For people who think shark when they hear hammerhead, the hammerhead ribozyme's mechanism of action sounds like a feeding frenzy gone bad: self-cleavage leading to destruction.

But it is a novel method for scientists to control gene expression in mammalian cells, and it has potential applications across a wide range of research and clinical areas.

Proof of principle of the method was published in the Sept. 23, 2004, issue of Nature, in a paper titled "Exogenous control of mammalian gene expression through modulation of RNA self-cleavage."

Current methods of controlling gene expression are usually focused at the level of transcription and "have found tremendous applications in basic science," said senior author Richard Mulligan, director of gene therapy research at Children's Hospital Boston and director of the Harvard Gene Therapy Initiative. Nevertheless, Mulligan and his colleagues at Harvard and the Children's Hospital wanted to find a simpler system that would be widely applicable in basic research, as well as in the clinic.

The researchers focused their efforts at translation, the process of making proteins from mRNA. They focused on ribozymes, self-cleaving RNA sequences that inhibit translation when they are part of an mRNA sequence. Previous research had shown that efficient cleavage could be generated by inserting ribozyme sequences in vitro. So-called allosteric ribozymes, which self-cleave in response to specific concentrations of a molecule, also had been generated in vitro, enabling fine control of gene expression.

It's Very Simple Until You Try To Make It Work

"In principle, one only needs to incorporate sequences into the RNA transcription unit and find a molecule to inhibit it," Mulligan said. Colleague Laising Yen, lead author of the study, translated "in principle" into the realities of the lab: "There was a lot of trial and error."

Success came with the use of a ribozyme, which has structural features that have earned it the hammerhead ribozyme moniker. The authors demonstrated that gene constructs bearing a hammerhead ribozyme coding sequence and the reporter gene lacZ were incorporated into host DNA and transcribed in cell cultures. But those gene constructs were not expressed because the hammerhead ribozyme cleaved itself after transcription, preventing translation into a protein.

The authors next began to search for molecules that could inhibit the ribozymes from self-cleaving, so that the proteins would actually be expressed. In test tubes, a number of compounds, including standard antibiotics, can inhibit such self-cleavage; in cell cultures, however, that approach proved to be a bust. The researchers tried more than 100 commonly available compounds, but "very few of them functioned in the context of an RNA transcript in a mammalian cell," Mulligan said.

After some searching, the researchers found two approaches that were successful in inhibiting ribozyme self-cleavage. Antisense oligonucleotides worked if the right sequence in the ribozyme was targeted. Additionally, high-throughput screening identified compounds capable of inhibiting self-cleavage. The details of that research are forthcoming in another publication, but one compound was used in the Nature paper to demonstrate the use of the approach in vivo.

In the experiments, mice were injected with viral vectors carrying a ribozyme coding sequence, along with the luciferase reporter gene. Some mice then received a compound to inhibit the ribozyme. Mice receiving the ribozyme gene plus the inhibitory compound expressed the luciferase reporter protein; mice receiving just the ribozyme did not, because the ribozyme sequence prevented expression of the protein. Mice receiving a construct with an inactive ribozyme also expressed the protein, showing that its non-expression in mice with active ribozyme was not due to a failure of the gene constructs to integrate into host cells.

Mulligan enumerated several advantages of the ribozyme approach. Simplicity is one. Unlike the regulation of gene expression at the transcription level, regulation via ribozymes can be achieved without the need for transactivators or specialized promoters. The need for those elements also has made it hard to control the expression of endogenous genes by transcription regulation, something that Mulligan believes could be done more easily with the ribozyme approach.

Another strength is that "if you are able to introduce self-cleaving systems, the default would be non-functional," potentially improving both the specificity and the safety of gene delivery, Mulligan said. He named "developing safer tools for gene therapy" as an overarching goal of his research.

Infinite Number Of Regulatory Systems'

As one way of improving safety, Mulligan pointed to the theoretical ability of incorporating "conditional ablation," i.e. a kill switch, into the system. For that, the gene construct simply would contain the coding sequence for a harmful protein with a ribozyme. Because the default of the ribozyme system is off, such a protein would normally not be expressed. However, if problems developed, the ribozyme could be inhibited and the cell induced to kill itself.

Given that the current Nature publication does no less, but also no more, than show proof of principle for the ribozyme approach, Yen, Mulligan and their colleagues are working to improve the system in practice. They also have entered a collaboration with researchers at Yale University attempting to bring allosteric ribozymes - ribozymes with self-cleaving activity that is dependent not only on the presence or absence of an inhibitory molecule, but can be influenced by its concentration - from test tubes into cells and, ultimately, whole animals.

Though enthusiastic about its possibilities, Mulligan cautions against expecting to see the method bear practical fruit too soon. "The approach is not ready for prime time in either basic research or clinical applications. But we have illustrated its possibilities," he said. "This kind of system would make it possible to make an infinite number of regulatory systems."