Fatal familial insomnia

A rare genetic mutation has two methods of killing: it deprives you of sleep, permanently, or it makes you demented. Both methods work.

FIRST COME THE LONG nights of unrelenting, untreatable insomnia. Hallucinations and loss of memory follow, then the physical symptoms: excessive sweating, high fever, a racing heart, and sudden muscle twitches. "It's not only that you don't get to sleep, but you keep up all the rhythms of wakefulness all the time," explains pathologist Pierluigi Gambetti of Case Western Reserve. "When you sleep your heartbeat goes down, your breathing rate goes down--your body goes into kind of an idle mode. This is like keeping the engine racing all the time. "The end comes in little more than a year, and it is always the same: stupor, followed by coma, followed by death.

Fatal familial insomnia, as this bizarre disease is called, is a genetic syndrome that was first described in 1986 by Gambetti and Elio Lugaresi of the University of Bologna. It attacks both men and women, usually in their late 50s. The disease is very rare. So far, only five families have been diagnosed with it, though three more are under scrutiny.

Yet researchers are intrigued by it because it appears to be another of the perplexing prion diseases--a group of invariably fatal illnesses in which a normal brain protein, called the prion protein, is somehow changed, so that it begins to form abnormal clumps. In fact, Gambetti and his collaborators have recently discovered that fatal familial insomnia is astonishingly close in origin to a prion disease called Creutzfeldt-Jakob. The early symptom of Creutzfeldt-Jakob is dementia rather than insomnia, and the two disease have different patterns of brain lesions. But they are associated with the same mutation at the same spot on the same gene--the one for the prion protein. Unlike fatal insomnia, Creutzfeldt-Jakob is usually transmitted through infection, by an agent whose identity and modus operandi are unclear. In about 10 percent of Creutzfeldt-Jakob patients, however, the disease is transmitted genetically.

Even as Gambetti and Robert Petersen at Case Western were zeroing in on the genetic cause of fatal familial insomnia, Lev Goldfrab and his colleagues at the National Institutes of Health were independently trying to identify the mutation that causes genetic Creutzfeldt-Jakob. Surprisingly, the two research teams met at the same point on the prion gene.

The mutation they found is at a site known as codon 178. (A codon is a sequence of three DNA nucleotides that codes for a single amino acid in the chain that constitutes a protein.) Normally codon 178 on the prion gene instructs the cell to make an amino acid called aspartic acid. But the mutation Gambetti and Goldfarb found turns that instruction into the code for asparagine.

A change of a single amino acid can alter the shape of the whole protein, disrupting its function--that's how most genetic diseases are caused. What's unusual is a single mutation causing two different diseases. "I wondered whether we might be dealing with two differently abnormal prion proteins," says Gambetti. "And if the protein is different, there must be a difference in the gene." The two teams decided to pool their data and look for that difference.

They found it at codon 129, which is the site of what geneticists call a polymorphism--a normal genetic variation. While in some parts of a protein the substitution of one amino acid for another can be lethal--as it is at codon 178--in other places different amino acids can act as harmless synonyms for one another. Thus codon 129 in the prion gene encodes the amino acid methionine in some people, and valine in others--and ordinarily, both types of people are perfectly healthy.

But when a person happens to have a mutation at codon 178, the polymorphism at codon 129 apparently takes on a different significance: it determines which prion disease the victim will die of. Of the 15 insomniacs from five families that Gambetti and Goldfrab studied, all had methionine at codon 129. Fifteen victims of genetic Creutzfeldt-Jakob all had valine.

Somehow the mutated part of prion protein must interact with the polymorphic part. Gambetti speculates that the mutation might bend the protein, bringing part of it into contact with the amino acid encoded by codon 129. Depending on whether that amino acid was methionine or valine, the mutant protein might form, say, a closed loop, or just a twisted chain. And somehow that small difference would lead to different patterns of brain lesions, and to different afflictions--but not, sadly, to any difference in the patients' fate.

COPYRIGHT 1993 Discover
COPYRIGHT 2004 Gale Group