PON1 has palmitoyl‐protein thioesterase (PPT) activity, and can affect the presence of SR‐B1 on the endothelial cell membrane

The high‐density lipoprotein (HDL)‐associated enzyme paraoxonase 1 (PON1) is expressed almost exclusively in the liver and is then transported by HDL to the peripheral tissues. The lipophilic nature of PON1 enables its easy exchange between the lipoprotein and cell membranes in a process that is dependent on the HDL receptor scavenger receptor class B, type 1 (SR‐B1). In endothelial cells, PON1 binding to the cell membrane leads to its internalization by endocytosis and subsequent lysosomal degradation. PON1 is a “promiscuous” enzyme with unusually broad substrate specificity in vitro, but its actual function and substrate are still unknown. The enzyme requires a lipid environment and becomes completely inactive upon delipidation. However, when PON1 binds HDL, its active site faces the lipoprotein's core and is inaccessible to external substrates. Hence, the HDL‐bound PON1 is inactive toward substrates outside the particle's lipid core and is rapidly degraded and becomes inactive upon internalization. Consequently, the enzyme is only active in the cell membrane during its transit from HDL to the cytoplasm. To assign a function to PON1, we investigated whether it is a palmitoyl‐protein thioesterase (PPT) that can hydrolyze the palmitoyl moieties of membrane proteins involved in HDL and cholesterol transport, such as SR‐B1, ABCA1, or their neighboring caveola proteins to facilitate the release of HDL or trigger its endocytosis. This study shows that PON1 can hydrolyze palmitoyl‐cysteine thioester bonds in vitro, has direct or indirect PPT activity in vivo, and can significantly decrease the presence of SR‐B1 in the endothelial membrane.

Paraoxonase 1 (PON1) is a small (43 kDa) high-density lipoprotein (HDL)-bound enzyme belonging to a threemember family of paraoxonases (reviewed in Reference 1).Most circulating PON1 is attached to the HDL fraction of small, spherical, lipid-poor HDL particles (HDL 3 ). 2 Although the enzyme is expressed almost exclusively in the liver and binds HDL with high affinity (K d ( 10 À9 M), 3 it can shuttle between its carrier HDL and membranes of the peripheral tissues. 4,5This shuttling is quantitatively dependent on the presence of the HDL receptor scavenger receptor class B, type 1 (SR-B1). 4ON1 binding to the endothelial membrane leads to its internalization by caveola-mediated endocytosis, lysosomal degradation, and complete inactivation. 6,7everal epidemiological studies have found an inverse relationship between serum PON1 activity and atherosclerosis and cardiovascular disease (extensively reviewed in References 8-10).The atheroprotective capability of PON1 has been conclusively demonstrated by using PON1 knockout (PON1 À/À ) and transgenic (PON1-Tg) mice.PON1 À/À mice are more prone to developing atherosclerotic lesions and plaques than control mice when fed a high-fat diet. 11Conversely, mice overexpressing human PON1 are dose-dependently more resistant to developing atherosclerotic lesions. 12However, despite decades of intensive research, the exact physiological function of PON1 and the identity of its endogenous substrate remain unknown. 9he ability of PON1 to protect low-density lipoproteins (LDLs) against metal ions-induced lipid peroxidation by hydrolyzing lipid peroxides in vitro has been construed as the basis for the antiatherogenic activity of HDL, 13 a notion that is supported by many publications. 14However, the exact mechanism of PON1 protection against lipid peroxidation has not been adequately elucidated.It is unclear whether the main catalytic site of PON1 is involved in its antioxidant activity, or whether LDL protection represents a second, independent activity, which relies on a free thiol group on cysteine 284. 15,16At least two publications seem to refute the peroxidase activity of PON1 altogether: Connelly et al. 17,18 and Teiber et al. 17,18 showed that purified active PON1, in the presence of lipids or tergitol, is incapable of preventing LDL or HDL oxidation by either metal ions or peroxynitrite.
Aside from lipid peroxides, PON1 can hydrolyze, with varying efficiency, a remarkably long list of potential substrates-the outcome of a decade-long quest to find its "true" endogenous substrate.The list includes numerous organophosphates, lactones, esters, and thioesters. 19his unique quality of PON1 has been termed "promiscuity." 20It reflects the exceptional versatility of residues of the enzyme's catalytic group to perform multiple tasks, depending on the substrate. 19It is currently believed that PON1 is capable of changing its conformation to adapt to new substrates. 20ON1 has a six-bladed beta-propeller architecture, with a catalytic site inside its central tunnel. 21The catalytic site holds two calcium ions that are essential for its activity. 21Access to the catalytic site is limited by a lid ("canopy") formed by two (out of three) alpha-helices. 21hese two helices create a high-affinity bond with HDL, with the catalytic site facing the lipoprotein core. 3,21he third, N-terminal alpha-helix protrudes above the catalytic site and is probably used to anchor PON1 to the HDL particle by penetrating its lipid envelope. 21he spatial position of the HDL-bound PON1 implies that it can hydrolyze only lipid-soluble substrates that can reach the active site by dissolving into the HDL core. 3 The binding of PON1 to HDL, and especially to the apolipoprotein A1 (ApoA1) component of HDL, dramatically increases its structural stability and catalytic efficiency toward lipophilic lactones.3 PON1 has an absolute requirement for a lipid environment; delipidation of the enzyme results in complete loss of activity.3 This dependence is probably why PON1 is always associated with phospholipids of either the plasma membrane or the lipoprotein envelope.17 As a result, the only place and time that PON1 can be active toward substrates outside the HDL's lipid core is inside the endothelial cell membrane, shortly after its release from its lipoprotein carrier and before it is internalized and inactivated by lysosomal degradation.
By carefully analyzing the various structural and functional oddities of PON1, we surmised that the enzyme could be a palmitoyl-protein thioesterase (PPT), whose substrate is the palmitoyl moieties anchoring proteins involved in cholesterol transport to the cell membrane.3][24] The hydrolysis of these palmitoyl groups by PON1 can destabilize these proteins, 25 facilitating the release of a tightly bound cholesterol-loaded HDL or accelerate its endocytosis.We believe that the PPT activity of PON1 is an essential component of the HDL's adaptation for collecting and removing excess cholesterol from the endothelia of lymph and blood vessels.In this study, we present evidence supporting this hypothetical function of PON1.We show that PON1 can hydrolyze palmitoyl-cysteine thioester bonds in vitro, has direct or indirect PPT activity in endothelial cell culture, and can significantly decrease the expression of SR-B1 in endothelial cells and its presence in their membrane.

