Biochimica et Biophysica Acta 1851 (2015) 566–576

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbalip

Caloric restriction and intermittent fasting alter hepatic lipid droplet proteome and diacylglycerol species and prevent diabetes in NZO mice Christian Baumeier a,b, Daniel Kaiser a,b, Jörg Heeren c, Ludger Scheja c, Clara John c, Christoph Weise d, Murat Eravci d, Merit Lagerpusch a,b, Gunnar Schulze a,b, Hans-Georg Joost a,b, Robert Wolfgang Schwenk a,b, Annette Schürmann a,b,⁎ a

Department of Experimental Diabetology, German Institute of Human Nutrition (DIfE), Arthur-Scheunert Allee 114-116, D-14558 Nuthetal, Germany German Center of Diabetes Research, Ingolstädter Landstraße 1, D-85764 Neuherberg, Germany c Department of Biochemistry and Molecular Cell Biology, University Medical Center Hamburg-Eppendorf, Martinistraße 52, D-20246 Hamburg, Germany d Freie Universität Berlin, Institut für Chemie und Biochemie, Thielallee 63, D-14195 Berlin, Germany b

a r t i c l e

i n f o

Article history: Received 18 November 2014 Received in revised form 8 January 2015 Accepted 21 January 2015 Available online 31 January 2015 Keywords: Caloric restriction Diacylglycerol Fatty acid oxidation Insulin resistance Intermittent fasting Lipid droplet proteome

a b s t r a c t Caloric restriction and intermittent fasting are known to improve glucose homeostasis and insulin resistance in several species including humans. The aim of this study was to unravel potential mechanisms by which these interventions improve insulin sensitivity and protect from type 2 diabetes. Diabetes-susceptible New Zealand Obese mice were either 10% calorie restricted (CR) or fasted every other day (IF), and compared to ad libitum (AL) fed control mice. AL mice showed a diabetes prevalence of 43%, whereas mice under CR and IF were completely protected against hyperglycemia. Proteomic analysis of hepatic lipid droplets revealed significantly higher levels of PSMD9 (co-activator Bridge-1), MIF (macrophage migration inhibitor factor), TCEB2 (transcription elongation factor B (SIII), polypeptide 2), ACY1 (aminoacylase 1) and FABP5 (fatty acid binding protein 5), and a marked reduction of GSTA3 (glutathione S-transferase alpha 3) in samples of CR and IF mice. In addition, accumulation of diacylglycerols (DAGs) was significantly reduced in livers of IF mice (P = 0.045) while CR mice showed a similar tendency (P = 0.062). In particular, 9 DAG species were significantly reduced in response to IF, of which DAG-40:4 and DAG-40:7 also showed significant effects after CR. This was associated with a decreased PKCε activation and might explain the improved insulin sensitivity. In conclusion, our data indicate that protection against diabetes upon caloric restriction and intermittent fasting associates with a modulation of lipid droplet protein composition and reduction of intracellular DAG species. © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction Type 2 diabetes (T2D) is a metabolic disorder that has become one of the major challenges of health care systems. The onset of this disease is dependent on the interplay of a subject's genetic background and the environment, and characterized by hyperglycemia as a consequence of inability of pancreatic beta-cells to secrete adequate amounts of insulin. This process is driven by overnutrition and inactivity, leading to overweight or obesity, and finally to insulin resistance and beta-cell dysfunction. In consideration of the fact of an increasing prevalence of T2D (592 million affected people estimated for 2035 [1]), it is of Abbreviations: AL, ad libitum; CR, caloric restriction; DAG, diacylglycerol; HOMA-IR, homeostatic model assessment for insulin resistance; IF, intermittent fasting; LD, lipid droplet; NZO, New Zealand Obese; T2D, type 2 diabetes ⁎ Corresponding author at: Department of Experimental Diabetology, German Institute of Human Nutrition, Arthur-Scheunert-Allee 114-116, D-14558 Nuthetal, Germany. Tel.: +49 33200 882368; fax: +49 33200 882334. E-mail address: [email protected] (A. Schürmann).

particular importance to develop prevention strategies to circumvent this pandemic development. One of the major theories linking obesity and insulin resistance is lipotoxicity, an increased accumulation of intracellular lipids in nonadipose tissues. Metabolically active lipid intermediates such as diacylglycerols (DAGs) [2,3], ceramides [4], long chain fatty acyl-CoAs [5] and acylcarnitines [6] are discussed to be involved in the suppression of insulin sensitivity. Increased accumulation of intracellular fat is dependent on storage and hydrolysis of neutral lipids, which in turn is regulated by lipid droplet (LD) associated proteins [7]. Since synthesis of toxic lipid intermediates is regulated by LD proteins, composition of LDs also plays a role in protection from lipotoxins [8,9]. However, in addition to some prominent LD proteins, such as perilipins [10], CIDE proteins [11], or PNLPA3 lipase [12], very little is known about the hepatic LD composition and its role in lipotoxicity and insulin resistance. As increased dietary intake is known to be a strong risk factor for insulin resistance and T2D, it is obvious that interventions such as a limited energy intake are promising approaches for improving insulin

http://dx.doi.org/10.1016/j.bbalip.2015.01.013 1388-1981/© 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

C. Baumeier et al. / Biochimica et Biophysica Acta 1851 (2015) 566–576

sensitivity and reducing T2D prevalence. Caloric restriction (CR) and intermittent fasting (IF) are two procedures of dietary restriction known for several beneficial effects on health and longevity [13,14]. Several studies in rodents and primates have shown that the reduction of daily caloric intake by 10–40% improves insulin sensitivity, reduces fasting glucose and insulin concentration and prevents obesity, T2D, hypertension and chronic inflammation [15–17]. In humans, 20% CR improves glucose tolerance and insulin action, and reduces risk factors for T2D, cardiovascular disease and cancer [18,19]. Furthermore, IF in rodents has been shown to prevent and reverse several indications of the metabolic syndrome, such as impaired insulin sensitivity, hyperglycemia, adiposity and T2D [20–22]. However, molecular mechanisms of CR and IF on lipotoxicity, LD composition, insulin sensitivity and T2D prevalence have not been identified. In the present study, we applied a moderate caloric restriction (10%) or an intermittent fasting regimen to New Zealand Obese (NZO) mice, a model for obesity, metabolic syndrome and T2D [23], and demonstrated that both interventions are sufficient to prevent the development of T2D, to modulate hepatic LD protein composition, and to reduce DAG levels.

567

Table 1 TaqMan assays used in quantitative real-time PCR. Gene

TaqMan probe

Fasn Acaca Scd1 Lipe Pnpla2 Eef2

Mm00662319_m1 Mm01304277_m1 Mm01197142_m1 Mm00495359_m1 Mm00503040_m1 Mm01171434_g1

2.6. Ex vivo palmitate oxidation Fatty acid oxidation of isolated extensor digitorum longus (EDL) and soleus muscles was measured essentially as described before [28,29]. 2.7. Glucose and insulin tolerance tests

2. Materials and methods

Eight to nine week old mice were either fasted overnight (oGTT) or for 2 h (ITT), and received an oral glucose bolus (3 g/kg lean mass) or an insulin injection (1.7 U/kg lean mass). Blood glucose and insulin levels were measured before and after administration.

2.1. Animals

2.8. Immunohistochemistry

Male NZO/HIBomDife mice were fed a high-fat diet (33 kcal% fat, 49 kcal% carbohydrate, 18 kcal% protein; S8022-E080 unsat.FA; Ssniff) ad libitum (AL), with 90% of the food consumed by AL mice (CR), or every other day (IF). Food was provided or removed daily at 4 p.m. Water was continuously available for all groups. The animals were kept in a 12 hour light/12 hour dark cycle and in accordance with EU Directive 2010/63/EU guidelines for the care and use of laboratory animals. All experiments were approved by the ethics committee of the State Agency of Environment, Health, and Consumer Protection (State of Brandenburg, Germany).

