|Home | About | Journals | Submit | Contact Us | Français|
Fragile X syndrome (FXS), an inherited disorder characterized by mental retardation and autism-like behaviors, is caused by the failure to transcribe the gene for fragile X mental retardation protein (FMRP), a translational regulator and transporter of select mRNAs. FXS model mice (Fmr1 KO mice) exhibit impaired neuropeptide release. Release of biogenic amines does not differ between wild-type (WT) and Fmr1 KO mice. Rab3A, an mRNA cargo of FMRP involved in the recruitment of vesicles, is decreased by ~50% in synaptoneurosomes of Fmr1 KO mice; however, the number of dense-core vesicles (DCVs) does not differ between WT and Fmr1 KO mice. Therefore, deficits associated with FXS may reflect this aberrant vesicle release, specifically involving docking and fusion of peptidergic DCVs, and may lead to defective maturation and maintenance of synaptic connections.
Fragile X syndrome (FXS) is the most common cause of inherited mental retardation, with an incidence of 1 in 4000 males and 1 in 8000 females (1). The molecular mechanism underlying FXS involves the expansion of an unstable polymorphic CGG triplet repeat to >200 in the 5′ untranslated region of the Fmr1 gene (2,3). This results in hypermethylation of the Fmr1 promoter and silencing of the Fmr1 gene, with consequent absence of the protein product, fragile X mental retardation protein (FMRP) (3). FMRP is an mRNA-binding protein that is enriched at the synapse and known to regulate the transport and localized translation of mRNA in response to mGluR receptor activation (4). Many of the mRNA cargoes associated with FMRP encode proteins crucial for spine maturation and synaptic plasticity (5−8).
In the absence of FMRP, there is defective regulation of localized mRNA translation. This absence affects synaptic plasticity in FXS, with abnormalities in long-term potentiation (LTP) and long-term depression (LTD) (9,10) in Fmr1 knockout (KO) mice, which exhibit characteristics of FXS (11,12). The absence of FMRP should lead to dysregulated local protein levels in both axons and dendrites, but previous reports have focused largely on translational regulation deficits at the postsynaptic site in FXS. Studies by Hanson and Madison (13) and Lauterborn et al. (14) recently suggested possible presynaptic effects caused by the loss of FMRP, prompting us to examine neuropeptide release in FXS. The mRNA cargoes of FMRP include presynaptic proteins that participate in the secretory pathway, in particular, vesicle exocytosis (5,6). One such protein is Rab3A, a GTPase that cycles between a soluble Rab3A-GDP form and a vesicle membrane-bound Rab3A-GTP form and is involved in activity-dependent vesicle docking and fusion at the synapse (15,16). Changes in Rab3A levels would be expected to affect activity-dependent release of transmitters and modulators. Using Western blot analyses, we characterized the levels of this protein in wild-type (WT) and Fmr1 KO mice.
Next, we used matrix-assisted laser desorption/ionization (MALDI) mass spectrometry (MS) to examine synaptoneurosomal preparations and probed the physiology of stimulus-evoked neuropeptide release in Fmr1 KO mice using live brain slices. Our results indicate that these mice are markedly deficient in neuropeptide release. In order to determine whether the neuropeptide release deficit in Fmr1 KO mice is a general deficit in dense-core vesicle (DCV) release, we used electrochemistry to examine the release of biogenic amines. We show that the release deficit is specific to peptides, because there is no significant difference in the release of dopamine (DA), serotonin (5-HT), and norepinephrine (NE) from the striatum of WT and Fmr1 KO mice. Lastly, using electron microscopy to quantify the number of peptide-housing DCVs, we do not observe significant differences between WT and Fmr1 KO mice, again suggesting a specific release deficit in FXS.
We characterized the amount of Rab3A at the synapse using synaptoneurosomal preparations, which are enriched in intact pinched-off synaptic processes, from postnatal 10−14 day (P10−14) WT and Fmr1 KO mice. Using Western blot analysis, we found Rab3A expression to be reduced by ~50% in isolated cortical synapses of Fmr1 KO mice (n = 8) compared with that in WT mice (n = 8; p < 0.01) (Figure (Figure1a,b).1a,b). The amount of Rab3A in total cortical homogenates was also decreased in the Fmr1 KO mice, however, to a smaller extent, because total homogenates contain more somatic material (Figure (Figure1c,1c, n = 4, WT; n = 4, Fmr1 KO; p < 0.01).
