- Top of page
- Experimental procedures
The appetite-suppressing effect of phenylpropanolamine (PPA) has been attributed to its inhibitory action on neuropeptide Y (NPY), an appetite stimulant. However, molecular mechanisms underlying this effect are not clear. This study aimed to investigate if cAMP response element binding protein (CREB) signaling was involved. Moreover, possible role of superoxide dismutase-2 (SOD-2) during PPA treatment was also examined. Rats were daily treated with PPA for 4 days. Changes in hypothalamic NPY, protein kinase A, CREB, and SOD-2 mRNA contents were measured and compared. Results showed that protein kinase A, CREB, and SOD-2 mRNA levels increased during PPA treatment, which is concomitant with decreases in NPY and feeding. Moreover, CREB DNA binding activity detected by electromobility shift assay increased during PPA treatment, revealing an involvement of CREB-dependent gene transcription. Furthermore, infusions of CREB antisense oligonucleotide (or missense control) into cerebroventricle were performed at 1 h before daily PPA treatment in free-moving rats, and results showed that CREB knockdown could block PPA-induced anorexia and modify NPY and SOD-2 mRNA content toward normal. It is suggested that CREB signaling may participate in the central regulation of PPA-mediated appetite suppression via the modulation of NPY gene expression and that an increase of SOD-2 may favor this modulation.
Hypothalamic neuropeptide Y (NPY) is a powerful orexigenic agent which plays critical roles in periodic eating behavior and maintenance of body weight (Wynne et al. 2005). Central administration of NPY can induce hyperphagia even under conditions of satiation, resulting in an increase of fat deposition, a decrease of energy expenditure, and a promotion of obesity (Williams et al. 2001). Phenylpropanolamine (PPA) is an over-the-counter anorectic drug that can be used in human dieters to reduce obesity (Schteingart 1992; Borovicka et al. 2002; Cooper et al. 2005). PPA is structurally and functionally related to amphetamine (AMPH)-like anorectic drugs, such as ephedrine, phentermine, diethylpropion, and methamphetamine, and is regarded as a sympathomimetic agent because of its effect on the brain (Colman 2005). Evidence reveals that the mechanism underlying the appetite-suppressing effect of PPA is implicated in the central release of catecholamine (Wellman and Davies 1992; Cheng and Kuo 2003), which can exert its effect on hypothalamic NPY neurons to suppress appetite (Hsieh et al. 2004).
It is unclear whether the intracellular signaling of cAMP response element binding protein (CREB) is required for NPY gene expression during PPA treatment. CREB is a downstream nuclear transcription factor of protein kinase A (PKA), which can be elicited by various physiological ligands and is critically involved in the regulation of energy metabolism (Habener et al. 1995; Griffioen and Thevelein 2002). Given the role in neuronal plasticity, CREB has emerged as a molecule that is important in modulating behavioral responses (Nair and Vaidya 2006). In the brain, CREB is required for behavioral sensitization (McDaid et al. 2006) and self-administration behavior (Choi et al. 2006). Moreover, several studies demonstrate that cAMP is involved in the regulation of NPY-induced feeding behavior (Akabayashi et al. 1994; Konradi et al. 1994; Sheriff et al. 2003) and that CREB is required for dopamine-dependent gene expression (Das et al. 1997; Andersson et al. 2001) and can regulate the expression of NPY gene that is implicated in anxiety and alcohol drinking behaviors (Pandey 2003). Thus, we hypothesized that CRE-mediated gene induction might involve the regulation of NPY gene expression during PPA treatment.
The enzyme superoxide dismutase (SOD), including CuZn-SOD (SOD-1) and Mn-SOD (SOD-2), is essential in destroying oxygen-based radicals and reported to play a role in the reduction of methamphetamine-induced neurotoxicity (Sheng et al. 1996). Moreover, SOD is activated during AMPH treatment, which is associated with the normalization of AMPH-induced appetite suppression (Hsieh et al. 2006). Thus, we hypothesized that SOD gene might be activated for the decrease of the oxidative stress and thus help to normalize the feeding behavior during PPA treatment. The enzyme SOD-2, a mitochondrial SOD, was detected as it was reported to implicate in the neuroprotective action in the brain (Hu et al. 2007).
Daily application of CREB antisense oligodeoxynucleotides (ODN) into brain was employed in the present study to disrupt CREB-dependent gene transcription or CREB translation in free-moving rats. Antisense technique was recently applied to study the effect of drug on behavioral response as it was preferentially taken up by neurons in the rodent brain after intracerebral administration (Yee et al. 1994; Ogawa et al. 1995). Moreover, antisense had been used to interrupt specific gene expression in the brain (Chiasson et al. 1992; Ghosh and Cohen 1992) or in the hypothalamus (Hulsey et al. 1995). Therefore, we chose antisense, which had been previously used to specifically decrease CREB translation in the brain (Chance et al. 2000), to examine its effect on PPA anorexia following central ventricular administration.
- Top of page
- Experimental procedures
Recently, the mechanism for appetite-suppressing effect of PPA has been attributed to its inhibitory effect on hypothalamic NPY neurons. However, molecular mechanisms underlying this effect are yet to be elucidated. In this study, we found that CREB was activated during PPA treatment and that CREB knockdown could block the decreasing effect of PPA on NPY mRNA level. These results suggest that CREB signaling is involved in the regulation of NPY gene expression.
This finding was supported by previous reports showing that cAMP agonist administered into hypothalamus could decrease NPY-induced feeding behavior (Sheriff et al. 2003) and that CREB signaling was involved in the modulation of NPY gene expression (Sheriff et al. 1997). However, some in vitro and in vivo studies had shown that NPY gene was activated by phorbol ester or cAMP analog (Magni and Barnea 1992; Akabayashi et al. 1994). Mechanisms underlying this contradictory effect of CREB on NPY gene expression are unknown. To clarify this contradiction, infusion of CREB antisense into brain was administered to PPA-treated rat, and results showed that CREB knockdown in the brain could interrupt the anorectic response of PPA with a restoration of NPY gene expression. This result supported our hypothesis that activation of CREB signaling was involved in the inhibition of NPY gene expression in PPA-treated rats.
Although the injected site and the method for detecting CREB were different, the present findings were consistent with the result shown in a previous report (Chance et al. 2000). Using a similar sequence of CREB antisense phosphorothioate-ODN injected directly into the hypothalamic perifornical area, it revealed that both NPY-stimulated feeding and ad libitum feeding were reduced and that both reductions were accompanied with a decrease in CREB protein. These results indicated that CREB signaling was involved in NPY-mediated feeding behavior and that perifornical area might be one of the sites where antisense exerted its effect.
An increase in CREB mRNA level on day 2 was accompanied with a significant decrease in NPY mRNA level on the same day, implying a consistent role for CREB signaling in the regulation of NPY gene. Indeed, a site resembling CRE had been shown to exist on the 5′-flanking region of the rat NPY gene (Larhammar et al. 1987). Thus, CREB phosphorylation and CRE-mediated gene expression were indispensable for NPY gene expression in fasted rats (Shimizu-Albergine et al. 2001). Consistent with a previous report indicating that a 48-h fasting in rats could increase CRE binding activity and NPY gene expression in hypothalamic nuclear extracts (Sheriff et al. 1997), our results suggested that CREB gene was markedly activated to decrease NPY gene expression after 1–2 days of daily PPA treatment.
Instead of being inhibited, CREB gene was activated following PPA treatment. This result suggests that CREB signaling may participate in the inhibition of NPY gene expression during PPA treatment. As the induction of CREB signaling normally served to activate gene transcription, including NPY gene (Sheriff et al. 1997), it was possible that PPA might activate CREB signaling in a distinct population of hypothalamic neurons, such as anorexigenic CART-producing neurons, and in turn inhibit NPY neurons. CART is a potent appetite-suppressing peptide closely associated with the action of NPY (Lambert et al. 1998). Evidence revealed that CART expression appeared to be regulated via CREB-mediated signaling in rat brain (Jones and Kuhar 2006) and that a CRE site in the area of CART proximal promoter was involved in cAMP/PKA/CREB signaling in neuron-like cells (Dominguez and Kuhar 2004). The present findings revealed that CART gene was activated following PPA treatment and that the alteration of CART mRNA levels during a 4-day period of PPA treatment is consistent with that of CREB mRNA levels. This result supports our view that PPA anorexia may be through CART. Moreover, our results shown that, except AMPH and cocaine, the drug PPA which is an AMPH-like anorectic drug can also activate the expression of CART gene.
