Virus Vision

Viruses are among a cell’s tiniest predators – and for a long time, their size and biology have kept them a mystery to observing microbiologists. But what if we had an easy way to identify not only which cells were infected, but also which proteins were affected by the virus? That’s precisely what Jens-Ola Ekström, Dan Hultmark, and their colleagues at Umeå University have developed with a new system called Munin (1). So far, they’ve used their new method to observe picornavirus infections in fruit flies – but the potential goes far beyond that.

Immunofluorescence of the stomach of a Nora virus-infected fruit fly, showing muscle cells (green), uninfected cell nuclei (blue), and infected cell nuclei (red). Credit: Jens-Ola Ekström

What?

Munin is a system for expression of any gene of interest in cells infected by a virus – and it can easily take advantage of the virus’ specific proteases to determine whether or not a particular cell is infected. The system is based on a ubiquitously expressed target protein for a protease produced by the virus of interest. That target protein, Gal4, is a transcription factor connected to a membrane anchor (a transmembrane domain preventing the protein from reaching the nucleus) via a cleavage site for the viral protease. In infected cells, the virus-encoded protease releases Gal4 from its membrane anchor, allowing it to enter the nucleus and control transcription of any gene equipped with the yeast promoter UAS. If an infected cell contains a fluorescent protein gene controlled by UAS, virus-released Gal4 can activate that protein – but if no infection is present, Gal4 will never reach the nucleus, and the cell will never fluoresce.

The system isn’t limited to detecting infected cells, though; it can be used to overexpress any gene of interest, including those that researchers suspect may be affected by the virus. Even more importantly, it’s also possible to knock down endogenous genes by expressing an RNA hairpin that feeds into the RNAi system. For fruit flies, the researchers’ model organism of choice, the research community freely provides thousands of transgenic RNAi constructs covering a majority of known genes – which allows the testing of almost any gene of interest in a virus-host relation, and may even permit genetic screening if a simple enough readout is available.

Why?

“The goal of our project is to understand the factors that control and limit persistent infections of RNA viruses,” say researchers Jens-Ola Ekström and Dan Hultmark. “With limited resources we may not be able to pursue this goal as far as we had hoped, but we are happy to share our plasmids and fly stocks with other researchers.”

Why choose Drosophila as a model system – and why picornavirus infections? “The very first step in viral immune defense in vertebrates is dependent on innate immunity,” explain the researchers, “and Drosophila is an excellent model for innate immunity.” Fruit flies have a record of discoveries with an impact on human innate immunity – for instance, the discoveries of toll-like receptors and defensins. The model organism also justifies the choice of virus; because some picorna-like viruses infect Drosophila species, Ekström and Hultmark thought they would be a good choice for detailed study of molecular virus-host interactions. “Specifically, we were curious about the phenomenon of persistent RNA virus infections such as those of the Nora virus, which infects the fruit fly and was discovered in our lab.”

Where next?

“One goal of the project was to create a system for fast and specific detection of infection, as a tool for genetic screening of host genes that affect viral infection. This goal was only partially achieved with the fluorescent readout,” say Ekström and Hultmark. “We can detect infection in living animals, but only when the Nora virus is injected – not when they are infected the natural way, via feeding. The normal gut infection can only be detected after dissection of the animal, which is not practical for screening purposes. We hope that the solution will be to use the Munin system to express a protein that gives rise to a phenotype that we can score from the outside. We have begun to work along those lines, but the problem is not solved yet.”

Munin builds on transgenic technology, which means it’s limited to model organisms and tissue cultures at the moment. That makes it useful for research, but not in clinical testing, because the need for models makes it less cost-effective than PCR- or sequencing-based analyses. But that doesn’t mean there’s no place for Munin in laboratory medicine. “We developed the system for research purposes, but it could probably have a role in broad analyses for certain groups of viruses, including clinically important ones,” say its creators, “as Munin is much less sensitive for mutations than primer- or probe-based methods.”

Where next?

“One goal of the project was to create a system for fast and specific detection of infection, as a tool for genetic screening of host genes that affect viral infection. This goal was only partially achieved with the fluorescent readout,” say Ekström and Hultmark. “We can detect infection in living animals, but only when the Nora virus is injected – not when they are infected the natural way, via feeding. The normal gut infection can only be detected after dissection of the animal, which is not practical for screening purposes. We hope that the solution will be to use the Munin system to express a protein that gives rise to a phenotype that we can score from the outside. We have begun to work along those lines, but the problem is not solved yet.”

Munin builds on transgenic technology, which means it’s limited to model organisms and tissue cultures at the moment. That makes it useful for research, but not in clinical testing, because the need for models makes it less cost-effective than PCR- or sequencing-based analyses. But that doesn’t mean there’s no place for Munin in laboratory medicine. “We developed the system for research purposes, but it could probably have a role in broad analyses for certain groups of viruses, including clinically important ones,” say its creators, “as Munin is much less sensitive for mutations than primer- or probe-based methods.”