J Neuropathol Exp Neurol Copyright Ó 2005 by the American Association of Neuropathologists, Inc.

Vol. 64, No. 11 November 2005 pp. 1018–1027

ORIGINAL ARTICLE

Serotonin Transporter Abnormality in the Dorsal Motor Nucleus of the Vagus in Rett Syndrome: Potential Implications for Clinical Autonomic Dysfunction David S. Paterson, PhD, Eric G. Thompson, MS, Richard A. Belliveau, BA, Bobbie A. Antalffy, BSc, Felicia L. Trachtenberg, PhD, Dawna D. Armstrong, MD, and Hannah C. Kinney, MD

Autonomic dysfunction is prevalent in girls with Rett syndrome, an X-chromosome-linked disorder of mental retardation resulting from mutations in the gene encoding methyl-CpG-binding protein 2 (MeCP2). This gene plays a role in regulating neuronal activitydependent gene expression, including brain-derived neurotrophic factor (BDNF), which is a potent serotonergic (5-HT) neuronal growth factor. We analyzed selected parameters of the 5-HT system of the medulla in autopsied patients with Rett syndrome because of the role of BDNF in 5-HT cell development and because 5-HT plays a key role in modulating autonomic control. 5-HT neurons were identified by immunostaining for tryptophan hydroxylase, the biosynthetic enzyme for 5-HT. We quantitated the number of 5-HT cells in the medulla at 2 standardized levels in 11 Rett and 7 control cases. There was no significant difference in 5-HT cell number between the groups. We analyzed binding to the serotonin transporter (SERT) using the radioligand [125I]-RTI-55 with tissue autoradiography in 7 Rett and 5 controls in 9 cardiorespiratory-related nuclei. In the dorsal motor nucleus of the vagus (DMX) (preganglionic parasympathetic outflow), SERT binding for the control cases decreased significantly over time (p = 0.049) but did not change in the Rett cases (p = 0.513). Adjusting for age, binding between the Rett and control cases differed significantly in this nucleus (p = 0.022). There was a marginally significant age versus diagnosis interaction (p = 0.06). Thus, altered 5-HT innervation and/or uptake in the DMX may contribute to abnormal 5-HT modulation of this major autonomic nucleus in patients with Rett syndrome. These data suggest hypotheses concerning 5-HT modulation of vagal function for testing in MeCP2 knockout mice to understand mechanisms underlying autonomic dysfunction in patients with Rett syndrome.

From the Department of Pathology (DSP, EGT, RAB, HCK), Children’s Hospital and Harvard Medical School, Boston, Massachusetts; Department of Pathology (BAA, DDA), Texas Children’s Hospital and Baylor College of Medicine, Houston, Texas; and New England Research Institutes (FLT), Watertown, Massachusetts. Send correspondence and reprint requests to: David S. Paterson, PhD, Department of Pathology (Neuropathology), Enders Room 11110, 320 Longwood Avenue, Boston, MA 02115; E-mail: david.paterson@childrens. harvard.edu This work was supported by a grant from the Rett Syndrome Research Foundation and the Mental Retardation Research Center, Children’s Hospital, Boston, Massachusetts (P30-HD18655).

1018

INTRODUCTION Rett syndrome (RTT) is a progressive neurodevelopmental disorder characterized by gradual cognitive decline following normal development through the first 6 to 18 postnatal months and stereotypic hand movements. It is one of the most common causes of mental retardation in females (1, 2). In addition, patients with RTT syndrome demonstrate a spectrum of respiratory and autonomic dysfunction, including hyperventilation, abnormal heart rate variability, and sudden death (3). The majority of RTT cases are caused by inactivating mutations in the X-chromosome-linked gene encoding methyl-CpG-binding protein 2 (MeCP2) (4, 5). MeCP2 plays a critical role in neuronal maturation and synaptogenesis (6, 7) by regulating activity-dependent expression of specific genes, including brain-derived neurotrophic factor (BDNF) and putatively S100b (7–9). BDNF and S100b are both neurotrophins that play important roles in serotonergic (5-HT) neuronal development, including stimulation of 5-HT synthesis, terminal sprouting, neurite outgrowth, dendrite formation, synaptic plasticity, and neuron survival (10–24). BDNF exerts these effects on 5-HT neurons through direct activation of tyrosine kinase b (TrkB) receptors (21, 23), which are known to be expressed by raphe´ 5-HT neurons (25). Serotonin is an important modulator of cardiorespiratory control through projections from 5-HT neurons in the midline (raphe´) and lateral (extra-raphe´) to ‘‘effector’’ nuclei elsewhere in the brainstem and spinal cord. These latter nuclei include the 2 major brainstem components of the autonomic nervous system, i.e. the dorsal motor nucleus of cranial nerve X (DMX) and the nucleus of the solitary tract, motor, and sensory components of the vagus nerve, respectively. 5-HT neurons in the medulla are thought to form a ‘‘medullary 5HT system’’ that modulates and potentially integrates autonomic and respiratory function according to the level of arousal (26). Indeed, medullary 5-HT abnormalities are associated with respiratory dysfunction (26–30) and with the sudden infant death syndrome (SIDS) (31), a putative disorder of central cardiorespiratory control. Significantly, BDNF and 5-HT are both involved in long-term

J Neuropathol Exp Neurol  Volume 64, Number 11, November 2005

Downloaded from http://jnen.oxfordjournals.org/ by guest on October 8, 2016

Key Words: Heart rate variability, Hyperventilation, MeCP2 mutation, Medulla oblongata, Parasympathetic, Raphe´ nucleus.

Abstract

J Neuropathol Exp Neurol  Volume 64, Number 11, November 2005

region. As a result of the availability of SERT-binding data from an infant data set in our laboratory (see subsequently), we were also able to determine the normative developmental profile of SERT binding from infancy through adolescence, allowing us to study SERT distribution in RTT relative to normal development.

MATERIALS AND METHODS Clinical Database The RTT and control brainstem tissue for this study was obtained from the McLean Brain Tissue Resources (Boston, MA), University of Maryland Brain Bank (Baltimore, MD), and the autopsy services of the Children’s Hospital Boston (Boston, MA), and Texas Children’s Hospital (Houston, TX). The whole brainstems of the RTT cases were not available, but rather, typically blocks of hemisected medulla, pons, and midbrain. Consequently, in the analysis, standardized levels of the brainstem were used for comparison among RTT and control cases (see subsequently). Limited clinical information was available for the patients with RTT syndrome in the SERT analysis; in particular, information about respiratory and autonomic dysfunction was not available, except for information reporting sudden and unexpected death in 2 cases (Table 1). The RTT cases in the SERT analysis were tested for MeCP2 mutations in the DNA Diagnostics Laboratory, Baylor College of Medicine (Houston, TX). All cases analyzed for SERT binding were tested for mutations in exons 1–4 of the MeCP2 gene or exhibited classic RTT clinical phenotype (Table 1). For each case in this study, the genetic information was coded and was not linked with the patient’s name. The control cases were children and adolescents who died of defined causes of autopsy without abnormalities in the brain, including the brainstem (Table 1).

