R

Rh

TDD

h

•••••

a

ARRAA

KRHIN

1

kbissRs

tl[

U8

h0

Neuroscience Letters 619 (2016) 1–7

Contents lists available at ScienceDirect

Neuroscience Letters

jo ur nal ho me page: www.elsev ier .com/ locate /neule t

esearch paper

eactive oxygen species mediate insulin signal transduction in mouseypothalamus

akeshi Onoue, Motomitsu Goto ∗, Takashi Tominaga, Mariko Sugiyama, Taku Tsunekawa,aisuke Hagiwara, Ryoichi Banno, Hidetaka Suga, Yoshihisa Sugimura, Hiroshi Arima

epartment of Endocrinology and Diabetes, Nagoya University Graduate School of Medicine, Showa-ku, Nagoya 466-8550, Japan

i g h l i g h t s

Insulin increased intracellular ROS in mouse hypothalamic explants.H2O2 by itself significantly increased p-IR� and p-Akt levels.Insulin-induced p-IR� and p-Akt increases were attenuated by NADPH oxidase inhibitor.NADPH oxidase inhibitor also attenuated insulin-induced intracellular ROS.Insulin-induced p-IR� and p-Akt were mediated via ROS produced by NADPH oxidase.

r t i c l e i n f o

rticle history:eceived 5 January 2016eceived in revised form 29 February 2016ccepted 7 March 2016

a b s t r a c t

In the hypothalamus, several reports have implied that ROS mediate physiological effects of insulin. Inthis study, we investigated the mechanisms of insulin-induced ROS production and the effect of ROS oninsulin signal transduction in mouse hypothalamic organotypic cultures. Insulin increased intracellularROS, which were suppressed by NADPH oxidase inhibitor. H2O2 increased phospho-insulin receptor � (p-

vailable online 9 March 2016

eywords:eactive oxygen species

IR�) and phospho-Akt (p-Akt) levels. Insulin-induced increases in p-IR� and p-Akt levels were attenuatedby ROS scavenger or NADPH oxidase inhibitor. Our data suggest that insulin-induced phosphorylationof IR� and Akt is mediated via ROS which are predominantly produced by NADPH oxidase in mousehypothalamus.

ypothalamus

nsulin signalADPH oxidase

. Introduction

Hydrogen peroxide (H2O2) and superoxide anion radicals,nown as reactive oxygen species (ROS), are regarded as toxicyproducts of aerobic metabolism. The accumulation of ROS result-

ng from nutrient overload and mitochondrial exhaustion areuggested to play a causal role in insulin resistance in peripheral tis-ues [1]. Contrary to this, a growing body of evidence suggests thatOS also function as intracellular second messengers that promoteignaling by hormones, including insulin [2–4].

Insulin plays a pivotal physiological role not only in peripheral

issues, but also in the central nervous system [5]. In the hypotha-amus, insulin regulates energy balance by reduction of food intake6,7], activation of sympathetic nerve outflow to the brown adi-

∗ Corresponding author at: Department of Endocrinology and Diabetes, Nagoyaniversity Graduate School of Medicine, 65 Tsurumai−cho, Showa−ku, Nagoya 466-550, Japan.

E-mail address: goto@med.nagoya-u.ac.jp (M. Goto).

ttp://dx.doi.org/10.1016/j.neulet.2016.03.011304-3940/© 2016 Elsevier Ireland Ltd. All rights reserved.

© 2016 Elsevier Ireland Ltd. All rights reserved.

pose tissue [8], or suppression of hepatic endogenous glucoseproduction [9,10]. Recently, it was reported that insulin-stimulatedmitochondrial H2O2 production triggers autophosphorylation ofinsulin receptors in cultured cerebellar granule neurons [11]. Itwas also reported that the increase of ROS levels within thehypothalamus decreases food intake [12]. Furthermore, centrallyadministered insulin reportedly increases hypothalamic ROS levels,and pharmacological suppression of ROS production by intracere-broventricular injection of NADPH oxidase inhibitor abolished theanorexigenic effect of insulin [13]. These evidences suggest thatROS are generated in response to insulin stimulation and enhancethe physiological effects of insulin in the hypothalamus. However,the molecular mechanisms by which insulin produces ROS andwhether or not ROS by themselves have an effect on the moleculesinvolved in insulin signal transduction in the hypothalamus is notclear so far.

In this study, we investigated the precise molecular mechanismsof insulin-induced ROS production and the effect of ROS on insulinsignal transduction in mouse hypothalamic organotypic cultures,which maintain the intrinsic properties [14,15].