| Cells, media, and reagents
The human umbilical vein endothelial cell (HUVEC) line CRL-1730 ATCC was purchased from the American Type Culture Collection (Rockville, MD, USA).The cells were cultured in Ham's F-12K medium supplemented with 10% (v/v) fetal bovine serum, 0.1 mg/mL heparin, 10 mg/mL streptomycin, 10 mg/mL penicillin, and endothelial cell medium (ECM).In the experiments, the cells (passage 4) were maintained in a 37 C incubator with 5% CO 2 and treated in a biological hood under sterile conditions.

| HDL isolation
HDL was isolated from commercial human serum using an ultracentrifugation-free HDL purification kit (STA-608, Cell Biolabs).Characterization of the isolated HDL in comparison to HDL isolated by ultracentrifugation is included in the product manual (https://www.cellbiolabs.com/sites/default/files/STA-607-hdl-purificationkit.pdf).The isolated HDL was stored at 4 C until use.Before initiating the experiments, the HDL was dialyzed twice, 1 h each time and once more overnight, against phosphate buffered saline (PBS) at 4 C.The PON1 protein percentage of the isolated HDL was calculated to be 0.14% (0.07 mg PON1 protein out of 50 mg HDL protein).

| Synthesis of the palmitoyl-Fmoc-cysteinyl thioester model substrate
The model substrate for testing rePON1's putative PPT activity was produced by the following procedure.Fmoc-L-cysteine (100 mg) was added to 5 mL trifluoroacetic acid.Palmitoyl chloride (200 μL) was added slowly, and the solution was mixed for 24 h at room temperature.Water (25 mL) was added, and the resulting precipitate was filtered and dried.The product was further purified using preparative direct chromatography on a silica gel column.The product was eluted with 30% ethyl acetate in hexane to obtain 25 mg product (14% yield with more than 95% purification as determined by HPLC and 1 H-NMR).The product mass and composition were determined by high-resolution LC-MS to be 581.806(C 34 H 47 NO 5 S) (Figure S1 and Figure S2).

| Measurement of rePON1 thioesterase activity
The ability of rePON1 to hydrolyze the thioester bond of the model substrate palmitoyl-Fmoc-cysteinyl thioester was measured using Ellman's reagent (DTNB).The substrate, dissolved in 100 μL (50 mM) Tris-HCl buffer pH 7.4 containing 1% (w/v) DMSO (1 mM), alone or with HDL at a final concentration of 50 μg/mL, was incubated overnight at 37 C. rePON1 was added to a final concentration of 100 μg/mL, followed by DTNB at a final concentration of 2.5 mM.Self-cleavage of the substrate was evaluated, and the enzymatic activity of rePON1 was measured by monitoring the absorbance at 412 nm, at 0.25 s intervals, in an Infinite M200 PRO spectrophotometer (Tecan, Männedorf, Switzerland).

| Cell viability assay
HUVECs were seeded in a 6-well cell culture plate at 500,000 cells per well in 2 mL ECM.The cells were treated with different concentrations of rePON1 (5, 10, and 50 μg/mL) or HDL (25 and 50 μg/mL) for 4 h.Cells were then washed with PBS, and 400 μL XTT solution was added.The plate was incubated for 5 h, and absorbance at 450 nm was measured every hour.

| Preparation of cell lysate
HUVECs were seeded in a 6-well cell culture plate at a density of 500,000 cells per well in 2 mL ECM and incubated overnight until full confluence.The cells were then treated for 4 h at 37 C and 5% CO 2 with rePON1 (10 μg/mL), HDL (50 μg/mL), or both together, with or without the addition of 1 μM of the rePON1 inhibitor 2-hydroxyquinoline (2HQ).The control included cells without any additions.After 4 h treatment, the cells were washed and detached using 1 mL of a 0.25% solution of trypsin-EDTA.The detached cells were washed with M-199 medium and centrifuged at 4 C, 1250g for 8 min.The cell pellets were resuspended in 300 μL cold lysis buffer comprised of 0.5 M Tris-HCl pH 7.4, 1.5 M NaCl, 2.5% (w/v) deoxycholic acid, 10% (w/v) nonyl phenoxypolyethoxylethanol (NP-40), and 10 mM EDTA, and supplemented with 1 mM phenylmethylsulfonyl fluoride (PMSF) protease inhibitor.The samples were then gently vortexed for 15 min.Cells were centrifuged at 14,000g for 15 min at 4 C.The supernatants containing the cell lysate were transferred to clean centrifuge tubes for further experimentation or storage at À80 C.

| Quantification of palmitoylated proteins by acyl-biotin exchange (ABE) assay
The ABE assay for the quantitative determination of palmitoylated cysteine residues was performed essentially as previously described. 26Briefly, 10 mM of the freethiol-blocking N-ethylmaleimide was added to the cell lysates at a final volume of 300 μL and sonicated for 15 min.A mixture of chloroform and methanol (1:4, v/v) was then added to the samples, and they were centrifuged at 4 C and 14,000g for 30 min.The pellets were dissolved in 4% (w/v) SDS and gently thawed in a refrigerator for 30 min.Hydroxylamine reagent was added at a final concentration of 0.6 M, followed by 200 μL of 1 mM biotin maleimide in 2% DMSO.A 96-well ELISA plate was coated with 100 μL protein lysate (10 μg/mL in pH 9.6 bicarbonate coating buffer).The plate was incubated overnight at 4 C and then washed three times with PBS containing 0.05% (w/v) Tween-20.After washing, the plate was blocked with a blocking buffer (PBS with 5% skim milk) and incubated for 1 h at 37 C.After blocking, avidin-horseradish peroxidase (HRP) conjugate (diluted 1:2000 in blocking buffer) was added to the plate and incubated for another 1 h at 37 C.The plate was then washed, and 100 μL of the HRP substrate 3,3 0 ,5,5 0tetramethylbenzidine (TMB) was added.The absorbance was measured at 650 nm and used to assess the number of biotinylated thiols, representing the number of palmitoylated cysteine residues of proteins in the cell lysates.

