Biochemistry. in the Article “The Med1 Subunit of the Yeast Neurobiology

3330 Corrections Proc. Natl. Acad. Sci. USA 96 (1999)

Biochemistry. In the article “The Med1 subunit of the yeast Neurobiology. In the article “Caspase-1 is activated in neural mediator complex is involved in both transcriptional activation cells and tissue with amyotrophic lateral sclerosis-associated and repression” by Darius Balciunas, Cecilia Ga¨lman,Hans mutations in copper-zinc superoxide dismutase” by Piera Ronne, and Stefan Bjo¨rklund, which appeared in number 2, Pasinelli, David R. Borchelt, Megan K. Houseweart, Don W. January 19, 1999, of Proc. Natl. Acad. Sci. USA (96, 376–381), Cleveland, and Robert H. Brown, Jr., which appeared in due to a printer’s error, the affiliation symbols were incorrect. number 26, December 22, 1998, of Proc. Natl. Acad. Sci. USA The correct author line, affiliation line, and address footnotes (95, 15763–15768), the following corrections should be noted. appear below. An erroneous version of Fig. 6 was published. The lane indicated as G41D represents lumbo-sacral spinal cord extract from G85R transgenic mice. In Fig. 7a, cell viability is ex- DARIUS BALCIUNAS*†,CECILIA GALMAN¨ ‡§, pressed as % of untreated cells and not as % of viability. HANS RONNE*†, AND STEFAN BJORKLUND¨ ‡¶

‡Department of Medical Biochemistry and Biophysics, Umeå University, Physiology. In the article “Inositol 1,4,5-tris-phosphate acti- S-901 87 Umeå, Sweden; and *Department of Medical Biochemistry and vation of inositol tris-phosphate receptor Ca2ϩ channel by Microbiology, Uppsala University Biomedical Center, Box 582, 751 23 2ϩ Uppsala, Sweden ligand tuning of Ca inhibition” by Don-On Daniel Mak, Sean McBride, and J. Kevin Foskett, which appeared in number 26, December 22, 1998, of Proc. Natl. Acad. Sci. USA †Present address: Department of Plant Biology, Uppsala Genetic Center, Swedish University of Agricultural Sciences, Box 7080, (95, 15821–15825), due to printer’s errors, the following cor- S-75007 Uppsala, Sweden. rections should be noted. The title of the article should read §Present address: CNT, Karolinska Institute at NOVUM, 141 57 “Inositol 1,4,5-trisphosphate activation of inositol trisphos- Huddinge, Sweden. phate receptor Ca2ϩ channel by ligand tuning of Ca2ϩ inhibi- ¶To whom reprint requests should be addressed. e-mail: ste@ tion.” On page 15821, the first line of the abstract should read panther.cmb.umu.se. “Inositol 1,4,5-trisphosphate (IP3)” instead of “Inositol 1,4,5- tris-phosphate (IP3),” and on page 15821, in the right column, Medical Sciences. In the article “Activation of ␧ protein kinase the first line of the abbreviations footnote should read “IP3, C correlates with a cardioprotective effect of regular ethanol inositol 1,4,5-trisphosphate” instead of “IP3, inositol 1,4,5-tris- consumption” by Masami Miyamae, Manuel M. Rodriguez, S. phosphate.” Albert Camacho, Ivan Diamond, Daira Mochly-Rosen, and Vincent M. Figueredo, which appeared in number 14, July 7, Physiology. In the article “Post-priming actions of ATP on 1998, of Proc. Natl. Acad. Sci. USA (95, 8262–8267), the Ca2ϩ-dependent exocytosis in pancreatic beta cells” by Noriko following correction should be noted. On page 8264, an Takahashi, Takashi Kadowaki, Yoshio Yazaki, Graham C. R. incorrect Fig. 1 was printed. The correct figure and its accom- Ellis-Davies, Yasushi Miyashita, and Haruo Kasai, which panying legend are reproduced below. appeared in number 2, January 19, 1999, of Proc. Natl. Acad. Sci. USA (96, 760–765), the following corrections should be noted. On page 761, left column, line 11 should read “2- nitrophenyl-EGTA” instead of “DMNPE-4.” Also, on page 765, left column, lines 11–15 should read “(ii) glucose in- 2ϩ creased insulin exocytosis, even when [Ca ]i was clamped at a high level and KATP channels were open (9, 11). These observations can be explained readily by the ATP-sensing mechanism identified in the present study.”

FIG. 1. LVDP prior to 45 min of global ischemia and during reperfusion in four groups of perfused guinea pig hearts (n ϭ 9 for each group): 1, following 8 wk 15% ethanol-derived calories (E); 2, pair-fed controls (‚); 3, following 8 wk of ethanol, before and after 10 mM chelerythrine (F); and 4, pair-fed controls, before and after chelerythrine (■). LVDP recovery is significantly greater in hearts from ethanol-treated animals (P Ͻ 0.05 at each 6-min interval). Chelerythrine abolished ethanol’s protective effect on LVDP recovery. Data are presented as mean Ϯ SEM (SEM not included for group 2 but lie well within SEM of groups 3 and 4). Downloaded by guest on September 27, 2021 Proc. Natl. Acad. Sci. USA Vol. 95, pp. 8262–8267, July 1998 Medical Sciences

Activation of ␧ protein kinase C correlates with a cardioprotective effect of regular ethanol consumption (chelerythrine͞ischemia͞reperfusion͞adenosine͞preconditioning)

ʈ MASAMI MIYAMAE*†‡,MANUEL M. RODRIGUEZ‡§,S.ALBERT CAMACHO*†,IVAN DIAMOND¶ **, DARIA MOCHLY-ROSEN§, AND VINCENT M. FIGUEREDO*†**††

Departments of *Medicine (Cardiology), ¶Neurology, ʈCellular and Molecular Pharmacology, and the **Ernest Gallo Clinic and Research Center, San Francisco General Hospital, †University of California, San Francisco, CA, 94110 and §Department of Molecular Pharmacology, School of Medicine, Stanford University, CA, 94305-5332

