LPS-induced autophagy is mediated by oxidative signaling in ...

Am J Physiol Heart Circ Physiol 296: H470–H479, 2009. First published December 19, 2008; doi:10.1152/ajpheart.01051.2008.

LPS-induced autophagy is mediated by oxidative signaling in cardiomyocytes and is associated with cytoprotection Hua Yuan,1 Cynthia N. Perry,1 Chengqun Huang,1 Eri Iwai-Kanai,1 Raquel S. Carreira,1 Christopher C. Glembotski,2 and Roberta A. Gottlieb1,2 1

Bioscience Center and 2Department of Biology, San Diego State University, San Diego, California

Submitted 30 September 2008; accepted in final form 8 December 2008

Yuan H, Perry CN, Huang C, Iwai-Kanai E, Carreira RS, Glembotski CC, Gottlieb RA. LPS-induced autophagy is mediated by oxidative signaling in cardiomyocytes and is associated with cytoprotection. Am J Physiol Heart Circ Physiol 296: H470 –H479, 2009. First published December 19, 2008; doi:10.1152/ajpheart.01051.2008.—Bacterial endotoxin lipopolysaccharide (LPS) is responsible for the multiorgan dysfunction that characterizes septic shock and is causal in the myocardial depression that is a common feature of endotoxemia in patients. In this setting the myocardial dysfunction appears to be due, in part, to the production of proinflammatory cytokines. A line of evidence also indicates that LPS stimulates autophagy in cardiomyocytes. However, the signal transduction pathway leading to autophagy and its role in the heart are incompletely characterized. In this work, we wished to determine the effect of LPS on autophagy and the physiological significance of the autophagic response. Autophagy was monitored morphologically and biochemically in HL-1 cardiomyocytes, neonatal rat cardiomyocytes, and transgenic mouse hearts after the administration of bacterial LPS or TNF-␣. We observed that autophagy was increased after exposure to LPS or TNF-␣, which is induced by LPS. The inhibition of TNF-␣ production by AG126 significantly reduced the accumulation of autophagosomes both in cell culture and in vivo. The inhibition of p38 MAPK or nitric oxide synthase by pharmacological inhibitors also reduced autophagy. Nitric oxide or H2O2 induced autophagy in cardiomyocytes, whereas Nacetyl-cysteine, a potent antioxidant, suppressed autophagy. LPS resulted in increased reactive oxygen species (ROS) production and decreased total glutathione. To test the hypothesis that autophagy might serve as a damage control mechanism to limit further ROS production, we induced autophagy with rapamycin before LPS exposure. The activation of autophagy by rapamycin suppressed LPSmediated ROS production and protected cells against LPS toxicity. These findings support the notion that autophagy is a cytoprotective response to LPS-induced cardiomyocyte injury; additional studies are needed to determine the therapeutic implications.

Although the proinflammatory cytokines can act directly or indirectly to cause cardiac myocyte injury, the etiology of and strategies to prevent/treat myocardial dysfunction in the setting of sepsis in humans remain elusive. Recently, LPS was reported to stimulate mitochondrial biogenesis and autophagy in neonatal rat cardiomyocytes (15). Autophagy is a highly regulated intracellular degradation process by which cells remove cytosolic long-lived proteins and damaged organelles (23, 25, 28). Activation of autophagy has been observed in a variety of heart diseases including cardiac hypertrophy and ischemia-reperfusion injury, suggesting that autophagy may play an important role in myocardial dysfunction (reviewed in Ref. 11). When autophagy is initiated, cytoplasmic constituents are sequestered into the autophagosome, a closed double membrane vacuole. The autophagosome eventually fuses with a lysosome, forming an autolysosome in which the contents are degraded and recycled for protein synthesis (4). Inappropriate stimulation of autophagy may facilitate cell death, referred to as autophagic cell death or type II programmed cell death (5), whereas other studies have pointed to a protective role for autophagy (12–14). Based on the latter observations, we elected to evaluate the signal transduction pathways involved in the induction of myocardial autophagy by LPS and to assess the biological role of autophagy in LPS-associated myocardial injury. MATERIALS AND METHODS

MYOCARDIAL DEPRESSION IS A common feature of endotoxemia in patients (31, 32) and is a major cause of morbidity and mortality in septic patients. This myocardial dysfunction is due, in part, to the production of proinflammatory cytokines induced by bacterial endotoxin lipopolysaccharide (LPS) (15). LPS is also responsible for the multiorgan dysfunction that characterizes septic shock (9, 18, 27, 29). Although the precise mechanism is incompletely understood, increasing evidence suggests that LPS-induced myocardial dysfunction is mediated by multiple proinflammatory mediators, such as TNF-␣, IL-1␤, and cytokine-inducible nitric oxide synthase (iNOS) (3, 19).

Reagents. LPS (from Escherichia coli) was purchased from Sigma. Rapamycin, tyrphostin AG126, Bafilomycin A1, chloroquine, NGmonomethyl-L-arginine (L-NMMA), and sodium nitroprusside (SNP) were purchased from EMD Biosciences. 5-(and 6-) Chloromethyl-2⬘,7⬘dichlorohydrofluorescein diacetate acetyl ester (CM-H2DCFDA) was obtained from Invitrogen. Cell culture and transfections. Rat neonatal cardiomyocytes were isolated and maintained as described previously (21, 41). The murine atrial-derived cardiac HL-1 cells (2) were plated in gelatin/fibronectin-coated culture vessels and maintained in Claycomb medium (JRH Biosciences) supplemented with 10% fetal bovine serum, 0.1 mM norepinephrine, 2 mM L-glutamine, 100 U/ml penicillin, 100 U/ml streptomycin, and 0.25 ␮g/ml amphotericin B. Cells were transfected with green fluorescent protein-microtubule-associated protein light chain 3 (GFP-LC3) using Effectene transfection reagents (Qiagen) according to the manufacturer’s instructions, achieving at least 40% transfection efficiency. Widefield fluorescence microscopy and autophagy determination. GFP-LC3-transfected cells were visualized by fluorescence microscopy. To determine the activation of autophagy, GFP-LC3-expressing cells were incubated with LPS or TNF-␣ in culture medium for the

