Re: Power, Sex, Suicide: Mitochondria and the Meaning of Life

Vend wrote:


Michael Olea wrote:

... rerun the tape and something like 9
times out of ten you get bacteria and nothing else.

Are you sure? Games Theory shows that in many situations individuals
can gain a significant advantage if they can follow a cooperative
strategy. Thus, environments which exert a strong pressure on
unicellular organisms to organize in cooperating colonies (the
ancestors of multicellular organisms) are probably not unlikely.

Not 100% sure, but reasonably sure - more sure today than I was a week ago.

The argument, in broad outline, runs like this: bacteria occur very early in
the fossil record, almost immediately after the period of saturation
bombardment*, but complex metazoans do not appear till more than 3,000
million years later. So making bacteria is "easy", and making metazoans is
"hard", either because 1) it depends on the chance occurance of low
probability events, or 2) because of some "bottle neck" in the
preconditions that took a long time to develop - for example the build-up
of oxygen in the atmosphere.

The evidence for the "low probability events" scenario is now pretty strong
- much stronger than it was in 1994, the time of wrting of Kauffman's book,
and stronger even than it was in 2001, the copyright date of Ridley's book.

The first point is that all complex metazoans, all plants, animals, and
fungi, are eukaryotes (their cells are eukaryotic cells). Secondly, all
known eukaryotes either have mitochondria, or once had mitochndria and
subsequently lost them. The second point is important to working out the
evolutionary history of the eukaryotic cell. There are about 1000 known
species of single celled eukaryotes that do not have mitochondria. Most of
them are parasites, or live in extreme conditions. At one time these looked
like good candidates to be modern descendents similar to some common,
mitochondria-free host, ancestor of the the eukaryotic line. One by one
they have proved not to be such - these eukaryotes have descended from
eukaryotes with mitochondria, and retain evidence of their
mitochondrial-hosting past (e.g. mitichondrial DNA that has migrated to the
nucleus).

Nowadays, I gather, there is just about universal agreement among
microbiologists, molecular biologists, and biochemists, on the
endosymbiotic origin of mitochondria in eukaryotes. In fact, mitochondria
are clearly related to alpha-proteobacteria. But the presence of
mitochondria within the eukaryotic cell is just one of many profound
differences between eukaryotic and prokaryotic cells. Discriminating
between "bottle-neck" and "low probability event" scenarios depends on,
among other things, details of the order and timing in which these
differences arose. One approach is to find which genes are common to all
eukaryotes and which occur only in subgroups. To make a long and
interesting story short: "This was the approach employed by Maria Rivera
and her colleagues at the University of California, Los Angeles, published
in 1998 and in more detail in Nature in 2004." The common genes fall into
two groups: "operational genes" inherited from alpha-proteobacteria - so
mitochondria were present from the begining, in the common ancestor of all
eukaryotes - and "informational genes" inherited, as expected, from a group
of archea, but, shockingly, from the specific and completely unexpected
group: *methanogens*, "those swamp lovers that shun oxygen and produce the
marsh gas methane."

Why is that "shocking"? Mitochondria do many things, but one of the more
significant things is respiration - they manufacture ATP, the main energy
currency of all cells. They are oxygen breathers. But oxygen is toxic to
methanogens. They get their energy by generating methane from hydrogen and
CO2. Hydrogen can only be found in low oxygen environments (since oxygen
oxidizes hydrogen into H2O). So how does a merger between an oxygen-hating
methanogen and an oxygen-loving alpha-proteobacter come about? Where would
they meet?

The answer, according to "the hydrogen hypothesis" is that they don't. The
bacterial ancestor of the mitochondria would have to have been an energy
"jack of all trades", like Rhodobacter (which, by the way, is an
alpha-proteobacter). Enter the "hydrogenosome", mitochondria-like
organelles that breathe anaerobically, and generate hydrogen and CO2 as
waste products - the life-blood of the methanogens. It turns out, according
to the genetic evidence, that the mitochondria and the hydrogenosomes share
a common ancestor, which, like the Rhodobacters of today, would perforce
have been metabolic sophistictates, that "could respire using oxygen or
other mloecules, or generate hydrogen gas, as the circumstances dictated."

So in the anaerobic sludges in the deep ocean, methanogens ekeing out an
anaerobic existence, encounter Rhodobacter-like bacteria in hydrogen and
CO2 generating mode and cozy up like white on rice - une belle matriminio
bilingue. Just how this association becomes one of endosymbiosis is
certainly open to question, but there do not seem to be any major obstacles
- the fact is, it seems, that it did happen, roughly 2000 million years
ago, right around the time of a sharp increase in atmospheric oxygen (as
evidenced by, among other things, the appearance of red-beds in the
stratigraphic record). But now, equipped with its symbiont capable of
breating oxygen, the methanogen host is no longer tied to its anerobic
environment. It can venture into oxygen-rich waters, reap the benefits of a
far more efficent metabolism, and conquer, or rather construct, the
metazoan world.

