Hacker News new | ask | show | jobs
by avancemos 2004 days ago
The question is, how did this start just now? In 2020? How is the COVID-19 vaccine the first to use mRNA? Anyone with who has taken AP Biology could conceive of and understand the idea behind making vaccines rapidly: take some mRNA, inject it, have in translated as the antigen in the body. Poof, that's it. I feel like the development of mRNA drugs should have started in the 70's or 80's. It isn't exactly high-tech or clever.
10 comments

Delivery of the RNA is hard. To the right cell type, not immediately degraded, not accidentally integrated into a critical part of the genome, with a payload that is actually effective, etc.

The original gene therapies (early 2000s) were essentially RNA therapies (adenovirus). And their unethical rush and subsequent failures caused a bit of a 'gene therapy winter' [1]. We've since made enormous progress on both the ability to safely deliver genes, but also our ability to generate/design new useful genes.

[1] https://www.labiotech.eu/in-depth/gene-therapy-history/

> In 1972, a paper titled ‘Gene therapy for human genetic disease?’ was published in Science by US scientists Theodore Friedmann and Richard Roblin, who outlined the immense potential of incorporating DNA sequences into patients’ cells for treating people with genetic disorders. However, they urged caution in the development of the technology, pointing out several key bottlenecks in scientific understanding that still needed to be addressed.

Wouldn't an adenovirus be delivering DNA? mRNA can't be incorporated into the genome (barring some crazy mixing due to a retrovirus) because it's RNA.
mRNA is not the only type of RNA that can be delivered. Another type could be interfering RNA targeting endogenous coding or non-coding RNA molecules.
"incorporating DNA sequences into patients’ cells" is a wildly more dramatic approach than temporarily tricking a number of cells into manufacturing some protein with mRNA. It's almost like the difference between getting the browser to run your page's js vs a full remote code execution vulnerability.
Can we please not spread the misinformation that mRNA can somehow integrate into the genome? It feeds crazy conspiracy theories around the vaccines.
The parent was probably confusing RNA with adenovirus which IIUC does deliver DNA that integrates itself into the genome. There are dozens of COVID vaccines under development right. Many of them are in fact DNA vaccines. To date they've only been approved for vaccinating dogs of rabies. The mRNA vaccines that companies like Pfizer are making have the advantage of not permanently changing the DNA in the target cells. Even with DNA vaccine it's not the end of the world. For example, herpes simplex (cold sores) is an example of a natural virus that integrates itself in the DNA. But it's localized and it's not something that your children are going to inherit. Another interesting fact is that the Pfizer mRNA vaccine and others are delivered using lipid nanobots rather than adenovirus which I think is cool. But DNA vaccines have even potentially cooler applications since it means the medical field might for once be able to offer cures to illnesses, rather than charging you for a pill every day.
How can you say we made enormous progress when it appears this is still an untested therapy.
The number of [nucleic-acid-delivered] gene therapies in Phase II & Phase III trials right now is huge - because of this progress in delivery [of nucleic acids]. Gene therapies for the eye, for hemophilia, for sickle cell, many many cancer therapies all rely on the ability to 'deliver' nucleic acid payloads to cells. Of those, only 3 or 4 have been approved - and all in the past 2 years, but there are a huge number that are behind that tip of the iceberg - quite precisely because it's relatively straightforward to do 'same thing but with a different sequence' once the first one works.
>take[1] some mRNA[2], inject it [3], have in translated as the antigen in the body[4]. Poof, that's it.

1) Develop synthesis technique for large scale high purity mRNA without base errors or truncations.

2) Discover pseudouridine modification to decrease innate immune response.

3) Discover and optimize lipid nanoparticles for encapsulation of mRNA to prevent its degradation.

4) Optimize LNPs and miRNA sites in UTRs for localization to desired cell type and to prevent aggregation in undesired or dangerous cells/organs.

Poof, that's it!

Another easily-searchable fact is that the COVID-19 vaccine is not the first RNA-mechanism drug.
The technology for synthesizing large quantities of specific RNA sequences has only been available recently. Same goes for forming the lipid nanoparticles that are used for encapsulating and delivering the mRNA. In fact if anyone has more detailed information on how these two processes are done I would love to learn more.
PCR, the Polymerase Chain Reation.

https://en.wikipedia.org/wiki/Polymerase_chain_reaction

the emphasis in the wiki article is on DNA, the same basic principles apply with RNA.

for example you can start with RNA and use reverse trancriptase to produce a DNA sequence, amplify that sequence to a large copy number by repeatedly replicating it then transcribe the DNA to produce large quantities of RNA.

or you can start with the DNA [in large quantities of purity] then create many copies of the RNA by repeatedly transcribing the DNA.

