How Long To Deplete Liver Glycogen

Nat Commun. 2013 Aug xiii; 4: 2316.

Glycogen shortage during fasting triggers liver–brain–adipose neurocircuitry to facilitate fat utilization

Yoshihiko Izumida,^{1, 2, 3} Naoya Yahagi,^{a, one, 2, 4} Yoshinori Takeuchi,^{ane, iv} Makiko Nishi,^{ane, ii, 4} Akito Shikama,^{1, 4} Ayako Takarada,¹ Yukari Masuda,^{i, 2} Midori Kubota,^{1, 2} Takashi Matsuzaka,^iv Yoshimi Nakagawa,^iv Yoko Iizuka,² Keiji Itaka,⁵ Kazunori Kataoka,⁵ Seiji Shioda,³ Akira Niijima,⁶ Tetsuya Yamada,⁷ Hideki Katagiri,⁷ Ryozo Nagai,² Nobuhiro Yamada,^four Takashi Kadowaki,² and Hitoshi Shimano⁴

Yoshihiko Izumida

^aneNutrigenomics Research Group, Faculty of Medicine, University of Tsukuba, Tsukuba 305-8575, Nippon

^twoDepartment of Internal Medicine, Graduate Schoolhouse of Medicine, Academy of Tokyo, Tokyo 113-8655, Japan

³Department of Anatomy, Showa University School of Medicine, Tokyo 142-8555, Japan

Naoya Yahagi

^aneNutrigenomics Research Group, Kinesthesia of Medicine, University of Tsukuba, Tsukuba 305-8575, Nihon

^twoSection of Internal Medicine, Graduate School of Medicine, University of Tokyo, Tokyo 113-8655, Nihon

⁴Department of Internal Medicine (Endocrinology and Metabolism), Faculty of Medicine, Academy of Tsukuba, Tsukuba 305-8575, Japan

Yoshinori Takeuchi

^aneNutrigenomics Research Grouping, Faculty of Medicine, University of Tsukuba, Tsukuba 305-8575, Japan

⁴Section of Internal Medicine (Endocrinology and Metabolism), Faculty of Medicine, University of Tsukuba, Tsukuba 305-8575, Japan

Makiko Nishi

¹Nutrigenomics Enquiry Grouping, Faculty of Medicine, University of Tsukuba, Tsukuba 305-8575, Japan

²Department of Internal Medicine, Graduate Schoolhouse of Medicine, University of Tokyo, Tokyo 113-8655, Nippon

⁴Department of Internal Medicine (Endocrinology and Metabolism), Faculty of Medicine, University of Tsukuba, Tsukuba 305-8575, Japan

Akito Shikama

ⁱNutrigenomics Inquiry Grouping, Faculty of Medicine, University of Tsukuba, Tsukuba 305-8575, Japan

^fourDepartment of Internal Medicine (Endocrinology and Metabolism), Faculty of Medicine, University of Tsukuba, Tsukuba 305-8575, Nihon

Ayako Takarada

¹Nutrigenomics Enquiry Group, Faculty of Medicine, Academy of Tsukuba, Tsukuba 305-8575, Nippon

Yukari Masuda

¹Nutrigenomics Research Group, Faculty of Medicine, Academy of Tsukuba, Tsukuba 305-8575, Japan

²Department of Internal Medicine, Graduate Schoolhouse of Medicine, University of Tokyo, Tokyo 113-8655, Japan

Midori Kubota

^aneNutrigenomics Inquiry Group, Faculty of Medicine, University of Tsukuba, Tsukuba 305-8575, Japan

²Department of Internal Medicine, Graduate School of Medicine, University of Tokyo, Tokyo 113-8655, Japan

Takashi Matsuzaka

^fourSection of Internal Medicine (Endocrinology and Metabolism), Faculty of Medicine, University of Tsukuba, Tsukuba 305-8575, Japan

Yoshimi Nakagawa

^ivDepartment of Internal Medicine (Endocrinology and Metabolism), Faculty of Medicine, University of Tsukuba, Tsukuba 305-8575, Nihon

Yoko Iizuka

²Department of Internal Medicine, Graduate School of Medicine, Academy of Tokyo, Tokyo 113-8655, Nippon

Keiji Itaka

⁵Centre for Disease Biology and Integrative Medicine, Graduate School of Medicine, University of Tokyo, Tokyo 113-0033, Japan

Kazunori Kataoka

^fiveMiddle for Disease Biology and Integrative Medicine, Graduate School of Medicine, University of Tokyo, Tokyo 113-0033, Japan

