PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Neurobiol Dis. Author manuscript; available in PMC Apr 1, 2013.
Published in final edited form as:
PMCID: PMC3299908
NIHMSID: NIHMS348745
MeCP2+/− mouse model of RTT reproduces auditory phenotypes associated with Rett syndrome and replicate select EEG endophenotypes of autism spectrum disorder
Wenlin Liao,1 Michael J. Gandal, Richard S. Ehrlichman, Steven J. Siegel, and Greg C. Carlson*
Department of Psychiatry, University of Pennsylvania Philadelphia, PA 19104
1Present Address: Institute of Neuroscience National Cheng-Chi University 64, Sec. 2, ZhiNan Rd., Wenshan District Taipei City 11605 Taiwan
*Corresponding Author Address: Center for Neurobiology and Behavior, Upenn Room 2226, TRL 125 S. 31st Street Philadelphia, PA 19104-3403 ; gregc/at/mail.med.upenn.edu
Impairments in cortical sensory processing have been demonstrated in Rett syndrome (RTT) and Autism Spectrum Disorders (ASD) and are thought to contribute to high-order phenotypic deficits. However, underlying pathophysiological mechanisms for these abnormalities are unknown. This study investigated auditory sensory processing in a mouse model of RTT with a heterozygous loss of MeCP2 function. Cortical abnormalities in a number of neuropsychiatric disorders, including ASD are reflected in auditory evoked potentials and fields measured by EEG and MEG. One of these abnormalities, increased latency of cortically sourced components, is associated with language and developmental delay in autism. Additionally, gamma-band abnormalities have recently been identified as an endophenotype of idiopathic autism. Both of these cortical abnormalities are potential clinical endpoints for assessing treatment. While ascribing similar mechanisms of idiopathic ASD to Rett syndrome (RTT) has been controversial, we sought to determine if mouse models of RTT replicate these intermediate phenotypes. Mice heterozygous for the null mutations of the gene MeCP2, were implanted for EEG. In response to auditory stimulation, these mice recapitulated specific latency differences as well as select gamma and beta band abnormalities associated with ASD. MeCP2 disruption is the predominant cause of RTT, and reductions in MeCP2 expression predominates in ASD. This work further suggests a common cortical pathophysiology for RTT and ASD, and indicates that the MeCP2+/− model may be useful for preclinical development targeting specific cortical processing abnormalities in RTT with potential relevance to ASD.
Event-related magnetoencephalography (MEG) and electroencephalography (EEG) studies of auditory and language processing have identified intermediate phenotypes associated with autism spectrum disorders (ASD) and abnormal responses in Rett syndrome (RTT) (Bader et al., 1989; Badr et al., 1987; Kalmanchey, 1990; Oram Cardy et al., 2008; Roberts et al., 2008; Stach et al., 1994). In autism, delayed middle latency components of the auditory-evoked response have been observed in the cortex (superior temporal gyrus) and have been linked to higher-order language impairments (Roberts et al., 2011). Likewise, abnormalities in cortical gamma-band (30–80 Hz) synchrony have been observed in ASD, are thought to reflect deficits in excitatory-inhibitory balance (Gandal et al.; Rojas et al.; Wilson et al.). Fewer and a less conclusive set or studies have been performed in girls with RTT (Kalmanchey, 1990) (Stach et al., 1994; Yamanouchi et al., 1993). The relative lack of preclinical studies investigating these auditory response deficits limits our ability to test for relationships between intermediate clinical phenotypes and neuronal circuit abnormalities in RTT.
Rett is a unique disorder, but shares proposed mechanistic and core symptoms of autism. In contrast to the complex genetic etiology of idiopathic ASD, RTT has a clear monogenetic basis with mutations in the X-linked gene MeCP2 occurring in approximately 90% of patients with RTT. Nevertheless, RTT is characterized by developmental regression, similar to that observed in a subset of severely affected autistic children, combined with the loss of age-appropriate social interaction and speech. Clinically, RTT patients often present with core autism-like behavioral deficits, along with RTT specific components that include severe motor abnormalities. In some cases, patients with RTT associated mutations in MeCP2 nevertheless present preserved but affected speech and limited motor abnormalities leading to a diagnosis of ASD that is clinically undifferentiated from idiopathic ASD (Young et al., 2007). Reduced MeCP2 expression is also found in forebrain post-mortem tissue from the majority of idiopathic ASD subjects, suggesting similar epigenetic dysregulation in many cases of idiopathic ASD and RTT (Samaco et al., 2005; Samaco et al., 2004). Such links between MeCP2 and ASD suggest that mice with reduced MeCP2 expression may have construct validity for ASD as well as RTT, and thus models of MeCP2 dysfunction may help understand the distinction between RTT and ASD mechanisms and symptoms. Owing to the lack of well-characterized models of idiopathic ASD yet strong clinical data, changes in auditory and visual evoked phenotypes in a mouse model of RTT may provide additional insight into auditory and visual sensory processing abnormalities found in ASD.
