Discussion
The evidence presented here, at both spectral and protein levels, confirmed that OEC of Z. marina was preferentially inactivated in response to light exposure: (1) PSII electron donation capacity characterized by L 1 and W Kwas significantly decreased already during initial illumination; (2) OEC activity represented by the contents of OEC peripheral proteins (Bricker and Frankel, 2008; Popelkova et al., 2011) greatly decreased, while the reduction degree of QA at the acceptor-side, represented by 1 - V J, slightly increased; (3) PSII photoinhibition occurred upon light exposure and the photoinhibition rate constant, determined via first-order kinetic fits, was positively correlated with PPFD at all tested light intensities. This was the first in vivo study of PSII donor-side photoinhibition with the direct experimental evidence in higher plants under visible light. The in vivo occurrence of this mechanism in higher plants enabled the study of both integrated photosynthetic characteristics and photoprotective mechanisms, which involved multiple coordinated systems.
The present study investigated the integrated photosynthetic regulatory mechanisms involved in light absorption, transfer and conversion, and energy dissipation via chloroplast proteome combined with chlorophyll fluorescence phenotype. The strong interactions of PsbO with Lhcb3, PGR5, HCF244, Psb29, PsaN, and PsaH, shown in the PPI network, suggested that OEC played a central role in the regulation of photosynthesis. The decreased size of the PSI and PSII antennas indicated a decline of the light-harvesting capacity inZ. marina . The interactions of PsbO with PsaH and Lhcb3 were interpreted as a need to decrease the light-harvesting capacity due to the partially impaired OEC. It should be realized that a decrease of PSII antenna size corresponds to fewer absorbed photons but also to a faster trapping rate (Croce and Van Amerongen, 2014). Simultaneously, PSII assembly was activated to retain PSII functionality. Because continuously impaired OEC led to a relatively oxidized PQ pool (Makarova et al., 2007), the PGR5-dependent PSI-CEF in Z. marina was activated to supplement the PSI linear electron transport. Activated PSI-CEF revealed by proteome data in the present study was in agreement with chlorophyll fluorescence physiological phenotype observed in our previous study (Yang et al., 2017). The interactions between PsbO and HCF244, as well as PGR5 and PsaN indicated the potential of the partially impaired OEC to enhance PSII repair and activate the PGR5-dependent PSI-CEF.
Continuous impairment of OEC in this mechanism suggested electron deficiency in the electron transport chain. Unchanged antioxidant levels and unactivated alternative electron flows, such as chlororespiration, Mehler reaction, malic acid synthesis and photorespiration, supported this viewpoint. The low PSII excitation pressure caused by electron deficiency made it difficult to form a strong trans-thylakoid proton gradient (ΔpH), and consequently, could not induce strong NPQ. This was verified by the unchanged ΔpH sensor PsbS protein, the low value of NPQmax (Schubert et al., 2015), and slightly upregulated ZEP and VDE. NPQ is usually dissected into at least three kinetic components, in which qE, induced within 20-60 s, was the dominant NPQ component (Horton et al., 1996; Nilkens et al., 2010). Further examination of the dynamics of NPQ induction inZ. marina indicated the absence of the fast qE component, which accounted for the slow development of NPQ. Moreover, the decrease of NPQ during the late illumination indicated that the demand for dissipation decreased with the increased OEC impairment. A low NPQ capacity has also been reported in the seagrass Posidonia sinuosa (Schubert et al., 2015).
The low levels of NPQ were conductive to the formation of the reduced form of NADPH and ATP derived from photosynthetic electron transport, which in turn contributed to carbon fixation. In C3 plants, photosynthesis is typically restricted by the Rubisco capacity due to its exceedingly low catalytic turnover rate and competition with oxygenation reaction (Farquhar et al., 1980). The significantly enhanced carboxylase activity of Rubisco and the upregulated NTRC (Nikkanen et al., 2016) suggested an increased CO2 fixation rate in response to light exposure. Recoveries of activities of GAPDH and PRK, the key enzymes of carbon fixation, were achieved by downregulated CP12 (Marri et al., 2009), suggesting the activation of carbon fixation pathway. The well photosynthetic performance during photoinhibition was also supported by the normal photosynthetic rate characterized by O2 evolution rate in Z. marina .
