Autophagy delivers cytoplasmic constituents to autophagosomes and is involved in innate and adaptive immunity. Cytosolic phospholipase (cPLA2)-initiated proinflammatory lipid mediator pathways play a critical role in host defense and inflammation. The crosstalk between the two pathways remains unclear. In this study, we report that cPLA2 and its metabolite lipid mediators induced autophagy in the RAW246.7 macrophage cell line and in primary monocytes. IFN-γ–triggered autophagy involves activation of cPLA2. Cysteinyl leukotrienes D4 and E4 and PGD2 also induced these effects. The autophagy is independent of changes in mTOR or autophagic flux. cPLA2 and lipid mediator-induced autophagy is ATG5 dependent. These data suggest that lipid mediators play a role in the regulation of autophagy, demonstrating a connection between the two seemingly separate innate immune responses, induction of autophagy and lipid mediator generation.

Recently published studies have strongly indicated that autophagy is a host defense mechanism by which cells respond to microbial invasion and promote cell survival (14). Autophagy plays an important role in innate and adaptive immunity (58). The signals that activate autophagy and molecular tags guide the formation of double-membrane cytosolic vesicles. These vesicles, designated autophagosomes, sequester invading pathogens and their products, portions of the cytosol and damaged organelles. The autophagosomes ultimately fuse with other vesicles in the endolysosomal pathway to deliver microbial ligands for adaptive or innate immune activation, or with the lysosome for subsequent degradation in autolysosomes (9, 10).

After an inflammatory stimulus, cells also may produce lipid mediators, such as leukotrienes (LTs) or PGs (11). Those lipid messengers are derived from the polyunsaturated fatty acid, arachidonic acid (AA). As the common precursor, AA is in turn converted to PGs by cyclooxygenase (COX) pathway enzymes or to LTs by the 5-lipoxygenase pathway (5-LO) (11, 12). The enzyme that hydrolyzes AA release from membrane phospholipids is cytosolic phospholipase A2 (cPLA2), which is a rate-limiting enzyme that plays a key role in initiating and regulating the multistage biosynthetic process of eicosanoid production. cPLA2 activation is involved in TLR-induced innate immune signaling (13). Therefore, a cPLA2-initiated proinflammatory lipid mediator pathway may play a pivotal role in the regulation of immune and inflammatory responses (14, 15).

Because autophagy is a cellular defense mechanism and the cPLA2-initiated lipid mediator pathway is important for the production of lipid mediators and the promotion of the inflammatory response, we investigated the role of cPLA2 and its lipid products in the induction of autophagy. Furthermore, we studied whether they may participate in IFN-γ–induced autophagy in the macrophage. In this study, we report that cPLA2 and its lipid metabolites induce autophagy in the RAW264.7 macrophage cell line and in primary human peripheral blood monocytes. This pathway may also be important in IFN-γ–induced autophagy in macrophages. The induction of autophagy may be via an ATG5-dependent pathway. Therefore, cPLA2-initiated lipid mediator generation may play a role in the autophagy response.

Earle’s balanced salt solution (EBSS medium) was purchased from Thermo Scientific (Waltham, MA). Murine and human IFN-γ were purchased from PeproTech (Rocky Hill, NJ). cPLA2 inhibitor, N-{(2S,4R)-4-(biphenyl-2-ylmethyl-isobutyl-amino)-1-[2-(2,4-difluorobenzoyl)-benzoyl]-pyrrolidin-2-ylmethyl}-3-[4-(2,4-dioxothiazolidin-5-ylidenemethyl)-phenyl]acrylamide HCl, a 1,2,4-trisubstituted pyrrolidine derivative, was from EMD Chemicals (San Diego, CA). MK866, indomethacin, 5-HETE, PGE2, PGD2, LTB4, LTD4, AA, and LTE4 were purchased from Cayman Chemical (Ann Arbor, MI). Rabbit polyclonal Abs against LC3 and actin were from Sigma-Aldrich (St. Louis, MO). Rabbit polyclonal Abs against beclin-1 and ATG5 were purchased from Novus (Littleton, CO). A rabbit polyclonal Ab against cPLA2, rabbit anti-phospho S6K mAb, and mouse anti-S6K mAb were purchased from Cell Signaling Technology (Beverly, MA). mAb against GST was from Santa Cruz Biotechnology (Santa Cruz, CA). E64d and pepstatin A were from Sigma-Aldrich (St. Louis, MO).

The murine macrophage cell line RAW264.7 was obtained from the American Type Culture Collection (Manassas, VA) and maintained in DMEM (BioSource International, Camarillo, CA) supplemented with 10% FCS and 1% antibiotics PenStrep (Life Technologies, Carlsbad, CA). Elutriated human peripheral blood monocytes were received from the Department of Transfusion Medicine, Clinical Center, under an institutional review board-approved protocol. Cells were washed with PBS and maintained in human monocyte medium (Amaxa, Gaithersburg, MD). For RAW264.7 cell starvation, cells were washed once, grown in EBSS for 2 h, and harvested for S6K immunoblotting.

pEGFP-LC3 was a gift of N. Mizushima (Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan) (16). For the cPLA2 plasmid construct, the cPLA2 gene insert was prepared by PCR from pEGFP-cPLA2 (a gift of J. Evans and C. Leslie, National Jewish Medical Center, Denver, CO) (17). The insert was then cloned into the mammalian expression GST plasmid. The correct sequence of the positive clones was confirmed by nucleotide sequencing based on the sequence of accession number P47712. The plasmids were then transiently transfected into the RAW264.7 cells using FuGENE HD Transfection Reagent from Roche (Nutley, NJ), according to the manufacturer’s protocol.

The small inhibitory RNAs (siRNAs) of cPLA2, beclin-1, ATG5, and control siRNA were purchased from Santa Cruz Biotechnology. Cells were transfected with 30 pmol/ml siRNA by using Nucleofection electroporating transfection (Amaxa) following the manufacturer’s directions. The interference of cPLA2, beclin-1, or ATG5 protein expression was compared with control nontargeting siRNA and confirmed by immunoblotting.

Total RNA was extracted from cells using QIA Shredder columns and RNeasy mini kit, and was treated with DNase (Qiagen, Valencia, CA). mRNA expression for IRG-47 (IFI-47) was measured using an ABI Prism 7900 sequence detection system (Applied Biosystems, Foster City, CA) and IFI-47 probe and primer sets (Applied Biosystems). Reverse transcription and PCR were performed using a reverse-transcription kit and TaqMan Universal PCR master mix (Applied Biosystems), according to the manufacturer’s directions. Relative gene expression was normalized to GAPDH transcripts, calculated as a fold-change compared with control.

