5 ms·

I work in industrial biotech, read the entire article and found it mostly compelling, aside from lacking discussion of the reasons meat production might increas

I work in industrial biotech, read the entire article and found it mostly compelling, aside from lacking discussion of the reasons meat production might increase in cost over the time frame of the analysis. The fundamental challenge is not a lack of understanding - it seems to be mostly centered around keeping bacteria out of giant 100,000 Liter vats, and out of all of the smaller upstream processes. Think of biotech as happening in successively larger tanks. Start in a vial, go up to a flask, go up to a tank fermenter, go up to a bigger and bigger vat. If competing bacteria get into your process at any of these steps or the transfers between them and start winning, you kill the batch and start over. And you probably take down your equipment for a deep clean as well. This takes you offline for some time. This is already a problem for fermentation of other bacterial cultures, but it’s an even bigger problem for animal cells because they grow so much slower. So there’s some lack of fundamental understanding on animal cell growth mechanisms. Given how much attention has been focused during the pandemic on understanding how to limit the spread of germs, it seems to me that there may be some synergistic advancements in keeping large scale processes contamination-free. A lot of it is human factors.. operators using proper PPE etc. Now we have a global population that understands all of that better.. maybe that moves the baseline.

kbenson 5y ago
It seems like it should be possible, eventually, to use a bioengineered organism to grow the tissue (and just cut down on unneeded things like extra brain matter, or eyes, etc), and let immune systems deal with bacteria and infections. I mean, that's what evolution has already come up with as a solution to this problem. What we would need though is an organism that's far more symbiotic, where we provide security, and food, and protection from the elements, and it provides lots of tissue and hopefully at a good rate if return for input calories (and why not flavor while we're at it). It's not that far a stretch from some of our domesticated animals, at least with the symbiotic relationships involved. We would just need to ramp that up to 11 and optimize out the parts that we can do more efficiently at scale or that we don't want (like cognitive function and much more than a pain response to keep it from hurting itself).
- spoonjim 5y ago
  Huh. Fascinating line of thought. What if we genetically engineered a cow that basically had no brain besides what was necessary to be alive? Would it be more ethical to kill it? Would it even be an animal?
  - meowface 5y ago
    It'd be nice to have something like lab-grown unfertilized chicken (or other bird) eggs. Not the same as meat, but it's kind of close to the goal: a biological protein source that's completely unconscious and doesn't require any death to create. It'd be very difficult, but nature already did a lot of the work there. I imagine it might be much tougher to design a full animal without a brain. You'd always be on the line between it not being able to stay alive due to loss of functionality, or, in the other direction, possibly having some degree of thinking/consciousness. This on top of the fact that we'd need a far better understanding of brains and genetics to ever hope to engineer such a thing. If it did exist, I think it'd not only be more ethical to kill it but that it'd be (hypothetically) completely ethical. I don't think it would be an animal. It gets a little weird, though, because one would have to also accept that you could hypothetically use a similar process to create brainless humans and that it'd be fine to kill and eat them if you wanted to. Or do whatever else you wanted to them. That human also wouldn't be an animal.
    - kbenson 5y ago
      It may not be that we'd have to design the full animal, and it may not be it has to not have any brain. Modifying an existing organism might be much easier, and removing higher functions might be enough, depending on what we know now or later about the brain. Is what makes eating an animal and a plant different for most people what constitutes their mass, or how we perceive them to think and feel? Making them not think and feel (and at the organism level, not the removal of that potential in an individual after birth) may be a path forward. As a hypothetical, to bring it back to humans, and divorce it from reproduction which might have emotional baggage in many peoples reasoning, imagine a person with a tuberous growth that was removed, but we can cause it to continue to grow outside the body. If it's confirmed safe to consume, are there ethical problems with that person eating their own cells? What about other people?
asdff 5y ago
its typical though to spike cell culture media with antibiotics to avoid bacterial contamination. i did a ctrl-f within the OP article as well as the 50 page pdf but i haven't seen 'antibiotic' mentioned at least. I suppose its always possible you get some bateria thats resistant to your antibiotic, but people have been using penstrep in their media for decades now and that hasn't been a huge concern thus far at least in lab environments following sterile technique.
- plaidfuji 5y ago
  Agreed, this was one of my first thoughts as well and it’s not discussed at all. Why not just dose with antibiotics? Too expensive? Regulatory issues (seems unlikely as all our livestock are already hopped up)? Consumer groups will balk?
- the8472 5y ago
  Antibiotics don't help with viruses or resistant bacteria. In live animals the antibiotics are applied on top of the immune system.
  - asdff 5y ago
    I don't think resistant bacteria is going to be an issue if you switch to other antibiotics. People have been using these antibiotics in media for decades and decades in cell culture already and this hasn't been a huge issue in this setting at least.
01100011 5y ago
Does it have to be a batch process? Can you have a continuous flow system where pathogens rate of travel is lower than the flow of nutrients and cells? Does the system have to rely on vats of cells? Can you make a believable meat substitute with GMO plants producing the raw materials which are constructed with bioprinters? Can we genetically engineer, say, a cow with a functioning immune system but with no brain or other un-consumed organs such that it is amenable to mass culturing? (Edit: Sorry, just saw sibling post says exactly this) I feel like the article is well informed and mostly correct. It's also the type of thinking that, while correct 99% of the time, prevents innovation by focusing on what can't be done. I think every engineer knows cynicism is correct 99% of the time. It's that rare chance when it's not, when you push against what seems like a law of reality and break through, that we achieve what we didn't think was possible. I have no doubt, given this article, that the current approach to lab grown meat will fail. I am not so sure about artificial meat in general, however.
yholio 5y ago
If rapid sterilization is a requirement of the process, than it makes sense to embedded it into the design. The 100 ton reactor could have electric heating elements surrounding it on every side, or magnetic loops that induce heating to the stainless steel vat. Do a dry pre-heat at 500K to vaporize any leftovers then fill it with clean water and bring it to a boiling point, establishing slight positive pressure against any pores or leaks. And voila, you have a sterile 100 ton reactor with zero bacteria. Similar sterilization systems would be installed on all ingress ports, using heat, ultraviolet or gamma ray, depending on the nature and heat sensitivity of the input, with slight overlap between the different systems. Since you are culturing organic life from dead raw materials, and life is so sensitive to heat, gamma rays etc., it seems relatively simple to make sure nothing alive enters other than what has been grown in upstream reactors. Recursively apply these principles down to the last cell inserted into the process inside a (small) clean room.
- plaidfuji 5y ago
  What you’re describing (at least the heat aspect) is referred to as “Sterilization in Place” and it is standard practice to design into bioreactors already. Contamination still happens. In general you’re right though, the concept of continuous automated sterilization will need to be reimagined, but like many of the topics addressed in this article, a lot of the low hanging fruit has been picked. https://www.flottweg.com/wiki/separation-technology/sterilization-in-place-sip/ https://www.flottweg.com/wiki/separation-technology/steriliz...
  - yholio 5y ago
    The description linked talks about a distinct sterilization phase, where hot steam is passed trough the installation prior to use. It seems exactly like the type of situation described in the article, where small welding gaps allow bacteria to stay alive - I would imagine, especially if the surrounding metal has enough thermal mass. My suggestion was a bit more nuanced, heat the very body of the reactor + reaction medium; but I could be venting hot air. Nonetheless, it's not a fundamental, thermodynamic limit, it's not like bacteria spontaneously materialize inside vats above a certain volume. It's a technical problem to solve, it seems solvable and once you solve it, all 100 ton reactors can use that design.

kbenson 5y ago

It seems like it should be possible, eventually, to use a bioengineered organism to grow the tissue (and just cut down on unneeded things like extra brain matter, or eyes, etc), and let immune systems deal with bacteria and infections. I mean, that's what evolution has already come up with as a solution to this problem. What we would need though is an organism that's far more symbiotic, where we provide security, and food, and protection from the elements, and it provides lots of tissue and hopefully at a good rate if return for input calories (and why not flavor while we're at it). It's not that far a stretch from some of our domesticated animals, at least with the symbiotic relationships involved. We would just need to ramp that up to 11 and optimize out the parts that we can do more efficiently at scale or that we don't want (like cognitive function and much more than a pain response to keep it from hurting itself).

spoonjim 5y ago
Huh. Fascinating line of thought. What if we genetically engineered a cow that basically had no brain besides what was necessary to be alive? Would it be more ethical to kill it? Would it even be an animal?
- meowface 5y ago
  It'd be nice to have something like lab-grown unfertilized chicken (or other bird) eggs. Not the same as meat, but it's kind of close to the goal: a biological protein source that's completely unconscious and doesn't require any death to create. It'd be very difficult, but nature already did a lot of the work there. I imagine it might be much tougher to design a full animal without a brain. You'd always be on the line between it not being able to stay alive due to loss of functionality, or, in the other direction, possibly having some degree of thinking/consciousness. This on top of the fact that we'd need a far better understanding of brains and genetics to ever hope to engineer such a thing. If it did exist, I think it'd not only be more ethical to kill it but that it'd be (hypothetically) completely ethical. I don't think it would be an animal. It gets a little weird, though, because one would have to also accept that you could hypothetically use a similar process to create brainless humans and that it'd be fine to kill and eat them if you wanted to. Or do whatever else you wanted to them. That human also wouldn't be an animal.
  - kbenson 5y ago
    It may not be that we'd have to design the full animal, and it may not be it has to not have any brain. Modifying an existing organism might be much easier, and removing higher functions might be enough, depending on what we know now or later about the brain. Is what makes eating an animal and a plant different for most people what constitutes their mass, or how we perceive them to think and feel? Making them not think and feel (and at the organism level, not the removal of that potential in an individual after birth) may be a path forward. As a hypothetical, to bring it back to humans, and divorce it from reproduction which might have emotional baggage in many peoples reasoning, imagine a person with a tuberous growth that was removed, but we can cause it to continue to grow outside the body. If it's confirmed safe to consume, are there ethical problems with that person eating their own cells? What about other people?

asdff 5y ago

its typical though to spike cell culture media with antibiotics to avoid bacterial contamination. i did a ctrl-f within the OP article as well as the 50 page pdf but i haven't seen 'antibiotic' mentioned at least. I suppose its always possible you get some bateria thats resistant to your antibiotic, but people have been using penstrep in their media for decades now and that hasn't been a huge concern thus far at least in lab environments following sterile technique.

plaidfuji 5y ago
Agreed, this was one of my first thoughts as well and it’s not discussed at all. Why not just dose with antibiotics? Too expensive? Regulatory issues (seems unlikely as all our livestock are already hopped up)? Consumer groups will balk?
the8472 5y ago
Antibiotics don't help with viruses or resistant bacteria. In live animals the antibiotics are applied on top of the immune system.
- asdff 5y ago
  I don't think resistant bacteria is going to be an issue if you switch to other antibiotics. People have been using these antibiotics in media for decades and decades in cell culture already and this hasn't been a huge issue in this setting at least.

01100011 5y ago

Does it have to be a batch process? Can you have a continuous flow system where pathogens rate of travel is lower than the flow of nutrients and cells? Does the system have to rely on vats of cells? Can you make a believable meat substitute with GMO plants producing the raw materials which are constructed with bioprinters? Can we genetically engineer, say, a cow with a functioning immune system but with no brain or other un-consumed organs such that it is amenable to mass culturing? (Edit: Sorry, just saw sibling post says exactly this) I feel like the article is well informed and mostly correct. It's also the type of thinking that, while correct 99% of the time, prevents innovation by focusing on what can't be done. I think every engineer knows cynicism is correct 99% of the time. It's that rare chance when it's not, when you push against what seems like a law of reality and break through, that we achieve what we didn't think was possible. I have no doubt, given this article, that the current approach to lab grown meat will fail. I am not so sure about artificial meat in general, however.

yholio 5y ago

If rapid sterilization is a requirement of the process, than it makes sense to embedded it into the design. The 100 ton reactor could have electric heating elements surrounding it on every side, or magnetic loops that induce heating to the stainless steel vat. Do a dry pre-heat at 500K to vaporize any leftovers then fill it with clean water and bring it to a boiling point, establishing slight positive pressure against any pores or leaks. And voila, you have a sterile 100 ton reactor with zero bacteria. Similar sterilization systems would be installed on all ingress ports, using heat, ultraviolet or gamma ray, depending on the nature and heat sensitivity of the input, with slight overlap between the different systems. Since you are culturing organic life from dead raw materials, and life is so sensitive to heat, gamma rays etc., it seems relatively simple to make sure nothing alive enters other than what has been grown in upstream reactors. Recursively apply these principles down to the last cell inserted into the process inside a (small) clean room.

plaidfuji 5y ago
What you’re describing (at least the heat aspect) is referred to as “Sterilization in Place” and it is standard practice to design into bioreactors already. Contamination still happens. In general you’re right though, the concept of continuous automated sterilization will need to be reimagined, but like many of the topics addressed in this article, a lot of the low hanging fruit has been picked. https://www.flottweg.com/wiki/separation-technology/sterilization-in-place-sip/ https://www.flottweg.com/wiki/separation-technology/steriliz...
- yholio 5y ago
  The description linked talks about a distinct sterilization phase, where hot steam is passed trough the installation prior to use. It seems exactly like the type of situation described in the article, where small welding gaps allow bacteria to stay alive - I would imagine, especially if the surrounding metal has enough thermal mass. My suggestion was a bit more nuanced, heat the very body of the reactor + reaction medium; but I could be venting hot air. Nonetheless, it's not a fundamental, thermodynamic limit, it's not like bacteria spontaneously materialize inside vats above a certain volume. It's a technical problem to solve, it seems solvable and once you solve it, all 100 ton reactors can use that design.