Abstract

Rett syndrome (RTT) is a neurodevelopmental disorder caused by mutations in MECP2, encoding methyl-CpG-binding protein 2 (MeCP2). Although MECP2 is ubiquitously transcribed, MeCP2 expression is developmentally regulated and heterogeneous in neuronal subpopulations, defined as MeCP2lo and MeCP2hi. To test the hypothesis that pathways affecting MeCP2 expression changes may be defective in RTT, autism and other neurodevelopmental disorders without MECP2 mutations, a high-throughput quantitation of MeCP2 expression was performed on a tissue microarray containing frontal cortex samples from 28 different patients with neurodevelopmental disorders and age-matched controls. Combined quantitative analyses of MeCP2 protein and alternatively polyadenylated transcript levels were performed by laser scanning cytometry and tested for significant differences from age-matched controls. Normal cerebral samples showed an increase in total MeCP2 expression and the percentage of MeCP2hi cells with age that could be explained by increased MECP2 transcription within the MeCP2hi population. A significant decrease in the relative usage of the long transcript in the MeCP2lo population was observed in postnatal compared to fetal brain, but alternate polyadenylation did not correlate with MeCP2 expression changes at the single cell level. Brain samples from several related neurodevelopmental disorders, including autism, pervasive developmental disorder, Prader–Willi and Angelman syndromes showed significant differences in MeCP2 expression from age-matched controls by apparently different transcriptional and post-transcriptional mechanisms. These results suggest that multiple pathways regulate the complex developmental expression of MeCP2 and are defective in autism-spectrum disorders in addition to RTT.

INTRODUCTION

Rett syndrome (RTT) is an X-linked dominant neurodevelopmental disorder and a leading cause of mental retardation and autistic behavior in females (1,2). Linkage analysis mapped the genetic defect in RTT to Xq28 (3) and mutations were identified in the gene MECP2 (4). MECP2 encodes the methylated CpG-binding protein 2 (MeCP2), a protein that binds to methylated CpGs and a putative transcriptional repressor of downstream genes (5,6). Subsequent studies have identified MECP2 coding region mutations in approximately 80% of sporadic RTT patients (712). Although X-linked dominant disorders such as RTT were originally thought to be incompatible with development in males (2), several recent reports have demonstrated MECP2 mutations in males with Rett syndrome, Angelman syndrome, X-linked mental retardation or severe neonatal encephalopathy (1317). Two recent studies suggest that the reason RTT is more frequent in females is not due to male lethality but because MECP2 mutations occur more frequently on the paternal X chromosome (18,19). These findings suggest that the scope of diseases associated with MECP2 mutations may extend beyond RTT to other forms of X-linked mental retardation and severe neurodevelopmental disorders.

Autistic disorder shares several characteristics in common with RTT, including a loss of social, cognitive and language skills in the first 3 years, coupled to a gain in repetitive stereotyped behavior (20,21). A higher incidence in males to females is observed in autistic disorder, suggesting X-linked inheritance. Mutations in two X-linked genes encoding neuroligins NLGN3 and NLGN4 have recently been associated with autism (22). Although a few cases of infantile autism with an MECP2 mutation have been described (2325), three large studies have determined that mutations or polymorphisms in the coding region of MECP2 occur at a low frequency in autistic patients (24,26,27).

Recent studies on normal human (2830), primate (31) and rodent (30,32) brain samples have revealed a complex developmental regulation of MeCP2 at the protein level in brain, suggesting that other types of mutations may affect normal MeCP2 expression. Cellular heterogeneity in MeCP2 expression was observed in normal brain with a subpopulation of cells exhibiting high expression (MeCP2hi) and the remainder exhibiting low expression (MeCP2lo) (28). In the cerebral cortex, the MeCP2hi phenotype is limited to neuronal cells, but neurons are heterogeneous for both the MeCP2lo and MeCP2hi phenotypes. Elevated MeCP2 expression and the MeCP2hi neuronal phenotype is acquired with increasing postnatal age, potentially explaining the delay of phenotype in RTT (29,30). By northern blot analysis, MECP2 is ubiquitously expressed in fetal and adult brain as 10 and 1.9 kb transcripts (30,3335). The 10 kb transcript includes the 1.9 kb coding region of MECP2 as well as an exceptionally long 8.5 kb 3′-untranslated region (UTR) produced by alternate polyadenylation (34,35). A decrease in usage of the long transcript was observed with age, inversely correlating with increased MeCP2 expression (29). A direct association between MeCP2 protein expression and alternate transcript use has not been previously performed, however, and is important to determine transcriptional and post-transcriptional mechanisms of MeCP2 tissue-specific and developmental regulation.

In this report, we performed combined analyses of MeCP2 protein and each alternative MECP2 transcript within individual cells of the human cerebral cortex. Analysis of a tissue microarray containing multiple neurodevelopmental disorder and control cerebral samples by laser scanning cytometry (LSC) allowed a unique approach to detecting and correlating multiple fluorescent probes and statistically quantitating MeCP2/MECP2 expression. Using this approach, we demonstrate significantly higher levels of both MECP2 transcripts in MeCP2hi neurons that may explain the acquired elevated protein expression in this subpopulation. In addition, significant defects in MeCP2 protein expression were observed in several neurodevelopmental brain samples that could be explained by different transcriptional or post-transcriptional mechanisms. The results demonstrate the feasibility of this novel approach to understand normal developmental expression of MeCP2 and to identify potential regulatory pathways deficient in related neurodevelopmental disorders.

RESULTS

Quantitation of MeCP2 immunofluorescence on a tissue microarray containing human neurodevelopmental disorders and control cerebral samples

In order to simultaneously analyze multiple human tissue samples for MeCP2 expression, a tissue microarray containing 12 normal control (three fetal, three infant, three juvenile, three adult) six RTT, five autism (AUT), three Prader–Willi syndrome (PWS), one Angelman syndrome (AS) and one pervasive developmental disorder (PDD) cerebrum samples was constructed. Samples were obtained from the frontal cortex, Brodman area 9 (BA9) and cores were extracted from layers III–V of the cerebral cortex from the individuals described in Table 1. Slides containing the tissue microarray were immunostained with anti-MeCP2, non-specific rabbit IgG as a negative control, or anti-histone H1 as a positive control, and Cy5 (long red) fluorescent detection. LSC quantitation and analysis of anti-MeCP2 immunofluorescence is shown in Figure 1 for a representative experiment. Figure 1A shows a representative LSC scattergram of x,y slide positions of cells and tissue cores, with each individual in triplicate and labeled according to sample category. Samples from each age category of control samples were individually gated and the histograms of anti-MeCP2 fluorescence are shown in Figure 1B. All tissues show positive expression of MeCP2 compared with the immunoglobulin G (IgG) control (black overlay) and are colored green for MeCP2lo and red for MeCP2hi according to the control adult samples. An increase in MeCP2 mean and MeCP2hi population with increasing age is observed, as reported previously (29).

Triplicate samples from each individual with a neurodevelopmental disorder were also independently gated and compared with age-matched controls for differences in mean MeCP2 expression or %MeCP2hi cells. Figure 1C shows representative histograms of RTT and AUT samples showing abnormal MeCP2 expression in the cerebral cortex compared with age-matched controls (blue overlay). Table 2 summarizes the results from 17 different experiments for MeCP2 expression and three for positive control histone H1. For MeCP2 expression, 10 replicates were performed with an N-terminal reactive antibody (Affinity) and seven with two different C-terminal specific antibodies, Affinity (five replicates) and Upstate (two replicates). All antibodies were determined to be specific to MeCP2 based on lack of immunoreactivity in Mecp2tm1.1Bird/y brain samples by LSC and immunoblot (Braunschweig et al., unpublished results). In addition, the C-terminal reactive antibodies showed lack of immunoreactivity (defined as completely overlapping with the IgG control histogram) in the male RTT 1238 sample with a MECP2 truncation mutation (Braunschweig et al., unpublished data) and in 23–46% of cells in samples from two RTT patients with MECP2 truncation mutations (28). Results from N- and C-terminal specific antibodies were analyzed separately because of inherent differences in MeCP2 histograms by LSC and subtle size differences by immunoblot, suggesting that they may recognize different post-translationally modified forms of MeCP2 (data not shown). Histone H1 was chosen as a control for normal variations between samples because it was shown to have the lowest variation between samples (19%) after analysis of three different control antibodies (GAPDH, 24%; transferrin receptor, 37%). In addition, histone H1 has been previously used as a control for MeCP2 in LSC analyses because it is nuclear and has similar DNA binding properties as MeCP2 (28,29).

