Could a virus that normally serves as a vehicle (vector) for delivering a cold be modified to “deliver” immunity? That’s the hope with Oxford/AstraZeneca’s ChadOx1 vaccine. And it’s not the only coronavirus vaccine taking such a “viral vector” strategy. Despite the quick movement and swift progress, there’s only been one proven-effective viral vector vaccine approved for human use (I think). But viral vectors do have a long history of use in the lab and a mixed history of use for gene therapy. So, instead of just telling you about the vaccine, I want to tell you some of the adenovirus vector background (hope you’ll stick around!)⠀

Yesterday we talked about how mRNA coronavirus vaccines (like those from Pfizer & Moderna) deliver the genetic instructions for making the coronavirus Spike protein (S) in the form of messenger RNA (mRNA). This can work because the cells that receive the instructions can make S protein from them – S is the one that juts out from the viral membrane and allows it to bind to and get into human cells. On its own, this protein is harmless, but it is foreign, so the body will learn to mount an immune response against it if it finds that protein again when the virus were to try to attack. However, since naked RNA is fragile and negatively-charged, this mRNA has to encapsulated in lipids (oily molecules) in the form of Lipid NanoParticles (LNPs) in order to get into cells and kept at really cold temperatures before they do so. 

Viral vector vaccines therefore take a different approach to delivering S protein instructions, sticking the S gene into a different, harmless, virus and using that virus as a “vector” (vehicle) for the gene delivery. This strategy makes a lot of sense, since viruses offer a lot of protection to the genetic info they contain, and they’re naturally really good at getting genetic info into cells (that’s their M.O.!) But, this strategy is not without problems of its own, so let’s dive into the pros, cons, and controversies

You’re probably very familiar with adenoviruses even if you don’t know it – they cause ~5% of what we refer to as “colds” – one of the reason’s there’s no cure for the common cold and you can keep getting colds is because “cold” is just a term we apply to that classic set of symptoms (stuffy nose, cough, sore throat, etc.) which can be caused by a number of different viruses. A couple of these “cold viruses” are coronaviruses (in the same family as SARS-CoV-2, the novel coronavirus everyone’s talking about these days, which causes the disease COVID-19, but tamer). Some of the other “cold viruses” are adenoviruses. And, although they may cause similar symptoms, they’re much different molecularly, starting where it all starts, their genetic blueprint (genome).⠀

Adenoviruses (Ads) are double-stranded DNA viruses, whereas coronaviruses are single-stranded RNA viruses. Lengthwise, their genomes are similar. Ads are ~36,000 letters long. SARS-CoV-2 is ~30,000. But coronaviruses have an oily lipid membrane around them, which they acquire when they bud out from cells, whereas adenoviruses don’t have one of these membranes – instead they have a protein “shell” called a capsid. ⠀

All those differences aside, there’s one thing both families of virus are really good at – getting genetic info inside of human cells. Are you thinking what scientists are thinking?… If we could put genetic info we want into those viruses, the viruses would get tricked into delivering them – like in the movies where someone hides a secret message in an egg that the dairy man delivers. ⠀

If this “using-virus-to-deliver-genetic-info” thing could be done, it would offer up a huge range of possibilities, because stretches of that genetic info called genes are like recipes for making proteins. If you could get a virus to deliver those recipes, you could get the cells to make those proteins you wanted. Even if the cell doesn’t normally make that protein. So you could get cells to make proteins that they have typo-ed recipes for (i.e. mutated genes), treating (or even curing depending on the circumstances) genetic diseases – or you could get cells to make viral proteins they could learn to recognize as foreign and produce antibodies against. ⠀

The former (giving human gene to human to compensate for missing and/or mutated gene) is called gene therapy and the latter is called a viral vector vaccine or a DNA vaccine. I’m going to tell you a bit about both ⠀

quick note: The viruses used to these vectors are usually modified making them “non-replicating.” Normally a virus can use your cells to make more copies of itself which can then infect more cells, but a non-replicating virus can get into your cells but doesn’t multiply once there. Adenoviruses need a viral protein called E1 to replicate. So if scientists cut out the E1 gene, they can make a virus non-replicating. Preventing the virus from making more of itself can make sticking it in people safer (and makes it safer to work with in a lab), but it also poses problems if you want to make more of the virus. So in order to produce it, manufacturers infect cells in a dish/flask that already have a copy of the E1 gene. Those cells can thus make the E1 protein for the virus, the virus will replicate and then burst out and then you can collect that virus (which doesn’t have the E1 gene and thus can’t replicate) and use it. Okay – that aside out of the way, let’s go back in time to 1999…⠀

