- Email: [email protected]
Hypoxia-Reoxygenation and Polyunsaturated Fatty Acids Modulate Adrenergic Functions in Cultured Cardiomyocytes
J Mol Cell Cardiol 31, 377–386 (1999) Article No. jmcc.1998.0871, available online at http://www.idealibrary.com on
Hypoxia-Reoxygenation and Polyunsaturated Fatty Acids Modulate Adrenergic Functions in Cultured Cardiomyocytes Philippe Delerive1, Fabien Oudot2, Blandine Ponsard3, Sylvie Talpin1, Jean Pierre Sergiel1, Catherine Cordelet1, Pierre Athias3 and Alain Grynberg2 1
Unite´ de Nutrition Lipidique, INRA, 21034 Dijon, France; 2INRA-UNHPI-UR914, Faculte´ des Sciences Pharmacologiques et Biologiques, 75270 Paris, France and 3LPPCE, Faculte´ de me´decine, 21034 Dijon, France (Received 27 April 1998, accepted in revised form 26 October 1998) P. D, F. O, B. P, S. T, J. P. S, C. C, P. A A. G. HypoxiaReoxygenation and Polyunsaturated Fatty Acids Modulate Adrenergic Functions in Cultured Cardiomyocytes. Journal of Molecular and Cellular Cardiology (1999) 31, 377–386. The polyunsaturated fatty acids (PUFAs) of the x3 series are known to modulate adrenergic functions in ventricular myocytes. This study evaluated the influence of hypoxia duration and PUFA composition on the ability of cultured rat cardiomyocytes in producing a- and badrenergic messengers (IPs and cAMP). After hypoxia (1.5, 2.5 or 3.5 h) followed by reoxygenation (1h), IP and cAMP production was induced by phenylephrine or isoproterenol stimulation, respectively. Hypoxia did not affect the basal level of messenger production in unstimulated cells, but decreased the cAMP production elicited by isoproterenol stimulation (up to 50%). The decrease in IP production after phenylephrine stimulation was observed only after long-term hypoxia duration close to irreversible cellular damages. The use of modified culture media supplemented with either arachidonic acid (AA) or docosahexaenoic acid (DHA) induced cardiomyocytes displaying either an arachidonic acid membrane profile (35% AA and 2% DHA in the phospholipids) or a docosahexaenoic acid membrane profile (15% AA and 20% DHA). These modifications did not alter the basal level of either messenger production in unstimulated cells nor the IP released after a-adrenergic stimulation. Conversely, the decrease in cAMP production was significantly more pronounced in docosahexaenoic acid-enriched cells than in arachidonic acid-enriched cells. This study suggests that hypoxia alters the b-adrenergic messenger production, and that the a-system may balance the depression of the b-system. The depression of the b-adrenergic function induced by the incorporation of docosahexaenoic acid in membrane phospholipids may contribute to the beneficial 1999 Academic Press effect of this fatty acid in the reperfused heart. K W: Polyunsaturated fatty acids; Hypoxia; Cardiomyocyte; Adrenergic function; Membrane.
Introduction From the early epidemiological studies by Bang and Dyerberg (1972), dietary long-chain polyunsaturated fatty acids (PUFA) of the x3 series have been known for their contribution to cardiovascular
disease prevention. This effect has been progressively associated to a reduction of the incidence of atherosclerosis, based on several sites of intervention of x3 PUFAs, including plasma cholesterol and triglycerides, regulation of the vascular tone, blood pressure and platelet aggregation (Kinsella et
Please address all correspondence to Alain Grynberg, INRA-UNHPI-UR914, Faculte´ des Sciences Pharmacologiques et Biologiques, 4 Av de l’Observatoire, 75270 Paris, France.
0022–2828/99/020377+10 $30.00/0
1999 Academic Press
378
P. Delerive et al.
al., 1990). But several data clearly show that beside this preventive effect on the cardiovascular risk factors, the x3 PUFAs are able to exert a beneficial effect on the heart itself during ischemia, through their incorporation in membrane phospholipids (PL). Gudbjarnason and Hallgrimsson (1976) reported that in sudden cardiac death, the ratio of arachidonic acid (AA, 20:4 x6) to docosahexaenoic acid (DHA, 22:6 x3) in cardiac membrane phospholipids is frequently much higher than expected. Dietary x3 PUFAs elicit a very low AA/DHA ratio in cardiac membranes (Charnock et al., 1986, Nalbone et al., 1989), which appears as a key factor that significantly affects the production of prostaglandin by the heart (Charnock et al., 1992) and the ventricular myocyte (Oudot et al., 1995). Other mechanisms have also been pointed out to explain the cardiac effects of long chain x3 PUFAs. McLennan and coworkers reported the beneficial effect of docosahexaenoic acid on ischemia- and reperfusioninduced arrhythmia (McLennan, 1992, McLennan et al., 1996). This effect was observed on several species, provided that the cardiac phospholipid content in docosahexaenoic acid was close to 20%, a feature that requires a dietary intake of docosahexaenoic acid, and not its precursors. In fish oil fed rats, the possibility of a decreased oxygen requirement for energy production from mitochondria was reported by several authors (Pepe et al., 1992; Demaison et al., 1994). The decrease of both the x6/x3 ratio and the AA/DHA ratio largely influences the contraction (spontaneous beating activity) and the electrical activity (action potential parameters) during hypoxia and reoxygenation in isolated ventricular myocytes (Grynberg et al., 1988, Durot et al., 1996), although the mechanisms involved remain largely unknown. The possible involvement of the adrenergic control of cardiac cell function was investigated by several authors and numerous studies outlined the modulation of adrenoceptors by x3 PUFAs. Studies on the badrenergic system showed that the activity of rat heart adenylate cyclase decreased in essential fatty acid deficiency and increased in x3 PUFA enriched hearts (Alam et al., 1987, 1988). In a previous paper, we reported that the adrenergic system could be specifically related to the docosahexaenoic acid content rather than the x3/x6 ratio (Grynberg et al., 1995). In isolated ventricular myocytes, the adenylate cyclase activity was lower in DHA enriched cells than in EPA enriched cells although both groups displayed the same arachidonic acid content. But in spite of this reduced activity, the efficiency of cAMP was higher in the DHA rich cardiomyocytes (Grynberg et al., 1996). The in-
fluence of x3 PUFAs on the a-adrenergic pathway is also a matter of debate. The a-adrenoceptors, considered as less important in the regulation of myocardial contractility, have been reported to be involved in the development of arrhythmia during ischemia (Endoh et al., 1991). Several results suggest that cardiac inositide phosphate production by phospholipase C as elicited by a-adrenergic agonist is not influenced by x3 PUFAs (Du et al., 1993; Grynberg et al., 1995). Conversely, x3 PUFAs were reported to reduce the inotropic a-adrenergic response in the isolated perfused rat heart (Reibel et al., 1988) and the a-adrenoceptor-mediated contractile response in rat femoral artery (Macleod et al., 1994). This fall was partly explained by the substitution of arachidonic acid by docosahexaenoic acid which provokes a decreased diacylglycerol efficiency for PKC activation (Bordoni et al., 1991). The objective of the present study was to investigate the maintenance during hypoxia and reoxygenation of the cardiac cell capacity to produce second messengers in response to adrenergic stimulation, and to evaluate the influence of the membrane phospholipid fatty acid composition.
Materials and Methods Cell culture Primary cultures of rat ventricular myocytes were prepared as previously described (Grynberg et al., 1986). The hearts were removed aseptically from 2–4 days-old Wistar rats. The ventricles were minced and the cardiac cells were dissociated during seven proteolytic treatments in 0.1% trypsin (Difco) at 30°C. The cells from the last six steps were resuspended in culture medium and the preparation was enriched in myocytes by two successive preplating periods (30 and 120 min). The final suspension was adjusted to 4×105 cells/ ml and seeded in 60 mm Petri dishes (Falcon Primaria 3002; 2×106 cells/dish). The cells were grown in a standard culture medium (Ham’ F10 medium supplemented with 10% fetal calf serum and 10% human serum and antibiotics) in a humidified atmosphere at 37°C containing 5% CO2. The medium was renewed after 24 h and every 48 h thereafter. After 24 h in standard medium, the cells were further grown either in standard medium or in one of the experimental media for four days and then submitted to experimental treatments.
379
Cardiac PUFAs and Second Messengers in Hypoxia Table 1 Fatty acid composition (as % of total fatty acids) of the standard medium (Stdmed) and the experimental media enriched with either arachidonic acid or docosahexaenoic acid (AAmed and DHAmed, respectively). (% of total fatty acids) Fatty acid C16:0 C18:0 C20:0 C22:0 RC16:1 RC18:1 RC20:1 C18:2x6 C18:3x6 C20:2x6 C20:3x6 C20:4x6 C22:4x6 C22:5x6 C18:3x3 C20:5x3 C22:5x3 C22:6x3
Stdmed 23.0 10.4 0.1 0.3 3.4 20.8 0.3 26.0 0.3 0.3 1.7 8.0 0.2 0.1 0.4 0.9 1.1 2.7
Membrane fatty acid alterations The experimental media, enriched with either arachidonic acid or docosahexaenoic acid (100 l), were prepared by addition of fatty acid bound to bovine serum albumin (Sigma) to the standard culture medium as previously described (Grynberg et al., 1992). The fatty acid composition of the standard medium and the experimental media is shown in Table 1. After one day in the standard culture medium, the cells were grown for a further 4 days in either of the experimental media to produce the AA or DHA enriched cardiomyocytes (AAcells and DHAcells, respectively). Lipid analysis The lipids were extracted from the media and the fatty acids were analysed as previously described (Liautaud et al., 1991). The cell lipids were extracted (Folch et al., 1957), the phospholipids were separated from non phosphorous lipids in silica cartridges (Sep-pack Waters), and the fatty acids were transmethylated with BF3-methanol (Morrison and Smith, 1964). The fatty acid methyl-esters were analysed by gas chromotography on a Carbowax 20M capillary column (Chevalier et al., 1990). Hypoxia-reoxygenation The experiments were realized in a specially designed gas-controlled device with culture dishes
DHAmed 20.4 8.1 0.1 0.2 3.0 19.8 0.3 23.0 0.3 0.2 1.5 6.7 0.2 0.1 0.4 0.8 1.0 13.9
AAmed 21.2 6.6 0.1 0.2 3.3 20.3 0.3 24.3 0.3 0.2 1.5 16.2 0.2 0.1 0.4 0.9 0.9 3.2
covered with a layer of paraffin oil (Chevalier et al., 1990). The culture medium was removed and replaced by 5 ml of glucose-free Puck’s F saline solution (pH 7.4) and the cultures were submitted to hypoxia and reoxygenation (HR) first by incubation under N2 at 37°C for hypoxia period (H) of 1.5, 2.5 or 3.5 h and then for one hour of reoxygenation (R). The time course evolution of oxygen pressure is presented in Figure 1. It was similar to that reported previously in similar conditions (Fantini et al., 1990). These conditions were chosen on the basis of previous data indicating that the functional recovery of the cardiomyocytes after hypoxia was satisfactory in terms of action potential and contraction (Fantini et al., 1990). This was again controlled in this study.
