Induced Autophagy −γ Mediate IFN- Tryptophan Depletion and the ...

of June 28, 2022.This information is current as

Induced Autophagy−γMediate IFN-Tryptophan Depletion and the Kinase GCN2

Beaune, Eric Thervet and Nicolas PalletSophie Fougeray, Iadh Mami, Gildas Bertho, Philippe

http://www.jimmunol.org/content/189/6/2954doi: 10.4049/jimmunol.1201214August 2012;

2012; 189:2954-2964; Prepublished online 15J Immunol 

Referenceshttp://www.jimmunol.org/content/189/6/2954.full#ref-list-1

, 20 of which you can access for free at: cites 60 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2012 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on June 28, 2022

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Tryptophan Depletion and the Kinase GCN2 MediateIFN-g–Induced Autophagy

Sophie Fougeray,*,† Iadh Mami,*,† Gildas Bertho,†,‡ Philippe Beaune,*,†,x Eric Thervet,*,†,{

and Nicolas Pallet*,†

IFN-g is a master regulator of the immune responses that occur in the transplanted kidney, acting both on the immune system and

on the graft itself. The cellular responses to IFN-g are complex, and emerging evidence suggests that IFN-g may regulate

autophagic functions. Conversely, autophagy modulates innate and adaptive immune functions in various contexts. In this study,

we identify a novel mechanism by which IFN-g activates autophagy in human kidney epithelial cells and provide new insights into

how autophagy regulates immune functions in response to IFN-g. Our results indicate that IFN-g promotes tryptophan depletion,

activates the eIF2a kinase general control nonderepressible-2 (GCN2), and leads to an increase in the autophagic flux. Further,

tryptophan supplementation and RNA interference directed against GCN2 inhibited IFN-g–induced autophagy. This process is of

functional relevance because autophagy regulates the secretion of inflammatory cytokines and growth factors by human kidney

epithelial cells in response to IFN-g. These findings assign to IFN-g a novel function in the regulation of autophagy, which, in turn,

modulates IFN-g–induced secretion of inflammatory cytokines. The Journal of Immunology, 2012, 189: 2954–2964.

The renal tubular epithelium plays a central role in the de-velopment of kidney allograft structural deterioration, par-ticularlybythegenerationof inflammatoryandprofibrogenic

mediators secreted in response to injury (1–5). The inflamed kidneytissue is a highly immunogenic microenvironment that activatesprofessional and nonprofessional APCs and triggers recipient T cellactivation and proliferation, ultimately leading to rejection, epithe-lium dedifferentiation, and fibrosis (6–9). IFN-g is a master regulatorof the homeostasis of the kidney transplant during rejection (8, 10).IFN-g is a cytokine that is produced mostly by activated T cells andNK cells and has complex effects on immune and nonimmune cells.IFN-g plays important roles in inflammation, making it particularlyrelevant to transplantation, with diverse and potentially contradictoryeffects on organ allograft survival (11). IFN-g coordinates a diversearray of cellular programs through the transcriptional regulation ofhundreds of genes (12). In the transplanted kidney, IFN-g protectsepithelial cells from necrosis (13), regulatesMHC expression and Agpresentation by epithelial and endothelial cells (14), and inducestryptophan (Trp) metabolism (15). Emerging evidence suggests thatIFN-g signaling and autophagy interact together (16).Macroautophagy (autophagy) is a major protective mechanism

that allows cells to survive in response to multiple stressors and

helps defend organisms against degenerative, inflammatory, in-fectious, and neoplastic diseases (17). A primordial function ofautophagy is the lysosomal degradation of cytoplasmic componentsin response to nutrient shortage. Autophagy also exerts numerouseffects on the control of immunity and inflammation: autophagy istriggered by immune signaling molecules, and autophagy proteinsnegatively regulate innate immunity functions such as inflamma-some activation, inflammatory cytokine production, and gene tran-scription. Autophagy proteins also enhance adaptive immunity,such as the development and homeostasis of the immune systemand Ag presentation (18).Autophagy and IFN-g interfere reciprocally. IFN-g promotes

autophagy in immune and nonimmune mice cells by mechanismsthat involve Beclin-1 expression (19, 20), and autophagy is re-quired for IFN-g–mediated antimicrobial efficacy, a process thatinvolves immunity-related guanosine triphosphatases (21, 22). Themechanism by which IFN-g functions to activate autophagy inhuman cells is not well understood, and whether IFN-g triggersautophagy in the kidney is currently unknown. In this study, weprovide evidence suggesting that autophagy is activated in humanrenal epithelial cells (HRECs) in response to IFN-g. Mechanis-tically, IFN-g induces Trp metabolism, which then activates thegeneral control nonderepressible-2 (GCN2) kinase, leading to thephosphorylation of the eukaryotic translation initiation factor 2a(eIF2a), an activator of autophagy. Conversely, Trp supplemen-tation reduces the activation of the GCN2–eIF2a pathway andinhibits autophagy. Further, targeting of GCN2 expression by RNAinterference also inhibits IFN-g–induced autophagy. The cellularimpact of autophagy in response to IFN-g is due to its capacity todampen the IFN-g–induced cytokine secretion by HRECs.

Materials and MethodsCell culture

Normal HRECs were harvested from human nephrectomy specimens re-moved for renal cell carcinoma and were isolated according to previouslypublished methods, with minor modifications (23, 24). Fragments ofnontumoral renal cortex were minced and digested with collagenase IV(250 IU/ml) for 3 h at 37˚C. Cells were centrifuged, and the pellets werewashed three times with PBS. Cells were then cultured in DMEM con-

*INSERM U775, Paris, France; †Universite Paris Descartes, Paris Sorbonne Cite,Paris, France; ‡Unite Mixte de Recherche 8601, Centre National de la RechercheScientifique, Paris, France; xPole de Biologie, Hopital Europeen Georges Pompidou,Paris, France; and {Service de Nephrologie, Hopital Europeen Georges Pompidou,Paris, France

Received for publication April 26, 2012. Accepted for publication July 16, 2012.

This work was supported by a grant from INSERM.

Address correspondence and reprint requests to Dr. Nicolas Pallet, INSERM U775,Centre Universitaire des Saints Peres, 45 Rue des Saints-Peres, 75006 Paris, France.E-mail address: [email protected]

Abbreviations used in this article: AO, acridine orange; eIF2a, eukaryotic translationinitiation factor 2a; ER, endoplasmic reticulum; GCN2, general control nonderepres-sible-2; HREC, human renal epithelial cell; NMR, nuclear magnetic resonance;PDGFB, platelet derived growth factor B; RPL13A, ribosomal protein L13A; RT-qPCR, real-time quantitative PCR; siRNA, small interfering RNA; TOCSY, totalcorrelation spectroscopy; Trp, tryptophan.

Copyright� 2012 by TheAmericanAssociation of Immunologists, Inc. 0022-1767/12/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1201214

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

taining 5 mg/ml insulin, 10 mg/ml human apotransferrin, 500 ng/ml hy-drocortisone, 10 ng/ml EGF, 6.5 ng/ml triiodothyronine, 5 ng/ml sodiumselenite, 1% FCS, 25 IU/ml penicillin, 25 mg/ml streptomycin, and10 mmol/L HEPES buffer. Cells were then incubated at 37˚C in 5% CO2

and 95% air. The characterization of this cellular model was performedby immunocytochemical peroxidase analysis and flow cytometry anal-ysis (data not shown). Immunocytochemical analysis revealed a positivestaining for cytokeratin (clone AE1/AE3) and a6 integrin (clone NKI-GoH3), two markers of epithelial cells, by the vast majority of cells, thusconfirming their epithelial nature. Cytometry analysis showed the ab-sence of staining for CD90 (clone 5E10), which is present on fibroblasts,but not on tubular epithelial cells. These results confirmed the proximaldescent of the vast majority of the cultured tubular epithelial cells.Experiments were not performed with cells beyond the third passagebecause it has been shown that no phenotypic changes occur up to thispassage number.