| Quantification of SR-B1 in endothelial cell lysates by ELISA
The wells of a 96-well ELISA plate were coated with 100 μL HUVEC lysate (10 μg protein/mL in pH 9.6 bicarbonate coating buffer).The plate was incubated overnight at 4 C and washed three times with PBS containing 0.05% Tween-20, and then blocked with a blocking buffer (PBS with 5% skim milk), incubated for 1 h at 37 C, and washed.The primary antibody, anti-SR-B1 monoclonal antibody, was added at a dilution of 1:1000 in blocking buffer (5% skim milk), and the plate was incubated again for 1 h at 37 C and washed.A secondary antibody, goat anti-rabbit HRP, was added at a dilution of 1:10,000 in blocking buffer, and the plate was incubated again, washed, and HRP substrate (TMB) was added.Absorbance was measured at 650 nm.

| Quantification of SR-B1 in endothelial cell lysates by Western blot analysis
The HUVEC lysate proteins were mixed with 4Â Laemmli sample buffer and denatured at 95 C for 5 min.Equal amounts of each protein sample (30 μg) were separated on a 10% SDS-polyacrylamide gel at a constant current of 50 mA.The proteins were then transferred to a polyvinylidene fluoride membrane using the TransBlot Turbo transfer system (Bio-Rad) according to the manufacturer's instructions.After the transfer, the membrane was blocked with 5% skim milk in 1Â TBST buffer pH 8 (0.24% Tris-HCl, 0.8% NaCl, and 0.5% Tween-20, w/v).The blocking reaction was carried out for 2 h with agitation at room temperature.The membrane was then washed three times for 5 min each with 1Â TBST buffer, and incubated overnight at 4 C in the dark with the primary antibodies rabbit anti-SR-B1 (diluted 1:1000) and mouse anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (diluted 1:1000).The next day, the membrane was washed three additional times for 5 min each with 1Â TBST buffer and then incubated for 1 h with the secondary antibody goat anti-rabbit HRP (diluted 1:3000) and rabbit anti-mouse HRP (diluted 1:2000) in 1Â TBST buffer.After the incubation, the membrane was developed using an enhanced chemiluminescence (ECL) substrate, and the signal was measured at different time points using the ChemiDoc MP imaging system (Bio-Rad).

| Identification and quantification of SR-B1 on endothelial cell membranes by direct ELISA
SR-B1 was identified and quantified on the endothelial cell membrane essentially as previously described. 27UVEC at a density of 4000 cells per well were seeded into each well of a 96-well plate and incubated overnight.The following day, the cells were treated with rePON1 (10 μg/mL), HDL (50 μg/mL), or both together, for 6 h at 37 C. Control wells contained cells without any additions.Each experiment was performed in triplicates (wells).After the incubation, the plate was washed with PBS, and the cells were fixed with 1% (w/v) paraformaldehyde for 15 min.The plate was then washed three times with PBS and blocked with 2% (w/v) bovine serum albumin for 2 h at 37 C.After another wash, the plate was incubated with the primary antibody, rabbit anti-SR-B1 (diluted 1:400), for 1 h at 37 C.The secondary antibody (goat anti-rabbit HRP diluted 1:10,000) was then added for 1 h at 37 C, followed by the substrate TMB.Signal intensity was measured at 650 nm.

| Heat inactivation of PON1
PON1 heat inactivation was performed by heating the enzyme at 65 C for 15 min, which decreased its lactonase activity by 80%.

| HDL-mediated endothelial efflux
HUVECs were seeded at a density of 25,000 cells/well in 96-well plates for 24 h.The cells were then treated with Bodipy-labeled fluorescent cholesterol for 1 h.Subsequently, the cells were washed with Dulbecco's modified Eagle's medium (DMEM) buffered with 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) and incubated with RPMI medium containing an acyl-CoA cholesterol acyltransferase (ACAT) inhibitor (1 μg/mL) for 18 h.The cells were rewashed with DMEM-HEPES and incubated with, or without, rabbit anti-mouse SR-BI antibody, diluted 1:1000 (v:v), for 2 h.After washing, the cells were incubated with DMEM-HEPES containing rePON1 (10 μg/mL), heateinactivated rePON1 (10 μg/mL), HDL (50 μg/mL), or both for 4 h.The fluorescence of the medium outside the cells was measured at an excitation/emission of 482/515 nm.The cells were lysed with 1% cholic acid at room temperature for 1 h, and the fluorescence of the lysate was measured.Cholesterol efflux was calculated by dividing the fluorescence intensity of the medium by the sum of the fluorescence values of the medium and cell lysate.

