I’m glad you wrote this review, now I don’t have to :-)
(I agree that it’s an excellent book and heartily endorse it.)
Nick Lane thinks the reason is because genes have to be close to a cell membrane in order to make ATP. The square-cube law means that a cell’s membrane grows slower than its volume. A big archaeon is limited in how much ATP it can create which limits how many genes it can have which limits its complexity.
Among the three domains of life, eukaryotes are the most complex. By offloading ATP production to mitochondria, they get around the square-cube law limiting ATP production. They can have more genes which enables more complexity.
The way I remember his argument is a bit different, and not involving the square-cube law: (1) if some little section of a membrane is producing ATP, it needs a nearby copy of all the genes for ATP production; (2) if there’s a big complicated cell doing lots of stuff, it needs a lot of surface area for producing ATP; (3) It follows from (1) + (2) that if there’s a big complicated cell doing lots of stuff, it needs a gazillion copies of its genome per cell; (4) if the genome is really huge, then it’s prohibitively expensive to make a gazillion copies of it per cell; (5) therefore genomes can’t be huge; (6) therefore cells can’t be too complicated.
The reason eukaryotes avoid this dilemma is that they have two genomes. One genome is tiny, and has just the genes for ATP production, and there are gazillion copies of that genome per cell. The other genome is huge, with massive numbers of genes, but it’s OK because there’s only one copy of that genome per cell.
So the way I remember the story, it doesn’t involve the square-cube law. Can’t a bacteria increase the effective surface area by, say, putting ATP production machinery onto a vesicle? It’s been a while since I read the book, I could be misremembering. :-)
I’m glad you didn’t write this review, now I got to instead :-)
I could be wrong and I would like to be corrected if I am. From your writings and from having met you, I would not be surprised if you know far more biology than I do.
(3) It follows from (1) + (2) that if there’s a big complicated cell doing lots of stuff, it needs a gazillion copies of its genome per cell;
Why can’t bacterium/archaeon just have many copies of the same ATP gene on a single genome? Duplicating a DNA segment is not an uncommon mutation.
On the other hand, packaging ATP genes in mitochondria lets a cell create more or less of them on demand, something it can’t do with a repeated sequence in DNA.
Can’t a bacteria increase the effective surface area by, say, putting ATP production machinery onto a vesicle?
If I understand correctly, mitochondria use cristae for this purpose instead of vesicles. Cristae are what gives mitochondria membranes their characteristic shape.
I think the “nearby” part is important … I thought his claim was basically that if I’m an ATP production machine, I need the genome to be really close to me distance-wise (like within X nanometers, for some number X that I don’t know), so that if I have a broken part then a replacement can be quickly manufactured on demand in the right location. So having 100 copies of the gene that are physically attached to each other doesn’t help at all towards solving the problem, and in fact makes the problem worse.
I really don’t know much biology, I’m just going off my memory of the book :-)
I was also halfway through a review of this book. Since I’ve only met one other person who’d read it I thought it was unlikely anyone else would! I guess LWers have more similar interests than I would have predicted.
Me too. I’d hate to find out Jemist’s work has gone to waste. A second review could add to what I have here. I skimmed at least half the book including sex, biochemistry and why eukaryotes appeared exactly once. I came to this book with my own perspective. Someone with a different perspective could draw different main ideas from it. Besides, I like the idea of more biological content on LW.
I’m glad you wrote this review, now I don’t have to :-)
(I agree that it’s an excellent book and heartily endorse it.)
The way I remember his argument is a bit different, and not involving the square-cube law: (1) if some little section of a membrane is producing ATP, it needs a nearby copy of all the genes for ATP production; (2) if there’s a big complicated cell doing lots of stuff, it needs a lot of surface area for producing ATP; (3) It follows from (1) + (2) that if there’s a big complicated cell doing lots of stuff, it needs a gazillion copies of its genome per cell; (4) if the genome is really huge, then it’s prohibitively expensive to make a gazillion copies of it per cell; (5) therefore genomes can’t be huge; (6) therefore cells can’t be too complicated.
The reason eukaryotes avoid this dilemma is that they have two genomes. One genome is tiny, and has just the genes for ATP production, and there are gazillion copies of that genome per cell. The other genome is huge, with massive numbers of genes, but it’s OK because there’s only one copy of that genome per cell.
So the way I remember the story, it doesn’t involve the square-cube law. Can’t a bacteria increase the effective surface area by, say, putting ATP production machinery onto a vesicle? It’s been a while since I read the book, I could be misremembering. :-)
I’m glad you didn’t write this review, now I got to instead :-)
I could be wrong and I would like to be corrected if I am. From your writings and from having met you, I would not be surprised if you know far more biology than I do.
Why can’t bacterium/archaeon just have many copies of the same ATP gene on a single genome? Duplicating a DNA segment is not an uncommon mutation.
On the other hand, packaging ATP genes in mitochondria lets a cell create more or less of them on demand, something it can’t do with a repeated sequence in DNA.
If I understand correctly, mitochondria use cristae for this purpose instead of vesicles. Cristae are what gives mitochondria membranes their characteristic shape.
I think the “nearby” part is important … I thought his claim was basically that if I’m an ATP production machine, I need the genome to be really close to me distance-wise (like within X nanometers, for some number X that I don’t know), so that if I have a broken part then a replacement can be quickly manufactured on demand in the right location. So having 100 copies of the gene that are physically attached to each other doesn’t help at all towards solving the problem, and in fact makes the problem worse.
I really don’t know much biology, I’m just going off my memory of the book :-)
He definitely did say something to that effect and it definitely is easier to have the genome near the cell wall of a small cell than a large cell.
You say “the genome” but note that one bacterium (i.e. one cell) can have more than one copy of its entire genome inside it, e.g. “many bacteria harbor multiple copies of their genome per cell”, “Enormous bacterium uses thousands of genome copies to its advantage”, etc. That’s what I was (implicitly) referring to. :-)
I did not realize that. Whoops.
I was also halfway through a review of this book. Since I’ve only met one other person who’d read it I thought it was unlikely anyone else would! I guess LWers have more similar interests than I would have predicted.
I suppose I’ll review another book instead!
I’d love to see two reviews of the book if you feel like it!
Me too. I’d hate to find out Jemist’s work has gone to waste. A second review could add to what I have here. I skimmed at least half the book including sex, biochemistry and why eukaryotes appeared exactly once. I came to this book with my own perspective. Someone with a different perspective could draw different main ideas from it. Besides, I like the idea of more biological content on LW.