Resistin differentially modulates neuropeptide gene expression and AMP-activated protein kinase activity in N-1 hypothalamic neurons

BR A IN RE S EA RCH 1 2 94 ( 20 0 9 ) 5 2 –60

available at www.sciencedirect.com

www.elsevier.com/locate/brainres

Research Report

Resistin differentially modulates neuropeptide gene expression and AMP-activated protein kinase activity in N-1 hypothalamic neurons Russell E. Brown a,b,⁎, Paul M.H. Wilkinson a , Syed A. Imran a,c , Michael Wilkinson a,b,c a

Department of Obstetrics and Gynaecology, Faculty of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada B3K 6R8 Department of Physiology and Biophysics, Faculty of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada B3K 6R8 c Division of Endocrinology and Metabolism, Faculty of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada B3K 6R8 b

A R T I C LE I N FO

AB S T R A C T

Article history:

Intraventricular resistin is known to reduce food intake, modify hypothalamic gene

Accepted 21 July 2009

expression (e.g. NPY, POMC) and influence the activity of novel metabolic enzymes (e.g. 5′

Available online 29 July 2009

AMP-activated protein kinase; AMPK) in the rodent brain. Previously we demonstrated that the hypothalamus, and the N-1 hypothalamic neuronal cell line, also expressed several

Keywords:

adipokines, including resistin and adiponectin (ADPN). These data suggested that they

Adiponectin

might also impact brain function and metabolism. We used the N-1 hypothalamic neuronal

NPY

cell line to examine NPY, AgRP, POMC, and ADPN expression following acute resistin

RNA interference

treatment (45 min; 100 ng/mL and 1000 ng/mL). The total and phosphorylated levels of

Overexpression

AMPKα and acetyl-CoA carboxylase (ACC) were subsequently assessed using Western blot

Acetyl-CoA carboxylase (ACC)

analysis. Parallel investigations were also conducted following a) resistin overexpression, or

AMPK

b) after the RNAi-mediated attenuation of resistin mRNA in N-1 neurons. Resistin overexpression lowered POMC (−35%, p < 0.01), ADPN (−23%, p < 0.05) and NPY (− 36%, p < 0.05) mRNA as evaluated using realtime RT-PCR, although AgRP remained unchanged, and significant increases in pAMPKα and pACC were detected (+ 47% and + 34% respectively, p < 0.001). In contrast recombinant resistin only significantly increased the level of pAMPKα (+ 31%; p < 0.05), but failed to significantly modify gene expression, in N-1 neurons. Conversely the RNAi-mediated silencing of resistin expression increased AgRP (+ 37%, p < 0.05), POMC (+ 66%, p < 0.0001), ADPN (+ 87%, p < 0.0001), whereas NPY was reduced (−22%, p < 0.01) along with pAMPKα and pACC (−43% and −35% respectively, p < 0.001). In summary, these in vitro data suggest that endogenous resistin might be capable of fine-tuning the expression and enzymatic activity of various hypothalamic targets previously implicated in the delicate homeostatic control of food intake. As such, resistin may be part of an autocrine/paracrine loop, which may in turn contribute to some of the reported effects of resistin on energy metabolism. © 2009 Elsevier B.V. All rights reserved.

⁎ Corresponding author. Department of Obstetrics and Gynaecology, IWK Health Centre, 5850/5980 University Avenue, PO Box 9700, Halifax, NS Canada B3K 6R8. E-mail address: [email protected] (R.E. Brown). 0006-8993/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2009.07.068

53

BR A IN RE S E A RCH 1 2 94 ( 20 0 9 ) 5 2 –6 0

1.

Wilkinson et al., 2007), and immunohistochemical investigations also confirmed the localization of resistin-like immunoreactivity within POMC neurons in the ARC (Wilkinson et al., 2005). Rstn mRNA was also detected in the N-1 hypothalamic neuronal cell line, which co-expresses several other adipokines (e.g. fasting-induced adipose factor (fiaf) and adiponectin (ADPN) (Brown et al., 2007). N-1 cells also express several neuropeptides known to be involved in the regulation of central energy metabolism, such as NPY, and thus provide a useful model for these preliminary investigations designed to elucidate some of the potential roles for brain-derived resistin (Belsham et al., 2004; Brown et al., 2007, 2008). Previously RNAi-mediated silencing of rstn in N-1 neurons was found to significantly increase the expression of socs-3, an intracellular inhibitor of leptin and insulin signaling, and of another adipokine, fiaf (Brown et al., 2007). This led us to hypothesise that centrallyderived adipokines may act in an autocrine/paracrine manner as part of a fine-tuning system that contributes to the delicate homeostatic control required for the normal hypothalamic regulation of body weight and energy metabolism. Following the recent demonstrations that resistin can influence hypothalamic gene expression and enzyme activity in vivo (e.g. AMPK) (Muse et al., 2007; Park et al., 2008; Singhal et al., 2007; Tovar et al., 2005; Vazquez et al., 2008), we investigated whether similar changes could be replicated in N1 hypothalamic neurons. However we also studied the potential role(s) of endogenous rstn in N-1 neurons by: a) inducing its overexpression, or b) using an RNAi-based approach to specifically attenuate basal rstn expression. Our initial data provide further support for an autocrine/paracrine role for brain-derived resistin in the regulation of central energy metabolism.

