Every last one of the gazillion cells in a person's body owes itsancestry to a single cell, generated by the union of a sperm and anegg. Nine months later, it gives rise to a person's multicellular body.
Along the way, as that first founder zygote divides and subdivides adinfinitum, it spawns a series of primitive stem cells _ founders in turnof all the emerging body's organs and tissues.
"We've known for 120 years," said molecular geneticist John Dick atthe University of Toronto, "that stem cells must exist. The problem ishow to identify them in a human." Ten years ago, Dick toldBioWorld Today, he and his associates "showed that one could put agene into a stem cell, mark it, and follow its repopulation" ofprogenitor cells and their ultimate progeny, in the mouse.
Dick is senior author of an article in this month's Nature Medicinetitled: "Identification of primitive human hematopoietic cells capableof repopulating NOD/SCID mouse bone marrow: Implications forgene therapy." This multi-center project, which he initiated, is co-sponsored by Vancouver-based Canadian Genetic Diseases Network
Since it became feasible to zero in on isolated stem cells in thebloodstream, gene therapists have been trying, with mixed success, toinsert therapeutic DNA into them. Stem cells make a tempting targetbecause, by definition, they are pluripotent _ capable of expressingany old gene product, but preprogrammed to none.
"Ultimately," Dick explained, "what a human hematopoietic stem celldoes is, it regrows the body's entire blood system _ all the myeloid[bone marrow] cells, all the lymphoid [immune system cells], all theerythroid [red-call precursors].
"Ideally," he went on, "if I wanted to study a human stem cell, I'dwant to watch it regrow the blood system in a human. But obviously,"he said, "no one wants to volunteer for my experiments."
NOD: More Immuno-Helpless Than SCID
So he and his co-authors use mice instead, a special breed of rodentthat's even less immunocompetent than the well-known SCID mouse,named for its `severe combined immunodeficiency.'
"A SCID mouse," Dick said, "though it doesn't have T cells and Bcells, still has other resistance mechanisms that are fully functional."His animals of choice are a cross between SCID and NOD _ "non-obese diabetic" mice.
When Dick's team first tried inserting human cells into SCID mice,"we had to give them very high cell doses, in order to dampen downthat resistance. Whereas, we can give very low doses to NOD miceand still have repopulation. It allowed us to transplant as few as 50 or100 cells, with no resistance to those cells going in.
"The Canadian approach," he observed, "is simply to inject a mouseintravenously with human stem cell preparations, and ask those cellsto function in the blood-producing environment of that mouse."
Two years ago, the group identified the abnormal stem cell thatcauses leukemia, and then set out to pin down its ancestor _ motherEve of all blood-forming elements in the body _ the humanhematopoietic stem cell.
To begin with, they subfractionated human bone marrow or umbilicalcord blood and tested each of their components in mice. "In thatway," Dick said, "we were able to narrow down the population ofcells with the unique ability to regrow an entire human blood systemin the mouse."
To visualize this strategy, Dick compared the blood-forming systemto a pyramid, with billions of blood cells at its base. Many of theseneed to be frequently replenished.
Halfway up are the committed _ preprogrammed _ progenitor cells.When stimulated with a cytokine hormone, such as erythropoietin(EPO) or granulocyte-macrophage colony-stimulating factor (GM-CSF), they will divide and differentiate, producing that turnover ofcells at the base. But they too don't last forever.
"Ultimately," Dick observed, "the whole system is driven by the stemcell at the apex of the pyramid, which has the unique property ofrenewing itself."
CD34 decorates those halfway progenitors, and probably theimmortal stem cell at the apex as well. "Statistically," he went on,"we can determine that the frequency of this cell is one in a million."
Mouse Bails Out Human Gene Therapy
Eventually, 25 frustrating experiments and a year and a half later, theToronto team came to realize that "while there are progenitor cellsthat we could infect, there must be another stem cell capable ofrepopulating the mice, but of low or zero infectability. As an alternateapproach to try to determine which cell was able to repopulate inmice," Dick added, "we attempted to transfer genes into the variousprogenitor cell types."
Meanwhile, human gene therapy trials all over the map, usingretroviral vectors as vehicles for delivering DNA to stem cells, werealso experiencing some sort of a block.
"The point is," Dick said, "the high gene transfer efficiency intoprogenitor cells or into mouse stem cells did not predict the poorresults coming out of the human trials, whereas our mouse modeldid."
Taking leukemia as an example of potential therapeutic utility, Dicksuggested: "If you treat somebody with high-dose chemotherapy orradiation, you'll kill tumor cells, but also normal cells. By removingsome of the cells prior to that high-dose conditioning, and givingthem back afterwards ex vivo, you could regrow the blood system,and so help the patient to survive.
"But," he pointed out, "if you're also reinfusing tumor cells, youaren't going to be doing the patient much good. So if you couldseparate out the stem cells alone, and transplant only them back, thatmay well be important in terms of progress from lab bench tobedside."
Picturing the assay's clinical benefit, Dick suggested: "I could comeinto the lab one day and ask: `Will this drug do something onleukemia? sickle-cell anemia? thalassemia?' Well, I can test its effecton human cells in the NOD/SCID mouse model, and either discardthat candidate compound or allow it to go forward _ in the space oftwo months or so.
"Obviously," he concluded, "I can't do that currently by trying thedrug in a patient. It would cost hundreds of millions of dollars and 12years to get it to the approval process." n
-- David N. Leff Science Editor
(c) 1997 American Health Consultants. All rights reserved.