We have tested various concentrations of IFN-g to define the workingconcentration of 10 ng/ml (see Fig. 1E), which induced autophagy ina more reproducible manner than 1 ng/ml, and we have tested variousincubation times, which led us to demonstrate that after 48 h of incubation,the LC3II signal was maximal.

Small interfering RNA transfections

BECN1, GCN2, and scramble (control) synthetic small interfering RNAs(siRNAs) were designed and obtained from Qiagen. Transfectionwas performed using HiPerFect (Qiagen) following the manufacturer’sprotocol.

Electron microscopy

For electron microscopy, samples were fixed in 2% glutaraldehyde–0.1 Msodium cacodylate, postfixed in 1% OsO4, dehydrated in alcohol, pro-cessed for flat embedding in Epon 812, and observed with the Zeiss CEM902 electron microscope.

RNA extraction and real-time quantitative PCR

Total RNAwas extracted using the RNeasy Mini Kit (Qiagen) following themanufacturer’s protocol. The yield and purity of the RNA were measuredusing a NanoDrop ND-1000 spectrophotometer (Nanodrop Technologies).Transcript expression levels were quantified by SYBR Green real-timequantitative PCR (RT-qPCR) using an ABI PRISM 7900 sequence detectorsystem (Applied Biosystems). Vehicle-treated samples were used as thecontrols, and fold changes for each tested gene were normalized to the ri-bosomal protein L13A (RPL13A) housekeeping gene. The relative expres-sion levels were calculated using the 2(2ΔΔC(T)) (threshold cycle number)method (25). For RT-qPCR, the following primers were used: Beclin-1 for-ward (59-AGGTTGAGAAAGGCGAGACA-39), Beclin-1 reverse (59-AA-TTGTGAGGACACCCAAGC-39); CHOP forward (59-TGGAAGCCTGG-TATGAGGAC-39), CHOP reverse (59-TGTGACCTCTGCTGGTTCTG-39);GRP78 forward (59-GGTGAAAGACCCCTGACAAA-39), GRP78 reverse(59-GTCAGGCGATTCTGGTCATT-39); ICAM-1 forward (59-GAGATCA-CCATGGAGCCAAT-39), ICAM-1 reverse (59-CTGACAAGTTGTGGGG-GAGT-39); IDO-1 forward (59-GGCACACGCTATGGAAAACT-39), IDO-1reverse (59-CGCTGTGACTTGTGGTCTGT-39); IP-10 forward (59-CCA-CGTGTTGAGATCATTGC-39), IP-10 reverse (59-CCTCTGTGTGGTCCA-TCCTT-39); PDGFB forward (59-CCGCCAGCGCCCATTTTTCA-39), PD-GFB reverse (59-CTTTGCAGCGAGGCTGGAGGG-39); RANTES forward(59-GCTGCAGTGAGCTGAGATTG-39), RANTES reverse (59-GCCAG-TAAGCTCCTGTGAGG-39); RPL13A forward (59-CCTGGAGGAGAAG-AGGAAAGAGA-39), RPL13A reverse (59-GAGGACCTCTGTGTATTTG-TCAA-39); TNFa forward (59-TCCTTCAGACACCCTCAACC-39), TNFareverse (59-CAGGGATCAAAGCTGTAGGC-39); sXBP1 forward (59-GCA-GGTGCAGGCCCAGTTGT-39), sXBP1 reverse (59-TGGGTCCAAGTTG-TCCAGAATGC-39).

Protein extraction and Western blot analysis

Total protein lysate from HRECs was separated by SDS-PAGE under de-naturing conditions and transferred to a polyvinylidene fluoride membrane(GE Healthcare). Primary Abs were visualized using HRP-conjugated poly-clonal secondary Abs (Dako) and detected by ECL reagent (GE Health-care). The following Abs were used: anti-LC3B (1:1000, no. 2775),anti-eIF2a (1:1000, no. 9722), anti–phospho-4E-BP1 (1:1000, no. 9459),anti–4E-BP1 (1:1000, no. 9452), anti–phospho-p70S6K (1:1000, no.9204), anti–p70S6K (1:1000, no. 9202) were from Cell Signaling Tech-nology; anti-p62/SQSTM1 (1:500, no. sc-28359) was from Santa CruzBiotechnology; anti–Beclin-1 (1:1000, no. NB5000-249) was from NovusBiologicals; anti–phospho-eIF2a (1:500, no. ab32157), anti-ATF4 (1:500,

no. ab50546) were from Abcam; anti–phospho-GCN2 (1:1000, no. AJ1318a),anti-GCN2 (1:500, no. AP7130a) were from Abgent; and anti–b-actin(1:1000, no. A2668) was from Sigma-Aldrich. Immunoblots were quan-tified using the ImageJ 1.44 software (http://imagej.nih.gov/ij). We useda graphical method that involves generating lane profile plots, drawing linesto enclose peaks of interest, and then measuring peak areas (i.e., definiteintegrals). After background subtraction, a rectangular selection (region ofinterest) is made to enclose the first band, and the rectangular region ofinterest is moved over the adjacent lanes. Next, lane profiles plots aregenerated, and the area measurements are obtained and compared betweenconditions after having indexed them to the loading control.

Immunofluorescence microscopy

HRECs were cultured on glass coverslips and fixed with 4% paraformal-dehyde, rinsed with PBS, and blocked with 50 mM NH4Cl. Cells werepermeabilized with Triton X-100 and incubated with primary Abs (anti-LC3B [1:20, no. 2775] was from Cell Signaling Technology; anti–HLA-DM [1:50, no. sc-32248] was from Santa Cruz Biotechnology). Sampleswere then incubated with a cyanine 3- or FITC-coupled secondary Ab(Jackson ImmunoResearch). Acridine orange (Sigma) was diluted in waterand incubated in cell culture medium at a final concentration of 2.5 ng/ml2 h before the end of the experiment. Lysotracker Probe (Invitrogen) wasdirectly incorporated in cell culture medium from the beginning of theexperiment at a 1 mM final concentration. Slides were mounted and viewedusing a Nikon Eclipse TE 2000E imaging fluorescence microscope.

Nuclear magnetic resonance spectroscopy analysis of culturesupernatants

The nuclear magnetic resonance (NMR) experiments were run at 500.13MHz for 1H on a Bruker AVANCE 500 spectrometer with a 5-mm 1H/13C/15NTXI probe equipped with a z-gradient axis. The spectra of the differentculture media of the epithelial cells were measured after the addition of 5%(v/v) 2H2O in standard 5-mm sample tubes. All experiments were per-formed at 300 K using excitation sculpting water suppression (26) toeliminate solvent signal in an H2O/

2H2O 95:5 solution. One-dimensionalspectra were measured with 512 scans. Two-dimensional NMR spectrawere detected in the phase-sensitive mode using the States–TPPI method(27). The two-dimensional correlated spectroscopy and the total correla-tion spectroscopy (TOCSY) spectra were recorded with 128 scans and 128points in the indirect dimension. TOCSY spectra used an MLEV-17 spin-lock sequence (28) with a mixing time (tm) of 70 ms. The heteronuclearspectra 1H-13C HSQC were recorded with 512 scans and 200 increments inthe F1 dimension using natural 13C abundance. Chemical shift assignmentsreferred to internal 3-(trimethylsilyl) propionic acid-2,2,3,3-d4, sodium salt.The signals of the compounds of interest were assigned and compared witha sample of pure kynurenine and Trp measured under the same conditions.

Cytokine arrays

Multiple cytokine expression levels from conditioned media were simul-taneously assayed by the protein array RayBio Human Cytokine Ab Array(RayBiotech). Subconfluent cells were grown in 6-well plates. Cytokineexpression was evaluated in the cell culture supernatant using the RayBioHuman Inflammation Ab Array 3 (AAH-IFN-3) according to the manu-facturer’s protocol. The signal intensities were quantified by densitometryafter background subtraction, and positive controls were used to normalizethe results from the different membranes being compared. Evaluations ofthe relative cytokine expression levels were made by comparing the signalintensities between the different conditions. The following ratios of ex-pression were measured: 1) intensity in IFN-g–treated cells transfectedwith control siRNA/intensity in vehicle-treated cells transfected withcontrol siRNA (IFNg-control siRNA); and 2) intensity in IFN-g–treatedcells transfected with BECN1 siRNA/intensity in vehicle-treated cellstransfected with BECN1 siRNA (IFNg-BECN1 siRNA).

Statistical analysis

All data are expressed as the means 6 SEM of three different experimentsunless otherwise specified. Biological and histological data were comparedusing Student test. Statistical analyses were performed using GraphPadPrism software. Calculated p values ,0.05 were considered significant.

ResultsIFN-g activates autophagy in HRECs

HRECs that are exposed to IFN-g for 48 h accumulate cytoplasmicvacuoles (Fig. 1A). To characterize the nature of these vacuoles,

The Journal of Immunology 2955

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

FIGURE 1. IFN-g activates autophagy in HRECs. (A) IFN-g promotes the cytoplasmic accumulation of vacuoles. HRECs were incubated with 10 ng/ml

IFN-g for 48 h or were left untreated and were analyzed by phase contrast microscopy. A representative photomicrograph of three independent experiments

is shown. Scale bar, 20 mm. (B) IFN-g promotes the cytoplasmic accumulation of acidic vacuoles, which may be at least partly of a lysosomal nature.

HRECs were incubated with 10 ng/ml IFN-g for 48 h or were left untreated. Cells were then stained with 2.5 ng/ml AO and analyzed by confocal mi-

croscopy (top) or were stained with 1 mM Lysotracker and analyzed by epifluorescence microscopy (bottom). Scale bar, 20 mm. A representative image of

each staining of three independent experiments is shown. *p , 0.05. (C) IFN-g promotes the cytoplasmic accumulation of LC3-positive puncta. HRECs

were incubated with 10 ng/ml IFN-g for 48 h or were left untreated and then stained with Abs to LC3. Cells were analyzed by epifluorescence microscopy.

Scale bar, 10 mm. A representative image of three independent experiments is shown. *p , 0.05. (D) IFN-g induces HLA-DM expression, and LC3 and

HLA-DM colocalize in IFN-g–treated HRECs. HRECs were transfected with siRNAs targeting BECN1 transcripts or control nontargeted (scramble)

siRNAs. Twenty-four hours after transfection, HRECs were incubated with 10 ng/ml IFN-g for 48 h or were left untreated and then stained with Abs to

LC3, HLA-DM, and DAPI. The cells were analyzed by confocal microscopy. Scale bar, 10 mm. Top, Representative images of three independent

experiments are shown. Bottom, The intensity of the green and red fluorescence according to the distance between the two intensities of the fluorescence is

represented. (E) IFN-g induces accumulation of LC3II in HRECs. HRECs were incubated with the indicated concentrations of (Figure legend continues)

2956 IFN-g–INDUCED AUTOPHAGY

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

we stained cells with acridine orange (AO). AO moves freelyacross biological membranes and accumulates in acidic compart-ments, such as lysosomes and autophagolysosomes, where it isvisualized as bright orange/red fluorescence. AO vital staining ofIFN-g–treated cells shows the cytoplasmic accumulation of acidicorganelles (Fig. 1B), which may be at least partly of a lysosomalnature because they also retained Lysotracker, a fluorescent dyethat specifically stains lysosomes (Fig. 1B). The Atg8/LC3 proteinassociates with the membranes of autophagic structures and isused to monitor autophagy through indirect immunofluorescencemicroscopy, in which autophagy is measured as an increase inpunctate LC3 (29). IFN-g–treated HRECs accumulated LC3 puncta,in contrast to vehicle-treated cells, suggesting that autophagosomesaccumulate in HREC cytoplasm under IFN-g exposure (Fig. 1C).Notably, the autophagosome marker LC3 colocalized with HLA-DM–positive compartments (Fig. 1D), suggesting that MHC class IIloading compartments can obtain input from autophagosomes inIFN-g–treated HRECs. IFN-g also increased the accumulation ofLC3II, the lipidated form of LC3, which is found specifically linkedwith autophagosome membranes. LC3 is initially synthesized in anunprocessed form, proLC3, which is converted into a form lackingamino acids from the C terminus, LC3I, and is converted into thephosphatidylethanolamine-conjugated form, LC3II. LC3II is theonly protein marker that is reliably associated with completedautophagosomes (Fig. 1E). Electron microscopy analysis confirmedthat autophagosomes, vesicles limited by two parallel membranebilayers separated by an electron-lucent cleft and that contain het-erogenous materials formed by cytosol and organelles, accumulatein the cytoplasm of HRECs exposed to IFN-g (Fig. 1F). Overall,these data suggest that IFN-g promotes autophagosome and auto-phagolysosome accumulation in HRECs. Notably, IFN-g did notpromote LC3II accumulation in HUVECs (data not shown), sug-gesting that the autophagic response to IFN-g is cell-specific.The cytoplasmic accumulation of autophagosomes can result

from either an increased autophagic flux (i.e., increased auto-phagosome production, accumulation, and destruction) or from theinhibition of the fusion between autophagosomes and lysosomes(29). Inhibition of lysosomal proteases by incubating HRECs withE64 and pepstatin led to an increased accumulation of LC3II inIFN-g–treated cells (Fig. 1G), suggesting that lysosomal functionsare intact under IFN-g exposure and that IFN-g does not inhibitthe fusion of autophagosomes with lysosomes or their degrada-tion. We also monitored p62/sequestosome 1 expression as a markerof increased autophagic flux (30). The p62 protein links unwantedcytoplasmic cargos to LC3 and targets them for degradation inautophagolysosomes, where they are degraded. Whereas p62 ex-pression increased when autophagy was inhibited by RNA inter-ference against BECN1, the gene encoding Beclin-1, a centralregulator of autophagy (Fig. 1H; see also Fig. 6A), its expressionwas reduced in IFN-g–treated HREC (Fig. 1H), suggesting thatIFN-g increases autophagic flux and p62 degradation. Overall, these

results demonstrate that the accumulation of autophagosomes inHRECs in response to IFN-g results from an increased autophagicflux.

IFN-g induces IDO expression and Trp metabolism

We next attempted to gain mechanistic insights into IFN-g–inducedautophagy in HRECs and reasoned that amino acid deprivation isa master inducer of autophagy (17) and that IFN-g promotes Trpdepletion after IDO production (31). IDO is the first and rate-limiting enzyme of Trp catabolism through the kynurenine path-way and causes the depletion of Trp. We confirmed that IFN-gpromotes strong transcriptional induction of the gene encodingIDO (Fig. 2A). NMR spectroscopy analysis of the culture mediaof HRECs collected 48 h after IFN-g exposure showed that theexpression of IDO is accompanied by the catabolism of Trp,which is characterized by Trp depletion and the production ofkynurenine metabolites (Fig. 2B). Thus, IFN-g promotes Trp de-pletion in HREC culture media.

IFN-g does not interfere with mTOR signaling

mTOR is a central regulator of autophagic flux and is inhibitedin response to amino acid deprivation (32, 33), which activatesautophagosome biogenesis (17). To test whether IFN-g inhibitsmTOR activity in our model, we analyzed the phosphorylationstatus of two downstream targets of mTOR, p70S6K and 4E-BP1,in response to IFN-g. Whereas the mTOR inhibitor rapamycininhibited p70S6K phosphorylation and enhanced 4E-BP1 phos-phorylation, a paradoxical effect observed after long exposures torapamycin (34) (Fig. 3), we did not observe any modification ofthe phosphorylation statuses of 4E-BP1 and p70S6K under IFN-gexposure compared with the vehicle, findings that suggest thatIFN-g does not interfere with mTOR signaling in HRECs.

IFN-g activates the eIF2a pathway

We next focused on other cellular nutrient status sensors that can beactivated in response to amino acid deprivation and can promoteautophagy. The eIF2a signaling pathway is a well-characterizedregulator of stress-induced translational control programs (35)activated during nutrient starvation (36, 37). The eIF2a signalingpathway is also involved in the regulation of autophagy (38). Inour model, IFN-g promoted eIF2a phosphorylation (Fig. 4A) andATF4 protein expression (Fig. 4B) and C/EBP homologous pro-tein (CHOP) transcription (Fig. 4C), suggesting that the eIF2apathway is activated in HRECs in response to IFN-g exposure.Because PERK, which is activated by endoplasmic reticulum (ER)stress, can activate the eIF2a pathway, we tested whether IFN-gcould promote ER stress in HRECs, as has been reported foroligodendrocytes (39). IFN-g induced neither the expression ofthe spliced form of X-box binding protein-1 (XBP-1) mRNA northe transcription of the chaperone glucose-related protein 78(GRP78); these results indicate that the unfolded protein response

IFN-g or 100 nM rapamycin (Rapa), a known autophagy inducer, for 48 h (left) or with 10 ng/ml IFN-g for the indicated periods (right), and then whole-

cell lysates were run on an SDS-PAGE gel. LC3I and LC3II protein expression was determined by anti-LC3 immunoblots. Actin blots show general protein

amounts. Representative immunoblots of three independent experiments are shown. (F) IFN-g induces the accumulation of autophagosomes. HRECs were

incubated with 10 ng/ml IFN-g for 48 h or were left untreated and then fixed and embedded in epon. Cells were analyzed by electron microscopy. Black

arrows denote autophagosomes. Scale bar, 250 nm. Images representative of three independent experiments are shown. (G) IFN-g activates autophagic flux.

HRECs were incubated with 5 mM E64 (cathepsin inhibitor) and 5 mg/ml pepstatin (acid proteases inhibitor) with or without 10 ng/ml IFN-g. Whole-cell

lysates were run on a 12% SDS-PAGE gel, and LC3I and LC3II protein expression was determined by anti-LC3 immunoblots. Actin blots show general

protein amounts. Left, A representative immunoblot of three independent experiments is shown. Right, The LC3II/b-actin ratio is presented as the mean 6SEM of three independent experiments. *p, 0.05. (H) IFN-g reduces p62 accumulation. HRECs were transfected with siRNA targeting BECN1 transcripts

or control nontargeting (scramble) siRNAs. Twenty-four hours posttransfection, HRECs were incubated with 10 ng/ml IFN-g for 48 h or were left un-

treated. Whole-cell lysates were run on an SDS-PAGE gel, and p62 protein expression was determined by anti-p62 immunoblots. Actin blots show general

protein amounts. A representative immunoblot of three independent experiments is shown.

The Journal of Immunology 2957

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

is not activated in our model (Fig. 4D). Overall, these findingssuggest that the eIF2a signaling pathway is activated in responseto IFN-g independently of ER stress.

Trp depletion mediates IFN-g–induced autophagy andactivates the eIF2a kinase GCN2

Given that IFN-g promotes Trp depletion and activates the eIF2apathway independently of ER stress, we tested whether Trp de-pletion could activate the GCN2 kinase, an eIF2a kinase that isactivated by uncharged tRNAs in amino acid-starved cells (40).GCN2 and eIF2a were phosphorylated in IFN-g–treated HRECs,

and this phosphorylation was reversed by Trp supplementation(Fig. 5A, 5B). Further, Trp supplementation reduced LC3II andLC3-positive puncta accumulation during IFN-g exposure (Fig.5C, 5D), which suggests that the accumulation of autophago-somes in IFN-g–treated HRECs depends on Trp availability andmight implicate the eIF2a kinase GCN2. Of note, L-kynurenine,the principal Trp metabolite product of IDO, did not activateautophagy (data not shown). To examine the role of GCN2 in theactivation of autophagy in response to IFN-g and Trp depletion,we inhibited GCN2 expression by siRNA-mediated RNA inter-ference (Fig. 5E). Inhibition of GCN2 expression significantly

FIGURE 2. IFN-g induces IDO expression and Trp depletion. (A) IFN-g induces IDO transcript expression. HRECs were incubated with 10 ng/ml IFN-g for

48 h. IDO transcript levels were measured by RT-qPCR and are presented as the mean 6 SEM relative to the levels in untreated cells for three independent

experiments. **p, 0.01. (B) Trp is metabolized during IFN-g treatment. HRECs were incubated with 10 ng/ml IFN-g for 48 h or were left untreated. Trp and

kynurenines were detected in the culture medium by NMR spectroscopy. Top, A representative spectrum is displayed. Bottom, Purified L-Trp and L-kynurenine

spectra are shown. The “IFNg-Ctrl” spectrum corresponds with the difference between IFN-g spectrum and Ctrl spectrum presented at the top.

2958 IFN-g–INDUCED AUTOPHAGY

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

reduced the accumulation of LC3II in response to IFN-g (Fig. 5F),suggesting that GCN2 regulates autophagy in HRECs in responseto IFN-g. Overall, our results demonstrate that the activation ofautophagy in response to IFN-g is promoted by Trp depletionand relies, at least in part, on the activation of the GCN2–eIF2apathway.

Autophagy interferes with IFN-g–associated secretoryphenotype

We next tested whether autophagy could interfere with relevantIFN-g–related immune functions in HRECs. Although they arenonprofessional immune cells, kidney epithelial cells can secretenumerous proinflammatory cytokines when the kidney tissue isinjured (1, 2), and cytokine release can amplify inflammation andfacilitate adaptive immunity. We generated an expression profile

of cytokines secreted by HRECs in response to IFN-g (the IFN-g–associated secretory phenotype) and compared this profile withthat produced when the expression of BECN1 is inhibited. siRNA-mediated RNA interference directed against BECN1 inhibitedBeclin-1 expression (Fig. 6A) and reduced LC3-positive punctaaccumulation in response to IFN-g (Fig. 6B). The inhibition ofBECN1 expression profoundly altered the IFN-g–associated se-cretory phenotype and resulted in the increased secretion ofproinflammatory and immunostimulatory mediators, including IP-10, MIP-1d, MCP-2, ICAM-1 (the secreted form), TNF-a, andprofibrogenic cytokines such as TGF-b1 and platelet-derivedgrowth factor B (PDGFB) (Fig. 6C). Conversely, BECN1 inhibi-tion reduced the secretion of some cytokines, such as sTNFR I andII, which act as TNF-a antagonists. This observation suggests thatthe activation of autophagy in HRECs in response to IFN-g couldresult in a reduction of the amplitude of the cytokinic secretoryphenotype. The inhibition of BECN1 expression did not alter IP10,TNFa, ICAM1 or PDGFB mRNA expression (Fig. 6D), whichsuggests that autophagy may regulate cytokine production at theposttranscriptional level.