To investigate the role of MeCP2 function on the integrity of sensory processing, this study measured auditory and visual event-related potentials (ERPs) in female mice carrying a single null allele of MeCP2, which replicates the genetic condition leading to RTT (Guy et al., 2001). A number of ERP features common to idiopathic ASD and RTT were observed, including delayed auditory-evoked responses, increased component amplitudes, and gamma-band abnormalities. Each of these differences have been identified in the ASD or RTT clinical population (Castren et al., 2003)These findings suggest that ASD and RTT may share a subset of underlying local circuit abnormalities that contribute to endophenotypic and behavioral abnormalities. Finally, by demonstrating intermediate phenotypic deficits in MeCP2+/− mice, this work helps bridge the divide between clinical and preclinical studies, providing a basis for future pathophysiological investigation and indicates targets for therapeutic development.
Event-related potentials
Animals and implantation surgery
Female heterozygous Mecp2 null mice (Mecp2tm1.1Bird/J) and littermate controls (n=9/group) were obtained from The Jackson Laboratory (Bar Harbor ME) and bred in house using males from the background strain (C57/B6J). At 4 months of age, mice underwent stereotaxic implantation of bipolar, twisted, stainless steel electrodes into region CA1 of the hippocampus (AP −2.2 mm, ML 2.0 mm, DV −1.9 mm; 100 um diameter, Plastics One, Roanoke VA). A reference skull screw was implanted over the primary visual cortex (AP −3.4 mm, ML −2.7 mm) and a ground screw was placed above the frontal cortex (AP 3.1 mm, ML −1.0 mm).
Recording of event-related potentials (ERPs)
Recording of auditory and visual ERPs was performed between 10AM to 4PM, after a minimum of two weeks recovery from surgery. All studies were performed in mice between the ages of 4 to 5 months, a period where these mice are largely presymptomatic (Stearns et al., 2007). The mice were tested in their home cages, which are fitted with special tops to accommodate speakers and electrode cables, and placed inside a Faraday cage. The mice were then acclimatized to the testing apparatus for 30 min before first stimulus onset. The head stage is connected to a 30 cm six channel electrode cable, which is in turn connected to a high-impedance differential AC amplifier (A-M Systems, Carlsborg WA). Auditory stimuli were generated by Micro1401 hardware and Spike 6 software (Cambridge Electronic Design) and delivered through speakers attached to the cage top. In the presence of background white noise of 55 dB, 150 single white-noise clicks (10 msec in duration) were issued at 82 dB 8 seconds apart. For gating experiments, 150 white-noise clicks pairs (S1, S2) were presented with a 500 msec inter-stimulus interval and a 9 second inter-trial interval. Auditory brain stem responses were derived from stimuli consisting of a 4000 white noise clicks (0.1 ms duration, 125 ms ISI) sampled at 10 kHz, repeated at 85 dB similar to as previously published (Connolly et al., 2003). Recordings were filtered 100–500 Hz offline (Digital IIR filter, butterworth bandpass, order 4) and grand averages were compared. Recording of visual evoked potentials was performed as previously published (Halene et al., 2009). Visual stimuli were delivered through a flash box (PS40/R Photic Stimulator, Grass Technologies, West Warwick, RI) 30 cm above home cages with transparent cage tops. During the 15 min acclimation period and subsequent stimulus presentation, mice were entirely in the dark. Average waves were created for the response to the visual stimulus for each mouse separately. Recording sessions consisted of an acclimation phase (15 min) and subsequent data collection. Analysis was performed as described above for auditory ERPs.