Seiji Shioda

³Department of Beefcake, Showa University School of Medicine, Tokyo 142-8555, Japan

Akira Niijima

⁶Department of Physiology, Niigata University Schoolhouse of Medicine, Niigata 951-8510, Japan

Tetsuya Yamada

⁷Division of Advanced Therapeutics for Metabolic Diseases, Tohoku University Graduate School of Medicine, Sendai 980-8575, Nippon

Hideki Katagiri

⁷Segmentation of Advanced Therapeutics for Metabolic Diseases, Tohoku University Graduate Schoolhouse of Medicine, Sendai 980-8575, Nihon

Ryozo Nagai

²Department of Internal Medicine, Graduate School of Medicine, University of Tokyo, Tokyo 113-8655, Japan

Nobuhiro Yamada

⁴Section of Internal Medicine (Endocrinology and Metabolism), Faculty of Medicine, Academy of Tsukuba, Tsukuba 305-8575, Nihon

Takashi Kadowaki

²Department of Internal Medicine, Graduate School of Medicine, University of Tokyo, Tokyo 113-8655, Japan

Hitoshi Shimano

⁴Department of Internal Medicine (Endocrinology and Metabolism), Faculty of Medicine, University of Tsukuba, Tsukuba 305-8575, Japan

Received 2013 Jan 26; Accustomed 2013 Jul 17.

Supplementary Materials

Supplementary Information Supplementary Figures S1-S8, Supplementary Tables S1-S4, Supplementary Methods and Supplementary References

GUID: 50879C27-C51C-4A23-9CDF-A9E97A05F43A

Abstract

During fasting, animals maintain their energy residuum by shifting their energy source from carbohydrates to triglycerides. However, the trigger for this switch has not withal been entirely elucidated. Here we evidence that a selective hepatic vagotomy slows the speed of fatty consumption by attenuating sympathetic nerve-mediated lipolysis in adipose tissue. Hepatic glycogen pre-loading by the adenoviral overexpression of glycogen synthase or the transcription factor TFE3 abolished this liver–brain–adipose axis activation. Moreover, the occludent of glycolysis through the knockdown of the glycogen phosphorylase factor and the resulting elevation in the glycogen content abolished the lipolytic point from the liver, indicating that glycogen is the primal to triggering this neurocircuitry. These results demonstrate that liver glycogen shortage activates a liver–brain–adipose neural axis that has an important function in switching the fuel source from glycogen to triglycerides under prolonged fasting conditions.

A sufficient energy supply is essential for life; consequently, multiple mechanisms have evolved to ensure both energy availability and conservation during fasting. During fasting, liver glycogen, a glucose-storage polymer, is initially mobilized to replenish claret glucose (glycogenolysis). Major changes in metabolism are known to occur as the glycogen supply dwindles; when glycogen storage in the liver is depleted, stored adipose tissue triglycerides are released into the apportionment every bit fat acids and glycerol. The released fat acids are directly oxidized every bit an energy source by some tissues (liver and muscle), or are metabolized by the liver to ketone bodies for use by tissues, notably the brain that cannot utilize fatty acids, while the glycerol is converted by the liver into glucose (gluconeogenesis).

Among the mechanisms that prompt this shift in energy source from liver glycogen to adipose triglycerides are changes in blood hormones; when an animal is fasting, the torso senses a drop in the glucose concentration at sites such as the pancreatic islets and encephalon and responds by reducing insulin secretion from islet β cells and by increasing glucagon secretion from islet α cells. Some other response is sympathetic adrenal stimulation causing increased epinephrine levels1.

However, whether the trigger that switches the major energy source from hepatic glycogen to adipose tissue triglycerides is comprised but of the drib in claret glucose level and the concomitant changes in metabolic hormone levels has not yet been determined.

Here we show that liver glycogen shortage direct activates a liver–brain–adipose neural axis independently of blood glucose and insulin/glucagon levels, having an important role in switching the fuel source from glycogen to triglycerides under prolonged fasting conditions (Supplementary Fig. S1).

Results

Selective hepatic vagotomy preserves fasting fat pad mass

To examine the role of the hepato-vagal nervus pathway in the regulation of lipolysis in adipose tissue, we dissected the hepatic branch of the vagus nerve and observed the metabolic changes in epididymal fat pads during fasting. Three weeks after a selective hepatic vagotomy (HVx) was performed using microsurgery, in one case the animals had recovered their trunk weight and their food intake had returned to normal2 (Supplementary Fig. S2a,b), the mice were killed in a fasting state. As shown in Fig. 1a–e, the vagotomized mice fasted for 24 h lost significantly less fat and displayed larger adipocytes than mice with a sham surgery (isolation of the nervus without resection). The same effect as that seen after the HVx was also observed after capsaicin treatment, which chemically ablates but the afferent vagus nerve composed of unmyelinated fibres (Fig. 1a–e, Supplementary Fig. S3a,b). This effect of HVx or capsaicin treatment was not seen in subcutaneous fat (Fig. 1e). Figure 1f shows that the body fat weight loss during fasting was significantly retarded in HVx mice after 24 h when analysed using DEXA (digital energy 10-ray analysis), while the full body weight in both the sham and HVx groups decreased similarly.

Selective HVx preserves fasting fat pad mass.

(a) Gross advent of epididymal fat in sham, hepatic vagotomized (HVx) and capsaicin-treated (Cap) mice. (b) Weight of remnant epididymal fatty tissue in sham, HVx and Cap model after 24 h of fasting. Scale bar, five mm. (n=seven). (c) Microscopic appearance of HE-stained epididymal fatty tissue. Calibration bar, 100 μm. (d) Morphometry of adipocytes in sham, HVx and Cap animals. (n=3, 10 fields per a section). (east) Total torso weight, weight of epididymal fat pad and subcutaneous fatty in 0, 20 and 24 h fasting mice. (northward=5–7). (f) Changes in torso weight (left), fat weight (middle) and lean body weight (right) during fasting measured using DEXA. (n=4–7). *P<0.05 versus sham grouping (by Tukey's mail-hoc exam). Mistake confined, s.e.thou.

Vagotomy suppresses sympathetic nerve-mediated lipolysis

Side by side, we examined the metabolic changes in adipose tissue caused past HVx. When campsite signalling, which is the second messenger mediating the lipolytic signal, was visualized using adenovirally delivered luciferase reporter driven by the campsite response element (CRE-luc) using an in vivo imaging system, campsite signalling in the epididymal fat was reduced by HVx (Fig. 2a). Accordingly, the activated form of hormone-sensitive lipase (HSL), phosphorylated at Ser 563, tended to be reduced by hepatic vagus nerve interference (Fig. 2b; Supplementary Fig. S4a). The mRNA levels of HSL and adipose triglyceride lipase (ATGL) as well as pyruvate dehydrogenase kinase 4, a cardinal enzyme regulating glyceroneogenesis that is induced by fasting or epinephrine treatment3, were also downregulated (Fig. 2c). There were no differences betwixt HVx and capsaicin-treated groups, indicating that afferent signal is of key importance. These changes in lipolytic activities resulted in lower levels of plasma non-esterified fatty acids and glycerol in the HVx groups (Fig. 2d), leading to higher respiratory quotient and lower fat utilization per body despite the equal energy consumption (Fig. 2e,f). These differences were cancelled by guanethidine administration, which reduces the release of catecholamines from sympathetic nerves (Fig. 2g, Supplementary Fig. S5a,b)4. The plasma glucose, insulin, glucagon, catecholamines and FGF21 concentrations during fasting were not significantly dissimilar between the HVx and sham groups (Fig. 2d, Supplementary Figs S6 and S7a,b). These data demonstrate that the hepatic vagal signals control the lipolytic activities in adipose tissues through the sympathetic nervous system.

Vagotomy suppresses sympathetic nerve-mediated lipolysis.

(a),Ad-CRE-luc activities in adipose tissue visualized using IVIS Imaging Arrangement after 20 h of fasting. *P<0.05 versus sham grouping (by Student's t-examination) (n=half-dozen). (b) Left, poly peptide expressions assessed by western absorb analysis using specific antibodies for Ser 563 phospho-HSL (p-HSL) and ATGL in epididymal adipose tissue of sham and HVx mice after 0 and 20 h of fasting. Right, quantification of the blot (ratio of p-HSL to total HSL) (n=5 pooled). (c) mRNA expressions in epididymal fatty of 24 h fasting animals, examined by real-time quantitative PCR (RT–qPCR). *P<0.05 versus sham grouping (past Tukey's post-hoc examination) (n=vii). (d) Circulating non-esterified fatty acrid (NEFA), glycerol, β-hydroxybutyrate, glucose and insulin levels later xx h of fasting. *P<0.05 versus sham group (by Student'southward t-test) (north=4–9). (east,f) Respiratory quotient (RQ), fat utilization and energy consumption measured using a calorimetric system. *P<0.05 versus sham group (by Student's t-test) (n=15–18). (chiliad) Weight of remnant epididymal fat tissue in sham or HVx mice with or without guanethidine (GD) treatment after 24 h fasting. *P<0.05 versus sham group (past Tukey'southward post-hoc test) (due north=6–10). NS, not significant. Error confined, s.eastward.thousand.