Comparing the changes between CREB and PKA mRNA levels in PPA-treated rats, it appears that PKA and CREB signaling may play a consistent role in the regulation of NPY gene expression. Indeed, CREB is one of the downstream nuclear transcription factors of PKA; therefore, it is rational that PKA and CREB signaling are co-activated during PPA treatment. However, the increase in CREB mRNA content on day 1 was about eightfold but that of PKA mRNA content on the same day was only about 2.5-fold. This difference might be because of the possibility that other signaling, such as protein kinase C and c-jun, might play a convergent role to regulate CREB gene expression. It has been suggested that several intracellular signaling, including PKA, protein kinase C, and c-jun signaling, may be activated together in rats treated repeatedly with moderate dose of AMPH. These co-activated signaling could effect together on CREB for the purpose to regulate SOD gene expression, which may reduce the oxidative stress in the brain and thus favored the restoration of NPY gene expression (Hsieh et al. 2005, 2006). PPA was reported to involve the oxidative stress-related brain disturbances following a chronic treatment of drug (Levin 2005). Moreover, an acute treatment of PPA was employed to improve the stress-related urinary incontinence because of the action of drug on CNS (Scott et al. 2002). The present findings revealed that SOD-2 mRNA levels were increased during moderate dose (75 mg/kg) of PPA treatment and their alterations were coincided with those of PKA and CREB mRNA levels. Moreover, CREB knockdown in PPA-treated rats could reduce SOD-2 mRNA level. This result suggests that PKA/CREB signaling may participate in the activation of SOD-2 gene, which favors the correction of NPY gene expression during moderate dose of PPA treatment.
The activation of PKA/CREB signaling may be a critical pathway for AMPH-like anorectic drugs to induce appetite-suppressing effect. Except the current finding that PKA/CREB is activated in PPA treatment, several evidence reveal that (i) the intracellular cAMP is increased following PPA treatment in minces of rat heart (Hull et al. 1993), (ii) PKA/CREB signaling is activated following AMPH treatment (Hsieh et al. 2007), (iii) intracellular cAMP is accumulated following ephedrine treatment in rat leukemia cell (Saito et al. 2004) or human adipose tissue (Diepvens et al. 2007), and (iv) daily oral administration of phentermine to rats can change the activity of adenylate cyclase-cAMP system in renal and hepatic tissues (Kacew et al. 1977). PPA, ephedrine, and phentermine are AMPH-like anorectic drugs and are classified as sympathomimetic agents that may exert their effect via the activation of monoaminergic system in the brain (Alexander et al. 2005; Nelson and Gehlert 2006). Except sympathomimetic agents, it has been suggested that psychostimulant and opiate drugs, such as AMPH and cocaine, may target at similar CREB gene to induce behavioral responses (Brenhouse et al. 2007; Hsieh et al. 2007). It was then rational to speculate that induction of PKA signaling by AMPH-like anorectic drugs should be viewed as a group of concerted events that occurred against a complex background of intra- and intercellular signal pathway.
The physiological state during the first 2 days in PPA-treated rats was similar to that of fasting, which is in a state of negative energy balance, resulting in the induction of NPY gene expression on subsequent days. Restoration of NPY mRNA level on day 4 during PPA treatment was accompanied with a gradual decrease in CREB mRNA level, implying a disinhibitory effect of CREB on NPY gene expression. Possibly, this disinhibitory effect of CREB might be relevant to a gradual decrease in catecholamine released from pre-synaptic nerve terminals during a repeated treatment of PPA (Cheng and Kuo 2003).
Although evidence revealed that substantial food deprivation might lead to increased NPY gene expression (Richard 1995), we ruled out the possibility that the change in 24-h NPY level following PPA treatment was simply secondary to reduced feeding, rather than the rapid action of PPA on hypothalamic NPY. It is because PPA-induced anorexia following PPA treatment was occurred only at 0–6 h time interval. We also ruled out the possibility that the change of NPY level was due to a disturbance of neuroendocrine in tolerant rats as they showed no change in feeding behavior compared with that in normal animals.
In summary, the present data provide a molecular message to understand the role of central CREB in the regulation of hypothalamic NPY gene expression in vivo in PPA-treated rats.