One month when the world changed

What a difference one month makes. When we posted our first blog post dedicated to the SARS-CoV-2 and sheepishly tried to bring attention to the virus (by piquing your curiosity with an interesting story on the origins of the Wuhan coronavirus), there were just barely over 100,000 cases in the world - mostly confined to Asia and hence vastly ignored by the rest of the world, who were hoping that the outbreak would be contained in that geographical location. Now we are approaching 1.5 million global cases, and the entire world has been infected with more than a third of the entire global population requested or forced into some form of social isolation.

Some people still think that this is an over-the-top attention reaction to the pandemic and it is overblown. Indeed, the world cannot remain in a lockdown forever, and no one can predict the economic aftermath to our global economy. The current goal is to slow the spread of the virus so that those who need to be hospitalized for treatment to potentially a fight for their life, have a chance at survival because we actually have the proper resources available to them. “Flatten the curve” concept.

In the meantime, there are still no specific treatments or a vaccine for COVID-19, and the world is rushing to try to understand if there are any drug interventions to either prevent or treat the COVID-19 disease. This is the focus of this article.

Any preventive drugs for COVID-19?

Perhaps this article is massively overdue. It is in fact a continuation of what we touched upon in our very first foray into studying the structure of SARS-CoV-2, when we mentioned one study which looked into potential compounds that could inhibit viral infection. The need for this story has since massively expanded, so in this post we will take you so deep that we peer into atomic structures and traverse the landscapes of biological molecules (like we never dared before) because we believe they offers views that are beautiful if you have the imagination and stamina to observe them.

But before we go on, let us first announce some good news.

Here are our favourite chemicals that we found that might have protective capabilities against COVID-19! We mention these because they already seem to be used everywhere. Then we explain everything so be ready for another journey down the scientific rabbit hole (aka, deep molecular hydrophobic pocket – it is a joke… for later, promise!)!

• (-)-Epigallocatechin gallate

Don’t let the wild name fool ya! We missed it the first time we studied this paper, but this is actually the famed EGCG compound of green teas, touted for its healing properties, especially anti-cancer. One of our favourite sources on this topic comes from CTOAM, a company dedicated to fighting cancer and with which Merogenomics has closely collaborated in the past. We produced promotional/educational material for cancer patients about DNA testing and precision oncology (one of those target groups for Merogenomics approved DNA tests), and we are still in frequent contact. So here is their info on green tea in cancer prevention.

• Theaflavin 3,3'-di-O-gallate

This compound is from the same source as the one above: Camellia sinensis, but the compound above is mostly seen in green teas, whereas this one dominates the black teas.

• Folic acid

Yes, this is the same compound used by pregnant women to minimize chances of developing certain birth defects in their baby. Can men take it? It appears so, but we did not investigate this ad nauseum so we cannot be promoting this with absolute certainty to men. But if you are a pregnant woman, you are requested to take it anyway and it is readily available and cheap.

• Aspartame

Wait, what? This is the artificial sweetener found in diet soda drinks. So diet soda drinkers enjoy!

• Procyanidin

In the study reporting it, the source of compound was listed as Vitis vinifera, basically the common grape you see being used in Europe to make wines. So does that mean you can find this compound in wine? Yes! Now you can enjoy your social isolation with a glass of wine in hand, and feel good about it! But there are even better sources of procyanidins. Oh dear, the ever-expanding (or perhaps continuously contracting) universe, it gets way better! One of the highest sources of procyanidin is chocolate! Women had it right all along! Maybe the only reason why we should not always listen to them is so that our doctors can actually keep their jobs (kidding, we love our doctors!). Another fantastic source is apples, but there is quite a variation depending on what kind of apple with Red Delicious and Granny Smith seeming to be the way to go. Of course, there is no problem in having all of these foods at the same time! ;)

However, bear in mind, there are no direct studies in human subjects to confirm these claims, and these suggestions come from molecular modeling that we will explain in painful (not!) detail below. There are so many others, as you will see, that it is an impossible amount to investigate fully. If you are an expert in nutraceuticals, the references below might interest you and perhaps you can let us know what else you uncover or you like the most that should make the list. Overall, these chemicals were pretty benign options that caught our attention, and are already super common. So let us begin our story.

Types of potential anti-COVID-19 measures

As a consequence of the rapid spread of COVID-19 disease around the world, the entire world’s scientific community has gathered its focus like never before to look for potential remedies against this devastating infection. It seems that every single known chemical compound in our databases is being reassessed for: the ability to interact with any of the SARS-CoV-2 viral proteins whose structure are known, or the possible inhibition of the interphase of the interaction between the viral proteins and the human proteins which act as receptors for these virus particles.

First, let us talk about some divisions that we can use for either treatments or prophylactic measures. We can divide these into three types:

• Chemical compounds
• Biologics
• Vaccines

The chemical compounds category seems self-explanatory - these are small molecules that have a specific structure (that have a certain three-dimensional shape, it’s all about the shape!) which happens to fit somewhere (in some crevices and grooves) of our large biological molecules (proteins being just some of them), and therefore affecting how these large biological molecules function. Think of the classic “lock-and-key” analogy. More appropriately, think of those perfectly shaped ear plugs that you can have custom made for yourself (if you can afford that, you should probably also sequence your genome!). In this case the ear plug is the drug, and your ear is the biological molecule that is targeted and affected (in this case, your ear no longer delivers outside noise to tickle your brain).

Biologics is a blanket term to refer to higher order biological molecules that are produced by living cells. Antibodies are a perfect example. In theory we can synthesize these structures (a process termed 3D bioprinting) but we cannot yet match what nature can deliver either in capabilities or because of cost. Thus we need other humans or animals to produce biologics, or even bacterial cells that have been reprogrammed for such purposes (by altering their genome of course).

Vaccines are in a bit of a different class, as they are modified infectious agents that end up mimicking the ones found in nature, which then produces a mock infection in our body so that we can build up our immunity against the real, natural threat. In other words, we use vaccines to command our bodies to produce those biologics (in the form of antibodies) ready for action to fight the infection when it arrives (sometimes this can also be done post infection). Vaccines have a variable efficacy - the efficacy of vaccination is quite dependent on individual biology (you will have to check out the next post to find out why).

• If we look at the compounds that are being investigated, we can divide these into five groups:
• Repurposing existing approved medications which could be used for inhibiting COVID-19 infection directly
• Novel compounds used against the viral components only or viral components involved in interactions with human molecules
• Novel compounds acting on human molecules independent of the virus but which could be use for inhibiting infection (in other words, drugs that help to properly employ our immune system)
• Repurposing existing compounds that were shown to be inhibiting against SARS or MERS viral epidemics (this could be a subcategory of any of the above three points)
• Creation of brand new chemicals from scratch using the understanding of the physical shape of our designated targets (it’s all about the shape)

Besides compounds that modulate our immune system, compounds targeting the viral system could be acting on one of the following areas:

• Blocking the virus’ binding to human cell receptors
• Preventing the synthesis of viral RNA
• Targeting critical enzymes of the virus to block virus replication
• Inhibiting the virus’s self-assembly process

So where do we start?

In terms of using existing drugs, such medications have already been previously rigorously tested for human toxicity, efficacy and dosages, so finding new methods for using them means they are of especially high interest because we could start using them right away without lengthy testing to see if they are safe. This is exactly what the first move was in fighting SARS-CoV-2.

Here are drugs already used for treatment of COVID-19 first started in China where the pandemic commenced.

This list could grow at a rapid pace, and for a medical guideline of COVID-19 patient treatment in a hospital setting, this is by far the best document we have come across so far, from Belgium. In physicians’ general opinions on the disease and its impact, here is a survey of over 6000 doctors about the response to COVID-19. In the meantime, there are over three hundred registered clinical trials related to COVID-19. Absolutely a stunning number! You have to understand that each one of these could be millions of dollars in cost because medical experiments are very expensive to execute.

And along the way, there is an equally stunning search in expanding our candidates.

Billion of compounds at a time?

In one example, a billion compounds were looked at as potential inhibitors of COVID-19! Just a mind boggling quantity. Any time you hear the number billion, it is truly a staggering amount. When you get to a trillion, well, that is not even comprehensible by any self respecting mind (which is why you may excuse bankers for tossing about trillions of future financial liabilities since they probably do not comprehend it either).

