Abstract

Defects in synaptic development and plasticity may lead to autism. Brain-derived neurotrophic factor (BDNF) plays a critical role in synaptogenesis and synaptic plasticity. BDNF is synthesized as a precursor, pro-BDNF, which can be processed into either a truncated form or into mature BDNF. Previous studies reported increased BDNF-immunoreactive protein in autism, but the mechanism of this increase has not been investigated. We examined BDNF mRNA by real-time reverse transcriptionYpolymerase chain reaction and BDNF protein by Western blotting and enzyme-linked immunosorbent assay in postmortem fusiform gyrus tissue from 11 patients with autism and 14 controls. BDNF mRNA levels were not different in the autism versus control samples, but total BDNF-like immunoreactive protein, measured by enzyme-linked immunosorbent assay, was greater in autism than in controls. Western blotting revealed greater pro-BDNF and less truncated BDNF in autism compared with controls. These data demonstrate that increased levels of BDNF-immunoreactive protein in autism are not transcriptionally driven. Increased pro-BDNF and reduced truncated BDNF are consistent with defective processing of pro-BDNF to its truncated form. Distortion of the balance among the 3 BDNF isoforms, each of which may exhibit different biological activities, could lead to changes in connectivity and synaptic plasticity and, hence, behavior. Thus, imbalance in proteolytic isoforms is a possible new mechanism for altered synaptic plasticity leading to autism.

Introduction

Autistic disorder (AD) is a life-long neurological disorder characterized by abnormal or impaired development in social interactions and communication and repetitive interests or behavior (1). Developmental anomalies of the limbic system and the cortex have been well documented (2–7). Areas of the brain affected in autism are associated with the behavioral characteristics observed in patients with the disorder. For example, patients with autism have abnormalities in face recognition skills and difficulty with face perception (8). Face processing is mediated by the fusiform gyrus, particularly the fusiform face area, an area of occipitotemporal cortex that shows activity during facial recognition tasks (9–11). Hypoactivation of the fusiform gyrus related to face processing in AD has been replicated in a variety of studies (12–14). Therefore, examination of this area using a molecular approach may yield information on neurobiological and neurochemical abnormalities in AD.

The etiology of AD is unknown, but epidemiological and twin studies point to a major role for genetic factors in its pathogenesis (2, 15). Inheritance shows a complex pattern of transmission thought to be the result of multiple interacting genes. Recent advances in the identification of autism-linked susceptibility genes have identified a wide variety of rare genetic events that carry an increased risk for AD. These include chromosomal rearrangements such as maternal duplication of 15q11–q13, copy number variations and mutations in specific synaptic genes (16–25). These genes code for proteins associated with synaptogenesis and excitation/inhibition imbalance. Recent copy number variation (17) and genome-wide association studies implicate additional sets of genes involved in neurogenesis, neurite outgrowth and guidance, and plasticity (20, 26, 27). These findings have focused attention on aberrant neurite outgrowth and synapse formation and connectivity during development as an underlying cause of AD (24, 28).

Brain-derived neurotrophic factor (BDNF) is a synaptic molecule that plays an important role in neuronal survival and differentiation, dendritic spine maturation, and neuritic branching and connectivity during development (29–33). It also plays a critical role in synaptic plasticity and long-term potentiation and is involved in learning, memory, and attention (34–39). Low serum or brain BDNF levels have been reported in schizophrenia (40–42), major depression (43), and Alzheimer disease (44–47).

Brain-derived neurotrophic factor is synthesized as a 32-kDa precursor that can be processed to the 14-kDa mature form by furin, matrix metalloprotease 7, or plasmin (48Y50). Both the precursor, pro-BDNF, and the mature form are found in human brain in areas demonstrated to be abnormal in patients with autism (4, 51, 52). Pro-BDNF exhibits distinct and opposite functions compared with mature BDNF. It reduces neuronal differentiation and dendritic spines and induces apoptosis and long-term depression in cultured neurons, whereas mature BDNF promotes spine formation, neuronal survival, and long-term potentiation (35, 53–55). A polymorphism in the prodomain of pro-BDNF that interferes with pro-BDNF processing and secretion of mature BDNF has been implicated in memory deficits (56, 57). Moreover, a genetic association has been found between this polymorphism and psychiatric disorders, including schizophrenia (58). This suggests that an imbalance of pro-BDNF and mature BDNF may contribute to neurodegenerative or psychiatric disorders. Pro-BDNF can also be processed at a different site by a calcium-dependent serine proteinase known as membrane-bound transcription factor protease site 1 (MBTPS-1), also known as subtilisin/kexin isozyme (SKI-1). This cleavage produces a 28-kDa truncated BDNF isoform (59). This isoform is not further processed into mature 14-kDa BDNF and, therefore, represents a true final proteolytic product. Although the biological role of this isoform has yet to be elucidated, increased pro-BDNF and mature BDNF and reduced truncated BDNF have been found recently in the serum of patients with schizophrenia (60). Therefore, relative levels of all isoforms of BDNF may be important in psychiatric disorders.

Because of its important role in neuronal development, higher than normal levels of BDNF could explain the abnormal synaptic connectivity and neurotransmitter imbalances and the abnormal trajectory of accelerated brain growth seen in subjects with AD. Several studies have reported an increase in BDNF-immunoreactive protein in autism. Neonatal cord blood of children subsequently diagnosed with autism showed a 36% increase in BDNF protein versus controls (61). However, a subsequent report from the same group using a Luminex assay reported no difference in BDNF levels between groups (62). Elevated levels of BDNF immunoreactivity have also been reported in serum (63, 64), in platelet-rich plasma (65), and in basal forebrain tissue of patients with autism (66). However, it is not clear whether this increased BDNF protein derives from transcriptional upregulation or from translational or posttranslational mechanisms because neither BDNF mRNA levels nor protein isoforms have been examined in brain tissue from individuals with autism. Only one study has examined BDNF mRNA levels in AD; elevated BDNF expression in blood lymphocytes of drug-naive male adults with AD compared with controls was reported (67). In this study, we examined postmortem tissue from the fusiform gyrus region to identify differences in BDNF mRNA and protein isoform levels between subjects with autism and controls.

Materials And Methods

Human Brain Tissue Samples

Postmortem brain samples from 11 subjects with autism and 14 controls were provided by the Autism Speaks' Autism Tissue Program (Princeton, NJ) via the Harvard Brain Tissue Resource Center (Belmont, MA) and the University of Maryland Brain and Tissue Bank (Baltimore, MD). High quality RNA could not be obtained from 2 autism samples (1 and 8). The 9 remaining AD and all control samples were used for mRNA analysis. All 11 autism and 14 control samples were used for ELISA; 9 autism and 9 control samples were used for Western blot analysis because 2 autism and 5 control samples were obtained at a later date and could not be analyzed simultaneously. All of the tissues were from the fusiform gyrus area of the brain. Clinical data on each tissue sample were obtained through the Autism Tissue Program online portal (http://www.atpportal.org) (Tables 1 and 2). The diagnosis of AD was confirmed using the Autism Diagnostic Interview-Revised (68) postmortem through interviews with the parent(s) and/or caregivers. Tissue samples were stored at −80°C before use.

