Male and female mice overexpressing the b -adrenergic receptor

Transcription

1 Cardiovascular Research 53 (00) locate/ cardiores Male and female mice overexpressing the b -adrenergic receptor exhibit differences in ischemia/ reperfusion injury: role of nitric oxide Abstract Heather R. Cross *, Elizabeth Murphy, Walter J. Koch, Charles Steenbergen a, c b a a Department of Pathology, Box 371, Duke University Medical Center, Durham, NC 7710, USA b Department of Surgery, Duke University Medical Center, Durham, NC 7710, USA c NIEHS, Research Triangle Park, Durham, NC 7709, USA Received 6 August 001; accepted 6 October 001 Objective: Cardiac overexpression of b-adrenergic receptors (bars) in male mice (MTG4) results in increased contractility and increased ischemic injury. Considering recent clinical data indicating that premenopausal women are protected from cardiovascular injury, we assessed the consequences of bar overexpression in females (FTG4). Since protection in females is mediated via estrogen, which activates endothelial and inducible nitric oxide synthases (enos and inos) we also examined the role of NOS in ischemia/ reperfusion injury in male and female TG4 and wild-type (WT) mice. Methods: Hearts from MTG4, FTG4, MWT and FWT mice were isolated and perfused in the Langendorff mode. Hearts were pretreated with either 1 mmol/ l of the nonspecific NOS inhibitor, L-NAME, or 100 nmol/ l of the specific inos inhibitor, 1400W. Control hearts received no treatment. All hearts were subjected to 0 min ischemia and 40 min 31 reperfusion while P-NMR spectra were acquired. Results: During ischemia, ATP and ph fell lower in MTG4 hearts than in FTG4 or WT hearts. Hearts from MTG4 mice exhibited increased ischemia/ reperfusion injury as indicated by lower recoveries of postischemic contractile function, ATP and PCr than WT. Despite contractility being elevated in FTG4 hearts to the same level as MTG4 hearts, ischemia/ reperfusion injury was not increased, as indicated by similar postischemic recoveries of contractile function, ATP and PCr in FTG4 hearts compared to WT. ATP and ph fell lower during ischemia in L-NAME-treated FTG4 hearts than in untreated FTG4 hearts, falling as low as untreated MTG4s. Recoveries of contractile function, ATP and PCr were as low in L-NAME-treated FTG4 hearts as in untreated MTG4 hearts and lower than untreated FTG4 hearts. In contrast, 1400W had no effect on FTG4 hearts. MTG4 hearts were unaffected by L-NAME or 1400W. Conclusions: b AR overexpression increased ischemia/ reperfusion injury in males but not females, thus females were protected from the detrimental effects of bar overexpression. Protection was abolished by treatment with L-NAME, but not 1400W, implying that protection was mediated by enos not inos. 00 Elsevier Science B.V. All rights reserved. Keywords: Energy metabolism; Gender; Ischemia; Nitric oxide; NMR 1. Introduction the sarcoplasmic reticulum (SR) Ca ATPase inhibitor phospholamban leads to increased SR Ca levels and Cardiac b-adrenergic receptors (bars) mediate the increased SR Ca release and phosphorylation of the contractile response to the sympathetic transmitters, epi- myofibrillar protein troponin I leads to altered Ca nephrine and norepinephrine. Binding of agonist to the sensitivity [1]. These phosphorylation events result in bar leads to activation of adenylate cyclase and protein increased cardiac contractility. Correspondingly, transgenic kinase A (PKA). In the heart, PKA phosphorylates several mice with a 00-fold cardiac-specific overexpression of the proteins. Phosphorylation of the L-type Ca channel by bar exhibit increased basal myocardial contractility []. PKA leads to increased Ca influx, phosphorylation of There was no evidence of cardiac abnormalities nor increased mortality in this transgenic model [], however, the male bar overexpressor mice exhibited greater *Corresponding author. Tel.: ; fax: address: cross017@mc.duke.edu (H.R. Cross). Time for primary review 9 days. Downloaded from by guest on 09 January / 0/ $ see front matter 00 Elsevier Science B.V. All rights reserved. PII: S (01)0058-4

2 H.R. Cross et al. / Cardiovascular Research 53 (00) myocardial energy utilization during ischemia and in- Animals published by the US National Institutes of Health creased ischemia/ reperfusion injury when compared to (NIH Publication No. 85-3, revised 1996). wild-type littermates [3]. Clinical findings indicate that females are protected from.. Ischemia/reperfusion protocol cardiovascular injury and that this protection is mediated via estrogen. Premenopausal women possess a lower risk Hearts were isolated and perfused in the Langendorff for ischemic heart disease than age-matched males [4] and mode as described previously [15]. Hearts were perfused this protection is lost after menopause [5]. Estrogen for 30 min before being subjected to 0 min no-flow replacement therapy has been shown to decrease the ischemia and 40 min reperfusion. Left-ventricular deoccurrence and mortality of myocardial infarction [6] and veloped pressure (LVDP), 6dP/dt and heart rate were increase survival from coronary artery bypass [7]. To monitored via a water-filled latex balloon inserted into the determine whether female bar overexpressor mice were left ventricle. Recovery of contractile function was asprotected from the exacerbation of ischemia/ reperfusion sessed by measurement of LVDP at the end of reperfusion injury observed in the males, we subjected isolated hearts and expressed as a percentage of preischemic LVDP. from male and female wild type and bar overexpressor mice to global ischemia and reperfusion. Ischemia/ re-.3. Nitric oxide synthase inhibitors perfusion injury was determined by the extent of recovery of postischemic contractile function and energy metabo- To inhibit both enos and inos, hearts from six of the lites. Interestingly, female hearts were protected from the male transgenic, six of the female transgenic, five of the detrimental effects of bar overexpression, with respect male wild-type and seven of the female wild-type mice to ischemia/ reperfusion injury. were perfused with 1 mmol/ l L-NAME, a non-specific Further studies were performed to determine the mecha- NOS inhibitor, 5 min prior to ischemia. The concentration nism of protection in the female bar overexpressor of L-NAME used corresponded to the concentration rehearts. Because the detrimental effects of bar over- quired to inhibit enos and inos without inducing vasexpression are likely to be mediated by calcium overload, oconstriction [16]. To inhibit specifically inos, five of the and nitric oxide (NO) is known to affect calcium homeo- male transgenic and seven of the female transgenic mice stasis [8 10], we studied the role of the estrogen effector, were perfused with 100 nmol/ l 1400W, 5 min prior to NO synthase (NOS). Expression of both inducible NOS ischemia. Specificity of 1400W was confirmed as de- (inos) and endothelial NOS (enos) is increased by scribed below. estrogen in the myocardium [11] and a number of studies have found NO produced via inos, enos or NO donors.4. Nuclear magnetic resonance spectroscopy to be cardioprotective [1 14]. By pretreating male and female bar overexpressor mouse hearts with either the Relative changes in concentrations of phosphorus menonspecific NOS inhibitor, Nw-nitro-L-arginine methyl tabolites were observed during the ischemia/ reperfusion 31 ester (L-NAME) or the specific inos inhibitor, N-[3- protocol by acquiring consecutive 5 min P nuclear (aminomethyl)benzyl]acetamidine (1400W), and studying magnetic resonance (NMR) spectra as described previously the ischemic response, we further aimed to determine the [3]. The areas of the spectral peaks were expressed as a role of NOS in the protection from ischemia/ reperfusion percentage of the peak areas of an initial, preischemic injury observed in female bar overexpressor hearts. control spectrum from each heart. The ratios of the PCr/ ATP peaks in the preischemic control spectra were lower in the male transgenic hearts, compared to male wild-type.. Methods The lower PCr/ATP ratio implies that preischemic PCr levels were lower in male transgenic mice. PCr values.1. Animals were therefore normalized by expressing PCr peak areas as a percentage of the ATP peak area in the initial pre- The transgenic mice (TG4) exhibited a 00-fold, car- ischemic control spectrum from each heart. Intracellular diac-specific overexpression of the bar; expression ph was estimated from the chemical shift of the Pi peak levels were determined as described previously []. relative to PCr using previously obtained titration curves. Twenty-four male adult heterozygous transgenic mice of body weight 3061 g and 6 female adult heterozygous.5. Specificity of 1400W and NO levels in male and transgenic mice of body weight 461 g were used. Twenty female hearts male and 1 female wild-type littermate mice of body weights 336 and 61 g, respectively, were employed as To confirm specificity of 1400W, 13 male wild-type controls. All animals were adult and reproductively viable mice were injected with.5 mg/ kg i.p. monophosphoryl at the time of experimentation. The investigation conforms lipid A (MLA) to induce inos expression, 8 h prior to with the Guide for the Care and Use of Laboratory heart perfusion, 10 min ischemia and freeze-clamping [17]. Downloaded from by guest on 09 January 019

3 664 H.R. Cross et al. / Cardiovascular Research 53 (00) Four of these hearts were perfused with 100 nmol/ l 1400W-treated male transgenic, 1400W-treated female 1400W, three hearts were perfused with 1 mmol/ l L- transgenic. NAME, three hearts were perfused with 1 mmol/ l L- NAME plus 100 nmol/ l 1400W, all for 5 min before ischemia; three hearts received no additional treatment. Total nitrate concentrations (NO ) were then determined 3. Results x using a nitrate/ nitrite colorimetric assay kit (Cayman Chemical, MI, USA) and compared to NOx levels in 3.1. bar density non-mla treated male wild-type hearts also subjected to 10 min ischemia before freeze-clamping. bar density was the same in female transgenic hearts, To determine NO levels in male vs. female hearts, five at pmol/ mg protein (0-fold increase over x untreated male wild-type and five untreated female wild- wild-type; n53), as in male transgenic hearts, at type hearts were isolated, perfused and freeze-clamped. A pmol/ mg protein (180-fold increase over wild-type; n53). further three male wild-type and three female wild-type untreated hearts were subjected to 10 min ischemia before freeze-clamping and determination of NO levels as de- 3.. Contractile function x scribed above Effects of overexpression of the bar.6. Statistics During the preischemic period, LVDP, 1dP/dt and dp/dt were higher in male transgenic than male wild- Results are expressed as means6s.e.m. Pre- and post- type hearts ( p,0.05; Table 1) and higher in female infusion contractile function values were compared within transgenic than female wild-type hearts ( p,0.05). Overeach treated group by a paired t-test. Discrete variables of expression of the bar, therefore, increased basal contraccontractile function were analyzed by a two-way analysis tility in both male and female hearts. of variance for sex, genotype and the interaction between Ischemic contracture, the sigmoidal increase in leftthe two factors or for sex, treatment and the interaction ventricular diastolic pressure during ischemia, began earbetween the two factors. Where the two-way analysis lier in male transgenic than male wild-type hearts ( p, revealed significant differences, pairwise comparisons 0.01; Table ). However, there was no difference in the were made using a Fisher s LSD test and significance was onset of contracture in female transgenic, compared to taken as p,0.05. For the energetic measurements, a two- female wild-type, hearts. way analysis of variance and Fisher s LSD test were By the end of reperfusion, recovery of contractile performed as above, comparing the end-point value of function was lower in male transgenic, at 8% initial LVDP, ischemia between groups and the end-point value of than male wild-type hearts, at 31% initial LVDP ( p, reperfusion between groups. In all cases where the two ); Fig. 1. However, there was no difference in way analysis was used, the following groups were com- functional recovery in female transgenic, at 31% initial pared: (a) male transgenic, female transgenic, male wild- LVDP, compared to female wild-type hearts, at 37% initial type, female wild-type; (b) male transgenic, female trans- LVDP. Overexpression of the bar, therefore, resulted in genic, L-NAME-treated male transgenic, L-NAME-treated increased susceptibility to ischemia/ reperfusion injury in female transgenic; (c) male transgenic, female transgenic, male, but not female, hearts. Table 1 Effect of NOS inhibitors on myocardial contractile function during normoxic perfusion Group Heart rate Pre-infusion Heart rate Post-infusion n (bpm) LVDP 1dP/dt dp/dt (bpm) LVDP 1dP/dt dp/dt (cmho) (cmho/ms) (cmho/ms) (cmho) (cmho/ms) (cmho/ms) MWT * * * 15 MTG FWT * * * 5 FTG MTG41L-NAME FTG41L-NAME MTG411400W FTG411400W Data are means6s.e.m. For preinfusion values, *: significant difference from MTG4; : significant difference from FTG4 ( p,0.05). For postinfusion values, : significant difference from corresponding preinfusion values ( p,0.05). For a description of statistical analyses see Methods. For abbreviations see Legend to Fig. 1. Downloaded from by guest on 09 January 019

4 H.R. Cross et al. / Cardiovascular Research 53 (00) Table There was no effect of L-NAME pretreatment on the Ischemic contracture parameters timing of contracture in male transgenic hearts (Table ). Group Ischemic contracture n However, ischemic contracture was earlier in L-NAMEtreated Start End Maximum female transgenic hearts than in untreated female (min) (min) (cm HO) transgenic hearts ( p,0.05). There was no significant MWT 1461* 1961* difference in timing of contracture between L-NAME- MTG treated female transgenic and untreated male transgenic FWT 1361* 1961* hearts. FTG4 1361* 1861* There was no difference in post-ischemic functional MTG41L-NAME recovery between L-NAME-treated male transgenic hearts, FTG41L-NAME MTG411400W at 10% initial LVDP, and untreated male transgenic hearts, FTG411400W at 8% initial LVDP (Fig. 1). However, by the end of Data are means6s.e.m. *: significant difference from MTG4; : reperfusion, recovery of contractile function was lower in significant difference from FTG4 ( p,0.05). For a description of statistical L-NAME-treated female transgenic hearts, at 11% initial analyses see Methods. For abbreviations see Legend to Fig. 1 LVDP than in untreated female transgenic hearts, at 31% initial LVDP ( p,0.001) and as low as in male transgenic 3... Effects of pretreatment with L-NAME hearts. There was no effect of L-NAME pretreatment on To inhibit both enos and inos, male and female functional recovery in male, at 3566% initial LVDP, and wild-type and male and female transgenic hearts were female, at 3164% initial LVDP, wild-type hearts. In perfused with 1 mmol/l L-NAME prior to ischemia. During summary, the protection from ischemia/ reperfusion injury normoxic perfusion, L-NAME had no effect on heart rate in female, compared to male, bar overexpressors was LVDP, 1dP/dt or dp/dt in male or female transgenic abolished by pretreatment with L-NAME. These results hearts (Table 1). Likewise, L-NAME had no effect on imply that the protection in female bar overexpressors is contractility in male and female wild-type hearts (data not mediated via enos or inos. shown). L-NAME also had no significant effect on coronary flow, ml/ min pre-infusion and Effects of pretreatment with 1400 W ml/ min post-infusion in male transgenic hearts and To confirm specificity of the inos inhibitor, 1400W, ml/min pre-infusion and ml/min postinos male wild-type mice were injected with MLA to induce expression and NOx was assayed in 1400W-treated, infusion in female transgenic hearts. L-NAME-treated, 1400W1L-NAME-treated hearts and untreated hearts from these mice. NOx was 663 pmol/ mg protein after 10 min ischemia in untreated hearts. NO x increased with MLA treatment, being pmol/ mg protein in MLA-only hearts, and was significantly lower in MLA11400W-treated hearts, at 865 pmol/ mg protein ( p#0.05). NOx in MLA11400W-treated hearts did not differ from that of untreated hearts, indicating that a 5 min perfusion with 100 nmol/l 1400W is sufficient to inhibit inos. It is possible that the full inhibition of MLAinduced NOx production by 1400W was not due to complete inhibition of inos, but to sub-maximal inhibition of inos plus partial inhibition of enos. However, Fig. 1. Postischemic recovery of LVDP in male wild-type (MWT, n5 based on the Ki values of Garvey et al. [18], 1400W is 15), male bar overexpressor (MTG4, n510), female wild-type (FWT, n55), female b AR overexpressor (FTG4, n510), L-NAME-treated 7000 times more specific for inos than enos, therefore it MTG4 and FTG4 (MTG41L-NAME, n56; FTG41L-NAME, n56) and is unlikely that inhibition of enos plays a significant role 1400W-treated MTG4 and FTG4 (MTG411400W, n55; FTG411400W, in our observations. The residual NOx in the MLA1 n57) mouse hearts. Data are means6s.e.m. Postischemic recovery was 1400W-treated hearts was presumably due to activity of compared between groups by a two-way analysis of variance for sex and other NOS isoforms. This was confirmed by treatment of genotype or sex and treatment followed by a Fisher s LSD test. For untreated hearts the two-way analysis revealed a significant sex and hearts with MLA1L-NAME, NOx being negligible, at 65 genotype interaction, that is, the effect of genotype differed depending on pmol/ mg protein, in MLA1L-NAME hearts. NOx was the sex. For L-NAME treated hearts the two-way analysis revealed a also negligible, at 368 pmol/ mg protein, in MLA1Lsignificant sex and treatment interaction, that is, the effect of L-NAME NAME11400W-treated hearts. differed depending on the sex. For 1400W-treated hearts the two-way When male and female transgenic hearts were perfused analysis revealed a significant sex effect but no treatment effect or sex/ treatment interaction. The results of the subsequent Fisher s analyses with 100 nmol/l 1400W, 5 min prior to ischemia, 1400W are indicated by *: significant difference from MTG4; : significant had no effect on heart rate, LVDP, 1dP/dt or dp/dt in difference from FTG4 ( p,0.05). either group during normoxic perfusion (Table 1). There Downloaded from by guest on 09 January 019

5 666 H.R. Cross et al. / Cardiovascular Research 53 (00) was no effect of pretreatment with 1400W on timing of ischemic contracture or postischemic recovery of contractile function in male or female transgenic hearts (Table ; Fig. 1). Therefore, inhibition of inos alone had no effect on ischemia/ reperfusion injury in female bar overexpressors, in contrast to inhibition of both inos and enos with L-NAME, implying that the protection in female bar overexpressors is mediated via enos Phosphate metabolite levels and intracellular ph Effects of overexpression of the bar During the preischemic period, the PCr/ATP ratio was lower in male transgenic, at , than in male wild-type hearts, at ( p,0.01). However, despite the increase in contractility in the female transgenic hearts, there was no difference in PCr/ATP ratio between female transgenic, at , and female wild-type hearts, at During ischemia, ATP levels fell lower in male transgenic hearts, reaching 16% initial ATP, than in either male wild-type, at 40% initial ATP ( p,0.0001), or female transgenic hearts, at 3% initial ATP ( p,0.05; Fig. a). There was no difference in ATP at the end of ischemia between female wild-type, at 36% initial ATP, and female transgenic hearts. During reperfusion, ATP levels remained lower in male transgenic hearts, reaching 9% initial ATP by the end of reperfusion, compared to 47% initial ATP in male wild-type ( p,0.01) and 54% initial ATP in female transgenic hearts ( p,0.01). There was no difference in end-reperfusion ATP levels between female wild-type, at 50% initial ATP, and female transgenic hearts. At the end of ischemia, there was no difference in PCr levels between any groups of hearts (Fig. b). At the end of reperfusion, PCr was lower in male transgenic hearts, at 54% initial ATP, than in either male wild-type, at 89% initial ATP ( p,0.001) or female transgenic hearts, at 89% initial ATP ( p,0.01). There was no difference in endreperfusion PCr levels between female wild-type, at 95% initial ATP, and female transgenic hearts. At the end of ischemia, ph was lower in male transgenic hearts, at ph 5.5, than in either male wild-type, at ph 5.85 ( p,0.0001), or female transgenic hearts, at ph 5.87 ( p,0.0001; Fig. 3). There was no difference in endischemic intracellular ph between female wild-type, at ph 5.85, and female transgenic hearts. During reperfusion, there were no differences in ph between any groups of hearts. Fig.. Myocardial intracellular levels of (a) ATP and (b) PCr during ischemia and reperfusion. Points are means6s.e.m. and represent data acquired over the preceding 5 min; for abbreviations see legend to Fig. 1. ATP and PCr levels at the end-point of ischemia and the end-point of reperfusion were compared between groups by a two-way analysis of variance for sex and genotype. Where the two-way analysis revealed significant differences, pairwise comparisons were made using a Fisher s LSD test which indicated that (i) end-ischemic ATP levels were lower in MTG4 than in either MWT, FWT or FTG4 ( p,0.05), (ii) end-reperfusion ATP levels were lower in MTG4 than either MWT, FWT or FTG4 ( p,0.01), (iii) end-reperfusion PCr levels were lower in MTG4 than either MWT, FWT or FTG4 ( p,0.01). In summary, male bar overexpressor hearts had a Fig. 3. Myocardial intracellular ph during ischemia and reperfusion. lower PCr/ATP ratio during basal perfusion than wild-type Points are means6s.e.m. and represent data acquired over the preceding hearts. During ischemia, these hearts exhibited a greater 5 min; for abbreviations see legend to Fig. 1. Intracellular ph at the loss of ATP and a lower intracellular ph than wild-type end-point of ischemia and the end-point of reperfusion was compared hearts. During reperfusion, recoveries of ATP and PCr between groups by a two-way analysis of variance for sex and genotype. Where the two-way analysis revealed significant differences, pairwise were lower in male bar overexpressor hearts than in comparisons were made using a Fisher s LSD test which indicated that (i) wild-type hearts, therefore correlating with the lower end-ischemic ph was lower in MTG4 than either MWT, FWT or FTG4 recovery of contractile function observed and indicating ( p,0.001). Downloaded from by guest on 09 January 019

6 H.R. Cross et al. / Cardiovascular Research 53 (00) greater injury. In contrast, despite contractility being elevated in the female bar overexpressor hearts to the same extent as in males, PCr/ATP ratios during basal perfusion were the same as in wild-type hearts and there were no differences in energy metabolites or intracellular ph during ischemia and reperfusion compared with wildtype hearts. Overexpression of the bar, therefore, resulted in increased energy demand and increased susceptibility to ischemia/ reperfusion injury in male, but not female, hearts Effects of pretreatment with L-NAME There was no effect of L-NAME treatment on energetics or intracellular ph in either male wild-type or female wild-type hearts during ischemia or reperfusion (data not shown). During ischemia, ATP levels fell lower in L-NAMEtreated female transgenic hearts, reaching 18% initial ATP, than in untreated female transgenic hearts, at 3% initial ATP (Fig. 4a; p50.05). ATP fell as low in L-NAMEtreated female transgenic hearts during ischemia as in untreated male transgenic hearts, at 16% initial ATP. During reperfusion, ATP remained lower in L-NAMEtreated female transgenic hearts, reaching 3% initial ATP, compared to 54% initial ATP in untreated female trans- Fig. 4. Myocardial intracellular levels of (a) ATP and (b) PCr during genic hearts ( p,0.05). There was no difference in endacquired over the preceding 5 min; for abbreviations see legend to Fig. 1. ischemia and reperfusion. Points are means6s.e.m. and represent data reperfusion ATP levels between L-NAME-treated female ATP and PCr levels at the end-point of ischemia and the end-point of transgenic hearts and untreated male transgenic hearts, at reperfusion were compared between groups by a two-way analysis of 9% initial ATP. At no time during the protocol were there variance for sex and treatment. Where the two-way analysis revealed differences in ATP levels between L-NAME-treated male significant differences, pairwise comparisons were made using a Fisher s transgenic and untreated male transgenic hearts (Fig. 4a). LSD test which indicated that (i) end-ischemic ATP levels were lower in At the end of ischemia, there was no difference in PCr FTG41L-NAME, MTG4 and MTG41L-NAME than FTG4 ( p,0.05), (ii) end-reperfusion ATP levels were lower in FTG41L-NAME, MTG4 levels between any groups of hearts (Fig. 4b). At the end and MTG41L-NAME than FTG4 ( p,0.05), (iii) end-reperfusion PCr of reperfusion, PCr was lower in L-NAME-treated female levels were lower in FTG41L-NAME, MTG4 and MTG41L-NAME than transgenic hearts, reaching 5% initial ATP, compared to FTG4 ( p,0.05). 89% initial ATP in untreated female transgenic hearts ( p,0.05) and as low as in untreated male transgenic hearts, at 54% initial ATP. At no time during the protocol were there differences in PCr levels between L-NAMEtreated male transgenic and untreated male transgenic hearts (Fig. 4b). At the end of ischemia, intracellular ph was lower in L-NAME-treated female transgenic hearts, at ph 5.64, than in untreated female transgenic hearts, at ph 5.87 ( p,0.05; Fig. 5). There was no difference in end-ischemic ph between L-NAME-treated female transgenic and untreated male transgenic hearts, at ph 5.5. During ischemia there was no significant difference in ph between L-NAMEtreated male transgenic and untreated male transgenic Fig. 5. Myocardial intracellular ph during ischemia and reperfusion. hearts (Fig. 5). During reperfusion, there were no signifi- Points are means6s.e.m. and represent data acquired over the preceding cant differences in ph between any groups of hearts. 5 min; for abbreviations see legend to Fig. 1. Intracellular ph at the To summarize, L-NAME-treated female b AR over- end-point of ischemia and the end-point of reperfusion was compared expressor hearts exhibited a greater loss of ATP during between groups by a two-way analysis of variance for sex and treatment. Where the two-way analysis revealed significant differences, pairwise ischemia than untreated female bar overexpressor hearts comparisons were made using a Fisher s LSD test which indicated that (i) and were not different from male bar overexpressor end-ischemic ph was lower in FTG41L-NAME, MTG4 and MTG41Lhearts. During reperfusion, recoveries of ATP and PCr NAME than FTG4 ( p,0.05). Downloaded from by guest on 09 January 019

7 668 H.R. Cross et al. / Cardiovascular Research 53 (00) were also lower in L-NAME-treated female b AR over- expressor hearts than in untreated female b AR over- expressor hearts, therefore correlating with the lower recovery of contractile function observed and indicating greater injury. There was no effect of L-NAME pretreatment, with respect to ischemic energetics or ph, on male b AR overexpressor hearts, consistent with the lack of functional effects. L-NAME pretreatment, therefore, abolished the protection of female hearts from the increased energy demand and ischemia/ reperfusion injury resulting from b AR overexpression, implying that protec- tion was mediated by inos or enos Effects of pretreatment with 1400 W Consistent with the lack of functional effects, there was no effect of pretreatment with 1400W on intracellular ph or levels of ATP or PCr during ischemia or reperfusion in male transgenic or female transgenic hearts (Figs. 6a, b and 7). Fig. 