TABLE 1. Clinicopathologic Information of Rett Syndrome and Control Cases in Analysis of Serotonin Transporter Binding Case No.

Age (years)

Gender

Diagnosis

Mutation

Mental Retardation

0766 1748 1790 1954 3085 8449 2211 2396 2457 8413 4812 4888 5071 7943 9168 8959 6975

12 6 24 11 8 10 5 20 7 9 16 3 14 9 11 20 26

F F F F F F F F F F NA NA F F F F F

RTT RTT RTT RTT RTT RTT RTT RTT-like RTT-like RTT-like Leukemia Sepsis Trauma Aspiration asphyxia Pending Ewing’s sarcoma Pulmonary hemorrhage

R270x P322 R255x R255x R106w R270x R255x Unknown Unknown Unknown NR NR NR NR NR NR NR

Profound Profound Profound Profound Profound Profound Profound Profound Profound Profound None None None None None None None

NA, information not available; NR, not relevant.

q 2005 American Association of Neuropathologists, Inc.

1019

Downloaded from http://jnen.oxfordjournals.org/ by guest on October 8, 2016

facilitation of respiration in response to episodic hypoxia (32), and BDNF is necessary for normal development of the putative respiratory rhythm generator (the pre-Bo¨tzinger complex) in mice (33). Taken together, these observations suggest that MeCP2 mutations resulting in altered BDNF expression, and subsequent abnormal development of 5-HT neurons, may be at least partly responsible for the respiratory and autonomic dysfunction associated with RTT. This idea is supported by the observations that brain levels of both BDNF and 5-HT are reduced in RTT patients (34–37). In this study, we tested the hypothesis that the number of 5-HT neurons and/or the density of serotonin transporter (SERT) binding are altered in the source nuclei and projection sites of the medullary 5-HT system in RTT cases compared with age-related controls. We focused on the medullary 5-HT system in particular because of its role in the modulation of autonomic and respiratory control, and because the basic pathology of cardiorespiratory dysfunction in patients with RTT syndrome is poorly understood. We counted the number of 5-HT neuronal cell bodies immunoreactive for tryptophan hydroxylase (TPOH) in the raphe´ and extra-raphe´ regions of the medulla. Tryptophan hydroxylase is the rate-limiting enzyme for 5-HT synthesis and is a specific marker of 5-HT neurons. We measured the quantitative distribution of SERT binding with tissue section autoradiography in the source nuclei and projection sites of the medullary 5-HT system. The SERT is responsible for the reuptake back into the neuron of released 5-HT and is considered the most critical element in the regulation of 5-HT neurotransmission by determining the levels of 5-HT at the synapse and thus, the magnitude and duration of 5-HT responses. It is located predominantly on 5-HT axons and at periterminal sites, and thus its distribution and density are considered a measure of the number of 5-HT terminals and/or the efficiency of 5-HT reuptake in a given

Altered Serotonin Transporter Binding in Rett Syndrome

Paterson et al

J Neuropathol Exp Neurol  Volume 64, Number 11, November 2005

Immunocytochemistry for 5-HT Neurons Using the Tryptophan Hydroxylase Antibody

Quantitation of Tryptophan Hydroxylase-Immunopositive Cell Number Two standardized levels of the medulla were selected for cell counts. The midmedulla level was defined by the landmark of the nucleus of Roller just ventral to the hypoglossal nucleus and corresponds to Plate XII in the human brainstem atlas of Olszewski and Baxter (38). A rostral level was defined by the landmarks of the nucleus prepositus (rostral to the hypoglossal nucleus), gigantocellularis, and paragigantocellularis lateralis and corresponds to Plate XIV in the atlas of Olszewski and Baxter (38). At the midlevel, TPOH-immunopositive neurons were located in the midline (raphe´ obscurus and raphe´ pallidus) and in lateral sites (intermediate reticular zone and nucleus subtrigeminalis). At the rostral level, these cells were located in the midline (raphe´ obscurus and raphe´ pallidus) and in lateral sites (paragigantocellularis lateralis, intermediate reticular zone, and gigantocellularis). At both sites, scattered TPOH-immunopositive neurons were located in the arcuate nucleus along the ventral medullary surface. For ease in cell counting, the midline TPOH-immunoreactive neurons were collectively called ‘‘raphe´’’ neurons, and the lateral TPOHimmunoreactive neurons were called ‘‘extra-raphe´’’ neurons. The TPOH-immunoreactive cells were not counted at the ventral medullary surface as a result of their very low number in both the RTT and control cases. Cells were counted in 2 sections per case by one examiner, with every section counted twice and the mean value used for analysis.

Serotonin Transporter Autoradiography With 125I-RTI-55 Unfixed brainstem samples were stored frozen at –70°C until use, after receipt from the national tissue banks and hospital autopsy services. The brainstem samples were em-

1020

Developmental Analysis of Serotonin Transporter Binding From Human Infancy Through Adolescence To compare SERT binding in the human infant medulla with that in the child and adolescent medulla, we referred to the dataset of SERT binding in human infant cases in our laboratory and from which SERT-binding information has been published (39). The use of the same 125I standards in the autoradiographic exposures of the medullary sections and the same incubation and exposure protocols allowed us to make comparisons in the quantitative SERT-binding data between the 2 studies. Seven infant specimens were available, ranging in age from 1 to 47 weeks (median, 23 weeks) postnatal age. Linear regression analysis was used to determine the effect of age on 125I-RTI-55 binding in the DMX. Given the extraordinary difficulty in collecting normative brainstems from human cases in the pediatric age range, we believe continuing analysis of the datasets in the laboratory that address different questions is justified. q 2005 American Association of Neuropathologists, Inc.

