Josh P. Roberts has an M.A. in the history and philosophy of science, and he also went through the Ph.D. program in molecular, cellular, developmental biology, and genetics at the University of Minnesota, with dissertation research in ocular immunology.
When they need to get cells to take up nucleic acids and express desired proteins, researchers typically turn to transfection. But when it comes to making the right protein, cells don’t always have an open-pore policy. Primary cells, stem cells and other slow- or non-dividing cells, for example, can be refractory to transfection.
That’s when you turn to nature’s natural vectors—sort of. Viruses have evolved alongside just about every species out there, and they have developed crafty ways of entering cells and turning them into viral factories. That makes viruses efficient delivery and expression systems—and also potential biohazards. Fortunately, few researchers outside of virology labs need work with pathogens in their native state. Here we take a look at some of your engineered options.
Don’t bug me
All things being equal, if you can avoid viral transduction for protein-expression work, you should. “Any time you can use bacteria to express protein, do that—you’re always going to make more, and it costs a fraction of what it takes to do eukaryotic expression,” says John Elder, professor of immunology and microbial science and director of protein expression at the Protein Expression and Proteomics Core facility at The Scripps Research Institute. But proteins for which the correct folding and post-translational modifications are important—such as glycoproteins—require a eukaryotic, and perhaps mammalian, expression system.
Nucleic acids can be delivered to many cell types by chemical means, with perhaps 10% to 15% of cells taking up the plasmid. Many others cell types require harsher, mechanical treatments like electroporation. “You end up with a lot of mortality; it’s hard on the cells,” Elder notes.
Some cells, though, are just inherently resistant. "In our lab, we use these primary cultures of airway epithelia, and they’re not transfectable by plasmids. There’s just no way to get it in,” says Patrick Sinn, research associate professor and director of the viral core at the University of Iowa. “To do simple things, we have to make viral vectors.”
The ‘Big Three’
Viral-expression vectors are created by replacing some or all of the genes needed for viral replication with an expression cassette for the gene of interest. The construct is transfected into specialized packaging cells, which supply the necessary replication machinery in trans (sometimes on separate plasmids), allowing replication-deficient viral particles to be produced in large numbers. “They won’t let us use viruses that self-replicate—they have to be one-hit viruses to be approved by the safety committees,” explains William Osborne, a research professor in the Department of Pediatrics at the University of Washington, who also “runs the viral part” of the university’s Diabetes Research Center Viral Vector and Transgenic Mouse Core.
Many different virus types have been explored as potential ways to introduce nucleic acids into cells. Issues such as immunogenicity, systemic toxicity or the ability to cross the blood-brain barrier, for example, may be of great concern for gene-therapy applications but do not generally come into play when choosing a vector for in vitro delivery. Other issues, such as the maximum payload size, host genome integration, impact on physiology and yield, are of more general concern and have led the field to largely coalesce around three types of virus: adenoviruses, adeno-associated viruses and lentiviruses.
“Between Ad [adenovirus], AAV [adeno-associated virus] and lenti [lentivirus], you’ve covered most of the bases of what you need. Those are usually superior for ease of use and protein production,” says Sinn. “The ‘Big Three’ are the big three for a reason.”
Each has a wide tropism (or can be pseudotyped) to allow them to enter and deliver a payload to nuclei of a wide variety of mammalian cells. Several vendors market kits to produce virus with various promoters and other regulatory elements, serotypes and tags or selection markers such as GFP and antibiotic resistance. Researchers also can take advantage of the many contract service providers and university core facilities with expertise in viral-vector production.
To integrate or not to integrate
Henry Chiou, senior product manager for expression systems at Thermo Fisher Scientific, says lentiviral expression vectors are the most frequently ordered viral kits at his company. “Lenti has become so popular because it’s really the only virus that can transduce into nonproliferating cells and get high levels of expression.” Lentiviruses integrate into the host genome, enabling stable cell lines expressing the transgene to be created.
Adenovirus, on the other hand, can be grown to a very high titer—typically around 1010 infectious units (IFUs) per ml, which is about two logs greater than lenti, Chiou notes—but it does not integrate into its host and so will be lost or diluted as the cell divides. “Ad typically will have a pretty good burst of expression but will start to cause cytotoxicity in the cell, so you have a limited time frame,” he explains.
For most applications, you need a multiplicity of infection of about 10 virus particles/target cell, says Osborne. Lenti can be concentrated by ultracentrifugation or transflow filtration, allowing for a final titer on par with adenovirus.
Both lenti and Ad have respectable carrying capacities—about 10 kb to 12 kb for lenti and between 7.5 kb and 35 kb (depending on the construct and the packaging cell) for adenovirus—enabling designs incorporating complex regulatory elements and possibly multiple genes. Both vectors require the use of Biosafety Level 2 (BSL-2) protocols.
Recombinant AAV—as long as it does not contain a tumorigenic or toxic insert and does not require a helper virus (which most modern constructs do not)—requires only BSL-1 protocols. AAV is a nontoxic, nonintegrating vector. It can be grown to similar titers to lenti. Because it is nonmutagenic and generally less immunogenic than adenovirus itself, AAV is often used in vivo. “The biggest drawback is you can only put in 4 kb of your genes. They have a very limited carrying capacity,” Osborne says.
Of the three viral systems, “lenti is probably the easiest one to produce by yourself. It’s just a three- or four-plasmid transfection, and [then] you collect the [supernatants] and you have a dirty, low-titer prep of virus,” says Sinn. “There’s a lot steeper learning curve to produce your own Ad and AAV, although it can be done.”
Other viral vectors, which can be found in the literature and in vendors’ and service providers’ catalogs, may offer their own advantages over the ‘Big Three’ for certain niche applications. But whichever you choose, beware: The ability to reliably deliver a nucleic-acid payload to recalcitrant cells can be infective.
Image: AAV particles. Credit: Dr Graham Beards at en.wikipedia [CC-BY-SA-3.0 (http://creativecommons.org/licenses/by-sa/3.0) or GFDL (http://www.gnu.org/copyleft/fdl.html)], from Wikimedia Commons (Source)