In a preclinical model, specific methylation of the Mecp2 promotor in the hippocampus was sufficient to reduce Mecp2 expression and induce most ASC-like behaviours (ref. Lu et al., 2020). ASC and schizophrenia have been associated with aberrant neuronal circuitry and anxiety (ref. Qiu, 2018).

Epigenetic mechanisms of anxiety have previously been reviewed (ref. Gottschalk and Domschke, 2016) and the importance of developing biomarkers for the epigenetic effects of stress to enable individualised treatment has been emphasised (ref. Gottschalk et al., 2020). Peripheral biomarkers (often blood and saliva) can only provide limited understanding of epigenetic changes within the CNS. Rett syndrome models may provide a method to develop epigenetic biomarkers – such as blood, saliva, and possibly cerebrospinal fluid – that may correlate with CNS mechanisms and anxiety symptoms. This could be used to enhance understanding of anxiety in those exposed to early life stress or with ASC, in addition to guiding anxiety treatment in Rett syndrome (ref. Bearer and Mulligan, 2018).

Translation to therapeutic applications

Mecp2/MECP2 mutations result in comparable effects in mice and humans. This increases the likelihood that biological mechanisms are conserved between species and that these may have value as translational models for developing targeted drug treatments.

At present drug treatments for anxiety disorders are largely limited to SSRIs and serotonin and noradrenaline reuptake inhibitors (SNRIs). Understanding the molecular and cell- and region-specific effects of MeCP2 loss will be important in developing effective treatments for anxiety in Rett syndrome. The mechanisms identified in this review highlight the potential breadth of targets, for example the sensory system, a wider range of neurotransmitters and regulation of the HPA axis. The current range of investigations do not allow us to identify reliable potential treatment targets, but potentially compounds targeting astrocytic functions may be relevant (ref. Lioy et al., 2011). Results from pre-clinical and clinical studies seem to suggest that only specific serotonergic agents may be effective (ref. Samaco et al., 2009; ref. De Filippis et al., 2014; ref. Barnes et al., 2015).

Given the potential overlapping mechanism of Rett syndrome and early life stress, treatments for anxiety in Rett syndrome may also be effective in those with anxiety and a history of early life stress.

Anxiety in patients with ASC is often difficult to treat but there have been few studies of the efficacy of pharmacotherapy. The current evidence from studies of children and adolescents is insufficient to support treatment with SSRIs (ref. Stepanova et al., 2017). This suggests the need for an alternative innovative approach to treating anxiety symptoms and disorders, possibly linked to MeCP2 and its isoforms (ref. Good et al., 2021).

As well as anxiety, disordered breathing is a common feature in Rett syndrome. Although anxiety can worsen respiratory function, it is possible that altered respiratory function can exacerbate anxiety (ref. Ren et al., 2012). Within the field of anxiety research, the 7.5% CO₂ inhalation experimental medicine model is well established. In this model humans inhale 7.5% CO₂ over 20 ​min, which induces subjective and autonomic responses and neurocognitive changes that are comparable to the features of generalised anxiety disorder. The mechanisms are only partially understood (ref. Baldwin et al., 2017). Given that carbon dioxide/acid sensing is involved in breathing and the CO₂ anxiety model, this prompts consideration of an overlap between anxiety and breathing mechanisms in Rett syndrome, for example via similar cellular or molecular sensing mechanisms. Exploration of these factors may extend knowledge of contributing mechanisms in the CO₂ model and so inform understanding of overall mechanisms of anxiety.

Conclusion and next steps

Rett syndrome is a complex disorder with multiple intertwining cellular and systems abnormalities. The combined findings of pre-clinical and clinical studies show that progress is being made towards improved understanding of anxiety mechanisms in Rett syndrome.

Future steps for pre-clinical work will need to include greater use of female mice within studies, clarification of whether whisker sensitivity (affecting interpretation of behaviour in the elevated plus maze) is found across Rett syndrome models, inclusion of tests that are not affected by motor impairment (such as ultrasonic vocalisation) and inclusion of physiological markers (corticosterone and heart rate) that are quantifiable and translatable. Standardisation of anxiety tests would enable improved comparison between studies. Given that environmental factors may modify anxiety, standard environment should be used as well as consideration of maternal temperament. In terms of added mechanistic understanding, it will be helpful to develop understanding about the effects of Mecp2 mutation on specific anxiety-circuitry connections.

