Dietary triglycerides act on mesolimbic structures to regulate the rewarding and motivational aspects of feeding

Abstract

Circulating triglycerides (TGs) normally increase after a meal but are altered in pathophysiological conditions, such as obesity. Although TG metabolism in the brain remains poorly understood, several brain structures express enzymes that process TG-enriched particles, including mesolimbic structures. For this reason, and because consumption of high-fat diet alters dopamine signaling, we tested the hypothesis that TG might directly target mesolimbic reward circuits to control reward-seeking behaviors. We found that the delivery of small amounts of TG to the brain through the carotid artery rapidly reduced both spontaneous and amphetamine-induced locomotion, abolished preference for palatable food and reduced the motivation to engage in food-seeking behavior. Conversely, targeted disruption of the TG-hydrolyzing enzyme lipoprotein lipase specifically in the nucleus accumbens increased palatable food preference and food-seeking behavior. Finally, prolonged TG perfusion resulted in a return to normal palatable food preference despite continued locomotor suppression, suggesting that adaptive mechanisms occur. These findings reveal new mechanisms by which dietary fat may alter mesolimbic circuit function and reward seeking.

Figure 1. Central triglycerides delivery specifically decreases nocturnal locomotor activity and abolishes feeding preference for palatable food

Experimental procedure (a). Daily variation of locomotor activity (b) and cumulative locomotor activity during light and dark periods (c) in control mice (black squares and bars) and triglycerides (TG) infused mice (red circles and hatched red bars). Daily variation in chow (black squares) and HFHS (gray squares) diet intake in control mice (d). Daily variation in chow (red circles) and HFHS (orange circles) diet intake in TG infused mice (e). Cumulative food intake during light and dark periods of chow diet (control mice: black bars; TG infused mice: hatched red bars) and HFHS diet (control mice: grey bars; TG infused mice: dotted orange bars) (f). Cumulative total food intake (chow+HFHS diet) during light and dark periods in control mice (black bars) and TG infused mice (hatched red bars) (g).Locomotor and feeding recording period are indicated by an hatched arrow .Data present the mean of days 14+15. Displayed values are means ± SEM. (n=5-6). *p<0,05 NaCl vs TG; #p<0,05 CHOW NaCl vs HFHS NaCl; $p<0,05 HFHS NaCl vs HFHS TG.

Figure 2. Central triglycerides delivery rapidly alters spontaneous locomotor activity and decreases Amphetamine-induced locomotion

Experimental procedure (a, d). Locomotor activity evolution (b) and cumulative activity (c) was monitored at the very beginning of the infusion at the onset of dark period (7pm-8pm) in control mice (black squares and bars) and TG infused mice (red circles and hatched red bars). Locomotor activity was recorded during 2h after an acute intraperitoneal injection of 3 mg/kg D-amphetamine or vehicle (saline solution)in control mice (black square) and TG infused mice (red circles) day 12 (e) and day 13 (f). Cumulative locomotor activity after vehicle or amphetamine injection in control mice (black bars) and TG infused mice (hatched red bars) (g). Mice were infused with NaCl or TG solution during 6h just before vehicle or D-amphetamine injection. Displayed values are means ± SEM. (n=5). *p<0,05 NaCl vs TG.

Figure 3. Central triglycerides delivery specifically decreases motivational aspect of reward seeking

Experimental procedure (a). Progressive ratio (PR) responding for sucrose pellet in control mice (black squares)and TG infused mice (red circle). During basal conditions both group were infused with saline solution. Numbers of rewards achieved (b) and active lever presses achieved (e) were recorded during 4 sessions of PR. Time course of number of rewards and active lever presses were recorded during the 1^st (c, f) and 3^rd PR session (d, g). Mice were infused with NaCl or TG solution during 6h just before session of PR. Displayed values are means ± SEM. (n=5). *p<0,05 NaCl vs TG; #p<0,05 TG S1 et S2 vs TG S3.Using repeated measure analysis we show an effect of group (NaCl vs TG) during TEST &p<0,001.

Figure 4. Prolonged central triglycerides delivery results in desensitization of feeding but not locomotor activity

Experimental procedure (a, f). Cumulative locomotor activity during dark period in control mice (black bars) and TG infused mice (hatched red bars) was recorded after 24h (b) or 7 days (c) infusion Cumulative food intake during dark period of chow (control mice: black bars; triglycerides infused mice: hatched red bars) and HFHS diet (control mice: grey bars; triglycerides infused mice: dotted orange bars) was recorded in a food choice procedure after 24h (d) or 7 days (e) of infusion. Daily variation of locomotor activity (g), cumulative locomotor activity (h) (insert represents the reduction of locomotor activity expressed in % of total activity during saline perfusion) and ratio between chow/HFHS kcal intake during a two choice procedure (i) in lean mice infused with NaCl (black squares and bars), in lean mice infused with TG (red circles and hatched red bars), in DIO mice infused with NaCl (grey square and bars)and in DIO mice infused with TG (orange circles and hatched orange bars).Data present the mean of days 14+15 (b, d, g, h) and days 19+20 (c, e). Displayed values are means ± SEM. (n=5-7). *p<0,05 NaCl vs TG; #p<0,05 CHOW NaCl vs HFHS NaCl; $p<0,05 HFHS NaCl vs HFHS TG; £p<0,05 CHOW TG vs HFHS TG; & p<0,05 NaCl-DIO vs TG-DIO; § p<0,05 DIO vs Lean.

Figure 5. NAc-specific Lpl knock down increases motivational aspect of reward seeking

Fluorescent in situ hybridization was performed for LPL mRNA (red) within NAc (a) revealing that the gene is expressed across multiple NAc neurons labeled with dapi (blue, b). No expression of LPL mRNA was observed in white matter tracks (anterior commissure, ac) or other brain regions, as reflected in the absence of LPL mRNA grains within the somatosensory cortex (c). Schematic and representative photo micrograph showing GFP expression after AAV-CRE-GFP virus injection in the NAc of Lpl^lox/lox mice (d). Schematic and representative photo micrograph showing GFP expression after AAV-CRE-GFP virus injection in the NAc of Lpl ^+/+ (NAc-Lpl ^+/+ black bars) and Lpl ^lox/lox mice (NAc-Lpl^Δ/Δ,green dotted bars) after viral injection (e). Progressive ratio responding for sucrose pellets in control mice (black squares and bars) and in mice with a NAc-specific Lpl knock down (green circles and green dotted bars). Numbers of rewards (f), active lever presses (g) and ratio of active/inactive lever presses achieved (i) were recorded during 4 sessions of PR. Time course of rewards acquisition in within the 3^rd PR session (h). Average chow intake was recorded at several time points during 2 months after NAc-specific AAV-Cre-GFP virus injection (j). Total cumulative food intake (chow+HFHS) and specific cumulative food intake of chow and HFHS diet was recorded in a 24h food choice procedure (k) in control mice (black bars) and mice with a NAc-specific Lpl knock down (green dotted bars). Displayed values are means ± SEM. (n=5-6 in each group). *p<0,05 NAc-Lpl^+/+ vs NAc-Lpl^Δ/Δ; $p<0,05 CHOW NAc-Lpl^+/+ vs CHOW+HFHS NAc-Lpl^+/+; #p<0,05 CHOW NAc-Lpl^+/+ vs HFHS NAc-Lpl^+/+; £p<0,05 CHOW NAc-Lpl^Δ/Δ vs HFHS NAc-Lpl^Δ/Δ. Using repeated measure analysis we show an effect of group (NAc-Lpl^+/+ vs NAc-Lpl^Δ/Δ) &p<0,001