Paraffin sections of livers were stained with antibody against PLIN2. Nuclei were stained with TO-PRO 3 iodide (1:500), and Alexa Fluor 488F(ab′)2 fragment of goat anti-guinea pig IgG(H + L) (Invitrogen) was used as secondary antibody. Sections were analyzed by a Leica TCS SP2 Laser Scan inverted microscope.

2.2. Body weight, body composition and blood glucose measurements Body weight and blood glucose were measured weekly in the morning (8–10 a.m.) of a feeding day. Body fat and lean mass were determined with a nuclear magnetic resonance spectrometer EchoMRI (Echo Medical Systems). 2.3. Indirect calorimetry Respiratory quotient (RQ) was measured by indirect calorimetry as described before [24]. Carbohydrate and lipid oxidation were calculated according to the following equations: Carbohydrate oxidation [g/min] = 4.55 ∗ VCO2 − 3.21 ∗ VO2; and Lipid oxidation [g/min] = 1.67 ∗VO2 − 1.67 ∗ VCO2 [25].

2.9. Quantitative real-time PCR QRT-PCR was performed as described before [24] using corresponding TaqMan probes (Table 1). Gene expression was calculated as 2−ΔΔCT and expressed relative to the AL control group. 2.10. Western blot analysis Western blot analysis was performed as described before [26] using appropriate antibodies (Table 2). 2.11. Tissue lipid measurements Liver and skeletal muscle triglycerides were measured as described before [29]. The DAG extraction and ceramide extraction were performed by a modified method of Merrill [30]. Tissue (80 mg) was supplemented with internal standards (glyceryltritridecanoate, Nheptadecanoyl-D-erythro-sphingosine, 1-heptadecanoyl-2-hydroxysn-glycero-3-phosphocholine, 1-o-pentadecanyl-3-(9Z-octadecenoyl)sn-glycerol, 1,3(d5)-dipentadecanoyl-glycerol, 1,2-di-O-tridecyl-sn-

2.4. Plasma parameters and pancreatic insulin Plasma triglycerides, glycerol, non-esterified fatty acids and insulin as well as pancreatic insulin were measured as described before [26]. 2.5. In vivo lipolysis assay In vivo lipolysis was measured as described before [27]. Briefly, 7 week old mice were fed ad libitum overnight, followed by a 2 hour fasting period. Thereafter, mice were ip injected with 0.1 mg CL 316,243 per kg body weight (Sigma-Aldrich), a β3-adrenergic receptor agonist. Plasma glycerol and non-esterified fatty acids were measured before and 5, 10, 15, 30 and 60 min after the administration of CL 316,243 and normalized to body fat mass.

Table 2 List of antibodies used in immunohistochemistry and Western blot experiments. Primary antibody

Dilution

Host species

Company

PLIN2 PKCε CGI 58 ATGL E-Cadherin Syntaxin-6 Calnexin GAPDH PSMD9 (Bridge-1) MIF FABP5

1/500 1/1000 1/250 1/1000 1/2500 1/2000 1/2000 1/10,000 1/200 1/2000 1/1000

Guinea pig Mouse Mouse Rabbit Mouse Mouse Rabbit Mouse Rabbit Rabbit Rabbit

Progen BD Transduction Santa Cruz Cell Signaling BD Transduction BD Transduction Abcam Ambion Santa Cruz Abcam Abcam

568

C. Baumeier et al. / Biochimica et Biophysica Acta 1851 (2015) 566–576

glycero-3-phosphocholine, N-heptadecanoyl-D-erythro-sphingosylphosphorylcholine (purchased from either Sigma Aldrich or Avanti Polar Lipids)), homogenized in MeOH/CHCl3 (2/1 (vol/vol)), incubated at 48 °C for 12 h and centrifuged at 3500 ×g (10 min, 4 °C). Supernatant was dried and resuspended in MeOH/IPA, 4/6 (vol/vol), 5 mM NH4Ac for analysis by LC–MS/MS. Samples were separated on a Kinetex XBC18 150 mm × 2.1 mm × 1.7 μm column (Phenomenex®) with a flow rate of 300 μl/min. Quantitative analysis was performed with the mass spectrometer (ESI-qToF, maXis® 3G (Bruker Daltonic GmbH)) operated in positive ionization mode. Lipid species were identified by means of standard substances, MS/MS-spectra and the Lipidmaps database. 2.12. Isolation of LD, cytosolic and membrane fractions from liver Subcellular fractionation of fresh livers (200 mg) was performed as described before [31]. Tissue was homogenized in 1600 μl of homogenization buffer A (20 mM Tris–HCl (pH 7.4), 1 mM EDTA, 0.25 mM EGTA, 250 mM sucrose, protease and phosphatase inhibitors (Roche)), overlayed with 400 μl of 3% sucrose and centrifuged at 100,000 ×g for 1 h at 4 °C. The lipid cake was removed with a 23-G1 needle, and cytoplasm was transferred to a fresh tube. The remaining pellet was resuspended in 700 μl buffer B (20 mM Tris HCl (pH 7.5 at 4 °C), 150 mM NaCl, 50 mM NaF, 5 mM NaPPi, 1 mM EDTA, 1 mM EGTA, 1 mM PMSF, protease and phosphatase inhibitors (Roche)). Subsequently, samples were homogenized with a 25-G needle and centrifuged at 20,800 × g for 15 min at 4 °C. The remaining LD fraction was removed with a 24-G needle and Triton-X was added to 2% (vol/vol) to the supernatant. Membrane samples were passed through a 27-G needle and incubated for 30 min on ice before being centrifuged at 20,800 × g for 15 min at 4 °C. Following, pellet was resuspended in 80 μl buffer B. Proteins in LD fraction were precipitated in acetone at −80 °C overnight, followed by a centrifugation at 20,800 × g for 30 min at 4 °C. Protein pellet was dried and resuspended in buffer B with 2% Triton X (vol/vol).

3. Results 3.1. Caloric restriction and intermittent fasting prevent the development of hyperglycemia Male NZO mice on a high-fat diet (HFD) were subjected to different feeding patterns (Fig. 1A). CR mice received 90% of the food consumed by ad libitum fed control mice (AL), and IF mice had access to food ad libitum only every other day. The average food intake of AL control mice was 4.8 ± 0.1 g/day and of CR mice 4.3 ± 0.1 g/day (Fig. 1B). On feeding days, IF mice consumed 6.9 ± 0.1 g/day (+ 41% compared to AL mice) resulting in a decline of cumulative food intake by 27% (Fig. 1C). Body mass development and lean mass development in CR and IF mice were reduced (Fig. 2A,B). IF mice displayed also lower body fat mass compared to AL controls (Fig. 2C). Already after 5 weeks on HFD, AL mice developed hyperglycemia (blood glucose concentration N 16.6 mM; Fig. 2D). At 14 weeks of age the diabetes prevalence was 43% for AL mice, whereas not a single mouse of the CR and IF groups became diabetic (Fig. 2D). 3.2. Intermittent fasting improved insulin sensitivity Glucose and insulin tolerance tests were performed 4 weeks after initiation of the interventions. In oral glucose tolerance tests (oGTTs) all mice showed similar blood glucose excursions (Fig. 2E). However, plasma insulin levels during oGTT were lower in the CR and IF groups (Fig. 2F). Insulin tolerance tests (ITTs) revealed higher insulin sensitivity only in animals of the IF group (Fig. 2G). Fasting insulin levels of IF mice were 37% lower than those of AL mice (0.59 ± 0.07 μg/l vs. 0.94 ± 0.15 μg/l, P = 0.051), whereas insulin levels of CR mice (1.08 ± 0.15 μg/l) were similar to AL controls (Fig. 2H). HOMA-IR index was higher in AL (17.9 ± 7.4) than in IF mice (5.0 ± 0.7; P b 0.05 vs. AL and CR mice) and did not show significant differences in CR mice (12.5 ± 3.2; P = 0.741 vs. AL mice).

