Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2023 Aug 2;18(8):1891–1904. doi: 10.1021/acschembio.3c00401

Small Molecule Activation of NAPE-PLD Enhances Efferocytosis by Macrophages

Jonah E Zarrow †, Abdul-Musawwir Alli-Oluwafuyi †, Cristina M Youwakim ∥, Kwangho Kim ‡,§, Andrew N Jenkins ⊥, Isabelle C Suero †, Margaret R Jones ‡, Zahra Mashhadi †, Ken Mackie #, Alex G Waterson †,‡,§, Amanda C Doran ∥, Gary A Sulikowski †,‡,§, Sean S Davies †,§,*
PMCID: PMC10443532  PMID: 37531659

Abstract

graphic file with name cb3c00401_0010.jpg

N-Acyl-phosphatidylethanolamine hydrolyzing phospholipase D (NAPE-PLD) is a zinc metallohydrolase that hydrolyzes N-acyl-phosphatidylethanolamines (NAPEs) to form N-acyl-ethanolamines (NAEs) and phosphatidic acid. Several lines of evidence suggest that reduced NAPE-PLD activity could contribute to cardiometabolic diseases. For instance, NAPEPLD expression is reduced in human coronary arteries with unstable atherosclerotic lesions, defective efferocytosis is implicated in the enlargement of necrotic cores of these lesions, and NAPE-PLD products such as palmitoylethanolamide and oleoylethanolamide have been shown to enhance efferocytosis. Thus, enzyme activation mediated by a small molecule may serve as a therapeutic treatment for cardiometabolic diseases. As a proof-of-concept study, we sought to identify small molecule activators of NAPE-PLD. High-throughput screening followed by hit validation and primary lead optimization studies identified a series of benzothiazole phenylsulfonyl-piperidine carboxamides that variably increased activity of both mouse and human NAPE-PLD. From this set of small molecules, two NAPE-PLD activators (VU534 and VU533) were shown to increase efferocytosis by bone-marrow derived macrophages isolated from wild-type mice, while efferocytosis was significantly reduced in Napepld–/– BMDM or after Nape-pld inhibition. Together, these studies demonstrate an essential role for NAPE-PLD in the regulation of efferocytosis and the potential value of NAPE-PLD activators as a strategy to treat cardiometabolic diseases.

Atherosclerotic cardiovascular disease (ASCVD) remains a major cause of death in the United States and throughout the world. Efferocytosis, the noninflammatory clearance of apoptotic cells by macrophages, is a critical step in the resolution of inflammation and defective efferocytosis has been implicated in the development and expansion of necrotic cores within atherosclerotic plaques.1,2 Plaques with large necrotic cores are highly vulnerable to rupture, thereby triggering potentially fatal thrombosis and myocardial infarction.3 Therefore, identifying the factors whose dysfunction impairs efferocytosis and the appropriate countermeasures against this dysfunction could lead to novel treatments for ASCVD.

N-Acyl-phosphatidylethanolamine-hydrolyzing phospholipase D (NAPE-PLD) is a zinc metallohydrolase within the metallo-β-lactamase superfamily.4,5 NAPE-PLD hydrolyzes N-acyl-phosphatidylethanolamines (NAPEs) to phosphatidic acid and N-acyl-ethanolamines (NAEs) such as palmitoylethanolamide (PEA) and oleoylethanolamide (OEA)4,5 (Figure 1). In rodents, high-fat or Western diets decrease Napepld expression and levels of PEA and OEA in a variety of tissues.6–8NAPEPLD expression is also reduced in atherosclerotic plaques (especially unstable plaques) of human coronary arteries.8 In mice, directly administering NAEs or NAE-boosting bacteria counteracts atherosclerosis as well as other cardiometabolic diseases including obesity, glucose intolerance, and nonalcoholic fatty liver disease.8–12 Importantly, these treatments inhibit enlargement of the necrotic core within atherosclerotic lesions.8,10,11 Cellular studies show that PEA and OEA enhance M2 polarization and the efferocytosis capacity of bone-marrow-derived macrophages (BMDM) via GPR55 and/or PPARα-dependent mechanisms.8,10 Together, these studies suggest that reduced macrophage NAPEPLD expression could lead to reduced efferocytosis by macrophages and, thereby, drive the expansion of the necrotic core and atherosclerosis.

Figure 1.

Figure 1

Open in a new tab

Biosynthesis of N-acyl-ethanolamines via NAPE-PLD. N-Acyl-ethanolamines (NAEs) including palmitoylethanolamide (PEA) and oleoylethanolamide (OEA) are formed by NAPE-PLD pathway in a two-step process. First, PE N-acyltransferases transfer an acyl chain from phosphatidylethanolamine (PC) to the nitrogen of phosphatidylethanolamine (PE) to generate N-acyl-phosphatidylethanolamines (NAPEs) and lysophosphatidylcholine (lysoPC). Then NAPE-PLD cleaves NAPE at the distal phosphodiester bond to generate NAE and phosphatidic acid (PA). NAEs then act on receptors, including PPARα, GPR119, and GPR55 to exert biological effects. NAEs are rapidly inactivated by fatty acid amide hydrolase (FAAH) and N-acylethanolamine acid amidase (NAAA) by their degradation to ethanolamine and free fatty acid.

If NAPE-PLD regulates efferocytosis, then small molecules that enhance macrophage enzyme activity should enhance macrophage efferocytosis and could, therefore, potentially inhibit the development of unstable atherosclerotic lesions. While several small molecule inhibitors of NAPE-PLD have been reported recently,13–15 there are currently no small molecule activators of NAPE-PLD. We, therefore, sought to identify small molecules that could enhance macrophage enzyme activity in order to test their effects on the macrophage efferocytosis capacity.

Results and Discussion

New NAPE-PLD Activator Chemotype Identified by HTS and Early SAR Studies

We screened 39,328 compounds from the Vanderbilt Discovery Collection, a chemical library of lead-like compounds, for their effects on Nape-pld activity using the commercially available fluorogenic NAPE analog, PED-A1 (Figure 2A), and recombinant mouse Nape-pld.14 The change in the measured rate of fluorescence after the addition of commercially available PED-A1 (Figure 2B) was used to calculate a B-score for each compound (Figure 2C). From the 39,328 compounds screened, 399 were judged as potential activators based on observed change in fluorescence by at least 3 standard deviations compared to vehicle.

Figure 2.

Figure 2

Open in a new tab

High throughput screening identifies potential NAPE-PLD activators. (A) Schematic representation of the HTS assay that detects Nape–Pld activity by the fluorescence resulting from the hydrolysis of the fluorogenic NAPE analog PED-A1. Intact PED-A1 only weakly fluoresces due to internal quenching by its dinitrophenyl moiety. Nape-pld hydrolysis of PED-A1 generates dinitrophenyl-hexanoylethanolamide (DNP-EA) and highly fluorescent BODIPY-labeled phosphatidic acid (BODIPY-PA). (B) Sample activity curves from HTS for controls, a representative activator hit, and a representative inhibitor hit. The shaded region represents the time period used in scoring fluorescence changes for all compounds. (C) B-scores of various test library compounds in the HTS assay. Compounds with B-scores of ≥3 (activators) or ≤−3 (inhibitors) were deemed to be potential hits.

Three of the 399 identified activators (Table 1, entries 1–3) shared a common benzothiazole phenylsulfonyl-piperidine carboxamide (BT-PSP) core structure. To determine if this series of benzothiazoles might serve as a tractable lead for the development of activators as chemical probes, we obtained additional structurally similar compounds using a combination of targeted purchasing of commercial molecules (Table 1, entries 4–13) and discrete chemical synthesis (Table 1, entries 14–22). The activity of this series of 22 benzothiazole phenylsulfonyl-piperidine carboxamides, thus, obtained was assessed for enzyme activation using recombinant mouse Nape-pld (Figure S1 and Table 1). Compounds VU534 and VU533 (entries 8 and 9) proved the most potent of the series of enzyme activators, both showing half-maximal activation concentrations (EC50) of 0.30 μM and a more than 2-fold maximal induction of Nape-pld activity relative to vehicle controls (Emax > 2.0). Nine other BT-PSPs (entries 3, 5–7, 10–14) had EC50 ≤ 1.1 μM and Emax > 1.7. The remaining members of the series had EC50 < 10 μM and Emax > 1.7, except that compound VU212 (entry 16) showed poor efficacy (Emax < 1.6), and VU205 and VU233 proved inactive (Emax < 1.2). With this preliminary structure–activity relationship (SAR) data, we selected VU534 and VU533 as chemical probes for the study of NAPE-PLD activation and VU233 as a negative control in studies outlined below.

Table 1. In Vitro Nape-pld Modulation by Benzothiazole Phenylsulfonyl-Piperidine Carboxamides (BT-PSPs)a.

graphic file with name cb3c00401_0009.jpg

Open in a new tab
a

Compounds were either purchased from a commercial source or prepared in house by chemical synthesis.

b

EC50: concentration (in μM) needed for 50% maximal activator effect.

c

Emax: maximal efficacy (expressed as fold increase in Nape-pld activity versus vehicle only).

d

95% Cl for EC50 and Emax indeterminant due to poor fit.

e

NA: not active.

Utility as Probes of NAPE-PLD Activation in Mouse and Human Cells

To determine if the selected small molecules could be used as probes of NAPE-PLD activation in cultured cells, we first evaluated their cytotoxicity in RAW264.7 mouse macrophages and HepG2 human hepatocytoma cells. Graded concentrations of VU534 showed minimal cytotoxicity up to 30 μM in either cell line (Figure S2A,B). Likewise, VU533 and VU233 also showed no cytotoxicity when tested at 30 μM in either RAW264.7 (Figure S2C) or HepG2 (Figure S2D).

We next examined the efficacy of activators VU534 and VU533 and inactive VU233 to increase the Nape-pld activity in RAW264.7 cells. Both activators VU534 and VU533 significantly increased the Nape-pld activity, while VU233 showed no significant effect (Figure 3A). Other analogs in the series of 22 benzothiazole phenylsulfonyl-piperidine carboxamides that increased the activity of recombinant Nape-pld in the biochemical assay (Table 1) also significantly increased Nape-pld activity in RAW264.7 cells in a concentration-dependent manner (Figure S3), with a correlation observed between the efficacy in the biochemical assay and their efficacy in RAW264.7 cells (Figure 3B). To confirm that the effect seen in RAW264.7 cells was due to Nape-pld modulation, we tested the effects of bithionol, a known irreversible inhibitor of NAPE-PLD.14 The increase in the RAW264.7 cellular Nape-pld activity induced by 20 μM VU534 was blocked in a concentration-dependent manner by bithionol (Figure 3C). These results are consistent with enzyme activation of VU534 on PED-A1 hydrolysis being dependent on Nape-pld rather than an alternate modulating pathway.

Figure 3.

Figure 3

Open in a new tab

VU534 and VU533 increase the NAPE-PLD activity of RAW264.7 macrophages. (A) Effect of graded concentrations of VU534 (left panel), VU533 (middle panel), and VU233 (right panel) on Nape-pld activity in RAW264.7 cells, measured using PED-A1. Each compound was tested on at least two separate days and the individual replicates from each day normalized to vehicle control and then combined (mean ± SEM, n = 4–11). 1-way ANOVA p < 0.0001 for VU534, p < 0.0001 for VU533, and p = not significant for VU233. *p < 0.05 vs 0 μM, Dunnet’s multiple comparison test. Nonlinear regression with variable slope (four parameter) was used to calculate EC50 and Emax for VU534 EC50 6.6 μM (95% CI 2.6 to 11.2 μM), Emax 1.6-fold (95% CI n.d.); and VU533 EC50 2.5 μM (95% CI 1.4 to 6.1 μM), Emax 2.2-fold (95% CI 2.0 to 2.7-fold). (B) Correlation between maximal efficacy of 19 BT-PSPs and analogs in the recombinant Nape-pld assay and their efficacy at 30 μM in cultured RAW264.7 cells. Simple linear regression. Slope = 0.5455, R2 0.3394 p = 0.0089 for slope significantly nonzero. (C) Bithionol, a Nape-pld inhibitor, blocks increased Nape-pld activity in RAW264.7 cells induced by VU534. 1-way ANOVA p < 0.0001 Sidak’s multiple comparison test, ap < 0.05 vs 0 μM Bith with 0 μM VU534 group, bp < 0.05 vs 0 μM Bith with 20 μM VU534 group.

We then determined the effects of NAPE-PLD modulators on purified human NAPE-PLD and human-derived culture cells. Both VU534 and VU533 invoked concentration-dependent increases in the activity of recombinant human NAPE-PLD, while the inactive VU233 had no significant effect (Figure 4A). The potency of VU534 and VU533 for activating human NAPE-PLD was somewhat less than that for activating mouse Nape-pld (Table 1). Other benzothiazoles (Table 1) also activated recombinant human NAPE-PLD (Figure S4). VU534 and VU533, but not the inactive compound VU233, increased the NAPE-PLD activity in HepG2 cells in a concentration-dependent manner (Figure 4B).

Figure 4.

Figure 4

Open in a new tab

VU534 and VU533 activate human NAPE-PLD. (A) Effect of graded concentrations of VU534, VU533, and VU233 on activity of recombinant human NAPE-PLD with PED-A1 as substrate. Mean ± SEM, n = 3. Nonlinear regression with variable slope (four parameter) was used to calculate EC50 and Emax. VU534 EC50 0.93 μM (95% CI 0.63 to 1.39 μM), Emax 1.8-fold (95% CI 1.8 to 1.9-fold); VU533 EC50 0.20 μM (95% CI 0.12 to 0.32 μM), Emax 1.9-fold (95% CI 1.8 to 2.0-fold); VU233 not calculable. (B) Effect of graded concentrations of VU534, VU533, and VU233 on NAPE-PLD activity of HepG2 cells measured using flame-NAPE as substrate. Each compound was tested on two separate days, and the individual replicates from each day normalized to vehicle control and then combined (mean ± SEM, n = 4–6); VU534 EC50 1.5 μM (95% CI 0.6 to 2.8 μM), Emax 1.6-fold activity (95% CI 1.5 to 1.8-fold); VU533 EC50 3.0 μM (1.4 to 5.7 μM), Emax 1.6-fold (95% CI 1.5 to 1.8) fold; VU233 not calculable.

Biochemical Characterization of Lead NAPE-PLD Activators

To confirm the results of our fluorescence-based assays, we employed an orthogonal biochemical Nape-pld assay based on LC/MS, similar to those employed previously.13,16,17 Recombinant mouse Nape-pld was pretreated with VU534, VU533, VU233, or vehicle for 30 min, then N-oleoyl-phosphatidylethanolamine (NOPE) was added for 90 min, and the resulting levels of OEA and NOPE measured by LC/MS. Both activators VU534 and VU533 significantly increased the OEA/NOPE ratio, while inactive VU233 had no significant effect (Figure 5A). To further characterize the effect of VU534 on Nape-pld activity, we performed a Michaelis–Menten analysis using recombinant mouse Nape-pld and graded concentrations of flame-NAPE, a selective NAPE-PLD fluorogenic substrate resistant to competing esterase activity.18 Compound VU534 lowered the K1/2 (12.4 μM with vehicle vs 5.9 μM with VU534) and increased the maximal velocity (227 RFU/min with vehicle vs 421 RFU/min with VU534) of Nape-pld (Figure 5B). To determine if VU534 reversibly activated Nape-pld, we performed a rapid dilution assay. Concentrated and diluted enzyme solutions of Nape-pld were incubated with or without 1 μM VU534 for 1 h. The concentrated enzyme solutions were then diluted to the same enzyme concentration as the initially diluted samples using reaction buffer and then flame-NAPE added immediately to all samples to measure Nape-pld activity. After this rapid dilution, VU534 failed to increase Nape-pld activity (Figure 5C), consistent with VU534 reversibly binding to Nape-pld.

Figure 5.

Figure 5

Open in a new tab

Additional characterization of Nape-pld activators. (A) Activity of recombinant mouse Nape-pld using N-oleoyl-phosphatidylethanolamine (NOPE) as substrate and measuring OEA and NOPE by LC/MS/MS. Ratio of OEA to NAPE was normalized to 0 μM compound control. The assays of VU534 and VU233 were performed using the same 0 μM compound replicates. 1-way ANOVA VU534p < 0.0001, VU533p = 0.007, VU233p = 0.0547; Dunnett’s multiple comparison test for individual compounds **p = 0.0074, ***p = 0.0005, ****p < 0.0001. (B) Michaelis–Menten analysis using flame-NAPE as a substrate for recombinant mouse Nape-pld with or without VU534. Nonlinear regression curves (allosteric sigmoidal) were used to calculate K1/2 and Vmax. (C) Effect of rapid dilution on VU534-induced activation. Concentrated (40×) and diluted (1×) Nape-pld solutions were incubated with or without 1 μM VU534 for 1 h. The concentrated enzyme solutions were then diluted to 1× and Nape-pld activity immediately determined using flame-NAPE. Values were normalized to mean of initially dilute vehicle treated group (n = 4 replicates per group, mean ± SEM). 1-way ANOVA p < 0.0001. Groups sharing letters do not significantly differ from each other in Sidak’s multiple comparisons test.

To characterize the site of the interaction of VU534 with Nape-pld, we performed a series of competition experiments with other compounds previously characterized to interact with Nape-pld. Phosphatidylethanolamine (PE) and deoxycholic acid (DCA) are endogenous lipids that may interact with Nape-pld within the cellular environment. In the absence of detergent, both PE and DCA have been reported to increase the activity of recombinant Nape-pld,19–21 while in the presence of optimized concentrations of detergent (e.g., 0.1% Triton X-100), they have been reported to slightly decrease its activity.15,19 The solved crystal structure of NAPE-PLD found phosphatidylethanolamine (PE) bound immediately adjacent to the catalytic zinc ions and deoxycholic acid (DCA) bound in a large hydrophobic cavity that opens on to the catalytic site.19 The contradictory effect of PE and DCA may be rationalized by nonspecific binding of PE and DCA helping to solubilize the enzyme in the absence of detergent and their specific binding near the catalytic site when the enzyme is already solubilized inhibiting NAPE access to the catalytic site. Because our screening studies were performed in the presence of optimized detergent concentrations (0.4% NOG), our activators did not act by enhancing the solubility of the enzyme. In the presence of 0.4% NOG, there was a slight, but nonsignificant trend for inhibition of Nape-pld activity by 100 μM PE and 1 mM DCA. The efficacy of VU534 to enhance Nape-pld activity was somewhat reduced in the presence of 100 μM PE compared to vehicle but was still highly significant (Figure 6A). There was also a slight but nonsignificant trend toward reduced efficacy for VU534 in the presence of DCA (Figure 6A).

Figure 6.

Figure 6

Open in a new tab

Effect of known Nape-pld modulators on VU534-induced activation of Nape-pld. (A) Effect of 0.1 mM phosphatidylethanolamine (PE) and 1 mM deoxycholic acid (DCA) on Nape-pld activity in the presence and absence of 1 μM VU534. 1-way ANOVA p < 0.0001. Groups sharing letters do not significantly differ from each other in Sidak’s multiple comparisons test. (B) Effect of the reversible NAPE-PLD inhibitor LEI-401 on the VU534 concentration response curve. Vehicle only pretreatment EC50 0.54 μM (95% CI 0.37–0.78 μM), Emax 2.3 (95% CI 2.2–2.5); versus 30 μM LEI-401 pretreatment EC50 1.7 μM (95% CI 1.41–2.17 μM), Emax 1.9 (95% CI 1.8–2.0). (C) Effect of irreversible NAPE-PLD inhibitor bithionol on the VU534 concentration response curve. Vehicle only pretreatment EC50 0.11 μM (95% CI 0.03–0.32 μM), Emax 1.6 (95% CI 1.4–1.7); versus 3 μM Bith pretreatment EC50 0.38 μM (95% CI 0.31–0.49 μM) Emax 0.9 (95% CI 0.8–0.9). (D) Effect of VU233 on the VU534 concentration response curve. Vehicle only pretreatment EC50 veh 0.06 μM (95% CI 0.04–0.07 μM), Emax 2.3 (2.2–2.3); vs EC50 15 μM VU233 pretreatment EC50 0.21 μM (95% CI 0.15–0.29 μM), Emax 2.3 (95% CI 2.2–2.4). For (B,C), n = 4 replicates per group.

LEI-401 and bithionol are reversible and irreversible inhibitors of Nape-pld, respectively, whose binding sites on Nape-pld remain to be characterized.22 Pretreatment with either 30 μM LEI-401 (Figure 6B) or 3 μM bithionol (Figure 6C), markedly reduced the efficacy of VU534 to increase Nape-pld activity but only modestly reduced its potency, consistent with VU534 being unable to competitively displace these inhibitors from Nape-pld. Pretreatment with inactive compound VU233 had no effect on the efficacy of VU534 to increase Nape-pld activity but slightly reduced its potency (Figure 6D), consistent with VU233 and VU534 competing for the same site. Both LEI-401 and VU233 reduced the ability of 5 μM VU534 to activate Nape-pld in a concentration-dependent manner (Figure S5). Altogether, these data are most consistent with binding of VU534 at an allosteric site on Nape-pld distinct from that of PE, DCA, or LEI-401.

A survey of the literature revealed that a series of benzothiazoles with structural features somewhat similar to VU534 and VU533 have been developed as dual inhibitors of fatty acid amide hydrolase (FAAH) and soluble epoxide hydrolase (sEH),23 although the “left-side fragment” of that series of compounds were benzothiazole-phenyl moieties rather than the benzothiazole moieties present in our series. Specifically, compounds 4–14 of that chemical series share the greatest similarity of VU534 and VU533, as the “aromatic right-side fragment” of all three compounds is the same para-fluoro- phenylsulfonyl moiety (Figure S6). Compound 4–14 was reported to inhibit both FAAH (IC50 = 77 nM) and sEH (IC50 = 1300 nM). Because inhibition of FAAH would increase cellular NAE levels independently of NAPE-PLD (Figure 1), we tested whether our activators modulated FAAH activity. Graded concentrations of VU534, VU533, and VU233 showed only weak inhibition of the FAAH activity (Figure 7A). We also tested whether our activators modulated sEH activity. While sEH does not lie in the biochemical pathway for NAE biosynthesis or metabolism, inhibition or genetic ablation of sEH increases the levels of epoxy fatty acids24 and, thereby, exerts biological effects similar to the known effects of NAEs such as reducing obesity, cardiovascular disease, pain, and inflammation.24–27 Graded concentrations of VU534 showed modest inhibition of sEH (IC50 = 1.2 μM, 95% CI = 0.5–2.4 μM, maximal inhibition = 55%), while neither VU533 nor VU233 significantly inhibited sEH (Figure 7B). Other compounds of our series showed variable effects on FAAH and sEH activity (Figures S7 and S8). These modest off-target effects should not significantly interfere with the use of these compounds to probe the contribution of Nape-pld activity as long as bona fide sEH inhibitors are also tested as controls.

Figure 7.

Figure 7

Open in a new tab

Evaluation of off-target effects on FAAH and sEH. (A) Effects of graded concentrations of VU534, VU533, or VU233 on activity of fatty acid amide hydrolase (FAAH). 1-way ANOVA VU534p = 0.0007, VU533p < 0.0001, and VU233p < 0.0001; *p < 0.05 vs 0 μM, Dunnett’s multiple comparison test for individual compounds. (B) Effects of graded concentrations of VU534, VU533, or VU233 on activity of soluble epoxide hydrolase (sEH). 1-way ANOVA VU534p < 0.0001, VU533p < 0.0001, and VU233p < 0.0001; *p < 0.05 vs 0 μM, Dunnett’s multiple comparison test for individual compounds.

Modulation of Nape-Pld Activity Modulates Efferocytosis by Macrophages

Rinne et al. showed that PEA, a Nape-pld product, enhanced the ability of bone-marrow derived macrophages (BMDM) to carry out efferocytosis.8 We, therefore, assessed the effects of Nape-pld modulation on efferocytosis by BMDM. Efferocytosis was measured by proportion of BMDM that took up labeled apoptotic Jurkat T cells during a 45 min coincubation, using flow cytometry to measure this uptake. Treatment of BMDM with 10 μM of NAPE-PLD inhibitor bithionol (Bith) for 6 h prior to initiating the efferocytosis assay markedly reduced efferocytosis compared to vehicle treated BMDM, while treatment with 10 μM of either NAPE-PLD activators VU534 or VU533 significantly enhanced efferocytosis (Figure 8A). In contrast, treatment with 10 μM compound VU233 modestly reduced efferocytosis. Given the modest inhibitory effects of compound VU534 on sEH, we examined the effect of two bona fide sEH inhibitors, AUDA28 and TPPU,29 on efferocytosis. Neither AUDA nor TPPU significantly enhanced efferocytosis (Figure 8B), indicating that sEH inhibition was not responsible for the effect of VU534 on efferocytosis. To further assess the contribution of Nape-pld modulation on efferocytosis, we isolated BMDM from wild-type (WT) and Napepld–/– (KO) mice and measured the extent of efferocytosis in the presence and absence of activator VU534. KO BMDM treated with vehicle (Veh) had significantly reduced efferocytosis compared with WT BMDM treated with Veh (Figure 8C). In WT BMDM, treatment with activator VU534 significantly increased efferocytosis but, in KO BMDM, treatment with activator VU534 had no effect. Thus, NAPE-PLD appears to play a critical role in maximizing the efferocytosis capacity of macrophages, and activator VU534 enhances efferocytosis in a NAPE-PLD-dependent manner.

Figure 8.

Figure 8

Open in a new tab

Modulation of NAPE-PLD modulates efferocytosis by macrophages. (A) BMDM from wild-type mice were treated with 10 μM VU534, VU533, VU233, or bithionol (Bith) for 6 h prior to initiation of efferocytosis assay. 1-way ANOVA p = 0.0004. Dunnett’s multiple comparison’s test p value is shown for each comparison. Data are shown from one representative experimental day. All compounds were tested in two-six separate experimental days. (B) BMDM from wild-type mice was treated with 10 μM VU534 or with sEH inhibitor AUDA (10 μM) or TPPU (10 μM) for 6 h prior to initiation of efferocytosis assay. 1-way ANOVA p = 0.0004. Dunnett’s multiple comparisons test p value shown for each comparison. Data shown from one representative experimental day of two total experimental days. (C) BMDM from wild-type (WT) or Napepld–/– (KO) mice were treated with vehicle (veh) or 10 μM VU534 for 6 h prior to initiation of the efferocytosis assay. 1-way ANOVA p < 0.0001, Dunnett’s multiple comparison p value shown for each comparison. Data shown from one representative experimental day of three total experimental days.

These studies demonstrate that select benzothiazole phenylsulfonyl-piperidine carboxamides significantly increase the activity of both mouse and human NAPE-PLD and that increasing NAPE-PLD activity increases efferocytosis by macrophages. The two most potent of the compound series tested, VU534 and VU533, incorporate a para-fluoro-group in the phenylsulfonyl moiety and dimethyl substitution of the benzothiazole aromatic ring. This preliminary SAR suggests that improved activators may be identified from a focused lead optimization campaign. Most of the tested analogs have minimal cytotoxicity against RAW264.7 and HepG2 cells, and their efficacy to enhance cellular Nape-pld activity correlates with their efficacy to enhance activity in the biochemical assay with purified recombinant Nape-pld. Although our lead series share structural features with some previously developed dual FAAH and sEH inhibitors,23 they show little inhibition of FAAH and only modest inhibition of sEH. Therefore, activators VU534 and VU533 should be useful tool compounds to assess the contribution of NAPE-PLD to various biological processes in cultured cells.

Our studies provide further evidence that the NAPE-PLD/NAE signaling pathway plays a critical role in regulating efferocytosis. While NAPEPLD expression was previously shown to be significantly reduced in unstable atherosclerotic plaques from human coronary arteries, and impaired efferocytosis has been implicated in the development of these plaques,8 our studies are the first to demonstrate that deletion of Nape-pld markedly diminished the ability of macrophages to carry out efferocytosis. Furthermore, we found that inhibiting Nape-pld activity using bithionol phenocopied the effect of Napepld deletion. Consistent with the previous finding that treatment of BMDM with PEA enhances efferocytosis,8 we found that treatment of wild-type BMDM with either VU534 or VU533 to increase Nape-pld activity also enhanced efferocytosis. This increase in efferocytosis required Nape-pld, as VU534 failed to enhance efferocytosis in Napepld–/– BMDM.

Previously described effects of NAEs suggest some mechanisms by which increased NAPE-PLD activity could enhance efferocytosis. PEA acts via Gpr55 to increase the expression of MerTK,8 a receptor that helps macrophages recognize and bind to apoptotic cells.1,2 Deletion of MerTK in macrophages markedly enhances necrotic core expansion in Apoe–/– mice.30 OEA also acts via PPARα to increase the expression of CD206 and TGFβ, two classic markers of the M2 macrophage phenotype with enhanced efferocytosis.10 A more complete elucidation of how NAPE-PLD regulates efferocytosis will require a variety of approaches, including the use of both NAPE-PLD inhibitors and activators. Efferocytosis is a complex process involving macrophage recognition of so-called “find me” and “eat me” signals, and requires the binding, internalization, and controlled degradation of apoptotic cells, followed by export of their constituent components like cholesterol,1,2,31 so the effect of NAPE-PLD modulation on each of these steps needs to be examined. NAPEs exert membrane-stabilizing effects32,33 and facilitate the lateral diffusion of cholesterol,34 while phosphatidic acids exert membrane-bending effects.35 Therefore, the effect of increased NAPE-PLD activity on the membrane topology as a mechanism to enhance efferocytosis also needs to be examined.

Although defective efferocytosis has been implicated in the progression to unstable atherosclerotic plaques,1,2 whether the enhanced efferocytosis induced by NAPE-PLD activators can translate to improved efferocytosis under atherogenic conditions requires future studies. Previous trials administering OEA, PEA, or NAE-boosting bacteria decrease the size of necrotic cores with atherosclerotic lesions.8,10,11 The poor pharmacokinetic properties of OEA and PEA have hampered their clinical use, while engineered bacteria that produce these bioactive lipids in situ still face significant regulatory hurdles for use for humans. While still at a very early stage, our work demonstrates that BT-PSP-based NAPE-PLD activators represent a potential alternative strategy to raise NAE levels and, thereby, achieve these same effects. It is worth noting that impaired efferocytosis has been implicated in a number of diseases besides atherosclerosis, including systemic lupus erythematosus, neurodegenerative diseases, retinal degeneration, pulmonary disorders, liver diseases, diabetes, inflammatory bowel disease, colon carcinoma, impaired wound healing, and rheumatoid arthritis.2 Therefore, future studies could also examine whether NAPE-PLD activators can protect against their development or progression. The value of NAPE-PLD activators as a therapeutic intervention may also extend beyond conditions of defective efferocytosis. For instance, Nape-pld expression and NAE levels are rapidly reduced by feeding a high-fat diet6–8 and administering OEA or PEA or their precursor NAPEs can markedly blunt the obesity, glucose intolerance, inflammation, and hepatosteatosis that results from these high-fat diets.8–9,12 Thus, future testing of NAPE-PLD activators should examine their potential to treat these conditions, as well.

Methods

Materials

Initial stocks of potential NAPE-PLD modulator compounds were purchased from Life Chemicals and provided by the Vanderbilt HTS screening facility. Additional compounds were synthesized by the Vanderbilt Chemical Synthesis core (Supporting Information on the Synthesis of Compounds). LEI-401, [2H4]PEA, and [2H4]OEA were purchased from Cayman Chemicals. N-palmitoyl-PE, 1,2-dioleoyl-PE and 1,2-dihexanoyl-PE were purchased from Avanti Polar Lipids. PED-A1 was purchased from Invitrogen. Flame-NAPE was synthesized as previously described.18 [2H4]N-palmitoyl-PE was synthesized using [2H4] palmitic acid (Cambridge Isotope Laboratories) and 1,2-dioleoyl-PE and N-oleoyl-PE were synthesized using 1,2-dihexanoyl-PE and oleoyl chloride (MilliporeSigma) (Supporting Information-Synthesis of Compounds). Recombinant mouse Nape-pld with a C-terminal hexahistidine tag was expressed in Escherichia coli and purified using cobalt affinity beads as previously described.14 The expression plasmid including the full-length human NAPEPLD gene with a C-terminal hexahistidine tag inserted in a pET plasmid was purchased from VectorBuilder, and the protein was expressed and purified in a manner identical to that of recombinant mouse Nape-pld. The sEH inhibitors, TPPU (N-[1-(1-Oxopropyl)-4-piperidinyl]-N′-[4-(trifluoromethoxy)phenyl]urea) and AUDA (12-[[(tricyclo[3.3.1.13,7]dec-1-ylamino)carbonyl]amino]-dodecanoic acid), were purchased from Cayman Chemical and Sigma Chemicals, respectively. The soluble epoxide hydrolase Inhibitor screening kit (item no. 10011671) and the fatty acid amide hydrolase inhibitor screening kit (item no. 10005196) were purchased from Cayman Chemical.

Biochemical Nape-pld Assays with Recombinant Enzyme

In vitro fluorescence Nape-pld activity assays using recombinant Nape-pld and either PED-A1 or flame-NAPE as fluorogenic substrate were performed as previously described,14,18 except with small modifications as noted below.

