Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Molecular evolution

A hydrogen-producing mitochondrion

Dark, damp (and sometimes smelly) places harbour some of nature's most curious eukaryotes. The likes of swamps and intestines are swarming with mostly single-celled eukaryotes (protists) that, like all cells, must produce ATP to survive. Yet these places lack enough oxygen to sustain ATP synthesis as it occurs in textbook mitochondria like our own. Some protists possess no mitochondria, surviving from anaerobic fermentation in the cytosol. Others have quite odd mitochondria that harbour anaerobic ATP-producing pathways. On page 527 of this issue, Akhmanova et al.1 report a gem of such an odd mitochondrion in the ciliate protist Nyctotherus ovalis.

The ciliate lives in the suffocatingly oxygen-poor confines of cockroach intestines, where it helps the insect to digest cellulose. Instead of consuming oxygen, Nyctotherus's mitochondrion has the bizarre property of excreting hydrogen as a by-product of ATP synthesis. Similar hydrogen-generating organelles — hydrogenosomes — have been studied in anaerobic eukaryotes for 25 years2. Hydrogenosomes have often been suspected of stemming from the same endosymbiotic bacterium that gave rise to mitochondria. But now, Akhmanova et al. report a hydrogenosome that has its own genome, directly betraying its endosymbiotic past.

Cells of Nyctotherus (which do not grow in culture and have to be carefully micromanipulated from cockroach hindguts) contain hydrogenosomes that can be labelled by antibodies against DNA. The authors found that the cell produces a ribosomal RNA which, although not proven by in situ hybridization to localize to the organelle, bears all the sequence characteristics expected of ciliate mitochondria. Plus, it may look like a mitochondrion, but this DNA-bearing organelle is unquestionably a hydrogenosome because it produces hydrogen — Akhmanova and colleagues found hydrogen-consuming3 methanogenic endosymbionts inside the cells of Nyctotherus. Finally, Nyctotherus expresses a nuclear-encoded gene for a hydrogenase (an enzyme that makes hydrogen) that is probably imported into the hydrogenosome with a transit peptide1, as is the case for most proteins in mitochondria.

The evolutionary significance of these findings is twofold. First, hydrogenosome-associated DNA was hitherto a genuine missing link4. Such discoveries are rare, and the genes in this DNA are likely to hold exciting surprises. The other, deeper, significance emerges when we remind ourselves that ciliate hydrogenosomes are just the tip of the iceberg (Fig. 1). Hydrogenosomes and the anaerobic (or micro-aerophilic5) lifestyle are widespread among contemporary eukaryotes4,6, including groups that are distantly related in conventional genealogies and, sometimes, arise from within otherwise aerobic groups. Although biologists do not agree which groups of eukaryotes might be the most primitive, most will find their favourite candidate somewhere in Fig. 1, and most biologists do believe that all eukaryotes share a single common ancestor.

Figure 1: Summary of ATP-producing organelles across a very limited, and arbitrarily chosen, spectrum of eukaryotic taxa.

The ciliate Nyctotherus ovalis (red star) contains hydrogenosomes, which Akhmanova et al.1 now show contain their own genome. A peculiar group of eukaryotes called microsporidia are related to fungi6, whence the ‘N’ for that group. Antarctic euglenids contain organelles that may be hydrogenosomes10. Some groups designated ‘N’ may have organelles, but their role, if any, in energy metabolism is unclear. (Data taken from refs 4,5 and references therein, and also refs 2,3, 610)

The key to understanding hydrogenosomes, and why they produce hydrogen, is energy metabolism4. All known eukaryotes generate energy (ATP) by one principle — the oxidative breakdown of reduced carbon compounds. Electrons removed during the oxidation process must be dumped onto an electron-accepting compound (an acceptor) that can be excreted from the cell. Otherwise, ATP production — and life — comes to a halt. Our mitochondria use oxygen as the acceptor and excrete water. But the mitochondria of anaerobic eukaryotes must resort to compounds other than oxygen (Fig. 2). Some use organic acceptors such as fumarate7, some use nitrate8,9. The mitochondria of Nyctotherus1, like other hydrogenosomes2,4, simply transfer the electrons onto protons, producing hydrogen.