6. Myocardial intracellular levels of (a) ATP and (b) PCr during ischemia and reperfusion. Points are means6s.e.m. and represent data acquired over the preceding 5 min; for abbreviations see legend to Fig. 1. ATP and PCr levels at the end-point of ischemia and the end-point of reperfusion were compared between groups by a two-way analysis of variance for sex and treatment. For all measurements, the two-way analysis revealed a significant sex effect but no treatment effect or sex/ treatment interaction. For the sex effect, a Fisher s LSD test indicated that (i) end-ischemic ATP levels were lower in MTG4 than FTG4 ( p,0.05) and lower in MTG411400W than FTG411400W ( p,0.05), (ii) end-reperfusion ATP levels were lower in MTG4 than FTG4 ( p, 0.01) and lower in MTG411400W than FTG411400W ( p,0.01), (iii) end-reperfusion PCr levels were lower in MTG4 than FTG4 ( p,0.01) and lower in MTG411400W than FTG411400W ( p,0.05). Fig. 7. Myocardial intracellular ph during ischemia and reperfusion. Points are means6s.e.m. and represent data acquired over the preceding 5 min; for abbreviations see legend to Fig. 1. Intracellular ph at the end-point of ischemia and the end-point of reperfusion was compared between groups by a two-way analysis of variance for sex and treatment. For all measurements, the two-way analysis revealed a significant sex effect but no treatment effect or sex/ treatment interaction. For the sex effect, a Fisher s LSD test indicated that (i) end-ischemic ph was lower in MTG4 than FTG4 ( p,0.01) and lower in MTG411400W than FTG411400W ( p,0.01). In summary, inhibition of inos alone had no effect on myocardial energetics during ischemia and reperfusion in female b AR overexpressors, in contrast to the effects of inhibition of both inos and enos with L-NAME. These results are consistent with the lack of an effect of 1400W on post-ischemic functional recovery in female b AR overexpressors and supports the hypothesis that protection in female b AR overexpressors is mediated via enos NO levels in male and female hearts Since estrogen increases expression of enos and inos, NO levels would be expected to be elevated in female hearts. We found that total nitrate levels were slightly but not significantly greater under basal conditions in female wild-type, at 536 pmol/ mg protein, than male wild-type hearts, at 364 pmol/ mg protein. After ischemia, total nitrate levels were significantly greater in female wild-type hearts, at pmol/mg protein than in male wild-type hearts, at 864 pmol/ mg protein ( p#0.05). NO production was therefore slightly greater under basal conditions and significantly greater during ischemia in female than male hearts, consistent with reports that expression of enos and inos is increased by estrogen. 4. Discussion 4.1. Effects of bar overexpression on contractility and ischemia/reperfusion injury in male and female hearts In the present study, perfused hearts from both male and female b AR overexpressor mice exhibited increased peak contraction (LVDP), rate of contraction (1dP/dt) and rate Downloaded from by guest on 09 January 019

8 H.R. Cross et al. / Cardiovascular Research 53 (00) of relaxation (dp/dt), compared to wild-type hearts. to result in protection from ischemia-reperfusion injury During ischemia, ATP levels fell lower in the male bar [3]. Estrogen increases expression of both inducible nitric overexpressor hearts than male wild-type, reflecting greater oxide synthase (inos) and endothelial nitric oxide synthischemic energy utilization. The faster rate of ATP utiliza- ase (enos) in the myocardium [11]. Consistent with an tion in the male bar overexpressor hearts was consistent estrogen-mediated increase in NOS expression, we found 1 with the earlier onset of contracture observed and, as H that NO levels were higher in female than male hearts. are produced by ATP degradation, the faster rate of ATP Since NO is known to affect Ca homeostasis, moduutilization was also consistent with the lower ph observed lating sarcolemmal Ca channels [10] and SR Ca in these hearts. However, ischemic ATP depletion, timing transport proteins [8,9], and since, although contradictory of contracture onset and ischemic intracellular ph in findings exist [16,4,5], a number of studies have found female bar overexpressor hearts were not different from NO to be cardioprotective [1 14], we studied the role of that in male or female wild-type. Ischemia/ reperfusion NOS in the protection of female bar overexpressor injury was increased in hearts from male bar over- mouse hearts. expressor mice, as indicated by lower postischemic re- Male and female bar overexpressor mouse hearts coveries of contractile function, ATP and PCr than wild- were pretreated with 1 mmol/ l of the nonspecific NOS type hearts. Despite contractility being elevated in female inhibitor, L-NAME, prior to ischemia. The concentration of bar overexpressor hearts to the same level as males, L-NAME used corresponded to the concentration required ischemia/ reperfusion injury was not increased, as indi- to inhibit enos and inos without inducing vasoconstriccated by higher postischemic recoveries of contractile tion [16]. Correspondingly, we observed no effect of 1 function, ATP and PCr in female vs. male bar over- mmol/l L-NAME on contractile function and coronary flow expressors. Notably, recoveries of contractile function, in any hearts during normoxic perfusion. During ischemia, ATP and PCr in female bar overexpressors were no ATP levels and intracellular ph fell lower and the onset of different from male or female wild-type. In summary, in contracture was earlier in the L-NAME-treated female