Determination of the native form of FadD, the Escherichia coli fatty acyl-CoA synthetase, and characterization of limited proteolysis by outer membrane protease OmpT.

Several studies have described FadD, the Escherichia coli fatty acyl-CoA synthetase [also known as fatty acid:CoA ligase (AMP-forming); EC 6.2.1.3], as a 42-50 kDa enzyme. Based on sequencing and expression data from the fadD gene, other reports have suggested that FadD is a 62 kDa protein and represents the sole fatty acyl-CoA synthetase in E. coli. We report that the 62 kDa FadD enzyme is a substrate for the outer membrane protease OmpT in vitro, producing a 43 kDa C-terminal fragment and a 19 kDa N-terminal fragment. Immunoblotting with a FadD antibody revealed that only the 62 kDa form of the enzyme is present in vivo, but we utilized the proteolytic sensitivity of FadD to investigate its structure. Photoaffinity labelling experiments revealed that both intact FadD and the 43 kDa fragment bound a long-chain fatty acid. Intact and cleaved FadD were also purified to determine the effect of cleavage on function. When using oleate as a substrate, cleaved FadD displayed 2-fold higher K(m) and V(max) values compared with intact FadD, but the catalytic efficiencies (k(cat)/K(m)) of the two forms were similar. This indicated that cleavage did not adversely affect enzyme activity. Proteolysis of FadD by OmpT was altered by the presence of oleate or ATP, both of which are ligands for the fatty acyl-CoA synthetase. This suggested that FadD undergoes ligand-induced conformational changes and implies that the region surrounding the cleavage site is mobile, a common characteristic of linker domains.

Identification of caveolin-1 as a fatty acid binding protein.

In an attempt to identify high affinity, fatty acid binding proteins present in 3T3-L1 adipocytes plasma membranes, we labeled proteins in purified plasma membranes with the photoreactive fatty acid analogue, 11-m-diazirinophenoxy[11-3H]undecanoate. A single membrane protein of 22 kDa was covalently labeled after photolysis. This protein fractionated with caveolin-1 containing caveolae and was immunoprecipitated by an anti-caveolin-1 monoclonal antibody. Furthermore, 2D-PAGE analysis revealed that both the alpha and beta isoforms of caveolin-1 could be labeled by the photoreactive fatty acid upon photolysis, indicating that both bind fatty acids. The saturable binding of the photoreactive fatty acid suggests caveolin-1 has a lipid binding site that may either operate during intracellular lipid traffic or regulate caveolin-1 function.

The effect of intracellular pH on long-chain fatty acid uptake in 3T3-L1 adipocytes: evidence that uptake involves the passive diffusion of protonated long-chain fatty acids across the plasma membrane.

To understand the mechanism of long-chain fatty acid permeation of the plasma membrane in mammalian cells, the effects of changes in the cytoplasmic pH on the internalization of physiologically relevant, submicromolar concentrations of uncomplexed long-chain fatty acids were investigated in 3T3-L1 adipocytes. The acidification of the cytoplasm upon NH4Cl prepulsing of intact cells was accompanied by a rapid reduction of cellular long-chain fatty acid uptake (measured as the total accumulation of [9,10-3H]oleate). This was followed by a slow recovery to normal levels of uptake as the cytoplasmic pH recovered. Conventional filtration assays do not distinguish between fatty acid movement across the plasma membrane and intracellular steps, such as binding to cytoplasmic fatty acid-binding proteins or metabolism. While the in vitro binding of a photoreactive fatty acid, 11-m-diazirinophenoxy[11-3H]undecanoate, to a cytoplasmic fatty acid-binding protein was insensitive to changes in pH from pH 7.5 to 5.5, the in vitro conversion of oleate into oleoyl-CoA by cellular acyl-CoA synthetase decreased dramatically. Therefore, the labelling of the 15 kDa cytoplasmic fatty acid-binding protein in intact cells by the photoreactive fatty acid was used as a more direct measure of the permeation of the probe across the plasma membrane. Acidification of the cytoplasm resulted in an immediate reduction in the labelling of this protein in intact adipocytes. Its photolabelling recovered, however, upon the recovery of the cytoplasmic pH to normal levels. This was due to effects of the cytoplasmic pH on the permeation of the photoreactive fatty acid across the plasma membrane rather than its binding to the 15 kDa protein or metabolism in vivo. This is the first demonstration that the movement of physiologically relevant, submicromolar concentrations of uncomplexed long-chain fatty acids across the plasma membrane of intact cells is coupled to the cytoplasmic pH and suggests that it occurs by the diffusion of the protonated long-chain fatty acid through the lipid bilayer.

