Murine Macrophages Mitochondrial Transcription Factor a in Induction

TLR4-Mediated AKT Activation Is MYD88/TRIF-Dependent and Critical for Induction of Oxidative Phosphorylation and Mitochondrial Transcription Factor A in This information is current as Murine Macrophages of September 27, 2021. Christian P. Bauerfeld, Ruchi Rastogi, Gaila Pirockinaite, Icksoo Lee, Maik Hüttemann, Bobby Monks, Morris J. Birnbaum, Luigi Franchi, Gabriel Nuñez and Lobelia

Samavati Downloaded from J Immunol published online 6 February 2012 http://www.jimmunol.org/content/early/2012/02/06/jimmun ol.1102157 http://www.jimmunol.org/

Supplementary http://www.jimmunol.org/content/suppl/2012/02/06/jimmunol.110215 Material 7.DC1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 27, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2012 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published February 6, 2012, doi:10.4049/jimmunol.1102157 The Journal of Immunology

TLR4-Mediated AKT Activation Is MYD88/TRIF-Dependent and Critical for Induction of Oxidative Phosphorylation and Mitochondrial Transcription Factor A in Murine Macrophages

Christian P. Bauerfeld,*,1 Ruchi Rastogi,† Gaila Pirockinaite,† Icksoo Lee,‡ Maik Hu¨ttemann,‡ Bobby Monks,x Morris J. Birnbaum,x Luigi Franchi,{ Gabriel Nun˜ez,{ and Lobelia Samavati†,‡

Mitochondria play a critical role in cell survival and death. Mitochondrial recovery during inﬂammatory processes such as sepsis is

associated with cell survival. Recovery of cellular respiration, mitochondrial biogenesis, and function requires coordinated expres- Downloaded from sion of transcription factors encoded by nuclear and mitochondrial genes, including mitochondrial transcription factor A (T-fam) and cytochrome c oxidase (COX, complex IV). LPS elicits strong host defenses in mammals with pronounced inﬂammatory responses, but also triggers activation of survival pathways such as AKT pathway. AKT/PKB is a serine/threonine protein kinase that plays an important role in cell survival, protein synthesis, and controlled inﬂammation in response to TLRs. Hence we investigated the role of LPS-mediated AKT activation in mitochondrial bioenergetics and function in cultured murine macro-

phages (B6-MCL) and bone marrow-derived macrophages. We show that LPS challenge led to increased expression of T-fam and http://www.jimmunol.org/ COX subunits I and IV in a time-dependent manner through early phosphorylation of the PI3K/AKT pathway. PI3K/AKT pathway inhibitors abrogated LPS-mediated T-fam and COX induction. Lack of induction was associated with decreased ATP production, increased proinﬂammatory cytokines (TNF-a), NO production, and cell death. The TLR4-mediated AKT activation and mitochondrial biogenesis required activation of adaptor protein MYD88 and Toll/IL-1R domain-containing adaptor-inducing IFN-b. Importantly, using a genetic approach, we show that the AKT1 isoform is pivotal in regulating mitochondrial biogenesis in response to TLR4 agonist. The Journal of Immunology, 2012, 188: 000–000.

ammalian immune cells evolved to recognize patho- cascade of cellular signaling leading to inflammatory gene ex- by guest on September 27, 2021 gens through pattern recognition receptors, such as pression and clearance of invading organisms. LPS, a major M TLRs that recognize a wide range of microbial path- bacteria-derived cell wall component recognized by TLR4, elicits ogens. Recognition of microbial components by TLRs triggers a strong host defenses in mammals with pronounced inflammatory responses (1). Such responses are characterized by activation of inflammatory cascades with release of soluble cytokines and *Department of Pediatrics, St. John Hospital and Medical Center, Detroit, MI 48236; †Division of Pulmonary, Critical Care, and Sleep Medicine, Department of Medicine, chemokines such as TNF-a, IL-6, and generation of reactive ox- Wayne State University School of Medicine, Detroit, MI 48201; ‡Center for Molec- ygen species (ROS) and NO that subsequently facilitate the ular Medicine and Genetics, Wayne State University School of Medicine, Detroit, MI 48201; xDepartment of Medicine, University of Pennsylvania School of Medicine, eradication of pathogens. Although inflammation is necessary to Philadelphia, PA 19104; and {Department of Pathology, University of Michigan eradicate pathogens, excessive inflammation may lead to host Medical School, Ann Arbor, MI 48109 tissue injury and cell death. This might be, at least in part, because 1Current Address: Division of Critical Care, Department of Pediatrics, University of of the effect of inflammation on mitochondrial bioenergetics and Texas Health Science Center, San Antonio, TX mitochondrial injury. Received for publication July 26, 2011. Accepted for publication January 3, 2012. Mitochondria provide the majority of cellular energy in the form This work was supported by the Department of Medicine, Wayne State University of ATP and are the major source of ROS (2). Cellular immune School of Medicine (L.S.); the Center for Molecular Medicine and Genetics, Wayne State University School of Medicine (L.S., M.H.); and National Institutes of Health response to pathogens requires high energy in the form of ATP, Grants R01 DK61707 (to G.N.) and R01 DK56886 (to M.J.B.). and it appears that TLR-mediated mitochondrial ROS (mROS) are Address correspondence and reprint requests to Dr. Lobelia Samavati, Division of important for macrophage bactericidal activity (3). However, ex- Pulmonary, Critical Care, and Sleep Medicine, Department of Medicine, Wayne State cessive reactive oxygen and nitrogen species generation during University School of Medicine, 3 Hudson, 3990 John R Street, Detroit, MI 48201. E-mail address: [email protected] inflammation may damage mitochondria, causing bioenergetic The online version of this article contains supplemental material. failure. This process is thought to be due to inhibition of the electron transfer chain (ETC) (4) through secondary modification Abbreviations used in this article: BMDM, bone marrow-derived macrophage; COX, cytochrome c oxidase; ETC, electron transfer chain; FCCP, carbonyl cyanide 4-(tri- of enzymes involved in oxidative phosphorylation (OxPhos) and fluoromethoxy) phenylhydrazone; MnSOD, manganese superoxide dismutase; uncoupling of electron transport from ATP synthesis. Using an mROS, mitochondrial reactive oxygen species; NRF-1, nuclear respiratory factor 1; OxPhos, oxidative phosphorylation; PGC-1a, peroxisome proliferator-activated inflammation model, we have shown that TNF-a inhibits the ac- receptor g coactivator-1-a; ROS, reactive oxygen species; T-fam, mitochondrial tivity of cytochrome c oxidase (COX) in bovine liver and in transcription factor A; TMRM, tetramethylrhodamine-methylester; TRIF, Toll/IL- murine hepatocytes (5). This inhibition was associated with de- 1R domain-containing adapter-inducing IFN-b; WT, wild-type. creased ATP production and defective electron transport to mo- Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00 lecular oxygen. Several reports suggest regeneration of mitochon-

