Influence of arachidonic acid metabolites on cardiac

Influence of arachidonic acid metabolites on cardiac al-adrenoceptor responses in diabetic rats

Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of British Columbia on 07/21/15 For personal use only.

HONGXIANGAND JOHN H . MCNEHLL~ Division of Phamcology and Toxicology, F~cultyof Phamaceutical Sciences, Biniversitg' of British Columbia, Vancouver, B. C . , Canada V6T 123 Received March 11, I991 XIANG,H., and MCNEILL,J. H. 1992. Influence of arachidsnic acid metabolites on cardiac a,-adrenoceptor responses in diabetic rats. Can. J. Physiol. Pharmacol. 70: 680 -486. There is evidence &at sne or more metabolites sf arachidsnic acid can produce positive inotropic effects and may also be implicated in the enhanced a,-adrenoceptor responses in hearts from diabetic rats. We therefore carried out a study t s investigate the possibility that arachidonic acid metabolites could be involved in the altered cardiac a,-effect of norepinephrine (in the presence of propranolol) in chronic streptozotmin-diabetic rats. Our results have shown that in the presence of the cyclooxygenase inhibitor iadomethacin or the thromboxane synthehse inhibitor imidazole, the norepinephrine-stimulated positive inotropic effect and the formation of inositol 1,4,5-trisphosphate were significantly increased in control hearts but were unaltered in hearts from diabetic rats. The addition of the prostacyclin synthetase inhibitor tranylcypromine reduced the norepinephrine-stimula8edpositive inotropic effect and inositol 1,4,5-trisphosphate formation only in diabetic hearts and had no effect in the controls. The nature and physiological significance of the ehanced positive inotropic effect and inositol 1,4,5-trisphosphate formation in the control heart with the addition of indormethacin and imidazole are still unclear. The effect of tranylcypromine may indicate the participation of prostacyclin in mediating the enhanced a,-inotropic effect of aorepinephrine in the chronic diabetic heart, Key words: norepinephrine, positive inotropic effect, inositol 1,4,5-trisphosphate. -

-

-

XIANG,H., et MCNEILL,J. M. 1992. Influence of arachidonic acid metabolites on cardiac a,-adrenoceptor responses in diabetic rats. Can. J. Physiol. Pharmacol. 70 : 680-686. Des faits indiquent qu'un ou plaasieurs m6tabolites de I'acide arachidonique pourraient provoquer des effets instropes positifs et Ctre impliqu6s dans les r6ponses accrues des r6cepteurs adr6nergiques a, darns les coeurs de rats diaEtiques. Tenant corrapte de ceci, nous avons examine la possibilit6 que les rn6tabolites de I'acide arachidonique pourraient Ctre impliqu6s dans I'dtQation de H'effet a, cardiaque de la norbpinephrine (en pr6sence de propranolol) chez des rats au dia&te chronique induit par la streptozotocine. Nos rksultats ornt rmontrC que, en pr6sence de l'inlaibiteur de cyclmxygkrnase, indomCthacine, ou de l'inhibiteur de thromboxane synthktase, imidazole, l'effet inotrope positif stirnu16 par la norepinkphrine et la formation d'inositol 1,4,5-trisphosphate ont augment6 significativementdans les coeurs tkmoins, mais qu'ils n'ont pas $tC modifi6s sans les coeurs des rats diabktiques. &'addition de l'ihibiteur de prostacycline synthCtase, tranylcypromine, a rCduit I'effet inotrope positif stirmu16 par la norkpinkphrine ainsi que la formation d'inositol 1,4,5-trisphosphate dans les coeurs diabktiques seulement; elle n'a pas eu d'effet sur les coeurs tCmoirns. La nature et l'importance physiologique de l'effet inotrope pssitif et de la formation d'inositol 1,4,5-trisphosphate dans le coeur tkmoin en pr6sence dqindom6thacineet d'irnidazole ne sont toujours pas expliqukes. E'effet de la tranylcypromine pourrait indiquer la participation de la prostacycline c o m e rn6diateur de I'augmentation de l'effet inotrope a, de la norkpinkphrine dans le coeur diabktique chronique. Moa cl6s : norkpinCphriwe, effet inotrope positif, inositoj 1,4,5-trisphosphate. [Traduit par la rkdaction]

