Fatty Acid Auxotroph Research Articles

Phase transitions of phospholipid bilayers from an unsaturated fatty acid auxotroph of Escherichia coli

Total phospholipids were extracted from cells of temperature sensitive unsaturated fatty acid auxotrophs of Escherichia coli (K-12 UFA ts) grown at 28°C (PL28), and at 42°C in the presence of 2% KCl as an osmotic stabilizer (PL42 (KCl)). From the analysis of fatty acids, it was shown that the content of unsaturated fatty acids of PL42 (KCl) is only 9% of the total fatty acids, while that of PL28 is 54%. The thermal phase transitions of the bilayers prepared from the phospholipid fractions were studied by proton magnetic resonance. The line widths of the methylene signals and the sums of the methylene and methyl signal intensities were plotted against reciprocal values of absolute temperature 1/ T or temperature itself. From the plots phase transitions were detected at about 19°C for PL28 and at 43°C for PL42 (KCl). In spite of its complex composition of fatty acids a highly cooperative transition was observed in the case of PL42 (KCl). It was also suggested that the phospholipids bilayers in the biomembranes of this strain at the growth temperature (42°C) are in the state where the gel and liquid crystalline phases coexist.

Phenethyl alcohol inhibition of sn-glycerol 3-phosphate acylation in Escherichia coli.

In vivo and in vitro experiments were performed to determine how phenethyl alcohol (PEA) inhibits phospholipid synthesis in Escherichia coli. This drug drastically reduced the rate of incorporation of sn-glycerol 3-phosphate into the phospholipids of an sn-glycerol 3-phosphate auxotroph. PEA also reduced the rate of fatty acid incorporation into the phospholipids of a fatty acid auxotroph. The kinetics of PEA inhibition of the rate of incorporation of sn-glycerol 3-phosphate were almost identical to those of PEA inhibition of the rate of fatty acid incorporation into phospholipids. The in vivo experiments suggested that the rate-limiting step(s) in phospholipid biosynthesis inhibited by PEA is at the level of the acylation of sn-glycerol 3-phosphate or beyond this step. PEA inhibited the sn-glycerol 3-phosphate acyltransferase with either palmitoyl coenzyme A or palmitoyl-acyl carrier protein as the acyl donor. This drug, however, had no effect on the cytidine 5'-diphosphate-diglyceride:glycerol 3-phosphate phosphatidyl transferase, cytidine 5'-diphosphate-diglyceride:L-serine phosphatidyl transferase, and acyl coenzyme A:lysophatidic acid acyltransferase. The in vitro findings suggested that PEA inhibits phospholipid synthesis primarily at the level of sn-glycerol 3-phosphate acyltransferase.

A confirmation of the phase behavior of Escherichia coli cytoplasmic membrane lipids by x-ray diffraction

The lipid fatty acid composition of the cytoplasmic membranes of Escherichia coli can be varied by growing an unsaturated fatty acid auxotroph in the presence of different fatty acid supplements. Electron spin resonance (ESR) studies of spin-label partitioning into the cytoplasmic membranes of different lipid fatty acid compositions as a function of temperature have been interpreted as indicating a broad order-to-disorder transition in the membrane lipids, the end points of the transition depending upon the fatty acid composition. We have utilized x-ray diffraction to confirm the ESR studies for three different fatty acid supplements (oleic, elaidic, and bromostearic). We found that the characteristic end-point temperatures detected by ESR were indeed the end-point temperatures of a broad order-to-disorder transition of the cytoplasmic membrane lipids. In addition, Patterson functions calculated from lamellar x-ray diffraction from partially oriented cytoplasmic membranes indicate a decrease in average membrane thickness upon fatty acid chain melting.

Membrane fluidity and chemotaxis: Effects of temperature and membrane lipid composition on the swimming behavior of Salmonella typhimurium and Escherichia coli

The effect of membrane fluidity on the tumbling frequency, velocity, attractant response times, and viability was studied in strains of Salmonella typhimurium and Escherichia coli. Lipid composition was altered (1) by varying the growth temperature, (2) by using the unsaturated fatty acid auxotroph of E. coli 30E, or (3) by growing S. typhimurium in the presence of the unsaturated fatty acid synthesis inhibitor 3-decynoyl-N-acetyl-cysteamine supplemented with elaidic or oleic acids. Velocity increased linearly with increasing observation temperature and was independent of growth temperature. Tumble frequency showed a bell-shaped dependence on temperature with the highest tumble frequency near the growth temperature. No significant differences in tumble frequency or response times were observed between cells grown on oleic or elaidic acid above or below the elaidic acid transition temperature of 30°C. E. coli strain 30E, when grown on elaidic acid, however, shows a distinct loss of viability when incubated below 30°C, which can explain the previously reported lack of chemotaxis in such cells. A three-component model including a tumble regulator as a control in chemotaxis is capable of explaining the results. Some membrane processes are critically dependent on membrane fluidity and others are not.