RAW264.7 cells were incubated overnight with 0.5 μCi (5,6,8,9,11,12,14,15-[3H]AA) (150–230 Ci/mmol/ml; GE Healthcare, Pascataway, NJ). Media were removed, and the cells were washed and incubated in control medium or in medium with IFN-γ for 6 h. Media were harvested, and an aliquot was assayed in an LS 6500 scintilation counter (Beckman Instruments, Miami, FL).

RAW264.7 cells and primary monocytes were grown in six-well plates with stimuli or other treatment, as indicated. The procedure of Western blot analysis was carried out, as previously described (13). The primary Abs were as indicated. For LC3 Western blot, the cell lysis buffer was changed to PBS and 2% Triton X-100 (18). Actin was used as a loading control.

Cells were transfected with GST-cPLA2 (1 μg/well) for 24 h, or incubated with lipids (100 nM) for 6 h, or alternatively, with IFN-γ (0.1 μg/ml) for 6 h in nutrient medium.

RAW264.7 cells were grown on 35-mm glass-bottom microwell dishes and transfected by pEGFP-LC3 and with the cPLA2 expression plasmid or siRNA for 24 h, or different treatments, as described. Cells were then fixed with −20°C methanol overnight in the dishes. The fixed cells were washed with PBS twice and kept in PBS. The green fluorescence-labeled cells were observed, and the numbers of LC3-positive autophagosomes in cells were quantified using a Leica DMIRB fluorescence-inverted microscope with an original magnification ×63 oil objective and TGS Sl confocal system. For the GFP-LC3 assay, a minimum of 100 GFP-positive cells per sample was counted, and the number of GFP-LC3 aggregates was enumerated. Cells were scored as positive if they had more than three large GFP-LC3 puncta, and the data were presented as a percentage of the total number of GFP-positive cells. The results are shown as the means ± SD from three independent experiments. Laser-scanning images were cropped using Adobe Photoshop 5.0.

Cultures grown in Permanox tissue culture dishes (Nunc Nalgene) were fixed in 2.5% glutaraldehyde, 1% paraformaldehyde, and 0.12 M sodium cacodylate buffer (pH 7.3); postfixed in 1% OsO4; and en bloc stained with 1% uranyl acetate. The cultures were then dehydrated in graded ethanol and propylene oxide and infiltrated with Epon (Embed-812; Electron Microscopy Sciences), and the polymerized blocks were sectioned parallel to the culture substrate. Thin sections were stained with uranyl acetate and lead citrate, and then viewed with a JEM-1200EX electron microscope (JEOL) equipped with an AMT XR-60 digital camera (Advanced Microscopy Techniques). Images for quantification were taken with a ×5000 electron microscope original magnification setting from sections cut approximately midway between the adherent surface and the upper surface of the cell, such that cytoplasm occupied at least two-thirds of the image area. Structures of interest were counted in 44–47 digitally enlarged images corresponding to approximately half the number of cells each for experimentally treated and control cultures.

All statistical analyses were performed in Excel using a two-tailed t test. Where appropriate, a Bonferroni adjustment was applied for multiple comparisons. A p value <0.05 was considered significant.

To detect autphagosome formation, GFP fused to LC3 (GFP-LC3) was used as a marker of autophagy (5). Microtubule-associated protein L chain (LC3) is a homolog of ATG8 protein in yeast (16). LC3 exists in two forms, as follows: the nonlipid form, cytosolic species LC3-I corresponding to the Mr of 18 kDa, and its membrane-associated form, LC3-II, conjugated C-terminally to phosphatidylethanolamine, with an apparent Mr of 16 kDa. The latter form, LC3-II, is found both inside and outside the autophagosome. It can be used to document induction of autophagy (16) with increased levels of the autophagosome protein LC3-II on immunoblots or with the appearance of the cytoplasmic fluorescent puncta formed by inserted GFP-LC3-II into the membrane of the autophagosome. To test whether cPLA2 can induce autophagy, GFP-LC3 was cotransfected with the GST vector or the GST-cPLA2 vector into RAW 264.7 macrophages for 24 h. Expression of the indicated proteins is shown on the immunoblot (Fig. 1A). Overexpression of cPLA2 induced autophagy in ∼60% of GFP-positive cells as compared with ∼18% of GST vector control cells (Fig. 1B). The confocal images of cells containing GFP-LC3–marked autophagosomes are shown in Fig. 1C. To further confirm this observation, we detected endogenous LC3-I and LC3-II in lysates from GST vector control cells or GST-cPLA2–treated cells by immunoblotting with anti-LC3 Abs. Overexpression of cPLA2 increased the amount of LC3-II, which indicates formation of autophagosomes (Fig. 1D). The above results suggest that overexpression of cPLA2 may facilitate autophagy in macrophages.

FIGURE 1.

cPLA2 overexpression induces an autophagic response in macrophages. A, RAW264.7 cells were transfected with GST-cPLA2 for 24 h. A mAb against GST was used for immunoblotting. B, Quantification of cells with GFP-LC3 autophagic organelles in RAW 264.7 macrophages. Results are shown as the mean ± SD (n = 3), percentage of cells with marker-positive autophagic organelles. *p < 0.001, comparing cells transfected with the control GST vector versus cells transfected with the GST-cPLA2 vector. C, RAW264.7 cells were transfected with GFP-LC3 and with or without GST-cPLA2 for 24 h. GFP-LC3–positive autophagic organelles in transfected cells were observed by confocal fluorescence microscopy. D, RAW264.7 cells were transfected with GST-cPLA2. After 24 h, cells were harvested for immunoblotting with Abs against LC3 and actin. The results shown are from one of two independent experiments that gave similar results.

FIGURE 1.

cPLA2 overexpression induces an autophagic response in macrophages. A, RAW264.7 cells were transfected with GST-cPLA2 for 24 h. A mAb against GST was used for immunoblotting. B, Quantification of cells with GFP-LC3 autophagic organelles in RAW 264.7 macrophages. Results are shown as the mean ± SD (n = 3), percentage of cells with marker-positive autophagic organelles. *p < 0.001, comparing cells transfected with the control GST vector versus cells transfected with the GST-cPLA2 vector. C, RAW264.7 cells were transfected with GFP-LC3 and with or without GST-cPLA2 for 24 h. GFP-LC3–positive autophagic organelles in transfected cells were observed by confocal fluorescence microscopy. D, RAW264.7 cells were transfected with GST-cPLA2. After 24 h, cells were harvested for immunoblotting with Abs against LC3 and actin. The results shown are from one of two independent experiments that gave similar results.