Downloaded from http://jnen.oxfordjournals.org/ by guest on October 8, 2016

The TPOH immunocytochemistry was performed on 20-mm frozen sections when possible. However, frozen tissue was adequate only in 5 cases because brainstem preparations had been dissected off the midline. The remaining cases were available from paraffin-embedded tissues from archival RTT cases for which permission for study had been obtained at the time of autopsy. Thus, immunocytochemistry could not be performed in the exact same cases with which SERT autoradiography was performed (see subsequently). Of note, the RTT cases used for cell counting had either documented MeCP2 mutations or exhibited the classic phenotype clinically (data not shown). All sections (frozen and paraffin-embedded) were cut at 20 mm. Sections were processed for antigen retrieval by heating in citrate buffer (pH 6.0) (Zymed, South San Francisco, CA) for three 4-minute bursts at 195°F and then cooled to room temperature before incubation with the TPOH antibody. A monoclonal anti-mouse TPOH antibody (SigmaAldrich, St. Louis, MO) was applied overnight at room temperature in a 1:8,000 dilution. Chromagen staining was developed using DAB (DAKO Cytomation, Carpinteria, CA) according to standard laboratory procedures.

bedded in OCT and sectioned in the transverse plane at 20 mm on a Leica motorized cryostat. We measured the quantitative distribution of SERT-binding density in 9 nuclei in the medulla: 1) raphe´ obscurus; 2) raphe´ pallidus; 3) gigantocellularis (containing 5-HT neurons in the extra-raphe´); 4) paragigantocellularis lateralis (containing 5-HT neurons in the extra-raphe´ ); 5) arcuate nucleus (containing 5-HT neurons and receptors at the ventral medullary surface); 6) intermediate reticular zone (containing 5-HT neurons in the extra-raphe´); 7) nucleus of the solitary tract (viscerosensory component of the vagus nerve); 8) dorsal motor nucleus of cranial nerve X (DMX) (preganglionic parasympathetic outflow); and 9) hypoglossal nucleus (upper airway control). As established in our laboratory, the nuclei were analyzed at 2 levels based on the landmarks defined in the atlas of Olszewski and Baxter (38). Two sections were analyzed at each level. To measure SERT-binding density, we used tissue section autoradiography with the radioligand 25I-RTI-55 according to a previously described procedure in our laboratory (39). Unfixed, slide-mounted sections of the medulla were incubated in 0.15 nM 125I-RTI-55 (PerkinElmer-NEN, Boston, MA) in 50 mM Tris buffer (pH 7.4) containing 120 mM NaCl for 90 minutes at room temperature. Nonspecific binding was determined by addition of 100 nM citalopram hydrobromide (Tocris, St. Louis, MO) to the buffer. Sections were then dried before being placed in cassettes and exposed to 125I-sensitive film (Kodak Biomax MR, Rochester, NY) for 5 hours along with a set of 125I standards (Amersham Biosciences, Piscataway, NJ) for conversion of optical density of silver grains to fmol/mg tissue. The transporter binding density (expressed as the specific activity of tissue-bound ligand) was analyzed in 9 nuclei of the medulla specified previously; all nuclei were not available for every case. All steps of the analysis, including the measurements of SERT binding in individual nuclei, were made in a blinded fashion without knowledge of the diagnosis or age of the patient. Quantitative densitometry of autoradiograms was performed using an MCID 5+ imaging system (Imaging Research, Ontario, Canada).

J Neuropathol Exp Neurol  Volume 64, Number 11, November 2005

Statistical Analyses The Wilcoxon test was used to test for significant differences in cell number and 125I-RTI-55 binding (p , 0.05) between RTT and control cases, not adjusting for any effects of age. Analysis of covariance (ANCOVA) was used to test for significant differences in cell number and 125I-RTI-55 binding (p , 0.05) between RTT and control cases controlling for the effects of age and to identify any age–diagnosis interactions. When a significant age–diagnosis interaction was present, linear regression analysis for the effects of age were preformed separately for RTT and control cases.

Photomicrograph Production

RESULTS Clinical Database The number of RTT and control cases differed between the immunocytochemical and autoradiographic studies as a result of the availability of appropriately fixed tissues for each method. The available clinical information for the cases analyzed in the SERT-binding study, including the specific mutations in the MeCP2 gene in the RTT cases, is summarized in Table 1. Of note, there were 3 cases with the clinical diagnosis of RTT in which mutations were not found in exons 1–4 and which may represent cases with the rare deletions (Table 1). The diagnosis of RTT was made in these 3 cases on the basis that the patients were girls who presented the characteristic phenotype of RTT, including mental retardation. These 3 cases had frozen brainstem tissue available for SERT binding and were analyzed with the genetically defined RTT cases versus controls in a separate analysis (see subsequently). As a result of the unavailability of detailed information about the cardiorespiratory status of patients with RTT syndrome, correlations between cardiorespiratory parameters and findings in the medullary 5-HT system could not be made. The postmortem interval was less than 24 hours in all cases and did not adversely affect the 5-HT cell immunocytochemical study or SERT-binding analysis (data not shown).

Quantitation of 5-HT Cell Number Using Immunocytochemistry With the Tryptophan Hydroxylase Antibody We quantified TPOH-immunoreactive neurons in 11 RTT (ages 5–24 years; median, 12 years) and 7 controls (ages 3–26 years; median, 14 years) at a mid- and rostral level of the medulla. The relative distribution of the TPOH-immunoreactive q 2005 American Association of Neuropathologists, Inc.

neurons in raphe´ and extra-raphe´ regions of the medulla (Fig. 1) was essentially identical between the RTT and control cases. In addition, this distribution was identical to that found by us in human infants in the first postnatal year of life (26). There was no evidence of altered migration of TPOH-immunoreactive neurons or incomplete morphologic differentiation in the RTT compared with control cases. The number of TPOHimmunoreactive neurons did not change with increasing age in either the RTT or control cases. There were no statistically significant differences in cell number at the mid- or rostral medullary levels or in the combined levels (Table 2).

Serotonin Transporter Autoradiography of the Medulla With the Radioligand 125I-RTI-55 The Baseline Distribution of SERT-Binding Density in Childhood, Adolescence, and Early Adulthood Although the relative distribution of SERT binding in the human brainstem has been defined in the first postnatal year of life (39) and in adults, essentially nothing is known about it in childhood and adolescence. This paucity of information prompted us to analyze in detail the distribution and developmental profile in the 7 childhood and adolescent controls available to us in this study (Table 3). These controls ranged in age from 3 to 26 years, with a median of 14 years (ages 3, 9, 11, 14, 16, 20, and 26 years) (Table 1). The binding levels of the 9 nuclei analyzed did not change with increasing age except for levels in the DMX (Fig. 2A). In the DMX, the binding decreased over time (2.35 fmol/mg tissue per year, regression p value = 0.049) (Fig. 2A). Of note, measurements in the DMX were available for 5 of the 7 control cases and ranged in age from 3 to 16 years. Except for this nucleus, the relative distribution patterns of binding did not change over the age range analyzed. On average, the regions with the very highest SERT binding levels (>35 fmol/mg tissue) were the raphe´ obscurus and gigantocellularis, an extra-raphe´ nucleus containing, like the raphe´ nuclei, 5-HT cell bodies (Table 3). High binding (30–34 fmol/mg) was present in the hypoglossal nucleus and raphe´ pallidus; intermediate binding (20– 29 fmol/mg) was present in the nucleus of the solitary tract, paragigantocellularis lateralis, and intermediate reticular zone (Table 3). The mean binding in the DMX in the control cases, unadjusted for age, was 30.8 6 6.0 fmol/mg tissue, which indicates that overall binding is in the high category. Binding was negligible in the principal inferior olive and arcuate nucleus along the ventral medullary surface. The relative distribution of SERT binding was essentially the same across development, with no age-related changes observed across infancy (39) or childhood and adolescence except for the DMX. In the DMX, SERT binding remains constant across infancy (linear regression, p = 0.540) (Fig. 2B); however, between 2 and 3 years of age, it increases to a peak level that is maintained throughout childhood; and on reaching adolescence (12–13 years) once more falls to a level comparable to that in infancy (Fig. 2B). These observations suggest that the level of SERT binding observed in the DMX is dependent on age and that development of the DMX continues into childhood and adolescence.