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1102157 2 LPS AND MITOCHONDRIAL BIOENERGETICS dria, upregulation of ETC capacity, and generation of new mito- (Ipswich, MA). Anti-mouse IgG and anti-rabbit IgG HRP-linked Abs were chondria later in the disease process after acute insults such as purchased from Cell Signaling Technology (Beverly, CA), and anti-goat hypoxemia or sepsis are important for survival (6–8). The exact HRP was purchased from Bio-Rad (Hercules, CA). mechanism underlying the recovery of bioenergetic failure after Cell culture inflammatory insult is not well studied. 2 2 The mouse macrophage cell line (B6-MCL) and MYD88/TRIF / cell line LPS stimulation under specific conditions leads to activation of were generous gifts of Dr. Eicke Latz (University of Massachusetts the prosurvival cascade followed by growth and differentiation. For Medical School, Worcester, MA). Both cell lines were generated from example, it has been shown that LPS may serve as a signal to wild-type (WT) C57BL6 mice as previously described (19), and MYD88/ 2/2 promote preconditioning and to protect against reperfusion in- TRIF macrophages were generated from the same background mice as previously described (20). B6-MCL and cells deficient in MYD88/TRIF jury (9). This bifunctionality of LPS is neither well understood were grown in IMDM supplemented with 10% heat-inactivated FCS, nor studied. Among numerous pathways activated through TLR- antibiotics, sodium pyruvate, nonessential amino acids, and 2-ME. mediated signaling, the PI3K/AKT pathway plays a central role in RAW 264.7 cells were obtained from American Type Culture Collection the regulation of cell survival and proliferation (9, 10). Evidence (Manassas, VA). Cells were maintained in a 95% air, 5% CO2 humidified suggests an important negative regulatory role for the PI3K/AKT atmosphere at 37˚C in RPMI medium supplemented with L-glutamine, penicillin-streptomycin, and 10% FCS (Invitrogen, Carlsbad, CA). pathway in innate immune responses to TLR agonists (11). Al- though growth factor-mediated PI3K activation and the subse- Mice and macrophages quent AKT activation is well characterized, TLR-mediated AKT MYD882/2 and TRIF2/2 mice were generated as previously described activation and its role in inflammation and mitochondrial recovery (21). AKT12/2 and AKT22/2 mice were generated as previously de- in response to TLR4 agonists is not well defined. It was shown that scribed (22, 23). All mice were backcrossed onto the C57BL/6 background Downloaded from certain TLRs, such as TLR4, require the adaptor protein MYD88 at least 10 times. WT C57BL/6 mice were maintained at animal facilities and Toll/IL-1R domain-containing adapter-inducing IFN-b (TRIF) at the University of Michigan (Ann Arbor, MI) and University of Penn- sylvania (Philadelphia, PA). Animal studies were approved by the Uni- to interact and activate the PI3K/AKT pathway (12, 13). The versity of Michigan and University of Pennsylvania Committee on Use and PI3K/AKT pathway plays both proinflammatory and anti-inflam- Care of Animals. matory roles in TLR signaling (11, 14). The PI3K/AKT pathway Isolation of bone marrow-derived macrophages may be a link between immune response through TLRs and re- http://www.jimmunol.org/ storing energy metabolism through mitochondrial biogenesis. Bone marrow-derived macrophages (BMDMs) from AKT12/2, AKT22/2, 2/2 2/2 Recovery of energy capacity by biogenesis requires coordinated MYD88 , and TRIF mice were prepared as described previously expression of genes encoded by two independent genomes (nuclear (24). In brief, femurs and tibias from 6- to 12-wk-old mice were dis- sected, and the bone marrow was flushed out. Bone marrow cells were and mitochondrial), including transcription factor activation and cultured with IMDM supplemented with 30% L929 supernatant containing their binding to promoter consensus sequences in nuclear genes macrophage-stimulating factor, glutamine, sodium pyruvate, 10% heat- encoding mitochondrial proteins (7, 15). Among the transcrip- inactivated FBS (Life Technologies-BRL), and antibiotics for 5–7 d. Mac- 3 6 tional regulators, the mitochondrial transcription factor A (T-fam), rophages were replated at a density of 2 10 cells/well the day before the experiment. a nuclear-encoded protein, governs mitochondrial gene expression (16, 17). In addition, the peroxisome proliferator-activated re- Protein extraction and immunoblotting by guest on September 27, 2021 ceptor g coactivator-1-a (PGC-1a) and nuclear respiratory factors Cells were harvested after the appropriate treatment and washed with PBS. 1 and 2 (NRF-1 and -2) are involved in nuclear respiratory gene Total cellular proteins were extracted after addition of a protease inhibitor expression and serve to integrate mitochondrial gene expression mixture and antiphosphatase I and II, by adding RIPA buffer (Sigma with a wide range of cellular functions (17, 18). Aldrich). Protein concentrations of samples were measured with the BCA The exact molecular mechanism of LPS-mediated bioenergetic assay (Bio-Rad). Equal amounts of protein (5–30 mg) were mixed with same volume of sample buffer (20% glycerol, 4% SDS, 10% 2-ME, 0.05% recovery is not known. In this study, using a murine macrophage bromophenol blue, and 1.25 M Tris-HCl, pH 6.8), loaded on to a 10% cell line and macrophages lacking MYD88/TRIF, we investigated SDS-polyacrylamide gel, and run at 40 mA for 3 h. Proteins were trans- the effect of LPS on mitochondrial bioenergetics. We show that ferred to a polyvinylidene difluoride membrane (Bio-Rad, Hercules, CA) recovery of cellular energy perturbation after LPS treatment is for 30 min at 20 V using a SemiDry Transfer Cell (Bio-Rad). The polyvinylidene difluoride membrane was blocked with 5% dry milk in TBST closely related to PI3K/AKT activation, and inhibition of this (TBS with 0.1% Tween 20) for 1 h, washed, and incubated with primary pathway is directly related to bioenergetic failure and cell death. Ab (diluted between 1/500 and 1/1000 in 5% dry milk in TBST) overnight Using a genetic approach, we show that the AKT1 isoform plays at 4˚C. The blots were washed four times with TBST and then incubated a critical role in mitochondrial bioenergetics in response to TLR4. for 1 h with HRP-conjugated secondary anti-IgG Ab using a dilution be- We present data that indicate that TLR4-mediated AKT activation tween 1/5000 and 1/10,000 in 5% dry milk in TBST. Membranes were washed four times in TBST. Immunoreactive bands were visualized using requires adaptor protein MYD88 and TRIF. Such activation is a chemiluminescent substrate (ECL-Plus; GE Healthcare, Pittsburgh, PA) critical for T-fam upregulation, and increased expression and ac- for 5 min. Images were captured on Hyblot CL film (Denville Scientific, tivity of both COX subunit I (mitochondrially encoded) and COX Metuchen, NJ). OD analysis of signals was performed using ImageQuant subunit IV (nuclear encoded). software from Molecular Dynamics (version 5). Equal loading of the blots was shown by either total AKT, PI3K, or b-actin. Materials and Methods Cell viability Chemicals Cell viability was measured using the MTT assay as described previously (25). Cells equivalent to 1 3 105 ml21 were seeded in 96-well cell culture All chemicals were purchased from Sigma Chemical (St. Louis, MO) unless plates and incubated for 24 h before treatment, as indicated in the Results. specified otherwise. Inhibitors (LY294002, wortmannin, okadaic acid) were Absorbance was measured at 550 nm. Relative cell viability was calculated purchased from Calbiochem (San Diego, CA). Ultrapure LPS was pur- according to the formula: cell viability (%) = absorbance experimental/ chased from Invivogen (San Diego, CA). Anti–phospho-AKT and PI3K, absorbance control 3 100. total AKT, AKT1, AKT2, and b-actin Abs were purchased from Cell Signaling Technology (Beverly, CA). Abs against PI3K and T-fam were ATP assay from Santa Cruz Biotechnology (Santa Cruz, CA). Abs against mitochondrial complexes, COX subunits I and IV (COX I and COX IV), were B6-MCL were cultured to 80% confluency and treated as indicated in each purchased from Mitosciences (Eugene, OR). Ab against manganese su- experiment. After completion of experiments, cells were collected by peroxide dismutase (MnSOD) was purchased from New England Biolab scraping and immediately stored in aliquots at 280˚C until measurement. The Journal of Immunology 3

ATP was released using the boiling method, by addition of 300 ml boiling PGC-1, NRF-1, COX IV, and a reference gene (b-actin) were used to buffer (100 mM Tris-Cl [pH 7.75], 4 mM EDTA) and immediate transfer amplify the corresponding cDNAs. Using iQ SYBR Green Supermix (Bio- to a boiling water bath for 2 min (5). Samples were put on ice and soni- Rad, Hercules, CA), we performed quantitative analysis of mRNA ex- cated for 10 s. Samples were diluted 300-fold, and 50 ml was used to pression with a MX3000p instrument (Stratagene, Santa Clara, CA). PCR determine the ATP concentration using the ATP bioluminescence assay kit amplification was performed in a total volume of 20 ml containing 2 ml HS II (Roche, Indianapolis, IN), according to the manufacturer’s protocol. total cDNA and 20 pg primers (Invitrogen). The PCR amplification pro- Experiments were performed in triplicate, and data were standardized to tocol consisted of an initial incubation step for 10 min at 95˚C, followed by the protein concentration using the DC protein assay kit (Bio-Rad). 45 cycles of denaturation at 95˚C for 10 s, annealing for 20 s at 60˚C, and extension at 72˚C for 20 s. Relative mRNA levels were calculated after Mitochondrial membrane potential measurements normalizing to b-actin. Data were analyzed using the unpaired, two-tailed Student t test, and the results were expressed as relative fold change. The mitochondrial membrane potential (DCm) of intact cells was mea- Primer sequences were: b-actin, forward 59-GATTACTGCTCTGGC- sured as described previously (26) with modifications. Cultured cells were TCCTAGC-39 and reverse 59-GACTCATCGTACTCCTGCTTGC-39; COX washed with PBS and trypsinized. The protein concentration of cells was IV, forward 59-AGTTCAGTTGTACCGCATCCAG-39 and reverse 59-GG- adjusted to 0.2 mg/ml in DMEM without phenol red (Life Technologies- GCCATACACATAGCTCTTCT-39; NRF-1, forward 59-CCATCTATCCG- Invitrogen, Carlsbad, CA) and not supplemented with FBS and antibiotics. AAAGAGACAGC-39 and reverse 59-GGGTGAGATGCAGAGTACAA- Tetramethylrhodamine-methylester (20 nM; TMRM; T-668; Molecular TC-39; PGC-1, forward 59-GAA GGC CGT GTG GTATAC ATT C-39 and Probes-Invitrogen) was added to the cell suspension. As a control, the reverse 59-CTG GCC TCT TTT ACT TCT CGT C-39; and T-fam, forward DCm was dissipated using 1 mM carbonyl cyanide 4-(trifluoromethoxy) 59-CCTGAGGAAAAGCAGGCATA-39 and reverse 59-TCACTTCGTCC- phenylhydrazone (FCCP) and 500 nM oligomycin. Cells were incubated at AACTTCAGC-39. 37˚C for 30 min in the dark under slow rotation. Yellow fluorescence (excitation, 532-nm laser; emission, 585 nm; band pass, 42 nm) was Statistical analysis measured using a BD FACS Array (BD Biosciences, San Jose, CA), and Downloaded from data were analyzed with WinMDI version 2.9 software. Statistical analyses were performed using SPSS software, version 18.0 (SPSS, Chicago, IL). One-way ANOVA test and post hoc repeated-measure Measurement of nitrite comparisons (least significant difference) were performed on all obtained data. ELISA, MTT, ROS, and quantitative real-time RT-PCR results were The nitrite concentration in the culture media was used as a measure of NO expressed as mean 6 SEM. For all analyses, two-tailed p values ,0.05 production (27). After stimulation/incubation, the generation of NO in the were considered significant. cell culture supernatants was determined by measuring nitrite accumula- tion in the medium using Griess reagent (1% sulfanilamide and 0.1% N-[1- http://www.jimmunol.org/ naphthyl]-ethylenediamine dihydrochloride in 5% H3PO4; Sigma). One Results hundred microliters of culture supernatant and 100 ml Griess reagent were LPS treatment induces COX upregulation at the gene and mixed and incubated for 5 min. The absorption was measured in an au- protein level tomated plate reader at 540 nm. Sodium nitrite (NaNO2; Sigma) was used to generate a standard curve for quantification. Background nitrite was Because macrophages play a central role in recognition of TLR subtracted from the experimental values. Results for each condition were obtained from three separate measurements of identically treated wells, ligands and initiating inflammatory responses, as well as ROS and the data are derived from four independent experiments. production, we have chosen murine macrophages to determine the effects of LPS on the ETC. To show reproducibility in independent ELISA systems, we conducted most experiments in two different cell Murine TNF-a and IL-6 cytokine levels were measured in cell culture lines. Murine macrophage cells (B6-MCL) were generated as by guest on September 27, 2021 supernatants according to the manufacturer’s instructions (ELISA DuoKits; previously described (19). We also confirmed the results using R&D Systems) as previously described (25). RAW 264.7 cells. In all experiments, cells were grown to reach Measurement of COX activity 80% confluency. B6-MCL cells were treated with 100 ng/ml ultrapure LPS (to assure only TLR4-mediated signaling) for After appropriate treatments of cell homogenates in incubation buffer (250 different periods as indicated. Because the exact effects of LPS mM sucrose, 20 mM K-HEPES [pH 7.4], 10 mM MgSO4, 2 mM KH2PO4, 1 mM PMSF, 10 mM KF, 2 mM EGTA, 2 mM oligomycin), COX activity on the ETC and mitochondrial transcription factors in immune was analyzed in a closed 200-ml chamber equipped with a micro Clark- cells are unknown, we first investigated the kinetic responses type oxygen electrode (Oxygraph system; Hansatech) as previously de- of important members of the ETC, including COX subunits I scribed (5). Measurements were performed in the absence of nucleotides. and IV, as well as mitochondrial transcription factors NRF-1 Measurements were carried out with 2 mM COX at 25˚C in the presence of 20 mM ascorbate and increasing amounts of cow heart cytochrome c from and T-fam, in response to LPS stimulation. Total cell lysates 0to40mM. Oxygen consumption was recorded and analyzed with the were subjected to immunoblotting using specific Abs against COX Oxygraph plus software. Turnover number is defined as consumed O2 subunits I (mitochondrially encoded) and IV (nuclear encoded), (mM)/min/protein (mg) (5). Protein concentration was determined with the NRF-1, and T-fam. As shown in Fig. 1A, LPS induced expression DC protein assay kit (Bio-Rad). of both subunits of COX in a time-dependent manner. We observed mROS measurements increased protein expression for COX I/IV starting at 3 h after LPS Cells were stimulated with LPS (100 ng/ml) in the presence or absence of challenge, whereas increased expression of T-fam was observed as wortmannin for different time points. As positive control, cells were treated early as 90 min after treatment. LPS stimulation had only a minimal with antimycin A (5 mM) for 30 min. To measure mROS, we incubated effect on NRF-1 protein expression at a later time point (after cells with MitoSOX (Invitrogen) at a final concentration of 1 mM in the 24 h; Fig. 1A). Fig. 1B summarizes the densitometric analysis of dark for 20 min at 37˚C. Cells were collected by centrifugation, washed four different experiments. These data indicate that LPS increased three times in PBS, and immediately resuspended in cold PBS and subjected to FACS analysis. MitoSOX Red was excited by laser at 488 nm, protein expression of COX, and that there was a time-dependent and the data were collected at forward scatter, side scatter, 585/42 nm protein expression of mitochondrial transcription factors in re- (FL2), and 670LP (FL3) channel. Experiments were performed at least in sponse to LPS. To assess the dose-dependent upregulation of COX triplicate. Data are presented as fold change in mean intensity of MitoSOX I, we treated cells with different log concentrations of LPS for 6 h. fluorescence when compared with PBS alone. Supplemental Fig. 1 shows a dose-dependent but nonlinear re- RNA extraction and quantitative reverse sponse to LPS in terms of COX I upregulation. Fig. 1C shows transcriptase/real-time PCR means of relative gene expression for T-fam, COX IV, and NFR-1 Total RNA was extracted using Stat 60 (Iso-Tex Diagnostics, Friendswood, after 2-h LPS treatment. Although relative gene expression in- TX) and reverse transcribed using the Reverse Transcription System creased in response to LPS for all these genes, there was more (Promega, Madison, WI). The primers for amplification of T-fam, AKT, variation in terms of protein expression as determined by immu- 4 LPS AND MITOCHONDRIAL BIOENERGETICS