Given the importance of anxiety as a symptom within Rett syndrome, clinical studies should increasingly aim to include anxiety measures, including RSBQ. The HPA axis in individuals with Rett syndrome is an area that also requires increased study.

Improved understanding of anxiety mechanisms in Rett syndrome (which may include development of biomarkers to identify the mechanisms present within an individual, given the heterogenous nature of the syndrome) could enable more effective treatments to be developed. Importantly, the known MECP2 mutation means that there is the potential for translation from mouse models to human therapeutic applications. Improved understanding of potential overlapping mechanisms between Rett syndrome, autism and early life stress may allow more effective anxiolytic treatments to be developed across these disorders.

Funding

Bethan Impey worked on this review whilst in position as an NIHR funded Academic Clinical Fellow.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: BI has a small number of stocks and shares in GlaxoSmithKline. TAN has no conflicts of interest to declare. DSB is Medical Patron of Anxiety UK, and President-Elect of the British Association for Psychopharmacology. He receives an editor fee from Wiley, for his editorship of Human Psychopharmacology. He was an unpaid consultant for Idorsia in July 2020.

References

1. M. Abellán-Álvaro, O. Stork, C. Agustín-Pavón, M. Santos. MeCP2 haplodeficiency and early-life stress interaction on anxiety-like behavior in adolescent female mice. J. Neurodev. Disord., 2021. [DOI | PubMed]
2. M. Adachi, A.E. Autry, H.E. Covington, L.M. Monteggia. MeCP2-mediated transcription repression in the basolateral amygdala may underlie heightened anxiety in a mouse model of Rett syndrome. J. Neurosci., 2009. [DOI | PubMed]
3. A. Anderson, K. Wong, P. Jacoby, J. Downs, H. Leonard. Twenty years of surveillance in Rett syndrome: what does this tell us?. Orphanet J. Rare Dis., 2014. [DOI | PubMed]
4. D.R. Bach. Cross-species anxiety tests in psychiatry: pitfalls and promises. Mol. Psychiatr., 2022
5. K. Bailey, J. Crawley. 2009
6. D.S. Baldwin, R. Hou, R. Gordon, N.T. Huneke, M. Garner. Pharmacotherapy in generalized anxiety disorder: novel experimental medicine models and emerging drug targets. CNS Drugs, 2017. [DOI | PubMed]
7. B. Bandelow, S. Michaelis. Epidemiology of anxiety disorders in the 21st century. Dialogues Clin. Neurosci., 2015. [DOI | PubMed]
8. K.V. Barnes, F.R. Coughlin, H.M. O’Leary, N. Bruck, G.A. Bazin, E.B. Beinecke. Anxiety-like behavior in Rett syndrome: characteristics and assessment by anxiety scales. J. Neurodev. Disord., 2015. [DOI | PubMed]
9. E.L. Bearer, B.S. Mulligan. Epigenetic changes associated with early life experiences: saliva, a biospecimen for dna methylation signatures. Curr. Genom., 2018. [DOI]
10. T. Bittolo, C.A. Raminelli, C. Deiana, G. Baj, V. Vaghi, S. Ferrazzo. Pharmacological treatment with mirtazapine rescues cortical atrophy and respiratory deficits in MeCP2 null mice. Sci. Rep., 2016. [DOI]
11. C.B. Buchanan, J.L. Stallworth, A.E. Scott, D.G. Glaze, J.B. Lane, S.A. Skinner. Behavioral profiles in Rett syndrome: data from the natural history study. Brain Dev., 2019. [DOI | PubMed]
12. C.B. Buchanan, J.L. Stallworth, A.E. Joy, R.E. Dixon, A.E. Scott, A.A. Beisang. Anxiety-like behavior and anxiolytic treatment in the Rett syndrome natural history study. J. Neurodev. Disord., 2022. [PubMed]
13. B.J. Byiers, A. Payen, T. Feyma, A. Panoskaltsis-Mortari, M.J. Ehrhardt, F.J. Symons. Associations among diurnal salivary cortisol patterns, medication use, and behavioral phenotype features in a community sample of rett syndrome. Am. J. Intellect. Dev. Disabil., 2020. [DOI | PubMed]
14. G.G. Calhoon, K.M. Tye. Resolving the neural circuits of anxiety. Nat. Neurosci., 2015. [DOI | PubMed]
15. S. Cassel, D. Carouge, C. Gensburger, P. Anglard, C. Burgun, J.-B. Dietrich. Fluoxetine and cocaine induce the epigenetic factors MeCP2 and MBD1 in adult rat brain. Mol. Pharmacol., 2006. [DOI | PubMed]
16. J. Castro, R.I. Garcia, S. Kwok, A. Banerjee, J. Petravicz, J. Woodson. Functional recovery with recombinant human IGF1 treatment in a mouse model of Rett Syndrome. Proc. Natl. Acad. Sci. USA, 2014. [DOI | PubMed]
17. H.-T. Chao, H. Chen, R.C. Samaco, M. Xue, M. Chahrour, J. Yoo. Dysfunction in GABA signalling mediates autism-like stereotypies and Rett syndrome phenotypes. Nature, 2010. [DOI | PubMed]
18. E.W. Chin, W.M. Lim, D. Ma, F.J. Rosales, E.L. Goh. Choline rescues behavioural deficits in a mouse model of Rett syndrome by modulating neuronal plasticity. Mol. Neurobiol., 2019. [DOI | PubMed]
19. R. Cianfaglione, R.P. Hastings, D. Felce, A. Clarke, M.P. Kerr. Psychological well-being of mothers and siblings in families of girls and women with Rett syndrome. J. Autism Dev. Disord., 2015. [DOI | PubMed]
20. A.V. Ciernia, J. LaSalle. The landscape of DNA methylation amid a perfect storm of autism aetiologies. Nat. Rev. Neurosci., 2016. [PubMed]
21. L. Cosentino, F. Bellia, N. Pavoncello, D. Vigli, C. D’Addario, B. De Filippis. Methyl-CpG binding protein 2 dysfunction provides stress vulnerability with sex-and zygosity-dependent outcomes. Eur. J. Neurosci., 2021. [DOI]
22. B. De Filippis, L. Ricceri, A. Fuso, G. Laviola. Neonatal exposure to low dose corticosterone persistently modulates hippocampal mineralocorticoid receptor expression and improves locomotor/exploratory behaviour in a mouse model of Rett syndrome. Neuropharmacology, 2013. [DOI | PubMed]
23. B. De Filippis, P. Nativio, A. Fabbri, L. Ricceri, W. Adriani, E. Lacivita. Pharmacological stimulation of the brain serotonin receptor 7 as a novel therapeutic approach for Rett syndrome. Neuropsychopharmacology, 2014. [DOI | PubMed]
24. N.C. Derecki, J.C. Cronk, Z. Lu, E. Xu, S.B. Abbott, P.G. Guyenet. Wild-type microglia arrest pathology in a mouse model of Rett syndrome. Nature, 2012. [DOI | PubMed]
25. J. Downs, J. Rodger, C. Li, X. Tan, N. Hu, K. Wong. Environmental enrichment intervention for Rett syndrome: an individually randomised stepped wedge trial. Orphanet J. Rare Dis., 2018. [DOI | PubMed]
26. N. El-Farhan, D.A. Rees, C. Evans. Measuring cortisol in serum, urine and saliva–are our assays good enough?. Ann. Clin. Biochem., 2017. [DOI | PubMed]
27. A.J. Esbensen, J. Rojahn, M.G. Aman, S. Ruedrich. Reliability and validity of an assessment instrument for anxiety, depression, and mood among individuals with mental retardation. J. Autism Dev. Disord., 2003. [DOI | PubMed]