1O2 is usually formed by a photosensitization reaction in which an oxygen molecule reacts with3P680 (Pospíšil, 2016). The sensitizer molecule3P680, produced by3[P680+Phe] which is generated either via spin conversion of primary radical pair P680+Phe or via recombination of P680+QA, is triggered by the over-reduction of QA (Vass et al., 1992; Hakala et al., 2005; Ohnishi et al., 2005; Vass, 2011). The reduction degree of QA was low in Z. marina , which was evidenced by a slight decrease of 1 - V J. Therefore, the conditions for 1O2 production were not appropriate. In fact, light exposure neither caused an upregulation of PsbH and GPX, nor an increase of1O2levels. The strong interaction between PsbO and PsbH indicated that there was no need for PsbH to stabilize the combination of excess β-carotene and PSII (Hall et al., 2016) to eliminate1O2. Moreover, the inhibition of alternative electron donation in our study resulted in a significant damage of PSII RC described by the net losses of D1 and CP43 proteins, indicating that the photodamage of PSII was attributed to the long-lived P680+rather than1O2. Thus, the long-lived P680+ resulted from the interrupted electron supply was usually considered as the damage source in the PSII donor-side photoinhibition, which has been suggested by Vass (2012) and Tyystjärvi (2008). Noticeably, a distinct low descending amplitude ofF v/F mwas observed during light exposure, which could be explained by most PSII RC remaining open to contribute to the depletion of the pool of high-potential P680+ via facilitating the direct charge recombination of the P680+QA. Because the competition between3[P680+Phe] formation and the direct recombination of the P680+QA was controlled by PSII excitation pressure (Vass and Cser, 2009), the low reduction degree of QAin Z. marina prevented the formation of3[P680+Phe].
The existence of alternative PSII electron donors has been reported in some species (Thompson and Brudvig, 1988; Tóth et al., 2009, 2011). Based on in vivo data, the increased AsA level and the upregulated GLDH content involved in AsA synthesis following light exposure provided a clue of AsA as alternative electron donor. Similarly, in contrast to 62 μM DMBQ enhancing PSII-CEF, 125 μM DMBQ uncoupling PSII-CEF induced the lower PSII electron donation capacity and PSII photochemical activity, providing a clue of PSII-CEF as another alternative donation pathway. Furthermore, a factorial design experiment with different combinations of AsA and PSII-CEF inhibitions demonstrated that the suppression of alternative electron donation damaged the PSII component, including a decrease inF v/F m, an increase inW K, as well as net losses of PSII RC proteins and OEC peripheral proteins. With the duration of light exposure, the inhibition effect became more significant, of which, the most severe damage of PSII was caused by the dual inhibition of AsA and PSII-CEF. Based on these results, we suggested that both AsA and PSII-CEF were important photoprotective mechanisms, which provided electrons to remit the oxidative stress from long-lived P680+ during light exposure.
In conclusion, the water-splitting dysfunction caused by OEC photoinactivation interrupted the electron supply from water to the oxidized primary donor P680+, resulting in the damage to the PSII RC ofZ. marina . At least PSII-CEF and the alternative donor AsA exerted photoprotective roles in the depletion of P680+ by donating electrons to PSII (Fig. 7). In contrast to acceptor-side photoinhibition caused by electron excess, continuous impairment of OEC resulted in electron deficiency in the electron transport chain during the PSII photoinhibition. For the efficient use of the limited electrons, NPQ was inefficient and the alternative electron flows associated with energy dissipation, such as chlororespiration, Mehler reaction, malic acid synthesis and photorespiration, were not significantly activated to decrease unnecessary consumption (Fig. 7). The extremely sensitivity of OEC to visible light was presumably a result of insufficient polyphenol levels caused by a lack of blue light photoreceptors in a habitat that is rich in blue light (Peng and Moriguchi, 2013; Jiang et al., 2016).