PSII structure - adapted from Barber et al., (1997) Physiol. Plantarum. 100:817-827

[ See also: Barber (2003) Quart. Rev. Biophys. 36:71-89 for much greater detail ]

The Psb proteins make up the core of PSII which is excitonically linked to an outer antenna system consisting of the Lhcb proteins in the case of higher plants and green algae (chlorophytes) or the phycobiliproteins in the case of cyanobacteria and red algae (cyanophytes and rhodophytes). Other types of secondary antenna systems occur in chromophytes composed of a variety of carotenoid-chlorophyll complexes (Gantt, 1996).

There is every reason to believe, however, that the basic structure of the PSII core is conserved amongst all types of oxygenic organisms except for some minor differences in subunit content. Low resolution information (20 Å or poorer) is available for the PSII core structure and has recently been reviewed (Hankamer et al. 1997a). Much of this information has emerged from electron microscopy. Indeed, single particle analyses of PSII cores isolated from higher plants (spinach) and from cyanobacteria (Synechococcus) have revealed almost identical structures (Boekema et al. 1995). In both cases the complexes had comparable sizes (approximately 170 x 100 Å), and had similar protein compositions. Also of significance, is that these cores were dimeric, having molecular masses of about 450 kDa. The dimers seem to represent the most stable form of PSII although monomerisation of the isolated dimer does not inhibit the water splitting function (Hankamer et al. 1997b). The isolated and solubilised PSII core dimer of spinach has been reconstituted with thylakoid lipids and induced to form ordered 2D crystals by dialysing out the detergent. These crystals have been used to obtain both 2D and 3D maps after negative staining (Morris et al. 1997). Since the extrinsic proteins had been lost during crystallization the lumenal exposed protein mass is likely to be mainly composed of the extrinsic loops of CP47 and CP43.

In the absence of high resolution data, molecular by analogies between the D1 and D2 proteins and the L and M subunits of the purple bacterial reaction centre (Ruffle et al. 1992, Svensson et al. 1990). More recently the emerging high resolution structure of PSI offers an additional opportunity to predict high resolution features for PSII structure. The work of Kraub et al. (1996) has shown that the 22 transmembrane helices of PsaA and PsaB are arranged with 10 forming a central core with structural analogy with the L and M subunits of purple bacteria and the remaining 12 transmembrane segments fan out with 6 helices on either side of the central core. This structural arrangement contains therefore also suggests homology with D1 and D2 protein organisation and thus with PSII. The additional 6 helices on each side of the central core could be, in e case of PSII, CP47 and CP43 (Fromme et al. 1996). As pointed out by Fromme et al. (1996), there are some sequence homologies between CP47, CP43 (transmembrane segments I and IV) and transmembrane segments a and d of PsaA and B. Moreover, there is crosslinking data which indicates that CP47 is more closely located to the D2 than the D1 protein (Moskalenko et al. 1992). Taking this into account, comparison of the PSI and PSII proteins is consistent with 8 Å data obtained from 2D crystals of the CP47-RC subcore using electron crystallography (Rhee et al. 1997).

The postulated similarities between PSI and PSII emphasise the fact that photosynthetic reaction centres are derived from a common origin whether it be by gene splitting or gene fusion. In the case of PSII, the evolution of the water splitting system has placed additional requirements on the structure: the necessity for binding the Mn cluster and creating the environment for water oxidation to occur and the requirement to degrade and replace the D1 protein. In the former case a gene addition has been made to create CP43 and CP47 with large lumenal located hydrophobic loops. In the context of turnover of the D1 protein, the location of CP43 next to D1 and the ease by which it can be removed from isolated cores compared to CP47 should be noted. Indeed, Barbato et al. (1992) have presented evidence that the turnover of the D1 protein requires not only a conversion of the core dimmer to a monomer but also the removal of CP43. We now await high-resolution structural information in order to appreciate further the dynamics of PSII function and the role of the many subunits which it contains.