2 T. Onoue et al. / Neuroscience Letters 619 (2016) 1–7

F lturefl , and 1

2

2

mpcl

2

Ntu

2

wdS

ig. 1. Insulin-induced intracellular ROS production in hypothalamic organotypic cuuorescence microscopy. Fluorescence change in control (A–C), 10−4 M H2O2 (D–F)

. Materials and methods

.1. Animals

All procedures were approved by the Animal Care and Use Com-ittee of Nagoya University Graduate School of Medicine, and

erformed in accordance with the institutional guidelines whichonform with the National Institutes of Health animal care guide-ines.

.2. Slice-explant culture procedure

Sixteen-day-old C57BL6/J mice (Chubu Science Materials,agoya, Japan) were sacrificed by decapitation, and hypothalamic

issue slice cultures were performed at 37 ◦C in 5% CO2 enriched airnder stationary conditions, as described previously [16].

.3. Dispersed cell culture procedure

To assess intracellular ROS production, dispersed cell culturesere performed. Five-day-old C57BL6/J mice were sacrificed by

ecapitation, and hypothalamic cells were dissociated using aUMITOMO Nerve-Cell culture system kit (MB-X0802, SUMITOMO

s. Intracellular ROS production was assessed by fluorescence of CM-H2DCF-DA with0−7 M insulin (G–I). 3 V, third ventricle. Scale bar 100 �m.

Bakelite, Tokyo, Japan) according to the manufacturer’s instruc-tions. Dissociated cells were cultivated for 5 days at 37 ◦C in 5% CO2enriched air.

2.4. Intracellular ROS measurement of hypothalamic explants

Intracellular ROS production of hypothalamic explantswas assessed with fluorescent probe 5,6-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate (CM-H2DCF-DA, LifeTechnologies, MD, USA) as described previously [17]. Hypothala-mic explants were serum-starved overnight, incubated in the darkwith CM-H2DCF-DA (10−5 M) for 60 min at 37 ◦C, and treated withinsulin (10−7 M), H2O2 (10−4 M, Wako, Osaka, Japan) or controlmedium for 15 min. Fluorescence of CM-H2DCF-DA was assessedwith fluorescence microscopy (BX-51, Olympus, Tokyo, Japan) atan excitation wavelength of 488 nm and emission at 515–540 nm.

2.5. Intracellular ROS measurement of dispersed hypothalamiccells

Intracellular ROS production of the dispersed hypothala-mic cells was assessed with fluorescent probe CellROX orange(Life Technologies, MD, USA) according to the manufacturer’s

T. Onoue et al. / Neuroscience Letters 619 (2016) 1–7 3

Fig. 2. Insulin-induced intracellular ROS production in dispersed hypothalamic cells. Intracellular ROS production was assessed by fluorescence of CellROX orange withconfocal fluorescence microscopy. Densitometry analysis of fluorescence and representative fluorescence images are shown. Relative fluorescence intensities were expressedas mean ± SE (n = 4). * P < 0.05 vs. control. Scale bar 20 �m. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of thisarticle.)

Fig. 3. Effects of insulin or H2O2 on phospho-IR� (Tyr1150/1151) and phospho-Akt (Ser473) in hypothalamic organotypic cultures. Hypothalamic explants were incubatedw e. Dosa ) on

ioOdb

ith the indicated concentration of insulin or H2O2 for the indicated periods of timnd p-Akt. Dose-response of H2O2 at 3 h (E, F) and time-course of 10−4 M H2O2 (G, H

nstructions. Hypothalamic dispersed cells were serum-starved

vernight, treated with insulin (10−7 M), H2O2 (10−4 M, Wako,saka, Japan) or control medium for 30 min, and incubated in theark with CellROX orange (5 �M) for 30 min at 37 ◦C, followedy washing twice with pre-warmed PBS. Fluorescence of CellROX

e-response of insulin at 3 h (A, B) and time-course of 10−7 M insulin (C, D) on p-IR�p-IR� and p-Akt. Results are expressed as mean ± SE (n = 4). * P < 0.05 vs. control.

orange was assessed with confocal fluorescence microscopy (A1R,

Nikon, Tokyo, Japan) at an excitation wavelength of 543 nm andemission at 565 nm. Relative fluorescence was quantified usingImageJ software (http://rsb.info.nih.gov/ij/).

4 T. Onoue et al. / Neuroscience Letters 619 (2016) 1–7

F -Akt

i 4). * P

2h

optcpst

2h

pmiTepst

2h

oeeUJtab

2t

AwiAa

ig. 4. Effects of coadministration of insulin and H2O2 on phospho-IR� and phosphonsulin (10−7 M) and H2O2 (10−4 M) for 3 h. Results are expressed as mean ± SE (n =

.6. Effects of insulin on insulin signal transduction inypothalamic organotypic cultures

To examine the dose-response or time-course effects of insulinn phospho-insulin receptor � subunit (p-IR�, Tyr 1150/1151) andhospho-Akt (p-Akt, Ser 473) in hypothalamic organotypic cul-ures, hypothalamic explants were incubated with the indicatedoncentration of insulin for the indicated periods of time. The totalroteins from the hypothalamic slices were extracted 72 h aftertarting the cultures, and subjected to analyses with Western blot-ing.

.7. Effects of H2O2 on insulin signal transduction inypothalamic organotypic cultures

To examine the dose-response or time-course effects of H2O2 on-IR� and p-Akt in hypothalamic organotypic cultures, hypothala-ic explants were serum-starved overnight and incubated with the

ndicated concentration of H2O2 for the indicated periods of time.he H2O2-containing medium and control medium were replacedvery 30 min to maintain appropriate H2O2 concentration. The totalroteins from the hypothalamic slices were extracted 72 h aftertarting the cultures, and subjected to analyses with Western blot-ing.

.8. Effects of ROS scavenger on insulin signal transduction inypothalamic organotypic cultures

To explore the role of ROS on insulin-induced phosphorylationf IR� or Akt in hypothalamic organotypic cultures, hypothalamicxplants were serum-starved overnight and treated with ROS scav-nger, N-Acetyl-L-cysteine (NAC, 5 × 10−3 M, Sigma-Aldrich, MO,SA) or vehicle (dimethyl sulfoxide, DMSO, 0.1%, Wako, Osaka,

apan) in the presence or absence of insulin (10−7 M) for 3 h. Theotal proteins from the hypothalamic slices were extracted 72 hfter starting the cultures, and subjected to analyses with Westernlotting.

.9. Effects of NADPH oxidase inhibitor on insulin signalransduction in hypothalamic organotypic cultures

To examine the effects of NADPH oxidase on p-IR� and p-kt in hypothalamic organotypic cultures, hypothalamic explants

ere serum-starved overnight and treated with NADPH oxidase

nhibitor, diphenyleneiodonium chloride (DPI, 10−6 M, Sigma-ldrich, MO, USA) or vehicle (DMSO, 0.1%) in the presence orbsence of insulin (10−7 M) or H2O2 (10−4 M) for 3 h. The total

in hypothalamic organotypic cultures. Hypothalamic explants were incubated with < 0.05 vs. control. N.S., not significant.

proteins from the hypothalamic slices were extracted 72 h afterstarting the cultures, and subjected to analyses with Western blot-ting.

2.10. Effects of mitochondrial complex II inhibitor on insulinsignal transduction in hypothalamic organotypic cultures

To examine the effects of mitochondrial ROS on p-IR� and p-Aktin hypothalamic organotypic cultures, hypothalamic explants wereserum-starved overnight and treated with mitochondrial complexII inhibitor, thenoyltrifluoroacetone (TTFA, 10−4 M, Sigma-Aldrich,MO, USA) or vehicle (ethanol, 0.05%, Wako, Osaka, Japan) in thepresence or absence of insulin (10−7 M) for 3 h. The total proteinsfrom the hypothalamic slices were extracted 72 h after starting thecultures, and subjected to analyses with Western blotting.

2.11. Effects of NADPH oxidase inhibitor or mitochondrialcomplex II inhibitor on insulin-induced intracellular ROSproduction of dispersed hypothalamic cells

To examine the effects of NADPH oxidase inhibitor or mito-chondrial complex II inhibitor on insulin-induced intracellularROS production, hypothalamic dispersed cells were serum-starvedovernight, treated with vehicle (DMSO, 0.1%, ethanol, 0.05%),insulin (10−7 M), DPI (10−6 M) or TTFA (10−4 M) for 30 min, andincubated in the dark with CellROX orange (5 �M) for 30 min at37 ◦C, followed by washing twice with pre-warmed PBS. Fluores-cence of CellROX orange was assessed as described above.

2.12. Western blotting analysis

Western blotting was performed as described previously [16].Antibodies for p-IR� (Tyr1150/1151, #3024), IR� (#3025), p-Akt(Ser473, #9271) and Akt (#4691) were purchased from Cell Signal-ing Technology (Beverly, MA, USA). The membranes were strippedand incubated with antibodies against each unphosphorylated pro-tein for normalization. Densitometric analyses of the bands wereperformed by ImageJ software (http://rsb.info.nih.gov/ij/).