Introduction

Resistin induces insulin resistance and alters peripheral glucose metabolism in mice (Qatanani et al., 2009; Steppan et al., 2001), though these actions remain controversial, especially in humans (Hivert et al., 2008; McTernan et al., 2006). Most adipokines, including resistin, are thought to exert at least part of their metabolic influence via a hypothalamic-dependent mechanism (Ahima and Lazar, 2008; Ahima et al., 2006). For example, intracerebroventricular (icv) delivery of resistin acutely reduced feeding in adult rats (Cifani et al., 2009; Park et al., 2008; Tovar et al., 2005; Vazquez et al., 2008) (Table 1), but this route of delivery also induced hepatic insulin resistance and increased peripheral glucose production (Muse et al., 2007; Park et al., 2008; Singhal et al., 2007). Nevertheless a substantial degree of uncertainty still remains with regard to the exact central mechanism(s) through which resistin is able to mediate these metabolic actions. Several hypothalamic neuropeptides are known to regulate the control of energy metabolism and body weight. Moreover many of these central targets, including anorectic (POMC), and orexigenic (NPY and AgRP) genes are clearly under the regulatory influence of several adipokines (Ahima and Lazar, 2008; Ahima et al., 2006). Additionally adipokines appear to modulate various central enzymatic targets, in particular 5′AMP-activated protein kinase (AMPK) and one of its many downstream targets, acetylCoA carboxylase (ACC), which can also lead to changes in appetite and peripheral gluconeogenesis (Gao et al., 2007; McCrimmon et al., 2006; Namkoong et al., 2005). However these effects are regionally dependent within the hypothalamus such that activation of different neuronal populations is capable of inducing divergent responses (Claret et al., 2007; Gao et al., 2007; Kahn et al., 2005; Minokoshi et al., 2008). Resistin (rstn) is ubiquitously expressed and its mRNA was detected in the gonads (Nogueiras et al., 2004), liver (Sheng et al., 2008) adrenals, stomach, small intestine and skeletal muscle (Nogueiras et al., 2003; Patel et al., 2003). However the detection of rstn mRNA within the central nervous system (CNS) is of notable interest (Morash et al., 2002; Tovar et al., 2005; Wilkinson et al., 2005, 2007). Its expression is enriched within the arcuate (ARC) nucleus of the mouse hypothalamus (Morash et al., 2002;

2.

Results

2.1. The effects of recombinant resistin on gene expression in N-1 neurons When N-1 neurons were acutely treated with recombinant murine resistin (100 ng/mL or 1000 ng/mL; 45 min) this

Table 1 – The reported effects of ICV resistin on food intake, hypothalamic gene expression and enzyme activity. Authors Vazquez et al. (2008) Rat

Treatment Acute icv resistin (10 μg)

Fed Fasted Chronic resistin (10ug/day Fed for 6 days) Fasted Park et al. (2008) Chronic resistin (1 μg/day Diabetic Rat for 4 weeks) Singhal et al. Acute icv resistin (6 μg) (2007) Mouse Tovar et al. Acute icv resistin (10 μg) Fed (2005) Rat Fasted N-1 neurons Recombinant resistin in vitro 100 + 1000 ng/mL Resistin overexpression Resistin knockdown

Food intake

cfos pSTAT3 CART POMC NPY AgRP ADPN pAMPK pACC

↓ ↓ ↓ O O

– – – – –

– – – – ↓

O ↑ – – –

O O – – –

O ↓ – – –

O ↓ – – –

–

↑

–

–

O

↑

O

↓ ↓

O ↑ –

– – –

– – –

– – O

– – O

– – O

– –

– –

– –

↓ ↑

↓ ↓

O ↑

Legend: – = not measured O = no effect, ↑ = increase, ↓ = decrease.

– – – – –

↑ O – – O

↑ O – – –

–

–

– – O

– – ↑

– – O

↓ ↑

↑ ↓

↑ ↓

–

54

BR A IN RE S EA RCH 1 2 94 ( 20 0 9 ) 5 2 –60

failed to significantly modify the expression of AgRP, POMC, NPY or ADPN, as assessed using realtime RT-PCR (Fig. 1).

2.2. Resistin overexpression modifies target gene expression in N-1 cells In contrast to the effects of exogenous resistin, overexpression of rstn mRNA in N-1 neurons using previously optimized conditions (Brown et al., 2007) significantly reduced the expres-

Fig. 2 – Resistin overexpression modifies the expression of other genes in N-1 neurons. (A) The plasmid-mediated (6 h) overexpression of resistin failed to modify AgRP mRNA levels in N-1 neurons relative to cells transfected with an empty vector alone. However resistin overexpression did induce significant reductions in the expression of (B) POMC, (C) NPY and (D) ADPN in the N-1 hypothalamic cell line as assessed using realtime RT-PCR. Data were obtained from duplicate experiments and are expressed as a percentage of the control ± SEM (n = 5–6). (*p < 0.05, **p < 0.01).

Fig. 1 – The effects of recombinant resistin on gene expression in N-1 neurons. Recombinant resistin (45 min; 100 or 1000 ng/mL) induced modest, but non significant, reductions in (A) AgRP, (B) POMC and (D) ADPN expression as analysed using realtime RT-PCR. In contrast acute resistin treatment had no clear effect on NPY expression in N-1 neurons (C). Data were obtained from triplicate experiments and are expressed as a percentage of the control ± SEM (n = 6–8).

sion of POMC (− 35%, p < 0.01), NPY (−36%, p < 0.05) and ADPN (− 23%; p < 0.05; 6 h). However there were no detectable changes in the expression of AgRP, relative to control cells transfected with an empty vector (Fig. 2).

2.3. Gene expression variations in N-1 neurons following RNAi-mediated silencing of rstn Resistin gene expression was specifically silenced in N-1 neurons using a previously optimized resistin-specific siRNA

BR A IN RE S E A RCH 1 2 94 ( 20 0 9 ) 5 2 –6 0

55

(Brown et al., 2007), which not only induced the previously reported reductions in rstn mRNA (−60%, p < 0.0001, data not shown) but was also accompanied by significant increases in the expression of AgRP (+ 36%, p < 0.05), POMC (+ 65%, p < 0.0001), and ADPN (+ 87%, p < 0.0001), relative to cells transfected with the non-specific control molecule (Fig. 3). Unexpectedly, NPY expression was also significantly attenu-

Fig. 4 – AICAR-induced increases in pAMPK in N-1 neurons. Preliminary investigations were performed to assess the responsiveness of AMPK in N-1 neurons. Following a 1 h treatment with 1 mM AICAR, the level of pAMPKα/total AMPKα was significantly increased (50%, p < 0.05), although no further increases were detected following a 6 h exposure to AICAR (50%, p < 0.05). Representative Western blots are shown for pAMPKα and total AMPKα, and data were obtained from triplicate experiments and are expressed as a percentage of the control ± SEM (n = 3–4). (*p < 0.05).

ated (− 22%, p < 0.01) in N-1 cells transfected with the rstnspecific siRNA.

2.4.

Fig. 3 – Rstn knockdown influences gene expression in N-1 neurons. Rstn expression was significantly reduced in cells transfected with STHR4 (−60%, p < 0.0001, data not shown) relative to those transfected with the non-specific control molecule as assessed using realtime RT-PCR. Significant increases in AgRP (A), POMC (B) and ADPN (D) were also detected 24 h following the initiation of rstn knockdown in N-1 neurons. (C) Conversely a significant reduction in NPY mRNA was also detected following the RNAi-mediated silencing of rstn in N-1 neurons. Data were obtained from triplicate experiments and are expressed as a percentage of the control ± SEM (n = 6–10). (*p < 0.05, **p < 0.01, ****p < 0.0001).

The regulation of pAMPKα and pACC in N-1 neurons

In preliminary investigations to determine whether AMPK responded appropriately in N-1 neurons cells were treated with AICAR (1 mM; 1 h and 6 h), a potent AMPK agonist (Dasgupta and Milbrandt, 2007), which led to a significant increase in the phosphorylation of AMPK that was sustained for the duration of the experiment (+50%; p < 0.05) (Fig. 4). Subsequently, treatment of N-1 neurons with recombinant resistin (45 min; 100 ng/mL) induced an increase (31%, p < 0.05) in the ratio of pAMPKα/total AMPKα, and no additional effect was observed with the higher dose (1000 ng/mL; Fig. 5A). In contrast neither dose of recombinant resistin had any significant effect on pACC levels (Fig. 5D; −16% and −18% respectively; NS). In contrast, N-1 neurons overexpressing resistin showed a significant increase in the ratio of pAMPKα/ total AMPKα (Fig. 5B; +47%, p < 0.001) and in the ratio of pACC/ total ACC, relative to cells transfected with an empty vector alone (Fig. 5E; +34%, p < 0.001). Conversely the specific RNAimediated silencing of resistin was associated with a greater than 40% reduction in the ratio of pAMPKα/total AMPKα (Fig. 5C; p < 0.001), and a parallel decrease in pACC levels was also measured in N-1 neurons (Fig. 5F; −35%, p < 0.001).

3.