TABLE 1

Autism Cases

BDNF P rotein (Isofo rm/Total BD NF Ratio)  
Sample ID No. Age, yr Gender Postmortem Interval, h Primary Cause of Death Drug Treatment History Pro-BDNF Truncated BDNF Mature BDNF Ratio Total BDNF/β-Actin BDNF Protein (ELISA),pg/mg protein BDNF mRNA (BDNF/β-Actin Ratio) 
AN12552 56 23.61 Traumatic asphyxia and crush injury 0.22 0.63 0.15 0.67 25.51 1.583 
AN17344 46 25.9 NA 0.41 0.31 0.28 2.05 22.80 1.523 
AN14771 30 23 Cardiac arrhythmia 0.29 0.49 0.22 1.29 32.99 1.699 
UMB818 27 10 Multiple injuries 0.29 0.59 0.12 0.45 31.94 1.772 
UMB1407 20 Asthma 0.20 0.67 0.12 0.65 41.23 1.920 
UMB1649 20 22 Multiple injuries 0.20 0.59 0.21 0.93 61.33 1.595 
UMB1706 20 Rejection of cardiac allograft transplantation 0.17 0.67 0.16 0.42 57.20 1.635 
UMB1708 20 Multiple traumatic injuries 0.26 0.56 0.18 1.05 83.21 1.618 
UMB1860 Cardiac Arrhythmia 0.25 0.46 0.30 1.30 42.70 1.625 
AN10606 56 23 MI ND ND ND ND 111.62 1.761 
AN12240 51 4.75 MI ND ND ND ND 162.14 1.697 
AN19760 28 23.25 NA ND ND ND ND 102.66 1.743 
AN15240 36 18.08 NA ND ND ND ND 123.01 1.829 
AN17425 Mean T SEM 16 28.5 T 4.8 26.16 18.8 T 1.8 NA ND 0.25 T 0.02 ND 0.55 T 0.04 ND 0.19 T 0.02 ND 4.89 T 0.46 91.65 71.50 T 11.15 1.742 1.70 T 0.03 
BDNF P rotein (Isofo rm/Total BD NF Ratio)  
Sample ID No. Age, yr Gender Postmortem Interval, h Primary Cause of Death Drug Treatment History Pro-BDNF Truncated BDNF Mature BDNF Ratio Total BDNF/β-Actin BDNF Protein (ELISA),pg/mg protein BDNF mRNA (BDNF/β-Actin Ratio) 
AN12552 56 23.61 Traumatic asphyxia and crush injury 0.22 0.63 0.15 0.67 25.51 1.583 
AN17344 46 25.9 NA 0.41 0.31 0.28 2.05 22.80 1.523 
AN14771 30 23 Cardiac arrhythmia 0.29 0.49 0.22 1.29 32.99 1.699 
UMB818 27 10 Multiple injuries 0.29 0.59 0.12 0.45 31.94 1.772 
UMB1407 20 Asthma 0.20 0.67 0.12 0.65 41.23 1.920 
UMB1649 20 22 Multiple injuries 0.20 0.59 0.21 0.93 61.33 1.595 
UMB1706 20 Rejection of cardiac allograft transplantation 0.17 0.67 0.16 0.42 57.20 1.635 
UMB1708 20 Multiple traumatic injuries 0.26 0.56 0.18 1.05 83.21 1.618 
UMB1860 Cardiac Arrhythmia 0.25 0.46 0.30 1.30 42.70 1.625 
AN10606 56 23 MI ND ND ND ND 111.62 1.761 
AN12240 51 4.75 MI ND ND ND ND 162.14 1.697 
AN19760 28 23.25 NA ND ND ND ND 102.66 1.743 
AN15240 36 18.08 NA ND ND ND ND 123.01 1.829 
AN17425 Mean T SEM 16 28.5 T 4.8 26.16 18.8 T 1.8 NA ND 0.25 T 0.02 ND 0.55 T 0.04 ND 0.19 T 0.02 ND 4.89 T 0.46 91.65 71.50 T 11.15 1.742 1.70 T 0.03 

BDNF, brain-derived neurotrophic factor; F, female; M, Male; MI, myocardial infarction; NA, Information not available; ND, not done.

TABLE 2

Control Patients

BDNF Protein (Isoform/Total BDNF Ratio) Sample ID No. Age, yr Gender Postmortem Interval, h Primary Cause of Death Pro-BDNF Truncated BDNF Mature BDNF Ratio Total BDNF/A-Actin BDNF Protein(ELISA), pg/mg protein BDNF mRNA(BDNF/A-Actin Ratio) 
AN01093 56 19.48 Anoxic encephalopathy NA 0.72 0.01 0.27 0.56 76.13 ND 
AN00493 27 8.3 Drowning Synthroid 0.34 0.42 0.24 1.83 166.44 1.683 
AN00764 20 23.7 Auto Trauma Minocin 0.30 0.45 0.25 2.04 182.46 1.575 
AN08792 30 20.3 Gastrointestinal bleeding Phenobarbital, 
Mysoline, Dilantin, Depakote, Cisapride, Clorazepate, Prolosec, Propulsid, Reglan, Tranxene 0.53 0.17 0.30 4.00 205.82 1.678 
AN16115 11 12.88 Seizure and drowning in tub Adderall, Dexadrine, Dilantin, Klonopin, Lamictal, Tegretol, Topomax 0.30 0.44 0.26 1.54 394.25 1.647 
AN16641 27 Seizure disorder Clonidine, Depakote, Dilantin, Lamictal, Ritalin, Tegretol 0.31 0.36 0.32 2.04 104.48 1.411 
UMB1174 14 Seizure, 
hypotension Depakote, Dilantin, Tegretol 0.51 0.10 0.38 0.83 53.11 1.638 
UMB1182 24 Smoke Inhalation NA 0.84 0.09 0.07 0.31 33.39 ND 
UMB797 13 Drowning Desipramine 0.25 0.58 0.18 0.86 26.18 1.652 
AN08873 25.5 Asphyxia due to drowning Prozac ND ND ND ND 233.65 1.599 
AN06420 39 13.95 Cardiac tamponade Synthroid, Depakote, Risperidol, Paxil, Blood pressure medication ND ND ND ND 133.35 1.740 
Mean ± SE 20.1 ± 4.9  
 18.3 ± 1.8  
  
