Structural biology

Dual approach to a light problem

The structure of the last of the major pigment-containing protein complexes involved in photosynthesis is now revealed. The details complete our picture of electron shuttling in this vital process.

It is not uncommon for notable scientific progress to be made simultaneously by two independent teams of researchers. The latest example concerns the long-awaited structure of the cytochrome b6f complex, described by Stroebel et al. on page 413 of this issue1 and by Kurisu et al. in Science2. Cytochrome b6f mediates the flow of electrons between photosystems II and I in the photosynthetic membranes of plants and cyanobacteria. It is the last of these large, pigment-containing protein complexes to yield to detailed crystallographic analysis, and the structures offer some surprising insights into how it works.

Stroebel and colleagues1 examined the b6f complex found in the chloroplasts — the photosynthetic organelles — of a unicellular alga1, whereas Kurisu and co-workers2 studied the same complex from a cyanobacterium. Both complexes have essentially the same structure. This in itself is astonishing, given that the two types of organisms are separated by an evolutionary distance of roughly 1,000 million years.

Plants and cyanobacteria have the unique ability to use solar energy to withdraw electrons from water, the most abundant substrate on Earth. Unlike the purple and green bacteria, which use more energy-rich substrates, plants and cyanobacteria need two big protein complexes — the photosystems — to bridge the large energy gap between water and the stable reducing agent NADPH. This agent enables the synthesis of organic substances such as sugars and starch.

Photosystem II is the complex that uses solar energy to withdraw electrons from water, generating oxygen as a waste product (Fig. 1). The immediately useful product of this photosystem is the unstable, reduced form of plastoquinone, a lipid-soluble molecule that has been endowed with two electrons by photosystem II. The cytochrome b6f complex recycles this reduced plastoquinone, stripping it of its electrons and releasing two protons that contribute to the electrochemical gradient across the photosynthetic membrane. This gradient is ultimately used to make ATP, the main energy-storing molecule in living cells. The oxidized plastoquinone goes back to photosystem II to be reduced again. Meanwhile, the b6f complex feeds one of the stripped electrons into photosystem I, which uses solar power to transfer it to the opposite membrane surface for NADPH synthesis. The other electron passes through the b6f complex and reduces another plastoquinone molecule.

Figure 1: Electron transport in oxygenic photosynthesis.

Photosystem II uses solar energy to withdraw electrons from water, generating oxygen as a waste product. Two electrons are accepted by plastoquinone (PQ), which then binds two protons for electroneutrality. The reduced plastoquinone (PQH2) diffuses in the membrane to the cytochrome b6f complex. One electron is transferred to plastocyanin (PC); the other passes to another PQ molecule. The PC diffuses to photosystem I, which uses solar energy to propel the electron against a potential gradient across the membrane. The electron is accepted by ferredoxin (Fd) and transferred to an enzyme (ferredoxin:NADP+ reductase, FNR) that converts NADP into NADPH, the primary product of photosynthesis. X-ray structures of the cytochrome b6f complex1,2 suggest that Fd can also deliver its electrons back to b6f in a cyclic electron flow (dashed line). Black arrows, electron transfer; red arrows, proton transfer. The inset shows a chloroplast.

In addition to this linear mode of electron flow, the system can switch to a cyclic mode, in which photosystem I returns some of its electrons to the b6f complex rather than feeding them into biosynthesis. This is necessary to balance the activities of the two photosystems, which depend on the variable amounts of solar energy absorbed by each. The cyclic mode of electron flow is poorly understood, but the X-ray structures of the b6f complex1,2 suggest a plausible mechanism.

Cytochrome b6f comprises several membrane-spanning proteins, namely cytochrome b6, cytochrome f, and a protein that harbours an iron–sulphur cluster. The complex owes its deep brown colour to pigment molecules attached to the proteins; these include four haems, a chlorophyll and a β-carotene. Stroebel et al. and Kurisu et al. now find that the membrane-embedded part of the b6f complex has an extensive hydrophobic cavity, which is partly filled with membrane lipids, but allows access to the two plastoquinone-binding sites of the complex.

As expected from knowledge of a related structure, the mitochondrial cytochrome bc1 complex3,4,5,6, the two b-type haems are located next to the plastoquinone-binding sites (Fig. 2). Unexpectedly, however, one of these sites contains an extra haem group, not present in the bc1 complex. This pigment (which Stroebel et al. call haem ci and Kurisu et al. call haem x) is covalently attached to the cytochrome b6 polypeptide — making it all the more astonishing that it has escaped discovery until now. Given its position, the most likely function of this extra haem is in cyclic electron flow (Fig. 1). The X-ray structures now make it possible to design experiments to study this elusive process in detail.

