Photoinhibition: molecular mechanisms and physiological significance

'Why do we always come here?' 'I guess we'll never know.' Waldorf and Statler Photoinhibition is defined as the light-induced loss of photosynthetic activity and is an unavoidable consequence of the light reactions. Photoinhibitory damages are repaired by dedicated mechanisms and as long as the rate of repair matches the rate of damage no net loss of activity is observed. In this special issue we focus on recent views on photoinhibition from mechanistic and physiological perspectives (the D1 degradation mechanism was recently reviewed by Huesgen et al. 2009, Mulo et al. 2008, Nixon et al. 2010). Photosynthetic organisms, from cyanobacteria to vascular plants, have developed a wide range of mechanisms to reduce photoinhibitory damage. These include short-term processes that modulate the structure and function of antenna complexes including non-photochemical quenching and state transition, reaction centre quenching, alternative electron transport processes and movement of chloroplasts, leaves or whole organisms away from intense light (Baker et al. 2007, Raven 2011, Rochaix 2010). Long-term acclimation responses include processes such as changes in pigment composition, light harvesting structures and the photosystem II (PSII)/photosystem I (PSI) ratio (Kehoe 2010, Walters 2005). In addition to the mechanisms found in well characterised model organisms (spinach, pea, Chlamydomonas, Synechocystis etc.) work on organisms collected from extreme environments (e.g. cyanobacterial sand crusts, desiccation tolerant lichens and mosses) show very efficient dissipation processes. Details of these mechanisms are only starting to emerge (Heber et al. 2011, Ohad et al. 2011). Furthermore, many cyanobacterial species have a number of copies of the D1 protein, the high light form having a modified redox potential of pheophytin promoting direct non-radiative charge recombination (Vass 2011). Nevertheless, under severe conditions where the protective mechanisms are non-functional or overwhelmed and the rate of repair is insufficient, photoinhibition can lead to a reduction in photosynthetic productivity. The molecular mechanism of photoinhibition has been a matter of debate for as long as this phenomenon has been investigated (Kok 1956). Currently, it is perceived that PSII is more susceptible to photoinhibition than PSI with the exception of some chilling sensitive species (Sonoike 2011). For PSII, a number of hypotheses have been put forward to explain the photoinhibitory process on a molecular basis. These mechanisms differ in their sequence of events and the site of primary damage (PSII cofactors and electron transport steps are shown in Fig. 1). Schematic representation of PSII structure and function. Photosystem II is a thylakoid membrane embedded pigment–protein complex composed of more than 20 subunits. The subunits of PSII are presented in a partially transparent manner so as to highlight the cofactors that participate in the electron transport processes. The QA and QB quinones, the non-heme iron (Fe) and the primary pheophytin acceptor (Ph) are the electron transport components of the acceptor side. The primary electron donor chlorophylls (P) and the Mn4Ca cluster are the electron transport components of the donor side (the redox active tyrosine-Z residue, bridging between P and the Mn-Ca cluster, is not depicted in this scheme). The cofactors located at the D1 protein are shown in green and the cofactors located at the D2 protein are shown in yellow. The catalytic cycle of the Mn-Ca cluster is performed in five consecutive steps (S0–S4, as presented on the bottom right hand side of the figure). In total, four photons are required to convert two H2O molecules into O2 + 4H+ + 4e−. Structural data were drawn based on the 3BZ1 PDB file. In broad terms, photoinhibition mechanisms can be divided into a number of categories (Table 1): mechanisms in which the activity of the acceptor side is limiting while the donor side is still active (acceptor side mechanisms); mechanisms occurring at low light and mechanisms in which the donor side is partially or completely inactive (donor side mechanisms). In these mechanisms, damage is a direct result of PSII photochemistry. Additional mechanisms suggest that the primary damage is a result of light absorption by the Mn cluster (two-step donor side mechanism) or by weakly coupled chlorophyll (free chlorophyll mechanism). Certain conditions were specifically designed to isolate specific mechanisms. Single turnover flash illumination is one example for cherry-picking photoinhibitory conditions (Keren et al. 1995). Using single turnover flashes, photodamage is a result of recombination events leading to 1O2 formation. The role of reactive species in photoinhibition is controversial. While some studies show a