Phosphatidylcholine formation by LPCAT1 is regulated by Ca2+ and the redox status of the cell
© Soupene and Kuypers; licensee BioMed Central Ltd. 2012
Received: 12 March 2012
Accepted: 25 May 2012
Published: 7 June 2012
Skip to main content
© Soupene and Kuypers; licensee BioMed Central Ltd. 2012
Received: 12 March 2012
Accepted: 25 May 2012
Published: 7 June 2012
Unsaturated fatty acids are susceptible to oxidation and damaged chains are removed from glycerophospholipids by phospholipase A2. De-acylated lipids are then re-acylated by lysophospholipid acyltransferase enzymes such as LPCAT1 which catalyses the formation of phosphatidylcholine (PC) from lysoPC and long-chain acyl-CoA.
Activity of LPCAT1 is inhibited by Ca2+, and a Ca2+-binding motif of the EF-hand type, EFh-1, was identified in the carboxyl-terminal domain of the protein. The residues Asp-392 and Glu-403 define the loop of the hairpin structure formed by EFh-1. Substitution of D392 and E403 to alanine rendered an enzyme insensitive to Ca2+, which established that Ca2+ binding to that region negatively regulates the activity of the acyltransferase amino-terminal domain. Residue Cys-211 of the conserved motif III is not essential for catalysis and not sufficient for sensitivity to treatment by sulfhydryl-modifier agents. Among the several active cysteine-substitution mutants of LPCAT1 generated, we identified one to be resistant to treatment by sulfhydryl-alkylating and sulfhydryl-oxidizer agents.
Mutant forms of LPCAT1 that are not inhibited by Ca2+ and sulfhydryl-alkylating and –oxidizing agents will provide a better understanding of the physiological function of a mechanism that places the formation of PC, and the disposal of the bioactive species lysoPC, under the control of the redox status and Ca2+ concentration of the cell.
The oxygen carrying function of the red blood cell (RBC) leads to the generation of reactive oxygen species in the cell, and despite an intricate system of antioxidants, free radical damage of un-saturated acyl chains of glycerophospholipids occurs continuously. These oxidized acyl chains lead to a breach in normal membrane lipid organization and need to be replaced to maintain integrity of the membrane. The oxidized phospholipids (PL) are de-acylated by phospholipase A2 (PLA2) action [1–4], and re-acylation of the resulting lysophospholipid (lysoPL) is achieved by a two-step process. Fatty acids are activated to acyl-CoAs by membrane-bound long-chain acyl-CoA synthetases (ACSL) [5–7] and the acyl group of acyl-CoA is then transferred to lysoPL by acyl-CoA:lysoPL acyltransferase (LPLAT) enzymes [8–11]. This repair mechanism is also known as the Lands’ cycle [8, 12].
Phosphatidylcholine (PC) is the most abundant glycerophospholipid in membranes , with unsaturated acyl chains, mainly found at the sn-2 position. Repair of oxidized PC and re-acylation of the lysoPC in RBC proceeds rapidly by utilizing fatty acids that are taken up from plasma, and the action of ACSL and LPCAT in the plasma membrane . We previously identified ACSL6 as the acyl-CoA synthetase in the RBC membrane and LPCAT1 as the acyl-CoA:lysoPC acyltransferase [7, 15–17]. LPCAT1 is also the enzyme for the re-acylation of PC in alveolar type II cells [18, 19]. Furthermore, LPCAT1 might play an essential role in production of lipid surfactant in lung [20, 21], and in regulating the level of inflammatory lipids, such as lysoPAF and lysoPC, in the retina [22, 23]. LPCAT1 also appears to mediate O-palmitoylation of histone H4 in the nuclei of lung epithelial cell .
LPCAT1 does not require Ca2+ for activity [16, 19] and was reported as a Ca2+-independent member of the LPCAT family of enzymes [25–27]. Activity of LPCAT2 is regulated by Ca2+[16, 25, 28] and was defined as a Ca2+-dependent member of the LPCAT family [25–28]. However, two EF-hand motifs, folding into hairpin structure coordinating Ca2+[29–32], are predicted in both LPCAT1 and LPCAT2 [16, 18, 28]. Although, we have confirmed that Ca2+was not required for activity of LPCAT1, i.e. Ca2+-independent in , we have established that Ca2+was in fact inhibitory on the LPCAT1 activity . At the millimolar Ca2+ concentration values found in plasma , acylation rate of LPC by LPCAT1 was reduced and showed dependency to Ca2+ concentration . Thus, as it is the case for LPCAT2, Ca2+ also regulates the activity of LPCAT1. These observations led us to investigate the role of the predicted EF-hand motifs in Ca2+-binding.
LPCAT1 activity is also sensitive to treatment by sulfhydryl-modifier agents, such as the alkylating thiol reductant N-ethyl maleimide (NEM) . Cursory observation indicated that of the 12 cysteines of LPCAT1, Cys-211 found at the +1 position of motif III, 207PEGT210, could be conserved among acyltransferase forms that are sensitive to NEM and may be responsible for their sensitivity to this agent . This residue was also proposed to define the ‘motif 3-cysteine acytransferases’ sub-family of LPLAT enzyme and to be crucial for catalysis . However, the role of Cys-211 in the sensitivity to NEM and in catalysis was never tested since even the substitution of Cys-211 to the arginine residue present at the end of motif III of LPAAT enzymes , rendered an inactive C211R form . Similarly, substitution of Arg-181 of motif III of the human LPAAT enzyme AGPAT1 to several other residues rendered inactive forms .
We report that the EFh-1 motif of LPCAT1 is a functional Ca2+-binding site and that Cys-211 is not essential for activity of LPCAT1. Up to six cysteines residues, including Cys-211, are responsible for the decrease activity of the enzyme after treatment by NEM and diamide. The sensitivity of LPCAT1 activity to thiol damage and to Ca2+ binding to the EFh-1 site establishes that acylation of the most abundant phospholipids of the cell membranes is under the control of the redox status and Ca2+ concentration of the cell.
List of primers used for mutagenesis
Primer sequence (5′ to 3′)
saturated mutagenesis of
Cys-211 to Ser
Glu-403 to Ala
Cys-216 to Ala
Cys-263 to Ala
Cys-278 to Ala
Cys-314 to Ala
Cys-330 to Ala
Cys-401 to Ala
Cys-443 to Ala
Cys-477 to Ala
Cys-501 to Ala
Cys-514 to Ala
Cys-530 to Ala
The majority of the mutants displayed a very low activity (Figure 6A and Additional file 1: Figure S1). However in addition to the C211T form described above, five mutants missing 2 to 6 cysteines were obtained that showed acyltransferase activity ( Additional file 2: Table S1 and Figure 6). These 6 active proteins establish that Cys-211, 216, 314, 443, 501 and 514 are not essential to the acyltransferase reaction ( Additional file 2: Table S1). We could not identify a pattern and determine which cysteine, if any, was essential for activity. As observed for many of the substitutions tested with the Cys-211 residue and resulting in inactive mutant forms, alanine substitution of other non-essential cysteines could render mutant proteins with low activity due to structural alterations and defects. Among the active mutants, the form lacking all six cysteines mentioned above, C211T C(216,314,443,501,514)A, was the most active and had a higher activity rate than LPCAT1 ( Additional file 2: Table S1 and Figure 6). The other active forms displayed 50 to 150% of the activity of LPCAT1 (Figure 6). With one exception, all active forms were as sensitive to NEM and diamide treatment as was C211T (Figure 5B and Additional file 1: Figure S1). Activity of the mutant C211T C(216,314,443,501,514)A was not affected by treatment with either reagent (Figure 5). Even under condition resulting in 80 to 90% inhibition of the activity of LPCAT1, this mutant protein was still fully active. Moreover, it reproducibly displayed greater activity after treatment by diamide (Figure 5B, last bars).
To define the role of the 12 cysteines in the alkylation reaction, mutant forms embedded in the membrane, and presumably correctly folded, were exposed to MPB. Following treatment and quenching with excess amount of DTT, membrane proteins were dissolved in CHAPS and mutant forms were purified, as described in the Experimental section. The label was revealed with streptavidin and the different forms (unlabeled and labeled) were detected with an antibody reacting against the hexa-histidine tag. Unexpectedly, all 12 cysteines were alkylated and all mutant forms containing at least one cysteine residue were labeled by MBP (Figure 7B). Even the mutant enzyme C211T C(216,314,443,501,514)A, whose activity was not affected by treatment with NEM, was also labeled by MPB (Figure 7B, lane NEMr). This result established that alkylation of some of the 6 cysteines still present in C211T C(216,314,443,501,514)A were susceptible to alkylation and that their alteration had no effect on activity of the enzyme (Figure 5B, last bars). It cannot be ruled out that substitution of one, or more, cysteine resulted in a structural change and the exposure of thiol groups that were otherwise protected from labeling in the native form. However, the findings obtained with the C211T C(216,314,443,501,514)A enzyme established that modification of LPCAT1 by alkylating agents is due to attack of several cysteine residues that are not essential for catalysis.
