Recent magnetic-resonance work on YŻ suggests that this species exhibits considerable motional flexibility in its functional site and that its phenol oxygen is not involved in a well-ordered hydrogen-bond interaction (Tang et al., submitted; Tommos et al., in press). Both of these observations are inconsistent with a simple electron-transfer function for this radical in photosynthetic water oxidation. By considering the roles of catalytically active amino acid radicals in other enzymes and recent data on the water-oxidation process in Photosystem II, we rationalize these observations by suggesting that YŻ functions to abstract hydrogen atoms from aquo- and hydroxy-bound managanese ions in the (Mn)4 cluster on each S-state transition. The hydrogen-atom abstraction process may occur either by sequential or concerted kinetic pathways. Within this model, the (Mn)4/YZ center forms a single catalytic center that comprises the Oxygen Evolving Complex in Photosystem II.

The reaction center of photosystem II of oxygenic photosynthesis contains two redox-active tyrosines called Z and D, each of which can act as an electron donor to the oxidized primary electron donor, P680+. These tyrosines are located in homologous positions on the third transmembrane alpha-helix of each of the two homologous polypeptides, D1 and D2, that comprise the reaction center. Tyrosine D of polypeptide D2 has been proposed, upon oxidation, to give up its phenolic proton to a nearby basic amino acid residue, forming a neutral radical. Modeling studies have pointed to His190 (spinach numbering) as a likely candidate for this basic residue. As a test of this hypothesis, we have constructed three site-directed mutations in the D2 polypeptide of the cyanobacterium Synechocystis sp. PCC6803. His189 (the Synechocystis homologue of His190 of spinach) has been replaced by glutamine, aspartate, or leucine. Instead of the normal D. EPR signal (g = 2.0046; line width 16-19 G), PSII core complexes isolated from these three mutants show an altered dark-stable EPR signal with a narrowed line width (11-13 G), and g values of 2.0046, 2.0043, and 2.0042 for the His189Gln, His189Asp, and His189Leu mutants, respectively. Despite the reduced line width, these EPR signals show g values and microwave-power saturation properties similar to the normal D. signal. Furthermore, specific deuteration in one of those mutants at the 3 and 5 positions of the phenol ring of the photosystem II reaction center tyrosines results in a loss of hyperfine structure of the EPR signal, proving that the signal indeed arises from tyrosine.2+ This observation provides support for a model in which an imidazole nitrogen of His189 accepts the phenolic proton of Tyr160 upon oxidation of D, forming a back hydrogen bond to the phenolic oxygen of the neutral tyrosyl radical.

Saturation-recovery EPR spectroscopy has been used to probe the location of the redox-active tyrosines, YD (tyrosine 160 of the D2 polypeptide, cyanobacterial numbering) and YZ (tyrosine 161 of the D1 polypeptide), relative to the non-heme Fe(II) in Mn-depleted photosystem II (PSII). Measurements have been made on PSII membranes isolated from spinach and on PSII core complexes purified from the cyanobacterium Synechocystis sp. PCC 6803. In the case of Synechocystis PSII, site-directed mutagenesis of the YD residue to either phenylalanine (Y160F) or methionine (Y160M) was done to eliminate the dark-stable YD.species and, thereby, allow direct spectroscopic observation of the YZ. EPR signal. The spin-lattice relaxation transients of both YD. and YZ. were non-single-exponential due to a dipolar interaction with one of the other paramagnetic species in PSII. Measurements on CN(-)-treated, Mn-depleted cyanobacterial PSII, in which the non-heme Fe(II) was converted into its low-spin, diamagnetic state, proved that the non-heme Fe(II) was the sole spin-lattice relaxation enhancer for both the YD. and YZ. radicals. This justified the use of a dipolar model in order to fit the saturation-recovery EPR data, which were taken over the temperature range 4-70 K. The dipolar rate constants extracted from the fits were identical in magnitude and had the same temperature dependence for both YD. and YZ.. The observation of identical dipolar interactions between YD. and YZ. and the non-heme Fe(II) shows that the distance from each tyrosine to the non-heme Fe(II) is the same.(ABSTRACT TRUNCATED AT 250 WORDS)

Endoplasmic reticulum (ER) membrane junctions are formed by the dynamin-like GTPase atlastin (ATL). Deletion of ATL results in long unbranched ER tubules in cells, and mutation of human ATL1 is linked to hereditary spastic paraplegia. Here, we demonstrate that COPII formation is drastically decreased in the periphery of ATL-deleted cells. ER export of cargo proteins becomes defective; ER exit site initiation is not affected, but many of the sites fail to recruit COPII subunits. The efficiency of cargo packaging into COPII vesicles is significantly reduced in cells lacking ATLs, or when the ER is transiently fragmented. Cargo is less mobile in the ER in the absence of ATL, but the cargo mobility and COPII formation can be restored by ATL R77A, which is capable of tethering, but not fusing, ER tubules. These findings suggest that the generation of ER junctions by ATL plays a critical role in maintaining the necessary mobility of ER contents to allow efficient packaging of cargo proteins into COPII vesicles.