 0.45 ± 0.07 0.29 ± 0.06 0.25 ± 0.03 6.08 ± 1.14 146.30 ± 31.62 1.63 ± 0.03 
BDNF Protein (Isoform/Total BDNF Ratio) Sample ID No. Age, yr Gender Postmortem Interval, h Primary Cause of Death Pro-BDNF Truncated BDNF Mature BDNF Ratio Total BDNF/A-Actin BDNF Protein(ELISA), pg/mg protein BDNF mRNA(BDNF/A-Actin Ratio) 
AN01093 56 19.48 Anoxic encephalopathy NA 0.72 0.01 0.27 0.56 76.13 ND 
AN00493 27 8.3 Drowning Synthroid 0.34 0.42 0.24 1.83 166.44 1.683 
AN00764 20 23.7 Auto Trauma Minocin 0.30 0.45 0.25 2.04 182.46 1.575 
AN08792 30 20.3 Gastrointestinal bleeding Phenobarbital, 
Mysoline, Dilantin, Depakote, Cisapride, Clorazepate, Prolosec, Propulsid, Reglan, Tranxene 0.53 0.17 0.30 4.00 205.82 1.678 
AN16115 11 12.88 Seizure and drowning in tub Adderall, Dexadrine, Dilantin, Klonopin, Lamictal, Tegretol, Topomax 0.30 0.44 0.26 1.54 394.25 1.647 
AN16641 27 Seizure disorder Clonidine, Depakote, Dilantin, Lamictal, Ritalin, Tegretol 0.31 0.36 0.32 2.04 104.48 1.411 
UMB1174 14 Seizure, 
hypotension Depakote, Dilantin, Tegretol 0.51 0.10 0.38 0.83 53.11 1.638 
UMB1182 24 Smoke Inhalation NA 0.84 0.09 0.07 0.31 33.39 ND 
UMB797 13 Drowning Desipramine 0.25 0.58 0.18 0.86 26.18 1.652 
AN08873 25.5 Asphyxia due to drowning Prozac ND ND ND ND 233.65 1.599 
AN06420 39 13.95 Cardiac tamponade Synthroid, Depakote, Risperidol, Paxil, Blood pressure medication ND ND ND ND 133.35 1.740 
Mean ± SE 20.1 ± 4.9  
 18.3 ± 1.8  
  
 0.45 ± 0.07 0.29 ± 0.06 0.25 ± 0.03 6.08 ± 1.14 146.30 ± 31.62 1.63 ± 0.03 

BDNF, brain-derived neurotrophic factor; F, female; M, male; NA, Information not available; ND, not done.

Reagents

The affinity-purified rabbit polyclonal BDNF antibody, N-20, and the peptide used to raise it (mapping at the amino terminus of the mature form of human BDNF) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Mouse monoclonal anti-β-actin was obtained from Sigma (Oakville, Ontario, Canada). Horseradish peroxidase-conjugated donkey anti-rabbit and sheep anti-mouse antibodies were obtained from Amersham Pharmacia Biotech (Baie d’Urfé, Quebec, Canada). Human recombinant BDNF was a generous gift from Amgen (Thousand Oaks, CA). The BDNF ELISA kits were purchased from Promega (Madison, WI), Chemicon (Temecula, CA), and R&D Systems (Minneapolis, MN). Trizol reagent was from Invitrogen (Burlington, Ontario, Canada). Aprotinin was obtained from Sigma, pepstatin was from Boehringer Mannheim (Laval, Québec, Canada), and phenylmethylsulfonyl fluoride was from Life Technologies (Burlington, Ontario, Canada). Hybond-P polyvinylidene diflouride membranes were purchased from Amersham Pharmacia Biotech. Prestained broad-range protein molecular weight markers and DC protein assay kit and reagents were from Bio-Rad (Mississauga, Ontario, Canada). DNase treatment kits were purchased from Ambion (Austin, TX). Reagents for reverse transcription (RT) were obtained from Invitrogen (Carlsbad, CA). Primers were synthesized by MOBIX, McMaster University. Platinum SYBR Green qPCR Supermix UDG was obtained from Invitrogen. All other chemicals were reagent grade.

RNA Isolation

RNA isolation, DNase treatment, and real-time RT-polymerase chain reaction were carried out as previously described with modifications (69). Total RNA was isolated using 0.08 to 0.13 g of tissue per sample. Each frozen sample was homogenized in 1 mL/0.1 g wet weight tissue of Trizol reagent on ice using either a Polytron homogenizer (Brinkmann Instruments) or a sonic dismembrator (Fisher Scientific, Ottawa, Ontario, Canada). When using the homogenizer, RNA was purified according to the Trizol reagent instructions. The brain homogenate obtained by the sonication method was purified over RNeasy spin columns (Qiagen, Mississauga, Ontario, Canada) instead of ethanol precipitated. Yield and purity of total cellular RNA were determined by absorbance at 260 and 280 nm. RNA integrity was examined using agarose gel electrophoresis. RNA samples were stored at 80°C before use.

DNase Treatment and cDNA Synthesis

RNA samples were treated with DNase before RT, according to the manufacturer's protocols (Ambion). cDNA was generated in a reaction mixture containing 2.5 μg of DNase-treated sample, 10 mmol/L each of dATP, dCTP, dGTP, and dTTP (Invitrogen), and 250 ng of random primers (Invitrogen). The samples were incubated at 65°C for 5 minutes followed by addition of 1 × First Strand Buffer (3 mmol/L MgCl2, 75 mmol/L KCl, 50 mmol/L Tris-HCl, pH 8.3), 0.01 mol/L dithiothreitol, 40 units of RNaseOUT, and 50 units of Superscript II. The reaction was carried out in a GeneAmp PCR system 2400 thermal cycler (Applied Biosystems, Foster City, CA) at 25°C for 10 minutes, followed by 42°C for 50 minutes, and heat inactivation at 70°C for 15 minutes. A negative control was run in parallel with each RT containing double-distilled water in place of RT (no-RT control).

BDNF Real-time PCR

Primers for BDNF, forward primer 5′-AAA CAT CCG AGG ACA AGG TG-3′ and reverse primer 5′-AGA AGA GGA GGC TCC AAA GG-3′, yielded a 249-bp fragment. Primers for A-actin, forward primer 5′-CTC TTC CAG CCT TCC TTC-3′ and reverse primer 5′-TGT TGG CGT ACA GGT CTT-3′, yielded a 110-bp fragment. Real-time PCR was carried out on a Stratagene MX3000P (La Jolla, CA). Total reaction volume was 20 μL, which included cDNA from 100 ng of RNA and the following components: 300 nmol/L of the BDNF forward and reverse primers, 10 μL of SYBR Green qPCR Supermix UDG (Invitrogen), and 30 nmol/L of Stratagene ROX reference dye. For β-actin, 150 nmol/L of both the forward and reverse primers was used. A negative control containing double-distilled water in place of cDNA was included to ensure that none of the reagents were contaminated. Levels of BDNF and β-actin were determined using absolute quantification with SYBR green. Standards for BDNF were from a BDNF plasmid (pGEM containing the entire BDNF coding region of 755 bp, NCBI Accession No. AY890649), and β-actin standards were from a plasmid obtained from Invitrogen. The thermal profile for BDNF and β-actin: 2 minutes at 50°C, 2 minutes at 95°C, followed by 40 cycles of 95°C for 15 seconds, 58°C for 30 seconds, and 72°C for 30 seconds. PCR product samples analyzed by agarose gel electrophoresis exhibited a single band of the expected size. A dissociation curve after 40 PCR cycles confirmed a single product for both BDNF and β-actin.

Copy numbers and real-time PCR efficiencies were calculated with MXPro MX3000P software (Stratagene). Only experiments in which the real-time PCR efficiency was between 90% and 100% and standard curves yielded R2 > 0.990 were used for analysis. Each sample was run in triplicate. Calculation of differences in BDNF levels between autism and control samples was carried out by 2-tailed Student t test on the means of triplicate Ct values using SPSS (SPSS Inc., Chicago, IL).

