Network analysis with quantum dynamics clarifies why photosystem II exploits both chlorophyll a and b

Abstract

In green plants, chlorophyll a and chlorophyll b are the predominant pigments bound to light-harvesting proteins. While the individual characteristics of these chlorophylls are well understood, the advantages of their coexistence remain unclear. In this study, we establish a method to simulate excitation energy transfer within the entire photosystem II supercomplex by using network analysis integrated with quantum dynamic calculations. We then investigate the effects of the coexistence of chlorophyll a and chlorophyll b by comparing various chlorophyll compositions. Our results reveal that the natural chlorophyll composition allows the excited energy to preferentially flow through specific domains that act as safety valves, preventing downstream overflow. Our findings suggest that the light-harvesting proteins in a photosystem II supercomplex achieve evolutionary advantages with the natural chlorophyll a/b ratio, capturing light energy efficiently and safely across various light intensities. Using our framework, one can better understand how green plants harvest light energy and adapt to changing environmental conditions.

Fig. 1.. EET network analysis of natural PSII SC.

(A) Protein compositions and Chl distribution of the PSII SC. (B) Schematic representations of the Chl domains generated via excitonic coupling between Chls. (C) Excitation energy transfer (EET) network and site energies of Chl domains in the natural PSII SC. The size and color of the circles represent the number of Chls in the domain and its averaged site energy. The direction and line width of the arrows between the circles represent the direction and the relative rate constants for EET. The rate constants larger than 0.005 ps⁻¹ (time constant of 200 ps) are shown to visualize dominant EET (fig. S1). (D) Excitation probability of domains at t = 0 due to random Chl excitation for EET simulation considering the charge separation by the special pairs and intrinsic dissipation processes of Chls. (E) Simulated ensemble probabilities of the excited states of Chls, charge separation, and intrinsic dissipation after initial excitation.

Fig. 2.. EET network analysis of all-a– and all-b–type PSII SCs.

(A and C) The EET network and site energies of Chl domains in the all-a– and all-b–type PSII SCs, respectively. The size and color of the circle represent the number of Chls in the domain and its averaged site energy. The direction and line width of the arrows between circles represent the direction and relative rate constants for EET. The rate constants larger than 0.005 ps⁻¹ (time constant of 200 ps) are shown to visualize dominant EET (fig. S1). (B and D) Simulated ensemble probabilities of the excited states of Chls, charge separation, and intrinsic dissipation after initial excitation of a Chl in the all-a– and all-b–type PSII SCs, respectively. Solid lines represent the simulation results of either all-a– or all-b–type PSII SCs, and dashed lines represent the results of the natural PSII SC. (E) Individual yield, Q(Y), of domains in the natural-, all-a–, and all-b–type PSII SCs. Q(Y) of the domain represents the yield of charge separation when the domain is excited. Q(Y) of RC domains was not included in the scale map of Q(Y).

Fig. 3.. Cumulative flow analysis of PSII SCs.

(A to C) Relative CI of each domain and NCF of each link in the natural-, all-a–, and all-b–type PSII SCs, respectively. The Chls in the high CI domains are shown in the red boxes for each domain.

Fig. 4.. Light-harvesting and photoprotective capabilities of PSII SCs.

(A) The relationship between the proportion of Chl b in the antenna of the PSII SCs compared with the relative absorption of solar radiation (gray dashed line) and the consequent relative net light-harvesting efficiency (green circles and line). (B) Non-photochemical quenching (NPQ) candidate domains (red circles) and RCs (green circles) and the major pathway of energy flow in the natural-type PSII SC. Major pathways selectively represent the top 1% NCF (fig. S8). (C) Simulated NPQ values [(yield without NPQ)/(yield with NPQ) − 1] of the natural-type PSII SC with various combinations of the rate constants of NPQ in trimeric LHCII and CP29 (table S5). (D) Simulated NPQ values of the natural-, all-a–, and all-b–type PSII SCs with the rate constants of NPQ in trimeric LHCII (tables S5 to S7).