contrast to the observation in males, bar overexpression bar overexpressor hearts than in the untreated female in female mouse hearts did not result in greater ischemic bar overexpressor hearts, reflecting greater ischemic energy utilization and increased ischemia/ reperfusion in- energy utilization. Notably, ischemic ATP and ph and jury. postischemic recoveries of contractile function and energy metabolites were as low in the L-NAME-treated female 4.. Role of NOS in protection from ischemia/ bar overexpressor hearts as in untreated male bar reperfusion injury in female bar overexpressor mouse overexpressor hearts. The post-ischemic recoveries of both hearts contractile function and ATP were less in the L-NAMEtreated female bar overexpressor hearts compared to Adrenergic stimulation leads to phosphorylation of the untreated female hearts, indicating greater injury. There- L-type Ca channel and the SR Ca ATPase inhibitor fore, the protection from ischemia/ reperfusion injury phospholamban, resulting not only in increased contractili- observed in the female bar overexpressors, compared to ty, but also in increased cytosolic and SR Ca levels, male bar overexpressors, was abolished by pretreatment respectively [1]. Since Ca overload is a major mediator with L-NAME. These results imply that the protection in of ischemia/ reperfusion injury [19 ], these increases in female bar overexpressors is mediated via enos or cytosolic and SR Ca may be the mechanism by which inos. overexpression of the bar leads to increased ischemia/ To determine the relative roles of enos and inos in reperfusion injury. Interestingly, under our experimental the observed female cardioprotection, we studied the conditions, we found no male/ female difference in suscep- ischemic response of bar overexpressor hearts pretreated tibility to ischemia/ reperfusion injury in wild-type mouse with 100 nmol/ l of the potent and specific inos inhibitor, hearts. However, male/female differences in susceptibility 1400W. The Ki of 1400W is 7 nmol/l for inos, mmol/l to ischemia/reperfusion injury have been observed under for nnos and 50 mmol/l for enos [18], therefore 100 our experimental conditions in mice which overexpress the nmol/ l 1400W should be sufficient to inhibit inos without 1 Na / Ca exchanger [15], another transgenic model affecting other NOS isoforms, as shown by Cavicchi and which exhibits increased cytosolic and SR Ca []. It Whittle [6]. Inhibition of inos by 100 nmol/ l 1400W appears, therefore, that females may be protected from the was also confirmed independently in the present study. consequences of cytosolic and SR Ca overload. Clinical There was no effect of pretreatment with 1400W on timing studies indicate that protection in females is mediated via of ischemic contracture, ischemic ATP levels and intracelestrogen [4 7]. Consistent with a role for estrogen, we lular ph or post-ischemic recovery of contractile function 1 have shown previously that protection of female Na / and energy metabolites in female bar overexpressor Ca exchange overexpressor hearts is abolished by hearts. Therefore, inhibition of inos alone had no effect ovariectomy [15] and chronic treatment of ovariectomized on ischemia/ reperfusion injury in female bar overrats with physiological levels of estrogen has been shown expressors, in contrast to inhibition of both inos and Downloaded from by guest on 09 January 019

9 670 H.R. Cross et al. / Cardiovascular Research 53 (00) enos with L-NAME, implying that the protection in Acknowledgements female bar overexpressors is mediated via enos. From our findings, we can speculate on a model of W.J.K. and C.S. thank the NIH for grant support. The female cardioprotection. When cytosolic and SR Ca authors thank Sandy J. Duncan and Rachel McAdam for overload are increased, such as by bar overexpression, mouse phenotyping, Robert E. London for use of NMR secondary Ca -mediated mechanisms of injury may be facilities and Joseph Haseman for assistance with statistical invoked during ischemia/ reperfusion resulting in injury analyses. being exacerbated in male transgenic hearts. Female hearts, however, exhibit increased NO production which appears to oppose the Ca -mediated injury induced by bar References overexpression, leading to cardioprotection in the female transgenics. In the wild-type hearts, the secondary mecha- [1] England PJ, Shahid M. Effects of forskolin on contractile responses nisms of injury may not be active thus the increased NO and protein phosphorylation in the isolated perfused rat heart. production in female wild-type hearts has no protective Biochem J 1987;46: role and, as observed in the present study, susceptibility to [] Milano CA, Allen LF, Rockman HA et al. Enhanced myocardial ischemia/ reperfusion injury remains equivalent in male function in transgenic mice overexpressing the b-adrenergic receptor. Science 1994;64: and female wild-type hearts. Interestingly, our observation [3] Cross HR, Steenbergen C, Lefkowitz RJ, Koch WJ, Murphy E. that NO has no protective role under basal conditions, but Overexpression of the b-adrenergic receptor and a bark1 inhibdoes have a role under stimulated conditions is similar to itor both increase contractility but have differential effects on reports that NO does not modulate contractility under basal susceptibility to ischemic injury. Circ Res 1999;85: conditions but does modulate contractility in stimulated [4] Isles CG, Hole DH, Hawthorne VM, Lever AF. Relation between coronary risk and coronary mortality in women of the Renfrew and hearts [7]. Paisley survey: comparison with men. Lancet 199;339: The findings of this