Marine biology

Genes in the glass house

Article metrics

Sequence data on a second species of diatom provide abundant insights into the evolution and metabolic capabilities of this group, as well as into mechanisms of gene acquisition and diversification.

Diatoms are one of life's big success stories. These silica-depositing microalgae have been abundant for at least the past 180 million years. Today they contribute an astonishing 40% of total ocean primary production, and about 20% of global primary production1,2, making them central players in the global carbon and silica cycles.

The main lineages of diatoms are the centrics and the pennates, the two having begun to diverge from their common ancestor about 90 million years ago3. The full genome sequence of a member of the centric lineage, Thalassiosira pseudonana, was reported4 in 2004, but since then we have been waiting for an equivalent treatment of one of the highly diverse pennates. That conspicuous gap has been filled by Bowler et al. (page 239 of this issue)5 with a study of the genome sequence of Phaeodactylum tricornutum (Fig. 1).

Figure 1


Phaeodactylum tricornutum — a quartet of the model pennate diatom.

It is no easy task to find the genes in the seemingly endless string of base pairs that constitutes a whole genome. To aid in mapping the P. tricornutum genome, Bowler et al. used more than 130,000 available expressed sequence tags; these are short DNA sequences that are derived from messenger RNA and so represent parts of genes that have been expressed. This and other information provided support for the existence of 10,402 genes in P. tricornutum, which is slightly fewer than the 11,776 genes identified in its centric cousin, T. pseudonana. Despite the similarity in overall numbers, the two species have only 57% of their genes in common, indicating an exceptional level of divergence since the split of the lineages.

Diatoms belong to a larger group called the chromalveolates (Fig. 2, overleaf). Along with their relatives, including the haptophytes, dinoflagellates and oomycetes, they are thought to have evolved from a secondary symbiosis between a photosynthetic red alga and a heterotrophic host6. The red alga involved is believed to have arisen from a primary symbiosis between a heterotroph and a cyanobacterium. In both instances, then, partnerships were forged between an organism that required organic substrates for energy generation (the heterotroph) and an organism that could produce such substrates from inorganic materials and light.

Figure 2: Relationships among the chromalveolates, to which diatoms belong.

Two subgroups are the alveolates (blue) and the stramenopiles, including the diatoms (red). The diatoms themselves consist of two main lineages; examples of the genomes of both of them, the centric4 and the pennate5, have now been sequenced. The numbers are indications of approximate divergence times in millions of years ago. (Graphic based on Fig. 1 of ref. 13.)

To date, there has been debate as to whether the secondary symbiosis occurred just once or at several points within the different lineages6,7. Genes associated with two of the main cell constituents of P. tricornutum, the chloroplast and mitochondrion, now provide some helpful clues.

For example, the genomic analyses strongly support the idea that the diatom chloroplast came from red algae: 171 genes of P. tricornutum are of red algal origin, and a total of 108 of the red algal genes occur in both centric and pennate diatoms. Of those 108 genes, 11 are also found in the oomycete Phytophthora sojae (a water mould)8, supporting the idea that diatoms and oomycetes shared a common ancestor possessing a red algal chloroplast that was later lost in the development of oomycetes. Furthermore, the suite of mitochondrial genes in P. tricornutum is nearly the same as that in the mitochondria of the earlier-diverging haptophytes and cryptophytes (Fig. 2). This provides some support for the chromalveolate hypothesis — the idea that all these groups arose from the same ancient secondary symbiosis event6. But there are still enough questions about the details of these phylogenetic associations to ensure that vigorous debate about the various evolutionary hypotheses will continue9.

The evolution of P. tricornutum, a eukaryote with the characteristic membrane-bound nucleus, seems to have been advanced by its acquisition of at least 587 genes (more than 5% of the total complement) from prokaryotic organisms — mostly from bacteria but with some from archaea too. Although this level of horizontal gene transfer is common among prokaryotes10, the number of prokaryote genes in P. tricornutum is unusually high for a eukaryote, and is suggestive of long-term intimate associations between bacteria and diatoms that led to transfer of useful genetic capabilities. The prokaryote genes in P. tricornutum support pathways for the use of organic carbon and nitrogen, and they include genes that encode the machinery for the diatom urea cycle, which are also present in T. pseudonana.