Bdnf Elisa

Protein isolation and BDNF ELISA were carried out using the BDNF Emax ImmunoAssay System (Promega), the human BDNF Quantikine ELISA (R&D Systems), and the ChemiKine BDNF sandwich ELISA kit (Chemicon) according to the manufacturers' protocols. Each sample was initially adjusted to the same protein concentration (0.5 mg/mL for Promega; 1 mg/mL for Chemicon and R&D Systems). In the case of the Promega kit, samples were acidified with 1 mol/L HCl to bring the pH below 2.5 for 15 minutes and then neutralized to pH 7.6 with NaOH. A standard curve using recombinant BDNF protein was run on each ELISA plate. Each sample was tested in triplicate. Two-tailed Student t test was carried out on the means of triplicates.

Protein Extraction From Brain Tissue

Protein extraction was performed as previously described with slight modifications (70). Approximately 50 to 100 mg of tissue was homogenized on ice without thawing using a Polytron homogenizer or sonicated on ice using a sonic dismembrator in 10 mL/g wet weight tissue of homogenization buffer (50 mmol/L Tris-HCl pH 7.5, 10 mmol/L ethylenedi-aminetetraacetic acid, 0.5% Tween-20, 2 μg/mL aprotinin, 2 μg/mL pepstatin, and 100 μg/mL phenylmethylsulfonyl fluoride). The homogenate was incubated for 15 minutes on ice and then centrifuged at 16,000g for 15 minutes at 4-C. Supernatants containing solubilized protein were aliquoted and stored at −80°C. Protein concentrations were determined using a DC protein assay kit.

Western Blotting

Western blotting was carried out as previously described with minor modifications (46, 47). Briefly, 20 μg protein of each sample was resolved in 12% sodium dodecyl sulfate-polyacrylamide gels under reducing conditions. After transfer onto polyvinylidene diflouride membranes, blots were blocked with 10% nonfat milk powder in TBS-T (50 mmol/L Tris-HCl pH 8.0, 133 mmol/L NaCl, 0.1% (v/v) Tween-20), then incubated with primary antibody N-20 (1:1000 dilution in TBS-T) overnight at 4-C and horseradish peroxidase-conjugated donkey anti-rabbit secondary antibody (1:5000 dilution in TBS-T) for 1 hour at room temperature. An enhanced chemiluminescence system (Amersham Pharmacia Biotech) was used for detection. The specificity of the antibody was tested by blocking immunoreactivity with a 5- or 10-fold molar excess of antigenic peptide, according to the manufacturer's protocol (Santa Cruz Biotechnology). All samples were analyzed in 2 to 3 independent Western blots.

Quantification of Western Blots

Each Western blot contained a standard curve consisting of different amounts of protein per lane serving as a control to ensure that the sample loading amount and x-ray film exposure time were in the linear range of detection for BDNF isoforms. The intensities of immunoreactive bands were measured using ImageJ 1.43 (Wayne Rasband, National Institutes of Health; http://rsb.info.nih.gov/ij) with local background subtracted. For each sample, the relative abundance of each BDNF isoform was expressed as a ratio of that BDNF isoform to the total BDNF signal. Total BDNF was calculated as the sum of pro-BDNF, truncated BDNF, and mature BDNF. For each isoform, the mean of ratios from 2 to 3 blots for each sample was used for statistical analysis. Differences between groups were calculated using 2-tailed Student t test.

Results

No Difference in BDNF mRNA Expression Between Autism and Control Samples

There was no significant difference in BDNF mRNA levels in the fusiform gyrus between control (n = 14) and autism (n = 9) samples (2-tailed t test, p = 0.35; Fig. 1). Each sample's BDNF Ct value was normalized to its β-actin Ct value. β-actin has been used in the past as a reference gene because its levels do not change in neurological disorders such as Alzheimer disease (45). Investigation of β-actin Ct values showed no significant difference between control and autism samples (p = 0.89, data not shown).

FIGURE 1

Real-time RT-PCR of BDNF mRNA in autism and control samples. BDNF mRNA expression was normalized to A-actin mRNA for each sample. Each sample was measured in triplicate, and the mean was used for statistical analysis. A second independent experiment gave similar results. Mean ± SE is shown for each group. Autism, n = 9; control, n = 14. p > 0.05, 2-tailed t test.

FIGURE 1

Real-time RT-PCR of BDNF mRNA in autism and control samples. BDNF mRNA expression was normalized to A-actin mRNA for each sample. Each sample was measured in triplicate, and the mean was used for statistical analysis. A second independent experiment gave similar results. Mean ± SE is shown for each group. Autism, n = 9; control, n = 14. p > 0.05, 2-tailed t test.

Increased BDNF-Like Immunoreactivity Measured by ELISA in Fusiform Gyrus of Autism Versus Control Subjects

Acid treatment has been found to alter the amount of neurotrophic factor detectable by ELISA in various tissue or sample types (72, 73). Acid treatment disrupts the interaction between BDNF and its receptor, TrkB, or other binding proteins, which may not allow total BDNF to be accurately measured. Therefore, we first tested the effect of acid treatment on our homogenized brain samples. Acidification and subsequent neutralization increased BDNF immunoreactivity detectable by ELISA approximately 4-fold above nonacidified samples (data not shown). Relative differences in BDNF levels between groups did not differ between acidified (Promega) and non acidified (Chemicon, R&D Systems) methods. Autism subjects exhibited BDNF protein levels in fusiform gyrus 2-fold higher than in controls (p = 0.026; Fig. 2). ELISAs from all 3 kits gave similar results.

FIGURE 2

BDNF protein levels in fusiform gyrus as measured by ELISA. Each sample was measured in triplicate, and the mean was used for statistical analysis. Data from acidified (Promega) and nonacidified (R&D Systems) assays were combined. Bars indicate mean ± SE. Autism, n = 11; control, n = 14. * p < 0.05 (2-tailed t test).

FIGURE 2

BDNF protein levels in fusiform gyrus as measured by ELISA. Each sample was measured in triplicate, and the mean was used for statistical analysis. Data from acidified (Promega) and nonacidified (R&D Systems) assays were combined. Bars indicate mean ± SE. Autism, n = 11; control, n = 14. * p < 0.05 (2-tailed t test).

BDNF Protein Isoform Balance Measured by Western Blotting

Western blotting using a rabbit polyclonal anti-BDNF immunoglobulin G detected multiple immunoreactive bands in human brain (Fig. 3). Blocking immunoreactivity with an excess of antigenic peptide demonstrated specificity of several bands. The 33-kDa band corresponds to the molecular weight of pro-BDNF. The 28-kDa band in our blots corresponds to a 28-kDa band previously identified as a truncated form of pro-BDNF (59). The 14-kDa-immunoreactive band was identified as mature BDNF because it comigrates with recombinant human BDNF (48, 59). The 24-kDa band has not previously been described and might be an intermediate in the proteolytic processing of pro-BDNF to mature BDNF. This band constituted only a very small fraction (1%–3%) of the total BDNF signal; including it or not in the analysis did not affect the results.

FIGURE 3

Representative Western blot of fusiform gyrus homogenates from autism (A) and control (C) samples. Lanes 1 to 4: different amounts of total protein loaded for standard curve. Lanes 5 to 13: 20 μg of total protein from each autism and control sample. Lane 14: 0.3 ng of recombinant human BDNF.

