Cannabinoids and COVID-19

“The phone has been ringing every couple of minutes! All these years we’ve been doing research we think is important and doing our best to get published – but very rarely does a project catch on so globally! It’s been amazing. We are currently on 402,000 views for the paper.”

Earlier this year, Richard van Breemen and his team published a paper stating that cannabinoids were capable of blocking cellular entry of SARS-CoV-2 (1). As you might imagine, people were excited. Very excited. In fact, you may recall reading headlines that claimed smoking weed stops COVID-19. These claims were outlandish – but where there’s smoke, there’s fire (sometimes). 

The team’s research focused specifically on the role of cannabinoid acids – CBD-A, CBG-A, and THC-A. And though smoking these compounds won’t do much (doing so would convert them into forms that are not known to prevent cellular entry of the virus), dietary supplements could be an effective route of administration – and this is what the team decided to focus on. Using affinity selection-mass spectrometry, the group were able to determine that the three cannabinoids were potential orthosteric or allosteric inhibitors of the viral spike protein. To further test their efficacy, live virus was incubated with 25 μg/mL of either CBD-A or CBG-A and then used to infect Vero 6 cells. After 24 hours, they found an absence of SARS-CoV-2 viral RNA in the cells. Importantly, they found that CBD-A and CBG-A blocked both infection of the original live virus, as well as variants of concern, including the B.1.1.7 and B.1.351 strains. We spoke to lead author, Richard van Breemen, to find out more. 

Your paper received an impressive reception. Were you surprised by that at all?
 

I was certainly surprised by just how positive the response has been. I was expecting at least some pushback, with people saying the research doesn’t really mean anything or wondering why it was getting so much attention. But people haven’t really been criticizing it. I’ve even had phone calls from people to say thank you because they now understand why they didn’t get sick after attending a gathering where everyone else got COVID – it’s because they were taking some sort of cannabinoid extract. I’ve heard a lot more of these anecdotal stories since our paper came out. 

You’ve been studying hemp for years. What made you turn your attention to COVID-19 specifically?
 

I’ve been interested in natural products as sources of drugs my entire career – and using mass spectrometry to explore these compounds. This forms the basis of the work I do at the Global Hemp Innovation Center at Oregon State University, where we are looking at the medicinal applications of hemp products including cannabinoids and other hemp secondary metabolites. 

There were two reasons I turned my attention to COVID-19. Firstly, as we all know, when COVID hit many laboratories had to shut down. The exception to this at Oregon State University was if you were doing work on COVID-19 – so it just made sense to start looking at the virus. At the same time, I’d started to recognize a pattern emerging in the US COVID deaths by state in 2020 and 2021; the top 10 states with the lowest death rates were those that had legalized marijuana. Conversely, those with no legal form of cannabis had the highest death rates. My wife – an expert in health data and informatics – has since made me well aware that there are too many variables to ever be able to confidently draw conclusions from such sparse data. But I can’t deny it having an influence on my decision to turn our research focus onto COVID-19! 

We knew the mechanism through which antibodies enveloped the spike protein and prevented it from interacting with the immune system, so we reasoned that a small molecule from a natural product source could potentially have the same effect. We began searching for ligands to the viral spike protein, and we decided to start with dietary supplements as they are already being used in humans – meaning if we did find something, they could potentially be approved for use very quickly. 

Using MS for natural product drug discovery is something you’re clearly passionate about. How does this technique enable the sort of research you’re doing? 
 

A long time ago, combinatorial chemistry became the latest and greatest innovation in drug discovery, but we recognized there was a need for higher throughput methods of screening and testing new compounds to see whether they are active in any particular pharmacological assay. We responded to this need by taking advantage of mass spectrometry’s sensitivity and specificity. While the pharmaceutical industry went in the direction of synthesizing one compound and testing it individually in assays, we took the approach of pooling a combinatorial library into mixtures of around 3000 compounds each and assaying them simultaneously using a technique we helped pioneer called affinity selection MS. By the early 2000s, we realized that this same approach could be applied to natural product mixtures. We’ve been doing this type of research ever since.

How has MS evolved since you began working with it?
 

I was first introduced to MS as a graduate student in Catherine Clarke Fenselau’s lab at Johns Hopkins School of Medicine in the 1980s. And Catherine introduced MS to her PhD mentor, Carl Djerassi – who wrote some of the early books about how we use MS for steroid structure determination and the like. So I’m sort of the grandchild of Carl Djerassi and the era of organic MS, and I represent the (current) era of biomedical MS. 

