July 8, 2002
MEDIA CONTACT: Joanna Downer
Hopkins Scientists Return to the Mouse to Overcome Some Obstacles In Working with Human Stem Cells
Editors' Note: See related release Normal Gene Control Increases Chances Human Stem Cells Will Be Safe
Learning about human stem cells requires working with them, but some Johns Hopkins researchers are turning to a clever new mouse model to learn things the human cells can't teach them.
The scientists were concerned with a control mechanism called "imprinting," in which the copy of a gene used by the cell to make proteins depends only on which parent passed it on. Some imprinted genes use the copy inherited from the father, while others use that from the mother.
"The inheritance rules Mendel observed in his pea garden aren't operating here," explains Andrew Feinberg, M.D., King Fahd Professor of Medicine in the McKusick-Nathans Institute of Genetic Medicine at Hopkins. "'Dominant' and 'recessive' don't explain imprinted genes."
Published online the week of July 8 in the Proceedings of the National Academy of Sciences, the Hopkins scientists' report shows that imprinting is normal in human embryonic germ cells -- a type of stem cell that can become virtually any type of cell in the body -- but the experiments left the researchers wanting more.
"You can't draw conclusions about imprinting in human embryonic germ cells without examining them directly, so we did," says Feinberg. "But there are many remaining questions about how imprinting is controlled in these primitive cells and as they form more specialized cells. A lot of those mysteries can't be solved using human cells because of legal, ethical and scientific reasons."
A mouse isn't a man when it comes to genes, but frequently the basic biology is the same or similar, making the mouse model a great starting place for understanding the details of how imprinting affects cell differentiation, he adds.
For what they want to learn, the human cells present a number of problems: they are hard to keep pure because they rapidly and spontaneously form specialized cells. Crucial information about the parents' genetics is missing. The cells can only be examined in laboratory dishes, not a living human. For human embryonic germ cells, which were first isolated and characterized at Hopkins, the parents gave informed consent to provide the fetal tissue, but no genetic samples from the parents were taken.
Their solution to the obstacles was to mate males of one kind of mouse to females of another kind. From the offspring, the scientists established four new mouse embryonic germ cell lines. Because the parental genes in the lines are distinct and easily identifiable, the scientists can more easily investigate imprinting and how cells recognize or ignore imprinting marks, which identify the parent that contributed a particular gene copy.
"We can manipulate lots of things with mouse cells we can't with human embryonic germ cells," says first author Patrick Onyango, a postdoctoral fellow in the McKusick-Nathans Institute. "We can get dishes of pure, undifferentiated mouse embryonic germ cells more easily and we'll have clear information about the parent genes. We can let the cells differentiate into various tissues and see what factors control imprinting during the process. It will be very powerful."
In addition to thoroughly characterizing the new mouse cell lines, the scientists tested their imprinting status, which is very close to normal. For one paternally imprinted gene (i.e. only the copy from the father is usually turned on), they discovered that the mother's copy was on instead, a finding they have not explained yet.
One question the scientists hope to answer using the mouse cells stems from their initial experiments: how do both mouse and human embryonic germ cells appear to have two active copies of the imprinted genes, but when the cells differentiate, correct imprinting is restored?
"It makes sense this would happen, because imprinting sometimes varies by tissue type, but we don't know how it works," says Feinberg. "We and others believe that even when imprinting appears to have been lost, the parental 'marks' aren't really gone, they just aren't active or aren't recognized by the cell. Stem cells must be able to reverse this, or the specialized cells they become couldn't have imprinted genes.
"Imprinting is just one way cells use 'epigenetic' processes to turn genes on or off, one way they modify their genes without changing the sequence of DNA," continues Feinberg, who co-organized the conference "Epigenetic Mechanisms in Disease," held May 30 and 31 at the National Institutes of Health. "Geneticists talk a lot about sequence and mutations, and those things are important, but they are just a small part of how DNA is manipulated in cells."
Other authors on the paper are Shan Jiang, Hiroshi Uejima, Michael Shamblott, John D. Gearhart and Hengmi Cui, all of The Johns Hopkins University School of Medicine.
Funding for the work was provided by the National Institutes of Health. Funding for the derivation of the human embryonic germ cells mentioned in this release was provided by Geron Inc. (Menlo Park, Calif.). Under a licensing agreement between Geron and The Johns Hopkins University, the University, Shamblott and Gearhart are entitled to a share of sales royalty from Geron. The University, Shamblott and Gearhart own stock in Geron, the sale of which is subject to certain restrictions under University policy. The terms of this arrangement are being managed by the University in accordance with its conflict of interest policies.
On the Web: