Roux-en-Y gastric bypass improves hepatic mitochondrial function in obese rats

1d

Original article

Roux-en-Y gastric bypass improves hepatic mitochondrial function inobese rats

Yanhua Peng, Ph.D., Michel M. Murr, M.D., F.A.C.S.*James A. Haley Veterans Affairs Medical Center, Department of Surgery, University of South Florida, Tampa, Florida

Received May 11, 2011; accepted June 20, 2011

Abstract Background: Obesity-related fatty liver disease is linked to mitochondrial dysfunction and oxidativestress. Nuclear factor (erythroid-derived 2)-like 2 (Nrf2) regulates mitochondrial function and is atranscriptor of multiple genes that produce antioxidants. Because Roux-en-Y gastric bypass (RYGB)improves fatty liver and decreases the oxidative stress in the liver, we hypothesized that RYGB activatesNrf2 and increases cytochrome C oxidase subunit II (COX-II) in the liver of obese rats.Methods: Sprague-Dawley rats were fed a high-fat diet for 16 weeks. The obese rats underwent eitherRYGB (n � 20) or a sham operation (n � 20). The tissues were harvested 13 weeks postoperatively. Thenuclear fraction and mitochondrial extracts were used for protein analysis with immunoblotting. Immu-nostaining was done on liver sections for COX-II, Nrf2, and the macrophage marker ED2 and F4/80. Thegels were quantified using densitometry; P � .05 was considered significant.Results: RYGB increased COX-II expression in the liver sections (3330 � 56 versus 2056 � 37for RYGB versus sham, P � .001). The total (nuclear and cytoplasmic) Nrf2 expression was highin the obese sham-operated control (2456 � 45 versus 4352 � 76, RYGB versus sham, P � .001).However, the nuclear fraction of Nrf2 was significantly increased in the RYGB liver (2341 � 46versus 1352 � 35, RYGB versus sham, P � .001). Furthermore, Nrf2 protein co-localized with themolecular markers of Kupffer cells.Conclusions: Diet-induced fatty liver is associated with mitochondrial dysfunction. RYGB in-creases COX-II, which is involved in mitochondrial respiration, and increases the nuclear translo-cation of the Nrf2 transcriptional factor, which is involved in mitochondrial biogenesis and function.Taken together, these data suggest that surgically induced weight loss is associated with improvedmitochondrial function in obese rats. (Surg Obes Relat Dis 2013;9:429–435.) © 2013 Published byElsevier Inc. on behalf of American Society for Metabolic and Bariatric Surgery.

Keywords: Oxidative stress; Nuclear factor (erythroid-derived 2)-like 2; Nrf2; Mitochondria function; Obesity; Diabetes;

Surgery for Obesity and Related Diseases 9 (2013) 429–435

Roux-en-Y gastric bypass; RYGB

Masnirnta

i

Obesity is associated with chronic inflammation, oxida-tive stress, and insulin resistance, as well as nonalcoholicfatty liver disease (NAFLD). Oxidative stress is reported toincrease with the increased severity of steatohepatitis [1].

Supported by the American Society for Metabolic and Bariatric Sur-geons (to M. M. Murr).

Presented at the Plenary Session of the American Society for Metabolicand Bariatric Surgeons Annual Meeting, Orlando, Florida, June 2011.

*Correspondence: Michel M. Murr, M.D., F.A.C.S., University of SouthFlorida, C/O Tampa General Hospital, P.O. Box 1289, Tampa, FL 33601.

rE-mail: [email protected].

550-7289/13/$ – see front matter © 2013 Published by Elsevier Inc. on behalfoi:10.1016/j.soard.2011.06.012

oreover, patients with NAFLD have additional metabolicbnormalities that are associated with increased oxidativetress. A substantial number of morbidly obese patients withonalcoholic steatosis, steatohepatitis, or cirrhosis exhibitncreased concentrations of hydroperoxides, suggestive of aole for oxidative stress in the pathogenesis of NAFLD/onalcoholic steatohepatitis [2]. Similarly, a close associa-ion has been found between the degree of insulin resistancend markers of oxidative stress [3].