And that was just one publication looking at the compounds for inhibiting only one of the viral proteins. Only just one! Those authors made a short list of the best 1000 compounds. For sake of simplicity we guess? The viral protein being targeted is called main protease. This is not the coronavirus protein that gets the most attention. That honor belongs to a SARS-CoV-2 spike protein (which the virus uses to interact with human cell receptors but more on that later). But main protease is also a very important mini-robot because it is a fancy cutting machine.

Because a virus is so tiny, all the information encoded for its replication and assembly has to be as small as possible and there are different ways to achieve this. One of them is that when the virus genome is translated into proteins (or basically these tiny molecular robots), the viral genome is actually translated into bunch of fused proteins in one go. What we mean is that all the proteins are all linked together as a one big messy polyprotein. These have to be cleaved apart into individual entities and that is where the main protease enters the picture. You knock that puppy out, then there is no viral replication. So it has been getting serious attention.

We really do need machine learning these days because how can people parse through so much data? This field is exploding so rapidly, this small subcomponent of SARS-CoV-2 scientific literature alone is becoming impossible to follow.

To pick some interesting highlights from the literature though, another group of researchers made copies of 26 viral proteins that are encoded by the SARS-CoV-2 virus and scrupulously went on a hunt to discover which human proteins actually managed to physically interact with these viral proteins. With this simple sentence, we have embodied countless hours of human endeavour. In this way, they identified 332 high quality SARS-CoV-2-human protein-protein interactions. And that was just their first step. Then they used these interactions to screen approved medications to see which of these medications could be disrupting such interaction between the virus and the human proteins. They found 69 drugs targeting 67 different human proteins. This is an entire arsenal of possibilities to try to figure out how we could be manipulating our biological response to this virus. The question is, when are we going to find the time to do all of this to discover which ones really work?

You literally would have to be testing these candidate drugs one by one in cell cultures to see which ones inhibit viral propagation. This is exactly what one group did, testing already FDA-approved drugs for SARS-CoV-2 inhibition, and they found 17 candidates. Among them, and what caught their interest, was hydroxychloroquine and chloroquine compounds that have garnered so much attention lately, authors calling attention to its potential use in human treatment. Currently hydroxychloroquine is rated as the most effective COVID-19 therapy by doctors around the world.

Now let us get to the literature that generated our top picks at the start of the article. It came from couple of published articles by the same research group from China.

Molecular deep dive

Here we will begin our molecular deep dive and so a tiny review is necessary. Recall from our previous coronavirus article that SARS-CoV-2 encodes a spike (also just termed S) protein that binds to ACE2 receptors on human cells. The part of the spike protein that interacts with the human receptor is referred to as the receptor binding domain, or RBD (it’s the only way to be). But the spike protein’s dirty little secret is that it is also important for virus-human cell membrane fusion (basically, a human cell ends up swallowing the virus).

In the first article of the two-part series, the authors proposed possible reasons why SARS-CoV-2 is more infectious than its predecessor SARS-CoV:

• SARS-CoV-2 RBD interacting with the human ACE2 receptor may have different conformations than SARS-CoV RBD
• SARS-CoV-2 spike protein can bind to additional receptors besides ACE2
• SARS-CoV-2 spike protein is more easily cleaved by host enzymes which allows the virus to more easily fuse with a host cell membrane

And they went nuts investigating that last point. So will we.

For you see, for the virus to effectively infect our cells, the spike protein has to be cleaved into two components: the human receptor-bound S1 subunit and the S2 subunit participating in the membrane-fusion. No cutting, no entry. This is why spike protein is the real star of the show, and the protein that determines, to some extent, the host and tissue range of the virus (that is referred to as viral tropism). In order for viral invasion of our cells to happen, the virus relies on this cleavage to be performed by our own proteins: the human host proteases. Yes, as we mentioned above, the virus encodes its own proteases, but they are used later on in the process once the virus has entered the cell and its genome is translated (again by hijacking our human host proteins to do so). This very important cleavage occurs at the S1/S2 cleavage site and there are number of host proteins that can do this type of cutting for the virus. The cleavage of a spike protein is absolutely indispensable for the membrane fusion of a S2 subunit to take place, while at the same time it promotes structural rearrangements of the RBD of the S1 subunit that make it more capable to interact with the human ACE2 receptor (this is why we call these players tiny molecular robots – they transform). So with one host snip, the virus kills two birds with one stone.