Fatty acid synthesis in Escherichia coli is indirectly inhibited by phenethyl alcohol.

Experiments were performed to determine how phenethyl alcohol inhibits phospholipid synthesis in E. coli. At a nonbacteriostatic concentration, the drug reduces the rate of de novo fatty acid and phospholipid synthesis by 60 to 70%. The inhibition of fatty acid synthesis was found to be a secondary consequence of the inhibition of phospholipid synthesis. Phenethyl alcohol reduces the rate of incorporation of exogenous fatty acids into the phospholipids of a fatty acid auxotroph by 60%. These results indicate that this drug controls phospholipid synthesis beyond the level of fatty acid synthesis. Phenethyl alcohol inhibits the synthesis of phospholipids containing saturated fatty acids to a greater extent than it does the synthesis of phospholipids containing unsaturated fatty acids. It controls the synthesis of phospholipids containing saturated fatty acids at both the level of fatty acid synthesis and the level of incorporation of the saturated fatty acids into phospholipids. The synthesis of phospholipids containing unsaturated fatty acids is inhibited at the level of incorporation of the fatty acids into phospholipids.

Conjugated polyene fatty acids as fluorescent probes: biosynthetic incorporation of parinaric acid by Escherichia coli and studies of phase transitions

The use of the fluorescent fatty acid, parinaric acid (9, 11, 13, 15-octadecatetraenoic acid) (PnA), was studied in cells of an unsaturated fatty acid auxotroph of Escherichia coli. Growth conditions were found that permitted biosynthetic incorporation of PnA (up to 3%) into membrane phospholipids during growth on oleic or elaidic acid. Fluorescence measurements of incorporated PnA revealed phase transitions in cells, membranes, and phospholipids at temperatures that reflected the fatty acid composition of the sample. Transitions had a well-defined onset from high temperature, while the lower and end point was less well defined. cis- and trans-PnA (cis, trnas, trans, cis, and all trans, respectively) gave comparable results. Similar phase transitions were detected with PnA, which was not biosynthetically incorporated. Fluorescence of tryptophan was measured in E. coli membranes as a function of concentration of PnA. Significant quenching of tryptophan fluorescence by PnA was observed.

Studies of energy linked reactions: A cofactor function for unsaturated fatty acids in oxidative phosphorylation; studies with a yeast auxotroph

Mitochondria from the yeast unsaturated fatty acid auxotroph KD115 when grown with a high unsaturated fatty acid supplement catalyse normal oxidative phosphorylation and dihydrolipate-dependent ATP synthesis in the absence of added cofactors. Mitochondria from unsaturated fatty acid depleted KD115 cells have half the unsaturated fatty acid content of supplemented cells and do not catalyse oxidative phosphorylation and dihydrolipoate dependent ATP synthesis. Dihydrolipoate-dependent ATP synthesis can be restored specifically by addition of cofactor amounts of oleic acid and oleoyl CoA. The results provide further evidence for a cofactor role for an unsaturated fatty acid in oxidative phosphorylation.

Structure and synthesis of a lipid-containing bacteriophage. Effects of lipids containing cis or trans fatty acids on the reconstitution of bacteriophage PM2.

Infectious PM2 virus paticles could be reconstituted in vitro from a mixture of nucleocapsid, phospholipids containing cis fatty acids, and proteins I and II. The presence or absence of acyl phosphatidylglycerol, a minor lipid component of thevirion, did not affect the reconstitution of infectious particles, even though it was incorporated into the particles when present. When phosphatidylglycerol was completely replaced by acyl phosphatidylglycerol in the reconstitution mixture, no infections particles were formed. Lipids containing either cis or trans fatty acids were also used for reconstitution in vitro of the lipid-containing bacteriophage PM2. Regardless of the ratio of phosphatidlyglycerol to phospatidylethanolamine in the reconstitution mixture, infectious particles were formed and had almost the same phospholipid composition when lipids containing cis-palmitoleic acid were used; no infectious particles were obtained when lipids containing trans-palmitoleic acid were used. In the latter case, virus-like particles were, however, formed. Reconstituted particles containing cis fatty acids were infectious when tested on wild type Pseudomonas BAL-31 as well as on the unsaturated fatty acid auxotroph grown in the presence of either cis or trans-palmitoleic acid. Reconstituted particles containing trans fatty acids were not infectious on any of these cells. When trans fatty acids as well as cis fatty acids were present in the reconstitution mixture, then there was a lower yield of infectious particles. Particles with either cis or trans fatty acids had all four viral proteins and adsorbed to BAL-31 host cells in a specific manner.

Lipid and protein segregation in Escherichia coli membrane: morphological and structural study of different cytoplasmic membrane fractions.

Lipid and protein segregations can be induced in E. coli cytoplasmic membranes by conformational transitions of their lipid hydrocarbon chains from a disordered to an ordered state. For E. coli strain K 1059 (an unsaturated fatty acid auxotroph) supplemented with linolenic acid, the segregation leads to large areas of membrane surfaces having distinctly different morphological characteristics (smooth compared with strongly particulated fracture faces, as visualized by freeze fracture electron microscopy). The different regions are physically separated by osmotic lysis of spheroplasts at temperatures below those of the order-disorder transition of the lipid hydrocarbon chains. The analysis of the different cytoplasmic membrane fractions provides a direct demonstration and allows a direct analysis of the segregation. As compared to the nonfractionated membranes, the membrane regions corresponding to the smooth fracture surfaces are poor in proteins, rich in lipids, and enriched in saturated fatty acids, while the membrane regions corresponding to the strongly particulated fracture surfaces are rich in proteins, poor in lipids, and enriched in unsaturated fatty acids. Quantitative information about the extent of these segregations is obtained from high-angle x-ray diffraction of the different membrane fractions and of the corresponding total lipid extracts.

Effect of lipids on the reconstitution of D-lactate oxidase in Escherichia coli membrane vesicles.

An unsaturated fatty acid auxotroph lacking D-lactate dehydrogenase activity has been isolated from Escherichia coli ML 308-225 dld-3. While NADH oxidase activity in membrane vesicles prepared from the mutant cells grown in a variety of unsaturated fatty acids is comparable to that of previously isolated fatty acid auxotrophs, D-lactate oxidase activity is absent. However, D-lactate oxidase ativity can be restored when vesicles are incubated with a purified preparation of D-lactate dehydrogenase obtained from wild type cells. The effect of altering the fatty acid composition of the membrane on the reconstitution of D-lactate oxidase activity was examined. Binding of purified D-lactate dehydrogenase was not affected by either the lipid composition of the membrane vesicles or the temperature during reconstitution. However, the reconstitution of D-lactate oxidase activity was strongly influenced by the fatty acid composition of the membrane lipids. The temperature dependence of the reconstituted activity was analyzed. Temperature transitions were not observed with membrane vesicles supplemented with oleic or linolenic acid but palmitelaidic acid-enriched vesicles exhibited a transition temperature at 30 degrees. Attempts to reconstitute the elaidic-supplemented vesicles at temperatures below 30 degrees failed to yield active D-lactate oxidase and the vesicles aggregated. At 42 degrees, aggregation did not occur and D-lactate oxidase activity was obtained to a level of 10% of that found with membranes from the parent strain (ML 308-225). These results suggest that although binding of D-lactate dehydrogenase is independent of the physical state of the membrane, reconstitution of the D-lactate oxidase activity in membrane vesicles is dependent upon the fatty acid composition of the phospholipid, hence the physical state of the membrane.

Independence of deoxyribonucleic acid replication and initiation from membrane fluidity and the supply of unsaturated fatty acids in Escherichia coli.

Mutant derivatives of the unsaturated fatty acid auxotroph K1062 were employed to investigate whether the supposedly membrane-bound bacterial replication machinery requires for its replicatory functions a fluid membrane environment as is known for several membrane-associated protein functions. Temperatures Tt for fluid reversible nonfluid phase transitions of membrane phospholipids are raised from below 18 to 38 degrees C when mutant cells are supplemented with elaidate instead of with oleate. In this experimental system current or synchroneously initiated new rounds of DNA replication are shown in vivo to continue 8 degrees below Tt, provided appropriate corrections for the concurrent cellular metabolic breakdown are considered. Temperature rate profiles for in vitro deoxyribonucleic acid replication rates measured in lysates of either oleate- or elaidate-supplemented cells yield congruent Arrhenius plots without discontinuities at corresponding Tt positions. We conclude that neither the start nor the propagation of replication forks depends on a fluid membrane. The capacity for the assembly of new replication complexes was studied in replication-aligned cells either shifted from oleate to elaidate (at temperatures below Tt for newly synthesized phospholipids) or starved for oleate. Regardless of whether unsaturated fatty acids are exchanged or completely withheld, new replication complexes can be normally assembled and initiated. These results do not support the conclusions reached by Fralick and Lark (1973) that the availability of unsaturated fatty acids is a prerequisite for the assembly of a functional replication complex.