Some of the prokaryote gene products seem to be involved in the perception of light and may be part of cell-signalling systems. More than 300 of these prokaryote genes are found in both centric and pennate diatoms, indicating that many of them were acquired before these groups split. But only 14 prokaryote sequences are shared between P. tricornutum and the oomycete P. sojae, indicating that most of the bacterial sequences were acquired after the divergence of oomycetes and the diatom lineage about 700 million years ago. It will be interesting to learn whether many bacterial genes occur in other stramenopile algae (Fig. 2), and whether the same bacterial genes have been retained or different ones have been acquired by the different lineages.

Of the more than 10,000 genes in each of the two diatom genomes, 1,338 are unique to the diatoms and will attract interest from researchers wanting to find out what sets this group apart from others. A defining characteristic of diatoms is their intricate silica frustules, or tests, which serve as multifunctional cell walls11. These exquisite glass houses are the result of nanofabrication of polymerized silicic acid laid on an organic matrix of long-chain polyamines such as spermine and proteins called silaffins. The resulting silica frustule is coated with a glycoprotein called frustulin. Not surprisingly, the P. tricornutum genome contains many genes involved in frustulin synthesis and polyamine metabolism. But there is evidence of just a single silaffin-like protein, indicating either that only one is needed or that other proteins (and their genes) in this family await discovery. There are many genes associated with cell-cycle regulation, which may be related to the need to coordinate the division of cells with rigid and unequal silica valves.

Further analysis of the genome should reveal many more diatom-specific functions, and as other relatives along this phylogenetic tree are sequenced we should learn more about when and how these genetic traits were acquired. Comparative analysis of diatom genomes may also reveal the basis for certain differences, such as the production of the biogeochemically important metabolite dimethylsulphoniopropionate (DMSP) by T. pseudonana but not by P. tricornutum12.

Finally, Bowler et al.5 provide evidence that diatom-specific genes in P. tricornutum and T. pseudonana seem to be evolving faster than other genes in these diatoms or in eukaryotes in general. Mechanisms for this rapid gene diversification vary. They include expansion of certain gene families; gains of non-protein-coding — intron — sequences (mainly in T. pseudonana); and, in P. tricornutum, an exceptionally high frequency of retrotransposons, which are mobile genetic elements that probably accelerated gene fragmentation. The rapid gene evolution in diatoms may help to explain their extensive diversification, and probably contributed to their niche specialization and ultimate success in the global ecosystem2.


  1. 1

    Falkowski, P. G., Barber, R. T. & Smetacek, V. Science 281, 200–206 (1998).

  2. 2

    Falkowski, P. G. et al. Science 305, 354–360 (2004).

  3. 3

    Kooistra, W. H. C. F., Gersonde, R., Medlin, L. K. & Mann, D. G. in Evolution of Primary Producers in the Sea (eds Falkowski, P. G. & Knoll, A. H.) 207–249 (Academic, 2007).

  4. 4

    Armbrust, E. V. et al. Science 306, 79–86 (2004).

  5. 5

    Bowler, C. et al. Nature 456, 239–244 (2008).

  6. 6

    Cavalier-Smith, T. J. Eukaryotic Microbiol. 46, 347–366 (1999).

  7. 7

    Bodyl, A. J. Phycol. 41, 712–719 (2005).

  8. 8

    Tyler, B. M. et al. Science 313, 1261–1266 (2006).

  9. 9

    Keeling, P. J. et al. Science 306, 2191b (2004).

  10. 10

    Woese, C. R. Proc. Natl Acad. Sci. USA 99, 8742–8747 (2002).

  11. 11

    Milligan, A. J. & Morel, F. M. M. Science 297, 1848–1850 (2002).

  12. 12

    Keller, M. D., Bellows, W. K. & Guillard, R. R. L. in Biogenic Sulfur in the Environment (eds Saltzman, E. & Cooper, W. J.) 167–182 (Am. Chem. Soc., 1989).

  13. 13

    Fehling, J. et al. in Evolution of Primary Producers in the Sea (eds Falkowski, P. G. & Knoll, A. H.) 75–107 (Academic, 2007).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kiene, R. Genes in the glass house. Nature 456, 179–181 (2008) doi:10.1038/456179a

Download citation


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.