Communicated by David M. Kipnis, Washington University School of Medicine, St. Louis, MO, April 30, 1998 (received for review November 20, 1997)

ABSTRACT In addition to decreasing the incidence of like experimental ischemic preconditioning (15–19). Ischemic myocardial infarction, recent epidemiological data suggest preconditioning occurs when brief periods of ischemia and that regular alcohol consumption improves survival after reperfusion protect hearts against injury from subsequent myocardial infarction. We recently found that chronic ethanol prolonged ischemia–reperfusion. A recent study demonstrated exposure induces long-term protection against cardiac isch- that hearts from rats fed ethanol for 8 wk could be precondi- emia–reperfusion injury, which improves myocardial recov- tioned with a single 5-min episode of ischemia prior to ery after infarction. Furthermore, this cardioprotection by prolonged ischemia–reperfusion, whereas hearts from control ethanol is mediated through myocyte adenosine A1 receptors. animals were not protected (20). However, the short-term We now determine the role of protein kinase C (PKC) in protective effect of ischemic preconditioning is not clinically ethanol’s protective effect against ischemia–reperfusion in- applicable, necessitating the search for long-term therapies jury. Using perfused hearts of ethanol-fed guinea pigs, we find which maintain protection against ischemia–reperfusion injury that improved contractile recovery and creatine kinase release in patients at risk for MI. after ischemia–reperfusion are abolished by PKC inhibition The protective effect of ischemic preconditioning has been with chelerythrine. Western blot analysis and immunofluo- associated with activation of protein kinase C (PKC) in several rescence localization demonstrate that regular ethanol con- animal models and human myocardium (21–27). Recent re- sumption causes sustained translocation (activation) of ports correlate activation of specific PKC isozymes, including ␧PKC, but not ␦ or ␣PKC. This same isozyme is directly ␧, ␦, and ␣PKC, with ischemic preconditioning (26–29). Gray implicated in ischemic preconditioning’s protection against et al. (26), directly implicated ␧PKC in ischemic precondition- ischemia–reperfusion injury. Our findings suggest (i) that ing’s protection of neonatal cardiomyocytes using PKC regular ethanol consumption induces long-term cardioprotec- isozyme-specific inhibitors. We now report evidence that PKC tion through sustained translocation of ␧PKC and (ii) that activity is necessary for mediating ethanol’s protective effect PKC activity is necessary at the time of ischemia to mediate against ischemia–reperfusion injury at the time of ischemia. ethanol’s protective effect against ischemia–reperfusion in- We also observed that regular ethanol consumption causes jury. Studying this selective effect of ethanol on ␧PKC acti- sustained translocation of a distinct PKC isozyme in myocytes, vation may lead to new therapies to protect against ischemia– potentially contributing to the long-term cardioprotection reperfusion injury in the heart and other organ systems. against ischemia–reperfusion injury induced by ethanol.

Epidemiological studies show that drinking alcohol decreases METHODOLOGY mortality due to ischemic heart disease, primarily by reducing the incidence of myocardial infarction (MI) (1–4). This de- Perfused Heart Study. Male Hartley guinea pigs (275–300 g) creased incidence of MI is likely due to ethanol’s pleotropic were fed a nutritionally supplemented liquid diet (Dyets, effects on lipids, platelets, and fibrinolytic activity (5–9). Bethlehem, PA) containing 15% or 0% ethanol-derived cal- Recent epidemiological data suggest that drinking alcohol may ories for 8 wk. This approximates the upper limits of moderate also improve survival after MI (10–12), but the mechanisms ethanol consumption in humans (Ͻ45 g of ethanol/day, at 7 underlying this cardioprotective effect of ethanol remain un- kcal/g ethanol, results in Ϸ15% ethanol-derived calories in a certain. 2000 kcal diet; see ref. 4). Control animals received the same Survival after MI is directly related to myocardial recovery number of calories as the paired ethanol-fed animals had (13). Reducing ischemia–reperfusion injury improves myocar- consumed over the previous 24 hr. Serum ethanol levels were dial recovery after MI. The increasing use of emergent reper- drawn after 8 wk in five animals. fusion therapies during acute MI, including thrombolysis and Hearts were isolated and perfused via the Langendorff coronary angioplasty, provides impetus for finding therapies to method as described previously (14) using a Krebs-Henseleit reduce ischemia–reperfusion injury. We recently found that perfusate (pH 7.4, 37°C) bubbled with 95%O2/5%CO2. Left regular ethanol consumption induces long-term protection ventricular (LV) pressures were measured using a 2-French, against ischemia–reperfusion injury in guinea pig hearts (14). high-fidelity micromanometer (Millar Instruments, Houston, We showed that this cardioprotective effect of ethanol requires TX) at a LV end diastolic pressure of 10 mmHg and pacing rate adenosine A1 receptor activation at the time of ischemia (14), Abbreviations: MI, myocardial infarction; PKC, protein kinase C; LV, The publication costs of this article were defrayed in part by page charge left ventricular; CK, creatine kinase; LVDP, LV-developed pressure; PMA, phorbol 12-myristate 13-acetate. payment. This article must therefore be hereby marked ‘‘advertisement’’ in ‡M.M. and M.M.R. contributed equally to this work. accordance with 18 U.S.C. §1734 solely to indicate this fact. ††To whom reprint requests should be addressed at: Division of © 1998 by The National Academy of Sciences 0027-8424͞98͞958262-6$2.00͞0 Cardiology, Lovelace Medical Center, 8400 Gibson SE, Albuquer- PNAS is available online at http:͞͞www.pnas.org. que, NM 87108. e-mail: [email protected].