EEG signal was bandpass filtered online between 1 and 500 Hz, and grand average waveforms were created from −500 ms to 1000 ms relative to the auditory stimulus. To remove movement artifacts, trials containing activity over 2 SD of the mean were rejected. Initial peak analysis was performed in Microsoft Excel (Redmond, WA) or Igor (Wavemetrics, OR) on the remaining averaged trials. The baseline was corrected at stimulus onset of S1 and S2 independently. Peak components were extracted from grand-average waveforms as follows: P1/P20 (most positive deflection between 10 and 30 msec), N1/N40 (most negative deflection between 25 and 60 msec) and P2/P80 (most positive deflection between 60 and 250 msec) in the averaged waves were analyzed by two-way analysis of variance (ANOVA). To also test for differences independent of peak identification each time point was compared to using a T-test and corrected for multiple comparisons using a non-parametric bootstrap method fully described bellow for time frequency analysis and additionally described by Carlson (2011).
Time-Frequency Analysis
Spectral decomposition of auditory-evoked response waveforms was performed using the EEGLab toolbox in Matlab (Delorme and Makeig, 2004), as published (Gandal et al., 2010). Single-trial epochs between −0.3 and 0.8 seconds relative to the first stimulus (S1) were extracted from the continuous EEG data sampled at 1667 Hz. For each epoch, total power (i.e., event-related spectral perturbation, ERSP) and phase-locking values (i.e., intertrial coherence, ITC) were calculated using Morlet wavelets in 100 linearly spaced frequency bins between 5.0 and 100 Hz, with wavelet cycles increasing from 3 (at low frequencies) to 6 (at high frequencies). Total power was calculated in decibels (dB) relative to baseline power (−200 to 0 ms) in each frequency band. Phase locking factor (PLF) is expressed as a unitless ratio between 0 and 1, where 1 represents complete phase synchrony at a given frequency and time across trials. Significance for total power and PLF differences between the group-derived time-frequency graphs was assessed by unpaired T-tests at each time-frequency bin (Carlson et al. 2011). Correction for multiple comparisons was achieved using a non-parametric permutation bootstrap method as implemented in EEGlab, using 500 permuted samples. The permutation bootstrap method involves pooling all of the single-subject 2D time-freque ncy images from both groups, randomly shuffling and partitioning the pooled samples into two “random” groups of the same size as the initial comparison. Unpaired t-tests are then conducted between these two randomly chosen groups at each time-frequency bin and the maximum T-statistic is retained. Repeating this procedure 500 hundred times generates a distribution of T-statistics. One can then compare T-statistics from the original group comparisons to this permuted bootstrap distribution, retaining any points that have P values less than 0.05. Such correction for multiple comparisons has been commonly used in the neuroimaging community (Singh et al., 2003).
Auditory and visual evoked responses produce a typical set of middle latency positive going (P) and negative going (N) components that are labeled P1-N1-P2 in humans as well as mice. In response to single stimuli spaced 8 seconds apart, the ERP responses in both MeCP2+/− and wildtype littermate mice showed this same typical pattern (Fig 1). Following the P2 there can be another wide positivity (P3) associated with novelty and often an even broader negativity sometimes described as the slow wave. These middle and late-latency events in mice are cortically generated and are thought to represent an accelerated but very similar set of potentials as found in humans (Ehrlichman et al., 2009; Umbricht et al., 2004).
Figure 1
Figure 1
Abnormal sensory evoked potentials found in MeCP2+/− mice
Increased N1 amplitude, delayed P2 latency
Short latency auditory brains stem responses (ABR) were test between unchanged, indicating that initial audition is not different between the MeCP2+/− and wild type mice (Fig 1 A, inset, at 85 db all components P>0.5; n=4). However, middle and late components of the auditory ERP were significantly different between groups (Fig 1A). In MeCP2+/− mice the N1 peak was 142% larger (WT N1 =−104.18±9.082 versus MeCP2+/− −148.15± 10.496 µV; P= 0.0068; n=8; Fig. 1C). A similarly increased N1 component was also found in response to visual stimuli (Fig 1B), suggesting the increase in N1 is a common cortical feature of the MeCP2+/− mouse (Fig 1C), and supports a general increase in cortical excitability or coherence across trials.
