Drug therapy for the Coronavirus

Last month, a new epidemic started in Wuhan, China, and quickly spread all over the world. As of now, there are reports of a few companies rushing to develop, or having already developed, a vaccine for this virus.¹ ² ³ ⁴ Thanks to recent technological progress, the development of new vaccines is fairly quick, and clinical trials can start only a few months after the pathogen was sequenced.

However, availability of a vaccine to the public will be delayed by the clinical trials necessary to ensure the vaccine works as intended and is not harmful. The expected time-span for these tests is one or two years, as animal testing must precede human testing. The issue is that vaccines behave slightly differently in humans and in animals (usually mice and/or monkeys). In the early stages of research, vaccines that worked well in animals were much weaker in humans. In addition, vaccines can cause inflammation. Slight inflammation is good, and makes the vaccine more effective, but strong inflammation can be counterproductive and outright harmful. Because of differences between humans and animals, a vaccine can be toxic for us but not for them (and vice-versa, but in this case the vaccine would never be tested in humans).

Unfortunately I could not find any scientific publications about these alleged vaccines (as of Jan. 30th), likely because of all the commercial interest behind them, ⁵ ⁶ but most news sources seem to agree on the fact that it will be either a mRNA or a DNA vaccine. These kind of vaccines are designed from scratch to match the biological signature of a pathogen, without including unnecessary, and potentially toxic, proteins. Developing these kind of vaccines is much easier, as it does not require interacting with live pathogens, and they can be manufactured cheaply.

January 31st

Several articles have been published on biorxiv on the topic,⁷ and the article that first shared the sequences of this virus has been read more than 150 000 times since it was posted seven days ago.⁸ As I said above, these sequences are fundamental to develop drugs. There were several works modeling the outbreak and studying how the virus spread, estimating the number of undiscovered cases and the infectivity of the virus.⁹ ¹⁰

Others are comparing this strain of Coronavirus with old strains that caused outbreaks in the past, finding good agreement between the sequences, or at least the parts essential for the life cycle of the virus ¹¹. This is great news: it means that drugs developed for those strains can possibly work directly on the new Coronavirus, or require little adaptation. (Addendum: I misunderstood what was said in that paper. Even though they argue that the viruses work in the same way, they point out that antibodies designed for the old strain are not very effective on the new strain). This in turn means that clinical testing is not necessary, as these drugs have already been tested. For now, several existing drugs have been identified as effective inhibitors or potential therapeutics,¹² ¹³ ¹⁴ ¹¹ which might speed up general availability as they have already undergone clinical trials.

In the next weeks, I will try to follow these developments and keep this post updated. I hope to learn more about how vaccines are developed, as it is very relevant to my PhD, and watching this process unfold live is fascinating.

How does the Coronavirus work?

Viruses are at the boundary of things considered to be “alive”. A virus only contains what is strictly necessary to infect a cell and trick it into producing copies of the virus itself. In a sense, the only purpose of a virus is to create more copies of itself. This process proceeds more or less as follows: first, the virus docks to the membrane of a cell, then it enters inside as a whole and gets rid of its envelope. This exposes the genetic material contained inside the virus, which will be copied by the host cell and assembled into a new copy. This new organism can now exit the cell, potentially destroying it, and infect others. There are lots of variations to this procedure, but it gives the basic picture.

Coronaviruses, specifically, use the so-called Spike protein, a know region in its genome, to enter inside cells. The Spike protein is divided into two regions, S1 (surface unit) and S2 (transmembrane unit); S1 binds to a receptor on the surface of the cell called Angiotensin-converting enzyme 2, or ACE2 for short. This receptor normally “listens” for a hormone called Angiotensin, which takes part in regulating blood pressure. After the Coronavirus is bound to ACE2, the S2 portion will take care of fusing the envelope of the virus with the cell’s membrane, which causes the content of the virus to be released inside the cell. More in detail, this happens because a protein of the host cell, called type II transmembrane serine protease (TTSP) “cuts” the spike protein, separating S1 and S2. This frees S2 to fuse with the membrane of the host cell, allowing the virus to enter and infect the cell. There is also a second mechanism that the Coronavirus can use to enter the host cell. In case TTSP is not present to separate S1 and S2, the cell “pulls” the virus inside through a process called endocytosis, in which the surface of the host cell folds inward, surrounding the virus with a sort of bag which is then pulled inside the cell. This causes a change in the chemical substances surrounding the virus, triggering separation of S1 and S2. S2 is now free to merge with the bag and free the content of the virus inside the host cell. This process is identical to what happens on the outside of the cell. ¹⁵

Note: I am a bit confused about this. Two papers ¹⁵ ¹⁶ indicate that these two mechanisms are separate, and the virus is able to perform both according to the presence or absence of TTSP. But another paper ¹⁷ explains that they actually happen one after the other. However, in their explanation they cite several articles that seem to confirm that the two mechanisms are separate. ¹⁸ ¹⁹ ²⁰ I do not feel competent enough to say who is right, as I might have misunderstood the literature, but it seems to me that there are indeed two different and separate ways for the Coronavirus to enter a cell.

