The Pharmacogenomics Journal (2006) 6, 381–387 & 2006 Nature Publishing Group All rights reserved 1470-269X/06 $30.00 www.nature.com/tpj ORIGINAL ARTICLE Hippocampal upregulation of the cyclooxygenase-2 gene following neonatal clomipramine treatment (a model of depression) P Cassano1, A Hidalgo1, Although a putative role has been attributed to inflammation in the 1 1 pathogenesis of depressive disorders, the relationship of prostaglandins, V Burgos , S Adris known mediators of inflammation, and depression has not been elucidated. 1 and P Argibay Clomipramine is an antidepressive drug with a pro-depressive paradoxical effect in adult rats when administrated neonatally. Using this effect as a 1Unit of Brain Research, Instituto de Ciencias Ba´sicas y Medicina Experimental del Hospital model of depression, we investigated the differential expression of the Italiano de Buenos Aires, Buenos Aires, Argentina cyclooxygenase (COX-2) gene in rat brains. Rats injected neonatally with clomipramine showed depressive-like symptoms in adulthood, as well as Correspondence: decreased levels of the brain-derived neurotrophic factor (BDNF) and a Dr P Cassano, Unit of Brain Research, Instituto quantitative differential expression of the COX-2 gene (Real Time PCR) and de Ciencias Ba´sicas y Medicina Experimental del Hospital Italiano de Buenos Aires, Potosı´ protein (immunohistochemistry) in the hippocampus. As evidenced, the 4240, 8th floor, (C1199ACL) Buenos Aires, relationship between a key enzyme in the prostaglandin synthesis and Argentina. biological and behavioral depression-like changes opens an interesting line of E-mail: [email protected] investigation regarding the molecular bases of depression and its potential treatment through immunomodulatory drugs. The Pharmacogenomics Journal (2006) 6, 381–387. doi:10.1038/sj.tpj.6500385; published online 28 March 2006 Keywords: COX-2; clomipramine; depressive disorders Introduction Clinical depression, known as major depression, is one of the most prevalent mental disorders nowadays. It is one of the most disabling illnesses in the world,1 affecting approximately 17% of United States population.2 Since many families have a predisposition to depression, a biological vulnerability to depression could be inherited. External events such as a serious loss, chronic illnesses, difficult relationships, financial problems, or any other stressor, often seem to initiate an episode of depression. Very often, a combination of genetic, psychological, and environmental factors are involved in the onset of a depressive disorder.3 The most acceptable theory to explain depression has been the ‘monoamine theory’, which proposes a loss or deficits in one or more monoaminergic neurotransmitters.4 On the other hand, there is great evidence that inflamma- tion has much to do with depression, and that many pro-inflammatory molecules and cytokines5 could participate in pro-depression mechanisms. Moreover, many antidepressants could attenuate the brain expression or actions of pro-inflammatory cytokines.6,7 Cytokines are believed to bind to glial cells, Received 12 August 2005; revised 27 January 2006; accepted 16 February 2006; published which produce cytokines and other mediators such as prostaglandins, particu- 8 online 28 March 2006 larly prostaglandin E2 (PGE2). Regarding prostaglandins, cyclooxygenase-2 Clomipramine-induced depression and COX-2 expression P Cassano et al 382 (COX-2), a rate limiting enzyme for prostanoids synthesis, is there were no differences between the groups (98.4715 induced during inflammation and participates in inflamma- vs 98.2710.4 s); however, during the evaluation of tion-mediated cytotoxicity9 and neuronal death.10 COX-2, climbing, there was a clear statistical difference between also modulates inflammation, glutamate-mediated cytotoxi- the groups (71.4735 vs 47.8719 s, Po0.01, ANOVA test). city,11 and synaptic plasticity.12,13 However, in spite of the 4. Sucrose consumption: this test was designed as an close relationship between inflammation and prostaglan- indirect assessment of anhedonia. Treated animals con- dins and on the other hand inflammation and depression, sumed lower levels of sucrose solution (Po0.05, ‘t’-test). the role of COX-2 and prostaglandins was clearly described 5. Body weight: the weight of clomipramine-treated and not in the clinic, neither in animal models of depression. control animals, equal at the beginning of the experi- In order to elucidate the putative changes in the ment and during the initial phase of the experiment expression of the enzyme COX-2, we chose a pharmacolo- (PN ¼ 8–15: P40.05, ‘t’-test), showed significant differ- gical model of depression consisting of the neonatal ences at the end of the experiment (PN ¼ 85: P40.001, administration of clomipramine in rats.14 In previous ‘t’-test) (Figure 1), reflecting lower food consumption. studies, this model resembled the behavioral and physiolo- gical abnormalities found in human depression.15 Qualitative and quantitative analysis of behavior In the treated animals, qualitative behavior such as social Results isolation, indifference towards novelty, and lower solid food consumption (Po0.05, ‘t’-test) were observed. In addition, Behavioral tests quantification of sexual behavior indicated a tendency All test performed showed either a statistical difference or a toward loss of sexual interest in treated animals (zero statistical tendency between treated and control animals. mating), as compared to controls (66% mating; Po0.05, ‘Z’-test). Assessment was carried out four times daily, by 1. Morris water maze test: treated rats spent more time animal facility technicians trained in experimental psychol- (Po0.05, ANOVA test) in the water until they found the ogy and ethology. Also, the behavior of animals showed platform. This showed that it was more difficult for ethological alterations. The cumulative time that treated treated animals than for controls to learn the location of rats actively engaged in social behavior was decreased the platform. (approximately 75%) as compared to control animals. In 2. Elevated plus maze: treated animals also showed a addition, the socially impaired rats had an increase in the significant difference (Po0.05, ANOVA test) at this test, relative amount of no contact interactions compared with more time spent in open arms reflecting less anxiety like controls. Also, a decreased locomotor activity of rats was behavior. It is accepted that anxious animals spend more observed. The analysis of behavior was carried out by animal time in closed arms as a measure of behavior associated facility personnel four times a day for periods of 30 min. with anxiety and fear. 3. Porsolt forced swim test: treated animals showed beha- Brain-derived neurotrophic factor quantification vioral differences during the steps of the test; there were Densitometry band analysis showed a significant (Po0.05, no differences in immobility (31.8715 vs 39.75717 s, ‘t’-test), decrease in the brain-derived neurotrophic factor P ¼ 0.09, ANOVA test), although differences in the (BDNF) bands of treated animals compared to control behavior were observed; during the step of mobility, animals (treated 0.85470.071; control 0.97970.071) (Figure Figure 1 Assessment of rat weight at the beginning of the treatment (PN8), and after the treatment (PN85). Adult treated animals showed less body weight than controls. At the beginning of the treatment there were no differences in the body weight. White bar, control animals; gray bar, treated animals. *Po0.01 (unpaired ‘t’-test). The Pharmacogenomics Journal Clomipramine-induced depression and COX-2 expression P Cassano et al 383 2). This result indicates that BDNF levels were reduced in COX-2 treatment could reverse the behavioral changes treated animals, which is thought to be involved in observed in clomipramine treated animals (data not shown). depressive disorders as previously described. There is evidence of a decreased serotonin level in adult animals treated with clomipramine during the neonatal 16,26 Cyclooxygenase expression period. In addition, there is evidence of behavioral and Quantitative COX-2 gene expression on different brain monoaminergic mechanism alterations in adult animals 19 regions was determined by Real Time RT-PCR. The hippo- neonatally treated with clomipramine. campus of treated animals showed significant differences On the other hand there are results that demonstrated compared with controls (Figure 3). Cyclooxygenase levels that the depression-like state induced by neonatal exposure were normalized using GPS1 and actin genes. Samples from to clomipramine are reverted during the adulthood by the neocortex showed a tendency without significant levels antidepressants drug which enhance the levels of sero- 27 of expression differences. tonin. To confirm that the increase in COX-2 gene expression In the present work, treated animals showed statistical was in concordance with that of protein localization, brains differences in the performance of behavioral tests. The were cut into 4-mm-thick slices and processed for immuhis- Morris water maze is a useful tool to assess the spatial 28 tochemistry. In concordance with molecular biology results, memory ability of rodents. Memory impairment and significant differences in the hippocampus of treated depression were closely associated in several animal models animals were found (Z ¼ 5.29, P 0.05). Additionally, treated and cognitive processes, such as learning and memory, o 29 animal