Close modal

IFN-γ induces autophagy in macrophages (1), a process that is important for eliminating intracellular bacterial infection and cell defense. However, the mechanism or pathway by which this occurs is not clear. Previous reports suggest that IFN-γ induces cPLA2 activity in human epithelial cells and in HL-60 cells (19, 20). Therefore, we next examined whether cPLA2 participates in IFN-γ–induced autophagy in macrophages. We first reduced the expression of endogenous cPLA2 by transfecting two siRNAs directed against cPLA2 into RAW264.7 macrophages and then treated the cells with or without IFN-γ. As demonstrated by immunoblotting, two cPLA2 siRNAs reduced cPLA2 expression in macrophages (Fig. 2A), and the reduction of cPLA2 expression impaired IFN-γ–induced autophagy based on the amount of endogenous LC3-II detected (Fig. 2B). Furthermore, RAW 264.7 macrophages were cotransfected with GFP-LC3 and cPLA2 siRNA or the control scrambled siRNA. The formation of GFP-LC3 autophagosomes was analyzed with or without IFN-γ stimulation by confocal fluorescence microscopy. These images are shown in Fig. 2C. IFN-γ induced autophagic vesicle formation. This response was reduced in cells treated with siRNA against cPLA2 compared with cells treated with a control siRNA (Fig. 2C, 2D). To further confirm that cPLA2 is involved in IFN-γ–induced autophagy in macrophages, we used a pyrrolidine derivative cPLA2 inhibitor, N-{(2S,4R)-4-(biphenyl-2-ylmethyl-isobutyl-amino)-1-[2-(2,4-difluorobenzoyl)-benzoyl]-pyrrolidin-2-ylmethyl}-3-[4-(2,4-dioxothiazolidin-5-ylidenemethyl)-phenyl]acrylamide HCl, to perform the above experiments (Fig. 3). Treatment of cells with this cPLA2 inhibitor blocked IFN-γ–induced AA release (Fig. 3B). Consistent with the siRNA data, cPLA2 inhibition partially blocked IFN-γ–induced autophagy, as measured by localization of LC3 to autophagosomes or changes in LC3-II by immunoblot (Fig. 3D). These results indicate that cPLA2 at least in part regulates IFN-γ–induced autophagy in macrophages. To determine whether the cPLA2 inhibitor altered IFN-γ–induced transcription, the effect of the cPLA2 inhibitor was studied on an IFN-γ–inducible gene. IRG-47 (IFI-47) is an IFN-γ–reducible gene that has distinct roles in immune defense against protozoan infections (21). RAW264.7 cells were treated with or without cPLA2 inhibitor (10 μM) for 2 h before treatment of IFN-γ for 6 h. Cells were harvested, and IRG-47 mRNA levels were determined by RT-PCR (Fig. 3E). IFN-γ induction of IRG-47 was not altered by treatment with the cPLA2 inhibitor, suggesting that cPLA2 does not regulate IFN-γ gene induction.

FIGURE 2.

cPLA2 is involved in IFN-γ–induced macrophage autophagy. A, RAW264.7 cells were transfected with two cPLA2-specific siRNAs or nonspecific siRNA (NC siRNA) for 24 h, and immunoblots were performed with polyclonal Abs against cPLA2 and actin. B, RAW264.7 cells were transfected with siRNAs against cPLA2 or NC siRNA. After 24 h, cells were incubated with or without IFN-γ (0.1 μg/ml) for 6 h. Immunoblots were performed using an Ab against LC3. The results shown are representative of one of two independent experiments that gave similar results. C, RAW264.7 cells were cotransfected with GFP-LC3 and siRNAs against cPLA2 or NC siRNA for 24 h. Cells were stimulated with or without IFN-γ (0.1 μg/ml) for 6 h. LC3 autophagic organelles in samples were observed by confocal fluorescence microscopy. D, The number of GFP-LC3 autophagic organelle-positive cells was quantified. Results are shown as the mean ± SD. n = 100 LC3-transfected cells from each of three independent experiments. *p < 0.01 for NC siRNA-transfected plus IFN-γ–treated cells versus NC siRNA-transfected cells. **p < 0.05 for NC siRNA-transfected plus IFN-γ–treated cells versus cells transfected and treated with either of two cPLA2 siRNAs plus IFN-γ.

FIGURE 2.

cPLA2 is involved in IFN-γ–induced macrophage autophagy. A, RAW264.7 cells were transfected with two cPLA2-specific siRNAs or nonspecific siRNA (NC siRNA) for 24 h, and immunoblots were performed with polyclonal Abs against cPLA2 and actin. B, RAW264.7 cells were transfected with siRNAs against cPLA2 or NC siRNA. After 24 h, cells were incubated with or without IFN-γ (0.1 μg/ml) for 6 h. Immunoblots were performed using an Ab against LC3. The results shown are representative of one of two independent experiments that gave similar results. C, RAW264.7 cells were cotransfected with GFP-LC3 and siRNAs against cPLA2 or NC siRNA for 24 h. Cells were stimulated with or without IFN-γ (0.1 μg/ml) for 6 h. LC3 autophagic organelles in samples were observed by confocal fluorescence microscopy. D, The number of GFP-LC3 autophagic organelle-positive cells was quantified. Results are shown as the mean ± SD. n = 100 LC3-transfected cells from each of three independent experiments. *p < 0.01 for NC siRNA-transfected plus IFN-γ–treated cells versus NC siRNA-transfected cells. **p < 0.05 for NC siRNA-transfected plus IFN-γ–treated cells versus cells transfected and treated with either of two cPLA2 siRNAs plus IFN-γ.

Close modal
FIGURE 3.

cPLA2 inhibition impairs IFN-γ–induced autophagy. A and C, RAW264.7 cells were transfected with GFP-LC3 for 24 h. Cells were treated with or without a cPLA2 inhibitor (10 μM) for 2 h, followed by incubation with or without IFN-γ (0.1 μg/ml) for 6 h. GFP-LC3 autophagic organelles were evaluated and quantified by confocal fluorescence microscopy. *p < 0.0014 for cells treated with IFN-γ versus control cells; **p < 0.015 for IFN-γ–treated cells versus cells treated with IFN-γ plus cPLA2 inhibitor. B, RAW264.7 cells were labeled with 0.5 μCi/ml [3H]AA overnight. The [3H]AA–labeled cells were treated with cPLA2 inhibitor (10 μM) for 2 h before stimulation with IFN-γ (0.1 μg/ml) for 6 h. AA release was then assessed in IFN-γ–treated and untreated cells. The data were presented as the mean values ± SD of fold stimulation compared with control supernatant in three independent assays. *p < 0.001 for control cells versus IFN-γ–treated cells alone; **p < 0.001 for cells treated with IFN-γ versus cells treated with cPLA2 inhibitor plus IFN-γ. D, Immunoblotting showing cPLA2 inhibition blocked the IFN-γ–induced endogenous LC3-II increase. The result is representative of two separate immunoblotting experiments. E, cPLA2 inhibition does not inhibit IFN-γ–induced gene expression. Cells were treated with or without cPLA2 inhibitor for 2 h, stimulated with or without IFN-γ (0.1 μg/ml) for 6 h, and RNA extracted for real-time PCR. Data are expressed as fold change (mean ± SD) from triplicate samples each assayed in duplicate.

FIGURE 3.

cPLA2 inhibition impairs IFN-γ–induced autophagy. A and C, RAW264.7 cells were transfected with GFP-LC3 for 24 h. Cells were treated with or without a cPLA2 inhibitor (10 μM) for 2 h, followed by incubation with or without IFN-γ (0.1 μg/ml) for 6 h. GFP-LC3 autophagic organelles were evaluated and quantified by confocal fluorescence microscopy. *p < 0.0014 for cells treated with IFN-γ versus control cells; **p < 0.015 for IFN-γ–treated cells versus cells treated with IFN-γ plus cPLA2 inhibitor. B, RAW264.7 cells were labeled with 0.5 μCi/ml [3H]AA overnight. The [3H]AA–labeled cells were treated with cPLA2 inhibitor (10 μM) for 2 h before stimulation with IFN-γ (0.1 μg/ml) for 6 h. AA release was then assessed in IFN-γ–treated and untreated cells. The data were presented as the mean values ± SD of fold stimulation compared with control supernatant in three independent assays. *p < 0.001 for control cells versus IFN-γ–treated cells alone; **p < 0.001 for cells treated with IFN-γ versus cells treated with cPLA2 inhibitor plus IFN-γ. D, Immunoblotting showing cPLA2 inhibition blocked the IFN-γ–induced endogenous LC3-II increase. The result is representative of two separate immunoblotting experiments. E, cPLA2 inhibition does not inhibit IFN-γ–induced gene expression. Cells were treated with or without cPLA2 inhibitor for 2 h, stimulated with or without IFN-γ (0.1 μg/ml) for 6 h, and RNA extracted for real-time PCR. Data are expressed as fold change (mean ± SD) from triplicate samples each assayed in duplicate.

Close modal

Inflammatory lipid mediators (eicosanoids) are final metabolites of cPLA2 activation. We examined the hypothesis that inflammatory lipid mediators are important in the induction of autophagy. A COX inhibitor, indomethacin, or 5-LO inhibitor, MK866, was used to treat RAW264.7 macrophages transfected with GFP-LC3 prior to IFN-γ stimulation. Inhibition of either COX or 5-LO partially inhibited IFN-γ–induced GFP-LC3 autophagosome formation as compared with IFN-γ stimulation alone (Fig. 4A), suggesting that the lipid mediators from either pathway may be involved in the induction of autophagy. We tested whether eicosanoids that are known to be produced by macrophages might induce these changes. Cysteinyl LT (CysLT) D4 and CysLTE4 from the 5-LO pathway and PGD2 from the COX pathway induced an increase in the level of GFP-LC3 autophagosome formation (Fig. 4B, 4C) and endogenous LC3-II formation (Fig. 4D). LTB4, 5-HETE, and PGE2 had no significant effect. AA added to the media induced LC3-II (Fig. 4F), suggesting that AA may be metabolized to an active eicosanoid capable of inducing the autophagic response.

FIGURE 4.

Lipid mediators induce autophagy. A, RAW264.7 cells were transiently transfected with GFP-LC3 for 24 h. Cells were treated with either indomethacin (10 μM) or MK866 inhibitors (10 μM) for 2 h before incubation with IFN-γ (0.1 μg/ml) for 6 h. GFP-LC3 autophagic organelle-positive cells were calculated from 100 green cells and the mean ± SD compared with control or IFN-γ–treated cells in three independent assays. *p < 0.001 comparing control cells with IFN-γ–treated cells. **p < 0.001 comparing IFN-γ–treated cells with IFN-γ–treated cells plus either MK886 or indomethacin. B and C, GFP-LC3–transfected RAW264.7 cells were incubated with lipid mediators (100 nM) for 6 h. Autophagic organelle-positive cells were quantified from 100 transfected cells as the mean ± SD in three independent assays. *p < 0.01 versus control. C, GFP-LC3 autophagic organelles were examined by confocal fluorescence microscopy in GFP-LC3–transfected cells with the lipid mediators that induced the strongest autophagic effects in cells. D, Endogenous LC3-II was detected in lipid mediator-treated RAW 264.7 cells by immunoblotting with the Ab against LC3. The results are from one of three independent experiments that gave similar results. E, Primary monocytes were incubated with lipid mediators or with IFN-γ for 6 h. In some experiments, the cPLA2 inhibitor was added 2 h prior to treatment with IFN-γ for 6 h. Cellular LC3-II was assayed by immunoblotting with the Ab against LC3. Actin was used as a loading control. The immunoblot is representative of three similar independent experiments. F, AA induced LC3-II formation. RAW 264.7 cells were treated with or without AA (100 nM) for 6 h, and LC3-II formation was assayed by Western blot. Actin was used as a loading control.

FIGURE 4.