FIGURE 3

Representative Western blot of fusiform gyrus homogenates from autism (A) and control (C) samples. Lanes 1 to 4: different amounts of total protein loaded for standard curve. Lanes 5 to 13: 20 μg of total protein from each autism and control sample. Lane 14: 0.3 ng of recombinant human BDNF.

Statistical analysis of the mean ratios of each BDNF isoform to total BDNF pixel values demonstrated a highly significant group × isoform interaction (2-way analysis of variance, p < 0.001). Western blotting revealed a significant decrease in the truncated BDNF/total BDNF ratio in autism samples versus controls (2-tailed t test, p < 0.01; Fig. 4) and an increase in pro-BDNF (2-tailed t test, p < 0.05; Fig. 4). The ratio of mature BDNF/total BDNF exhibited a trend toward increased values in autism samples versus controls (2-tailed t test, p = 0.15; Fig. 4).

FIGURE 4

Quantificationof BDNFin autism and control samples by Western blotting. For each sample, the pixel value of each BDNF isoform was expressed as a ratio to total BDNF pixel values, where total BDNF is the sum of pro-BDNF + 28-kDa truncated BDNF + mature BDNF. ** p < 0.01, * p < 0.05, 2-tailed t test. Bars represent mean ± SE. A = autism, n = 9; C = control, n = 9. Data are averaged from 2 to 3 independent Western blots per sample.

FIGURE 4

Quantificationof BDNFin autism and control samples by Western blotting. For each sample, the pixel value of each BDNF isoform was expressed as a ratio to total BDNF pixel values, where total BDNF is the sum of pro-BDNF + 28-kDa truncated BDNF + mature BDNF. ** p < 0.01, * p < 0.05, 2-tailed t test. Bars represent mean ± SE. A = autism, n = 9; C = control, n = 9. Data are averaged from 2 to 3 independent Western blots per sample.

No Influence of Age at Death or Postmortem Interval

There was no statistically significant difference observed between the age at death of the control group and age at death of the autism group (p = 0.12 for RNA studies, p = 0.25 for ELISA, p = 0.64 for Western blotting) (Tables 1 and 2). Similarly, there was no significant difference between mean postmortem interval (PMI) for the control group and mean PMI for the autism group (p = 0.67 for RNA studies, p = 0.85 for ELISA, p = 0.81 for Western). Although the PMIs were longer than the 12-hour period required for preservation of mature BDNF protein as reported by Ferrer et al (71), there was no correlation observed between PMI and BDNF protein levels measured by ELISA (R2 = 0.19, p = 0.36) or by Western blotting (R2 = 0.0015, p = 0.62). Therefore, PMI had little to no influence on BDNF protein levels in our study. Regression analysis showed that there was no correlation between BDNF mRNA levels and either age (R2 = 0.026, p = 0.46) or PMI (R2 = 0.065, p = 0.24). We did not find a correlation between BDNF protein levels measured by ELISA and age (R2 = 0.007, p = 0.97) or between BDNF protein levels measured by Western blotting and age (R2 = 0.001, p = 0.88).

Seizure and Drug History

Because seizures have been found to upregulate BDNF mRNA and protein (74–76), subjects who experienced seizures were excluded from the sample population when the tissue was first acquired. This exclusion criterion was to control for the effect of seizures on BDNF expression. However, after the experiments were conducted, subjects 4, 5, 6 and 7 were discovered to have comorbid seizure disorder. Analysis excluding these subjects did not change the group mean values or significance for either mRNA or protein levels measured by all 3 techniques. Comparison of mRNA and protein levels between subjects with autism who had experienced seizures versus subjects free of seizures (t tests) did not show any significant differences for mRNA (seizures, n = 4; no seizures, n = 5; p = 0.43), BDNF protein determined by ELISA (seizures, n = 4; no seizures, n = 7; p = 0.34), pro-BDNF (seizures, n = 4; no seizures, n = 5; p = 0.62), or truncated BDNF (seizures, n = 4; no seizures, n = 5; p = 0.79). However, levels of mature BDNF were increased in the group with seizures compared with the group with no seizures (seizures, n = 4; no seizures, n = 5; p = 0.04) and to the control group (n = 9; p = 0.005).

Anticonvulsant medications have been shown to correct BDNF upregulation caused by seizures (77). There were 5 AD subjects (4–7, 11) who were on anticonvulsant medication. A comparison between the AD subjects who were treated with anticonvulsant medication and those who were not (t tests) showed no significant differences in the mean ratio of BDNF mRNA (anticonvulsant, n = 5; none, n = 4; p = 0.97), BDNF protein determined by ELISA (anticonvulsant, n = 5; none, n = 6; p = 0.40), pro-BDNF (anticonvulsant, n = 4; none, n = 5; p = 0.62), or truncated BDNF (anticonvulsant, n = 4; none, n = 5; p = 0.79). However, there was a significant difference between groups for mature BDNF. The autism group treated with anticonvulsants exhibited significantly more mature BDNF/total than either the untreated autism group (p = 0.038) or the normal control group (p = 0.008).

Discussion

In this study, we found increased BDNF-immunoreactive protein in fusiform gyrus of subjects with autism by ELISA, consistent with earlier reports. Previous studies reported greater BDNF protein in postmortem basal forebrain tissue (66) and in blood and serum (61, 63–65) of autism samples compared with controls. Our study is the first to examine BDNF protein and mRNA in the fusiform gyrus, an area known to be functionally abnormal in autism. Our study is also the first to examine the mechanism of this increase. Using real-time RT-PCR, we found no difference in levels of BDNF mRNA in the fusiform gyrus of autism patients compared with controls, thereby demonstrating that the increase in BDNF-immunoreactive protein in the same samples is not transcriptionally driven. Importantly, Western blotting demonstrated that the increased BDNF immunoreactivity measured by ELISA was caused by significant changes in the relative levels of pro-BDNF and truncated BDNF isoforms between the 2 groups, suggesting that there may be changes in proteolytic processing of pro-BDNF to truncated BDNF in autism. Previous studies have reported increases in other neurotrophins and cytokines in autism brain tissue (78, 79), indicating that many secreted peptides are altered and suggesting that this could be a general mechanism leading to defective synaptic development and plasticity in autism.

It is unlikely that translation or stability could account for these differences in isoforms. Increased translation could explain the increase in pro-BDNF but cannot explain the decrease in truncated BDNF unless coupled with a corresponding defect in conversion to truncated BDNF. The lack of correlation between BDNF isoform levels and PMI in this study (in addition to our previously published experiments demonstrating no effect of PMI on BDNF levels [46, 47]) argues against postmortem degradation or protein instability as a mechanism. Previous studies indicate that seizures upregulate both BDNF mRNA and mature BDNF protein (74–76), but we found no significant differences in pro-BDNF or truncated BDNF levels between subjects with and without comorbid seizure disorder and subjects on anticonvulsant medication versus those not on such medication. Therefore, it is unlikely that seizures could account for increased pro-BDNF protein in our study without a concurrent increase in BDNF mRNA. Dysregulation of proteolytic processing in autism, unrelated to seizures, is a possible alternative mechanism. The protease SKI-1 converts pro-BDNF to truncated BDNF (59) and is, therefore, a good candidate for the cause of this dysregulation.