Address for reprint requests and other correspondence: R. A. Gottlieb, Bioscience Center, San Diego State Univ., 5500 Campanile Dr., San Diego, CA 92182-4650 (e-mail: [email protected]).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

lipopolysaccharide; HL-1 cardiac myocyte; green fluorescent proteinmicrotubule-associated protein light chain 3; oxidative stress

H470

0363-6135/09 $8.00 Copyright © 2009 the American Physiological Society

http://www.ajpheart.org

H471

LPS-INDUCED AUTOPHAGY IN CARDIOMYOCYTES

indicated times. Cells were fixed with 4% paraformaldehyde in PBS (pH 7.4) at room temperature for 15 min. For analysis of autophagy activation in a cell population, cells were inspected at ⫻60 magnification and the percentage of cells showing numerous GFP-LC3 puncta (⬎20 dots/cell) was scored as previously described (1). To analyze autophagic flux, cells were treated in the absence or presence of the vacuolar H⫹-ATPase inhibitor Bafilomycin A1 (or the lysosomotropic alkalinizing agent chloroquine) to inhibit autophagosomelysosome fusion (20, 44). A minimum of 200 cells were scored for each condition in at least three independent experiments. To quantify the number of GFP-LC3 puncta in a single cell, Z-stack images of single GFP-LC3-expressing cells were collected using 60⫻ oil objective and deconvolved to each stack image. Maximal projection images were produced, and the number of GFP-LC3 puncta in a cell was determined using ImageJ Software (National Institutes of Health). More than 10 cells were analyzed for each condition. Measurement of reactive oxygen species and terminal deoxynucleotidyl transferase dUTP-mediated nick-end labeling. Cells were plated at 10,000 cells/well in 96-well plates and grown to confluence. Cells were pretreated with indicated reagents for indicated time and then LPS (1 ␮g/ml) or TNF-␣ (50 ng/ml) was added for an additional 4 h. Cells were incubated with CM-H2DCFDA (10 ␮M) for 30 min, and fluorescence was then measured by a SPECTRAmax fluorimeter (Molecular Devices) set to 490 nm excitation and 520 nm emission at 37°C. Treatments were done in replicates of six, and three independent experiments were performed. Terminal deoxynucleotidyl trans-

ferase dUTP-mediated nick-end labeling (TUNEL) assay was performed as previously described (19, 21). Transgenic mCherry-LC3 mice. LC3-mCherry mice (20) were injected with AG126 (1.5 mg/kg ip) or chloroquine (10 mg/kg ip) or vehicle; one hour later LPS (1.5 mg/kg ip) was injected. Four hours after LPS injection, mice were euthanized with pentobarbital and the hearts were excised and embedded in optimal cutting temperature medium for cryosectioning and fluorescence microscopy. Where indicated, monodansylcadaverine (MDC; 1.5 mg/kg ip) was injected 60 min before euthanization. All procedures were carried out in accordance with Institutional guidelines and approved by the Institutional Animal Care and Use Committee. Western blotting. After indicated treatments, cells were harvested and solubilized in lysis buffer or hearts from pentobarbital-anesthetized mice were excised and Polytron homogenized in lysis buffer (see supplemental methods; all supplemental material can be found with the online version of this article). After clarification by centrifugation, samples were resolved by SDS-PAGE and transferred to nitrocellulose membranes. Blots were probed with anti-phospho-p38 antibody or total p38 antibody. Statistics. The probability of statistically significant differences between two experimental groups was determined by both paired Student’s t-test and ANOVA. Values are expressed as means ⫾ SE of at least three independent experiments unless stated otherwise, and P values are reported for ANOVA. A value of P ⬍ 0.05 was considered significant.

Fig. 1. LPS induced autophagy in cardiac cells. Green fluorescent protein (GFP)-microtubule-associated protein light chain 3 (LC3)-expressing HL-1 cells were exposed to LPS in medium, and autophagy was assessed. A: representative images of GFPLC3-labeled puncta in HL-1 cells 4 h after exposure to 1 ␮g/ml LPS. B: histogram showing the number of autophagosomes per cell under basal conditions and after LPS exposure (x-axis, number of autophagosomes per cell; y-axis, number of cells observed with the indicated number of autophagosomes). C and D: autophagic flux in HL-1 cells was determined in the absence (black bars) or presence (hatched bars) of 100 nM Bafilomycin A1 (C) for 4 h or 3 ␮M chloroquine (hatched bars; D) for 2 h. The percentage of cells with numerous GFP-LC3 dots is represented as “%autophagy.” E: HL-1 cells were treated with 1 ␮g/ml LPS for 4 h or starved (stv) for 3.5 h in Krebs-Henseleit buffer in the absence or presence (⫹) of Bafilomycin A1. Cell lysates were separated on 10 –20% SDS-PAGE and analyzed by Western blot using anti-LC3 antibody. F: GFP-LC3 expressing cells were treated with indicated concentrations of LPS for 4 h. Data represented as means ⫾ SE from at least 3 independent experiments. *P ⬍ 0.01; **P ⬍ 0.001. Scale bar, 10 ␮m. Con, control.

