Product inhibition of secreted phospholipase A2 may explain lysophosphatidylcholines' unexpected therapeutic properties
© Cunningham et al. 2008
Received: 20 June 2008
Accepted: 22 October 2008
Published: 22 October 2008
Skip to main content
© Cunningham et al. 2008
Received: 20 June 2008
Accepted: 22 October 2008
Published: 22 October 2008
Lysophosphatidylcholines (lysoPCs) are products of phospholipase A2 (PLA2) enzyme activity, and like the enzyme, have a direct role in toxic inflammatory responses in variety of organ systems. Paradoxically, reduced plasma lysoPC levels have been noted in sepsis patients and systemic treatment with lysoPCs is therapeutic in rodent models of sepsis and ischemia. These observations suggest that elevation of plasma levels of these lipids can actually help to relieve serious inflammatory conditions. We demonstrate that specific lysoPCs act as uncompetitive product inhibitors of plasma secreted PLA2 enzymes (sPLA2s), especially under conditions of elevated enzyme activity, thus providing a feedback mechanism for the observed anti-inflammatory effects of these compounds.
Thin layer chromatography and mass spectroscopy were used to estimate total lysoPC concentration and the relative contributions of different lysoPC species in rat plasma samples. Kinetic studies of sPLA2 enzyme activity were conducted on these samples ex vivo and on purified group IA sPLA2 in vitro after addition of specific lysoPC species to the reaction mixture. Enzyme activity was also measured in plasma samples of rats injected with these same lysoPCs.
Palmitoyl (16:0), stearoyl (18:0) are the most abundant lysoPCs in rat plasma consistent with other reports. Kinetic studies demonstrated that both were uncompetitive inhibitors of plasma sPLA2 enzyme activity. In vitro experiments with group IA sPLA2 confirmed the inhibition and the kinetic properties of these lysoPC species. Decanoyl lysoPC (10:0), which was not detected in plasma, did not inhibit enzyme activity in vitro. LysoPC injections into normal rats resulted in "buffering" of plasma sPLA2 activity in a narrow low range, consistent with the activity-dependent inhibition suggested by the ex vivo and in vitro experiments.
The results may explain the efficacy of lysoPC therapy during periods of elevated inflammatory activity and further highlight the utility uncompetitive enzyme inhibitors. In this case, the inhibitor is a product of the enzyme reaction, and therefore represents an example of activity-driven feedback inhibition.
Upregulation of circulating phospholipase A2 enzymes, principally the secreted isoforms (sPLA2s), is associated with the activity of the innate immune system and a number of inflammatory disorders [1–4]. Experimental and correlative studies suggest increased levels of sPLA2s in the blood contribute to or are predictive of the tissue destruction that occurs following trauma, heart and lung disease, local and systemic infections, brain damage, and autoimmune disorders [4–9]. PLA2 enzyme activity and the lipid mediators regulated by that activity are directly linked to a variety of cell death effector mechanisms including the production of reactive oxygen species (both directly and via inflammatory cells), excitotoxicity, and expression of the death receptor family members in a variety of cells [10–12]. Many of these activities have been studied relative to the fatty acid branch of PLA2 pathways, which includes arachidonic acid and its metabolites (prostanoids, leukotrienes). The other products of PLA2 hydrolysis, the lysophosphatides, are also potent biological mediators but their mechanisms are more enigmatic and their actions are often contradictory. There is current interest in lysophosphatidylcholines (lysoPCs) in particular, as some of these are proposed for treatment of systemic inflammatory disorders. This suggestion is based on reports that plasma lysoPC levels are diminished with the onset of sepsis in human patients [15, 16], and in rodent models of sepsis and ischemia, lysoPC treatment is an effective therapy [17–21].
In previous studies, we investigated sPLA2 enzyme activity in plasma of rats and humans because inhibition of this group of enzymes, and subsequent tissue repair and protection, is a long-standing goal of pharmacotherapeutics . Our studies were expedited by the fact that plasma sPLA2 enzymes will follow Michaelis-Menten kinetics when incubated with a broad-spectrum substrate, even though the specific classes of sPLA2s active in plasma are diverse. This property made it possible to characterize a peptide inhibitor of sPLA2 activity (called CHEC-9), also with broad-spectrum activity, as well as demonstrate the therapeutic potential of sPLA2 inhibition in vivo. CHEC-9 treatment of both traumatic and autoimmune models of neurodegeneration resulted in significant cytoprotection and reduction of cell-mediated inflammation [5, 6, 23]. These results have been supported by other experimental studies both in and outside of the nervous system. For example, infusion of sPLA2 substrate-like compounds reduces circulating sPLA2 activity and also has both neuroprotective and anti-inflammatory effects . Conversely, transgenic models that express high systemic levels of group II or group V sPLA2s demonstrate pro-atherogenic pathologies and exaggerated lung disease[8, 25, 26].
