FWIW - I really do not expect a bacterial enodsymbiont that has been
passed down in isolation from other bacteria and in intimate contact
with its eukaryotic host for billions of years would look exactly
like the bacterial group from where it came. I think it is fortunate
that we can recognize it as being bacterially derived, not just
that it does not fit with the rest of the eukaryotic sequences.
>Interesting that those espousing these theories of endosymbiosis do
>not look at the differences between bacteria and mitochondria or
>chloroplasts. How many of you have ever read a paper on this aspect?
Actually, many of the issues you bring up are mentioned in the
Cavalier-Smith paper I referenced. Also, I would be greatly indebted
if you could help me locate recent papers in the scientific literature
that actively dispute the endosymbiotic theory of mitochondrial and
chloroplast origins. As time and interest has permitted, I've been
casually looking for such papers myself.
>What about just for instance, the presence of two very specialized
>membranes in the organelles whereas bacterial membrane is single
>and also very specialized, but very different from the organelles.
>I suppose it would make as much sense to suggest the nucleus was
>also an endosymbiont, or maybe that a bacterial cell infected
>another bacterial cell, and thus the double membrane, and the
>cytoplasm just accumulated in between the two membranes, and thus
>the origin of eukaryotes...further infection by other bacteria infected
>by other bacteria, resulted in the double membraned organelles. (This
>theory is copyrighted, just in case it catches on).
*grin*
Nice idea, but it might already have been done. Lynn Margulis is
the one who tosses in endosymbiotic events left and right. But
returning to mitochondria and chloroplasts...
Gram-negative bacteria, which are the proposed progenitors of the
mitochondria and chloroplasts, already have _two_ membranes, an inner
and an outer. Gram-positive bacteria, with which these organelles
do not group, are the ones with a single membrane. Further, there
is some question whether the outer membrane of the mitochondrion
was derived from the outer membrane of the gram-negative progenitor
or whether it came from a food-vacuole membrane. When colonizing
the interior space of a eukaryotic cell, bacteria can live either
within these vacuole spaces (ie. within a eukaryotic-derived membrane)
or freely within the cytoplasm. Basically, I do not see the number of
membranes or their nature as being problematic for the theory. As to
the evidence for the origins of membranes, Cavalier-Smith mentions
(p 97 of his paper that I cited previously) that "the chloroplasts
of Cyanophora and most other gluacophytes still have the peptidoglycan
murien between their two outer membranes." Granted, this trait is not
found in all chloroplasts, but its loss can be understood as a later
derived trait and its presence is strikingly indicative of its bacterial
origins. Recall in another letter that I presented the example of algae
(cryptomonads being one group) where it was found that their chloroplasts
have an extra (third) membrane. From this it was proposed that this
membrane arose as a result of a second endosymbiosis event between two
eukaryotes (one of which carried a chloroplast). Later analysis of the
small nucleus-like organelle (called a nucleomorph) found in the space
between the inner and outer plastid membrane pairs placed its origins
among the eukaryotes and helped confirm this proposal. This is a
potential example of a endosymbiotically-derived organelle undergoing
the process of "coding absorption" into its host's chromosomes.
>While neither I nor anyone else on this planet has a clue yet why
>mitochondria make a few proteins on their own,
...or especially, why their components need to resemble those of specific
bacterial groups if they _didn't_ arise from them...
>the answer may lie in the specialized nature of these proteins. Most
>of them form subunits of larger polymeric proteins, whose other
>subunits are cytoplasmic proteins encoded for in the cell nucleus.
>Without, for the moment worrying about how such a feature could be
>orchestrated between an endosymbiont and its host, lets look at what
>is involved in getting these cytoplamic components into the
>mitochondrion.[list removed...]
Again, this is discussed in the Cavalier-Smith paper and in the
references cited by him. Note that in many symbiotic interactions
with bacteria (or eukaryotes, FWIW), signalling and transfer mechanisms
already exist and are certainly highly tuned and coordinated. This
area is probably being most hotly pursued in the field of bacterial
pathology and in the study of nitrogen-fixing symbiosis with crop plants.
>One question we might want to ask the endosimbiont believers is the same
>question I posed above: why do the organelles make any proteins at all?
>why not transfer all of the protein synthesizing process to the nucleus?
>The nucleus carries about 100 genes specifically required just for the
>organelle to be able to synthesize the proteins that it makes, an
>exceedingly costly arrangement. To quote from Alberts, et. al. "The
>Molecular Biology of the Cell"
>
> "The reason for this costly arrangement is not clear, and the
>hope that the nucleotide sequence of the mitochondrial and chloroplast
>genomes would provide the answer has proved unfounded. We cannot think
>of compelling reasons why the proteins made in mitochondria and
>chloroplasts should be made there rather than in the cytosol.
> [This whole problem] is difficult to explain by any hypothesis
>that postulates a specific evolutionary advantage of presentday
>mitochondrial or chloroplast genetic systems." p715.
is a cost advantage to maintaining separate genetic systems. As you
mentioned in your reply, there remains the possibility that some
things might simply be impossible (or extremely difficult) to transport
into these organelles. Whether this is an absolute barrier or something
like a frozen accident is difficult to assess, particularly with respect
to the mitochondria, whose genome sizes can vary significantly between
groups. This is discussed briefly in the paragraphs below the section
you cited of Albert's (et al) textbook (It's p. 400-401 in the second,
1989 edition). Note however, that he does not dispute the theory of
endosybiosis but instead whether there is an advantage to maintaining
separate genomes.
I think we should make clear that this also does not mean there is no
advantage to having mitochondria or chloroplasts in the first place
(where would we be without wood for a cozy fire on a winter's night? ;^),
but that it is not clear why all the genes haven't moved to the
eukaryotic chromosome. FWIW, At least many (most?) of them have been
moved so far.
Regards, Tim Ikeda (timi@mendel.berkeley.edu)