Chmges in sensitivity $rand responsiveness to at-adrenoceptor stimulation have been observed in hearts isolated from diabetic animals. Most reports have suggested an enhanced inotropic response to al-adrenoceptor stimulation in diabetic atria (Cauga and Sterin-Borda 1986; Jackson et d. 1986; Goyd et d. 1987), ventricle (Wdd et al. 1988), and isolated heart (Downing et al. 1983) in both acute and chronic diabetics. Ira a study performed by Cauga and Sterin-Borda (19861, it was suggested that the observed supersensitivity to methoxamine seen in atria isolated from short-term diabetic rats could be mediated by a mechanism that couples activation of cardiac al-adrenoceptors to changes in the metabolism of arachidonic acid towards an increased production of thromboxane synthetase and llipoxygenase derived products. Furthermore, they have shown that during short-term diabetes, arachdonic acid metabolism in the heart is shifted towards thromboxanes and lipoxygenase metabolites, while in normal atria it is directed towards prostaglandin production (Cauga et d . 1985). 'Author for correspondence. Printed in Canada / Imprim6 au Canada

In view of these findings, the present study was undeflaken to explore the involvement of different arachidonic acid metabolites in the enhanced al-adrenoceptor mediated inotrspic responses in the hearts of chronic diabetic model. Since the al-adrenoceptor mediated positive inotropic effect has been suggested to be linked to activation of phosphoinositide turnover, inositol B ,4,5-trisphosphate (Ins( 1,4,5)P3)formation in these hearts was also studied.

Materisls and methods Biochemical reagents were obtained from Sigma Chemical Co. (St. Louis, Ms.) and laboratory grade chemicals were obtained from BDH (Vancouver, B .C .). Insulin radioimunoassay and [3H]lns(1,4,5)P, kits were obtained from Amersham Canada Ltd. (Oakville, Ont.). Glucose kits were purchased from Boehringer-Mannheim Canada (Dcsrval, Que.). Male Wistar rats (175-200 g) were used. All animals were anesthetized transiently with diethyl ether to allow injection of either streptozotocin (STZ) or its vehicle into the tail vein. Diabetes was induced by a single intravenous injection of STZ (55 mglkg) dissolved in citrate buffer (pH 4.5). The citrate buffer was made by mixing 0.1 M citric acid and 0.1 M sodium citrate until a pH of 4.5 was

68 l

Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of British Columbia on 07/21/15 For personal use only.