8262 Medical Sciences: Miyamae et al. Proc. Natl. Acad. Sci. USA 95 (1998) 8263 of 240 beats per minute. Coronary flow was measured by an Kraftbruhe buffer (pH 7.2), triturated with a Pasteur pipette, in-line flow meter (Gilmont Instruments, Barrington, IL). filtered through a stainless mesh, and centrifuged for 5 min at Creatine kinase (CK) release during reperfusion was mea- 1000 ϫ g. Myocytes were isolated from nonmyocytes by sured with a commercially available kit (Sigma). Values were resuspending the pellet in 20 ml of 4% Ficoll-400 (Sigma)/ corrected for dry heart weight and coronary flow rates and Kraftbruhe buffer and centrifuging for 5 min at 40 rpm. The expressed in units per ml per g dry weight (units/ml ϫ gdw). myocyte pellet was resuspended in 5 ml of Kraftbruhe buffer. Ischemia–Reperfusion Protocol. Ethanol was removed from Isolated myocytes were divided into three groups and the liquid diet 12–16 hr prior to sacrifice to avoid a direct effect treated for 15 min with active phorbol ester to maximally of ethanol on hearts. After a 20-min equilibration period, translocate PKC [4␤-phorbol 12-myristate 13-acetate (4␤- hearts were subjected to 45 min of no-flow ischemia and 48 min PMA), 100 nM; LC Laboratories, Woburn, MA), inactive of reperfusion as previously described (14) (n ϭ 9 ethanol; n ϭ phorbol ester as a control (4␣-PMA, 100 nM), or vehicle 9 control). After reperfusion, hemodynamic measurements (dimethyl sulfoxide, 0.04%). Myocytes were then washed three were repeated every 6 min for 48 min. Hearts were then dried times in 15 ml of PBS and pelleted after each wash by for 24 hr at 80°C and weighed. To exclude the possibility that centrifugation at 800 ϫ g for 30 sec. Cell viability was deter- ethanol withdrawal might contribute to findings, additional mined by trypan blue exclusion. experiments were performed using animals allowed to con- Western Blot Analysis. Vehicle and PMA-treated myocytes sume ethanol until sacrifice (n ϭ 5). were homogenized in 1.0 ml of PBS with protease inhibitors Additional experiments were performed with chelerythrine (20 mg/ml each of phenylmethylsulfonyl fluoride, soybean chloride (10 ␮M; Sigma) added to the perfusate 10 min prior trypsin inhibitor, leupeptin, and aprotinin) by trituration 15 to ischemia (n ϭ 9 ethanol, n ϭ 9 control). This concentration times with a syringe and 22-gauge needle and then centrifuged of chelerythrine, a potent and specific inhibitor of PKC (30), for 30 min at 100,000 ϫ g (4°C). Supernatants were collected was used as described previously (28, 31, 32). Chelerythrine as the soluble (cytosolic) fraction and the pellets were reho- decreased baseline LV developed pressure (LVDP) by 14% mogenized by identical trituration in 1.0 ml of PBS/protease and increased coronary flow by 7% in both ethanol and control inhibitor solution with 1% Triton X-100. Homogenates were hearts (Table 1). Similar changes in hemodynamics have been centrifuged and the resultant Triton X-100-soluble fraction reported in previous studies with PKC inhibitors (28, 31, 32). was collected. Equal amounts of cytosolic and Triton X-100- An additional group of hearts were studied from guinea pigs soluble fractions were loaded onto SDS/PAGE gels after consuming 1.25% ethanol in their drinking water for 8 wk to addition of SDS–Laemmli sample buffer. Western blots were document that milder levels of regular ethanol consumption probed with either anti-␧PKC (1:100)- or anti-␦PKC (1:100)- also produce protection against ischemia–reperfusion injury specific polyclonal IgG (Santa Cruz Biotechnology) or anti- (n ϭ 10 ethanol; n ϭ 8 control). These experiments were ␣PKC (1:300) monoclonal IgG (Seikagaku America, Rock- performed in an identical manner to those described above. ville, MD). Quantitation of Western blots was performed by Cardiomyocyte Studies. Male Hartley guinea pigs (275–300 densitometric analysis of digitized enhanced chemilumines- g; n ϭ 9 ethanol; n ϭ 9 control) were given 10% (a dose more cence-exposed films using NIH Image v1.58. Accuracy of this consistent with moderate consumption) or 0% ethanol and method and linearity of detection within the measured sample solid food ad libitum (Lab Diet: PMI Feeds, St. Louis) for 8 range was confirmed by quantitation of the same Western blots wk as described previously (14). Isolated hearts were perfused by enhanced chemifluorescence detection using a Molecular for 10 min with Krebs-Henseleit perfusate (pH 7.4, 37°C) Dynamics STORM 850 fluorescence scanner and Molecular bubbled with 95%O2/5%CO2. Perfusate was changed to a Dynamics ImageQuant v1.1 software (Sunnyvale, CA). nominally Ca2ϩ-free Krebs-Henseleit perfusate (pH 7.4) with Immunofluorescence Localization of ␧PKC. Myocytes were 0.5 mg/ml bovine albumin for 10 min and then Krebs-Henseleit fixed immediately after vehicle or PMA treatment with 5% perfusate containing 0.7 mg/ml collagenase B (Boehringer gluteraldehyde for 15 min. Fixed myocytes were incubated for Mannheim), 25 mmol/liter CaCl2, and 0.5 mg/ml BSA at a 1 hr with 1% normal goat serum in PBS containing 0.1% constant flow of 5 ml/min for 20 min. Two hearts (one ethanol Triton X-100 and then incubated with anti-␧PKC IgG (Re- and one control) were deemed unusable due to insufficient search and Diagnostic Antibodies, Berkeley, CA) diluted 1:100 collagenase digestion and were eliminated from analysis with in PBS containing 0.1% Triton X-100 and 2 mg/ml BSA the heart from their paired animal. Ventricles were minced in overnight at 4°C. Myocytes were washed three times with PBS

Table 1. PKC activity is important in mediating ethanol’s cardioprotective effect against ischemia–reperfusion injury Preischemia Reperfusion Control Ethanol Control Ethanol Developed pressure (mmHg) 117 Ϯ 4 117 Ϯ 225Ϯ 449Ϯ 4* Diastolic pressure (mmHg) 10 Ϯ 010Ϯ 058Ϯ 535Ϯ 6* Perfusion pressure (mmHg) 70 Ϯ 070Ϯ 076Ϯ 176Ϯ 1 Coronary flow (ml͞min) 35 Ϯ 236Ϯ 123Ϯ 423Ϯ 2 (ϩ) CHE Developed pressure (mmHg) 104 Ϯ 4† 102 Ϯ 4† 24 Ϯ 330Ϯ 3† Diastolic pressure (mmHg) 10 Ϯ 010Ϯ 053Ϯ 449Ϯ 4† Perfusion pressure (mmHg) 70 Ϯ 070Ϯ 077Ϯ 176Ϯ 1 Coronary flow (ml͞min) 39 Ϯ 1† 39 Ϯ 1† 19 Ϯ 220Ϯ 2 Hemodynamic data from experiments in perfused guinea pig hearts subjected to 45 min of global ischemia and 48 min of reperfusion. Hearts from guinea pigs consuming 15% ethanol-derived calories for 8 wk were compared to hearts from pair-fed controls (ethanol, n ϭ 9; control, n ϭ 9). Experiments were repeated in the presence of the PKC antagonist, chelerythrine [(ϩ)CHE; ethanol, n ϭ 9; control, n ϭ 9]. Data are presented as mean Ϯ SEM. P Ͻ 0.05 for all reperfusion data in controls and ethanol-treated animals versus preischemia values. *P Ͻ 0.05 ethanol versus control. †P Ͻ 0.05 (ϩ)CHE versus (Ϫ)CHE. 8264 Medical Sciences: Miyamae et al. Proc. Natl. Acad. Sci. USA 95 (1998) and incubated for 2 hr with fluorescein-conjugated rabbit IgG secondary antibody (Organon Technika, West Chester, PA) diluted 1:1000. Myocytes were washed three times with PBS, mounted on glass slides using Vecta Shield (Vector Labora- tories), and viewed with a Zeiss IM35 microscope (Zeiss) with a ϫ40 water immersion objective. Images were recorded on Kodak Tmax 400 film with an exposure time of 45 sec for photomicrographs. The percentage of myocytes showing a predominantly cytosolic/perinuclear fluorescence (nontrans- located ␧PKC) or predominantly cross-striated fluorescence (translocated ␧PKC) were determined by blind counting of 100 myocytes in each treatment group for each heart. The preab- sorbed anti-␧PKC IgG used in this study do not exhibit any specific staining in similarly treated and fixed myocytes (26, 33). Statistical Analysis. All data are expressed as mean Ϯ SEM. Comparisons between groups were made using repeated mea- sures analysis of variance with multiple grouping factors. A FIG. 2. CK release during the first 18 min of postischemic reper- Student–Newman–Keuls post hoc test was used to confirm the fusion from hearts of ethanol-treated (shaded bars) and control (black ϫ n ϭ Ϯ significance of differences between groups. P Ͻ 0.05 was bars) animals (units/ml gdw; 9 for each group; mean SEM). CK release was significantly less from hearts of ethanol-treated P Ͻ 0.05). Chelerythrine abolished ethanol’s reduction of ,ء) considered to be statistically significant. animals CK release during reperfusion. RESULTS or coronary perfusion pressure after ischemia–reperfusion Serum ethanol levels were 10 Ϯ 2 mg/dl (Ϸ2 mM; 9–11 a.m.) (Table 1), suggesting that ethanol’s protection did not involve after 8 wk of feeding with a nutritionally supplemented liquid vasodilation but is mediated at the myocyte level. diet containing 15% ethanol-derived calories. Body weights In the above experiments, ethanol feeding was discontinued (728 Ϯ 9 g vs. 728 Ϯ 11 g) and dry heart weight to body weight 12–16 hr before sacrifice to avoid a direct effect of ethanol on Ϫ Ϫ ratios (4.02 Ϯ 0.09 ϫ 10 4 vs. 4.11 Ϯ 0.09 ϫ 10 4) were the hearts. To determine whether ethanol withdrawal contributed same after 8 wk in ethanol-fed and control animals, suggesting to the observed cardioprotection, additional experiments were no development of LV hypertrophy as is seen with heavy also performed using hearts from five animals consuming ethanol consumption (34–36). Baseline LVDP, coronary flow, ethanol until sacrifice. LVDP recovered to 45% of preischemic and perfusion pressure were the same in hearts from ethanol- levels (LVDP ϭ 118 Ϯ 4 mmHg preischemia; 54 Ϯ 8 mmHg fed and control animals (Table 1). at 48 min reperfusion), LV end diastolic pressure increased to Regular Ethanol Consumption Reduces Ischemia– 290% of preischemic levels, and CK release during reperfusion Reperfusion Injury. LVDP recovered to 42% of preischemic was 279 Ϯ 59 units/ml ϫ gdw. These data suggest that ethanol’s levels in hearts from ethanol-fed animals compared with 22% cardioprotective effect is the same whether ethanol is present Ͻ in controls (P 0.05; Fig. 1). Increased LV end diastolic in the serum or withdrawn 12–16 hr before the ischemic insult. pressure, an index of myocyte contracture and irreversible Furthermore, the presence of ethanol is not deleterious to the injury, was lower during postischemic reperfusion in hearts heart during ischemia–reperfusion. from ethanol-fed animals (350%) compared with controls To demonstrate that ethanol was cardioprotective at milder Ͻ (580%; P 0.05; Table 1). CK release, a measure of myocyte levels of consumption, hearts from guinea pigs drinking 1.25% necrosis and/or loss of membrane integrity, was significantly ethanol for 8 wk were also studied. Weights were the same in lower from hearts of ethanol-fed animals compared with ethanol-treated and control animals (850 Ϯ 20 vs. 870 Ϯ 20 g). Ϯ Ϯ ϫ Ͻ controls (260 40 vs. 469 74 units/ml gdw, P 0.05; Fig. LVDP recovery was greater (50 Ϯ 7 vs. 34 Ϯ 5 mmHg at 48-min 2). There were no differences between groups in coronary flow reperfusion, P Ͻ 0.05), LV end diastolic pressure rise was less (21 Ϯ 4 vs. 42 Ϯ 5 mmHg at 48-min reperfusion; P Ͻ 0.05), and

