Building a Biobank to Explore Mysteries of the Genome
GTEx is collecting multiple tissue samples from an estimated 1,000 individual donors for genetic research. (Image from the National Disease Research Interchange GTEx Team)
The architects of the biobank wanted nothing left to chance and everything well documented.
That’s why they developed 150 standard operating procedures to ensure that tissue samples were collected, processed, and stored in exactly the same way. And that’s why they are collecting data on the best temperatures for shipping the samples across the United States.
All that planning is paying off for the Genotype-Tissue Expression (GTEx) project, which will use the samples to investigate how genes are regulated in health and disease. Sponsored by the National Institutes of Health (NIH), the project has nearly 4,400 samples of “normal” human tissue from about 175 donors. By collecting many more samples, the project aims to be a resource for studying genetic variation and the regulation of genes in specific tissues.
Normal tissue—that is, tissue with no signs of disease—is not routinely collected for research. GTEx is the first large-scale project to collect high-quality samples of up to 32 tissue types from many individual donors.
During a pilot study that began last year, investigators with NCI’s cancer Human Biobank (caHUB) and their collaborators were responsible for acquiring and managing the biospecimens. On average, each organ and tissue donor has contributed 25 types of postmortem tissue, including heart, muscle, and skin. Surgical donors have also contributed tissues.
Based on the success of the pilot, the NIH Common Fund is scaling up GTEx, with a goal of reaching 1,000 donors in 3 years. This larger number of donors and samples will provide the statistical power that is needed to ask fundamental questions about the genome, the researchers said.
All cells in the human body contain essentially the same complement of genes, but these genes are activated, or expressed, differently in different types of cells. For the first time, GTEx will allow researchers to investigate how common genetic variants influence the regulation of gene expression using a set of reference tissues.
The relationship between genetic variants and gene regulation in different tissues is “a fascinating biology question, but it is also relevant in medicine,” said Dr. Barbara Stranger of Harvard Medical School and Brigham and Women’s Hospital, who studies genes and complex diseases but is not involved in GTEx.
“Healthy individuals will have some of the same regulatory processes as people with a complex disease,” Dr. Stranger continued. Understanding these processes could provide information about a variety of diseases.
Donors and Families
“No one really knew if this would work,” said Dr. Sherilyn Sawyer, who co-led the development of the GTEx biobank and until recently was part of NCI’s Office of Biorepositories and Biospecimen Research (OBBR). “So we’re ecstatic that the infrastructure we put in place through caHUB has resulted in the specimens that have produced an amazing collection of data.”
GTEx works with organizations involved in tissue and organ donation in several cities. When an organ or tissue donor who is eligible for GTEx passes away, one of these organizations contacts the deceased person’s family to ask permission to collect tissues from their loved one for the study.
“It’s a sensitive time, so care is taken,” said Anna M. Smith of the Frederick National Laboratory for Cancer Research, who works on ethical, legal, and regulatory affairs for GTEx. “It goes without saying that there would be no biobank without the donors and families.”
GTEx minimizes changes to tissue by using standard procedures to collect, process, and store the samples. (Image from the National Disease Research Interchange GTEx Team)
Once specimens are collected, they are preserved and shipped to the Van Andel Institute in Grand Rapids, MI, which is the central repository of materials. By collecting, processing, and storing the tissue samples in a controlled and uniform way, the researchers hope to maintain them in a nearly natural state and minimize alterations.
“The quality of the biospecimen is extraordinarily important,” said Dr. Sawyer. Collecting tissues with uniformly high RNA and DNA quality helps ensure that the information gained from analyzing the samples will “be as accurate to the biology of the tissues as possible,” she explained.
GTEx staff regularly scrutinize tissue samples to be sure that they are of high quality. “GTEx has a higher degree of quality-control management over the whole process than other projects we have done,” noted Dr. Scott Jewell, who directs the Program for Biospecimen Science at the Van Andel Institute.
Whether the same amount of quality control that went into GTEx will be required for all biobank projects is not yet known, Dr. Jewell added. But the lessons learned from GTEx could guide future efforts. Toward this end, researchers have released approximately 150 standard operating procedures related to biobanking.
“This project turned out to be a hugely complex logistical challenge,” said Dr. Jim Vaught, deputy director of OBBR. But with these challenges have come opportunities to ask important questions about biobanking.
For example, no one knows the best temperatures for shipping biospecimens. The GTEx team developed shipping containers that have “data loggers” to record the temperature at every minute. These data could help reveal the optimal temperatures for shipping biospecimens.
A Universe of Data
The GTEx project will more than triple its pool of donors in the next 3 years. Success will depend on a robust and flexible information technology system, noted Charles Shive, director of software development at the Frederick National Laboratory for Cancer Research.
A web-based application developed by Shive and his colleagues allows project members with various levels of access to data to communicate and monitor the progress of specimen processing and data collection in real time. “We know where specimens are at all times in the GTEx pipeline,” said Shive.
Pathologists review images of all tissue samples to assess the quality of biospecimens. Using the web application, pathologists can add their comments to “the universe of data associated with each sample,” Shive said.
“Unless the research is done and the data are generated, no one is ever going to know the answers,” said Smith.
Another question GTEx could help answer is how long tissues remain scientifically useful after the blood flow to an organ has been cut off. During the pilot study, GTEx researchers at the Broad Institute, in Cambridge, MA, observed a drop-off in RNA quality that was linked to the interval between death and tissue collection. Degradation depends on the type of tissue and how much time has passed since blood flow to the tissue ceased, the researchers found.
“For some tissues, the RNA will stay good for many hours after blood flow to the organ stops,” said Dr. Wendy Winckler, co-leader of the Broad GTEx team, which conducts molecular studies, including RNA sequencing. “But in the pancreas and the spleen, for example, if you don’t get the tissue within a few hours [after death] you’re not going to get [useful RNA].”
Despite these challenges, the researchers are optimistic. “Getting biospecimens for any research project is a tremendous challenge,” said Dr. Kristin Ardlie, also a leader of the GTEx work at the Broad. “We’ve been surprised by how well the process has worked in this project.”
The Broad researchers deposit their data quarterly in the database of Genotypes and Phenotypes (dbGP). Investigators can apply for access to the data, which have been stripped of identifying information to protect the anonymity of the donors. In addition, the project plans to make extra samples available to researchers in 2013.
Many Questions to Explore
GTEx data will likely be used to interpret genome-wide association studies. In recent years, these studies have identified inherited genetic variants associated with common diseases. The variants, however, often map to regions of the genome that lack genes, raising questions among researchers about how the variants influence disease risk.
A logical explanation could be that these variants influence the regulation of genes and thereby increase (or decrease) the risk of disease.
“Many common variants associated with common human diseases may be more about affecting the regulation of gene expression than about changing protein structure,” said Dr. Nancy Cox of the University of Chicago, who is developing statistical tools for analyzing data generated by GTEx. “This is one reason we believe the GTEx project is so important.”
With samples of so many tissues, GTEx could also help researchers look at changes in gene expression throughout the body. “A disease process might not be limited to one particular cell type or tissue type,” said Dr. Stranger.
“People will be using data from this project for a long time and in different ways,” she added. “There are all kinds of people waiting to see the GTEx results.”