2.13. Statistical analysis

The statistical significance of the differences between groups

was analyzed with one-way ANOVA followed by Dunnett’s test (vs.control) or Bonferroni’s test (multiple comparisons). Results wereexpressed as means ± SE, and differences were considered signifi-cant at P < 0.05.

T. Onoue et al. / Neuroscience Letters 619 (2016) 1–7 5

F II inhil vehiclD essed

3

3h

flhs(u(flca

ig. 5. Effects of ROS scavenger, NADPH oxidase inhibitor or mitochondrial complexamic explants were treated with NAC (5 × 10−3 M), DPI (10−6 M), TTFA (10−4 M) or

) and TTFA (E, F) on insulin-induced increases in p-IR� and p-Akt. Results are expr

. Results

.1. Insulin-induced intracellular ROS production in theypothalamus

Intracellular ROS production was assessed by observation ofuorescence of CM-H2DCF-DA with fluorescence microscopy onypothalamic explants. Representative fluorescence images arehown in Fig. 1. Fluorescence change in control was only slightFig. 1A-C). As expected, administration of H2O2 (10−4 M) grad-ally increased fluorescence ubiquitously in hypothalamic slicesFig. 1D-F). Administration of insulin (10−7 M) by itself increased

uorescence compared to control (Fig. 1G-I). Increase of fluores-ence compared to control was more apparent 15 min after insulindministration.

bitor on insulin signal transduction in hypothalamic organotypic cultures. Hypotha-e in the presence or absence of insulin (10−7 M) for 3 h. Effects of NAC (A, B), DPI (C,

as mean ± SE (n = 4). * P < 0.05 vs. control. # P < 0.05 vs. insulin. N.S., not significant.

To quantify intracellular ROS production, dispersed hypothala-mic cell cultures were performed and intracellular ROS productionwas assessed by observing fluorescence of CellROX orangewith confocal fluorescence microscopy. Administration of H2O2(10−4 M) or insulin (10−7 M) significantly increased fluorescencecompared to control. Representative fluorescence images areshown (Fig. 2).

3.2. Effects of insulin and H2O2 on insulin signal transduction

Dose response experiments showed that insulin (from 10−7 to

10−6 M) significantly increased p-IR� (Tyr1150/1151) and p-Akt(Ser473) levels at 3 h (Fig. 3A and B). Time course experimentsshowed that insulin (10−7 M) significantly increased p-IR� andp-Akt levels at 1 h, and phosphorylation of IR� was significantly

6 T. Onoue et al. / Neuroscience

Fig. 6. Effects of NADPH oxidase inhibitor or mitochondrial complex II inhibitoron insulin-induced intracellular ROS production of dispersed hypothalamic cells.Hypothalamic dispersed cells were treated with vehicle, insulin (10−7 M), insulin(10−7 M) with DPI (10−6 M) or insulin (10−7 M) with TTFA (10−4 M) for a total of 1 h.Intracellular ROS production was assessed by observing fluorescence of CellROXorange with confocal fluorescence microscopy. Densitometry analysis of fluores-cence and representative fluorescence images are shown. Relative fluorescencei(r

iseosHsTiwel

3m

llasDAHScd(

3Id

adoTrDp

ntensities were expressed as mean ± SE (n = 4). * P < 0.05 vs. control. Scale bar 20 �m.For interpretation of the references to colour in this figure legend, the reader iseferred to the web version of this article.)

ncreased until 6 h (Fig. 3C), while phosphorylation of Akt wasignificantly increased until 3 h (Fig. 3D). We next examined theffects of H2O2 on phosphorylation of IR� and Akt in hypothalamicrganotypic cultures. Similar to insulin, dose response experimentshowed that p-IR� levels were significantly increased at 3 h by2O2 (from 10−4 to 10−3 M) (Fig. 3E), while p-Akt levels were

ignificantly increased by H2O2 (from 10−5 to 10−3 M) (Fig. 3F).ime course experiments showed that H2O2 (10−4 M) significantly

ncreased p-IR� and p-Akt levels at 1 h, and these phosphorylationsere significantly increased until 6 h (Figs. 3 G and 3H). No additive

ffects of insulin (10−7 M) and H2O2 (10−4 M) on p-IR� and p-Aktevels were observed in this study (Fig. 4A and B).