HTS Assays

For the HTS assays, test compounds were incubated with recombinant enzyme for 1 h prior to adding PED-A1 (final concentration of 0.4 μM) mixed with N-palmitoyl-dioleoyl-PE (final concentration of 3.6 μM) to adjust for the high sensitivity of the Panoptic instrument (WaveFront Biosciences). Assays used black-wall, clear-bottom, nonsterile, and nontreated 384-well plates (Greiner Bio-One 781906). The assay was read in kinetic fluorescence mode (488 nm excitation/530 nm emission) on the Panoptic instrument for 4 min, and the slope of the signal from 30 to 100 s was used for analysis. A total of 39,328 compounds from the Vanderbilt Discovery Collection were tested, each at 10 μM. The tested compounds were chosen to represent a structurally diverse selection from the full library. Each 384-well plate included 320 test compound wells and 64 control wells. Unlike our previous pilot screening assay,14 we used bithionol (10 μM final) in place of lithocholic acid (100 μM final) as the inhibitor control. Before performing high-throughput screening, a checkerboard assay was performed to validate the assay parameters.14,36 This yielded a Z′ score of 0.676. Z′ scores were also calculated for each plate during screening, and plates with scores <0.5 were rerun. The average Z′ across all screening plates was 0.52, and the total hit rate was 3.6%. B-scores were calculated from the initial slopes across each plate using WaveGuide software (WaveFront Biosiences).36 Modulator hits were defined as compounds with absolute B-scores of 3 or higher. The number of compounds in various B-score ranges were as follows: −21 to −10, 12 compounds; −10 to −5, 221 compounds; −5 to −3, 770 compounds; −3 to 3, 37,924 compounds; 3 to 4, 314 compounds; 4 to 6, 70 compounds; and 6 to 10, 15 compounds. Activator hits with B-scores of ≥3 were selected for the replication assay and a selection of analogs of the activators.

Concentration Response Curve Studies

Concentration response curve (CRC) studies shown in Table 1 and Figure 4 used the same assay conditions as the HTS assay, except that graded concentrations of each test compound were used, with total amount of vehicle (DMSO) kept constant. CRC experiments with purified recombinant mouse Nape-pld were performed on two separate days, with values from each day normalized to vehicle only controls on the same plate and then all normalized values from both days averaged together. CRC experiments with human NAPE-PLD represent a value from only a single day, due to limited amounts of this recombinant enzyme.

LC/MS Assays

N-oleoyl-PE (22.5 nmol) was added as substrate after rNAPE-PLD (0.57 μg per reaction) was preincubated for 1 h with 0, 10, or 20 μM compound VU534, VU533, VU233, or vehicle. 90 min after N-oleoyl-PE was added, the reaction was quenched by adding 3 volumes of ice-cold methanol containing 0.67 nnmol [2H4]OEA and 1.3 nmol [2H4]N-palmitoyl-PE and then 6 volumes of ice-cold chloroform.37 The lower phase was dried under nitrogen gas and dissolved in 100 μL of mobile phase A and sample injected for analysis by high performance liquid chromatography tandem mass spectrometry (LC/MS/MS) using an Acquity UPLC coupled to a ThermoQuantum triple quadrupole mass spectrometer operating in multiple reaction monitoring mode. Chromatography utilized a 2.1 mm C18 guard column (Phenomenex AJ0-8782) and a rapid gradient ramp. Mobile phase A was 5:1:4 (v/v/v) isopropanol/methanol/water, with 0.2% v/v formic acid, 0.66 mM ammonium formate, and 3 μM phosphoric acid included as additives. Mobile phase B was 0.2% (v/v) formic acid in isopropanol. Initial column conditions were 5% mobile phase B, followed by gradient ramp to 95% B over 0.5 min, held at 95% B for 2 min, and then returned to initial conditions (5% B) over 1 min. Flow rate throughout was 100 μL/min. Injection volume was 2 μL. The sample injector needle was washed before each injection using a strong wash of methanol and a weak wash of 1:1:1:1 (v/v/v/v) isopropanol: methanol: acetonitrile: water, with 0.2% formic acid, 0.3 mM ammonium formate, and 0.37 mM phosphoric acid included as additives. Multiple reaction monitoring was carried out for the following ion transitions: OEA [M + H]+: m/z 326.3 → m/z 62.1; [2H4]OEA [M + H]+: m/z 330.3 → m/z 66.1; N- oleoyl-PE [M + NH4]+: m/z 693.5 → m/z 308.3 (quantifier), m/z 693.5 → m/z 271.2 (qualifier); [2H4]N-palmitoyl-PE [M + NH4]+: m/z 1003.8 → m/z 286.3 (quantifier), m/z 1003.8 → m/z 603.5 (qualifier). Chromatographs were collected, and the resulting peaks were analyzed utilizing Thermo XCaliber software. The ratio of peak height for OEA to [2H4]OEA was used to calculate the amount of OEA in the sample and the ratio of peak height for N-oleoyl-PE to [2H4]N-palmitoyl-PE was used to calculate the amount of N-oleoyl-PE in the sample. These values were then used to calculate the OEA/N-oleoyl-PE ratio for the sample. This ratio was then normalized to the average OEA/N-oleoyl-PE ratio for the 0 μM activator control.

Michaelis–Menten Assays

rNape-pld was incubated with 10 μM VU534 or vehicle (0.2% final concentration of DMSO) for 60 min and then added with reaction buffer to individual wells of black-walled clear-bottom 96-well plates (final concentration 0.4 μg enzyme, 50 mM Tris, 0.4% N-octylglucoside) where graded concentrations of flame-NAPE had been added. The resulting fluorescence (488 nm excitation/530 nm emission) was read in a BioTek Synergy H1 plate reader with the slope of the fluorescence signal from 0 to 4 min (linear phase) used to determine enzyme reaction rate (ΔRFU/min) and the nonlinear regression of flame-NAPE concentration vs reaction rate performed using GraphPad Prism 9.

VU534 Competition Assays with Compounds Known to Bind Nape-pld

In individual wells of black-walled clear-bottom 96-well plates, rNAPE-PLD enzyme solution (final concentration in assay 0.4 μg enzyme, 50 mM Tris, 0.4% N-octylglucoside) was preincubated with the appropriate compound or its vehicle. Final concentrations of dioleoylphosphatidylethanolamine (PE), deoxycholic acid (DCA), biothionol (Bith), LEI-401, and VU233 were 100 μM, 1 mM, 3 μM, 30 μM, and 15 μM, respectively. For PE, DCA, Bith, and LEI-401, preincubation was for 30 min, and for VU233 preincubation was 5 min. Graded concentrations of VU534 were then added and allowed to incubate with an enzyme solution for 60 min. A substrate mixture of flame-NAPE (0.5 μM final concentration) and N-oleoyl-PE (9.5 μM final concentration) was then added to enzyme solution, and fluorescence (488 nm excitation/530 nm emission) was measured in a BioTek Synergy H1 plate reader with the change in fluorescence over 5 min used to measure Nape-pld activity. For each compound, Nape-pld activity was normalized to the average value of the vehicle pretreated, 0 μM VU534 samples. GraphPad Prism 9 was used to calculate the VU534 concentration response curves in the presence of each potential competitor compound. All experiments were repeated on at least two separate days, and the results from one representative day shown.

Rapid Dilution Assay

Replicate concentrated (40×) solutions of rNape-pld (0.4 μg of enzyme in 5 μL of 50 mM Tris HCl with 0.4% NOG) or diluted (1×) solutions of rNape-pld (0.4 μg in 195 μL of 50 mM Tris HCl with 0.4% NOG) were incubated with or without 1 μM VU534 (initial concentration) in individual wells of a black-walled, clear-bottom 96-well plate for 60 min 190 μL of reaction buffer (50 mM Tris HCl with 0.4% NOG) was then added to the initially 40x concentrated solutions of rNape-pld. Immediately following this dilution, 5 μL of NAPE substrate mixture (0.5 μM flame-NAPE and 9.5 μM N-oleoyl-PE, final concentration) was added to both the initially concentrated and the initially diluted sample wells. The resulting fluorescence (488 nm excitation/530 nm emission) was read in a BioTek Synergy H1 plate reader with the fluorescence after 5 min used to determine Nape-pld activity. Values from individual wells were normalized, with 100% Nape-pld activity being the average value of the group with the enzyme initially diluted and no VU534. Experiments were repeated on two separate days, and the results from one representative day shown.

Other Biochemical Assays

Nonspecific Fluorescence Modulators Screen

To identify false HTS hits that modulated the fluorescence of the BODIPY moiety itself (rather than altering rate of PED-A1 hydrolysis by rNape-pld), potential hit compounds were incubated with BODIPY-FL C5 (ThermoFisher Scientific), a BODIPY-labeled free fatty acid, rather than PED-A1 and their effect on fluorescence measured (488 nm excitation/530 nm emission). One potential hit compound from the original HTS screen directly modulated fluorescence was therefore eliminated from further evaluation (Figure S9).