Lipid mediators induce autophagy. A, RAW264.7 cells were transiently transfected with GFP-LC3 for 24 h. Cells were treated with either indomethacin (10 μM) or MK866 inhibitors (10 μM) for 2 h before incubation with IFN-γ (0.1 μg/ml) for 6 h. GFP-LC3 autophagic organelle-positive cells were calculated from 100 green cells and the mean ± SD compared with control or IFN-γ–treated cells in three independent assays. *p < 0.001 comparing control cells with IFN-γ–treated cells. **p < 0.001 comparing IFN-γ–treated cells with IFN-γ–treated cells plus either MK886 or indomethacin. B and C, GFP-LC3–transfected RAW264.7 cells were incubated with lipid mediators (100 nM) for 6 h. Autophagic organelle-positive cells were quantified from 100 transfected cells as the mean ± SD in three independent assays. *p < 0.01 versus control. C, GFP-LC3 autophagic organelles were examined by confocal fluorescence microscopy in GFP-LC3–transfected cells with the lipid mediators that induced the strongest autophagic effects in cells. D, Endogenous LC3-II was detected in lipid mediator-treated RAW 264.7 cells by immunoblotting with the Ab against LC3. The results are from one of three independent experiments that gave similar results. E, Primary monocytes were incubated with lipid mediators or with IFN-γ for 6 h. In some experiments, the cPLA2 inhibitor was added 2 h prior to treatment with IFN-γ for 6 h. Cellular LC3-II was assayed by immunoblotting with the Ab against LC3. Actin was used as a loading control. The immunoblot is representative of three similar independent experiments. F, AA induced LC3-II formation. RAW 264.7 cells were treated with or without AA (100 nM) for 6 h, and LC3-II formation was assayed by Western blot. Actin was used as a loading control.

Close modal

We next examined the autophagic response in primary cells, using monocytes separated from human peripheral blood. Human monocyte expression of endogenous LC3-II was assayed by immunoblotting (Fig. 4E). Although no LC3-II was detected in untreated monocytes, IFN-γ induced an increase in LC3-II, which was partially inhibited by pretreatment with a cPLA2 inhibitor. When monocytes were stimulated with the lipid mediators CysLTD4, CysLTE4, or PGD2, LC3-II formation was again noted. Thus, a cPLA2-initiated lipid mediator pathway for the induction of autophagy is present in human monocytes.

Next, we examined the RAW264.7 macrophages by electron microscopy to look for early autophagosomes and autolysosomes (Fig. 5). Early autophagosomes were defined by the presence of a double membrane envelope surrounding a region of cytoplasm, whereas autolysosomes, also called late autophagosomes, were defined by a single membrane envelope with cytoplasmic content that appeared degraded or condensed. In addition to autophagosomes and autolysosomes, we observed tubular cytoplasmic structures with two membranes separated by a space of ∼10 nm with an electron-dense core. Profiles of these tubular structures often occurred in clusters, suggesting that the structures could assume a curled configuration. We also noted many examples of these tubular structures surrounding areas of cytoplasm, which appeared to undergo degradation. Such compound structures were counted as autophagosomes or autolysosomes. We counted the total number of autophagosomes and autolysosomes in 44–47 fields from ∼20 cells each from control and CysLTD4-treated cultures. CysLTD4-treated cells had 1.8-fold more autophagosomes or autolysosomes per field than control cells. Moreover, 49% of the fields had three or more autophagic structures in CysLTD4-treated cells compared with 18% in controls. These results are consistent with the increased abundance of LC3-positive autophagosomes (Fig. 4B) in CysLTD4-treated cells. We also counted the number of widely separated individual profiles and clusters of profiles of the tubular dense-core structures. We found 1.7-fold more of these structures in CysLTD4-treated cells than in controls.

FIGURE 5.

Electron micrographs of the cytoplasm of RAW246.7 cells. Both control cells (A) and cells treated with CysLTD4 for 6 h (B, C) contained early autophagosomes with an irregular double membrane envelope (large arrows, not shown in control), late autophagosomes or autolysosomes with a single membrane envelope (large arrowheads), and irregular tubular structures with a double membrane and a dense core (small arrows). The latter structures often appear to wrap around areas of cytoplasm, forming an autophagosome (small arrowheads). Both the autophagosomes/autolysosomes and the tubular dense core structures were more abundant in the CysLTD4-treated cells than in controls (see 13Results). Scale bars, 1 μm.

FIGURE 5.

Electron micrographs of the cytoplasm of RAW246.7 cells. Both control cells (A) and cells treated with CysLTD4 for 6 h (B, C) contained early autophagosomes with an irregular double membrane envelope (large arrows, not shown in control), late autophagosomes or autolysosomes with a single membrane envelope (large arrowheads), and irregular tubular structures with a double membrane and a dense core (small arrows). The latter structures often appear to wrap around areas of cytoplasm, forming an autophagosome (small arrowheads). Both the autophagosomes/autolysosomes and the tubular dense core structures were more abundant in the CysLTD4-treated cells than in controls (see 13Results). Scale bars, 1 μm.

Close modal

Beclin-1 and ATG5 are central regulators in autophagy (22, 23). Beclin-1 is a component of the class III PI3K complex, which involves vesicle nucleation in the early stage of autophagy. ATG5 participates in the vesicle elongation of autophagy (24). We reduced the expression of beclin-1 and ATG5 in RAW264.7 macrophages by transfection with beclin-1 or ATG5 siRNA (Fig. 6A, 6B). GFP-LC3 autophagosomes were observed with cPLA2, siRNAs targeting beclin-1 or ATG5 and GFP-LC3 cotransfection for 24 h. Knockdown of ATG5 appeared to inhibit cPLA2-induced changes in GFP-LC3 autophagosome formation (Fig. 6C). Similarly, cells were stimulated with CysLTD4 after transfection with siRNAs against beclin-1 or ATG5 and GFP-LC3 cotransfection for 24 h. GFP-LC3 autophagosomes were again increased by CysLTD4 treatment. This increase was reduced in cells treated with siRNA against ATG5 relative to the cells transfected with control nonspecific siRNA (NC siRNA) or siRNA against beclin-1 (Fig. 6D). Endogenous LC3-II protein was decreased in cells transfected with siRNA against ATG5 (Fig. 6E). These data suggest that the cPLA2–initiated pathway–induced autophagy is ATG5 dependent.

