Aubrey's "survival of the slowest" hypothesis

Aubrey de Grey ag24 at
Fri Apr 13 14:11:43 EST 2001

Paul Brookes wrote:

> Why is it so hard to swallow that ATP availability might not be the 
> only signal for mitochondrial proliferation.

I don't think it's hard to swallow.  I do think ATP availability is a
pretty good candidate though (see below).

> In fact, when you think about it, it's a pretty poor signal, as
> mitochondria are not the only things that make ATP, and steady state
> ATP level doesn't really change that much, even in a dynamic tissue
> like the heart (Bob Balaban's NMR studies show this quite neatly).
> Certainly ATP does not change over the orders of magnitude that an
> archetypal cell signal would change over (think of calcium,
> phosphorylation,or caspase activity as examples - at least 10 fold
> changes involved).   I'm unaware of an ATP-sensitive factor that can
> sense 1.2-1.3 fold changes in concentration.

All agreed, but all rather circumstantial I'd say.  The example of
calcium is a good one, because of course mitochondria are responsible
for cellular calcium homeostasis: thus, reduced mitochondrial numbers
would impair that homeostasis, so spikes in calcium concentrations are
also a candidate trigger for mitochondrial biogenesis.  But one of the
best reasons to favour ATP availability is the hyperproliferation of
mitochondria that Iuval alluded to.  Since mutant mitochondria are
still capable of generating a reasonable proton gradient, and are thus
able to continue in all their mtDNA-independent functions such as Ca++
homeostasis, this hyperproliferation would be rather paradoxical if
the reason weren't ATP shortfall.  One could argue that other functions
are compromised even if they're not abolished, but that's a bit weak.

> As for what the signal is, why does it have to be the cell itself that 
> controls mito' number.   Why not imagine mitochondria as the endosymbiont 
> bacteria they are, all competing for food, and having autonomous control 
> over their own replication - just like bacteria in a culture dish.  The 
> cell can only support a certain number of mito's, and certain cells, 
> because they supply more substrates and remove more waste products, can 
> support a larger population of mito's.  Thus, it is not so much that a 
> signal tells mitochondria to replicate, but that mitochondria will 
> replicate un-checked if left to do so, and the important signal is the
> one telling them to stop.

Right; exactly this was proposed by Shoubridge back in 1990 (Cell 62:43)
as an explanation for clonal expansion of mutant mitochondria.  Actually
he proposed two alternatives.  The first was that there are sequences in
the mtDNA that are mediators of inibition of replication, and when those
sequences are mutated/deleted you get expansion.  Doesn't really work,
because we now know of point mutations all over the mt genome that are
expanded.  The other idea was that mitochondria replicate when they want
to: slightly unlike what you suggest, the mechanism was supposed to be
that they replicate themselves out of trouble, i.e. that when they get
into a sub-optimal state (maybe low matrix ATP, maybe high membrane
damage) they divide and that gives daughters that are healthier (due to
more ATP synthesis capacity or dilution of damage, respectively).  But
this relies on the availability of "food" in the form of nuclear-coded
mt proteins, which are not (at least not all) present in the cytosol in
advance of mt biogenesis.  An additional problem with the whole idea of
mitochondria dividing autonomously is the hyperproliferation phenomenon:
why would they have more capacity to divide when the cell was devoid of

> Has anyone looked for repressors of mitochondrial replication?

Not that I know of.  Interesting idea.

Aubrey de Grey

More information about the Ageing mailing list