AJP-Heart Circ Physiol • VOL

296 • FEBRUARY 2009 •

www.ajpheart.org

H472


RESULTS

LPS induces autophagy in cardiomyocytes. HL-1 cardiomyocytes were transfected with GFP-LC3. Cultures were later (48 h) exposed to LPS for 4 h and then observed for autophagy as indicated by the redistribution GFP-LC3 from a diffuse distribution to a punctate pattern. Under baseline conditions, transfected cells displayed diffuse cytoplasmic GFP-LC3; LPS exposure triggered the formation of numerous GFP-LC3 puncta in HL-1 cells (Fig. 1A). The number of autophagosomes per cell was scored to establish a threshold. We observed a distinctive difference in the number of autophagosomes per cell when comparing unstimulated cells with those exposed to LPS. In unstimulated conditions, the majority of cells in culture medium had 0 –10 GFP dots/cell and none had ⬎20 dots/cell. In contrast, most cells exposed to LPS had ⬎20 dots/cell, with a median of 30 – 60 dots/cell (Fig. 1B). Autophagy was scored as the percentage of cells in the population that exhibited numerous puncta (⬎20 dots/cell) and expressed as “%autophagy” for simplicity. To measure autophagic flux, cells were incubated with LPS in the absence or presence of vacuolar

H⫹-ATPase inhibitor Bafilomycin A1 or chloroquine to prevent lysosomal acidification (30, 38), resulting in the accumulation of autophagosomes. This allows the estimation of autophagic flux, which is important to distinguish increased autophagosome formation from impaired degradation. Here the addition of Bafilomycin A1 or chloroquine increased the percentage of cells exhibiting numerous puncta in cells exposed to LPS (Fig. 1, C and D). These results indicate that LPS increases autophagic flux. In addition, the conversion of LC3-I to LC3-II, an indicator of autophagy activation, was assessed by immunoblot analysis using LC3 antibody (Fig. 1E). LC3-II abundance was increased by LPS. Starvation of HL-1 cells was done in parallel as a positive control. Because autophagosomes turn over rapidly, LC3-II levels do not appear to change substantially unless lysosomal degradation is blocked, consistent with the flux measurements made by microscopy. Autophagy increased dose dependently in relation to LPS (Fig. 1F). We obtained essentially identical results with primary neonatal cardiomyocytes (supplemental Fig. 1A). LPS has no effect on cell viability during the 4-h incubation (data not shown). These

Fig. 2. LPS induced autophagy via TNF-␣ production. Cells were cultured under the same conditions as described in Fig. 1 and treated with 50 ng/ml TNF-␣. A: representative images showing activation of autophagy in HL-1 cells 4 h after addition of 50 ng/ml TNF-␣. Scale bar, 10 ␮m. B: population distribution of cells containing various numbers of autophagosomes after exposure to TNF-␣. C: determination of autophagic flux in response to TNF-␣ in the absence (black bars) or presence (hatched bars) of Bafilomycin A1. *P ⬍ 0.01. D: Western blot of LC3-I/LC3-II processing under the same conditions. E: dose-dependent stimulation of autophagy by TNF-␣. F: effect of tyrosine kinase inhibitor tyrphostin AG126 on LPSor TNF-␣-induced autophagy. HL-1 cells were incubated with LPS or TNF-␣ in the absence or presence of 30 ␮M AG126 for 4 h in the presence of Bafilomycin A1, and autophagy was scored by fluorescence microscopy. ns, Not significant. *P ⬍ 0.01.


296 • FEBRUARY 2009 •

www.ajpheart.org


H473

(supplemental Fig. 1B). We also determined the number of autophagosomes per cell and observed a pattern similar to LPS-induced autophagy (Fig. 2B). In the presence of Bafilomycin A1, the percentage of cells showing numerous GFPLC3 puncta was further increased, indicating the increased formation of autophagosomes rather than impaired clearance (Fig. 2C). Consistent with the microscopy findings, Western blotting revealed increased LC3-II under the same conditions (Fig. 2D). Autophagy in response to TNF-␣ was concentration dependent (Fig. 2E). To determine whether LPS-mediated autophagy required TNF-␣, we used the tyrosine kinase inhibitor tyrphostin AG126, which inhibits JAK2 and thereby prevents TNF-␣ production (3). When cultures were exposed to TNF-␣ or LPS in the presence of AG126, we observed that AG126 significantly inhibited autophagy induced by LPS but not by TNF-␣ (Fig. 2F). These data suggest that LPS-induced autophagy is mediated by TNF-␣. p38 MAPK and nitric oxide synthase are required for LPS-induced autophagy. LPS and TNF-␣ activate multiple molecular pathways in cardiomyocytes and other cell types. To test whether p38 MAPK is required for the activation of autophagy by LPS, HL-1 cells were stimulated with 1 ␮g/ml LPS in the presence or absence of the p38 MAPK inhibitor SB203580. Activation of p38 MAPK was examined by immunoblotting. As early as 10 min after the addition of LPS, phosphorylated p38 MAPK could be detected (Fig. 3A),

Fig. 3. MAPK pathway and nitric oxide synthase (NOS) activity is involved in LPS-mediated autophagy. A: p38 MAPK is activated by LPS. HL-1 cells were preincubated with 10 ␮M SB203580 (SB) for 2 h followed by 1 ␮g/ml LPS for the indicated time. Cell lysates were resolved on 10 –20% SDS-PAGE and blotted with anti-phospho (p)-p38 or anti-p38 antibodies. B: effect of p38 MAPK inhibitor and NOS inhibitor on autophagy. HL-1 cells were pretreated with 10 ␮M SB or 10 ␮M NG-monomethyl-L-arginine (L-NMMA) for 2 h followed by LPS for 4 h in the absence (black bars) or presence (hatched bars) of Bafilomycin A1. *P ⬍ 0.001 compared with control; **P ⬍ 0.001 compared with LPS treated. C: effect of p38 MAPK inhibitor and NOS inhibitor on TNF-␣ induced autophagy in HL-1 cells. *P ⬍ 0.01 compared with control; **P ⬍ 0.001 compared with TNF-␣.