Discussion

There is increased interest in defining the mechanism(s) through which resistin might exert its regulatory effects on hypothalamic function. Although icv delivery of resistin in rodents led to a reduction in appetite, in addition to stimu-

56

BR A IN RE S EA RCH 1 2 94 ( 20 0 9 ) 5 2 –60

Fig. 5 – Resistin modulates AMPK and ACC activity in N-1 hypothalamic neurons. (A) Treating serum-starved N-1 neurons for 45 min with recombinant resistin induced a significant increase in pAMPKα (+31%). (D) Resistin treatment tended to reduce the level of pACC, relative to the untreated control cells, but this change was statistically nonsignificant. (B and E) Overexpression of rstn in N-1 cells for 24 h resulted in a 47% increase in AMPKα phosphorylation, which was also associated with a 34% increase in the ratio pACC/total ACC. (C and F) Following the siRNA-mediated silencing of rstn a 40% reduction in the relative ratio of pAMPKα/total AMPKα was detected, in addition to a 35% reduction in the level of pACC, relative to cells treated with the nonspecific siRNA. Representative Western blots are shown for pAMPKα and total AMPKα (A–C) and for pACC and total ACC (D–F). Data were obtained from duplicate or triplicate experiments and are expressed as a percentage of the control ± SEM (n = 6–8). (*p < 0.05, ***p < 0.001).

lating peripheral glucose production, the molecular mechanisms involved still remain uncertain (Cifani et al., 2009; Muse et al., 2007; Park et al., 2008; Singhal et al., 2007; Tovar et al., 2005). For example a single icv injection of resistin was reported to induce hypothalamic expression of cfos and NPY, but failed to modify AgRP or POMC mRNA (Singhal et al., 2007; Tovar et al., 2005). Others reported that chronic resistin infusion had little effect on hypothalamic STAT3 and AMPK phosphorylation, though it did attenuate the stimulatory effect of leptin on these signals (see Table 1) (Park et al., 2008). A further investigation confirmed the anorectic effects of icv resistin, but added to the complexity of its effects within the hypothalamus, since pAMPK and pACC were acutely increased, but there was no effect on AgRP and NPY expression unless the rats were fasted first (Vazquez et al., 2008). Some of the paradoxical effects of icv resistin might be explained by species-specific variations (mouse vs. rat), in addition to differences in metabolic status (i.e. fed vs. fasted), and require further investigation. Results from the in vitro treatment of N-1 neurons with recombinant murine resistin (100 and 1000 ng/mL; 45 min) were also inconsistent with some of the previous studies; i.e., recombinant resistin failed to modify the expression of AgRP, POMC, NPY or ADPN (Table 1). Nevertheless when rstn was

overexpressed in N-1 neurons we observed a significant attenuation in the expression of POMC, NPY and ADPN, whereas AgRP was unaffected. Although these data provide additional support for an inhibitory effect of resistin on hypothalamic function, it remains uncertain why recombinant and endogenous resistin had these differing effects, though this likely reflects differences in intracellular resistin concentrations. However of particular interest is how the resistin-specific siRNA not only attenuated the endogenous expression of resistin in N-1 neurons (Brown et al., 2007), but this led to significant increases in AgRP, POMC, and ADPN mRNA. NPY mRNA was unexpectedly attenuated in the rstn silencing studies which might be attributed to a biphasic or temporally-dependent effect of resistin on NPY gene expression in N-1 neurons since this trend was consistently observed in both the resistin overexpression and silencing experiments, whereas recombinant resistin had no significant effect. In contrast to the absence of an effect of exogenous resistin on gene expression, we observed an acute stimulatory effect of recombinant murine resistin on AMPK activation in N-1 neurons, as assessed by its phosphorylation status. In addition we showed that the RNAi-mediated reduction of endogenous resistin had the expected opposite effect. Unexpectedly resistin treatment tended to lower the ratio of pACC/

BR A IN RE S E A RCH 1 2 94 ( 20 0 9 ) 5 2 –6 0

total ACC, a downstream target of AMPK which is inactivated by phosphorylation, whereas resistin overexpression had the opposing effect. This paradoxical result may reflect the difference between endogenous and exogenous sources of resistin. However, as expected, the RNAi-mediated silencing of rstn also lowered pACC levels in N-1 neurons. Overall these studies in the N-1 hypothalamic cell line provide further support for the capacity of endogenous resistin to modulate the expression of several key neuropeptide genes (AgRP, POMC and NPY), and other adipokines (ADPN), in addition to regulating the activity of critical metabolic enzymes (AMPK and ACC), all of which are targets implicated in the control of glucose metabolism, appetite and bodyweight regulation (Ahima and Lazar, 2008; Ahima et al., 2006; Gao et al., 2007; Kahn et al., 2005; McCrimmon et al., 2006; Namkoong et al., 2005). More importantly our data demonstrate the potential for brain-derived resistin to regulate local gene expression and enzyme activity via possible autocrine/paracrine signaling pathways, and may have implications for the regulation of energy metabolism in vivo, which will now need to be clarified in the intact hypothalamus in future investigations. Overexpression of resistin in N-1 neurons induced reductions in AgRP and NPY gene expression. This result suggests that the in vivo anorectic effect following the icv administration of resistin (Cifani et al., 2009; Tovar et al., 2005; Vazquez et al., 2008) might also be partially mediated by these changes in orexigenic gene expression. However, overexpression of resistin in N-1 neurons induced a reduction in POMC mRNA. Such an effect in vivo would be expected to have the opposing effect on food intake (i.e. increase appetite). One interpretation of this anomaly is that the consequences of reduced POMC expression are relatively less significant than that of AgRP/NPY such that there is still an overall net anorexigenic effect. The decrease in POMC mRNA in N-1 neurons might also represent a compensational change that aims to limit the duration of the resistin effect. Such a mechanism is supported by the fact that acute treatment with resistin (10 μg) in vivo reduced food intake (Cifani et al., 2009; Tovar et al., 2005; Vazquez et al., 2008), whereas the chronic 4-week icv infusion of resistin (1 μg/day) had no significant effect on food intake or bodyweight (Park et al., 2008). Note, however, that the daily injection of a higher resistin dose (10 μg/day) continued to suppress food intake in rats even after 6 days of treatment (Vazquez et al., 2008). However, as expected, RNAi-mediated silencing of rstn in N-1 neurons led to an increase in POMC mRNA. We conclude that endogenous resistin appears to modulate the expression of orexigenic and anorexigenic neuropeptides in N-1 neurons in a temporally-dependent manner. Insofar as these cells may represent a useful model system, these data raise further questions about the existence of a possible autocrine/paracrine system within the normal intact hypothalamus. Significant advancements have been made towards understanding the roles of AMPK and ACC in both peripheral and central energy metabolism, especially in relation to their regulation by adipokines (Gao et al., 2007; Hardie, 2008; Kahn et al., 2005). Initially AMPK was found to be activated by phosphorylation of threonine 172 in the alpha subunit which occurred in response to subtle increases in the ratio of AMP/ ATP, but was also found to phosphorylate and inactivate downstream targets such as acetyl-CoA carboxylase (ACC)