Importance of membrane fluidity in the induction of alkaline phosphatase, a periplasmic enzyme, in [formula omitted

An unsaturated fatty acid auxotroph was supplemented with either elaidate or oleate. After derepression of alkaline phosphatase by phosphate limitation at 38°C, the cells were shifted to incubation at various temperatures. Arrhenius plots of the rate of enzyme induction gave a steeper negative slope in the temperature range from 30°C to 35°C with elaidate-supplemented cells than with oleate-supplemented cells. At 25°C the induction was arrested in the former cells, while it was continued at a considerable rate in the latter. The arrest was released upon shift-back to 38°C, and precursors convertible to the active enzyme were not accumulated during incubation at 25°C. There was no marked difference in slope of Arrhenius plots of the rate of bulk protein synthesis between both types of cells, and the slope was almost equal to that of the rate of enzyme induction in the oleate-supplemented cells. The rate of β-galactosidase induction in the elaidate cells showed a similar temperature dependence to that of bulk protein synthesis.

Function of phosphatidylglycerol molecular species in membranes activation of membrane-bound SN-glycerol 3-phosphate acyltransferase in Escherichia coli

sn-Glycerol 3-phosphate acyltransferase (EC 2.3.1.15) bound to the elaidateenriched membranes of an unsaturated fatty acid auxotroph of Escherichia coli had a lower specific activity than the acyltransferase associated with the wildtype membranes. The 1-saturated-2- cis-unsaturated and 1,2-di- cis-unsaturated molecular species of phosphatidylglycerol activated this enzyme. However, these molecular species did not change the original temperature profile obtained by Arrhenius plots of the enzyme activity bound to the elaidate-enriched membranes.

Positional specificity of cyclopropane ring formation from cis-octadecenoic acid isomers in Escherichia coli

An unsaturated fatty acid auxotroph of Escherichia coli was grown with a series of cis-octadecenoate isomers in which the location of the double bond varied from positions 3 to 17. Each of these fatty acid isomers was incorporated into the cellular lipids, but cyclopropane derivatives were formed to at least a 3-fold greater extent from the cis-9 and cis-11 isomers than from any other positional isomers. The extent of cyclopropane acid formation was observed to be highly dependent on the rate of shaking of the culture. A culture shaking at 340 rev./min converted 8.7% of its oleate to the cyclopropane derivative at stationary phase, whereas a parellel culture shaken at 110 rev./min converted 66% of the oleate to a cyclopropane acid. The inability to observe selectivity or form derivatives from isomers other than the is-9 and cis-11 isomers seems to be due to enzyme specificity rather than a secondary affect of the abnormal unconverted fatty acids on the cell, because the cis-9 isomer is converted to its cyclopropane derivative even in cells grown with abnormal unreactive positional isomers. The preferred substrates for cyclopropanecarboxylic acid formation contained a cis ethylenic bond at either the 9 position or the -7 position. In combination with results of previous studies the specificity reported here supports a concept that two different enzymes may participate in cyclopropane ring synthesis. One enzyme activity may recognize its substrate by the distance from the π-bond to the carboxyl group and the other by the distance to the methyl group.

Phase Transitions in the Membrane of a Marine Bacterium, Pseudomonas BAL‐31

An unsaturated fatty acid auxotroph, strain UFA, isolated from the marine pseudomonad Pseudomonas BAL-31, host cell of the lipid-containing bacteriophage PM2, was grown in media supplemented with different unsaturated fatty acids. Under these conditions the fatty acid composition of the cell could be altered drastically. The phase transition in the native membrane and in the extracted lipids was analyzed by electron spin resonance using a nitroxide spin probe. Membranes prepared from strain UFA grown in cis16:1 or trans16:1 showed one transition at 9.4 degrees C and 12.4 degrees C respectively. Extracted lipids in both cases had almost the same transition temperature as that of the intact membrane. Membranes prepared from Pseudomonas BAL-31 had one transition at approximately 12 degrees C, on the other hand there was no clear cut phase transition using extracted lipids. Replication of bacteriophage PM2 took place below the transition temperature of the membrane lipids in the case where strain UFA was grown in tran16:1. Other cases were not studied.

Conjugated polyene fatty acids as fluorescent membrane probes: model system studies.

The use of conjugated polyene fatty acids as probes of membrane structure is examined, alpha- and beta-parinaric acid (cis, trans, trans, cis- and all trans-9, 11, 13, 15-octadecatetraenoic acid) and synthetic lecithins containing an alpha-parinaric acid chai in position 2 have been prepared, and their absorption and fluorescence properties have been determined. Their absorption spectra are at sufficiently long wavelength to be unobscured by cellular chromophores such as nucleotides and aromatic amin acids. Parinaric acid absorption does, however, overlap tryptophan emission which allows fluorescence energy transfer. Potential uses of these fluorescent probes are presented with studies on mode systems with known physical properties. Dipalmitoyl phosphatidylcholine exhibits a sharp phase transition 1 degree wide at 42 degrees C, as monitored by the fluorescence intensit of parinaric acid. The magnitude of the transition is independent of probe concentration, but the width of the transition and hysteresis are dependent upon such factors as the probe concentration and whether or not sonication is used in sample preparation. Using both fluorescence and absorption properties of the probe, we show that the addition of cholesterol to the dispersion broadens and decreases the magnitude of the transition. These results are interpreted in terms of a change in the polarizability of the acyl chains of a lipid bilayer undergoes a thermal transition. Lipid-protein interactions are studied by the binding of alpha-parinaric acid to bovine serum albumin. Fluorescence enhancement, absorption spectral shifts, and quenching of tryptophan fluorescnece are observed when alpha-parinaric acid binds to bovine serum albumin. Calculations based on these measurements are consistent with two binding sites of KB approximately 10(8) (M-1) and three to four binding sites of KB approximately 10(6)-10(7) (M-1), similar to known values for the binding of other long-chain fatty acids. Biosynthetic incorporation of beta-parinaric acid into the E. coli fatty acid auxotroph 30E betaox has been accomplished and phase transitions in cells and isolated phospholipids are shown.

NMR Studies of Biophysical Problems

Nuclear magnetic resonance (NMR) has become more and more actively used in various biophysical and biochemical studies. Under such circumstances it seems timely to publish a special number of this Journal containing papers which introduce recent attainments in this field to non-specialist readers.The importance of NMR in the studies of biological molecules and higher-order systems stems from the following features. (1) Thanks to the developments of high-field NMR spectrometers with superconducting magnets, it is now possible to perform detailed investigations of the microstructure of biopolymers. (2) Possibilities of gaining a deeper insight into molecular motions are opened by the advent of pulse. Fourier transform techniques through accurate measurements of the spin-lattice relaxation times. (3) New aspects of biomolecular structures are being revealed by the application of recent experimental methods such as the use of paramagnetic probes. (4) The observation of resonances of various nuclei other than 1H (e. g., 13C, 15N, 31P, 19F, and 2H) provides a powerful tool for solving many problems which have not been so far tractable.In Japan also, NMR studies of biological systems are steadily increasing in number and advancing in quality. We find that quite a high percentage of the papers reported at the domestic meetings and conferences during 1974-1975 are concerned with either measurements of the spin-lattice relaxation times or the use of paramagnetic probes. Therefore, The papers published in this special number are also more or less related to these topics. The title and author (s) of each paper are as follows.1. Introduction (M. Tasumi & T. Iizuka)2. Molecular motions and spectral assignments based on the NMR relaxation mea- surements (R. Chujõ)3. Use of nuclear magnetic relaxation times to the determination of molecular conformation in solution (K. Akasaka & S. Shibata)4. Side-chain conformation of polyamino acids (K. Nagayama)5. NMR studies on the phospholipids from an unsaturated fatty-acid auxotroph of E. Coli (H. Akutsu & Y. Kyogoku)6. Structural studies on membrane models by NMR spectroscopy (M. Kainosho)7. Studies on the structure and function of hemoproteins and heme-enzymes by FT-NMR with superconducting magnet (I. Morishima, T. Inubushi & T. Iizuka)8. Structural studies of biomolecules by the lanthanide-ion probe method (M. Tasumi & F. Inagaki)

Effectiveness of various unsaturated fatty acids in supporting growth and respiration in Saccharomyces cerevisiae

The saturated fatty acid auxotroph of Saccharomyces cerevisiae, KD115, was used to determine the efficiency of various unsaturated fatty acids in supporting growth. The efficiency, as the number of cells produced per fmol of unsaturated fatty acid, ranged from zero for a number of acids to over 26 cells per fmol of eicosapentaenoic acid. Efficiencies tended to be higher for acids with fewer carbons or more double bonds. In a series of positional isomers of cis-octadecenoic acid, the delta9 isomer had the greatest efficiency (12 cells per fmol). Exogenous oleic acid was taken up and incorporated into cellular lipid early in the growth of the cells. Further growth proceeded with a decrease in the relative content of oleate in lipids until a minimum value of 9 mol % was reached at stationary phase. The initial concentration of supplemental acid did not affect the final mole % value. Other unsaturated fatty acids reached limiting values of mole % in phospholipid at stationary phase that were characteristic for the acid used. When cells were grown with glycerol as the carbon source, the efficiencies of most acids in supporting growth were one-third to one-fifth the value with glucose and the final mole % of supplement acid in phospholipid at stationary phase was two to five times greater. Apparently, mitochondrial energy transduction necessary for glycerol utilization requires higher levels of unsaturated fatty acids in membrane lipids than do extramitochondrial functions. The respiratory rate of mitochondria was not decreased at lower levels of oleic or palmitoleic acid in lipids, although respiratory control was lower when the mole % of unsaturated fatty acid was lower. Mitochondria from cells supplemented with eicosaenoic acid were found to have both low respiration and respiratory control. The decreased respiration of these mitochondria coincided with a decreased cytochrome content, not a decrease in respiration per mol of cytochrome.

Requirements for unsaturated fatty acids for the induction on respiration in Saccharomyces cerevisiae

Unsaturated fatty acids provided during the release from glucose repression were shown to be essential for derepression of respiration in an unsaturated fatty acid auxotroph of Saccharomyces cerevisiae (KD115). Cells derepressed in the presence of oleic acid contained three to six times as much cytochrome per cell as those derepressed in the absence of unsaturated fatty acid or those derepressed with eicosaenoic acid. The delta9 isomer was the most efficient of the cis-octadecenoic acid isomers in supporting that increase, and eicosaenoic acid supported an increase at only 15% the rate observed with oleic acid. Derepression, even in the presence of oleic acid, proceeded only after a lag of 3 hours. When glucose was removed prior to the addition of oleate, the lag was reduced by the time of the preincubation with glycerol. This result suggests that some processes necessary for increased respiration can proceed in the absence of an added unsaturated fatty acid, but these processes apparently require certain levels of unsaturated acids in the pre-existing lipids, since they occurred in cells whose membranes contained 50 mol % oleate, but not in cells containing only 20 mol %. These processes leading to eventual increased respiration were inhibited by cycloheximide but not chloramphenicol, suggesting that protein synthesis on cytoplasmic ribosomes but not mitochondrial ribosomes were required. Derepression in the absence of oleate for 3 hours lessened the inhibition or respiration induction by ethidium bromide. This result indicates that the transcription of mitochondrial DNA necessary for the induction of respiration may have occurred in the absence of added unsaturated fatty acid, but that some subsequent event required added esterified unsaturated fatty acid.

Phosphatidylethanolamine molecular species of fatty acid auxotroph of Escherichia coli grown with elaidate

Monoacetyldiglycerides derived from the phosphatidylethanolamine molecular species of the fatty acid auxotroph of Escherichia coli grown with elaidate at 37 °C were fractionated on thin-layer plates of silica impregnated with silver nitrate and were identified by gas chromatography-mass spectrometry with an OV-17 column and gas chromatography with a Silar-10C column. Phosphatidylethanolamine was made up of the following molecular species: 1–16 : 0-2-16 : 0 (1.2%), 1–14 : 0-2-trans-16 : 1 (1%), 1–16 : 0-2-trans-16 : 1 (3.5%), 1–16 : 0-2-trans-18 : 1 (26.4%), 1–16 : 0-2-cis-16 : 1 (3.8%), 1-trans-18 : 1-2-trans-16 : 1 (13.2%), 1-trans-18 : 1-2-trans-18 : 1 (44.9%), 1-trans-18 : 1-2-cis-16 : 1 (4.5%) and trans-18 : 1-cis-18 : 1 (1.5%).