Chronic Imipramine but Not Bupropion Increases Arachidonic Acid
Total Page:16
File Type:pdf, Size:1020Kb
Molecular Psychiatry (2010) 15, 602–614 & 2010 Nature Publishing Group All rights reserved 1359-4184/10 www.nature.com/mp ORIGINAL ARTICLE Chronic imipramine but not bupropion increases arachidonic acid signaling in rat brain: is this related to ‘switching’ in bipolar disorder? H-J Lee1, JS Rao, L Chang, SI Rapoport and H-W Kim1 Brain Physiology and Metabolism Section, National Institute on Aging, National Institutes of Health, Bethesda, MD, USA Agents effective against mania in bipolar disorder are reported to decrease turnover of arachidonic acid (AA) in phospholipids and expression of calcium-dependent AA-selective cytosolic phospholipase A2 (cPLA2) in rat brain. In contrast, fluoxetine, an antidepressant that is reported to switch bipolar depressed patients to mania, increases cPLA2 expression and AA turnover in rat brain. We therefore hypothesized that antidepressants that increase switching to mania generally increase cPLA2 and AA turnover in brain. To test this hypothesis, adult male CDF-344 rats were administered imipramine and bupropion, with reported high and low switching rates, respectively, at daily doses of 10 and 30 mg kgÀ1 i.p., respectively, or i.p. saline (control) for 21 days. Frontal cortex expression of different PLA2 enzymes and AA turnover rates in brain when the rats were unanesthetized were measured. Compared with chronic saline, chronic imipramine but not bupropion significantly increased cortex cPLA2 mRNA activity, protein and phosphorylation, expression of the cPLA2 transcription factor, activator protein-2a (AP-2a) and AA turnover in phospholipids. Protein levels of secretory phospholi- pase A2, calcium-independent phospholipase A2, cyclooxygenase (COX)-1 and COX-2 were unchanged, and prostaglandin E2 was unaffected. These results, taken with prior data on chronic fluoxetine in rats, suggest that antidepressants that increase the switching tendency of bipolar depressed patients to mania do so by increasing AA recycling and metabolism in brain. Mania in bipolar disorder thus may involve upregulated brain AA metabolism. Molecular Psychiatry (2010) 15, 602–614; doi:10.1038/mp.2008.117; published online 4 November 2008 Keywords: arachidonic acid; bipolar disorder; bupropion; cPLA2; imipramine; switching Introduction vant to BD, reduce arachidonic acid (AA; 20:4nÀ6) turnover in phospholipids and expression of AA Persons with bipolar disorder (BD) experience epi- cascade enzymes in rat brain.4 AA is a polyunsatu- sodes of depression more frequently and for longer rated fatty acid found predominately in the stereo- periods than episodes of mania.1 Bipolar depression specifically numbered (sn)-2 position of membrane is a major cause of psychiatric morbidity and phospholipids, and can be released from this position mortality and poses a major public health concern. by certain phospholipase A (PLA ) enzymes.5 These A challenge in treating bipolar depression is the 2 2 enzymes can be activated through their coupling to tendency of many antidepressants, particularly tri- certain neuroreceptors by a G-protein or by calcium,6–10 cyclic antidepressants and serotonin selective reup- or in conditions of neuroinflammation through their take inhibitors, to induce episodes of mania or to coupling to cytokine receptors.11–13 The unesterified increase cycle frequency or symptom intensity.2,3 The AA that is released by a PLA can be converted to causes for the tendency of antidepressants to ‘switch’ 2 eicosanoids or other active metabolites, subjected to depressed bipolar patients to mania are not under- b-oxidation, or reesterified into the sn-2 position of stood. membrane phospholipids via a long-chain acyl-CoA Antimanic mood stabilizers, lithium, valproic acid synthetase and an acyltransferase.14–19 Released AA and carbamazepine, when given chronically to rats to and its metabolites can influence a number of brain produce plasma concentrations therapeutically rele- processes, including signal transduction, transcrip- tion, neuronal activity, apoptosis, inflammation and 20–26 Correspondence: Dr H-W Kim, Brain Physiology and Metabolism blood flow. Section, National Institute on Aging, National Institutes of Health, The agents lithium, carbamazepine (5H-dibenz[b, 9000 Rockville Pike, Building 9, 1S -126, Bethesda, MD 20892, f]azepine-5-carboxamide; 5-carbamoyl-5H-dibenz[b, USA. f]azepine) and valproic acid (2-propylpentanoic acid), E-mail: [email protected] which are effective in the manic phase of BD, are 1These authors contributed equally to this work. Received 29 July 2008; revised 22 September 2008; accepted 7 reported to decrease brain AA turnover rate in October 2008; published online 4 November 2008 membrane phospholipids in unanesthetized rats