Research Proposal Summary
Title: Exploratory Gene Expression Analysis of Childhood Autism
Principal Investigator: Roger Kurlan, MD, Neurology, Atlantic Neuroscience Institute, Overlook Medical Center, Summit, NJ
Co-Investigators: Christopher Eckman, PhD and Elizabeth Eckman, PhD, Neuroscience,
Morristown Medical Center, Morristown, NJ
Harvey Bennett, MD, Pediatric Neurology,
Andrew Shenkman, MD, Neonatology,
Kathleen Selvaggi-Fadden, Developmental Pediatrics,
Tara Gleeson, RNP, Goryeb Children’s Hospital, Morristown, NJ
Timothy Mhyre, PhD and Howard Federoff, MD, PhD, Neuroscience,
Georgetown University School of Medicine, Washington, DC
Background and Specific Aims:
Although a large number of hypotheses regarding the etiology and pathogenesis of childhood autism have been presented, little definitive information exists. Little is known about the brain neurodevelopmental disturbances that underlie autism. High throughput quantitative analysis of the expression of a large number of genes using microarrays is a relatively new technology that can be used to quantify mRNA synthesis for the entire genome on a single chip. Rigorous statistical/bioinformatics methods are used to determine a threshold above which the “signal” of differentially expressed genes is considered well above possible “noise” due to variability associated with the sampling process. Statistical methods serve to rank genes in order of differential expression and to estimate the rate of false positive results induced by tens of thousands of comparisons. Generally, and in our research plans, genes found to be of interest in the microarray data analyses then undergo a validation process. With this approach, gene expression profiles that distinguish a disease state (e.g., autism) can be identified and, in turn, suggest critical disease-causing mechanisms. Gene expression analysis is a powerful screening tool for conditions like autism where few mechanistic clues are present. Gene expression is indeed the basic responsible process underlying all biological activities.
A published gene expression profiling study involved monozygotic twins discordant with respect to the severity of autism and/or language impairment and found a differential gene expression pattern. Most of the differentially expressed genes are known to be important in the development, structure and function of the nervous system [Hu et al, 2006]. Gene expression analysis was used as part of a mutimodal genetic investigation of autism that identified CNTNAP2 (a member of the neurexin family of genes that are known to be involved with early brain development) as a susceptibility gene for language deficits in autism [Alarcon et al, 2008].
We are proposing to study blood samples since they are readily attainable from young patients and an ultimate test for autism (a potential result of gene expression analysis) should be accessible from easy to obtain tissue such as blood. This is a reasonable approach since it is possible that some diseases like autism considered to be of the brain might result from systemic processes or have systemic components that can be identified in peripheral tissues like blood. Similar to the central nervous system, the blood compartment contains long-lived cells that share genetic programs and systemic pathological processes that may be reflected in circulating blood cells. As evidence for this possibility, it has been shown that selective reduction in mitochondrial complex I is seen in both the brain and circulating platelets of patients with Parkinson’s disease (PD). Brain dopamine dysfunction in PD is reflected by evidence of reduction of dopamine transporter in white blood cells from patients with the illness. Our own and other studies have identified distinct gene expression patterns in blood samples from patients with PD, ALS, Huntington’s disease, and Alzheimer’s disease. A small study involving 35 children with autism and 12 normal controls used gene expression analysis of blood samples and found 7 genes that were differentially expressed between these groups [Gregg et al, 2008]. The same group was able to find gene expression patterns that distinguished autism cases with early onset and those with developmental regression.
Candidate genes identified by our proposed screening study can be followed up by a logical sequence of further investigation that may include study in animal models (such as the creation of transgenic animals) and high throughput screening of drugs that influence the expression of these genes or their products and have other properties, such as the ability to cross the blood brain barrier.
We now propose a case-control exploratory study to determine if there are discovery gene expression profiles linked to patients with autism. We plan to begin with a pilot investigation involving a relatively small number of subjects. The purposes of the pilot study are to obtain experience in conducting this kind of research in the local school districts, determine parents’ willingness to provide informed consent, work out logistical issues in the conduct of the study (such as interviewing children in the school during the school day, obtaining blood samples in children with behavioral problems), determine the availability of subjects with specific diagnoses and the extent of the documentation in school and medical records available to confirm those diagnoses, conduct preliminary data analyses to assist with accurate power (sample size) analyses and determine the best subject groups to include in the larger, more definitive study. In the pilot project, we propose the inclusion of 20 subjects in 2 diagnostic groups: 1) Autistic disorder, 2) age, and gender matched normals. In the pilot study, we are looking for samples with fairly homogeneous demographic features. Once the pilot study is completed, we expect to extend the investigation to a larger sample of subjects (estimated total of 50-100 per group) with possibly more heterogeneous demographics.
The specific aims are:
1. Using blood samples, to determine if there are discovery gene expression profiles as generated by microarray analyses that distinguish cases with autism from controls without autism. This work will take place in 2 phases. The first is a pilot study involving a small number of subjects and the second will be a more definitive study using a larger sample size. A secondary aim will be to determine if there is a specific gene expression profile linked to the presence of the abnormal movements tics or stereotypies. In the future, we can also assess other clinical subtypes.