1021

Downloaded from http://jnen.oxfordjournals.org/ by guest on October 8, 2016

Images of SERT binding in the human medulla were generated as TIFF files from the autoradiography film using the MCID 5+ imaging system (Imaging Research). The binding levels in the images were normalized to allow comparison of binding density between cases. All images were then imported into PhotoShop 6.0 (Adobe Systems, San Jose, CA) where they were scaled relative to each other. Appropriate labels were added to form composite images.

Altered Serotonin Transporter Binding in Rett Syndrome

Paterson et al

J Neuropathol Exp Neurol  Volume 64, Number 11, November 2005

Comparison of Serotonin Transporter Binding in Selected Medullary Nuclei Between Rett Syndrome and Control Cases The major abnormality found by us in the RTT cases compared with controls was in the DMX. In this site, SERT binding for the control cases (n = 5) decreased significantly over time (2.35 fmol/mg tissue per year, linear regression p = 0.049), but did not change significantly in the RTT cases (n = 7) (0.25 fmol/mg tissue per year, linear regression p = 0.513) (Fig. 2A). Adjusting for this effect of age, binding is significantly lower in this nucleus in RTT compared with control cases (ANCOVA, p = 0.022) (Fig. 3); there was also a marginally significant age versus diagnosis interaction (p = 0.06) (Table 3), indicating that age differentially influences binding in RTT and control cases. When the 3 cases with the RTT phenotype but incomplete genetic testing for deletions in the MeCP2 gene were added to the RTT group (see previously), the age versus diagnosis interaction became significant (ANCOVA, p = 0.028) with a sample of 9 RTT cases and 5 controls. The effect of age on SERT binding in RTT and controls, therefore, is significantly different, i.e. decreasing with age in controls, while increasing very slightly with age in RTT. A significant age versus diagnosis interaction was noted in the raphe´ obscurus (ANCOVA, p = 0.046) (Table 3), indicating that there is a similar differential effect of age on binding in RTT and controls cases in this nucleus, with a marginally significant difference in age-adjusted binding between the RTT and control cases (p = 0.074). When the 3 RTT-like cases were added to the RTT group, the findings were essentially the same, especially considering the age versus diagnosis interaction (ANCOVA, p = 0.021). The biologic significance of this latter finding is uncertain given that regressions on age for this nucleus were not significant for either the RTT cases (p = 0.101) or controls (p = 0.180). There were no significant differences between the RTT and control cases in any of the other nuclei examined (Table 3).

1022

Comparison of Serotonin Transporter Binding in Rett Syndrome Relative to Baseline (control) Serotonin Transporter Expression During Development To provide further insight into the developmental abnormalities in SERT binding in the DMX in RTT, we compared this binding to the binding in infants and childhood– adolescence in the control cases. During development, SERT binding in the DMX remains constant across infancy, increases to peak during childhood, and subsequently falls again during adolescence (Fig. 2B). In RTT, SERT binding in the DMX also appears to increase with age during childhood, but this increase begins at an older age (i.e. 8 years vs 3 years, respectively, for RTT and controls), peaks at a lower level than in controls (i.e. 29 fmol/mg vs 45 fmol/mg, respectively, for RTT and controls), and does not appear to display a significant age-related reduction thereafter. SERT binding in the DMX in RTT is also higher on average than that observed in infant controls, such that it appears to be midway between infant and child levels (Fig. 2B). Although we have no data on SERT

TABLE 2. Cell Counts of Tryptophan HydroxylaseImmunoreactive Neurons in Rett Syndrome (RTT) and Control Cases at 2 Standardized Medullary Levels in the Raphe´ and Extra-Raphe´ Regions Combined Mean (SE) Medullary Level

RTT (n = 11)

Controls (n = 7)

p Value*

Midmedulla Rostral medulla Combined levels

89.5 (11.5) 115.4 (16.8) 214.8 (25.3)

80.7 (6.8) 101.5 (16.7) 186.7 (12.3)

NS NS NS

No significant difference in 5-HT cell number was observed in mid-, rostral, or the combined total of the levels between Rett and control cases. *, Wilcoxon test; SE, standard error of the mean; NS, not significant at p , 0.05 level.

q 2005 American Association of Neuropathologists, Inc.

Downloaded from http://jnen.oxfordjournals.org/ by guest on October 8, 2016

FIGURE 1. Distribution and morphology of medullary raphe´ and extra raphe´ 5-HT neurons. (A) Distribution of tryptophan hydroxylase (TPOH)-immunoreactive 5-HT neurons in cells in the midline raphe´ and (B) lateral extra-raphe´ region in human midmedulla (43). (C) Highpower photomicrograph of human medullary 5-HT neuron (403). There was no difference in relative distribution or morphology of medullary 5-HT neurons in Rett syndrome and control cases. Arrows indicated location of TPOH-immunoreactive 5-HT neurons.

J Neuropathol Exp Neurol  Volume 64, Number 11, November 2005

TABLE 3. 125I-RTI-55 Binding to Serotonin Transporter in Rett Syndrome (RTT) and Controls Cases Nucleus DMX HG NTS Rob ARC GC PGCL IRZ RPa

RTT

Controls

Diagnosis p Value

Age p Value

Interaction p Value

NA 32.3 (4.1) 18.4 (3.1) NA 6.0 (1.0) 29.1 (6.6) 25.3 (3.6) 23.3 (3.8) 20.2 (26.1)

NA 34.3 (4.8) 23.6 (3.4) NA 5.9 (1.0) 42.4 (6.6) 27.7 (3.6) 26.4 (3.8) 30.3 (26.1)

0.022 NS NS 0.074 NS NS NS NS NS

0.018 NS NS NS NS NS NS NS NS

0.060 NS NS 0.046 NS NS NS NS *

expression in RTT infants, these observations suggest that SERT-binding expression in the DMX in RTT is arrested at an immature or ‘‘infant-like’’ level. Comparison of binding in RTT cases to that in control infants and children–adolescents revealed that the DMX was the only nucleus in which this developmental pattern of binding was observed (data not shown).