The major difference between previous studies using immunoaffinity chromatography and ELISA and our study including Western assays is that the ELISA measures antibody reactivity in solution, whereas in Western blotting, proteins are separated based on size, allowing one to visualize individual protein isoforms. Using Western blotting, we demonstrated a significant increase in pro-BDNF and a significant decrease in truncated BDNF in autism subjects compared with controls. Thus, the sandwich ELISAs that measure increased BDNF in autism likely detect pro-BDNF but are less sensitive toward truncated BDNF. Consistent with this, total BDNF as measured by Western blotting (pro-BDNF + truncated BDNF + mature BDNF, normalized to β-actin) (Tables 1, 2) did not differ between autism and control groups (t test, p = 0.35) and did not exhibit the 2-fold difference seen using ELISA. On the other hand, the ratio of pro-BDNF/β-actin measured in Western blots exhibited a trend toward an increase in autism compared with that in control subjects (t test, p = 0.07), suggesting again that the antibodies used for ELISA detect predominantly pro-BDNF.

Increased pro-BDNF and reduced truncated BDNF in AD strongly suggest the importance of proper balance among all 3 BDNF isoforms. The dramatic increase in pro-BDNF likely has undesirable consequences. The biological properties of pro-BDNF suggest that increases in this isoform may be responsible for reduced neuronal differentiation and dendritic spines and altered synaptic connectivity and neurotransmitter levels found in subjects with autism (35, 53–55). Distortion of the balance between BDNF isoforms could lead to changes in connectivity and synaptic plasticity and, hence, behavior. Our results focus attention on defective proteolytic maturation as a possible new mechanism for altered synaptic plasticity leading to autism.

Indeed, several studies support the hypothesis that deficits or imbalances in BDNF isoforms may lead to neuropsychiatric disorders. A polymorphism in the BDNF prodomain, which reduces pro-BDNF processing and BDNF secretion, correlates with deficits in episodic memory (56, 57); reduced proBDNF and BDNF correlate with cognitive decline in mild cognitive impairment and Alzheimer disease (47); and decreased truncated BDNF correlates with cognitive impairment in schizophrenia (60). Animal studies show that decreased BDNF impairs long-term potentiation, learning, and memory (80–82). Therefore, increased pro-BDNF and reduced truncated BDNF might have equally serious effects on learning, memory, and attention in autism. The biological role of the truncated isoform is unknown, and therefore, the full impact of this change in isoform ratios has yet to be determined.

In summary, we have shown that BDNF mRNA levels are unchanged in fusiform gyrus of subjects with autism compared with controls. BDNF-immunoreactive protein measured by ELISA is greater in subjects with autism than in controls, and Western blotting identifies the responsible molecular species as pro-BDNF. Increased pro-BDNF coupled with decreased truncated BDNF suggests that there may be a deficit in the processing of pro-BDNF into the truncated form in subjects with autism. Such a distortion of the balance of BDNF isoforms may have serious neuropsychiatric consequences.

References

1.
Rapin
I
.
Current concepts: Autism
.
N Engl J Med
 
1997
;
337
:
97
102
2.
Bailey
A
Le Couteur
A
Gottesman
I
et al
.
Autism as a strongly genetic disorder: Evidence from a British twin study
.
Psychol Med
 
1995
;
25
:
63
77
3.
Bailey
A
Luthert
P
Dean
A
et al
.
A clinicopathological study of autism
.
Brain
 
1998
;
121
:
889
905
4.
Kemper
TL
Bauman
M
.
Neuropathology of infantile autism
.
J Neuropathol Exp Neurol
 
1998
;
57
:
645
52
5.
Blatt
GJ
Fitzgerald
CM
Guptill
JT
et al
.
Density and distribution of hippocampal neurotransmitter receptors in autism: An autoradiographic study
.
J Autism Dev Disord
 
2001
;
31
:
537
43
6.
Bauman
ML
Kemper
TL
.
The neuropathology of the autism spectrum disorders: What have we learned?
Novartis Found Symp
 
2003
;
251
:
112
22
7.
Herbert
MR
Ziegler
DA
Deutsch
CK
et al
.
Dissociations of cerebral cortex, subcortical and cerebral white matter volumes in autistic boys
.
Brain
 
2003
;
126
:
1182
92
8.
Boucher
J
Lewis
V
.
Unfamiliar face recognition in relatively able autistic children
.
J Child Psychol Psychiatry
 
1992
;
33
:
843
59
9.
Allison
T
Ginter
H
McCarthy
G
et al
.
Face recognition in human extrastriate cortex
.
J Neurophysiol
 
1994
;
71
:
821
25
10.
Gauthier
I
Anderson
AW
Tarr
MJ
et al
.
Levels of categorization in visual recognition studied using functional magnetic resonance imaging
.
Curr Biol
 
1997
;
7
:
645
51
11.
Kanwisher
N
McDermott
J
Chun
MM
.
The fusiform face area: A module in human extrastriate cortex specialized for face perception
.
J Neurosci
 
1997
;
17
:
4302
11
12.
Schultz
RT
Gauthier
I
Klin
A
et al
.
Abnormal ventral temporal cortical activity during face discrimination among individuals with autism and Asperger syndrome
.
Arch Gen Psychiatry
 
2000
;
57
:
331
40
13.
Pierce
K
Muller
RA
Ambrose
J
et al
.
Face processing occurs outside the fusiform ‘face area’ in autism: Evidence from functional MRI
.
Brain
 
2001
;
124
:
2059
73
14.
Grelotti
DJ
Klin
AJ
Gauthier
I
et al
.
fMRI activation of the fusiform gyrus and amygdala to cartoon characters but not to faces in a boy with autism
.
Neuropsychologia
 
2005
;
43
:
373
85
15.
Fombonne
E
.
Epidemiological surveys of autism
. In:
Volkmar
F
, ed.
Autism and Pervasive Developmental Disorders
 .
Cambridge University Press
,
Cambridge
,
1998
16.
Cook
EH
Jr
Scherer
SW
.
Copy-number variations associated with neuropsychiatric conditions
.
Nature
 
2008
;
455
:
919
23
17.
Pinto
D
Pagnamenta
AT
Klei
L
et al
.
Functional impact of global rare copy number variation in autism spectrum disorders
.
Nature
 
2010
;
466
:
368
72
18.
Kumar
RA
KaraMohamed
S
Sudi
J
et al
.
Recurrent 16p11.2 microdeletions in autism
.
Hum Mol Genet
 
2008
;
17
:
628
38
19.
Marshall
CR
Noor
A
Vincent
JB
et al
.
Structural variation of chromosomes in autism spectrum disorder
.
Am J Hum Genet
 
2008
;
82
:
477
88
20.
Weiss
LA
Arking
DE
Gene Discovery Project of Johns Hopkins & the Autism Consortium. A genome-wide linkage and association scan reveals novel loci for autism
.
Nature
 
2009
;
461
:
802
8
21.
Jamain
S
Quach
H
Betancur
C
et al
.
Mutations of the x-linked genes encoding neurolignins NLGN3 and NLGN4 are associated with autism
.
Nat Genet
 
2003
;
34
:
27
29
22.
Durand
CM
Betancur
C
Boeckers
TM
et al
.
Mutations in the gene encoding the synaptic scaffolding protein SHANK3 are associated with autism spectrum disorders
.
Nat Genet
 
