All protocols for the use of animals are in accordance with the University of Western Ontario animal care guidelines. These protocols conform to the guidelines of the Canadian Council on.

Animal Care (Ottawa, ON, Canada) and the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). The study has been approved by the University of Western Ontario Council on Animal Care (protocol number 2009–020).

Rats for neonatal ventricular myocyte studies were bred in the Health Sciences Animal Care Facilities at the University of Western Ontario. Adult male Sprague-Dawley rats were purchased from Charles River Canada (St. Constant, Quebec, Canada), and maintained in the Animal Care Facilities in accordance with the guidelines of the Canadian Council on Animal Care (Ottawa, Ontario, Canada).

Neonatal cardiomyocytes were prepared from hearts of 1- to 3-day-old Sprague-Dawley rats. Ventricles were washed, minced with scissors, and subsequently digested with collagenase II. The reaction was terminated with the addition of 20% FBS. The cellular extracts were strained and centrifuged at 514 g for 5 min. The supernatant was then aspirated and the resulting pellet was resuspended in cell culture media. The suspended cell mixture was subjected to two cycles of preplating to eliminate fibroblasts and other nonmyocyte components. The final myocyte preparation was plated in cell culture media containing Dulbecco’s modified Eagle medium/Ham’s F-12 supplemented with 10% FBS, 10 µg/ml transferrin, 10 µg/ml insulin, 10 ng/ml selenium, 50 U/ml penicillin, 2 mg/ml BSA, 5 µg/ml linoleic acid, 3 mM pyruvic acid, 0.1 mM minimum essential medium nonessential vitamins, 10% minimum essential medium vitamin solution, 0.1 mM bromodeoxyuridine, 100 µM L-ascorbic acid, and 30 mM HEPES, pH 7.2. The number of myocytes of each set was counted using a Countess® Automated Cell Counter (Life Technologies Inc, Burlington, Ontario, Canada) in order to achieve a final cell concentrations of 10⁶/ml.

For studies involving pharmacological modulations the following drugs were used: the rat leptin receptor antagonist SHLA (100 ng/ml, Protein Laboratories Rehovot, Israel), the Janus kinase 2 (JAK2) inhibitor AG490 (50 µM, Santa Cruz Biotechnology Inc., Santa Cruz, CA), the Signal Transducer and Activator of Transcription 3 (STAT3) inhibitor S31–201 (10 µM, Santa Cruz) and a cathepsin L inhibitor peptide (10 µM. Sigma-Aldrich Canada Co, Oakville, Ontario, Canada). For each study, the respective agent was added 30 min before leptin (3.1 mM) administration. Additional studies were done to compare the effect of leptin with two other pro-hypertrophic agents, endothelin-1 (10 nM, Sigma-Aldrich) or angiotensin II (100 nM, Sigma-Aldrich). For all experiments, myocytes were exposed to leptin, endothelin-1 or angiotensin II for 24 hours.

We employed a small interfering RNA (siRNA) approach to knockdown either FTO or cut-like homeobox 1 (CUX1) gene expression in cardiac cells. Following appropriate plating, myocyte transfections were performed as recommended by the manufacturer. siRNA was designed to target the FTO gene with the oligonucleotide sequence 5′-CAACGTGACTTTGCTAAACTT-3′ corresponding to nucleotides 588–608 of rat FTO (GenBank accession no. NM_001039713). FTO knockdown was performed by transfecting cells with a pair of siRNA constructs specific for FTO mRNA (5′-ACGUGACUUUGCUAAACUUTT-3′ and 5′-AAGUUUAGCAAAGUCACGUTG-3′). siRNA targeted the CUX1 gene with the oligonucleotide sequence 5′-AGGCCGATGAAACAAATGCAA-3′ corresponding to nucleotides at 2270–2290 of rat CUX1 (GenBank accession no: XM-001070410). CUX1 knockdown was performed by transfecting cells with a pair of siRNA constructs specific for CUX1 mRNA (5′-GCCGAUGAAACAAAUGCAATT-3′ AND 5′-UUGCAUUUGUUUCAUCGGCCT-3′). All siRNA was purchased from QIAGEN Inc. (Mississauga, Ontario. Canada). Each siRNA was transfected into cells with HiPerFect reagent (QIAGEN) for 48 hours. A universal scrambled negative control siRNA (QIAGEN) was used for control studies. The silencing efficacy was confirmed by both AF488-tagged control siRNA and Western blotting at various time points after transfection. Transfection efficiency determined by immunofluorescence analysis ranged between 75–85%.

Myocytes were grown on collagen-pre-coated glass coverslips in medium. Following 24 hour incubation with treatments, cells were fixed in 4% paraformaldehyde for 20 min at 40C. After 5 min permeabilization in 0.1% Triton X-100 PBS, myocytes were blocked for 1 h in 1% BSA in PBS buffer and then incubated overnight with the primary antibody against either FTO (1∶50 dilution), CUX1 (1∶500) or cathepsin L (1∶100) in 1% BSA in PBS buffer, then for 1 hour with a secondary antibody (AF488-conjugated anti-mouse, Life Technologies). Nuclei were counterstained with the nuclear stain Hoechst 33342 (1 µg/ml, Sigma-Aldrich) for 5 min. Cardiomyocytes were visualized using a Zeiss Axio Observer D1 fluorescence microscope (Zeiss, Gottingen, Germany) and quantification performed using SigmaScan Pro v.5 software (Systat Software Inc., San Jose, CA, USA).