For now, we will also add that even within the RBD itself, only a portion of it is actually dedicated to interact with our host receptors. Below is our favourite representation of the spike protein demonstrated so far, because it shows you which sections of that spike protein are conserved (meaning they are immutable because they likely would destroy the entire architecture, and basically kill the virus), shown in purple, versus those that are hyper mutable, shown in teal.

Adapted from Walls AC et al. 2020. Cell pii: S0092-8674(20)30262-2

Observe from the image, the top of the spike protein - which has the section involved in binding the human ACE2 receptor - is the most variable among different coronaviruses spike proteins. That makes sense because the ability for this area to easily mutate means that the virus can constantly evolve to adapt to new receptors.

Ignore the fusion peptide as we are still following the simplified version of everything. Although we have to come clean and tell that you that what you are actually looking at is not a single protein, but three identical proteins put together! Yes, the spike protein is a trimer, meaning the same protein is so intelligently designed that three identical pieces will fit together to make up the final product (individually they are useless). Hence the tri-symmetry you can see from the top view of the spike protein. If you want to get super technical, because the trimer is composed of three identical units, it is called homotrimer. 

The representation you see here is referred to as space-filled model, or our way of pretending what the proteins would physically look like. In reality there is actually pretty much nothing there, but our brains are so adapted to perceiving the world in a physical manner they have a hard time understanding that it is a form of illusion. In reality, these shapes and contours you see as a physical space actually designate the confines of electrons of all the atoms that make up this protein. But these electrons appear and disappear in the space around the atom nucleus so fast that one moment they are here, another moment there and so on, in a manner so fast that in the end it creates a certain electromagnetic charge that dictates how different groups of atoms are attracted or repulsed to each other. So in reality, what we define as matter, is probably closer to a pulsating cloud, ever so imperceptibly changing shape on account of these electrons doing their crazy “appearance and disappearance” magic show around each atomic nucleus. We recommend you try that as a form of compliment next time. It really works among scientists!

This pulsating leads to the final shape of best probability and as you can see it is complex and intricate with many bulges and crevices. Every single bulge can in theory stick itself into another crevice elsewhere, if they spatially fit together and have complementary charges (meaning the atoms are not repulsing each other). Every crevice could be a welcoming sanctuary for whatever might fit in there, provided that they are not repulsing each other with their charges. As you will soon see, that is lots of playground to play with, as in theory, drugs could be snuck in anywhere, and antibodies could be binding to anything it all just has to have the right three-dimensional fit and complimentary charges. The closer these criteria are met, the happier and stronger the interaction shall be.

But wait there is more!

What you see here is also a naked version of the spike protein. Yes, this emperor with a crown forgot his clothes. In reality, this protein is also adorned with additional chemicals, basically different types of carbohydrates, or sugar molecules if you will, linked together in intricate chains, so the final three-dimensional space is even more complex. This is why you might come across the SARS-CoV-2 Spike protein referred to as glycoprotein. It is one way that viruses try to avoid detection by our immune system.

Here is one super cool way to actually show you this.

Glycosylated SARS-Cov-2 Spike Trimer. Biantennary LacNAc N-glycans (18 per protomer) are shown in dark magenta. Protein is from Swiss-Model (based on 6VSB), glycans built using https://t.co/lX8uga8Tzz. #glycotime#COVIDー19@SWISS_MODEL@glycampic.twitter.com/6iE8CrzjZc

— Oliver Grant (@Olivercgrant) March 27, 2020

Furin: an accidental traitor

As we noted in our previous post, what makes SARS-CoV-2 so unusual is that its spike protein has extra few amino acids inserted in its sequence (amino acids are the building blocks of proteins) which happens to be one of the best restriction sites for another of our own proteases. And not just any protease but one of the most ubiquitously found proteases: furin. These suckers are found almost everywhere in our body. SARS-2-CoV-2 could not find better sucker to do its bidding than that. As a consequence, the spike protein is far more likely to be cleaved, promoting the viral entry into our cells.

This is believed to be the magic sauce behind SARS-CoV-2 infectivity (although the virus could still have other tricks up its sleeve, and could quite possibly be interacting with additional receptors as well). This might also help explain why we see additional organ failures with SARS-CoV-2 infection, because as mentioned in our last post, ACE2 receptors are also observed in many tissues. So a bad combo all together, ACE2 everywhere and furin everywhere, and a virus that uses both to infect these cells.

So the authors thought that combined administration of drugs that target all the different SARS-CoV-2 proteases, including furin, could be an effective therapeutic strategy. They set out to search for potential furin inhibitors. They scanned a database of over 4000 compounds and measured in predictive models how these compounds would perform as inhibitors of furin. This is how we got the top three chemicals mentioned above as potential prophylactic measures against COVID-19 - the two tea compounds and the folic acid.

Below is a picture of the structure of furin protease with a couple compounds modeled onto it, one of which is the EGCG compound. This time however, the atoms are color coded instead for their overall charge. So the more red the color, the more negative charge of that molecular environment is. The more blue the color, the more positive the collective charge of that particular shape. You can see that a big chunk of furin is pretty negative.

Adapted from http://www.chinaxiv.org/abs/202002.00062 preprint

This super negatively charged area occupied by the compounds is in fact the active site of the protein, or the place where the protein does the cutting. The other proteins that will go inside this active site and be cut, are referred to as substrates. If you guessed that the substrates of furin that would go into this active site would probably have to be pretty positive, then you guessed right. The substrate cleavage site must be “arginine-X-X-arginine ↓" where X is any amino acid and ↓ is the cleavage site. Arginines are quite positive in their charge. Furthermore, if that first X position is either lysine or arginine, the cleavage efficiency is improved 10-fold. If you also guessed that in the SARS-CoV-2 spike protein it is either one of those, you got it right again. It is another arginine.

Now what else you see in that image is a known furin substrate in purple, and the EGCG tea compound in yellow, overlaid on top of each other (they would not be able to occupy that space at the same time). You can right away appreciate how it blocks the active site of furin, and why it could act as a good inhibitor.

All you tea lovers rejoice!

What about that folic acid then?

All the pregnant women rejoice?

Perhaps not just them. Folic acid is also known as vitamin B if that puts you at ease. Another group of scientists that proposed folic acid as a safe drug to prevent or treat COVID-19 were from Iran, another hard-hit area early on in the global pandemic (when WHO was still pretending otherwise). Folic acid is readily available in many foods, and among its functions is the inhibition of DNA damage associated with cancer development. When these authors did folic acid modelling into the active site of furin, they found many favourable interactions between the atoms of folic acid and those of furin.

Adapted from https://chemrxiv.org/articles/The_Role_of_Folic_Acid_in_the_Management_of_Respiratory_Disease_Caused_by_COVID-19/12034980 preprint

In the image above, we are seeing the entire furin protein, but this time presented in a ribbon model, or basically just following the direction of the chain of all amino acids. Remember, when proteins are synthesized, it is through the painstaking gluing of one amino acid to another, over and over. An amino acid is just a small chemical compound of a specific design, and our cells use about twenty different kinds. The information on which amino acids are to be pieced together come from RNA which itself is like a xerox copy of the master sheet, which is our genome, the DNA. SARS-CoV-2 genome takes a shortcut in that its genome is already an RNA, ready to be used to produce its viral proteins to produce more of itself. Once you make this long chain of amino acids, because they all have their different electromagnetic forces, they all start to attract or repulse each other, and form a three-dimensional shape. It is hard to overstate the genius of the design necessary to be able to produce complex three-dimensional robots from a contiguous chain of chemicals.

Scientists love the ribbon structures because it allows you to see a kind of “see through” perspective of the protein - to better grasp what is what and where. In this ribbon model, you can also see three basic patterns of how amino acids can arrange themselves: helixes, stacking sheets (where the arrow points the direction of the contiguous chain of amino acids) and random loops. The colors in this case are the authors’ preference, and clearly some scientists love rainbows and unicorns too, no shame in that. The active site of furin, the same one we just saw above in a space filled model, is where you find the folic acid shown as a pink chemical (on the left hand side). The authors also studied folinic acid, which is a metabolic by-product of folic acid, shown on the right hand side. You can see that the furin structure was tilted a bit (and where handy rainbow colors help you orient yourself when comparing such images, clever, right?).

And what about that aspartame?

The final countdown (drug screen countdown)

Here comes the second investigation from the same Chinese group that looked at the furin inhibition options. But this time they looked at structural models of 18 coronavirus proteins bound to a couple of human proteins, and did a similar SARS-CoV-2 drug screen as before.

Surprisingly they found only one target that would inhibit the spike protein and ACE2 interaction. Although that might not be a desired choice anyway, as we previously pointed out in our last post that ACE2 could be involved in lung injury protection. Still, to let you know, that compound was hesperidin, a type of plant pigment with antioxidant and anti-inflammatory properties, found primarily in citrus fruits like oranges, grapefruits, lemons, or tangerines.

Compounds that bind other areas of the spike protein could be important though because they could also prevent the cutting of the protein.

The authors also found compounds against SARS-CoV-2 papain-like protease, yes, another viral protease. SARS-CoV-2 encodes two of these, the one already mentioned early in this post, and this one; it is these two proteases that cleave the long polyprotein into the subcomponents that become different viral proteins important for the transcription and replication of viral RNAs (other areas of the SARS-CoV-2 genome encode individually the proteins responsible for the formation of the viral coat, and the packaging of the RNA genome). What is tricky about this papain-like protease, is that it does not just clip and snip the viral polyprotein. Like an unloved bully, it harasses proteins of our own immune system, cutting them up to reduce our chances of a fight. At least you can appreciate the incredible evolutionary interplay between the infectious agent and the host; that these viral proteins are so well designed that they execute a multitude of complex functions. That’s like the most advanced molecular, multi-functional kitchen appliance you could dream of.

Guess which compound inhibits this protease? It is another benefit of EGCG! This is also the enzyme inhibited by aspartame and procyanidins. Grapes have also another inhibitor of this protein as well, piceatannol, which the authors also mentioned. Drink up!

The other viral proteins for which they provided detailed lists of candidates were the main protease mentioned earlier, the RNA-dependent RNA polymerase (this protein replicates the viral genome to produce more of it, and just to make sure you really consider that glass of wine, the above mentioned procyanidin also happens to target this protein), and bunch of the structural proteins (ones involved in making the final package complete, ready and loaded for more human cell colonization).

Another drug that has gotten some attention as a promising treatment for COVID-19 is Remdesivir (GS-5734), and although not yet approved, it has been used for treatment of patients. This compound also targets the RNA-dependent RNA polymerase. From the authors’ modelling, it appears that it might interact with a few more viral proteins, as well as target one of the human proteases involved in spike protein cleavage. This drug is being investigated in clinical trials right now.

Thus as the world is scrambling to find a solution to one of the worst pandemics experienced in a century, we are still without any protective remedy measures. However, the scientific community has come together like never before to study this new threat to our global health, and science is pouring in at an astonishing pace. Here we demonstrate just a flavour of that science dedicated specifically to uncover any information that could provide any clues towards any prophylactic measures. Here we presented some examples of what has been published, however, since much of the information has not even been peer reviewed, yet alone actually studied for efficacy, we focused on some examples where the proposed compounds are already in daily use in our society.

Take advantage of this information in whatever way it suits you, we do wish you to stay healthy, happy, and come back one day to sequence your genome to delve into your own molecular biology instead some virus.

This article has been produced by Merogenomics Inc. and edited by Jason Chouinard, B.Sc. Reproduction and reuse of any portion of this content requires Merogenomics Inc. permission and source acknowledgment. It is your responsibility to obtain additional permissions from the third party owners that might be cited by Merogenomics Inc. Merogenomics Inc. disclaims any responsibility for any use you make of content owned by third parties without their permission.

Products and Services Promoted by Merogenomics Inc.

Select target group for DNA testing

Healthy screening	Undiagnosed diseases	Cancer	Prenatal

Or select popular DNA test

Pharmaco-genetic gene panel	Non-invasive prenatal screening	Cancer predisposition gene panel	Full genome