10.1: Nucleic acid delivery

In a lab course or in a research lab, you may have transformed bacteria such as E. coli with a gene on a plasmid. The most common ways to do this involve using chemicals or electricity to open transient pores in the cell, allowing plasmids in the solution to enter. Unfortunately, getting DNA or RNA into human cells reliably using these chemical and physical methods is very difficult! In this section, we'll describe several ways that are currently used.

Viral delivery

Fortunately, there is a biological system that has evolved to be very good at transferring nucleic acids into cells, and that is viruses. So, many gene therapies, both early efforts as well as modern licensed therapies, use viruses to transfer either DNA or RNA into cells.

What does this look like from a practical standpoint? How do you “make” a custom virus? The two key tools are modern recombinant DNA technology and modern cell culture methods. We usually need two or three plasmids – one called the “packaging” plasmid, which encodes the proteins require to assemble a virus, and another with the gene that we want to transfer – called the transgene – into the cell. We transfect these two plasmids into mammalian cells growing in culture, which read the packaging genes from one plasmid and the transgene from another. Then, the virii spontaneously assemble and the cells lyse, producing ready-to-use gene therapy virii bearing the transgene.

Why do we have to do this complicated dance? Well, we don’t want these virii to be able to subsequently reproduce and “spread” the transgene! If we separate the transgene from the packaging genes, the virii don’t end up with copies of the packaging genes – that makes them replication incompetent. Once the patient is transfected with the virii, the transgene gets delivered but genes required to make more of the virus don’t.

Types of viral vectors

All this is well and good in theory, but there are a LOT of viruses we could engineer like this. Which one to choose? 

When it comes to deciding which virus to use for a gene therapy, there are two practical considerations that I think are important. First, not all cells in the patient can be infected by every virus! And this makes sense, right – a virus that causes a cold infects cells in the airway, not liver cells or muscle cells. Which cells can be infected by a virus is called the virus’ tropism, and this can actually be used to a therapeutic advantage. For example, if the goal is to cure cystic fibrosis by introducing a working copy of the CFTR gene, we could use a virus that specifically infects cells of the airway.

The second practical matter is what happens to the transgene once it’s introduced to the cell. One commonly used type of virus is a retrovirus, which integrates the transgene into the genome. In some cases, this is a good thing – if the cells are dividing, such as bone marrow cells, then you want the daughter cells to also carry the transgene. However, where the transgene is integrated is random! What if, for example, it integrated into the middle of a tumor suppressor and shut it off? Or even worse, if it integrated in the promoter of a proto-oncogene, activating it and turning it into an oncogene? In some trials, up to 10% of the patients got cancer because of the gene therapy. For a really terrible disease like severe combined immunodeficiency, that may be a risk that a patient is willing to run, but it’s not acceptable for a broad therapy.

So using retroviruses comes with risks. However, not all virii lead to genomic integration of the transgene. Adenovirii and adeno-associated virii or AAV are also commonly used in gene therapy, and in these virii the transgene is transported into the nucleus but remains episomal. It’s not integrated into the genome. However, that means that when infected cells divide, the transgene isn’t copied and is lost. This makes adenovirus and AAV vectors suitable for transfecting cells that are long-lived, like neurons and muscle cells, and less suitable for rapidly dividing cells like haematopoetic stem cells.

Viruses have one final disadvantage – our immune system detects them! In particular, this can make repeated administration difficult because patients become effectively immune to the therapy. But what if there was another delivery mechanism that didn’t rely on viruses at all? 

 

Lipid nanoparticles

We mentioned earlier that getting DNA and RNA into cells reliably using chemical methods is hard. That doesn’t mean it’s impossible, though, and recent advancements have made chemical delivery of nucleic acids into cells a viable platform for gene therapy. Modern lipids used for this purpose are ionizable – they’re positively charged at low pH and neutral at physiological pH. Lipid-nucleic acid complexes are transported into cells in endosomes, which then become acidic. This causes the lipids to protonate, which seems to destabilize the endosome and release the nulceic acids into the cell, though the details are still a little hazy.

It’s also important to note that these nucleic acids are release into the cytosol, not the nucleus. That means that DNA like plasmids isn’t much good in quiescent non-dividing cells, because it won’t make it into the nucleus until the nuclear membrane disassembles and reassembles over the course of mitosis. RNA, however, is a great candidate for lipid nanoparticle delivery, particularly if what you want is a short-lived dose of transgene expression. And this makes RNA and lipid nanoparticles an excellent platform for vaccine development. This is why the Pfizer and Moderna COVID vaccines rely on mRNA wrapped up in lipid nanoparticles – they cause a short-lived expression of the COVID spike protein on transfected muscle cells, which is enough to teach the immune system to recognize the “real” virus when you encounter it.