Closer to making blood donation redundant

Illustration showing blood stem and progenitor cells emerging from hemogenic endothelial cells. The left blue cells are emerging hematapoietic stem and progenitor cells. The red cells are red blood cells.
Illustration showing blood stem and progenitor cells emerging from hemogenic endothelial cells. The left blue cells are emerging hematapoietic stem and progenitor cells. The red cells are red blood cells.
Credit: O’Reilly Science Art

When human embryonic stem cells lit up the world’s headlines in 1998, it seemed the era of spare organs and tissues would soon follow.

One of the hopes was that these “blank slate” cells would fill the gap in the fraught supplies of blood banks by generating an endless supply of the different blood-cell types.

Almost two decades on this dream is approaching reality, as two papers in Nature suggest.

One team, led by George Daley of Boston Children’s Hospital’s Stem Cell Program, have trained human pluripotent stem cells (either derived from embryos or induced from skin cells) to do a fair job of replenishing a mouse’s blood supply.

The other team, led by Shahin Rafii of Cornell University in New York, began with endothelial cells scraped from the lining of mouse blood vessels. After training, these cells did an even better job of replenishing a mouse’s blood supply.

Both teams relied on genes introduced by viruses to train the cells. While the introduction of these foreign genes raises concerns as to the long-term safety of such cells, the scientists nevertheless achieved what has till now been impossible: repopulating the blood supply of a mouse.

“It’s great; we’re part of the way there,” says Andrew Elefanty, who has been attempting the same feat without adding foreign genes at the Murdoch Childrens Research Institute in Melbourne.

Elefanty, like his colleagues, found the key to educating stem cells to become blood-cell precursors was to be a keen and patient student of nature.

Whether mice or humans, developing embryos proceed through a precisely choreographed set of stages, cued by specific signals, to form different tissues. In the case of blood cells, the first cue comes at a primitive stage when a tissue called the mesoderm appears. Under further cues, mesoderm cells give rise to what is known as “haemogenic endothelium”. After several weeks and more cues, these cells give rise to haematopoietic stem cells (HSC) that ultimately reside in the bone marrow. Just one of these cells is powerful enough to replenish the entire blood supply of a mouse.

Repeating this performance in a culture dish starting from stem cells, however, has been difficult.  Over the past two decades, researchers have often gone down the wrong path – producing yolk blood cells, for instance, that have little ability to regenerate a whole blood supply.

Just last year, Elefanty and colleagues were able to reproduce the first part of the program, training embryonic stem cells to go from the mesoderm stage to haemogenic endothelium using growth factors added to the medium bathing the cells. So far his group has not been able to take the cells the final step of the way.

Daley’s group followed a similar formula but took the cells the final step by pushing them with seven genes introduced by viruses.

Rafii’s group began with endothelial cells derived from the lining of mouse blood vessels, made them stem-cell-like by introducing four genes, and then completed their education them by growing them on a layer of cells derived from umbilical cord.

Both produced cells that, like HSC, were able to repopulate the blood supply of a mouse.

Elefanty notes the blood stem cells produced by the two groups are “not perfect” at matching the capabilities of what real HSC can do. Given both methods insert foreign genes, he also notes there are safety issues to deal with: “These are just the first papers; there are still lots of questions to answer.”

The goal, as Daley et al acknowledge, remains “the derivation of bona fide transgene-free HSCs for applications in research and therapy”.

Please login to favourite this article.