I know that mitochondria of eukaryotes have their own DNA, more similar to that of bacteria than to the rest of the cell's DNA. I also know that a cell can have plenty of mitochondria, and I understand each of them has it's own DNA. Is it the same DNA?

I know that when a human is conceived, the mitochondria of the sperm are destroyed and don't contribute to the new cell. If there were no mutations and the process started with a cell with homogenous mDNA, that would mean that the mDNA not only within one cell but among all of the descendants was the same. But there are mutations because we have different mDNAs. Then shouldn't the mutations cause variety within single cells? It's hard for me to believe that there can even be a single human cell in the world that has homogenous mDNA because, if I understand correctly, an ovum retains many mitochondria during conception, not just one, so at least some of the variety within the ovum should be passed over to the conceived human. But perhaps it's extremely difficult for a cell with variously DNAed mitochondria to survive? Can it be estimated what the percentage of human cells with homogenous or nearly homogenous mDNA is across the world?

I'm asking this because I've heard that the mDNA can be considered another chromosome. I wonder if then every single piece of mitochondrial DNA should be considered another chromosome.

Please note that all I know about biology is what I learned in high school several years ago and the bits I know from Wikipedia and other internet sources. I would like to ask for possibly accessible answers.

EDIT: Reading up on this, I came across the term heteroplasmy. So it does happen that mitochondria of an organism can store different genetic information. But the article seems to suggest that it is an anomaly rather than the normal course of things. That seems strange to me. I understand of course that some mutations have to be harmful, but shouldn't some mutations and some kinds of in-cell variety turn out to be neutral or even advantageous?

I've also just read that a mitochondrion stores more than one copy of its DNA. That seems to allow for variation even within a single mitochondrion.

EDIT 2: I think it's possible that my question is unclear. There seem to be four levels on which mDNA variety can be considered: a species, an organism, a cell, a mitochondrion. I ask the answerers to focus on the third one. I will be happy to learn about the other levels too and I will consider such information on-topic, but my main question is about what happens in a cell.

  • $\begingroup$ Why would be mtDNA any different from nuclear DNA in that sense? They are both sensible to mutations. $\endgroup$
    – nico
    Aug 19, 2012 at 18:02
  • $\begingroup$ @nico Yes, of course there are mutations in mitochondrial DNA. Otherwise there would be no mDNA variety across the human population. But nuclear DNA is different from mDNA in that the latter occurs in very many (possibly thousands as I understand) copies within a single cell so there is room for variety within that single cell. That's impossible with nuclear DNA. $\endgroup$
    – ymar
    Aug 19, 2012 at 18:08
  • $\begingroup$ sure, but what I wanted to say is that by the same principle you may as well ask "does every nucleus contain the same DNA"? Surely, having thousand of copies increases by 3 orders of magnitude the probability of having mutations, no doubt about that. I am wondering, however, if there are repair mechanisms for mtDNA... I would assume so but I am not sure $\endgroup$
    – nico
    Aug 19, 2012 at 18:12
  • $\begingroup$ @nico That wouldn't be exactly the same kind of question, although certainly related. I'm asking about what happens in a single cell. There is only one nucleus in a single cell so your question has to be about a larger structure like an organism. I will edit the question to make it clear. $\endgroup$
    – ymar
    Aug 19, 2012 at 18:20
  • $\begingroup$ oh, OK, I am sorry I read the question too quickly and I misunderstood what you were asking :) $\endgroup$
    – nico
    Aug 19, 2012 at 18:25

1 Answer 1


This link seems to have good information that answers most of your questions.

In my mind, there are two types of mitochondria: ones that work and ones that don't. Mitochondria do have DNA but that mDNA is there to encode proteins for their specific functions (e.g. to create ATP). So, although the mDNA may not be uniform for every mitochondrion in your body, it is most likely functional unless you have a mitochondrial disease. If you do have a mitochondrial disease then it suggests that some of your mitochondria are functional and some are not (otherwise you would be inviable).

As discussed in the comments, all DNA undergoes mutations so that even identical mDNA genomes may vary after undergoing some mutations. It's possible that only one genetic line of mitochondria was passed to you from your mom. In this case all the mitochondria in your body would be identical (except for minor mutations). It's also possible that your mom inherited and passed on to you more than one genetic line (as is the case in viable mitochondrial disease).

Mitochondrial repair mechanisms do exist but they differ in various ways from nuclear DNA repair and may be more prone to damage.

  • $\begingroup$ Thank you. The first document seems to be concerned only with the mutations that make genes faulty. But surely there must be mutations that are not harmful to the cell or the organism? Is it more difficult for an mDNA mutation to be neutral or advantageous than for nuclear DNA mutations? Is it common for a cell to have several different but (nearly) all perfectly good sets of mDNA in its mitochondria? $\endgroup$
    – ymar
    Aug 19, 2012 at 19:49
  • 1
    $\begingroup$ 1) You're correct, there are mutations that are not harmful to the mitochondria or cell. 2) Mitochondrial mutation is probably more likely to be harmful than nuclear DNA mutation since it is much less vast. Nuclear DNA is much larger and thus there is less change to mutate something important. 3) I don't know how common it is, but it is possible. $\endgroup$
    – Conner
    Aug 19, 2012 at 19:55

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .