PAC-MAN vs. COVID-19

Image credit: “Pacman Games” by Piotr Siedlecki is under a CC0 Public Domain license.

This study was conducted by researchers from Stanford University.

This post was reviewed by Dr. Phil Lange, one of our subject matter experts.

TL;DR: Stanford researchers investigated using PAC-MAN, a technique that destroys virus genes, to attack SARS-CoV-2. PAC-MAN was shown to destroy SARS-CoV-2 fragments and a strain of swine flu. Database research showed that PAC-MAN could theoretically be used to target a range of coronaviruses.

The paper we’ll be demystifying can be found here, if you’d like to follow along!

Pac-Man is the beloved yellow circle that roams the mazes of his arcade game, eating pellets and scaring ghosts. Ever the little soldier, Pac-Man has been enlisted in the fight against COVID-19 in his own way.

PAC-MAN, or Prophylactic Antiviral CRISPR in huMAN cells, is a technique that targets the genetic material of viruses. It does so with CRISPR RNAs, or crRNAs, small molecules that target specific parts of the virus’ genetic material, marking them for destruction. Then, a protein called Cas chops up the tagged section of genetic material like the yellow circle himself, effectively killing the virus. It’s like wrapping a pill in ham so that your dog knows to eat it.

A team at Stanford university has turned this technique against SARS-CoV-2, using a specific Cas protein called Cas13d.     

crRNA works by binding to specific parts of a virus’ genetic code using its own complementary sequence. To make crRNA sequences against SARS-CoV-2 genes, the researchers first compared the genome of SARS-CoV-2 with that of its relatives, SARS-CoV and MERS-CoV, They found two regions common to all three that contained the RNA-dependent RNA polymerase (RdRP) and nucleocapsid (N) genes. RdRP is necessary the make more viral genomes, while N encodes the proteins that package its genes, making both vital to SARS-Cov-2 and excellent targets for destruction. The team digitally made a collection of all the crRNAs that could target them, then ruled out any crRNAs that also target human genes. They ended up creating 40 crRNAs, with 20 crRNAs targeting each viral gene. The team fused a glowing marker, green fluorescent protein (GFP), to SARS-CoV-2 fragments to track if pieces of SARS-CoV-2’s genetic material were targeted by these crRNAs.

Think of it as testing an entire deli’s variety of cold cuts for a bunch of Cas dogs in need of their medication. Thanks to GFP, each pill now glows a Ghostbusters green. Testing with one type of ham at a time, we could tell how tasty each one was by looking at the leftovers, characterized by the amount of green pills left glowing on the floor. Likewise, anything that destroys the SARS-CoV-2 fragment would also turn off the GFP glow and measuring the reduction in fluorescence tells us how much was eliminated.

They started by pinpointing the regions of the SARS-CoV-2 fragments that were the most vulnerable to Cas13d, as the researchers didn’t have access to the full virus. Human lung epithelial cells were made to produce Cas13d and a pool of four of the crRNAs, essentially a group of four crRNA that all target the same thing. Pooling four crRNAs together prevents escape via single mutation where a crRNA would target. Imagine stacking a bunch of cold cuts together in case one of them fell off.

The cells were then given the glowing fragments of SARS-CoV-2. A crRNA pool labelled “G4” ferociously devoured RdRP, reducing the GFP by 86%. As such, it was singled out as the ideal candidate to attack RdRP. Likewise, a pool of N-targeting crRNAs named “G6” caused a 71% decrease in GFP, and was selected as the best pool against the N region. Quantitative real-time PCR (qRT-PCR) analysis, another method of gene analysis, corroborated this story.

Next, the team delivered the SARS-CoV-2 fragments by lentiviral transduction, a technique that uses a virus to deliver genetic material to better mimic the conditions of a real COVID-19 infection. This time, only a few crRNA pools were tested: G4 and G6, as well as the third best pool, G5, and a less efficient pool, G1. Again they found that G4, G5, and G6 were all able to target their fragments. Additionally, there was no difference observed in the health of normal cells and those adjusted to produce Cas13d and crRNAs, suggesting that it could be a safe treatment.

The team didn’t have access to live SARS-CoV-2 virus to prove the effectiveness of their treatment on a whole virus, so they tested their strategy on a strain of H1N1 influenza A virus (IAV), or swine flu, as it targets respiratory tract epithelial cells like SARS-CoV-2 and has been stopped by Cas13b in previous studies. Unlike SARS-CoV-2’s single continuous genome, IAV has its genome contained in eight separate segments.

The team made 48 crRNAs for IAV in largely the same way as done for SARS-CoV-2, this time targeting the ends of the eight segments. The ends are essential for forming new IAVs and interfering with one segment end messes with the other seven. Additionally, these segment ends are common to many strains, giving this treatment a broad potential range.

Each genome segment was again tagged with fluorescent proteins and was targeted by a pool of six crRNAs. The pool targeting segment 6 (S6) was consistently highly effective in targeting the gene for neuraminidase, a protein helps IAV travel around the body. The crRNA pool for segment 4 (S4) was also consistently moderately effective in targeting the gene that encodes hemagglutinin, the protein that lets IAV stick to our cells. Further testing of the S6 pool showed that it could also target the S4 fragment. The same trend was seen when the amount of virus was lowered, suggesting that Cas13d PAC-MAN could be used to stop an infection before it even truly starts.

Inspired by this, the team investigated any crRNAs that could target different strains of IAV at once and found one, crRNA-S6g2, that could target 32% of them. As such, using these crRNAs could allow Cas13d PAC-MAN to be a general strategy against IAV.

Seeing how Cas13d PAC-MAN could potentially be a broad-strokes targeting strategy, the team looked into using it on a broad range of coronaviruses. After all, SARS-CoV-2 isn’t the only one out there; MERS-CoV and SARS-CoV are also well-known coronaviruses, and more than 3000 other coronavirus genomes have been sequenced. By searching coronavirus genome databases, the team found that a pool of 22 crRNAs could target all known coronavirus strains without error. Narrowing it down even further, they found a collection of the top six RNAs, termed PACMAN-T6, that could target all human coronaviruses while broadly targeting animal coronaviruses.

All that aside, testing Cas13d PAC-MAN on a bunch of cells is one thing; getting it to work in a patient is a different set of problems entirely. Altering the genetics of a patient’s cells to make Cas13d and crRNA pools can carry a lot of risk, as with any experimental genetic therapy. Using a method that only temporarily gives these effects could lower the associated risks. Finding a way of safely deliver Cas13d and crRNA into cells will be the first hurdle to overcome. Even if the technique is deemed safe, getting a patient’s cells to produce the exact quantity needed to destroy SARS-CoV-2 has its own host of difficulties.

Regardless, Cas13d PAC-MAN is a very promising strategy against SARS-CoV-2. Someday soon, we may have Cas13d PAC-MAN firing down our bloodstream, chomping down viruses.

And that’s game over for COVID-19.