At 1 year after Roux-en-Y gastric bypass (RYGB), anmprovement in antioxidant protection, associated with a
eduction in inflammatory and oxidative markers, was ob-
of American Society for Metabolic and Bariatric Surgery.

mailto:[email protected]

hs

cuicsil

dpN

M

A

Ua

H

ll

dosRs

O

mm8wg

R

sflmi

mBlitap

430 Y. Peng and M. M. Murr / Surgery for Obesity and Related Diseases 9 (2013) 429–435

served, including a significant reduction in serum malondi-aldehyde, C-reactive protein, superoxide dismutase, gluta-thione disulfide, �1-acid glycoprotein, and leptin [4].

In our rat model of diet-induced obesity, RYGB reducedthe body weight and liver weight and improved the histo-logic aspects of NAFLD. Specifically, RYGB reduced thefat content in the liver by reducing cell ballooning [5,6].Additionally, obesity increased the oxidative stress and cy-tokine production in the livers of mice and rats [7,8] andRYGB reversed these changes and was associated withincreased 5=-adenosine monophosphate-activated proteinkinase (AMPK), sirtuin (silent mating type information reg-ulation 2 homolog) 1 (SIRT1) and a reduction in reactiveoxygen species (ROS) within the liver and Kupffer cells [9].

Mitochondria are essential for fat oxidation and can be amajor source of oxidative stress under conditions of sepsisor lipotoxicity [10]. Moreover, excess adiposity in the liveris associated with mitochondrial dysfunction.

Of particular interest to mitochondrial function is nuclearfactor (erythroid-derived 2)-like 2 (Nrf2). Low to modestROS production increases Nrf2 expression, and overpro-duction of ROS inhibits Nrf2 expression. Nrf2 has severalfunctions, including detoxification; regulation of transcrip-tional activation of antioxidant genes and electrophiles, andregulation of mitochondrial biogenesis and lipid oxidation[11]. More recently, Nrf2 was implicated in the obesityformation of mice by regulation of the peroxisome prolif-erator-activated receptor-� [12]. Because Nrf2 regulatesigh-fat diet-induced steatosis and adiposity, it has beenuggested as a molecular target to prevent obesity [13].

Although Nrf2 is involved in adipogenesis and mito-hondrial biogenesis [11], the cytochrome C oxidase sub-nit II (COX-II) is a marker of mitochondrial function ands the rate-limiting step in oxidative phosphorylation. Mito-hondrial COX-II expression can be an indicator of up-tream genes that regulate mitochondrial function; moremportantly, the mammalian COX-II complex can be regu-ated by Nrf2 [14].

Because RYGB improves fatty liver and decreases oxi-ative stress in the liver, we hypothesized that RYGB im-roves glucose tolerance, reduces oxidative stress, activatesrf2, and increases COX-II levels in the liver of obese rats.

ethods

nimals and animal care

The institutional animal care and use committee of theniversity of South Florida College of Medicine approved

ll experiments.

igh-fat diet

The 4-week-old Sprague-Dawley rats were maintained inight- and temperature-controlled environments (12 hours
ight/12 hours dark, 20°-24°C). The rats were fed a high-fat i
iet containing 60% fat for 13 weeks. Subsequently, thebese rats underwent RYGB or a sham operation, as de-cribed previously [9]. The rats were killed 13 weeks afterYGB or the sham procedure, and the tissues and blood

amples were harvested.

ral glucose tolerance test

After an overnight fast, the serum glucose levels wereeasured using a commercially available clinical glucoseeter at 0, 30, 60, 90, and 120 minutes (n � 8 sham, n �RYGB) after oral glucose was administered (2 mg/g bodyeight). We calculated the mean glucose levels for eachroup and plotted the glucose tolerance accordingly.

OS levels

The ROS levels were determined using the oxidationensitive fluorogenic precursor dihydrodichlorocarboxyuorescein diacetate. In brief, each well of a 96-wellicrotiter plate was filled with respiration buffer contain-

ng 1 �M dihydrodichlorocarboxy fluorescein diacetateand 0.5 mg/mL of mitochondrial particles and 5 mMglutamate. Fluorescence was measured with a fluores-cence plate reader at an excitation of 485 nm and anemission of 530 nm.