In the calmodulin superfamily of Ca2+-binding proteins, repeats of EF-hand motifs are often present in tandem but few forms, such as the members of the S100 family, contain a single pair of sites . Member 1 and 2 of the LPCAT family have two EF-hand motifs and thus, belong to the S100 Ca2+-binding protein family [16, 18, 28]. In the helix-loop-helix structure of such calcium-binding sites, the loop coordinates the divalent cation to seven oxygen atoms of semi-conserved residues [30–32, 38]. There is a great variability of sequence but the first and last residues of the loop are almost always an aspartate and glutamate, respectively. The EFh-2 motif of LPCAT1 is lacking these two important residues . It was suggested that the EF-hand motifs of LPCAT1 could be structurally different with divergent functions as compared to those present in LPCAT2 . Our findings establish that activity of LPCAT1 is controlled by Ca2+ and that the EFh-hand motifs of LPCAT1 represent active Ca2+-binding sites, as in LPCAT2.
The residue found at the +1 position of motif III of LPLAT was thought to confer preference to different acceptor species (lysoPA, lysoPC or lysoCardiolipin) for acylation with acyl-CoAs. With the identification of many more members of the large LPLAT family, it now appears that presence of a cysteine, arginine or aspartate residue at that position does not define specificity to the substrate (Figure 1B). Nevertheless, these residues must play some role in catalysis since their substitution often rendered inactive forms. As shown previously by Shimizu, T and colleagues , we confirmed that Cys-211 of LPCAT1 could not be replaced by an arginine residue. An arginine is present at that position in the three characterized LPAAT enzymes (AGPAT1, AGPAT2 and AGPAT3) (Figure 1B). Substitution of Cys-211 to many other residues rendered inactive forms. Substitution of Arg-181 of AGPAT1 to an alanine or a lysine also rendered inactive enzymes . Thus, these residues are important for activity of LPLAT enzymes but activity of the C211T mutant of LPCAT1 established that contrary to previous report , Cys-211 is not essential for catalysis. In addition, removal of Cys-211 did reduce, but did not eliminate sensitivity to NEM. It also did not prevent labeling of the protein by MPB. These results demonstrate that Cys-211 is also not crucial for the sensitivity of LPCAT1 activity to alkylating agents. The C211T C(216,314,443,501,514)A enzyme was unaffected by treatment with diamide and NEM suggesting that a combination of several cysteines among Cys-211, Cys-216, Cys-314, Cys-443, Cys-501 and Cys-514 confer sensitivity of the acyltransferase reaction to these reagent but that none of these residues is essential for catalysis.
The RBC is exposed to high oxidant stress from the cytosolic side due to the hemoglobin-mediated transport of oxygen and is exposed to high concentrations of Ca2+ in plasma [33, 46, 47]. Both an increase in cytosolic calcium and oxidant stress lead to a loss of the membrane phospholipid asymmetry with the exposure of phosphatidylserine and removal of the damaged cells. The sensitivity of the de-acylation/re-acylation repair cycle suggests that the integrity of their membrane will also be compromised. In hemoglobinopathies, such as sickle cell disease and thalassemia, the presence of damaged RBCs in the circulation plays a significant aggravating role in the vasculopathy that characterizes these disorders. Thus, regulation of LPCAT1 activity might contribute to the removal of damaged-RBCs from the circulation and could represent a mechanism for the dismissal of stressed cells in other tissues.
[1-14C]C18:1-CoA (55.0 mCi/mmole) was from from Amersham Corp., (Arlington Heights, IL, U.S.A) and 1 acyl-lysoPC from Avanti Polar Lipids, Inc. (Alabaster, AL, U.S.A). TLC silica plates were obtained from Analtech. Inc. (Newark, DE, U.S.A). N-Ethylmaleimide (NEM) and diamide were from Sigma-Aldrich, Inc. (St. Louis, MO, USA). Na-(3-maleimidylpropionyl)biocytin (MPB) was from Invitrogen. All other compounds used were reagent grade.
