Mitochondrial and cell-surface F<Subscript>0</Subscript>F<Subscript>1</Subscript>ATPsynthase in innate and acquired cardioprotection

Mitochondria are central to heart function and dysfunction, and the pathways activated by different cardioprotective interventions mostly converge on mitochondria. In a context of perspectives in innate and acquired cardioprotection, we review some recent advances in F₀F₁ATPsynthase structure/function and regulation in cardiac cells. We focus on three topics regarding the mitochondrial F₀F₁ATPsynthase and the plasma membrane enzyme, i.e.: i) the crucial role of cardiac mitochondrial F₀F₁ATPsynthase regulation by the inhibitory protein IF₁ in heart preconditioning strategies; ii) the structure and function of mitochondrial F₀F₁ATPsynthase oligomers in mammalian myocardium as possible endogenous factors of mitochondria resistance to ischemic insult; iii) the external location and characterization of plasma membrane F₀F₁ ATP synthase in search for possible actors of its regulation, such as IF₁ and calmodulin, at cell surface.     

Manipulating the Length of the <Emphasis Type="Italic">b</Emphasis> Subunit F<Subscript>1</Subscript> Binding Domain in F<Subscript>1</Subscript>F<Subscript>0</Subscript> ATP Synthase from <Emphasis Type="Italic">Escherichia coli</Emphasis>

The peripheral stalk of F₁F₀ ATP synthase is composed of a parallel homodimer of b subunits that extends across the cytoplasmic membrane in F₀ to the top of the F₁ sector. The stalk serves as the stator necessary for holding F₁ against movement of the rotor. A series of insertions and deletions have been engineered into the hydrophilic domain that interacts with F₁. Only the hydrophobic segment from {val-121} to {ala-132} and the extreme carboxyl terminus proved to be highly sensitive to mutation. Deletions in either site apparently abolished enzyme function as a result of defects is assembly of the F₁F₀ complex. Other mutations manipulating the length of the sequence between these two areas had only limited effects on enzyme function. Expression of a b subunit with insertions with as few as two amino acids into the hydrophobic segment also resulted in loss of F₁F₀ ATP synthase. However, a fully defective b subunit with seven additional amino acids could be stabilized in a heterodimeric peripheral stalk within a functional F₁F₀ complex by a normal b subunit.     

Functional and stoichiometric analysis of subunit e in bovine heart mitochondrial F<Subscript>0</Subscript>F<Subscript>1</Subscript>ATP synthase

The role of the integral inner membrane subunit e in self-association of F₀F₁ATP synthase from bovine heart mitochondria was analyzed by in situ limited proteolysis, blue native PAGE/iterative SDS-PAGE, and LC-MS/MS. Selective degradation of subunit e, without disrupting membrane integrity or ATPase capacity, altered the oligomeric distribution of F₀F₁ATP synthase, by eliminating oligomers and reducing dimers in favor of monomers. The stoichiometry of subunit e was determined by a quantitative MS-based proteomics approach, using synthetic isotope-labelled reference peptides IAQL*EEVK, VYGVGSL*ALYEK, and ELAEAQEDTIL*K to quantify the b, γ and e subunits, respectively. Accuracy of the method was demonstrated by confirming the 1:1 stoichiometry of subunits γ and b. Altogether, the results indicate that the integrity of a unique copy of subunit e is essential for self-association of mammalian F₀F₁ATP synthase. Elena Bisetto and Paola Picotti contributed equally to this work.     

Interconversion Between Dimers and Monomers of Endogenous Mitochondrial F<Subscript>1</Subscript>-Inhibitor Protein Complexes and the Release of the Inhibitor Protein. Spectroscopic Characteristics of the Complexes

The F₁-inhibitor protein complex (F₁-IP) was purified from heart submitochondrial particles. Size exclusion chromatography of the endogenous complex showed that it contains dimers (D) and monomers (M) of F₁-IP. Further chromatographic analysis showed that D and M interconvert. At high protein concentrations, the interconversion reaction is shifted toward the D species. The release of the inhibiting action of IP is faster at low than at high protein concentrations. During activation of F₁, the M species accumulates through a process that is faster than the release of IP from F₁. These findings indicate that the activation of F₁-IP involves the transformation of D into M, which subsequently loses IP. The spectroscopic characteristics of D, M, and free F₁ show that the binding of IP and dimerization modifies the fluorescence intensity of tyrosine residues and that of the single tryptophan of F₁ which is far from the IP binding site.     