FIGURE 1. Time-dependent effect of LPS on mitochondrial transcription factors (NRF-1, T-fam) and COX subunits I and IV. Murine macrophages (B6- MCL) were cultured in a density of 2 3 106/well 24 h before LPS treatment. Cells were treated with LPS (100 ng/ml) for different periods. (A) Whole-cell extracts were prepared, and 30 mg total protein was subjected to SDS-PAGE and Western blot analysis using Abs against COX subunits I and IV, NRF-1, and T-fam. Equal loading was determined using b-actin. Cells responded to LPS stimulation with an early increase in expression of T-fam followed by increased expression of both COX subunits (I and IV). NRF-1 increased at a later time point (24 h). (B) Densitometric analysis (mean) of NRF-1, COX I, and T-fam of four independent experiments. (C) Relative gene expression for T-fam, COX IV, and NRF-1 in response to LPS. Cells were treated with LPS (100 ng/ml) for 2 h, RNA was isolated, and quantitative real-time RT-PCR was carried out for T-fam, NRF-1, and COX IV. Values were normalized to b-actin. Results represent mean values of four independent experiments each performed in triplicate. Using ANOVA Mann–Whitney U test, p , 0.05 was Downloaded from considered significant. *p , 0.05, **p , 0.001. noblotting, suggesting posttranscriptional regulation, a regulatory LPS treatment activates both PI3K and AKT mechanism that seems to be important in mitochondria-related The PI3K and AKT pathways are involved in many important genes (28). cellular processes including cell growth, cell survival, and apo- http://www.jimmunol.org/ LPS alters mitochondrial membrane potential, COX activity, ptosis (30). Because of the central role of mitochondria as medi- and ATP production ators of cell survival and apoptosis, we investigated the effect of LPS-mediated AKT activation on the upregulation of T-fam, as The mitochondrial membrane potential (DCm) is generated well as COX I. First, we examined the time-dependent phos- through pumping of protons across the mitochondrial inner mem- phorylation of PI3K and AKT. B6-MCL cells were challenged brane by ETC complexes I, III, and IV (COX) producing the with LPS (100 ng/ml) for different time points as indicated (Fig. proton motive force used by ATP synthase to generate ATP. Pre- 3A). Total cell lysates were subjected to Western blot analysis vious studies have shown that DCm of monocytes is decreased in using a specific Ab against the phosphorylated form of the regu- severe sepsis, and that this is associated with cell death (29). We latory domain of PI3K (phospho-p85) and an Ab detecting the by guest on September 27, 2021 analyzed the kinetic response of DCm in response to LPS treat- Ser473-phosphorylated form of AKT. Fig. 3A shows rapid phos- ment using a membrane potential-sensitive fluorescent probe phorylation of p85 5 min after LPS challenge followed by AKT (TMRM) and flow cytometry. As controls, the DCm was dissi- phosphorylation on Ser473. Next, we explored specific inhibitors of pated using 1 mM FCCP, resulting in a left shift of TMRM fluo- the PI3K/AKT pathway and their effect on LPS-mediated p85 and rescence, or increased DCm using 0.5 mM oligomycin with a right AKT phosphorylation. B6-MCL cells were treated with either shift of fluorescence (Fig. 2A). LPS treatment led to an early (at LY294002 or wortmannin 30 min before LPS challenge or kept in 30 min) reduction in TMRM fluorescence indicating a decreased media without treatment. Fig. 3B shows that both inhibitors DCm (Fig. 2B) followed by a recovery of DCm starting at 3 h LY294002 and wortmannin prevented phosphorylation of p85 and after LPS treatment (Fig. 2C). At 6 h after LPS treatment, DCm AKT, indicating that LPS-induced AKT phosphorylation is PI3K recovered and was even slightly higher compared with untreated dependent. cells (Fig. 2D), consistent with increased protein expression for COX subunits at this time point (Fig. 1A, 1B). Inhibition of the PI3K/AKT pathway abrogates LPS-mediated Next, we asked whether LPS-mediated increases in protein ex- upregulation of COX and T-fam pression for COX I/IV are associated with a change in both the We next assessed the effect of inhibition of the PI3K/AKT pathway activity of COX and changes of ATP content of cells. Murine on protein expression for COX I and T-fam. B6-MCL cells were macrophages (B6-MCL) were cultured in the presence of LPS (100 treated with vehicle (DMSO), LY294002, or wortmannin 30 min ng/ml) for 6 h as COX I and T-fam levels peaked at that time point before LPS challenge. Immunoblotting on total cell lysates was (Fig. 1B). After solubilization, COX activity was measured by performed using specific Abs against NRF-1, COX I, and T-fam. adding increasing amounts of its substrate cytochrome c.As Fig. 4 shows that pretreatment with wortmannin inhibited LPS- shown in Fig. 2E, LPS treatment led to a 56% increase of COX mediated upregulation of both COX I and T-fam with a minimal activity at maximal turnover. effect on NRF-1 expression. Similar results were observed in the To further assess the link between the change of protein ex- presence of LY294002 (Supplemental Fig. 2). We observed an pression in response to LPS treatment and changes in DCm, we increase in expression for both proteins when cells were exposed measured the ATP content of B6-MCL cells. Fig. 2F shows the to wortmannin or LY294002 alone. In the presence of AKT in- time-dependent changes of ATP levels after LPS treatment of B6- hibitors, however, LPS-mediated protein expression for T-fam and MCL cells with a significant increase of ATP after 6 h and a fur- COX I was strongly diminished. These data suggest that the ac- ther increase after 12 h. At 3 h, ATP levels were slightly de- tivation of the PI3K/AKT pathway is required for LPS-mediated creased, then increased .3-fold and 5-fold after 6 and 12 h, upregulation of COX subunit I and T-fam. It is known that protein respectively. These data confirm the relationship between cellular phosphatase 2A plays a pivotal role in dephosphorylation of AKT, ATP and mitochondrial OxPhos capacity represented in the form and thereby diminishes its activity (31). To confirm that phos- of COX activity and expression. phorylation of AKT plays a critical role in LPS-mediated upreg- The Journal of Immunology 5 Downloaded from