But there is more than one catch - the timing of events has to play out just
right. Oxygen has to be sufficiently scarce to drive some Rhodobacter-like
bacteria into the anaerobic mud, where methanogens are established in
sufficient numbers to lead to a merger, which certainly must take some time
for the chimeric "hopeful monster" to consolidate to the point where it can
venture out of the mud, yet must not take so long that the genes for oxygen
base resperiation are jettisoned - a case of use-it-or-lose-it. Maybe this
point needs a comment: retaining genes bears a metabolic cost, not to
mention slowing the rate of reproduction. So there is selection pressure
against genes not used. The hydrogenosomes, for example, have lost the
ability to breath oxygen, and the mitochondria have lost the ability to
breath anaerobicaly. Stay too long in the mud, and our progenitors would
have been tied to it forever. Don't stay long enough and there is no
mereger. The sequence of events is delicate: there had to be enough oxygen
in some environments for oxygen-based resperiation to evolve in bacteria,
but scarce enough to drive some versatile metabolizers into the mud, where
they form an alliance with an archeon, itself driven to its niche by having
been outcompeted by the rise of sulphate-reducing bacteria, remain there
just long enough to consolidate a mereger, but not long enough to
stream-line it, so as to be able to exploit the sudden, fortuitously
well-timed, sudden rise in available oxygen. In the interplay between
chance and necessity, chalk this one up to chance.

But wait, there's more, way more. In the 4,000 million year history of
bacteria the eukaryotes arose exactly once. If metazoans were a likely
outcome, whose long delay was due only to the slow development of
preconditions - a bottleneck, but not a low probability coincidence of
chance events, then where are the competitors? What of convergent
evolution? Flight evolved independently 4 times: pterodactyls, insects,
birds, and bats. Sight evolved independently some 40 times. Sonar evolved
independently in dolphins and in bats. Eukaryotes, the ancestors of all
complex metazoans, birds, bats, moss, grass, sponges, mushrooms, and
lizards, evolved only once in 4,000 million years. But prehaps there were
many proto-metazoans, outcompeted and driven to extinction by the
eukaryotes? Not likely. THye would have to have dominated every single
niche, driving every single alternative into an extinction so profound it
left no record. This rarely happens at taxenomic levels higher than a
class, let alone a *domain*. Sulphate-reducing bacteria outcompeted
methanogens for available hydrogen, yet, thank our lucky stars, the
methanogens found low sulfer niches to inhabit. Mammals, after the
Cretaceous-Tertiary boundary event, out-competed reptiles, as did birds,
but reptiles are still with us. There is no trace anywhere of any sort of
eukaryote that does not hale from the methanogen/alpha-proteobacter merger.
And unlike sight, flight, sonar, wing, fin, tail, teeth, and claw, there
have been NO new entrants in the metazoan game since that chance event long
ago, not one in 4,000 million years.

So, much as I admire Kauffman's autocatalytic sets, the interplay between
selection and self-organization, order for free at the edge of chaos, and
evolutionary stable strategies, when he chants "we, the expected", if he
means life, microbial life, then yes, I agree; but if he means metazoans,
then I put my money on "we, the improbable", not alone, perhaps, but rare.
Roll the dice, re-run the tape, play it again, Sam, and something like 9
times out of 10 you get microbe world. Of this I am reasonably certain.

* "Saturation bombardment" happens when the rate of impact (by asteroids and
comets) is so high that it destroys as many craters as it creates. This was
the case in the lunar highlands (as opposed to the maria). The highlands
were so intensely bombarded that every new impact destroyed as many craters
as it created. In fact, the bombardment was so intense that it completely
melted the lunar crust, creating a magma "ocean" about 100 km deep. The
highlands consist almost entirely of a single mineral, a Ca-rich
plagioclase called "anorthosite". The lunar maria, on the other hand,
genrally only slightly younger than the highlands, are so smooth as to have
at one time been mistaken for oceans. They are flood basalts, not unlike
the Deccan traps of Africa and Brazil, or the Modoc plateau of the
northwestern US, or the entire oceanic crust. Indeed, the moon is, except
for the complete loss of volatiles, almost identical in bulk and trace
element, and isotopic composition, to the Earth's mantle - it is a hunk of
earth mantle. The fact that the maria are smooth, the highlands saturated
by craters, and that the oldest maria are close in age to the youngest
highlands means that the transition from intense bombardment to a slow
meteoric drizzle was sudden and swift. almost like a phase transition. The
period of intense bombardment lasted from about 4.5x10^9 to 4.0x10^9 years
ago. The oldest known bacterial fossils (though there is some contention
about this) are about 3.85x10^9 years old. Life on Earth was around almost
as soon as the rain of impactors let up. At that time, by all best
evidence, conditions on Venus, Earth, and Mars were very similar...

Cheers,
Michael






.