57

(Hardie, 2008; Kahn et al., 2005; Lane et al., 2008; Minokoshi et al., 2008). Subsequently the regulation of AMPK was found to be tissue-, and even cell type-specific (Claret et al., 2007; Gao et al., 2007; Kahn et al., 2005; Minokoshi et al., 2008), and its activation in various discrete hypothalamic regions had differential effects on peripheral energy metabolism. For example activation of AMPK within the ARC induced hyperphagia whereas in the ventral medial hypothalamus (VMH) this increased peripheral glucose production (McCrimmon et al., 2006; Minokoshi et al., 2008; Namkoong et al., 2005). We show here that resistin induced modest increases in pAMPK in N-1 neurons following an acute treatment, or after cells were made to overexpress rstn. Conversely the opposite effect was observed following rstn knockdown. However pAMPK levels were unchanged in N-1 neurons following a 24 h resistin treatment (data not shown). Thus our data are in agreement with recent reports that: (a) acute icv resistin modestly increased pAMPK levels in the intact hypothalamus after 1.5 h (Vazquez et al., 2008), whereas (b) no effects on pAMPK were detected following the prolonged central infusion of resistin (Park et al., 2008) (Table 1). Although the induction of hypothalamic pAMPK activation might generally be expected to increase appetite, evidently this was not the case following icv resistin in vivo (Cifani et al., 2009; Tovar et al., 2005; Vazquez et al., 2008). However the hypothalamic activation of AMPK was also shown to increase peripheral glucose production, as did icv resistin (Muse et al., 2007). This discrepancy may reflect the regionally dependent variations in hypothalamic AMPK activation and function, and that resistin may influence appetite and glucose production through different hypothalamic structures (Cifani et al., 2009; Gao et al., 2007), and needs further investigation. Though the nonsignificant reduction in pACC detected in N-1 cells following resistin treatment was unexpected, this might be attributed to other regulatory mechanisms, in addition to AMPK, which are also known to modulate its activity (Brownsey et al., 2006). Nonetheless, resistin overexpression led to the expected increase in pACC, whereas the opposite effect was observed following rstn knockdown which again further suggests the potential capacity for brain-derived resistin to modulate hypothalamic energy metabolism. Emerging evidence suggests that adiponectin (ADPN) is also implicated in hypothalamic function. Both isoforms of the ADPN receptor, Adipo R1 and R2, have been detected in the mouse, rat and human hypothalamus (Ahima and Lazar, 2008; Coope et al., 2008; Neumeier et al., 2007; Qi et al., 2004), and we previously detected ADPN mRNA within the rodent hypothalamus using realtime RT-PCR (Wilkinson et al., 2007), now confirmed by our data from N-1 neurons. Multiple reports have also suggested that icv ADPN influences appetite, although its proposed effects on food intake and bodyweight were conflicting (Coope et al., 2008; Kubota et al., 2007; Qi et al., 2004). Similarly ADPN inhibited gonadotropin releasing hormone (GnRH) secretion from the GT1-7 neuronal cell line further suggesting its potential involvement in the maintenance of normal hypothalamic function (Wen et al., 2008). Rstn overexpression also significantly reduced ADPN gene expression in N-1 neurons (−22%, p < 0.05), although recombinant resistin only tended to inhibit ADPN expression (−18%, p = N/S). Conversely rstn knockdown led to a significant increase in