Given the therapeutic advantages of sPLA2 inhibition, we considered the possibility that under certain conditions the lysoPC product of sPLA2 activity actually inhibits the enzyme, thus explaining one aspect of the paradoxical behaviour of these lipids. Product inhibition is a widely recognized phenomenon in enzymology , but one that has received scattered experimental attention with respect to regulation of enzyme activity in mammalian systems in vivo [28, 29]. As for sPLA2, product inhibition of phosphatidylcholine hydrolysis was suggested to occur in vitro during experiments with venomous group IA sPLA2 [30, 31]. However there were different conclusions in these studies as to the source of this inhibition. Furthermore, serum albumin binding to lysoPC was suggested to neutralize its inhibitory activity, a property that might limit the applicability of the results to systemic therapies given the sheer abundance of albumin in plasma. In the present paper, we first documented the principal lysoPC species in rat plasma relative to each other and to other abundant plasma lipids using thin layer chromatography and mass spectroscopy. Then the effects of exogenous lysoPCs were studied with plasma sPLA2 activity ex vivo and with group IA sPLA2 in vitro. Finally, lysoPCs were injected into rats to determine the influence of elevating systemic levels of these compounds on circulating sPLA2 activity. The results suggest that the two most abundant lysoPC species found in plasma are activity-dependent uncompetitive inhibitors of sPLA2 enzyme activity, even in the presence of albumin.
Blood was obtained from the trunk of 32 female Sprague Dawley rats (200–250 g) after decapitation. Blood samples were treated with citrate-phosphate-dextrose anticoagulant (1:10, Sigma), and plasma prepared by centrifugation after which samples were frozen at -80° until used. For 16 of these animals, plasma samples were pooled from 4–6 rats for ex vivo enzyme assays. The other 16 rats received lysoPC treatments (see below). The Institutional Animal Care and Use Committee of Drexel University College of Medicine approved all specific procedures of this study. Purified group IA secreted phospholipase A2 from the venom of the Mozambique spitting cobra (Naja mossambica) was obtained from Sigma (St. Louis, MO).
Samples from the Folch-extracted lipid were dried and re-dissolved in chloroform-methanol-300 mM ammonium acetate in water (60:133:7, v/v/v). Lysophosphatidylcholines were analyzed by electrospray ionization triple quadrupole mass spectrometry at the Kansas Lipidomics Research Center, as previously described.
Enzyme assays were conducted as in previous studies  using a Victor 3 fluorescent reader (Perkin Elmer, Waltham, MA). The substrate was 1-palmitoyl-2-pyrenedecanoyl phosphatidylcholine ("10-pyrene"), Caymen Chemical, Ann Arbor MI) a substrate for all calcium dependent PLA2s with the exception of cPLA2 and PAF-AH (LP-PLA2). Substrate solutions were prepared in reaction buffer consisting of 50 mM tris (pH = 7.4), 0.1 M NaCl, 2 mM CaCl2, 0.25% fatty acid-free albumin. The substrate forms phospholipids vesicles in this solution , and upon hydrolysis, releases fluorescent 10-pyrenyldecanoic acid (PDA). Plasma samples were 10% final and mixed with lysoPCs by gentle shaking for 100 sec, or they were derived from lysoPC-injected rats. All enzyme reactions were initiated with the addition of the substrate solution to the sPLA2 containing samples. Kinetic parameters including the properties of lysoPC inhibitors were determined by recording the initial maximum velocities (Vo) of enzyme reactions, obtained within 2 minutes of initiation. A detailed presentation of the calculations used in our kinetic analysis has been presented . Relative fluorescent units (RFU) were converted to product concentration using a standard curve made with pyrenyldecanoic acid (Invitrogen, Eugene, OR). Individual reactions were carried out in duplicate or triplicate and kinetic curves were produced using at least 6 substrate concentrations, with lysoPC or its vehicle (PBS with 8% ethanol). Representative Lineweaver-Burk and Michaelis-Menten plots as well as nonlinear regression analyses are presented in Results. Experiments were repeated at least 4 times with similar results. Statistical analyses were by the Mann-Whitney test or Welch test for the in vivo data since control data for the latter was Gaussian with a variance that was significantly different from the experimental samples. These calculations, and Km, Vmax and R2 were made with statistical (INSTAT) and regression software (Prism), both from Graphpad (San Diego, CA).