Following N1, the P2 component in the MeCP2+/− mice did not show an increase in amplitude (MeCP2+/− was 87% of controls at P=0.57) but did show a significantly delayed latency (MeCP2+/−: 212±5.4; WT: 135±27; P=0.019). In contrast to the auditory stimuli, increased P2 latency was not found in response to the visual stimuli (MeCP2+/−; 167±4.4 versus wildtype: 174±16 msec), demonstrating that delayed sensory processing is specific to the auditory modality.
To further examine differences in the sensory evoked responses we also directly compared each time point between groups and corrected for multiple comparisons using a non-parametric bootstrap method. This technique identified contiguous sections in time, where group differences exceeded significance at P<0.05 (see x-axis Fig 1 A, B). This analysis supported the peak-based analysis and also highlighted a reduction in the “slow wave” negativity that follows the P2/3 component. This was found in both the auditory and visual record and is seen as significant difference group amplitude between these late grand average time points (Fig 1 A, B).
Frequency differences
The above auditory and visual ERPs are well-validated clinical and preclinical measures. More recently, analysis of evoked responses in the frequency domain has provided biomarkers associated with ASD, most notably a disruption in gamma frequency activity following auditory stimuli in autistic adults, children, and parents of autistic children (Rojas et al., 2008). To test for similar differences in the MeCP2+/− mice, auditory-evoked responses were analyzed in the frequency domain for differences in power (i.e., event-related spectral perturbation, ERSP; Fig 2A) and phase-locking (i.e., intertrial coherence, ITC; Fig 2C). Examination of these plots demonstrated two primary differences in MeCP2+/− mice: an increase in induced gamma power (> 30Hz), with the most significant differences centering around 45Hz, and a similar increase in beta-band activity spanning 10 to 20 Hz (Fig 2B). Theta and delta-band components appeared unchanged between groups, suggesting that these differences where specific to the two frequency bands. A second aspect of the spectral response that is often abnormal in neuropsychiatric disorders is phase-locking — the consistency of the phase of an oscillatory response across repeated trials. This measure can provide insight into ERP amplitude changes, as elevated amplitude can reflect either increases in evoked spectral power or how coherent the phase of the spectral response `line up' the frequency components trial-to-trial. Functionally, the appropriate consistency of phase is critical to the processing of sensory information (Fontanini and Katz, 2008). MeCP+/− mice showed much higher phase-locking (Fig 2D), again in the gamma-range. This is a unique finding that is not indicative of idiopathic ASD, but may indicates unique circuit abnormalities that may play a specific role in RTT.
Figure 2
Figure 2
Auditory stimulus associated changes found in EEG gamma and beta bands
This work demonstrates that MeCP2 heterozygous for a loss of function mutation causes a number of impairments in sensory processing, as measured by auditory and visual evoked-responses. These differences include: increased N1 amplitude, increased latency of auditory P2 component, reduced suppression of induced gamma, and increased suppression of beta-band activity. We identified no differences in ABR, indicating normal initial auditory processing through the brain stem. These results are consistent with studies in ASD and RTT, demonstrating that ABR responses are preserved in most cases (Akshoomoff et al., 1989; Pelson and Budden, 1987; Pillion et al., 2003). We also found stimulus-linked low-frequency responses (e.g., theta range) to be unaffected, demonstrating that the gamma and beta-band changes are specific abnormalities not tied to gross alterations in neural synchrony. Baseline EEG activity in these mice were also very similar, but long term EEGs were not included in this study.
The N1, P2, and N400 represent independent components, yet it is easy to imagine how changes in one peak could lead to changes in the latency or amplitude of another (Luck, 2005). In particular, an increased N1 could make the P2 component appear later. In this case, our analysis is assisted by the visual ERP data. If the increased P2 latency were linked with N1 size, we would expect similar findings in the visual responses where N1 is similarly increased (Fig 1), yet there is no change in visually-evoked P2 latency. Thus, the increased latency seems to be an independent component of the auditory ERP that covaries with genotype.
We selected the bipolar recording paradigm pairing hippocampal depth and cortical surface electrodes. This electrode arrangement, unlike scalp EEGs, captures activity between the hippocampal and surface probes. While different from scalp EEG, this arrangement was chosen because similar paradigms replicate the peak/component progression found in humans and replicate abnormalities and drug effects in other neuropsychiatric disorders (Siegel et al., 2003; Umbricht et al., 2004). Specificity is achieved by repeated modality-specific external stimulation followed by time-locked average, which removes “random” EEG signals and preserves only those relevant to sensory processing for the given stimulus. Thus, despite this recording configuration's limited ability to localize response, responses from the present configuration recapitulate common temporal EEG themes that are found in the clinical ASD population, indicating similar cortical abnormalities (Jeste and Nelson, 2009; Roberts et al., 2008; Rojas et al., 2008).