| PON1 hydrolyzes a synthetic palmitoyl thioester substrate in vitro
The PPT activity of rePON1 was measured in vitro using the synthetic substrate palmitoyl-cysteinyl thioester, in which the cysteine's amine is protected by the Fmoc (fluorenylmethoxycarbonyl) protection group.The substrate was synthesized, purified, and identified using LC-MS/MS and 1 H-NMR (Figures S1 and S2).
The ability of rePON1, with and without HDL, to cleave the thioester bond of the synthetic substrate was measured spectrophotometrically using Ellman's assay.This substrate has very limited solubility and incubating it at a low concentration (1 mM) with either rePON1 or HDL alone did not result in its cleavage.However, the addition of rePON1 after overnight incubation of HDL with the substrate to allow its diffusion into the lipoprotein, resulted in a statistically significant increase in substrate hydrolysis, probably due to its incorporation into the lipoprotein and the increased stability and activity of PON1 in its phospholipid envelope (Figure 1).

| PON1 affects HUVEC viability
The effect of rePON1 and HDL on HUVEC viability was measured using the XTT assay.HUVECs were incubated with different concentrations of rePON1 (5, 10, and 20 μg/mL) or HDL (25 and 50 μg/mL) at 37 C for 4 h.At the end of the incubation, the cells were washed, and the number of living cells was counted.Incubation of F I G U R E 1 rePON1 hydrolysis of a synthetic substrate.Palmitoyl-protein thioesterase (PPT) activity of rePON1 (100 μg/mL) with palmitoyl-Fmoc-cysteinyl thioester at 1 mM concentration.rePON1 was added separately or after incubating the substrate with HDL overnight at 37 C. PPT activity measured using Ellman's reagent and calculated as μM TNB produced per min; extinction coefficient of TNB at 412 nm is 14,150 M À1 cm À1 .Results are presented as mean ± standard error (n = 3).****Significant difference between substrate incubated with rePON1 and HDL versus substrate incubated with rePON1 or HDL separately ( p < 0.00001).HDL, high-density lipoprotein; rePON1, recombinant PON1; TNB, thiobis(2-nitrobenzoic acid).the cells with rePON1 at a concentration of 20 μg/mL was toxic and killed the cells (Figure 2).However, lower concentrations of 10 and 5 μg/mL rePON1 did not affect the cells' viability relative to the control.Unlike rePON1, HDL at 25 or 50 μg/mL did not affect the cells' viability.Based on these results, we conducted subsequent experiments with concentrations not exceeding 10 and 50 μg/ mL for rePON1 and HDL, respectively.

| PON1 decreases the number of palmitoylated cysteine residues in HUVEC
HUVECs were incubated with rePON1 (10 μg/mL), HDL (50 μg/mL), or both for 4 h.The cells were then washed, and protein lysates with the same adjusted protein concentration were prepared.The number of palmitoylated cysteine residues was assessed spectrophotometrically using the ABE-ELISA method. 26Incubating the cells with rePON1 alone significantly reduced the number of palmitoylated residues (Figure 3).Addition of the PON1 inhibitor 2HQ with rePON1 abolished this activity.
Incubating the cells with HDL alone significantly increased the number of palmitate moieties.In contrast, the addition of rePON1 with HDL significantly this number compared to the addition of HDL alone.Here, again, addition of the PON1 inhibitor abolished this effect.These results suggest that rePON1 triggers the hydrolysis of palmitoyl moieties in vivo or somehow inhibits the HDL-induced increase in their number.

| PON1 affects the endothelial expression of SR-B1
The effect of rePON1, HDL, or both on SR-B1 expression in HUVEC lysates was measured by ELISA using rabbit anti-SR-B1 monoclonal antibodies.Incubation of the cells with rePON1 alone significantly increased the expression of SR-B1 compared to the control cells without treatment.Addition of rePON1 inhibitor 2HQ abolished this increase (Figure 4).Incubation of the cells with HDL alone led to a significantly larger increase in SR-B1 expression relative to the control and incubation with rePON1 alone.Adding HDL together with rePON1 led to a statistically significant decrease in the amount of expressed SR-B1 relative to cells treated with HDL alone, and again, the use of 2HQ abolished this effect.A similar effect of rePON1, HDL, or both on SR-B1 expression in HUVEC lysates was also measured by Western blot analysis (Figure 5).