.3. Effects of ROS scavenger, NADPH oxidase inhibitor anditochondrial complex II inhibitor on insulin signal transduction

While the ROS scavenger NAC did not affect p-IR� and p-Aktevels by themselves, insulin-induced increases in p-IR� and p-Aktevels were significantly attenuated in the presence of NAC (Fig. 5And B). Insulin-induced increases in p-IR� and p-Akt levels wereignificantly attenuated in the presence of NADPH oxidase inhibitorPI (10−6 M) at the dose where DPI did not decrease p-IR� and p-kt levels by itself (Fig. 5C and D). DPI (10−6 M) did not suppress2O2 (10−4 M)-induced phosphorylations of IR� nor Akt (Supp. Fig.1 in the online version at DOI: 10.1016/j.neulet.2016.03.011). Inontrast, mitochondrial complex II specific inhibitor TTFA (10−4 M)id not suppress insulin-induced phosphorylation of IR� and AktFig. 5E and F).

.4. Effects of NADPH oxidase inhibitor or mitochondrial complexI inhibitor on insulin-induced intracellular ROS production ofispersed hypothalamic cells

To clarify the source of insulin-induced intracellular ROS, wedditionally examined the insulin-induced intracellular ROS pro-uction of dispersed hypothalamic cells in the presence of NADPHxidase inhibitor DPI or mitochondrial complex II specific inhibitor

TFA. Intracellular ROS production was assessed by observing fluo-escence of CellROX orange with confocal fluorescence microscopy.PI (10−6 M) significantly decreased insulin-induced ROSroduction, whereas TTFA (10−4 M) did not (Fig. 6). Neither

Letters 619 (2016) 1–7

DPI nor TTFA by itself affected fluorescence of CellROX (data notshown).

4. Discussion

This study demonstrated that administration of insulinincreased intracellular ROS production in mouse hypothalamicexplants, which was significantly attenuated in the presence ofNADPH oxidase inhibitor, but not with mitochondrial complex IIinhibitor. Administration of H2O2 by itself significantly increasedp-IR� and p-Akt levels. Insulin-induced increases in p-IR� and p-Akt levels were significantly attenuated in the presence of ROSscavenger or NADPH oxidase inhibitor, but not with mitochondrialcomplex II inhibitor.

It was indicated that the physiological effects of insulin weremediated via ROS production by intracerebroventricular injectionof insulin or NADPH oxidase inhibitor in vivo [13], but it wasnot clear whether insulin by itself increased ROS production orROS by itself enhanced insulin signaling within the hypothala-mus. The present study clearly demonstrated that administrationof insulin is sufficient for increasing intracellular ROS production,and that administration of H2O2 is sufficient for enhancing insulinsignaling in the hypothalamus. Furthermore, administration ofROS scavenger or NADPH oxidase inhibitor completely abolishedinsulin-induced phosphorylation of IR� and Akt. Our data thusindicate that insulin enhances ROS production, and ROS positivelyregulate insulin signaling at the level of IR or Akt phosphorylationin mouse hypothalamus.

To date, two distinct insulin-sensitive cellular ROS sources havebeen identified: NADPH oxidase in adipocytes [2,18–21] or vascu-lar smooth muscles cells [22,23]; and the mitochondrial respiratorychain in liver [24,25], heart [25] or cerebellar granule neurons [11].To clarify which source of ROS is predominant in insulin-inducedROS production in the hypothalamus, we used NADPH oxidaseinhibitor DPI or mitochondria complex II specific inhibitor TTFA.Our data showed that DPI completely abolished insulin-inducedROS production and insulin-induced phosphorylation of IR� or Akt.On the other hand, TTFA did not suppress insulin-induced ROSproduction nor insulin-induced phosphorylation of IR� or Akt atthe dose of 10−4 M. Although DPI has been indicated as possiblysuppressing mitochondria ROS production [26,27], our data implythat insulin predominantly produces ROS via NADPH oxidase, andthat the contribution of ROS produced by mitochondria is relativelysmall in insulin-induced ROS production in the hypothalamus.

In the liver, it was reported that ROS demonstratedconcentration-dependent dual effects on insulin signal trans-duction; high ROS concentration suppressed insulin signaltransduction whereas low ROS concentration enhanced it [28]. Inour study, high doses of H2O2 (10−3 M), under which insulin signaltransduction was reportedly suppressed in the liver [28], did notsuppress insulin-induced IR� nor Akt phosphorylation. Our dataindicate that in the hypothalamus, ROS enhance insulin signalingeven at relatively high ROS concentrations.