I taught spectroscopy for many years to medicinal chemists – I tackled MS and infrared spectroscopy (IR) and my colleagues taught things like nuclear magnetic resonance spectroscopy (NMR) and X-ray crystallography. What we learned early on was that two of the following methods could be used for the complete structural determination of a new molecule: either MS and NMR, or MS and X-ray crystallography. So for me, MS has always been a fundamental part of natural product structure determination. 

In recent years, we’ve seen an explosion in the applications of MS – and most of these were unimaginable 40 years ago. I remember arguments about whether MS-based peptide sequencing could ever compete with approaches such as Edman degradation. Today, MS dominates this field. I remember submitting grant proposals in the 1980s and 1990s to use MS for sequencing proteins. One of the critiques I got back was that MS would never be capable of determining the structure of large molecules like  proteins. I ended up backing away from that area – today we call it proteomics! I received similar critiques of my grant proposals for using MS to show how a small molecule would prevent cell entry of a virus – I won’t be backing down this time.

Could you tell us about the magnetic microbead approach you used in this study?
 

We often hear about finding drugs from combinatorial libraries as finding the needle in the haystack. Think of the mass spectrometer as the magnet that finds the needle in the haystack. In the case of MagMass (Magnetic Microbead Affinity Selection Screening), this can be taken quite literally. MagMass involves tethering the pharmacological target (for example, a receptor) to a magnet microbead, incubating that with a mixture of small molecules, and using a magnetic field to pull the beads to the bottom of a well plate. You are then able to wash anything away that isn’t bound, and inject the bound molecules into the mass spectrometer for screening. 

The great thing about affinity selection MS (AS-MS) is that we can discover both orthosteric and allosteric ligands – in other words, ligands that don’t necessarily bind to the active site. Once we find the ligand, we don’t know how it binds. We usually add a known orthosteric ligand and see if it competes with the new molecule for this site. If we are testing for inhibition of an enzyme, we can carry out a functional enzyme assay to determine whether it’s an allosteric inhibitor. We think that CBD-A and CBG-A may work better when combined, and this is something we are going to be investigating further.

Another key benefit of AS-MS is speed – the bond between ligand and receptor is non-covalent, meaning the longer the complex is isolated, the more ligand is lost. Ultrafiltration takes a few minutes, size-exclusion chromatography takes a little longer, but MagMass can be done very rapidly. It’s the fastest method we’ve found to separate the bound from the free and then isolate the ligand complex with the macromolecular target.

Any surprises along the way?
 

It was surprising that CBD-A, CBG-A, and THC-A were the best ligands for the spike protein. There are so many cannabinoids, but hardly any of them bind to the spike protein – for example, we know that CBD and CBG in their decarboxylated forms do not block cell entry. It was very exciting to find that not only could these acidic forms bind, but they could function to block infection at the cell entry step. Our live virus testing also showed that there was equal activity against two variants of the virus that had just become available – that was great news. 

You mentioned some of the challenges around research in cannabis – like being able to get your hands on enough THC for your research. 
 

Extremely small levels of controlled substances (like THC-A) can be purchased for analytical qualitative analysis. So, this is what we did. As many will know, the legal difference between marijuana and hemp is the amount of THC, meaning THC levels are very low in hemp. Our initial experiments showed THC-A was binding to the spike protein, but unfortunately THC-A is a controlled substance in the US as it can be easily decarboxylated to form THC. We would have needed to apply for a special license to get the levels of THC-A we needed for live virus testing, and our main focus was to get the experiments done as quickly as possible so we could share the results and get the ball moving on some viable treatments. 

We are really hoping we can move to establish clinical trials very quickly – whether we know it or not, we have been doing human testing of CBD-A and CBG-A for a long time already, so they have a good safety profile. Hopefully this will save some time on the usual few years it takes to establish a safety profile in humans. 

Is THC-A something you are looking to explore in the future? 
 

Absolutely, we’d love to continue our studies with THC-A and see if it’s effective at preventing cell entry as well. Every successful experiment brings up new questions to explore. There are so many experiments we’d love to do off the back of this paper – we are only constrained by regulations and funding. As we overcome those hurdles, I am certain we will continue to publish more exciting work.