Phil Osborn wrote:
I had a couple of problems with that right off. Not that I challenged their
data, just that it raised some obvious questions. How is it that the
mitochondria in the cells in general mutates so much while the mitochondria
passed on in the egg does not? (If it did, then each generation would have
more defective eggs, which does not generally appear to be the case.)
### Actually, as far as I know, the germ line mitos degrade as well, but
this is a stochastic process - if you have 450 000 oocytes (the approximate
number in a female newborn, if I remember correctly), some of them will by
chance keep all of their mitos intact. During the maturation of an oocyte
there are checkpoints, eliminating all the oocytes with insufficient
energy-producing capability by apoptosis. This is not a fully foolproof
method, as there are some specific mitochondrial mutations (as in MELAS,
LHON) which apparently escape this mechanism, resulting in maternally
inherited "classical" mitochondrial diseases.
Secondly, why is it that the proportion of defective mitochondria
increases? What happens to select for them as opposed to selecting for
cells with the best mitochondrial genes?
### It is possible that there is not so much a selection for mutated mitos,
but rather an absence of intracellullar selection for the healthy ones
(except in the germ line). Mutations keep accumulating, because every now
and then a normal genome gets hit but the mutated ones do not spontaneausly
revert to wild type. Even with 10 000 mtDNA genomes per cell, after some
time the fraction of mutated copies is high enough to cause diminished
energy output, and eventually death of the cell (usually by apoptosis). If
this process goes on in postmitotic tissues, the result is e.g. Parkinson's
disease and possibly even Alzheimer's disease.
One hypothesis that I arrived at finally, which I would appreciate comment
on, is that a higher level of energy production for a cell means a shorter
doubling time and consequently a faster arrival at the Hayflick limit.
I.e., assuming that cellular reproduction occurs in response to a signal and
that the response likelihood is dependent upon available energy - which may
not be the case, but it does seem reasonable - then those cells with an
abundance of energy would be more likely to respond. Thus, over time, cells
that have poorer mitochondria tend to dominate, simply because their
telomeres are longer, having not divided as many times.
### I think that so far there was no clear proof that the Hayflick limit is
important for human aging. The most quickly dividing cells (as in the small
bowel, or the hematopoietic system) are not most limiting for survival -
this seems to be most affected by postmitotic cell attrition, in the brain,
skeletal muscle, and myocardium, as well as problems with the nuclear genome
(leading to cancer). It is possible that all you need is the accumulation of
mtDNA mutations, and the absence of a strong selective mechanism *for*
healthy ones (as in the germ line).
If this is the case, it is interesting to speculate upon the effects of
different populations of mitochondria. If one had only mitochondria of a
particular kind, how would this influence lifespan, as opposed to having a
variety of widely varying mitochondria, ranging from high output to useless.
A person with a wide variety of mitochondria might be expected to age rather
dramatically as various cell populations with varying proportions hit the
Hayflick limit early or late. A person with only one breed of mitochondria
could be expected to age uniformly, with less likelihood of catastrophic
###There are dramatic examples of variability in the severity of
mitochondrial disease between siblings - they can inherit varying ratios of
the mutated genomes (it's called heteroplasmy), so that some can die as
infants while others survive into adulthood. Variability between tissues in
an individual is also observed, and this is possibly another reason for the
heterogenous presentation of mitochondrial disease.
Rafal Smigrodzki, MD-PhD
This archive was generated by hypermail 2b30 : Sat May 11 2002 - 17:44:14 MDT