We report low temperature (T) optical spectra of the isolated CP47 antenna complex from Photosystem II (PSII) with a low-T fluorescence emission maximum near 695 nm and not, as previously reported, at 690-693 nm. The latter emission is suggested to result from three distinct bands: a lowest-state emission band near 695 nm (labeled F1) originating from the lowest-energy excitonic state A1 of intact complexes (located near 693 nm and characterized by very weak oscillator strength) as well as emission peaks near 691 nm (FT1) and 685 nm (FT2) originating from subpopulations of partly destabilized complexes. The observation of the F1 emission is in excellent agreement with the 695 nm emission observed in intact PSII cores and thylakoid membranes. We argue that the band near 684 nm previously observed in singlet-minus-triplet spectra originates from a subpopulation of partially destabilized complexes with lowest-energy traps located near 684 nm in absorption (referred to as AT2) giving rise to FT2 emission. It is demonstrated that varying contributions from the F1, FT1, and FT2 emission bands led to different maxima of fluorescence spectra reported in the literature. The fluorescence spectra are consistent with the zero-phonon hole action spectra obtained in absorption mode, the profiles of the nonresonantly burned holes as a function of fluence, as well as the fluorescence line-narrowed spectra obtained for the Q(y) band. The lowest Q(y) state in absorption band (A1) is characterized by an electron-phonon coupling with the Huang-Rhys factor S of approximately 1 and an inhomogeneous width of approximately 180 cm(-1). The mean phonon frequency of the A1 band is 20 cm(-1). In contrast to previous observations, intact isolated CP47 reveals negligible contribution from the triplet-bottleneck hole, i.e., the AT2 trap. It has been shown that Chls in intact CP47 are connected via efficient excitation energy transfer to the A1 trap near 693 nm and that the position of the fluorescence maximum depends on the burn fluence. That is, the 695 nm fluorescence maximum shifts blue with increasing fluence, in agreement with nonresonant hole burned spectra. The above findings provide important constraints and parameters for future excitonic calculations, which in turn should offer new insight into the excitonic structure and composition of low-energy absorption traps.

In this work, we present simulated steady-state absorption, emission, and nonresonant hole burning (HB) spectra for the CP47 antenna complex of photosystem II (PS II) based on fits to recently refined experimental data (Neupane et al. J. Am. Chem. Soc. 2010, 132, 4214). Excitonic simulations are based on the 2.9 Å resolution structure of the PS II core from cyanobacteria (Guskov et al. Nat. Struct. Mol. Biol. 2009, 16, 334), and allow for preliminary assignment of the chlorophylls (Chls) contributing to the lowest excitonic states. The search for realistic site energies was guided by experimental constraints and aided by simple fitting algorithms. The following experimental constraints were used: (i) the oscillator strength of the lowest-energy state should be approximately ≤0.5 Chl equivalents; (ii) the excitonic structure must explain the experimentally observed red-shifted (∼695 nm) emission maximum; and (iii) the excitonic interactions of all states must properly describe the broad (non-line-narrowed, NLN) HB spectrum (including its antihole) whose shape is extremely sensitive to the excitonic structure of the complex, especially the lowest excitonic states. Importantly, our assignments differ significantly from those previously reported by Raszewski and Renger (J. Am. Chem. Soc. 2008, 130, 4431), due primarily to differences in the experimental data simulated. In particular, we find that the lowest state localized on Chl 526 possesses too high of an oscillator strength to fit low-temperature experimental data. Instead, we suggest that Chl 523 most strongly contributes to the lowest excitonic state, with Chl 526 contributing to the second excitonic state. Since the fits of nonresonant holes are more restrictive (in terms of possible site energies) than those of absorption and emission spectra, we suggest that fits of linear optical spectra along with HB spectra provide more realistic site energies.

The PSI-CP43 supercomplex of the cyanobacterium Synechocystis PCC 6803, grown under iron-starvation conditions, consists of a trimeric core Photosystem I (PSI) complex and an outer ring of 18 CP43 light-harvesting complexes. We have investigated the electronic structure and excitation energy transfer (EET) pathways within the CP43 (also known as the isiA gene product) ring using low-temperature absorption, fluorescence, fluorescence excitation, and holeburning (HB) spectroscopies. Analysis of the absorption spectra of PSI, CP43, and PSI-CP43 complexes suggests that there are 13 chlorophylls (Chls) per CP43 monomer, i.e., a number that was observed in the CP43 complex of Photosystem II (PSII) (Umena, Y. et al. Nature 2011, 473, 55-60). This is in contrast with the recent modeling studies of Zhang, Y. et al. (Biochim. Biophys. Acta 2010, 1797, which suggested that IsiA likely contains 15 Chls. Modeling studies of various optical spectra of the CP43 ring using the uncorrelated EET model (Zazubovich, V.; Jankowiak, R. J. Lumin. 2007, 127, 245-250), suggest that CP43 monomers (in analogy to the CP43 complexes of PSII core) also possess two quasi-degenerate low-energy states, A and B. The site distribution functions of state A and B maxima/full width at half maximum (FWHM) are at 684 nm/180 cm , respectively. Our analysis shows that pigments mostly contributing to the lowest-energy A and B states must be located on the side of the CP43 complex facing the PSI core, a finding that contradicts the model of Zhang et al., but is in agreement with the model suggested by Nield et al. (Biochemistry 2003, 42, 3180-3188). We demonstrate that the A-A and B-B EET between different monomers is possible, though with a slower rate than intra-monomer A-B and/or B-A energy transfer.