| PON1 affects the presence of SR-B1 on the endothelial cell membrane
The effect of rePON1 (10 μg/mL), HDL (50 μg/mL), or both on SR-B1 presence on endothelial cell membranes was measured by direct ELISA.HUVECs, grown on an ELISA plate, were incubated with rePON1, HDL, or both and reacted with rabbit anti-SR-B1 antibodies after washing and fixation.Incubation with HDL alone significantly increased SR-B1 presence on the cell membrane compared to control cells without treatment.Adding rePON1 and HDL together significantly decreased SR-B1 membrane presence compared to the addition of HDL alone (Figure 6).

| PON1 affects HUVEC cholesterol efflux
The effect of rePON1 (10 μg/mL), HDL (50 μg/mL), or both on HUVEC cholesterol efflux was measured using fluorescently labeled cholesterol.The results (Figure 7A) closely follow the effect of PON1 and HDL on the membrane presence of SR-B1.The incubation of the cells with anti-SR-B1 antibodies for 2 h before the initiation of cholesterol efflux decreased it by $30%, indicating that, at least, this fraction of the total observed HDL-mediated efflux is dependent on SR-B1 (Figure 7B).In this experiment, we replaced the PON1 inhibitor 2HQ by heat inactivation since the addition of this inhibitor doubled the efflux by PON1 alone and by PON1 together with HDL (not shown).We suspected that 2HQ itself, internalized into the HUVEC either directly by PON1 or by PON1-induced endocytosis, could stimulate cholesterol efflux through activation of liver X receptor (LXR) α for which quinols are known agonists. 28

| DISCUSSION
This article provides evidence for a PPT activity of PON1, whose substrate is the palmitoyl moieties of membrane proteins.We show that HDL-bound PON1 can hydrolyze palmitoyl-cysteine thioester bonds of an artificial substrate in vitro.We also demonstrate that PON1 has direct or indirect PPT activity in vivo, manifested by an increased number of free protein thiol groups after incubation of endothelial cells with the enzyme.In addition, the toxic effect of the unbound, free PON1 toward HUVEC at concentrations above 10 μg/mL, observed in this work, could result from this PPT activity.It is possible that the free enzyme damages cells by indiscriminately hydrolyzing palmitate moieties as it enters the endothelial cell membrane.As PON1 bears no sequence or structural similarity to the known human PPT (PPT1), its observed in-vivo effects could also be the consequence of indirect activation of the endogenous PPT1. 29 Elucidation of this possibility requires further investigation.
Incubating the HUVECs with either PON1 or HDL alone resulted in increased expression of SR-B1 compared Effects of HDL and rePON1 on the presence of SR-B1 on cell membranes.The effect of rePON1, HDL, or both on the presence of SR-B1 on endothelial cell membranes was measured by direct ELISA.Endothelial cells were incubated with rePON1 (10 μg/mL), HDL (50 μg/mL), or both for 6 h.The cells were then fixed, blocked, and reacted with primary anti-SR-B1 and secondary goat anti-rabbit HRP antibodies.Results are presented as mean ± standard error (n = 3).* and ****Significant difference at p < 0.05 and 0.0001, respectively.ELISA, enzyme-linked immunosorbent assay; HDL, high-density lipoprotein; HRP, horseradish peroxidase; rePON1, recombinant PON1; SR-B1, scavenger receptor class B, type 1. to the control, with the effect of PON1 being smaller than that of HDL.However, incubating the cells with HDLbound PON1 significantly decreased SR-B1 expression compared to incubation with HDL alone.The ability of HDL and PON1 to increase the expression of SR-B1 has already been described by others. 27,30The stimulatory effect of HDL can be construed as the manifestation of a positive feedback loop.However, we cannot adequately explain why incubation of HUVECs with rePON1 alone also increases SR-B1 expression and cholesterol efflux, albeit to a smaller extent than HDL.Since only the HDLbound enzyme can act specifically on SR-B1 or other nearby caveola proteins, as HDL binds the receptor, the effect of PON1 alone may be nonspecific and merely a consequence of indiscriminate depalmitoylation of membrane proteins.
The results of the endothelial cholesterol efflux experiments matched the effect of PON1 and HDL on the expression of SR-B1 and its membrane presence.While HDL alone increased the endothelial cholesterol efflux compared to the control, adding PON1 to the lipoprotein significantly decreased it.These results contrast a previous report showing the stimulation of HDL-mediated efflux in macrophages by PON1 and the independence of this effect on SR-B1. 31The incubation of HUVECs with anti-SR-B1 antibodies for 2 h before initiating cholesterol efflux, decreased this efflux by $30%, indicating, contrary to the work of O'Connell et al., 32 that approximately onethird of the observed HDL-mediated efflux is SR-B1-dependent.3][34] Therefore, a more thorough evaluation of the function of PON1 requires measurement of the rates of both processes, which is beyond the scope of this work.
Nonetheless, whether direct or indirect, the PON1-associated PPT activity may be involved in either or both processes by destabilizing palmitoylated membrane proteins such as SR-B1, ABCA1, and Cav1.The larger diameter of the cholesterol-loaded HDL particle probably allows PON1 to contact the cell membrane and use its ability to shuttle between membranes 4,35 to move into the endothelial cell membrane and make its way toward its inner leaflet, where it can hydrolyze the palmitoyl moieties of SR-B1 or critical components of the caveolae such as Cav1.This hydrolysis can destabilize the receptor or even the entire caveola structure, causing SR-B1 to release its bound lipoprotein or trigger the endocytosis and subsequent recycling of HDL.
Once PON1 completes its task in the membrane, it becomes fully internalized in the cytoplasm by caveolamediated endocytosis and is then directed to lysosomal destruction. 6The mechanism of PON1 internalization and its dependence on the enzyme's catalytic activity are not yet clear and warrant further investigation.
Instead of being an evolutionary oddity, the exceptional promiscuity of PON1 20 can now be viewed as a perfect adaptation to a large and virtually immobile substrate situated in the restrictive environment of the inner membrane leaflet.Under such conditions, the enzyme must maneuver around the substrate to correctly position the thioester bond of the palmitoyl moiety with respect to the catalytic group.The versatility of residues of the catalytic group to perform multiple tasks 20 now relaxes the restrictions on the correct relative spatial positions of the enzyme and its substrate and increases PON1's ability to carry out the hydrolysis.
A decrease in PON1 activity could slow the RCT and cholesterol turnover, resulting in longer retention of LDL in the blood, with the inevitable consequence of increased LDL lipid peroxidation, as observed in PON1 À/À mice. 36Therefore, there is no need to invoke a direct peroxidase activity of PON1 to account for its observed effect on LDL peroxidation in vivo.In the long run, PON1 malfunction and slower RCT can lead to cholesterol accumulation in the intima and the formation of an atherosclerotic plaque, as has also been found in PON1 À/À mice and epidemiological studies. 9,36n summary, our results indicate potentially new enzymatic activity, natural substrate, and function of PON1, which together provide a coherent account of its unusual characteristics.However, the exact function of PON1 and the mechanisms involved in its transit from the HDL particle to the cell membrane and its entry into the cytoplasm remain to be elucidated, and conclusive evidence of its ability to directly hydrolyze palmitoyl residues in vivo must be provided by future research.