DISCUSSION The major finding of this study is a significant alteration in SERT binding in the DMX in patients with RTT syndrome dying in childhood or adolescence compared with control cases dying in the same age range. This finding is of potential importance because abnormalities of the DMX, the preganglionic origin of the parasympathetic nervous system through the vagus nerve, may explain, at least in part, autonomic dysfunction in patients with RTT. In this study, we found that SERT binding in the DMX of control cases decreased significantly during adolescence from a peak during childhood but remained relatively constant over the same period in the RTT cases. Adjusting for age, SERT binding between the RTT and control cases differed significantly in this nucleus, and there was a marginally significant age versus diagnosis interaction, indicating a differential effect of age on binding in RTT and control cases. A significant diagnosis versus age interaction was also noted in the raphe´ obscurus, a key nucleus involved in the modulation of respiratory, autonomic, pain, and temperature regulation through projections to both sympathetic and parasympathetic preganglionic targets, including the DMX. In the raphe´ obscurus, we also found a marginally significant difference in age-adjusted binding between the RTT and control cases. We are uncertain of the q 2005 American Association of Neuropathologists, Inc.

biologic significance of the findings in the raphe´ obscurus, however, given that regressions on age for this nucleus were not significant for either the RTT cases or controls. Yet, in both the DMX and raphe´ obscurus, the pattern of SERT binding appeared more ‘‘infant-like’’ than ‘‘child/adolescent-like,’’ i.e. developmentally delayed. In contrast, there were no significant-binding differences in the 7 other nuclei sampled that contain the source 5-HT neurons of the medullary 5-HT system, and/or receive projections from them, and are involved in central respiratory and/or autonomic control, i.e. hypoglossal nucleus (upper airway control), nucleus of the solitary tract (visceral sensory input). In addition, there was no significant difference in the number of TPOH-immunoreactive neurons (5-HT source neurons) in the raphe´ and extra-raphe´ regions in standardized levels between the RTT and control medullae. Given that SERT is localized to the perisynaptic terminals of 5-HT neurons, its expression can be interpreted as a marker of 5-HT innervation of a particular nucleus. Because SERT is responsible for 5-HT reuptake into the presynaptic terminals of 5-HT neurons, its expression can also be interpreted as a measure of 5-HT reuptake efficiency. Thus, the finding of altered SERT binding in the DMX in patients with RTT without an alteration in the number of 5-HT neurons that project to the DMX (i.e. medullary raphe´ ) suggests that altered 5-HT innervation and/or uptake in the DMX contributes to abnormal 5-HT modulation of parasympathetic function and secondary autonomic dysfunction in these patients.

Baseline Development of Serotonin Transporter Binding in Childhood and Adolescence A significant age-related reduction in SERT binding was observed across childhood and adolescence in the DMX and raphe´ obscurus, but not in any of the other 7 medullary nuclei analyzed. Of note, lipid quenching is not a factor in the developmental analysis because we used an iodinated, and not tritiated, radioligand for analysis of SERT binding. The functional significance of the reduction of SERT binding with increasing age in the DMX and raphe´ obscurus is completely unknown. Nevertheless, the neurochemical change implies a functional change, and its occurrence during childhood and adolescence suggests that 5-HT-related function in the DMX and raphe´ obscurus is still undergoing maturation. The SERTbinding data in these 2 sites suggest that 1) 5-HT terminals are ‘‘pruned back’’ across childhood and adolescence, and thereby decrease in number; and/or 2) the number of SERT proteins decrease over the same period without necessarily a loss of the 5-HT terminals themselves. These processes, either singly or in combination, likely cause developmental modifications in autonomic function regulated by the DMX and raphe´ obscurus in a critical period that extends beyond early life into childhood. In essence, MeCP2- and BDNF-related regulatory programs are likely important for 5-HT maturation in the DMX and raphe´ obscurus beyond embryonic and fetal epochs of the proliferation and migration of 5-HT neurons, well into the epoch of the organization of appropriately matched synaptic terminals, transporter, and receptor proteins.

1023

Downloaded from http://jnen.oxfordjournals.org/ by guest on October 8, 2016

Age-adjusted 125I RTI-55 binding to SERT (mean 6 standard error of mean fmol/mg) to 9 medullary nuclei in Rett and control cases. Binding values are presented for each nucleus sampled, as determined by analysis of covariance. NA, not available—the ageadjusted means are not provided when the age versus diagnosis interaction is significant because the difference in means depends on the specific age in question. *, Not enough cases to fit a model with an interaction; NS, not significant; p , 0.05 is considered significant; 0.05 . p , 0.10 is considered marginally significant. DMX, dorsal motor nucleus of cranial nerve X; HG, hypoglossal nucleus; NTS, nucleus of the solitary tract; Rob, raphe´ obscurus; ARC, arcuate nucleus; GC, gigantocellularis; PGCL, paragigantocellularis lateralis; IRZ, intermediate reticular zone; RPa, raphe´ pallidus.

Altered Serotonin Transporter Binding in Rett Syndrome

Paterson et al

J Neuropathol Exp Neurol  Volume 64, Number 11, November 2005

to reach this peak at a later age, and does not display a significant age-related reduction thereafter. In the absence of SERT-binding data for RTT cases in infancy, it is impossible to determine if the attenuation of SERT expression observed in RTT cases in childhood and adolescence becomes manifest only after infancy (i.e. on entering childhood) or if it is also present during gestation and infancy. The question arises in this study, why is the abnormality in SERT binding restricted to certain medullary regions and not all regions sampled? The answer is of course unknown, but it is noteworthy that the 2 nuclei with abnormal SERT binding in the patients with RTT syndrome, i.e. DMX and raphe´ obscurus, were nuclei in which the pattern of binding was decreasing over the age range analyzed, suggesting that the affected nuclei in childhood and adolescence were those still undergoing maturation.

FIGURE 2. Serotonin transporter (SERT) binding in the dorsal motor nucleus of cranial nerve X (DMX) changes differently with age in Rett syndrome (RTT) compared with controls. (A) Age-related distribution of 125I-RTI-55 binding (fmol/mg) to SERT in the DMX of RTT and control cases from childhood through adolescence. Binding decreases significantly with age (p = 0.049, linear regression) in the DMX of control patients, but does not change with age in patients with Rett syndrome (p = 0.513, linear regression). Furthermore, binding in the DMX is significantly lower in patients with RTT syndrome when age is taken into consideration (p = 0.022, analysis of covariance). (B) Age-related distribution of 125I-RTI-55 binding (fmol/mg) to SERT in the DMX of RTT and control cases from infancy through adolescence. During development, SERT binding in the DMX remains constant across infancy (linear regression, p = 0.540), increases to peak during childhood, and subsequently falls again during adolescence. SERT binding in the DMX in RTT is midway between control infant and adolescent levels, suggesting retarded development.