Specific molecular targets of ROS identified to date includethe protein tyrosine phosphatase 1B (PTP1B) [2,29,30], whichnegatively regulates IR and insulin receptor substrate (IRS), andthe lipid phosphatase PTEN [31–33], which dephosphorylatesphosphatidylinositol-3,4,5-triphosphate (PIP3) to terminate PI3 Ksignaling. ROS reportedly enhanced IR, IRS or PI3 K phosphorylationby suppressing these phosphatases in peripheral tissues [30,32,33].

Our data showed that ROS enhanced phosphorylation of IR�, thefirst step of PI3 K/Akt signaling. Suppression of PTP1B activity byROS might be one of the possible mechanisms which explain thisenhancement.

ience

bppbanco

5

lp

A

i0

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[41] S. Czernichow, A. Couthouis, S. Bertrais, A.C. Vergnaud, L. Dauchet, P. Galan, S.

T. Onoue et al. / Neurosc

Oxidative stress is thought to be increased in patients with dia-etes or obesity [34–36], and to contribute to the pathogenesis androgression of glucose metabolism disorder [37–39]. However, therotective effect of antioxidants for diabetes in human has not yeteen established [40,41]. Our data indicate that the existence of anppropriate quantity of ROS is necessary to maintain insulin sig-al transduction in the hypothalamus. Further study is needed tolarify the role of ROS in the pathological conditions of diabetes orbesity.

. Conclusion

Insulin-induced autophosphorylation of IR� and phosphory-ation of Akt are mediated via ROS which are predominantlyroduced by NADPH oxidase in mouse hypothalamus.

ppendix A. Supplementary data

Supplementary data associated with this article can be found,n the online version, at http://dx.doi.org/10.1016/j.neulet.2016.03.11.

eferences

[1] N. Houstis, E.D. Rosen, E.S. Lander, Reactive oxygen species have a causal rolein multiple forms of insulin resistance, Nature 440 (2006) 944–948.

[2] K. Mahadev, H. Motoshima, X. Wu, J.M. Ruddy, R.S. Arnold, G. Cheng, J.D.Lambeth, B.J. Goldstein, The NAD(P)H oxidase homolog Nox4 modulatesinsulin-stimulated generation of H2O2 and plays an integral role in insulinsignal transduction, Mol. Cell. Biol. 24 (2004) 1844–1854.

[3] B.J. Goldstein, K. Mahadev, X. Wu, Redox paradox: insulin action is facilitatedby insulin-stimulated reactive oxygen species with multiple potentialsignaling targets, Diabetes 54 (2005) 311–321.

[4] B.J. Goldstein, K. Mahadev, X. Wu, L. Zhu, H. Motoshima, Role ofinsulin-induced reactive oxygen species in the insulin signaling pathway,Antioxid. Redox Signal. 7 (2005) 1021–1031.

[5] A. Kleinridders, H.A. Ferris, W. Cai, C.R. Kahn, Insulin action in brain regulatessystemic metabolism and brain function, Diabetes 63 (2014) 2232–2243.

[6] S.C. Woods, E.C. Lotter, L.D. McKay, D. Porte Jr., Chronicintracerebroventricular infusion of insulin reduces food intake and bodyweight of baboons, Nature 282 (1979) 503–505.

[7] L.M. Brown, D.J. Clegg, S.C. Benoit, S.C. Woods, Intraventricular insulin andleptin reduce food intake and body weight in C57BL/6J mice, Physiol. Behav.89 (2006) 687–691.

[8] K. Rahmouni, D.A. Morgan, G.M. Morgan, X. Liu, C.D. Sigmund, A.L. Mark, W.G.Haynes, Hypothalamic PI3 K and MAPK differentially mediate regionalsympathetic activation to insulin, J. Clin. Invest. 114 (2004) 652–658.

[9] S. Obici, B.B. Zhang, G. Karkanias, L. Rossetti, Hypothalamic insulin signaling isrequired for inhibition of glucose production, Nat. Med. 8 (2002) 1376–1382.

10] L. Koch, F.T. Wunderlich, J. Seibler, A.C. Konner, B. Hampel, S. Irlenbusch, G.Brabant, C.R. Kahn, F. Schwenk, J.C. Bruning, Central insulin action regulatesperipheral glucose and fat metabolism in mice, J. Clin. Invest. 118 (2008)2132–2147.

11] N.A. Persiyantseva, T.P. Storozhevykh, Y.E. Senilova, L.R. Gorbacheva, V.G.Pinelis, I.A. Pomytkin, Mitochondrial H2O2 as an enable signal for triggeringautophosphorylation of insulin receptor in neurons, J. Mol. Signal 8 (2013) 11.