Soluble Epoxide Hydrolase Activity Assay

Assays were performed according to the manufacturer’s (Cayman Chemical) specifications. Vehicle (DMSO), graded concentrations of activator compound (1, 3.3, 10, and 33 μM final concentration), or AUDA (1 μM final concentration) were incubated with the sEH enzyme solution for 5 min prior to addition of sEH substrate, with fluorescence read at 330 nm excitation/465 nm emission. Values from all wells were normalized, with 0% inhibition set as the average of the vehicle treated wells and 100% inhibition set as the average of the AUDA-treated wells.

Fatty Acid Amide Hydrolase Activity Assay

Assays were performed according to manufacturer’s (Cayman Chemical) specifications. Vehicle (DMSO), graded concentrations of activator compound (1, 3.3, 10, and 33 μM final concentration), or FAAH inhibitor JZL195 (1 μM final concentration) were incubated with the FAAH enzyme solution for 5 min prior to addition of FAAH substrate, with fluorescence read at 350 nm excitation/460 nm emission. Values from all wells were normalized, with 0% inhibition set as the average of the vehicle treated wells and 100% inhibition set as the average of the JZL195 treated wells.

Cell-Based Nape-Pld Assays

NAPE-PLD Activity

NAPE-PLD activity was measured in cells cultured in sterile 96-well plates as previously described,18 except that FluoroBrite DMEM (Gibco A1896701) was used as media. Activator compounds were incubated with cells for 1 h prior to the addition of PED-A1 or flame-NAPE to measure NAPE-PLD activity. For RAW264.7 assays, 10 μM orlistat was added (to inhibit PLA1 activity18) at the same time as activator and then Nape-pld activity measured using PED-A1 (3.6 μM final). For studies to assess the effect of NAPE-PLD inhibitor bithionol on the ability of VU534 to increase Nape-pld activity in RAW264.7 cells, 10 μM orlistat and graded concentrations of biothionol (0, 1, 5, and 15 μM final concentration) with or without 20 μM VU534 were incubated with cells for 1 h prior to the addition of PED-A1. For HepG2 cells, no orlistat was used, and NAPE-PLD activity was measured using flame-NAPE was used. Fluorescence (488 nm excitation/530 nm emission) was read in a BioTek Synergy H1 plate reader.

Cytotoxicity

Cytotoxicity was measured using MTT as previously described14 except that the studies used 96-well plates with 100 μL of 0.3 mg mL–1 MTT solution was added after 24 h of treatment and then replaced after 3 h with 0.1 M HCl in isopropanol. Viability was expressed as the percent absorbance at 560 nm relative to vehicle controls.

Efferocytosis Assays

Male C57BL6/j wild-type or Napepld–/– mice38 were euthanized with isoflurane and hind legs were removed. Marrow was flushed from the femurs and tibias using DMEM containing 4.5 g/L glucose and a 26-gauge needle. Cell suspensions were passed over a 40 μm filter, centrifuged at 500g, and resuspended in 50 mL of DMEM containing 4.5 g/L glucose, 20% L-cell conditioned media, 10% heat-inactivated FBS, and 1% penicillin/streptomycin. Ten ml of cell suspension was plated into each of five 100 mm dishes and incubated for 4 days at 37 °C and 5% CO2. On day four, nonadherent cells and debris were aspirated from the plates and replaced with fresh media. After 7 days of differentiation, cells were harvested for use in experiments.

Assays were performed according to previously established protocols.39,40 Bone marrow-derived macrophages were seeded at 0.25 × 106 cells/well in a nontissue-culture-treated 24-well plate and allowed to adhere overnight. Macrophages were treated with various compounds at a final concentration of 10 μM or DMSO as a vehicle for 6 h prior to each experiment. Jurkat cells were exposed to UV light (254 nm) for 5 min to induce apoptosis and then incubated in a 37 °C incubator with 5% CO2 for 2 h. Surveillance staining of these cells routinely yields approximately 80–90% apoptosis (Annexin V+) using this method. Apoptotic Jurkat cells were labeled with either CellVue Claret (Millipore Sigma) or Cell Trace Violet (Invitrogen) according to the manufacturer’s instructions. After staining, cells were resuspended in macrophage medium at a density of 0.75 × 106 cells/ml and 500 μL of this suspension was added to the drug-containing media on the macrophages to achieve a cell ratio of 3:1 Jurkats:macrophages. After incubating for 45 min at 37 °C and 5% CO2, the medium was aspirated, and the macrophages were gently washed twice with PBS to remove unbound apoptotic cells. Macrophages were then removed from the plate using Cell Stripper (Sigma), washed, resuspended in staining buffer consisting of 2% FBS in PBS with 2 mM EDTA, and blocked with antimouse CD16/32 antibodies for 15 min on ice. After blocking, cells were pelleted and resuspended with anti-F4/80 antibody. Cells were incubated for 45 min on ice in the dark, then washed and resuspended in staining buffer for analysis. Cells were analyzed using an Attune NxT cytometer (ThermoFisher) and data were analyzed using FlowJo software to quantify the proportion of total F4/80+ macrophage population that also costained for apoptotic cells (Cell Trace Violet or CellVue Claret). The % efferocytosis was set as the % of the F4/80+ cells that costained with the apoptotic cell marker.

Statistical Analyses

All statistical analyses and nonlinear regression analyses were performed using GraphPad Prism 9 software, except for calculation of B-scores for high throughput screening assays, which were calculated using WaveGuide software (WaveFront Biosiences).

Acknowledgments

This work was supported by National Institutes of Health Grants P01HL116263 (S.S.D.), T32GM065086 (J.E.Z.), U01DA056242 (K.M.), and 5R01DA047858 (K.M.); American Heart Association fellowship 835504 (J.E.Z.); a Discovery Grant from the Vanderbilt Diabetes Center (S.S.D.); the Vanderbilt Institute for Chemical Biology; and the Vanderbilt University Department of Pharmacology Academic Support (S.S.D.). Experiments were performed in the Vanderbilt High-Throughput Screening (HTS) Core Facility with assistance provided by P. Vinson, J. Bauer, and C. Whitwell. The Discovery Collection was distributed by the Vanderbilt HTS Core. The HTS Core receives support from the Vanderbilt Institute of Chemical Biology and the Vanderbilt Ingram Cancer Center (P30 CA68485). The WaveFront Biosciences Panoptic kinetic imaging plate reader is housed and managed within the Vanderbilt HTS Core Facility, an institutionally supported core, and was funded by National Institutes of Health Shared Instrumentation Grant 1S10OD021734. I. Romaine assisted in identifying additional compounds in the library with the BT-PSP core structure. K. Tallman synthesized and provided the [2H4]N-palmitoyl-PE. The Vanderbilt Mass Spectrometry Research Core receives support from Vanderbilt University.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acschembio.3c00401.

  • Additional experimental details, materials, and methods, including individual concentration response curves for biochemical and cellular assays (PDF)

Author Contributions

J.E.Z. performed the majority of experiments including HTS, bioactivity characterization studies, cellular assays, and selectivity studies. He also performed the synthesis of N-oleoyl-PE, analyzed data, created figures, interpreted results, and assisted in writing of the initial manuscript. A.-M.A.-O extracted and cultured BMDM, performed most of the competition assays and the rapid dilution assay, and assisted with efferocytosis studies. C.M.Y. performed efferocytosis studies and analysis of data. K.K. synthesized and characterized BT-PSP analogs and flame-NAPE. A.N.J. assisted in development and performance of LC/MS assays. M.R.J. synthesized BT-PSP analogs. Z.M. assisted with conception of the project, purification of enzyme, LC/MS assays, and cell culture studies. I.C.S. assisted with cellular and LC/MS assays. K.M. supervised the husbandry of Napepld–/– and control mice and femur extraction and provided guidance on Nape-pld biology to the project. A.G.W. provided guidance on medicinal chemistry to the project and edited the manuscript. A.C.D. assisted in conception of the project, supervised, and performed efferocytosis assays, provided guidance on macrophage biology to the project, created figures, and assisted in writing the initial manuscript. G.A.S. assisted with the conception of the project, supervised the synthesis and characterization of compounds, provided guidance of the project, obtained financial support, and edited the manuscript. S.S.D. conceived and guided the overall project, supervised various studies, obtained financial support, oversaw interpretation of the data, created figures, and wrote the manuscript. All authors reviewed the manuscript.