Adenoviral vectors were first used in a gene therapy attempt in 1999 at the University of Pennsylvania by a scientist named James Wilson for a genetic liver disease called ornithine transcarbamylase (OTC) deficiency. As the name suggests, the disease is caused by a lack of OTC, which is a protein enzyme (reaction mediator/speed-upper) the liver needs to break down certain molecules. ⠀

Patients with OTC deficiency have a problem with just that one gene for making that enzyme, so scientists hoped that by giving patients a good version of the gene, they could treat the disease. The disease can be deadly severe in some patients, but the teenager Jesse Gelsinger had a relatively mild case. He volunteered to be the one of the first humans to receive the therapy. And he received a LOT of it. 38 trillion viruses. Tragically, his body had a massive immune over-reaction and he died four days later. ⠀

Gene therapy attempts in people screeched to a halt for a while, but since the versatility was so promising, scientists continued to work on how to make safe gene therapy vectors. That initial therapy had used an adenovirus vector (Ad5 to be precise). Since most humans are used to that virus cuz it causes some colds, they caused the immune system to go into overdrive. So scientists started looking into other types of viruses that wouldn’t induce such a strong immune response. And they found Adeno-Associated Viruses (AAVs) – they’re only related to adenoviruses in being present in the same sample as adenoviruses when they were discovered in 1965 – biology-wise they’re unrelated. ⠀

AAVs are generally “safer” for gene therapy. And they’ve proven successful. The first FDA-approved viral-based gene therapy was Luxturna. It was approved in December 2017 for treating a form of inherited blindness. This was followed by Zolgensma – a gene therapy treatment for Spinal Muscular Atrophy (SMA) – in May 2019. ⠀

With gene therapy you *don’t* want an immune response to be generated, but with a vaccine you *do* (in fact, with many vaccines you actually have to include helper chemicals called adjuvants to help stimulate an immune response). So, while adenoviruses might not be good for gene therapy, could they be used for vaccines? We sure hope so!⠀

To explain this I first want to give a really quick, overly simplified, explanation of part of “adaptive immunity” – how our body learns to recognize and attack a virus or other invader using little proteins called antibodies that specifically bind to parts of that invader (such as pieces of viral proteins). Each time your body is confronted with a new viral infection, it has to develop antibodies specific to that virus. Your body can’t predict what viruses it will need to combat, so it uses a trial and error approach, mixing and matching constant and variable regions of antibodies and seeing what works. ⠀

A certain type of immune cell (antigen-presenting cells) engulfs viral particles, chops up their proteins and then displays those protein parts (peptides) on the surface of the cell. Antibody-making cells (which each produce a different antibody thanks to random recombination) then compete to see which cell has an antibody that can bind this antigen. The “winners” get selected for and more of that antibody is made. The antibody helps the body do things like flag infected cells for destruction, call for backup, and keep watch after the virus has been conquered in case it tries to return (which is why antibody tests can detect past infection). ⠀

That process of finding and making the right antibodies takes a while, so the goal of a vaccine is to pre-introduce your body to viral parts well before your body actually meets a virus, so you can build up antibodies against them without actually having to get sick. Classically, vaccines do this by using a weakened (aka attenuated) or killed (aka inactivated) version of the virus. Some other vaccines do this by introducing just individual viral proteins that have been made in a lab. But that takes a lot of work (expressing the protein, purifying it, etc.), so the idea of nucleic acid (DNA or RNA-based) vaccines is to get the cells to do the hard work – deliver the recipe for making the viral protein and let the cells make the protein. Much easier. But also not-yet-approved-to-work-as-a-vaccine-in-humans (the only commercial adenovirus vector vaccine is a rabies vaccine for wild animals) ⠀

As far as I can tell, there’s only one proven-effective human viral vector vaccines approved – for *any* disease *yet* – but some groups have gotten really close. And one such group is Oxford/AstraZeneca – they’re one of the frontrunners in the vaccine “race”, and one reason why is because they were able to “just” modify a vaccine they have in late-stage trials for one of the other coronaviruses, MERS (Middle Eastern Respiratory Syndrome). They stuck the gene for the SARS-CoV-2 Spike protein into a modified adenovirus. When the virus is injected, it infects the injected-person’s cells, those cells make spike protein, and the immune system makes antibodies against. it. Sounds great, and it can be, but this strategy isn’t without some potential problems. And here are a couple scientists are worried about.⠀

A lot of times with vaccines, the immunity doesn’t last forever, so you need a follow-up “booster shot.” Problem is, if the initial vaccination causes your body to develop antibodies against the vector, and not just the viral part, when you get a booster shot your immune system will attack the vaccine before it even has a chance to introduce more viral protein. This can also be a problem if you want to mix and match to use the same viral vector to deliver protein from a different virus to vaccinate against a different disease. ⠀

It’s kinda like your viral vector is a creepy white van – and you first use it to get people to learn to attack vampires. So you deliver some vampires – in the creepy white van that gives the people a heads-up that there’s something dangerous so they learn to attack a vampire if they see one again. And then you want to teach people to attack werewolves. So you do the same thing but stick werewolves instead of vampires in the van. Problem is, when people see the van, they attack the van itself without even knowing that a werewolf is inside. So if the werewolf shows up in some other form of transportation, the people won’t know it’s dangerous. ⠀

Because adenoviruses are so common, many people have antibodies against various parts of various ones. And this can cause a similar creepy-white-van problem – your body may have already learned to recognize the van before you try to deliver the vampires, so your body never even meets the vampires and won’t know how to greet them in the future. ⠀

This can be seen with failed HIV vaccines and with CanSino’s Ebola vaccine. CanSino got Chinese approval for their Ad5-based Ebola vaccine in 2017, but only for emergency use and they never actually proved it would prevent Ebola infection; additionally, the antibodies made after injection didn’t last very long, potentially because of pre-existing immunity to Ad5 (that white van problem).

CanSino Biologics started human trials of an Ad5-based Covid-19 vaccine in China in March (the first company to take a Covid-19 vaccine to clinical trial). It’s currently in Phase III studies. Because of that pre-existing immunity problem, doses given will likely need to be higher which could cause more side effects ⠀⠀

So scientists have tried a few different strategies to get around these problems. One strategy is to modify the adenovirus genes to camouflage it a bit better – like giving it a paint job. Biochemically, they can do this by deleting or altering more adenovirus genes. A company called ImmunityBio is using this approach They’ve also included a second coronavirus gene, N (which has instructions for making the Nucelocapsid protein which normally functions to coat and protects coronavirus RNA and is also very immunogenic (capable of prompting an immune response). It’s currently in Phase 1 trials, checking for safety. 

Another strategy is to turn to different vans all together – less common coronaviruses, like Ad26, which Johnson & Johnson acquired and made an Ebola vaccine from (it received approval in Europe in July 2020, and I think this makes it the first proven human adenoviral vector vaccine)

Trials of the coronavirus version of it were halted for a bit because there was an “unexplained illness” in a participant that turned out to be unrelated to the vaccine, but phase 3 testing is back on track A benefit of J&J’s vaccine (if their strategy works) is they’re just giving a single shot (no booster). This makes production and distribution a lot easier, and it avoid the white van problem, except…

Ad26 is a human coronavirus, and lots of people in sub-Saharan Africa & Southeast Asia may have preexisting immunity to it, so some groups are looking to viruses from other animals. And they found some promising ones in chimpanzee poop. This is the strategy taken by Oxford with their ChAdOx1 vector (name comes from CHimpanzee ADenovirus OXford). They took a chimp poop adenovirus, and cut out the E1 – they also removed a gene called E3 which they don’t need in order to free up more space in the genome for sticking in a protein of interest – in this case the SARS-CoV-2 spike protein. They teamed up with AstraZeneca for production and distribution so now you usually hear this vaccine referred to as the Oxford/AstraZeneca vaccine. And I think the ChadOx name has been replaced with “AZD1222.” Sorry chimps! Similarly to J&J, they had to halt trials to make sure an adverse event was not vaccine-caused (it was not) and they’re continuing their Phase III trials 

In addition to viral vector vaccines and mRNA vaccines, you have the more traditional vaccines in the works. These other strategies include:

  • inactivated (killed) virus⠀
  • weakened (live attenuated) virus⠀
  • viral proteins (aka subunit vaccines)⠀
  • virus-like particles (VLPs)⠀

And you can learn more about them here: ⠀ 

More on how vaccines are tested: ⠀

More coronavirus-related posts: ⠀

Leave a Reply

Your email address will not be published.