February 4th

Research continues on the applicability of existing drugs for the novel Coronavirus. ²¹ ²² It is now well understood that this virus works in essentially the same way as the old SARS- and MERS- Coronaviruses that caused outbreaks in 2003 and 2012, with several confirmatory results. ¹¹ ²³ ²⁴ ²⁵ The virus is able to infect cells in the lungs,²⁴ explaining the symptom of pneumonia (lung inflammation), but the digestive system could be another possible route for infection. ²⁵ There are projects that test the efficacy of known antibodies¹¹ or develop new ones specifically tailored to the new strain. ²⁶ Computer simulations have concluded that the novel Coronavirus binds less strongly to ACE2, resulting in lower infectivity. ²⁷

February 14th

The volume of publications on biorxiv on the topic seems to have diminished, and several are now focusing on the design of vaccines for the Coronavirus. These are most interesting to me, so I will cover them in slightly more detail.

Ahmed et al. ²⁸ propose several B-cell and T-cell epitopes that can be included in a vaccine. These epitopes are derived from the structural proteins of the virus, i.e. the parts that keep it together. Besides the spike protein (S), the virus also uses the envelope protein (E), membrane protein (M), and the nucleocapsid protein (N). All of them take part in building the shell of the virus, so a vaccine designed to look like these proteins can train the body to spot the Coronavirus. This work essentially started from epitopes that were confirmed experimentally for the SARS virus, and searched for them inside the novel Coronavirus, then reduced the candidate set by choosing the epitopes that cover as many people as possible. People respond differently to these epitopes, so an effective vaccine should contain epitopes that are recognized by most people.

Abdelmageed et al. ²⁹ also looked for epitopes for a potential vaccine. Interestingly, they focused entirely on the envelope protein, while the previous work focused on the spike and nucleocapsid proteins. This study used bioinformatics tools to predict potential epitopes from conserved regions. Conserved regions are those that do not mutate; this indicates importance for the correct functioning of the virus, because mutation is an important mechanism used to evade the defenses of the body. This work also filters epitopes for high population coverage. However, despite the title, they do not design a vaccine, but only propose epitopes to include in a vaccine.

Sarkar et al. ³⁰ proposes three complete vaccines for the Coronavirus. The initial stages are very similar to the two previous works, ²⁸ ²⁹ using the same bioinformatics tools to identify epitopes. This work carries things further by filtering epitopes based on allergenicity and toxicity, and computing the ease by which these epitopes dock to the MHC by analyzing how well the folds of the two sequences (epitope and MHC) match. Binding between epitopes and MHC is fundamental for the body to learn to recognize the epitopes. Next, they assemble these epitopes in three vaccines, that differ by which adjuvant they use. Adjuvants are substances that further increase the sensitivity of the body to the epitopes in the vaccine. Then they again analyze toxicity and allergenicity of the vaccines, predict how they are going to fold, and improve their stability to make the protein less brittle. Finally, they adapt the sequence so that it can be easily produced.

Another paper ³¹ looked for potential epitopes in the Coronavirus. Interestingly, they found that the spike protein of 2019-nCov is quite different from that of SARS, and suggest that antibodies tailored to SARS might not be effective for the novel Coronavirus. This is consistent with what reported previously, however some drugs seem to work with lower effectivity. This work also postulates that the Chinese population might be more vulnerable to the novel Coronavirus, and that the virus might mutate in the same regions where SARS and MERS did. Nothing new, but it is good to have several confirmations of the same facts.

Several papers focus on the entry mechanism of the Coronavirus. As described earlier, entry in the host cells is mediated by the surface unit of the spike protein docking to the ACE2 receptor, and TTSP cleaving the spike protein, priming S2 to fuse with the cell membrane. A paper ¹⁶ identifies where the cells expressing ACE2 and TTSP are located in the body, and shows that a mutation in the novel Coronavirus that is believed to increase the activity of a specific type of TTSP, making the virus more infective compared to SARS. This type of TTSP is also involved in SARS and influenza infections. ³² Interestingly, this seems to contradict a previous result ²⁷ that used computer simulations to conclude that this virus is less infective. This conclusion was supported by the fact that the spike protein binds less strongly to ACE2, but they did not consider the activity of TTSP. These two papers actually studied the two different entry mechanisms of the Coronavirus, so they are not directly contradicting each other. With a similar approach to ¹⁶, another paper ³³ shows that the ACE2 receptor is also found in certain liver cells, explaining the symptoms observed in some patients.

A paper ³⁴ designs anti-Coronavirus drugs that inhibit the activity of its protease. The viral protease helps in producing the final proteins needed to assemble a new copy of the virus, after its genetic material has been copied by the infected cell. The same authors also use the same tools to identify protease inhibitors designed for SARS that are effective for the novel Coronavirus. ³⁵ Finally, a study ³⁶ identifies another antibiotic used for Ebola, SARS and MERS that works for the Coronavirus, preventing cellular entry.

Conclusion

After this point, I have not seen any more new interesting findings related to vaccines, and, to be honest, I stopped following preprints so closely. I focused on more reputable and informative sources and I still learned a lot even though I lost the novelty aspect of preprints. Anyway, the follow-up to this post is here, enjoy.

January 31st

How does the Coronavirus work?

February 4th

February 14th

Conclusion

References