Preclinical significance
The most interesting overlap with auditory evoked findings in ASD is with changes in P2 latency and gamma-band activity both of which are endophenotypes of ASD, and have been suggested as useful treatment outcome measures (Rojas et al., 2008). With its large effect size, latency may be particularly significant, as increased latency of evoked auditory correlates with the level of language impairment and developmental delay in ASD (Gandal et al., 2010; Roberts et al., 2010).
Gamma-band activity is generated across cortical structures and is modulated, and likely driven, by local interneuronal activity (Fries et al., 2007; Uhlhaas and Singer, 2010; Yizhar et al., 2011). In response to sensory stimulation, two distinct types of gamma rhythms are generated: brief phase locked (evoked) gamma occurring immediately following the initial cortical input, and a later rise in averaged gamma-activity that is not necessarily phase locked to the stimuli (induced gamma) associated with the cognitive activity (Tallon-Baudry et al., 1998). The presence of the induced gamma differences that parallel spectral features associated with ASD (Rojas et al., 2008), suggests that both ASD and RTT may share similar local circuit insults. By replicating these gamma differences, the MeCP2+/− mouse may be useful in understanding cortical abnormalities in a range of neuropsychiatric disorders where gamma-band activity is affected (Uhlhaas and Singer, 2010). Our data does not reproduce the reduction in phase-locked responses at gamma frequencies reported (i.e., “evoked gamma” deficits) in ASD (Rojas et al., 2008). This difference may relate to the increased activity during the N1 in the MeCP2+/− mice. This increased N1 is rare in idiopathic ASD, but is associated with Fragile × Syndrome, another genetic disorder conferring increased risk for ASD, mental retardation, and epilepsy (Berry-Kravis, 2002; Castren et al., 2003). Indeed, there have been reports of elevated evoked cortical activity in RTT (Yoshikawa et al., 1991). Thus while the non-phase locked (e.g., “induced”) gamma power deficits are likely a marker for cortical auditory processing abnormalities, the increased N1 may be a marker for a subset of ASD more strongly associated with epilepsy (Pillion and Naidu, 2000). Interestingly, an increased N1, a late positivity following P2 (a P3), and increased induced-gamma within a subject is a classic pattern of novelty detection. A failure to respond appropriately to novelty may be a common feature of ASD (Gomot et al., 2006). This possibility indicates the need for further work determining whether these mice have abnormal abilities to identify or orientate to novelty, and thus provide additional links to other ASD-associated deficits. At the mechanistic level further understanding of the circuit disorders underlying these ERP and ERSP changes should be investigated.
The late component, which was also found to be different between groups, has not been extensively studied in mice. Yet, this negativity may be analogous in sequence to the N400, which in humans is associated with higher-order cognitive activity that appears to include sources in the sensory cortex (Verbaten et al., 1991). Across several cognitive paradigms, changes in the N400 response are reduced in ASD (Fishman et al., 2010). As these mice were not actively engaged in a cognitive task during recording, we can only speculate that similar circuitry may be disrupted in the MeCP2+/− mice.
These findings are congruent with growing indications that epigenetic mechanisms, such as those associated with MeCP2 and likely in ASD are associated with sensory processing abnormalities. (Schanen, 2006). These results also demonstrate a potential mechanistic link between specific autism-like features of RTT and idiopathic ASD in the auditory processing domain. If this is the case, the MeCP2+/− mouse provides unique access to RTT- and ASD-associated changes in local cortical circuit function. These findings suggest that MeCP2+/− mice may find a role similar to other validated models of non-idiopathic disorders with ASD phenotypes, such as Fragile × Syndrome (Yan et al., 2005) and Fetal Valproate Syndrome (Gandal et al., 2010; Schneider and Przewlocki, 2005; Wagner et al., 2006)). Thus, auditory and other sensory ERPs in the MeCP2+/− mouse provide unique opportunities to test potential therapeutics in treating and understanding pathphysiologies of RTT and those features shared in ASD and other neuropsychiatric disorders.
