Fig. 1: A look at the nanoworld of the DNA of the cell’s power stations: Super-resolution STED microscopy enables a more detailed view of the organization structures of mitochondrial DNA. In the cell shown, the mitochondria are stained red, the cell nucleus blue and the mitochondrial DNA green (Credit: Max Planck Institute for Biophysical Chemistry Göttingen, Max Planck Institute for Biology of Ageing, Cologne / Christian Wurm, Corinna Schwierzy).
The evolutionary precursors of today’s mitochondria gave up their independent existence as bacteria about 2 billion years ago in order to live on symbiotically inside other cells (endosymbiosis). Meanwhile they supply most of the energy for nearly all the cells in our body: Most of the energy in nutrients is utilized inside the mitochondria by oxidation to form the ATP molecule that can be used everywhere in the cell as an all-purpose energy supplier. The symbiosis can also be seen in the fact that the genes for nearly all the 750–1,000 different mitochondrial proteins are localized in the nucleus. The proteins encoded by these genes are produced in the cell cytosol and then imported into the mitochondria.
However, a few sub-units of the protein apparatus of the respiratory chain, in which the final steps of ATP production take place, are still directly produced in the mitochondria. The genes for these 13 proteins and for the tRNA molecules necessary for genetic transcription (and also for two other RNA molecules) are located inside the mitochondria on a circular DNA molecule, the mitochondrial DNA (mtDNA).
Unlike nuclear DNA, which is present in every cell exactly twice (one set of chromosomes from the father and one from the mother) there are about 1,000 to 2,000 mtDNA molecules in every cell. What happens when some of these mtDNA molecules are mutated, how the mtDNA molecules are distributed to the daughter cells upon cell division, and what happens to the mtDNA when an oocyte is fertilized and develops further are exciting questions that are being researched by scientists like Christian Kukat and other research groups all over the world.
Unlike gene mutations in the nucleus, which can occur exactly once or twice because of the double set of chromosomes (heterozygosity or homozygosity, respectively), mitochondrial mutations can theoretically occur in any percentage from 0 to 100 % due to the thousand or more mtDNA copies per cell. This phenomenon is called heteroplasmy. Many diseases caused by mutations of the mitochondrial DNA only manifest themselves when the proportion of mutated mtDNA molecules in the cells exceeds a certain threshold from about 30 % to over 80 %, depending on the disease and the tissue concerned. And generally, not all cells of the particular tissue are affected, but a mosaic of cells develops consisting of cells with a functioning respiratory chain and cells where the respiratory chain is defective. It is not yet known exactly how the proportion of different mtDNA versions is regulated and how it changes during an organism’s lifetime. It is now thought that certain mtDNA mutations increase over time and thus contribute to the ageing process.
As opposed to nuclear DNA, which comes from the father and mother in equal halves, mitochondrial DNA comes from the mother only – the sperm mitochondria are completely digested after the oocyte is fertilized. However, it also seems that only a small portion of the mtDNA molecules of the mother’s germ cell are transferred as the oocyte develops. This so-called bottleneck effect is an important and as yet unresolved phenomenon of mitochondria genetics. It may serve to sort out unfavorable mtDNA copies, so that, for example, oocytes with too large a proportion of unfavorable mtDNA molecules do not survive.
Due to the appearance of mitochondrial sections in electron-microscopic images, and maybe also to the knowledge of their bacterial ancestors, mitochondria were depicted as bacteria- or bean-shaped objects for decades. Unfortunately, this representation is still often found in many of today’s biology text books. The first images of specifically stained mitochondria in living cells finally proved that for many cell types, this representation is not true. Depending on the cell type and tissue, the mitochondria are frequently in the form of a "mitochondrial network": a tubular, branched system inside the cell that is constantly changing due to fusion and fission.
It has been known for several years that the mtDNA molecules in the mitochondrial network are not randomly distributed but organized in structures called "mitochondrial nucleoids", analogous to the situation in bacteria. An important component of these nucleoids is the protein TFAM (the mitochondrial transcription factor A), which has been the subject of the research group of Nils-Göran Larsson for many years. This work has contributed to the knowledge today that TFAM performs a similar task in the nucleoids to the one performed by the histone proteins in the nucleus, which combine with the nuclear DNA to form the nucleosome: practically the entire length of the mtDNA is covered with TFAM molecules, and the mtDNA is bent into a U shape and thus "packaged" into a more compact space by binding to TFAM. TFAM also plays a further role: as the name suggests, it acts as a transcription factor, i.e. it makes sure that the mtDNA can be read. Here too, the U-shaped bending of the mtDNA due to the binding to TFAM is important.
But how precisely are the nucleoids structured? Are there other components besides TFAM and mtDNA? The new results of Christian Kukat, Christian Wurm and colleagues have made a key contribution to answering these questions.
