"Gene therapy is a promising field that offers fundamentally new ways of curing human illness," said NHGRI Director Francis S. Collins, M.D., Ph.D. "A decade of clinical studies has demonstrated the complexity of the biology behind gene therapy and the technical problems researchers have experienced. This study provides insight into one of the most serious, current technical hurdles."
The discovery may lead to safer gene therapy techniques.
"Provided this new insight, researchers can now aim to improve the design of gene therapy techniques that can insert genes in less risky areas of the genome," said NHGRI Scientific Director Eric D. Green, M.D., Ph.D., who is director of the Division of Intramural Research and chief of the Genome Technology Branch where the work was performed. "The viruses used therapeutically can be altered in various ways and now tested in the lab to see if they insert curative genes more safely."
In January 2003, the U.S. Food and Drug Administration (FDA) placed a "clinical hold" on 27 gene therapy studies after two children being treated by French researchers for a form of severe combined immunodeficiency disease developed a leukemia-like condition.
Most current gene therapy experiments attempt to cure genetic illnesses by inserting normal, functional copies of a gene into target cells of the body. Researchers have learned to genetically engineer different types of viruses so they can infect target cells and integrate their genes into the chromosomes of those cells. Among the most widely used in gene therapy studies is the Moloney murine Leukemia Virus (MoMuLV), a mouse retrovirus that also can infect human cells. Physicians now believe that the children in the French study developed leukemia because the MoMuLV inserted therapeutic genes next to a gene known to promote blood cancer.
Previously, researchers believed that MoMuLV randomly integrated into the genome of target cells. But scientists lacked the means to study these integration events in a large-scale fashion. Now, the team of Shawn Burgess, Ph.D., and Xiaolin Wu, Ph.D., from the NHGRI Genome Technology Branch, and Yuan Li, and Bruce Crise, Ph.D., from the AIDS Vaccine Program of the National Cancer Institute, Frederick, Md., has developed a laboratory technique that allows them to rapidly sort through the entire genome of hundreds of individual cells to see where the retroviruses have inserted. The results were surprising.
"MoMuLV does not insert its genes randomly, as had been previously thought," said Dr. Burgess, head of the Developmental Genomics Section of the Genome Technology Branch and senior author on the paper. "It likes to land right at the beginning of a gene, potentially affecting the way the gene works. The virus is eight times more likely than random to land at the beginning of a gene. We don't know why it has this specificity."
In addition to targeting the beginning of genes, Dr. Burgess said, "MoMuLV seems to have a clear preference for more actively expressed genes." That may explain why the two French gene therapy patients showed signs that the vector had integrated next to the LMO2 gene, a gene that is turned on in bone marrow cells to promote the growth of white blood cells. By altering LMO2's function, the retrovirus may have caused the patients' leukemia. "Based on the number of cells that physicians are infecting during a gene therapy treatment, estimated at 5 to 6 million cells, we calculated that every time they do a treatment, something like 70 bone marrow cells will have the new genes integrated close to the LMO2 gene," Burgess said. "That seems to be a dangerous spot."
The team also studied the insertion preferences for the Human Immunodeficiency Virus (HIV) the virus that causes AIDS. "About half the time, HIV likes to land in the middle of a gene," said Xiaolin Wu, Ph.D., a postdoctoral fellow in Dr. Burgess' laboratory and lead author of the study. "This is a very different pattern than the mouse virus. This is the first time that the integration patterns of two different retroviruses have been compared, showing definitively that their insertion in the genome is not random."
"We wanted to figure out a fast and easy way to find out where viruses land in the human genome," Dr. Burgess said,. "Not just one or two gene integrations, but thousands of them."
The NHGRI researchers developed a simple laboratory trick for easily identifying the insertion site by capturing and then sequencing a small bit of the human genome immediately adjacent to where the retrovirus inserted.
"Now that a high-quality, finished copy of the human genome is available in public databases, it only takes the sequence of 30 base pairs (the chemical subunits that make up the human genome) to know exactly where you are in the genome," Dr. Burgess said. The scientists literally went to the computerized databases of the human genome sequence and looked it up in each case.
"Without the success of the Human Genome Project, knowing precisely where the retroviruses inserted would have been nearly impossible," Dr. Collins said. "These are the kinds of laboratory applications for which the finished genome sequence was intended, applications that will end up improving the practice of medicine."
NHGRI is one of the 27 institutes and centers at the National Institutes of Health, an agency of the Department of Health and Human Services. The NHGRI Division of Intramural Research develops and implements technology to understand, diagnose and treat genomic and genetic diseases. Additional information about NHGRI can be found at its Web site: www.genome.gov.