FIGURE 2. LPS effect on mitochondrial membrane potential, COX activity, and ATP production. (A) As control, cells were either treated with FCCP (1 m DC m DC

M) to dissipate m, resulting in a left shift of TMRM fluorescence, or treated with 0.5 M oligomycin to increase m showing a right shift of http://www.jimmunol.org/ fluorescence probe. All measurements were performed with a FACScan (Becton-Dickinson) flow cytometer equipped with a yellow fluorescence (excitation, 532-nm laser; emission, 585 nm; band pass, 42 nm) analyzing 10,000 cells in each run. Data obtained were analyzed with the Cell Quest software (n = 4). TMRM fluorescence was normalized to MitoTracker red fluorescence, a membrane potential-independent mitochondrial marker in these cells. (B–D) Kinetic of DCm in response to LPS treatment. Cells were incubated with LPS (100 ng/ml), and relative membrane potentials were determined using the fluorescent probe TMRM. Thirty-minute LPS treatment led to an early decrease of DCm(B) and recovery starting after 3-h treatment (C). After 6 h of LPS treatment, cells showed an increase in fluorescent signal suggesting recovery of mitochondrial mass and function (D). (E) LPS increased COX activity. B6- MCL cells were cultured with or without LPS (100 ng/ml). COX activity was measured 6 h after LPS treatment in solubilized cells by the addition of increasing amounts of cytochrome c. Specific activity (TN, turnover number) is defined as consumed O2 ([mmol]/(min × total protein [mg]). Shown are representative results of three independent experiments. (F) Time-dependent changes of ATP levels in response to LPS. B6-MCL cells were incubated with

LPS (100 ng/ml) for different time points as indicated. ATP concentrations were measured using the bioluminescence method. As indicated, LPS treatment by guest on September 27, 2021 led to a time-dependent increase of ATP concentration. Using ANOVA Mann–Whitney U test, p , 0.05 was considered significant. **p , 0.001. ulation of COX I and T-fam, we evaluated the effect of a protein effect of LPS on MnSOD expression in the presence or absence of phosphatase 2A inhibitor (okadaic acid) to prevent the dephos- LY294002 or wortmannin. B6-MCL cells were cultured and pre- phorylation of AKT. Cells were incubated with okadaic acid (10 treated with either inhibitor for 30 min before LPS treatment. A nM) 30 min before LPS treatment for different time points. As 6-h LPS treatment duration was chosen because it produced a ro- expected, okadaic acid treatment led to AKT phosphorylation bust induction of MnSOD and COX I expression. MnSOD ex- alone and to a prolonged activation of AKT in the presence of LPS pression was analyzed by immunoblotting of total cell lysates (data not shown). Next, we evaluated the effect of okadaic acid in using a specific Ab. As shown in Fig. 4D (lower panel), neither of the presence and absence of LPS on COX I. Okadaic acid pre- the two inhibitors prevented the LPS-mediated protein expression treatment led to an enhanced LPS effect in terms of COX I up- for MnSOD. Interestingly, the presence of wortmannin alone in- regulation (Fig. 4C), supporting the role of active AKT in this duced MnSOD, and a similar effect was found for LY294002. process. These data suggest that LPS-mediated upregulation of MnSOD To further confirm that the AKT pathway is a distinct intra- versus COX I and T-fam occurs through distinct intracellular signal cellular signal and specific to LPS-mediated upregulation of COX I transduction pathways independent of the PI3K/AKT pathway. and T-fam, we assessed the effect of LPS challenge in the presence and absence of LY294002 or wortmannin on MnSOD. MnSOD, AKT inhibition and failure of COX I and T-fam upregulation in a a vital antioxidant enzyme localized to the mitochondrial matrix, response to LPS is associated with increased TNF- , NO, and cell death, and decreased ROS and ATP levels catalyzes the dismutation of superoxide anions to H2O2. It has been shown that LPS challenge leads to rapid induction of MnSOD at It has been suggested that the AKT pathway can play a positive and the transcriptional and protein levels (32, 33). Because rapid up- a negative role in TLR-mediated cytokine production (11, 13). We regulation of the mitochondrial respiratory chain proteins in re- investigated the effect of AKT inhibition on the response of LPS sponse to LPS is associated with increased ROS production (34), stimulation in our cell system. LPS stimulation (100 ng/ml for 24 we investigated the effect of LPS on MnSOD protein expression. h) enhanced TNF-a secretion in supernatants, but secretion of We confirmed in our experimental model that LPS challenge leads TNF-a was significantly higher (p , 0.05) in pretreated cells with to rapid induction of MnSOD (Fig. 4D, upper panel). Although it wortmannin (Fig. 5A). Some studies suggested that NO may in- has been shown that MnSOD expression is associated with mROS duce mitochondrial biogenesis (NRF-1, T-fam, and COX) and production during inflammation, the signaling pathways respon- MnSOD expression (35, 36). Hence we investigated whether the sible for such a response are still unclear. We thus assessed the inhibition of the AKT pathway in LPS-stimulated macrophages 6 LPS AND MITOCHONDRIAL BIOENERGETICS Downloaded from

FIGURE 3. LPS treatment activates the PI3K/AKT pathway in murine macrophages. (A) Murine macrophages (B6-MCL) were cultured and treated with LPS (100 ng/ml) for different time points. Total cell lysates were prepared, and 20 mg total cellular proteins was subjected to SDS- FIGURE 4. Inhibition of the PI3K/AKT pathway abrogates LPS-medi- PAGE and Western blot analysis using phospho-epitope–speciﬁc PI3K p85 ated upregulation of COX I and T-fam, but not MnSOD. (A) Murine 458 199 473