DiscussionAccumulating evidence indicates that autophagy plays a criticalrole in kidney maintenance, disease, and aging. Ischemic, toxic,immunological, and oxidative insults can cause an induction ofautophagy in renal epithelial cells, modifying the course of variouskidney diseases (41–44). The effects of autophagy on the reg-ulation of innate and adaptive immunity and cell viability areparticularly relevant in transplanted tissue, which faces a greatnumber of stresses that challenge its viability and immunogenicity(9). In this study, we have identified a new mechanism by whichIFN-g promotes autophagy in the human kidney epithelium anddemonstrated that autophagy modulates IFN-g–induced immune

FIGURE 3. IFN-g does not interfere with mTOR signaling. HRECs

were incubated with 10 ng/ml IFN-g or 100 nM rapamycin or were left

untreated for 24 or 48 h. Whole-cell lysates were run on an SDS-PAGE gel,

and the protein expression levels of phospho-4E-BP1, 4E-BP1, phospho-

p70S6K, and p70S6K were evaluated by immunoblotting. A representative

immunoblot of two independent experiments is shown.

FIGURE 4. IFN-g activates the eIF2a pathway. (A)

IFN-g induces eIF2a phosphorylation. HRECs were

incubated with 10 ng/ml IFN-g for 24 or 48 h or were

left untreated. Whole-cell lysates were run on an SDS-

PAGE gel, and the protein expression levels of phos-

pho-eIF2a and eIF2a were evaluated by immuno-

blotting. Left, A representative immunoblot of three

independent experiments is shown. Right, Densitometric

analysis of phospho-eIF2a/eIF2a of three independent

immunoblots. *p , 0.05. (B) IFN-g increases ATF4

expression. HRECs were incubated with 10 ng/ml IFN-g

for 24 or 48 h or were left untreated. Whole-cell lysates

were run on an SDS-PAGE gel, and ATF4 expression

was evaluated by anti-ATF4 immunoblotting. A repre-

sentative immunoblot of three independent experiments

is shown. (C) IFN-g induces CHOP mRNA expression.

HRECs were incubated with 10 ng/ml IFN-g for 24 or

48 h or were left untreated. CHOP transcript levels

were measured by RT-qPCR and are presented as the

mean 6 SEM relative to the transcript levels in un-

treated cells for three independent experiments.

*p , 0.05. (D) IFN-g does not promote XBP-1 splicing

and GRP78 mRNA expression induction. HRECs were

incubated with 10 ng/ml IFN-g for 24 and 48 h or with 2

mg/ml tunicamycin (“Tunica”; a protein glycosylation

inhibitor known to induce ER stress) for 24 h or were

left untreated. Spliced XBP1 and GRP78 transcript

levels were measured by qRT-PCR and are presented as

the means6 SEM relative to levels in untreated cells for

three independent experiments.

The Journal of Immunology 2959

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

FIGURE 5. Trp depletion mediates IFN-g–induced autophagy and activates the eIF2a kinase GCN2 . (A) Trp supplementation reduces GCN2 phos-

phorylation. HRECs were incubated with 10 ng/ml IFN-g alone or with 80 mg/l Trp for 48 h or were left untreated. Whole-cell lysates were run on an SDS-

PAGE gel, and the protein expression levels of phospho-GCN2 and GCN2 were evaluated by immunoblotting. Left, An immunoblot representative of three

independent experiments is shown. Right, Densitometric analysis of phospho-GCN2/GCN2 of three independent immunoblots. *p , 0.05. (B) Trp sup-

plementation reduces eIF2a phosphorylation. HRECs were incubated with 10 ng/ml IFN-g alone or with 80 mg/l Trp for 48 h or were left untreated.

Whole-cell lysates were run on an SDS-PAGE gel, and the protein expression levels of phospho-eIF2a and eIF2a were evaluated by immunoblotting. An

immunoblot representative of three independent experiments is shown. The phospho-eIF2a/eIF2a ratio is presented as the mean 6 SEM of three inde-

pendent experiments. *p , 0.05. (C) Trp supplementation reduces LC3II accumulation. HRECs were incubated with 10 ng/ml IFN-g alone or with 80 mg/l

Trp for 48 h or were left untreated. Whole-cell lysates were run on an SDS-PAGE gel, and the LC3I and LC3II protein expression levels were determined by

anti-LC3 immunoblots. Actin blots show general protein amounts. Left, A representative immunoblot of three independent experiments is shown. Right,

The LC3II/b-actin ratio is presented as the mean 6 SEM of three independent experiments. *p , 0.05. (D) Trp supplementation reduces LC3-positive

puncta accumulation. HRECs were incubated with 10 ng/ml IFN-g alone or with 80 mg/l Trp for 48 h or were left untreated and then were stained with Abs

to LC3. Cells were analyzed by epifluorescence microscopy. Scale bar, 10 mm. Images representative of three independent (Figure legend continues)

2960 IFN-g–INDUCED AUTOPHAGY

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

responses in HRECs. We provide evidence that the activation ofautophagy by IFN-g in human epithelial cells relies on Trp de-pletion and the activation of the GCN2–eIF2a pathway. Ourresults also show that autophagy dampens the IFN-g–inducedsecretion of cytokines by HRECs.Our results do not exclude that other mechanisms might be

involved in the regulation of autophagy by IFN-g. In mice, oneimmunity-related GTPase protein in particular, Irgm1, has beenshown to exert critical functions in IFN-g–induced autophagy.However, the human ortholog of Irgm1, IRGM, is not elicited byIFN-g. Data regarding the regulation of autophagy in human cellsare very scarce. A very recent study demonstrated that IFN-g in-duces autophagy and does so independently of Irgm1 describedearlier. This novel pathway used JAK1/2 and p38 MAPK signalingbut does not require STAT1 (45). An in vivo model of IFN-g–mediated inflammatory renal injury, like allograft rejection, inautophagy defective mice would be helpful to understand betterthe interplay between IFN-g and autophagy.The eIF2a-signaling pathway is an important regulator of

autophagy (17). The mechanisms by which eIF2a activatesautophagy involve the isolation of ER membrane and the induc-tion of the expression of the transcription factors ATF4 and CHOP,which in turn promote the expression of the autophagy genes LC3and ATG5 (46, 47). There are four eIF2a kinases in mammals:PERK; PKR, which is activated by dsRNA during viral infection;HRI, which is activated by iron deficiency in erythrocytes (48);and GCN2. IFN-g can regulate the expression or the activity ofPERK, PKR, and GCN2. IFN-g induces ER stress and activatesPERK through ill-defined mechanisms that involve the increasedsynthesis of membrane-spanning proteins (39). PKR expressionis induced by IFN-g, but in an inactive form, and requires viraldsRNA to be activated (12). GCN2 is activated when the level ofany amino acid, including Trp, diminishes sufficiently to cause theaccumulation of uncharged tRNAs, which are direct activators ofthe kinase (37, 40). Because ER stress occurs in the transplantedkidney (49), there is little doubt that the activation of PERK, inaddition to GCN2, also promotes autophagy in the injured allo-graft (44).The fact that Trp depletion does not interfere with mTOR

signaling suggests that mTOR and GCN2 are nutrient availabilitysensors with different levels of sensitivity and/or specificity.Whereas intracellular amino acid depletion is sensed directly byGCN2, which binds uncharged tRNAs and can thus theoreticallydetect deficiencies in any essential amino acid or nonessentialamino acid (50), the regulation of mTORC1 is most responsiveto specific individual amino acids such as leucine and arginine(32), which diminish the capacity of Rheb to bind and activatemTOR (33). However, other factors related to kidney trans-plantation can interfere with mTOR signaling, including theuse of the immunosuppressive drug rapamycin, and one cannotexclude the possibility that the inhibition of mTOR signalingcould be an additive process that contributes to macroautophagyin the transplanted kidney. Therefore, the respective contributionsof the various inducers of macroautophagy occurring during

transplant damage, which depend on the nature, the timing ofoccurrence, and the intensity of the injury, remain to be charac-terized.Our findings underscore the importance of amino acid avail-