2.13. Protein identification by LC–MS

3.3. Intermittent fasting improved metabolic flexibility

Equal amounts of protein from LD fractions (20 μg) were digested with LysC and trypsin. After digestion peptide samples were desalted by solid phase extraction (SPE), using C18 empore disc cartridges (Supelco). Desalted peptide mixtures were separated by reversed phase chromatography using the Dionex Ultimate 3000 nanoLC on inhouse manufactured 25 cm fritless silica microcolumns with an inner diameter of 75 μm. Columns were packed with the ReproSil-Pur C18AQ 3 μm resin. Peptides were separated on a 5–60% acetonitrile gradient (90 min) with 0.1% formic acid at a flow rate of 200 nl/min. Eluting peptides were ionized online by electrospray ionization and transferred into an LTQ Orbitrap Velos mass spectrometer. After MS and MS/MS acquisition raw files were analyzed using MaxQuant version 1.3.0.5. The MS/MS spectra were matched against Uniprot mouse database and 248 frequently observed laboratory contaminants as provided by the MaxQuant software package, and calculated using the MaxQuant search engine Andromeda [32].

AL mice showed a nearly stable respiratory quotient (RQ) of about 0.94 ± 0.04 indicating the predominant oxidation of carbohydrates (Fig. 3A). CR mice displayed slightly increased RQ in the dark and lower RQ at the end of the light phase (Fig. 3A). IF mice had a constantly low RQ (≈0.7) at fasting, and high RQ (≥1.0) at feeding times suggesting increased lipogenesis from carbohydrates (Fig. 3A). The calculation of carbohydrate and lipid oxidation performed as described by Frayn [25], confirmed the exclusive utilization of carbohydrates during feeding and lipids during fasting periods, and shows a fast switch between carbohydrate and lipid metabolism in IF animals (Fig. 3B,C). In comparison to AL, CR mice showed also an improved rhythm of carbohydrate and lipid metabolism (Fig. 3B,C). In order to test whether the rapid switch of metabolic pathways is only due to the severe fasting/feeding rhythm or an adaptation of IF mice, we subjected AL mice to a 24 hour fasting/feeding pattern. Although AL mice were responsive to fasting (RQ ≈ 0.7), they did not show a similar change from lipid to carbohydrate oxidation (or vice versa) as IF animals (Fig. 3D). Thus, IF results in an adaptive metabolic response to feeding/fasting cycles and a pronounced rhythm in substrate utilization, a so-called metabolic flexibility. Accordingly, mRNA of lipogenic acetyl-Coenzyme A carboxylase alpha (Acaca) and stearoyl-Coenzyme A desaturase 1 (Scd1) was strongly up-regulated in the white adipose tissue (WAT) of the IF group, but was not altered by CR in comparison to the AL control (Fig. 3E). Levels of lipolytic hormone sensitive lipase (Lipe) were higher in WAT of IF animals, whereas those of adipose triglyceride lipase (Pnpla2) were not changed by CR or IF (Fig. 3E). Elevated Lipe expression indicates an increased release of free fatty acids from WAT. In fact, injection of the β3-adrenergic receptor agonist CL-316,243 resulted in an increase of plasma glycerol and free fatty

2.14. Microarray analysis Liver transcriptome analysis was performed by OakLabs (Berlin, Germany) using a SurePrint G3 Mouse GE 8 × 60 k chip (Agilent Technologies). 2.15. Statistical analysis Statistical analysis was performed by either one-way ANOVA or two-way ANOVA, or by the unpaired Student t test or Mann–Whitney test using the software Prism 6 from GraphPad Software (USA). Significance levels were set for P-values b 0.05 (*), 0.01 (**) and 0.001 (***).

C. Baumeier et al. / Biochimica et Biophysica Acta 1851 (2015) 566–576

A

569

Access to food

AL CR

IF Light

6:00

Dark

Light

6:00

18:00

Dark

18:00

6:00

Time (h)

B 1 .2

AL

0 .9 0 .6 0 .3 0 .0 1 6 :0 0

1 6 :0 0

1 6 :0 0

1 .2

CR

0 .9 0 .6

*

0 .3 0 .0 1 6 :0 0

*

*

1 6 :0 0

1 6 :0 0

1 .2

* * * *

0 .9

C u m u la t iv e F o o d In t a k e ( g )

F o o d In ta k e ( g ) F o o d In ta k e ( g ) F o o d In ta k e ( g )

C 400 300

* ***

200 100 0 AL

CR

IF

IF

0 .6 0 .3 0 .0 1 6 :0 0

*

**

* *

1 6 :0 0

1 6 :0 0

T im e ( h ) Fig. 1. Different feeding pattern cause changes in food intake. (A) Schematic illustration of the three applied feeding regimens. (B, C) Food intake is displayed hourly over a period of 48 h (B), or cumulative for the whole study period of 10 weeks (C). Gray areas represent 12 h dark phase. Food access is depicted on top. Data are mean ± SEM of 6 (B) and 14 (C) mice per group. *P b 0.05, ***P b 0.001 vs. AL group.

acids, which was most pronounced in animals of the IF group, but also slightly elevated in CR mice (Fig. 3F,G). Thus, our data demonstrate an increased capacity of adipocyte lipolysis in animals subjected to IF. 3.4. Intermittent fasting reduced specific diacylglycerols by increasing fatty acid oxidation in glycolytic muscles We next analyzed the lipid profile of muscle and liver at the age of 11–13 weeks. In muscle, the levels of triglycerides (Fig. 4A) as well as of ceramides (Fig. 4B) were not affected by CR or IF, whereas the diacylglycerol (DAG) concentrations were significantly reduced in IF mice and showed a similar tendency but no significant effects in the CR group (P = 0.114) (Fig. 4C). Among 23 DAG species analyzed, 16 were detected in skeletal muscle of NZO mice (Fig. 4D). The majority of DAGs was represented by 5 distinct species (DAG-16:0-18:1, DAG-16:0-18:2, DAG-18:1-18-1, DAG-18:1-18:2, DAG-20:4-16:0), which were all significantly reduced in IF muscles. Muscles from CR mice exhibited a tendency of reduced concentrations of these particular DAG species (P-values 0.100–0.147), however, DAG-20:4-18:0 was significantly reduced by both, CR and IF (Fig. 4D).

To elucidate whether the reduction of DAGs is the result of an increased fatty acid oxidation, we measured palmitate oxidation in isolated extensor digitorum longus (EDL) and soleus muscles. In soleus muscles, no difference in palmitate oxidation between the groups was observed (Fig. 4E). In EDL muscles, however, beta-oxidation of IF mice was increased by 114% (P = 0.076) at basal conditions, and by 60% (P b 0.01) upon AMPK activation by AICAR (Fig. 4E). CR tended to increase palmitate oxidation in EDL muscle, albeit to a lower extent (48% increment basal, P = 0.392; 41% increment AICAR-stimulated, P = 0.057; Fig. 4E). 3.5. Intermittent fasting reduced hepatic diacylglycerol levels and PKCε activity In comparison to AL, CR and IF mice exhibited much smaller hepatic lipid droplets (LDs) as indicated by staining of PLIN2 (Fig. 5A). Hepatic triglyceride content was significantly reduced in IF mice, whereas CR animals showed only a similar tendency but did not reveal significant effects (P = 0.175) (Fig. 5B). Ceramide levels were not affected by CR or IF (Fig. 5C), but a 2-fold decrease in DAG content in livers of IF mice

570

C. Baumeier et al. / Biochimica et Biophysica Acta 1851 (2015) 566–576

B 60

AL

50

CR

30

L e a n M a s s (g )

B o d y W e ig h t ( g )

A IF

40 30 20

**

10 0

AL CR IF

20

10

0 3

6

9

12

15

3

6

A g e (w e e k s )

CR IF

20

**

10

** 0 3

6

9

25 20

% d ia b .

F a t M a s s (g )

AL

B lo o d G lu c o s e ( m M )

D 30

***

25 0 AL CR

15

**

AL

0 3

20 10

30

60

IF

90

AUC

5 4

CR

IF

0 60

IF

3 2

*

1

** **

AL

0 0

90

120

F a s tin g In s u lin ( g /l)

*

30

** CR

30

*** CR

60

IF

90

120

H

10

0

250

T im e ( m in )

15

AL

15

500

AL

120

ITT

5

12

0

T im e ( m in )

G

IF

9

oGTT P la s m a In s u lin ( g /l)

B lo o d G lu c o s e ( m M )

30

0

CR

6

A g e (w e e k s )

40

0

**

5

F

CR

IF

10

12

oGTT

AL

12

50

A g e (w e e k s )

E

9

A g e (w e e k s )

C

B lo o d G lu c o s e ( m M )

*

**

P = 0 .0 5 1

2 .0

*

1 .5 1 .0 0 .5 0 .0

AL

CR

IF

T im e ( m in ) Fig. 2. Caloric restriction and intermittent fasting reduce body weight gain, prevent hyperglycemia and improve insulin sensitivity. (A–D) Development of body weight (A), lean mass (B), fat mass (C), and blood glucose as well as diabetes prevalence at 14 weeks of age (D) of NZO mice on different feeding regimens (n = 14 mice/group). (E, F) Oral glucose tolerance test (oGTT) (n = 6 mice/group). (G) Insulin tolerance test (ITT) (n = 7–10 mice/group). (H) Fasting insulin (n = 7 mice/group). Data are mean ± SEM. *P b 0.05, **P b 0.01, ***P b 0.001 vs. AL group, or as indicated.

was detected (Fig. 5D). DAG levels of CR livers were similarly but not significantly reduced (P = 0.062). Among 24 detected DAG species we found 9 (DAG-16:0-18:0, DAG-16:0-18:2, DAG-18:0-18:0, DAG-18:018:1, DAG-18:1-18:1, DAG-18:1-18:2, DAG-20:1-18:0, DAG-40:4, DAG-40:7) to be significantly reduced by IF, of which DAG-40:4 and DAG-40:7 showed also significant effects after CR (Fig. 5E). Among the major changes in DAG species, 5 (DAG-16:0-18:0, DAG-16:0-18:2, DAG-18:0-18:1, DAG-18:1-18:1, DAG-18:1-18:2) were overlapping

and simultaneously reduced in liver and muscle of IF mice. A known consequence of increased hepatic DAG levels is the activation of PKCε by its translocation to the plasma membrane where it inhibits insulinreceptor kinase activity [3]. Actually, decreased hepatic DAG concentration of IF mice was linked to lower PKCε levels in the membrane fractions indicating a reduced PKCε activity (Fig. 5F). CR also reduced PKCε content in membranes, however, to a lower not significant extent (P = 0.085).

C. Baumeier et al. / Biochimica et Biophysica Acta 1851 (2015) 566–576

B

RQ

1 .5

1 .0

0 .8

0 .6 1 6 :0 0

0 .5

T im e ( h )

( g /m in )

0 .8 *

1 .0

*

***

0 .5

CR *

0 .2

*

0 .0 1 6 :0 0

1 6 :0 0

1 6 :0 0

0 .0 1 6 :0 0

1 6 :0 0

1 6 :0 0

1 .5

IF

0 .3

IF

IF

( g /m in )

**

1 .0

0 .8

***

1 .0

0 .5

0 .2

0 .1

**

0 .6 1 6 :0 0

*

0 .0 1 6 :0 0

1 6 :0 0

1 6 :0 0

T im e ( h )

1 6 :0 0

R e la tiv e e x p r e s s io n ( fo ld )

E 1 .2 ***

***

1 .0

0 .8 AL IF 1 6 :0 0

1 6 :0 0

1 6 :0 0

T im e ( h )

1 6 :0 0

1 6 :0 0

8

AL

***

CR

6

IF

*** 4

***

2

*** 0

Fasn A caca S cd1 L ip e P n p la 2 (F A S ) (A C C 1 ) (S C D 1 ) (H S L ) (A T G L )

L ip o g e n e s is

L ip o ly s is

G *****

*

100

AL CR IF

50 0 0

20

40

60

80

Time after CL 316,243 injection (min)

)

* -1

200 150

0 .4

***

(m M g

250

FFA / Fat M ass

Glycerol / Fat Mass ( g ml-1 g-1)

F

*

**

T im e ( h )

RQ (V C O 2 /V O 2 )

0 .0 1 6 :0 0

1 6 :0 0

T im e ( h )

D

0 .6 1 6 :0 0

1 6 :0 0

1 6 :0 0

T im e ( h )

T im e ( h )

1 .2

**

0 .1

*

*

1 6 :0 0

1 6 :0 0

0 .3

CR

( g /m in )

(V C O 2 /V O 2 )

*

T im e ( h )

(V C O 2 /V O 2 )

0 .1

T im e ( h )

1 .5

CR

0 .6 1 6 :0 0

0 .2

0 .0 1 6 :0 0

1 6 :0 0

1 6 :0 0

AL

T im e ( h )

1 .2

1 .0

AL

1 .0

0 .0 1 6 :0 0

1 6 :0 0

1 6 :0 0

Lipid Oxidation 0 .3

( g /m in )

AL

( g /m in )

(V C O 2 /V O 2 )

1 .2

C

CH Oxidation

( g /m in )

A

571

***

**

*

0 .3 0 .2 AL

0 .1

CR IF

0 .0 0

20

40

60

80

T im e a fte r C L 3 1 6 ,2 4 3 in je c tio n ( m in )

Fig. 3. Intermittent fasting increases metabolic flexibility by enhancing lipid metabolism. (A, D) Respiratory quotient (RQ) was measured in 9 week old mice (n = 6 mice/group). In addition, VO2 and VCO2 were used to calculate carbohydrate (CH) and lipid oxidation rates (B, C). (D) RQ of AL and IF mice subjected to 24 h feeding/24 h fasting cycles (n = 4–6 mice/group). Gray areas represent 12 hour dark phase. Food access is depicted on top. (E) Real-time PCR analysis of lipogenic and lipolytic gene expression in gonadal white adipose tissue (n = 9–13 mice/ group). Tissues were taken from 14 week old mice after a 6 h fasting period. (F, G) In vivo lipolysis. Plasma glycerol (F) and free fatty acids (FFA) (G) were measured before and after injection of the β3-adrenergic receptor agonist CL-316,243 and corrected to body fat mass (n = 10–13 mice/group). Data are mean ± SEM. *P b 0.05, **P b 0.01, ***P b 0.001 vs. AL group.

572

C. Baumeier et al. / Biochimica et Biophysica Acta 1851 (2015) 566–576

2 .0 1 .5 1 .0 0 .5 0 .0

CR

IF

D A G (n m o l / g tis s u e )

TG ( g /

2 .5

AL

C C e ra m id e s (n m o l / g tis s u e )

B

g p ro te in )

A

150

100

50

0

AL

CR

2000

1500 P = 0 .1 1 4

1000

*

500

0

AL

IF

CR

IF

630

AL

430

CR IF

230

*

*

30 30

*

*

*

20

1 6 :0

2 2 :5

1 8 :0

2 0 :4

2 2 :3

1 8 :1

2 0 :1

1 8 :0

2 0 :1

1 6 :0

2 0 :4

1 8 :2

1 8 :1

1 8 :1

1 8 :1

1 8 :1

1 8 :0

1 8 :2

1 6 :1

1 8 :2

1 6 :0

1 8 :1

*

*

* 1 6 :0

1 8 :0

1 6 :0

1 6 :1

1 6 :0

1 6 :1

1 6 :0

1 6 :0

0

**

*

*

*

1 6 :0

10

1 6 :1

D A G (n m o l / g tis s u e )

D

P a lm ita te O x id a tio n (fo ld )

E EDL

6 5

*

*

*

AL CR

*

IF

**

4 3

S o le u s

P = 0 .0 7 6

2 1 0

B asal

A IC A R

B asal

A IC A R

Fig. 4. Intermittent fasting decreases muscular diacylglycerols by increasing lipid oxidation in glycolytic muscles. (A) Concentrations of triglycerides (TG; n = 8–12 mice/group), (B) ceramides (n = 4–5 mice/group) and (C) diacylglycerols (DAG; n = 4–5 mice/group) were analyzed in quadriceps muscles of 11–13 week old AL, CR and IF mice. (D) Profile of single DAG species with corresponding fatty acid residues. (E) Ex vivo palmitate oxidation in isolated EDL (more glycolytic) and soleus (more oxidative) muscles from 11–13 week old mice (n = 4–8 mice/group). Palmitate oxidation was measured under basal conditions or after AICAR stimulation. Data are mean ± SEM. *P b 0.05, **P b 0.01 vs. AL control, or as indicated.

3.6. Caloric restriction and intermittent fasting affected hepatic LD composition As LD proteins are involved in maintaining lipid homeostasis and protection from lipotoxicity, we isolated LDs from livers of the three groups at the age of 11–13 weeks and analyzed their proteome by LC– MS. Fig. 6A shows Western blot analysis of the isolated fractions (LDs, membranes, cytosol) and demonstrates that each fraction was enriched with corresponding marker proteins. CGI-58 and ATGL were detected in LD fractions. Plasma membrane, ER and Golgi-specific proteins were detected in the membrane fraction and the cytosolic GAPDH was found in the cytosolic fraction. However, each fraction was moderately contaminated with proteins from other compartments. By proteomics of LD fractions we detected 613 proteins. Increased abundance (P b 0.05, fold change ≥ 2.0) at LDs from CR and IF livers was found for 18 proteins when compared to AL LDs (Fig. 6B). Of these PSMD9, MIF, TCEB2, ACY1 and FABP5 showed significantly higher levels in LDs of the CR and IF groups. KPNBP1, ATG3, DAP, MPST, ADH4, TTPA, EIF5, CALD1, CORO1b and ISOC1 were significantly elevated in LD

fractions of IF livers and showed a similar tendency in CR fractions. Three proteins (GAMT, THRSP and CDV3) exhibited higher abundance in the LD compartment after CR with a similar tendency in IF samples. Four proteins (GSTA3, UOX, CROT and A2LD1) were reduced in the LD fraction of livers from CR and IF mice (Fig. 6C). Array-based transcriptomics of the same livers showed that only five of the transcripts (Psmd9, Kpnb1, Cdv3, Uox and Crot) exhibited significantly altered expression between the three groups, of which only three (Psmd9, Kpnb1 and Crot) were altered in the same direction as detected by the proteomic approach (Fig. 6D,E). However, Western blot analyses performed with liver lysates demonstrated that the total amount of PSMD9, MIF and FABP5 proteins was not elevated in response to CR and IF. It rather appears that the protein expression of FABP5 is slightly reduced in IF livers (Fig. 6F). 4. Discussion The present study demonstrated that a moderate caloric restriction by 10% has the same beneficial effect in preventing T2D in the NZO mice as a very strict intervention, the intermittent fasting. However,

C. Baumeier et al. / Biochimica et Biophysica Acta 1851 (2015) 566–576

AL

A

573

IF

CR

B

0 .6 P = 0 .1 7 5

TG ( g /

0 .2

*** 0 .0 AL

CR

IF

D A G (n m o l / g tis s u e )

0 .8

0 .4

D C e ra m id e s (n m o l / g tis s u e )

g p ro te in )

C 2500 2000 1500 1000 500 0

AL

CR

800

600

400 P = 0 .0 6 2

*

200

0

IF

AL

CR

IF

165

AL

115

CR

*

65

IF

*

15 15

*

10

*

IF PKCε E-Cadherin

4 0 :7

4 0 :6

4 0 :5

4 0 :4

4 0 :3

4 0 :2

4 0 :1

2 2 :6

1 6 :0

1 6 :0

2 2 :5

2 0 :4

1 8 :0

1 6 :0

2 2 :3

1 8 :1

1 8 :0

2 0 :1

/ E -C a d h e rin

CR

PKC

AL

2 0 :1

1 6 :0

2 0 :4

1 8 :2

1 8 :1

1 8 :1

1 8 :1

1 8 :1

F

**

**

*

* 1 8 :0

1 8 :0

1 8 :2

1 8 :2

1 6 :0

1 8 :1

1 6 :0

1 8 :0

1 6 :0

1 6 :1

1 6 :0

1 6 :0

* *

*

1 8 :0

*

0

1 6 :1

5 1 6 :0

D A G (n m o l / g tis s u e )

E

1 .2 P = 0 .0 8 5

0 .8

*

0 .4 0 .0

AL

CR

IF

Fig. 5. Intermittent fasting reduces hepatic lipid accumulation and PKCε activation. (A) Representative confocal immunofluorescence images of liver sections from 6 h fasted, 14 week old AL, CR and IF mice stained with antibody against PLIN2 (green). Nuclei were stained with TO-PRO (blue). (B–E) Concentrations of hepatic triglycerides (TG) (B), ceramides (C) and diacylglycerols (DAG) (D, E) in livers of 14 week old AL, CR and IF mice (n = 8–12 mice/group). (E) Profile of single DAG species with corresponding fatty acid residues. Composition of very long-chain fatty acid containing DAG (≥40) is ambiguous and therefore represented as total DAG (40:1-40:7). (F) Immunoblot analysis and densitometry of PKCε in the membrane fraction from livers of AL, CR and IF mice (n = 3 mice/group). Data are mean ± SEM. *P b 0.05, ***P b 0.001 vs. AL control.

574

C. Baumeier et al. / Biochimica et Biophysica Acta 1851 (2015) 566–576

A

AL LD

M C

CR

IF

LD M C

LD M C CGI-58 (Lipid Droplet) ATGL (Lipid Droplet/Cytosol) E-Cadherin (Membrane) Syntaxin-6 (Golgi) Calnexin (ER) GAPDH (Cytosol)

Proteomics (liver LDs)

* *

10

* 5

* *

*

*

* * *

*

* *

*

* **

*

*

*

*

*

*

1 .5

* 0 .5

* * *

*

0 .0 G S T A3

C D V3

THR SP

G AMT

IS O C 1

C O R O 1b

C ALD 1

E IF 5

TTPA

AD H4

MP S T

D AP

AT G 3

KPNB1

F AB P 5

AC Y1

TC EB2

M IF

0

1 .0

A2LD 1

IF

15

1 .5

CROT

CR

UO X

AL

R e la tiv e A b u n d a n c e (fo ld )

20

R e la tiv e E x p r e s s io n ( fo ld )

C

P S MD 9

R e la tiv e A b u n d a n c e (fo ld )

B

Transcriptomics (whole liver) E

1 .5

*

F

*

* 1 .0

*

AL

CR

C dv 3

T h rs p

G am t

Is o c 1

C o ro 1 b

C a ld 1

E if5

T tp a

M ps t

A dh4

D ap

A tg 3

K pnb1

Fabp5

Acy1

Tc eb2

0 .0

M if

0 .5

* 1 .0

* 0 .5

nd

0 .0

A 2 ld 1

IF

C ro t

CR

U ox

AL

G s ta 3

2 .0

P s m d9

R e la tiv e E x p r e s s io n ( fo ld )

