Structural biology

The purple heart of photosynthesis

The structure of a photosynthetic complex from a purple bacterium reveals a new class of light-harvesting protein and the channels that might allow electron-transporting molecules to escape this otherwise closed system. See Article p.228

Photosynthesis is one of the most fundamentally important biological processes on Earth. It provides all the oxygen we breathe, the food we eat and the fossil fuels we so greedily consume. Purple photosynthetic bacteria have proved to be excellent model organisms with which to study the reactions that occur at the early stages of photosynthesis. In this issue, Niwa et al.1 (page 228) report the X-ray crystal structure of part of the photosynthetic machinery from the bacterium Thermochromatium tepidum, at near-atomic resolution. Their findings greatly advance our understanding of the detailed molecular mechanisms involved in primary photosynthetic reactions.

The pigment–protein complexes that catalyse these reactions are located in and on intracytoplasmic membranes — infoldings of the cell membrane. The major light-absorbing pigments in purple bacteria are bacteriochlorophyll and carotenoid molecules (the orange pigments found in carrots). These pigments are non-covalently bound to well-defined membrane proteins known as the light-harvesting complexes (LH1 and LH2) and reaction centres (RCs). In most species of purple bacterium, light energy is initially absorbed by LH2 and then transferred by LH1 to the RC, in which it is used to drive the primary redox reactions of photosynthesis2 (Fig. 1). These reactions involve the conversion of ubiquinone molecules to their completely reduced form, which carries electrons to the cytochrome bc1 complex as part of a cyclic electron-transport pathway.

Figure 1: Typical organization of a photosynthetic unit in the intracellular membrane of purple bacteria.

Light-harvesting complex 2 (LH2) contains two groups of bacteriochlorophyll (BChL) pigments, B800 and B850. Light energy absorbed by B800 is transferred to B850 (blue arrows). It is then passed to another group of BChLs (B880) in light-harvesting complex 1 (LH1), and finally into the reaction centre (RC). The excited RC fully reduces ubiquinone molecules (Q), which leave the LH1–RC complex and pass into the membrane, transferring electrons (e) to the cytochrome bc1 complex; this transfer forms part of a cyclic electron-transport pathway (red arrows) that drives photosynthesis. This pathway is completed by a soluble protein, cytochrome c2 (Cyt c2). The periplasm is the region between the cell membrane and the cell wall of the bacterium.

The value of this model system is clearly illustrated by two landmark studies. The first determined the X-ray crystal structure of the RC from the bacterium Blastochloris viridis3; this was the first crystal structure of a membrane protein to be determined, and led to the award of the Nobel Prize in Chemistry in 1988. The second study solved the crystal structure of the LH2 complex from the bacterium Rhodopseudomonas acidophila4. These two structures revealed fundamental principles that underlie our current understanding of the energy-transfer processes of photosynthesis. Niwa and colleagues' study describes the structure of the T. tepidum LH1–RC complex, and promises to be just as influential as the two earlier papers.

Earlier low-resolution structures had revealed that LH1–RC complexes can occur as dimers in Rhodobacter sphaeroides5 and as monomers in Rhodopseudomonas palustris6 (Fig. 2a, b). In these bacteria, the basic structures are elliptical and contain gaps in the LH1 ring. It has been suggested that reduced ubiquinone passes through these gaps to leave the RC and connect to the cytochrome bc1 complex. In both of these structures, the gaps are introduced by a protein: PufX in the dimeric complex and protein W in the monomeric complex.

Figure 2: X-ray crystal structures of three types of the LH1–RC complex.

a, A model of the dimeric RC–LH1–PufX complex from Rhodobacter sphaeroides5 (8 ångströms resolution), in which PufX proteins create gaps in the LH1 rings. Yellow and green cylinders represent helical portions of protein subunits of LH1 (α- and β-apoproteins, respectively); blue cylinders are helices of the RC. Non-helical portions of protein chains are shown as thin coils. Bacteriochlorophyll pigments of the LH1 complex are shown in red. No LH1 carotenoid pigments or RC cofactors are shown, for clarity. b, A model of the monomeric RC–LH1–protein W complex from Rhodopseudomonas palustris6 (4.8 Å resolution). Here, protein W creates a gap in the LH1 ring. c, Niwa et al.1 report this 3-Å model of the monomeric RC–LH1 complex from Thermochromatium tepidum. In the absence of a protein such as PufX or protein W, there is no visible break in the LH1 ring.

