Fig 1: Peanut butter consumption blunts the ability of exendin-4 to decrease aggressive behaviors in male mice.Panel A Illustrates the outline of the resident intruder experiment when mice were fed peanut butter (, day 0–6) and treated repeatedly with exendin-4 (Ex4; 2.4 µg/kg, IP, n = 8) or vehicle (Veh, IP, n = 8). At day 1, the mice were trained in the resident intruder paradigm and the time to first attack was scored. This allows stratification (*; with similar attack score between future treatment groups) into Ex4 injections throughout the rest of training (day 2–5) and at the test day (day 6). During training the latency to attack was recorded (Panel B), and at the test day aggressive behaviors (attack, threat, Panel C) and non-aggressive behaviors (social and non-social, Panel D) were scored for 10 min. Directly after the test the mice were euthanized, and the brains from each mouse were collected for ex-vivo analyses (Panel F). A In these mice the baseline latency to attack was similar between future treatment groups (t(14) = 0.23, P = 0.8186). As described above, mice fed with chow display less aggressive behaviors after repeated administration of Ex4 (Fig. 2 and Fig. 3). In contrast, mice fed with HFD (i.e., peanut butter) neither exhibit altered latency to attack during training (Panel B), nor reduced aggressive behaviors at test (Panel C) after repeated administration of Ex4 (2.4 µg/kg, IP). A Specifically, the latency to attack was similar in HFD fed mice treated with either vehicle of Ex4 (treatment F(1,14) = 0.1470, P = 0.7072, time F(3,42) = 1.66, P = 0.1892, interaction F(3,42) = 0.3804, P = 0.7676; n = 8 per group). B Besides, the area under curve of attack latency is similar between the two groups of mice (t(14) = 0.27, P = 0.7931). C Moreover, the 24-h peanut butter intake was similar in vehicle and Ex4 treated mice prior to (treatment F(1,14) = 0.05, P = 0.8343, time F(1,14) = 1.83, P = 0.1981, interaction F(1,14) = 0.10, P = 0.7517) and during (treatment F(1,14) = 0.02, P = 0.8789, time F(3,42) = 3.36, P = 0.0275, interaction F(3,42) = 2.78, P = 0.0530) training. At test day, there were no differences between treatments when it comes to (D) attack duration (t(14) = 0.03, P = 0.9758), (E) attack frequency (t(14) = 0.43, P = 0.6758), (F) attack latency (t(14) = 0.56, P = 0.5863), (G) threat duration (t(14) = 0.54, P = 0.5984), (H) threat frequency (t(14) = 0.87, P = 0.8178) or (I) threat latency (t(14) = 0.29, P = 0.7769). Neither was there a difference in (J) social behavior duration (t(14) = 0.86, P = 0.4020), (K) social behavior frequency (t(14) = 1.09, P = 0.2940), (L) social behavior latency (t(14) = 0.47, P = 0.6467) or (M) social behavior duration (t(14) = 0.87, P = 0.4003) between vehicle and Ex4 treated mice. Western blot (Panel E) show (N) the protein levels of tryptophan hydroxylase (TPH2) of raphe from mice of the above resident intruder test, (O) and the TPH2 levels are higher after a low dose of Ex4 (n = 6) compared to vehicle (n = 6) in mice fed peanut butter. Western blots show the protein levels of dopamine-β-hydroxylase (DBH), noradrenalin reuptake transporter (NET), serotonin reuptake transporter (SERT), serotonin-1B-receptor (5HT1B) of (P) nucleus accumbens (NAc) core, (Q) where no differences are evident between treatments, or (R) NAc shell. (S) Compared to vehicle, Ex4 reduces the protein levels of DBH, and increases NET, without altering the protein levels of GLP1R, or 5HT1B in NAc shell. β-actin was always used as a loading control. The relative density of the target protein band was normalized to the density of the β-actin band to represent the relative expression of the target protein. The real time PCR data (Panel F) reveal that the expression of the GLP-1R in (T) NAc (t(10) = 0.33, P = 0.7437) or (U) dorsal raphe (t(10) = 1.22, P = 0.2516) was similar in mice exposed to chow compared to those exposed to HFD. Data are presented as mean ± SEM; significant data are illustrated by *P < 0.05, **P < 0.01.