Serotonin Transporter Binding in Rett Syndrome Relative to Normative Distribution of Serotonin Transporter in Human Infancy Through Adolescence In an effort to provide further insight into the developmental abnormalities in SERT binding in RTT, we compared binding in DMX to the developmental expression of SERT during infancy and childhood/adolescence in the DMX of the normative medulla. During normative development, SERT binding in the DMX remains constant across infancy, increases to a peak during childhood, and subsequently falls again during adolescence. In RTT, SERT binding in the DMX peaks at a lower density than controls during childhood, appears

1024

A spectrum of respiratory and autonomic abnormalities has been reported in patients with Rett syndrome at different ages. The respiratory abnormalities include episodic hyperventilation at all ages (40), apneustic breathing prominent among 5- to 10-year-old patients, and Valsalva breathing prominent in patients over 18 years (29). Autonomic manifestations include flushing, pupillary dilatation, constipation, and cold extremities relieved by sympathectomy (40). In addition, low resting cardiac vagal tone, reduced cardiac sensitivity to the baroreceptor reflex, and weak vagal responses to hyperventilation and breathholding have been reported, suggesting inadequate parasympathetic control (41). Moreover, a loss of physiological heart rate variability has been reported, suggesting exaggerated sympathetic tone (41). The alteration in heart rate variability is progressive with age and is postulated to predispose to ventricular arrhythmias and sudden death observed in patients with Rett syndrome (41). Plasma levels of 5-HT are lower in untreated patients with Rett syndrome compared with patients with Rett syndrome who receive anticonvulsants, with correlation between plasma 5-HT levels and sympathetic overactivity (as measured by the lowfrequency to high-frequency ratio in the assessment of heart rate variability) observed. This observation suggests that the 5-HT system is specifically involved in the pathogenesis of the cardiac dysautonomia in RTT. Our finding of an imbalance in SERT binding at different ages in the DMX in patients with Rett syndrome compared with controls is directly relevant to the clinical evidence for reduced parasympathetic tone and increased sympathetic tone in RTT, given that the DMX is the preganglionic origin of parasympathetic outflow. The DMX innervates through the vagal nerve the lungs and bronchi, gastrointestinal tract from the esophagus through the small intestine, liver, gallbladder, and pancreas; it does not, however, innervate blood vessels, and thus does not play a role in blood pressure control. Although the nucleus ambiguus was previously considered the main origin of the parasympathetic innervation of the heart, recent morphologic evidence suggests a dual innervation of this organ from the DMX and nucleus ambiguus (both vagal q 2005 American Association of Neuropathologists, Inc.

Downloaded from http://jnen.oxfordjournals.org/ by guest on October 8, 2016

Altered Serotonin Transporter Binding in the Dorsal Motor Nucleus of Cranial Nerve X in Patients With Rett Syndrome and Cardiorespiratory Dysfunction

J Neuropathol Exp Neurol  Volume 64, Number 11, November 2005

Altered Serotonin Transporter Binding in Rett Syndrome

FIGURE 3. Color-coded autoradiograms of 125I-RTI-55 binding to serotonin transporter (SERT) in medullary tissue sections at the level of the dorsal motor nucleus of cranial nerve X (DMX) in Rett syndrome (RTT) and controls. (A) Cartoon of a human midmedulla section showing the localization of the DMX (box B). (B) An enlargement of the box in (A), which corresponds to the area covered by the autoradiograms in (C) and (D) illustrating the location of the DMX in relation to the hypoglossal nucleus (HG) and nucleus of the solitary tract (NTS). SERT binding in the region of the DMX in control cases (C) is significantly higher than in RTT cases (D) during childhood (p = 0.022, analysis of covariance).

motor subnuclei) in the control of heart rate and ventricular contractility (42, 43). Indeed, stimulation of the DMX elicits bradycardia and reduces myocardial contractility (42), and stimulation of vagal fibers from both the DMX and nucleus ambiguus evokes bradycardia, arterioventricular block, and reduced myocardial contractility (42, 43), whereas the nucleus ambiguus appears especially critical for the baroreceptor reflex (42, 43). The DMX and nucleus ambiguous appear to innervate the same intrinsic cardiac ganglia, but their axons target separate, nonoverlapping subpopulations of interspersed ganglionic neurons (44). Because the DMX and nucleus ambiguus project to distinct subpopulations of cardiac principal neurons, it is postulated that these 2 nuclei play different roles in controlling cardiac function given that neurons in different cardiac ganglia control heart rate, arterioventricular conduction, and myocardial contractility (44). Of major relevance to the findings reported here in patients with Rett syndrome, microinjections of 5-HT1A receptor agonists into the DMX produce bradycardia in rats pretreated q 2005 American Association of Neuropathologists, Inc.

Postulated Factors in the Pathogenesis of Altered Serotonin Transporter Binding in Rett Syndrome MeCP2 is an activity-dependent inhibitor of BDNF gene transcription (9). Mutations in the MeCP2 gene produce a loss of protein function and in neurons under resting (basal) conditions, a 2-fold overexpression of BDNF (7, 9). The association of the majority of RTT cases with MeCP2 gene mutations suggests that there is a chronic overexpression of BDNF in RTT. BDNF is a potent 5-HT growth factor, but also plays a role in neuronal plasticity, destabilizing the cytoskeleton allowing it the ‘‘fluidity’’ necessary for neurite growth and extension (22, 23, 48). Overexpression of BDNF in RTT, therefore, may potentially cause not only a general disruption of 5-HT neuronal development, but also a specific abnormality in 5-HT synapse formation and thus the altered SERT expression observed in this study. Recent evidence suggests that MeCP2 may also regulate the transcription of the neurotrophic factor S100b (49). Significantly, S100b promotes 5-HT neuronal maturation and neurite formation by stabilizing the cytoskeleton, i.e. in opposition to the effects of BDNF (17, 18, 50). Thus, MeCP2-induced abnormalities in S100b expression could also putatively contribute to aberrant 5-HT synapse formation and SERT binding in RTT. Interestingly, synthesis and release of S100b in astrocytes is stimulated by BDNF and 5-HT1A receptor activation, respectively (17, 18, 50). Maintenance of appropriate levels of BDNF and 5-HT is therefore likely to be necessary for optimal 5-HT neuronal growth and synapse formation during development. We propose that MeCP2 mutations in RTT disrupt the levels of BDNF (and potentially S100b) resulting in abnormal 5-HT neuron development and synapse formation, which is putatively responsible for the age-related alterations in SERT expression observed in this study.