58

BR A IN RE S EA RCH 1 2 94 ( 20 0 9 ) 5 2 –60

ADPN gene expression (+90%, p < 0.0001). These findings mirror the previous antagonistic relationship seen between fiaf and rstn, and further confirms our previous hypothesis that adipokine cross talk occurs in N-1 neurons (Brown et al., 2007). However until the role(s) of ADPN in the hypothalamus is entirely resolved, the significance of this crosstalk remains uncertain. Accumulating evidence supporting a ubiquitous pattern of adipokine expression suggests that their regulation and function might be tissue-dependent. This could also further suggest the potential existence of autocrine/paracrine adipokine systems within the body. Although their significance and purpose remains less certain, the local expression of resistin may have an important influence on various tissue functions and metabolism. These preliminary data derived from N-1 hypothalamic neurons further support our hypothesis that such an autocrine/paracrine system may exist in the brain which could have implications for the acute fine-tuning regulation required for the precise control of central energy metabolism. Endogenous resistin was not only shown to acutely inhibit the expression of several key hypothalamic neuropeptide genes in N-1 hypothalamic neurons, but it also modified the activity of novel enzymatic targets implicated in the regulation of central energy metabolism and warrants further investigation in vivo. However future investigations should also consider the possible importance of dynamic and temporal changes in local resistin levels in vivo, as opposed to absolute concentrations alone, which cannot be accomplished using standard knockout mice models. As such, hypothalamic adipokines may exert an acute regulatory influence on central energy metabolism (i.e. hour by hour), as opposed to peripheral adipokines that could be providing more of a long term report on the status of systemic metabolic energy stores. However several important questions remain including: a) is this endogenous adipokine system functional in the intact hypothalamus?; (b) is the N-1 neuronal cell line a good model for the study of hypothalamic function?; and (c) are these findings relevant to human physiology?

4.

Experimental procedures

4.1.

N-1 neurons

Immortalized mouse N-1 hypothalamic neurons (Belsham et al., 2004) were maintained in DMEM culture medium containing 10% fetal bovine serum (FBS) at 37 °C in 5% CO2/ 95% air. Cells were plated at 100,000 cells/well in Nunc 6-well plates and cells were cultured in serum free DMEM overnight before treatment with recombinant murine resistin (45 min or 24 h; 100 and 1000 ng/mL, PeproTech; Rocky Hill, NJ), and control cells received vehicle alone. In subsequent studies cells were serum-starved overnight prior to being transfected with either 3.2 μg/well with either a resistin expressing plasmid (6 h or 24 h) or an empty plasmid, and in knockdown experiments cells were treated with 100 nM of either the rstnspecific STEALTH siRNA (STH R4; 24 h) or the non-specific siRNA control using previously optimized conditions for Lipofectamine 2000 (2.25 μL:1 μg STH R4; Invitrogen; Burlington, ON) (Brown et al., 2007).

4.2.

Realtime RT-PCR

Following experimentation RNA was isolated from N-1 neurons using the GenElute Mammalian Total RNA isolation Kit (Sigma-Aldrich, Oakville, ON), and treated with DNAse, according to the manufacturer's protocol. Total RNA (2 μg) was reverse transcribed (RT) using the SuperScript™ III (Invitrogen) as described by the manufacturer. In brief, RNA was diluted to 16 μL and then heat denatured for 5 min at 70 °C. Samples were returned to ice prior to the addition of 20 μL of the 2× reverse transcription master mix and 4 μL of the SuperScript™ III RT enzyme master mix (Invitrogen). The RT reaction consisted of a 10 min incubation at 25 °C, 45 min at 42 °C, followed by 5 min at 85 °C to terminate reactions, and the resulting complementary DNA (cDNA; 40 μL) was stored at − 20 °C. Gene expression for NPY, AgRP, POMC, ADPN, rstn and cyclophilin were analysed in triplicate using the Platinum SYBR Green qPCR SuperMix for iCycler (Invitrogen). Individual PCR reactions consisted of a 2× SYBR Green qPCR SuperMix, 7 pmol of the sense and antisense primers (Table 2), 2 μL of cDNA, to a final volume of 25 μL in sterile water. Reactions were initiated by incubating at 95 °C for 5 min, followed by 60 amplification cycles of 95 °C for 20 s annealing at 60 °C for 10 s, extension at 72 °C for 45 s, and data collection at 80 °C for 10 s using a Bio-Rad iCycler. A standard curve, that was prepared using a serial dilution of a reference sample, was included in each realtime run to correct for possible variations in product amplification. Relative copy numbers were obtained from standard curve values, and were normalized to the values obtained for our house keeping gene, cyclophilin. Data are expressed as a percentage of the control ± SEM. Note that no significant variations in cyclophilin expression were observed between groups.