ability in the regulation of immune functions. The immunoreg-ulatory functions of the local metabolism of Trp, and to a lesserextent L-arginine, are well known (51). Trp depletion is tolero-genic, as it inhibits Th cell proliferation and promotes regulatoryT cell amplification, and kynurenines, the Trp metabolites, pro-mote T cell death. IDO-mediated Trp depletion can activateGCN2 in T cells, leading to proliferative arrest, anergy, andregulatory T cell production (52, 53). The fact that IFN-g–inducedTrp depletion promotes autophagy expands the spectrum of theimmunoregulatory properties of IDO-induced Trp depletion. In-deed, autophagy proteins function in the inactivation of immunesignaling by negatively regulating inflammasome activation, NF-kB signaling, and inflammatory cytokine production (54). In linewith our findings showing that autophagy dampens IFN-g–inducedinflammatory cytokine secretion, mice kidneys in which ATG5 hasbeen selectively knocked out in the tubular epithelium developa more severe inflammation in response to ischemia–reperfusioninjury than their wild-type counterparts (55). A particularly im-portant step will be to reconcile the apparently opposite con-sequences of autophagy and Trp depletion on innate immunity andadaptive immunity, as autophagy also facilitates adaptive functionsincluding Ag presentation, thymic education, and lymphocyte pro-liferation (18).p62 is an important signaling adapter protein that interacts

with TNF receptor-associated factor 6 and promotes NF-kB acti-vation (30). p62 is targeted to be degraded by autophagy duringstressful conditions (56). Consequently, autophagy defects resultin NF-kB induction. In other words, autophagy is required to sup-press p62 accumulation and inappropriate activation of NF-kB.This corroborates with our findings, which demonstrate that inconditions of impaired autophagy, the secretion of cytokines inresponse to IFN-g is higher than in condition of nonaltered auto-phagy, which indicates that the degradation of p62 by IFN-g–in-duced autophagy could reduce NF-kB signaling and dampensecretion of cytokines. Conversely, when autophagy is altered (e.g.,when Beclin-1 is inhibited by RNA interference), NF-kB could beactivated and the secretion of cytokines increased.Recent results indicate a novel role for autophagy in noncon-

ventional secretion pathways of molecules, including cytokines(57–60), which does not involve lysosomal degradation of auto-phagosomal contents but instead involves their redirection towardthe extracellular delivery. Autophagy is involved in mountingintercellular communication networks, which could be of broadimmunological importance (57, 60). For example, autophagy isessential for the immunogenic release of ATP from dying cells(60), which will activate purinergic P2RX7 receptors and willpromote the production of IL-1b by dendritic cells. Conversely,our results indicate that autophagy negatively regulates inflamma-tory cytokine secretion in response to IFN-g, which corroborateswith previous studies demonstrating that autophagy reduces IL-

experiments are shown. *p , 0.05. (E) Inhibition of GCN2 expression by siRNA-mediated RNA interference. HRECs were transfected with siRNA

targeting GCN2 transcripts or control nontargeting (scramble) siRNAs. Twenty-four hours posttransfection, the GCN2 and phospho-GCN2 protein levels

were measured by anti-GCN2 and anti–phospho-GCN2 immunoblots. A representative immunoblot of three independent experiments is shown. (F) The

inhibition of GCN2 expression reduces LC3II accumulation. HRECs were transfected with siRNAs targeting GCN2 or control, nontargeted (scramble)

siRNAs. Twenty-four hours after transfection, HRECs were incubated with 10 ng/ml IFN-g for 48 h or were left untreated. Whole-cell lysates were run on

an SDS-PAGE gel, and LC3I and LC3II protein expression levels were determined by anti-LC3 immunoblots. Actin blots show general protein amounts.

Left, A representative immunoblot of three independent experiments is shown. Right, The LC3II/b-actin ratio is presented as the mean 6 SEM of three

independent experiments. *p , 0.05.

The Journal of Immunology 2961

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

1b production, as a consequence by a mechanism that involvesthe destruction of inflammasomes and altered mitochondria(61, 62).In conclusion, we have identified a new mechanism by which

IFN-g, a master regulator of kidney allograft injury, activatesautophagy in kidney epithelial cells, and our study provides newinsights into how autophagy regulates immune functions in re-sponse to IFN-g. It might be useful to investigate the potential

of therapeutic targeting of autophagic processes to modulate theevolution of active kidney allograft deterioration.

AcknowledgmentsWe thank Dr. Jean-Pierre Denizot (Unite de Neurosciences, Information

et Complexite, Centre National de la Recherche Scientifique, Gif-sur-

Yvette, France) for technical assistance in performing the electron micros-

copy study.

FIGURE 6. Autophagy interferes with the IFN-g–associated secretory phenotype. (A) Inhibition of BECN1 expression by siRNA-mediated RNA in-

terference. HRECs were transfected with siRNA targeting BECN1 transcripts or control nontargeting (scramble) siRNAs. Twenty-four hours post-

transfection, the Beclin-1 protein level was measured by anti–Beclin-1 immunoblots. A representative immunoblot of three independent experiments is

shown. (B) Inhibition of BECN1 expression reduces the cytoplasmic accumulation of LC3-positive puncta. HRECs were transfected with siRNA targeting

BECN1 transcripts or control nontargeting (scramble) siRNAs. Twenty-four hours after transfection, HRECs were incubated with 10 ng/ml IFN-g for 48 h

or were left untreated and then were stained with Abs to LC3. Cells were analyzed by epifluorescence microscopy. Scale bar, 5 mm. Images representative

of three independent experiments are shown. *p, 0.05. (C) BECN1 expression inhibition modifies the IFN-g–associated secretory phenotype. HRECs were

transfected with siRNAs targeting BECN1 or control, nontargeted (scramble) siRNAs. Twenty-four hours after transfection, HRECs were incubated with 10

ng/ml IFN-g for 48 h or were left untreated. Culture media were analyzed by Human Cytokine Ab Arrays. The cytokines for which expression levels were

detectable compared with negative controls are shown. For each protein, the ratio between expression levels in IFN-g–treated and vehicle-treated cell was

calculated. The baseline represents the averaged signals from vehicle-treated conditions. Signals above the baseline are yellow; signals below the baseline

are blue. The heat map key shows fold changes from baseline. One of two representative experiments is shown. *p , 0.05. (D) BECN1 expression in-

hibition does not modify cytokine transcript expression. HRECs were transfected with siRNAs targeting BECN1 or control, nontargeted (scramble) siRNAs.

Twenty-four hours after transfection, HRECs were incubated with 10 ng/ml IFN-g for 48 h or were left untreated. IP10, ICAM1, PDGFB, and TNFa

transcript levels were measured by RT-qPCR and are presented as the means 6 SEM relative to the levels in untreated cells for three independent

experiments.

2962 IFN-g–INDUCED AUTOPHAGY

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

DisclosuresThe authors have no financial conflicts of interest.

References1. Bonventre, J. V., and A. Zuk. 2004. Ischemic acute renal failure: an inflamma-

tory disease? Kidney Int. 66: 480–485.2. Bonventre, J. V. 2003. Molecular response to cytotoxic injury: role of inflam-

mation, MAP kinases, and endoplasmic reticulum stress response. Semin.Nephrol. 23: 439–448.

3. Eddy, A. A., and E. G. Neilson. 2006. Chronic kidney disease progression. J. Am.Soc. Nephrol. 17: 2964–2966.

4. Leemans, J. C., G. Stokman, N. Claessen, K. M. Rouschop, G. J. Teske,C. J. Kirschning, S. Akira, T. van der Poll, J. J. Weening, and S. Florquin. 2005.Renal-associated TLR2 mediates ischemia/reperfusion injury in the kidney. J.Clin. Invest. 115: 2894–2903.