XIANG AND MCNEILL

obtained. Control rats were injected with citrate buffer alone. All rats injected with ST% survived and were housed two to three per cage, All rats had access to Purim rat chow and water ad libitum. The lights were left on from 06:W to 18:W and room temperature was maintained at 22°C. Diabetes was detected 3 days following STZ injection by estimating the extent of glucosuria with the aid of enzymatic test strips (TesTape, Eli Eilly, Toronto, Ont .) , and rats displaying glucosuria greater than 2 % were used. No detectable glucose was present in urine of control animals. The animals were sacrificed 6 weeks after diabetes induction. The 6-week duration of diabetes was chosen based on a report from our laboratory which showed that alterations in cardiac performance occur 6 weeks after the onset s f diabetes (Taaniliani et al. 1983). At the time of death, whole blood (arterial and venous) samples were collected in heparidzed tubes from wonfasting animals and centrifuged at 1300 x g for 28 min to separate a l s from plasma. The plasma obtained was stored at -20°C until assayed for glucose and insulin. Hearts were excised imediately and u s d for functional studies in the isolated tissue bath. Isolation of cardiac tissues and measurement of the concentration response curves were done according to the method of Ramnadham and Tenner (1983). At the time of killing, the heart was excised and imediately placed in a beaker containing ice-cold oxygenated Chenoweth-Koelle (CK) solution (pH 7.4) consisting of the following (mM): NaCl, 120; KCl, 5.6; CaCl,, 2.18; MgCl,, 2.1; NaHCO,, 19; and glucose, 10. After expressing blood from the heart, atrial tissue, fat, and connective tissue were trimmed off. The right ventricle was then carefully dissected and transferred to another beaker containing cold fresh CK solution. Each right ventricle was f b f i e r dissected into three to four triangular-shaped pieces. One end of each right ventricular piece was attached to a tissue-electrode assembly with two platinum electrodes. The other end was mounted to a Grass FT.03 force-displacement transducer that was connected to a Grass model 79D polygraph eight-channel recorder (Quincy, Mass.). The tissue preparations were equilibrated at 37°C for $0 min in 20-mL tissue baths of CK solution under constant oxygenation (95 % O2 and 5 9% CCO,). During this period, CK solution in the tissue bath was replaced every 15 min. Westing tensions in right ventricular pieces of 1 g were maintained throughout the experiment. After 38 min of equilibration, the right ventricular pieces were stimulated electrically via the punctate platinum electrodes (at a frequency of 1.0 Hz, two times threshold voltage, and a duration of 5 ms). At the end of the equilibration period, basal developed force in right ventricular tissues was recorded. Indomethacin was dissolved in 70% ethanol. The drug was diluted in the tissue bath to achieve a final concentration of 18 pM, a concentration that was determined in a preliminary experiment not to affect the basal tension. Ethanol done had no effect on cardiac contractility or hs(1,4,S)P3 level in respnse to norepinephrine in control and diabetic hearts (results not shown), After a 60-min equilibration period, isolated right ventricular pieces were incubated for 15 min with the b-adrenoceptor antagonist propranolol and the cyclooxygenase inhibitor indomethacin before increasing concentrations of norepinephrine were added. Following the determination of the concentration -response curves, tissues were freeze clamped in liquid nitrogen and stored at -88°C prior to the measurement of Ins(1 ,4,5)P3. Effects of iranidazole m d tranylcypromine were determined simanilarly except that imidazole was dissolved in distilled H 2 0 at a final concentration of l d (Cauga and Sterin-Borda 1986) and tranylcypromine was dissolved in distilled water and diluted to a find eoncentration of 1.25 x 10-%h.I.These concentrations s f drugs by themselves had no inotropic effect. All experiments were performed in the presence of propranolol to block any interference from /3-adrenoceptor activation. Pt=opranoloI at a concentration of 10 pM did not show any influence on basaldeveloped force. A preliminary study showed that propranolol given at higher concentrations would exert a negative effect on the basaldeveloped force.

s -0

Control Control with Incfometkocin

A-

Diabetic with IndomeBhocin

-e

-7

-6

-5

-4

(n = 6)

-3

Norepinephrine (tog

-2

-1

M)

Fro. 1. Concentration -response curves for the effect of norepinephrine (in the presence of propranolol) on change in tension (percent of maximum) in right ventricles of control and diabetic rats. Influence of the cyclooxygenase inhibitor indomethacin (I0 pM). Norepinephrine was introduced into the tissue bath 15 min after the addition of propranolol (10 pM)and indomethacin. At each concentration, the inotropic response to norepinephrine was recorded after it was stabilized, which took approximately 3 min. Points represent the mean f SEM for the number of ventricles shown in parentheses. *, significantly different from the other three groups, p < 8.05.

