Hacker News Reader

Comment by gilleain | Hacker News Reader

Slightly misleading headline, as PET (polyethylene terephthalate) hydrolases were already known - see https://www.chm.bris.ac.uk/motm/PETase/PETaseh.htm

This is a study of an archaeal enzyme (from the deep-sea) that can degrade both long-chain and short-chain polymers that is more efficient at 70 deg C than other enzymes.

The details are that this is a feruloyl esterase with a conserved alpha/beta-hydrolase fold but with an additional flexible 'lid' domain that covers the active site. The authors suggest this lid domain replaces the conserved Trp that is an Ala in the archaeal enzyme.

View on HN

"this is a feruloyl esterase with a conserved alpha/beta-hydrolase fold but with an additional flexible 'lid' domain that covers the active site"

As educated as I think I am, it is always humbling to read a hacker news comment that I cannot understand.

ELI5?

Apologies, that was meant for the specialists :) It is mostly just technical jargon so do not feel too humble ...

So I'll break it down:

- "feruloyl esterase" : an 'esterase' is an enzyme that makes or breaks ester bonds. In this case, 'feruloyl' which I've never heard of but apparently is some small molecule that is normally attached to a sugar. https://en.wikipedia.org/wiki/Feruloyl_esterase

- "conserved alpha/beta-hydrolase fold" : a structural pattern (fold) that is shared (conserved) between a set of structures. In other words, these enzymes all have roughly the same shape. https://www.cathdb.info/version/v4_3_0/superfamily/3.40.50.1...

- "additional flexible 'lid' domain that covers the active site" : a 'domain' is just a compact part of a protein, without going into too much detail. They are calling it a 'flexible lid' as the idea is that it moves out of the way to bind the substrate to the active site (where the reaction is carried out) and back again when fully bound.

Thank you.

I've been using ChatGPT4 for this situtations, with excellent results. Check out this conversation about it (I used a PDF reader plugin, also): https://chat.openai.com/share/a682c5a7-f2c5-4849-a999-5a7218...

Enzymes are kind of like a tool for breaking apart lego blocks, or a jig for assembling them. This sounds like a moving part that covers the "teeth" and makes it more effective (or less likely to get jammed by the wrong molecules?).

Two things.

Lot of reactions are energetically favorable. But the reactants are stuck in a local minima. So either doesn't happen or is too slow. Enzymes mediate the reaction so it can happen or happen faster.

Lot of reactions there are side reactions that produce stuff you don't want. Resulting a terrible yield or maybe none. Enzymes can promote the reaction you want increasing the yield.

Interesting thing to watch for is eletrocatalysts. Where you use electricity and a catalist to push a reaction that is energetically unfavorable. Like CO2 + H2O --> methanol. These may become very important as we stop pumping oil and gas out of the ground.

I don't think there's much to say besides "slightly different chemical structure makes it better". At some point you hit jargon rock bottom.

Yeah, it's not bad news at all but it appears to be one of many that still require considerable heat, industrial infra, etc.

Has anyone attempted to put these enzymes to the test in at some scale in an industrial setting, I wonder?

I was recently at a party and spent some time talking with someone who has a stealth startup working on this exact problem. He said that they aren't the first, but getting the scale right (in an industrial setting) and finances right enough that someone is willing to spend the money to set it up is far harder than just finding an enzyme that can break down PET / similar polymers.

No idea. From skimming some university press releases, I see a lot of "potential for!" and similar wording.

This https://news.utexas.edu/2022/04/27/plastic-eating-enzyme-cou... has "Up next, the team plans to work on scaling up enzyme production to prepare for industrial and environmental application. " which is more promising.

As far as I understand, yes it is a very different challenge to make a bioreactor for actually using some random new enzyme than the (also hard) problem of getting the structure in the first place. How to immobilize it on some surface, how to pre-process the material to pass it through the reactor (?) I guess ...

Feels like it's always "potential for"... sigh

I mean my processor hits 70° and has for a decade or more, and my car engine thinks 87° is the happy temperature.

70° is maybe a little out of the reach of a home system but nothing for a municipal system. Right?

A hot water tank is not far from 70C

Should work well in landfills since there's heat buildup

https://www.purecycle.com/who-we-are. ?

> Slightly misleading headline, as PET (polyethylene terephthalate) hydrolases were already known

It doesn't say it was the only one, just a newly discovered one

Sure, fair enough. I just thought people might read it as "Enzyme with new ability (to break down PET) discovered", but maybe I was being pessimistic.

It's also mentioned in the article body:

> Until now, about 80 different PET-degrading enzymes were known, most of which were found in bacteria or fungi.

Most of us aren't reading the body though.

Reply on news.ycombinator.com