2007
;
39
:
25
27
23.
Autism Genome Project Consortium
Szatmari
P
Paterson
AD
et al
.
Mapping autism risk loci using genetic linkage and chromosomal rearrangements
.
Nat Genet
 
2007
;
39
:
319
28
24.
Bourgeron
T
.
A synaptic trek to autism
.
Curr Opin Neurobiol
 
2009
;
19
:
231
34
25.
Romorini
S
Piccoli
G
Jiang
M
et al
.
A functional role of postsynaptic density-95-guanylate kinase-associated protein complex in regulating Shank assembly and stability to synapses
.
J Neurosci
 
2004
;
24
:
9391
404
26.
Hussman
JP
Chung
RH
Griswold
AJ
et al
.
A noise-reduction GWAS analysis implicates altered regulation of neurite outgrowth and guidance in autism
.
Mol Autism
 
2011
;
2
:
1
27.
Anney
R
Klei
L
Pinto
D
et al
.
A genome-wide scan for common alleles affecting risk for autism
.
Hum Mol Genet
 
2010
;
19
:
4072
82
28.
Berkel
S
Marshall
CR
Weiss
B
et al
.
Mutations in the SHANK2 synaptic scaffolding gene in autism spectrum disorder and mental retardation
.
Nat Genet
 
2010
;
42
:
489
91
29.
Lewin
GR
Barde
YA
.
Physiology of the neurotrophins
.
Annu Rev Neurosci
 
1996
;
19
:
289
317
30.
Huang
EJ
Reichardt
LF
.
Neurotrophins: Roles in neuronal development and function
.
Annu Rev Neurosci
 
2001
;
24
:
677
736
31.
Chapleau
CA
Larimore
JL
Theibert
A
et al
.
Modulation of dendritic spine development and plasticity by BDNF and vesicular trafficking: Fundamental roles in neurodevelopmental disorders associated with mental retardation and autism
.
J Neurodev Disord
 
2009
;
1
:
185
96
32.
McAllister
AK
Lo
DC
Katz
LC
.
Neurotrophins regulate dendritic growth in developing visual cortex
.
Neuron
 
1995
;
15
:
791
803
33.
Horch
HW
.
Local effects of BDNF on dendritic growth
.
Rev Neurosci
 
2004
;
15
:
117
29
34.
Bramham
CR
Messaoudi
E
.
BDNF function in adult synaptic plasticity: The synaptic consolidation hypothesis
.
Prog Neurobiol
 
2005
;
76
:
99
125
35.
Pang
PT
Teng
HK
Zaitsev
E
et al
.
Cleavage of pro-BDNF by tPA/plasmin is essential for long-term hippocampal plasticity
.
Science
 
2004
;
306
:
487
91
36.
Tyler
WJ
Alonso
M
Bramham
CR
et al
.
From acquisition to consolidation: On the role of brain-derived neurotrophic factor signaling in hippocampal-dependent learning
.
Learn Mem
 
2002
;
9
:
224
37
37.
Zakharenko
SS
Patterson
SL
Dragatsis
I
et al
.
Presynaptic BDNF required for a presynaptic but not postsynaptic component of LTP at hippocampal CA1-CA3 synapses
.
Neuron
 
2003
;
39
:
975
90
38.
Lu
B
.
Pro-region of neurotrophins: Role in synaptic modulation
.
Neuron
 
2003
;
39
:
735
38
39.
McAllister
AK
.
Subplate neurons: A missing link among neurotrophins, activity, and ocular dominance plasticity?
Proc Natl Acad Sci USA
 
1999
;
96
:
13600
2
40.
Weickert
CS
Hyde
TM
Lipska
BK
et al
.
Reduced brain-derived neurotrophic factor in prefrontal cortex of patients with schizophrenia
.
Mol Psychiatry
 
2003
;
8
:
592
610
41.
Durany
N
Michel
T
Zochling
R
et al
.
Brain-derived neurotrophic factor and neurotrophin 3 in schizophrenic psychoses
.
Schizophr Res
 
2001
;
52
:
79
86
42.
Green
MJ
Matheson
SL
Shepherd
A
et al
.
Brain-derived neurotrophic factor levels in schizophrenia: A systematic review with meta-analysis
.
Mol Psychiatry
 
2010
;
16
:
960
72
43.
Karege
F
Perret
G
Bondolfi
G
et al
.
Decreased serum brain-derived neurotrophic factor levels in major depressed patients
.
Psychiatry Res
 
2002
;
109
:
143
48
44.
Connor
B
Young
D
Yan
Q
et al
.
Brain-derived neurotrophic factor is reduced in Alzheimer's disease
.
Brain Res Mol Brain Res
 
1997
;
49
:
71
81
45.
Holsinger
RM
Schnarr
J
Henry
P
et al
.
Quantitation of BDNF mRNA in human parietal cortex by competitive reverse transcription-polymerase chain reaction: Decreased levels in Alzheimer's disease
.
Brain Res Mol Brain Res
 
2000
;
76
:
347
54
46.
Michalski
B
Fahnestock
M
.
Pro-brain-derived neurotrophic factor is decreased in parietal cortex in Alzheimer's disease
.
Brain Res Mol Brain Res
 
2003
;
111
:
148
54
47.
Peng
S
Wuu
J
Mufson
EJ
et al
.
Precursor form of brain-derived neurotrophic factor and mature brain-derived neurotrophic factor are decreased in the preclinical stages of Alzheimer's disease
.
J Neurochem
 
2005
;
93
:
1412
21
48.
Mowla
SJ
Farhadi
HF
Pareek
S
et al
.
Biosynthesis and posttranslational processing of the precursor to brain-derived neurotrophic factor
.
J Biol Chem
 
2001
;
276
:
12660
66
49.
Seidah
NG
Benjannet
S
Pareek
S
et al
.
Cellular processing of the neurotrophin precursors of NT3 and BDNF by the mammalian proprotein convertases
.
FEBS Lett
 
1996
;
379
:
247
50
50.
Lee
R
Kermani
P
Teng
KK
et al
.
Regulation of cell survival by secreted proneurotrophins
.
Science
 
2001
;
294
:
1945
48
51.
Murer
MG
Boissiere
F
Yan
Q
et al
.
An immunohistochemical study of the distribution of brain-derived neurotrophic factor in the adult human brain, with particular reference to Alzheimer's disease
.
Neuroscience
 
1999
;
88
:
1015
32
52.
Fahnestock
M
Garzon
D
Holsinger
RM
et al
.
Neurotrophic factors and Alzheimer's disease: Are we focusing on the wrong molecule?
J Neural Transm Suppl
 
2002
;
62
:
241
52
53.
Teng
HK
Teng
KK
Lee
R
et al
.
Pro-BDNF induces neuronal apoptosis via activation of a receptor complex of p75NTR and sortilin
.
J Neurosc
 
2005
;
25
:
5455
63
54.
Woo
NH
Teng
HK
Siao
CJ
et al
.
Activation of p75NTR by pro-BDNF facilitates hippocampal long-term depression
.
Nat Neurosci
 
2005
;
8
:
1069
77
55.
Koshimizu
H
Kiyosue
K
Hara
T
et al
.
Multiple functions of precursor BDNF to CNS neurons: Negative regulation of neurite growth, spine formation and cell survival
.
Mol Brain
 