Glycogen loading cancels liver–brain–adipose axis activation

Next, we explored the mechanism by which the liver–brain–adipose neurocircuitry is activated during fasting. Beginning, the effect of the forced pre-loading of glycogen was tested; when glycogen synthase 2 (Gys2) or the transcription factor TFE3, which promotes glycogen synthesis5, was overexpressed by adenoviral cistron delivery to the liver, the hepatic glycogen contents and consequently the adipose tissue mass increased to the levels of those seen in vagotomized mice (Fig. 3a,b,east,f, Supplementary Fig. S8a,b). Consequent with this finding, the expressions of the lipolytic enzymes HSL and ATGL as well as the phosphorylated HSL levels were suppressed by Gys2 or TFE3 overexpression in the liver (Fig. 3c,d,m,h; Supplementary Fig. S4b,c). These data indicate that hepatic glycogen loading abolishes the activation of the liver–encephalon–adipose axis during fasting.

Hepatic glycogen loading cancels liver–brain–adipose axis activation.

(a,eastward) Gross appearance and weight of epididymal fatty in Ad-CMV-Gys2 (a) or Ad-CMV-TFE3 (eastward) injected mice (filled bars) compared with control (grey confined; Ad-CMV-LacZ/GFP). Scale bar, 5 mm. *P<0.05 versus Advertisement-CMV-LacZ/GFP-treated sham grouping (by Tukey'southward post-hoc examination) (n=5–9). (b,f) Northern blots verifying Gys2 or TFE3 overexpression (3 private liver RNA samples pooled) and hepatic glycogen content in mice infected with Ad-CMV-Gys2 or Ad-CMV-TFE3. *P<0.05 versus Ad-CMV-LacZ/GFP-treated group (past Student's t-exam) (n=5–14). (c,g) RT–qPCR analyses of HSL, ATGL and pyruvate dehydrogenase kinase 4 (PDK4) mRNA expression of epididymal fatty in Gys2- or TFE3-overexpressing mice in starvation. *P<0.05 versus Advertising-CMV-LacZ/GFP-treated sham group (by Tukey'due south post-hoc test) (n=4–9). (d,h) Protein expression of Ser 563 p-HSL and total HSL in epididymal adipose tissue of Gys2- or TFE3-overexpressing mice. (north=3 or three individual protein samples pooled). All mice are analysed on 24-h fasting condition. *P<0.05 versus Ad-CMV-LacZ-treated sham group (by Tukey's post-hoc test). NS, not significant. Error confined, s.eastward.chiliad.

Glycolysis blockade abolishes lipolytic indicate from liver

Conversely, when Gys2 expression was knocked down by adenovirally transferred curt hairpin RNA (shRNA), and liver glycogen shortage was concomitantly exacerbated, the adipose tissue tended to shrink faster, and this tendency was cancelled past the HVx (Fig. 4a,c–eastward, Supplementary Fig. S4d, Supplementary Table S1). Next, to distinguish which of the parameters, glycogen itself or its downstream metabolites, is the key to triggering liver–brain–adipose axis activation when depleted, the glycogen phosphorylase liver type gene (Pygl) was knocked down using shRNA (Fig. 4b–e, Supplementary Fig. S4d, Supplementary Tabular array S1); when glycolysis was suppressed by this RNA interference, the liver glycogen content was elevated and this in turn led to a decrease in lipolysis in the adipose tissue. This result demonstrates that the shortage of glycogen, just not that of the downstream metabolites, is the primal to triggering the neurocircuitry. Meanwhile, AMP-activated kinases (AMPKs) were prone to exist activated past the knockdown of glycogen phosphorylase (Fig. 4f,k, Supplementary Fig. S4e).

Glycolysis occludent abolishes lipolytic signal from liver.

