By David N. Leff

Before a contractor starts to build a brick house, he calculates exactly how many bricks of various sizes the job will take. His total might work out at precisely 21,016 bricks. But he orders 23,000, to allow for miscalculation, loss and breakage.

That's the simplest model of programmed cell death in embryonic brain development, explained neurobiologist Pasko Rakic, at Yale University Medical School, in New Haven, Conn.

"The developing nervous system," he told BioWorld Today, "doesn't hook up all of its myriad neuronal connections as it grows along, but makes a much broader structure, which it then trims down to size by apoptosis."

More than one smoking gun can trigger this cellular suicide, Rakic pointed out. "For example, in cell-to-cell interaction, some connections may not be made properly. Or cells fail to get trophic [nerve-supply] influences from other cells, so they don't have a signal to live. Instead, that activates a signal to die."

Many genes in the mammalian genome transmit those apoptotic signals. Among the proteins they express, one family of proteases, called caspases, plays a heavy hand in enabling cells to declare themselves dispensable. So far, scientists have identified 11 human caspase enzymes, some with tissue-specialized functions. Of their number, the ninth caspase (C9) is "a sensor of the cell-death process, which it initiates," said molecular biologist and immunologist Richard Flavell, a Howard Hughes Medical Institute investigator at Yale.

"When a cell's mitochondrion is damaged, it involves caspase 9 in the release of cytochrome c, which triggers the apoptosis cascade. Cytochrome c," Flavell explained, "is a key component of the cell's energy transduction chain. And once it gets going, it activates caspase 3 (C3), which then chops up a number of other substrates, and the cell dies very very rapidly.

"So C9," he observed, "is at the control point of programmed cell death, and C3 is what we would call the effector point."

Flavell and Rakic are co-senior authors of a paper in the current issue of Cell, dated Aug. 7, 1998. Its title is "Reduced apoptosis and cytochrome c-mediated caspase activation in mice lacking caspase 9."

The article's first author is molecular biologist Keisuke Kuida, a former postdoctoral fellow in Flavell's lab, now a staff investigator at Vertex Pharmaceuticals Inc., of Cambridge, Mass. Vertex and the Yale scientists have a long-standing, hands-on research collaboration in the mechanism and application of apoptosis. Kuida characterizes their report in Cell as "the first to describe the in vivo function of caspase 9, and to suggest apparent tissue-selective activity among the 11 reported human caspases."

C9, Kuida told BioWorld Today, "is essential for apoptosis in neuronal cells, as is C3 to a lesser extent."

Caspase-Minus Mice Can't Trim Brain Cells

Early this year, the co-authors raised a colony of mice completely devoid of genes for C9, and studied the effect on their embryonic brain development.

Kuida compared this effect to hydrocephalus, in which intracranial fluid can inflate the cranium and cause atrophy of the brain. "In mice without C9, we usually saw the condition called exencephaly in which the brain cells get out of the brain structure and create an abnormal structure on the top of the head. That's what happened in our C9 knockout mice."

"Normal mice," Rakic pointed out, "have a cortex that is smooth, whereas humans and monkeys, with far bigger brains, have convolutions — indentations — in their cortex. Our C9 knockout mouse," he continued, "because there are more and more neurons growing in the absence of apoptosis, start to develop convolutions. Their cortex, we saw, carries indentations and gyri and sulci formations.

"That's very exciting," Rakic said, "because it shows to some extent how in evolution a larger cortex can be formed by just manipulating genes, which determine cell proliferation or morphogenetic cell death."

He pictured such a C9 knockout rodent telling itself: "I'm going to be a mouse; I'd better stop producing more cells for a convoluted cortex."

"Now that we have new molecular methods for looking at caspase genes in a knockout mouse," Rakic observed, "maybe we will go to the next stage and look for the molecules or genes that are more closely involved in normal development."

"However," he went on, "we already know that there is a genetical trigger of cell death in neurodegenerative diseases, such as Alzheimer's disease (AD), Parkinson's disease (PD) and Huntington's disease (HD), as well as amyotrophic lateral sclerosis (ALS). In each of these, there could be a different type of apoptosis gene involved.

"In many such diseases," Rakic observed, "brain neurons have to start a molecular chain of events that lead to cell death. When we know that caspases are involved in that, if we can develop a drug that would prevent or manipulate that apoptosis, maybe we could inhibit that process somehow. Either the cells would not die, or die slower.

"Take ALS, for the sake of argument," Rakic suggested. If you develop and test in animals a small molecule that inhibits a caspase, when you give it to the patients, it either slows down the disease or totally stops it.

"In PD," Rakic went on, "maybe you could engineer cells in which you knock out the caspase. Insert those neurons in the patient; they would stay indefinitely, not die. And therefore the transplant would be more successful than the current fetal tissue implants for treating PD.

"This is one of the projects we are now doing," he pointed out, "trying to find whether cells from this C9 knockout mouse would survive better if we transplant them to the artificially induced, Parkinsonian rat."

Potential Stroke Treatment Under Study

Flavell foresees a more immediate form of therapy for a more acute disease — stroke.

"During an acute cerebral ischemia situation," he told BioWorld Today, "you know that the patient has suffered an event, and what's going to happen is that neurons are going to die.

"As for adding the inhibitor of such an enzyme as caspase 9," he continued, "it hasn't yet been proved that the C9 pathway is used in stroke. That's an essential next step, which needs to be done. We're already doing quite a lot of work in that general area."

If C9 is validated for stroke, Flavell suggested, "conceptually, one would add the inhibitors during the period in which the event has occurred, and then hope to reduce the damage to the neurons. Fewer of the cells would die.

"In neurodegenerative diseases," he went on, "one could administer these materials on an ongoing basis. But the more chronic the administration," he cautioned, "the more difficult it is. Therefore the therapy would be more problematic with a long-term entity like Alzheimer's disease. But," he concluded, "the concepts are the same — and an interesting avenue for therapy." *