Protein profiling in F<Subscript>1</Subscript> and F<Subscript>2</Subscript> generations of two tomato genotypes differing in ripening time

Pericarp polypeptide profiles were analyzed at three ripening stages in the F₁ hybrid and the F₂ population from the cross between the accessions: LA1385 (Lycopersicon esculentum var. cerasiforme) and 804627 (L. esculentum, a homozygous genotype for the nor mutant). Six polymorphic polypeptides were observed in LA1385, while no polymorphic polypeptides among ripening stages was observed in 804627. On the other hand, some polypeptides in the F₁ hybrid were not observed in the parents whereas others were present in both parental genotypes and were unnoticeable in the hybrid genotype. From a cluster analysis on the protein profiles of the F₂ population, the differential expression of proteins allowed to distinguish mature green (MG) stage from the others two stages, while for breaker stage (BR) and red ripe stage, the genetic background was more important in forming groups. The differential expression of proteins could be associated with fruit morphology traits such as a 72 kDa polypeptide present in MG stage with fruit diameter, height and mass and a 47 kDa polypeptide found in BR with fruit shelf life.     

Solution Structure of a Novel C<Subscript>2</Subscript>-Symmetrical Bifunctional Bicyclic Inhibitor Based on SFTI-1

A novel bifunctional bicyclic inhibitor has been created that combines features both from the Bowman–Birk inhibitor (BBI) proteins, which have two distinct inhibitory sites, and from sunflower trypsin inhibitor-1 (SFTI-1), which has a compact bicyclic structure. The inhibitor was designed by fusing together a pair of reactive loops based on a sequence derived from SFTI-1 to create a backbone-cyclized disulfide-bridged 16-mer peptide. This peptide has two symmetrically spaced trypsin binding sites. Its synthesis and biological activity have been reported in a previous communication [Jaulent and Leatherbarrow, 2004, PEDS 17, 681]. In the present study we have examined the three-dimensional structure of the molecule. We find that the new inhibitor, which has a symmetrical 8-mer half-cystine CTKSIPP′I′ motif repeated through a C₂ symmetry axis also shows a complete symmetry in its three-dimensional structure. Each of the two loops adopts the expected canonical conformation common to all BBIs as well as SFTI-1. We also find that the inhibitor displays a strong and unique structural identity, with a notable lack of minor conformational isomers that characterise most reactive site loop mimics examined to date as well as SFTI-1. This suggests that the presence of the additional cyclic loop acts to restrict conformational mobility and that the deliberate introduction of cyclic symmetry may offer a general route to locking the conformation of β-hairpin structures. Electronic supplementary material Electronic supplementary material is available for this article at and accessible for authorised users.     

Determination of the Amino Acid Sequence of a New Phospholipase A<Subscript>2</Subscript> (MIDCA1) Isolated from <Emphasis Type=Italic>Micrurus dumerilii carinicauda</Emphasis> Venom

A new Phospholipase A₂ (PLA₂) from Micrurus dumerilii carinicauda venom was isolated and its primary structure determined. This new PLA₂ showed a low enzymatic activity when compared with other PLA₂s and it is moderately basic with an isoelectric point of 8.0. Its amino acid sequence showed the presence of 120 amino acid residues and its sequence was: NLIQFLNMIQCTTPGREPLVAFANYGCYCGRGGSGTPVDELDRCCQVHDNCYDTAKKVFGCSPYFTMYSYDCSEGKLTCKDNNTKCKAAVCNCDRTAALCFAKAPYNDKNYKIDLTKRCQ. The structural model of MIDCA1, when compared with other strong neurotoxic PLA₂s, such as Naja naja, showed significant differences in the β-wing and neurotoxic sites, despite the high level of amino acid sequence similarity. These observations indicate a dissociation between the biological and catalytic activity of this new PLA₂, supporting the view that other regions of the protein are involved in the biological effects.     