(Tyr /p55 Tyr ) and AKT (Ser ) Abs. Equal loading of protein was macrophages (B6-MCL) were pretreated with wortmannin (100 ng/ml) for http://www.jimmunol.org/ confirmed using total p85 and AKT Abs. The results shown are repre- 30 min before adding LPS (100 ng/ml) to the media for 6 h. After incu- sentative of three independent experiments. LPS treatment led to an early bation, whole-cell extracts were prepared and 20 mg total protein was phosphorylation of the PI3K p85/AKT pathway. (B) Murine macrophages subjected to SDS-PAGE and Western blot analysis for NRF-1, COX I, (B6-MCL) were treated with LY294002 (50 ng/ml) or wortmannin (100 T-fam, and b-actin as loading control. (B) Densitometric analysis (mean 6 ng/ml) 30 min before LPS (100 ng/ml) stimulation. Thirty min after LPS SEM) of COX I in four independent experiments in response to LPS stimulation, total cell lysates were prepared and 20 mg total cellular protein stimulation in the presence of wortmannin. Using ANOVA Mann–Whitney was subjected to Western blot analysis using phospho-epitope–specific U test, p , 0.05 was considered significant. *p , 0.05, **p , 0.001. (C) PI3K detecting dual-phosphorylation sites on p85 (Tyr458/p55 Tyr199) and Murine macrophages (B6-MCL) were pretreated with okadaic acid (10 AKT (Ser473). Equal loading of protein was confirmed using total AKT and nM) 30 min before adding LPS (100 ng/ml) to the media for 6 h. After p85 Abs. Both inhibitors blocked the effect of LPS on phosphorylation incubation, whole-cell extracts were prepared, and 20 mg total protein was by guest on September 27, 2021 of p85 and AKT. (C and D) Densitometric analyses (mean 6 SEM) subjected to SDS-PAGE and Western blot analysis using an Ab against of phosphorylated form of p85 and AKT in response to LPS treatment COX I; b-actin was used to confirm equal loading. Figure is representative in presence and absence of wortmannin and LY294002. Using ANOVA of four independent experiments. (D, upper panel) Time-dependent ex- Mann–Whitney U test, p , 0.05 was considered significant. *p , 0.05, pression of MnSOD in B6-MCL cells. Cells were treated with LPS (100 **p , 0.001. ng/ml) for different periods as indicated. Total cell lysates were prepared and 20 mg protein was subjected to SDS-PAGE and Western blot analysis using an Ab against MnSOD. (D, lower panel) Murine macrophages (B6- changes the NO levels in these cells. Fig. 5B shows the effect of MCL) were pretreated with wortmannin (100 ng/ml) or LY294002 (50 ng/ LPS in the presence and absence of wortmannin on cumulative ml) for 30 min before adding LPS (100 ng/ml) to the media for 6 h. After nitrite production 24 h after LPS treatment as an indirect measure incubation, whole-cell extracts were prepared, and 20 mg total protein was of NO production. As expected, LPS treatment alone resulted in subjected to SDS-PAGE and Western blot analysis using an Ab against a 50% increase of nitrite levels. Interestingly, wortmannin alone MnSOD. Equal loading was assessed using a b-actin Ab. The presence of had no effect on nitrite production, whereas wortmannin in wortmannin or LY294002 did not abrogate LPS-mediated upregulation of combination with LPS led to a 2-fold increase in nitrate levels MnSOD. (Fig. 5B). LPS stimulation of macrophages induces production of ROS such as peroxides and superoxide. ROS play an important generation is diminished, leading to decreased mROS production role in cellular functions via the activation of signaling cascades, and an increase in cell death. Next, we assessed whether the in- and it has been suggested that ROS affect mitochondrial biogen- hibition of the AKT pathway translates into changes of ATP esis, morphology, and function (37). Hence we investigated the content of the cells. B6-MCL cells were cultured and pretreated effect of LPS stimulation in the presence and absence of wort- for 30 min with either LY294002 or wortmannin before stimula- mannin on the production of mROS in form of superoxide. To tion with LPS (100 ng/ml) for 6 h. As shown in Fig. 5D, ATP detect mROS, we further investigated the effect of LPS in the levels were significantly decreased in the presence of both in- presence or absence of wortmannin using a fluorescence dye hibitors (LY294002 or wortmannin), whereas LPS treatment led (MitoSox) that measures predominantly mitochondrial superoxide to an increase in cellular ATP content. To further explore the bi- (3). As shown in Fig. 5C, LPS treatment led to a time-dependent ological effect of inhibition of AKT and subsequent failure of augmentation in mROS, which coincided with increased COX, as upregulation of protein expression for COX I and T-fam, we as- well as mitochondrial membrane potential (Figs. 1, 2). The pres- sessed cell viability using the tetrazolium dye (MTT) assay (25). ence of wortmannin reduced mROS. However, this mROS re- Fig. 5E shows that LPS treatment alone at the concentration ap- duction was associated with increased cell death (Fig. 5E). These plied throughout this study did not change viability. Using AKT data suggest that mitochondrial biogenesis is required for mROS pathway inhibitors LY294002 or wortmannin led to a minimal production. By inhibiting the AKT pathway, mitochondrial re- decrease in cell viability, whereas LPS challenge in the presence The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/ FIGURE 5. Effect of LPS and PI3K inhibitor on cellular ATP and nitrite levels, ROS production, and cell viability. (A) B6-MCL cells were pretreated with wortmannin (100 ng/ml) for 30 min before adding LPS (100 ng/ml) or kept in media for 24 h. Supernatants were analyzed for TNF-a via ELISA. Data are presented as means of four independent experiments and error bars indicate the SEM. p , 0.01. (B) Cumulative nitrite levels in the supernatant were measured as described in Materials and Methods. Data presented as mean 6 SEM of percentage changes of untreated cells. Using ANOVA Mann–Whitney U test, p , 0.05 was considered significant. *p , 0.05, **p , 0.001. (C) mROS were measured using the probe MitoSox. As positive control, cells were treated with antimycin A (AA) for 30 min before treatment with the MitoSox fluorescence probe. Cells were treated with LPS (100 ng/ml) in presence and absence of wortmannin (100 ng/ml). Data are presented as fold change in mean intensity of MitoSOX fluorescence compared with PBS plus MitoSOX alone (negative control). LPS treatment led to a time-dependent increase of mROS after 6 and 24 h. Pretreatment of cells 30 min before LPS stimulation led to a significant decrease in mROS production. Using ANOVA Mann–Whitney U test, p , 0.05 was considered significant. *p , 0.05, **p , 0.001. (D) by guest on September 27, 2021 Murine macrophages (B6-MCL) were treated with LPS (100 ng/ml) for 6 h in the presence or absence of wortmannin (100 ng/ml) or LY294002 (50 ng/ml) 30 min before LPS treatment. ATP concentrations were measured using the bioluminescence method. ATP levels increased in response to LPS. Pretreatment with wortmannin or LY294002 abrogated the LPS-mediated ATP upregulation. Data presented are the result of three independent experiments measured in triplicates. Using ANOVA Mann–Whitney U test, p , 0.05 was considered significant. *p , 0.05, **p , 0.001. (E) Cell viability. A total of 2 3 105 B6-MCL cells was plated in 96-well plate and treated with LPS (100 ng/ml) for 24 h in the presence or absence of wortmannin (100 ng/ml) or LY294002 (50 ng/ml). Cell viability was assessed using MTT assay. Data presented as mean 6 SEM of percentage changes of untreated cells. Using ANOVA Mann– Whitney U test, p , 0.05 was considered significant. *p , 0.05, **p , 0.001. of AKT pathway inhibitor led to a significant loss of cell viability side WT and MYD88/TRIF2/2 cells with LPS for different time (p , 0.001). Collectively, these data suggest that LPS-mediated points to assess AKT activation. Fig. 6A shows phosphorylation of upregulation of COX requires PI3K/AKT activation, and that this p85 and AKT in response to LPS stimulation in WT and MYD88/ activation is important in the prevention of cell death and exces- TRIF2/2. As shown, WT cells responded to LPS stimulation with sive proinflammatory mediators (NO and TNF-a). phosphorylation of PI3K and AKT on Ser473, whereas no phosphorylation of either kinase was observed in MYD88/TRIF2/2 LPS-mediated AKT activation and subsequent COX I and macrophages. These data suggest that LPS-induced activation of T-fam upregulation are MYD88/TRIF dependent the PI3K/AKT pathway is dependent on the presence of MYD88/ Although several studies have shown that LPS stimulation leads to TRIF adaptor protein. To determine whether the lack of AKT PI3K/AKT activation, the mechanism by which LPS activates AKT activation in MYD88/TRIF2/2 would parallel the effect on LPS- is not well understood and remains controversial. In addition, PI3K/ mediated T-fam and COX I expression, we stimulated WT and AKT pathway activation can occur in response to diverse growth MYD88/TRIF2/2 cells side by side with LPS for different time factors and ligand stimulation through diverse receptor activation points (90 min, 3 h, and 6 h). As shown in Fig. 6B, WT cells (30). Macrophages from MYD88/TRIF2/2 mice are defective in responded to LPS treatment with increased expression of both many TLR4-mediated responses, such as LPS-induced secretion T-fam and COX I, whereas MYD88/TRIF2/2 cells failed to up- of cytokines (e.g., IL-6, TNF-a, and IL-1b) (38). A previous study regulate either one of them. These data suggest that LPS-induced has shown that LPS-mediated AKT activation is TLR4/MYD88 phosphorylation of AKT on Ser473 requires the presence of dependent and requires PI3K activation (13). Therefore, we as- MYD88 and TRIF adaptor proteins, and that the LPS-induced sessed whether the LPS-mediated increased expression for COX expressions of T-fam and COX I require activation of the PI3K/ I and T-fam requires MYD88/TRIF adaptor proteins. Phenotypes AKT pathway. Next, we asked whether absence of either MYD88 of WT and MYD88/TRIF2/2 cells were verified by a range of or TRIF would be sufficient to prevent LPS-mediated PI3K/AKT functional parameters, including responsiveness to TLR4 and cy- pathway activation and subsequent T-fam and COX expression. tokine production (Supplemental Fig. 3). We stimulated side-by- BMDMs from WT and macrophages deficient for either MYD88 8 LPS AND MITOCHONDRIAL BIOENERGETICS Downloaded from

FIGURE 6. LPS-induced COX I and T-fam expression is MYD88/TRIF dependent. (A) WT murine macrophages and MYD882/2 TRIF2/2 cells were http://www.jimmunol.org/ cultured side by side in the presence and absence of LPS (100 ng/ml) for 30 or 60 min. Total cell lysates were prepared, and 20 mg total protein was subjected to SDS-PAGE and Western blot analysis using phospho-epitope–specific Abs for p85 and AKT (Ser473). Equal loading of protein was confirmed using total p85 and AKT Abs. (B) WT murine macrophages and MYD88/TRIF2/2 cells were cultured side by side in the presence and absence of LPS for 90 min, 3 h, and 6 h. Total cell lysates were prepared and 20 mg total protein was subjected to SDS-PAGE and Western blot analysis using Ab against T-fam and COX I. Equal loading of protein was confirmed using b-actin Ab. WT cells responded to LPS stimulation with a significant increase in T-fam starting at 90 min throughout 6 h after treatment with LPS. Upregulation of COX I in WT was seen starting at 3 h after LPS stimulation. In contrast, MYD88/TRIF2/2 cells did not respond to LPS with increased expression for T-fam and COX I. (C) BMDMs isolated from WT, TRIF-, or MYD88-deficient mice were cultured in parallel with LPS for 5, 30, and 60 min. Total cell lysates were prepared, and 20 mg total protein was subjected to SDS-PAGE and Western blot 473