results clearly demonstrate that LPS exposure triggers autophagy in cardiomyocytes. LPS induces autophagy via TNF-␣. It has been shown that LPS induces the expression of proinflammatory cytokines such as TNF-␣, which is associated with the pathogenesis of sepsis (3). To examine whether TNF-␣ is involved in LPS-induced autophagy, cultures expressing GFP-LC3 were exposed to TNF-␣ for 4 h. TNF-␣ induced the formation of GFP-LC3 puncta in HL-1 cells (Fig. 2A) and neonatal cardiomyocytes AJP-Heart Circ Physiol • VOL

Fig. 4. LPS-induced autophagy is mediated by oxidative stress. A: GFP-LC3expressing HL-1 cells were preincubated with 2 mM N-acetyl-cysteine (NAC) for 2 h before 4 h incubation with 1 ␮g/ml LPS. Cells were fixed, and GFP-positive cells were scored for autophagy. Data are means ⫾ SE of 3 independent experiments. *P ⬍ 0.01 compared with control; **P ⬍ 0.001 compared with LPS. B: GFP-LC3 HL-1 cells were preincubated without or with NAC before addition of 100 ␮M H2O2 or 200 ␮M sodium nitroprusside (SNP) for 4 h. Data are means ⫾ SE of 3 independent experiments. *P ⬍ 0.01; **P ⬍ 0.001 compared with H2O2 or SNP.

296 • FEBRUARY 2009 •

www.ajpheart.org

H474


whereas SB203580 inhibited the phosphorylation of p38. These results indicated that LPS activated p38 MAPK in HL-1 cells. This phosphorylation is a quick and transient process, since p38 phosphorylation returned to baseline after 60 min. We next tested whether SB203580 had any effect on the induction of autophagy. Pretreatment with SB203580 significantly inhibited LPS-induced autophagy in HL-1 cells (Fig. 3B) as well as in neonatal cardiomyocytes (supplemental Fig. 2A). In addition, SB203580 similarly inhibited TNF-␣-induced autophagy (Fig. 3C and supplemental Fig. 2B). In cardiomyocytes, LPS exposure induces the upregulation of nitric oxide (NO) synthase (NOS) (19), leading to increased free radical NO generation (22). We reasoned that NOS may also be involved in LPS signaling to induce autophagy. Therefore, we examined the effect of the NOS inhibitor L-NMMA on LPS- or TNF-␣-induced autophagy in cardiac cells. Cells expressing GFP-LC3 were pretreated for 2 h with the NOS inhibitor and were then exposed to LPS or TNF-␣ for 4 h. Pretreatment with 10 ␮M L-NMMA significantly reduced LPSor TNF-␣-mediated autophagy in HL-1 cells (Fig. 3, B and C)

and in neonatal cardiomyocytes (supplemental Fig. 2, A and B). These data suggest that the p38 MAPK pathway and NOS are involved in LPS- or TNF-␣-induced autophagy. LPS-mediated autophagy is induced by oxidative stress. LPS rapidly stimulates TNF-␣ secretion and production of ROS in macrophages (16, 43). Several lines suggest ROS production follows the activation of p38 MAPK. Upregulation of NOS also increases oxidative stress (22). To determine whether oxidative stress is involved in LPS-mediated induction of autophagy in HL-1 cells, we tested the effect of the sulfhydryl antioxidant reagent N-acetyl-cysteine (NAC). Preincubation of HL-1 cells or neonatal cardiomyocytes with NAC dramatically blocked LPS-induced autophagy (Fig. 4A and supplemental Fig. 3A). This suggests that oxidative stress plays a role in inducing autophagy. To further confirm that reactive oxygen species (ROS)/reactive nitrogen species (RNS) induce autophagy directly, we treated HL-1 cells with 100 ␮M H2O2 or 200 ␮M SNP as a NO donor. As shown in Fig. 4B, H2O2 or SNP triggered autophagy, whereas preincubation with NAC blocked the effects of H2O2 and SNP.

Fig. 5. Rapamycin (Rap or Rapa) inhibits reactive oxygen species (ROS) production and protects against LPS-mediated cell death. A: HL-1 cells were treated with LPS with or without 5 ␮M rapamycin for 4 h, and ROS production was determined using 5-(and 6-) chloromethyl-2⬘,7⬘-dichlorohydrofluoroscein diacetate acetyl ester. B: HL-1 cells were incubated without or with LPS for 4 h, and then total glutathione (GSH) was measured spectrophotometrically. C: HL-1 cells were exposed to LPS with or without 5 ␮M rapamycin pretreatment. After 48 h, cell death was scored with propidium iodide (PI) and Hoechst 33342. Representative images of nuclear staining are shown. D: quantification of cell death. Results are presented as fold increase compared with untreated cells. Data are means ⫾ SE of 3 independent experiments. *P ⬍ 0.01 compared with control; **P ⬍ 0.001 compared with LPS. E: neonatal rat cardiac myocytes were cultured in serum-free media in the presence or the absence of wortmannin (Wm; 10⫺7 M) and/or LPS (10⫺4 M) as indicated for 48 h. Values are means of 4 independent experiments. **P ⬍ 0.001 compared with control; *P ⬍ 0.01 compared with LPS. TUNEL Pos, terminal deoxynucleotidyl transferase dUTP-mediated nickend labeling positive; RFU, relative fluorescent units.