4.3. Western blot analysis of phosphorylated AMPKα and ACC In preliminary experiments we used a well-described activator of AMPK, AICAR (Dasgupta and Milbrandt, 2007); 5-aminoimidazole-4-carboxamide-1-β- D -ribofuranoside; Sigma-Aldrich), to confirm that N-1 cells responded appropriately. Following the various resistin treatment(s) (i.e. recombinant resistin, resistin overexpression and resistin knockdown), cells were washed twice in Dulbecco's phosphate buffered saline (1× D-PBS; Invitrogen) prior to harvesting and

Table 2 – Sequences of realtime PCR primers. Gene Cyclophilin Resistin NPY AgRP POMC Adiponectin

Primer sequences Sense: 5′ TTCCTTGTCCCTGAACTGCT 3′ Antisense: 5′ TGCTGTCCAGTCTATCCTTG 3′ Sense: 5′ TTCCTTGTCCCTGAACTGCT 3′ Antisense: 5′ TGCTGTCCAGTCTATCCTTG 3′ Sense: 5′ CTC CGC TCT GCG ACA CTA CA 3′ Antisense: 5′ AATCAG TGT CTC AGG GCT GGA 3′ Sense: 5′ TGCTACTGCCGCTTCTTCAA 3′ Antisense: 5′ CTTTGCCCAAACAACATCCA 3′ Sense: 5′ AGGCCTGACACGTGGAAGAT 3′ Antisense: 5′ AGGCACCAGCTCCACACAT 3′ Sense: 5′ ACTCCTGGAGAGAAGGGAGAGAAA 3′ Antisense: 5′ GCTCCTGTCATTCCAACATCTCCT 3′

BR A IN RE S E A RCH 1 2 94 ( 20 0 9 ) 5 2 –6 0

lysing in radioimmunoprecipitation (RIPA) buffer containing 10 μg/mL antipain, 10 μg/mL leupeptin, 10 μg/mL pepstatin, 1 mM phenylmethylsulphonylfluoride, 5 mM EDTA and 5 mM EGTA. Cellular lysate concentrations were determined using the DC protein assay (Bio-Rad; Mississauga, ON). Samples (40 μg) were separated on a 10% SDS-PAGE gel and transferred overnight onto HYBOND-C nitrocellulose membrane (Amersham Bioscience; Baie d'Urfé, QC). Membranes were washed with Tween-20 tris buffered saline (TTBS) prior to blocking in TTBS containing 10% milk powder. Blots were probed overnight with an anti-phosphorylated (Threonine 172)-AMPKα antibody, total AMPKα, anti-phosphorylated (Serine 79) ACC or total ACC (all 1:1000; Cell Signaling; Danvers, MA), followed by a donkey anti-rabbit horseradish peroxidase (HRP) conjugated antibody (1:5000; Amersham Bioscience). Bands were detected by chemiluminescence using SuperSignal™ West Femto Chemiluminescent substrate (Pierce; Rockford, IL). Band intensity was quantified using NIH Image (v1.63), and relative ratios of pAMPKα/total AMPKα, or pACC/total ACC were obtained and data are expressed as a percentage of controls ± SEM.

4.4.

Statistics

The normalized data were analysed either by Student's t-test or analysis of variance (ANOVA) followed by the Newman– Keuls multiple comparisons test. Significance was set at p < 0.05.

Acknowledgments These studies were funded by the NSHRF, the IWK Health Centre, the Atlee Endowment and UIMRF/Capital Health. RB was the recipient of an IWK Summer Studentship. We are indebted to Diane Wilkinson for her technical assistance and to Dr. Denise Belsham (University of Toronto) who generously provided the N-1 hypothalamic cell line.

REFERENCES

Ahima, R.S., Lazar, M.A., 2008. Adipokines and the peripheral and neural control of energy balance. Mol. Endocrinol. 22, 1023–1031. Ahima, R.S., Qi, Y., Singhal, N.S., 2006. Adipokines that link obesity and diabetes to the hypothalamus. Prog. Brain Res. 153, 155–174. Belsham, D.D., Cai, F., Cui, H., Smukler, S.R., Salapatek, A.M., Shkreta, L., 2004. Generation of a phenotypic array of hypothalamic neuronal cell models to study complex neuroendocrine disorders. Endocrinology 145, 393–400. Brown, R., Imran, S.A., Belsham, D.D., Ur, E., Wilkinson, M., 2007. Adipokine gene expression in a novel hypothalamic neuronal cell line: resistin-dependent regulation of fasting-induced adipose factor and SOCS-3. Neuroendocrinology 85, 232–241. Brown, R., Imran, S.A., Ur, E., Wilkinson, M., 2008. Valproic acid and CEBPalpha-mediated regulation of adipokine gene expression in hypothalamic neurons and 3T3-L1 adipocytes. Neuroendocrinology 88, 25–34. Brownsey, R.W., Boone, A.N., Elliott, J.E., Kulpa, J.E., Lee, W.M., 2006. Regulation of acetyl-CoA carboxylase. Biochem. Soc. Trans. 34, 223–227.