28. J. Flores Gutiérrez, C. De Felice, G. Natali, S. Leoncini, C. Signorini, J. Hayek. Protective role of mirtazapine in adult female Mecp2+/− mice and patients with Rett syndrome. J. Neurodev. Disord., 2020. [PubMed]
29. S.L. Fyffe, J.L. Neul, R.C. Samaco, H.-T. Chao, S. Ben-Shachar, P. Moretti. Deletion of Mecp2 in Sim1-expressing neurons reveals a critical role for MeCP2 in feeding behavior, aggression, and the response to stress. Neuron, 2008. [DOI | PubMed]
30. F. Gambino, M. Khelfaoui, B. Poulain, T. Bienvenu, J. Chelly, Y. Humeau. Synaptic maturation at cortical projections to the lateral amygdala in a mouse model of Rett syndrome. PLoS One, 2010. [DOI]
31. T. Gemelli, O. Berton, E.D. Nelson, L.I. Perrotti, R. Jaenisch, L.M. Monteggia. Postnatal loss of methyl-CpG binding protein 2 in the forebrain is sufficient to mediate behavioral aspects of Rett syndrome in mice. Biol. Psychiatr., 2006. [DOI]
32. D. Goffin, E.S. Brodkin, J.A. Blendy, S.J. Siegel, Z. Zhou. Cellular origins of auditory event-related potential deficits in Rett syndrome. Nat. Neurosci., 2014. [DOI | PubMed]
33. W.A. Gold, R. Krishnarajy, C. Ellaway, J. Christodoulou. Rett syndrome: a genetic update and clinical review focusing on comorbidities. ACS Chem. Neurosci., 2017. [DOI]
34. K.V. Good, J.B. Vincent, J. Ausió. MeCP2: the genetic driver of Rett syndrome epigenetics. Front. Genet., 2021. [DOI]
35. R. Goodman. The Strengths and Difficulties questionnaire: a research note. JCPP (J. Child Psychol. Psychiatry), 1997. [DOI | PubMed]
36. M.G. Gottschalk, K. Domschke. Novel developments in genetic and epigenetic mechanisms of anxiety. Curr. Opin. Psychiatr., 2016. [DOI]
37. M.G. Gottschalk, K. Domschke, M.A. Schiele. Epigenetics underlying susceptibility and resilience relating to daily life stress, work stress, and socioeconomic status. Front. Psychiatr., 2020. [DOI]
38. A. Hamilton, M.P. Marshal, G.S. Sucato, P.J. Murray. Rett syndrome and menstruation. J. Pediatr. Adolesc. Gynecol., 2012. [DOI | PubMed]
39. S. Hao, B. Tang, Z. Wu, K. Ure, Y. Sun, H. Tao. Forniceal deep brain stimulation rescues hippocampal memory in Rett syndrome mice. Nature, 2015. [DOI | PubMed]
40. L-j He, N. Liu, T-l Cheng, X-j Chen, Y-d Li, Y-s Shu. Conditional deletion of Mecp2 in parvalbumin-expressing GABAergic cells results in the absence of critical period plasticity. Nat. Commun., 2014. [DOI]
41. A. Ito-Ishida, K. Ure, H. Chen, J.W. Swann, H.Y. Zoghbi. Loss of MeCP2 in parvalbumin-and somatostatin-expressing neurons in mice leads to distinct Rett syndrome-like phenotypes. Neuron, 2015. [DOI | PubMed]
42. X.-R. Jin, X.-S. Chen, L. Xiao. MeCP2 deficiency in neuroglia: new progress in the pathogenesis of Rett syndrome. Front. Mol. Neurosci., 2017. [DOI | PubMed]
43. D.M. Katz, J.E. Berger-Sweeney, J.H. Eubanks, M.J. Justice, J.L. Neul, L. Pozzo-Miller. Preclinical research in Rett syndrome: setting the foundation for translational success. Dis Model Mech, 2012. [DOI | PubMed]
44. B. Kerr, P.A. Silva, K. Walz, J.I. Young. Unconventional transcriptional response to environmental enrichment in a mouse model of Rett syndrome. PLoS One, 2010. [DOI]
45. M.A. Kondo, L.J. Gray, G.J. Pelka, S.K. Leang, J. Christodoulou, P.P. Tam. Affective dysfunction in a mouse model of Rett syndrome: therapeutic effects of environmental stimulation and physical activity. Dev Neurobiol, 2016. [DOI | PubMed]