296 • FEBRUARY 2009 •

www.ajpheart.org


LPS, TNF, and ROS/RNS all activate autophagy, and it is well known that LPS triggers ROS production. We confirmed this in HL-1 cells (Fig. 5A). Furthermore, LPS exposure caused a reduction of total glutathione (reduced ⫹ oxidized) by 49% compared with untreated cells (n ⫽ 4; P ⬍ 0.001; Fig. 5B). Oxidative stress is not the only mechanism to trigger autophagy. To determine whether rapamycin, an agent that induces autophagy through the inhibition of mammalian target of rapamycin (mTOR), also triggers ROS production, we treated cells with rapamycin and measured ROS production with H2DCFDA, which is converted to a fluorescent product upon the reaction with H2O2. HL-1 cells were preincubated with 5 ␮M rapamycin for 1 h before 4 h exposure to LPS. Rapamycin not only decreased the basal level of ROS production in HL-1 cells but also significantly suppressed LPS-mediated ROS production (Fig. 5A). Taken together, these results suggest that oxidative stress induced by LPS triggers autophagy and the induction of autophagy (e.g., by rapamycin) helps to mitigate oxidative stress. It is consistent with the notion that autophagy may be a protective response elicited by the cell in response to injury mediated by LPS and TNF-␣. Autophagy modulates LPS-mediated cell death. Previous studies have shown that LPS and TNF-␣ trigger apoptosis in adult cardiomyocytes (3). To determine whether autophagy protects against cell death caused by LPS, we determined the effect of rapamycin on cell death. HL-1 cells were incubated with LPS in the absence or presence of rapamycin for 48 h, and cell death was examined using nuclear staining dyes Hoechst 34422 and propidium iodide (Fig. 5C). LPS induced cell death in a dose-dependent manner, whereas rapamycin remarkably protected cells from LPS toxicity (Fig. 5D). In contrast, the inhibition of autophagy by wortmannin increased LPS-medi-

H475

ated cell death (Fig. 5E). Thus our results indicate that the induction of autophagy protects cells against LPS toxicity, potentially through the amelioration of ROS production. The mechanism by which autophagy might suppress ROS production remains to be elucidated. LPS induces mitochondrial damage and autophagy in vivo. LPS was administered to mice via intraperitoneal injection at doses ranging from 0.5 to 1.5 mg/kg. Electron microscopy of heart sections obtained 4 h after injection revealed widespread damage to mitochondria and numerous autophagosomes (Fig. 6). To more definitively investigate autophagy in vivo, we used the previously described transgenic ␣-MCH-mCherry-LC3 mice, which express mCherry-LC3 specifically in cardiomyocytes (20). LPS injection resulted in a dramatic accumulation of autophagosomes within 4 h (Fig. 7A). To distinguish between increased autophagosome formation and decreased lysosomal clearance, we injected mice with chloroquine, which prevents autophagosome-lysosome fusion. Under this condition, if the increased number of autophagosomes was due to impaired clearance, the addition of chloroquine would not increase the number of autophagosomes. If, however, autophagosome formation is increased by LPS, the number of autophagosomes will be even greater after chloroquine administration. Comparison of autophagosome number under the two conditions allows an estimation of flux. As shown in Fig. 7B, LPS stimulates increased autophagic flux. These findings were confirmed by western blotting for LC3 (Fig. 7C). Cathepsin D, a major lysosomal protease, is often upregulated in parallel with autophagy. We examined cathepsin D immunoreactivity in control and LPS-treated hearts. Immunofluorescence revealed partial colocalization of cathepsin D and mCherry-LC3 puncta, indicating the formation of autophagolysosomes (Fig. 7D).

Fig. 6. LPS causes mitochondrial damage in heart tissue. Mice received LPS (1.5 mg/kg ip), and hearts were removed 4 h later and processed for electron microscopy. Representative images show swollen mitochondria with ruptured outer membranes (arrowheads) and autophagosomes (arrows). Control heart tissue is shown for comparison.


296 • FEBRUARY 2009 •

www.ajpheart.org

H476


Fig. 7. LPS stimulated autophagic flux and upregulated cathepsin D in vivo. Hearts of mCherry-LC3 transgenic mice were harvested 4 h after LPS injection (1.5 mg/kg ip injection) with or without concurrent administration of chloroquine (CQ) to prevent lysosomal degradation of autophagosomes. A: fluorescence micrographs of heart tissues under indicated conditions. Bright dots indicate autophagosomes. B: quantitative analysis of mCherry puncta from each condition. *P ⬍ 0.001 compared with control. C: immunoblot of LC3 in heart tissue lysates prepared from mice treated as described above. D: immunodetection of cathepsin D in sections of mCherry-LC3 transgenic mouse hearts treated as above (no CQ). Cathepsin D (green) colocalized with mCherry-LC3 puncta (red) as indicated by arrows. SS, steady state.

To determine whether the induction of autophagy in vivo is due to LPS or the downstream cytokine TNF-␣, we treated transgenic mice with the tyrosine kinase inhibitor AG126 or vehicle 1 h before LPS injection. After 4 h LPS, mice were euthanized and heart tissue was processed for fluorescence microscopy to visualize autophagosomes containing mCherryLC3. LPS-treated mice exhibited increased the abundance of mCherry-LC3 and numerous puncta compared with control hearts (Fig. 8A). The administration of AG126, which prevents TNF-␣ induction by LPS, greatly decreased mCherry-LC3 fluorescence and puncta formation (Fig. 8B). Our results suggest that LPS induces autophagy in vivo via TNF-␣, consistent with our cell culture results. DISCUSSION

LPS, a major component of bacterial outer walls, can have profound and diverse effects in mammals. It is responsible for the multiorgan dysfunction that characterizes septic shock. It is well documented that LPS induces TNF-␣ release from cardiomyocytes, and TNF-␣ in turn triggers apoptosis in cardiomyocytes (3). In this study we investigated the signal transduction pathway leading from LPS to autophagy, as well as the functional signifAJP-Heart Circ Physiol • VOL

icance of autophagy in myocytes. The role of autophagy in the heart has been controversial, and whether it is protective or harmful in a particular setting is unclear. In this study, we used primary neonatal cardiomyocytes, the HL-1 cardiomyocyte cell line, and mCherry-LC3 transgenic mice to explore these questions. It is important to note that HL-1 cells and neonatal cardiomyocytes responded similarly under all the conditions we studied. We therefore consider the HL-1 cell line to be quite useful for studies of signal transduction in a cardiomyocyte model system; as an immortalized and easily transfectable cell line it offers obvious advantages over neonatal cardiomyocytes. Our studies showed that LPS or TNF-␣ can induce autophagy in cardiomyocytes in cell culture and in vivo. Using pharmacological agents and molecular targeting, we probed the signaling pathway of LPS-induced autophagy. The interactions are depicted in Fig. 9. We showed that autophagy induced by LPS in cardiac cells is mediated by p38 MAPK, NOS, and the formation of ROS, with resultant depletion of glutathione. Pharmacological inhibition of p38 and NOS suppressed ROS/RNS production and induction of autophagy. Importantly, rapamycin, which activates autophagy through the inhibition of mTOR, suppressed ROS production and protected HL-1 cells against LPS toxicity.