analysis using phospho-epitope–specific Abs for p85 and AKT (Ser ). Equal loading was confirmed using total AKT and total p85 Abs. LPS treatment by guest on September 27, 2021 evoked a weak and delayed AKT phosphorylation in TRIF2/2 cells, but MYD88-deficient cells did not show any response to LPS treatment. (D) BMDMs isolated from WT, TRIF-, or MYD88-deficient mice were either kept in media or treated with LPS for 3 h. Twenty micrograms of total protein was subjected to SDS-PAGE and Western blot analysis using T-fam Ab. Equal loading was confirmed using b-actin Ab. Although LPS treatment led to an increase in T-fam expression in WT, neither TRIF2/2 nor MYD882/2 cells responded with an increased expression for T-Fam. (E) BMDMs isolated from WT, TRIF-, or MYD88-deficient mice were either kept in media or treated with LPS for 3 and 6 h. Twenty micrograms of total protein was subjected to SDS-PAGE and Western blot analysis using a COX IV Ab. Equal loading was confirmed using total AKT Ab. Although LPS treatment led to an increase in COX IV expression in WT after 6 h, TRIF2/2 and MYD882/2 cells did not show any expression of COX IV. or TRIF were stimulated with LPS for the indicated times. As WT mice were cultured side by side under equal conditions. Cells shown, LPS-mediated PI3K/AKT pathway activation required were treated with LPS (100 ng/ml) for different time points, and both MYD88 and TRIF. In macrophages lacking TRIF, we ob- whole-cell extracts were subjected to immunoblotting. As seen is served an attenuated and delayed response in AKT phosphor- Fig. 7A, LPS stimulation of BMDMs isolated from WT AKT12/2 ylation, whereas cells lacking MYD88 did not respond to LPS and AKT22/2 led to rapid phosphorylation of p85 and AKT with AKT phosphorylation (Fig. 6C). Similarly, WT macrophages phosphorylation on phospho-Ser473. This Ab detects phosphory- responded to LPS with an increased expression for T-fam (Fig. lated AKT on Ser473 regardless of the isoforms expressed (either 6D) and COX IV (Fig. 6E), whereas macrophages deficient in AKT1, AKT2, or AKT3). Next, we used isoform-specific AKT1 MYD88 or TRIF did not increase expression for T-fam (Fig. 6D) and AKT2 Abs to detect the level of AKT1 and AKT2 expression and COX IV (Fig. 6E). in WT, AKT12/2, and AKT22/2 BMDMs. Fig. 7B confirmed the lack of expression of AKT1 protein in AKT12/2 and of AKT2 in Expression of AKT1 is pivotal in LPS-mediated COX AKT22/2 macrophages, respectively. As described earlier, in our upregulation in BMDMs cell model, we detected the highest expression for COX 6 h after Three AKT isoforms have been described with distinct biological LPS treatment. Thus, immunoblotting was done using the 6 h time functions. To further investigate mechanistically which AKT iso- point after LPS treatment. Fig. 7C shows that the cells lacking form plays a major role in the LPS response and mitochondrial AKT1 had severe impairment in COX I and COX IV expression biogenesis in macrophages, we used cells derived from knockout at baseline, as well as in response to LPS. In contrast, the induc- mice lacking AKT1 or AKT2. Whereas AKT1 and AKT2 are tion of COX I was unimpaired in response to LPS treatment in ubiquitously expressed, including immune cells, endothelial cells, AKT22/2 macrophages (Fig. 7C). Furthermore, the basal level of and in hematopoietic cells, AKT3 expression is largely confined to COX IV was lower in these macrophages as compared with those the testes and brain (39–41). Thus, we focused on AKT1 and of WT, but AKT22/2 macrophages responded to LPS with an AKT2 isoforms. BMDMs from mice deficient in AKT1, AKT2, or enhanced expression for COX IV. These findings indicate that the The Journal of Immunology 9

FIGURE 7. Critical role of AKT1 in LPS-mediated COX expression. BMDMs were isolated from WT, AKT1-, or AKT2-deficient mice, cultured under equal conditions, and treated with LPS for different time points as indicated. (A) Detection of phospho-p85 and -AKT. Twenty micrograms total protein was subjected to SDS-PAGE and Western blot analysis using phospho-epitope–specific Abs for p85 and AKT (Ser473). Equal loadings were confirmed using either total p85 or total AKT Abs. (B) Detection of AKT isoforms. Immunoblot analysis of whole-cell lysate performed using Abs detecting AKT1 and AKT2. Equal loading was confirmed using b-actin Ab. As shown, AKT12/2 macrophages lack expression of AKT1, whereas AKT22/2 macrophages lack AKT2. (C) BMDMs isolated from WT, AKT12/2, and AKT22/2 mice kept in media or treated with LPS for 6 h. Immunoblot analysis of whole-cell lysate performed using Abs detecting COX I (upper panel) and COX IV (lower panel). Equal loading was confirmed using b-actin Ab. Cells lacking AKT1 had severe impairment in COX I and COX IV expression at baseline, as well as in response to LPS. Induction of COX I was unimpaired in response to LPS treatment in AKT22/2 macrophages. Furthermore, the basal level of COX IV was lower in these macrophages as compared with those of WT, but AKT22/2 macrophages responded to LPS with an enhanced expression for COX IV. Downloaded from

AKT1 isoform plays a central role in LPS-mediated COX I and IV important in the response of macrophages to bacterial products expression, whereas AKT2 contributes only to COX IV upregu- and may facilitate the required energy for cytokine production, as lation by LPS in BMDMs. well as phagocytic activity of these cells. One recent study found

