By David N. Leff

Science Editor

A post-mortem of a person who died with senile dementia is the only way to pin down -- in retrospect -- a certifiable diagnosis of Alzheimer's disease (AD).

Only an autopsy of the brain can reveal the exclusive hallmarks of AD -- neuritic amyloid plaques and fibrillary tangles. But to reveal the root cause of the disease would take discovery of a mutant gene.

Such a candidate gene has surfaced in today's Proceedings of the National Academy of Sciences (PNAS), dated April 29, 1997. Its title: "Mutations in mitochondrial cytochrome c oxidase genes segregate with late-onset Alzheimer's disease." That paper's lead author is neuroscientist Robert Davis, president and chief scientific officer of MitoKor Inc., in San Diego.

"I think the major contribution of our work reported in PNAS," Davis told BioWorld Today, "is that a large proportion of late-onset AD has a genetic etiology. That this leads to a defect in energy metabolism in the brain, and that it provides us with new targets for diagnosis and drug therapy."

That "genetic cause" assigns the inheritance of AD to the maternal line of descent, because the mitochondrial genome is transmitted exclusively via the mother's chromosomes. Nuclear genomes come from both parents.

Lesions On Mitochondrial DNA Probed

The mitochondrial organelles, which supply a cell's energy needs, float free in the cytoplasm. Their genome is a circular DNA molecule of 16,569 base pairs. "The 13 polypeptides it encodes," Davis said, "express critical subunits of ETC -- the electron transport chain -- which is the main cellular energy-generating pathway.

"Lesions in that mtDNA," he pointed out, "could account for the mitochondrial and bioenergetic pathogenesis of AD."

A key molecular player in the mitochondrial energy game is cytochrome c oxidase (CO), a terminal enzyme of the electron transport complex. "It's involved," said neuroscientist and clinical neurologist M. Flint Beal, "in generating a proton gradient in the mitochondria, which is then utilized to make ATP -- adenosine triphosphate." Beal, a co-author of the PNAS paper, teaches and does research at Harvard Medical School and Massachusetts General Hospital. He serves on MitoKor's scientific advisory board.

Researchers first suspected a mitochondrial connection to AD, Beal recounted, "when they found that there are cytochrome oxidase (CO) defects in blood platelets, which have mitochondria but no nucleus. And platelets have a number of enzymes and transporters similar to those that occur in the brain.

"That led to the query," Beal added: "Does the CO defect originate as a mitochondrial DNA defect or merely some secondary dysfunction of the enzyme?"

Seeking answers, the MitoKor contingent extracted mitochondria from the platelets of AD patients and transferred them into human neuroblastoma cells that they had stripped of their own mitochondrial DNA. That maneuver put those platelet mitochondria, which had started life with a blood-cell nuclear DNA background, into a new nuclear setting.

There, if they still exhibited the CO defect, it meant that the defect had been encoded on the mitochondrial genome.

"They did indeed see this defect," Beal recalled, "so, by gene sequencing, they went on to discover specific alterations in the CO gene. Those point mutations occur on nucleotide 7650 of the mitochondrial genome. By population screening, the team was then able to show that those mutations were indeed associated with AD."

The company's collaborators at five academic and clinical centers across the U.S. collected 506 patients with a clinical diagnosis of "probably Alzheimer's disease," and 95 age-matched control individuals.

Correlating for the presence of the mutation in the blood of these cohorts, they found that it was rare in non-AD controls, but common in "probable AD" patients. Conversely, it was common for AD cases to have high levels of the mutant base, but rare for controls. "The presence of high levels of this mutant mtDNA molecule is 100 percent specific for AD," the article reported. "Approximately 60 percent of AD cases had levels of this mutant base exceeding 20.3 percent, but only 20 percent of controls did."

Davis and his group also studied the relative abundance of the disease-associated mutation in asymptomatic children of affected mothers and fathers. They found "significantly higher rates of mutants" in the offspring of mothers. These findings, they suggest, may explain why the risk of AD increases if a maternal relative has the disease.

"Because we can model the disease in those neuroblastoma cells," Davis pointed out, "that gives us a very nice avenue for screening for novel therapeutics. And we've used those cells to do that. The CO defect we have leads to a generation of oxygen-free radicals. We can suppress those genetic lesions and normalize CO activity, with small orally available molecules that have the potential to get into the brain. We are already testing these in the cell lines."

Diagnostics Now, Pills Later

While not at liberty to describe the mode of action of those candidate drugs, Davis allowed that "they are not classical antioxidants, free radical-scavenging molecules. They work specifically on selected sites on the mitochondria."

He surmised that "the first in vivo studies will probably start before the end of this year. We still want to select the best molecules to take forward. We're in the optimization stage right now. Then, after those preclinicals," Davis went on, "will come clinical trials, and some day there'll be a pill against AD. That's where we're heading."

Even before a therapeutic, the aim is to develop a diagnostic. "Clearly," he pointed out, "screening for the mutation has the ability to confirm a diagnosis of AD in about 60 percent of the AD population," and a simple blood test may prove practical. MitoKor has just hired a vice president of diagnostics, Davis said. "Her job is to commercialize the diagnostic test."

MitoKor has already developed an Alzheimer's assay that is currently being used by two major pharmaceutical companies to stratify patients for late-stage clinical trials of their anti-dementia therapeutics. *