Cloning of mouse LPCAT1 was previously reported . Full-length cDNA was cloned in pET28a vector (Novagen), with a unique in-frame hexahistidine tag at the N-terminus, to yield plasmid pFK192. Site-directed mutagenesis experiments were performed with the QuikChange Multi Site-directed Mutagenesis kit (Stratagene) according to the manufacturer instruction. Primers were designed with the QuikChange® Primer Design Program (Stratagene) and are listed in Table 1. The presence of the intended nucleotide change(s) and the absence of unwarranted mutations were verified by full-length sequencing of the constructs. Position of the amino acid residues is given relative to the full-length mouse LPCAT1 protein NP_663351 (534 amino acids).
Expression of (His)6-LPCAT1 was previously reported . Production of (His)6-LPCAT1 was obtained by growing E. coli strain, Rosetta™ 2(DE3) transformed with the different recombinant constructs in presence of 500 μM of IPTG for 3 hours. Cells were collected, disrupted and membrane fractions were obtained by centrifugation as previously described . Samples were stored at −80°C in Tris–HCl 0.2 M pH 7.4 with 10% glycerol. For analysis, protein samples were denatured for 20 min at 37°C in SDS-PAGE loading buffer and separated on SDS-PAGE 12% gel. Proteins were visualized by staining (GelCode Blue stain reagent, Pierce) and recombinant (His)6-LPCAT1 proteins were detected with a commercial HRP-conjugated anti-histidine antibody (INDIA-HisProbe-HRP antibody, Pierce), diluted a thousand fold. HRP detection was performed with SuperSignal West Pico Chemiluminescent kit (Thermo Fisher Scientific Inc., Rockford, IL).
Incorporation of 14C]acyl-CoA into egg lysoPC by recombinant LPCAT1 protein in E. coli membranes was determined as previously described . Reactions were performed in glass tubes at 37°C in a shaking water bath, in 200 μl of (Tris–HCl 20 mM, pH 7.4; Tween-20 0.8 mg/ml) containing 20 μM lysoPC and 5 μM 14C]acyl-CoA. Reactions were initiated by addition of 1 to 15 μg of membrane protein fractions and incubated from 0 to 8 min. Three to four time points, in triplicate, were used to determine the PC formation rate by LPCAT1 enyzmes. Preliminary experiments were performed to determine the correct amount of microsomes necessary to obtain a linear dependency of PC formation by the different mutant forms ( Additional file 2: Table S1). For forms with very low activity up to 15 μg of microsomes were used in each reaction. Control experiments were performed with membrane fractions obtained from E. coli strains transformed with the empty pET28a vector. Under our growth condition, no detectable E. coli acyl-CoA: 1-acyl-lysoPC acyltransferase activity was detected (first lane of Figure 4A and Figure 6A). Reactions were stopped by the addition of 200 μl of CHCl3:MeOH:12 N HCl (40:40:0.26, v/v) and vigorous vortexing. Phases were separated by centrifugation at 1,000 g for 5 min and the lipid-containing chloroform phase was dried down under N2, and dissolved by vortexing in 20 μl of CHCl3:MeOH (2:1, v/v). Samples were applied to TLC silica plates and developed with chloroform/methanol/acetic acid/0.9% NaCl (100:50:16:5, v/v). TLC plates were air-dried for 20 min and exposed to a PhosphoImager screen (Storm 840, Molecular Dynamics). Quantification of PC formation was performed with ImageQuant software subtracting the plate background.
To determine the effect of divalent cations on LPCAT1 activity the rate of [14C]PC formation was measured in absence or presence of 10 mM calcium chloride in the incubation mixture. Assays were performed in triplicate. The samples to be compared were applied and developed on the same plate. The relative activity rate is expressed as the ratio of the rates in the presence compared to the rate in the absence of the divalent cation. To determine the effect of N-ethylmaleimide and of diamide on LPCAT1 activity membrane fractions were incubated with the chemical for 30 minutes on ice in Tris–HCl 20 mM pH 7.4. The incubation in 10 μl, was followed by a 20 times dilution into the reaction mixture (200 μl). In some assays, membrane samples were first treated with 0.5 mM NEM or diamide and one half of the mixture was then incubated with 20 mM dithiothreitol (DTT) for 20 min at room temperature. Samples were then assayed for LPCAT activity as described above.
