Photosystem

A photosystem is a large complex of proteins and light-absorbing pigments, embedded in a cell membrane, that uses light energy to drive the first steps of photosynthesis. In plants, green algae and cyanobacteria, two photosystems, called photosystem I (PSI) and photosystem II (PSII), work together in the thylakoid membranes, the internal membranes where the light-dependent reactions of photosynthesis take place.^[1:1]

Photosystem II oxidizes water, which supplies electrons and releases molecular oxygen, while photosystem I produces the most negative redox potential known in nature.^[2:1] Both belong to a family of photosynthetic reaction centers that appear to share a common ancestor.^[3:1]

Overview

Oxygenic photosynthesis, the form of photosynthesis that releases oxygen, depends on four multi-subunit membrane-protein complexes: photosystem I, photosystem II, the cytochrome b6f complex and ATP synthase (F-ATPase).^[2:1] The two photosystems sit at opposite ends of the light-driven electron flow. Photosystem II generates an oxidant strong enough to oxidize water, a substrate abundant enough to provide an effectively unlimited source of electrons. Photosystem I generates the most negative redox potential in nature, and so largely determines the amount of chemical energy stored in living systems.^[2:1]

Photosystem II

Photosystem II is the site of photosynthetic water oxidation. A crystal structure of the cyanobacterial complex at 1.9 Å resolution showed that it contains 20 subunits with a total molecular mass of about 350 kDa.^[4:1]

Water is split at a cluster of four manganese atoms and one calcium atom, written Mn4CaO5. The 1.9 Å structure located every metal atom of the cluster together with its ligands. Five oxygen atoms act as oxo bridges linking the five metal atoms, and four water molecules are bound to the cluster; some of these may serve as the substrates for forming dioxygen. The structure also identified more than 1,300 water molecules in each PSII monomer, some of which form hydrogen-bonded networks that may act as channels for protons, water or oxygen.^[4:2]

The S-state cycle

Photosystem II oxidizes water through a four-step cycle of intermediate states, labeled S0 to S4, at the Mn4CaO5 cluster. An extra oxygen atom, called O6, is incorporated at the S3 state and may form a dioxygen bond. Time-resolved X-ray crystallography with femtosecond X-ray pulses has been used to follow structural changes in the complex from nanoseconds to milliseconds after illumination with one or two flashes of light.^[5:1]

Photosystem I

The first 2.5 Å crystal structure of photosystem I, from the thermophilic cyanobacterium Synechococcus elongatus, resolved 12 protein subunits and 127 cofactors: 96 chlorophylls, 2 phylloquinones, 3 Fe4S4 clusters, 22 carotenoids, along with lipids, a probable calcium ion and water molecules.^[1:2]

In plants, photosystem I is a supercomplex: a reaction center combined with light-harvesting complexes that collect light and pass the energy inward. A 3.4 Å structure of plant PSI revealed 17 protein subunits and 168 chlorophylls, along with 2 phylloquinones, 3 Fe4S4 clusters and 5 carotenoids.^[6:1]

Evolution

The reaction centers of photosystems I and II, and those of photosynthetic bacteria such as proteobacteria, show a conserved core of five transmembrane helices, even though their residual sequence identity is under 10%. Sadekar, Raymond and Blankenship conclude that this strongly implies common ancestry, and their phylogenetic trees from structural alignments suggest that the earliest reaction center was a homodimeric complex (built from two identical halves) that did not produce oxygen.^[3:1]

When the ability to split water arose remains uncertain. Cardona and colleagues used sequence and structural comparisons with Bayesian relaxed molecular clocks to infer that a homodimeric photosystem with enough oxidizing power to split water had appeared in the early Archean, roughly a billion years before the most recent common ancestor of all described oxygenic cyanobacteria. They hypothesize, on structural and functional grounds, that this early photosystem could already oxidize water.^[7:1] This is an inference from molecular-clock models rather than a direct observation, so the date depends on the assumptions built into those models.

Open questions

The S-state cycle has been captured in structural snapshots, but the chemistry of O-O bond formation, including the role of the extra oxygen incorporated at the S3 state, is described in the literature as a "possible" dioxygen-forming step, not a settled mechanism.^[5:1] The timing of the origin of water oxidation could be tested by further comparative sequence and structural analyses, and by new calibrations for the molecular clocks used to date it.