The practice of artificial [in vitro] gene synthesis can create arbitrary sequences for input to the process[es].

https://en.wikipedia.org/wiki/Artificial_gene_synthesis

liposomes are manufactured via biochemical-mechanical process.

https://en.wikipedia.org/wiki/Liposome#Manufacturing

> the emphasis in the wiki article is on DNA, the same basic principles apply with RNA.

This is not my understanding. PCR works to produce DNA because the DNA polymerase enzyme creates DNA copies from DNA templates, and those copies become templates themselves, feeding back into the chain reaction.

I'm not aware of similar enzymes capable of making RNA copies from RNA templates (RNA polymerases use DNA templates AFAIK).

I was imagining the RNA vaccines would have used a fully synthetic oligo production method, to get the extreme purity required. But it might be that some sort of PCR-like amplification process is used. Would love to read more if someone has details.

the viral RNA polymerase replicates the viral RNA and this can be exploited for direct RNA to RNA amplification.

to wit : https://chempedia.info/info/viral_rna_polymerase/

I was making reference to the process of cyclic amplification, in a manner palatable to non biology literate individuals

The same _basic_ principles

as can be read in the chempedia example the process of RNApol replication of RNA from RNA template is error prone, thus it is superior to create a large copy number of DNA template corresponding to the mRNA desired as DNA amplification is self correcting for the most part.

dependent upon experimental or procedural requirements it may be desirable to produce error prone replication variants of the RNA template, however it is quite possible to go from RNA template to RNA product with RNApol.

due to the fragile nature of RNA there must be some shielding or , a low cycle rate is used in conjunction with immediate harvest and stabilization of the product.

the sars-2 covid19 virus replicates by way of an RNA dependent RNApol [rdRNApol]

https://pubmed.ncbi.nlm.nih.gov/32277040/

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7164392/

Thanks for the additional links. I guess you would need temperature stable viral RNApol to do direct 'RNA' amplification. Not sure if that exists or if it's even possible due to RNA instability. As you say not so worthwhile due to high errors.

Producing RNA from DNA amplicons (PCR products) makes sense, I'm curious if that is the source of material from in the vaccines, or if it's artificially synthesized?

Would that be a trade secret?

the exact sequences could be proprietary secrets.

when you have determined a sequence of RNA corresponding to a desired protien you may construct a DNA template with modification to suit your purpose.

the DNA polymerase used for DNA PCR is originally from a high tempurature tolerant organism, [thermophillic bcterium] and it is possible via searching high and low to find an organism that bears high temperature tolerant rdRNApol

https://www.sciencedirect.com/topics/biochemistry-genetics-a...

and BTW the original inventor of PCR [kary mullis]

was awake AF on LSD when he conceived of this technology while sitting on a park bench.

https://en.wikipedia.org/wiki/Kary_Mullis

i believe the original PCR was performed MANUALLY! with glassware and perfectionist tedium.

https://en.wikipedia.org/wiki/Kary_Mullis#Use_of_hallucinoge...

you may find this to be interesting:

https://en.wikipedia.org/wiki/Ion_semiconductor_sequencing

I'd like to know more about the potential for lipid nanoparticles to cross the BBB [1], and therefore consideration of long-term effects of mRNA and resulting proteins in brain cells. Am I alone in this concern? [1] https://pubmed.ncbi.nlm.nih.gov/29886842/
> it isn’t exactly high-tech or clever

Read the history section of the rna vaccine wiki: https://en.m.wikipedia.org/wiki/RNA_vaccine

You’re right, we’ve understood that this could be done for a long time. Not quite the 70s, but.. at least the 90s. But believing something is possible and knowing how to do it are different:

1989 - injected rna maybe goes into cells

1990 - proof injected rna creates proteins

1994 - proof injected rna creates immune responses

Then:

“2005 they published a joint paper that solved one of the key technical barriers by using modified nucleosides to get mRNA inside human cells without setting off the body's defense system”