Cell surface area was assessed using a Leica inverted microscope equipped with an Infinity 1 digital camera at ×200 magnification and measured using SigmaScan Software (Systat Software Inc., Chicago, Il). At least 50 randomly selected cells per experiment were used to determine cell size and averaged to provide an N value of one.

RNA was extracted using Trizol (Life Technologies) according to the manufacturer’s instructions. RNA (1 µg) was used to synthesize the first strand of cDNA using M-MLV reverse transcriptase according to the manufacturer’s protocols and was used as a template in the PCR reactions. The expression of JAK2, STAT3, cathepsin L, FTO, CUX1, ANP, 18S rRNA (loading internal control) genes was determined in 10 µl reaction volumes using EvaGreen qPCR Mastermix (Applied Biological Materials, Richmond, British Columbia, Canada). Fluorescence was measured and quantified using a DNA Engine Opticon 2 System (Bio-Rad Laboratories, Mississauga, Ontario, Canada). Primer sequences for all genes are shown in Table 1.

Myocytes were washed with PBS and lysed with 100 µl of lysis buffer following homogenization, and centrifuged at 10 000 g for 5 minutes at 4°C. The supernatant was transferred to a fresh tube and the protein concentration was determined by the Bradford protein assay method (BioRad). Thirty micrograms of protein were resolved on a 10% SDS-polyacrylamide gel and transferred to nitrocellulose membranes. The membranes were blocked in 5% milk for 1 hour and incubated with primary antibodies for FTO (Cayman Chemicals, Ann Arbor, MI, 1∶500 dilution), p200CUX1 (Santa Cruz or Abcam, Toronto, Ontario, Canada, 1∶500), p110CUX1 (Thermo Scientific, Rockford, Il, 1∶500) cathepsin L (LifeSpan BioSciences, Seattle, WA,1∶500), JAK2 (Santa Cruz, 1∶1000), STAT3 (Santa Cruz, 1∶1000) or β-actin (EMD Millipore, Billerica, MA, 1∶1000) overnight, followed by either a secondary antibody for 1 hour or by adding the IRDye 800-conjugated secondary antibody (1∶5000; LI-COR Biosciences, Lincoln, NE) to the primary antibody. The bound complex was then detected either by enhanced chemiluminescence reagent (Amersham Biosciences Inc., Piscataway, NJ) or using the Odyssey Clx Infrared Imaging System (Li-COR). The images were analyzed for density using the Odyssey Application Software (Li-COR).

Cathepsin L activity in cell lysates was assayed using a Cathepsin L Activity Fluorometric Assay Kit according to the manufacturer’s instructions (BioVision Inc, Milpitas, CA). This is a fluorescence-based assay that utilizes the preferred cathepsin-L substrate sequence FR labeled with AFC (amino-4-trifluoromethyl coumarin). Cell lysates that contain cathepsin-L cleave the synthetic substrate FR-AFC to release free AFC. Fluorescence intensity was determined at excitation/emission wavelengths of 400 nm/505 nm using a Spectramax M5 spectral fluorimeter (Molecular Devices, Sunnyvale, CA).

Experiments were performed to determine if endogenous leptin regulates cardiac FTO levels in normal rats or rats made obese by a high fat diet. For these studies adult male Sprague-Dawley rats weighing 225 to 250 g were fed either a standard diet or a high fat diet for 12 weeks. The breakdown of calories in the standard diet was 17% from protein, 73% from carbohydrate and 10% from fat. In the high fat diet 17% of calories were from protein, 35% from carbohydrate and 48% from fat (Research Diets, New Brunswick, NJ). To examine the role of leptin in regulating cardiac FTO protein expression animals from each dietary group were injected with leptin receptor antibody (OBR-Ab) (4 ug/kg bw) every other day for the duration of the 12 week feeding period. This dose of OBR-Ab was previously found to prevent leptin-induced cardiac STAT activation [14]. At the end of the 12 week period the animals were sacrificed by decapitation and hearts removed for FTO analysis by Western blotting.

Results are presented as means ± standard error. The data were analyzed by a 1-way ANOVA and group differences were detected using a Student-Newman-Keuls post hoc test when initial ANOVA analysis revealed statistically significant differences. P values of <0.05 were considered significant.

Our first set of experiments was aimed to identify the presence of FTO in cultured myocytes and determine whether this can be modulated by leptin treatment. Figure 1 summarizes our results in which FTO gene expression was determined as well as protein levels assessed by Western blotting. As evident from Figure 1A, 24 h leptin treatment resulted in a three-fold increase in FTO gene expression. Moreover, Western blot analysis (Figure 1B and 1C) showed substantial abundance of FTO in myocytes as well as the ability of 24 h leptin treatment to produce approximately a two-fold elevation in FTO levels. These data also demonstrate that siRNA directed at FTO substantially decreased FTO protein levels (Figure 1B and 1C). Although leptin tended to increase FTO gene and protein expression in siFTO-treated myocytes no significant effect of leptin was seen under this condition. Moreover, all effects of leptin on FTO were, as would be expected, completely abrogated by the highly potent and specific leptin receptor antagonist SHLA.