Sensitivity of the right ventricular tissues towards the positive inotropic effect of norepinephrine was detemined by calculating the estimated pD2 value. The pD2 value is defined as the negative log of the concentration of the drug required to produce 50% of the maximum change in tension. For the determination of Ins(1,4,5)P3 levels, the frozen right ventricular strips were weighed and imediately put into 0.8 mL of 6 % triehloroacetic acid on ice. The tissues were then homogenized with a polytron PT-10 homogenizer for 10 s at setting 5, The homogenates were centrifuged at 6000 X g for 15 min at 4OC. The supernatants were transferred into glass tubes, to which 2 mE of H28-saturated diethyl ether was added. The tubes were shaken well and the top diethyl ether layer was removed. The diethyl ether extraction procedure was repeated five times. The tubes were then put in a rack that was placed in warm tap water for a few minutes to evaporate the dieahyl ether, A 108pE portion of h e ether-extracted supernatant was used for subsequent analysis using the Amersham ['H]Ins(I ,4,5)P3 kit. Results are expressed as means f SEM. The statistical differences between mean values of the various groups were evaluated by twoway analysis of variance followed by a Newman-Keul's test. For isolated tissue function curves, repeated measures of analysis of variance was used. All other analyses were carried out using one-way analysis of variance where appropriate. The level of statistid significance was set at a probability of less than 0.05 (' 6 0.05).

Results Rats injected with STZ exhibited symptoms characteristic of the diabetic state. Diabetic rats did not gain as much weight as age-matched controls (457 f 8 g) and had significantly lower body weights 63-47 f 8 g) at the time of sacrifice. This occurred despite the fact that diabetic animals had dramatically elevated fluid (polydipsia) and f w d (hypeqhagia) intake. Diabetic rats d s s had increased fecal mQ urine output, which were not measured quantitatively. Six-week plasma glucose levels of diabetic animals were found to be elevated by threefold (25.27 f 0.61 m o l / L )

Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of British Columbia on 07/21/15 For personal use only.

682

CAN. J. PHYSIBL. PHARMACOL. VOL. 70, 1992

0Control

Diabetic

a Control with indomethc~cin

LX3 Diabetic w i t h indomethclein

FIG. 2.Influence of indomethacin on the estimated pD2 values for norepinephrine (in the presence of propranolol) in control and diabetic right ventricles. Results are the mean k SEM for the number of ventricles shown in parentheses. *, significantly different from pD, values for norepinephrine in control rat ventricles, p < 0.05.

0ConhroB

H Diabetic

Control with indomethacin

Diabetic with indomethocin

FIG. 4. Norepinephrine (lo-") stimulated changes in Ins(1,4,5)P3 levels in right ventricles of control and diabetic rats. Influence of indomethacin. Norepinephrine was introduced into the tissue bath 15 min after addition of propranolol (10 yM) and indomaethacin. At the end of the determination of csncentrationresponse curves, tissues were fieeze clamped and Ins(1 ,4,5)P3 levels were determined using a protein binding assay kit. The basal Ins(1,4,5)P3 levels @ns(1,4,5)P3levels in the absence of norepinephrine) were 0.14 f 0.02 in the control group, 0.12 f 0.01 in the control with imdomehacin group, 0. H 3 k 0.02 in the diabetic group, and 0.12 f 0.0%in the diabetic with indomehacin group. Points represent the mean f SEM for the number of ventricles shown in parentheses. *, significantly different from controls without the inhibitor, p < 0.05.

compared with controls (7.25 & 0.61 mmol/l). Plasma insulin levels measured at the time of sacrifice were significantly depressed in diabetic animals (26 5 pU/mE) compared with controls (48 6 pUlrnE). Figure 1 depicts concentration -response curves for the change in tension (percent of maximum) evoked by norepinephrine in right ventricles of control and diabetic rats. Norepinephrine produced a positive inotropic effect in both control and diabetic groups. The increase in tension was concentration dependent and was significantly greater in the diabetic group compared with controls.

+

+

D Control ESlI Control with imidazole

% Diabetic

Diabetic with imidanale

FIG. 4. Change in tension in response to 10-% of norepinephrine (in the presence of propmnolol) in control and diabetic right ventricles. Influence of imiduole (10 pM). Norepinephrine was introduced into the tissue bath 15 min after addition of propranolol (10 pM) and imidazole. The inotropic response to norepinephrine was recorded after it had stabilized, which took approximately 3 min. Points represent the mean f §EM for the number of ventricles shown in parentheses. *, significantly different from controls without imidazole, p < 0.05.