Figure 2: Electron-transfer routes in the cytochrome b6f complex, incorporating the new findings1,2.

The complex is a dimer, with two identical sets of components. At the PQH2-binding site, two protons are released on the lumenal side of the membrane. The iron–sulphur cluster (Fe2S2) takes one electron from PQH2 and passes it to haem f in cytochrome f, where it is picked up by PC. The Fe2S2 cluster is attached to an iron–sulphur protein (Rieske protein; ISP in ref. 2), which moves on a hinge to bridge the gap between haem bL and f. The second electron passes via haem bL and bH (bp and bn in ref. 2) to a PQ bound at a site near the stromal membrane surface. PQ binds a proton, adding to the pH gradient across the membrane. The newly discovered haem ci (x in ref. 2) is well placed for the re-uptake of electrons in cyclic electron flow. Purple squares, haems; green squares, chlorophylls; black arrows, electron transfer; red arrows, proton transfer.

Together, the haems mark the path that electrons take between the two plastoquinone-binding sites. The other path to the surface-exposed cytochrome f leads via an iron–sulphur cluster, which is found in a protein domain that moves on a hinge to deliver the electron to haem f (Fig. 2). This movement has been inferred from the different positions of the analogous iron–sulphur domains in bc1 complexes3,4,5,6. Kurisu and colleagues' structure2 provides evidence for the same mechanism in the b6f complex.

The new haem is not the only surprising feature of the b6f complex. Even though their function in the complex was not clear, it was widely expected that the chlorophyll and carotene would be in close contact, allowing the carotene to defuse the chlorophyll — excitation of an isolated chlorophyll molecule by light would produce oxygen radicals. But both structures show that the two pigments are too far apart to prevent this potentially lethal reaction. It was also hoped that the structure would explain why the chlorophyll is there, but its role remains a mystery. Kurisu et al.2 suggest that it may simply be a space filler (although it would be a highly dangerous one!), whereas Stroebel et al.1 propose that it acts as a sensor in the interaction with photosystem I — which might then provide the absent carotene. Such large functional assemblies, or 'supercomplexes', of cytochrome b6f and the photosystems have been postulated7,8, and might indeed exist in the crowded photosynthetic membrane, but they have not yet been seen.

Both teams1,2 have been working towards a high-resolution structure of the b6f complex for well over ten years. Why did it take so long? Crystallizing membrane proteins is nearly always problematic, because detergents are required, which often make it difficult to grow diffraction-quality crystals. To make matters worse, the b6f complex is easily disrupted once it is removed from the photosynthetic membrane. Each team came up with a different ruse to get around this problem. Stroebel et al.1 engineered a tag into the complex, using it to get hold of the complex and purify it quickly — so avoiding prolonged exposure to damaging detergents. Kurisu et al.2 chose the inherently more stable complex from a thermophilic organism, but found that they needed to add back lipids removed during purification before they could obtain well-ordered crystals9.

The structure of cytochrome b6f completes our picture of the molecular events that produce the oxygen in the air we breathe. This is one of the first membrane processes that we can follow in its entirety. In the future, we can hope to understand many other, no less fundamental and fascinating membrane systems, including those that enable us to see, hear, taste and think. For all of these we will need information on the structure of membrane proteins at similarly high resolution. But this information cannot be obtained without similar high-risk, long-term efforts.


  1. 1

    Stroebel, D., Choquet, Y., Popot, J.-L. & Picot, D. Nature 426, 413–418 (2003).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Kurisu, G., Zhang, H., Smith, J. L. & Cramer, W. A. Science 302, 1009–1014 (2003).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Xia, D. et al. Science 277, 60–66 (1997).

    CAS  Article  Google Scholar 

  4. 4

    Zhang, Z. et al. Nature 392, 677–684 (1998).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Iwata, S. et al. Science 281, 64–71 (1998).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Hunte, C. et al. Structure Folding Design 8, 669–684 (2000).

    CAS  Article  Google Scholar 

  7. 7

    Zhang, H. et al. J. Biol. Chem. 276, 38159–38165 (2001).

    CAS  Google Scholar 

  8. 8

    Joliot, P. & Joliot, A. Proc. Natl Acad. Sci. USA 99, 10209–10214 (2002).

    ADS  CAS  Article  Google Scholar 

  9. 9

    Zhang, H., Kurisu, G., Smith, J. L. & Cramer, W. A. Proc. Natl Acad. Sci. USA 100, 5160–5163 (2003).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kühlbrandt, W. Dual approach to a light problem. Nature 426, 399–400 (2003).

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.


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