Imipramine increases brain AA turnover and cPLA2 H-J Lee et al 603 without changing the turnover rate of docosahexa- bupropion given to rats had an antidepressant effect enoic acid (DHA; 22:6nÀ3),27–31 another polyunsat- by reducing immobility time in the forced swim urated fatty acid usually found in the sn-2 position. test,53–55 and improved behavior in another animal The effect of chronically administered lithium and model of depression.56 Because imipramine but not carbamazepine on AA turnover was ascribed to a bupropion is reported to increase switching in bipolar selective decrease in the mRNA, protein and activity depressed patients,43,47,57–59 we hypothesized that 2 þ levels of AA-selective Ca -dependent cytosolic PLA2 chronic imipramine but not bupropion would in- 32,33 (cPLA2), whereas chronic valproate inhibited the crease markers of AA metabolism, including AA activity of an AA-selective acyl-CoA synthetase that is turnover in phospholipids and cPLA2 expression, in involved in recycling AA back into membrane rat brain. phospholipids.27,34 In contrast to the three antimanic agents, another Materials and methods mood stabilizer, lamotrigine (3,5-diamino-6-(2,3- dichlorophenyl)-as-triazine), which is preferred for Animals treating bipolar depression and rapid cycling, neither Chemicals and reagents, including bupropion and decreased AA turnover in brain phospholipids of imipramine, were purchased from Sigma Chemicals 35 unanesthetized rats nor changed cPLA2 expression, (St Louis, MO, USA) unless otherwise indicated. The whereas it decreased transcription and protein of study was conducted following the National Insti- cyclooxygenase (COX)-2.36 Topiramate (2,3:4,5-bis-O- tutes of Health Guidelines for the Care and Use of isopropylidene-b-d-fructopyranose sulfamate), which Laboratory Animals (Publication no. 80-23) and was had been thought effective against BD37 and shown to approved by the Animal Care and Use Committee of reduce hyperactivity in a quinpirole rat model,38 did the Eunice Kennedy Shriver National Institute of not significantly change expression of any of the Child Health and Human Development Animal Care measured enzymes in the AA cascade, including and Use Committee. Male CDF-344 rats, weighing cPLA2, or alter AA or DHA turnover in rat brain 180–190 g (Charles River Laboratories; Wilmington, phospholipids.39,40 Supporting this lack of effect, four MA, USA), were acclimatized for 1 week in an animal double-blind placebo-controlled trials later demon- facility with controlled temperature, humidity and strated that topiramate was ineffective in BD,41 and it light cycle, ad libitum access to food (NIH-31) was withdrawn as a treatment. and water. Rats were assigned randomly to an A number of antidepressants when administered to acute or a chronic treatment (imipramine and bupro- bipolar depressed patients cause them to ‘switch’ to pion) group or to a control group. Chronic drug- mania more frequently than expected.2,3,42,43 Because treated rats received 10 mg kgÀ1 dayÀ1 imipramine or antimanic mood stabilizers when given chronically to 30 mg kgÀ1 dayÀ1 bupropion dissolved in saline (0.9%) rats downregulate brain AA turnover and AA en- once daily for 21 days intraperitoneally (i.p.), whereas zymes, we thought it possible that the increased controls received the same volume of saline once switching tendency caused by some of these anti- daily i.p., also for 21 days. For acute studies (to depressants might be due to their upregulating the separate chronic from acute drug action), rats were brain AA cascade. This hypothesis is consistent with randomized into three treatment groups that received evidence that lamotrigine, which does not upregulate a single i.p. injection of imipramine (10 mg kgÀ1)or AA turnover but reduces COX-2 expression in rat bupropion (30 mg kgÀ1) or an equivalent volume of brain,35,36 does not increase switching, whereas saline 3 h before a study. fluoxetine, which upregulates AA turnover and The chronic dosing regimens for the two drugs are 44,45 cPLA2 mRNA, protein and activity in rat brain, reported to decrease depression scores in rodents, as is reported to do so.42,46,47 measured by a variety of behavioral and biochemical To further test the hypothesis, in the present study, tests.48–51,60 On the last day of a dosing regimen, a rat we examined brain AA metabolism and expression was injected with its appropriate treatment 3 h before of brain PLA2 and COX enzymes in rats treated its brain was removed for enzyme, protein, prosta- chronically with two additional antidepressants, glandin E2 (PGE2) or kinetic analysis. For molecular imipramine (5-[3-(dimethylamino) propyl] -10,11- analysis, 3 h after the last injection, the rat was dihydro-5H-dibenz [b,f]-azepine monohydrochloride) anesthetized with CO2 and decapitated. The frontal