A direct role for serum albumin in the cellular uptake of long-chain fatty acids.

The interaction of long-chain fatty acids with cells is important for their uptake and metabolism, as well as their involvement in signalling processes. The majority of long-chain fatty acids circulating in plasma exist as complexes with serum albumin. Thus an understanding of the involvement of serum albumin in these processes is vitally important. The effect of serum albumin on the uptake of long-chain fatty acids was studied in 3T3-L1 adipocytes. Serum albumin had a stimulatory effect on oleate uptake at all ratios of oleate: serum albumin tested. Furthermore, the rate of oleate uptake was saturable with increasing concentrations of serum albumin when the oleate: serum albumin ratio, and therefore the concentration of uncomplexed oleate, remained constant. This was not due to uptake being limited by dissociation of oleate from serum albumin, because oleate did not appear to be limiting. Furthermore, at very high ratios of oleate: serum albumin, when the concentration of uncomplexed oleate was predicted to be large relative to the amount of oleate taken up by cells, the rate of oleate uptake was still dependent on the albumin concentration. Serum albumin, covalently labelled with the photoreactive fatty acid 11-m-diazirinophenoxy[11-3H]undecanoate, bound to cells in a manner exhibiting both saturable (Kd 66.7 microM) and non-saturable processes. These results indicate that the stimulatory effect of serum albumin on the rate of oleate uptake is due to a direct interaction of serum albumin with the cells and point to an involvement of albumin binding sites in the cell surface in the cellular uptake of long-chain fatty acids.

Membrane permeation and intracellular trafficking of long chain fatty acids: insights from Escherichia coli and 3T3-L1 adipocytes.

Long chain fatty acids are important substrates for energy production and lipid synthesis in prokaryotes and eukaryotes. Their cellular uptake represents an important first step leading to metabolism. This step is induced in Escherichia coli by growth in medium containing long chain fatty acids and in murine 3T3-L1 cells during differentiation to adipocytes. Consequently, these have been used extensively as model systems to study the cellular uptake of long chain fatty acids. Here, we present an overview of our current understanding of long chain fatty acid uptake in these cells. It consists of several distinct steps, mediated by a combination of biochemical and physico-chemical processes, and is driven by conversion of long chain fatty acids to acyl-CoA by acyl-CoA synthetase. An understanding of long chain fatty acid uptake may provide valuable insights into the roles of fatty acids in the regulation of cell signalling cascades, in the regulation of a variety of metabolic and transport processes, and in a variety of mammalian pathogenic conditions such as obesity and diabetes.

Specific labeling of Candida tropicalis peroxisomal proteins with photoreactive fatty-acid derivatives.

The labeling of Candida tropicalis peroxisomal proteins with photoreactive fatty-acid derivatives was investigated. Proteins having molecular masses of 70 kDa, 48 kDa and 15 kDa were labeled with 11-m-diazirinophenoxy-[11-3H]undecanoate while 11-m-diazirinophenoxy-[11-3H]undecanoyl-CoA labeled proteins of 70 kDa and 55 kDa. The 70 kDa protein labeled with both photoreactive probes was resolved into two bands by electrophoresis on a gradient polyacrylamide gel; immunoprecipitation with anti-fatty acyl-CoA oxidase showed that these proteins are fatty-acyl-CoA oxidases. In purified peroxisomal membranes, two proteins of 36 kDa and 25 kDa were labeled with the photoreactive fatty-acid probe, whereas very little labeling of the above proteins or other proteins was observed with the fatty-acyl-CoA probe. The photoaffinity labeling method described is, thus, clearly capable of identifying and distinguishing between proteins having an affinity for fatty acid or fatty-acyl-CoA. The labeling also identified a fatty-acid-binding site on the 16 kDa peroxisomal matrix protein as well as on two peroxisomal acyl-CoA oxidases. This approach thus provides a general means for the identification of fatty-acid metabolizing enzymes, as well as for the identification of fatty-acid-binding sites on known enzymes.

Identification of high affinity membrane-bound fatty acid-binding proteins using a photoreactive fatty acid.

A photoaffinity labeling method was developed to identify and characterize high affinity fatty acid-binding proteins in membranes. The specific labeling of these sites requires the use of low concentrations (nanomolar) of the photoreactive fatty acid 11-m-diazirinophenoxy-[11-3H]undecanoate. It was delivered as a bovine serum albumin (BSA) complex which serves as a reservoir for fatty acid and thus allows precise control of unbound fatty acid concentrations. The fadL protein of E. coli, which is required for fatty acid permeation of its outer membrane, was labeled by the photoreactive fatty acid neither specifically nor saturably when the probe was added in the absence of BSA; however when a nanomolar concentration of the uncomplexed probe was maintained in the presence of BSA, the labeling of the fadL protein was highly specific and saturable. This photoaffinity labeling method was also used to characterize a 22 kDa, high affinity fatty acid-binding protein which we have recently identified in the plasma membrane of 3T3-L1 adipocytes. This protein bound the probe with a Kd of 216 nM. The approach described is easily capable of identifying membrane-bound fatty acid-binding proteins and can distinguish between those of high and low affinities for fatty acids. It represents a general method for the identification and characterization of fatty acid-binding proteins.

Fatty acid uptake in Escherichia coli: regulation by recruitment of fatty acyl-CoA synthetase to the plasma membrane.

Fatty acid uptake in Escherichia coli has been shown to be inhibited by starvation and to be reversed by a short preincubation of the starved cells with D- or L-lactate, succinate, and acetate; these effects on oleate uptake were due to regulation of the rate-limiting step which involves fatty acyl-CoA synthetase. Investigation into the mechanism of regulation of fatty acyl-CoA synthetase showed that D-lactate did not affect the activity of the enzyme directly. Fatty acyl-CoA synthetase was found to be activated by about 20-fold by Triton X-100 and by another 4-fold by the addition of bacterial membranes. D-Lactate treatment was shown to result in coisolation of fatty acyl-CoA synthetase with the plasma membrane; these results are consistent with the interpretation that recruitment of the enzyme to the plasma membrane by D-lactate results in its activation and consequently in the increased level of fatty acid uptake.

Photoaffinity labeling of fatty acid-binding proteins involved in long chain fatty acid transport in Escherichia coli.

The photoreactive fatty acid 11-m-diazirinophenoxy-[11-3H]undecanoate was shown to be taken up specifically by the fatty acid transport system expressed in Escherichia coli grown on oleate. This photoreactive fatty acid analogue was therefore used to identify proteins involved in fatty acid uptake in E. coli. The fadL protein was labeled by the probe, confirmed to be exclusively in the outer membrane and to exhibit the heat modifiable behavior typical of outer membrane proteins. The apparent pI of the incompletely denatured form of the protein having the mobility of a 33-kDa protein was 4.6 while that of the fully denatured form was consistent with the calculated value of 5.2. The denaturation was reversible depending upon the protein to detergent ratios. The photoreactive fatty acid partitions into the outer membrane, resulting in extensive photolabeling of the lipid; a high affinity fatty acid-binding site is not apparent in total membranes labeled using free fatty acids due to this large binding capacity of the outer membrane. However, when the free fatty acid concentration was controlled by supplying it as a bovine serum albumin complex, the fadL protein exhibited saturable high affinity fatty acid binding, having an apparent Kd for the probe of 63 nM. The methods described very readily identify fatty acid-binding proteins: the fact that even when the sensitivity was increased 500-fold, no evidence was found for the presence of a fatty acid-binding protein in the inner membrane is consistent with the proposal that fatty acid permeation across the plasma membrane is not protein mediated but occurs by a simple diffusive mechanism.

Fatty acid uptake in Candida tropicalis: induction of a saturable process.

The rates of oleate uptake by Candida tropicalis cells grown on a high oleate concentration (3.5 mM oleate in the presence of 0.50% Brij 58) were higher than those observed in cells grown on glucose; however, oleate uptake was not saturable with substrate concentration. Cells grown at a low oleate concentration (1.0 mM oleate and 2.5% Brij 58) grew to a lower density and at a slightly slower rate; these cells were found to take up oleate at a rate 43-fold higher than cells grown on high oleate concentration. Furthermore, oleate uptake by the cells grown in low oleate medium was a saturable process with Kt and Vmax values of 56 microM and 15 nmol/(min.mg cell protein), respectively. The growth of C. tropicalis under low fatty acid concentration thus clearly results in the induction of a saturable process for its uptake. The total level of acyl-CoA synthetase activity in cells grown on the low oleate concentrations was only twofold higher than in high oleate or glucose grown cells; the level of this enzyme thus does not account for the saturable process and suggests that either the enzyme is regulated in vivo or else a hitherto unidentified enzyme is induced by growth in low concentrations of oleate.

Photoaffinity labeling and fatty acid permeation in 3T3-L1 adipocytes.

Long chain fatty acid uptake was investigated in 3T3-L1 cells. Differentiation of these cells from fibroblasts to adipocytes was accompanied by an 8.5-fold increase in the rate of oleate uptake. This was saturable in adipocytes with apparent Kt and Vmax values of 78 nM and 16 nmol/min/mg cell protein, respectively. A number of proteins in various subcellular fractions of differentiated cells were labeled with the photoreactive fatty acid 11-m-diazirinophenoxy[11-3H]undecanoate. A 15-kDa cytoplasmic protein was induced upon differentiation to adipocytes. This protein was labeled with the photoreactive fatty acid in cytoplasm isolated from differentiated adipocytes, but not in cytoplasm from undifferentiated, fibroblastic cells. Furthermore, a high affinity fatty acid binding protein of 22 kDa was identified in plasma membranes of undifferentiated cells, and its level of labeling increased 2-fold upon differentiation. These results indicate the usefulness of the photoreactive fatty acid in identifying cellular fatty acid binding proteins, and its potential to elucidate the spatial and temporal distribution of fatty acids in intact cells.

Water solubilization of membrane proteins. Extensive derivatization with a novel polar derivatizing reagent.

A reagent and a derivatization procedure have been developed which result in trimesylation of all -OH and -NH2 groups on proteins: serine, threonine, tyrosine, and lysine side chains are all completely derivatized. The parameters affecting the kinetics of trimesylation of serine, threonine, and tyrosine peptides were studied using dinitrophenylated peptides as model systems. Conditions for solubilization of proteins in anhydrous organic solvents for the derivatization are described; removal of blocking groups results in the polar, highly water-soluble protein derivative which behaves as a monomer during gel permeation chromatography in simple aqueous buffers even in the absence of detergents. Synthesis of the tritiated reagent is described, and this reagent was used to monitor the protein derivatization.

Identification of the phosphorylation sites in early region 1A proteins of adenovirus type 5 by amino acid sequencing of peptide fragments.

We have mapped the positions of three of the phosphorylation sites on the 289 and 243 residue (289R and 243R) early region 1A (E1A) proteins of human adenovirus type 5 (Ad5). These proteins, which play roles in both transcriptional control and oncogenic transformation, have identical sequences except for the presence in 289R of 46 additional internal amino acids. Phosphorylation was detected exclusively at serine residues. E1A proteins purified from [35S]methionine- or [32P]orthophosphate-labeled Ad5-infected cells were digested with trypsin, and two phosphopeptides were isolated by reverse-phase chromatography and subjected to automated Edman degradation. The major species was shown to contain a single phosphorylation site at Ser-219. The second phosphopeptide was shown to contain at least one phosphorylation site at Ser-231. A third phosphorylated tryptic peptide could not be eluted from the column but was isolated using an E1A-specific rat monoclonal antibody. Following subcleavage by Staphylococcus aureus V-8 protease, this peptide was shown to contain at least one phosphorylation site at Ser-89. The present data indicate that both the 289R and 243R E1A proteins are phosphorylated at the same sites, at least one in the amino terminal half of the molecule, and at least two toward the carboxyl terminus.

Amino acid analysis by dinitrophenylation and reverse-phase high-pressure liquid chromatography.

The determination of amino acids has been achieved by reverse-phase high-pressure liquid chromatography of their dinitrophenyl derivatives. The methods developed permit the quantitation of all amino acids commonly encountered in a protein hydrolysate and the effect of various parameters on this separation was systematically evaluated. The procedure eliminates the need for specialized postcolumn equipment as employed in conventional amino acid analysis and can be obtained by a simple gradient high-pressure chromatograph. The sensitivity obtained is comparable to that available by methods in common usage, being able to determine amino acids quantitatively in the low picomole range.

Complete nucleotide and derived amino acid sequence of cDNA encoding the mitochondrial uncoupling protein of rat brown adipose tissue: lack of a mitochondrial targeting presequence.

A cDNA clone spanning the entire amino acid sequence of the nuclear-encoded uncoupling protein of rat brown adipose tissue mitochondria has been isolated and sequenced. With the exception of the N-terminal methionine the deduced N-terminus of the newly synthesized uncoupling protein is identical to the N-terminal 30 amino acids of the native uncoupling protein as determined by protein sequencing. This proves that the protein contains no N-terminal mitochondrial targeting prepiece and that a targeting region must reside within the amino acid sequence of the mature protein.

Biosynthetic utilization of photoreactive fatty acids by rat liver microsomes.

The photoreactive omega-diazirinophenoxy derivatives of nonanoate, undecanoate, tridecanoate, and pentadecanoate were shown to be activated by rat liver microsomes to the corresponding acyl-CoA derivatives. The Km and Vmax for these fatty acid analogues were determined: the values obtained indicate that the addition of a photoreactive group to an alkyl chain has an effect similar to that of elongation of the chain by about seven carbons. Incubation of microsomes in the presence of lysophospholipids resulted in the incorporation of the photoreactive fatty acids into the corresponding phospholipids. The ability of mammalian systems to utilize these photoreactive fatty acids for phospholipid synthesis establishes their suitability as photoaffinity analogues of fatty acids.

Localization of membrane proteins by the use of a photoreactive fatty acid incorporated in vivo into vesicular stomatitis virus.

Vesicular stomatitis virus grown in the presence of omega-[9-3H]diazirinophenoxy nonanoate resulted in the biosynthetic incorporation of this photoreactive fatty acid into viral phospholipids as well as into the membrane anchoring domain of the viral G glycoprotein. Photolysis of the isolated virus at 360 nm resulted in extensive labeling of the G protein but none of the other proteins by the viral phospholipids. In addition, a new product was obtained and was identified as a G-G dimer by its molecular weight and its reaction with anti-G antibody. These results demonstrate the use of photoreactive fatty acid to identify integral membrane proteins and to make photoaffinity probes of fatty acylated membrane proteins.

Synthesis and biosynthetic utilization of radioactive photoreactive fatty acids.

A synthesis of a homologous series of highly radioactive photoreactive fatty acids has been developed. Oxidative cleavage of the cis double bonds of unsaturated fatty acids by NaIO4/OsO4 produced the omega-oxo fatty acids; subsequent reduction by [3H]NaBH4 afforded a series of highly radioactive [omega-3H]hydroxy fatty acids to which a variety of photoreactive probes can be readily attached. In this study, the m-diazirinophenoxy group was coupled to the omega-hydroxy groups of the fatty acids, yielding the omega-(m-diazirinophenoxy) derivatives of nonanoic, undecanoic, tridecanoic, and pentadecanoic acid. These photoreactive derivatives were incorporated by L-cells into their major phospholipids; in addition, a number of proteins were shown to be acylated by these fatty acids. The photoreactive group was shown to remain intact as indicated by the cross-linking of the resulting photoreactive phospholipids to other phospholipids as well as to a number of L-cell proteins upon photolysis. This procedure thus identifies integral membrane proteins and directly produces photoaffinity derivatives of those proteins normally acylated by fatty acids.

Interaction of cholesterol with photoactivable phospholipids in sonicated vesicles.

Photoactivable phospholipids containing either alpha-diazo-beta-trifluoropropionyloxy or m-diazirinophenoxyl groups in the omega-positions of sn-2 fatty acyl chains were synthesized and incorporated into sonicated vesicles containing 33 mol% of cholesterol. Photolysis of the vesicles at 350 nm produced covalent cross-links between the synthetic phospholipids and cholesterol. The cross-linked products obtained using [14C]cholesterol were characterized by their chromatographic behavior, cleavage on phospholipase A2 treatment, base-catalyzed transesterification and mass spectral measurements. The cross-linking was shown not to involve the 3-beta-hydroxyl group of cholesterol, and it was concluded that the reactive carbene intermediates formed from the photolabels inserted into the hydrocarbon skeleton of cholesterol in the bilayer. The extent of cross-linking obtained was comparable to that observed previously using phospholipids alone, indicating that no lateral phase separation occurred. The present approach is promising for further precise studies of the molecular interactions between cholesterol and phospholipids in biological membranes.

Amino acid sequence of the membranous segment of rabbit liver cytochrome b5. Methodology for separation of hydrophobic peptides.

The primary structure of the membranous segment of rabbit liver cytochrome b5 has been determined to be as follows: (Formula: see text), New methods were developed for the separation of peptide fragments. These methods, which are particularly useful for work with hydrophobic peptides, involved gel permeation chromatography and reverse phase high performance liquid chromatography (HPLC) using ethanol/formic acid mixtures. Sequences of the purified fragments were determined by automated sequence analysis.