Immunoblotting

The cells were lysed in radioimmunoprecipitation as-say buffer (phosphate-buffered saline [PBS] with .1%sodium dodecyl sulfate [SDS], 1% NP-40, and .5% so-dium deoxcholate); 50 –100-�g samples of protein werefractionated by 10% SDS-polyacrylamide gel electropho-resis, transferred to nitrocellulose membrane, blocked for1 hour with PBS (5% instant nonfat dry milk, .1% Tween20), and incubated for 2 hours with antibodies (.05 �g/

L) to Nrf2 or COX-II (Cell Signaling Technology,everly, MA) and �-actin for the whole cell protein

oading control, histone H1 for the nuclear extract load-ng control, or the cytochrome C oxidase subunit V forhe mitochondrial protein loading control. Bound primaryntibody was detected by incubating with horseradisheroxide goat anti-mouse or anti-rabbit IgG (.0125 �g/

mL). The membranes were developed using Super Signal(Pierce, Rockford, IL) ECL reagent, and quantified bydensitometry.

Co-immunoprecipitation and western blotting

Co-immunoprecipitation and Western blotting was usedto determine whether any change occurred in the physicalassociation between insulin receptor substrate 2 and phos-phoinositide kinase-3 (PI3K). In brief, liver tissue was lysed(50 mm Tris-HCl, pH 8.0, 5 mm ethylene diamante tet-racetic acid, 150 mM NaCl, .5% Nonidet P-40, 1 mmphenylmethylsulfonyl fluoride); 1000 �g of protein was
mmunoprecipitated with PI3K antibody and protein A sep-

wm

R

G

wl1ttr1

431RYGB improves hepatic function in obese rats / Surgery for Obesity and Related Diseases 9 (2013) 429–435

harose beads. The beads were then washed with lysis bufferto remove nonspecific binding; the immunoprecipitate wasfractionated by 10% SDS-polyacrylamide gel electrophore-sis, and transferred to Hybond ECL nitrocellulose mem-brane. Subsequently, the “pulled-down” PI3K proteins wereimmunoblotted with an insulin receptor substrate 2 antibodyto detect the insulin receptor substrate-PI3K complex (CellSignaling Technology). Representative gels were quantifiedusing densitometry.

Nuclear and cytoplasmic level of Nrf2

The nuclear and cytoplasmic levels of Nrf2 were pre-pared using an active motif kit as described in our previouspublications [7]. Nuclear and whole cell Nrf2 were deter-mined by immunoblotting; Histone H1 and actin were usedas a nuclear extract and whole cell loading controls, respec-tively.

Immunofluorescent staining for protein co-localization

In brief, formalin-fixed liver sections were deparaffinizedand hydrated with xylene, ethanol, PBS, and treated with.1–.2% trypsin in .4% CaCl2 for 1 hour. The slides werethen incubated with antibody for Nrf2, ED2, or F4/80 (1:100 in PBS plus 10% normal goat serum) for 2–4 hours.The slides were washed with PBS plus .1% Triton X-100,incubated with fluorescent isothiocyanate goat anti-mouseor anti-rabbit IgG, and mounted with antifade solution con-taining 4,6=-diamidino-2-phenylindole, a blue fluorescentdye for staining nuclei (double-strand DNA). The slideswere examined using a Nikon microscope; the images weremerged using Image-Pro Express software (Image Process-ing Solutions, North Reading, MA).

Mitochondrial COX-II

Mitochondria were prepared as described previously[15,16]. In brief, the mitochondria were purified by differ-ential centrifugation. The liver samples (200 mg) werechilled on ice cold .85% NaCl in a small dish, cut into smallpieces, and homogenized in a buffer (.25 m sucrose, 5 mMHEPE, 1 mM ethylenediaminetetraacetic acid), then fol-lowed by repeated centrifuging and washing steps. Themitochondrial layers were collected and stored at �190°Cin preservation buffer. COX-II expression from mitochon-drial samples was measured using SDS-polyacrylamide gelelectrophoresis separation and immunoblotting.

Statistical analysis

All experiments were repeated at least in triplicate. Thet test was used to compare the mean values, and P � .05

as considered significant. The data are presented as the
ean � standard deviation.
esults

lucose tolerance

RYGB decreased the serum glucose levels comparedith those in the obese sham controls. The serum glucose

evel at 0, 30, 60, 90, and 120 minutes was 75, 165, 135,25, 118 mg/dL and 86, 196, 181, 157, and 142 mg/dL inhe RYGB and sham groups, respectively. The area underhe curve of plasma glucose was decreased in the RYGBats compared with the sham obese control rats (11.6 �03 � .3 � 103 versus 17.9 � 103 � .5 � 103 mg/dL/

min; P � .001, for RYGB versus sham control; Fig. 1).The difference in serum glucose levels between theRYGB and sham control groups was apparent within 30minutes and was sustained throughout the duration of the120-minute test.