The decrease in Rab3A may be accompanied by a decrease in Rab3 interacting proteins. Liao et al. (17) recently reported reduced protein levels of other Rab isoforms in Fmr1 KO mice, including several proteins involved in vesicle exocytosis. Rab3A KO mice have previously been shown to have altered activity-dependent vesicle release (15) and complete loss of LTP at the hippocampal CA3 mossy fiber synapses (18). Although the mechanism behind Rab3A function in FXS is currently unknown, we speculate that there may be an imbalance in Rab3A-GDP and Rab3A-GTP cycling. It has previously been shown that an accumulation of membrane-bound Rab3A-GTP leads to a reduction in docked DCVs (19). Van Weering et al. (20) have observed that neither Rab3A-GTP- nor Rab3A-GDP-locked mutant chromaffin cells exhibited any differences in the total amount of DCVs as compared with WT chromaffin cells.
To directly investigate whether there is a peptide release deficit associated with reduced Rab3A, we developed a MS-based protocol to examine stimulation-dependent release of peptides from WT and Fmr1 KO mouse cortical synaptoneurosomes. Previously reported methods primarily used electrophysiology to observe the effects of stimulation at Fmr1 KO mouse synapses (21). Synaptosomes have been used for many studies on neurotransmitter release (22). Here, we adapted small-volume (23,24) and neuropeptide release collection approaches (25−28) to enable MS-based peptide characterization of released peptides in Fmr1 KO mice. Briefly, synaptoneurosomal preparations from WT and Fmr1 KO mice (n = 4, WT; n = 5, Fmr1 KO) were stimulated with KCl (55 mM), and the releasates were concentrated and separated into fractions using reversed-phase liquid chromatography (LC). The fractions were analyzed using MALDI-MS. The resulting mass profiles of releasates from WT cortical synaptoneurosomes revealed stimulation-dependent release of several peptide signaling molecules as well as protein fragments (Figure (Figure2a).2a). Little or no stimulated peptide release was observed in cortical synaptoneurosomes from Fmr1 KO mice; the reduction in peptide release is at least 90% and perhaps greater (Figure (Figure2c),2c), with the large decrease making quantitative measures of decrease difficult. To consider whether the KCl stimulation data reflect presynaptic or postsynaptic peptide release events, (S)-3,5-dihydroxyphenylglycine (DHPG) was administered to stimulate postsynaptic group I metabotropic glutamate receptors in the WT synaptoneurosomes; this stimulation did not elicit synaptic peptide release. Thus, we suggest, but have not yet fully confirmed, that this deficit is presynaptic in nature. Several peptides and protein fragments observed in the releasates of stimulated WT synaptoneurosomes were present in sufficient amounts to confirm their identity with tandem MS (MS/MS) analysis. These peptides are primarily derived from structural and related proteins (such as actin, tubulin, and stathmin) and neuropeptide prohormones (Supporting Information, Table S1).
To confirm changes in the stimulation-induced release in a more intact preparation, we used brain slices obtained from P10−15 WT and Fmr1 KO mice, as described earlier (25). We chose two regions to study: (1) the cerebral cortex, because this was the source of our synaptoneurosomes, and (2) the suprachiasmatic nucleus (SCN), a well-defined nucleus that is highly peptidergic and for which we previously characterized robust peptide release (25). Using an on-tissue collection strategy, we collected releasates from the cortex and SCN of WT and Fmr1 KO mice using a micropipet tip packed with C18 solid-phase extraction beads. We then performed MALDI-MS analysis on releasates obtained before and after stimulation with 55 mM KCl. The MS profiles showed a significant increase in the number and intensity of peaks detected in releasates after stimulation of WT, but not KO slices, compared with the prestimulated collections (Figure (Figure3,3, n = 4, WT; n = 4, Fmr1 KO). In addition to arginine vasopressin, somatostatin, and proSAAS, which have been shown earlier to be released from the SCN (25), we identified several peptides from the cortex that were also identified from the synaptoneurosomal releasates.
Although we have shown that the impaired release mechanisms involve peptides, it is reasonable to speculate that other regulated release pathways might be affected, such as cytokine and growth factor release. For example, brain-derived neurotrophic factor (BDNF) release from CA3 neurons was shown to recruit a presynaptic component of plasticity (29). Lauterborn et al. (14) recently showed that the LTP deficiency observed in the hippocampal field CA1 of Fmr1 KO mice can be rescued by exogenous application of BDNF, even though there was no significant difference in the basal level of BDNF in homogenates of Fmr1 KO versus WT hippocampus; therefore, those results also imply a deficit in BDNF synaptic release in FXS.
To analyze the specificity of the observed release deficit, we used electrochemistry-based measurements to analyze the stimulus-evoked release of biogenic amines. Brain slices from P70 WT and Fmr1 KO mice containing the caudate putamen (CP), ventral bed nucleus of the stria terminalis (vBNST), or substantia nigra pars reticulata (SNr) were used to analyze the electrically stimulated release of DA, NE, and 5-HT, respectively, because these particular regions are enriched in these biogenic amines (30−32). DA release in striatal brain slices was evoked by a single-pulse electrical stimulation (Figure (Figure4a).4a). Stimulated release of 5-HT (20 pulses at 20 Hz) in slices of the SNr had a similar profile; however, the maximal release values and rate of uptake are both lower than those seen for DA (Figure (Figure4b).4b). NE release was evoked by 60 pulses at 60 Hz stimulation and demonstrated the slowest uptake of all three neurotransmitters (Figure (Figure4c).4c). Electrically stimulated DA release monitored by fast-scan cyclic voltammetry was unchanged in Fmr1 KO mice compared with WT mice (n = 6, WT; n = 7, Fmr1 KO; p > 0.05). Additionally, NE release in the vBNST stimulation was not different (n = 4, WT; n = 6, Fmr1 KO; p > 0.05) between Fmr1 KO and WT control animals. This trend was repeated in the SNr where stimulation evoked the same amount of 5-HT in both animals (n = 6, WT; n = 7, Fmr1 KO; p > 0.05). Similar results were found when monitoring exocytosis from chromaffin cells (Supporting Information, Figure S1 and Table S2). The lack of differences in release of biogenic amines between WT and Fmr1 KO mice shows that the presynaptic release deficit is not general but is transmitter specific.
It is known that both clear-core vesicles, which house many classical neurotransmitters, and large DCVs, which house peptides and perhaps the catecholamines, are mobilized and released in an activity-dependent manner (33,34). Because DCVs contain neuropeptides, we considered whether the decrease in the amount of peptides observed in Fmr1 KO mice was due to a decrease in the number of DCVs. We examined electron micrographs containing layer IV of the visual cortex from P15 and P25 WT and Fmr1 KO mice, utilizing unbiased stereological methods to look for differences in the density of DCVs (DCVs/μm3). While a decrease in DCVs/μm3 between WT and Fmr1 KO mice at both P15 and P25 was observed (Supporting Information, Figure S2), these differences did not reach significance. A significant decrease in DCVs/μm3 from P15 to P25 was observed for both genotypes. Specifically, our analysis revealed a main effect of age (p < 0.01) but no main effect of genotype (p = 0.220) and no genotype−age interaction (p = 0.996).
MS-based assays that are capable of analyzing release enable characterization of peptide secretion following chemical stimulation, without having to preselect the peptides (25,26,35). The ability to collect and characterize peptide release allows us to present the first report of impaired peptidergic release in the FXS mouse model. Although there is not a significant reduction of DCVs in FXS, there is a ~50% reduction in synaptic levels of Rab3A. These results suggest that some of the deficits in Fmr1 KO mice may be directly correlated with deficits in the peptide release machinery. Intriguingly, the release deficit observed in the Fmr1 KO mice involves the release of neuropeptides in the neocortex and does not involve biogenic amines in the striatum, suggesting a deficit in the release of peptidergic DCVs, at least in several brain regions of FXS mice. Because similar reductions in peptide release are observed in both functional synapses (synaptoneurosomes) and brain slices from the cerebral cortex and the SCN, this impairment appears to be fairly general and may be associated with synaptic machinery, perhaps aberrant DCV docking and fusion, in which Rab3A normally participates. Stimulated release of DA, NE, and 5-HT is unaffected in the striatum, and basal synaptic transmission in Fmr1 KO mice is similar to WT mice (36), results that point to a specific release defect in the brain.
In conclusion, there appears to be a role for cell-to-cell signaling in the etiology of FXS, possibly affecting one or more phases of the synaptogenesis and synaptic stabilization sequence (37,38). Peptides characterized as released from the synapses in an activity-dependent manner include fragments of structural proteins such as actin, tubulin, and stathmin, which are involved in exocytosis. Their dynamics play a role in the morphological plasticity of dendritic spines (39), and release of these peptides is reduced in Fmr1 KO mice. The resilience of this defect, as observed in synaptoneurosomes and slice preparations from several brain regions, suggests that this impaired peptide-specific release may be a fundamental biological deficit and may contribute to defective synaptic maturation and maintenance and the behavioral deficits known to be associated with FXS. The correlation between FXS and the neuropeptide release deficit suggests that this is a unique pharmacological target, which, if restored, may ameliorate some of the effects of FXS.
WT and sighted Fmr1 KO mice bred in the FVB.129P2-Fmr1tm1Cgr background were housed at the Beckman Institute at the University of Illinois at Urbana−Champaign (UIUC) or the University of North Carolina at Chapel Hill (UNC−CH) animal facility. Mice were maintained under controlled environmental conditions, with standard alternating periods of 12-h light and darkness. Rodent chow and water were provided ad libitum. Animal care and procedures in accordance with the National Institutes of Health regulations and protocols were approved by the UIUC and UNC−CH Institutional Animal Care and Use Committees.
Synaptoneurosomes were prepared from cortices of P10−14 WT and Fmr1 KO mice. For the MS measurements of synaptoneurosome release only, six experiments were performed using rat cortices to validate our ability to stimulate peptides and collect releasates (n = 6). Once the ability to collect peptide release was shown to be robust, we modified the procedures for smaller mouse samples. Briefly, mice were sacrificed by decapitation. Brains were removed and dissected, and several cerebral cortices were pooled to obtain adequate cellular material for each experiment (n = 4, WT; n = 5, Fmr1 KO with each sample consisting of two or three cortices). Cortices were homogenized in a glass−Teflon homogenizer in 1 mL of ice-cold homogenizing buffer (50 mM N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES), pH 7.5, 125 mM NaCl, 100 mM sucrose, 2 mM potassium acetate), filtered through a series of nylon mesh filters (149, 62, 30, and 10 μm; Small Parts, Inc., Miramar, FL) and finally filtered through a 10-μm polypropylene filter (Gelman Sciences, Ann Arbor, MI). Filters were washed at each step with the ice-cold homogenizing buffer. The final filtrate was spun briefly (4000 × g, 4 °C, 1 min) to remove cellular debris; final supernatant volume was ~1 mL. This synaptoneurosomal preparation was either used directly for stimulation and collection of releasates by MS or was lysed (for Western blot analysis) in a lysis buffer [final concentration 50 mM Tris (pH 8), 150 mM NaCl, 1% Nonidet P-40, 20 nM okadaic acid, 20 μM sodium orthovanadate, phosphatase inhibitor mixture (Pierce, Rockford, IL), and protease inhibitor mixture (Roche, Indianapolis, IN)]. NP-40 soluble total cortical homogenate was prepared in the following way: cortices from a single mouse, P10−14, were homogenized in NP-40 lysis buffer [final concentration 50 mM Tris (pH 8), 150 mM NaCl, 1% Nonidet P-40, 20 nM okadaic acid, 20 μM sodium orthovanadate, phosphatase inhibitor mixture (Pierce), and protease inhibitor mixture (Roche)] and spun (15000 × g, 4 °C, 15 min), and then the supernatant was collected.
For each synaptoneurosomal preparation and total cortical homogenate sample, 30 μg of lysed protein samples was run on 12% polyacrylamide gels, blotted to nitrocellulose membranes, and stained with rabbit polyclonal antibody specific for Rab3A (Abcam, Cambridge, MA). HRP-labeled secondary antirabbit antibody (Cell Signaling Technology, Inc., Danvers, MA) was detected by enhanced chemiluminescence (Pierce). To quantify and standardize protein levels, blots were stained for β-actin (Sigma Aldrich, St. Louis, MO) and with Amido Black (J. T. Baker, Phillipsburg, NJ) for total protein. Chemiluminescence was scanned in a FluorChem 8900 Imager (Alpha Innotech, San Leandro, CA) and relative optical densities were determined by using AlphaEaseFC software, version 4.0.1 (Alpha Innotech), normalized to β-actin level and total protein loaded.
The synaptoneurosomal preparations, obtained from two to three mouse cerebral cortices for each preparation, were divided into two equal aliquots used as a control (unstimulated) or as the stimulated sample. For the stimulated samples, 1 M KCl (55 mM final concentration) or 1 M DHPG (Tocris Cookson, Ellisville, MO; 5 mM final concentration), was added to one of the aliquots and incubated for 15 min. For the control samples, 50 μL of homogenizing buffer was added. After stimulation, the aliquots were centrifuged (4 min at 4000 × g at 4 °C), and the supernatant was collected. The peptides present in the supernatant were desalted and concentrated using a spin-column (Pierce) according to the manufacturer’s protocols. These measurements were repeated for nine samples (from 21 animals). In two instances, after chemical stimulation, no peptides (including the normally ubiquitous β-thymosin) were detected via MS, indicating an issue with the preparation; in such cases, the data were excluded.
