The Evolution of Photosystem II Subunit S across the Green Lineage

Abstract

Light absorbed by plants can have three different fates: it can be used to drive photosynthetic electron transport, be re-emitted as fluorescence or be dissipated as heat (non-photochemical quenching, NPQ). When light levels are too high, NPQ is an important protective process that prevents damage to membranes and proteins in the chloroplast. In vascular plants, the major trigger of fast-inducing energy-dependent NPQ is the Photosystem II subunit S protein (PsbS), which is activated in response to high light and interacts with photosystem II to cause quenching. Algae and moss, however, are thought to predominantly use another protein to regulate NPQ, the Light Harvesting Complex Stress-Related protein (LHSCR), despite having gene copies of PsbS. Therefore, there has been a broad-scale transition from LHCSR-dependent NPQ to PsbS-dependent NPQ during plant evolution but how and why PsbS became the major trigger of NPQ and whether this required neofunctionalisation is still unclear. To answer these questions, we performed a phylogenetic analysis that uses a newly collected and curated dataset of PsbS sequences corresponding to 1,007 species to analyse both overall structural similarity and the conservation of key residues that have been identified to be important for PsbS function (Chapter 2). We found that PsbS is very well-conserved both at a structural and residue level across the whole green lineage. In fact, 92% of the protein is under negative selection, while only one residue was found to be both under positive selection and undergo recurrent mutations. Based on these findings, we saw no evidence of pronounced sequence changes aligning with PsbS’ rise to become the dominant quenching regulator. To support the theoretical analyses in Chapter 2, we tested functional complementation of PsbS sequences from 11 species across the green lineage via expression in the PsbS-knockout mutant in Arabidopsis thaliana in Chapter 3. Significant restoration of NPQ was observed with ten of these PsbS sequences, supporting the predictions of Chapter 2 that PsbS has likely been functional as an NPQ trigger since its early origins in the green lineage one billion years ago. In Chapter 4, the impact of key residues was evaluated using the same complementation approach as in Chapter 3. Work focused on the effect of the residue identified in the selection analyses in Chapter 2 to be under positive selection and recurrent mutation (alignment position 175 corresponding to Q183 in Arabidopsis) as well as two glutamates in the lumenal loops which are closely situated to the two already identified in previous studies to be instrumental to function. Excitingly, mutation to the ancestral state, glycine (Q183G), significantly reduced the maximal NPQ, while mutation to most recurrently evolved residue, glutamate (Q183E), significantly increased the maximal amount of NPQ compared to the control, without changing protein accumulation. This is the first time that a novel residue is identified based on recurrent mutation evolutionary analyses and subsequently validated, as well as the only gain-of-function mutation in PsbS identified to date. Furthermore, mutation of E131Q was found to strongly impair NPQ induction to about 50% compared to the control lines, while mutation of E129Q led to no reduction in maximal NPQ but slowed down relaxation of NPQ upon return to darkness, consistent with recent work pertaining to this lumenal domain. Finally, Chapter 5 is a physiological characterisation of PsbS-knockout lines in maize (Zea mays B104), a C4 species. C4 species represent a significant example of convergent evolution and have been observed to have an increased contribution of the fast-quenching component of NPQ, but little is known about the role of PsbS in these species. Here, we aimed to determine the impact of knocking out PsbS in a C4 species and to find out whether this varies significantly from a C3 plant (Arabidopsis). CRISPR-Cas9 gene editing was used to create several PsbS-knockout lines in maize. Characterisation of these lines showed that NPQ is strongly decreased but assimilation is not significantly affected by PsbS-knockout in maize compared to wildtype in both steady state and fluctuating light conditions, which aligns with previous observations in C3 plants. Altogether, this thesis leverages the massive proliferation in publicly available sequence information to provide the first largescale evolutionary analysis of PsbS across the green lineage. The findings represent major advances in the field by challenging the current consensus about the ancestral role of PsbS in the algae, validating novel methodology to identify residues under adaptive evolution, demonstrating two loss-of-function and one first gain-of-function mutations and providing the first characterisation of the impact of knocking out PsbS in a C4 species.