ROS production

RYGB decreased ROS production in the livers of theobese rats (68% � 4% versus 267% � 5%, P � .001,RYGB versus sham obese control; Fig. 2).

Nrf2 expression

RYGB decreased whole cell expression of Nrf2 com-pared with the sham obese controls (1414 � 66 versus 2987 �89; P � .001, RYGB versus sham obese control; Fig. 3).However, the nuclear level of Nrf2 increased in the RYGB ratlivers compared with that in the sham obese controls (immu-noblotting 2788 � 67 versus 1876 � 56; all P � .001, RYGBversus sham obese control; Fig. 3).

We confirmed the nuclear shift of Nrf2 by immuno-

Fig. 1. RYGB attenuates obesity-induced glucose intolerance (0-120 min)in obese rats compared with sham obese control (all P � .001).

fluorescent staining (2110 � 67 versus 3056 � 87; all

Lfaiai

C

mtPc(sefhcK

N

tflps

tf


P � .001, RYGB versus sham obese control; Fig. 3,ower). Blue staining indicates the nuclear DNA staining

rom well-arranged parenchymal cells; Nrf2 stains greennd is localized in nonparenchymal cells. The intensity ofmmunostaining for Nrf2 was not changed; however, thereas stained were smaller after RYGB, consistent withncreased nuclear translocation of Nrf2.

Fig. 2. RYGB attenuates obesity-induced oxidative stress. Bar graph showsRYGB decreased obesity-increased ROS production in liver comparedwith sham obese control (*P � .001).

Fig. 3. Representative blots showing that RYGB increased nuclear transshowing immunofluorenscent staining for Nrf2 in green is decreased in
phenylindole (DAPI) in blue indicates nuclear staining.
OX-II expression

RYGB increased the expression of the mitochondrialarker COX-II compared with that in the sham obese con-

rol rats (immunoblotting: 3631 � 91 versus 1342 � 14; all� .001, RYGB versus sham obese control; Fig. 4). We

onfirmed these results with immunofluorescent staining3662 � 78 versus 1611 � 51; all P � .001, RYGB versusham obese control; Fig. 4). Red staining indicates COX-IIxpression; blue indicates nuclear DNA staining, mostlyound in large well-arranged (lined) cells, indicative ofepatocytes. The cells with red color are nonparenchymalells that might indicated stellale cells, immune cells andupffer cells.

rf2 expression in kupffer cells

To further characterize the type of nonparenchymal cellshat stain for Nrf2 and COX-II, we used double immuno-uorescent staining. Nrf2 co-localized with cells that stainositive with the Kupffer cell marker ED2 [17] in the liverections of rats that underwent RYGB (Fig. 5, Upper).

Similarly, Nrf2 co-localized in the cells that stain posi-ive with the macrophage cell marker F4/80 in liver sectionsrom rats that underwent RYGB (Fig. 5, Lower). These data

of Nrf2 in liver compared with sham obese control. Photomicrographrom RYGB rats compared with sham obese controls; 4,6=-diamidino-2-

locationcells f

damplbvwi

g

wtppgt

Dc


suggest that Nrf2 is localized within Kupffer cells in thisparticular surgical weight loss model.

Discussion

The liver is central to glucose and lipid homeostasis.Emerging concepts indicate that insulin resistance ismanifest in the liver because of diet-induced adiposity.Once adiposity is established in the liver, parenchymalinjury ensues and is subsequently worsened by Kupffercell-derived lytokines and oxidative stress [18]. There isample evidence to implicate Kupffer cells in the progres-sion of steatosis to steatohepatitis because of their role ininnate immunity and their proinflammatory response tofatty acids [7].

RYGB improves fatty liver and insulin resistance inanimal models and humans [19]. Recent studies haveemonstrated that lipid oxidation, cytokine production,nd oxidative stress play important roles in the develop-ent of obesity-induced type 2 diabetes [20]. More im-

ortantly, how RYGB attenuate obesity-induced fattyiver and insulin resistance is not clear. Our focus haseen to investigate mechanistic pathways that are in-olved in improving insulin sensitivity after RYGB, asell as the mechanistic pathways implicated in obesity-

nduced liver injury.In our rat model of obesity, a high-fat diet induced

Fig. 4. Representative blots showing that RYGB increases mitochondrimmunofluorescent staining showing that RYGB increases COX-II expre

NA staining, these well-arranged cell nuclear can come from parencells.

lucose intolerance and downregulated insulin signaling in

the liver [21]. Therefore, we hypothesized that RYGB im-proves insulin resistance and decreases oxidative stress inthe liver.