AUTHOR CONTRIBUTIONS
Rasha Ashkar performed the main experiments and analyses and drafted the manuscript.Ali Khattib contributed to the methodology.Sanaa Musa contributed to the compound synthesis.Doron Goldberg and Soliman Khatib designed and supervised the study, reviewed, and edited the manuscript.All authors have read and approved the final version of the manuscript.

F
I G U R E 2 PON1 affects the viability of HUVEC.HUVEC viability relative to control was measured by XTT assay.HUVECs were incubated with rePON1 (5, 10, and 20 μg/mL) or HDL (25 and 50 μg/mL) at 37 C for 4 h.Then cells were then treated with XTT and the absorbance was measured at 450 nm.Results are presented as mean ± standard error (n = 3).***Significant difference at p < 0.0001 relative to control (cells without any additions).HDL, high-density lipoprotein; HUVEC, human umbilical vein endothelial cell; rePON1, recombinant PON1.U R E 3 rePON1 decreases the quantity of palmitoylated cysteine residues in HUVEC.HUVECs were incubated with PON1, HDL, or both for 4 h and then collected and lysed.All cell lysates had the same adjusted protein concentration.The number of thioester bonds in the cell lysates was measured spectrophotometrically at 650 nm by the ABE-ELISA assay.Results are presented as mean ± standard error (n = 3).*, **, ***, and ****Significant difference at p < 0.05, 0.01, 0.001, and 0.0001, respectively.ABE, acyl-biotin exchange; ELISA, enzyme-linked immunosorbent assay; HDL, high-density lipoprotein; HUVEC, human umbilical vein endothelial cell; rePON1, recombinant PON1.