Cells were grown and proteins were induced as described above. Proteins were purified using His GraviTrap and His buffers kit of Amersham (GE Healthcare) in presence of 1% CHAPS. Frozen cell pellets obtained from 1 liter of culture were thawed on ice in 10 ml of breakage buffer (20 mM Sodium Phosphate pH 7.4, 0.5 M NaCl, 20 mM Imidazole, 1 mM PMSF and 10% glycerol). Cells were disrupted with a French-press cell at 12,000 psi. The resulting lysate was cleared by centrifugation at 16,000 g for 20 min at 4°C. ß-Mercaptoethanol was added to a final concentration of 5 mM. Membranes were collected by centrifugation at 100,000 g for 1 hr at 10°C. The pellet was slowly suspended in 1 ml of breakage buffer containing 2% CHAPS with a stir bar on a magnetic plate for about 30 min. One ml of breakage buffer was added, and 5 ml of breakage buffer containing 1% CHAPS was then mixed to obtain 7 ml of sample in BB with 1% CHAPS. It was loading on pre-washed and equilibrated Nickel column. The sample was passed 2 to 3 times on the column to improve binding of the protein. The column was washed with 20 volume of breakage buffer containing 1% CHAPS. Stepwise elution was performed with 3 ml BB containing 1% CHAPS and 50 mM, 200 mM and 500 mM imidazole. As shown on Additional file 3: Figure S2, LPCAT1 and mutant forms were detected in the third elution with few contaminant proteins.
Na-(3-maleimidylpropionyl) biocytin and all mixtures containing it were kept protected from light. MPB was dissolved in DMSO at 2 mM and kept at −20°C. Membrane fractions (25 μg) and purified proteins (2 μg) were labeled in 25 μl of 10 mMK-Phosphate pH 7.4 with 100 μM of MBP. Reactions were performed for 15 min at room temperature. Reactions were stopped by quenching un-reacted MBP with 25 mM DTT for 5 min at room temperature. For purified proteins, samples were mixed with SDS-PAGE loading buffer, denatured and separated on 12% gel SDS-PAGE. For membrane fractions, the (His)6-proteins were extracted and purified before analysis on SDS-PAGE and detection of the MPB label. After quenching by DTT, membranes in 25 μl were solubilized in 200 μl of buffer B (urea 7 M, NaH2PO4 100 mM and Tris–HCl 20 mM pH8.0) for 20 min at room temperature. Samples were cleared by centrifugation at 14,000 g for 5 min. The supernatant was mixed with 50 μl of washed Ni-NTA slurry 50% and incubated for 10 min. The resin was then washed twice with buffer C (urea 8 M, NaH2PO4 100 mM and Tris–HCl 20 mM pH6.3). The (His)6-protein bound to the resin was then eluted in 50 μl buffer E (urea 8 M, NaH2PO4 100 mM and Tris–HCl 20 mM pH4.5). This extraction procedure was efficient and very little (His)6-proteins were lost in the insoluble pellet or in the washes. Eluted samples were then mixed with SDS-PAGE loading buffer and separated on 12% gel SDS-PAGE. When necessary, Tris–HCl 2 M pH 8.0 was used to increase the pH of the eluted sample before loading. After electrophoresis, proteins were transferred onto a PVDF membrane. The blot was blocked with 1% BSA (fraction V) in TBS with Tween-20 0.1% for 60 min at room temperature. Hexahistidine tag of the recombinants proteins was detected with an HRP-conjugated anti-histidine antibody at a 1/3,000 dilution or the MPB-label was revealed by incubation with an HRP-conjugated streptavidin ligand at a 1/6,000 dilution in TBS with Tween-20 0.1% for 2 to 4 hr at room temperature. HRP detection was performed with SuperSignal West Pico Chemiluminescent kit (Thermo Fisher Scientific Inc., Rockford, IL).
Long-chain acyl-CoA synthetase
We thank Dr. Trudy Forte, CHORI, for critical reading of the manuscript. This work was supported in part by a grant from the National Institutes of Health [R21HL092535] as well as funds from the CHORI Red Blood Cell Laboratory and the CHORI foundation (to FAK) and Jordan Family Fund (to ES).
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.