D

IF PSMD9 MIF FABP5 GAPDH

Fig. 6. Effects of caloric restriction and intermittent fasting on hepatic lipid droplet composition. (A) Immunoblot analysis of lipid droplet (LD), membrane (M) and cytosol (C) fractions isolated from livers of AL, CR and IF mice. Antibodies against specific markers for LDs (CGI-58, ATGL), plasma membrane (E-Cadherin), Golgi apparatus (Syntaxin-6), ER (Calnexin) and cytosol (GAPDH, ATGL) were used. (B, C) Proteomic analysis of liver LDs (n = 4–7 mice/group). List of proteins which were either more (B), or less abundant (C) in LD fractions of CR and/or IF mice (P b 0.05, fold change ≥ 2.0, vs. AL mice). (D, E) mRNA expression of proteins discovered by LD proteomics was analyzed in the same livers by array-based transcriptomics approach (n = 4–7 mice/group). (F) Western blot analysis of PSMD9, MIF, and FABP5 on total liver lysates of AL, CR and IF mice (n = 3 mice/group). LD proteome, Western blotting and liver transcriptome were performed with samples of 11–13 week old, 4 h fasted mice. nd, not detected. Data are mean ± SEM. *P b 0.05 vs. AL control.

C. Baumeier et al. / Biochimica et Biophysica Acta 1851 (2015) 566–576

the molecular mechanisms responsible for preventing hyperglycemia are still not entirely clarified. Beneficial effects of IF are significantly associated with alterations in hepatic LD protein composition and the reduction of DAG species in liver and muscle, and these effects show at least the same tendency in the CR group. It can be speculated that an altered LD composition in IF livers participates in the reduction of DAG concentrations which prevent the activation of PKCε and inhibition of insulin signaling. Another reason for the prevention of T2D by CR and IF could be the slightly reduced body weight. Storage and hydrolysis of intracellular fat are regulated by proteins associated with LDs. As lipid-derived metabolites cause progressive lipotoxicity and insulin resistance, LD proteins are important targets in protection from lipotoxic events and consequently from non-alcoholic fatty liver disease (NAFLD), insulin resistance and T2D [33]. Earlier studies have already identified LD proteins in rat and mice livers [34, 35] of which 84% overlap with proteins we identified in our approach. The majority of identified proteins were not classical LD associated proteins like lipases or PLINs. Since LDs closely interact with other subcellular organelles, some of the identified proteins may actually be contaminants during preparation, but we assume that such contaminations are similar in all groups. However, a number of proteins in the LD fraction were markedly affected by CR and IF. Interestingly, most of these proteins were altered in their LD abundance rather than in their mRNA expression indicating an impact of caloric restriction and fasting on the localization, stability or degradation of particular proteins. As it is suggested that the proteome of individual LDs is different [36], it is possible that the metabolic state determines either the pattern of proteins associated with particular LDs or the composition of different subpopulations of LDs within the liver. Among the differentially abundant proteins, 18 were elevated at LDs of CR and IF mice, of which five (PSMD9, MIF, TCEB2, ACY1 and FABP5) exhibited similar and significant effects in both groups and are therefore the most likely candidates to play a role in mediating protective effects when localized at hepatic LDs. GSTA3 (glutathione S-transferase alpha 3) is a candidate that might exhibit a negative effect on LDs, because its abundance on LDs of CR and IF livers was significantly decreased. The strongest effect was seen with the co-activator Bridge-1 (PSMD9), which is in linkage with T2D, T2D-nephropathy and macrovascular pathology in Italian families [37]. PSMD9 transduces signals that regulate insulin production, proliferation, and survival of pancreatic beta-cells [38], and pancreatic overexpression of Psmd9 results in insulin deficiency and T2D [39]. However, hepatic PSMD9 and its role at LDs are not yet known. As PSMD9 is a proteasomal chaperone it can be speculated that in the CR and IF states LDs are associated with the proteasome to be degraded. Along this line we also identified ATG3, an ubiquitin-like-conjugating enzyme to be elevated in LD fractions of IF mice which could indicate that prolonged fasting regimens accelerate the degradation of lipids through the lysosomal degradative pathway of autophagy, designated lipophagy [40]. Macrophage migration inhibitor factor (MIF) was also elevated at LDs of CR and IF livers. Mif-deficient mice develop hyperglycemia and glucose intolerance [41], and administration of MIF to human hepatocytes reduced lipid accumulation [42]. In addition, there is evidence that exercise as well as CR induces Mif expression in murine livers [42,43]. As we did not observe differences in hepatic Mif mRNA and protein expression after CR and IF, our data suggest a role of MIF in regulating hepatic lipid metabolism by its recruitment to LDs. Also the fatty acid binding protein 5 (FABP5) was more abundant at LDs of CR and IF mice. In contrast, its protein expression was rather decreased in total liver lysates of IF mice, supporting its increased localization at LDs in these animals. FABP5 is a member of fatty acid carrier proteins that facilitate the transfer of fatty acids between extra- and intracellular membranes. Increased FABP5 association with LDs after CR and IF suggests an elevated transport of fatty acids to/ from LDs and thereby indicates an increased lipid turnover. Another interesting effect we detected especially in the IF group, was the reduced level of muscular and hepatic DAGs. Previous studies

575

have found that reducing calories rapidly decreases hepatic fat content and improves insulin sensitivity in T2D patients [2,3]. In these studies, similar to our findings, DAGs but not ceramides correlated with insulin sensitivity. Moreover, we showed that the reduction of hepatic DAGs is linked to a decreased PKCε activation in livers of CR and IF mice suggesting a direct effect to improved insulin sensitivity in these animals. Among the different DAG species DAG-20:4-18:0 in muscle and DAG40:4 and DAG-40:7 in liver were significantly reduced by both interventions, and several DAG-16 and DAG-18 species showed similar effects in these tissues. This might indicate that particularly these species mediate negative effects on insulin sensitivity under conditions of overnutrition. The diminished DAG content appears to be the result of a combination of a decreased energy intake, and an increased mitochondrial betaoxidation as detected in EDL muscles of IF mice. In agreement, previous studies in mice [44] and humans [45] demonstrated an induced lipid oxidation by IF. Thus, our data link an increased lipid oxidation by IF to the reduction of ectopic accumulation of intracellular DAGs. In summary, we showed that extended fasting enhances lipid oxidation in glycolytic muscles, which in turn affects DAG accumulation and insulin sensitivity, presumably via reduced activation of novel PKCs. Beneficial anti-diabetic effects of CR and IF can also be the consequence of other changes that only reach significance after the strict intermittent fasting procedure. This could be the case for the improved metabolic flexibility that might participate in the enhancement of insulin sensitivity. The change of caloric intake and feeding rhythm is known to influence metabolism and source of energy which is preferentially utilized. Feeding a HFD ad libitum results in a disturbed feeding behavior with frequent food intake throughout day and night. In addition, NZO mice are genetically predisposed for hyperphagia [46]. A study in HFD-fed C57BL/6J mice showed a protection against diet-induced obesity, hyperinsulinemia, hepatic steatosis, and inflammation by restricting feeding to an 8 hour period per day without reducing calories [47]. In our study CR and IF also limited feeding to a particular time interrupted by different fasting periods in between. Thus, it is likely that CR and IF improve feeding rhythm – for instance by affecting satiety – leading to protective effects on metabolism. However, we did not expect that (1) the CR mice did not react with a reduced fat mass (Fig. 2) and (2) glucose clearance during oral glucose tolerance tests was not improved by CR and IF (Fig. 3). The fact that insulin concentrations during the glucose challenge were significantly lower in IF with a similar tendency in CR indicates that limiting food intake improves insulin sensitivity, presumably as a consequence of lower ectopic fat accumulation in the liver (Fig. 5). Accordingly, an improved insulin tolerance could be detected at least for the IF mice (Fig. 2). 5. Conclusion In summary, we found that a moderate caloric restriction as well as an intermittent fasting are suitable to prevent the onset of type 2 diabetes in NZO mice. Although suppression of diabetes by these interventions has been shown before, this is the first study linking changes in hepatic lipid droplet proteome and levels of specific diacylglycerols to improved insulin sensitivity and protection from hyperglycemia and beta-cell loss, however, without showing causality. This finding may prompt further studies aiming to understand the role of particular lipid droplet proteins in the regulation of hepatic insulin sensitivity. Duality of interest The authors declare that there is no duality of interest associated with this manuscript. Author contributions C.B. performed data acquisition and analysis, and drafted the article. R.W.S., D.K., M.L., M.E., C.W., L.S., C.J. and J.H. performed data acquisition