Brain slices were prepared from P10−P15 WT and Fmr1 KO mice. Briefly, mice were quickly decapitated, brains were removed, and a Vibratome (Vibratome 3000 Series; Ted Pella, Redding, CA) was used to obtain 400-μm sections. The sections were placed into a brain slice chamber (AutoMate Scientific, Inc., Berkeley, CA) equipped with a proportional temperature controller. Each slice was perfused with Earle's balanced salt solution (EBSS; without phenol red), supplemented with 24.6 mM glucose, 26.2 mM NaHCO3, and 2.5 mg/L gentamycin and saturated with 95% O2/5% CO2 at 37 °C, pH 7.4. Collection of releasates from each slice (n = 4, WT; n = 4, Fmr1 KO) was performed as described earlier (25) and analyzed by MALDI-MS.
The synaptoneurosomal releasates were separated using a capLC system (Micromass, U.K.) coupled to a robotic spotter (Proteineerfc, Bruker Daltonics, Billerica, MA). Briefly, the samples were initially loaded onto a 5-μL loop using a manual injector (VALCO Instruments Co. Inc., Houston, TX) and subsequently trapped onto a reversed-phase trap column (PepMap, C18, 5 μm, 100 Å, LC Packings, a Dionex Company, Sunnyvale, CA) using solvent C (90% water, 10% acetonitrile (ACN), 0.1% formic acid (FA), and 0.01% trifluoroacetic acid (TFA)) at a flow rate of 8 μL/min and washed for 5 min. Each sample was then eluted and separated on a reversed-phase column (LC Packings 300-μm i.d. × 15 cm, C18 PepMap100, 100 Å) using a gradient at 2 μL/min flow rate for 45 min. Gradients were generated using solvent A (95% water, 5% ACN, 0.1% TFA) and solvent B (95% ACN, 5% water, 0.1% FA, and 0.01% TFA). Gradients were changed linearly in three different steps; 5−30% solvent B in 15 min, 30−80% solvent B in 35 min, 80−5% solvent B in 40 min. The chromatographic profile was monitored using a UV detector at 210 nm. The capLC system was connected online to the Proteineerfc spotter to collect the fractions onto a MALDI target. Prior to spotting the fractions, the Proteineerfc was programmed to deposit 1 μL of α-cyano-4-hydroxy cinnamic acid MALDI matrix (10 mg/mL, 60% aqueous ACN, 0.1% TFA). For each sample, a total of 24 fractions were collected at 1 min intervals, corresponding to samples eluted from 18−42 min of the chromatographic run.
Mass spectrometric analysis of the samples was performed using a MALDI-TOF/TOF mass spectrometer (Ultraflex II, Bruker Daltonics). The instrument was externally calibrated using a mixture of standard peptides (angiotensin II, angiotensin I, substance P, bombesin, ACTH clip 1−17, ACTH clip 18−39, somatostatin 28, bradykinin fragment 1−7, renin substrate tetradecapeptide porcine) in the mass range 700−3200 Da (Bruker Daltonics). For each of the synaptoneurosomal releasate fractions, data was collected using 400 laser shots in 10 short steps by panning across the spot and combining the spectra. Following data collection, the mass spectra were smoothed and baseline corrected, and the peaks with high peak intensity and signal/noise ratio were tabulated. Releasates from the slice were directly analyzed using MALDI-MS.
MS/MS of selected molecular ion species was performed using the TOF/TOF feature of the Bruker Ultraflex II instrument. The MS/MS spectra were processed using FlexAnalysis 2.4 and Biotools 3.0 software packages (Bruker Daltonics). Partial de novo analysis was also performed on the MS/MS spectra using Peaks software (Bioinformatics Solutions Inc., Waterloo, Ontario, Canada) to assist in peptide identification. For the release of peptides from the SCN, previously sequenced peptides characterized in the SCN and supraoptic nucleus were used for peptide identification (25,40).