FIGURE 6.

cPLA2–initiated autophagy is ATG5 dependent. A and B, RAW264.7 cells were transiently transfected with control siRNA (NC siRNA) or siRNA directed at beclin-1 or ATG5 for 24 h and assessed by immunoblotting for beclin-1 or ATG5. C, Quantification of LC3 autophagic organelles in cPLA2, GFP-LC3, and beclin-1 siRNA or ATG5 siRNA-cotransfected RAW264.7 cells for 24 h. *p < 0.005 for NC siRNA-transfected control cells versus cells transfected with GST-cPLA2 and NC siRNA. **p > 0.05 for cells treated with NC siRNA plus GST-cPLA2 versus cells treated with beclin-1 siRNA plus GST-cPLA2. ***p < 0.01 for cells treated with NC siRNA plus GST-cPLA2 versus cells treated with ATG5 siRNA plus GST-cPLA2. D, Cells were transfected with GFP-LC3 and NC siRNA, beclin-1 siRNA, or ATG5 siRNA for 24 h; treated with LTD4 for 6 h; and then analyzed for LC3 autophagic organelles. Results are shown as the mean ± SD. n = 100 LC3-transfected cells each for three independent experiments. *p < 0.001 for cells treated with NC siRNA versus cells treated with NC siRNA plus LTD4; ** p > 0.05 for cells treated with NC siRNA plus LTD4 versus cells treated with beclin-1 siRNA plus LTD4. ***p < 0.001 for cells treated with NC siRNA plus LTD4 versus cells treated with ATG5 siRNA plus LTD4. E, Immunoblotting was performed to assess LC3 protein expression in cells that had been transfected with siRNA against beclin-1 or ATG5 for 24 h, followed by treatment of LTD4 for 6 h. Actin was used as a loading control. The blot is representative of two separate experiments with similar results.

FIGURE 6.

cPLA2–initiated autophagy is ATG5 dependent. A and B, RAW264.7 cells were transiently transfected with control siRNA (NC siRNA) or siRNA directed at beclin-1 or ATG5 for 24 h and assessed by immunoblotting for beclin-1 or ATG5. C, Quantification of LC3 autophagic organelles in cPLA2, GFP-LC3, and beclin-1 siRNA or ATG5 siRNA-cotransfected RAW264.7 cells for 24 h. *p < 0.005 for NC siRNA-transfected control cells versus cells transfected with GST-cPLA2 and NC siRNA. **p > 0.05 for cells treated with NC siRNA plus GST-cPLA2 versus cells treated with beclin-1 siRNA plus GST-cPLA2. ***p < 0.01 for cells treated with NC siRNA plus GST-cPLA2 versus cells treated with ATG5 siRNA plus GST-cPLA2. D, Cells were transfected with GFP-LC3 and NC siRNA, beclin-1 siRNA, or ATG5 siRNA for 24 h; treated with LTD4 for 6 h; and then analyzed for LC3 autophagic organelles. Results are shown as the mean ± SD. n = 100 LC3-transfected cells each for three independent experiments. *p < 0.001 for cells treated with NC siRNA versus cells treated with NC siRNA plus LTD4; ** p > 0.05 for cells treated with NC siRNA plus LTD4 versus cells treated with beclin-1 siRNA plus LTD4. ***p < 0.001 for cells treated with NC siRNA plus LTD4 versus cells treated with ATG5 siRNA plus LTD4. E, Immunoblotting was performed to assess LC3 protein expression in cells that had been transfected with siRNA against beclin-1 or ATG5 for 24 h, followed by treatment of LTD4 for 6 h. Actin was used as a loading control. The blot is representative of two separate experiments with similar results.

Close modal

We further studied the effect of LTD4 on cellular processes that might affect autophagosome accumulation. To determine whether LC3-II accumulation is an effect of increased autophagy or a result of a block in autophagic flux, RAW264.7 cells were treated with or without lysozome inhibitors E64d and pepstatin A, and changes in LC3-II were analyzed by Western blot (Fig. 7A). Pepstatin A and E64d treatment appeared to increase LC3-II accumulation in cells not treated with LTD4, but to a greater degree in cells treated with LTD4. Therefore, LC3-II accumulation in LTD4–treated cells does not appear to be a result of inhibition of autophagic flux. To determine whether the effect of LTD4 on LC3-II accumulation might be the result of mTOR inhibition, phospho-S6K was determined by Western blot. Treatment of RAW264.7 cells with LTD4 was not associated with a decrease in the phosphorylation of the mTOR substrate, S6K. As a control, amino acid starvation of these cells was associated with inhibition of S6K phosphorylation (Fig. 7B).

FIGURE 7.

LTD4-induced autophagy is not regulated at the level of autophagic flux or by mTOR. A, RAW264.7 cells were treated with or without lysozome inhibitors E64d (5 μg/ml) and pepstatin A (5 μg/ml) and with or without LTD4 (100 nM) for 6 h. Cells were harvested for immunoblotting of LC3. Actin is used as a loading control. The induction of LC3-II appeared to be enhanced by pretreatment with E64d and pepstatin A. The blot is representative of two blots with similar results. The average and range of the ratio of the density of blots of LC3-II/LC3-I are presented below the blot. B, mTOR activity was assayed by Western blot of phospho-S6K. RAW264.7 cells were treated with or without LTD4 for the indicated times or with amino acid starvation with EBSS for 2 h as a control and then harvested for immunoblotting for LC3.

FIGURE 7.

LTD4-induced autophagy is not regulated at the level of autophagic flux or by mTOR. A, RAW264.7 cells were treated with or without lysozome inhibitors E64d (5 μg/ml) and pepstatin A (5 μg/ml) and with or without LTD4 (100 nM) for 6 h. Cells were harvested for immunoblotting of LC3. Actin is used as a loading control. The induction of LC3-II appeared to be enhanced by pretreatment with E64d and pepstatin A. The blot is representative of two blots with similar results. The average and range of the ratio of the density of blots of LC3-II/LC3-I are presented below the blot. B, mTOR activity was assayed by Western blot of phospho-S6K. RAW264.7 cells were treated with or without LTD4 for the indicated times or with amino acid starvation with EBSS for 2 h as a control and then harvested for immunoblotting for LC3.

Close modal

In this study, we explored whether a cPLA2–initiated lipid mediator pathway participates in the induction of autophagy. We observed that a cPLA2-initiated lipid mediator pathway induces autophagy in both a murine macrophage cell line (RAW264.7) and human primary monocytes. The induction may be ATG5 dependent and independent of autophagic flux or mTOR inhibition. The autophagy induction may be due to the action of lipid mediators generated downstream of cPLA2. In addition, lipid mediators appear to be involved in IFN-γ–induced autophagy, suggesting that cPLA2–initiated lipid mediators may be downstream effectors of IFN-γ signaling for autophagy.

Autophagy has been implicated previously in both health-promoting and disease-associated states. It has been thought to be a cellular homeostatic mechanism. Autophagy has been described in a variety of processes, including neoplasia, neurodegeneration, myopathies, development, aging, and innate and adaptive immune responses (5, 8). Lipid mediators also may play a role in regulating immune and inflammatory responses (25). Our results have linked two events, autophagy and cPLA2-initiated lipid mediator generation, which may bridge two aspects of the innate immune response. It is suggested that lipid mediator–induced inflammation may in part regulate autophagy induction.

Lipid mediators are most likely key participants in the pathogenesis of inflammatory diseases (26). Our inhibitor and lipid mediator stimulation data suggest that autophagy was induced by the lipid products from two multienzyme pathways downstream of cPLA2. LTB4 and LTC4 are products of the 5-LO pathway of AA metabolism. LTC4 is converted extracellularly to LTD4 and LTE4. Consequently, LTC4, LTD4, and LTE4 are together referred to as CysLTs. On the basis of multiple assays, including Western blots, immunofluorescence microscopy, and electron microscopy, we have shown that a subset of lipid mediators can induce autophagy in murine macrophages. Autophagy induction was observed with CysLTD4/E4 and PGD2 treatment, but not with LTB4, 5-HETE, or PGE2. These data suggest that specific receptor expression signaling is associated with the lipid mediator induction of autophagy in murine macrophages (27).

CysLTs and PGD2 exert their actions through activation of their receptors, such as CysLT subtype 1 receptor (CysLT1) (28), CysLT2 (29), and PGD2 receptor (DP) (30).

These are seven transmembrane domain G protein-coupled receptors that bind ligands to mediate intercellular signaling of inflammatory and other cells. The mechanism by which the signaling of lipid mediators may induce autophagy is not clear. Monocytes and macrophages can produce proinflammatory eicosanoids and express the CysLT receptors on the cell surface (31). CysLTs modify macrophage functions. For instance, LTD4 primes alveolar macrophages to release macrophage inflammatory protein 1α, TNF-α, and NO on exposure to inflammatory mediators (31). It may induce autophagy as well.

The autophagy induced by lipid mediators may be an aspect of cellular responses against microbial invasion. Previous research suggested that several lipids could modulate the macrophage innate immune response against mycobacteria and enhance their killing (32). Both NF-κB–dependent and NF-κB–independent mechanisms are involved in macrophage killing of mycobacteria, and both mechanisms can be enhanced by selected lipids.

The cPLA2-initiated lipid mediator pathway that induces autophagy appears to be ATG5 dependent. It may involve autophagic vesicle elongation. Several other proteins have been shown to bind ATG5. For example, in hepatitis C virus infection, ATG5 is an interacting protein for the hepatitis C virus NS5B protein (33). ATG5 may also contribute to autophagic cell death by interacting with Fas-associated protein with death domain (34). The mechanism by which ATG5 may participate in the cPLA2-initiated lipid mediator pathway–induced autophagy remains to be determined. In conclusion, cPLA2, acting via generation of lipid mediators, appears to be capable of inducing or amplifying an autophagic response in monocytes and macrophages.

We are grateful to Shervin Esfahani for assistance with electron microscopy.

This work was supported by the intramural research program of the Clinical Center, National Institutes of Health.

Abbreviations used in this article:

AA

arachidonic acid

COX

cyclooxygenase

cPLA2

cytosolic phospholipase

CysLT

cysteinyl leukotriene

EBSS

Earle’s balanced salt solution

5-LO

5-lipoxygenase pathway

LT

leukotriene

NC siRNA

control nonspecific small inhibitory RNA

siRNA

small inhibitory RNA.