(a,b) Weight of epididymal fat pad after fasting. Effects of brusk hairpin RNA expression adenovirus vector targeted for Gys2 (hatched bar, a) or Pygl (filled bar, b). (n=4–eight). (c) Immunoblot analysis of glycogen synthase (GS) and glycogen phosphorylase (PYGL) expression in the livers of Advertizing-Gys2-i or Ad-Pygl-i injected mice. (n=3–4). (d) RT–qPCR analysis of mRNA expression of glycogen synthase (Gys2) and glycogen phosphorylase (Pygl) genes. (northward=iii). (e) Liver glycogen content of mice infected with adenovirus expressing short hairpin RNA targeted for LacZ, Gys2 or Pygl (Ad-LacZ-i, Ad-Gys2-i or Ad-Pygl-I, respectively) (due north=3–iv). (f) Immunoblot analysis of Ser485-phosphorylated AMPKα (pAMPKα), total AMPKα, Ser181-phosphorylated AMPKβ (pAMPKβ) and total AMPKβ. Whole-cell lysates from the livers of mice infected with adenovirus expressing short hairpin RNA targeted for LacZ, Gys2 or Pygl (Ad-LacZ-i, Advertizing-Gys2-i or Advertizement-Pygl-i). (one thousand) Quantified results of the data shown in (f), shown as the ratio of pAMPKα to total AMPKα and pAMPKβ to total AMPKβ. (northward=3–4). All mice are analysed on 24-h fasting condition. *P<0.05 versus Ad-LacZ-i sham group (past Tukey's postal service-hoc test). NS, not significant. Error bars, s.e.m.

Discussion

The results of this paper reveal that glycogen shortage in the liver triggers the liver–brain–adipose neural axis, promoting fatty acid and glycerol release from white adipose tissue every bit a response to fasting (Supplementary Fig. S1). The presence of the liver–encephalon–adipose axis for the regulation of lipolysis in adipose tissue was offset described by Katagiri et al.half-dozen They found that adenovirus-mediated expression of peroxisome proliferator-activated receptor γ2 in the liver induces astute hepatic steatosis, while markedly decreasing peripheral adiposity through afferent vagal signals originating in the liver, and that this neural pathway functions to protect confronting metabolic perturbation induced past pathologically excessive energy storage. By contrast to their finding, we accept institute that this neural axis has a physiological role and mediates the fasting signal from the liver to the adipose tissue under non-pathological conditions. Moreover, nosotros have clarified that glycogen shortage in the liver triggers the neural axis during fasting.

The liver is widely known to respond to a fasting state by activating various pathways, such equally cAMP-activated kinase7, AMPK8, peroxisome proliferator-activated receptor α9 and glucocorticoid receptor signalling10. Still, the known upstream triggers of these pathways are all external signals derived outside the liver, such as changes in the levels of claret glucose, fatty acids, insulin, glucagon, epinephrine, glucocorticoids and then on. By contrast, we have shown that the neural axis is driven by an internal signal reflecting glycogen shortage in the liver. Although the question 'What is the molecular identity of the glycogen sensor?' remains unanswered, it is conspicuously demonstrated that liver itself senses glycogen shortage during fasting independently of changes in claret glucose or metabolic hormone levels.

Every bit a candidate glycogen sensor, it has been reported that the AMPKβ subunit binds to and is inhibited by glycogen in muscle11. Still, our data from glycogen phosphorylase knockdown experiments (Fig. 4f,g) indicate that the blockade of glycolysis and the resultant higher glycogen levels suppresses the lipolytic activeness through the liver–brain–adipose axis despite the activation of AMPK, suggesting that AMPK is not involved in this regulation. Consequent with this finding, AMPK is also reported to exist hyperactivated in subjects with McArdle affliction, who are unable to mobilize glycogen because of an inherited defect in glycogen phosphorylase, despite the presence of high glycogen levels12. Therefore, in our experimental model, AMPK was non idea to be the primary glycogen sensor in sensing the glycogen shortage and triggering the neural axis.

It has been reported that glucocorticoid signals are also mediated past afferent vagal nerve pathway to control systemic insulin sensitivity and blood pressure2. Although nosotros focused on adipose tissue as the effecter of the neurocircuitry in the nowadays study, the liver-derived fasting signal transmitted through the vagal nerve might possibly have wider effects on the regulation of systemic energy metabolism. Further piece of work is required to understand the overall system triggered by hepatic vagal nerve signals.

In conclusion, our results demonstrated that the hepatic sensing of glycogen shortage activates a liver–brain–adipose neural axis to ensure energy availability by shifting the energy source from carbohydrates to fat.