study have clinical relevance. [5] Godsland IF, Wynn V, Crook D, Miller NE. Sex, plasma lipoproteins Overexpression of the bar has been proposed as a and atherosclerosis: prevailing assumptions and outstanding quescontractile therapy for heart failure [], therefore our tions. Am Heart J 1987;114: observation that basal contractility can be increased by [6] Heckbert SR, Weiss NS, Koepsell TD et al. Duration of estrogen expression of the b AR in females without a concomitant replacement therapy in relation to the risk of incident myocardial infarction in postmenopausal women. Arch Intern Med increase in ischemia/ reperfusion injury has important 1997;157: implications for sex-specific therapies for heart failure. In [7] Sullivan JM, El-Zeky F, Van der Zwaag R, Ramanathan KB. Effect addition, our finding that females are protected from the on survival of estrogen replacement therapy after coronary bypass detrimental effects of adrenergic stimulation and Ca grafting. Am J Cardiol 1997;79: overload suggest that under conditions of adrenergic [8] Viner RI, Ferrington DA, Williams TD, Bigelow DJ, Schoneich C. Selective tyrosine nitration of the SERCAa isoform in skeletal stimulation or enhanced cytosolic or SR Ca overload, muscle. Biochem J 1999;340: such as occur clinically during cardioplegia, hypercalcemia [9] Zahradnikova A, Minarovic I, Venema RC, Meszaros LG. Inactivaor treatment with adrenergic agonists, females may be tion of the cardiac ryanodine release channel by nitric oxide. Cell protected from cardiovascular injury via an enos-me- Calcium 1997;: diated mechanism. [10] Mery PF, Pavoine C, Bellhassen L, Pecker F, Fischmeister R. Nitric oxide regulates cardiac Ca current. J Biol Chem 1993;68:686 In summary, by comparing the ischemic response of 695. hearts from male and female bar overexpressor mice, [11] Nuedling S, Kahlert S, Loebbert K et al. 17b-Estradiol stimulates we demonstrated that females were protected from the expression of endothelial and inducible NO synthase in rat myocardium in vitro and in vivo. Cardiovasc Res 1999;43: detrimental effects of b AR overexpression observed in males, namely a greater ischemic energy utilization and [1] Sumeray MS, Rees DD, Yellon DM. Infarct size and endothelial nitric oxide synthase in murine myocardium. J Mol Cell Cardiol increased ischemia/ reperfusion injury. By pretreating male 000;3:35 4. and female bar overexpressor mouse hearts with the [13] Xi L, Jarrett NC, Hess ML, Kukreja RC. Essential role of inducible nonspecific NOS inhibitor, L-NAME, and studying the nitric oxide synthase in monophosphoryl lipid A-induced late ischemic response, we demonstrated that the protection in cardioprotection. Circulation 1999;99: female b AR overexpressors was mediated via either [14] Hoshida S, Yamashita N, Igarashi J et al. Nitric oxide synthase protects the heart against ischemia-reperfusion injury in rabbits. J enos or inos. By also studying the ischemic response of Pharm Exp Ther 1995;74: bar overexpressor hearts pretreated with the specific [15] Cross HR, Lu L, Steenbergen C, Philipson KD, Murphy E. 1 inos inhibitor, 1400W, we found evidence that it was Overexpression of the cardiac Na / Ca exchanger increases enos, not inos, which mediated protection in the female susceptibility to ischemic injury in male, but not female, transgenic b AR overexpressors. These results highlight the role of mice. Circ Res 1998;83: [16] Depre C, Vanoverschelde J-L, Goudemant J-F, Mottet I, Hue L. enos in female cardioprotection and may be relevant to Protection against ischemic injury by nonvasoactive concentrations the protection from cardiovascular injury observed clinical- of nitric oxide synthase inhibitors in the perfused rabbit heart. ly in premenopausal females. Circulation 1995;9: Downloaded from by guest on 09 January 019

10 H.R. Cross et al. / Cardiovascular Research 53 (00) [17] Gyorgy K, Muller B, Vegh A, Kleschyov AL, Stoclet J-C. Triggering [3] Kolodgie FD, Farb A, Litovsky SH et al. Myocardial protection of role of nitric oxide in the delayed protective effect of mono- contractile function after global ischemia by physiologic estrogen phosphoryl lipid A in rat heart. Br J Pharmacol 1999;17:189 replacement in the ovariectomized rat. J Mol Cell Cardiol ;9: [18] Garvey EP, Oplinger JA, Furfine ES et al. 1400W is a slow, tight [4] Ronson RS, Nakamura M, Vinten-Johansen J. The cardiovascular binding and highly selective inhibitor of inducible nitric oxide effects and implications of peroxynitrite. Cardiovasc Res synthase in vitro and in vivo. J Biol Chem 1997;7: ;44: [19] Gao WD, Atar D, Liu Y et al. Role of troponin I proteolysis in the [5] Woolfson RG, Patel VC, Neild GH, Yellon DM. Inhibition of nitric pathogenesis of stunned myocardium. Circ Res 1997;80: oxide synthesis reduces infarct size by an adenosine-dependent [0] Marban E, Kitakaze M, Koretsune Y et al. Quantification of [Ca ] i mechanism. Circulation 1995;91: in perfused hearts. Circ Res 1990;66: [6] Cavicchi M, Whittle BJ. Potentiation of cytokine-induced inos [] Murphy E, Perlman M, London RE, Steenbergen C. Amiloride expression in the human intestinal epithelial cell line, DLD-1, by delays the ischemic-induced rise in cytosolic free calcium. Circ Res cyclic AMP. Gut 1999;45: ;68: [7] Hare JM, Colucci WS. Role of nitric oxide in the regulation of [] Terracciano CM, Souza AI, Philipson KD, MacLeod KT. Na1 myocardial function. Prog Cardiovasc Dis 1995;38: Ca exchange and sarcoplasmic reticular Ca regulation in ventricular myocytes from transgenic mice overexpressing the Na1 Ca exchanger. J Physiol Lond 1998;51: Downloaded from by guest on 09 January 019