The Roots of Progress

The 300-year journey to the covid vaccine

From inoculation to RNA

A covid vaccine has demonstrated 90% efficacy and no significant safety concerns in preliminary data from Phase 3 trials, according to an announcement today from Pfizer and BioNTech SE. The trials aren’t yet complete and the data hasn’t yet been released for independent verification, but this is very good news. (More from STAT News.)

Pfizer/BioNTech’s vaccine, like Moderna’s, is based on “mRNA” technology. If approved by the FDA, it will be the first such vaccine to reach that milestone. From a long-term progress perspective, this is a big deal.

Immunization technology has existed since the early 1700s (and the folk practices it originated in go back centuries further.) We can see the whole 300-year history of the technology as a quest to achieve immunity with ever-more safety and ever-fewer side effects. More recently, it has also become important to be able to react quickly to new epidemics, such as covid.

Here’s how immunization has advanced in stages:


All immunization is based on the observation that exposure to a disease often grants immunity (temporary if not permanent) to subsequent exposure. Long before we knew anything about antibodies or T-cells, people had noticed this simple correlation. Many people got smallpox in the past, but almost no one got it twice. The goal of immunization technology is to achieve that same immunity, but without having to suffer the disease or to risk death or other side effects.

The earliest form of immunization, then, was not a vaccine, but a method in which the patient was given the actual disease itself, in a manner that would cause a mild rather than a severe case of the illness. This was done with smallpox, and the technique was called inoculation or variolation.

This worked with smallpox for two reasons. One, infectious material was easy to obtain, from the pustules caused by the disease itself. Second, contracting the disease through a scratch on the skin caused a much more mild form than contracting it more naturally through inhalation.

Inoculation saved many people from smallpox. But there were downsides. First, the patient still had to contract the disease, causing mild symptoms. Second, there was still a small risk of a severe case; even the best inoculation methods had about a 0.2% death rate. Third, the patient was still contagious while going through the illness, and anyone who caught the disease naturally from an inoculated patient would get the full, severe version. Inoculation thus risked outbreaks.


These problems were solved by the next stage: vaccination. It was observed that cowpox infection granted some form of cross-immunity to smallpox. Thus, the inoculation procedure could be performed using cowpox material, rather than smallpox. Cowpox was a milder and non-lethal disease. This reduced the symptoms and the risk of death, and eliminated the risk of smallpox outbreaks as a result of immunization. This new technique, invented by Edward Jenner in 1796, was called vaccination (from vacca, the Latin word for cow).

So far, however, the technique only worked for smallpox—not for tuberculosis, malaria, influenza, cholera, or any of the other major diseases that caused something like half of all deaths in that era.

Engineered vaccines

The next stage would wait almost ninety years. Louis Pasteur, a pioneer of microbiology who along with Robert Koch established the germ theory, was the first to discover how to create vaccines for any disease other than smallpox.

Cowpox can be seen as a “natural vaccine” against smallpox: a natural virus that grants smallpox immunity but produces milder side effects. Pasteur’s accomplishment was to create artificial, engineered vaccines.

There are essentially two ways to do this. The germ that causes the disease, or pathogen, can be modified chemically, “killing” or inactivating it. This can be done through heat, through chemicals such as formaldehyde, or through other means. Or it can be modified biologically, attenuating (i.e., weakening) it. This is done by evolving the virus or bacterium for many generations in an animal or tissue culture that is sufficiently different from the target patient. For instance, Pasteur found that “passing” a disease called swine erysipelas through many generations of rabbits caused it to be less virulent in pigs.

These techniques allowed vaccines to be created for more diseases, and many were created in the decades that followed. The other advantage was a reduction in side effects. By weakening or inactivating the pathogen, the patient no longer had to suffer through a full infection in order to receive immunity.

But attenuated and killed vaccines still had risks. If a killed vaccine was not properly manufactured, it could contain some portion of live germs, as happened with one of the makers of the first polio vaccine. And a live attenuated vaccine could always mutate back into a virulent form. In either case, the vaccine would cause the very disease it was designed to prevent, not only in the unlucky patient but potentially in a new contagious outbreak.

Subunit vaccines

A way to prevent this risk is by giving the patient, not an entire virus or bacterium, not even a weakened or inactivated one, but just a portion of the pathogen.

This works because of the way the immune system functions. In essence, it detects foreign substances in the body and produces new molecules, called antibodies, that bind to these substances and get in their way, preventing them from doing damage. This process takes time for a new, never-before-seen infection, but after the first encounter, a record of the antibody is stored in the body’s immunological memory, which enables a quicker reaction to subsequent infection. A foreign substance that stimulates the production of antibodies is called an antigen.

All immunization works by this process of priming the immunological memory using antigens. The key observation is that even a piece of the pathogen can be used as an antigen, and the antibodies thus generated are effective against the full pathogen itself. An antigen that is not a pathogen is exactly what we want: a substance that produces immunity without producing disease.

For example, consider the SARS-CoV-2 virus that causes covid. You’ve probably seen it rendered as a spiky ball. Those spikes are crucial to the virus’s function: they stick to your body’s cells like tentacles, as the first step of the infection. A subunit covid vaccine, then, works by injecting just the spike into the body, rather than the full virus. The body learns to generate antibodies against the spike, and those antibodies are effective against covid itself. The big advantage of this, of course, is that a single piece of a pathogen cannot replicate and thus cannot cause an infection or become contagious.

But how is the subunit antigen to be manufactured? Inactivated or attenuated vaccines can start with the original virus or bacterium and grow it in culture. Subunit vaccines can be created in a similar way, by culturing the pathogen and then breaking it apart in order to extract the desired piece. With modern biotech, however, there are other ways. If the antigen is a protein (as in the case of the covid spike), it can be manufactured in genetically engineered microbes. Start with a single-celled organism such as E. coli or baker’s yeast. Insert the DNA that codes for the subunit protein into its genome using recombinant DNA technology. Replicate these cells until you have a whole vat of them creating vaccine proteins for you. (The same technology makes other synthetic biologics, such as insulin.)

RNA vaccines take this idea the next logical step.

RNA vaccines

A yeast cell can function as a biological factory, producing proteins according to a programmed genetic code. But every cell in your body is also such a factory, with the same fundamental machinery.

An RNA vaccine skips the step of programming single-celled organisms to produce the antigen for us: it sends the genetic code for the antigen directly to your own cells, and they produce the antigen. These vaccines, then, are the only kind that do not inject the antigen directly into the body; genetic instructions are injected instead. (Note that, unlike with recombinant DNA technology, the DNA of your cells is not modified.)

To my (limited) understanding, this does not produce a significantly different immune response than injecting the antigen directly. However, it makes a big difference in how these vaccines are designed, developed, and manufactured, which affects our ability to respond quickly to new outbreaks such as covid. Once the virus’s genetic code is sequenced, the virus itself does not need to be handled in order to create a vaccine. The vaccine is based entirely on genetic material, and can be created using genetic synthesis techniques. Every pathogen is different in how it can be grown in culture, and in what it takes to inactivate it, weaken it, or break it apart into subunits. Genetic techniques, by contrast, can be much more standardized. This doesn’t make the development of these vaccines trivial; there are still many problems to be solved for each one (such as the delivery mechanism to get it into the cells, where the genetic program will be executed). But as we are seeing, their development can be significantly faster than traditional techniques.

When you get your covid shot (probably in 2021), take a moment to think back on the 300 years of progress that got us to this point.

Comment: LessWrong, Reddit

Get posts by email:

Become a patron

Get posts by email