Plants are able to dissipate excess absorbed energy as heat through several mechanisms collectively termed non-photochemical quenching (NPQ). Because these pathways can compete with photochemistry, they must be tightly regulated in order to optimize photosynthesis. In high light, the primary component of NPQ is feedback de-excitation (qE), which is known to require the protein PsbS. In order to investigate additional NPQ pathways and to identify novel proteins involved, I performed a suppressor screen in PsbS-deficient (npq4) mutants of Arabidopsis thaliana. One of the mutants identified, called suppressor of quenching 1 (soq1), displayed considerably higher NPQ relative to npq4 plants after illumination with high light.
The characteristics of the NPQ demonstrated by the soq1 mutant point to a new component of NPQ that has not been previously described. The NPQ seen in soq1 plants had slower induction and relaxation kinetics relative to qE, relaxing over the course of several hours. Additionally, fluorescence measurements of soq1 single mutants in an otherwise wild-type background demonstrated an additive capacity for quenching, and this quenching was dependent on the light intensity. The additional quenching formed independently of both changes in the pH gradient and zeaxanthin formation based on treatments with nigericin and dithiothreitol, respectively. The additional NPQ in soq1 was also still observed in xanthophyll mutants, and the pigment composition was identical in wild type and soq1 leaves before and after illumination. State transitions are also not involved in soq1 quenching, as quenching occurred in the absence of the STN7 kinase and did not cause changes in phosphorylation of thylakoid proteins or the 77 K fluorescence spectra when compared to wild type. Increased damage to reaction centers is unlikely to be the cause of quenching, as determined by treatment with lincomycin, light curves, and a seemingly wild-type growth phenotype in high light. Using time-resolved fluorescence, it was shown that quenching in soq1 mutants was the result of a decrease in average lifetime of excited chlorophyll.
The soq1 mutation was mapped to an uncharacterized chloroplast protein containing three domains: a haloacid dehalogenase-like hydrolase (HAD), a thioredoxin-like (Trx-like) fold and a beta-propeller. The NPQ phenotype of T-DNA lines and complementation with the wild-type SOQ1 gene confirmed that this locus was responsible for the soq1 phenotype, and that the soq1 mutants obtained by mutagenesis were loss-of-function alleles. SOQ1 is found in all green plants with the exception of prasinophytes and certain species of algae and monocots that have unique domain architecture. Trx-like and beta-propeller domains are also found together in diverse groups of organisms including animals, bacteria and archaea. I demonstrated that SOQ1 is a thylakoid membrane protein with both Trx-like and beta-propeller domains located in the lumen and the HAD domain in the stroma. Successful complementation with a truncated SOQ1 protein lacking the HAD domain indicated that it did not contribute to the formation of the NPQ seen in soq1 plants.
To determine the role of SOQ1 in the thylakoid and how it prevents NPQ, biochemical and high-resolution imaging techniques were used to explore differences between soq1 and wild-type plants. SOQ1 was shown to bind thylakoid proteins in plants treated with high light, however these interactions did not affect the composition of supercomplexes. To identify these interacting proteins, co-immunoprecipitation and yeast two-hybrid analysis were performed and indicated that SOQ1 might interact with HCF136 and At2G26340. Fluorescent labeling of reduced thiols using monobromobimane could lead to additional targets of SOQ1. Electron microscopy (EM) showed that soq1 plants did not appear to have significantly different chloroplast or thylakoid structures, except for a possible accumulation of plastoglobules in the soq1 background. When investigated using atomic force microscopy (AFM), the organization of supercomplexes within the grana was unique in soq1 mutants in both dark-adapted and high light-treated plants.
The role of the HAD domain of SOQ1 has also been investigated. I have shown that both the full-length protein and the HAD domain alone can act as phosphatases on small molecules. Initially, I demonstrated that SOQ1 removed phosphate groups from a general phosphate substrate, p-nitrophenyl phosphate. Using the malachite green assay, I narrowed SOQ1's substrate specificity to a pentose sugar phosphate. The highest activity was seen using ribose-5-phosphate as the substrate, and this activity was redox regulated and had a pH optimum near 7. Initial characterization of the sugar content of wild type and soq1 mutants may improve our understanding of the role of the HAD domain in vivo.
Other aspects of NPQ that I have investigated include the dynamic mechanical changes to the thylakoid that occur upon illumination. Using a modified AFM instrument, changes in the stiffness and height of thylakoid membranes were followed in real time. Upon illumination with photosystem II-specific light, an immediate increase in membrane stiffness was observed that was attributed to lumen expansion. Visco-elastic measurements suggest that some undetermined interaction between opposing lumen membranes was involved in this change. After a short delay, the change in stiffness was followed by an increase in the height of the membrane. Both the change in stiffness and height required formation of the pH gradient as determined by the addition of an uncoupler during the measurements. Using measurements on the stn7 mutant that is incapable of phosphorylating antenna complexes, the height change was shown to result from membrane unstacking.
The identification and characterization of a novel quenching pathway and the use of new AFM techniques to monitor dynamic changes in the thylakoid membrane will lead to a better understanding of how plants regulate photosynthesis.