Nrf2 is a transcription factor that plays an important rolein mediating antioxidative cellular functions [22]. Nrf2forms heterodimers with small macrophage-activating fac-tor proteins that upregulate genes that encode antioxidantenzymes and detoxifying proteins. Although Nrf2 is impor-tant for growth and development in the mouse, Nrf2-defi-cient mice are sensitized to oxidative stress-related diseasesin various organs. Nrf2 is highly expressed in the liver, lung,kidney, and intestine [13,23,24].

Activation of the proinflammatory transcription factorsnuclear factor-�B and activating protein-1 in obese patients

ith nonalcoholic steatohepatitis is associated with oxida-ive stress and insulin resistance [25]. Because Nrf2-driveneroxisome proliferator-activated receptor-� inductionlays a protective role in lung oxidant injury [26], investi-ators have suggested that the activation of nuclear fac-or-�B and peroxisome proliferator-activated receptor-�regulators as the principal hepatic adaptive mechanism tofat-rich diets [12,27].

Moreover, the downstream components of AMPK sig-naling, which include antioxidative stress genes, hemeoxygenase 1, and nuclear respiratory factor 1 are bothregulated by Nrf2, and peroxisome proliferator-activatedreceptor-� co-activator-1� regulates Nrf2 expression/ac-

ker COX-II compared with sham obese controls. Photomicrograph ofompared with sham obese control (all P � .001). Blue indicates nuclearcells. Red (arrow) indicates COX-II expression from nonparenchymal

ial marssion chymal

tivation as well [28]. The transcriptional factor peroxi-

ttBbpiosf

RRpegow

htr

pwp

mpiostvr

tfNlcecu

tpOEafl

d organ


some proliferator-activated receptor-� co-activator-1�regulates mitochondrial biogenesis by Nrf2 [28]. Hemeoxygenase-1 regulates mitochondrial biogenesis by wayof Nrf2-mediated transcriptional control of nuclear respi-ratory factor 1 [29]. Recent proteomic studies have sug-gested that lipid metabolic regulation is Nrf2 dependentin the liver [30].

Nrf2 is involved in antioxidation and lipid deposition inhe liver and therefore is important for mitochondrial func-ion. Another marker of mitochondrial function is COX-II.oth Nrf2 and COX-II exhibit antioxidative and antidia-etic properties. COX-II is an oligomeric enzymatic com-lex that is a component of the respiratory chain and isnvolved in the transfer of electrons from cytochrome C toxygen. Therefore, the study of Nrf2, COX-II, and ROSheds light on the effect of lipotoxicity on mitochondrialunction.

RYGB improved the glucose tolerance and decreasedOS production in the liver. Although the incretin effect ofYGB is established, little is known about the mechanisticathways that are involved. Our model reflects the chronicffects of weight loss; we chose the 13-week point to studylucose tolerance, because it mimics the long-term benefitsf RYGB. These data confirm our hypothesis that surgicaleight loss improves glucose homeostasis.Previously, we have shown that RYGB improves the

istologic features of NAFLD and decreases the fat con-ent in the liver of obese rats compared with sham obese

Fig. 5. Nrf2 expression and both macrophage markers ED2 and F4/80 exp(red indicates Nrf2 expression; green indicates ED2 expression; and organcells (red indicates Nrf2 expression; green indicates F4/80 expression; an

ats [31,32]. Our findings have shown that RYGB im- p

roves glucose tolerance and reduces oxidative stresshen weight loss has stabilized, consistent with previousublications [21].

At a cellular level, ROS originates from either theitochondria or cell membrane. We have previously re-

orted that expression of AMPK and SIRT1 are increasedn the rat liver after RYGB compared with the shambese control rats [9]. Because Nrf2 is the downstreamignaling molecule of SIRT1/peroxisome proliferator-ac-ivated receptor-� co-activator-1�, we suggest that acti-ation of the AMPK/SIRT1 pathway might positivelyegulate the activation of Nrf2.