that mROS production in response to specific TLRs is required for http://www.jimmunol.org/ Discussion phagocytic activity of macrophages (3). Another study found that In this study, we analyzed the molecular mechanisms underlying AKT1 and AKT2 isoforms regulate the intracellular ROS content the LPS-mediated upregulation of ETC components focusing on in hematopoietic stem cells necessary for their function and pro- COX and mitochondrial transcription factors (T-fam and NRF-1) as liferation (40). markers of mitochondrial biogenesis. We show that LPS-mediated COX biogenesis is a critical part of mitochondrial biogenesis. upregulation of COX and T-fam is dependent on AKT phos- Among the 13 subunits of COX, the largest subunits, COX I–III, are phorylation (Figs. 1, 3), and that inhibition of the PI3K/AKT encoded by the mitochondrial DNA and regulated at the tran- pathway leads to abrogation of this effect (Fig. 3). Increased ex- scriptional level by T-fam (45). COX is the terminal enzyme of the pression of COX subunits I and IV in response to LPS challenge ETC and plays a pivotal role in electron transfer to molecular by guest on September 27, 2021 was associated with restoration of COX activity and increased oxygen, and generation and maintenance of DCm and ATP. COX cellular ATP levels (Fig. 2). Inhibition of the PI3K/AKT pathway activity, as well as expression, is tightly regulated by metabolic abrogated the effect of LPS on both COX and T-fam, and led to cellular demand (5, 42). Our laboratory has shown that TNF-a decreased ATP production and increased cell death. We also show challenge leads to rapid inhibition of COX, and this inhibition is that the effect of LPS on both COX and T-fam is dependent on due to tyrosine phosphorylation of the catalytic subunit I of COX MYD88 and TRIF adaptor proteins. LPS challenge of MYD88/ (5). This finding may explain, at least in part, the observed met- TRIF knockout cells failed to phosphorylate AKT and upregulate abolic switch from aerobic to glycolytic energy metabolism during T-fam and COX (Fig. 6A, 6B). Cells lacking MYD88 did not sepsis. However, not all septic patients experience organ failure respond to LPS with AKT activation, whereas macrophages defi- with a poor outcome. Thus, additional mechanisms may be es- cient in TRIF showed a delayed AKT activation (Fig. 6C). Neither sential for the recovery from sepsis. Our current data indicate that MYD88 nor TRIF deficient cells showed an increase in T-fam or LPS stimulation of murine macrophages leads to 20-fold increase COX expression (Fig. 6D, 6E). These data suggest that MYD88/ in expression of COX I (Fig. 1A), and that such an increase was TRIF is required both for AKT phosphorylation and for LPS- associated with a 56% increase in COX activity. mediated increased expression of COX and T-fam. Importantly, Evidence suggest that nonsurvivors of sepsis exhibit more ex- macrophages from AKT12/2 mice showed a lack of response to tensive mitochondrial damage and failure of cellular energetics LPS in terms of upregulation of COX I and IV (Fig. 7C). These compared with survivors as determined from muscle biopsies of data collectively indicate that AKT1 regulates and maintains mi- critically ill patients (46). Interestingly, Staphylococcus aureus- tochondrial function in response to TLR4 agonists. induced sepsis in mice with a lower bacterial load led to a recov- Mitochondria are highly dynamic organelles and essential in ery from bioenergetic failure with orderly (sequential) induction providing most cellular ATP. They have emerged as important of NRF-1, NRF-2, PGC-1, and T-fam m-RNAs in the liver of integrators of several vital signaling cascades (42). Mitochondrial septic animals (47), whereas animals with sepsis induced with membrane potential (DCm) builds the proton motive force for a higher bacterial count failed to recover and exhibited lack of generation of ATP through OxPhos and is critical for the integrity mitochondrial biogenesis. We have observed that treatment of of mitochondrial function. Bacteria and several bacterial products macrophages with a higher dose of LPS (.1 mg/ml) failed to such as LPS or bacterial toxins have been shown to affect DCm increase expression for COX I (Supplemental Fig. 1) and T-fam. and alter the OxPhos, as well as mROS production (3, 43, 44). We In contrast with Haden et al.’s observation (47), we show ex- show in this study the dynamic changes of DCm of murine pression at both the gene and protein levels that murine macro- macrophages in response to LPS, an initial decrease of DCm phages exhibit higher baseline NRF-1 mRNA expression and followed by a delayed restoration of DCm that mirrored the up- minimal change in response to LPS treatment as compared with regulation of COX (Figs. 1, 2). This process may be specifically PGC-1a (data not shown), T-fam, and COX I and IV. It should be 10 LPS AND MITOCHONDRIAL BIOENERGETICS noted, however, that tissue and cell-specific differences may exist, been shown that MYD88-deficient animals exhibit poorer outcome and that various mitochondria-related transcription factors likely when infected with various microorganisms that were associated play different roles in specific tissues and cells. In addition, diverse with an enhanced bacterial burden (11, 56–58). This paradigm, at posttranscriptional modifications of mitochondrial transcription fac- least in part, may be explained by their inability to enhance mi- tors such as phosphorylation or redox regulation (e.g., in the case of tochondrial biogenesis in response to microbial stimuli. NRF-1) have been reported (15). Such events may be important in regulation of mitochondrial transcripts in response to LPS. Acknowledgments Recognition of LPS by TLR4 plays a pivotal role in the initiation We thank Dr. Eicke Latz (University of Massachusetts Medical School, of Gram-negative sepsis and septic shock through diverse in- Worcester, MA) for providing the macrophage cell lines. flammatory processes in mammals. LPS challenge leads to activation of diverse cell signaling pathways such as inflammatory, Disclosures apoptotic, but also prosurvival pathways. It has been shown that The authors have no financial conflicts of interest. LPS challenge results in activation of the PI3K/AKT pathway, which is required for activation of survival pathways (48). Both PI3K and AKT pathway activation have been reported to be as- References sociated with TLR2, TLR3, TLR4, and TLR5 in different cells 1. Doyle, S. L., and L. A. O’Neill. 2006. Toll-like receptors: from the discovery of (13, 49). Although it has been suggested that the activation of the NFkappaB to new insights into transcriptional regulations in innate immunity. Biochem. Pharmacol. 72: 1102–1113. PI3K/AKT pathway has a negative regulatory function on TLR 2. Rich, P. 2003. Chemiosmotic coupling: The cost of living. Nature 421: 583. ligation, the level of interaction between the TLRs and the PI3K/ 3. West, A. P., I. E. Brodsky, C. Rahner, D. K. Woo, H. Erdjument-Bromage, Downloaded from AKT pathway is not well understood (13, 49, 50). Several studies P. Tempst, M. C. Walsh, Y. Choi, G. S. Shadel, and S. Ghosh. 2011. TLR signalling augments macrophage bactericidal activity through mitochondrial ROS. suggested that the inhibition of this pathway leads to decreased Nature 472: 476–480. phosphatase activity, resulting in an increased production of in- 4. Levy, R. J., and C. S. Deutschman. 2007. Cytochrome c oxidase dysfunction in sepsis. Crit. Care Med. 35(9 Suppl.): S468–S475. flammatory mediators such as IL-1b and IL-6 with an adverse 5. Samavati, L., I. Lee, I. Mathes, F. Lottspeich, and M. Hu¨ttemann. 2008. Tumor effect on cell survival (11). Our data indicate that inhibition of necrosis factor alpha inhibits oxidative phosphorylation through tyrosine phosphorylation at subunit I of cytochrome c oxidase. J. Biol. Chem. 283: 21134– AKT pathway in presence of LPS increases proinflammatory http://www.jimmunol.org/ 21144. cytokines (IL-6 and TNF-a), ROS, NO, and decreased cell sur- 6. Carre´, J. E., and M. Singer. 2008. Cellular energetic metabolism in sepsis: the vival in murine macrophages, and these findings paralleled the need for a systems approach. Biochim. Biophys. Acta 1777: 763–771. lack of mitochondrial biogenesis. Although we have observed in- 7. Scarpulla, R. C. 2008. Nuclear control of respiratory chain expression by nuclear respiratory factors and PGC-1-related coactivator. Ann. N. Y. Acad. Sci. 1147: creased ROS (both cellular and mROS) in response to LPS, the 321–334. presence of wortmannin led to an increased fluorescence in CM- 8. Brealey, D., S. Karyampudi, T. S. Jacques, M. Novelli, R. Stidwill, V. Taylor, R. T. Smolenski, and M. Singer. 2004. Mitochondrial dysfunction in a long-term H2DCFDA (data not shown) but a decrease in the MitoSox signal. rodent model of sepsis and organ failure. Am. J. Physiol. Regul. Integr. Comp. We propose that the enhanced CM-H2DCFDA probe in the pres- Physiol. 286: R491–R497. ence of wortmannin is due to increased peroxynitrite in the cy- 9. Wang, Z. J., F. M. Zhang, L. S. Wang, Y. W. Yao, Q. Zhao, and X. Gao. 2009. Lipopolysaccharides can protect mesenchymal stem cells (MSCs) from oxidative by guest on September 27, 2021 toplasm because of cell death processes. As shown in Fig. 5C, stress-induced apoptosis and enhance proliferation of MSCs via Toll-like wortmannin led to a decreased mROS, but also increased cell receptor(TLR)-4 and PI3K/Akt. Cell Biol. Int. 33: 665–674. death (Fig. 5E). 10. Schabbauer, G., J. Luyendyk, K. Crozat, Z. Jiang, N. Mackman, S. Bahram, and P. Georgel. 2008. TLR4/CD14-mediated PI3K activation is an essential com- Moreover, in macrophages, TLR ligation initiates autophagy, ponent of interferon-dependent VSV resistance in macrophages. Mol. Immunol. a process that provides metabolic intermediates to maintain bio- 45: 2790–2796. 11. Medina, E. A., I. R. Morris, and M. T. Berton. 2010. Phosphatidylinositol 3- energetics necessary to carry out the killing of pathogens and to kinase activation attenuates the TLR2-mediated macrophage proinflammatory promote survival (51, 52). It has been shown that mammalian cells cytokine response to Francisella tularensis live vaccine strain. J. Immunol. 185: with hyperactive AKT accumulate ATP, whereas cells with re- 7562–7572. 12. Kenny, E. F., and L. A. O’Neill. 2008. Signalling adaptors used by Toll-like duced AKT activity exhibit markedly reduced ATP content (53). receptors: an update. Cytokine 43: 342–349. Such effects have been explained by the role of activated AKT on 13. Laird, M. H., S. H. Rhee, D. J. Perkins, A. E. Medvedev, W. Piao, M. J. Fenton, mitochondria-associated hexokinase and glycolysis (54). Our data and S. N. Vogel. 2009. TLR4/MYD88/PI3K interactions regulate TLR4 signaling. J. Leukoc. Biol. 85: 966–977. indicate that presence of AKT1 and its activation in response to 14. Rajaram, M. V., L. P. Ganesan, K. V. Parsa, J. P. Butchar, J. S. Gunn, and LPS is critical for the upregulation of mitochondrial complexes, in S. Tridandapani. 2006. Akt/Protein kinase B modulates macrophage inflammatory response to Francisella infection and confers a survival advantage in mice. particular, COX. Previously, AKT1 has been implicated as a crit- J. Immunol. 177: 6317–6324. ical regulator in homeostasis of hematopoietic stem cell function 15. Scarpulla, R. C. 2008. Transcriptional paradigms in mammalian mitochondrial and ROS (40). Interestingly, another study associated the loss of biogenesis and function. Physiol. Rev. 88: 611–638. 16. Gleyzer, N., K. Vercauteren, and R. C. Scarpulla. 2005. Control of mitochondrial AKT1 with severe atherosclerosis in a mice model of coronary transcription specificity factors (TFB1M and TFB2M) by nuclear respiratory artery disease (55). Macrophages of animals lacking AKT1 were factors (NRF-1 and NRF-2) and PGC-1 family coactivators. Mol. Cell. Biol. 25: more susceptible to apoptosis and exhibited defective clearance 1354–1366. 17. Scarpulla, R. C. 2006. Nuclear control of respiratory gene expression in mam- of dead foam cells (55). Our study sheds some light on these malian cells. J. Cell. Biochem. 97: 673–683. observations because our data mechanistically suggest that sup- 18. Dhar, S. S., S. Ongwijitwat, and M. T. Wong-Riley. 2008. Nuclear respiratory factor 1 regulates all ten nuclear-encoded subunits of cytochrome c oxidase in pressed mitochondrial energetics cause an impaired stress re- neurons. J. Biol. Chem. 283: 3120–3129. 2/2 sponse in AKT1 macrophages. PI3K is involved in many 19. Hornung, V., F. Bauernfeind, A. Halle, E. O. Samstad, H. Kono, K. L. Rock, important cellular pathways including cell growth, migration, and K. A. Fitzgerald, and E. Latz. 2008. Silica crystals and aluminum salts activate the NALP3 inflammasome through phagosomal destabilization. Nat. Immunol. 9: apoptosis (30). It has been shown that the PI3K/AKT pathway is 847–856. activated in response to growth factors but also in response to TLR 20. Nagpal, K., T. S. Plantinga, C. M. Sirois, B. G. Monks, E. Latz, M. G. Netea, and ligands (13). However, PI3K activation in absence of AKT1 was D. T. Golenbock. 2011. Natural loss-of-function mutation of myeloid differentiation protein 88 disrupts its ability to form Myddosomes. J. Biol. Chem. 286: not sufficient to upregulate COX in response to LPS. Our results 11875–11882. indicate that the PI3K/AKT pathway is required for upregulation 21. Park, J. H., Y. G. Kim, C. McDonald, T. D. Kanneganti, M. Hasegawa, M. Body- Malapel, N. Inohara, and G. Nuń˜ez. 2007. RICK/RIP2 mediates innate immune of mitochondrial OxPhos and T-fam, and both processes require responses induced through Nod1 and Nod2 but not TLRs. J. Immunol. 178: the adaptor protein MYD88/TRIF (Fig. 6A, 6B). Previously, it has 2380–2386. The Journal of Immunology 11

