N-acylphosphatidylethanolamine, a gut- derived circulating factor induced by fat ingestion, inhibits food intake

Abstract

N-acylphosphatidylethanolamines (NAPEs) are a relatively abundant group of plasma lipids of unknown physiological significance. Here, we show that NAPEs are secreted into circulation from the small intestine in response to ingested fat and that systemic administration of the most abundant circulating NAPE, at physiologic doses, decreases food intake in rats without causing conditioned taste aversion. Furthermore, (14)C-radiolabeled NAPE enters the brain and is particularly concentrated in the hypothalamus, and intracerebroventricular infusions of nanomolar amounts of NAPE reduce food intake, collectively suggesting that its effects may be mediated through direct interactions with the central nervous system. Finally, chronic NAPE infusion results in a reduction of both food intake and body weight, suggesting that NAPE and long-acting NAPE analogs may be novel therapeutic targets for the treatment of obesity.

Figure 1

Regulation of NAPE levels in rat plasma, lymph and small intestine by fat feeding and intraduodenal lipid infusion. (A) High-fat chow feeding increases total plasma NAPE (P<0.025) at four hours relative to fasting conditions (n=8/group). (B) Time course of intraduodenal lipid (solid squares), dextrin (open triangles), or protein (open circles, dashed line) infusions’ effects on relative lymph NAPE concentration (P<0.0025) (n=4–8/group). (C) Intraduodenal infusion of lipid significantly increases absolute NAPE concentration in lymph four hours after the start of the experiment (P<0.02) (n=7–8/group). (D) Intraduodenal infusion of lipid, but not dextrin, increases plasma NAPE concentration (P<0.02) four hours after the start of the experiment (n=4/group). (E) N-acyl species profile of plasma NAPE at four hours after the start of the lipid infusion (solid bars) compared with fasting levels in the same animals (open bars) (n=4/group). (F) High fat feeding significantly increases NAPE content in the small intestine relative to fasting (P<0.05) (n=4/group). (G) High fat feeding significantly increases plasma NAPE at four hours relative to fasting in rats maintained on regular chow (open bars), but this effect is absent in animals fed high fat chow for 35 days prior to the experiment (P<0.008) (n=6–14/group). All data are expressed as mean ± SEM.

Figure 2

C16:0 NAPE treatment suppresses food intake dose-dependently without causing conditioned taste aversion. (A) Systemic (i.p) NAPE administration significantly reduces the rate of food intake in a dose-dependent manner at 200 mg/kg, 500 mg/kg, and 1000 mg/kg (P<0.01) in free-feeding mice (n=4–14/group). (B) Systemic (i.p) NAPE administration reduces overnight cumulative food intake dose-dependently in free-feeding mice (n=4–11/group). (C) 100 mg/kg C16:0 NAPE (i.p.) significantly reduces overnight food intake in ad libitum fed rats at t=3 hrs (P<0.03) and overnight t=13 (P<0.004) (n=6/group) (D) 100 mg/kg C16:0 NAPE (i.p.) in rats, a dose sufficient to reduce overnight food intake, does not produce conditioned taste aversion in contrast to 100 mg/kg LiCl. Open bars indicate the amount of saccharin solution consumed when presented to 24 hour water deprived, naïve rats, and filled bars show the amount of saccharin solution consumed in 24 hour water deprived rats trained to associate the sweet taste with either saline, C16:0 NAPE, or LiCl (n=6/group). (E) Total plasma NAPE time course in fasted rats injected with 100 mg/kg C16:0 NAPE at t=0 (n=5–9/group). (F) Total plasma NAPE time course in fasted rats given an intraduodenal infusion of lipid (Liposyn II) at t=0 (n=4/group). (G) C16:0 NAPE (500 mg/kg) dramatically suppresses overnight food intake in ad libitum fed ob/ob mice at t=3 (P<0.002) and t=12 (P<0.0001) (n=4/group). (H) C16:0 NAPE (500 mg/kg) suppresses overnight food intake in fasted CB1−/− mice (P<0.0005) (n=5/group). All data are expressed as mean ± SEM.

Figure 3

C16:0 NAPE treatment suppresses food intake by a central mechanism. (A) Central (i.c.v.) administration of C16:0 NAPE (80 nmol) significantly (P<0.0001) reduces overnight food intake in mice relative to lipid control (Liposyn II) (n=6–7/group). (B) i.c.v. administration of 80 nmol dioleoylphosphatidylethanolamine (DOPE), a phospholipid, to mice does not reduce food intake relative to lipid control (Liposyn II) (n=4/group). (C) C16:0 NAPE (100 mg/kg i.p.) reduces overnight food intake in vagotomized rats, indicating that its suppressive effects on food intake do not require intact vagal afferents (P<0.05) (n=5–8/group). (D) Intravenous infusion of 2.5μCi ¹⁴C NAPE significantly increases counts in the brain (P<0.0001) and hypothalamus (P<0.03) of treated animals relative to controls at t=4 (n=5–6/group). (E) Central administration of PEA, the C16:0 NAPE hydrolysis product (100 nmol) does not affect overnight food intake in mice (n=6–7/group). (F) Systemic pre-treatment with URB597 (0.3 mg/kg), an inhibitor of central NAE degradation, which increases brain NAE concentrations, does not affect food intake in fasted mice (n=4/group). All data are expressed as mean ± SEM.

Figure 4

C16:0 NAPE administration (i.p.) reduces locomotor activity without causing motor deficits, and central administration of a small quantity of i.c.v. NAPE is sufficient to reproduce this effect. (A) Systemic C16:0 NAPE administration produces a dose-dependent reduction in motor activity relative to lipid control at 100 mg/kg (P<0.01), 500 mg/kg (P<0.001), and 1000 mg/kg (P<0.001) (n=4–12/group). (B) Systemic (i.p.) NAPE administration (1000 mg/kg) does not affect time to fall from the accelerating rotarod, a test of motor coordination and ability to respond to a motor challenge in pre-trained mice (n=5/group). (C) Central administration (i.c.v.) of C16:0 NAPE, 80 nmol, or .25% of the maximum dose (1000 mg/kg) given peripherally, causes a similar reduction in locomotor activity (P<0.0001) (n=3/group). All data are expressed as mean ± SEM.

Figure 5

Systemic C16:0 NAPE treatment (1000 mg/kg i.p.) reduces fasting-induced activation of NPY neurons in the hypothalamic arcuate nucleus (ARC) of transgenic mice expressing GFP in NPY neurons. cFOS is stained in red. (A) cFOS-GFP co-localization in the arcuate nucleus of free-feeding mice is low, indicating few transcriptionally active NPY neurons. (Small scale bar, inset, 10 microns, large scale bar, panel, 100 microns.) (B) Overnight fasting dramatically increases co-localization, indicating activated NPY neurons. (C) C16:0 NAPE treatment of fasted animals results in reduction of NPY-GFP co-localization to fed levels. (D) Diffuse expression of cFOS is low in the ARC of free-feeding animals. (E) Overnight fasting also stimulates cFOS expression in GFP-negative cells throughout the ARC. (F) C16:0 NAPE treatment markedly reduces cFOS expression in the ARC of fasted animals. (G) Quantification of percent cFOS-GFP co-localization in the arcuate nucleus of all fasted, fed, and NAPE treated animals (P<0.001) (n=4–5/group). (H) Quantification of NAPE-induced reductions in the number of cFOS-NPY positive neurons/0.1mm³ in the ARC relative to fasting (P<0.0001) (n=4–5/group). (I) Quantification of NAPE-induced reductions in the number of cFOS positive cells/0.1mm³ in the ARC relative to fasting (P<0.0001) (n=4–5). All data are expressed as mean ± SEM.

Figure 6

Systemic C16:0 NAPE treatment (1000 mg/kg i.p.) increases cFOS expression in the paraventricular nucleus (PVN) relative to lipid vehicle. (A) Representative section at 10X stained for cFOS in a vehicle-treated animal shows low cFOS expression in the PVN. (B) NAPE treatment significantly increases cFOS staining in the PVN. (C) Slice shown in (A) at 20X magnification. (D) Slice shown in (B) at 20X magnification. (E) Quantification of percent cFOS induction in the PVN by 1000 mg/kg C16:0 NAPE in all animals (P<0.008) (n=4–5/group). All data are expressed as mean ± SEM.

Figure 7

Chronic intravenous infusion of C16:0 NAPE (100 mg/kg-day or .07 mg/kg-min) significantly reduces cumulative food intake and body weight in ad libitum fed rats. (A) Time course of cumulative food intake in rats treated either with vehicle solution or 0.07 mg/kg-min over five days (P<0.0001) (n=7–15/group). (B) Comparison of weight change of vehicle treated controls and NAPE treated animals after five days of infusion (P<0.0001) (N=7–15/group). (C) Cumulative food intake after five days in vehicle treated controls and NAPE treated animals after five days of infusion (P<0.0001) (n=7–15). All data are expressed as mean ± SEM.