Adrenergic messengers production After hypoxia-reoxygenation (HR), the cells were rinsed in saline and 3-isobutyl-1 methylxanthine (0.5×10−3 M) (Sigma) was added. The b-adrenergic system was stimulated for 3 min at 37°C by l-Isoproterenol HCl (ISO, Sigma) (10−6 M). The cells were then harvested at 0°C. The intracellular cyclic 3′5′-adenosine monophosphate (cAMP) content was determined by radioimmunoassay using a commercial kit (Amersham). The experiment was realized on three control and three ISO-treated dishes in each of the 3 groups (Stdcells, DHAcells and AAcells), and the whole experiment was repeated
380
P. Delerive et al.
on three different culture preparations. The a-adrenergic-induced intracellular messenger production was measured as total IPn (Grynberg et al., 1995, 1996). The culture dishes were incubated overnight in serum-free Ham’s F10 medium containing 1 lCi/ml myo 3H-inositol (NEN), before hypoxia and reoxygenation treatment. Alphaadrenergic stimulation was achieved by a 60 min activation with phenylephrine 10−4 M (PHE; Sigma) and the phosphatidylinositol breakdown was quantified in the presence of lithium (10 m) according to a method adapted from Berridge et al. (1983) and Masters et al. (1984). The cells were then harvested and sonicated in chloroform/methanol/ water (5:10:4). The solvent mixture was brought to 10:10:5 and the extract was centrifuged (3000 g, 3 min). The radiolabelled IPs were extracted from the upper phase by a batch addition of anion exchange resin (AG1-X8, Biorad), and the resin was cleared from 3H-inositol by rinsing with unlabelled myoinositol (5 m). The 3H-inositide phosphates were eluted from the resin with ammonium formate (1 M) in formic acid (0.2 M) and the radioactivity was measured in aqueous scintillation liquid (ACSII, Amersham). The experiment was realized on three control dishes and three PHE-treated dishes for the each of the 3 groups, and the whole experiment was repeated on three different culture preparations. The biochemical constants (Bmax and Kd) of the membrane b-adrenergic receptors were evaluated on 150 lg of cell homogenate protein by conventional binding techniques, using 125I-iodocyanopindolol (2125 Ci/mMole, Amersham) in TrisMg buffer as specific ligand and alprenolol (5×10−5 M) for non-specific binding. The adrenergic stimulation concentration and duration used throughout this study were standardized from previous studies (Ettaiche et al., 1985, Courtois et al., 1992, Grynberg et al., 1995) to take into account the functional response in terms of contraction, the biochemical response in terms of second messenger, and the requirements of a single drug addition in each dish to avoid interference with the fast desensitization in vitro.
Statistics The data were expressed as mean ± ..., and submitted to a two-way analysis of variance (ANOVA) with the fatty acids and the HR times considered as fixed factors. When significantly different, the means were compared by the NewmanKeuls test (Dagnelie, 1975).
Figure 1 Evolution of oxygen pressure in the bathing fluid in the experimental conditions.
Results Cell phospholipid fatty acid composition The fatty acid composition of the cardiomyocytes was modified to obtain cells displaying either a x6 or a x3 profile. The fatty acid composition of the phospholipid fraction of these cells is presented in Table 2. Increasing the PUFA content in the media did not alter the amount of total saturated fatty acids in the PL fraction. Conversely, growing the cardiomyocytes in a PUFA-enriched medium resulted in an increased proportion of long chain PUFAs in the PL fraction (in AAcells and DHAcells, as compared with Stdcells), partly compensated by a slight decrease in MUFAs. In these conditions, the p/s ratio (PUFA/SFA) was raised from 1 (Stdcells) to 1.4 (AAcells and DHAcells). As evidenced in Table 2, the main modifications induced by growing the myocytes in the experimental media were the individual fatty acid alterations in the PUFA composition of the PL. AAcells displayed a very high level of arachidonic acid (35%) and its elongation product C22:4 x6 (9%), which was poorly desaturated in C22:5 x6 (0.4%). All these fatty acids were significantly higher than in Stdcells. Interestingly, the increase in arachidonic acid was also associated with a decrease in linoleic acid. Moreover, the x3 PUFAs (C22:5 and C22:6) were slightly but significantly reduced (−50%, approximately) in AAcells as compared to Stdcells. Overall, the x6/x3 ratio was increased from 5 in Stdcells to 12 in AAcells and mainly represented the AA/DHA ratio (5 and 16, respectively). On the other hand, growing the cells on docosahexaenoic acid supplemented medium led to a sharp increase of docosahexaenoic acid in the phospholipids (up to 21%) without major increase in the other x3 PUFAs (EPA and DPA). All the x6 PUFAs, including arachidonic and linoleic acids, were reduced in the
381
Cardiac PUFAs and Second Messengers in Hypoxia Table 2 Fatty acid composition of the phospholipids of Stdcells, DHAcells and AAcells after 4 days in experimental media (Mean ± ..., n=9 as 3 dishes×3 cultures) Fatty acid
Stdcells
DHAcells
AAcells
ANOVA
C16:0 RC16:1 C18:0 RC18:1 C18:2x6 C20:2x6 C20:3x6 C20:4x6 C22:4x6 C22:5x6 C20:5x3 C22:5x3 C22:6x3 RSFA RMUFA RPUFA Rx6 Rx3 x6/x3 PUFA/SFA
18.5±3.95 1.7±0.60 21.3±1.60 13.9±3.03a 11.0±1.31a 0.8±0.30a 1.8±0.21a 17.8±1.93a 2.6±0.48a 0.4±0.22a 0.7±0.30a 2.4±0.64a 3.9±0.49a 9.8±0.80 15.6±1.45a 41.4±4.74a 34.4±3.64a 7.0±1.36a 4.9±0.71a 1.0
13.4±0.88 1.1±0.22 22.7±1.11 12.4±0.23b 7.8±0.86b 0.3±0.06b 1.5±0.38a 15.4±0.28b 1.1±0.54b trb 2.5±0.53b 1.1±0.33b 20.5±3.65b 36.1±1.98 13.5±0.01b 50.2±1.87b 26.1±1.58b 24.1±3.45b 1.1±0.22b 1.4
13.1±0.57 0.9±0.07 23.2±0.38 10.5±0.94c 3.3±0.66c 0.3±0.10b 0.7±0.13b 34.8±2.28c 8.7±2.49c 0.4±0.24a 0.2±0.05c 1.5±0.23c 2.2±0.68c 36.3±0.18 11.4±0.87c 52.1±0.72b 48.2±0.23c 3.9±0.95c 12.4±3.08c 1.4
ns ns ns ns P<0.01 P<0.05 P<0.05 p<0.01 P<0.01 p<0.05 P<0.01 P<0.05 P<0.01 ns P<0.05 P<0.01 P<0.01 P<0.01 P<0.01
The means with different letters are significantly different according to the P value appearing in the ANOVA column (ns: not significant, tr: traces, SFA: saturated fatty acids, MUFA: monounsaturated fatty acids, PUFA: polyunsaturared fatty acids).
phospholipids, resulting in very low x6/x3 and AA/DHA ratios (both close to 1).
Intracellular messenger production during hypoxia and reoxygenation The intracellular formation in cyclic 3′5′ adenosine monophosphate (cAMP) was analysed in isoproterenol stimulated and unstimulated cells in the standard cell group after increasing durations of hypoxia followed by 1 h of reoxygenation. As shown on Figure 2, the ventricular myocytes display a very significant biochemical response to b-adrenergic stimulation in normoxic conditions. After 1.5 h of hypoxia, the stimulated cAMP production was slightly reduced to 80% of the value measured in normoxia. This difference with normoxia control was not significant, possibly because of the large ... which may be associated with individual sensitivity in the resistance to hypoxia. Increasing the hypoxia duration further decreased the capacity in cAMP production to 40% after 2.5 h (P<0.01) and 30% after 3.5 h (P<0.01). The basal production of cAMP as measured in unstimulated cells was not significantly affected by hypoxia-reoxygenation, whatever the duration of hypoxia. The Bmax values of the b-adrenergic receptors as determined in control cells and after hypoxia (2.5 h) and reoxygenation
Figure 2 Influence of hypoxia duration on the production of cAMP induced by b-adrenergic stimulation (Isoproterenol, 10−6 M) 1 h after reoxygenation in cardiomyocytes grown on standard medium. The data (mean ± ..., n=9) are expressed as % of the response to stimulation in normoxic conditions. The mean of each ISO groups was significantly different from its unstimulated control (P<0.01). The means of the ISO groups with different letters are significantly different (P<0.01).
were 29.2±12.3, 19.0±2.1 and 18.9±2.7 fmol/ mg protein (mean±), respectively. These values were not significantly different. The Kd values were 31.8±11.2, 21.2±2.2, and 57.6±24.3 pM, respectively, and were also not significantly different. The phosphatidylinositol breakdown was analysed in phenylephrine stimulated and unstimulated
382
P. Delerive et al.
Figure 3 Influence of hypoxia duration on the production of 3H-[IP] induced by a-adrenergic stimulation (Phenylephrine, 10−4 M) 1 h after reoxygenation in cardiomyocytes grown on standard medium. The data (mean ± ..., n=9) are expressed as % of the response to stimulation in normoxic conditions. The mean of each PHE treated group was significantly different from its unstimulated control (P<0.01). The means with different letters are significantly different (P<0.05).
cells (Stdcells) after increasing duration of hypoxia followed by 1 h of reoxygenation. The data are presented in Figure 3 and show that the ventricular myocytes display a significant biochemical response to a-adrenergic stimulation in normoxic conditions, as described above for b-adrenergic stimulation. However, the ability of the cells to produce 2nd messengers during a-adrenergic stimulation was less affected by hypoxia, since after 2.5 h hypoxia the response was still 100% of the value measured in normoxia. The decrease in phospholipase C efficiency (−50%) was observed only after 3.5 h hypoxia. Similarly to the observation reported for cAMP, the basal production of IP as measured in unstimulated cells was not significantly affected by hypoxia-reoxygenation, whatever the duration of hypoxia.