12] A. Benani, S. Troy, M.C. Carmona, X. Fioramonti, A. Lorsignol, C. Leloup, L.Casteilla, L. Penicaud, Role for mitochondrial reactive oxygen species in brainlipid sensing: redox regulation of food intake, Diabetes 56 (2007) 152–160.

13] T. Jaillard, M. Roger, A. Galinier, P. Guillou, A. Benani, C. Leloup, L. Casteilla, L.Penicaud, A. Lorsignol, Hypothalamic reactive oxygen species are required forinsulin-induced food intake inhibition: an NADPH oxidase-dependentmechanism, Diabetes 58 (2009) 1544–1549.

14] M. Goto, H. Arima, M. Watanabe, M. Hayashi, R. Banno, I. Sato, H. Nagasaki, Y.Oiso, Ghrelin increases neuropeptide Y and agouti-related peptide geneexpression in the arcuate nucleus in rat hypothalamic organotypic cultures,Endocrinology 147 (2006) 5102–5109.

15] H. Arima, S.B. House, H. Gainer, G. Aguilera, Neuronal activity is required forthe circadian rhythm of vasopressin gene transcription in thesuprachiasmatic nucleus in vitro, Endocrinology 143 (2002) 4165–4171.

16] Y. Ito, R. Banno, S. Hagimoto, Y. Ozawa, H. Arima, Y. Oiso, TNF� increaseshypothalamic PTP1B activity via the NF(B pathway in rat hypothalamicorganotypic cultures, Regul. Pept. 174 (2012) 58–64.

17] S.I. Yamagishi, D. Edelstein, X.L. Du, Y. Kaneda, M. Guzmán, M. Brownlee,Leptin induces mitochondrial superoxide production and monocytechemoattractant protein-1 expression in aortic endothelial cells by increasingfatty acid oxidation via protein kinase A, J. Biol. Chem. 276 (2001)25096–25100.

Letters 619 (2016) 1–7 7

18] J. Kreuzer, B. Nurnberg, H.I. Krieger-Brauer, Ligand-dependentautophosphorylation of the insulin receptor is positively regulated byGi-proteins, Biochem. J. 380 (2004) 831–836.

19] H.I. Krieger-Brauer, P.K. Medda, H. Kather, Insulin-induced activation ofNADPH-dependent H2O2 generation in human adipocyte plasma membranesis mediated by Galphai2, J. Biol. Chem. 272 (1997) 10135–10143.

20] H.I. Krieger-Brauer, H. Kather, Human fat cells possess a plasmamembrane-bound H2O2-generating system that is activated by insulin via amechanism bypassing the receptor kinase, J. Clin. Invest. 89 (1992)1006–1013.

21] S.P. Mukherjee, W.S. Lynn, Reduced nicotinamide adenine dinucleotidephosphate oxidase in adipocyte plasma membrane and its activation byinsulin. Possible role in the hormone’s effects on adenylate cyclase and thehexose monophosphate shunt, Arch. Biochem. Biophys. 184 (1977) 69–76.

22] M. Yang, Y. Yang, S. Zhang, A.M. Kahn, Insulin-stimulated hydrogen peroxideincreases guanylate cyclase activity in vascular smooth muscle, Hypertension42 (2003) 569–573.

23] M. Yang, E. Foster, A.M. Kahn, Insulin-stimulated NAD(P)H oxidase activityincreases migration of cultured vascular smooth muscle cells, Am. J.Hypertens. 18 (2005) 1329–1334.

24] I.A. Pomytkin, O.E. Kolesova, Key role of succinate dehydrogenase ininsulin-induced inactivation of protein tyrosine phosphatases, Bull. Exp. Biol.Med. 133 (2002) 568–570.

25] I.A. Pomytkin, O.E. Kolesova, Effect of insulin on the rate of hydrogen peroxidegeneration in mitochondria, Bull. Exp. Biol. Med. 135 (2003) 541–542.

26] Y. Li, M.A. Trush, Diphenyleneiodonium an NAD(P)H oxidase inhibitor, alsopotently inhibits mitochondrial reactive oxygen species production, Biochem.Biophys. Res. Commun. 253 (1998) 295–299.

27] E. Aldieri, C. Riganti, M. Polimeni, E. Gazzano, C. Lussiana, I. Campia, D. Ghigo,Classical inhibitors of NOX NAD(P)H oxidases are not specific, Curr. DrugMetab. 9 (2008) 686–696.