The authors declare the following competing financial interest(s): J.E.Z., K.K., A.W.G., A.M.D., G.A.S., and S.S.D. are named as inventors on a patent application for the use of benzothiazole phenylsulfonyl-piperidine carboxamides as small molecule NAPE-PLD activators. This work was funded in part by grants from the NIH/NHLBI P01HL116263 (S.S.D. and A.C.D.) and by an American Heart Association fellowship 835504 (J.E.Z.).

Supplementary Material

cb3c00401_si_001.pdf (432.5KB, pdf)

References

  1. Doran A. C.; Yurdagul A. Jr.; Tabas I. Efferocytosis in health and disease. Nat. Rev. Immunol. 2020, 20 (4), 254–267. 10.1038/s41577-019-0240-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Boada-Romero E.; Martinez J.; Heckmann B. L.; Green D. R. The clearance of dead cells by efferocytosis. Nat. Rev. Mol. Cell Biol. 2020, 21 (7), 398–414. 10.1038/s41580-020-0232-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bentzon J. F.; Otsuka F.; Virmani R.; Falk E. Mechanisms of plaque formation and rupture. Circ. Res. 2014, 114 (12), 1852–1866. 10.1161/CIRCRESAHA.114.302721. [DOI] [PubMed] [Google Scholar]
  4. Ueda N.; Okamoto Y.; Morishita J. N-acylphosphatidylethanolamine-hydrolyzing phospholipase D: a novel enzyme of the beta-lactamase fold family releasing anandamide and other N-acylethanolamines. Life Sci. 2005, 77 (14), 1750–1758. 10.1016/j.lfs.2005.05.018. [DOI] [PubMed] [Google Scholar]
  5. Wang J.; Okamoto Y.; Morishita J.; Tsuboi K.; Miyatake A.; Ueda N. Functional analysis of the purified anandamide-generating phospholipase D as a member of the metallo-beta-lactamase family. J. Biol. Chem. 2006, 281 (18), 12325–12335. 10.1074/jbc.M512359200. [DOI] [PubMed] [Google Scholar]
  6. Igarashi M.; DiPatrizio N. V.; Narayanaswami V.; Piomelli D. Feeding-induced oleoylethanolamide mobilization is disrupted in the gut of diet-induced obese rodents. Biochim. Biophys. Acta 2015, 1851 (9), 1218–1226. 10.1016/j.bbalip.2015.05.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Diep T. A.; Madsen A. N.; Holst B.; Kristiansen M. M.; Wellner N.; Hansen S. H.; Hansen H. S. Dietary fat decreases intestinal levels of the anorectic lipids through a fat sensor. FASEB J. 2011, 25 (2), 765–774. 10.1096/fj.10-166595. [DOI] [PubMed] [Google Scholar]
  8. Rinne P.; Guillamat-Prats R.; Rami M.; Bindila L.; Ring L.; Lyytikäinen L. P.; Raitoharju E.; Oksala N.; Lehtimäki T.; Weber C.; van der Vorst E. P. C.; Steffens S. Palmitoylethanolamide promotes a proresolving macrophage phenotype and attenuates atherosclerotic plaque formation. Arterioscler., Thromb., Vasc. Biol. 2018, 38 (11), 2562–2575. 10.1161/ATVBAHA.118.311185. [DOI] [PubMed] [Google Scholar]
  9. Fu J.; Oveisi F.; Gaetani S.; Lin E.; Piomelli D. Oleoylethanolamide, an endogenous PPAR-alpha agonist, lowers body weight and hyperlipidemia in obese rats. Neuropharmacology 2005, 48 (8), 1147–1153. 10.1016/j.neuropharm.2005.02.013. [DOI] [PubMed] [Google Scholar]
  10. Chen Z.; Zhuo R.; Zhao Y.; Yang L.; Zhou Y.; Cheng X.; Peng L.; Jin X.; Wang Y. Oleoylethanolamide stabilizes atherosclerotic plaque through regulating macrophage polarization via AMPK-PPARα pathway. Biochem. Biophys. Res. Commun. 2020, 524 (2), 308–316. 10.1016/j.bbrc.2020.01.103. [DOI] [PubMed] [Google Scholar]
  11. May-Zhang L. S.; Chen Z.; Dosoky N. S.; Yancey P. G.; Boyd K. L.; Hasty A. H.; Linton M. F.; Davies S. S. Administration of N-Acyl-Phosphatidylethanolamine expressing bacteria to Low Density Lipoprotein Receptor(−/−) mice improves Indices of cardiometabolic disease. Sci. Rep. 2019, 9 (1), 420. 10.1038/s41598-018-37373-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gillum M. P.; Zhang D.; Zhang X. M.; Erion D. M.; Jamison R. A.; Choi C.; Dong J.; Shanabrough M.; Duenas H. R.; Frederick D. W.; Hsiao J. J.; Horvath T. L.; Lo C. M.; Tso P.; Cline G. W.; Shulman G. I. N-acylphosphatidylethanolamine, a gut- derived circulating factor induced by fat ingestion, inhibits food intake. Cell 2008, 135 (5), 813–824. 10.1016/j.cell.2008.10.043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Mock E. D.; Mustafa M.; Gunduz-Cinar O.; Cinar R.; Petrie G. N.; Kantae V.; Di X.; Ogasawara D.; Varga Z. V.; Paloczi J.; Miliano C.; Donvito G.; van Esbroeck A. C. M.; van der Gracht A. M. F.; Kotsogianni I.; Park J. K.; Martella A.; van der Wel T.; Soethoudt M.; Jiang M.; Wendel T. J.; Janssen A. P. A.; Bakker A. T.; Donovan C. M.; Castillo L. I.; Florea B. I.; Wat J.; van den Hurk H.; Wittwer M.; Grether U.; Holmes A.; van Boeckel C. A. A.; Hankemeier T.; Cravatt B. F.; Buczynski M. W.; Hill M. N.; Pacher P.; Lichtman A. H.; van der Stelt M. Discovery of a NAPE-PLD inhibitor that modulates emotional behavior in mice. Nat. Chem. Biol. 2020, 16 (6), 667–675. 10.1038/s41589-020-0528-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Aggarwal G.; Zarrow J. E.; Mashhadi Z.; Flynn C. R.; Vinson P.; Weaver C. D.; Davies S. S. Symmetrically substituted dichlorophenes inhibit N-acyl-phosphatidylethanolamine phospholipase D. J. Biol. Chem. 2020, 295 (21), 7289–7300. 10.1074/jbc.RA120.013362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Castellani B.; Diamanti E.; Pizzirani D.; Tardia P.; Maccesi M.; Realini N.; Magotti P.; Garau G.; Bakkum T.; Rivara S.; Mor M.; Piomelli D. Synthesis and characterization of the first inhibitor of N-acylphosphatidylethanolamine phospholipase D (NAPE-PLD). Chem. Commun. 2017, 53 (95), 12814–12817. 10.1039/C7CC07582K. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fu J.; Astarita G.; Gaetani S.; Kim J.; Cravatt B. F.; Mackie K.; Piomelli D. Food intake regulates oleoylethanolamide formation and degradation in the proximal small intestine. J. Biol. Chem. 2007, 282 (2), 1518–1528. 10.1074/jbc.M607809200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Chen Z.; Zhang Y.; Guo L.; Dosoky N.; de Ferra L.; Peters S.; Niswender K. D.; Davies S. S. Leptogenic effects of NAPE require activity of NAPE-hydrolyzing phospholipase D. J. Lipid Res. 2017, 58 (8), 1624–1635. 10.1194/jlr.M076513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Zarrow J. E.; Tian J.; Dutter B.; Kim K.; Doran A. C.; Sulikowski G. A.; Davies S. S. Selective measurement of NAPE-PLD activity via a PLA1/2-resistant fluorogenic N-acyl-phosphatidylethanolamine analog. J. Lipid Res. 2022, 63 (1), 100156. 10.1016/j.jlr.2021.100156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Magotti P.; Bauer I.; Igarashi M.; Babagoli M.; Marotta R.; Piomelli D.; Garau G. Structure of human N-acylphosphatidylethanolamine-hydrolyzing phospholipase D: regulation of fatty acid ethanolamide biosynthesis by bile acids. Structure 2015, 23 (3), 598–604. 10.1016/j.str.2014.12.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Wang J.; Okamoto Y.; Tsuboi K.; Ueda N. The stimulatory effect of phosphatidylethanolamine on N-acylphosphatidylethanolamine-hydrolyzing phospholipase D (NAPE-PLD). Neuropharmacology 2008, 54 (1), 8–15. 10.1016/j.neuropharm.2007.06.001. [DOI] [PubMed] [Google Scholar]