Rats were injected subcutaneously under the loose skin at the shoulders with synthetic stearoyl (n = 4) or palmitoyl (n = 4) lysoPCs (25 mg/kg) two hrs before sacrifice and compared to vehicle-injected rats (n = 8). The relative concentrations of active sPLA2 enzymes were measured in plasma samples of these animals by determining the total hydrolysis of substrate in 1 hr.
The two-dimensional thin layer chromatography procedure we adapted was developed to reveal several of the most abundant phospholipid and sphingolipid species on a single TLC plate  (Fig. 1A). Phosphatidylcholine, sphingomyelin and lysophosphatidylcholine, were the major phospholipids appearing on the plates with plasma samples. Ceramides, cholesterol, free fatty acids, and phosphatidylinositol also appeared in significant quantities while lower levels of phosphatidylethanolamine and galactoceramide were found with this procedure. Therefore, these plates showed that lysoPCs were a significant proportion of the lipid component of blood plasma in rodents and humans, as reported by others [16, 38–43]. However, published reports or calculations made from published reports provide a fairly wide range of estimates of the total lysoPC concentration in plasma (145 to 270 μM). Our estimates of total lysoPC, including from standard curves generated on the TLC plates (Fig. 1B), were 207.5 ± 44 μM (TLC, n = 3) and 196 ± 64 μM (mass spectroscopy, n = 3). Mass spectroscopy analysis also showed that palmitoyl (16:0) and stearoyl (18:0), calculated as mol percent of total lysoPC present, were the most abundant species in plasma, followed by linoleoyl (18:2), and oleoyl (18:1) lysoPCs, in general agreement with the previous studies cited above, and similar for rats and humans (Figs. 1C, D).
Kinetic data for lysoPC inhibition of sPLA2 activity in plasma
Ki ± s.e.m. (μM)
25.9 ± 4.0
34.2 ± 17.4
Kinetic data for lysoPC inhibition of group IA sPLA2
Ki ± s.e.m. (μM)
Albumin binding by the lysoPCs may influence some of their biological activities (see Discussion), but of special interest in the present study was the possibility that albumin also neutralizes group IA sPLA2 product inhibition in vitro [30, 31]. In the present ex vivo experiments with plasma, significant inhibition sPLA2 activity was observed even though albumin is major constituent of plasma (approximately 3% or ~400 μM) . Since one albumin molecule can bind two lysoPC molecules , and given the estimated concentrations of lysoPCs in plasma (see above), it appears that all lysoPC molecules with a propensity for albumin binding could be bound. It is suggested therefore that albumin does not influence lysoPC inhibitory activity at least under physiological conditions. We observed attenuated inhibition of group IA sPLA2 with albumin, but only after preincubation of 40 μM lysoPC 18:0 with a large excess of this protein (5% which is > 600 μM). In these cases, the percent inhibition (i.e., the percent reduction in Vo) was 29.3 ± 1.6% (n = 4) compared to 63 ± 6% (n = 8) with standard buffer (p < 0.01). This difference was significant (p = 0.029).
The principal finding of this study is that the two most abundant lysoPC species in plasma are capable of inhibition of plasma sPLA2 activity. We documented this inhibition ex vivo and in vivo with plasma samples from rats after treatment with palmitoyl or stearoyl lysoPC species. The fate of these exogenous lysoPCs after they were added to the plasma or injected into the animals is uncertain. It is likely that they ultimately reside in one or more of the functionally available pools of lysoPC, either free, micellar, bound to LDL or serum proteins such as albumin or immunoglobulin (anti-phospholipid-immune complexes). The relative abundance of these forms and of the total lysoPC concentration in plasma is expected to be highly regulated, including by the balanced activity of sPLA2s and lysophosphatidylcholine acyltransferases , enzymes that cleave and attach the A2 fatty acid. Despite these uncertainties as to the fate of the exogenous lysoPCs, it appears that a 10–20% increase in their total molar concentration in plasma is sufficient to inhibit sPLA2 activity ex vivo. In addition, the fact that we used a general sPLA2 substrate further suggests these lysoPCs have the potential to inhibit several of the sPLA2 isoforms found in plasma [46, 47].
