March 1, 2011 (Vol. 31, No. 5)
With the ongoing interest in the latest iteration of next-gen sequencing, we thought it would be a good time to take a look back at some early developments in the automated diagnostic DNA seqeuncing field. Thus, this issue’s article commemorating GEN’s 30th anniversary is reprinted from July 1993. The story reports the opening of the world’s first clinical diagnostic laboratory for automated DNA sequence analysis.
The new lab was largely the result of a collaboration between the Eye Research Institute of Canada (ERIC) and Pharmacia Biotech. Now that’s a noble company name in bioindustry lore! Long one of the most innovative life science instrument and fine chemical firms, Pharmacia Biotech later became Amersham Biosciences until it was acquired by GE Healthcare in 2004.
The essence of the article revolves around the fact that to diagnose retinoblastoma in children, ERIC relied on the Pharmacia Biotech A.L.F. Automated DNA Sequencer. This marked the first time that a sequencer, primarily developed as a research tool, was to be used in a clinical setting.
This story is the latest example of our decision to reprint an article from one of GEN’s early issues in each issue in 2011. All these stories demonstrate a particular significance and relevance for the life science research community.
—John Sterling, Editor in Chief
“As Seen in GEN—Flashback” Volume 13, Number 13, July 1993
First Automated Diagnostic DNA Sequence Lab Opens Doors
By Stephanie Yanchinski
The Eye Research Institute of Canada (ERIC) has opened its doors in Toronto as the world’s first clinical diagnostic laboratory for automated DNA analysis. At the opening, ERIC, in partnership with the Swedish firm Pharmacia Biotech and the Canadian Genetic Diseases Network, unveiled a system for routinely diagnosing retinoblastoma (RB), a relatively rare inherited tumor that can result in total blindness or death in young children.
The Diagnostic Laboratory of the Twenty-First Century (DL21C) will use the Uppsala-based company’s Sequence-Based Diagnosis (SBD™) for analyzing human DNA base by base. SBD integrates a package of technologies for the entire process, from extracting the DNA from blood or biopsy to presenting results in a computer printout.
However, according to John K. Stevens, Ph.D., ERIC’s director, “This is just the beginning. It provides a foundation for diagnosing many other diseases in a cost-effective way.” He believes that eventually SBD may have the capability of detecting up to 4,000 genetically linked conditions long before symptoms appear. “This makes early intervention—and cures—much more likely,” he says, with substantial cost savings to the healthcare system.
The Eye Research Institute of Canada was conceived in 1984 by a group of ophthalmologists to pursue innovative research programs to help cure eye disease and prevent blindness. The Institute’s facility was created with a C$4.5 million grant from the Ontario government. Corporate sponsors such as the Royal Bank of Canada and the Canadian Imperial Bank of Commerce and a number of foundations contributed $3.5 million, paving the way for the establishment of the DL21C.
The project to develop automated sequencing—the key to DL21C—began three years ago when Dr. Stevens and Brenda Gallie, M.D., director of molecular genetics at the Institute and senior scientist of the department of ophthalmology, immunology and cancer at the Hospital for Sick Children in Toronto, became convinced that automation was the best route for diagnosing inherited diseases like RB.
They turned to Pharmacia Biotech for help and the company worked closely with the Institute’s team of ten clinical researchers. Pharmacia supplied sequencers, reagents and the support of their systems experts to help make SBD a reality.
Scientists now know that many genes may be responsible for an inherited disease. For example, when the cystic fibrosis gene was first discovered it was thought to be associated with a single mutation. Today, more than 300 different mutations have been documented. In the case of retinoblastoma, each RB family has a new and unique genetic error.
Other genes, related to breast or colon cancer, for instance, may have many more mutations. Testing by conventional methods might require many hundreds of specific tests. In these cases, sequencing the suspected disease gene, base by base, makes more sense than developing a specific test for each possible mutation.
The ability to read the genetic code of a new unidentified gene has been possible for years. But conventional methods relying on the x-ray reading of gel banding are labor-intensive and time-consuming. An enormous number of RB patients, for instance, are awaiting testing. Automated DNA sequencers for reading the genetic code have been available since the mid-1980s, but were developed largely as research tools and have not received wide acceptance, even in the research community. Few scientists believed sequencing could be adapted for clinical use, and fewer still were willing to invest the time and resources to try. But retinoblastoma provided a powerful incentive to clinical researchers such as Drs. Stevens and Gallie.
Serves as Model
Retinoblastoma requires early detection if the child’s sight is to be saved. But not every child in an RB family will develop tumors—it depends on whether he or she inherits certain mutations. Currently, each child in a family suspected to have a mutation in the RB gene must be examined every three months to see if a tumor has developed.
Under the age of three, the child must be anesthetized, and each exam can cost as much as $1,000, or about $50,000 for an average family over the first three years of their children’s lifetimes. Clearly, a better alternative is to detect which children have inherited the disease-causing mutation and monitor only those at risk. But developing a cost-effective way of detecting the RB gene suitable for routine hospital laboratory practice posed major challenges.
The RB gene consists of 27 stretches of DNA scattered among a total of 200,000 base pairs. The current procedure for detecting RB, the single stranded conformation polymorphism technique, is laborious and time consuming—one test involving one family can take from 6 to 12 months. For many hospitals, this is unaffordable on a routine basis. This method also “seemed to us to require a high level of interpretation,” says Dr. Gallie. “We simply couldn’t see how this technology could ever become a service tool.”
The integrated system developed by Pharmacia Biotech and ERIC collaborators consists of five elements: DNA extraction and purification; gene amplification; a Decipher™ kit for processing the sequence reactions; a SBD A.L.F. automated DNA analyzing unit for reading the sequence; and SBD presentation of the results. SBD incorporates major technical advances, says Pharmacia’s Steven Chackowicz, who worked with the ERIC team in developing the technology. Key steps have been automated, and Pharmacia has provided software prompting the clinical technologist on what follows next.
The development of special kits has all but eliminated laborious pipetting and the chance of human error from many stages in the process. For instance, the Decipher kits are based on a patented Pharmacia process of vitrification which involves “glassifying” the reaction enzymes and other heat labile products. This ensures their stability at room temperatures and eliminates the need to store them in the freezer.
Pharmacia links these reaction enzymes to plastic microtiter plates as well as specially designed “combs” bearing primers. The technologist then adds water to the wells and dips the primer-bearing comb into the wells where the reaction then takes place. To read the sequence, the technologist places the comb in the A.L.F. automated sequencer.
Altogether, it takes about one day for one RB patient (24 samples) to be processed, a far cry from the months it used to take. Where it now takes $50,000 to assess one family for three years, the cost could drop to as little as $1,000, according to Dr. Gallie.
But SBD solves another major problem with automated sequencing: the high background noise obscuring the signal, leading to inaccuracies.
Scientists have tended to distrust such results, and Chackowicz believes this is a major reason why the technology has not flourished within the research community. In other processes, the labeled samples contain all four fluorescent tags—one each for adenine, cytosine, thymine and guanine. “These four labels have overlapping emissions,” says Chackowicz, creating the bothersome background noise.
The SBD system, instead, divides the sample into four lanes, one for each type of base. This advance, along with a fixed laser beam reading the results, provides better signal and less noise.
Not all parts of the system are ready for commercial use, says Chackowicz, but he anticipates a complete package ready for reading the RB gene in about 12 months. The company has begun submissions to the FDA and other regulatory agencies for approval.
Dr. Gallie sees as future goals developing SBD for more common diseases, such as genetically-linked forms of breast cancer and colon tumors. Other applications may include congenital adrenal hyperplasia, certain forms of heart disease, and tissue typing. She emphasizes, however, that SBD is not designed for screening healthy populations. The gene must be well characterized first, she says, and patients should be at some risk of developing the disease or show early symptoms.
Per Lindstrom, Ph.D., director of molecular diagnostic market development at Pharmacia, says, “We start with the clinicians—they want better diagnostic procedures, and better treatments to put in place.” SBD is linked to treatment, no population genetics.
“There has to be an outcome for the patient, either in lifestyle changes or in better therapies. It’s the only way to get the technology accepted by the clinician,” adds Maggie Bywater, M.D., Ph.D., deputy director, molecular diagnostic market development at Pharmacia.