576

C. Baumeier et al. / Biochimica et Biophysica Acta 1851 (2015) 566–576

and analysis. G.S. performed bioinformatics data analysis. H-G.J. participated in the study conception and advised on the study concept. A.S. was responsible for the study conception and drafted the article. A.S. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Acknowledgements and notice of grant support

[23]

[24]

[25] [26]

The authors thank Michaela Rath and Andrea Teichmann for their skillful technical assistance. This work was supported by the German Federal Ministry of Education and Research (BMBF, DZD, grant 01GI0922) and the German Research Foundation (DFG, SFB-958). J.H. is supported by the EU FP7 project RESOLVE (FP7-HEALTH-2012-305707).

[27]

[28]

References [1] International Diabetes Federation, IDF Diabetes Atlas, 6th ed., 2013. [2] S.J. Heydrick, N.B. Ruderman, T.G. Kurowski, H.B. Adams, K.S. Chen, Enhanced stimulation of diacylglycerol and lipid synthesis by insulin in denervated muscle. Altered protein kinase C activity and possible link to insulin resistance, Diabetes 40 (1991) 1707–1711. [3] F.R. Jornayvaz, G.I. Shulman, Diacylglycerol activation of protein kinase Cε and hepatic insulin resistance, Cell Metab. 15 (2012) 574–584. [4] J.M. Adams, T. Pratipanawatr, R. Berria, E. Wang, R.A. DeFronzo, M.C. Sullards, L.J. Mandarino, Ceramide content is increased in skeletal muscle from obese insulinresistant humans, Diabetes 53 (2004) 25–31. [5] B.A. Ellis, A. Poynten, A.J. Lowy, S.M. Furler, D.J. Chisholm, E.W. Kraegen, G.J. Cooney, Long-chain acyl-CoA esters as indicators of lipid metabolism and insulin sensitivity in rat and human muscle, Am. J. Physiol. Endocrinol. Metab. 279 (2000) E554–E560. [6] S.J. Mihalik, B.H. Goodpaster, D.E. Kelley, D.H. Chace, J. Vockley, F.G.S. Toledo, J.P. DeLany, Increased levels of plasma acylcarnitines in obesity and type 2 diabetes and identification of a marker of glucolipotoxicity, Obesity (Silver Spring) 18 (2010) 1695–1700. [7] A.S. Greenberg, R.A. Coleman, F.B. Kraemer, J.L. McManaman, M.S. Obin, V. Puri, Q.-W. Yan, H. Miyoshi, D.G. Mashek, The role of lipid droplets in metabolic disease in rodents and humans, J. Clin. Invest. 121 (2011) 2102–2110. [8] L. Liu, X. Shi, C.S. Choi, G.I. Shulman, K. Klaus, K.S. Nair, G.J. Schwartz, Y. Zhang, I.J. Goldberg, Y.-H. Yu, Paradoxical coupling of triglyceride synthesis and fatty acid oxidation in skeletal muscle overexpressing DGAT1, Diabetes 58 (2009) 2516–2524. [9] C. Wang, Y. Zhao, X. Gao, L. Li, Y. Yuan, F. Liu, L. Zhang, J. Wu, P. Hu, X. Zhang, Y. Gu, Y. Xu, Z. Li, H. Zhang, J. Ye, Perilipin 5 improves hepatic lipotoxicity by inhibiting lipolysis, Hepatology (2015). http://dx.doi.org/10.1002/hep.27409. [10] M. Bell, H. Wang, H. Chen, J.C. McLenithan, D.-W. Gong, R.-Z. Yang, D. Yu, S.K. Fried, M.J. Quon, C. Londos, C. Sztalryd, Consequences of lipid droplet coat protein downregulation in liver cells: abnormal lipid droplet metabolism and induction of insulin resistance, Diabetes 57 (2008) 2037–2045. [11] J.Z. Li, J. Ye, B. Xue, J. Qi, J. Zhang, Z. Zhou, Q. Li, Z. Wen, P. Li, Cideb regulates dietinduced obesity, liver steatosis, and insulin sensitivity by controlling lipogenesis and fatty acid oxidation, Diabetes 56 (2007) 2523–2532. [12] M.K. Basantani, M.T. Sitnick, L. Cai, D.S. Brenner, N.P. Gardner, J.Z. Li, G. Schoiswohl, K. Yang, M. Kumari, R.W. Gross, R. Zechner, E.E. Kershaw, Pnpla3/Adiponutrin deficiency in mice does not contribute to fatty liver disease or metabolic syndrome, J. Lipid Res. 52 (2011) 318–329. [13] L. Fontana, L. Partridge, V.D. Longo, Extending healthy life span — from yeast to humans, Science 328 (2010) 321–326. [14] V.D. Longo, M.P. Mattson, Fasting: molecular mechanisms and clinical applications, Cell Metab. 19 (2014) 181–192. [15] J.A. Mattison, G.S. Roth, T.M. Beasley, E.M. Tilmont, A.M. Handy, R.L. Herbert, D.L. Longo, D.B. Allison, J.E. Young, M. Bryant, D. Barnard, W.F. Ward, W. Qi, D.K. Ingram, R. de Cabo, Impact of caloric restriction on health and survival in rhesus monkeys from the NIA study, Nature 489 (2012) 318–321. [16] R.M. Anderson, D. Shanmuganayagam, R. Weindruch, Caloric restriction and aging: studies in mice and monkeys, Toxicol. Pathol. 37 (2009) 47–51. [17] S.Y. Park, G.H. Choi, H.I. Choi, J. Ryu, C.Y. Jung, W. Lee, Calorie restriction improves whole-body glucose disposal and insulin resistance in association with the increased adipocyte-specific GLUT4 expression in Otsuka Long–Evans Tokushima fatty rats, Arch. Biochem. Biophys. 436 (2005) 276–284. [18] E.P. Weiss, S.B. Racette, D.T. Villareal, L. Fontana, K. Steger-May, K.B. Schechtman, S. Klein, J.O. Holloszy, Improvements in glucose tolerance and insulin action induced by increasing energy expenditure or decreasing energy intake: a randomized controlled trial, Am. J. Clin. Nutr. 84 (2006) 1033–1042. [19] L. Fontana, S. Klein, Aging, Adiposity, and Calorie Restriction, 297 (2007). [20] R. Wan, S. Camandola, M.P. Mattson, Intermittent fasting and dietary supplementation with 2-deoxy-D-glucose improve functional and metabolic cardiovascular risk factors in rats, FASEB J. 17 (2003) 1133–1134. [21] C.R. Pedersen, I. Hagemann, T. Bock, K. Buschard, Intermittent feeding and fasting reduces diabetes incidence in BB rats, Autoimmunity 30 (1999) 243–250. [22] R.M. Anson, Z. Guo, R. de Cabo, T. Iyun, M. Rios, A. Hagepanos, D.K. Ingram, M.A. Lane, M.P. Mattson, Intermittent fasting dissociates beneficial effects of dietary

