Hypothalamic-specific manipulation of Fto, the ortholog of the human obesity gene FTO, affects food intake in rats

Abstract

Sequence variants in the first intron of FTO are strongly associated with human obesity and human carriers of the risk alleles show evidence for increased appetite and food intake. Mice globally lacking Fto display a complex phenotype characterised by both increased energy expenditure and increased food intake. The site of action of FTO on energy balance is unclear. Fasting reduces levels of Fto mRNA in the arcuate nucleus (ARC) of the hypothalamus, a site where Fto expression is particularly high. In this study, we have extended this nutritional link by demonstrating that consumption of a high fat diet (45%) results in a 2.5 fold increase in Arc Fto expression. We have further explored the role of hypothalamic Fto in the control of food intake by using stereotactic injections coupled with AAV technology to bi-directionally modulate Fto expression. An over expression of Fto protein by 2.5-fold in the ARC results in a 14% decrease in average daily food intake in the first week. In contrast, knocking down Arc Fto expression by 40% increases food intake by 16%. mRNA levels of Agrp, Pomc and Npy, ARC-expressed genes classically associated with the control of food intake, were not affected by the manipulation of Fto expression. However, over expression of Fto resulted in a 4-fold increase in the mRNA levels of Stat3, a signalling molecule critical for leptin receptor signalling, suggesting a possible candidate for the mediation of Fto's actions. These data provide further support for the notion that FTO itself can influence key components of energy balance, and is therefore a strong candidate for the mediation of the robust association between FTO intronic variants and adiposity. Importantly, this provide the first indication that selective alteration of FTO levels in the hypothalamus can influence food intake, a finding consistent with the reported effects of FTO alleles on appetite and food intake in man.

Figure 1. Endogenous Fto expression in the ARC.

(A) Fto expression, quantified by real-time RT-PCR, is up-regulated in the arcuate nucleus following a high-fat diet demonstrated by the relative arcuate Fto mRNA expression in rat either fed on a standard chow or on a high-fat diet for 10 weeks. Data is represented as the mean±S.E.M of at least 6 independent rats per group; **P<0.01. (B) Fto mRNA is diffusely expressed throughout the arcuate nucleus and overlaps with Pomc neurons. Double in situ hybridization detecting Fto (³⁵S labeled) and Pomc (DIG labeled) mRNA in the cells of the arcuate nucleus. Insert in figure shows a high-magnification (scale bar, 20 µm) of a Pomc containing neurons colocalised with Fto (marked with *). (C) Intra-nuclei bilateral injections of adeno-associated virus (AAV2/7) mediated transfer of GFP cDNA precisely targets the hypothalamic arcuate (ARC) and paraventricular (PVN) nuclei; as demonstrated by photomicrographs of representative coronal sections showing localization of GFP 7 days after injection; right panels, higher magnification of area indicated by red box. Scale bar, 100µm. 3v: third ventricle; ME: median eminence.

Figure 2. Manipulation of Fto expression within the ARC and PVN.

(A) Knockdown of Fto in GT1-7 cells. Fto short hairpin RNA expressed from U6 promoter were transfected into cultured hypothalamic GT1-7 cells resulting in efficient knockdown of Fto mRNA (quantified by real time RT-PCR, left panel) and protein (western blot analysis, right panel) compare with cells transfected with scramble shRNA sequence. (B–C) Confirmation of AAV mediated transfer of full length cDNA (AAV-Fto) for over expression and shRNA (AAV-shRNA) for knockdown of Fto 3 weeks following intra-nuclei injection. Fto expression is increased by AAV-Fto and decreased by AAV-shRNA as compared to controls (empty vector for AAV-Fto and scrambled sequence for AAV-shRNA). Bar graphs show the quantified change in expression of Fto. Response is expressed in terms of fold induction over the control. (B) Fto mRNA relative to expression of B-actin was determined by real-time quantitative PCR. Left panel shows representative amplification curve from PVN micro-punches of AAV-Fto and AAV-shRNA injected rats. (C) Fto protein levels were measured by semi-quantitative western blot analysis. Left panel shows representative blots from micro-punched AAV-shRNA injected ARC and from AAV-Fto injected PVN. Western blots were probed with mouse monoclonal antibodies directed against the C-terminal Fto. B-actin was used as a loading control and bands were visualized and semi-quantified using Chemiluminescence. P-value was calculated using a two-tailed distribution unpaired Student's t-test. Data is represented as the mean ±S.E.M of at least 6 independent rats per group. *p<0.05; **p<0.01; ***p<0.001.

Figure 3. Effects of AAV-shRNA and AAV-Fto microinjection in hypothalamic ARC.

Effects of manipulating Fto expression in rat hypothalamic ARC. The 2-way repeated measures ANOVA indicated that ARC injection of (A) AAV-Fto significantly reduces daily food intake (F_1,140 = 15.28; p = 0.002) and (B) 2 weeks cumulative food intake (F_1,140 = 13.26; p = 0.003). Whilst (C,D) AAV-shRNA had a significant effect on induction of food intake (daily: F_1,140 = 8.823; p = 0.01; cumulative: F_1,140 = 13.73; p = 0.002). (E) Significant effects of Fto over expression on food intake last for 2 weeks whilst the effect of Fto knockdown was only significant in the first week. All values are expressed as mean±S.E.M. Statistical comparison between control and treated groups for each site were performed by two-tailed unpaired Student's t-test; **p<0.01.There were no significant effects of ARC Fto over expression and knockdown on (F) fat mass normalized to final body weight or (G–H) body weight gain expressed as percentage change from pre-surgical weight.

Figure 4. Effects of AAV-shRNA and AAV-Fto microinjection in hypothalamic PVN.

Effects of manipulating Fto expression in rat hypothalamic PVN. PVN injection of AAV-Fto reduces (A) daily (F_1,130 = 7.986; p = 0.01) and (B) cumulative (F_1,130 = 10.70; p = 0.006) food intake while (C) AAV-shRNA has no impact. (D) Food reducing effects of Fto over expression in the PVN was significant for 2 weeks. All values are expressed as mean ±S.E.M. Statistical comparison between control and treated groups for each site were performed by two-tailed unpaired Student's t-test. *p<0.05; **p<0.01. For the analysis of food intake over time, two-way repeated measure ANOVA was used with time and treatment as variables for comparison.

Figure 5. Effects of manipulating ARC Fto on genes involved in energy homeostasis.

Bar graphs show no change of Agrp, Npy and Pomc expression in response to arcuate injection of AAV-Fto or AAV-shRNA, while TH expression in the ARC decreases by 5- fold and Stat3 expression increases by 4-fold in response to Fto overexpression. Response of each gene was measured 3 weeks following intra-ARC injection, data normalized to B-actin and expressed in terms of fold induction over its expression in controls. P-value was calculated using a two-tailed distribution unpaired Student's t-test. Data is represented as the mean ±S.E.M of at least 6 independent rats per group; *p<0.05.