FIG. 1. LVDP prior to 45 min of global ischemia and during reperfusion in four groups of perfused guinea pig hearts (n ϭ 9 for each group): 1, following 8 wk 15% ethanol-derived calories (⅜); 2, pair-fed controls (‚); 3, following 8 wk of ethanol, before and after 10 mM chelerythrine (F); and 4, pair-fed controls, before and after FIG. 3. Representative Western blots depicting ␧PKC transloca- chelerythrine (■). LVDP recovery is significantly greater in hearts tion in vehicle- (group 1), 100 nM 4␣-PMA- (group 2), or 100 nM from ethanol-treated animals (P Ͻ 0.05 at each 6-min interval). 4␤-PMA-treated (group 3) myocytes following isolation from one pair Chelerythrine abolished ethanol’s protective effect on LVDP recov- of control and ethanol-fed animals. Myocytes were subjected to ery. Data are presented as mean Ϯ SEM (SEM not included for group fractionation by centrifugation to buffer-soluble (cytosol) and Triton 2 but lie well within SEM of groups 3 and 4). X-100-soluble (particulate) fractions. Medical Sciences: Miyamae et al. Proc. Natl. Acad. Sci. USA 95 (1998) 8265

implicated in ischemic preconditioning (26–28). Translocation of PKC isozymes from the soluble to the particulate myocyte fraction correlates with activation (37). We determined the subcellular distribution of ␧, ␦, and ␣PKC using Western blot analysis in myocytes isolated from hearts of ethanol-fed and control animals. Isolated myocytes were exposed to ␤-PMA (phorbol ester activator of PKC), ␣-PMA (inactive phorbol ester, control), or dimethyl sulfoxide (control). ␤-PMA induced translocation of PKC isozymes from the cytosolic (inactive PKC) to particulate fractions (active PKC), as shown previously (26, 38). Representative autoradiographs of West- ern blots probed for ␧PKC are shown in Fig. 3. Individual autoradiographs of particulate fractions for each animal pair, as well as the mean data, are shown in Fig. 4. Response to ␤-PMA was similar in myocytes from ethanol-fed and control animals. Subcellular distribution of ␦ and ␣PKC did not differ in myocytes from ethanol-fed and control animals in the FIG. 4. PKC isozyme translocation in vehicle-treated myocytes. presence of dimethyl sulfoxide vehicle or ␣-PMA (Fig. 4). In (Left) Depicted are Western blots for ␧, ␣, and ␦PKC from the contrast, the ratio of particulate to cytosolic ␧PKC was greater particulate fraction of each of the seven control (C) and ethanol-fed in myocytes from ethanol-fed compared with control animals (E) animal pairings used in this study (numbered 1 through 7). (Right) (2.7 Ϯ 0.1:1 vs. 1.5 Ϯ 0.4:1, P Ͻ 0.05; Fig. 4). Total ␧PKC was Depicted is each average corresponding PKC isozyme level in both the Ϯ ϭ ␧ not significantly different from total ␧PKC in myocytes from cytosolic and particulate fractions of these pairings ( SEM, n 7). , Ϯ ␣, and ␦PKC levels for all treatment groups are normalized to the control animals (82 15% of control levels, P, not significant). vehicle-treated paired control for each group and to the average cell To corroborate ␧PKC Western blot data, we used immu- viability of each treatment group following PMA (or vehicle) treat- nofluorescence localization to assess the degree of ␧PKC P Ͻ 0.05. translocation in myocytes from ethanol-fed animals compared ,ء .ment with controls. Representative microphotographs of ␧PKC Ϯ Ϯ CK release during reperfusion was reduced (178 26 vs. 313 antibody fluorescence are shown in Fig. 5. As shown previously ϫ Ͻ 29 units/ml gdw, P 0.05) in hearts from ethanol-treated (26, 33, 37–39), immunolocalization of ␧PKC antibody fluo- animals compared with controls. These data suggest that rescence from the perinuclear and cytosolic regions of the ethanol’s cardioprotective effect is also present at milder levels myocyte to the cross-striations (possibly myofilaments) is of regular ethanol consumption. indicative of ␧PKC translocation to activation sites. Translo- Role of PKC in Ethanol’s Cardioprotective Effect: Perfused cation by immunofluorescence was not discernable in cells Heart Study. We next determined whether PKC activity is stained for either ␣ or ␦PKC. As shown in Fig. 6, blinded important in mediating ethanol’s protective effect against scoring revealed that the percentage of myocytes demonstrat- ischemia–reperfusion injury at the time of ischemia. In the ing a predominant cross-striated pattern of fluorescence was presence of the PKC inhibitor chelerythrine, LVDP and end greater in hearts from ethanol-fed animals compared with diastolic pressure (Table 1) and CK release (415 Ϯ 42 units/ controls (61 Ϯ 3% vs. 22 Ϯ 5%, P Ͻ 0.05). ␣-PMA, an inactive ml ϫ gdw vs. 484 Ϯ 44 units/ml ϫ gdw) were similar during phorbol ester, did not cause redistribution of ␧PKC in either postischemic reperfusion in hearts from ethanol-fed and con- group of myocytes. In contrast, ␤-PMA increased the percent- trol animals. Furthermore, these parameters were similar to age of control myocytes with cross-striated fluorescence (62 Ϯ data from hearts of control animals not exposed to cheleryth- 2% vs. 22 Ϯ 5%, P Ͻ 0.05), but had no significant effect on the rine (Figs. 1 and 2), suggesting that chelerythrine abolished percentage of myocytes from ethanol-fed animals with cross- ethanol’s protective effect. striated fluorescence (68 Ϯ 3% vs. 61 Ϯ 3%, P, not significant). Role of PKC in Ethanol’s Cardioprotective Effect: Cardio- These data suggest that ␧PKC is already significantly translo- myocyte Studies. We next determined whether ethanol’s long- cated or redistributed with regular ethanol consumption and term cardioprotective effect against ischemia–reperfusion in- that ethanol and ␤-PMA may induce ␧PKC translocation by jury is due to sustained activation of selective PKC isozymes similar mechanisms. Note also that the translocation to cellular