Highlights
  • > 
    We assayed sensory evoked potentials in a female mouse model of Rett syndrome.
  • > 
    Auditory evoked potentials demonstrated increased N1 and increased P2 latency.
  • > 
    Specific abnormalities were also found in gamma-band responses.
  • > 
    These novel findings demonstrated a selective reproduction of ASD biomarkers.
Footnotes
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
  • Akshoomoff N, et al. Brainstem auditory evoked potentials in receptive developmental language disorder. Brain Lang. 1989;37:409–18. [PubMed]
  • Bader GG, et al. Neurophysiological findings in the Rett syndrome, II: Visual and auditory brainstem, middle and late evoked responses. Brain Dev. 1989;11:110–4. [PubMed]
  • Badr GG, et al. Brain stem and spinal cord impairment in Rett syndrome: somatosensory and auditory evoked responses investigations. Brain Dev. 1987;9:517–22. [PubMed]
  • Berry-Kravis E. Epilepsy in fragile × syndrome. Dev Med Child Neurol. 2002;44:724–8. [PubMed]
  • Carlson GC, et al. From the Cover: Dysbindin-1 mutant mice implicate reduced fast-phasic inhibition as a final common disease mechanism in schizophrenia. Proc Natl Acad Sci U S A. 2011;108:E962–70. [PubMed]
  • Castren M, et al. Augmentation of auditory N1 in children with fragile × syndrome. Brain Topogr. 2003;15:165–71. [PubMed]
  • Connolly PM, et al. Inhibition of auditory evoked potentials and prepulse inhibition of startle in DBA/2J and DBA/2Hsd inbred mouse substrains. Brain Res. 2003;992:85–95. [PubMed]
  • Delorme A, Makeig S. EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. J Neurosci Methods. 2004;134:9–21. [PubMed]
  • Ehrlichman RS, et al. N-methyl-d-aspartic acid receptor antagonist-induced frequency oscillations in mice recreate pattern of electrophysiological deficits in schizophrenia. Neuroscience. 2009;158:705–12. [PubMed]
  • Fishman I, et al. Contrasting patterns of language-associated brain activity in autism and Williams syndrome. Soc Cogn Affect Neurosci. 2010 [PMC free article] [PubMed]
  • Fontanini A, Katz DB. Behavioral states, network states, and sensory response variability. J Neurophysiol. 2008;100:1160–8. [PubMed]
  • Fries P, et al. The gamma cycle. Trends Neurosci. 2007;30:309–16. [PubMed]
  • Gandal MJ, et al. Validating gamma Oscillations and Delayed Auditory Responses as Translational Biomarkers of Autism. Biol Psychiatry. 2010;68:1100–6. [PubMed]
  • Gomot M, et al. Change detection in children with autism: an auditory event-related fMRI study. Neuroimage. 2006;29:475–84. [PubMed]
  • Guy J, et al. A mouse Mecp2-null mutation causes neurological symptoms that mimic Rett syndrome. Nat Genet. 2001;27:322–6. [PubMed]
  • Halene TB, et al. Assessment of NMDA receptor NR1 subunit hypofunction in mice as a model for schizophrenia. Genes Brain Behav. 2009;8:661–75. [PMC free article] [PubMed]
  • Jeste SS, Nelson CA., 3rd Event related potentials in the understanding of autism spectrum disorders: an analytical review. J Autism Dev Disord. 2009;39:495–510. [PubMed]
  • Kalmanchey R. Evoked potentials in the Rett syndrome. Brain Dev. 1990;12:73–6. [PubMed]
  • Luck SJ. Ten simple rules for designing ERP experiments. In: Handy TC, editor. Event-related potentials : a methods handbook. MIT press; Cambridge, MA: 2005. p. 404.
