Development of an anti-bleeding agent for recombinant hirudin induced skin bleeding in the pig.

Recombinant hirudin (rH) is a highly specific thrombin inhibitor which is under clinical investigation for various thrombotic disorders. However, one of its potential risks during clinical use might be hemorrhage, especially when combined with other agents interfering with the coagulation system like antiplatelet or fibrinolytic agents. In this experimental study we investigated whether Haemate, a von Willebrand Factor (vWF) and factor VIII containing product, could correct rH/aspirin induced bleeding in an experimental pig study. Skin bleeding time was evaluated in three open, placebo-controlled, randomized studies following comparable designs. A total of 62 animals were given a short-term infusion of aspirin (20 mg/kg) followed by a three-hour infusion of a high or low dose (0.3 or 0.5 mg/kg/h) of rH. At cessation of rH infusion, animals were allocated to treatment with either Haemate (30 FVIII U/kg) or the recombinant factor VIII, Helixate, which is devoid of vWF. The skin bleeding time (SBT, given as times of baseline) as measured four hours after the start of the rH infusion was defined as the prospective endpoint. In study 1 (low dose rH + Haemate) 4 h SBT was 2.18 (placebo) and 1.61 (Haemate, p = 0.0111). In study No. 2 (high dose rH + Haemate) SBT was 2.58 in placebo and 1.73 in Haemate (p = 0.0001). No significant difference between placebo and treatment were detected in study No. 3 (low dose rH + Helixate). Haemate but not Helixate significantly decreased bleeding time as compared to placebo at termination of the study (7 hours) which was defined as the secondary endpoint. No effect on either aPTT nor rH plasma levels were observed with any of the study drugs. It was concluded that Haemate decreases excess bleeding induced by rH/aspirin treatment without altering rH's anticoagulant effect.

Effects of permeation enhancers on the transport of a peptidomimetic thrombin inhibitor (CRC 220) in a human intestinal cell line (Caco-2).

PURPOSE: The effects of five different permeation enhancer systems on the transport properties of a peptidomimetic thrombin inhibitor. CRC 220, were investigated in monolayers of a human intestinal cell line (Caco-2). METHODS: The transepithelial transport rates and additionally the cytotoxic properties of these enhancers were characterized using the following tests: measurement of the transepithelial electrical resistance (TEER), the MTT-transformation, the protein content and the release of cytosolic lactate dehydrogenase (LDH), as well as FITC-phalloidin and propidium iodide staining. RESULTS: All permeation enhancer systems showed concentration-dependent effects on cell permeability and toxicity. The most prominent effects on peptide transport were seen at the highest concentration (40 mM), yielding the rank order, NaTC > NaTC/Cholesterol > Solulan C24 > NaTC/Oleic acid > NaTC/PC18. Using the TEER after 120 min exposure as the most sensitive parameter describing cytotoxicity, the following order was obtained: Solulan C24 > NaTC > NaTC/PC18 = NaTC/Cholesterol > NaTC/Oleic acid > NaTC/PC. Generally, efficient enhancement of peptide transport was associated with a noticeable influence on cell viability under in-vitro conditions. CONCLUSIONS: Taking into account permeation and cytotoxicity as a function of concentration, both NaTC at 15 mM and the mixed micellar system NaTC/oleic acid at 0.75 mM offer interesting enhancement properties, showing an 18-fold increase in CRC 220 transport rates. The effects on cell viability and cytotoxicity were comparatively low and of reversible nature.

The peptide-based thrombin inhibitor CRC 220 is a new substrate of the basolateral rat liver organic anion-transporting polypeptide.

The peptidomimetic thrombin inhibitor CRC 220, 4-methoxy-2,3,6-trimethylphenylsulfonyl-L-aspartyl-D-4-amidinop henylalanyl- piperidide, is taken up into isolated rat hepatocytes through active, carrier-mediated transport. This uptake is inhibited by bile acids. Functional expression in Xenopus laevis oocytes was performed to identify the transport system responsible for the hepatocellular CRC 220 uptake. Injection of poly(A)+RNA in X. laevis oocytes resulted in a two- to three-times higher uptake of CRC 220, compared with uninjected or water-injected control oocytes. Taurocholate (200 mumol/L) inhibited this uptake completely. No uptake of the peptidomimetic thrombin inhibitor was observed, when X. laevis oocytes were injected with complementary RNA (cRNA) encoding either the cloned rat liver Na(+)-dependent taurocholate transporter Ntcp, the renal oligopeptide carrier rhaPT or the intestinal oligopeptide transporter PepT1. However, after injection of cRNA of the cloned rat liver Na(+)-independent organic anion transporting polypeptide oatp, a specific and saturable CRC 220 uptake was observed (Michaelis-Menten constant 29.5 mumol/L). Cis-inhibition with known oatp-substrates, e.g., 20 mumol/L Bromsulphalein (BSP), 2007 mumol/L taurocholate and 2007 mumol/L cholate, occurred in oatp-expressing X. laevis oocytes, whereas substrates of the two peptide carriers as well as dipeptide- and single-amino acid constituents of the thrombin inhibitor itself lacked any significant inhibitory effects. These data show that the modified dipeptide CRC 220 is a highly selective substrate of the organic anion transporting polypeptide oatp in the basolateral plasma membrane of rat hepatocytes.

First-pass elimination of a peptidomimetic thrombin inhibitor is due to carrier-mediated uptake by the liver. Interaction with bile acid transport systems.

CRC 220 (4-methoxy-2, 3, 6-trimethylphenylsulfonyl-L-aspartyl-D-4-amidinophenylalanyl -piperidide) is a competitive peptide-based trombin inhibitor with high affinity to human alpha-thrombin (Ki 2.5 nM). The amphiphilic compound exhibits virtually no systemic bioavailability despite proteolytic stability and proven enteral absorption. After intravenous application (V. jejunalis) in rats CRC 220 is almost completely excreted into bile. Simultaneous administration of bile acids considerably decreases this first-pass elimination. CRC 220 is extensively taken up in isolated rat hepatocytes by a saturable carrier-mediated transport with Km 23.7 microM and Vmax 775 pmol x mg-1 x min-1. A large part of this transport is energy-dependent. At temperatures above 20 degrees C, the uptake is accelerated exponentially. The activation energy amounts to 82 kj/mol. A minor portion of CRC 220 uptake occurs by physical diffusion with a permeability coefficient of 7.83 x 10(-7) cm/sec at 12 degrees C. Sodium ions energize CRC 220 uptake. Replacement of sodium by choline or lithium decreases the transport rate of 23-40%. In addition, a negative membrane potential facilitates the uptake. CRC 220 transport is only observed in hepatocytes: it is absent in BHK, FAO, HepG2, HPCT 1E3, and HPCT 1E3-TC cells. In the presence of 4-amidinophenylalanine derivatives, CRC 220 uptake is considerably decreased. Inhibition also occurs with bile acids and bromosulfophthalein, but less with bumetanide. Because CRC 220 inhibits bile acid uptake into hepatocytes and vice versa, the results suggest that the first-pass elimination of this amphiphilic thrombin inhibitor is due to an active carrier-mediated transport process in the basolateral plasma membrane of rat hepatocytes, and that this transport occurs via a bile acid transport system.

Reduction of r-hirudin induced bleeding in pigs by the administration of von Willebrand factor.

To prevent r-hirudin induced excess bleeding an animal model was established in pigs for the investigation of an anti-bleeding strategy. We used the Simplate® device to monitor skin bleeding time (SBT) at the inner site of the ear. r-Hirudin infused in a dose of 0.3 mag per h induced a prominent increase of SBT. The aim of our studies was to reverse r-hirudin induced bleeding by enhancing platelet adhesion to the endothelium via the administration of von Willebrand Factor (vWF). Pigs were treated with vWF containing solutions (Haemate® and a vWF-concentrate) at 3h after the start of the r-hirudin infusion. Both compounds suppressed SBT 1h after administration and significantly prevented bleeding until the termination of the experiment. SBT values (given in times of baseline) in the placebo group were 3.32 ± 0.9, 1.51 ± 0.14 in the Haemate® and 1.85 ± 0.42 in the vWF concentrate group (P = 0.008 or 0.032, in a two-sided Kruskall-Wallis-test). Coagulation parameters (aPTT, PT) were unaltered by the treatment, as were the r-hirudin plasma levels suggesting that vWF is not an antidote in its strict sense. It is concluded that vWF reverses bleeding without altering the anticoagulant effect of r-hirudin. Addition of 20 mg/kg per h aspirin resulted in a further increase of SBT. Aspirin, moreover, suppressed platelet aggregation but did not alter platelet counts. In a further study, bleeding induced by r-hirudin and aspirin was antagonized by Haemate® (66 Ukg i.v. bolus + 187 Ukg per h for 2 h infusion) and a significant reduction of bleeding time occurred.

Transepithelial transport properties of peptidomimetic thrombin inhibitors in monolayers of a human intestinal cell line (Caco-2) and their correlation to in vivo data.

Peptidomimetic thrombin inhibitors (TI), derived from L-Asp-D-Phe were examined in confluent monolayers of a human colon carcinoma cell line (Caco-2) to elucidate their transepithelial transport properties. Effect availabilities, based on activated partial thromboplastin time (aPTT) measurements in rats, after peroral administration of five TI correlated reasonably well with permeability coefficients obtained from in vitro transport studies in Caco-2 monolayers, whereas physicochemical properties, such as molecular mass, solubilities, pKa and octanol-buffer partition coefficients failed to yield meaningful relationships. Substitution of the beta-carboxylic group of L-Asp leads to analogues which are mainly transported by passive diffusion, while an unsubstituted carboxylic group favours carrier-mediated active transport. The effects of concentration, temperature, competitive inhibitors and direction dependence on in vitro transport were investigated. The results obtained are compatible with a saturable carrier-mediated transport, operating parallel to a passive paracellular route. The Michaelis-Menten parameters for the active transport component (Km = 1.67 mM, Vmax = 26.5 pmol min-1 mg protein-1) indicate an involvement of the intestinal di/tripeptide transport system for one of the TI. The Caco-2 transport model may be helpful for the design of perorally active peptidomimetics.

Preparation and evaluation of PEG-bound thrombin inhibitors based on 4-amidinophenylalanine.

The dipeptide Mtr-Asp-D-Adf-Pip 10 represents a potent thrombin inhibitor. In comparison to NAPAP, 10 exhibited improved tolerability and a longer half-life in vivo, i.e., 20 +/- 5 min. We have coupled aminopolyethyleneglycolmonomethylether of various molecular weights to the carboxyl moiety of 10 and evaluated their biological properties. First, Mtr-Asp-OBut was coupled to the amino group of the PEG employing TOTU as an activating agent. This was followed by the removal of the OBut protecting group and coupling of D-Adf-Pip using TOTU as well. The PEG-bound thrombin inhibitors showed inhibition constants vs. thrombin in the subnanomolar range, i.e., they were more active than the parent molecule 10. Moreover, the pegylated inhibitors exhibited a longer lasting effect in vivo. In rats the half-life of Mtr-Asn (PEG10000-OMe)-D-Adf-Pip 14 was determined to be 63 min. Mtr-Asn(PEG10000-OMe)-D-Adf-Pip 14 showed a half-life of 120 min in pigs. It could be concluded that these PEG-bound thrombin inhibitors may be employed as versatile drugs for parenteral administration in treating thrombotic disorders.

Pharmacological characterization of a new 4-amidinophenyl-alanine thrombin-inhibitor (CRC 220).

The new thrombin inhibitor CRC 220 was characterized in vivo for its antithrombotic effects. CRC 220 led to a dose-dependent prolongation of clotting parameters as determined in rats, rabbits, dogs, sheeps, pigs and monkeys. We evaluated the efficacy of CRC 220 to prevent thrombus formation in arteries and in the microcirculation in different animal models. In a rabbit model of tissue factor-induced coagulation activation, infusion of 0.5 mg/kg x h CRC 220 (3 hours) led to a significant prevention of fibrinogen decrease. In a rat model of lethal LPS-induced DIC CRC 220 significantly prevented the mortality rate after a 4h-infusion of 0.75 mg/kg x h. Thrombin-induced platelet aggregation in rat lungs could be prevented by the i.v. bolus injection of CRC 220. A dose of 0.3 mg/kg leads to a reduction of more than 80% of platelet deposition in the lung, significant inhibition was still observed 90 minutes after CRC 220 administration; at this time the inhibitor had already been cleared from plasma. Arterial thrombosis was induced in rabbits by squeezing and stenosis of the A. carotis. The i.v. bolus administration of CRC 220 dose-dependently prevented thrombus formation, an ED50 of 0.03 mg/kg was calculated. This dose was associated with only a minor prolongation of aPTT.

Synthesis and characterisation of novel thrombin inhibitors based on 4-amidinophenylalanine.

Thrombin inhibitors have been thought to play a pivotal role in myocardial infarct and stroke incidences and their aftermath. Quite some time ago potent synthetic thrombin inhibitors became known based on peptide derivatives D-Phe-Pro-Arg and benzamidine. One of them, fairly well characterised was beta-naphthylsulphonylglycyl-D,L-4-amidino-phenylalanylpiperidi de (NAPAP). NAPAP was prone to being administered intravenously due to its short plasma half life. Drawbacks to this compound such as effects on histamine release and blood pressure may have obstructed its clinical use. Long half life and oral bioavailability would be desirable for prophylactic treatment of thrombotic disorders. We have used NAPAP as a template for new synthetic compounds to improve some characteristics of its profile. For screening purposes we have investigated fairly simple surrogate parameters, aspects that were considered to contribute to pharmacological effects. Potency was correlated to thrombin inhibition, side effects were addressed by specificity toward thrombin as well as reduction in basicity, and plasma half life was considered to be modulated by plasma stability of the compound. Oral bioavailability would be affected by instability during the passage through the gut wall. Chemical introduction of a carboxylic group and exchange of the naphthyl group for 4-methoxy-2,3,6-trimethylphenyl led to a compound that when compared to NAPAP, exhibited a 4-fold increase in thrombin inhibitory activity and a 3-fold increase in trypsin specificity. Plasma stability decreased to 22 h, however, sufficient enough not to play a major role in plasma half life. Gut homogenate stability of the compound has not changed. The potency increase did not translate into a reduction in IC50-values for the coagulation assay aPTT and TT, in contrast to the IC50-values for thrombin-induced platelet aggregation.

Comparison of in vitro and in vivo properties of rhirudin (HBW 023) and a synthetic analogous peptide.

Recombinant hirudin and a shortened synthetic analogue, with the amino acid sequence of D-Phe-Pro-Arg-Pro-(Gly)4-Asn-Gly-Asp-Phe-Glu- Glu-Ile-Pro-Glu-Glu-Tyr-Leu, are specific thrombin inhibitors which in a concentration-dependent manner inhibit thrombus formation as well as clot propagation both in vitro and in vivo. In comparison to the analogue, lower molar concentrations of rhirudin affected doubling of aPTT and TT as well as inhibition of thrombin amidolytic activity or thrombin-induced platelet aggregation in vitro. In the rat wire coil-induced thrombosis model, a 50% thromboprotective effect may be brought about with doses of 0.043 mumol/kg of rhirudin and 1.43 mumol/kg of the synthetic peptide. However, doubling of bleeding times is caused, on average, by dosages of between 0.143 and 0.43 mumol/kg rhirudin or approximately 0.143 mumol/kg of the analogue. Treatment groups included animals revealing significant prolongation of bleeding times as well as nonresponders. Despite the 10-fold longer impact on aPTT after application of rhirudin, the extent of mean bleeding time prolongation is identical to that of the analogue.

Na+ efflux mechanisms in ventricular myocytes: measurement of [Na+]i with Na(+)-binding benzofuran isophthalate.

We characterized the Na(+)-sensitive dye benzofuran isophthalate (SBFI) with fluorescence microscopy in isolated, adult rat ventricular myocytes. When cells were loaded with SBFI by incubation with 10 microM of the acetoxymethyl ester, fluorescence excitation spectra were markedly attenuated below 340 nm and the isoemissive point was blue shifted by approximately 25 nm when compared with spectra from SBFI acid in buffered solutions. Fluorescence intensity (49 +/- 3%) was partially released by permeabilization of the sarcolemma with digitonin, suggesting that one-half of the dye molecules are sequestered subcellularly in a compartment shown most likely to be mitochondrial. Intracellular Na+ concentration ([Na+]i) was determined by in situ calibration using cation-selective ionophores and was found to be 14 +/- 2 mM in cells studied at 37 degrees C. The relative importance of Na+ efflux mechanisms in myocytes was investigated. Substitution of Ca2+ and Mg2+ with EGTA in the superfusing medium resulted in a reversible rise of [Na+]i from 13 +/- 2 to 31 +/- 5 mM, which was blocked by 1 microM verapamil. Cellular efflux of Na+ after loading in this manner was found to be insensitive to blockade of Na(+)-Ca2+ exchange but was abolished when Na(+)-K(+)-ATPase was inhibited with zero extracellular K+ concentration. We conclude that SBFI can be used to measure [Na+]i in ventricular cells nondestructively and without impalement. Na+ efflux after loading by Ca2+ and Mg2+ withdrawal is mediated by the Na+ pump with no measurable contribution from Na(+)-Ca2+ exchange.

J-aggregate formation of a carbocyanine as a quantitative fluorescent indicator of membrane potential.

The spectral properties of a novel membrane potential sensitive probe (JC-1) were characterized in aqueous buffers and in isolated cardiac mitochondria. JC-1 is a carbocyanine with a delocalized positive charge. It formed under favorable conditions a concentration-dependent fluorescent nematic phase consisting of J-aggregates. When excited at 490 nm, the monomers exhibited an emission maximum at 527 nm and J-aggregates at 590 nm. Increasing concentrations of JC-1 above a certain concentration caused a linear rise in the J-aggregate fluorescence, while the monomer fluorescence remained constant. The membrane potential of energized mitochondria (negative inside) promoted a directional uptake of JC-1 into the matrix, also with subsequent formation of J-aggregates. The J-aggregate fluorescence was sensitive to transient membrane potential changes induced by ADP and to metabolic inhibitors of oxidative phosphorylation. The J-aggregate fluorescence was found to be pH independent within the physiological pH range of 7.15-8.0 and could be linearly calibrated with valinomycin-induced K+ diffusion potentials. The advantage of JC-1 over rhodamines and other carbocyanines is that its color altered reversibly from green to red with increasing membrane potentials. This can be exploited for imaging live mitochondria on the stage of a microscope.

Intracellular heterogeneity in mitochondrial membrane potentials revealed by a J-aggregate-forming lipophilic cation JC-1.

By using a potential-dependent J-aggregate-forming delocalized lipophilic cation, 5,5',6,6'-tetrachloro-1,1',3,3'-tetraethylbenzimidazolocarbocyanine++ + iodide (JC-1), we find that membrane potentials across mitochondria in a living cell can be heterogeneous. Remarkably, even within a long contiguous mitochondrion, regional heterogeneity in membrane potentials appears to be possible.

Endothelin enhances the contractile responsiveness of adult rat ventricular myocytes to calcium by a pertussis toxin-sensitive pathway.

It has long been assumed that the primary influences regulating cardiac contractility are the extent of mechanical loading of muscle fibers and the activity of the autonomic nervous system. However, the vasoactive peptide endothelin, initially found in vascular endothelium, is among the most potent positively inotropic agents yet described in mammalian myocardium. In isolated adult rat ventricular cells, endothelin's action was slow in onset but very long lasting with an EC50 of 50 pM that approximates the reported KD of the peptide for its receptor in rat heart. When the calcium activity of the buffer superfusing isolated single fura-2-loaded myocytes paced at 1.5 Hz was varied from 0.1 to 0.9 mM [Ca2+]o, 100 pM endothelin increased contractile amplitude with no significant change in diastolic or systolic [Ca2+]i, thus appearing to sensitize the myofilaments to intracellular calcium. Pertussis toxin, or prior exposure to a beta-adrenergic agonist, reduced or abolished the increase in myocyte contractility induced by endothelin. This novel and potent pharmacologic action of endothelin points to the potential importance of local, paracrine factors, perhaps derived from microvascular endothelium or endocardium, in the control of the contractile function of the heart.

In situ calibration of fura-2 and BCECF fluorescence in adult rat ventricular myocytes.

Quantitation of Ca+ and H+ activities within cells using presently available fluorescent probes is optimal when the fluorescence signal is calibrated in situ after each experiment. Fura-2 and 2',7'-bis(2-carboxy-ethyl)-5,6-carboxyfluoroscein (BCECF) are difficult to calibrate in freshly dissociated adult cardiac myocytes because calibration procedures produce cellular hypercontracture. In situ calibration was accomplished in rat ventricular cells by saturating fura-2 with La3+, an agent known to produce myocardial relaxation. Since fura-2 has different spectral properties when complexed with La3+ than with Ca2+, scaling factors were defined in vitro and then verified by experiments in cultured neonatal myocytes. In adult rat myocytes using the La3+ method, intracellular Ca2+ concentration ([Ca2+]i) was 131 +/- 47 nM (n = 14) in quiescent cells; diastolic [Ca2+]i and systolic [Ca2+]i in myocytes stimulated at 1 Hz were 140 +/- 56 and 1,088 +/- 211 nM (n = 5), respectively. BCECF fluorescence was calibrated in situ by a method that prevented cellular hypercontracture and reported a pH value of 7.10 +/- 0.10 in cells stimulated at 1.5 Hz. An additional advantage of both methods is that the buffers employed prevented large changes in the redox state of intracellular pyridine nucleotides, thus preventing a change in cellular autofluorescence during the calibration procedure.

Calcium and proton activities in rat cardiac mitochondria. Effect of matrix environment on behaviour of fluorescent probes.

The ionic composition of the mitochondrial matrix, under both physiological and pathophysiological conditions, remains controversial. Although fura-2 and 2',7'-bis-(2-carboxyethyl)-5(6)-carboxyfluorescein (BCECF), fluorescent probes for [Ca2+] and [H+] respectively, have successfully been loaded into mitochondria [Lukács & Kapus (1987) Biochem. J. 248, 609-613; Davis, Altschuld, Jung & Brierley (1987) Biochem. Biophys. Res. Commun. 49, 40-45], the adaptation of fluorescence-ratio spectroscopy to the study of the matrix ion content poses unique problems. In this report, we describe a method for successfully attaching viable rat cardiac mitochondria to glass coverslips, allowing continuous superfusion of isolated organelles during fluorescence microscopy. This technique obviated the need to correct for the accumulation of ion-sensitive and -insensitive fluorescent species of dye both within the matrix and outside of mitochondria in suspension in a cuvette, a particular problem with fura-2. By using this technique for superfusion of immobilized mitochondria, we found the pKa of BCECF for H+ at 25 degrees C shifted from 6.8 in buffer to 7.2 in rat cardiac mitochondria, with a marked hysteresis effect noted for intramitochondrial BCECF calibration curves. At higher pH, photobleaching of BCECF was enhanced. The dissociation constant (Kd) of fura-2 for Ca2+ was found to be 315 nM at 25 degrees C, pH 8.0, but only at [Ca2+] below 1 microM. At matrix [Ca2+] greater than 1 microM, the Kd shifted into the micromolar range, an effect that appeared to be pH-dependent. Importantly, the matrix [Ca2+] was determined to be between 10 and 100 nM at perfusion buffer [Ca2+] below 500 nM, but rose rapidly at the higher extramitochondrial [Ca2+] reported to occur in ischaemic cardiac myocytes. Importantly, mitochondrial transmembrane H+ and Ca2+ gradients both appeared to be maximal at perfusion buffer [H+] and [Ca2+] that approximate those of the cytosol of many resting cells.

Inhibition of mitochondrial phospholipase A2 by mono- and dilysocardiolipin.

Phospholipase A2 extracted from the acetone powder of previously frozen rat liver mitochondria is strongly inhibited compared to the activity manifest before acetone powder preparation. Activity is substantially recovered upon partial purification of the enzyme by gel filtration chromatography. Inhibitor activity elutes in the void volume from the column and is obtained in the chloroform layer when void volume fractions are subjected to a Folch extraction. Structural studies support the inhibitor being monolysocardiolipin. Under the assay conditions employed, 1 molecule of the inhibitor per 5000 substrate molecules or 40 nM on a nominal concentration basis is I50 for the mitochondrial enzyme. The agent is similarly effective against pancreatic and snake venom phospholipases A2. Monolysocardiolipin and dilysocardiolipin prepared enzymatically from bovine heart cardiolipin are less potent than the material arising from rat liver cardiolipin by factors of 10- and 30-fold, respectively, yet are still highly potent compared to the other known inhibitors of this enzyme. Differences in acyl group composition, in the degree of acyl group oxidation, or in structural isomerism between the sn-1 and sn-2 positions of the lyso compounds may account for the difference in potency between the materials derived from rat liver and bovine heart.

Separation of phosphatidylethanolamine from its oxidation and hydrolysis products by high-performance liquid chromatography.

1-Palmitoyl-2-linoleoyl-phosphatidylethanolamine degrades relatively quickly when subjected to common storage and handling procedures. The degradation products consist of compounds in which double bonds in the sn-2 position acyl chain are partially oxidized and of products arising from the hydrolysis of the acyl ester bonds. Thin-layer chromatography (TLC), which is widely utilized to isolate and to ascertain the purity of phospholipids, does not readily separate the oxidation products from the parent lipid class. High-performance liquid chromatography (HPLC), however, employing a normal phase column and an isocratic, UV-transparent solvent system, can be employed to produce a rapid analytical or preparative of phosphatidylethanolamine (PE) from these degradative impurities.