1025

Downloaded from http://jnen.oxfordjournals.org/ by guest on October 8, 2016

with atenolol, a drug that blocks the effects of any changes in sympathetic drive to the heart (45), and intravenous injections of 5-HT1A agonists cause large increases in parasympathetic cardiac drive (45). Microinjections of a 5-HT1A agonist into the DMX also causes significant increases in phrenic nerve activity in rats, suggesting that activation of 5-HT1A receptors in this nucleus results in an increase in central respiratory drive (46). These observations indicate that central excitatory 5-HT pathways through 5-HT1A receptors are involved in the control of parasympathetic tone to the heart (45) and are critical for increases in central respiratory drive (46, 47). 5-HT neurons in the caudal raphe´ are known to project to the DMX and nucleus of the solitary tract; the latter site contains 5-HT1A receptors. The finding in our laboratory that 5-HT receptors, including the 5-HT1A subtype, by tissue section autoradiography are located in the DMX of the human child and adolescent (unpublished observations) support these concepts that are based on animal data. The precise mechanism(s) whereby altered 5-HT input in the DMX causes abnormal autonomic function in affected patients with Rett syndrome is unknown, but the identification of a key autonomic-related region and neurotransmitter system directly affected in patients with Rett syndrome is an important first step toward deciphering autonomic pathophysiology in this disorder.

J Neuropathol Exp Neurol  Volume 64, Number 11, November 2005

Paterson et al

Potential Impact of Hypoxic-Ischemic Injury on Serotonin Transporter Binding

Conclusions Although this study is limited by an unavoidably small sample size of autopsied RTT and control brainstem specimens, it suggests that 5-HT modulation of the DMX is abnormal in patients with Rett syndrome in childhood and adolescence and may contribute to autonomic dysfunction in this disorder. These data from rare autopsied cases of RTT generate hypotheses for testing in the MeCP2 knockout mice to understand the role of abnormal 5-HT regulation of the DMX in causing abnormal heart rate variability and other autonomic dysfunction in the patients, as well as the underlying molecular mechanisms involving MeCP2, BDNF, and 5-HT.

REFERENCES 1. Trevathan E, Naidu S. The clinical recognition and differential diagnosis of Rett syndrome. J Child Neurol 1988;3(suppl):S6–16 2. Hagberg B, Aicardi J, Dias K, et al. A progressive syndrome of autism, dementia, ataxia, and loss of purposeful hand use in girls: Rett’s syndrome: Report of 35 cases. Ann Neurol 1983;14:471–79 3. Weaving LS, Ellaway CJ, Gecz J, et al. Rett syndrome: Clinical review and genetic update. J Med Genet 2005;42:1–7 4. Van den Veyver IB, Zoghbi HY. Mutations in the gene encoding methylCpG-binding protein 2 cause Rett syndrome. Brain Dev 2001;23(suppl 1): S147–51 5. Amir RE, Van den Veyver IB, Wan M, et al. Rett syndrome is caused by mutations in X-linked MECP2, encoding methyl-CpG-binding protein 2. Nat Genet 1999;23:185–88 6. Jugloff DG, Jung BP, Purushotham D, et al. Increased dendritic complexity and axonal length in cultured mouse cortical neurons overexpressing methyl-CpG-binding protein MeCP2. Neurobiol Dis 2005;19: 18–27

1026

q 2005 American Association of Neuropathologists, Inc.

Downloaded from http://jnen.oxfordjournals.org/ by guest on October 8, 2016

The incidence of conditions with agonal phases associated with hypoxic–ischemic injury in both the control and RTT cases in this study raises the possibility that hypoxic– ischemic insult may have affected SERT-binding values. Studies of this type on human postmortem brain tissue are chronically hampered by the collection of appropriate control tissue, because the agonal phase of any control case could potentially influence the measurements being made. Furthermore, the unavoidable heterogeneity in the terminal pathology of controls introduces a multitude of potential influences that may differ from one control case to another. It is therefore difficult to identify for certain the specific effects of any one influence over another, particularly in a study such as this in which there is an unavoidably low sample number. No examples exist in the literature examining the effect of hypoxia–ischemia on SERTbinding values in humans or animals; we cannot, therefore, state for certain what influence hypoxia–ischemia may have on SERT-binding values in this study. However, the SERTbinding density values measured in the controls in this study are consistent with our measurements of SERT binding in the medullae of control cases in which the incidence of hypoxia– ischemia was not appreciably associated with the agonal phase (39). We propose, therefore, that hypoxic–ischemic injury is unlikely to have had a significant effect on SERT binding in this study.