  21. Petersen G.; Pedersen A. H.; Pickering D. S.; Begtrup M.; Hansen H. S. Effect of synthetic and natural phospholipids on N-acylphosphatidylethanolamine-hydrolyzing phospholipase D activity. Chem. Phys. Lipids 2009, 162 (1–2), 53–61. 10.1016/j.chemphyslip.2009.08.005. [DOI] [PubMed] [Google Scholar]
  22. Mock E. D.; Kotsogianni I.; Driever W. P. F.; Fonseca C. S.; Vooijs J. M.; den Dulk H.; van Boeckel C. A. A.; van der Stelt M. Structure-activity relationship studies of pyrimidine-4-carboxamides as inhibitors of N-acylphosphatidylethanolamine phospholipase D. J. Med. Chem. 2021, 64 (1), 481–515. 10.1021/acs.jmedchem.0c01441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Wilt S.; Kodani S.; Le T. N. H.; Nguyen L.; Vo N.; Ly T.; Rodriguez M.; Hudson P. K.; Morisseau C.; Hammock B. D.; Pecic S. Development of multitarget inhibitors for the treatment of pain: Design, synthesis, biological evaluation and molecular modeling studies. Bioorg. Chem. 2020, 103, 104165. 10.1016/j.bioorg.2020.104165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. McReynolds C.; Morisseau C.; Wagner K.; Hammock B. Epoxy fatty acids are promising targets for treatment of pain, cardiovascular disease and other indications characterized by mitochondrial dysfunction, endoplasmic stress and inflammation. Adv. Exp. Med. Biol. 2020, 1274, 71–99. 10.1007/978-3-030-50621-6_5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Leuillier M.; Duflot T.; Ménoret S.; Messaoudi H.; Djerada Z.; Groussard D.; Denis R. G. P.; Chevalier L.; Karoui A.; Panthu B.; Thiébaut P. A.; Schmitz-Afonso I.; Nobis S.; Campart C.; Henry T.; Sautreuil C.; Luquet S. H.; Beseme O.; Féliu C.; Peyret H.; Nicol L.; Henry J. P.; Renet S.; Mulder P.; Wan D.; Tesson L.; Heslan J. M.; Duché A.; Jacques S.; Ziegler F.; Brunel V.; Rautureau G. J. P.; Monteil C.; do Rego J. L.; do Rego J. C.; Afonso C.; Hammock B.; Madec A. M.; Pinet F.; Richard V.; Anegon I.; Guignabert C.; Morisseau C.; Bellien J. CRISPR/Cas9-mediated inactivation of the phosphatase activity of soluble epoxide hydrolase prevents obesity and cardiac ischemic injury. J. Adv. Res. 2023, 43, 163–174. 10.1016/j.jare.2022.03.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Abdalla H. B.; Alvarez C.; Wu Y. C.; Rojas P.; Hammock B. D.; Maddipati K. R.; Trindade-da-Silva C. A.; Soares M. Q. S.; Clemente-Napimoga J. T.; Kantarci A.; Napimoga M. H.; Van Dyke T. E. Soluble epoxide hydrolase inhibition enhances production of specialized pro-resolving lipid mediator and promotes macrophage plasticity. Br. J. Pharmacol. 2023, 180, 1597–1615. 10.1111/bph.16009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Burmistrov V.; Morisseau C.; Babkov D. A.; Golubeva T.; Pitushkin D.; Sokolova E. V.; Vasipov V.; Kuznetsov Y.; Bazhenov S. V.; Novoyatlova U. S.; Bondarev N. A.; Manukhov I. V.; Osipova V.; Berberova N.; Spasov A. A.; Butov G. M.; Hammock B. D. Anti-inflammatory activity of soluble epoxide hydrolase inhibitors based on selenoureas bearing an adamantane moiety. Int. J. Mol. Sci. 2022, 23 (18), 10710. 10.3390/ijms231810710. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Morisseau C.; Goodrow M. H.; Newman J. W.; Wheelock C. E.; Dowdy D. L.; Hammock B. D. Structural refinement of inhibitors of urea-based soluble epoxide hydrolases. Biochem. Pharmacol. 2002, 63 (9), 1599–1608. 10.1016/S0006-2952(02)00952-8. [DOI] [PubMed] [Google Scholar]
  29. Rose T. E.; Morisseau C.; Liu J. Y.; Inceoglu B.; Jones P. D.; Sanborn J. R.; Hammock B. D. 1-Aryl-3-(1-acylpiperidin-4-yl)urea inhibitors of human and murine soluble epoxide hydrolase: structure-activity relationships, pharmacokinetics, and reduction of inflammatory pain. J. Med. Chem. 2010, 53 (19), 7067–7075. 10.1021/jm100691c. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Thorp E.; Cui D.; Schrijvers D. M.; Kuriakose G.; Tabas I. Mertk receptor mutation reduces efferocytosis efficiency and promotes apoptotic cell accumulation and plaque necrosis in atherosclerotic lesions of apoe–/– mice. Arterioscler., Thromb., Vasc. Biol. 2008, 28 (8), 1421–1428. 10.1161/ATVBAHA.108.167197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Elliott M. R.; Koster K. M.; Murphy P. S. Efferocytosis signaling in the regulation of macrophage inflammatory responses. J. Immunol. 2017, 198 (4), 1387–1394. 10.4049/jimmunol.1601520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Domingo J. C.; Mora M.; Africa de Madariaga M. Incorporation of N-acylethanolamine phospholipids into egg phosphatidylcholine vesicles: characterization and permeability properties of the binary systems. Biochim. Biophys. Acta Biomembr. 1993, 1148 (2), 308–316. 10.1016/0005-2736(93)90144-O. [DOI] [PubMed] [Google Scholar]
  33. Mora M.; Mir F.; de Madariaga M. A.; Sagrista M. L. Aggregation and fusion of vesicles composed of N-palmitoyl derivatives of membrane phospholipids. Lipids 2000, 35 (5), 513–524. 10.1007/s11745-000-551-9. [DOI] [PubMed] [Google Scholar]
  34. Térová B.; Petersen G.; Hansen H. S.; Slotte J. P. N-acyl phosphatidylethanolamines affect the lateral distribution of cholesterol in membranes. Biochim. Biophys. Acta 2005, 1715 (1), 49–56. 10.1016/j.bbamem.2005.07.004. [DOI] [PubMed] [Google Scholar]
  35. Kooijman E. E.; Chupin V.; de Kruijff B.; Burger K. N. Modulation of membrane curvature by phosphatidic acid and lysophosphatidic acid. Traffic 2003, 4 (3), 162–174. 10.1034/j.1600-0854.2003.00086.x. [DOI] [PubMed] [Google Scholar]
  36. Brideau C.; Gunter B.; Pikounis B.; Liaw A. Improved statistical methods for hit selection in high-throughput screening. J. Biomol. Screen 2003, 8 (6), 634–647. 10.1177/1087057103258285. [DOI] [PubMed] [Google Scholar]
  37. Guo L.; Amarnath V.; Davies S. S. A liquid chromatography-tandem mass spectrometry method for measurement of N-modified phosphatidylethanolamines. Anal. Biochem. 2010, 405 (2), 236–245. 10.1016/j.ab.2010.06.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Leishman E.; Mackie K.; Luquet S.; Bradshaw H. B. Lipidomics profile of a NAPE-PLD KO mouse provides evidence of a broader role of this enzyme in lipid metabolism in the brain. Biochim. Biophys. Acta 2016, 1861 (6), 491–500. 10.1016/j.bbalip.2016.03.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Doran A. C.; Ozcan L.; Cai B.; Zheng Z.; Fredman G.; Rymond C. C.; Dorweiler B.; Sluimer J. C.; Hsieh J.; Kuriakose G.; Tall A. R.; Tabas I. CAMKIIγ suppresses an efferocytosis pathway in macrophages and promotes atherosclerotic plaque necrosis. J. Clin. Invest. 2017, 127 (11), 4075–4089. 10.1172/JCI94735. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Proto J. D.; Doran A. C.; Gusarova G.; Yurdagul A. Jr.; Sozen E.; Subramanian M.; Islam M. N.; Rymond C. C.; Du J.; Hook J.; Kuriakose G.; Bhattacharya J.; Tabas I. Regulatory T cells promote macrophage efferocytosis during inflammation resolution. Immunity 2018, 49 (4), 666–677.e6. 10.1016/j.immuni.2018.07.015. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

cb3c00401_si_001.pdf (432.5KB, pdf)

Articles from ACS Chemical Biology are provided here courtesy of American Chemical Society

RESOURCES