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37] [38]

[39]

[40] [41]

[42]

[43]

[44]

[45]

[46]

[47]

restriction on glucose metabolism and neuronal resistance to injury from calorie intake, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 6216–6220. R. Kluge, S. Scherneck, A. Schürmann, H.-G. Joost, Pathophysiology and genetics of obesity and diabetes in the New Zealand obese mouse: a model of the human metabolic syndrome, Methods Mol. Biol. 933 (2012) 59–73. N. Schulz, H. Himmelbauer, M. Rath, M. van Weeghel, S. Houten, W. Kulik, K. Suhre, S. Scherneck, H. Vogel, R. Kluge, P. Wiedmer, H.-G. Joost, A. Schürmann, Role of medium- and short-chain L-3-hydroxyacyl-CoA dehydrogenase in the regulation of body weight and thermogenesis, Endocrinology 152 (2011) 4641–4651. K.N. Frayn, Calculation of substrate oxidation rates in vivo from gaseous exchange, J. Appl. Physiol. Respir. Environ. Exerc. Physiol. 55 (1983) 628–634. O. Kluth, F. Mirhashemi, S. Scherneck, D. Kaiser, R. Kluge, S. Neschen, H.-G. Joost, A. Schürmann, Dissociation of lipotoxicity and glucotoxicity in a mouse model of obesity associated diabetes: role of forkhead box O1 (FOXO1) in glucose-induced beta cell failure, Diabetologia 54 (2011) 605–616. K.T. Uysal, L. Scheja, S.M. Wiesbrock, S. Bonner-Weir, G.S. Hotamisligil, Improved glucose and lipid metabolism in genetically obese mice lacking aP2, Endocrinology 141 (2000) 3388–3396. A. Chadt, K. Leicht, A. Deshmukh, L.Q. Jiang, S. Scherneck, U. Bernhardt, T. Dreja, H. Vogel, K. Schmolz, R. Kluge, J.R. Zierath, C. Hultschig, R.C. Hoeben, A. Schürmann, H.-G. Joost, H. Al-Hasani, Tbc1d1 mutation in lean mouse strain confers leanness and protects from diet-induced obesity, Nat. Genet. 40 (2008) 1354–1359. S. Keipert, M. Ost, A. Chadt, A. Voigt, V. Ayala, M. Portero-Otin, R. Pamplona, H. Al-Hasani, S. Klaus, Skeletal muscle uncoupling-induced longevity in mice is linked to increased substrate metabolism and induction of the endogenous antioxidant defense system, Am. J. Physiol. Endocrinol. Metab. 304 (2013) E495–E506. A.H. Merrill, M.C. Sullards, J.C. Allegood, S. Kelly, E. Wang, Sphingolipidomics: highthroughput, structure-specific, and quantitative analysis of sphingolipids by liquid chromatography tandem mass spectrometry, Methods 36 (2005) 207–224. J.L. Cantley, T. Yoshimura, J.P.G. Camporez, D. Zhang, F.R. Jornayvaz, N. Kumashiro, F. Guebre-Egziabher, M.J. Jurczak, M. Kahn, B.A. Guigni, J. Serr, J. Hankin, R.C. Murphy, G.W. Cline, S. Bhanot, V.P. Manchem, J.M. Brown, V.T. Samuel, G.I. Shulman, CGI-58 knockdown sequesters diacylglycerols in lipid droplets/ER-preventing diacylglycerolmediated hepatic insulin resistance, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 1869–1874. J. Cox, N. Neuhauser, A. Michalski, R.A. Scheltema, J.V. Olsen, M. Mann, Andromeda: a peptide search engine integrated into the MaxQuant environment, J. Proteome Res. 10 (2011) 1794–1805. A.S. Greenberg, R.A. Coleman, F.B. Kraemer, J.L. Mcmanaman, M.S. Obin, V. Puri, Q. Yan, H. Miyoshi, D.G. Mashek, Review Series: The Role of Lipid Droplets in Metabolic Disease in Rodents and Humans, 121 (2011). S. Turró, M. Ingelmo-Torres, J.M. Estanyol, F. Tebar, M.A. Fernández, C.V. Albor, K. Gaus, T. Grewal, C. Enrich, A. Pol, Identification and characterization of associated with lipid droplet protein 1: a novel membrane-associated protein that resides on hepatic lipid droplets, Traffic 7 (2006) 1254–1269. A.E. Crunk, J. Monks, A. Murakami, M. Jackman, P.S. Maclean, M. Ladinsky, E.S. Bales, S. Cain, D.J. Orlicky, J.L. McManaman, Dynamic regulation of hepatic lipid droplet properties by diet, PLoS One 8 (2013) e67631. H. Zhang, Y. Wang, J. Li, J. Yu, J. Pu, L. Li, H. Zhang, S. Zhang, G. Peng, F. Yang, P. Liu, Proteome of skeletal muscle lipid droplet reveals association with mitochondria and apolipoprotein a-I, J. Proteome Res. 10 (2011) 4757–4768. C. Gragnoli, Hypercholesterolemia and a candidate gene within the 12q24 locus, Cardiovasc. Diabetol. 10 (2011) 38. M.K. Thomas, S.W. Tsang, M.-L. Yeung, P.S. Leung, K.-M. Yao, The roles of the PDZcontaining proteins bridge-1 and PDZD2 in the regulation of insulin production and pancreatic beta-cell mass, Curr. Protein Pept. Sci. 10 (2009) 30–36. J.L. Volinic, J.H. Lee, K. Eto, V. Kaur, M.K. Thomas, Overexpression of the coactivator bridge-1 results in insulin deficiency and diabetes, Mol. Endocrinol. 20 (2006) 167–182. K. Liu, M.J. Czaja, Regulation of lipid stores and metabolism by lipophagy, Cell Death Differ. 20 (2013) 3–11. I. Nikolic, M. Vujicic, T. Saksida, T. Berki, S. Stosic-Grujicic, I. Stojanovic, The role of endogenous glucocorticoids in glucose metabolism and immune status of MIF-deficient mice, Eur. J. Pharmacol. 714 (2013) 498–506. H.Y. Moon, P. Song, C.S. Choi, S.H. Ryu, P.-G. Suh, Involvement of exercise-induced macrophage migration inhibitory factor in the prevention of fatty liver disease, J. Endocrinol. 218 (2013) 339–348. R.A. Miller, Y. Chang, A.T. Galecki, K. Al-Regaiey, J.J. Kopchick, A. Bartke, Gene expression patterns in calorically restricted mice: partial overlap with long-lived mutant mice, Mol. Endocrinol. 16 (2002) 2657–2666. E. Krízová, V. Simek, Effect of intermittent feeding with high-fat diet on changes of glycogen, protein and fat content in liver and skeletal muscle in the laboratory mouse, Physiol. Res. 45 (1996) 379–383. L.K. Heilbronn, S.R. Smith, C.K. Martin, S.D. Anton, E. Ravussin, Alternate-day fasting in nonobese subjects: effects on body weight, body composition, and energy metabolism, Am. J. Clin. Nutr. 81 (2005) 69–73. H.S. Jürgens, A. Schürmann, R. Kluge, S. Ortmann, S. Klaus, H.-G. Joost, M.H. Tschöp, Hyperphagia, lower body temperature, and reduced running wheel activity precede development of morbid obesity in New Zealand obese mice, Physiol. Genomics 25 (2006) 234–241. M. Hatori, C. Vollmers, A. Zarrinpar, L. DiTacchio, E.A. Bushong, S. Gill, M. Leblanc, A. Chaix, M. Joens, J.A.J. Fitzpatrick, M.H. Ellisman, S. Panda, Time-restricted feeding without reducing caloric intake prevents metabolic diseases in mice fed a high-fat diet, Cell Metab. 15 (2012) 848–860.