Blood extraction and processing
With the exception of the microbiome, all other datasets are derived from a single blood draw that is carefully processed into various fractions. A goal of the pilot study in year 1 is to determine the minimal volume required to produce each dataset from smaller blood samples obtained from children. After extraction by phlebotomists, blood will be immediately transported to the Sample Processing Center at the Crnic Institute, where specialists will process the samples to a point where they can be safely cryopreserved until shipment to the research teams.
Our goal is to produce the deepest phenotyping of individuals with T21 to date. Health and medical records and family histories will be entered into a research-friendly database via an online survey with > 500 fields of clinical metadata. Different modules will be completed by relatives and research assistants embedded at the various recruitment sites. The survey will be hosted in a RedCap database and will be accessible remotely from the various clinics. A tiered-access system will enable different users to access specific modules. The survey is organized with ‘branching logic’, whereby the depth of the phenotyping varies with the fields enabled by the various participants, as defined by their medical condition. For example, participants that report ‘autism’ would qualify for deeper phenotyping by a specialized team. Adult participants who are determined to suffer mild cognitive impairment by the RMADC staff will qualify for deeper phenotyping of the brain pathology of AD. Cognitive assessment will be performed at the various clinics employing customized tests. Data will be stored in secure, HIPAA-compliant servers administered by BIPM.
For whole genome sequencing, a fraction of peripheral blood cells (PBMCs) will be collected, their DNA will be extracted with standard protocols, and shipped to a provider selected through a competitive RFP (e.g. Macrogen, Hudson Alpha). A fraction of DNA will be provided to the Rocky Mountain Biorepository, where it would be subject to Illumina GWAS analysis prior to storage.
EpigENOME and transcriptome analysis
For analysis of the epigenomes and transcriptomes, the naïve cell type of choice is monocytes, for several reasons: i) they are easy to purify; ii) they constitute 20-30% of PBMCs, iii) unlike B and T cells, they do not undergo genome rearrangements or hypersomatic mutations; iv) they are considered to be a ‘sentinel’ cell type that can report on the physiological state or natural history of an individual; v) they are involved in neuroinflammatory processes relevant to AD. Monocytes will be purified using ‘positive selection’ protocols. We will explore the epigenome by complementary DNA methylation (MCC-seq. and llumina Human Methylation450 beadchip) and chromatin accesibility assays (ATAC-seq).
Analysis of the transcriptome will be performed from the same monocyte fraction subject to epigenome analysis, as both datasets are extremely synergistic to inform on patterns of gene activity. RNA depleted of ribosomal RNAs will be analyzed via RNAseq using established protocols. This technology has been successfully employed by members of our team to identy RNA signatures unique to T21 cells that could explain diverse aspects of DS. In the future, epigenomic and RNA signatures could be determined from immortalized cell lines, iPSCs, and cell types derived from iPSCs.
Red blood cells (RBCs) will be employed for quantitative measurements of >400 metabolites using well established massspectrometry technology. The choice of RBCs is based on the fact that they act as ‘metabolite sinks’, and have been sucessfully employed by our team to identify metabolic signatures specific to disease states. In fact, we already employed this technology to identify metabolites that are different between disomic and T21 fibroblasts, which could explain some aspects of DS pathology.
With the advent of genome-wide tools for gene knockdown (shRNA) and knockout (CRISPR), it is now possible to perform loss-of-function screens to test for the contribution of >20,000 human genes to various cellular functions. A powerful use of this technology is to determine ‘essentiality scores’ for each gene, as defined by their contribution to cell survival and proliferation. Members of our team have extensive experience and have employed this approach to identify gene products that are differentially required by disomic versus T21 cells. This dataset will be generated originally from LCLs but could be later derived from iPSCs or differentiated cells derived from iPSCs.
A deep characterization of the blood is warranted for several reasons: i) many important phenotypes in the T21 population have a molecular or cellular correlate in the blood (e.g. thyroid hormone levels and hypothyroidism, various cell counts for leukemia and hematological conditions); ii) since most datasets are derived from blood, it is critical to characterize in detail the abundance and signaling state of as many blood cell types as possible. Therefore, the ‘Bloodwork’ includes quantification of dozens of common analytes, plus deep characterization of all major peripheral cell types using CyTOF Fluidigm technology
There is ample evidence that the host microbiome plays instructive roles in health and disease. We hypothesize that several co-morbidities observed in DS can be modulated by the gut microbiome and our team has extensive experience linking gut dysbiosis to host genotypes and phenotypes in a variety of chronic human disease contexts. Therefore, we will examine the gut microbiome using established protocols for determination of microbial taxa from stool samples.
Immortalized cell lines and Stem Cells
We will generate immortalized lymphoblastoid cell lines (LCLs) from circulating B-cells using EBV transformation and standard protocols. Team members have significant expertise in this area, and ~60 such cell lines derived from individuals with T21 are already available at UCD-SOM through the Nexus biobank (see letter from Dr. Robinson Rosenberg). LCLs will be an enormous resource for future investigations requiring large cell numbers and a key component of the Biobank.
We will generate iPSCs from circulating vascular endothelial progenitor cells (VEPCs) using a novel vector-free, RNA-based reprogramming protocol developed by the Bilousova team at the Gates Center for Regenerative Medicine. Although iPSCs can be generated from other cell types in the blood, the efficiency is low and it requires prohibitively large cell volumes. In contrast, VEPCs can be easily captured by adherence to plastic and expanded in culture. The Bilousova protocol can generate iPSCs from single adherent cells.