2. Should a gene expression profile that is linked to autism be identified, the discovery microarray results will be validated with a directed quantitative method, quantitative real-time polymerase chain reaction (qRT-PCR).
3. Blood samples will also be analyzed for beta-amyloid and tau, two proteins critically involved with brain development and disease.
4. Blood samples will be processed and stored for future analyses that might stem from the initial results.
The identification of a gene expression profile unique to autism would potentially generate a diagnostic test for the condition (autism is notorious for its heterogeneous clinical features resulting in problems making an accurate diagnosis on clinical grounds alone) and holds promise for elucidating underlying mechanisms of the illness and identifying new therapies.
Methods and Procedures
I. Subjects
A. Enrollment Criteria
1. Autistic disorder: age 14-21, male gender, meets DSM-IV-TR criteria for Autism Disorder, IQ >70 (not diagnosed with mental retardation), is a student in the Morris-Union Jointure system, no other known medical/neurological diagnosis is responsible for autism (e.g., Rett syndrome, head trauma, postencephalitis), dysmorphic features are absent, signed informed consent provided by parent or guardian, subject provides assent to participate in the procedures of the study.
2. Normals: age 14-21, male gender, no diagnosis of a neurodevelopmental disorder, friend or sibling of a case, a student in the Morris-Union County school districts, signed informed consent provided by parent of guardian, subject provides assent to participate in the procedures of the study.
B. Number
Pilot study: 20 cases in each of the 2 diagnostic groups.
Definitive study: total of 50-100 subjects in each of the 2 groups.
C. Procedures
1. Parents invited to an evening or Saturday informational meeting at the Developmental Learning Center in Warren with the investigators at which time the study rationale and procedures will be explained and questions will be answered.
2. Obtain informed consent from parent/guardian of each subject and assent from the subject.
3. Investigator meets with each consented student in the school. Parents are encouraged to attend. At the meeting, the following activities will take place:
a. Finalize informed consent if needed.
b. Review developmental history, school and medical records, and current symptoms, observe the child, in order to determine whether or not subject satisfies DSM criteria. If there is insufficient data in the records to estimate IQ, additional testing will be done using the Kaufman Brief Intelligence Test (K-BIT).
c. Record whether or not tics or stereotypies are observed.
d. Administer the Childhood Autism Rating Scale (CARS) in order to assess the severity of autism.
e. Obtain blood samples. Whenever possible, blood samples will be obtained in an overnight fasting state between 8:00 a.m. and 10:00 a.m. For gene expression (RNA) analysis, 7.5 cc of blood will be collected by venipuncture into three PAXgene tubes (2.5 cc/tube; Qiagen, Inc.) and inverted 12 times. For other biomarker assays (beta-amyloid, tau), we will obtain 7.5 cc of blood in red/gray top tubes and 6 cc of blood in lavender top tubes which will be processed and stored (aliquots of serum and plasma) in the Eckman laboratory.
4. After the school interview and phlebotomy, the following activities will take place:
a. Transport the PAXgene tubes to the Eckman laboratory for processing and storage. The processed samples will be stored at -80◦C in Dr. Eckman’s laboratory until shipped to the Mhyre/Federoff laboratory at Georgetown University.
b. Batch shipping of some of the processed samples to the microarray laboratory at Georgetown University.The PAXgene tubes will be processed at Georgetown according to the manufacturer’s directions using the PAXgene Blood RNA Kit (Qiagen) following a 2-hour room temperature incubation. A small aliquot of each subject’s RNA will be analyzed using the Agilent 2100 Bioanalyzer to assess overall quantity and quality of each sample. Global transcript expression in peripheral leukocytes will be analyzed using the Affymetrix Human Genome U133 Plus 2.0 Set, which contains >47,000 unique human gene/transcript targets. The Mhyre/Federoff laboratories have extensive experience using these arrays in human and animal model studies of Parkinson disease [Miller, Callahan et al. 2004; Miller and Federoff 2005]. Using the biostatistical methods detailed below, a candidate list of up to 100 genes that are differentially expressed between the autism and control groups will be prepared.
Differentially expressed gene transcripts will be validated using the TaqManÒ Low Density Arrays and the 7900HT Fast Real-Time polymerase chain reaction System (Applied Biosystems). First strand cDNA for each sample will be produced from 1 ug total RNA using the High Capacity cDNA Archive Kit, following the manufacturer’s protocol using random hexamer primers. cDNA and 2X TaqMan® Universal polymerase chain reaction Master Mix will be loaded onto the Low Density Arrays (preconfigured based upon the Affymetrix data) using the manufacturer’s standard protocol. Each target will be analyzed in quadruplicate and all data will be normalized for equal array loading based upon 18s rRNA expression using the SDS software.
h. In the Eckman lab, plasma amyloid-beta-40 and amyloid-beta-42 will be analyzed by sandwhich ELISA (Waco, USA) using the well-characterized BNT77/BA27 and BNT77/BC05 antibody pairs [Scheuner et al, 1996]. Serum total tau protein will be analyzed by sandwhic ELISA (Invitrogen).
i. Initial and follow-up data analyses.