Influence of membrane PUFAs on intracellular messenger production The same experiments were realized with PUFAenriched cells (AAcells and DHAcells) grown in arachidonic acid and docosahexaenoic acid supplemented media, respectively. The data are presented in Figure 4. In these groups of cells, the evolution of cAMP production ability during hypoxia was similar to that of the control cells. However, the decrease was delayed since the cAMP production was still not significantly reduced after 1.5 h hypoxia. The cAMP synthesis decreased significantly after 2.5 h hypoxia and this diminution
Figure 4 Influence of the phospholipid PUFA composition (AAcells v DHAcells) on the effect of hypoxia duration on the cardiomyocyte production of cAMP (panel A) induced by b-adrenergic stimulation (Isoproterenol, 10−6 M) and 3H-[IP] (panel B) induced by aadrenergic (Phenylephrine, 10−4 M) stimulation. Adrenergic stimulation was done 1 hour after post-hypoxic reoxygenation. The data (mean ± ..., n=9) are expressed as % of the response to stimulation in normoxia conditions. According to ANOVA, all the means after 2.5 and 3.5 h were significantly different from normoxia control (P<0.01). (∗ significant difference between AAcells and DHAcells, P<0.05).
was more pronounced in DHAcells (−60%) than in AAcells (−45%). This difference between DHAcells and AAcells was maintained after 3.5 h hypoxia (−70%, and −50%, respectively), suggesting that the PUFA composition of the membranes may influence the evolution of the adenylate cyclase activity of the cardiomyocyte during hypoxia, and its efficiency at reoxygenation. The modification of the membrane PUFA composition elicited a moderate difference in the b-adrenergic receptor Kd in normoxic condition (50.4±11.1 v 26.5±3.9 pM in DHAcells and AAcells, respectively, P<0.05), but not in Bmax (26.0±3.3 v 19.15±2.3 fmol/mg protein, respectively). Moreover, no further significant difference appeared between DHAcells and AAcells in Kd or Bmax as a consequence of hypoxia and reoxygenation. As evidenced in Figure 4, the evolution of the cellular capacity in producing IP as a response to a-adrenergic stimulation was also delayed in cells
Cardiac PUFAs and Second Messengers in Hypoxia
enriched with PUFAs, since no significant decrease could be measured throughout the study, even after 3.5 h hypoxia. This result contrasts with the control cells (Stdcells), which displayed a reduced capacity in these conditions. The results were however quite different from those presented above for the badrenergic system, since the evolution of the aadrenergic induced messenger production was similar in DHAcells and AAcells. Whatever the system considered, (a- or b-adrenergic) the fatty acid composition of the cardiomyocyte membranes did not influence the basal level of cAMP or 3H-IP, in unstimulated cells.
Discussion This study was carried out to investigate the cross interaction between the time-dependent effect of hypoxia (1.5, 2.5 and 3.5 h) followed by reoxygenation (1 h) and the membrane PUFA composition on a- and b-adrenergic intracellular messenger production in cultured rat ventricular myocytes. The manipulation of the fatty acid composition of the media resulted in a corresponding modification of the phospholipid fatty acid profile of the cardiomyocytes similar to those reported in the literature (Courtois et al., 1992; Grynberg et al., 1992; Oudot et al., 1995). Incubation with arachidonic acid led to myocytes displaying high content of arachidonic acid (35%) and its elongation product C22:4 x6 (9%) and a very low total x3 PUFA content (4%). Conversely, the incubation in docosahexaenoic acid enriched medium induced a large increase in docosahexaenoic acid in the phospholipids (20%) with a very low retroconversion to 20:5 x3 (EPA). This treatment of the media resulted in cardiomyocytes that appear as a satisfactory model regarding the DHA/AA balance in vitro, since the ratio fits the values reported for the antiarrhythmic effects of docosahexaenoic acid in vivo (McLennan, 1992, McLennan et al., 1996). As observed in vivo as well as in vitro, such a docosahexaenoic acid content in the phospholipids (20%) can only be reached when docosahexaenoic acid is supplied itself in the diet. This could be related to the absence of D4-desaturation in the cardiac cell (Liautaud et al., 1991, Grynberg et al., 1996). The results of this study indicate that the cAMP synthesis capacity is significantly affected in the course of the hypoxic treatment. The decrease appeared roughly at the same stage as the lactate dehydrogenase release in the same conditions
383
(Fantini et al., 1990). In some cultures, the decrease appeared earlier and was already observed after 1.5 hours. This variability may explain the large ... at this time, which could be close to the transition point from the unaffected cAMP synthesis to the reduced cAMP synthesis capacity. Moreover, the time course evolution was roughly similar to that of ATP content (Chevalier et al., 1990) and electrical parameters (Grynberg et al., 1988, Fantini et al., 1990, Durot et al., 1997). In these various studies, the parameters were followed during the course of hypoxia, because reoxygenation elicited a recovery in electromechanical parameters (Fantini et al., 1990), although in the present study the adrenergic stimulation was induced 1 h after the onset of reoxygenation. The persistence of the low level of cAMP after reoxygenation was reported in a model of isolated rat heart submitted to global ischemia, which induced a decrease in cAMP synthesis (−40%) without recovery during reoxygenation (Van den Ende et al., 1991). The data reported here are roughly similar (−50%), although they were obtained from isolated cells in glucose-free hypoxia. Moreover, the amplitude of the decrease in cAMP synthesis capacity increased with the hypoxia duration. The decline of cAMP synthesis capacity appeared to reach a plateau after 3.5 h of hypoxia. Interestingly, this hypoxia duration was considered in this model as the transition time from reversible to irreversible damage, on the basis of electrophysiological parameters (Fantini et al., 1990). Among the several components of the b-adrenergic system that can modulate the synthesis of cAMP, the molecular basis for the hypoxia-induced decrease is still debated. Van den Ende et al. (1991) showed that the decrease they observed in isolated heart was not related to the receptor itself, but was associated with a decrease of adenylate cyclase activity and an increase of the G protein of the inhibitor type. Similarly, Homcy et al. (1991) reported that neither the receptor density (Bmax) nor its affinity (Kd) were affected by ischemia in the heart. The present study led to similar results in ventricular myocytes, since we did not observe any hypoxia-induced alteration in b-adrenoceptor Bmax and Kd. Other data report that ligation of the left coronary artery in isolated rat hearts induces the b-adrenoceptor-coupled G protein dysfunction (Yamamoto et al., 1994). Other hypotheses were raised including the effect of hypoxia-induced acidosis on the adenylate cyclase catalytic site (Van den Ende et al., 1991) and a feed-back mechanism depressing cAMP synthesis through hypoxia-induced ATP deficiency (Hugtenburg et al., 1991). The IP production was affected in ventricular
384
P. Delerive et al.
myocytes only after a long time exposure to hypoxia (3.5 hours). Similar results were reported in isolated hearts in which a long ischemia induced a reduction of IP production (Muntz et al., 1993) and a sharp decrease of DAG production (Kawai et al., 1990). All these data support the observation that long ischemic (or hypoxic) treatments are required to affect the a-adrenergic receptor mediated IP production in the cardiac muscle. The comparison with other hypoxia-induced events reported in the same model suggests that this decrease could be associated to the “irreversible state” corresponding to a shift to irreversible functional damages (Fantini et al., 1990). The decrease in a-adrenergic receptor mediated IP production would then parallel the cell death process, and this outlines the difference in behavior during ischemia, between the ventricular myocyte b-adrenergic system which displayed a very fast decrease in cAMP production capacity, and the a-adrenergic system which displayed a delayed decrease in IP production capacity. Several reports pointed out the independence of adenylate cyclase activity towards membrane PUFA profile (see for review McMurchie, 1988). Recently, we reported that the b-adrenergic response of the ventricular myocytes and the cAMP production was influenced by the DHA content in the membrane. This effect was shown to be specific of the presence of DHA rather than related on the x6/x3 ratio (Grynberg et al., 1996; Ponsard et al., 1998). Enrichment of the cells with PUFAs appeared to induce a delay in the hypoxia-induced decrease in cAMP synthesis capacity, since no decrease was observed after 1.5 h in the 2 PUFA-enriched groups whereas at the same time, the synthesis in the control cells began to be reduced in some of the control cells. As suggested above, this could indicate the balance point between resistance to hypoxia and failure. Beyond this point (after 2.5 and 3.5 h of hypoxia), the amplitude of the reduction was significant, and significantly more pronounced in docosahexaenoic acid rich cells than in arachidonic acid rich cells. In a previous study, we reported that the isoproterenol-induced cAMP production was lower in docosahexaenoic acid rich cells than in cells containing other x3 PUFAs, although the 2 groups displayed the same content in arachidonic acid (Grynberg et al., 1995). Docosahexaenoic acid appears thus to decrease the b-adrenergic biochemical response in normoxia and to trigger this decrease in hypoxia, through pathways which do not involve receptor occupancy. This moderate but significant “b-blocking-like” effect could contribute to explain the beneficial effect of this fatty acid in the heart.