In the presence of the cyclooxygenase inhibitor indomethacin, the positive inotropic effect induced by norepinephrine was significantly increased in control right ventricles. In fact, the change in tension (percent s f m x i m m ) in control right ventricles pretreated with indornethacin was similar t s that in diabetic right ventricles. On the other hand, indornethacin failed to modify the contractile stimulatory action of norepinephrine in the diabetic right ventricles. The increase in sensitivity to norepinephrine in diabetic right ventricles was reflected by higher estimated pD2 (-log ECS0) values obtained in the diabetic group (3.25 0.2%) relative to the age-matched control group (2.43 f 0.25) (Fig. 2). Estimated pD2 values in the control group treated with indomethacin were significantly increased (3.5 1 & 0.25) compared with controls, whereas estimated pB2 values remained elevated in the diabetic group pretreated with indornethacin (3.36 +_ 8.23). At the end of the concentration -response curves, right ventricles were freeze clamped and Ins(1 ,4,5)P3 levels were measured in these tissues. The changes in Ins(l,4,5)P3 corresponded with the above changes in the tension development. Ins(1,4,5)P3 levels were elevated in the diabetic group (2.98 8.59 pmol/mg tissue) as well as in the diabetic group 0.42) compared with pretreated with indomethacin (3.36 the control group (0.56 ) 0.17) (Fig. 3). Ins(l ,4,5)P3 levels in the control groups treated with indomethacin (2.79 f 8.68) were, however, similar to the Ins(B,4,5)P3 values in diabetic hearts. M norepinephrine, the diabetic group as well as At the diabetic group pretreated with imiduole had enhanced positive inotropic effect compared with the control group. Interestingly, imidazole enhanced the positive inotropic effect significantly in the control animals (Fig. 4). Ins(1,4,S)P3 levels were significantly elevated in the con0.27 pmollmg tistrol group treated with imidazole (3.67 sue) compared with the control group (8.56 0.17) (Fig. 5). Alsa, Ins(1 ,4,5)P3 levels were increased in the diabetic group (2.98 & 0.27). Hns(l,4,5)P3 levels in the diabetic group

+

+

+

+

+

Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of British Columbia on 07/21/15 For personal use only.

Control

.s[

5 Control with irnidomole

nControl

Diabetic

C X I Diabetic with imidomsle

FIG. 5. Norepinephrine (1W2 M) stimulated changes in Ins(1,4,5)P3 levels in right ventricles from control and diabetic rats. Influence of imidazole. Norepinephrine was introduced into the tissue bath It5 min after addition of propranolol (18 yM) and imidazole. At the end of the determination of concentration-response curves, tissues were freeze clamped and Ins(It94,5)P, levels were determined using a protein binding assay kit. The basal values of Bns(B,4,5)P3 (Hns(1,4,5)P3 levels in the absence sf norepinephrine) were 0.14 0.01 in the control group, 0.13 f 0.02 in the control with imidazole group, 0.13 f 0.01 in the diabetic group, and 0.12 f 0.02 in the diabetic with imidazole group. Points represent the mean f SEM for the number of ventricles shown in parentheses. *, significantly different from controls, p < 0.05; , significantly different from controls with imidazole and diabetic group, p < 0.05.

+

Control with Irenylsyprornine

Diabetic

Diabetic with tronylcypromine

FIG. 6. Change in tension in response to 10-' M of worepinephrine (in the presence of propranolol) in control and diabetic right ventricles. Influence of the prostacyclin synthetase inhibitor tranylcypromine (1.25 x M). Norepinephrine was introduced into the tissue bath 15 min after addition of propranolol (10 pM) and tranylcypromine. The iwotropic response to norepinephrine was recorded after it was stabilized, which took approximately 3 min. Points represent the mean f SEM for the number of ventricles shown in parentheses. * , significantly different from the other three groups, g