5. Wu, H., G. Chen, K. R. Wyburn, J. Yin, P. Bertolino, J. M. Eris, S. I. Alexander,A. F. Sharland, and S. J. Chadban. 2007. TLR4 activation mediates kidneyischemia/reperfusion injury. J. Clin. Invest. 117: 2847–2859.

6. Gago, M., L. D. Cornell, W. K. Kremers, M. D. Stegall, and F. G. Cosio. 2012.Kidney allograft inflammation and fibrosis, causes and consequences. Am. J.Transplant. 12: 1199–1207.

7. Sellares, J., D. G. de Freitas, M. Mengel, B. Sis, L. G. Hidalgo, A. J. Matas,B. Kaplan, and P. F. Halloran. 2011. Inflammation lesions in kidney transplantbiopsies: association with survival is due to the underlying diseases. Am. J.Transplant. 11: 489–499.

8. Mengel, M., J. Chang, D. Kayser, W. Gwinner, A. Schwarz, G. Einecke,V. Broecker, K. Famulski, D. G. de Freitas, L. Guembes-Hidalgo, et al. 2011. Themolecular phenotype of 6-week protocol biopsies from human renal allografts:reflections of prior injury but not future course. Am. J. Transplant. 11: 708–718.

9. Nankivell, B. J., and D. R. Kuypers. 2011. Diagnosis and prevention of chronickidney allograft loss. Lancet 378: 1428–1437.

10. Famulski, K. S., G. Einecke, B. Sis, M. Mengel, L. G. Hidalgo, B. Kaplan, andP. F. Halloran. 2010. Defining the canonical form of T-cell-mediated rejection inhuman kidney transplants. Am. J. Transplant. 10: 810–820.

11. Hidalgo, L. G., and P. F. Halloran. 2002. Role of IFN-gamma in allograft re-jection. Crit. Rev. Immunol. 22: 317–349.

12. Schroder, K., P. J. Hertzog, T. Ravasi, and D. A. Hume. 2004. Interferon-gamma:an overview of signals, mechanisms and functions. J. Leukoc. Biol. 75: 163–189.

13. Afrouzian, M., V. Ramassar, J. Urmson, L. F. Zhu, and P. F. Halloran. 2002.Transcription factor IRF-1 in kidney transplants mediates resistance to graftnecrosis during rejection. J. Am. Soc. Nephrol. 13: 1199–1209.

14. Takei, Y., T. N. Sims, J. Urmson, and P. F. Halloran. 2000. Central role forinterferon-gamma receptor in the regulation of renal MHC expression. J. Am.Soc. Nephrol. 11: 250–261.

15. Mohib, K., S. Wang, Q. Guan, A. L. Mellor, H. Sun, C. Du, and A. M. Jevnikar.2008. Indoleamine 2,3-dioxygenase expression promotes renal ischemia-reperfusion injury. Am. J. Physiol. Renal Physiol. 295: F226–F234.

16. Harris, J. 2011. Autophagy and cytokines. Cytokine 56: 140–144.17. Kroemer, G., G. Marino, and B. Levine. 2010. Autophagy and the integrated

stress response. Mol. Cell 40: 280–293.18. Levine, B., N. Mizushima, and H. W. Virgin. 2011. Autophagy in immunity and

inflammation. Nature 469: 323–335.19. Shi, C. S., and J. H. Kehrl. 2010. TRAF6 and A20 regulate lysine 63-linked

ubiquitination of Beclin-1 to control TLR4-induced autophagy. Sci. Signal. 3: ra42.20. Tu, S. P., M. Quante, G. Bhagat, S. Takaishi, G. Cui, X. D. Yang, S. Muthuplani,

W. Shibata, J. G. Fox, D. M. Pritchard, and T. C. Wang. 2011. IFN-g inhibitsgastric carcinogenesis by inducing epithelial cell autophagy and T-cell apoptosis.Cancer Res. 71: 4247–4259.

21. Singh, S. B., A. S. Davis, G. A. Taylor, and V. Deretic. 2006. Human IRGM inducesautophagy to eliminate intracellular mycobacteria. Science 313: 1438–1441.

22. Kim, B. H., A. R. Shenoy, P. Kumar, R. Das, S. Tiwari, and J. D. MacMicking.2011. A family of IFN-g-inducible 65-kD GTPases protects against bacterialinfection. Science 332: 717–721.

23. Tang, S., J. C. Leung, K. Abe, K. W. Chan, L. Y. Chan, T. M. Chan, andK. N. Lai. 2003. Albumin stimulates interleukin-8 expression in proximal tubularepithelial cells in vitro and in vivo. J. Clin. Invest. 111: 515–527.

24. Detrisac, C. J., M. A. Sens, A. J. Garvin, S. S. Spicer, and D. A. Sens. 1984.Tissue culture of human kidney epithelial cells of proximal tubule origin. KidneyInt. 25: 383–390.

25. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expressiondata using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method.Methods 25: 402–408.

26. Nguyen, B. D., X. Meng, K. J. Donovan, and A. J. Shaka. 2007. SOGGY:solvent-optimized double gradient spectroscopy for water suppression. A com-parison with some existing techniques. J. Magn. Reson. 184: 263–274.

27. States, D., R. A. Haberkorn, and D. J. Ruben. 1982. A two-dimensional nuclearOverhauser experiment with pure absorption phase in four quadrants. J. Magn.Reson. 48: 286–292.

28. Bax, A., and D. G. Davis. 1985. MLEV-17-based two-dimensional homonuclearmagnetization transfer spectroscopy. J. Magn. Reson. 65: 355–360.

29. Klionsky, D. J., H. Abeliovich, P. Agostinis, D. K. Agrawal, G. Aliev,D. S. Askew, M. Baba, E. H. Baehrecke, B. A. Bahr, A. Ballabio, et al. 2008.Guidelines for the use and interpretation of assays for monitoring autophagy inhigher eukaryotes. Autophagy 4: 151–175.

30. Mathew, R., C. M. Karp, B. Beaudoin, N. Vuong, G. Chen, H. Y. Chen, K. Bray,A. Reddy, G. Bhanot, C. Gelinas, et al. 2009. Autophagy suppresses tumori-genesis through elimination of p62. Cell 137: 1062–1075.

31. Yasui, H., K. Takai, R. Yoshida, and O. Hayaishi. 1986. Interferon enhancestryptophan metabolism by inducing pulmonary indoleamine 2,3-dioxygenase: itspossible occurrence in cancer patients. Proc. Natl. Acad. Sci. USA 83: 6622–6626.

32. Hara, K., K. Yonezawa, Q. P. Weng, M. T. Kozlowski, C. Belham, and J. Avruch.1998. Amino acid sufficiency and mTOR regulate p70 S6 kinase and eIF-4E BP1through a common effector mechanism. J. Biol. Chem. 273: 14484–14494.

33. Long, X., S. Ortiz-Vega, Y. Lin, and J. Avruch. 2005. Rheb binding to mam-malian target of rapamycin (mTOR) is regulated by amino acid sufficiency. J.Biol. Chem. 280: 23433–23436.

34. Choo, A. Y., S. O. Yoon, S. G. Kim, P. P. Roux, and J. Blenis. 2008. Rapamycindifferentially inhibits S6Ks and 4E-BP1 to mediate cell-type-specific repressionof mRNA translation. Proc. Natl. Acad. Sci. USA 105: 17414–17419.

35. Harding, H. P., Y. Zhang, H. Zeng, I. Novoa, P. D. Lu, M. Calfon, N. Sadri,C. Yun, B. Popko, R. Paules, et al. 2003. An integrated stress response regulatesamino acid metabolism and resistance to oxidative stress.Mol. Cell 11: 619–633.