296 • FEBRUARY 2009 •

www.ajpheart.org


H477

Fig. 8. Effect of AG126 on induction of autophagy by LPS in vivo. mCherry-LC3 transgenic mice were pretreated with the tyrosine kinase inhibitor AG126 (0.5 mg/kg ip) or vehicle 1 h before LPS (1.5 mg/kg ip); hearts were harvested 4 h later. A: fluorescence micrographs of heart tissues under indicated conditions. Bright dots indicate autophagosomes. B: quantitative analysis of mCherry puncta from each condition, normalized to the number of nuclei in each field. *P ⬍ 0.001 compared with control; **P ⬍ 0.0001 compared with LPS.

Fig. 9. Scheme of LPS-induced autophagy in cardiomyocytes. In cardiomyocytes, LPS induces activation of p38 MAPK and induces the expression of TNF-␣ and NOS. Multiple signaling pathways converge to ROS/reactive nitrogen species (RNS) production, which results in activation of autophagy through Atg4. AJP-Heart Circ Physiol • VOL

The p38 MAPK pathway is involved in a number of pathological stress conditions including ischemia-reperfusion, inflammation, and LPS. Signaling through p38 can trigger apoptosis (24, 33, 42), and here we showed that LPS-activated p38 MAPK plays a role in regulating autophagy. SB203580, a selective and potent inhibitor of p38, suppressed autophagy after exposure to LPS or TNF-␣. A limitation of this study is that we cannot distinguish between the different isoforms of p38, which may have different biological effects in certain pathways (6). In cardiomyocytes, LPS induces iNOS (19) as well as TNF-␣ (10). LPS stimulates the production of NO and superoxide in rat hearts (22). Inhibition of NOS by L-NMMA suppressed the induction of autophagy by LPS as well as TNF-␣ induced autophagy, suggesting that NOS also functions downstream of TNF-␣. A limitation of our study is that L-NMMA does not allow us to distinguish between the isoforms of NOS that may be responsible for NO production in this context. Inhibition of TNF-␣ secretion by AG126 suppressed LPS-mediated autophagy both in vitro and in vivo, suggesting LPS induces autophagy through a TNF-␣-dependent pathway. The results suggest that p38 MAPK activation is an early and required event, which may be upstream and downstream of TNF-␣ production and which is clearly required for ROS generation and the induction of autophagy. LPS induces oxidative stress in macrophage cell lines and cardiomyocytes (15, 17, 34). TNF-␣ also shifts the intracellular

296 • FEBRUARY 2009 •

www.ajpheart.org

H478


redox potential (8, 37). The production of ROS is attributed to mitochondrial injury, which follows the production of ceramide mediated by TNF-␣ signaling (7, 35). LPS and TNF-␣ triggered ROS production in HL-1 cells. Both H2O2 and NO triggered autophagy in HL-1 cells. The antioxidant reagent NAC significantly suppressed the induction of autophagy by LPS, H2O2, and NO. This is consistent with the findings of Elazar and colleagues (36), who showed that autophagy is induced through the oxidative inactivation of Atg4. These studies do not reveal whether p38 and NOS directly lead to autophagy or whether they lead to mitochondrial damage with a secondary repair response of autophagy. Further work will be needed to answer this question. Autophagy, like a double-edged sword, plays a role in cell survival as well as cell death. Although autophagy is a protective response in some settings (26), it also can promote cell death in others (39, 40, 45). The induction of autophagy could be part of a cellular program leading to cell death, or it could reflect attempts by the cell to repair itself through the removal of damaged organelles including mitochondria. Autophagy may be induced to aid in the removal of damaged mitochondria. Hickson-Bick et al. (15) described mitochondrial biogenesis in cardiomyocytes after LPS exposure; this may reflect an effort to replace the mitochondria that were eliminated by autophagy. In HL-1 cells, the induction of autophagy by rapamycin administration before LPS exposure suppressed ROS production and protected cells against LPS-mediated cell death. In contrast, interfering with autophagy by wortmannin administration exacerbated LPS-mediated cell death. These results are consistent with the notion that autophagy is a protective mechanism in this setting and in some manner limits the production of harmful ROS, either by removing damaged mitochondria or by supporting de novo glutathione biosynthesis through the delivery of amino acids. Our studies here provide some insights for understanding molecular signaling and suggest an important protective role for autophagy in LPS-induced toxemia. Further elucidation of the relationship between autophagy, contractile function, and cell death in the setting of LPS-induced myocardial dysfunction could lead to the development of new therapeutic modalities to treat patients with sepsis-induced heart failure in humans. GRANTS This work was supported by NIH Grants P01-HL085577 and R01AG033283 (to R. A. Gottlieb) and a grant from the Ministry of Education, Science, and Culture of Japan (to E. Iwai-Kanai). REFERENCES 1. Brady NR, Hamacher-Brady A, Yuan H, Gottlieb RA. The autophagic response to nutrient deprivation in the HL-1 cardiac myocyte is modulated by Bcl-2 and sarco/endoplasmic reticulum calcium stores. FEBS J 274: 3184 –3197, 2007. 2. Claycomb WC, Lanson NA Jr, Stallworth BS, Egeland DB, Delcarpio JB, Bahinski A, Izzo NJ Jr. HL-1 cells: a cardiac muscle cell line that contracts and retains phenotypic characteristics of the adult cardiomyocyte. Proc Natl Acad Sci USA 95: 2979 –2984, 1998. 3. Comstock KL, Krown KA, Page MT, Martin D, Ho P, Pedraza M, Castro EN, Nakajima N, Glembotski CC, Quintana PJ, Sabbadini RA. LPS-induced TNF-alpha release from and apoptosis in rat cardiomyocytes: obligatory role for CD14 in mediating the LPS response. J Mol Cell Cardiol 30: 2761–2775, 1998. 4. Cuervo AM. Autophagy: in sickness and in health. Trends Cell Biol 14: 70 –77, 2004. AJP-Heart Circ Physiol • VOL

