A fraction of plasma transthyretin (TTR) circulates in HDL through binding to apolipoprotein A-I (apoA-I). Moreover, TTR is able to cleave the C terminus of lipid-free apoA-I. In this study, we addressed the relevance of apoA-I cleavage by TTR in lipoprotein metabolism and in the formation of apoA-I amyloid fibrils. We determined that TTR may also cleave lipidated apoA-I, with cleavage being more effective in the lipid-poor preb-HDL subpopulation. Upon TTR cleavage, discoidal HDL particles displayed a reduced capacity to promote cholesterol efflux from cholesterol-loaded THP-1 macrophages. In similar assays, TTR-containing HDL from mice expressing human TTR in a TTR knockout background had a decreased ability to perform reverse cholesterol transport compared with similar particles from TTR knockout mice, reinforcing the notion that cleavage by TTR reduces the ability of apoA-I to promote cholesterol efflux. As amyloid deposits composed of N-terminal apoA-I fragments are common in the atherosclerotic intima, we assessed the impact of TTR cleavage on apoA-I aggregation and fibrillar growth. We determined that TTR-cleaved apoA-I has a high propensity to form aggregated particles and that it formed fibrils faster than full-length apoA-I, as assessed by electron microscopy. Our results show that apoA-I cleavage by TTR may affect HDL biology and the development of atherosclerosis by reducing cholesterol efflux and increasing the apoA-I amyloidogenic potential.-Liz, M. A., C. M. Gomes, M. J. Saraiva, and M. M. Sousa. ApoA-I cleaved by transthyretin has reduced ability to promote cholesterol efflux and increased amyloidogenicity. J. Lipid Res. 2007Res. . 48: 2385Res. -2395

A seven-iron ferredoxin was isolated from aerobically grown cells of the hyperthermoacidophilic archaleon Desulfurolobus ambivalens (DSM 3772). The protein is monomeric, with an apparent molecular mass of 15 kDa and contains 7 iron atoms/molecule. The N-terminal sequence shows a large similarity (70% identity) with that of the ferredoxin isolated from the archaeon Sulfolobus acidocaldarius. The EPR isharacteristics in both the native (oxidized) and dithionite-reduced states of this protein allowed an unequivocal identification of a [3Fe-4SI1+" center, with a reduction potential of -270-+ 20 mV, at pH 7.5. The protein also contains a [4Fe-4S]*'"+ center with a very low reduction potential (Eo = -540 mV, pH 7.1D), which yields a rhombic EPR spectrum upon reduction with sodium dithionite at high pH. The reduction potentials of both centers are slightly pH dependent between pH 6 and 9. The [3Fe-4S] ferredoxin center is able to accept electrons from pyruvate oxidase and NADH oxidase isolated from D. ambivalens. This ferredoxin is present in large amounts (at least 130 mgkg wet cells), which allowed the unequivocal observation of oxidized [3Fe-4S] clusters in intact D. ambivalens cells.Keywords. Iron-sulfur centers ; Archaea; EPR; thermophilic.Proteins containing iron-sulfur clusters with a wide range of stoichiometries are ubiquitously found to be involved in fundamental biological processes like nitrogen fixation, CO, fixation, hydrogen metabolism, the citric-acid cycle, detoxification and membrane-bound (energy-transduction processes [l -31. The simplest iron-sulfur proteins, the ferredoxins, are involved in these pathways as one-electron carriers. Due mainly to the extensive work on model compounds, it has been shown that under anaerobic conditions iron-sulfur structures may self-assemble in reaction mixtures containing iron, sulfide and thiols [4]. Thus, since iron-sulfur proteins are present in the three major urkingdoms, they were proposed as the most ancient electron-transfer agents to have appeared during biological evolution [5]. Therefore, in evolutionary terms it is particularly interesting to screen ffor iron-sulfur proteins in archaea, the most ancient living organisms [6].

The aa 3 quinol oxidase has been purified from the thermoacidophilic archaea Acidianus ambivalens as a three-redox-centers enzyme. The functional properties of this oxidase both as purified and in its most integral form (i.e. in native membranes and in intact cells) were investigated by stopped-flow spectrophotometry. The results suggest that the enzyme interacts in vivo with a redox-active molecule, which favours the electron entry via heme a and provides the fourth electron demanded for catalysis.We observe that the purified enzyme has two hemes with apparent redox potentials 215Ϯ 20 mV and 415 Ϯ20 mV at pH 5.4, showing redox-Bohr effect, and a heme a 3 -Cu B center with an affinity for carbon monoxide (K a ϭ 5.7ϫ104 M Ϫ1 at 35°C) much lower than that reported for the mammalian enzyme (K a ϭ 4ϫ10 6 M Ϫ1 at 20°C). The reduction by dithionite is fast and monophasic when the quinol oxidase is in the native membranes, whereas it is slow and biphasic in the purified enzyme (with heme a 3 being reduced faster than heme a). The oxygen reaction of the reduced purified enzyme is fast (few milliseconds), but yields an intermediate (likely ferryl) clearly different from the fully oxidized enzyme. In contrast, the same reaction performed in intact cells leads to the fully oxidized enzyme.We postulate that caldariella quinol, the physiological electron donor, is in vivo tightly bound to the enzyme, providing the fourth redox active center lacking in the purified enzyme.

To fulfil the title hypothesis, such a protein would have to harbour a binuclear transition metal site that would allow the complete reduction of dioxygen to water, thus playing an important role in early oxygen defence mechanisms in primordial anaerobes. The hypothesis here raised is based on data concerning the oxygen reductase activity of the four‐helix diiron protein rubrerythrin (see schematic picture), and on sequence‐ and structure‐based phylogenetic relations between this and other diiron proteins. Interestingly, an evolutionary relationship between such an early system and the alternative oxidases present in extant eukaryotes can be depicted.