In the PUFA enriched cardiomyocytes, the aagonist-induced IP production was not significantly decreased, even after long-term hypoxia, whereas it was reduced after 3.5 h in control cells. The increase of membrane PUFAs was able to delay the hypoxia-induced decrease in IP production, as well as in cAMP production as discussed above. Conversely, no significant differences could be observed between docosahexaenoic acid enriched cells and AA-enriched cells. Similarly, Meij et al. (1990) reported that the total amount of inositide phosphates produced in rat heart cells after a-adrenergic stimulation was not affected by changes in PUFA composition of membrane phospholipids. In conclusion, the mechanisms of action of dietary fish oil long chain PUFAs on the heart remains largely unknown. The available data suggest that this influence should be multifactorial. In previous papers, we have shown that the beneficial effects of x3 PUFAs could be related to their influence on electromechanical and biochemical characteristics of the ventricular myocyte (Grynberg et al., 1988; Chevalier et al., 1990; Durot et al., 1997). The present study extends these earlier results and suggests a “b-blocking-like” effect of these PUFAs since the presence of docosahexaenoic acid in membrane phospholipids favored the decrease of cAMP production capacity after hypoxia-reoxygenation, but did not affect the IP production capacity. Although observed during reoxygenation, this phenomenon was related with the duration of hypoxia. Moreover, this study reported a modification by hypoxia of the a/b system balance in the hypoxic heart which raises the question of the role of the a-adrenergic system in pathological situations.
Acknowledgement The skillful technical assistance of Simone Almanza and Martine Degois is gratefully acknowledged. This work was supported by a grant from the Conseil Re´gional de Bourgogne.
References A SQ, A BS, R YF. 1987. Adenylate cyclase activity, membrane fluidity and fatty acid composition of rat heart in essential fatty acid deficiency. J Mol Cell Cardiol 19: 465–475. A SQ, R YF, A BS. 1988. (3H) Forskolin and (3H) dihydroalprenolol-binding sites and adenylate cyclase activity in heart of rats fed diets containing different oils. Lipids 23: 207–213.
Cardiac PUFAs and Second Messengers in Hypoxia B HO, D J. 1972. Plasma lipids and lipoproteins in Greenlandic west coast eskimos. Acta Med Scand 192: 85–94. B MJ, D RMC, D CP, H P, I F. 1983. Changes in the levels of inositol phosphates after agonist-dependent hydrolysis of membrane phosphoinositides. Biochem J 212: 473–482. B A, B PL, R CA, H S. 1991. a1 stimulated phosphoinositide breakdown in cultured cardiomyocytes: Diacylglycerol production and composition in docosahexaenoic acid supplemented cells. Biochem Biophys Res Commun 174: 869–877. C JS, A MY, ML PL. 1986. Comparative changes in the fatty acid composition of rat cardiac phospholipids after long term feeding of sunflower seed oil- or tuna fish oil-supplemented diets. Am Nutr Metab 30: 393–406. C JS., ML PL, A MY. 1992. Dietary modulation of lipid metabolism and mechanical performance of the heart. Mol Cell Biochem 116: 19–25. C A, D L, G A, A P. 1990. Influence of the phospholipid polyunsaturated fatty acid composition on some metabolic disorders induced in rat cardiomyocytes by hypoxia and reoxygenation. J Mol Cell Cardiol 22: 1177–1186. C M, K, S, F E, A P, M P, G A. 1992. Polyunsaturated fatty acids in cultured cardiomyocytes: effect on physiology and badrenoceptor function. Am J Physiol 262, H451–H456. D P. 1975. The´ories et me´thodes statistiques, Vol. 2. Gembloux, Belgium: Presses Agronomiques de Gembloux. D L, S JP, M D, G A. 1994. Influence of the phospholipid n-6/n-3 polyunsaturated fatty acid ratio on mitochondrial oxidative metabolism before and after myocardial ischemia. Biochim Biophys Acta 1227: 53–59. D XJ, D AM, R RA. 1993. Lack of modulation by dietary unsaturated fats on sympathetic neurotransmission in rat hearts. Am J Physiol 265: 886–892. D I, A P, O F, G A. 1997. Influence of phospholipid long chain polyunsaturated fatty acid composition on rat cardiomyocyte function in physiological conditions and during hypoxia-reoxygenation. Mol Cell Biochem 175: 253–262. E M. 1991. Myocardial adrenoceptors: multiplicity of subcellular coupling process. Asia Pacific J Pharmacol 6: 171–186. F E, A P, C M, K S, G A, D M. 1990. Oxygen and substrate deprivation on isolated rat cardiac myocytes: close correlation between electromechanical and biochemical consequences. Can J Physiol Pharmacol 68: 1148–1156. F J, L M, S-S GH. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem 226: 497–509. G A, A P, D M. 1986. Effect of change in growth environment on cultured myocardial cells investigated in a standardized medium. In Vitro Cell Develop Biol 22: 44–50. G A, F E, A P, D M, G L, C M, K S. 1988. Modification of the n-6/n-3 fatty acid ratio in the phospholipids of rat ventricular myocytes in culture by the use of synthetic media: functional and biochemical consequences in
385
normoxic and hypoxic conditions. J Mol Cell Cardiol 20: 863–874. G A, N G, L J, L H, A P. 1992. Eicosapentaenoic and docosahexaenoic acids in cultured rat ventricular myocytes and hypoxiainduced alterations of phospholipase-A activity. Mol Cell Biochem 116: 75–78. G A, F A, S JP, A P. 1995. Effect of docosahexaenoic acid and eicosapentaenoic acid in the phospholipids of rat heart muscle cells on adrenoceptor responsiveness and mechanism. J Mol Cell Cardiol 27: 2507–2520. G A, F A, S JP, A P. 1996. Membrane DHA versus EPA and the beating function of the cardiomyocytes through the adrenergic receptors. Lipids 31: S205–S210. G S, H J. 1976. Prostaglandins and polyunsaturated fatty acids in heart muscle. Acta Biol Med Germ 35: 1069–1080. H CJ, V SF, V DE. 1991. b-adrenergic receptor regulation in the heart in pathophysiological state: abnormal adrenergic responsiveness in cardiac disease. Ann Rev Physiol 53: 137–159. H JG, M MJ, H N, B JJ, V Z PA. 1991. The influence of calcium antagonists on the adenine nucleotide metabolism in the guinea pig working heart during ischaemia and reperfusion. Arch Pharmacol 343: 496–504. Kı¨ T, O K, H H, S T. 1990. Alteration of 1,2 diacylglycerol content in ischemic and reperfused heart. Mol Cell Biochem 99: 1–8. K JE, L B, S RA. 1990. Dietary n-3 polyunsaturated fatty acids and amelioration of cardiovascular disease: possible mechanisms. Am J Clin Nutr 52: 1–28. L S, G A, M J, A P. 1991. Contribution to the use of cultured cardiomyocytes as a model in nutritional studies: fatty acid composition of the hearts of rats fed linseed oil or sunflower oil and cardiomyocytes grown on their sera. Cardioscience 2: 55–61. ML DC, H AM, B SJ, L TS, R RA. 1994. Effect of dietary polyunsaturated fatty acids on contraction and relaxation of rat femoral resistance arteries. J Cardiovasc Pharmacol 23: 92– 98. M SB, H TK, B JH. 1984. Relationship between phosphoinositides and calcium responses to muscarinic agonists in astrocytoma cells. Mol Pharmacol 26: 149–155. ML PL. 1993. Relative effects of dietary saturated, monounsaturated and polyunsaturated fatty acids on cardiac arrhythmias in the rat. Am J Clin Nutr 57: 207–212. ML PL, H PR, A MY, M R, R D, M M, R T, H RJ. 1996. The cardiovascular protective role of docosahexaenoic acid. Eur J Pharmacol 300: 83–89. MM EJ. 1988. Dietary lipids and the regulation of membrane fluidity and function, in Advances in membrane fluidity, Vol 3, Physiological Regulation of membrane fluidity (Aloia, Curtain and Gordon) New York: Alan R. Liss Inc. 189–237. M JTA, B A, D DHW, L JMJ, G C. 1990. Alterations in polyunsaturated fatty acid composition of cardiac membrane phospholipids
386
P. Delerive et al.
and a1 adrenoceptor mediated phosphatidylinositol turnover. Cardiovasc Res 24: 94–101. M WR, S LM. 1964. Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron-fluoride-methanol. J Lipid Res 5: 600–608. M KH, N SL, M JC. 1993. Alterations in a-1 adrenergic receptor mediated phosphatidylinositol turnover in hypoxic cardiac myocytes. J Mol Cell Cardiol 25: 1187–1202. N G, L J, T E, P H, L` P, P AM, L H. 1989. Effects of fish oil, corn oil and lard diets on lipid peroxidation status and glutathione peroxidase activities in rat heart. Lipids 24: 179–186. O F, G A, S JP. 1995. Synthesis of eicosanoids in cardiomyocytes during hypoxia and reoxygenation: influence of the polyunsaturated fatty acid composition. Am J Physiol 268: H308–H315. P S, ML PL. 1992. Modulation of myocardial oxygen requirements by dietary lipids in the isolated
erythrocyte perfused working rat heart. J Mol Cell Cardiol 24: S115. P B, D I, F A, O F, A P, G A. 1998. Long chain polyunsaturated fatty acids influence both b- and a- adrenergic function of rat cardiomyocytes. J Am Oil Chem Soc 75: 247– 254. R DK, H MA, H CE. 1988. Effects of dietary fish oil on cardiac responsiveness to adrenoceptor stimulation. Am J Physiol 254: 494– 499. V E R, B HD, P M, V Z PA. 1991. Discrepancies between inotropic responses and b adrenoceptor characteristics after global ischemia in isolated hearts J Cardiovasc Pharmacol 18: 679–686. Y J, O M, M M, I T. 1994. b-adrenoceptor-G protein-adenylate cyclase complex in rat heart failure produced by coronary artery ligation. J Mol Cell Cardiol 26: 617–626.
Hypoxia-Reoxygenation and Polyunsaturated Fatty Acids Modulate Adrenergic Functions in Cultured Cardiomyocytes Philippe Delerive1, Fabien Oudot2, Blandine Ponsard3, Sylvie Talpin1, Jean Pierre Sergiel1, Catherine Cordelet1, Pierre Athias3 and Alain Grynberg2 1
Unite´ de Nutrition Lipidique, INRA, 21034 Dijon, France; 2INRA-UNHPI-UR914, Faculte´ des Sciences Pharmacologiques et Biologiques, 75270 Paris, France and 3LPPCE, Faculte´ de me´decine, 21034 Dijon, France (Received 27 April 1998, accepted in revised form 26 October 1998) P. D, F. O, B. P, S. T, J. P. S, C. C, P. A A. G. HypoxiaReoxygenation and Polyunsaturated Fatty Acids Modulate Adrenergic Functions in Cultured Cardiomyocytes. Journal of Molecular and Cellular Cardiology (1999) 31, 377–386. The polyunsaturated fatty acids (PUFAs) of the x3 series are known to modulate adrenergic functions in ventricular myocytes. This study evaluated the influence of hypoxia duration and PUFA composition on the ability of cultured rat cardiomyocytes in producing a- and badrenergic messengers (IPs and cAMP). After hypoxia (1.5, 2.5 or 3.5 h) followed by reoxygenation (1h), IP and cAMP production was induced by phenylephrine or isoproterenol stimulation, respectively. Hypoxia did not affect the basal level of messenger production in unstimulated cells, but decreased the cAMP production elicited by isoproterenol stimulation (up to 50%). The decrease in IP production after phenylephrine stimulation was observed only after long-term hypoxia duration close to irreversible cellular damages. The use of modified culture media supplemented with either arachidonic acid (AA) or docosahexaenoic acid (DHA) induced cardiomyocytes displaying either an arachidonic acid membrane profile (35% AA and 2% DHA in the phospholipids) or a docosahexaenoic acid membrane profile (15% AA and 20% DHA). These modifications did not alter the basal level of either messenger production in unstimulated cells nor the IP released after a-adrenergic stimulation. Conversely, the decrease in cAMP production was significantly more pronounced in docosahexaenoic acid-enriched cells than in arachidonic acid-enriched cells. This study suggests that hypoxia alters the b-adrenergic messenger production, and that the a-system may balance the depression of the b-system. The depression of the b-adrenergic function induced by the incorporation of docosahexaenoic acid in membrane phospholipids may contribute to the beneficial 1999 Academic Press effect of this fatty acid in the reperfused heart. K W: Polyunsaturated fatty acids; Hypoxia; Cardiomyocyte; Adrenergic function; Membrane.
Introduction From the early epidemiological studies by Bang and Dyerberg (1972), dietary long-chain polyunsaturated fatty acids (PUFA) of the x3 series have been known for their contribution to cardiovascular
disease prevention. This effect has been progressively associated to a reduction of the incidence of atherosclerosis, based on several sites of intervention of x3 PUFAs, including plasma cholesterol and triglycerides, regulation of the vascular tone, blood pressure and platelet aggregation (Kinsella et
Please address all correspondence to Alain Grynberg, INRA-UNHPI-UR914, Faculte´ des Sciences Pharmacologiques et Biologiques, 4 Av de l’Observatoire, 75270 Paris, France.
0022–2828/99/020377+10 $30.00/0
1999 Academic Press
378
P. Delerive et al.
al., 1990). But several data clearly show that beside this preventive effect on the cardiovascular risk factors, the x3 PUFAs are able to exert a beneficial effect on the heart itself during ischemia, through their incorporation in membrane phospholipids (PL). Gudbjarnason and Hallgrimsson (1976) reported that in sudden cardiac death, the ratio of arachidonic acid (AA, 20:4 x6) to docosahexaenoic acid (DHA, 22:6 x3) in cardiac membrane phospholipids is frequently much higher than expected. Dietary x3 PUFAs elicit a very low AA/DHA ratio in cardiac membranes (Charnock et al., 1986, Nalbone et al., 1989), which appears as a key factor that significantly affects the production of prostaglandin by the heart (Charnock et al., 1992) and the ventricular myocyte (Oudot et al., 1995). Other mechanisms have also been pointed out to explain the cardiac effects of long chain x3 PUFAs. McLennan and coworkers reported the beneficial effect of docosahexaenoic acid on ischemia- and reperfusioninduced arrhythmia (McLennan, 1992, McLennan et al., 1996). This effect was observed on several species, provided that the cardiac phospholipid content in docosahexaenoic acid was close to 20%, a feature that requires a dietary intake of docosahexaenoic acid, and not its precursors. In fish oil fed rats, the possibility of a decreased oxygen requirement for energy production from mitochondria was reported by several authors (Pepe et al., 1992; Demaison et al., 1994). The decrease of both the x6/x3 ratio and the AA/DHA ratio largely influences the contraction (spontaneous beating activity) and the electrical activity (action potential parameters) during hypoxia and reoxygenation in isolated ventricular myocytes (Grynberg et al., 1988, Durot et al., 1996), although the mechanisms involved remain largely unknown. The possible involvement of the adrenergic control of cardiac cell function was investigated by several authors and numerous studies outlined the modulation of adrenoceptors by x3 PUFAs. Studies on the badrenergic system showed that the activity of rat heart adenylate cyclase decreased in essential fatty acid deficiency and increased in x3 PUFA enriched hearts (Alam et al., 1987, 1988). In a previous paper, we reported that the adrenergic system could be specifically related to the docosahexaenoic acid content rather than the x3/x6 ratio (Grynberg et al., 1995). In isolated ventricular myocytes, the adenylate cyclase activity was lower in DHA enriched cells than in EPA enriched cells although both groups displayed the same arachidonic acid content. But in spite of this reduced activity, the efficiency of cAMP was higher in the DHA rich cardiomyocytes (Grynberg et al., 1996). The in-
fluence of x3 PUFAs on the a-adrenergic pathway is also a matter of debate. The a-adrenoceptors, considered as less important in the regulation of myocardial contractility, have been reported to be involved in the development of arrhythmia during ischemia (Endoh et al., 1991). Several results suggest that cardiac inositide phosphate production by phospholipase C as elicited by a-adrenergic agonist is not influenced by x3 PUFAs (Du et al., 1993; Grynberg et al., 1995). Conversely, x3 PUFAs were reported to reduce the inotropic a-adrenergic response in the isolated perfused rat heart (Reibel et al., 1988) and the a-adrenoceptor-mediated contractile response in rat femoral artery (Macleod et al., 1994). This fall was partly explained by the substitution of arachidonic acid by docosahexaenoic acid which provokes a decreased diacylglycerol efficiency for PKC activation (Bordoni et al., 1991). The objective of the present study was to investigate the maintenance during hypoxia and reoxygenation of the cardiac cell capacity to produce second messengers in response to adrenergic stimulation, and to evaluate the influence of the membrane phospholipid fatty acid composition.
Materials and Methods Cell culture Primary cultures of rat ventricular myocytes were prepared as previously described (Grynberg et al., 1986). The hearts were removed aseptically from 2–4 days-old Wistar rats. The ventricles were minced and the cardiac cells were dissociated during seven proteolytic treatments in 0.1% trypsin (Difco) at 30°C. The cells from the last six steps were resuspended in culture medium and the preparation was enriched in myocytes by two successive preplating periods (30 and 120 min). The final suspension was adjusted to 4×105 cells/ ml and seeded in 60 mm Petri dishes (Falcon Primaria 3002; 2×106 cells/dish). The cells were grown in a standard culture medium (Ham’ F10 medium supplemented with 10% fetal calf serum and 10% human serum and antibiotics) in a humidified atmosphere at 37°C containing 5% CO2. The medium was renewed after 24 h and every 48 h thereafter. After 24 h in standard medium, the cells were further grown either in standard medium or in one of the experimental media for four days and then submitted to experimental treatments.
379
Cardiac PUFAs and Second Messengers in Hypoxia Table 1 Fatty acid composition (as % of total fatty acids) of the standard medium (Stdmed) and the experimental media enriched with either arachidonic acid or docosahexaenoic acid (AAmed and DHAmed, respectively). (% of total fatty acids) Fatty acid C16:0 C18:0 C20:0 C22:0 RC16:1 RC18:1 RC20:1 C18:2x6 C18:3x6 C20:2x6 C20:3x6 C20:4x6 C22:4x6 C22:5x6 C18:3x3 C20:5x3 C22:5x3 C22:6x3
Stdmed 23.0 10.4 0.1 0.3 3.4 20.8 0.3 26.0 0.3 0.3 1.7 8.0 0.2 0.1 0.4 0.9 1.1 2.7
Membrane fatty acid alterations The experimental media, enriched with either arachidonic acid or docosahexaenoic acid (100 l), were prepared by addition of fatty acid bound to bovine serum albumin (Sigma) to the standard culture medium as previously described (Grynberg et al., 1992). The fatty acid composition of the standard medium and the experimental media is shown in Table 1. After one day in the standard culture medium, the cells were grown for a further 4 days in either of the experimental media to produce the AA or DHA enriched cardiomyocytes (AAcells and DHAcells, respectively). Lipid analysis The lipids were extracted from the media and the fatty acids were analysed as previously described (Liautaud et al., 1991). The cell lipids were extracted (Folch et al., 1957), the phospholipids were separated from non phosphorous lipids in silica cartridges (Sep-pack Waters), and the fatty acids were transmethylated with BF3-methanol (Morrison and Smith, 1964). The fatty acid methyl-esters were analysed by gas chromotography on a Carbowax 20M capillary column (Chevalier et al., 1990). Hypoxia-reoxygenation The experiments were realized in a specially designed gas-controlled device with culture dishes
DHAmed 20.4 8.1 0.1 0.2 3.0 19.8 0.3 23.0 0.3 0.2 1.5 6.7 0.2 0.1 0.4 0.8 1.0 13.9
AAmed 21.2 6.6 0.1 0.2 3.3 20.3 0.3 24.3 0.3 0.2 1.5 16.2 0.2 0.1 0.4 0.9 0.9 3.2
covered with a layer of paraffin oil (Chevalier et al., 1990). The culture medium was removed and replaced by 5 ml of glucose-free Puck’s F saline solution (pH 7.4) and the cultures were submitted to hypoxia and reoxygenation (HR) first by incubation under N2 at 37°C for hypoxia period (H) of 1.5, 2.5 or 3.5 h and then for one hour of reoxygenation (R). The time course evolution of oxygen pressure is presented in Figure 1. It was similar to that reported previously in similar conditions (Fantini et al., 1990). These conditions were chosen on the basis of previous data indicating that the functional recovery of the cardiomyocytes after hypoxia was satisfactory in terms of action potential and contraction (Fantini et al., 1990). This was again controlled in this study.