  • Oram Cardy JE, et al. Auditory evoked fields predict language ability and impairment in children. Int J Psychophysiol. 2008;68:170–5. [PubMed]
  • Pelson RO, Budden SS. Auditory brainstem response findings in Rett syndrome. Brain Dev. 1987;9:514–6. [PubMed]
  • Pillion JP, Naidu S. Auditory brainstem response findings in Rett syndrome: stability over time. J Pediatr. 2000;137:393–6. [PubMed]
  • Pillion JP, et al. Prevalence of hearing loss in Rett syndrome. Dev Med Child Neurol. 2003;45:338–43. [PubMed]
  • Roberts TP, et al. Auditory magnetic mismatch field latency: a biomarker for language impairment in autism. Biol Psychiatry. 2011;70:263–9. [PMC free article] [PubMed]
  • Roberts TP, et al. MEG detection of delayed auditory evoked responses in autism spectrum disorders: towards an imaging biomarker for autism. Autism Res. 2010;3:8–18. [PMC free article] [PubMed]
  • Roberts TP, et al. Electrophysiological signatures: magnetoencephalographic studies of the neural correlates of language impairment in autism spectrum disorders. Int J Psychophysiol. 2008;68:149–60. [PMC free article] [PubMed]
  • Rojas DC, et al. Reduced neural synchronization of gamma-band MEG oscillations in first-degree relatives of children with autism. BMC Psychiatry. 2008;8:66. [PMC free article] [PubMed]
  • Samaco RC, et al. Epigenetic overlap in autism-spectrum neurodevelopmental disorders: MECP2 deficiency causes reduced expression of UBE3A and GABRB3. Hum Mol Genet. 2005;14:483–92. [PMC free article] [PubMed]
  • Samaco RC, et al. Multiple pathways regulate MeCP2 expression in normal brain development and exhibit defects in autism-spectrum disorders. Hum Mol Genet. 2004;13:629–39. [PubMed]
  • Schanen NC. Epigenetics of autism spectrum disorders. Hum Mol Genet. 2006;15(Spec No 2):R138–50. [PubMed]
  • Schneider T, Przewlocki R. Behavioral alterations in rats prenatally exposed to valproic acid: animal model of autism. Neuropsychopharmacology. 2005;30:80–9. [PubMed]
  • Siegel SJ, et al. Effects of strain, novelty, and NMDA blockade on auditory-evoked potentials in mice. Neuropsychopharmacology. 2003;28:675–82. [PubMed]
  • Singh KD, et al. Group imaging of task-related changes in cortical synchronisation using nonparametric permutation testing. Neuroimage. 2003;19:1589–601. [PubMed]
  • Stach BA, et al. Auditory evoked potentials in Rett syndrome. J Am Acad Audiol. 1994;5:226–30. [PubMed]
  • Stearns NA, et al. Behavioral and anatomical abnormalities in Mecp2 mutant mice: a model for Rett syndrome. Neuroscience. 2007;146:907–21. [PubMed]
  • Tallon-Baudry C, et al. Induced gamma-band activity during the delay of a visual short-term memory task in humans. J Neurosci. 1998;18:4244–54. [PubMed]
  • Uhlhaas PJ, Singer W. Abnormal neural oscillations and synchrony in schizophrenia. Nat Rev Neurosci. 2010;11:100–13. [PubMed]
  • Umbricht D, et al. Midlatency auditory event-related potentials in mice: comparison to midlatency auditory ERPs in humans. Brain Res. 2004;1019:189–200. [PubMed]
  • Verbaten MN, et al. Abnormal visual event-related potentials of autistic children. J Autism Dev Disord. 1991;21:449–70. [PubMed]
  • Wagner GC, et al. A new neurobehavioral model of autism in mice: pre- and postnatal exposure to sodium valproate. J Autism Dev Disord. 2006;36:779–93. [PubMed]
  • Wilson TW, et al. Children and adolescents with autism exhibit reduced MEG steady-state gamma responses. Biol Psychiatry. 2007;62:192–7. [PMC free article] [PubMed]
  • Yamanouchi H, et al. Abnormal cortical excitability in Rett syndrome. Pediatr Neurol. 1993;9:202–6. [PubMed]
  • Yan QJ, et al. Suppression of two major Fragile × Syndrome mouse model phenotypes by the mGluR5 antagonist MPEP. Neuropharmacology. 2005;49:1053–66. [PubMed]
  • Yizhar O, et al. Neocortical excitation/inhibition balance in information processing and social dysfunction. Nature. 2011 [PubMed]
  • Yoshikawa H, et al. Giant somatosensory evoked potentials in the Rett syndrome. Brain Dev. 1991;13:36–9. [PubMed]
  • Young DJ, et al. The diagnosis of autism in a female: could it be Rett syndrome? Eur J Pediatr. 2007 [PubMed]