7. Martinowich K, Hattori D, Wu H, et al. DNA methylation-related chromatin remodeling in activity-dependent BDNF gene regulation. Science 2003;302:890–93 8. Stancheva I, Collins AL, Van den Veyver IB, et al. A mutant form of MeCP2 protein associated with human Rett syndrome cannot be displaced from methylated DNA by notch in Xenopus embryos. Mol Cell 2003;12: 425–35 9. Chen WG, Chang Q, Lin Y, et al. Derepression of BDNF transcription involves calcium-dependent phosphorylation of MeCP2. Science 2003; 302:885–89 10. Erickson JT, Conover JC, Borday V, et al. Mice lacking brain-derived neurotrophic factor exhibit visceral sensory neuron losses distinct from mice lacking NT4 and display a severe developmental deficit in control of breathing. J Neurosci 1996;16:5361–71 11. Spenger C, Hyman C, Studer L, et al. Effects of BDNF on dopaminergic, serotonergic, and GABAergic neurons in cultures of human fetal ventral mesencephalon. Exp Neurol 1995;133:50–63 12. Lauder JM. Neurotransmitters as morphogens. Progr Brain Res 1988;73: 365–87 13. Lauder JM. Neurotransmitters as morphogens. In: Boer GJ, Feenstra MGP, Mirmiran M, et al., eds. Biochemical Basis of Functional Neuroteratology: Permanent Effects of Chemicals on the Developing Brain. Amsterdam: Elsevier, 1988:365–87 14. Lauder JM. Ontogeny of the serotonergic system in the rat: Serotonin as a developmental signal. Ann N YAcad Sci 1990;600:297–313; discussion 314 15. Azmitia EC, Whitaker-Azmitia PM. Developmental and neuroplasticity of central serotonergic neurons. In: Baumgarten HG, Gothert M, eds. Handbook of Experimental Pharmacology: Serotonergic Neurons and 5-HT Receptors in the CNS. Berlin, Heidelberg, New York: SpringerVerlag, 1997:1–39 16. Azmitia EC. Serotonin neurons, neuroplasticity, and homeostasis of neural tissue. Neuropsychopharmacology 1999;21(suppl):33S–45S 17. Nishiyama H, Knopfel T, Endo S, et al. Glial protein S100B modulates long-term neuronal synaptic plasticity. Proc Natl Acad Sci U S A 2002;99: 4037–42 18. Nishiyama H, Takemura M, Takeda T, et al. Normal development of serotonergic neurons in mice lacking S100B. Neurosci Lett 2002;321: 49–52 19. Pollock GS, Robichon R, Boyd KA, et al. TrkB receptor signaling regulates developmental death dynamics, but not final number, of retinal ganglion cells. J Neurosci 2003;23:10137–45 20. Rothermundt M, Peters M, Prehn JH, et al. S100B in brain damage and neurodegeneration. Microsc Res Tech 2003;60:614–32 21. Mamounas LA, Blue ME, Siuciak JA, et al. Brain-derived neurotrophic factor promotes the survival and sprouting of serotonergic axons in rat brain. J Neurosci 1995;15:7929–39 22. Galter D, Unsicker K. Brain-derived neurotrophic factor and trkB are essential for cAMP-mediated induction of the serotonergic neuronal phenotype. J Neurosci Res 2000;61:295–301 23. Galter D, Unsicker K. Sequential activation of the 5-HT1(A) serotonin receptor and TrkB induces the serotonergic neuronal phenotype. Mol Cell Neurosci 2000;15:446–55 24. Eaton MJ, Staley JK, Globus MY, et al. Developmental regulation of early serotonergic neuronal differentiation: The role of brain-derived neurotrophic factor and membrane depolarization. Dev Biol 1995;170:169–82 25. Merlio JP, Ernfors P, Jaber M, et al. Molecular cloning of rat trkC and distribution of cells expressing messenger RNAs for members of the trk family in the rat central nervous system. Neuroscience 1992;51: 513–32 26. Kinney HC, Belliveau RA, Rava LA, et al. The medullary serotonergic system in early human life. Part II. Developmental changes in serotonergic neuronal topography. J Comp Neurol, in press 27. Kerr CL, Ito Y, Manwell SE, et al. Effects of surfactant distribution and ventilation strategies on efficacy of exogenous surfactant. J Appl Physiol 1998;85:676–84 28. Julu PO, McCarron MO, Hansen S, et al. Selective defect of baroreflex blood pressure buffering with intact cardioinhibition in a woman with familial aniridia. Neurology 1997b;49:1705–8 29. Julu PO, Kerr AM, Hansen S, et al. Functional evidence of brain stem immaturity in Rett syndrome. Eur Child Adolesc Psychiatry 1997;6(suppl 1): 47–54

J Neuropathol Exp Neurol  Volume 64, Number 11, November 2005

q 2005 American Association of Neuropathologists, Inc.

41. Guideri F, Acampa M, Blardi P, et al. Cardiac dysautonomia and serotonin plasma levels in Rett syndrome. Neuropediatrics 2004;35:36–38 42. Cheng Z, Zhang H, Guo SZ, et al. Differential control over postganglionic neurons in rat cardiac ganglia by NA and DmnX neurons: Anatomical evidence. Am J Physiol Regul Integr Comp Physiol 2004; 286:R625–33 43. Persson PB, Armour JA. Dual vagal cardiac efferent pathways. Am J Physiol Regul Integr Comp Physiol 2004;286: R624 44. Cheng Z, Powley TL. Nucleus ambiguus projections to cardiac ganglia of rat atria: An anterograde tracing study. J Comp Neurol 2000;424:588– 606 45. Sporton SC, Shepheard SL, Jordan D, et al. Microinjections of 5-HT1A agonists into the dorsal motor vagal nucleus produce a bradycardia in the atenolol-pretreated anaesthetized rat. Br J Pharmacol 1991;104:466–70 46. Teng YD, Bingaman M, Taveira-DaSilva AM,. et al. Serotonin 1A receptor agonists reverse respiratory abnormalities in spinal cord-injured rats. J Neurosci 2003;23:4182–89 47. Lalley PM, Bischoff AM, Richter DW. 5-HT-1A receptor-mediated modulation of medullary expiratory neurones in the cat. J Physiol 1994;476: 117–30 48. Rumajogee P, Madeira A, Verge D, et al. Up-regulation of the neuronal serotoninergic phenotype in vitro: BDNF and cAMP share Trk Bdependent mechanisms. J Neurochem 2002;83:1525–28 49. Namihira M, Nakashima K, Taga T. Developmental stage dependent regulation of DNA methylation and chromatin modification in an immature astrocyte specific gene promoter. FEBS Lett 2004;572:184–88 50. Djalali S, Holtje M, Grosse G, et al. Effects of brain-derived neurotrophic factor (BDNF) on glial cells and serotonergic neurones during development. J Neurochem 2005;92:616–27

1027

Downloaded from http://jnen.oxfordjournals.org/ by guest on October 8, 2016

30. Wilken B, Lalley P, Bischoff AM, et al. Treatment of apneustic respiratory disturbance with a serotonin-receptor agonist. J Pediatr 1997;130:89–94 31. Panigrahy A, Filiano J, Sleeper LA, et al. Decreased serotonergic receptor binding in rhombic lip-derived regions of the medulla oblongata in the sudden infant death syndrome. J Neuropathol Exp Neurol 2000;59:377–84 32. Baker-Herman TL, Mitchell GS. Phrenic long-term facilitation requires spinal serotonin receptor activation and protein synthesis. J Neurosci 2002;22:6239–46 33. Balkowiec A, Katz DM. Brain-derived neurotrophic factor is required for normal development of the central respiratory rhythm in mice. J Physiol 1998;510:527–33 34. Kaufmann WE, Taylor CV, Hohmann CF, et al. Abnormalities in neuronal maturation in Rett syndrome neocortex: Preliminary molecular correlates. Eur Child Adolesc Psychiatry 1997;6(suppl 1):75–77 35. Nomura Y, Segawa M, Higurashi M. Rett syndrome—an early catecholamine and indolamine deficient disorder? Brain Dev 1985;7:334–41 36. Riederer P, Brucke T, Sofic E, et al. Neurochemical aspects of the Rett syndrome. Brain Dev 1985;87:351–60 37. Vanhala R, Korhonen L, Mikelsaar M, et al. Neurotrophic factors in cerebrospinal fluid and serum of patients with Rett syndrome. J Child Neurol 1998;13:429–33 38. Olszewski J, Baxter D. Cytoarchitechture of the Human Brain Stem. Basel: S. Karger, 1954 39. Paterson DS, Belliveau RA, Trachtenberg F, et al. Differential development of 5-HT receptor and the serotonin transporter binding in the human infant medulla. J Comp Neurol 2004;472:221–31 40. Julu PO, Kerr AM, Apartopoulos F, et al. Characterisation of breathing and associated central autonomic dysfunction in the Rett disorder. Arch Dis Child 2001;85:29–37

Altered Serotonin Transporter Binding in Rett Syndrome