For each patient sample, the mean MeCP2 and percentage MeCP2hi values were compared back to the three closest age-matched controls and tested for significance by t-test (asterisks). Three of the samples (PWS 970, AUT 1315 and RTT 1420) were obtained from fixed rather than frozen tissue and showed significant deficiencies in control histone H1 immunofluorescence, making them unsuitable for MeCP2 expression studies (ND). Of the remaining samples, all neurodevelopmental disorders included on the tissue array showed either lower or higher levels of MeCP2 immunofluorescence compared with controls. Six individuals including three RTT patients with MECP2 truncation mutations (RTT 3393, Fig. 1C), the AS sample, and two of the four autism samples (AUT 797, AUT 967, Fig. 1C) showed significant deficiencies in both MeCP2 mean expression and the percentage of MeCP2hi cells using both anti-MeCP2 antibodies. Unexpectedly, significant increases in both mean MeCP2 and percentage MeCP2hi were also observed in the PDD sample and one autism sample (AUT 732) compared with age-matched controls using both antibodies. The remaining neurodevelopmental disorder samples, including both PWS samples, AUT 1174, and the two RTT without detectable MECP2 mutations showed significant deficiencies in MeCP2 expression using the C-terminal, but not the N-terminal specific antibodies.

Combined quantitative analyses of MeCP2 immunofluorescence and MECP2 alternate transcripts by fluorescence in situ hybridization

Previous studies comparing results of northern and western blots of different human and mouse tissues for MECP2/Mecp2 expression found no obvious correlation between protein and RNA levels (30). Furthermore, our previous in situ analysis comparing MeCP2 immunofluorescence with the long MECP2 transcript suggested an inverse correlation between protein level and alternative long 3′-UTR usage (29). In order to directly investigate whether a transcriptional or post-transcriptional mechanism could explain the developmental change from MeCP2lo to MeCP2hi phenotype in individual neurons, we developed a combined staining approach for MeCP2 immunofluorescence and RNA fluorescence in situ hybridization (RNA-FISH) using two different riboprobes: a 1.5 kb probe from the coding region (CDS) and a 2.4 kb probe for the alternate long transcript (3′-UTR) of MECP2. Serial sections of the tissue array shown in Figure 1A were hybridized with riboprobes of either the MECP2 antisense or sense strand control and detected with FITC, then stained with anti-MeCP2 and detected with Cy5. Analysis of MeCP2 protein was performed as in Figure 1B and C for each control age category and individual patient sample, using colored gating to separately analyze MeCP2lo and MeCP2hi populations. The LSC analysis of total MECP2 RNA expression in control samples is shown in Figure 2A and C for CDS and 3′-UTR riboprobes, respectively. In all cases, the antisense fluorescence in the MeCP2lo (green), MeCP2hi (red), or total population (black) showed a significant shift to the right of the sense control probe (blue), demonstrating positive expression. For infant, juvenile and adult samples containing a detectable MeCP2hi population, the red population showed significantly higher levels of MECP2 CDS fluorescence compared with the green population, suggesting higher transcription of MECP2 in the MeCP2hi compared to the MeCP2lo population (Fig. 2A). Similarly, the MECP23′-UTR fluorescence levels were higher for the MeCP2hi than the MeCP2lo population for juvenile and adult samples (Fig. 2C), suggesting higher transcription of both long and short MECP2 transcripts in the MeCP2hi population.

As positive controls for RNA quality, serial sections of the tissue microarray were hybridized with an ACTIN or TFRC antisense riboprobes in place of MECP2 riboprobes for a combined analysis with MeCP2 immunofluorescence. TFRC was chosen for normalization of MECP2 values as it showed a lower variability between samples compared with ACTIN (15% and 20%, respectively). The mean fluorescence values shown in Figure 2 were normalized to TFRC, and the ratios are shown in Table 3. The MECP2/TFRC normalized results are also graphed in Figure 3 to show the developmental changes in MECP2 expression in the normal control samples. Figure 3A demonstrates significantly higher levels of total MECP2 transcript in the MeCP2hi than the MeCP2lo population for infant, juvenile and adult cerebral cortex, but MECP2 transcript levels did not change significantly in the MeCP2lo population with increasing age. The MECP2 long transcript was also significantly higher in the MeCP2hi population for juvenile and adult samples (Fig. 3B). Lastly, the ratio of long transcript over total MECP2 is graphed for the control age categories in Figure 3C, demonstrating a significant decrease in the relative usage of long MECP2 transcript in juvenile and adult samples compared with fetal. In contrast to total transcript levels, no significant differences in MECP2 long/total transcript ratio were observed between MeCP2lo and MeCP2hi populations. These results demonstrate that the age-related changes in MeCP2 expression could be explained by changes in transcription within the MeCP2hi population. In contrast, the age-related changes in long MECP2 transcript usage do not appear to be directly related to protein expression.

Neurodevelopmental disorders demonstrate different types of abnormalities in MeCP2/MECP2 expression

To determine if transcriptional and/or post-transcriptional defects could explain the MeCP2 protein deficiencies in the neurodevelopmental disorders demonstrated in Figure 1 and Table 2, the neurodevelopmental disorder samples were compared with age-matched controls in the combined analysis of MeCP2 protein and alternative MECP2 transcripts. Figure 2B shows histogram analyses of MECP2 CDS transcript levels in representative patient samples showing significant decreases in MeCP2 protein expression from Figure 1C and Table 2, with the inclusion of an MECP2 CDS antisense control from age-matched control samples (orange overlay). The two RTT samples showed relatively normal levels of both total and long MECP2 transcripts that were not significantly different from age-matched controls after TFRC-normalization (Table 3). In contrast, one autism sample with MeCP2 protein deficiency (AUT 797, Fig. 1C and Table 2) had significantly lower levels of both MECP2 CDS and 3′-UTR antisense fluorescence compared with age-matched controls [compare green and orange histograms (Fig. 2B and D), and TFRC-normalized results, (Table 3)]. Another autism sample (AUT 967) had slightly lower levels of MECP2 CDS fluorescence by LSC histogram analysis (Fig. 2B) that was not significantly different than age-matched controls following TFRC-normalization (Table 3). The normalized MECP2/TFRC, 3′-UTR/TFRC and long/total MECP2 ratio results for all samples are shown in Table 3. Unlike the protein analyses in Table 2, fewer neurodevelopmental disorders showed significant differences in MECP2 transcript levels from age-matched controls. AUT 797 showed significantly lower levels of MECP2 CDS, 3′-UTR, and ratio of long/total MECP2 only within the MeCP2hi population. Unexpectedly, PWS and AS samples showed significantly higher MECP2 levels (only 3′-UTR level for AS) in the MeCP2hi population (Table 3), despite deficiencies at the protein level (Table 2).

For simplicity in showing the types of abnormalities observed in the neurodevelopmental disorders, the significant results of Tables 2 and 3 are summarized in Figure 4. The neurodevelopmental disorders showing differences in MeCP2/MECP2 expression are categorized by the type of abnormalities observed. Figure 4A includes the three RTT samples (one male and two female) with MECP2 truncation mutations that showed significant deficiencies in MeCP2 protein expression and the MeCP2hi population. The non-RTT brain sample with the most conclusive abnormality in MeCP2 expression from these combined analyses was AUT 797, which showed significantly lower levels of MeCP2 protein and both MECP2 transcripts only in the MeCP2hi population (Fig. 4B). These combined results suggest a defect in transcriptional regulation of MECP2 during the progression to the MeCP2hi phenotype in this individual. An unexpected finding was observed for PWS 865 and 1290 (Fig. 4C), which showed significantly lower MeCP2 expression with the C-terminal anti-MeCP2, but higher levels of both MECP2 transcripts.

The last four categories of neurodevelopmental disorders showed differences in MeCP2 protein expression without corresponding differences in MECP2 transcript levels (Fig. 4D–G). Two samples, AUT 967 and AS 293, had significant defects in mean MeCP2 and percentage MeCP2hi for both antibodies as well as a higher ratio of long to total MECP2 transcript usage in the MeCP2hi population. These results would support the original hypothesis that the short form of MECP2 is translated more efficiently than the long form (29), so an increase in the relative use of the long MECP2 transcript may result in lower MeCP2 expression. The next category (Fig. 4E) is more difficult to explain because these two samples (PDD 144 and AUT 732) showed higher than expected levels of MeCP2, but the only transcriptional defect was a significant difference in the ratio of MECP2 long/total transcripts between the MeCP2lo and MeCP2hi populations that was not observed in controls (Fig. 3C). In addition, a RTT sample (RTT 3381) that showed a lower percentage MeCP2hi population than age-matched controls with both antibodies also showed a significant difference in the ratio of MECP2 long/total transcripts between the MeCP2lo and MeCP2hi populations because of a lower ratio in the MeCP2lo population. Lastly, the two remaining samples RTT 4321 and AUT 1174 only showed MeCP2 protein deficiency with the C-terminal specific antibody, perhaps suggesting a post-translational modification defect. Although these latter analyses are not conclusive in determining the precise defects in MeCP2 expression for each individual, they suggest that different transcriptional and post-transcriptional mechanisms are involved in regulating normal MeCP2 expression.