FIG. 5. Confocal indirect immunofluorescence images of vehicle-treated myocytes of a control animal (A), 4␤-PMA-treated myocytes of a control animal (B), and vehicle-treated myocytes of an ethanol-fed animal (C). (B and C) Depict activated ␧PKC translocated to myofibrillar structures; such cells were scored as having a translocated ␧PKC pattern of immunofluorescence. A, however, depicts inactive ␧PKC in a diffuse cytosolic staining pattern; such cells were scored as inactive. Images were acquired at ϫ60, 0.42-mm resolution in the z axis. Insets, ϫ3.5. 8266 Medical Sciences: Miyamae et al. Proc. Natl. Acad. Sci. USA 95 (1998)

␧PKC in preconditioning’s protection of neonatal cardiomyocytes using PKC isozyme-specific inhibitors. PKC isozyme activation is associated with translocation to sites in the myocyte where isozymes bind to anchoring mole- cules termed receptors for activated C kinases (RACKs; ref. 37). Isozyme-selective RACKs are located on a variety of subcellular structures, including membranes and cytoskeletal elements (37). Since we have shown that the selective PKC inhibitor chelerythrine inhibits ethanol-induced protection, and because our previous studies showed that translocation is required and sufficient for activation (26, 47–49), these data are most consistent with translocation correlated with activation of ␧PKC. After binding to their selective RACKs, activated PKC isozymes phosphorylate protein substrates which may ultimately mediate the cardioprotection against ischemia– reperfusion injury (50). Potential mediators of ischemic preconditioning’s first window of protection (up to 3 hr) include FIG. 6. Quantitative ␧PKC translocation assessed by immunoflu- PKC activation of ATP-sensitive potassium channels causing orescence localization. One hundred myocytes from each treatment reduced calcium influx (31, 51–53) and activation of vacuolar group were scored using the criteria described in the legend to Fig. 5 proton ATPase reducing intracellular acidification (54). Po- (also see Results) as either having an activated ␧PKC translocation Ϯ tential mediators of ischemic preconditioning’s second window pattern or an inactive pattern. Data are mean SEM from the same of protection (24–72 hr) include transcription factors activated seven animal pairs used in the Western blot analysis. via a PKC-mediated pathway which regulate expression of heat structures occurs from discrete sites in the cell where inactive shock/stress proteins (55, 56). ␧PKC resides (Fig. 5). Because regular ethanol consumption reduces ischemia– reperfusion injury in a manner analogous to ischemic preconditioning, we searched for evidence that PKC activation at the DISCUSSION time of ischemia is important in ethanol’s cardioprotective Our data support the hypothesis that regular ethanol con- effect. Perfusion with chelerythrine prior to ischemia abolished protection, suggesting a role for PKC in ethanol-induced sumption improves cardiac recovery after ischemia– cardioprotection. We are aware that chelerythrine may have reperfusion. In this study, ischemia–reperfusion injury is re- other effects on myocyte function. For example, treatment duced in hearts from guinea pigs drinking ethanol compared with chelerythrine did produce a small decrease of LVDP prior with isocalorically matched controls, suggesting that a starva- to ischemia. Similar changes in hemodynamics have been tion effect is not responsible for ethanol’s cardioprotective reported in prior ischemia–reperfusion studies with several effect. A major new finding of this study is that the PKC PKC inhibitors (28, 31, 32) and are thought to be due to antagonist chelerythrine abolishes ethanol’s cardioprotection, blocking PKC potentiation of adrenergic ␣1-associated slow suggesting that PKC activity is important for mediating pro- Ca2ϩ channels (57). A recent report in aortic rings suggests that tection at the time of ischemia. A second major finding is that chelerythrine can also affect cyclic nucleotide phosphodies- ␧ ␮ regular ethanol consumption selectively translocates PKC, terases with an IC50 ranging from 18 to 206 M (58) (we used but not ␦ or ␣PKC, as shown by Western blot analysis and 10 ␮M). Importantly, in our study, contractile recovery and CK immunofluorescence localization. These data suggest an release after ischemia–reperfusion were the same in control isozyme-specific role for ␧PKC in producing ethanol’s long- hearts in the presence or absence of chelerythrine. This term cardioprotective effect against ischemia–reperfusion in- suggests that chelerythrine did not alter cardiac recovery in jury. This observation is consistent with recent studies in other ischemia–reperfusion in control hearts. These data are also cell types demonstrating ethanol-induced activation of ␧PKC consistent with chelerythrine inhibiting PKC-induced cardio- (40–44). protection against ischemia–reperfusion injury in hearts from In addition to decreasing the incidence of MI (1–4), recent ethanol-fed animals. epidemiological data suggest that regular ethanol consumption We next examined the possibility that regular ethanol con- improves survival after MI (10–12). One potential mechanism sumption causes sustained activation of PKC isozymes implicated in ischemic preconditioning. We find that regular eth- by which regular drinking may improve survival after MI is to ␧ ␦ ␣ reduce ischemia–reperfusion injury, analogous to experimen- anol consumption translocates PKC, but not or PKC. tal ischemic preconditioning (15–19). The studies in guinea Thus, in guinea pig hearts, regular ethanol consumption may induce long-term cardioprotection by causing sustained trans- pigs described here show that regular ethanol consumption ␧ mimics ischemic preconditioning and reduces ischemia– location of PKC to its RACK (37). Such sustained translocation was proposed to occur by Downey and coworkers (59). reperfusion injury. Yet ethanol’s protection was present after Ethanol-induced activation of ␧PKC has been observed in 8 wk of ethanol feeding, whether animals were allowed to drink other cell types (40–44). For example, in neural crest-derived until sacrifice or were taken off ethanol 12–16 hr before. This PC12 cells, ethanol-induced activation of ␧PKC enhances suggests that it may be possible to use pharmacological agents nerve growth factor-induced signaling and neurite outgrowth to produce long-term cardioprotection against ischemia– (40). reperfusion injury in patients at risk for MI. We propose that regular ethanol consumption causes sus- Studies in animal and human myocardium demonstrate that tained translocation of ␧PKC to its RACK, where it is poised cardioprotection by ischemic preconditioning is abolished by at the time of ischemia to phosphorylate an effector protein(s) PKC inhibitors given prior to ischemia–reperfusion (21–23, which protects against ischemia–reperfusion injury. This hy- 45), mimicked by PKC activators (22, 23), and that membrane pothesis is supported by recent findings of Gray et al. (26) in PKC activity is enhanced during ischemia, suggesting PKC a myocyte model of ischemia–reperfusion injury. This study translocation (24, 46). Recent reports correlate activation of showed that a translocation antagonist specific for ␧PKC (a specific PKC isozymes, including ␧, ␦ and ␣PKC, with ischemic peptide derived from ␧PKC or its RACK) causes inhibition of preconditioning (26–29). Gray et al. (26), directly implicated translocation of ␧PKC and prevents protection induced by Medical Sciences: Miyamae et al. Proc. Natl. Acad. Sci. USA 95 (1998) 8267