Up to now, the size of mitochondrial nucleoids measured with confocal microscopes was given as 200–300 nm. These values lie within the range of the physical resolution limit of light microscopy of approx. 260 nm. Consequently, this technique – and conventional light microscopy in general – is not suitable for gaining more detailed information on the structure and the exact size of the nucleoids.
Due to the development of new super-resolution microscopy techniques in the nineteen nineties and the first decade of the 2000s, it became possible to break through the barrier imposed by the physical resolution limit. Using commercially available microscopes equipped with STED technology (STED: stimulated emission depletion), for instance, a resolution of less than 50 nm is currently attainable. In the department of the inventor of STED microscopy, Prof. Stefan Hell, at the Max Planck Institute for Biophysical Chemistry in Göttingen, resolutions of 20 nm and even less have been achieved with experimental set-ups. When Kukat, Wurm and colleagues looked at the nucleoids with a STED microscope of this kind in Göttingen, they discovered that nucleoids labeled with anti-DNA antibodies only had an average diameter of 100 nm in the STED images. In the confocal microscope, the nucleoids had an apparent diameter of 300 nm. The newly found size was consistently corroborated in all the STED images for all cells and cell lines examined from different types of tissue and mammals, whereas the nucleoid sizes determined at the resolution limit in confocal microscope images were considerably larger. In addition, many structures that looked like a single nucleoid in the conventional confocal microscope image were resolved into two or even three and more separate nucleoids in the STED microscope. So the cell contains more nucleoids than previously assumed.
Fig. 4: Mitochondrial nucleoids in the confocal microscope and in the STED microscope. The nucleoids stained with anti-DNA antibodies in human fibroblasts look like compact structures with an average diameter of about 300 nm in the confocal microscope, but often turn out to be agglomerates of several nucleoids with an average diameter of about 100 nm when viewed in the STED microscope. The scale bars correspond to 500 nm. (Source: Kukat et al., PNAS Vol. 108 no. 33 13534–13539, 2011, www.pnas.org/cgi/doi/10.1073/pnas.1109263108).
The researchers then applied molecular-biological methods to determine the number of mtDNA molecules per cell in cultivated primary human fibroblasts and correlate them with the newly determined larger number of nucleoids. Instead of the previously published 2 to 10 mtDNA molecules per nucleoid, there was now a number of 1.45 mtDNA molecules per nucleoid. The conclusion that can be drawn from this is totally new: Most nucleoids only contain one single mtDNA molecule.
Analysis of the TFAM molecules per cell in cultivated primary human fibroblasts revealed a number of about 1,500 molecules per nucleoid or 1,000 per mtDNA molecule. These figures closely match the measurements the group previously made in mouse tissue. The three-dimensional structure of the TFAM molecule bound to mtDNA was then projected with modeling techniques and the total volume of this structure calculated, as no x-ray structure data were available at the time. Given a nucleoid diameter of 100 nm, a nucleoid could then contain approximately 5,000 mtDNA-bound TFAM molecules. Therefore the 1,500 molecules measured fit into the volume of the nucleoid.
These discoveries have direct implications for the general concept of the nucleoid, which up to now had "rather a philosophical aura", as Kukat describes it. Unable to look into the nucleoid, researchers thought it contained a wide variety of proteins for gene regulation, transcription and replication. They are undoubtedly present in the nucleoid, but are not the main component. It is now extremely likely that the nucleoid is the packaging and organization unit of the mtDNA, and that the smallest inheritable unit of the mtDNA is therefore precisely one mtDNA molecule, not a number from 2 to 10. And the newly determined size of 100 nm or less indicates that mtDNA and TFAM are the main components. Maybe the typical nucleoid really does consist of only one mtDNA molecule coated with TFAM – other required proteins would then be recruited as needed, such as the above-mentioned proteins for transcription, translation or replication.
The consequences of the now found size of the smallest inheritable unit for transmission of hereditary diseases, its significance in the ageing process, and whether the number and size of the nucleoids change in the event of disease and ageing – these are all issues calling for further research. A new target has been defined and found. "The exciting thing about the new super-resolution microscopy techniques is that they provide much more detailed images of cell structures," says Kukat. "The STED microscope has given us completely new insights into mitochondria, and this naturally raises a whole lot of new questions and ideas for new experiments."
Kukat C, Wurm CA, Spåhr H, Falkenberg M, Larsson NG, Jakobs S: Super-resolution microscopy reveals that mammalian mitochondrial nucleoids have a uniform size and frequently contain a single copy of mtDNA. Proc Natl Acad Sci USA 108:33 (2011) 13534–9. Epub 2011 Aug 1. (www.pnas.org/cgi/doi/10.1073/pnas.1109263108)