direct damaging effect on PSII (Krieger-Liszkay et al. 2011, Song et al. 2006), others report on damage only to the repair process (Nishiyama et al. 2011). Under ultraviolet (UV) illumination there is a general agreement that the Mn cluster is damaged by direct absorption (Vass et al. 2002). However, there is an ongoing debate over whether this type of damage extends into the visible light range. Published action spectra of photoinhibition do not allow for a clear distinction between the acceptor side, donor side or two-step donor side mechanisms (for definitions refer to Table 1). Certain mechanisms are mutually exclusive. For example, acceptor side and donor side-induced photoinhibition cannot happen simultaneously in the same reaction centre. On the other hand, all proposed mechanisms can finally lead to a partial or complete disintegration of chlorophyll protein complexes resulting in an increase in the free pigments content and 1O2 production. Based on the mechanistic properties of the proposed pathways, different quantum efficiencies are expected. In the visible range, the low light acceptor side mechanism has the highest quantum yield (Keren et al. 1995); the two-step donor side mechanism has a low quantum yield. As a function of exposure time, the two-step mechanism is expected to show a linear behaviour (Hakala et al. 2005). Photoinhibition caused by free pigments will rise exponentially, while the acceptor side mechanisms are expected to reflect the reduction state of the plastoquinone acceptors. Using single turnover flashes, an oscillation pattern depending on the quinone reduction states involved was shown (Keren et al. 1995). The photoinhibitory process is a function of time and light intensity. Different mechanisms may operate during the time course of photoinhibition depending on the physiological state of the sample. Within a leaf, there will be areas with higher and lower light exposure based on the morphology and optical properties of the leaf. Processes with different quantum efficiencies are likely to occur in different regions at the same time (Oguchi et al. 2011). Moreover, with the progression of light stress, one mechanism may be replaced by the next one. In order to systematically compare the different mechanisms, attention has to be paid to the choice of organisms, growth conditions and experimental parameters. The response to different light intensities for a single plant species can vary according to its growth conditions. Plants grown under high light intensities will exhibit higher resistance towards photoinhibition than plants of the same species grown under low light. These differences can be partially resolved by providing the light saturation point of photosynthetic electron transport, for the plants used in the experiment. Light intensity normalised to the light saturation point, should be provided in order to make data comparable. Furthermore, to allow distinction between the different photoinhibitory mechanisms their quantum efficiencies, using the initial rate of photoinhibition, together with the related action spectra would be required. Resolution of the action spectra should take into account the light intensity (measured at both subsaturating and oversaturating intensities). Finally, to elucidate the nature of the primary damaging species (initiation of the D1 degradation), the putative site of protein oxidation has to be identified. This identification may also allow differentiating between oxidative damage by 1O2, O or OH• radicals or P+/Y. Attempts to resolve the problem were undertaken in the past (Barber 1998) but failed because of the inadequacy of the analytical techniques available at that time. Nowadays, with current advances in peptide mass spectrometry, identification of the primary target should be possible. On a physiological level, the exact price of photoinhibition is very difficult to estimate (Raven 2011) and the question which needs to be resolved is whether PSII photoinhibition should be regarded as a damaging process or rather as a protection mechanism (Somersalo and Krause 1990). In PSII, the photodamage of the PSII D1 protein can be regarded as a predetermined breaking point, a biological fuse. A specialised breakdown and repair machinery has been developed during evolution to guarantee a fast degradation and resynthesis of the D1 protein. Turnover of PSII is 'cheaper' and faster than the repair of PSI (Kudoh and Sonoike 2002, Sonoike 2011). The rate of degradation of the D1 protein can reach close to 80% of its full extent below light saturation (Edelman and Mattoo 2008) demonstrating the importance of this process under physiologically relevant conditions. The fact that recovery is inhibited by reactive oxygen species (Nishiyama et al. 2011) fits well with a protective role of PSII photoinhibition for the whole photosynthetic apparatus.