That kicked off a ton of research, but:

“Up until 2020, these mRNA biotech companies had poor results testing mRNA drugs for cardiovascular, metabolic and renal diseases; selected targets for cancer; and rare diseases like Crigler–Najjar syndrome”

But why did they spend from 2005 to 2020 working on mRNA drugs and not vaccines? Capitalism. Vaccines are not generally profitable (take once, you’re done) - so vaccines are not an appealing target for a startup with investors wanting big returns. (Source: https://www.statnews.com/2017/01/10/moderna-trouble-mrna/)

Yeah seems like vaccine development only started after the high profit objectives suffered failure. That doesn't mean we can't go back to working on those objectives in the future but they needed SOMETHING that would work to show mrna promise. Vaccines you only have to take once or twice so the side effects of immune response aren't too bad and you have an immune response to a vaccine anyways.
Even for the staunchest capitalist, covid has clearly outlined that we need medical research that is separated from market forces. After sars 1 was controlled, research in this area was all but dropped - despite virologists warning that it was just a matter of time before... well, this.

Medical research needs to be driven by what can help people, not by what can make the most money.

PPE too. We still don't have N95s for the public or in some areas even the medical professionals.

If we had a different administration in the USA we may have been able to conquer that quickly with sheer cash and coordination since the knowledge how to build melt blown N95 machines is there. For whatever reason we just didn't.

"Poof, that's it."

The immune system is incredibly dangerous to its own host if mishandled. By stimulating response, you are trying to light a cigarette using a white phosphorus flamethrower, so to say.

It took a lot of time to find the optimal way of mRNA delivery that a) really does something but b) does not provoke a massive, counterproductive response. This is a very narrow rocky ledge with precipices on both sides to walk.

I think the theory behind it is straightforward enough (and indeed, Moderna was founded 10 years ago with this sort of mRNA vaccine as their explicit goal), but the practice is more complicated.

Figuring out what sequence of mRNA will be the right one to get a cell to produce the right antibodies for the job, getting that sequence sliced out of the viral RNA, getting that all into a form where it can be absorbed by cells and not just instantly degrade are all nontrivial tasks.

The devil’s in the details: compare to the field of software or cpu engineering... seems straightforward enough to just have more instruction decoders, but due to complexities only ditching x86 has actually made it possible for Apple to do this.

> Anyone with who has taken AP Biology could conceive of and understand the idea behind making vaccines rapidly: take some mRNA, inject it, have in translated as the antigen in the body.

There was quite a lot of development on basic techniques of working with RNA necessary before that could even in isolated circumstances be easier than, or even competitive with, “isolate the antigen, inject it, done”.

> The question is, how did this start just now?

It didn't.

Getting a treatment to market isn't the start of application of a new technique in medicine; its usually something that happens many years, often decades, into work using the technique.

It’s not new. Work had been happening on mRNA. But the vaccine, I assume, is humans working under pressure and finally making it happen.

I found this video (from 2013) in the other thread today about the vaccine Moderna Vaccine taking two days to make: https://news.ycombinator.com/item?id=25468959

Given that Moderna is the first ever company to bring such a vaccine to market and they’ve been working on mRNA for the past 10 years, I imagine there’s a lot of technical complexity to actually deliver a therapy and then mass produce it beyond just the basic concept.

Reading how Pfizer and Moderna worked on it together, they needed detailed gene sequencing to understand how to design a potential vaccine. Even then they were left with a lot of potential options they still had to whittle down. Finally even with all that work they’re left with a vaccine with complex storage requirements.

So it’s entirely possible that we just didn’t have the surrounding technical ability even if theoretically it was possible. The gene sequencing to sequence it quickly and share that across the entire world, the compute needed to do try different experiments at scale, the manufacturing capabilities, Moderna having invested in the space for the preceding 10 years, existing experience with developing a SARS vaccine, etc.

Pfizer worked with BioNTech, not Moderna.

Moderna's vaccine is a great deal easier to store than the BioNTech/Pfizer vaccine (requires 'normal' freezers for storage, can be at refrigerator temperatures for a longer period).

BioNTech and Moderna both have links to Katalin Kariko so while they didn't explicitly work together on this the knowledge comes from similar research and sources.
I think you mean BioNTech not Moderna.