2009
;
2
:
27
56.
Egan
MF
Kojima
M
Callicott
JH
et al
.
The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and human memory and hippocampal function
.
Cell
 
2003
;
112
:
257
69
57.
Hariri
AR
Goldberg
TE
Mattay
VS
et al
.
Brain-derived neurotrophic factor val66met polymorphism affects human memory-related hippocampal activity and predicts memory performance
.
J Neurosci
 
2003
;
23
:
6690
94
58.
Hong
CJ
Yu
YW
Lin
CH
et al
.
An association study of a brain-derived neurotrophic factor Val66Met polymorphism and clozapine response of schizophrenic patients
.
Neurosci Lett
 
2003
;
349
:
206
20
59.
Seidah
NG
Mowla
SJ
Hamelin
J
et al
.
Mammalian subtilisin/kexin isozyme SKI-1: A widely expressed proprotein convertase with a unique cleavage specificity and cellular localization
.
Proc Natl Acad Sci USA
 
1999
;
96
:
1321
26
60.
Carlino
D
Leone
E
Di Cola
F
et al
.
Low serum truncated-BDNF isoform correlates with higher cognitive impairment in schizophrenia
.
J Psychiatr Res
 
2011
;
45
:
273
79
61.
Nelson
KB
Grether
JK
Croen
LA
et al
.
Neuropeptides and neurotrophins in neonatal blood of children with autism or mental retardation
.
Ann Neurol
 
2001
;
49
:
597
606
62.
Nelson
PG
Kuddo
T
Song
EY
et al
.
Selected neurotrophins, neuropeptides, and cytokines: Developmental trajectory and concentrations in neonatal blood of children with autism or Down syndrome
.
Int J Dev Neurosci
 
2006
;
24
:
73
80
63.
Miyazaki
K
Narita
N
Sakuta
R
et al
.
Serum neurotrophin concentrations in autism and mental retardation: A pilot study
.
Brain Dev
 
2004
;
26
:
292
95
64.
Connolly
AM
Chez
M
Streif
EM
et al
.
Brain-derived neurotrophic factor and autoantibodies to neural antigens in sera of children with autistic spectrum disorders, Landau-Kleffner syndrome, and epilepsy
.
Biol Psychiatry
 
2006
;
59
:
354
63
65.
Correia
CT
Coutinho
AM
Sequeira
AF
et al
.
Increased BDNF levels and NTRK2 gene association suggest a disruption of BDNF/TrkB signaling in autism
.
Genes Brain Behav
 
2010
;
9
:
841
48
66.
Perry
EK
Lee
ML
Martin-Ruiz
CM
et al
.
Cholinergic activity in autism: Abnormalities in the cerebral cortex and basal forebrain
.
Am J Psychiatry
 
2001
;
158
:
1058
66
67.
Nishimura
K
Nakamura
K
Anitha
A
et al
.
Genetic analyses of the brain-derived neurotrophic factor (BDNF) gene in autism
.
Biochem Biophys Res Comm
 
2007
;
356
:
200
6
68.
Lord
C
Rutter
M
Le Couteur
A
.
Autism diagnostic interview-revised: A revised version of a diagnostic interview for caregivers of individuals ith possible pervasive developmental disorders
.
Autism Dev Dis
 
1994
;
24
:
659
85
69.
Garzon
DJ
Fahnestock
M
.
Oligomeric amyloid decreases basal levels of brain-derived neurotrophic factor (BDNF) mRNA via specific down-regulation of BDNF transcripts IV and V in differentiated human neuroblastoma cells
.
J Neurosci
 
2007
;
27
:
2628
35
70.
Fahnestock
M
Michalski
B
Xu
B
et al
.
The precursor pro-nerve growth factor is the predominant form of nerve growth factor in brain and is increased in Alzheimer's disease
.
Mol Cell Neurosci
 
2001
;
18
:
210
20
71.
Ferrer
I
Marin
C
Rey
MJ
et al
.
BDNF and full-length and truncated TrkB expression in Alzheimer disease. Implications in therapeutic strategies
.
J Neuropathol Exp Neurol
 
1999
;
58
:
729
39
72.
Okragly
AJ
Haak-Frendscho
M
.
An acid-treatment method for the enhanced detection of GDNF in biological samples
.
Exp Neurol
 
1997
;
145
:
592
96
73.
Zettler
C
Bridges
DC
Zhou
XF
et al
.
Detection of increased tissue concentrations of nerve growth factor with an improved extraction procedure
.
J Neurosci Res
 
1996
;
46
:
581
94
74.
Gall
C
Lauterborn
J
Bundman
M
et al
.
Seizures and the regulation of neurotrophic factor and neuropeptide gene expression in brain
.
Epilepsy Res Suppl
 
1991
;
4
:
225
45
75.
Rudge
JS
Mather
PE
Pasnikowski
EM
et al
.
Endogenous BDNF protein is increased in adult rat hippocampus after a kainic acid-induced excitotoxic insult but exogenous BDNF is not neuroprotective
.
Exp Neurol
 
1998
;
149
:
398
410
76.
Binder
DK
Croll
SD
Gall
CM
et al
.
BDNF and epilepsy: Too much of a good thing?
Trends Neurosci
 
2001
;
24
:
318
19
77.
Lavebratt
C
Trifunovski
A
Persson
AS
et al
.
Carbamazepine protects against megencephaly and abnormal expression of BDNF and Nogo signaling components in the mceph/mceph mouse
.
Neurobiol Dis
 
2006
;
24
:
374
83
78.
Sajdel-Sulkowska
EM
Xu
M
Koibuchi
N
.
Increase in cerebellar neurotrophin-3 and oxidative stress markers in autism
.
Cerebellum
 
2009
;
8
:
366
72
79.
Vargas
DL
Nascimbene
C
Krishnan
C
et al
.
Neuroglial activation and neuroinflammation in the brain of patients with autism
.
Ann Neurol
 
2005
;
57
:
67
81
80.
Korte
M
Carroll
P
Wolf
E
et al
.
Hippocampal long-term potentiation is impaired in mice lacking brain-derived neurotrophic factor
.
Proc Natl Acad Sci USA
 
1995
;
92
:
8856
60
81.
Patterson
SL
Abel
T
Deuel
TA
et al
.
Recombinant BDNF rescues deficits in basal synaptic transmission and hippocampal LTP in BDNF knockout mice
.
Neuron
 
1996
;
16
:
1137
45
82.
Blurton-Jones
M
Kitazawa
M
Martinez-Coria
H
et al
.
Neural stem cells improve cognition via BDNF in a transgenic model of Alzheimer disease
.
Proc Natl Acad Sci USA
 
2009
;
106
:
13594
99

Author notes

We gratefully acknowledge the Harvard Brain Tissue Resource Center, the University of Maryland Brain and Tissue Bank and the Autism Speaks' Autism Tissue Program for providing brain tissues for this study.
This work was supported by a grant from the Ontario Mental Health Foundation to Margaret Fahnestock. Kristine L.P. Garcia was supported by a studentship from Autism Speaks. Diego J. Garzon was supported by a studentship from the Alzheimer Society of Canada.