Methods

Materials

Capsaicin were purchased from Wako Chemical (Tokyo, Japan). Guanethidine hemisulfate was purchased from Fluorochem (Derbyshire, U.k.). Anti-HSL, anti-phospho-HSL (ser 563), anti-ATGL, anti- phospho CRE-bounden poly peptide, anti- glycogen synthase, anti-AMPKα, anti- phospho AMPKα (Thr172), anti-AMPKβ1/2 and anti-phospho AMPKβ1 (Ser108) antibodies (rabbit polyclonal) were purchased from Cell Signaling Technology. Anti-PYGL (goat polyclonal) antibody was purchased from Santa Cruz Biotechnology.

Animals

Half dozen to seven-week-erstwhile ICR mice were purchased from CLEA (Tokyo, Japan). All animals were maintained in a temperature-controlled environment with a 12 h-light/nighttime cycle and were given free access to standard laboratory nutrition and water. Mice were killed in the early lite phase in a fasted or nonfasted (advertising libitum) state. All animals studied were anaesthetised and euthanized according to protocol approved by the Tokyo University Animal Care and Use Committee. All experiments were repeated at least twice.

Autopsy of hepatic branch of the vagus

14 days before adenovirus administration, vii-calendar week-old male mice were subjected to a selective HVx or to a sham operation (Sham). A laparotomy incision was made on the ventral midline and the intestinal muscle wall was opened with a second incision. The gastrohepatic ligament was severed using fine forceps, and the stomach was gently retracted, revealing the descending ventral oesophagus and the ventral subdiaphragmatic vagal trunk. The hepatic co-operative of this vagal torso was then transected using fine forceps.

Selective occludent of hepatic vagal afferent nerve

For perivagal awarding of capsaicin, the hepatic branch of the vagal trunk was exposed equally described above, and then loosely tied with a cotton cord immersed with or without capsaicin (Sigma Chemical) dissolved in olive oil (17%wt/vol) according to the modified Uno et al.6 method. The cotton string was removed thirty min later on and the abdominal incision was closed.

Sympathetic nerve blockade

For sympathetic nerve occludent experiments, Guanethidine (Fluorochem, Derbyshire, U.k.) was administered to 7-week-one-time male mice by a unmarried i.p. injection (100 mg kg^−one) xl h before kill4.

Preparation and transduction of recombinant adenoviruses

The fragment containing CRE with SV40 promoter linked to luciferase reporter gene was inserted into the Gateway entry vector pENTR4 (Invitrogen) and adenoviral constructs were generated by homologous recombination between the entry vector and the pAd promoterless vector (Invitrogen). The sequence of CRE was as follows: v′-CAGCCTGACGTCAGAGGCCTGACGTCAGAGAGCCTGACGTCAGAGAGCCTGACGTCAGA-three′. For adenovirus overexpressing Gys2 factor, cDNA encoding mouse Gys2 was PCR-amplified using primers 5′-GGTACCGCCACCATGCTCAGAGGCCGCTCC-3′ and 5′-CTCGAGATCATTTCGTTTGAGCTCAGTCAGTTC-3′ and subcloned into pENTR4 and then adenoviral constructs were generated by homologous recombination betwixt the entry vector and the pAd/CMV/V5-DEST vector (Invitrogen). Adenoviruses encoding shRNA targeting LacZ, Gys2 and Pygl genes (LacZ-i, Gys2-i and Pygl-i, respectively) were constructed by cloning synthetic DNA into pENTR/U6 entry vector followed by homologous recombination with the pAd promoterless vector (Invitrogen). The target sequences are as follows: v′-GCTAACGACATGCTCATAT-3′ for Gys2 and 5′-GGATAGTTATAATTGGTGG-3′ for Pygl. Recombinant adenoviruses were propagated in HEK293 cells and purified past CsCl gradient centrifugation as described previously13. Adenoviruses were injected intravenously into mice at the following doses: for expression of GFP, LacZ and Gys2, 5 × 10⁸ plaque-forming units (p.f.u.); for LacZ-i, Gys2-i and Pygl-i, 7.5 × 10⁸ p.f.u.

In vivo imaging of luciferase activeness

In vivo imaging was performed every bit described previously14. Briefly, i × 10⁸ p.f.u. of Advertizement-CRE-luc were delivered to epididymal fat in mice by micro syringe injection (Hamilton, Illinois, Us). Three to six days after the adenovirus transduction, three.0 mg of D-luciferin potassium salt (Wako Chemical) was injected into mice and the luminescence in the adipose tissue was captured using an IVIS Imaging Organisation (Xenogen) 15 min after the luciferin injection. Relative photon emission over the epididymal fat region was quantified using LivingImage software (Xenogen).