36. Bi, M., C. Naczki, M. Koritzinsky, D. Fels, J. Blais, N. Hu, H. Harding, I. Novoa,M. Varia, J. Raleigh, et al. 2005. ER stress-regulated translation increases toler-ance to extreme hypoxia and promotes tumor growth. EMBO J. 24: 3470–3481.

37. Ye, J., M. Kumanova, L. S. Hart, K. Sloane, H. Zhang, D. N. De Panis,E. Bobrovnikova-Marjon, J. A. Diehl, D. Ron, and C. Koumenis. 2010. TheGCN2-ATF4 pathway is critical for tumour cell survival and proliferation inresponse to nutrient deprivation. EMBO J. 29: 2082–2096.

38. Talloczy, Z., W. Jiang, H. W. Virgin, IV, D. A. Leib, D. Scheuner, R. J. Kaufman,E. L. Eskelinen, and B. Levine. 2002. Regulation of starvation- and virus-induced autophagy by the eIF2alpha kinase signaling pathway. Proc. Natl.Acad. Sci. USA 99: 190–195.

39. Lin, W., S. L. Bailey, H. Ho, H. P. Harding, D. Ron, S. D. Miller, and B. Popko.2007. The integrated stress response prevents demyelination by protecting oli-godendrocytes against immune-mediated damage. J. Clin. Invest. 117: 448–456.

40. Dong, J., H. Qiu, M. Garcia-Barrio, J. Anderson, and A. G. Hinnebusch. 2000.Uncharged tRNA activates GCN2 by displacing the protein kinase moiety froma bipartite tRNA-binding domain. Mol. Cell 6: 269–279.

41. Isaka, Y., T. Kimura, and Y. Takabatake. 2011. The protective role of autophagyagainst aging and acute ischemic injury in kidney proximal tubular cells.Autophagy 7: 1085–1087.

42. Koesters, R., B. Kaissling, M. Lehir, N. Picard, F. Theilig, R. Gebhardt,A. B. Glick, B. Hahnel, H. Hosser, H. J. Grone, and W. Kriz. 2010. Tubularoverexpression of transforming growth factor-beta1 induces autophagy and fi-brosis but not mesenchymal transition of renal epithelial cells. Am. J. Pathol.177: 632–643.

43. Hartleben, B., M. Godel, C. Meyer-Schwesinger, S. Liu, T. Ulrich, S. Kobler,T. Wiech, F. Grahammer, S. J. Arnold, M. T. Lindenmeyer, et al. 2010. Auto-phagy influences glomerular disease susceptibility and maintains podocyte ho-meostasis in aging mice. J. Clin. Invest. 120: 1084–1096.

44. Pallet, N., N. Bouvier, C. Legendre, J. Gilleron, P. Codogno, P. Beaune,E. Thervet, and D. Anglicheau. 2008. Autophagy protects renal tubular cellsagainst cyclosporine toxicity. Autophagy 4: 783–791.

45. Matsuzawa, T., B. H. Kim, A. R. Shenoy, S. Kamitani, M. Miyake, andJ. D. Macmicking. 2012. IFN-g elicits macrophage autophagy via the p38MAPK signaling pathway. J. Immunol. 189: 813–818.

46. Milani, M., T. Rzymski, H. R. Mellor, L. Pike, A. Bottini, D. Generali, andA. L. Harris. 2009. The role of ATF4 stabilization and autophagy in resistance ofbreast cancer cells treated with Bortezomib. Cancer Res. 69: 4415–4423.

47. Rouschop, K. M., T. van den Beucken, L. Dubois, H. Niessen, J. Bussink,K. Savelkouls, T. Keulers, H. Mujcic, W. Landuyt, J. W. Voncken, et al. 2010.The unfolded protein response protects human tumor cells during hypoxiathrough regulation of the autophagy genes MAP1LC3B and ATG5. J. Clin. In-vest. 120: 127–141.

48. Proud, C. G. 2005. eIF2 and the control of cell physiology. Semin. Cell Dev. Biol.16: 3–12.

49. Pallet, N., S. Fougeray, P. Beaune, C. Legendre, E. Thervet, and D. Anglicheau.2009. Endoplasmic reticulum stress: an unrecognized actor in solid organtransplantation. Transplantation 88: 605–613.

50. Hinnebusch, A. G. 2005. Translational regulation of GCN4 and the generalamino acid control of yeast. Annu. Rev. Microbiol. 59: 407–450.

51. Grohmann, U., and V. Bronte. 2010. Control of immune response by amino acidmetabolism. Immunol. Rev. 236: 243–264.

52. Munn, D. H., M. D. Sharma, B. Baban, H. P. Harding, Y. Zhang, D. Ron, andA. L. Mellor. 2005. GCN2 kinase in T cells mediates proliferative arrest and anergyinduction in response to indoleamine 2,3-dioxygenase. Immunity 22: 633–642.

53. Fallarino, F., U. Grohmann, S. You, B. C. McGrath, D. R. Cavener, C. Vacca,C. Orabona, R. Bianchi, M. L. Belladonna, C. Volpi, et al. 2006. The combinedeffects of tryptophan starvation and tryptophan catabolites down-regulate T cellreceptor zeta-chain and induce a regulatory phenotype in naive T cells. J.Immunol. 176: 6752–6761.

54. Saitoh, T., and S. Akira. 2010. Regulation of innate immune responses byautophagy-related proteins. J. Cell Biol. 189: 925–935.

55. Liu, S., B. Hartleben, O. Kretz, T. Wiech, P. Igarashi, N. Mizushima, G. Walz,and T. B. Huber. 2012. Autophagy plays a critical role in kidney tubule main-tenance, aging and ischemia-reperfusion injury. Autophagy 8. In press.

56. White, E. 2012. Deconvoluting the context-dependent role for autophagy incancer. Nat. Rev. Cancer. In press.

The Journal of Immunology 2963

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from

57. Dupont, N., S. Jiang, M. Pilli, W. Ornatowski, D. Bhattacharya, and V. Deretic.2011. Autophagy-based unconventional secretory pathway for extracellular de-livery of IL-1b. EMBO J. 30: 4701–4711.

58. Duran, J. M., C. Anjard, C. Stefan, W. F. Loomis, and V. Malhotra. 2010. Un-conventional secretion of Acb1 is mediated by autophagosomes. J. Cell Biol.188: 527–536.

59. Manjithaya, R., C. Anjard, W. F. Loomis, and S. Subramani. 2010. Unconven-tional secretion of Pichia pastoris Acb1 is dependent on GRASP protein, per-oxisomal functions, and autophagosome formation. J. Cell Biol. 188: 537–546.

60. Michaud, M., I. Martins, A. Q. Sukkurwala, S. Adjemian, Y. Ma, P. Pellegatti,S. Shen, O. Kepp, M. Scoazec, G. Mignot, et al. 2011. Autophagy-dependent

anticancer immune responses induced by chemotherapeutic agents in mice.Science 334: 1573–1577.

61. Nakahira, K., J. A. Haspel, V. A. Rathinam, S. J. Lee, T. Dolinay, H. C. Lam,J. A. Englert, M. Rabinovitch, M. Cernadas, H. P. Kim, et al. 2011. Autophagyproteins regulate innate immune responses by inhibiting the release of mito-chondrial DNA mediated by the NALP3 inflammasome. Nat. Immunol. 12: 222–230.

62. Shi, C. S., K. Shenderov, N. N. Huang, J. Kabat, M. Abu-Asab, K. A. Fitzgerald,A. Sher, and J. H. Kehrl. 2012. Activation of autophagy by inflammatory signalslimits IL-1b production by targeting ubiquitinated inflammasomes for destruc-tion. Nat. Immunol. 13: 255–263.

2964 IFN-g–INDUCED AUTOPHAGY

by guest on June 28, 2022http://w

ww

.jimm

unol.org/D

ownloaded from