Inhibitory role of TGIF in the As<Subscript>2</Subscript>O<Subscript>3</Subscript>-regulated p21<Superscript><Emphasis Type="Italic">WAF1/CIP1</Emphasis></Superscript> expression

Although arsenic is an infamous carcinogen, it has been effectively used to treat acute promyelocytic leukemia, and can induce cell cycle arrest or apoptosis in human solid tumors. Previously, we had demonstrated that opposing effects of ERK1/2 and JNK on p21 expression in response to arsenic trioxide (As₂O₃) are mediated through the Sp1 responsive elements of the p21 promoter in A431 cells. Presently, we demonstrate that Sp1, and c-Jun functionally cooperate to activate p21 promoter expression through Sp1 binding sites (−84/−64) by using DNA affinity binding, chromatin immunoprecipitation, and promoter assays. Surprisingly, As₂O₃-induced c-Jun(Ser63/73) phosphorylation can recruit TGIF/HDAC1 to the Sp1 binding sites and then suppress p21 promoter activation. We suggest that, after As₂O₃ treatment, the N-terminal domain of c-Jun phosphorylation by JNK recruits TGIF/HDAC1 to the Sp1 sites and then represses p21 expression. That is, TGIF is involved in As₂O₃-inhibited p21 expression, and then blocks the cell cycle arrest.     

Performance of a generalist grasshopper on a C<Subscript>3</Subscript> and a C<Subscript>4</Subscript> grass: compensation for the effects of elevated CO<Subscript>2</Subscript> on plant nutritional quality

The increasing CO₂ concentration in Earths atmosphere is expected to cause a greater decline in the nutritional quality of C₃ than C₄ plants. As a compensatory response, herbivorous insects may increase their feeding disproportionately on C₃ plants. These hypotheses were tested by growing the grasses Lolium multiflorum C₃) and Bouteloua curtipendula C₄) at ambient (370 ppm) and elevated (740 ppm) CO₂ levels in open top chambers in the field, and comparing the growth and digestive efficiencies of the generalist grasshopper Melanoplus sanguinipes on each of the four plant × CO₂ treatment combinations. As expected, the nutritional quality of the C₃ grass declined to a greater extent than did that of the C₄ grass at elevated CO₂; protein levels declined in the C₃ grass, while levels of carbohydrates (sugar, fructan and starch) increased. However, M. sanguinipes did not significantly increase its consumption rate to compensate for the lower nutritional quality of the C₃ grass grown under elevated CO₂. Instead, these grasshoppers appear to use post-ingestive mechanisms to maintain their growth rates on the C₃ grass under elevated CO₂. Consumption rates of the C₃ and C₄ grasses were also similar, demonstrating a lack of compensatory feeding on the C₄ grass. We also examined the relative efficiencies of nutrient utilization from a C₃ and C₄ grass by M. sanguinipes to test the basis for the C₄ plant avoidance hypothesis. Contrary to this hypothesis, neither protein nor sugar was digested with a lower efficiency from the C₄ grass than from the C₃ grass. A novel finding of this study is that fructan, a potentially large carbohydrate source in C₃ grasses, is utilized by grasshoppers. Based on the higher nutrient levels in the C₃ grass and the better growth performance of M. sanguinipes on this grass at both CO₂ levels, we conclude that C₃ grasses are likely to remain better host plants than C₄ grasses in future CO₂ conditions.     

Cloning a Truncated Fragment (stx2a<Subscript>1</Subscript>) of the Shiga-Like Toxin 2A<Subscript>1</Subscript> Subunit of EHEC O157:H7: Candidate Immunogen for a Subunit Vaccine

Shiga toxins consist of enzymatically active A and B subunit multimers. The A subunit of shiga-like toxins can be proteolytically cleaved into two parts, A₁ and A₂, with A₁ being responsible for toxic activity. Antibody neutralizing the A₁ subunit of shiga toxin may protect against infection of the enterohemorrhagic Escherichia coli (EHEC O157:H7). It was difficult to express the full-length A₁ subunit of shiga toxin 2 (stx2A₁) in a previous study. We have now analyzed the full-length of stx2A₁ using bioinformatics software. The data show that the carboxyl terminal (of ~15 amino-acid residues) has strong hydrophobicity and low antigenicity. We cloned and expressed a truncated fragment of stx2A₁ (15 amino-acid residues of the carboxyl terminal being removed), designated stx2a₁, which can evoke a humoral immune response. Anti-Stx2a₁ antibodies can neutralize the native shiga toxin 2 both in vivo and in vitro, which suggests that Stx2a₁ serves as a candidate immunogen for a subunit vaccine that can also be used as the antigen to screen phage anti-shiga toxin antibody libraries. L. Liu and H. Zeng contributed equally to this study.     

Fumonisin B<Subscript>1</Subscript>, a sphingoid toxin,is a potent inhibitor of the plasma membrane H<Superscript>+</Superscript>-ATPase

Fumonisin B₁ (FB₁) is an amphipathic toxin produced by the pathogenic fungus Fusarium verticillioides which causes stem, root and ear rot in maize (Zea mays L.). In this work, we studied the action of FB₁ on the plasma membrane H⁺-ATPase (EC 3.6.1.34) from germinating maize embryos, and on the fluidity and lipid peroxidation of these membranes. In maize embryos the toxin at 40 M inhibited root elongation by 50% and at 30 M decreased medium acidification by about 80%. Irrespective of the presence and absence of FB₁, the H⁺-ATPase in plasma membrane vesicles exhibited non-hyperbolic saturation kinetics by ATPH-Mg, with Hill number of 0.67. Initial velocity studies revealed that FB₁ is a total uncompetitive inhibitor of this enzyme with an inhibition constant value of 17.5±1 M. Thus FB₁ decreased V_max and increased the apparent affinity of the enzyme for ATP-Mg to the same extent. Although FB₁ increased the fluidity at the hydrophobic region of the membrane, no correlation was found with its effect on enzyme activity, since both effects showed different FB₁-concentration dependence. Peroxidation of membrane lipids was not affected by the toxin. Our results suggest that, under in vivo conditions, the plasma membrane H⁺-ATPase is a potentially important target of the toxin, as it is inhibited not only by FB₁ but also by its structural analogs, the sphingoid intermediates, which accumulate upon the inhibition of sphinganine N-acyltransferase by this toxin.     

Overexpression of the Inhibitor Protein IF1 in AS-30D Hepatoma Produces a Higher Association with Mitochondrial F1F0 ATP Synthase Compared to Normal Rat Liver: Functional and Cross-Linking Studies

According to functional studies, the higher IF(1) content reported in mitochondria of cancerous cells is supposed to induce a higher association with the F(1)F(0) complex than in normal cells and therefore a better inhibition of its ATPase activity. The first structural evidence supporting this prediction is here presented. Densitometric analyses of Western blotting experiments indicated a 2-fold increase in IF(1) content of AS-30D submitochondrial particles compared to normal rat liver controls. The ratio of IF(1)/F(1) alpha subunit increased similarly as judged by Westernblot analyses. This IF(1) overexpression correlated with a slower rate of IF(1) release (F(1)F(0)-ATPase activation) from the F(1)F(0) complex in AS-30D than in normal rat liver submitochondrial particles. The IF(1)-IF(1), gamma-IF(1), and alpha-IF(1) cross-linkages previously formed with dithiobis(succinimidylpropionate) in bovine F(1)F(0)I and IF(1) complexes were reproduced in the F(1)F(0)I-ATP synthase of hepatoma AS-30D cells. However, a much lower yield of IF(1) cross-linkages was found in normal rat liver particles which made them almost undetectable in SMP as well as in the immunoprecipitated F(1)F(0)I complex. Modeling in vivo IF(1) overexpression of cancerous cells by in vitro reconstitution of excess recombinant IF(1) with rat liver submitochondrial particles devoid of IF(1) reproduced the same IF(1) cross-linkages observed in AS-30D particles.     

Identification of the 7,8-didemethyl-8-hydroxy-5-deazariboflavin synthase required for coenzyme F<Subscript>420</Subscript> biosynthesis

The hydride carrier coenzyme F₄₂₀ contains the unusual chromophore 7,8-didemethyl-8-hydroxy-5-deazariboflavin (FO). Microbes that generate F₄₂₀ produce this FO moiety using a pyrimidine intermediate from riboflavin biosynthesis and the 4-hydroxyphenylpyruvate precursor of tyrosine. The fbiC gene, cloned from Mycobacterium smegmatis, encodes the bifunctional FO synthase. Expression of this protein in Escherichia coli caused the host cells to produce FO during growth, and activated cell-free extracts catalyze FO biosynthesis in vitro. FO synthase in the methanogenic euryarchaeon Methanocaldococcus jannaschii comprises two proteins encoded by cofG (MJ0446) and cofH (MJ1431). Both subunits were required for FO biosynthesis in vivo and in vitro. Cyanobacterial genomes encode homologs of both genes, which are used to produce the coenzyme for FO-dependent DNA photolyases. A molecular phylogeny of the paralogous cofG and cofH genes is consistent with the genes being vertically inherited within the euryarchaeal, cyanobacterial, and actinomycetal lineages. Ancestors of the cyanobacteria and actinomycetes must have acquired the two genes, which subsequently fused in actinomycetes. Both CofG and CofH have putative radical S-adenosylmethionine binding motifs, and pre-incubation with S-adenosylmethionine, Fe²⁺, sulfide, and dithionite stimulates FO production. Therefore a radical reaction mechanism is proposed for the biosynthesis of FO.Abbreviations AdoMet (SAM) S-adenosyl-l-methionine - Compound 6 5-Amino-6-ribitylamino-2,4(1H,3H)-pyrimidinedione - FO 7,8-Didemethyl-8-hydroxy-5-deazariboflavin - HPP 4-Hydroxyphenylpyruvate     