DISCUSSION

Several recent studies have demonstrated that MeCP2 expression is complex and developmentally regulated (2830,34). The central challenge to further investigating the developmental changes at the transcriptional and post-transcriptional levels is that standard molecular analyses rely on extracts from whole brains or brain regions and do not account for differences in cellular subpopulations within the brain. In this report, we have performed a quantitative approach for combined analyses of MeCP2 protein and alternate RNA transcripts in individual cells of post-mortem human brain sections. The novel high throughput approach of combining multiple samples on a tissue microarray followed by simultaneous quantitation of multiple fluorescent probes by LSC has revealed a complex relationship between MECP2 transcription, alternate polyadenylation, and protein expression during normal human brain development as well as identified defects in the normal developmental expression of MeCP2 in several neurodevelopmental disorders.

During normal brain development, elevated MeCP2 expression is acquired in individual neurons of the postnatal brain, resulting in a progressive increase in the MeCP2hi population from infancy to adulthood (Fig. 1B), consistent with our previous report (29). Because the addition of a violet laser to the LSC allowed for the first time a combined analysis of MeCP2 immunofluorescence and RNA-FISH with MECP2 riboprobes, we now demonstrate a correlation of MeCP2 protein and transcript within the MeCP2lo and MeCP2hi subpopulations. The correlation between transcript and protein has not been previously apparent by comparing multiple tissue northern and western blots (30) because of the complications of tissue specific differences in alternate polyadenylation, developmental dynamics, and cellular heterogeneity. In addition, quantitative PCR analyses of brain samples cannot discriminate between MeCP2lo and MeCP2hi subpopulations, but do show slight increases in both MECP2 transcripts and a decrease in long/total MECP2 ratio with increasing age (data not shown).

Because of the significant decrease in the relative usage of the alternative long 3′-UTR with increasing age, we originally predicted that the MeCP2hi population would show decreased usage of the long MECP2 transcript compared to the MeCP2lo population (29). This original hypothesis was disproved by our current study, as the MeCP2hi population exhibited a greater amount of MECP2 long transcript compared with the MeCP2lo population in juvenile and adult controls. These results suggest that the developmental decrease in MECP2 long 3′-UTR usage is independent from the transcriptional increase in total MECP2 expression that appears to drive the transition from MeCP2lo to MeCP2hi phenotype in maturing neurons. This interpretation is consistent with the observance of a developmental decrease in the MECP2 long transcript in non-CNS tissues that do not exhibit the MeCP2hi population (29,34). In addition, our results are consistent with the previous northern blot data showing increased usage of the MECP2 long transcript in fetal compared with adult tissues (34) as well as in adult brain compared with other adult tissues (34,35). Perhaps the MECP2 long transcript has a unique function that is required both in early development and later again in mature neurons. Our results and those of others, however, argue against a simple role for the long 3′-UTR of MECP2 in regulating the level of protein expression.

The addition of several brain samples from neurodevelopmental disorders, including RTT and autism, allowed the simultaneous quantitative analysis of the level of MeCP2 protein and alternative transcripts in these patient samples compared with age-matched controls. Despite the use of an N-terminal reactive antibody that detects the truncated protein product of the MECP2 R168X and 1154del32 mutations, significant deficiencies in MeCP2 protein expression were observed in one male and two female RTT patients with all anti-MeCP2 antibodies. No differences in MECP2 transcript levels were detected in these individuals, except for the lack of a MeCP2hi population in RTT 1238 and 3393. These results suggest that stability, translation, or localization of MeCP2 may be affected by MECP2 mutation.

Two of the RTT brain samples included in our tissue microarray had no detectable mutation in the coding region of MECP2 by direct sequencing (data not shown). Both samples showed MeCP2 expression differences with C-terminal specific anti-MeCP2, and RTT 3381 showed defects in the ratio of long to total MECP2 transcripts. This is the first description to our knowledge of a deficiency in MeCP2 protein expression in non-MECP2 mutation RTT patients. Further understanding of additional genes or proteins involved in the regulation of MECP2 transcription, alternate polyadenylation and post-translational modification may identify additional candidate genes to screen in the absence of MECP2 mutations.

In addition to the potentially more predictable deficiencies in MeCP2 expression in RTT samples, a significant decrease in MeCP2 expression was also observed in three of four autism samples, one AS and two PWS samples. In addition, significant increases in MeCP2 expression were observed in one autism and one PDD sample, suggesting that both decreased and increased expression of MeCP2 may be associated with abnormal neurodevelopmental phenotypes. Unexpectedly, some neurodevelopmental disorder samples showed discrepant results when analyzed with either the N- or C-terminal specific anti-MeCP2 antibodies. As MeCP2 expression was generally lower than controls for these samples analyzed for MeCP2 N-terminus, the discrepancy between the antibodies could be simply due to these values not reaching significance. Alternatively, the different epitopes of MeCP2 may reflect differences in post-translational modifications.

To our knowledge, this is the first description of significant abnormalities in MeCP2 expression in brain samples from autism-spectrum disorders. As both RTT and autism are characterized by abnormalities in early brain growth (21,36,37), perhaps dysregulation of the normal developmental pattern of MeCP2 expression may be either a primary or secondary defect in autism. The detection of abnormal MeCP2 levels in all of the neurodevelopmental disorder samples suggests that dysregulation of MeCP2 expression may be a downstream consequence of many mental retardation syndromes. One major limitation to these studies is the small numbers of postmortem brain samples currently available for each neurodevelopmental disorder. Our pilot studies on a small number of samples suggest that additional brain samples from patients with autism, autism-spectrum disorders, and mental retardation syndromes without autism should be tested in the future for abnormalities in MeCP2 protein expression.

The novel approach used in this study to quantitate MeCP2 protein and transcript levels of control and patient samples on a tissue microarray should be amenable to larger studies including hundreds of patient samples. This report shows the proof-of-principle that human postmortem brain samples can be used for quantitative protein and RNA studies with appropriate controls. The few fixed tissue samples in our microarray did not allow proper quantitation of MeCP2, as control histone H1 levels were also significantly lower than frozen age-matched controls (Table 2). Good quality of nuclear protein and cellular RNA as assessed by histone H1 and TRFC, respectively, was obtained in all control and patient brain samples obtained frozen and with PMIs up to 30 h. Variability between RNA quality in human postmortem brain samples was controlled in our study by normalization to TFRC control hybridizations. Therefore the majority of patient and control brain samples obtained fresh frozen up to 30 h postmortem are expected to reveal informative data using the approach outlined in this report.

Although MeCP2 has been implicated in the etiology of autism-spectrum disorders, mutations in the coding region of MECP2 are not found as a frequent cause of autism (24,26,27). Here, we demonstrate that multiple pathways regulating the transcription and alternate polyadenylation of MECP2 are involved in the complex developmental expression of MeCP2 during normal postnatal human brain development. Multiple genes or pathways may therefore be deficient in different autism cases and may act to modify MeCP2 expression in cis or trans. By identifying autism cases with defects in MeCP2 expression by the approach outlined here, DNA samples may then be screened for potential mutations in regulatory regions of MECP2. Potential cis mutations may include the promoter region in individuals with transcriptional defects, or the long 3′-UTR in those with defects in the ratio of long to total MECP2 transcripts. In addition, gaining a better understanding of the molecular pathways regulating MECP2 transcriptional, polyadenylation, and post-translational changes during normal development will help identify potential new candidate genes that act to regulate MeCP2 expression in trans. As autism is expected to be a complex multigenic disorder with environmental influences, further understanding of the role for MeCP2 in the epigenetic control of postnatal neurodevelopment is expected to be beneficial.

MATERIALS AND METHODS

Tissue samples and microarray construction

Frozen frontal cerebral cortex samples (BA 9) were obtained from multiple brain bank sources from human cadavers who died of non-neurologic causes and were obtained less than 30 h post-mortem. Information regarding cause of death is provided in Table 1. Tissue was fixed in 10% formalin overnight then embedded in paraffin. A 5 µm section was stained with hematoxalin and eosin (H&E) and examined for tissue histochemistry. Regions containing cerebral cortical layers III–V were circled to guide sampling from corresponding tissue block. Triplicate 600 µm diameter cores were removed from paraffin-embedded tissue blocks using a Beecher Instruments tissue microarrayer and inserted into a new paraffin block according to the array plan in Figure 1A. Serial 5 µm sections were made from the tissue microarray paraffin block. Representative sections of the tissue microarray were stained with H&E to examine for completeness and tissue sampling, while the remaining unstained sections were baked overnight at 56°C. Methods for tissue section deparaffinization and epitope exposure were performed as described previously (28).