Eight P15 WT and ten P15 Fmr1 KO mice and eight P25 WT and ten P25 Fmr1 KO mice were used to analyze DCV quantity in layer IV of the visual cortex. Perfusion and tissue preparation procedures for transmission electron microscopy ultrastructural analyses of brain have been described in a previous report (41). Within layer IV of the visual cortex, a region of neuropil was selected based only on proximity to easily recognizable landmarks (capillaries, nuclei, etc.) and on the absence of folds in adjacent sections (aside from these two criteria, the region was selected randomly within layer IV). Micrographs from identical regions of consecutive (approximately 60-nm thick) serial sections were taken at a magnification of 8800× (final magnification 26,400×) by a Phillips CM-200 transmission electron microscope and analyzed by raters blind to the experimental group. The density of DCVs was estimated using the physical disector (42), in which disector pairs, composed of a “reference” section and a “look-up” section, are analyzed. For each disector pair, the fractionator (43) is used to establish one sampling site that is the area of the electron micrograph in layer IV of the visual cortex. Between 6 and 15 serial electron micrographs per tissue block were used for stereological analysis. DCVs were characterized by the presence of an electron-dense core and at least 50% of circumference surrounded by an intact identifiable double-membrane (44); clear-core vesicles were not quantified. The sections were analyzed for DCVs present in the reference section but not the look-up section. These DCVs were counted. DCVs present on both the reference section and look-up section were not counted. The formulas DCVv = Q/Vdis and Vdis = AframeTN were used to obtain DCV density, where DCVv is DCV density, Q is the number of DCVs present in the reference section but not the look-up section, Vdis is the disector volume of the tissue through which the DCVs are counted, Aframe is the area of the superimposed counting frame, T is section thickness, and N is the number of sections through which the counting is done.
The SPSS (SPSS Inc., Chicago, IL) statistical package (version 16.0) was used to conduct statistical analyses. Independent Student’s t-tests were used to determine statistical significance. For DCV quantification, a two-way ANOVA was conducted to determine statistical significance for effects of age (P15, P25) and genotype (WT, Fmr1 KO) on 415 micrographs from 36 animals. For these analyses, p < 0.05 was considered statistically significant.
Chromaffin cells were prepared as previously described (45,46). Plates were maintained in a humidified, 5% CO2 atmosphere at 37 °C for at least 24 h prior to experimentation. Brain slices were prepared containing the CP (31), SNr (30), and BNST (32). Slices were superfused in artificial cerebrospinal fluid at 37 °C for 35−40 min prior to recording. Measurements were made with carbon-fiber microelectrodes prepared using T650 carbon fibers (47). At chromaffin cells, recordings were made by amperometry with disk electrodes (46). Measurements of DA, NE, and 5-HT release in brain slices were made using fast-scan cyclic voltammetry at cylindrical electrodes as previously described (30).
Exocytosis at chromaffin cells was triggered by 0.5 s pressure ejection of 60 mM K+ buffer from a pipet located 30 μm from the cell. Ten stimulations, separated by 30 s, were applied to each cell. In brain slices, release was electrically evoked (31). Amperometric spikes were analyzed with MiniAnalysis (Synaptosoft, Decatur, GA). Individual amperometric spike characteristics determined were the spike frequency, quantal size (Q), t1/2, and amplitude (48). In the slice experiments, the maximal neurotransmitter concentrations measured following stimulation are reported.
We thank Dr. G. Govindaiah for technical assistance with brain slice techniques. We also thank Dr. Deepa Venkitaramani and Shawn Kohler for assistance with statistical analysis of DCV quantification. We gratefully acknowledge Stephanie Baker for help with manuscript preparation.
National Institutes of Health, United States
S.P.A., S.H.K, J.V.S., B.M.K., K.T.T., R.M.W., W.T.G., and I.J.W. designed the research, S.P.A., A.E.L., S.H.K., S.R., B.M.K., K.T.T., and N.G.H. performed the research, S.P.A, B.M.K., K.T.T., and A.E.L. analyzed the data, and S.P.A., A.E.L., S.H.K., I.J.W., J.V.S., R.M.W., and W.T.G. wrote the paper.
The projected described was supported by the National Institute on Drug Abuse (NIDA) under Award Nos. P30DA018310 and DA017940, the National Institute of Neurological Disorders and Stroke (NINDS) under Award Nos. NS031609 and NS038879, and the National Institute of Mental Health (NIMH) under Award No. MH035321. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIDA, NINDS, NIMH, or the National Institutes of Health.
A list of the most prominent secreted peptides identified from WT synaptoneurosomes, vesicular release events at chromaffin cells from WT and Fmr1 KO mice, and representative EM images from visual cortex of WT and Fmr1 KO mice, together with quantification data. This material is available free of charge via the Internet at http://pubs.acs.org.