Interisland Mutation of a Novel Phospholipase A<Subscript>2</Subscript> from <Emphasis Type="Italic">Trimeresurus flavoviridis</Emphasis> Venom and Evolution of Crotalinae Group II Phospholipases A<Subscript>2</Subscript>

Trimeresurus flavoviridis (Crotalinae) snakes inhabit the southwestern islands of Japan: Amami-Oshima, Tokunoshima, and Okinawa. Affinity and conventional chromatographies of Amami-Oshima T. flavoviridis venom led to isolation of a novel phospholipase A₂ (PLA₂). This protein was highly homologous (91%) in sequence to trimucrotoxin, a neurotoxic PLA₂, which had been isolated from T. mucrosquamatus (Taiwan) venom, and exhibited weak neurotoxicity. This protein was named PLA-N. Its LD₅₀ for mice was 1.34 µg/g, which is comparable to that of trimucrotoxin. The cDNA encoding PLA-N was isolated from both the Amami-Oshima and the Tokunoshima T. flavoviridis venom-gland cDNA libraries. Screening of the Okinawa T. flavoviridis venom-gland cDNA library with PLA-N cDNA led to isolation of the cDNA encoding one amino acid-substituted PLA-N homologue, named PLA-N(O), suggesting that interisland mutation occurred and that Okinawa island was separated from a former island prior to dissociation of Amami-Oshima and Tokunoshima islands. Construction of a phylogenetic tree of Crotalinae venom group II PLA₂s based on the amino acid sequences revealed that neurotoxic PLA₂s including PLA-N and PLA-N(O) form an independent cluster which is distant from other PLA₂ groups such as PLA₂ type, basic [Asp⁴⁹]PLA₂ type, and [Lys⁴⁹]PLA₂ type. Comparison of the nucleotide sequence of PLA-N cDNA with those of the cDNAs encoding other T. flavoviridis venom PLA₂s showed that they have evolved in an accelerated manner. However, when comparison was made within the cDNAs encoding Crotalinae venom neurotoxic PLA₂s, their evolutionary rates appear to be reduced to a level between accelerated evolution and neutral evolution. It is likely that ancestral genes of neurotoxic PLA₂s evolved in an accelerated manner until they had acquired neurotoxic function and since then they have evolved with less frequent mutation, possibly for functional conservation. The nucleotide sequences reported in this paper are available from the GenBank/EMBL/DDBJ databases under accession numbers AB102728 and AB102729.     

New Phospholipase A<Subscript>1</Subscript>-producing Bacteria from a Marine Fish

Phospholipase A1 is a hydrolytic enzyme that catalyzes the removal of the acyl group from position 1 of glycerophospholipids to form 2-acyl lysophospholipids. Lysophospholipids are used in foods, cosmetics, and pharmaceuticals as surfactants. Novel forms of phospholipase A1 that function at low temperatures are desirable for use in lipophilic systems in food processing. However, there is currently little variety in the available sources of phospholipase A1. Given this situation, we screened the intestinal contents of marine animals for phospholipase A1-producing bacteria. Colonies that formed a halo on K28CP screening medium and that grew in K28 medium were cultured in liquid K28 medium, and the supernatant was retrieved for analysis. Phosphatidylcholine was added to the culture supernatant, and the product of the reaction was analyzed by using TLC. For culture supernatants that were able to generate lysophosphatidylcholine, synthetic phosphatidylcholines were added, and the site of the reaction was determined by analyzing the fatty acid compositions of the lysophosphatidylcholines generated by GLC. A bacterial isolate from a flatfish, which we named HFKI0020, was found to have phospholipase A1 activity at low temperatures. We determined that the isolate HFKI0020 is closely related to Pseudomonas by using 16S rDNA sequence analysis and by characterizing the isolate with respect to its physiologic and biochemical properties. From the intestinal contents of a marine fish, we successfully isolated a bacterium that secretes phospholipase A1 that is active at low temperatures.     