Immunofluorescence and LSC

Anti-MeCP2 N-terminus and C-terminus (Affinity Bioreagents, Golden, CO, USA), anti-histone H1, anti-MeCP2 C-terminus and mouse IgG (Upstate Biotechnology, Lake Placid, NY, USA) were each diluted 1/100 in IF stain buffer (PBS/1% FCS/0.5% Tween-20) and incubated on slides for 2 h, followed by three washes in PBS/0.5% Tween. Cy5-labeled secondary antibodies to rabbit or mouse IgG (Jackson Immunochemicals, West Grove, PA, USA) or an Oregon Green labeled anti-mouse IgG (Molecular Probes, Eugene, OR, USA) were diluted 1/100 in IF stain buffer containing 10 µg/ml RNAse and 10 µg/ml PI incubated on slides for 1 h at 37°C, followed by three washes in PBS/0.5% Tween. Slides were mounted in a 50% glycerol solution containing 10 µg/ml PI, coverslipped and sealed with nail polish. Analysis of immunofluorescence by LSC and digital deconvolution microscopy was performed as described previously (28).

Combined RNA FISH and immunofluorescence

A 2.4 kb probe specific to the MECP2 long 3′UTR and a 1.5 kb probe specific to the MECP2 coding region were RT–PCR amplified from normal lymphocyte cDNA, cloned into pGEM T-Easy vector (Stratagene) and sequenced. Biotin- and digoxygenin-labeled sense and anti-sense probes were synthesized commercially (Lofstrand Laboratories). Control TFRC antisense (GeneDetect.com) and ACTIN antisense (Roche) were obtained commercially. Slides containing sections of the tissue microarray were treated for deparaffinization and epitope exposure as described previously (28), post-fixed in Histochoice for 1.5 h, then washed 5 min in 1× PBS. Following post-fixation, slides were dehydrated for 5 min each in 50, 70 and 90% ethanol, then air-dried at 50°C. Dried slides were incubated with 1× hybridization solution (Sigma) containing 50% deionized formamide/10% dextran sulfate, 0.1 µg/ml tRNA, and 30 µg/ml riboprobe overnight at 50°C for the MECP2 3′-UTR riboprobe, 55°C for the combined MECP2 CDS and 3′-UTR riboprobes and 60°C for the MECP2 CDS. Post-hybridization washes were performed twice in 4× SSC and once in 2× SSC and 2× SSC/0.1% Tween. Slides were air-dried and incubated in blocking solution (28) in preparation for riboprobe detection. For detection of riboprobes, FITC-labeled anti-digoxygenin (Boehringer Mannheim) was diluted 1/100 in detection solution (28). Slides were incubated at 42°C for 2 h, washed in 4×SSC/0.5% Tween and mounted in 50% glycerol/1×PBS with either 500 nM Sytox Green or 10 µg/ml propidium iodide (PI). Fluorescent signals were quantitated by LSC, comparing antisense to sense riboprobes. Slides with successful hybridizations indicated by positive shifts between sense and antisense hybridizations and by paired t-test analysis of triplicate samples from each individual were subsequently stained for indirect immunofluorescence of N-terminal MeCP2 (Affinity Bioreagents) or rabbit IgG (Upstate Biotechnology) as described previously (28).

ACKNOWLEDGEMENTS

This work was supported in part by the UC Davis M.I.N.D. Institute, the Rett Syndrome Research Foundation, and the NIH (1R01HD/NS41462). Human tissue samples were generously provided by the Autism Tissue Program, the University of Maryland Brain and Tissue Bank for Developmental Disorders (supported by NIH N01-HD-1-3138), Harvard Brain Tissue Resource Center (supported in part by PHS MH/NS 31862), and M. Lalande.

Figure 1. LSC quantitative analysis of MeCP2 expression in multiple human cerebral samples on a tissue microarray. (A) A representative LSC analysis of x,y slide position of individually contoured cells within each tissue on a microarray is shown. Each circle represents 600 µm diameter tissue cores arranged in triplicate for each individual cerebral cortex sample. Three different fetal, infant, juvenile, and adult control samples (solid boxes) were arranged together with six RTT, five autism, three PWS, one AS, and one PDD samples. (B) Each control age category in (A) was independently gated for histograms of anti-MeCP2 fluorescence. The x-axes represent maximum pixel intensities per cell on a binned scale of 0–16 400. Black overlay histograms were derived from a duplicate slide stained with nonspecific rabbit IgG as a negative control. Solid histograms represent anti-MeCP2 fluorescence and are colored in two regions based on the right half max of the major peak of the adult control histogram. MeCP2hi cells are colored in red and MeCP2lo cells are colored in green for (A–C). (C) Examples of histograms from neurodevelopmental disorder samples showing significant defects in MeCP2 expression. Analysis is as in (B) except solid histograms represent independent gating of each individual (triplicate cores) compared with a histogram of the three closest age-matched controls for each sample (blue overlay histogram).

Figure 1. LSC quantitative analysis of MeCP2 expression in multiple human cerebral samples on a tissue microarray. (A) A representative LSC analysis of x,y slide position of individually contoured cells within each tissue on a microarray is shown. Each circle represents 600 µm diameter tissue cores arranged in triplicate for each individual cerebral cortex sample. Three different fetal, infant, juvenile, and adult control samples (solid boxes) were arranged together with six RTT, five autism, three PWS, one AS, and one PDD samples. (B) Each control age category in (A) was independently gated for histograms of anti-MeCP2 fluorescence. The x-axes represent maximum pixel intensities per cell on a binned scale of 0–16 400. Black overlay histograms were derived from a duplicate slide stained with nonspecific rabbit IgG as a negative control. Solid histograms represent anti-MeCP2 fluorescence and are colored in two regions based on the right half max of the major peak of the adult control histogram. MeCP2hi cells are colored in red and MeCP2lo cells are colored in green for (A–C). (C) Examples of histograms from neurodevelopmental disorder samples showing significant defects in MeCP2 expression. Analysis is as in (B) except solid histograms represent independent gating of each individual (triplicate cores) compared with a histogram of the three closest age-matched controls for each sample (blue overlay histogram).

Figure 2. Combined quantitative analysis of MeCP2 immunofluorescence and alternative MECP2 transcripts by LSC. (A, C) The tissue microarray in Figure 1A was hybridized with a riboprobe detecting the MECP2 coding region (CDS, A) or alternative long 3′-UTR (3′-UTR, C) in either the sense (blue histogram) or antisense (black/green/red histograms) orientation. MeCP2 immunofluorescence was performed following in situ hybridization. LSC analysis to determine and separate MeCP2lo (green) and MeCP2hi (red) populations from the total population (black) was performed as in Figure 1 (data not shown). The MeCP2hi population (red histogram) showed significantly higher MECP2 antisense fluorescence than the MeCP2lo population (green histogram) in juvenile and adult samples for both CDS and 3′-UTR riboprobes. (B, D) Examples of histograms from neurodevelopmental disorder samples showing defects in MeCP2 expression. Analysis is as in (A, C) except black, red and green histograms represent independent gating of each individual (triplicate cores) compared with the appropriate antisense histogram of the three closest age-matched controls for each sample (orange histogram). AUT 797 showed significantly lower levels of MECP2 CDS and 3′-UTR antisense fluorescence compared with age-matched controls.

Figure 2. Combined quantitative analysis of MeCP2 immunofluorescence and alternative MECP2 transcripts by LSC. (A, C) The tissue microarray in Figure 1A was hybridized with a riboprobe detecting the MECP2 coding region (CDS, A) or alternative long 3′-UTR (3′-UTR, C) in either the sense (blue histogram) or antisense (black/green/red histograms) orientation. MeCP2 immunofluorescence was performed following in situ hybridization. LSC analysis to determine and separate MeCP2lo (green) and MeCP2hi (red) populations from the total population (black) was performed as in Figure 1 (data not shown). The MeCP2hi population (red histogram) showed significantly higher MECP2 antisense fluorescence than the MeCP2lo population (green histogram) in juvenile and adult samples for both CDS and 3′-UTR riboprobes. (B, D) Examples of histograms from neurodevelopmental disorder samples showing defects in MeCP2 expression. Analysis is as in (A, C) except black, red and green histograms represent independent gating of each individual (triplicate cores) compared with the appropriate antisense histogram of the three closest age-matched controls for each sample (orange histogram). AUT 797 showed significantly lower levels of MECP2 CDS and 3′-UTR antisense fluorescence compared with age-matched controls.