22. Cho, H., J. Mu, J. K. Kim, J. L. Thorvaldsen, Q. Chu, E. B. Crenshaw, III, 41. Calamito, M., M. M. Juntilla, M. Thomas, D. L. Northrup, J. Rathmell, K. H. Kaestner, M. S. Bartolomei, G. I. Shulman, and M. J. Birnbaum. 2001. M. J. Birnbaum, G. Koretzky, and D. Allman. 2010. Akt1 and Akt2 promote Insulin resistance and a diabetes mellitus-like syndrome in mice lacking the peripheral B-cell maturation and survival. Blood 115: 4043–4050. protein kinase Akt2 (PKB beta). Science 292: 1728–1731. 42. Hu¨ttemann, M., I. Lee, L. Samavati, H. Yu, and J. W. Doan. 2007. Regulation of 23. Cho, H., J. L. Thorvaldsen, Q. Chu, F. Feng, and M. J. Birnbaum. 2001. Akt1/ mitochondrial oxidative phosphorylation through cell signaling. Biochim. Bio- PKBalpha is required for normal growth but dispensable for maintenance of phys. Acta 1773: 1701–1720. glucose homeostasis in mice. J. Biol. Chem. 276: 38349–38352. 43. Hickson-Bick, D. L., C. Jones, and L. M. Buja. 2008. Stimulation of mito- 24. Franchi, L., A. Amer, M. Body-Malapel, T. D. Kanneganti, N. Ozo¨ren, chondrial biogenesis and autophagy by lipopolysaccharide in the neonatal rat R. Jagirdar, N. Inohara, P. Vandenabeele, J. Bertin, A. Coyle, et al. 2006. Cy- cardiomyocyte protects against programmed cell death. J. Mol. Cell. Cardiol. 44: tosolic flagellin requires Ipaf for activation of caspase-1 and interleukin 1beta in 411–418. salmonella-infected macrophages. Nat. Immunol. 7: 576–582. 44. Stavru, F., F. Bouillaud, A. Sartori, D. Ricquier, and P. Cossart. 2011. Listeria 25. Samavati, L., R. Rastogi, W. Du, M. Hu¨ttemann, A. Fite, and L. Franchi. 2009. monocytogenes transiently alters mitochondrial dynamics during infection. Proc. STAT3 tyrosine phosphorylation is critical for interleukin 1 beta and interleukin- Natl. Acad. Sci. USA 108: 3612–3617. 6 production in response to lipopolysaccharide and live bacteria. Mol. Immunol. 45. Qi, Z., J. He, Y. Su, Q. He, J. Liu, L. Yu, O. Al-Attas, T. Hussain, S. Ding, L. Ji, 46: 1867–1877. and M. Qian. 2011. Physical exercise regulates p53 activity targeting SCO2 and 26. Lee, I., A. Pecinova, P. Pecina, B. G. Neel, T. Araki, R. Kucherlapati, increases mitochondrial COX biogenesis in cardiac muscle with age. PLoS ONE A. E. Roberts, and M. Hu¨ttemann. 2010. A suggested role for mitochondria in 6: e21140. Noonan syndrome. Biochim. Biophys. Acta 1802: 275–283. 46. Carre´, J. E., J. C. Orban, L. Re, K. Felsmann, W. Iffert, M. Bauer, H. B. Suliman, 27. Miranda, K. M., M. G. Espey, and D. A. Wink. 2001. A rapid, simple spectro- C. A. Piantadosi, T. M. Mayhew, P. Breen, et al. 2010. Survival in critical illness photometric method for simultaneous detection of nitrate and nitrite. Nitric is associated with early activation of mitochondrial biogenesis. Am. J. Respir. Oxide 5: 62–71. Crit. Care Med. 182: 745–751. 28. Tian, Q., S. B. Stepaniants, M. Mao, L. Weng, M. C. Feetham, M. J. Doyle, 47. Haden, D. W., H. B. Suliman, M. S. Carraway, K. E. Welty-Wolf, A. S. Ali, E. C. Yi, H. Dai, V. Thorsson, J. Eng, et al. 2004. Integrated genomic and H. Shitara, H. Yonekawa, and C. A. Piantadosi. 2007. Mitochondrial biogenesis proteomic analyses of gene expression in Mammalian cells. Mol. Cell. Proteo- restores oxidative metabolism during Staphylococcus aureus sepsis. Am. J. mics 3: 960–969. Respir. Crit. Care Med. 176: 768–777. 29. Adrie, C., M. Bachelet, M. Vayssier-Taussat, F. Russo-Marie, I. Bouchaert, 48. Yao, Y., F. Zhang, L. Wang, G. Zhang, Z. Wang, J. Chen, and X. Gao. 2009. Downloaded from M. Adib-Conquy, J. M. Cavaillon, M. R. Pinsky, J. F. Dhainaut, and B. S. Polla. Lipopolysaccharide preconditioning enhances the efficacy of mesenchymal stem 2001. Mitochondrial membrane potential and apoptosis peripheral blood cells transplantation in a rat model of acute myocardial infarction. J. Biomed. monocytes in severe human sepsis. Am. J. Respir. Crit. Care Med. 164: 389–395. Sci. 16: 74. 30. Koyasu, S. 2003. The role of PI3K in immune cells. Nat. Immunol. 4: 313–319. 49. Molnarfi, N., L. Gruaz, J. M. Dayer, and D. Burger. 2007. Opposite regulation of 31. Liao, Y., and M. C. Hung. 2004. A new role of protein phosphatase 2a in ade- IL-1beta and secreted IL-1 receptor antagonist production by phosphatidyl- noviral E1A protein-mediated sensitization to anticancer drug-induced apoptosis inositide-3 kinases in human monocytes activated by lipopolysaccharides or in human breast cancer cells. Cancer Res. 64: 5938–5942. contact with T cells. J. Immunol. 178: 446–454.

32. Rogers, R. J., J. M. Monnier, and H. S. Nick. 2001. Tumor necrosis factor-alpha 50. Tsukamoto, K., K. Hazeki, M. Hoshi, K. Nigorikawa, N. Inoue, T. Sasaki, and http://www.jimmunol.org/ selectively induces MnSOD expression via mitochondria-to-nucleus signaling, O. Hazeki. 2008. Critical roles of the p110 beta subtype of phosphoinositide 3- whereas interleukin-1beta utilizes an alternative pathway. J. Biol. Chem. 276: kinase in lipopolysaccharide-induced Akt activation and negative regulation of 20419–20427. nitrite production in RAW 264.7 cells. J. Immunol. 180: 2054–2061. 33. Tsan, M. F., R. N. Clark, S. M. Goyert, and J. E. White. 2001. Induction of TNF- 51. Carloni, S., S. Girelli, C. Scopa, G. Buonocore, M. Longini, and W. Balduini. alpha and MnSOD by endotoxin: role of membrane CD14 and Toll-like receptor- 2010. Activation of autophagy and Akt/CREB signaling play an equivalent role 4. Am. J. Physiol. Cell Physiol. 280: C1422–C1430. in the neuroprotective effect of rapamycin in neonatal hypoxia-ischemia. Auto- 34. White, J. E., and M. F. Tsan. 2001. Differential induction of TNF-alpha and phagy 6: 366–377. MnSOD by endotoxin: role of reactive oxygen species and NADPH oxidase. Am. 52. Xu, Y., C. Jagannath, X. D. Liu, A. Sharafkhaneh, K. E. Kolodziejska, and J. Respir. Cell Mol. Biol. 24: 164–169. N. T. Eissa. 2007. Toll-like receptor 4 is a sensor for autophagy associated with 35. Larosche, I., P. Lette´ron, A. Berson, B. Fromenty, T. T. Huang, R. Moreau, innate immunity. Immunity 27: 135–144. D. Pessayre, and A. Mansouri. 2010. Hepatic mitochondrial DNA depletion after 53. Robey, R. B., and N. Hay. 2009. Is Akt the “Warburg kinase”?-Akt-energy

an alcohol binge in mice: probable role of peroxynitrite and modulation by metabolism interactions and oncogenesis. Semin. Cancer Biol. 19: 25–31. by guest on September 27, 2021 manganese superoxide dismutase. J. Pharmacol. Exp. Ther. 332: 886–897. 54. Robey, R. B., and N. Hay. 2006. Mitochondrial hexokinases, novel mediators of 36. Nisoli, E., E. Clementi, C. Paolucci, V. Cozzi, C. Tonello, C. Sciorati, R. Bracale, the antiapoptotic effects of growth factors and Akt. Oncogene 25: 4683–4696. A. Valerio, M. Francolini, S. Moncada, and M. O. Carruba. 2003. Mitochondrial 55. Fernańdez-Hernando, C., E. Ackah, J. Yu, Y. Sua´rez, T. Murata, Y. Iwakiri, biogenesis in mammals: the role of endogenous nitric oxide. Science 299: 896– J. Prendergast, R. Q. Miao, M. J. Birnbaum, and W. C. Sessa. 2007. Loss of Akt1 899. leads to severe atherosclerosis and occlusive coronary artery disease. Cell Metab. 37. Irrcher, I., V. Ljubicic, and D. A. Hood. 2009. Interactions between ROS and 6: 446–457. AMP kinase activity in the regulation of PGC-1alpha transcription in skeletal 56. Kissner, T. L., E. D. Cisney, R. G. Ulrich, S. Fernandez, and K. U. Saikh. 2010. muscle cells. Am. J. Physiol. Cell Physiol. 296: C116–C123. Staphylococcal enterotoxin A induction of pro-inflammatory cytokines and le- 38. Kawai, T., O. Adachi, T. Ogawa, K. Takeda, and S. Akira. 1999. Unrespon- thality in mice is primarily dependent on MYD88. Immunology 130: 516–526. siveness of MYD88-deficient mice to endotoxin. Immunity 11: 115–122. 57. Loof, T. G., O. Goldmann, A. Gessner, H. Herwald, and E. Medina. 2010. 39. Easton, R. M., H. Cho, K. Roovers, D. W. Shineman, M. Mizrahi, M. S. Forman, Aberrant inflammatory response to Streptococcus pyogenes in mice lacking V. M. Lee, M. Szabolcs, R. de Jong, T. Oltersdorf, et al. 2005. Role for Akt3/ myeloid differentiation factor 88. Am. J. Pathol. 176: 754–763. protein kinase Bgamma in attainment of normal brain size. Mol. Cell. Biol. 25: 58. Shen, Y., I. Kawamura, T. Nomura, K. Tsuchiya, H. Hara, S. R. Dewamitta, 1869–1878. S. Sakai, H. Qu, S. Daim, T. Yamamoto, and M. Mitsuyama. 2010. Toll-like 40. Juntilla, M. M., V. D. Patil, M. Calamito, R. P. Joshi, M. J. Birnbaum, and receptor 2- and MYD88-dependent phosphatidylinositol 3-kinase and Rac1 ac- G. A. Koretzky. 2010. AKT1 and AKT2 maintain hematopoietic stem cell tivation facilitates the phagocytosis of Listeria monocytogenes by murine mac- function by regulating reactive oxygen species. Blood 115: 4030–4038. rophages. Infect. Immun. 78: 2857–2867.