5. Debnath J, Baehrecke EH, Kroemer G. Does autophagy contribute to cell death? Autophagy 1: 66 –74, 2005. 6. Degousee N, Martindale J, Stefanski E, Cieslak M, Lindsay TF, Fish JE, Marsden PA, Thuerauf DJ, Glembotski CC, Rubin BB. MAP kinase kinase 6-p38 MAP kinase signaling cascade regulates cyclooxygenase-2 expression in cardiac myocytes in vitro and in vivo. Circ Res 92: 757–764, 2003. 7. Di Paola M, Cocco T, Lorusso M. Ceramide interaction with the respiratory chain of heart mitochondria. Biochemistry 39: 6660 – 6668, 2000. 8. Garg AK, Aggarwal BB. Reactive oxygen intermediates in TNF signaling. Mol Immunol 39: 509 –517, 2002. 9. Gates DM. Cardiac dysfunction in septic shock and multiple organ dysfunction syndrome. Crit Care Nurs Q 16: 39 – 48, 1994. 10. Grandel U, Fink L, Blum A, Heep M, Buerke M, Kraemer HJ, Mayer K, Bohle RM, Seeger W, Grimminger F, Sibelius U. Endotoxin-induced myocardial tumor necrosis factor-alpha synthesis depresses contractility of isolated rat hearts: evidence for a role of sphingosine and cyclooxygenase2-derived thromboxane production. Circulation 102: 2758 –2764, 2000. 11. Gustafsson AB, Gottlieb RA. Eat your heart out: role of autophagy in myocardial ischemia/reperfusion. Autophagy 4: 416 – 421, 2008. 12. Hamacher-Brady A, Brady NR, Gottlieb RA. Enhancing macroautophagy protects against ischemia/reperfusion injury in cardiac myocytes. J Biol Chem 281: 29776 –29787, 2006. 13. Hamacher-Brady A, Brady NR, Gottlieb RA, Gustafsson AB. Autophagy as a protective response to Bnip3-mediated apoptotic signaling in the heart. Autophagy 2: 307–309, 2006. 14. Hamacher-Brady A, Brady NR, Logue SE, Sayen MR, Jinno M, Kirshenbaum LA, Gottlieb RA, Gustafsson AB. Response to myocardial ischemia/reperfusion injury involves Bnip3 and autophagy. Cell Death Differ 14: 146 –157, 2006. 15. Hickson-Bick DL, Jones C, Buja LM. Stimulation of mitochondrial biogenesis and autophagy by lipopolysaccharide in the neonatal rat cardiomyocyte protects against programmed cell death. J Mol Cell Cardiol 44: 411– 418, 2008. 16. Hsu HY, Twu YC. Tumor necrosis factor-alpha-mediated protein kinases in regulation of scavenger receptor and foam cell formation on macrophage. J Biol Chem 275: 41035– 41048, 2000. 17. Hsu HY, Wen MH. Lipopolysaccharide-mediated reactive oxygen species and signal transduction in the regulation of interleukin-1 gene expression. J Biol Chem 277: 22131–22139, 2002. 18. Hung J, Lew WY. Temporal sequence of endotoxin-induced systolic and diastolic myocardial depression in rabbits. Am J Physiol Heart Circ Physiol 265: H810 –H819, 1993. 19. Iwai-Kanai E, Hasegawa K, Fujita M, Araki M, Yanazume T, Adachi S, Sasayama S. Basic fibroblast growth factor protects cardiac myocytes from iNOS-mediated apoptosis. J Cell Physiol 190: 54 – 62, 2002. 20. Iwai-Kanai E, Yuan H, Huang C, Sayen MR, Perry-Garza CN, Kim L, Gottlieb RA. A method to measure cardiac autophagic flux in vivo. Autophagy 4: 322–329, 2008. 21. Iwaki K, Sukhatme VP, Shubeita HE, Chien KR. Alpha- and betaadrenergic stimulation induces distinct patterns of immediate early gene expression in neonatal rat myocardial cells. Fos/jun expression is associated with sarcomere assembly; Egr-1 induction is primarily an alpha 1-mediated response. J Biol Chem 265: 13809 –13817, 1990. 22. Khadour FH, Panas D, Ferdinandy P, Schulze C, Csont T, Lalu MM, Wildhirt SM, Schulz R. Enhanced NO and superoxide generation in dysfunctional hearts from endotoxemic rats. Am J Physiol Heart Circ Physiol 283: H1108 –H1115, 2002. 23. Klionsky DJ, Ohsumi Y. Vacuolar import of proteins and organelles from the cytoplasm. Annu Rev Cell Dev Biol 15: 1–32, 1999. 24. Lee JC, Young PR. Role of CSB/p38/RK stress response kinase in LPS and cytokine signaling mechanisms. J Leukoc Biol 59: 152–157, 1996. 25. Levine B. Autophagy in development, tumor suppression, and innate immunity. Harvey Lect 99: 47–76, 2003. 26. Lum JJ, DeBerardinis RJ, Thompson CB. Autophagy in metazoans: cell survival in the land of plenty. Nat Rev Mol Cell Biol 6: 439 – 448, 2005. 27. Meng X, Ao L, Meldrum DR, Cain BS, Shames BD, Selzman CH, Banerjee A, Harken AH. TNF-␣ and myocardial depression in endotoxemic rats: temporal discordance of an obligatory relationship. Am J Physiol Regul Integr Comp Physiol 275: R502–R508, 1998. 28. Mizushima N, Ohsumi Y, Yoshimori T. Autophagosome formation in mammalian cells. Cell Struct Funct 27: 421– 429, 2002.