Figure 3. Developmental changes in MECP2 alternative transcripts in MeCP2lo and MeCP2hi populations in control samples. (A) The fluorescent values of MECP2 CDS hybridization were divided by the values for TFRC hybridization individually for each core of the tissue microarray and the results represent the mean±SEM of the nine samples for the MeCP2lo (hatched bars) and MeCP2hi populations (solid bars) for each age category. MECP2 transcript levels were significantly higher in MeCP2hi compared with MeCP2lo cells in all postnatal age categories. (B) TFRC-normalized quantitation of MECP2 3′-UTR transcript in control samples. MeCP2hi cells exhibited significantly higher levels of MECP2 long transcript compared with MeCP2lo cells in both the juvenile and adult samples. The MeCP2hi population for the fetal samples contained <25 cells per tissue core and was therefore not analyzed. (C) The ratio of long to total MECP2 transcripts was calculated for each core of each sample and the results graphed as mean±SEM. A significant decrease in the usage of the MECP2 long transcript was observed in the MeCP2lo population of juvenile and adult samples compared with the fetal MeCP2lo population, demonstrating a decreased usage of the long transcript with age. No significant differences were observed in the MECP2 long/total ratio between the MeCP2lo and MeCP2hi populations in any control sample.

Figure 3. Developmental changes in MECP2 alternative transcripts in MeCP2lo and MeCP2hi populations in control samples. (A) The fluorescent values of MECP2 CDS hybridization were divided by the values for TFRC hybridization individually for each core of the tissue microarray and the results represent the mean±SEM of the nine samples for the MeCP2lo (hatched bars) and MeCP2hi populations (solid bars) for each age category. MECP2 transcript levels were significantly higher in MeCP2hi compared with MeCP2lo cells in all postnatal age categories. (B) TFRC-normalized quantitation of MECP2 3′-UTR transcript in control samples. MeCP2hi cells exhibited significantly higher levels of MECP2 long transcript compared with MeCP2lo cells in both the juvenile and adult samples. The MeCP2hi population for the fetal samples contained <25 cells per tissue core and was therefore not analyzed. (C) The ratio of long to total MECP2 transcripts was calculated for each core of each sample and the results graphed as mean±SEM. A significant decrease in the usage of the MECP2 long transcript was observed in the MeCP2lo population of juvenile and adult samples compared with the fetal MeCP2lo population, demonstrating a decreased usage of the long transcript with age. No significant differences were observed in the MECP2 long/total ratio between the MeCP2lo and MeCP2hi populations in any control sample.

Figure 4. A summary of the different types of MeCP2 expression abnormalities observed in the neurodevelopmental disorder samples analyzed in this study. Samples showing MeCP2/MECP2 expression differences from age-matched controls are sorted into seven categories by transcriptional and posttranscriptional changes. (A) RTT samples from individuals with MECP2 mutations showed significant differences in MeCP2 mean fluorescence (mean) and the percentage of MeCP2hi cells (percentage high), but no significant differences in MECP2 transcripts from age-matched controls. Two RTT samples (1238 and 3393) had less than 10% of cells in the MeCP2hi population, so a separate analysis was not performed for this population. (B) One autism sample, AUT 797, showed significant differences in MeCP2 expression that could be explained by lower levels of both MECP2 transcripts and a decreased ratio of long to total MECP2 in the MeCP2hi population. (C) PWS 865 and 1290 showed significantly lower MeCP2 expression with only the C-terminal anti-MeCP2 but significant increases in MECP2 transcripts, suggesting that increased MECP2 transcription does not equate with increased protein. (D). AUT 967 and AS 293 showed significantly lower MeCP2 expression that could potentially be explained by significant increases within the MeCP2hi population of the long/total MECP2 transcript ratio, as the long MECP2 transcript is predicted to be translated less efficiently than the short form. (E) Two samples, PDD 144 and AUT 732, showed significant increases in MeCP2 expression from controls but no change in MECP2 transcripts except that MeCP2lo and MeCP2hi cells showed significant differences in the ratio of MECP2 alternative transcripts, unlike controls. (F) RTT 3381 had no detectable MECP2 mutation, but a lower percentage of MeCP2hi cells compared with age-matched controls using both antibodies, and a significant difference in the ratio of MECP2 long/short transcripts between the MeCP2lo and MeCP2hi populations because of a lower ratio in the MeCP2lo population. (G) The remaining two samples RTT 4321 and AUT 1174 showed deficiencies in MeCP2 only with the C-terminal antibody and no transcriptional changes, perhaps suggesting post-translational alterations.

Figure 4. A summary of the different types of MeCP2 expression abnormalities observed in the neurodevelopmental disorder samples analyzed in this study. Samples showing MeCP2/MECP2 expression differences from age-matched controls are sorted into seven categories by transcriptional and posttranscriptional changes. (A) RTT samples from individuals with MECP2 mutations showed significant differences in MeCP2 mean fluorescence (mean) and the percentage of MeCP2hi cells (percentage high), but no significant differences in MECP2 transcripts from age-matched controls. Two RTT samples (1238 and 3393) had less than 10% of cells in the MeCP2hi population, so a separate analysis was not performed for this population. (B) One autism sample, AUT 797, showed significant differences in MeCP2 expression that could be explained by lower levels of both MECP2 transcripts and a decreased ratio of long to total MECP2 in the MeCP2hi population. (C) PWS 865 and 1290 showed significantly lower MeCP2 expression with only the C-terminal anti-MeCP2 but significant increases in MECP2 transcripts, suggesting that increased MECP2 transcription does not equate with increased protein. (D). AUT 967 and AS 293 showed significantly lower MeCP2 expression that could potentially be explained by significant increases within the MeCP2hi population of the long/total MECP2 transcript ratio, as the long MECP2 transcript is predicted to be translated less efficiently than the short form. (E) Two samples, PDD 144 and AUT 732, showed significant increases in MeCP2 expression from controls but no change in MECP2 transcripts except that MeCP2lo and MeCP2hi cells showed significant differences in the ratio of MECP2 alternative transcripts, unlike controls. (F) RTT 3381 had no detectable MECP2 mutation, but a lower percentage of MeCP2hi cells compared with age-matched controls using both antibodies, and a significant difference in the ratio of MECP2 long/short transcripts between the MeCP2lo and MeCP2hi populations because of a lower ratio in the MeCP2lo population. (G) The remaining two samples RTT 4321 and AUT 1174 showed deficiencies in MeCP2 only with the C-terminal antibody and no transcriptional changes, perhaps suggesting post-translational alterations.

Table 1.

Postmortem brain samples used in tissue microarray

Sample Sex Mutation Sample Age PMI Cause of death 
Control fetal  279 18 weeks Elective termination 
  33 18 weeks Elective termination 
  31 18 weeks Elective termination 
Control infant  35 1 day 22 Trauma 
  759 35 days Idiopathic pulmonary hemorrage 
  86 56 days 11 Congenital heart defect 
Control juvenile  118 2.5 years 22 No anatomical cause 
  1185 5 years 17 Drowning 
  26 18.5 years 16 Accident, multiple injuries 
Normal adult  285 31 years 20 Accident, multiple injuries 
  632 34 years Accident, multiple injuries 
  772 36 years 13 Accident, multiple injuries 
Prader–Willi NM 865 43 years 28 Ruptured stomach 
 NM 1290 44 years 29 Complications of PWS 
 ND 970 46 years 15 Complications of PWS 
PDD NM 144 10 years 22 Drowning 
Angelman 15q11–13del 293 20 years 30.5 Complications of AS 
Autism ND 1315 8 years 24 Drowning 
 NM 1174 8 years 14 Multi-system organ failure 
 NM 797 9 years 13 Drowning 
 NM 732 15 years 28 Suicide, hanging 
 NM 967 32 years 30 Complications of cancer 
RTT 1154del32 1238 2 years Complications of RTT 
 R168X 3393 12 years 2.8 Complications of RTT 
 ND 1420 21 years 18 Complications of RTT 
 NM 3381 22 years 3.5 Complications of RTT 
 NM 4321 23 years 10.5 Complications of RTT 
 R168X 4312 24 years 14 Complications of RTT 
Sample Sex Mutation Sample Age PMI Cause of death 
Control fetal  279 18 weeks Elective termination 
  33 18 weeks Elective termination 
  31 18 weeks Elective termination 
Control infant  35 1 day 22 Trauma 
  759 35 days Idiopathic pulmonary hemorrage 
  86 56 days 11 Congenital heart defect 
Control juvenile  118 2.5 years 22 No anatomical cause 
  1185 5 years 17 Drowning 
  26 18.5 years 16 Accident, multiple injuries 
Normal adult  285 31 years 20 Accident, multiple injuries 
  632 34 years Accident, multiple injuries 
  772 36 years 13 Accident, multiple injuries 
Prader–Willi NM 865 43 years 28 Ruptured stomach 
 NM 1290 44 years 29 Complications of PWS 
 ND 970 46 years 15 Complications of PWS 
PDD NM 144 10 years 22 Drowning 
Angelman 15q11–13del 293 20 years 30.5 Complications of AS 
Autism ND 1315 8 years 24 Drowning 
 NM 1174 8 years 14 Multi-system organ failure 
 NM 797 9 years 13 Drowning 
 NM 732 15 years 28 Suicide, hanging 
 NM 967 32 years 30 Complications of cancer 
RTT 1154del32 1238 2 years Complications of RTT 
 R168X 3393 12 years 2.8 Complications of RTT 
 ND 1420 21 years 18 Complications of RTT 
 NM 3381 22 years 3.5 Complications of RTT 
 NM 4321 23 years 10.5 Complications of RTT 
 R168X 4312 24 years 14 Complications of RTT 

NM, no detectable MECP2 mutation. ND, not determined, fixed tissue.