Adrenergic messengers production After hypoxia-reoxygenation (HR), the cells were rinsed in saline and 3-isobutyl-1 methylxanthine (0.5×10−3 M) (Sigma) was added. The b-adrenergic system was stimulated for 3 min at 37°C by l-Isoproterenol HCl (ISO, Sigma) (10−6 M). The cells were then harvested at 0°C. The intracellular cyclic 3′5′-adenosine monophosphate (cAMP) content was determined by radioimmunoassay using a commercial kit (Amersham). The experiment was realized on three control and three ISO-treated dishes in each of the 3 groups (Stdcells, DHAcells and AAcells), and the whole experiment was repeated
380
P. Delerive et al.
on three different culture preparations. The a-adrenergic-induced intracellular messenger production was measured as total IPn (Grynberg et al., 1995, 1996). The culture dishes were incubated overnight in serum-free Ham’s F10 medium containing 1 lCi/ml myo 3H-inositol (NEN), before hypoxia and reoxygenation treatment. Alphaadrenergic stimulation was achieved by a 60 min activation with phenylephrine 10−4 M (PHE; Sigma) and the phosphatidylinositol breakdown was quantified in the presence of lithium (10 m) according to a method adapted from Berridge et al. (1983) and Masters et al. (1984). The cells were then harvested and sonicated in chloroform/methanol/ water (5:10:4). The solvent mixture was brought to 10:10:5 and the extract was centrifuged (3000 g, 3 min). The radiolabelled IPs were extracted from the upper phase by a batch addition of anion exchange resin (AG1-X8, Biorad), and the resin was cleared from 3H-inositol by rinsing with unlabelled myoinositol (5 m). The 3H-inositide phosphates were eluted from the resin with ammonium formate (1 M) in formic acid (0.2 M) and the radioactivity was measured in aqueous scintillation liquid (ACSII, Amersham). The experiment was realized on three control dishes and three PHE-treated dishes for the each of the 3 groups, and the whole experiment was repeated on three different culture preparations. The biochemical constants (Bmax and Kd) of the membrane b-adrenergic receptors were evaluated on 150 lg of cell homogenate protein by conventional binding techniques, using 125I-iodocyanopindolol (2125 Ci/mMole, Amersham) in TrisMg buffer as specific ligand and alprenolol (5×10−5 M) for non-specific binding. The adrenergic stimulation concentration and duration used throughout this study were standardized from previous studies (Ettaiche et al., 1985, Courtois et al., 1992, Grynberg et al., 1995) to take into account the functional response in terms of contraction, the biochemical response in terms of second messenger, and the requirements of a single drug addition in each dish to avoid interference with the fast desensitization in vitro.
Statistics The data were expressed as mean ± ..., and submitted to a two-way analysis of variance (ANOVA) with the fatty acids and the HR times considered as fixed factors. When significantly different, the means were compared by the NewmanKeuls test (Dagnelie, 1975).
Figure 1 Evolution of oxygen pressure in the bathing fluid in the experimental conditions.
Results Cell phospholipid fatty acid composition The fatty acid composition of the cardiomyocytes was modified to obtain cells displaying either a x6 or a x3 profile. The fatty acid composition of the phospholipid fraction of these cells is presented in Table 2. Increasing the PUFA content in the media did not alter the amount of total saturated fatty acids in the PL fraction. Conversely, growing the cardiomyocytes in a PUFA-enriched medium resulted in an increased proportion of long chain PUFAs in the PL fraction (in AAcells and DHAcells, as compared with Stdcells), partly compensated by a slight decrease in MUFAs. In these conditions, the p/s ratio (PUFA/SFA) was raised from 1 (Stdcells) to 1.4 (AAcells and DHAcells). As evidenced in Table 2, the main modifications induced by growing the myocytes in the experimental media were the individual fatty acid alterations in the PUFA composition of the PL. AAcells displayed a very high level of arachidonic acid (35%) and its elongation product C22:4 x6 (9%), which was poorly desaturated in C22:5 x6 (0.4%). All these fatty acids were significantly higher than in Stdcells. Interestingly, the increase in arachidonic acid was also associated with a decrease in linoleic acid. Moreover, the x3 PUFAs (C22:5 and C22:6) were slightly but significantly reduced (−50%, approximately) in AAcells as compared to Stdcells. Overall, the x6/x3 ratio was increased from 5 in Stdcells to 12 in AAcells and mainly represented the AA/DHA ratio (5 and 16, respectively). On the other hand, growing the cells on docosahexaenoic acid supplemented medium led to a sharp increase of docosahexaenoic acid in the phospholipids (up to 21%) without major increase in the other x3 PUFAs (EPA and DPA). All the x6 PUFAs, including arachidonic and linoleic acids, were reduced in the
381
Cardiac PUFAs and Second Messengers in Hypoxia Table 2 Fatty acid composition of the phospholipids of Stdcells, DHAcells and AAcells after 4 days in experimental media (Mean ± ..., n=9 as 3 dishes×3 cultures) Fatty acid
Stdcells
DHAcells
AAcells
ANOVA
C16:0 RC16:1 C18:0 RC18:1 C18:2x6 C20:2x6 C20:3x6 C20:4x6 C22:4x6 C22:5x6 C20:5x3 C22:5x3 C22:6x3 RSFA RMUFA RPUFA Rx6 Rx3 x6/x3 PUFA/SFA
18.5±3.95 1.7±0.60 21.3±1.60 13.9±3.03a 11.0±1.31a 0.8±0.30a 1.8±0.21a 17.8±1.93a 2.6±0.48a 0.4±0.22a 0.7±0.30a 2.4±0.64a 3.9±0.49a 9.8±0.80 15.6±1.45a 41.4±4.74a 34.4±3.64a 7.0±1.36a 4.9±0.71a 1.0
13.4±0.88 1.1±0.22 22.7±1.11 12.4±0.23b 7.8±0.86b 0.3±0.06b 1.5±0.38a 15.4±0.28b 1.1±0.54b trb 2.5±0.53b 1.1±0.33b 20.5±3.65b 36.1±1.98 13.5±0.01b 50.2±1.87b 26.1±1.58b 24.1±3.45b 1.1±0.22b 1.4
13.1±0.57 0.9±0.07 23.2±0.38 10.5±0.94c 3.3±0.66c 0.3±0.10b 0.7±0.13b 34.8±2.28c 8.7±2.49c 0.4±0.24a 0.2±0.05c 1.5±0.23c 2.2±0.68c 36.3±0.18 11.4±0.87c 52.1±0.72b 48.2±0.23c 3.9±0.95c 12.4±3.08c 1.4
ns ns ns ns P<0.01 P<0.05 P<0.05 p<0.01 P<0.01 p<0.05 P<0.01 P<0.05 P<0.01 ns P<0.05 P<0.01 P<0.01 P<0.01 P<0.01
The means with different letters are significantly different according to the P value appearing in the ANOVA column (ns: not significant, tr: traces, SFA: saturated fatty acids, MUFA: monounsaturated fatty acids, PUFA: polyunsaturared fatty acids).
phospholipids, resulting in very low x6/x3 and AA/DHA ratios (both close to 1).
Intracellular messenger production during hypoxia and reoxygenation The intracellular formation in cyclic 3′5′ adenosine monophosphate (cAMP) was analysed in isoproterenol stimulated and unstimulated cells in the standard cell group after increasing durations of hypoxia followed by 1 h of reoxygenation. As shown on Figure 2, the ventricular myocytes display a very significant biochemical response to b-adrenergic stimulation in normoxic conditions. After 1.5 h of hypoxia, the stimulated cAMP production was slightly reduced to 80% of the value measured in normoxia. This difference with normoxia control was not significant, possibly because of the large ... which may be associated with individual sensitivity in the resistance to hypoxia. Increasing the hypoxia duration further decreased the capacity in cAMP production to 40% after 2.5 h (P<0.01) and 30% after 3.5 h (P<0.01). The basal production of cAMP as measured in unstimulated cells was not significantly affected by hypoxia-reoxygenation, whatever the duration of hypoxia. The Bmax values of the b-adrenergic receptors as determined in control cells and after hypoxia (2.5 h) and reoxygenation
Figure 2 Influence of hypoxia duration on the production of cAMP induced by b-adrenergic stimulation (Isoproterenol, 10−6 M) 1 h after reoxygenation in cardiomyocytes grown on standard medium. The data (mean ± ..., n=9) are expressed as % of the response to stimulation in normoxic conditions. The mean of each ISO groups was significantly different from its unstimulated control (P<0.01). The means of the ISO groups with different letters are significantly different (P<0.01).
were 29.2±12.3, 19.0±2.1 and 18.9±2.7 fmol/ mg protein (mean±), respectively. These values were not significantly different. The Kd values were 31.8±11.2, 21.2±2.2, and 57.6±24.3 pM, respectively, and were also not significantly different. The phosphatidylinositol breakdown was analysed in phenylephrine stimulated and unstimulated
382
P. Delerive et al.
Figure 3 Influence of hypoxia duration on the production of 3H-[IP] induced by a-adrenergic stimulation (Phenylephrine, 10−4 M) 1 h after reoxygenation in cardiomyocytes grown on standard medium. The data (mean ± ..., n=9) are expressed as % of the response to stimulation in normoxic conditions. The mean of each PHE treated group was significantly different from its unstimulated control (P<0.01). The means with different letters are significantly different (P<0.05).
cells (Stdcells) after increasing duration of hypoxia followed by 1 h of reoxygenation. The data are presented in Figure 3 and show that the ventricular myocytes display a significant biochemical response to a-adrenergic stimulation in normoxic conditions, as described above for b-adrenergic stimulation. However, the ability of the cells to produce 2nd messengers during a-adrenergic stimulation was less affected by hypoxia, since after 2.5 h hypoxia the response was still 100% of the value measured in normoxia. The decrease in phospholipase C efficiency (−50%) was observed only after 3.5 h hypoxia. Similarly to the observation reported for cAMP, the basal production of IP as measured in unstimulated cells was not significantly affected by hypoxia-reoxygenation, whatever the duration of hypoxia.