RNA isolation and Northern blotting

Full RNA preparation and absorb hybridization with cDNA probes were performed equally previously described15,16. Briefly, total RNA was isolated with Trizol reagent (Invitrogen), and 10 μg RNA samples every bit pooled from mice of each grouping were run on a 1% agarose gel containing formaldehyde and transferred to a nylon membrane. The cDNA probes used for hybridization are listed in Supplementary Tabular array S2. The probes were labelled with [α-32P]dCTP using Megaprime DNA Labelling Organisation kit (Amersham Biosciences). The membranes were hybridized with the radiolabeled probe in Rapid-hyb Buffer (Amersham Biosciences) at 65 °C, and done three times in 0.1 × SSC, 0.i% SDS at 65 °C. Blots were exposed to BAS imaging plate for the BAS2000 BIO Imaging Analyser (Fuji Photo Picture).

Quantitative reverse transcription PCR

Two micrograms of full RNA was opposite-transcribed using the High Capacity cDNA reverse transcription kit (Applied Biosystems). qRT–PCR was performed using SYBR Light-green Dye (Roche Applied Scientific discipline) on a LightCycler480II (Roche Practical Scientific discipline). The primer sets used for qRT–PCR are listed in Supplementary Table S3.

Metabolic measurements

Levels of glucose, insulin, free fatty acids, glycerol and β-hydroxybutyrate in plasma and glycogen in the liver were measured as previously described5,15. Glucagon and leptin were measured in plasma using a Bio-Plex Panel (Bio-Rad) according to the manufacturer'south instructions. FGF21 in plasma was measured by FGF21 ELISA kit (R&D organisation, Minneapolis, United states of america).

Immunoblotting

Immunoblotting was performed as described previously14. Briefly, whole-cell lysates or tissue extracts were fractionated by SDS–PAGE and transferred to a polyvinylidene difluoride membrane using a transfer apparatus according to the manufacturer's protocols (Bio-Rad). After blocking with five% nonfat milk in TBS-T (10 mM Tris at pH viii.0, 150 mM NaCl, 0.5% Tween twenty) for 30 min, the membrane was incubated with each master antibody at iv °C for 12 h. Membranes were done 3 times for 10 min and incubated with a 1:two,000 dilution of horseradish peroxidase-conjugated second antibody for i h. The primary antibodies used for immunoblotting are listed in Supplementary Table S4. Signals were detected using the ECL western blotting Detection System kit (Amersham Biosciences) and exposed to Kodak XAR-v pic and/or analysed using LAS-3000 imager (Fujifilm).

DEXA analysis

PIXImus2 DEXA (GE Medical Systems, LUNAR) was used to measure weight and per cent of lean body tissue and fat mass. Statistical differences between groups were analysed past ANOVA analysis of variance.

Metabolic cage analysis

After a 1-day acclimation period, oxygen and bicarbonate expired past 7-week-former male mice were measured every seven min for 25 h by a calorimetric arrangement (Alco Organization model, Chiba, Nihon). Fat utilization and energy consumption were calculated by followed formula; fat utilization=1.67 × (VO₂−VCO_ii) (mg min^−one), energy consumption=3.816 × VO₂+1.231 × VCO₂ (cal min^−ane).

Histological analysis

White adipose tissue was removed and fixed with 10% formalin and embedded in paraffin. Tissue sections were stained with hematoxylin eosin. Total adipocyte areas were traced manually and quantified with the NIH epitome program.

Statistical analyses

Data are expressed as means±s.eastward.m. Differences between 2 groups were assessed using the unpaired ii-tailed Student's t-test unless otherwise noted. Data sets involving more than two groups were assessed by Tukey's post-hoc test. The differences were considered to be meaning if P<0.05.

Author contributions

Y.I. and Due north.Y. conceived the experiments. Y.I. performed the experiments, and analysed the data together with North.Y. T.Y., H.K., G.I., K.M., A.Northward. and H.South. provide valuable help. All authors discussed the results and commented on the manuscript.

Additional data

How to cite this article: Izumida, Y. et al. Glycogen shortage during fasting triggers liver–brain–adipose neurocircuitry to facilitate fat utilization. Nat. Commun. 4:2316 doi: ten.1038/ncomms3316 (2013).