The expression of Nrf2 is increased in obese sham con-rol rats, which might reflect the high requirement of Nrf2 inatty livers that promotes expression of Nrf2. Nonetheless,rf2 activation (represented by the nuclear amount of Nrf2

evel) is increased in RYGB rats compared with sham obeseontrols. These data demonstrate that activation and notxpression of Nrf2 is a better reflection of the antioxidativeapability because RYGB decreased ROS production andpregulated COX-II.

Previously, we reported that ROS can also originate fromhe membrane protein nicotinamide adenine dinucleotidehosphate-oxidase subunit 2 production in Kupffer cells [7].ur current data have shown that Nrf2 co-localizes withD2 and F4/80, thereby suggesting that Kupffer cells playvital role in oxidative stress, similar to their role in in-

ammatory stress [33]. These data are consistent with our

are co-localized. (Upper) Nrf2 and ED2 co-localization in Kupffer cellss both co-localization). (Lower) Nrf2 and F4/80 co-localization in Kupfferindicates both co-localization).

ressionindicate

revious findings and with other reports implicating Kupffer

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[


cells in the pathogenesis of NAFLD/nonalcoholic steato-hepatitis.

Conclusion

RYGB improves glucose tolerance and reduces oxidativestress compared with that in sham obese rats. The activity ofantioxidative transcriptional factor Nrf2 and the expressionlevel of mitochondrial marker COX-II are increased afterRYGB. Additional studies to understand how Nrf2 affectsthe local and systemic lipid metabolism and insulin sensi-tivity are warranted.

Disclosures

The authors have no commercial associations that mightbe a conflict of interest in relation to this article.

References

[1] Nobili V, Parola M, Alisi A, et al. Oxidative stress parameters inpaediatric non-alcoholic fatty liver disease. Int J Mol Med 2010;26:471–6.

[2] Allard JP, Aghdassi E, Mohammed S, et al. Nutritional assessmentand hepatic fatty acid composition in non-alcoholic fatty liver disease(NAFLD): a cross-sectional study. J Hepatol 2008;48:300–7.

[3] Tinahones F, Murri-Pierri M, Garrido-Sánchez L, et al. Oxidativestress in severely obese persons is greater in those with insulinresistance. Obes Silver Spring 2009;172:240–6.

[4] João CE, Valezi AC, Delfino VD, Lavado EL, Barbosa DS. Reduc-tion in plasma levels of inflammatory and oxidative stress indicatorsafter Roux-en-Y gastric bypass. Obes Surg 2010;20:42–9.

[5] Furuya CK, de Oliveira CP, de Mello ES, et al. Effects of bariatricsurgery on nonalcoholic fatty liver disease: preliminary findings after2 years. J Gastroenterol Hepatol 2007;22:510–4.

[6] Liu X, Lazenby AJ, Clements RH, Jhala N, Abrams GA. Resolutionof nonalcoholic steatohepatits after gastric bypass surgery. Obes Surg2007;17:486–92.

[7] Peng Y, Rideout D, Rakita S, et al. Roux-en-Y gastric bypass reducesoxidative stress and down regulates pro-inflammatory genes in liversof obese rats and in Kupffer cells via an AMPK-dependent pathway.J Surg Res 2009;151:244–5.

[8] Peng Y, Rideout D, Rakita S, Li J, You M, Mur M. Diet-inducedobesity is associated with steatosis, oxidative stress and inflammationin the liver. Surg Obes Relat Dis in press.

[9] Peng Y, Rideout DA, Rakita SS, Gower WJ, You M, Murr MM. DoesLKB1 mediate activation of hepatic AMP-protein kinase (AMPK)and sirtuin1 (SIRT1) after Roux-en-Y gastric bypass in obese rats? JGastrointest Surg 2010;14:221–8.

10] Murrhy MP. How mitochondria produce reactive oxygen species.Biochem J 2009;417:1–13.

11] Pejznochova M, Tesarova M, Hansikova H, et al. MitochondrialDNA content and expression of genes involved in mtDNA transcrip-tion, regulation and maintenance during human fetal development.Mitochondrion 2010;10:321–9.

12] Pi J, Leung L, Xue P, et al. Deficiency in the nuclear factor E2-relatedfactor-2 transcription factor results in impaired adipogenesis andprotects against diet-induced obesity. Chem Biol 2010;17:537–47.