46. M. Lang, R.G. Wither, J.M. Brotchie, C. Wu, L. Zhang, J.H. Eubanks. Selective preservation of MeCP2 in catecholaminergic cells is sufficient to improve the behavioral phenotype of male and female Mecp2-deficient mice. Hum. Mol. Genet., 2012. [DOI | PubMed]
47. C.R. Lewis, R.M. Bastle, T.B. Manning, S.M. Himes, P. Fennig, P.R. Conrad. Interactions between early life stress, nucleus accumbens MeCP2 expression, and methamphetamine self-administration in male rats. Neuropsychopharmacology, 2016. [PubMed]
48. D.T. Lioy, S.K. Garg, C.E. Monaghan, J. Raber, K.D. Foust, B.K. Kaspar. A role for glia in the progression of Rett’s syndrome. Nature, 2011. [DOI | PubMed]
49. Z. Lu, Z. Liu, W. Mao, X. Wang, X. Zheng, S. Chen. Locus-specific DNA methylation of Mecp2 promoter leads to autism-like phenotypes in mice. Cell Death Dis., 2020. [PubMed]
50. J.G. Mantis, C.L. Fritz, J. Marsh, S.C. Heinrichs, T.N. Seyfried. Improvement in motor and exploratory behavior in Rett syndrome mice with restricted ketogenic and standard diets. Epilepsy Behav., 2009. [DOI | PubMed]
51. D. Marano, S. Fioriniello, M. D’Esposito, F. Della Ragione. Transcriptomic and epigenomic landscape in rett syndrome. Biomolecules, 2021. [DOI | PubMed]
52. B.E. McGill, S.F. Bundle, M.B. Yaylaoglu, J.P. Carson, C. Thaller, H.Y. Zoghbi. Enhanced anxiety and stress-induced corticosterone release are associated with increased Crh expression in a mouse model of Rett syndrome. Proc. Natl. Acad. Sci. USA, 2006. [DOI | PubMed]
53. N. Mellios, J. Woodson, R.I. Garcia, B. Crawford, J. Sharma, S.D. Sheridan. β2-Adrenergic receptor agonist ameliorates phenotypes and corrects microRNA-mediated IGF1 deficits in a mouse model of Rett syndrome. Proc. Natl. Acad. Sci. USA, 2014. [DOI | PubMed]
54. X. Meng, W. Wang, H. Lu, L-j He, W. Chen, E.S. Chao. Manipulations of MeCP2 in glutamatergic neurons highlight their contributions to Rett and other neurological disorders. Elife, 2016. [DOI]
55. D. Moher, A. Liberati, J. Tetzlaff, D.G. Altman, P. Group∗. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann. Intern. Med., 2009. [DOI | PubMed]
56. J.M. Mossner, R. Batista-Brito, R. Pant, J.A. Cardin. Developmental loss of MeCP2 from VIP interneurons impairs cortical function and behavior. Elife, 2020. [DOI]
57. K.J. Motil, R.J. Schultz, S. Abrams, K.J. Ellis, D.G. Glaze. Fractional calcium absorption is increased in girls with Rett syndrome. J. Pediatr. Gastroenterol. Nutr., 2006. [DOI | PubMed]
58. R.H. Mount, T. Charman, R.P. Hastings, S. Reilly, H. Cass. The rett syndrome behaviour questionnaire (RSBQ): refining the behavioural phenotype of rett syndrome. JCPP (J. Child Psychol. Psychiatry), 2002. [DOI | PubMed]
59. H. Mun, T. Lipina, J. Roder. Ultrasonic vocalizations in mice during exploratory behavior are context-dependent. 9: 316. Front. Behav. Neurosci., 2015. [DOI]
60. C. Murgatroyd, A.V. Patchev, Y. Wu, V. Micale, Y. Bockmühl, D. Fischer. Dynamic DNA methylation programs persistent adverse effects of early-life stress. Nat. Neurosci., 2009. [DOI | PubMed]
61. J.L. Neul, A.K. Percy, T.A. Benke, E.M. Berry-Kravis, D.G. Glaze, S.U. Peters. Design and outcome measures of LAVENDER, a phase 3 study of trofinetide for Rett syndrome. Contemp. Clin. Trials, 2022. [DOI | PubMed]
