By David N. Leff

Eleven years ago, molecular geneticist Haig Kazazian found a strange DNA sequence wedged into the Factor VIII gene of a boy with hemophilia. That disease is famously hereditary, but in this case the patient¿s clotting factor gene had been disrupted by a wandering, wild-card stretch of genomic DNA.

Four or five years later, Kazazian recalled, ¿I was sent a sample for analysis on a young man with muscular dystrophy. We found a DNA insertion that had jumped into an exon [protein-coding sequence] of his dystrophin gene.¿

These weird visitations may remind some of aliens from outer space snatching unsuspecting earthlings into their unidentified flying saucers.

In fact, these random insertions come not from outer space but from inner space ¿ the mammalian genome. They consist of retrotransposons ¿ jumping genes ¿ which are short DNA sequences that copy and paste new copies of themselves by reverse transcription of an RNA go-between sequence.

Recalling his Factor VIII discovery, Kazazian, who chairs the department of genetics at the University of Pennsylvania in Philadelphia, told BioWorld Today: ¿After finding that insertion, we had painstakingly isolated its precursor retrotransposon.¿

As for the muscular dystrophy case, Kazazian went on, ¿The clinical phenotype was a little bit different, in the sense that his disease wasn¿t as severe as Duchenne¿s, but seemed a little bit worse than a straightforward Becker¿s muscular dystrophy. That wasn¿t the first retrotransposon insertion that we¿d seen going into a gene, but the first time we¿d seen one that carried flanking sequences along with it ¿ 600 base pairs¿ worth.

¿In this case,¿ he pointed out, ¿that flanking material really didn¿t have any function. Now we know that there is the possibility at least, that when these flanking sequences jump, they could exchange exons. And if those exons jump into genes, they could make new genes.¿

Retrotransposons Could Be Evolutionary Agents

Making new genes, and hence potentially new forms of life, is the essence of Darwinian evolution. Kazazian is one of many researchers in the field of retrotransposons who suspect these free-floating elements that prowl the genome are indeed agents of evolutionary change. ¿They can insert into genes in two different orientations,¿ he observed, ¿having their sequence flipped one way or the other. And there are a lot of other possibilities for altering genes and gene sequences in this way.

¿A typical retrotransposon element,¿ Kazazian explained, ¿would have an internal promoter driving its transcription at one end, for synthesis of RNA. And it would have two protein-coding regions. The first one seems to make a protein that binds to the element¿s RNA transcript. The second one makes an enzyme that is able to cut DNA. And it also expresses a reverse transcriptase, which is the way the element makes the DNA from its RNA copy, to get back into the genome again.

¿We were really surprised,¿ he continued, ¿when colleagues found that the element had endonuclease activity. That it wasn¿t just waiting for the endogenous enzymes in the genome to do the job of getting it back into the genome. It actually had its own activity, and was ready and raring to go, to replicate itself.¿

¿Presumably,¿ Kazazian suggested, ¿these retrotransposable elements are becoming more frequent, and then dying down in numbers. And it turns out that the numbers of active ones that we humans have in our genomes are maybe 60 times less than the number of active retrotransposons that the mouse has. The mouse is loaded. We have about 50 that are active. The mouse has on the order of 3,000.¿

Kazazian is senior author of a paper in the current issue of Science, dated March 5, 1999. Its title is ¿Exon shuffling by L1 retrotransposition.¿ L1 stands for LINE1, which stands for ¿long interspersed nuclear elements,¿ the most abundant retrotransposons in the human genome.

¿We were looking to see if there was a possibility of moving genomic sequences from one site in the genome to another, via these retrotransposons,¿ Kazazian said, ¿as we¿d seen in that dystrophin patient. We wanted to ask two main questions: How frequently do retrotransposition events occur into the genes? And secondly, how efficient is the process by which flanking sequences go along with those events?¿

To assay that frequency and efficiency, the co-authors engineered an L1 retrotransposon to include a drug-resistance reporter gene that would turn on only when the mobile L1 jumped into a working gene. The results they reported in Science indicate that at least 6 percent of all retrotransposition events effectively zeroed in on genes.

¿What our paper shows,¿ Kazazian pointed out, ¿is that there does not seem to be any bias against insertions into genes. That¿s an important point. It looks as though, at least in our assay system, that retrotransposition events into genes ¿ generally into introns [non-coding stretches between exons] but occasionally into exons ¿ are not selected against. They do occur with about the same frequency as one would expect from the fraction of genes in the genome. That fraction of genomic space occupied by genes is on the order of maybe 15 percent. And we found that the fraction of retrotransposon insertions that went into genes was about 6 percent.¿

Farfetched Clinical Application: Gene Therapy

The only practical implication of this phenomenon that Kazazian sees is that ¿these insertions can go into genes and cause disease mutations. They¿re probably not that frequent. We had earlier found that pathological frequency may be on the order of under 1 percent of all new mutations. But it¿s interesting,¿ he added, ¿that in the mouse we have found that retrotransposition events account for on the order of 10 percent of all spontaneous mutations. Those rodents are under much more of a burden than we are.

¿From the human point of view,¿ he observed, ¿we¿ve isolated about 20 percent of the 50 that are active in the human genome. Some of them are more active than others. And we¿re hoping, from a practical point of view, that we¿re going some day to be able to use these elements for two things:

¿One, to make insertional mutations some time in the future to isolate genes, by putting them into the mouse and seeing them jump into the mouse genome.

¿The other possibility, a bit more farfetched, is some day being able to use these elements, which are endogenous in our genome, as vectors in gene therapy. Instead of inserting a drug-resistance marker gene, have some functional gene we wanted to replace, and drop that back into the genome. But we¿ve got a long way to go,¿ Kazazian concluded, ¿before we can do that ¿ modify the element in different ways to make it more palatable. I think that¿s a whole big frontier.¿ n