By David N. Leff

Here is the verbatim testimony of patients who “hear” sounds in their heads that aren’t there:

“All at once, in my left ear I heard a motor, similar to one that runs a washing machine. First, it was a click, like valve springs adink, adink, adink. Then it went faster and louder, like whackety-whackety.”

“A clicking of horses’ hooves.”

“I heard buzzing or bells ringing, and whooshing, as if I was in a wind tunnel.”

“Mainly the thumping of a helicopter.”

“There was a noise that sounded like a popgun. It got louder and louder till I blacked out with a grand mal epileptic seizure.”

“More than one patient said that those unreal sounds were triggered by a loud voice,” observed genetic epidemiologist Ruth Ottman at Columbia University in New York, who focuses on epilepsy. “We were reporting these auditory hallucinations as part of the manifestation of their seizures. If you talk to most neurologists about it, they will say it’s uncommon.”

Ottman is senior author of a paper in Nature Genetics, released online, Jan. 28, 2002. Its title: “Mutations in LGI1 cause autosomal-dominant partial epilepsy with auditory features.”

“We’ve identified a gene that causes a particular form of epilepsy,” Ottman told BioWorld Today. “This gene, LGI1, works differently from genes found in other forms of epilepsy.”

The LGI1 gene’s acronym has historical but puzzling origins in brain cancer.

“It stands for leucine-rich glioma inactivated 1.’ This gene had previously been studied by cancer researchers, who were interested in glioma a form of brain malignancy. When they looked at cell lines and tissue from glioma tumors, they found that this gene, LGI1, was completely inactivated. That’s different from what we see,” Ottman recounted, “because what I mean by completely inactivated’ is that both of the gene’s two alleles were inactive. Whereas members of the families with this rare type of auditory epilepsy that we studied have one inactivated allele in that gene.”

Leucine Not Really Involved In LGI1

“That doesn’t explain the leucine part of it,” she continued. “The proteins encoded by this class of genes are rich in the amino acid leucine. Proteins that look like that have been studied in other organisms and tend to be involved in neuronal migration. During embryonic development, the neurons migrate to positions where they are going to play their role in the mature brain.

“We don’t know how common this auditory variant of epilepsy is,” Ottman pointed out. “It’s never been systematically studied. I’m just starting a new research project where I will look in detail at the clinical characteristics of epilepsies that occur in the general population so we can try to get a handle on how common or rare the auditory variant is.

“Here at Columbia,” Ottman recalled, “when we started doing this work, there was a database where neurologists would enter certain infomation about their patients. We searched for words that sounded as if they would pick up auditory things like whooshing or helicopters, and we came up with a small proportion. But those auditory symptoms were important because they gave us an indication of where the seizure activity is occurring in the brain.

“In fact,” she added, “they point to the lateral temporal lobe. LGI1 is the first gene that’s been found for a form of temporal lobe epilepsy. But more specifically, it’s the lateral temporal lobe, not the mesial, and that’s important for neurologists.”

Then she and her co-authors set out aggressively to find auditory epilepsy families. They queried the Epilepsy Foundation at its national meeting, and ascertained one family that way. Then they did a mailing to all members of the American Epilepsy Society, which netted one additional family.

“And we ended up with these five families described in our paper,” Ottman observed. “We have two others that we haven’t studied yet. People around the world are beginning to describe this rare epilepsy,” she noted. “One family [was] reported in Italy, one in Germany, one from Norway, and a very large extended family from the Basque country, which was reported on in 1998. And that Basque pedigree helped us narrow the chromosome 10 neighborhood locus for the gene; it overlapped by only 3 centiMorgans.

“Partial epilepsy,” Ottman explained, “is synonymous with focal. It means the seizures originate in a localized area of the brain. Their onset is in a narrowly defined region in this case the lateral temporal lobe. That’s opposed to generalized epilepsy, which involves the entire brain both hemispheres from the very start of the seizure. So the fact that these auditory hallucinating people hear sounds is an indication that their epilepsy has a focal onset. Otherwise, they would just lose consciousness immediately.”

Schizophrenia Connection Not Likely

“We’ve seen onset of the auditory seizures,” she observed, “reported anywhere between age 8 and probably in the 20s. But we don’t know where the primary problem arises perhaps during fetal development.”

Ottman sees no hard evidence that the auditory variant is related to other hallucinatory disorders such as schizophrenia. “Most of what we hear reported does not have the same kind of cognitive feeling of schizophrenia. The auditories are not complex hallucinations. One patient did report that the noise in his head sounded like people talking backwards.”

In the Nature Genetics paper, the co-authors reported identifying the LGI1 gene by studying the DNA it encodes from their five affected families, and finding mutations in the gene. However, Ottman pointed out, “inheriting a mutant gene copy did not guarantee getting the rare epilepsy. Now that we knew what LGI1 was, we could actually see if they had the gene. And only 71 percent penetrance of those who did were actually affected with that epilepsy.”

As for putative therapeutic applications, Ottman suggested, “I don’t think it’s the auditoriness so much as what we will learn from the basic brain mechanism. All eight of the epilepsy genes identified so far have involved ion channels. So the fact that LGI1 has a different modus operandi,” she concluded, “might suggest that there are therapies that could be directed to other types of mechanisms than ion channels.”