Influence of membrane PUFAs on intracellular messenger production The same experiments were realized with PUFAenriched cells (AAcells and DHAcells) grown in arachidonic acid and docosahexaenoic acid supplemented media, respectively. The data are presented in Figure 4. In these groups of cells, the evolution of cAMP production ability during hypoxia was similar to that of the control cells. However, the decrease was delayed since the cAMP production was still not significantly reduced after 1.5 h hypoxia. The cAMP synthesis decreased significantly after 2.5 h hypoxia and this diminution
Figure 4 Influence of the phospholipid PUFA composition (AAcells v DHAcells) on the effect of hypoxia duration on the cardiomyocyte production of cAMP (panel A) induced by b-adrenergic stimulation (Isoproterenol, 10−6 M) and 3H-[IP] (panel B) induced by aadrenergic (Phenylephrine, 10−4 M) stimulation. Adrenergic stimulation was done 1 hour after post-hypoxic reoxygenation. The data (mean ± ..., n=9) are expressed as % of the response to stimulation in normoxia conditions. According to ANOVA, all the means after 2.5 and 3.5 h were significantly different from normoxia control (P<0.01). (∗ significant difference between AAcells and DHAcells, P<0.05).
was more pronounced in DHAcells (−60%) than in AAcells (−45%). This difference between DHAcells and AAcells was maintained after 3.5 h hypoxia (−70%, and −50%, respectively), suggesting that the PUFA composition of the membranes may influence the evolution of the adenylate cyclase activity of the cardiomyocyte during hypoxia, and its efficiency at reoxygenation. The modification of the membrane PUFA composition elicited a moderate difference in the b-adrenergic receptor Kd in normoxic condition (50.4±11.1 v 26.5±3.9 pM in DHAcells and AAcells, respectively, P<0.05), but not in Bmax (26.0±3.3 v 19.15±2.3 fmol/mg protein, respectively). Moreover, no further significant difference appeared between DHAcells and AAcells in Kd or Bmax as a consequence of hypoxia and reoxygenation. As evidenced in Figure 4, the evolution of the cellular capacity in producing IP as a response to a-adrenergic stimulation was also delayed in cells
Cardiac PUFAs and Second Messengers in Hypoxia
enriched with PUFAs, since no significant decrease could be measured throughout the study, even after 3.5 h hypoxia. This result contrasts with the control cells (Stdcells), which displayed a reduced capacity in these conditions. The results were however quite different from those presented above for the badrenergic system, since the evolution of the aadrenergic induced messenger production was similar in DHAcells and AAcells. Whatever the system considered, (a- or b-adrenergic) the fatty acid composition of the cardiomyocyte membranes did not influence the basal level of cAMP or 3H-IP, in unstimulated cells.
Discussion This study was carried out to investigate the cross interaction between the time-dependent effect of hypoxia (1.5, 2.5 and 3.5 h) followed by reoxygenation (1 h) and the membrane PUFA composition on a- and b-adrenergic intracellular messenger production in cultured rat ventricular myocytes. The manipulation of the fatty acid composition of the media resulted in a corresponding modification of the phospholipid fatty acid profile of the cardiomyocytes similar to those reported in the literature (Courtois et al., 1992; Grynberg et al., 1992; Oudot et al., 1995). Incubation with arachidonic acid led to myocytes displaying high content of arachidonic acid (35%) and its elongation product C22:4 x6 (9%) and a very low total x3 PUFA content (4%). Conversely, the incubation in docosahexaenoic acid enriched medium induced a large increase in docosahexaenoic acid in the phospholipids (20%) with a very low retroconversion to 20:5 x3 (EPA). This treatment of the media resulted in cardiomyocytes that appear as a satisfactory model regarding the DHA/AA balance in vitro, since the ratio fits the values reported for the antiarrhythmic effects of docosahexaenoic acid in vivo (McLennan, 1992, McLennan et al., 1996). As observed in vivo as well as in vitro, such a docosahexaenoic acid content in the phospholipids (20%) can only be reached when docosahexaenoic acid is supplied itself in the diet. This could be related to the absence of D4-desaturation in the cardiac cell (Liautaud et al., 1991, Grynberg et al., 1996). The results of this study indicate that the cAMP synthesis capacity is significantly affected in the course of the hypoxic treatment. The decrease appeared roughly at the same stage as the lactate dehydrogenase release in the same conditions
383
(Fantini et al., 1990). In some cultures, the decrease appeared earlier and was already observed after 1.5 hours. This variability may explain the large ... at this time, which could be close to the transition point from the unaffected cAMP synthesis to the reduced cAMP synthesis capacity. Moreover, the time course evolution was roughly similar to that of ATP content (Chevalier et al., 1990) and electrical parameters (Grynberg et al., 1988, Fantini et al., 1990, Durot et al., 1997). In these various studies, the parameters were followed during the course of hypoxia, because reoxygenation elicited a recovery in electromechanical parameters (Fantini et al., 1990), although in the present study the adrenergic stimulation was induced 1 h after the onset of reoxygenation. The persistence of the low level of cAMP after reoxygenation was reported in a model of isolated rat heart submitted to global ischemia, which induced a decrease in cAMP synthesis (−40%) without recovery during reoxygenation (Van den Ende et al., 1991). The data reported here are roughly similar (−50%), although they were obtained from isolated cells in glucose-free hypoxia. Moreover, the amplitude of the decrease in cAMP synthesis capacity increased with the hypoxia duration. The decline of cAMP synthesis capacity appeared to reach a plateau after 3.5 h of hypoxia. Interestingly, this hypoxia duration was considered in this model as the transition time from reversible to irreversible damage, on the basis of electrophysiological parameters (Fantini et al., 1990). Among the several components of the b-adrenergic system that can modulate the synthesis of cAMP, the molecular basis for the hypoxia-induced decrease is still debated. Van den Ende et al. (1991) showed that the decrease they observed in isolated heart was not related to the receptor itself, but was associated with a decrease of adenylate cyclase activity and an increase of the G protein of the inhibitor type. Similarly, Homcy et al. (1991) reported that neither the receptor density (Bmax) nor its affinity (Kd) were affected by ischemia in the heart. The present study led to similar results in ventricular myocytes, since we did not observe any hypoxia-induced alteration in b-adrenoceptor Bmax and Kd. Other data report that ligation of the left coronary artery in isolated rat hearts induces the b-adrenoceptor-coupled G protein dysfunction (Yamamoto et al., 1994). Other hypotheses were raised including the effect of hypoxia-induced acidosis on the adenylate cyclase catalytic site (Van den Ende et al., 1991) and a feed-back mechanism depressing cAMP synthesis through hypoxia-induced ATP deficiency (Hugtenburg et al., 1991). The IP production was affected in ventricular
384
P. Delerive et al.
myocytes only after a long time exposure to hypoxia (3.5 hours). Similar results were reported in isolated hearts in which a long ischemia induced a reduction of IP production (Muntz et al., 1993) and a sharp decrease of DAG production (Kawai et al., 1990). All these data support the observation that long ischemic (or hypoxic) treatments are required to affect the a-adrenergic receptor mediated IP production in the cardiac muscle. The comparison with other hypoxia-induced events reported in the same model suggests that this decrease could be associated to the “irreversible state” corresponding to a shift to irreversible functional damages (Fantini et al., 1990). The decrease in a-adrenergic receptor mediated IP production would then parallel the cell death process, and this outlines the difference in behavior during ischemia, between the ventricular myocyte b-adrenergic system which displayed a very fast decrease in cAMP production capacity, and the a-adrenergic system which displayed a delayed decrease in IP production capacity. Several reports pointed out the independence of adenylate cyclase activity towards membrane PUFA profile (see for review McMurchie, 1988). Recently, we reported that the b-adrenergic response of the ventricular myocytes and the cAMP production was influenced by the DHA content in the membrane. This effect was shown to be specific of the presence of DHA rather than related on the x6/x3 ratio (Grynberg et al., 1996; Ponsard et al., 1998). Enrichment of the cells with PUFAs appeared to induce a delay in the hypoxia-induced decrease in cAMP synthesis capacity, since no decrease was observed after 1.5 h in the 2 PUFA-enriched groups whereas at the same time, the synthesis in the control cells began to be reduced in some of the control cells. As suggested above, this could indicate the balance point between resistance to hypoxia and failure. Beyond this point (after 2.5 and 3.5 h of hypoxia), the amplitude of the reduction was significant, and significantly more pronounced in docosahexaenoic acid rich cells than in arachidonic acid rich cells. In a previous study, we reported that the isoproterenol-induced cAMP production was lower in docosahexaenoic acid rich cells than in cells containing other x3 PUFAs, although the 2 groups displayed the same content in arachidonic acid (Grynberg et al., 1995). Docosahexaenoic acid appears thus to decrease the b-adrenergic biochemical response in normoxia and to trigger this decrease in hypoxia, through pathways which do not involve receptor occupancy. This moderate but significant “b-blocking-like” effect could contribute to explain the beneficial effect of this fatty acid in the heart.