PMA or preconditioning against myocyte death after hypoxia– 24. Prasad, M. R. & Jones, R. M. (1992) Basic Res. Cardiol. 87, 19–26. reoxygenation. Future studies will further elucidate the role of 25. Strasser, R. H., Braun-Dullaeus, R., Walendzik, H. & Marquet- ␧PKC in this long-term cardioprotective effect of regular ant, R. (1992) Circ. Res. 70, 1304–1312. 26. Gray, M. O., Karliner, J. S. & Mochly-Rosen, D. (1997) J. Biol. ethanol consumption and its relationship to adenosine A1 receptor activation (14). Chem. 272, 30945–30951. 27. Henry, P., Demolombe, S., Puceat, M. & Escande, D. (1996) Circ. Res. 78, 161–165. We thank Dr. Hui Zhong-Zhou for his excellent technical assis- 28. Miyawaki, H., Zhou, X. & Ashraf, M. (1996) Circ. Res. 79, tance. This work was supported by funds from National Institutes of 137–146. Health Grants RO1-AA11135 (to V.M.F.), KO8-02883 (to V.M.F.), 29. Ping, P., Zhang, J., Tang, X.-L., Manchikalapudi, S., Cao, X. & RO1- AA11147 (to D.M.-R.) and a grant from the Alcoholic Beverage Bolli, R. (1997) Circ. Res. 81, 404–414. Medical Research Foundation (to V.M.F.). 30. Herbert, J. M., Augereau, J. M., Gleye, J. & Maffrand, J. P. (1990) Biochem. Biophys. Res. Commun. 172, 993–999. 1. Klatsky, A. L., Armstrong, M. A. & Friedman, G. D. (1990) Am. J. 31. Speechly-Dick, M. E., Grover, G. J. & Yellon, D. M. (1995) Circ. Cardiol. 66, 1237–1242. Res. 77, 1030–1035. 2. Rimm, E. B., Giovannucci, E. L., Willett, W. C., Colditz, G. A., 32. Bugge, E. & Ytrehus, K. (1995) Cardiovasc. Res. 29, 401–406. 338, Ascherio, A., Rosner, B. & Stampfer, M. J. (1991) Lancet 33. Csukai, M., Chen, C.-H., De Matteis, M. A. & Mochly-Rosen, D. 464–468. (1998) J. Biol. Chem. 272, 29200–29206. 3. Fuchs, C. S., Stampfer, M. J., Colditz, G. A., Giovannucci, E. L., 34. Chan, T. C. & Sutter, M. C. (1982) Can. J. Physiol. Pharmacol. 60, Manson, J. E., Kawachi, I., Hunter, D. J., Hankinson, S. E., 777–782. Hennekens, C. H., Rosner, B., Speizer, F. E. & Willett, W. C. 35. Segel, L. D. (1988) Alcohol. Clin. Exp. Res. 12, 523–530. (1995) N. Engl. J. Med. 332, 1245–1250. 36. Capasso, J. M., Li, P., Guideri, G. & Anversa, P. (1991) Am. J. 4. Stampfer, M. J., Colditz, G. A., Willett, W. C., Speizer, F. E. & Physiol. 261, H212–H219. Hennekens, C. H. (1988) N. Engl. J. Med. 319, 267–273. 37. Mochly-Rosen, D. (1995) Science (Washington DC) 268, 247–251. 5. Gaziano, J. M., Buring, J. E., Breslow, J. L., Goldhaber, S. Z., 38. Johnson, J. A., Gray, M. D., Chen, C.-H. & Mochly-Rosen, D. Rosner, B., VanDenburgh, M., Willett, W. & Hennekens, C. H. (1996) J. Biol. Chem. 271, 24962–24966. (1993) N. Engl. J. Med. 329, 1829–1834. 39. Disatnik, M. H., Buraggi, G. & Mochly-Rosen, D. (1994) Exp. 6. Fumeron, F., Betoulle, D., Luc, G., Behague, I., Ricard, S., Cell Res. 210, Poirier, O., Jemaa, R., Evans, A., Arveiler, D., Marques-Vidal, P., 287–297. Bard, J., Fruchart, J., Ducimetiere, P., Apfelbaum, M. & Cam- 40. Hundle, B., McMahon, T., Dadgar, J., Chen, C.-H., Mochly- bien, F. (1995) J. Clin. Invest. 96, 1664–1671. Rosen, D. & Messing, R. O. (1997) J. Biol. Chem. 272, 15028– 7. Pikaar, N. A., Wedel, M., van der Beek, E. J., van Dokkum, W., 15035. Kempen, H. J., Kluft, C., Ockhuizen, T. & Hermus, R. J. (1987) 41. Gordon, A. S., Yao, L., Wu, Z.