28] S. Iwakami, H. Misu, T. Takeda, M. Sugimori, S. Matsugo, S. Kaneko, T.Takamura, Concentration-dependent dual effects of hydrogen peroxide oninsulin signal transduction in H4IIEC hepatocytes, PLoS One 6 (2011) e27401.

29] T.C. Meng, D.A. Buckley, S. Galic, T. Tiganis, N.K. Tonks, Regulation of insulinsignaling through reversible oxidation of the protein-tyrosine phosphatasesTC45 and PTP1B, J. Biol. Chem. 279 (2004) 37716–37725.

30] K. Mahadev, A. Zilbering, L. Zhu, B.J. Goldstein, Insulin-stimulated hydrogenperoxide reversibly inhibits protein-tyrosine phosphatase 1b in vivo andenhances the early insulin action cascade, J. Biol. Chem. 276 (2001)21938–21942.

31] J. Kwon, S.R. Lee, K.S. Yang, Y. Ahn, Y.J. Kim, E.R. Stadtman, S.G. Rhee,Reversible oxidation and inactivation of the tumor suppressor PTEN in cellsstimulated with peptide growth factors, Proc. Natl. Acad. Sci. U. S. A. 101(2004) 16419–16424.

32] J.H. Seo, Y. Ahn, S.R. Lee, C. Yeol Yeo, K. Chung Hur, The major target of theendogenously generated reactive oxygen species in response to insulinstimulation is phosphatase and tensin homolog and not phosphoinositide-3kinase (PI-3 kinase) in the PI-3 kinase/Akt pathway, Mol. Biol. Cell 16 (2005)348–357.

33] K. Loh, H. Deng, A. Fukushima, X. Cai, B. Boivin, S. Galic, C. Bruce, B.J. Shields, B.Skiba, L.M. Ooms, N. Stepto, B. Wu, C.A. Mitchell, N.K. Tonks, M.J. Watt, M.A.Febbraio, P.J. Crack, S. Andrikopoulos, T. Tiganis, Reactive oxygen speciesenhance insulin sensitivity, Cell Metab. 10 (2009) 260–272.

34] G. Paolisso, A. D’Amore, C. Volpe, V. Balbi, F. Saccomanno, D. Galzerano, D.Giugliano, M. Varricchio, F. D’Onofrio, Evidence for a relationship betweenoxidative stress and insulin action in non-insulin-dependent (type II) diabeticpatients, Metabolism 43 (1994) 1426–1429.

35] J. Vessby, S. Basu, R. Mohsen, C. Berne, B. Vessby, Oxidative stress andantioxidant status in type 1 diabetes mellitus, J. Intern. Med. 251 (2002)69–76.

36] K. Ohmori, S. Ebihara, S. Kuriyama, T. Ugajin, M. Ogata, A. Hozawa, T. Matsui,Y. Tsubono, H. Arai, H. Sasaki, I. Tsuji, The relationship between body massindex and a plasma lipid peroxidation biomarker in an older, healthy Asiancommunity, Ann. Epidemiol. 15 (2005) 80–84.

37] H. Kaneto, G. Xu, K.H. Song, K. Suzuma, S. Bonner-Weir, A. Sharma, G.C. Weir,Activation of the hexosamine pathway leads to deterioration of pancreaticbeta-cell function through the induction of oxidative stress, J. Biol. Chem. 276(2001) 31099–31104.

38] H. Sakuraba, H. Mizukami, N. Yagihashi, R. Wada, C. Hanyu, S. Yagihashi,Reduced beta-cell mass and expression of oxidative stress-related DNAdamage in the islet of Japanese Type II diabetic patients, Diabetologia 45(2002) 85–96.

39] K. Sakai, K. Matsumoto, T. Nishikawa, M. Suefuji, K. Nakamaru, Y. Hirashima, J.Kawashima, T. Shirotani, K. Ichinose, M. Brownlee, E. Araki, Mitochondrialreactive oxygen species reduce insulin secretion by pancreatic beta-cells,Biochem. Biophys. Res. Commun. 300 (2003) 216–222.

40] S. Liu, I.M. Lee, Y. Song, M. Van Denburgh, N.R. Cook, J.E. Manson, J.E. Buring,Vitamin E and risk of type 2 diabetes in the women’s health studyrandomized controlled trial, Diabetes 55 (2006) 2856–2862.

Hercberg, Antioxidant supplementation does not affect fasting plasmaglucose in the Supplementation with Antioxidant Vitamins and Minerals(SU.VI.MAX) study in France: association with dietary intake and plasmaconcentrations, Am. J. Clin. Nutr. 84 (2006) 395–399.