Table 2.

Significant differences in MeCP2 expression and the percentage of MeCP2hi neurons in neurodevelopmental disorders

Sample type Sample no. Age N-terminal MeCP2 C-terminal MeCP2 Histone H1 
   Mean±SEM Percentage MeCP2hi Mean±SEM Percentage MeCP2hi Mean±SEM Percentage H1hi 
Control fetal 279 18 weeks 4360±286  7.8±3.1 2137±173  2.5±1.1 3222±158 23.9±9.0 
 33 18 weeks 4150±269  7.2±3.0 1690±211  0.1±0.1 3366±263 22.9±5.3 
 31 18 weeks 4115±289 10.7±6.0 2715±328  0.9±0.4 3427±240 17.6±0.8 
Fetal average   4215±157  8.4±2.2 2092±151  1.2±0.5 3327±112 22.0±3.3 
Control infant 35  1 day 5261±166 13.3±3.0 2539±223  4.6±1.6 4501±240 46.1±12.8 
 759 35 days 6034±279 28.1±5.8 3168±310  4.6±2.7 5646±596 62.7±1.1 
 86 56 days 5986±229 33.1±4.5 3684±367 20.5±2.9 5036±483 49.9±13.4 
Infant average   5761±144 24.8±3.0 3130±198 12.0±2.0 5061±285 52.9±5.9 
Control juvenile 118 2.5 years 6296±213 36.6±5.2 3730±393 22.8±2.7 5528±723 60.2±3.0 
 1185  5 years 5668±214 21.2±3.5 3558±421 18.9±2.0 5824±702 63.4±2.0 
 26 18.5 years 6524±209 40.2±4.6 4011±398 25.4±2.4 5306±681 56.0±4.6 
Juvenile average   6162±136 34.7±2.7 3766±225 22.4±1.4 5409±345 59.9±2.0 
Control adult 285 31 years 5987±176 30.8±4.7 4081±378 27.5±2.6 5974±421 62.7±8.7 
 632 34 years 6664±243 43.2±5.8 5901±609 49.6±1.0 4543±590 41.9±5.0 
 772 36 years 6800±166 43.6±5.1 4918±468 36.9±1.1 5471±1293 53.9±6.1 
Adult average   6484±129 39.2±3.0 4911±307 36.8±2.2 5329±477 60.9±4.4 
Prader–Willi 865 43 years 6107±362 28.4±7.7 3720±379a20.0±2.9a** 5979±1288 63.5±5.8 
 1290 44 years 6238±298 31.9±5.3 3411±377a15.5±2.3a** 6414±1677 65.2±9.4 
 970 46 years ND ND ND ND 4329±1161a34.2±14.7a
PDD 144 10 years 7146±296b54.3±7.6b5651±401b37.8±1.4b*** 4743±830 42.2±7.6 
Angelman 293 20 years 5298±203a*** 18.5±4.7a2822±304a 8.4±2.0a*** 5482±1209 60.6±8.0 
Autism 1315  8 years ND ND ND ND 2240±1247a***  7.1±2.5a*** 
 1174  8 years 6147±269 35.7±4.2 2707±335a 6.3±1.4a*** 5736±799 62.7±1.7 
 797  9 years 5413±266a19.4±4.6a2871±236a 6.4±2.9a5959±923 65.3±2.5 
 732 15 years 7068±274b50.9±5.9b5801±546b44.5±1.6b*** 5163±1227 56.3±3.6 
 967 32 years 5468±140a*** 17.0±3.8a*** 3466±351a14.4±2.3a*** 5619±1081 62.2±6.2 
RTT 1238  2 years 4419±268a***  3.9±1.4a*** 1944±275a**  1.1±0.9a*** 5692±1120 59.7± 6.0 
 3393 12 years 4326±298a***  5.5±3.1a*** 2065±221a***  1.2±0.5a*** 6109±1106 64.8±3.4 
 1420 21 years ND ND ND ND 4328±956a38.0±5.0a
 3381 22 years 5694±330 22.6±4.4a3289±297a13.9±1.2a*** 5368±880 60.9±4.4 
 4321 23 years 6050±378 31.7±5.9 3151±323a12.8±3.8a5904±806 68.1±2.1 
 4312 24 years 5483±257a18.6±3.6a*** 2746±229a**  9.2±3.4a*** 5721±680 63.1±3.0 
Sample type Sample no. Age N-terminal MeCP2 C-terminal MeCP2 Histone H1 
   Mean±SEM Percentage MeCP2hi Mean±SEM Percentage MeCP2hi Mean±SEM Percentage H1hi 
Control fetal 279 18 weeks 4360±286  7.8±3.1 2137±173  2.5±1.1 3222±158 23.9±9.0 
 33 18 weeks 4150±269  7.2±3.0 1690±211  0.1±0.1 3366±263 22.9±5.3 
 31 18 weeks 4115±289 10.7±6.0 2715±328  0.9±0.4 3427±240 17.6±0.8 
Fetal average   4215±157  8.4±2.2 2092±151  1.2±0.5 3327±112 22.0±3.3 
Control infant 35  1 day 5261±166 13.3±3.0 2539±223  4.6±1.6 4501±240 46.1±12.8 
 759 35 days 6034±279 28.1±5.8 3168±310  4.6±2.7 5646±596 62.7±1.1 
 86 56 days 5986±229 33.1±4.5 3684±367 20.5±2.9 5036±483 49.9±13.4 
Infant average   5761±144 24.8±3.0 3130±198 12.0±2.0 5061±285 52.9±5.9 
Control juvenile 118 2.5 years 6296±213 36.6±5.2 3730±393 22.8±2.7 5528±723 60.2±3.0 
 1185  5 years 5668±214 21.2±3.5 3558±421 18.9±2.0 5824±702 63.4±2.0 
 26 18.5 years 6524±209 40.2±4.6 4011±398 25.4±2.4 5306±681 56.0±4.6 
Juvenile average   6162±136 34.7±2.7 3766±225 22.4±1.4 5409±345 59.9±2.0 
Control adult 285 31 years 5987±176 30.8±4.7 4081±378 27.5±2.6 5974±421 62.7±8.7 
 632 34 years 6664±243 43.2±5.8 5901±609 49.6±1.0 4543±590 41.9±5.0 
 772 36 years 6800±166 43.6±5.1 4918±468 36.9±1.1 5471±1293 53.9±6.1 
Adult average   6484±129 39.2±3.0 4911±307 36.8±2.2 5329±477 60.9±4.4 
Prader–Willi 865 43 years 6107±362 28.4±7.7 3720±379a20.0±2.9a** 5979±1288 63.5±5.8 
 1290 44 years 6238±298 31.9±5.3 3411±377a15.5±2.3a** 6414±1677 65.2±9.4 
 970 46 years ND ND ND ND 4329±1161a34.2±14.7a
PDD 144 10 years 7146±296b54.3±7.6b5651±401b37.8±1.4b*** 4743±830 42.2±7.6 
Angelman 293 20 years 5298±203a*** 18.5±4.7a2822±304a 8.4±2.0a*** 5482±1209 60.6±8.0 
Autism 1315  8 years ND ND ND ND 2240±1247a***  7.1±2.5a*** 
 1174  8 years 6147±269 35.7±4.2 2707±335a 6.3±1.4a*** 5736±799 62.7±1.7 
 797  9 years 5413±266a19.4±4.6a2871±236a 6.4±2.9a5959±923 65.3±2.5 
 732 15 years 7068±274b50.9±5.9b5801±546b44.5±1.6b*** 5163±1227 56.3±3.6 
 967 32 years 5468±140a*** 17.0±3.8a*** 3466±351a14.4±2.3a*** 5619±1081 62.2±6.2 
RTT 1238  2 years 4419±268a***  3.9±1.4a*** 1944±275a**  1.1±0.9a*** 5692±1120 59.7± 6.0 
 3393 12 years 4326±298a***  5.5±3.1a*** 2065±221a***  1.2±0.5a*** 6109±1106 64.8±3.4 
 1420 21 years ND ND ND ND 4328±956a38.0±5.0a
 3381 22 years 5694±330 22.6±4.4a3289±297a13.9±1.2a*** 5368±880 60.9±4.4 
 4321 23 years 6050±378 31.7±5.9 3151±323a12.8±3.8a5904±806 68.1±2.1 
 4312 24 years 5483±257a18.6±3.6a*** 2746±229a**  9.2±3.4a*** 5721±680 63.1±3.0 

ND, not determined, fixed tissue.

aSignificantly lower compared with age-matched control; bsignificantly higher compared with age-matched control.