The pro-inflammatory effects of lysoPCs have been well documented, both as exogenous agents of toxicity and as products of sPLA2 hydrolysis. For example, the contribution of lysoPCs to atherosclerotic lesions is a prime example of ability of sPLA2 enzyme activity to direct a local injurious inflammatory response. LysoPC has been considered a principal agent of this activity [52, 53]. In the nervous system, exogenous lysoPCs are used in demyelination studies and as a trigger for responses of the innate immune system in order to model neuroinflammatory disorders . Glutamate toxicity after cerebral ischemia is exaggerated by local infusion of lysoPC , an effect attributed to the "detergent action" of the lipid . In the plasma however, where most lysoPCs are bound to albumin [39, 43, 57], it is assumed that endogenous lysoPCs are much less toxic, perhaps because of this binding . Our data indicate that at physiological concentrations, albumin does not effect lysoPC inhibition of plasma sPLA2 activity ex vivo or in vivo. Considering these results and the model presented above, it is can be suggested that: 1) albumin binding does not effect the binding to PC containing particles and subsequently the ability of the lysoPC to inhibit enzyme activity, or that 2), in some situations the affinity of the lysoPC for these particles is stronger than it is for albumin. The latter condition has been suggested to occur for oxidized low density lipoprotein (ox-LDL) particles both normally and during certain pathologies [57, 59]. One interesting possibility is that the affinity of the PC-containing particles for lysoPCs is regulated by sPLA2 activity; for example, as activity is increased, the concentration of desorbed lysoPCs also increases, making the particles more receptive to incorporating the lysoPC product, which in turn limits enzymatic activity. The present results suggest that exogenous lysoPCs also increase this affinity, thereby accomplishing sPLA2 inhibition and cytoprotection. In this circumstance however, the potential toxicity of excess lysoPCs cannot be ignored, so there is likely an upper limit of lysoPC dosages that will accomplish the inhibition without cytotoxicity. Dose limitations of synthetic lysoPCs have been demonstrated for treatment of rodent endotoxaemia . Along these same lines, the bactericidal properties of sPLA2s suggest a similar constraint exists regarding the extent of enzyme inhibition tolerated in models of fulminate bacterial infections .
The present results may complement recent studies that explore other mechanisms for the anti-inflammatory activities of lysoPCs. For example, the experiments of Chen, et al  showed the specific influence of LysoPC 18:0 on pro-inflammatory HMGB1 release from monocytes and macrophages. Although this effect appears partially mediated by the G2A receptor, we suggest that the well-known relationship between sPLA2 activity and macrophage activity is also relevant, especially because of the late stage at which the lysoPC was administered [2, 45, 61, 62]. It is expected that during this late period, the cascading inflammatory response would make conditions ideal for uncompetitive product inhibition of sPLA2 and resulting attenuation of monocyte/macrophage functions. Likewise, there is substantial cross talk between sPLA2 enzymes and the activity of pro-inflammatory cytokines that have been implicated in sepsis models . The suggestion that lysoPC 18:0 also enhances microbial elimination during sepsis induced by cecal ligation and puncture  might reflect direct effects of the lysoPCs on bacterial sPLA2 enzymes, since bacterial sPLA2s with remarkable similarities to the mammalian enzymes have now been characterized . Finally, the diminished LysoPC/PC (product/substrate) ratio found in sepsis patients that are more likely to recover  is entirely consistent with inhibition of sPLA2 activity described here. The inhibition might slow the inflammatory cascade and increase the probability of recovery.
The diverse, sometimes contradictory properties of LysoPCs suggest that the different functional forms of this class of lipids are highly regulated in the plasma. Under conditions of severe inflammatory stress and subsequent elevation of plasma sPLA2 activity, systemic lysoPCs will help to shift the balance towards sPLA2 inhibition and cytoprotection. LysoPC therapies are therefore possible if these lipids are introduced systemically at the appropriate levels and at the appropriate point in the inflammatory cascade.
We would like to thank Mary Roth and Ruth Welti of the Kansas Lipidomics Research Center for mass spectrometric analysis of the lysophosphatidylcholines and for reading the manuscript. The Kansas Lipidomics Research Center was supported by National Science Foundation (EPS 0236913, MCB 0455318, DBI 0521587), Kansas Technology Enterprise Corporation, K-IDeA Networks of Biomedical Research Excellence (INBRE) of National Institute of Health (P20RR16475), and Kansas State University. This work was also supported by grants from the Craig Neilsen Foundation and the Amyotrophic Lateral Sclerosis Association to TJC.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.