1
Gutierrez
M. G.
,
Master
S. S.
,
Singh
S. B.
,
Taylor
G. A.
,
Colombo
M. I.
,
Deretic
V.
.
2004
.
Autophagy is a defense mechanism inhibiting BCG and Mycobacterium tuberculosis survival in infected macrophages.
Cell
119
:
753
766
.
2
Levine
B.
2005
.
Eating oneself and uninvited guests: autophagy-related pathways in cellular defense.
Cell
120
:
159
162
.
3
Nakagawa
I.
,
Amano
A.
,
Mizushima
N.
,
Yamamoto
A.
,
Yamaguchi
H.
,
Kamimoto
T.
,
Nara
A.
,
Funao
J.
,
Nakata
M.
,
Tsuda
K.
, et al
.
2004
.
Autophagy defends cells against invading group A Streptococcus.
Science
306
:
1037
1040
.
4
Singh
S. B.
,
Davis
A. S.
,
Taylor
G. A.
,
Deretic
V.
.
2006
.
Human IRGM induces autophagy to eliminate intracellular mycobacteria.
Science
313
:
1438
1441
.
5
Deretic
V.
2005
.
Autophagy in innate and adaptive immunity.
Trends Immunol.
26
:
523
528
.
6
Paludan
C.
,
Schmid
D.
,
Landthaler
M.
,
Vockerodt
M.
,
Kube
D.
,
Tuschl
T.
,
Münz
C.
.
2005
.
Endogenous MHC class II processing of a viral nuclear antigen after autophagy.
Science
307
:
593
596
.
7
Levine
B.
,
Deretic
V.
.
2007
.
Unveiling the roles of autophagy in innate and adaptive immunity.
Nat. Rev. Immunol.
7
:
767
777
.
8
Schmid
D.
,
Münz
C.
.
2007
.
Innate and adaptive immunity through autophagy.
Immunity
27
:
11
21
.
9
Mizushima
N.
2007
.
Autophagy: process and function.
Genes Dev.
21
:
2861
2873
.
10
Shintani
T.
,
Klionsky
D. J.
.
2004
.
Autophagy in health and disease: a double-edged sword.
Science
306
:
990
995
.
11
Henderson
W. R.
 Jr
.
1994
.
The role of leukotrienes in inflammation.
Ann. Intern. Med.
121
:
684
697
.
12
Funk
C. D.
2001
.
Prostaglandins and leukotrienes: advances in eicosanoid biology.
Science
294
:
1871
1875
.
13
Qi
H. Y.
,
Shelhamer
J. H.
.
2005
.
Toll-like receptor 4 signaling regulates cytosolic phospholipase A2 activation and lipid generation in lipopolysaccharide-stimulated macrophages.
J. Biol. Chem.
280
:
38969
38975
.
14
Nagase
T.
,
Uozumi
N.
,
Ishii
S.
,
Kume
K.
,
Izumi
T.
,
Ouchi
Y.
,
Shimizu
T.
.
2000
.
Acute lung injury by sepsis and acid aspiration: a key role for cytosolic phospholipase A2.
Nat. Immunol.
1
:
42
46
.
15
Nagase
T.
,
Uozumi
N.
,
Ishii
S.
,
Kita
Y.
,
Yamamoto
H.
,
Ohga
E.
,
Ouchi
Y.
,
Shimizu
T.
.
2002
.
A pivotal role of cytosolic phospholipase A(2) in bleomycin-induced pulmonary fibrosis.
Nat. Med.
8
:
480
484
.
16
Kabeya
Y.
,
Mizushima
N.
,
Ueno
T.
,
Yamamoto
A.
,
Kirisako
T.
,
Noda
T.
,
Kominami
E.
,
Ohsumi
Y.
,
Yoshimori
T.
.
2000
.
LC3, a mammalian homologue of yeast Apg8p, is localized in autophagosome membranes after processing.
EMBO J.
19
:
5720
5728
.
17
Evans
J. H.
,
Spencer
D. M.
,
Zweifach
A.
,
Leslie
C. C.
.
2001
.
Intracellular calcium signals regulating cytosolic phospholipase A2 translocation to internal membranes.
J. Biol. Chem.
276
:
30150
30160
.
18
Tanida
I.
,
Ueno
T.
,
Kominami
E.
.
2008
.
LC3 and autophagy.
Methods Mol. Biol.
445
:
77
88
.
19
Wu
T.
,
Levine
S. J.
,
Lawrence
M. G.
,
Logun
C.
,
Angus
C. W.
,
Shelhamer
J. H.
.
1994
.
Interferon-gamma induces the synthesis and activation of cytosolic phospholipase A2.
J. Clin. Invest.
93
:
571
577
.
20
Visnjić
D.
,
Batinić
D.
,
Banfić
H.
.
1997
.
Arachidonic acid mediates interferon-gamma-induced sphingomyelin hydrolysis and monocytic marker expression in HL-60 cell line.
Blood
89
:
81
91
.
21
Collazo
C. M.
,
Yap
G. S.
,
Sempowski
G. D.
,
Lusby
K. C.
,
Tessarollo
L.
,
Woude
G. F.
,
Sher
A.
,
Taylor
G. A.
.
2001
.
Inactivation of LRG-47 and IRG-47 reveals a family of interferon gamma-inducible genes with essential, pathogen-specific roles in resistance to infection.
J. Exp. Med.
194
:
181
188
.
22
Cao
Y.
,
Klionsky
D. J.
.
2007
.
Physiological functions of Atg6/Beclin 1: a unique autophagy-related protein.
Cell Res.
17
:
839
849
.
23
Pua
H. H.
,
Dzhagalov
I.
,
Chuck
M.
,
Mizushima
N.
,
He
Y. W.
.
2007
.
A critical role for the autophagy gene Atg5 in T cell survival and proliferation.
J. Exp. Med.
204
:
25
31
.
24
Levine
B.
,
Kroemer
G.
.
2008
.
Autophagy in the pathogenesis of disease.
Cell
132
:
27
42
.
25
Brock
T. G.
,
Peters-Golden
M.
.
2007
.
Activation and regulation of cellular eicosanoid biosynthesis.
ScientificWorldJournal
7
:
1273
1284
.
26
Christie
P. E.
,
Henderson
W. R.
 Jr.
2002
.
Lipid inflammatory mediators: leukotrienes, prostaglandins, platelet-activating factor.
Clin. Allergy Immunol.
16
:
233
254
.
27
Peters-Golden
M.
,
Gleason
M. M.
,
Togias
A.
.
2006
.
Cysteinyl leukotrienes: multi-functional mediators in allergic rhinitis.
Clin. Exp. Allergy
36
:
689
703
.
28
Lynch
K. R.
,
O’Neill
G. P.
,
Liu
Q.
,
Im
D. S.
,
Sawyer
N.
,
Metters
K. M.
,
Coulombe
N.
,
Abramovitz
M.
,
Figueroa
D. J.
,
Zeng
Z.
, et al
.
1999
.
Characterization of the human cysteinyl leukotriene CysLT1 receptor.
Nature
399
:
789
793
.
29
Heise
C. E.
,
O’Dowd
B. F.
,
Figueroa
D. J.
,
Sawyer
N.
,
Nguyen
T.
,
Im
D. S.
,
Stocco
R.
,
Bellefeuille
J. N.
,
Abramovitz
M.
,
Cheng
R.
, et al
.
2000
.
Characterization of the human cysteinyl leukotriene 2 receptor.
J. Biol. Chem.
275
:
30531
30536
.
30
Spik
I.
,
Brénuchon
C.
,
Angéli
V.
,
Staumont
D.
,
Fleury
S.
,
Capron
M.
,
Trottein
F.
,
Dombrowicz
D.
.
2005
.
Activation of the prostaglandin D2 receptor DP2/CRTH2 increases allergic inflammation in mouse.
J. Immunol.
174
:
3703
3708
.
31
Ogawa
Y.
,
Calhoun
W. J.
.
2006
.
The role of leukotrienes in airway inflammation.
J. Allergy Clin. Immunol.
118
:
789
798, quiz 799–800
.
32
Gutierrez
M. G.
,
Gonzalez
A. P.
,
Anes
E.
,
Griffiths
G.
.
2009
.
Role of lipids in killing mycobacteria by macrophages: evidence for NF-kappaB-dependent and -independent killing induced by different lipids.
Cell. Microbiol.
11
:
406
420
.
33
Guévin
C.
,
Manna
D.
,
Bélanger
C.
,
Konan
K. V.
,
Mak
P.
,
Labonté
P.
.
2010
.
Autophagy protein ATG5 interacts transiently with the hepatitis C virus RNA polymerase (NS5B) early during infection.
Virology
405
:
1
7
.
34
Pyo
J. O.
,
Jang
M. H.
,
Kwon
Y. K.
,
Lee
H. J.
,
Jun
J. I.
,
Woo
H. N.
,
Cho
D. H.
,
Choi
B.
,
Lee
H.
,
Kim
J. H.
, et al
.
2005
.
Essential roles of Atg5 and FADD in autophagic cell death: dissection of autophagic cell death into vacuole formation and cell death.
J. Biol. Chem.
280
:
20722
20729
.

The authors have no financial conflicts of interest.