Figure 2

Some electron acceptors, and the corresponding reduced end-products associated with energy metabolism in mitochondria.

Why are anaerobic ATP-producing pathways so widespread in eukaryotes2,4,7,10? There are two popular hypotheses for their origin which, in principle, can be tested through gene-by-gene phylogenetic analysis. One view is that the genes were acquired by eukaryotes through horizontal gene transfer from one or more prokaryotic donors, other than the antecedent of mitochondria. If this were the case, the genes for these pathways in anaerobic eukaryotes should trace to different prokaryotic (eubacterial or archaebacterial) sources. An alternative view is that the eukaryotic genes involved in anaerobic ATP synthesis were inherited from a single common ancestor of mitochondria and hydrogenosomes. These genes are thought to have been transferred to the host's chromosomes, because they are not found in any known mitochondrial genome. In this case, the common ancestor of contemporary eukaryotes would have acquired, from a facultatively anaerobic protomitochondrion, a genome's worth of genes for all-purpose survival. These genes were then left to the workings of selection and common descent. From this we can predict that, across the anaerobic eukaryotes in Fig. 1, each gene should ultimately trace to a single eubacterial source.

We know that the antecedent of mitochondria brought with it the ability to respire oxygen — much of the respiratory chain is still encoded in mitochondrial DNA−1. But we don't know what else was in that symbiont's biochemical repertoire. It probably existed two billion years ago12 but, because it's no longer around, we cannot sequence its genome to find out. However, molecular fossils of its lifestyle might be preserved in nuclear chromosomes, including our own, allowing us to piece together the bacterial part of our heritage.

Is anaerobic energy metabolism in eukaryotes a telling relict of our history, or an oddity with little significance? Eukaryotes now regarded as primitive tend to be anaerobes, so these are important questions. The road to answers leads straight to the genomes of eukaryotes from the anoxic world. For scientists studying oxygen-shunning eukaryotes with unconventional mitochondria, with hydrogen-producing mitochondria or with no mitochondria at all, these are exciting times.


  1. 1

    Akhmanova, A. et al. Nature 396, 527–528 (1998).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Müller, M. J. Gen. Microbiol. 139, 2879–2889 (1993).

    Google Scholar 

  3. 3

    Fenchel, T. & Finlay, B. J. Ecology and Evolution in Anoxic Worlds (Oxford Univ. Press, 1995).

  4. 4

    Müller, M. in Evolutionary Relationships among Protozoa (eds Coombs, G. H., Vickermann, K., Sleigh, M. A. & Warren, A.) 109-132 (Chapman & Hall, London, 1998).

  5. 5

    Lloyd, D. Nature 381, 121 (1996).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Embley, T. M. & Hirt, R. P. Curr. Opin. Genet. Dev. 8, 624–629 (1998).

    Google Scholar 

  7. 7

    Tielens, A. G. M. & Van Hellemond, J. J. Biochim. Biophys. Acta 1365, 71–78 (1998).

    Google Scholar 

  8. 8

    Kobayashi, M. et al. J. Biol. Chem. 271, 16263–16267 (1996).

    Google Scholar 

  9. 9

    Finlay, B. J., Span, A. S. W. & Harman, J. M. P. Nature 303, 333–336 (1983).

    ADS  CAS  Article  Google Scholar 

  10. 10

    Simpson, A. G. B., van den Hoff, J., Bernard, C., Burton, H. R. & Patterson, D. J. Arch. Protistenkd. 147, 213–225 (1997).

    Google Scholar 

  11. 11

    Gray, M. et al. Nucleic Acids Res. 26, 865–878 (1998).

    Google Scholar 

  12. 12

    Doolittle, W. F. Proc. Natl Acad. Sci. USA 94, 12751–12753 (1997).

    Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Embley, T., Martin, W. A hydrogen-producing mitochondrion. Nature 396, 517–519 (1998).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing