Phospholipids in inflammatory synovial effusions

1H NMR Metabolic Profiling of Synovial Fluid from Patients with Anterior Cruciate Ligament Tears and Hemarthrosis

OBJECTIVE

To compare the metabolic profiles of synovial fluid (SF) from patients with anterior cruciate ligament tears and hemarthrosis (HA) with that of normal controls, using ¹H NMR spectroscopy (NMRS).

METHODS

Synovial fluid was collected from eleven patients undergoing arthroscopic debridement within fourteen days following an anterior cruciate ligament (ACL) tear and hemarthrosis. Ten additional SF samples were obtained from the knees of osteoarthritis-free volunteers to serve as normal controls. The relative concentrations of twenty-eight endogenous SF metabolites (hydroxybutyrate, acetate, acetoacetate, acetone, alanine, arginine, choline, citrate, creatine, creatinine, formate, glucose, glutamate, glutamine, glycerol, glycine, histidine, isoleucine, lactate, leucine, lysine, phenylalanine, proline, pyruvate, threonine, tyrosine, valine, and the mobile components of glycoproteins and lipids) were evaluated using NMRS and quantified using CHENOMX metabolomics analysis software. Mean differences between groups were evaluated with t-tests controlling for multiple comparisons at an overall error rate of 0.10.

RESULTS

Statistically significant increases in the levels of glucose, choline, the branched-chain amino acids leucine, isoleucine, and valine, and the mobile components of N-acetyl glycoproteins and lipids were observed in ACL/HA SF as compared with normal controls; lactate levels were reduced.

CONCLUSIONS

Marked changes occur in the metabolic profiles of human knee fluid following ACL injury and hemarthrosis, suggestive of increased demand and accompanying inflammatory response; potentially increased lipid and glucose metabolism; and possible hyaluronan degradation within the joint following trauma.

Composition of joint fluid in patients undergoing total knee replacement and revision arthroplasty: correlation with flow properties

The protein, phospholipid and hyaluronic acid (HA) contents of joint fluid samples were determined in specimens obtained from patients undergoing total knee arthroplasty (TKA) and revision TKA. It was hypothesized that these components would vary widely among patients undergoing TKA, and that the composition of joint fluid in patients undergoing revision would differ from that in patients undergoing revision. It was further hypothesized that HA concentration and molecular weight would principally determine the flow properties previously reported. Biochemical assays were used to assess protein and phospholipid content, and size exclusion chromatography was used to determine HA concentration and molecular weight. Sixty samples were included in the study. HA, protein, and phospholipid concentrations all varied widely in patients undergoing index TKA and revision TKA. HA concentration was lower in patients undergoing revision arthroplasty due to wear-related failure compared to patients undergoing the index procedure (0.9 +/- 0.4 mg/ml versus 1.3 +/- 0.5 mg/ml, mean +/- standard deviation, p = 0.04). Other components were not different between the groups. Flow properties at high shear rates were correlated with HA concentration and, to a lesser extent, HA molecular weight, but neither protein nor phospholipid concentration. The composition of joint fluid is highly variable in the context of arthroplasty. Much of the variation in flow properties, especially at high shear rate, is explained by large variation in HA concentration and small variation in HA molecular weight. The variation in composition and lower HA concentration in joints necessitating revision may relate to variation in arthroplasty lubrication leading to highly variable wear rates and clinical outcomes.

Electron microscope and biochemical observations of the surface active phospholipids on the articular surfaces and in the synovial fluid of the temporomandibular joint: a preliminary investigation

PURPOSE

The goal of this article is to investigate the surface-active phospholipids located on the articular surfaces and in the temporomandibular joint (TMJ) synovial fluid (SF) by means of electron microscopy and biochemical analysis.

MATERIALS AND METHODS

Synovial fluids and articular cartilage samples taken from 6 normally functioning TMJs were studied. The osmiophilic lining of human TMJ articular surfaces has been studied by using special nondestructive fixation procedures. To study the SF, negative staining technique has been used. In addition, thin-layer chromatography has been used to identify the phospholipids extracted from synovial fluid of human TMJs.

RESULTS

In the SF, granular bodies were identified with diameter of between 170 and 280 nm. Their diameter decreased dramatically when exposed to phospholipase-A(2). The amorphous and highly osmophilic material on the articular surface include membrane-bound vesicles (270 nm in diameter) with lamellated pattern surrounding the amorphous-dense core. Biochemical extraction revealed phosphatidylcholine as the major component of the polar lipids.

CONCLUSIONS

This preliminary study presents findings that suggest that phospholipids present in the TMJ may provide an efficient boundary lubrication that enables the disc to slide down the slope of the eminence on joint function.

Phospholipid composition of articular cartilage boundary lubricant

The mechanism of lubrication in normal human joints depends on loading and velocity conditions. Boundary lubrication, a mechanism in which layers of molecules separate opposing surfaces, occurs under severe loading. This study was aimed at characterizing the phospholipid composition of the adsorbed molecular layer on the surface of normal cartilage that performs as a boundary lubricant. The different types of phospholipid adsorbed onto the surface of cartilage were isolated by extraction and identified by chromatography on silica gel paper and mass spectroscopy. The main phospholipid classes identified were quantified by a phosphate assay. Gas chromatography and electrospray ionization mass spectrometry were used to further characterize the fatty acyl chains in each major phospholipid component and to identify the molecular species present. Phosphatidylcholine (41%), phosphatidylethanolamine (27%) and sphingomyelin (32%) were the major components of the lipid layer on the normal cartilage surface. For each lipid type, a mixture of fatty acids was detected, with a higher percentage of unsaturated species compared to saturated species. The most abundant fatty acid observed with all three lipid types was oleic acid (C18:1). Additional work to further quantify the molecular species using electrospray ionization mass spectrometry is recommended.

The contribution of synovial fluid lipoproteins to the chronic synovitis of rheumatoid arthritis

Lipids in the synovial fluid of patients with active rheumatoid arthritis are elevated compared to normal synovial fluid and that of other inflammatory arthropathies. Various assumptions about the role of these lipids have been made. This study offers evidence that these lipids may contribute to the synovitis in rheumatoid arthritis through participation in the arachidonic pathway within the joint space. Phospholipase A2 activity, phospholipids, prostaglandin E2, and leukotriene B4 have been correlated in the synovial fluid and plasma of untreated rheumatoid patients and compared with that of patients with osteoarthritis.

Role of protein kinase C in the phosphatidylserine-induced inhibition of DNA synthesis in blood mononuclear cells

The mechanism of immunosuppressant activity of phosphatidylserine has been studied in peripheral blood mononuclear cells depleted or not of monocytes. After the addition of phosphatidylserine, mass determinations and uptake of labeled compound demonstrate its transfer into the cells. Phosphatidylserine incorporation causes a 2.5-fold increase of membrane-bound protein kinase C activity. The activation of translocated enzyme is indicated by the inhibition of phosphoinositide hydrolysis, and early feedback effect induced by activated protein kinase C. This action of phosphatidylserine is reproduced by tetradecanoylphorbolacetate and is prevented by the protein kinase C inhibitor, staurosporine. Consistently, phosphatidylserine (8 nmol/10(6) cells) decreases by 46% the production of inositol phosphates in cells responding to phytohemagglutinin. The decrease of phosphoinositide signal pathway as well as the inhibition of mitogen-induced DNA synthesis are produced at the same phosphatidylserine concentration and are equally manifest in total mononuclear cells or in preparations depleted of monocytes. However, only in the presence of monocytes does tetradecanoylphorbolacetate enhance the action of phospholipid, decreasing its IC50 from 13-15 microM to 7 microM. Thus, the data suggest that a reaction driven by protein kinase-C and a factor released by activated monocytes are involved in the phosphatidylserine-induced inhibition of lymphocyte DNA synthesis.

Inhibition of DNA synthesis in peripheral blood mononuclear cells treated with phosphatidylserines containing unsaturated acyl chains

The immunosuppressive action of phosphatidylserine has been studied in mitogen-activated human peripheral blood mononuclear cells. The addition of phospholipid (10-60 nmol/10(6) cells) causes a dose-dependent inhibition of DNA synthesis induced by PHA, anti-CD3 mAb, allogeneic lymphocytes and tetradecanoylphorbol acetate plus ionomycin. In contrast, the interleukin-2-dependent DNA synthesis is less affected. Flow cytometric analysis and binding of radioiodinated interleukin-2 show that the phospholipid prevents the expression of interleukin-2 and transferrin receptors. Removal of monocytes by adherence does not change the action of phosphatidylserine. Furthermore, the phospholipid is equally effective in preparations depleted of CD4+ or CD8+ lymphocytes. Phosphatidylinositol partly reproduces the action of phosphatidylserine. Phosphatidic acid, phosphatidylglycerol and phosphatidylcholine are inactive. Also unsaturated phosphatidylserine analogues inhibit DNA synthesis whereas saturated phosphatidylserines do not. The data suggest that phosphatidylserine mainly affect the steps of T cell activation preceding the production of interleukin-2 and the expression of its receptor. The phosphorylserine headgroup and the unsaturated acyl chains contribute to this effect.

Phospholipid metabolism in rat intestinal mucosa after oral administration of lysophospholipids

The present results indicate that PS, a phospholipid contained in small amounts in the human diet and not included in plasma lipoproteins, may be used to influence phospholipid metabolism in intestinal mucosal cells. Since PS influx into absorptive cells occurs after its hydrolysis to lysoPS, this metabolite may be used to increase the absorption of this phospholipid. These data show that lysoPS, after diffusion into intestinal cells, is sequentially converted into PS and PE (which make up a minor fraction of the lipids present in lipoproteins). As expected, lysoPS given together with radiolabeled unsaturated fatty acids was unable to promote their transfer into plasma lipoproteins. In this respect lysoPS differed from lysoPC, the latter increasing the appearance of dietary fatty acids in plasma. When given together, lysoPS decreased the lysoPC-induced transfer of unsaturated fatty acids into plasma. This effect required addition of triglycerides to the lipid mixture. In attempting to explain this triacylglycerol-dependent inhibition by lysoPS, we found that this phospholipid increased the incorporation of glycerol into mucosal cell PC. In contrast, lysoPC was inhibitory. Furthermore, in the presence of labeled inositol, lysoPC (but not lysoPS) promoted the appearance of labeled phosphatidylinositol. The data thus suggest that the two lysophospholipids differ in promoting the two main pathways of PC synthesis in the intestinal cells. While lysoPC favors PC synthesis by reacylation, lysoPS enhances the CDP-choline pathway of PC synthesis.

pH-dependent lipid packing, membrane permeability and fusion in phosphatidylcholine vesicles

We have studied the rate of membrane fusion, the lipid dynamics and order and the membrane permeability of phosphatidylcholine vesicles as a function of pH. Acidification induced very different effects depending on the state of the bilayer. In liquid-crystalline bilayers, acidification decreased the rate of membrane fusion, the acyl chain motion and disorder and the rate of K+ release, whereas in solid bilayers acidification increased the rate of membrane fusion, the lipid acyl chain disorder and the rate of K+ release. These pH-dependent modifications are interpreted in terms of conformational and/or packing changes of the phosphatidylcholine head group in the membrane. In solid bilayers, these changes are not easily accommodated by the rigid structure, and the resulting stress leads to an unstable bilayer.

Pharmacokinetic characterization of phosphatidylserine liposomes in the rat

1. The plasma decay, tissue uptake and biotransformation of radiolabelled phosphatidylserine (PS) liposomes have been investigated in rats following bolus i.v. injection (2 mg kg-1). 2. PS plasma concentration showed a biexponential decay with half-lives of 0.85 and 40 min. The following interpretation of the biphasic decay is proposed: (1) The rapid initial decline is due to the irreversible uptake of PS liposomes by the mononuclear phagocyte system, as demonstrated by the almost exclusive accumulation of PS in liver and spleen. (2) The slow decay phase reflects the elimination of that fraction of PS that has been incorporated into high density plasma lipoproteins (HDL). A kinetic model has been developed to describe these phenomena and a good agreement has been observed between experimental data and theoretical values. 3. Evidence has been obtained that a large fraction of PS is hydrolyzed at the injection site, probably by phospholipase A2 and other hydrolytic enzymes released by platelets. Hydrolysis at the injection site has also been observed following intraperitoneal and intramuscular injections. 4. As shown by the comparative analysis of the biotransformation products found in tissues after administration of either [3H]-glycerol-PS or [14C]-serine-PS, parenterally administered PS follows two distinct metabolic pathways: (1) decarboxylation to phosphatidylethanolamine and (2) extensive hydrolytic degradation with release of the individual components of the molecule. These pathways probably reflect the two main mechanisms of PS uptake, incorporation into the plasma membrane and internalization by endocytosis, respectively.

Internalization of phosphatidylserine by adherent and non-adherent rat mononuclear cells

Energy-dependent, protein-mediated incorporation of radiolabeled phosphatidylserine vesicles is observed in casein-elicited rat peritoneal cells. Cell fractionation and a comparison with other phospholipids demonstrate the selective interaction of phosphatidylserine with the mononuclear fraction of these cells. During 60 min of incubation, unchanged phosphatidylserine accumulates in the cells whereas lysophosphatidylserine is released in the medium. When adherence is used to fractionate the mononuclear cells, phosphatidylserine uptake is detected in the macrophage-enriched fraction (adherent cells) and in the lymphocyte-enriched fraction (non-adherent cells). Evidence of stereoselective uptake and of phosphatidylserine internalization in both cells is obtained by the use of phosphatidyl-D-serine and by digestion of the extracellular phospholipid with phospholipase A2. Only in lymphocytes is the uptake of phospholipid substantially inhibited by cytochalasin B, metabolic poisons and a low incubation temperature (17 degrees C). Phosphatidylserine deacylation-reacylation is instead detected in both cells. It is concluded that lymphocytes actively concur in the uptake of phosphatidylserine by rat mononuclear cells.

Characterisation of soluble and cell associated phospholipase A2 from rheumatoid synovial fluid

The hydrolysis of radiolabelled Escherichia coli phospholipids, and micellar dispersions of phosphatidylethanolamine and phosphatidylcholine, were used to characterise the phospholipase A2 activity in synovial fluid from patients with rheumatoid arthritis. Cell free fractions of synovial fluid contain a phospholipase A2 enzyme that preferentially releases [14C]oleic acid from E coli biomembranes (specific activity 291.3 (SEM 27.6) pmol/min/mg). This enzyme requires calcium and is optimally active at neutral pH. Purified dispersions of phosphatidylethanolamine are also readily degraded by the soluble enzyme, but it is not active against phosphatidylcholine. The substitution of [14C]oleic acid by [3H]arachidonic acid for the labelling of E coli allowed differentiation between the soluble phospholipase A2 and the cell associated phospholipase A2 present in sonicates of mononuclear cells and neutrophils from peripheral blood and synovial fluid. The cell associated phospholipase A2 preferentially releases [3H]arachidonic acid from E coli cardiolipin. In this paper the phospholipid substrate specificity of phospholipase A2 from rheumatoid synovial fluid, the optimal assay conditions for its detection, and a standardised expression of activity in terms of pmol per minute per mg of protein are established.

Preferential hydrolysis of oxidized phospholipids by peritoneal fluid of rats treated with casein

1-Palmitoyl-2-azelaoyl-PC, which is one of the possible cytotoxic products generated by the oxyhemoglobin-induced lipid peroxidation of 1-palmitoyl-2-linoleoyl-PC, was found to be efficiently hydrolyzed by the peritoneal fluid of rats treated with casein. The rate of hydrolysis of 1-palmitoyl-2-azelaoyl-PC was approx. 15-fold higher than that observed with 1-palmitoyl-2-linoleoyl-PC. When 1-palmitoyl-2-linoleoyl-PC pretreated with oxyhemoglobin was incubated with the peritoneal fluid, oxidized products of PC were hydrolyzed more efficiently than the intact 1-palmitoyl-2-linoleoyl-PC. When 1-[(1-)14C]palmitoyl-2-azelaoyl-PC was incubated with the peritoneal fluid, radiolabeled lysoPC was formed, whereas radiolabeled neutral lipids were not formed, indicating that the hydrolytic activity was of the 'phospholipase A2' type. We previously found and purified an extracellular phospholipase A2 (Chang, H.W. et al. (1987) J. Biochem. 102, 147-154) in the peritoneal fluid of rats injected intraperitoneally with casein. Hydrolysis of 1-palmitoyl-2-azelaoyl-PC by this purified phospholipase A2 was as low as that of 1-palmitoyl-2-linoleoyl-PC. These two phospholipase A2 activities showed different pH optima and Ca2+ requirements. The present phospholipase A2 activity, which preferentially hydrolyzes oxidized products of PC, may play an important role in detoxification or repair of damaged membrane in inflamed sites.

Lysophosphatidylserine-dependent interaction between rat leukocytes and mast cells

The production of lysophosphatidylserine has been studied in a population of rat peritoneal cells; 67% polymorphonuclear and 33% mononuclear leukocytes. Pulse-chase experiments with L-[U-14C]serine reveal a net lysophosphatidylserine production of 0.33 nmol/mg protein in 2 h of incubation. The source of lysophosphatidylserine is probably the phosphatidylserine of cells damaged during the incubation, since plasma membrane fragments obtained from the leukocytes yield higher lysophosphatidylserine production (1.9 nmol/mg protein in 1 h of incubation). Both leukocytes and plasma membranes show phosphatidylserine splitting activity when tested with vesicles of this phospholipid. In the presence of albumin a fraction of produced lysophosphatidylserine is recovered in the incubation medium. Under these conditions efficient incorporation of lysoderivative into surrounding leukocytes and conversion to phosphatidylserine requires cell activation by tetradecanoylphorbol acetate. In agreement with radiochemical data it is found that a suspension of leukocytes elicits histamine release when rat peritoneal mast cells and nerve growth factor are subsequently added. This typical, lysophosphatidylserine-dependent mast cell response is retained when leukocyte plasma membranes substitute the whole cells. These results suggest that leukocyte lysis at sites of tissue injury results in the production of a sufficient amount of lysophosphatidylserine to reach and activate surrounding mast cells.

Effect of lysophosphatidylserine on immunological histamine release

The effect of lysophosphatidylserine on immunological histamine release has been studied in rat peritoneal mast cells actively sensitized with horse serum and in human basophils challenged with anti-IgE. In contrast to other lysophospholipids, lysophosphatidylserine enhances the immunological histamine release in rat mast cells. The effect shows the kinetics of a saturable process with an apparent Km for lysophosphatidylserine of 0.26 microM. A similar Km value (0.21 microM) is found when measuring the non-immunological histamine release activated by lysophosphatidylserine plus nerve growth factor. A comparison with phosphatidylserine shows that a half-maximal response to lysophosphatidylserine occurs at a concentration 4-times lower. In addition, the magnitude of the response is higher. At variance with rat mast cells, lysophosphatidylserine does not influence the histamine release elicited by immunological and non-immunological stimuli in human basophils. The histamine secretion in these cells is instead affected by a calcium ionophore or tetradecanoylphorbolacetate, a compound producing activation of protein kinase C.