Identification of agronomically favorable quantitative trait loci alleles from a dent corn inbred Dan232 using advanced backcross QTL analysis and comparison with the F<Subscript>2:3</Subscript> population in popcorn

Specific traits are an important consideration in plant breeding. In popcorn, inferior agronomic traits could be improved using dent or flint corn backcrossed with popcorn. In this study, we used advanced backcross quantitative trait locus (AB-QTL) analysis to identify trait-improving QTL alleles from a dent maize inbred Dan232, and compared the detection of QTL in the BC₂S₁ population with QTL results using F_2:3 families of the same population. Two hundred and twenty BC₂S₁ families developed from a cross between Dan232 and an elite popcorn inbred N04 were evaluated for nine plant traits in replicated field trials under two environments. Using composite interval mapping (CIM), a total of 28 significant QTL were detected, and of these, 23 (82.14%) had favorable alleles contributed by the dent corn parent Dan232. Nine QTL (32.14%) detected in the BC₂S₁ population were also located in or near the same chromosome intervals in the F_2:3 population. All of the favorable QTL alleles from Dan232 could be used in marker-assisted selection (MAS) to improve the respective plant traits in popcorn breeding. In addition, their near isogenic lines (QTL-NILs) could be obtained through selfing or another 1–2 backcross with N04. Also, N04 improved for the studied plant traits could be developed from the BC₂S₁ families used in this study. This study demonstrated that the AB-QTL method can be applied to identify favorable QTL from dent corn inbred in popcorn breeding and, once identified, the alleles could be used in marker-assisted selection to improve the respective plant traits.     

Cellular expression of C<Subscript>3</Subscript> and C<Subscript>4</Subscript> photosynthetic enzymes in the amphibious sedge<Emphasis Type="Italic"> Eleocharis retroflexa</Emphasis> ssp.<Emphasis Type="Italic"> chaetaria</Emphasis>

The amphibious leafless sedge Eleocharis retroflexa ssp. chaetaria expresses C₄-like biochemical characteristics in both the terrestrial and submerged forms. Culms of the terrestrial form have Kranz anatomy, whereas those of the submerged form have Kranz-like anatomy combined with anatomical features of aquatic plant leaves. We examined the immunolocalization of C₃ and C₄ enzymes in culms of the two forms. In both forms, phosphoenolpyruvate carboxylase; pyruvate, Pi dikinase; and NAD-malic enzyme were compartmentalized between the mesophyll (M) and Kranz cells, but their levels were somewhat reduced in the submerged form. In the terrestrial form, ribulose-1,5-bisphosphate carboxylase/oxygenase (rubisco) occurred mainly in the Kranz cells, and weakly in the M chloroplasts. In the submerged form, the rubisco occurred at higher levels in the M cells than in the terrestrial form. In both forms, the C₄ pattern of enzyme expression was clearer in the M cells adjacent to Kranz cells than in distant M cells. During the transition from terrestrial to submerged conditions, the enzyme expression pattern changed in submerged mature culms that had been formed in air before submergence, and matched that in culms newly developed underwater. It seems that effects of both environmental and developmental factors overlap in the C₄ pattern expression in this plant.     

Interaction of ADP and ATP with noncatalytic sites of isolated and membrane-bound chloroplast coupling factor CF<Subscript>1</Subscript>

This study of ATP and ADP binding to noncatalytic sites of membrane-bound CF1 (ATP synthase) revealed two noncatalytic sites with different specificities and affinities for nucleotides. One of these is characterized by a high affinity and specificity to ADP (Kd=2.6+/-0.3 microM). However, a certain increase in ADP apparent dissociation constant at high ATP/ADP ratio in the medium allows a possibility that ATP binds to this site as well. The other site displays high specificity to ATP. When the ADP-binding site is vacant, it shows a comparatively low affinity for ATP, which greatly increases with increasing ADP concentration accompanied by filling of the ADP-binding site. The reported specificities of these two sites are independent of thylakoid membrane energization, since both in the dark and in the light the ratios of ATP/ADP tightly bound to the noncatalytic sites were very close. The difference in noncatalytic site affinity for ATP and ADP is shown to depend on the amount of delta subunit in a particular sample. Thylakoid membrane ATP synthase, with stoichiometric content of delta-subunit (one delta-subunit per CF1 molecule), showed the maximal difference in ADP and ATP affinities for the noncatalytic sites. For CF1, with substoichiometric delta subunit values, this difference was less, and after delta subunit removal it decreased still more.