I'm not sure I completely understand your question. One part seems to wonder why mtDNA is kept but then the last part of your question seems to suggest you are asking something different. Also, which 'micronucleus' are you talking about which sits outside mitochondria? If you could clarify I hope I can help.

I'm with Mark on this. Instead of asking why the mitochondrial genome is exclusive, the question is more on why there are still DNA left in mitochondria in light of the higher mtDNA mutation rate and limited repair capability in the organelle. Over time most of the mitochondrial genome has been re-located to the nuclear genome. The remaining ones are presumed to produce proteins of extreme hydrophobicity and therefore could not be delivered into the organelle if they are coded by the nuclear genome. There is some work going on to re-locate the last few genes to the nuclear genome funded by the SENS Foundation.

Why micronucleus didn't integrate into the mtDNA genome? Actually, the reverse is evolutionally preferred in that most mitochondrial genes have integrated in to the genome of eukaryotic cells. The 13 proteins and 22 tRNAs encoded by mtDNA are essential to OHPHOS, and, as such, it is believed that they are retained in the mitochondrial genome (prokaryotic) to facilitate the efficient functioning of that process.

Many thanks to you all! Micronucleus is the erratic nucleus formed during the anaphase of mitosis or meiosis. It's often evident in cells under oxidative stress or in some cancer cell lines. It's a small piece of the chromosome drifting around in the cytosol.

What I aim to ask is that since DNA is much smaller than peptides in the cytosol, could the micronucleus pass into the mitochondrial matrix and recombine with mtDNA? So far no evidence showed that, so why can't?

Is the outer membrane of mitochondria prevent the enter of DNA such as micronucleus? Or the exogenous DNA is effectively degraded in the matrix?

The same questions could also apply to the RNA in cytosol.

Plant mitochondria take up both DNA and several essential tRNAs, and you find lots of pieces af plastid and nuclear-origin DNA in the plant mtDNA. However, mammalian mtDNA is very different from most organisms in being extremely slim, as if an active process constantly eliminate non-essential pieces.

Thanks Zhou Xin, I didn't know that about this micronucleus. As others said, there is a lot of evidence indicating that mtDNA is moving into the nucleus. Either 'historically' as part of the original endosymbiosis (endosymbiotic gene transfer) and later on in the form of Numts (small bits of mtDNA ending up in the nucleus). So, if this micronucleus DNA is floating around in the cytosol, I'd guess it would be most likely to end up in the nucleus if anywhere at all. Also, remember, there is only one nucleus that stays with the cells but many mitochondria. So, incorporation into mtDNA is more likely to end up in a dead end as mitochondria 'die' as the cell can afford to lose some. Obviously, if the nucleus 'dies', then it is game over for the cell. So, the nucleus is a much safer place for DNA to end up.

However, as a response to Xinkun Wang, current evidence suggests that the hydrophobicity theory might not be correct. For example, mitochondrial carrier proteins are extremely hydrophobic and consist (only) of six transmembrane spanning regions. It is nuclear encoded and has no problem coming from the nucleus and inserting itself into the inner membrane. A theory that is gaining acceptance is CoRR for co-location of redox regulation. It was postulated by Professor John Allen and there is a Wiki page explaining the theory briefly: http://en.wikipedia.org/wiki/CoRR_hypothesis. His papers explain it obviously in more detail: http://rstb.royalsocietypublishing.org/content/358/1429/19

(One disclaimer, I have published with John)

I know that micronuclei generally degraded in cytoplasm in a short times like 2 and 3 days..

To Dear Mark: Thanks. Very informative answer. I don't doubt that mtDNA is evolutionary favored “output” of its codes in mammals and birds. There is hypothesis assuming that mtDNA in mammals and birds tend to be “slim”, so allowing a prompt response to the changing physiological state of mammals and birds (such as sprint, external temperature changes). Also, smaller mtDNA genome renders its advantage during replication.

But to another aspect, if we look into the individuals during aging process, the question remains. As man grows old, the heterogeneity in the mtDNA increases. But as mini insertions, deletions, base substitutions and large deletions are comprehensively reported; there are no reports about large insertions in human. If micronucleus could recombine with mtDNA by chance, at least certain level of nuclear DNA insertions could be observed.

To Dear Allan: Many thanks! So that the passive barrier of mitochondrial membrane is out of the question. Now we need to know how mitochondria to distinguish the exogenous DNA and mtDNA; and how to prevent the combination of nDNA and mtDNA.

To Dear Nilufer: Thanks. I think 2 and 3 days is long enough.

Zhou: Ref for plants mito-uptake of DNA is Koulintchenko et al. The EMBO Journal Vol. 22 No. 6 pp. 1245±1254, 2003. It has been shown for mammals later, but I cannot recall the ref.

Mark: john Allen's theory should certainly be valid for chloroplast, but for mitochondria with their fusion and fission all the time, there is no real individuality, which the theory depends on. PS. i do not agree with "the nucleus dies". The nucleus dissolves every cell division, and is not really a stable organell, as I see it.

One more thing. Due to either RNA editing or changes in the genetic code, there appears to be no eukaryote where gene sequences can be functional in both the nucleus and the mitochondrion. This will slow down evolutionary movement of genes between compartments.

Regarding ageing of mtDNA and the accumulation of deleterious mutations, it are not the male mtDNAs that are the template for the next generation. The maternal cells destined to become eggs are dormant very early on and their mtDNA are transcriptionally not active and do not accumulate mutations and are therefore the perfect template for the next generations. See: http://www.ncbi.nlm.nih.gov/pubmed/24068653

Thanks, Allan! The reference you provided is very helpful to me!