13] Shin S, Wakabayashi J, Yates MS, et al. Role of Nrf2 in preventionof high-fat diet-induced obesity by synthetic triterpenoid CDDO-
imidazolide. Eur J Pharmacol 2009;620:138–44.
14] Lenka N, Vijayasarathy C, Mullick J, Avadhani NG. Structural or-ganization and transcription regulation of nuclear genes encoding themammalian cytochrome c oxidase complex. Prog Nucleic Acid ResMol Biol 1998;61:309–44.

15] Spremulli LL. Large-scale isolation of mitochondrial ribosomes frommammalian tissues. Methods Mol Biol 2007;372:265–75.

16] Ubuka T, Okada A, Nakamura HTU, Okada A, Nakamura H. Pro-duction of hypotaurine from L-cysteine sulfinate by rat liver mito-chondria. Amino Acids 2008;35:53–8.

17] Peng Y, Murr M. Establishment of immortalized rat Kupffer celllines. Cytokines 2007;37:185–91.

18] Jaeschke H. Reactive oxygen and mechanisms of inflammatory liverinjury: present concepts. J Gastroenterol Hepatol 2011;1:173–9.

19] Thaler P, Cummings D. Minireview: hormonal and metabolic mech-anisms of diabetes remission after gastrointestinal surgery. J ClinEndocrinol Metab 2009;94:2208–9.

20] Dandona P, Ghanim H, Chaudhuri A, Dhindsa S, Kim SS. Macro-nutrient intake induces oxidative and inflammatory stress: potentialrelevance to atherosclerosis and insulin resistance. Exp Mol Med2010;42:245–53.

21] Peng Y, Rideout D, Rakita S, et al. Roux-en-Y gastric bypass reducesoxidative stress and down regulates pro-inflammatory genes in liversof obese rats and in Kupffer cells via an AMPK-dependent pathway.J Surg Res 2009;151:244–5.

22] Pi J, Zhang Q, Fu J, et al. ROS signaling, oxidative stress and Nrf2in pancreatic beta-cell function. Toxicol Appl Pharmacol 2010;244:77–83.

23] Zhang DD. The Nrf2-Keap1-ARE signaling pathway: the regulationand dual function of Nrf2 in cancer. Antioxid Redox Signal 2010;13:1623–6.

24] Koenitzer JR, Freeman BA. Redox signaling in inflammation: inter-actions of endogenous electrophiles and mitochondria in cardiovas-cular disease. Ann N Y Acad Sci 2010;1203:45–52.

25] Esposito K, Giugliano D. Diet and inflammation: a link to metabolicand cardiovascular diseases. Eur Heart J 2006;27:15–20.

26] Cho HY, Gladwell W, Wang X, et al. Nrf2-regulated PPAR{gamma}expression is critical to protection against acute lung injury in mice.Am J Respir Crit Care Med 2010;182:170–82.

27] Li W, Khor TO, Xu C, et al. Activation of Nrf2-antioxidant signalingattenuates NFkappaB-inflammatory response and elicits apoptosis.Biochem Pharmacol 2008;76:1485–9.

28] Ventura-Clapier R, Garnier A, Veksler V. Transcriptional control ofmitochondrial biogenesis: the central role of PGC-1a. Cardiovasc Res2008;79:208–17.

29] Piantadosi CA, Carraway MS, Babiker A, Suliman HB. Heme oxy-genase-1 regulates cardiac mitochondrial biogenesis via Nrf2-medi-ated transcriptional control of nuclear respiratory factor-1. Circ Res2008;103:1232–40.

30] Kitteringham NR, Abdullah A, Walsh J, et al. Proteomic analysis ofNrf2 deficient transgenic mice reveals cellular defense and lipidmetabolism as primary Nrf2-dependent pathways in the liver. Pro-teomics 2010;73:1612–31.

31] Rideout D, Peng Y, Rakita S, Gower W, You M, Murr M. Roux-en-Ygastric bypass improves obesity-related steatosis and does not in-crease serum adiponectin. J Am Coll Surg 2008;207(Suppl 3):53.

32] Rideout D, Peng Y, Rakita S, Gower W, You M, Murr M. Roux-en-Ygastric bypass downregulates hepatic stearoyl-CoA desaturase 1 in-dependently of serum leptin in obese rats. J Surg Res 2009;151:244–245.

33] Peng Y, Sigua CA, Murr MM. Protein kinase C-zeta mediates apo-ptosis of mouse Kupffer cells via ERK-1/2: a novel mechanism.
Surgery 2011;149:135–42.

Documents

Roux-en-Y gastric bypass improves hepatic mitochondrial function in obese rats