-L., Coe, I. R. & Diamond, I. Metabolism 36, 538–543. (1997) Mol. Pharmacol. 52, 554–559. 8. Renaud, S. C., Beswick, A. D., Fehily, A. M., Sharp, D. S. & 42. DePetrillo, P. B. & Liou, C. S. (1993) Alcohol. Clin. Exp. Res. 17, Elwood, P. C. (1992) Am. J. Clin. Nutr. 55, 1012–1017. 351–354. 9. Ridker, P. M., Vaughan, D. E., Stampfer, M. J., Glynn, R. J. & 43. Kharbanda, S., Nakamura, T. & Kufe, D. (1993) Biochem. Hennekens, C. H. (1994) J. Am. Med. Assoc. 272, 929–933. Pharmacol. 45, 675–681. 10. Dufour, M. C., Caces, M. F., Whitmore, C. C. & Hanna, E. Z. 44. Skwish, S. & Shain, W. (1990) Life Sci. 47, 1037–1042. (1996) Alcohol. Clin. Exp. Res. 20, 97A (abstr.). 45. Sakamoto, J., Miura, T., Goto, M. & Iimura, O. (1995) Cardio- 11. Nakamura, Y., Kawai, C., Kinoshita, M. & Moss, A. J. (1995) vasc. Res. 29, 682–628. Circulation 92, I-708 (abstr.). 46. Strasser, R. H., Weinbrenner, C., Rothele, J., Kubler, W. & 12. Wannamethee, G., Whincup, P. H., Shaper, A. G., Walker, M. & Marquetant, R. (1994) Drug Dev. Res. 31, 325 (abstr.). MacFarlane, P. W. (1995) Br. Heart J. 74, 324–331. 47. Ron, D., Luo, J. & Mochly-Rosen, D. (1995) J. Biol. Chem. 270, 13. Multicenter Postinfarction Research Group (1983) N. Engl. 24180–24187. J. Med. 309, 331–336. 48. Ron, D. & Mochly-Rosen, D. (1995) Proc. Natl. Acad. Sci. USA 14. Miyamae, M., Diamond, I., Weiner, M. W., Camacho, S. A. & 92, 492–496. Figueredo, V. M. (1997) Proc. Natl. Acad. Sci. USA 94, 3235– 49. Mochly-Rosen, D. & Gordon, A. (1998) FASEB J. 12, 35–42. 3239. 50. Puceat, M. & Vassort, G. (1996) Mol. Cell. Biochem. 157, 65–72. 15. Armstrong, S. & Ganote, C. E. (1994) Cardiovasc. Res. 28, 51. Liang, B. T. (1997) Am. J. Physiol. 273, H847–H853. 1049–1056. 52. Hu, K., Duan, D., Li, G. R. & Nattel, S. (1996) Circ. Res. 78, 16. Liu, G. S., Thornton, J., Van Winkle, D. M., Stanley, A. W., 492–498. Olsson, R. A. & Downey, J. M. (1991) Circulation 84, 350–356. 53. Liu, Y., Gao, W. D., O’Rourke, B. & Marban, E. (1996) Circ. Res. 17. Liu, G. S., Richards, S. C., Olsson, R. A., Mullane, K., Walsh, 78, 443–454. R. S. & Downey, J. M. (1994) Cardiovasc. Res. 28, 1057–1061. 54. Gottlieb, R. A., Gruol, D. L., Zhu, J. Y. & Engler, R. L. (1996) 18. Walker, D. M., Walker, J. M., Pugsley, W. B., Pattison, C. W. & J. Clin. Invest. 97, 2391–2398. Yellon, D. M. (1995) J. Mol. Cell Cardiol. 27, 1349–1357. 55. Yellon, D. M., Pasini, E., Cargnoni, A., Marber, M. S., Latchman, 19. Forman, M. B., Velasco, C. E. & Jackson, E. K. (1993) Cardio- D. S. & Ferrari, R. (1992) J. Mol. Cell. Cardiol. 24, 895–907. vasc. Res. 27, 9–17. 56. Donnelly, T. J., Sievers, R. E., Vissern, F. L., Welch, W. J. & 20. McDonough, K. H. (1997) Alcohol. Clin. Exp. Res. 21, 869–873. Wolfe, C. L. (1992) Circulation 85, 769–778. 21. Armstrong, S., Downey, J. M. & Ganote, C. E. (1994) Cardiovasc. 57. Otani, H., Prasad, M. R., Engelman, R. M., Otani, H., Cordis, Res. 28, 72–77. G. A. & Das, D. K. (1988) Circ. Res. 63, 930–936. 22. Ytrehus, K., Liu, Y. & Downey, J. M. (1994) Am. J. Physiol. 266, 58. Eckly-Michel, A. E., Le Bec, A. & Lugnier, C. (1997) Eur. H1145–H1152. J. Pharmacol. 324, 85–88. 23. Speechly-Dick, M. E., Mocanu, M. M. & Yellon, D. M. (1994) 59. Lawson, C. S. & Downey, J. M. (1993) Cardiovasc. Res. 27, Circ. Res. 75, 586–590. 542–550.