The ability of leptin to potently increase FTO levels in cultured myocytes prompted us to examine whether endogenous leptin regulates cardiac FTO levels in vivo in normal rats as well as in rats made obese by high fat diet consumption for 12 weeks. Body weights for control animals at the end of the feeding period were 623±9 g whereas for rats fed a high fat diet mean body weights were 781±19 g (P<0.05), representing approximately a 25% increase. Moreover, plasma leptin levels in these animals were nearly doubled from 5.3±1.1 ng/mL in control animals to 10.3±0.8 ng/mL (P<0.05) in animals treated with the high fat diet. As shown in Figure 1D and 1E, the high fat diet and the accompanying increase in both body weight and plasma leptin concentrations had no effect on cardiac FTO protein expression. However, treatment with an antibody directed against the leptin receptor, at a dose which significantly inhibited STAT3 activation (Figure 1F and 1G), resulted in a marked and significant decrease in cardiac FTO levels in both dietary groups (Figure 1D and 1E) although body weights were not significantly affected (not shown).

As shown in the immunofluorescence images (Figure 2), FTO expression under control conditions as well as in myocytes treated with leptin was restricted to nuclei with FTO intensity increasing more than three-fold by leptin in those myocytes not treated with siFTO (Figure 2A and 2C)). However, myocytes treated with siFTO demonstrated significantly reduced FTO fluorescence intensity with no significant increase in response to leptin (Figure 2B and 2C). The ability of leptin to upregulate FTO expression within nuclei was abrogated by the leptin receptor antagonist (Figure 2A and 2C).

Figure 3 (top six panels) summarizes the responses of myocytes to either endothelin-1 or angiotensin II in terms of FTO expression as well as hypertrophy and the latter following FTO downregulation. Both endothelin-1 and angiotensin II produced a significant hypertrophic response as evidenced by increased myocyte surface area as well as both atrial natriuretic peptide (ANP) and β- myosin heavy chain (β-MHC) gene expression although these responses were unaffected by siRNA directed at FTO. As also evident from Figure 3 (bottom two panels) neither endothelin-1or angiotensin II had any effect on FTO protein expression.

We next determined the potential cellular mechanisms by which leptin increases FTO abundance. As leptin exerts many of its effects via JAK2/STAT3 activation we determined the contribution of this system to FTO upregulation. Leptin produced a marked increase in JAK2 gene (Figure 4A) and protein expression (Figure 4B), STAT3 gene expression (Figure 4C) as well as protein levels of phosphorylated STAT3 (Figure 4D). Moreover, inhibition of either JAK2 with AG490 (Figure 4E and 4F) or STAT3 with S31–201 (Figure 4G and 4H) completely supressed leptin- induced upregulation in either gene or protein FTO expression.

As CUX1 has been suggested as a possible regulator of FTO (please see Discussion), we sought to determine the effect of leptin on CUX1 and the potential role of the latter on leptin-induced FTO upregulation in cardiomyocytes. Total CUX1 as determined by immunofluorescence staining was significantly increased by leptin (Figure 5A and 5B). Individual CUX1 isoforms were measured using Western blotting (Figures 5C to 5E) which showed a significant increase in p110CUX1 protein expression with leptin treatment (Figure 5C and 5D). Although p200 CUX1protein levels tended to increase in response to leptin there were no significant differences between treatment groups (Figure 5C and 5E). The leptin-induced increase in p110 CUX1 protein expression was completely supressed by the cathepsin L peptide inhibitor (Figure 5C and 5D). siRNA directed at CUX1 produced a 80% reduction in basal FTO levels and a complete prevention of the ability of leptin to increase FTO expression (Figure 6).

The possibility that leptin-induced increase in FTO levels reflected p200CUX1 proteolysis to p110CUX1 was studied by first determining whether leptin affects cathepsin L and then whether a cathespin L inhibitor exerts any influence on leptin-induced FTO elevation. Leptin produced marked and significant increases in cathepsin L gene and protein expression as well as activity which were abrogated by Jak2 and STAT3 inhibition (Figure 7A–7D). Moreover, a peptide cathepsin L inhibitor completely supressed the ability of leptin to increase both FTO gene and protein expression (Figure 7E and 7F).

Finally, we determined whether approaches aimed at supressing cathepsin L, CUX1 or FTO share an equal ability to attenuate the hypertrophic effect of leptin. As summarized in Figure 8, pharmacological inhibition of cathepsin L or knockdown of either CUX1 or FTO with siRNA prevented the hypertrophic response of cardiomyocytes to leptin. Moreover, these effects of leptin as well as its ability to activate JAK2/STAT3 were completely prevented by the leptin receptor antagonist SHLA (Table 2).