296 • FEBRUARY 2009 •

www.ajpheart.org

LPS-INDUCED AUTOPHAGY IN CARDIOMYOCYTES 29. Natanson C, Eichenholz PW, Danner RL, Eichacker PQ, Hoffman WD, Kuo GC, Banks SM, MacVittie TJ, Parrillo JE. Endotoxin and tumor necrosis factor challenges in dogs simulate the cardiovascular profile of human septic shock. J Exp Med 169: 823– 832, 1989. 30. Ohkuma S, Poole B. Fluorescence probe measurement of the intralysosomal pH in living cells and the perturbation of pH by various agents. Proc Natl Acad Sci USA 75: 3327–3331, 1978. 31. Parker MM, Shelhamer JH, Bacharach SL, Green MV, Natanson C, Frederick TM, Damske BA, Parrillo JE. Profound but reversible myocardial depression in patients with septic shock. Ann Intern Med 100: 483– 490, 1984. 32. Parrillo JE, Burch C, Shelhamer JH, Parker MM, Natanson C, Schuette W. A circulating myocardial depressant substance in humans with septic shock. Septic shock patients with a reduced ejection fraction have a circulating factor that depresses in vitro myocardial cell performance. J Clin Invest 76: 1539 –1553, 1985. 33. Peng T, Lu X, Feng Q. Pivotal role of gp91phox-containing NADH oxidase in lipopolysaccharide-induced tumor necrosis factor-alpha expression and myocardial depression. Circulation 111: 1637–1644, 2005. 34. Sanlioglu S, Williams CM, Samavati L, Butler NS, Wang G, McCray PB Jr, Ritchie TC, Hunninghake GW, Zandi E, Engelhardt JF. Lipopolysaccharide induces Rac1-dependent reactive oxygen species formation and coordinates tumor necrosis factor-alpha secretion through IKK regulation of NF-kappa B. J Biol Chem 276: 30188 –30198, 2001. 35. Sawada M, Kiyono T, Nakashima S, Shinoda J, Naganawa T, Hara S, Iwama T, Sakai N. Molecular mechanisms of TNF-alpha-induced ceramide formation in human glioma cells: P53-mediated oxidant stress-dependent and -independent pathways. Cell Death Differ 11: 997–1008, 2004. 36. Scherz-Shouval R, Shvets E, Fass E, Shorer H, Gil L, Elazar Z. Reactive oxygen species are essential for autophagy and specifically regulate the activity of Atg4. EMBO J 26: 1749 –1760, 2007.


H479

37. Schreck R, Rieber P, Baeuerle PA. Reactive oxygen intermediates as apparently widely used messengers in the activation of the NF-kappa B transcription factor and HIV-1. EMBO J 10: 2247–2258, 1991. 38. Sewell RB, Barham SS, LaRusso NF. Effect of chloroquine on the form and function of hepatocyte lysosomes. Morphologic modifications and physiologic alterations related to the biliary excretion of lipids and proteins. Gastroenterology 85: 1146 –1153, 1983. 39. Shimizu S, Kanaseki T, Mizushima N, Mizuta T, Arakawa-Kobayashi S, Thompson CB, Tsujimoto Y. Role of Bcl-2 family proteins in a non-apoptotic programmed cell death dependent on autophagy genes. Nat Cell Biol 6: 1221–1228, 2004. 40. Shimomura H, Terasaki F, Hayashi T, Kitaura Y, Isomura T, Suma H. Autophagic degeneration as a possible mechanism of myocardial cell death in dilated cardiomyopathy. Jpn Circ J 65: 965–968, 2001. 41. Shubeita HE, Martinson EA, Van Bilsen M, Chien KR, Brown JH. Transcriptional activation of the cardiac myosin light chain 2 and atrial natriuretic factor genes by protein kinase C in neonatal rat ventricular myocytes. Proc Natl Acad Sci USA 89: 1305–1309, 1992. 42. Tomaru Ki K, Arai M, Yokoyama T, Aihara Y, Sekiguchi Ki K, Tanaka T, Nagai R, Kurabayashi M. Transcriptional activation of the BNP gene by lipopolysaccharide is mediated through GATA elements in neonatal rat cardiac myocytes. J Mol Cell Cardiol 34: 649 – 659, 2002. 43. van Lenten BJ, Fogelman AM. Lipopolysaccharide-induced inhibition of scavenger receptor expression in human monocyte-macrophages is mediated through tumor necrosis factor-alpha. J Immunol 148: 112–116, 1992. 44. Yamamoto A, Tagawa Y, Yoshimori T, Moriyama Y, Masaki R, Tashiro Y. Bafilomycin A1 prevents maturation of autophagic vacuoles by inhibiting fusion between autophagosomes and lysosomes in rat hepatoma cell line, H-4-II-E cells. Cell Struct Funct 23: 33– 42, 1998. 45. Yu L, Alva A, Su H, Dutt P, Freundt E, Welsh S, Baehrecke EH, Lenardo MJ. Regulation of an ATG7-beclin 1 program of autophagic cell death by caspase-8. Science 304: 1500 –1502, 2004.

296 • FEBRUARY 2009 •

www.ajpheart.org