*P<0.05; **P<0.001; ***P<0.0001.

Table 3.

Significant differences in MECP2 transcript usage in MeCP2lo and MeCP2hi populations and in several neurodevelopmental disorders

Sample No. Age CDS/TFRC 3′-UTR/TFRC 3′-UTR/CDS 
   MeCP2lo MeCP2hi MeCP2lo MeCP2hi MeCP2lo MeCP2hi 
Control fetal 279 18 weeks 0.69 0.79 0.72 NA 0.85 NA 
 33 18 weeks 0.65 0.62 0.70 NA 1.08 NA 
 31 18 weeks NA NA 1.02 NA 1.07 NA 
Fetal average   0.7 0.7 0.8 NA 1.14 NA 
Control infant 35  1 day 0.65 0.81 0.64 0.75 0.99 1.04 
 759 35 days 0.64 0.78 0.77 0.92 1.21 1.17 
 86 56 days 0.63 0.91a0.43 0.63a0.70 0.69 
Infant average   0.64 0.84a** 0.62 0.77 0.96 0.92 
Control juvenile 118 2.5 years 0.81 1.08a0.53 0.82a0.66 0.76 
 1185  5 years 0.84 1.47 0.64 1.31a0.78 0.93 
 26 18 years 0.83 1.23a0.64 1.15a*** 0.77 0.95 
Juvenile average   0.83 1.26a** 0.60 1.10a0.73 0.87 
         
Control adult 285 31 years 1.02 1.21 1.00 1.10 0.88 0.82 
 632 34 years 0.72 1.03a0.55 0.90a0.75 0.87 
 772 36 years 0.62 0.86 0.54 0.88a0.88 1.04 
Adult average   0.79 1.01a0.66 0.92a0.83 0.91 
PWS 865 43 years 0.97c1.41a*c** 0.71 1.16a*c0.73 1.19a*c
 1290 44 years 0.88 1.43a*c0.64 1.21a*c0.73 0.87 
PDD 144 10 years 0.87 1.28a0.63 1.23a0.73 0.96a
Angelman 293 20 years 0.72 1.38a0.61 1.41a*c0.83 1.02a*c
Autism 1174  8 years 0.74 1.22a** 0.60 1.50 0.81 1.23 
 797  9 years 0.81 0.97b0.55 0.58b0.68 0.58b
 732 15 years 0.81 1.20a** 0.59 1.13a*** 0.73 0.94a** 
 967 32 years 0.80 1.21 0.75 1.46a0.93 1.21c
RTT 1238  2 years 0.73 NA 0.67 NA 0.91 NA 
 3393 12 years 0.87 NA 0.63 NA 0.72 NA 
 3381 22 years 0.83 1.17a0.59 1.22 0.71b1.02a
 4321 23 years 0.86 1.20a0.63 1.16a0.74 1.03 
 4312 24 years 0.77 1.10 0.65 1.36a0.86 1.29 
Sample No. Age CDS/TFRC 3′-UTR/TFRC 3′-UTR/CDS 
   MeCP2lo MeCP2hi MeCP2lo MeCP2hi MeCP2lo MeCP2hi 
Control fetal 279 18 weeks 0.69 0.79 0.72 NA 0.85 NA 
 33 18 weeks 0.65 0.62 0.70 NA 1.08 NA 
 31 18 weeks NA NA 1.02 NA 1.07 NA 
Fetal average   0.7 0.7 0.8 NA 1.14 NA 
Control infant 35  1 day 0.65 0.81 0.64 0.75 0.99 1.04 
 759 35 days 0.64 0.78 0.77 0.92 1.21 1.17 
 86 56 days 0.63 0.91a0.43 0.63a0.70 0.69 
Infant average   0.64 0.84a** 0.62 0.77 0.96 0.92 
Control juvenile 118 2.5 years 0.81 1.08a0.53 0.82a0.66 0.76 
 1185  5 years 0.84 1.47 0.64 1.31a0.78 0.93 
 26 18 years 0.83 1.23a0.64 1.15a*** 0.77 0.95 
Juvenile average   0.83 1.26a** 0.60 1.10a0.73 0.87 
         
Control adult 285 31 years 1.02 1.21 1.00 1.10 0.88 0.82 
 632 34 years 0.72 1.03a0.55 0.90a0.75 0.87 
 772 36 years 0.62 0.86 0.54 0.88a0.88 1.04 
Adult average   0.79 1.01a0.66 0.92a0.83 0.91 
PWS 865 43 years 0.97c1.41a*c** 0.71 1.16a*c0.73 1.19a*c
 1290 44 years 0.88 1.43a*c0.64 1.21a*c0.73 0.87 
PDD 144 10 years 0.87 1.28a0.63 1.23a0.73 0.96a
Angelman 293 20 years 0.72 1.38a0.61 1.41a*c0.83 1.02a*c
Autism 1174  8 years 0.74 1.22a** 0.60 1.50 0.81 1.23 
 797  9 years 0.81 0.97b0.55 0.58b0.68 0.58b
 732 15 years 0.81 1.20a** 0.59 1.13a*** 0.73 0.94a** 
 967 32 years 0.80 1.21 0.75 1.46a0.93 1.21c
RTT 1238  2 years 0.73 NA 0.67 NA 0.91 NA 
 3393 12 years 0.87 NA 0.63 NA 0.72 NA 
 3381 22 years 0.83 1.17a0.59 1.22 0.71b1.02a
 4321 23 years 0.86 1.20a0.63 1.16a0.74 1.03 
 4312 24 years 0.77 1.10 0.65 1.36a0.86 1.29 

NA, <25 cells in population.

aSignificantly higher compared with MeCP2lo population; bsignificantly lower compared with age-matched control; csignificantly higher compared with age-matched control.

*P<0.05; **P<0.001; ***P<0.0001.