Niwa and co-workers' structure shows that a third class of LH1–RC complex is possible. In their case, the LH1 ring is more uniformly elliptical than the other two structures, but there is no analogue of PufX or protein W (Fig. 2c). Nevertheless, channels are present at a level corresponding to the middle of the transmembrane region of their complex (see Fig. 5 of the paper1). The authors suggest that these channels provide the same function as the gaps in the previously reported structures, allowing reduced ubiquinone molecules to connect with the cyclic electron-transport pathway. Their structure confirms conclusions from previous molecular-dynamic simulations that also suggested this type of mechanism for ubiquinone diffusion through the LH1 ring7,8. This escape of the reduced ubiquinone from the 'core' complex is essential for cyclic electron transport in purple bacteria; without it, they cannot grow through photosynthesis9.

Thermochromatium tepidum was discovered in hot springs, and has an optimum growth temperature of 50 °C. Its photosynthetic complex is therefore more stable at higher temperatures than the complexes of other purple bacteria. Niwa et al. describe in detail the amino-acid residues and bonds involved in binding calcium ions to the carboxy termini of the LH1 protein subunits, and suggest that these interactions are crucial for the increased thermostability of the complex.

Another interesting quirk of the LH1–RC complex in T. tepidum is that the longest-wavelength band (called the Qy band) in the near-infrared absorption spectrum of bacteriochlorophyll is considerably redshifted compared with that of the other complexes shown in Figure 2. The energy transfer from LH1 to RC takes place from this Qy band. The authors conclude that the calcium-ion binding observed in their structure, and the effects that this has on the bacteriochlorophyll molecules in LH1, causes this large redshift.

Niwa and colleagues' structure also shows, for the first time in an LH1 complex, the exact position and orientation of the carotenoid molecules; in this case the carotenoid is spirilloxanthin. Carotenoids are required for both light harvesting and photoprotection — in their absence, illumination in the presence of oxygen destroys light-harvesting complexes and kills the photosynthetic organism in which they reside. Niwa et al. observe that the end of each spirilloxanthin molecule in the LH1 complex is in close proximity to a histidine amino-acid residue that is bound to the central magnesium atom of a bacteriochlorophyll within the same protein subunit. In LH1 complexes from other species of purple bacterium, removal of this carotenoid results in a blueshift of the bacteriochlorophyll Qy absorption band10. Loss of the carotenoid interactions described above probably explains this long-standing observation.

The report of a new structure of a membrane protein is always an event. That is particularly true in this case, when the protein involved is at the heart of a process as vital as photosynthesis. It will be exciting to see if comparing theoretical modelling of the spectroscopic properties and energy-transfer reactions of this LH1–RC complex, based on Niwa and co-workers structure, with experimental measurements produces a consistent, detailed mechanistic understanding of photosynthesis.


  1. 1

    Niwa, S. et al. Nature 508, 228–232 (2014).

  2. 2

    Cogdell, R. J., Gall, A. & Köhler, J. Q. Rev. Biophys. 39, 227–324 (2006).

  3. 3

    Deisenhofer, J., Epp, O., Miki, K., Huber, R. & Michel, H. Nature 318, 618–624 (1985).

  4. 4

    McDermott, G. et al. Nature 374, 517–521 (1995).

  5. 5

    Qian, P. et al. Biochemistry 52, 7575–7585 (2013).

  6. 6

    Roszak, A. W. et al. Science 302, 1969–1972 (2003).

  7. 7

    Aird, A., Wrachtrup, J., Schulten, K. & Tietz, C. Biophys. J. 92, 23–33 (2007).

  8. 8

    Walz, T. & Ghosh, R. J. Mol. Biol. 265, 107–111 (1997).

  9. 9

    Barz, W. P. et al. Biochemistry 34, 15235–15247 (1995).

  10. 10

    Sistrom, W. R., Griffiths, M. & Stanier, R. Y. J. Cell. Comp. Physiol. 48, 473–515 (1956).

Download references

Author information

Correspondence to Richard J. Cogdell or Aleksander W. Roszak.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Cogdell, R., Roszak, A. The purple heart of photosynthesis. Nature 508, 196–197 (2014).

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.