Editor's Note: This is part two of a two-part series on stem cell research. Part one ran in the Jan. 31 issue.

The political restrictions on human embryonic stem cell research do not make their already-difficult scientific investigation any easier.

But as those political restrictions spawn arguments and grab headlines, it can be easy to forget that human embryonic stem cells still are a clinically unproven medical approach, and that there remain formidable scientific obstacles to controlling their differentiation. Like children, stem cells seem capable of developing into just about anything; but those trying to get them to develop with specificity soon tend to find that they have their own, often mysterious, agenda.

Elias Zerhouni, director of the National Institutes of Health, which controls federal stem cell research funding, stressed that fact last week when he published an editorial in the Jan. 25, 2005, issue of USA Today, in which he commented on a research article published in the February 2005 issue of Nature Medicine. The Nature Medicine paper showed that the use of mouse feeder layers, as well as serum replacements containing animal products, has led to immunogenic contamination of human embryonic stem cell lines with the sialic acid Neu5Gc. (See BioWorld Today, Jan. 31, 2005.)

Zerhouni pointed out that "we knew that there was the possibility that something might be transferred from either animal serum or a feeder layer, and that this issue would need to be addressed at the time of clinical implementation." He said the authors "should be commended for identifying a potential issue for future clinical applications, which may be several years away."

But for today, Zerhouni said, "the scientific challenges are more fundamental. We need to understand how to make embryonic stem cells become a particular specialized cell," adding that while great progress has been made, "there are still many challenges to be met. The most basic scientific questions remain to be answered."

In research published in the February 2005 issue of Nature Biotechnology, scientists from the University of Wisconsin at Madison report success in doing just that. They have elucidated the conditions necessary to coax one particular specialized cell from human embryonic stem cells: motor neurons. Though their tenacity ultimately was crowned by success, the unexpected twists along the experimental way also bear out Zerhouni's assertion that the basic work has not yet been done.

Mental processes are more complex in some than others, but brain development is complex in everyone, since it has hundreds of different cell types. In the research reported in Nature Biotechnology, the scientists concentrated on spinal motor neurons, which so far had eluded researchers. "The process has three steps," said senior author Su-Chun Zhang, assistant professor of anatomy and neurology in the stem cell research program at the Waisman Center at UW-Madison. "The first step is to tell the na ve cell to become a neural cell. The second step is to tell the neural progenitor cell to become a ventral spinal cord cell. And the final step is to get them to become a mature spinal motor neuron."

In the journey from embryonic stem cell to spinal motor neuron, cells express a complex pattern of multiple transcription factors over time. Three of those transcription factors can be used to mark different stages along the developmental path: early neural stem cells express Pax6, late neural stem cells express Sox1 as well as Pax6, and motor neurons express HB9. Two growth factors are known to be critical to inducing the formation of motor neurons: retinoic acid and sonic hedgehog.

For Human Cells, Get Em While They're Young

Zhang and his colleagues initially tried to manipulate neural stem cells at a relatively late stage in their development, based on previous animal work. But it turned out that "we cannot simply translate studies from animal to humans," Zhang said. "The earliest neural cells behaved quite differently than in other animal species. [With human embryonic stem cells], if you tell the cell too late, you will not get anywhere. You have to tell them at exactly the right time [and] that is the major finding of this paper."

When the scientists tried to coax late neural stem cells to develop into spinal motor neurons through retinoic acid and sonic hedgehog, the cells developed some markers typical of motor neurons but never expressed the critical HB9. That led the scientists to conclude that by the time cells express Sox1, the developmental window in which they turn into spinal motor neurons has passed. They next treated early neural stem cells with the retinoic acid/sonic hedgehog cocktail, as well as several other growth factors. That led to success: after a few weeks in culture, up to 20 percent of cells expressed HB9. The cells ultimately began expressing enzymes that motor neurons use to make their neurotransmitter. Electrophysiologial experiments showed that they were capable of forming functional synapses with muscle in culture, causing that muscle to contract.

Zhang's group plans to conduct in vivo experiments in baby chickens next to see whether the cells are functional in vivo. The electrophysiological experiments demonstrate that "in a Petri dish, these are bona fide motor neurons," he said. "But a Petri dish is just a Petri dish." He also noted that while the cells might be used fairly soon as model systems in drug discovery efforts, transplantation for motor neuron diseases such as spinal cord injury or amyotrophic lateral sclerosis remains "in the relatively distant future. There are many other factors still to be figured out about motor neuron diseases."

Immunogenic Contamination: An Expected' Concern

Asked whether he is concerned about the recent results showing contamination of human embryonic stem cell lines with sialic acids, Zhang said "That's a concern, but it's what we all expected. You would expect that because of the way these cells were derived. That's why we were proposing to the NIH to derive new cell lines not cultured on animal cells."

Zhang noted that the current NIH-approved cell lines are fine for basic research purposes, but that if and when they are transplanted into patients, they would need to be cleaned.

"It's possible to do that," he said, "but it probably would be wise to generate new cell lines."