References

1
Shahbazian, M.D. and Zoghbi, H.Y. (
2002
) Rett syndrome and MeCP2: linking epigenetics and neuronal function.
Am. J. Hum. Genet.
 ,
71
,
1259
–1272.
2
Van den Veyver, I.B. and Zoghbi, H.Y. (
2000
) Methyl-CpG-binding protein 2 mutations in rett syndrome.
Curr. Opin. Genet. Dev.
 ,
10
,
275
–279.
3
Sirianni, N., Naidu, S., Pereira, J., Pillotto, R.F. and Hoffman, E.P. (
1998
) Rett syndrome: confirmation of X-linked dominant inheritance, and localization of the gene to Xq28.
Am. J. Hum. Genet.
 ,
63
,
1552
–1558.
4
Amir, R.E., Van den Veyver, I.B., Wan, M., Tran, C.Q., Francke, U. and Zoghbi, H.Y. (
1999
) Rett syndrome is caused by mutations in X-linked MECP2, encoding methyl- CpG-binding protein 2.
Nat. Genet.
 ,
23
,
185
–188.
5
Lewis, J.D., Meehan, R.R., Henzel, W.J., Maurer-Fogy, I., Jeppesen, P., Klein, F. and Bird, A. (
1992
) Purification, sequence, and cellular localization of a novel chromosomal protein that binds to methylated DNA.
Cell
 ,
69
,
905
–914.
6
Meehan, R.R., Lewis, J.D. and Bird, A.P. (
1992
) Characterization of MeCP2, a vertebrate DNA binding protein with affinity for methylated DNA.
Nucl. Acids Res.
 ,
20
,
5085
–5092.
7
Amir, R.E., Van den Veyver, I.B., Schultz, R., Malicki, D.M., Tran, C.Q., Dahle, E.J., Philippi, A., Timar, L., Percy, A.K., Motil, K.J. et al. (
2000
) Influence of mutation type and X chromosome inactivation on Rett syndrome phenotypes.
Ann. Neurol.
 ,
47
,
670
–679.
8
Cheadle, J.P., Gill, H., Fleming, N., Maynard, J., Kerr, A., Leonard, H., Krawczak, M., Cooper, D.N., Lynch, S., Thomas, N. et al. (
2000
) Long-read sequence analysis of the MECP2 gene in Rett syndrome patients: correlation of disease severity with mutation type and location.
Hum. Mol. Genet.
 ,
9
,
1119
–1129.
9
Huppke, P., Laccone, F., Kramer, N., Engel, W. and Hanefeld, F. (
2000
) Rett syndrome: analysis of MECP2 and clinical characterization of 31 patients.
Hum. Mol. Genet.
 ,
9
,
1369
–1375.
10
Xiang, F., Buervenich, S., Nicolao, P., Bailey, M.E., Zhang, Z. and Anvret, M. (
2000
) Mutation screening in Rett syndrome patients.
J. Med. Genet.
 ,
37
,
250
–255.
11
Bienvenu, T., Carrie, A., de Roux, N., Vinet, M.C., Jonveaux, P., Couvert, P., Villard, L., Arzimanoglou, A., Beldjord, C., Fontes, M. et al. (
2000
) MECP2 mutations account for most cases of typical forms of Rett syndrome.
Hum. Mol. Genet.
 ,
9
,
1377
–1384.
12
Buyse, I.M., Fang, P., Hoon, K.T., Amir, R.E., Zoghbi, H.Y. and Roa, B.B. (
2000
) Diagnostic testing for Rett syndrome by DHPLC and direct sequencing analysis of the MECP2 gene: identification of several novel mutations and polymorphisms.
Am. J. Hum. Genet.
 ,
67
,
1428
–1436.
13
Orrico, A., Lam, C., Galli, L., Dotti, M.T., Hayek, G., Tong, S.F., Poon, P.M., Zappella, M., Federico, A. and Sorrentino, V. (
2000
) MECP2 mutation in male patients with non-specific X-linked mental retardation.
FEBS Lett.
 ,
481
,
285
–288.
14
Villard, L., Cardoso, A.K., Chelly, P.J., Tardieu, P.M. and Fontes, M. (
2000
) Two affected boys in a Rett syndrome family: clinical and molecular findings.
Neurology
 ,
55
,
1188
–1193.
15
Clayton-Smith, J., Watson, P., Ramsden, S. and Black, G.C. (
2000
) Somatic mutation in MECP2 as a non-fatal neurodevelopmental disorder in males.
Lancet
 ,
356
,
830
–832.
16
Couvert, P., Bienvenu, T., Aquaviva, C., Poirier, K., Moraine, C., Gendrot, C., Verloes, A., Andres, C., Le Fevre, A.C., Souville, I. et al. (
2001
) MECP2 is highly mutated in X-linked mental retardation.
Hum. Mol. Genet.
 ,
10
,
941
–946.
17
Watson, P., Black, G., Ramsden, S., Barrow, M., Super, M., Kerr, B. and Clayton-Smith, J. (
2001
) Angelman syndrome phenotype associated with mutations in MECP2, a gene encoding a methyl CpG binding protein.
J. Med. Genet.
 ,
38
,
224
–228.
18
Girard, M., Couvert, P., Carrie, A., Tardieu, M., Chelly, J., Beldjord, C. and Bienvenu, T. (
2001
) Parental origin of de novo MECP2 mutations in Rett syndrome.
Eur. J. Hum. Genet.
 ,
9
,
231
–236.
19
Trappe, R., Laccone, F., Cobilanschi, J., Meins, M., Huppke, P., Hanefeld, F. and Engel, W. (
2001
) MECP2 mutations in sporadic cases of Rett syndrome are almost exclusively of paternal origin.
Am. J. Hum. Genet.
 ,
68
,
1093
–1101.
20
Lamb, J.A., Moore, J., Bailey, A. and Monaco, A.P. (
2000
) Autism: recent molecular genetic advances.
Hum. Mol. Genet.
 ,
9
,
861
–868.
21
Armstrong, D.D. (
1997
) Review of Rett syndrome.
J. Neuropathol. Exp. Neurol.
 ,
56
,
843
–849.
22
Jamain, S., Quach, H., Betancur, C., Rastam, M., Colineaux, C., Gillberg, I.C., Soderstrom, H., Giros, B., Leboyer, M., Gillberg, C. et al. (
2003
) Mutations of the X-linked genes encoding neuroligins NLGN3 and NLGN4 are associated with autism.
Nat. Genet.
 ,
34
,
27
–29.
23
Lam, C., Yeung, W., Ko, C., Poon, P., Tong, S., Chan, K., Lo, I., Chan, L., Hui, J., Wong, V. et al. (
2000
) Spectrum of mutations in the MECP2 gene in patients with infantile autism and Rett syndrome.
J. Med. Genet.
 ,
37
,
E41
.
24
Beyer, K.S., Blasi, F., Bacchelli, E., Klauck, S.M., Maestrini, E., Poustka, A. and International Molecular Genetic Study of Autism (
2002
) Mutation analysis of the coding sequence of the MECP2 gene in infantile autism.
Hum. Genet.
 ,
111
,
305
–309.
25
Carney, R.M., Wolpert, C.M., Ravan, S.A., Shahbazian, M., Ashley-Koch, A., Cuccaro, M.L., Vance, J.M. and Pericak-Vance, M.A. (
2003
) Identification of MeCP2 mutations in a series of females with autistic disorder.
Pediatr. Neurol.
 ,
28
,
205
–211.
26
Vourc'h, P., Bienvenu, T., Beldjord, C., Chelly, J., Barthelemy, C., Muh, J.P. and Andres, C. (
2001
) No mutations in the coding region of the Rett syndrome gene MECP2 in 59 autistic patients.
Eur. J. Hum. Genet.
 ,
9
,
556
–558.
27
Lobo-Menendez, F., Sossey-Alaoui, K., Bell, J.M., Copeland-Yates, S.A., Plank, S.M., Sanford, S.O., Skinner, C., Simensen, R.J., Schroer, R.J. and Michaelis, R.C. (
2003
) Absence of MeCP2 mutations in patients from the South Carolina autism project.
Am. J. Med. Genet.
 ,
117B
,
97
–101.
28
LaSalle, J., Goldstine, J., Balmer, D. and Greco, C. (
2001
) Quantitative localization of heterologous methyl-CpG-binding protein 2 (MeCP2) expression phenotypes in normal and Rett syndrome brain by laser scanning cytometry.
Hum. Mol. Genet.
 ,
10
,
1729
–1740.
29
Balmer, D., Goldstine, J., Rao, Y.M. and LaSalle, J.M. (
2003
) Elevated methyl-CpG-binding protein 2 expression is acquired during postnatal human brain development and is correlated with alternative polyadenylation.
J. Mol. Med.
 ,
81
,
61
–68.
30
Shahbazian, M.D., Antalffy, B., Armstrong, D.L. and Zoghbi, H.Y. (
2002
) Insight into Rett syndrome: MeCP2 levels display tissue- and cell- specific differences and correlate with neuronal maturation.
Hum. Mol. Genet.
 ,
11
,
115
–124.
31
Akbarian, S., Chen, R.Z., Gribnau, J., Rasmussen, T.P., Fong, H., Jaenisch, R. and Jones, E.G. (
2001
) Expression pattern of the Rett syndrome gene MeCP2 in primate prefrontal cortex.
Neurobiol. Dis.
 ,
8
,
784
–791.
32
Jung, B.P., Jugloff, D.G., Zhang, G., Logan, R., Brown, S. and Eubanks, J.H. (
2003
) The expression of methyl CpG binding factor MeCP2 correlates with cellular differentiation in the developing rat brain and in cultured cells.
J. Neurobiol.
 ,
55
,
86
–96.
33
D'Esposito, M., Quaderi, N.A., Ciccodicola, A., Bruni, P., Esposito, T., D'Urso, M. and Brown, S.D. (
1996
) Isolation, physical mapping, and northern analysis of the X-linked human gene encoding methyl CpG-binding protein, MECP2.
Mamm. Genome
 ,
7
,
533
–535.
34
Coy, J.F., Sedlacek, Z., Bachner, D., Delius, H. and Poustka, A. (
1999
) A complex pattern of evolutionary conservation and alternative polyadenylation within the long 3′-untranslated region of the methyl-CpG-binding protein 2 gene (MeCP2) suggests a regulatory role in gene expression.
Hum. Mol. Genet.
 ,
8
,
1253
–1262.
35
Reichwald, K., Thiesen, J., Wiehe, T., Weitzel, J., Poustka, W.A., Rosenthal, A., Platzer, M., Stratling, W.H. and Kioschis, P. (
2000
) Comparative sequence analysis of the MECP2-locus in human and mouse reveals new transcribed regions.
Mamm. Genome
 ,
11
,
182
–190.
36
Courchesne, E., Carper, R. and Akshoomoff, N. (
2003
) Evidence of brain overgrowth in the first year of life in autism.
JAMA
 ,
290
,
337
–344.
37
Lainhart, J.E. (
2003
) Increased rate of head growth during infancy in autism.
JAMA
 ,
290
,
393
–394.