62. U.A. Nuber, S. Kriaucionis, T.C. Roloff, J. Guy, J. Selfridge, C. Steinhoff. Up-regulation of glucocorticoid-regulated genes in a mouse model of Rett syndrome. Hum. Mol. Genet., 2005. [DOI | PubMed]
63. T. Okonogi, R. Nakayama, T. Sasaki, Y. Ikegaya. Characterization of peripheral activity states and cortical local field potentials of mice in an elevated plus maze test. Front. Behav. Neurosci., 2018. [DOI | PubMed]
64. L.L. Orefice, A.L. Zimmerman, A.M. Chirila, S.J. Sleboda, J.P. Head, D.D. Ginty. Peripheral mechanosensory neuron dysfunction underlies tactile and behavioral deficits in mouse models of ASDs. Cell, 2016. [DOI | PubMed]
65. S. Pejhan, V.M. Siu, L. Ang, M. Del Bigio, M. Rastegar. Differential brain region-specific expression of MeCP2 and BDNF in Rett Syndrome patients: a distinct grey-white matter variation. Neuropathol. Appl. Neurobiol., 2020. [DOI | PubMed]
66. S. Pejhan, M.R. Del Bigio, M. Rastegar. The MeCP2E1/E2-BDNF-miR132 homeostasis regulatory network is region-dependent in the human brain and is impaired in rett syndrome patients. Front. Cell Dev. Biol., 2020. [DOI | PubMed]
67. T.J. Philippe, F. Vahid-Ansari, Z.R. Donaldson, B. Le François, A. Zahrai, V. Turcotte-Cardin. Loss of MeCP2 in adult 5-HT neurons induces 5-HT1A autoreceptors, with opposite sex-dependent anxiety and depression phenotypes. Sci. Rep., 2018. [DOI | PubMed]
68. Z. Qiu. Deciphering MECP2-associated disorders: disrupted circuits and the hope for repair. Curr. Opin. Neurobiol., 2018. [DOI | PubMed]
69. J. Ren, X. Ding, G.D. Funk, J.J. Greer. Anxiety-related mechanisms of respiratory dysfunction in a mouse model of Rett syndrome. J. Neurosci., 2012. [DOI | PubMed]
70. M.C. Ribeiro, J.L. MacDonald. Sex differences in Mecp2-mutant Rett syndrome model mice and the impact of cellular mosaicism in phenotype development. Brain Res., 2020
71. L. Ricceri, B. De Filippis, A. Fuso, G. Laviola. Cholinergic hypofunction in MeCP2-308 mice: beneficial neurobehavioural effects of neonatal choline supplementation. Behav. Brain Res., 2011. [DOI | PubMed]
72. J. Richetto, U. Meyer. Epigenetic modifications in schizophrenia and related disorders: molecular scars of environmental exposures and source of phenotypic variability. Biol. Psychiatr., 2021
73. L. Robertson, S.E. Hall, P. Jacoby, C. Ellaway, Nd Klerk, H. Leonard. The association between behavior and genotype in Rett syndrome using the Australian Rett Syndrome Database. Am J Med Genet B Neuropsychiatr Genet, 2006. [DOI]
74. J.-C. Roux, L. Villard. Biogenic amines in Rett syndrome: the usual suspects. Behav. Genet., 2010. [DOI | PubMed]
75. R.C. Samaco, C. Mandel-Brehm, H.-T. Chao, C.S. Ward, S.L. Fyffe-Maricich, J. Ren. Loss of MeCP2 in aminergic neurons causes cell-autonomous defects in neurotransmitter synthesis and specific behavioral abnormalities. Proc. Natl. Acad. Sci. USA, 2009. [DOI | PubMed]
76. R.C. Samaco, C.M. McGraw, C.S. Ward, Y. Sun, J.L. Neul, H.Y. Zoghbi. Female Mecp2+/− mice display robust behavioral deficits on two different genetic backgrounds providing a framework for pre-clinical studies. Hum. Mol. Genet., 2013. [DOI | PubMed]
77. M.A. Schiele, M.G. Gottschalk, K. Domschke. The applied implications of epigenetics in anxiety, affective and stress-related disorders-A review and synthesis on psychosocial stress, psychotherapy and prevention. Clin. Psychol. Rev., 2020
78. M.D. Shahbazian, B. Antalffy, D.L. Armstrong, H.Y. Zoghbi. Insight into Rett syndrome: MeCP2 levels display tissue-and cell-specific differences and correlate with neuronal maturation. Hum. Mol. Genet., 2002. [DOI | PubMed]
79. J. Singh, P. Santosh. Key issues in Rett syndrome: emotional, behavioural and autonomic dysregulation (EBAD)-a target for clinical trials. Orphanet J. Rare Dis., 2018. [DOI | PubMed]
80. E. Stepanova, S. Dowling, M. Phelps, R.L. Findling. Pharmacotherapy of emotional and behavioral symptoms associated with autism spectrum disorder in children and adolescents. Dialogues Clin. Neurosci., 2017. [DOI | PubMed]
81. S.-H. Su, F.-C. Kao, Y.-B. Huang, W. Liao. MeCP2 in the rostral striatum maintains local dopamine content critical for psychomotor control. J. Neurosci., 2015. [DOI | PubMed]
82. S. Thapa, A. Venkatachalam, N. Khan, M. Naqvi, M. Balderas, J.K. Runge. Assessment of the gut bacterial microbiome and metabolome of girls and women with Rett Syndrome. PLoS One, 2021. [DOI]
83. K. Ure, H. Lu, W. Wang, A. Ito-Ishida, Z. Wu, L-j He. Restoration of Mecp2 expression in GABAergic neurons is sufficient to rescue multiple disease features in a mouse model of Rett syndrome. Elife, 2016. [DOI]
84. D. Vigli, L. Cosentino, C. Raggi, G. Laviola, M. Woolley-Roberts, B. De Filippis. Chronic treatment with the phytocannabinoid Cannabidivarin (CBDV) rescues behavioural alterations and brain atrophy in a mouse model of Rett syndrome. Neuropharmacology, 2018. [DOI | PubMed]
85. D. Vigli, L. Cosentino, M. Pellas, B. De Filippis. Chronic treatment with cannabidiolic acid (CBDA) reduces thermal pain sensitivity in male mice and rescues the hyperalgesia in a mouse model of rett syndrome. Neuroscience, 2021. [DOI | PubMed]
86. C. Villani, M. Carli, A.M. Castaldo, G. Sacchetti, R.W. Invernizzi. Fluoxetine increases brain MeCP2 immuno-positive cells in a female Mecp2 heterozygous mouse model of Rett syndrome through endogenous serotonin. Sci. Rep., 2021. [DOI | PubMed]
87. J. Wang, J.E. Wegener, T.-W. Huang, S. Sripathy, H. De Jesus-Cortes, P. Xu. Wild-type microglia do not reverse pathology in mouse models of Rett syndrome. Nature, 2015. [DOI | PubMed]
88. R.G. Wither, M. Lang, L. Zhang, J.H. Eubanks. Regional MeCP2 expression levels in the female MeCP2-deficient mouse brain correlate with specific behavioral impairments. Exp. Neurol., 2013. [DOI | PubMed]
89. Y. Zhang, R.-R. Mao, Z.-F. Chen, M. Tian, D.-L. Tong, Z.-R. Gao. Deep-brain magnetic stimulation promotes adult hippocampal neurogenesis and alleviates stress-related behaviors in mouse models for neuropsychiatric disorders. Mol. Brain, 2014. [DOI | PubMed]
90. Y. Zhang, S.-X. Cao, P. Sun, H.-Y. He, C.-H. Yang, X.-J. Chen. Loss of MeCP2 in cholinergic neurons causes part of RTT-like phenotypes via α7 receptor in hippocampus. Cell Res., 2016. [DOI | PubMed]
91. H. Zhou, W. Wu, Y. Zhang, H. He, Z. Yuan, Z. Zhu. Selective preservation of cholinergic MeCP2 rescues specific Rett-syndrome-like phenotypes in MeCP2stop mice. Behav. Brain Res., 2017. [DOI | PubMed]
92. A.S. Zigmond, R.P. Snaith. The hospital anxiety and depression scale. Acta Psychiatr. Scand., 1983. [DOI | PubMed]
93. C.A. Zimmermann, A. Hoffmann, F. Raabe, D. Spengler. Role of mecp2 in experience-dependent epigenetic programming. Genes, 2015. [DOI | PubMed]