Supplementary Material

Supplementary Information:

Supplementary Figures S1-S8, Supplementary Tables S1-S4, Supplementary Methods and Supplementary References

Acknowledgments

This piece of work was supported by Grants-in-assistance from the Ministry building of Science, Education, Civilisation and Technology of Japan (to Y.I. and N.Y.), especially past a MEXT Grant-in-Aid Project (Scientific Research on Innovative Areas) 'Crosstalk between transcriptional control and energy pathways, mediated by hub metabolites' (to N.Y.). Information technology was also supported by research Grants from the Uehara Memorial Foundation, ONO Medical Research Foundation, Takeda Science Foundation, Suzuken Memorial Foundation, Nihon Heart Foundation, Kanae Foundation for the Promotion of Medical Scientific discipline, Senri Life Science Foundation, Japan Diabetes Foundation, Japan Foundation for Practical Enzymology and Okinaka Memorial Institute for Medical Research (to N.Y.).

References

• Reitman One thousand. 50. FGF21: a missing link in the biology of fasting. Cell Metab. v, 405–407 (2007). [PubMed] [Google Scholar]
• Bernal-Mizrachi C. et al.. An afferent vagal nerve pathway links hepatic pparalpha activation to glucocorticoid-induced insulin resistance and hypertension. Cell Metab. 5, 91–102 (2007). [PMC free article] [PubMed] [Google Scholar]
• Wan Z. et al.. Epinephrine-mediated regulation of PDK4 mRNA in rat adipose tissue. Am. J. Physiol. Jail cell Ph. 299, C1162–C1170 (2010). [PubMed] [Google Scholar]
• Haque K. S. et al.. Function of the sympathetic nervous system and insulin in enhancing glucose uptake in peripheral tissues later intrahypothalamic injection of leptin in rats. Diabetes 48, 1706–1712 (1999). [PubMed] [Google Scholar]
• Nakagawa Y. et al.. TFE3 transcriptionally activates hepatic IRS-ii, participates in insulin signaling and ameliorates diabetes. Nat. Med. 12, 107–113 (2006). [PubMed] [Google Scholar]
• Uno K. et al.. Neuronal pathway from the liver modulates free energy expenditure and systemic insulin sensitivity. Science 312, 1656–1659 (2006). [PubMed] [Google Scholar]
• Jiang Yard. & Zhang B. B. Glucagon and regulation of glucose metabolism. Am. J. Physiol. Endoc. M. 284, E671–E678 (2003). [PubMed] [Google Scholar]
• Kahn B. B., Alquier T., Carling D. & Hardie D. Yard. AMP-activated protein kinase: Ancient free energy gauge provides clues to modern understanding of metabolism. Cell Metab. ane, xv–25 (2005). [PubMed] [Google Scholar]
• Desvergne B. & Wahli West. Peroxisome proliferator-activated receptors: nuclear control of metabolism. Endocr. Rev. xx, 649–688 (1999). [PubMed] [Google Scholar]
• Schoneveld O. J. L. One thousand., Gaemers I. C. & Lamers West. H. Mechanisms of glucocorticoid signalling. Biochimica et Biophysica Acta 1680, 114–128 (2004). [PubMed] [Google Scholar]
• McBride A., Ghilagaber S., Nikolaev A. & Hardie D. G. The glycogen-binding domain on the AMPK beta subunit allows the kinase to human activity as a glycogen sensor. Cell Metab. nine, 23–34 (2009). [PMC free commodity] [PubMed] [Google Scholar]
• Nielsen J. N. et al.. Role of 5′-AMP-activated protein kinase in glycogen synthase activity and glucose utilization: insights from patients with McArdle'due south disease. J. Physiol. 541, 979–989 (2002). [PMC free article] [PubMed] [Google Scholar]
• Takeuchi Y. et al.. In vivo promoter assay on refeeding response of hepatic sterol regulatory element-binding protein-1c expression. Biochem. Biophys. Res. Commun. 363, 329–335 (2007). [PubMed] [Google Scholar]
• Takeuchi Y. et al.. Polyunsaturated fatty acids selectively suppress sterol regulatory element-binding protein-1 through proteolytic processing and autoloop regulatory circuit. J. Biol. Chem. 285, 11681–11691 (2010). [PMC free article] [PubMed] [Google Scholar]
• Yahagi N. et al.. Absence of sterol regulatory element-bounden protein-i (SREBP-1) ameliorates fat livers but not obesity or insulin resistance in Lep(ob)/Lep(ob) mice. J. Biol. Chem. 277, 19353–19357 (2002). [PubMed] [Google Scholar]
• Yahagi N. et al.. p53 activation in adipocytes of obese mice. J. Biol. Chem. 278, 25395–25400 (2003). [PubMed] [Google Scholar]

Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3753545/

Posted by: loweandlever.blogspot.com

Share this post