In the PUFA enriched cardiomyocytes, the aagonist-induced IP production was not significantly decreased, even after long-term hypoxia, whereas it was reduced after 3.5 h in control cells. The increase of membrane PUFAs was able to delay the hypoxia-induced decrease in IP production, as well as in cAMP production as discussed above. Conversely, no significant differences could be observed between docosahexaenoic acid enriched cells and AA-enriched cells. Similarly, Meij et al. (1990) reported that the total amount of inositide phosphates produced in rat heart cells after a-adrenergic stimulation was not affected by changes in PUFA composition of membrane phospholipids. In conclusion, the mechanisms of action of dietary fish oil long chain PUFAs on the heart remains largely unknown. The available data suggest that this influence should be multifactorial. In previous papers, we have shown that the beneficial effects of x3 PUFAs could be related to their influence on electromechanical and biochemical characteristics of the ventricular myocyte (Grynberg et al., 1988; Chevalier et al., 1990; Durot et al., 1997). The present study extends these earlier results and suggests a “b-blocking-like” effect of these PUFAs since the presence of docosahexaenoic acid in membrane phospholipids favored the decrease of cAMP production capacity after hypoxia-reoxygenation, but did not affect the IP production capacity. Although observed during reoxygenation, this phenomenon was related with the duration of hypoxia. Moreover, this study reported a modification by hypoxia of the a/b system balance in the hypoxic heart which raises the question of the role of the a-adrenergic system in pathological situations.
Acknowledgement The skillful technical assistance of Simone Almanza and Martine Degois is gratefully acknowledged. This work was supported by a grant from the Conseil Re´gional de Bourgogne.
References A SQ, A BS, R YF. 1987. Adenylate cyclase activity, membrane fluidity and fatty acid composition of rat heart in essential fatty acid deficiency. J Mol Cell Cardiol 19: 465–475. A SQ, R YF, A BS. 1988. (3H) Forskolin and (3H) dihydroalprenolol-binding sites and adenylate cyclase activity in heart of rats fed diets containing different oils. Lipids 23: 207–213.
Cardiac PUFAs and Second Messengers in Hypoxia B HO, D J. 1972. Plasma lipids and lipoproteins in Greenlandic west coast eskimos. Acta Med Scand 192: 85–94. B MJ, D RMC, D CP, H P, I F. 1983. Changes in the levels of inositol phosphates after agonist-dependent hydrolysis of membrane phosphoinositides. Biochem J 212: 473–482. B A, B PL, R CA, H S. 1991. a1 stimulated phosphoinositide breakdown in cultured cardiomyocytes: Diacylglycerol production and composition in docosahexaenoic acid supplemented cells. Biochem Biophys Res Commun 174: 869–877. C JS, A MY, ML PL. 1986. Comparative changes in the fatty acid composition of rat cardiac phospholipids after long term feeding of sunflower seed oil- or tuna fish oil-supplemented diets. Am Nutr Metab 30: 393–406. C JS., ML PL, A MY. 1992. Dietary modulation of lipid metabolism and mechanical performance of the heart. Mol Cell Biochem 116: 19–25. C A, D L, G A, A P. 1990. Influence of the phospholipid polyunsaturated fatty acid composition on some metabolic disorders induced in rat cardiomyocytes by hypoxia and reoxygenation. J Mol Cell Cardiol 22: 1177–1186. C M, K, S, F E, A P, M P, G A. 1992. Polyunsaturated fatty acids in cultured cardiomyocytes: effect on physiology and badrenoceptor function. Am J Physiol 262, H451–H456. D P. 1975. The´ories et me´thodes statistiques, Vol. 2. Gembloux, Belgium: Presses Agronomiques de Gembloux. D L, S JP, M D, G A. 1994. Influence of the phospholipid n-6/n-3 polyunsaturated fatty acid ratio on mitochondrial oxidative metabolism before and after myocardial ischemia. Biochim Biophys Acta 1227: 53–59. D XJ, D AM, R RA. 1993. Lack of modulation by dietary unsaturated fats on sympathetic neurotransmission in rat hearts. Am J Physiol 265: 886–892. D I, A P, O F, G A. 1997. Influence of phospholipid long chain polyunsaturated fatty acid composition on rat cardiomyocyte function in physiological conditions and during hypoxia-reoxygenation. Mol Cell Biochem 175: 253–262. E M. 1991. Myocardial adrenoceptors: multiplicity of subcellular coupling process. Asia Pacific J Pharmacol 6: 171–186. F E, A P, C M, K S, G A, D M. 1990. Oxygen and substrate deprivation on isolated rat cardiac myocytes: close correlation between electromechanical and biochemical consequences. Can J Physiol Pharmacol 68: 1148–1156. F J, L M, S-S GH. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem 226: 497–509. G A, A P, D M. 1986. Effect of change in growth environment on cultured myocardial cells investigated in a standardized medium. In Vitro Cell Develop Biol 22: 44–50. G A, F E, A P, D M, G L, C M, K S. 1988. Modification of the n-6/n-3 fatty acid ratio in the phospholipids of rat ventricular myocytes in culture by the use of synthetic media: functional and biochemical consequences in
385
normoxic and hypoxic conditions. J Mol Cell Cardiol 20: 863–874. G A, N G, L J, L H, A P. 1992. Eicosapentaenoic and docosahexaenoic acids in cultured rat ventricular myocytes and hypoxiainduced alterations of phospholipase-A activity. Mol Cell Biochem 116: 75–78. G A, F A, S JP, A P. 1995. Effect of docosahexaenoic acid and eicosapentaenoic acid in the phospholipids of rat heart muscle cells on adrenoceptor responsiveness and mechanism. J Mol Cell Cardiol 27: 2507–2520. G A, F A, S JP, A P. 1996. Membrane DHA versus EPA and the beating function of the cardiomyocytes through the adrenergic receptors. Lipids 31: S205–S210. G S, H J. 1976. Prostaglandins and polyunsaturated fatty acids in heart muscle. Acta Biol Med Germ 35: 1069–1080. H CJ, V SF, V DE. 1991. b-adrenergic receptor regulation in the heart in pathophysiological state: abnormal adrenergic responsiveness in cardiac disease. Ann Rev Physiol 53: 137–159. H JG, M MJ, H N, B JJ, V Z PA. 1991. The influence of calcium antagonists on the adenine nucleotide metabolism in the guinea pig working heart during ischaemia and reperfusion. Arch Pharmacol 343: 496–504. Kı¨ T, O K, H H, S T. 1990. Alteration of 1,2 diacylglycerol content in ischemic and reperfused heart. Mol Cell Biochem 99: 1–8. K JE, L B, S RA. 1990. Dietary n-3 polyunsaturated fatty acids and amelioration of cardiovascular disease: possible mechanisms. Am J Clin Nutr 52: 1–28. L S, G A, M J, A P. 1991. Contribution to the use of cultured cardiomyocytes as a model in nutritional studies: fatty acid composition of the hearts of rats fed linseed oil or sunflower oil and cardiomyocytes grown on their sera. Cardioscience 2: 55–61. ML DC, H AM, B SJ, L TS, R RA. 1994. Effect of dietary polyunsaturated fatty acids on contraction and relaxation of rat femoral resistance arteries. J Cardiovasc Pharmacol 23: 92– 98. M SB, H TK, B JH. 1984. Relationship between phosphoinositides and calcium responses to muscarinic agonists in astrocytoma cells. Mol Pharmacol 26: 149–155. ML PL. 1993. Relative effects of dietary saturated, monounsaturated and polyunsaturated fatty acids on cardiac arrhythmias in the rat. Am J Clin Nutr 57: 207–212. ML PL, H PR, A MY, M R, R D, M M, R T, H RJ. 1996. The cardiovascular protective role of docosahexaenoic acid. Eur J Pharmacol 300: 83–89. MM EJ. 1988. Dietary lipids and the regulation of membrane fluidity and function, in Advances in membrane fluidity, Vol 3, Physiological Regulation of membrane fluidity (Aloia, Curtain and Gordon) New York: Alan R. Liss Inc. 189–237. M JTA, B A, D DHW, L JMJ, G C. 1990. Alterations in polyunsaturated fatty acid composition of cardiac membrane phospholipids
386
P. Delerive et al.
and a1 adrenoceptor mediated phosphatidylinositol turnover. Cardiovasc Res 24: 94–101. M WR, S LM. 1964. Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron-fluoride-methanol. J Lipid Res 5: 600–608. M KH, N SL, M JC. 1993. Alterations in a-1 adrenergic receptor mediated phosphatidylinositol turnover in hypoxic cardiac myocytes. J Mol Cell Cardiol 25: 1187–1202. N G, L J, T E, P H, L` P, P AM, L H. 1989. Effects of fish oil, corn oil and lard diets on lipid peroxidation status and glutathione peroxidase activities in rat heart. Lipids 24: 179–186. O F, G A, S JP. 1995. Synthesis of eicosanoids in cardiomyocytes during hypoxia and reoxygenation: influence of the polyunsaturated fatty acid composition. Am J Physiol 268: H308–H315. P S, ML PL. 1992. Modulation of myocardial oxygen requirements by dietary lipids in the isolated
erythrocyte perfused working rat heart. J Mol Cell Cardiol 24: S115. P B, D I, F A, O F, A P, G A. 1998. Long chain polyunsaturated fatty acids influence both b- and a- adrenergic function of rat cardiomyocytes. J Am Oil Chem Soc 75: 247– 254. R DK, H MA, H CE. 1988. Effects of dietary fish oil on cardiac responsiveness to adrenoceptor stimulation. Am J Physiol 254: 494– 499. V E R, B HD, P M, V Z PA. 1991. Discrepancies between inotropic responses and b adrenoceptor characteristics after global ischemia in isolated hearts J Cardiovasc Pharmacol 18: 679–686. Y J, O M, M M, I T. 1994. b-adrenoceptor-G protein-adenylate cyclase complex in rat heart failure produced by coronary artery ligation. J Mol Cell Cardiol 26: 617–626.