59

Cifani, C., Durocher, Y., Pathak, A., Penicaud, L., Smih, F., Massi, M., Rouet, P., Polidori, C., 2009. Possible common central pathway for resistin and insulin in regulating food intake. Acta Physiol. (Oxf) 196, 395–400. Claret, M., Smith, M.A., Batterham, R.L., Selman, C., Choudhury, A.I., Fryer, L.G., Clements, M., Al-Qassab, H., Heffron, H., Xu, A.W., Speakman, J.R., Barsh, G.S., Viollet, B., Vaulont, S., Ashford, M.L., Carling, D., Withers, D.J., 2007. AMPK is essential for energy homeostasis regulation and glucose sensing by POMC and AgRP neurons. J. Clin. Invest. 117, 2325–2336. Coope, A., Milanski, M., Araujo, E.P., Tambascia, M., Saad, M.J., Geloneze, B., Velloso, L.A., 2008. AdipoR1 mediates the anorexigenic and insulin/leptin-like actions of adiponectin in the hypothalamus. FEBS Lett. 582, 1471–1476. Dasgupta, B., Milbrandt, J., 2007. Resveratrol stimulates AMP kinase activity in neurons. Proc. Natl. Acad. Sci. U. S. A. 104, 7217–7222. Gao, S., Kinzig, K.P., Aja, S., Scott, K.A., Keung, W., Kelly, S., Strynadka, K., Chohnan, S., Smith, W.W., Tamashiro, K.L., Ladenheim, E.E., Ronnett, G.V., Tu, Y., Birnbaum, M.J., Lopaschuk, G.D., Moran, T.H., 2007. Leptin activates hypothalamic acetyl-CoA carboxylase to inhibit food intake. Proc. Natl. Acad. Sci. U. S. A. 104, 17358–17363. Hardie, D.G., 2008. AMPK: a key regulator of energy balance in the single cell and the whole organism. Int. J. Obes. (Lond) 32 (Suppl. 4), S7–S12. Hivert, M.F., Sullivan, L.M., Fox, C.S., Nathan, D.M., D'Agostino Sr., R.B., Wilson, P.W., Meigs, J.B., 2008. Associations of adiponectin, resistin, and tumor necrosis factor-alpha with insulin resistance. J. Clin. Endocrinol. Metab. 93, 3165–3172. Kahn, B.B., Alquier, T., Carling, D., Hardie, D.G., 2005. AMPactivated protein kinase: ancient energy gauge provides clues to modern understanding of metabolism. Cell Metab. 1, 15–25. Kubota, N., Yano, W., Kubota, T., Yamauchi, T., Itoh, S., Kumagai, H., Kozono, H., Takamoto, I., Okamoto, S., Shiuchi, T., Suzuki, R., Satoh, H., Tsuchida, A., Moroi, M., Sugi, K., Noda, T., Ebinuma, H., Ueta, Y., Kondo, T., Araki, E., Ezaki, O., Nagai, R., Tobe, K., Terauchi, Y., Ueki, K., Minokoshi, Y., Kadowaki, T., 2007. Adiponectin stimulates AMP-activated protein kinase in the hypothalamus and increases food intake. Cell Metab. 6, 55–68. Lane, M.D., Wolfgang, M., Cha, S.H., Dai, Y., 2008. Regulation of food intake and energy expenditure by hypothalamic malonyl-CoA. Int. J. Obes. (Lond). 32 (Suppl. 4), S49–S54. McCrimmon, R.J., Fan, X., Cheng, H., McNay, E., Chan, O., Shaw, M., Ding, Y., Zhu, W., Sherwin, R.S., 2006. Activation of AMP-activated protein kinase within the ventromedial hypothalamus amplifies counterregulatory hormone responses in rats with defective counterregulation. Diabetes 55, 1755–1760. McTernan, P.G., Kusminski, C.M., Kumar, S., 2006. Resistin. Curr. Opin. Lipidol. 17, 170–175. Minokoshi, Y., Shiuchi, T., Lee, S., Suzuki, A., Okamoto, S., 2008. Role of hypothalamic AMP-kinase in food intake regulation. Nutrition 24, 786–790. Morash, B.A., Wilkinson, D., Ur, E., Wilkinson, M., 2002. Resistin expression and regulation in mouse pituitary. FEBS Lett. 526, 26–30. Muse, E.D., Lam, T.K., Scherer, P.E., Rossetti, L., 2007. Hypothalamic resistin induces hepatic insulin resistance. J. Clin. Invest. 117, 1670–1678. Namkoong, C., Kim, M.S., Jang, P.G., Han, S.M., Park, H.S., Koh, E.H., Lee, W.J., Kim, J.Y., Park, I.S., Park, J.Y., Lee, K.U., 2005. Enhanced hypothalamic AMP-activated protein kinase activity contributes to hyperphagia in diabetic rats. Diabetes 54, 63–68. Neumeier, M., Weigert, J., Buettner, R., Wanninger, J., Schaffler, A., Muller, A.M., Killian, S., Sauerbruch, S., Schlachetzki, F., Steinbrecher, A., Aslanidis, C., Scholmerich, J., Buechler, C.,

60

BR A IN RE S EA RCH 1 2 94 ( 20 0 9 ) 5 2 –60

2007. Detection of adiponectin in cerebrospinal fluid in humans. Am. J. Physiol. Endocrinol. Metab. 293, E965–E969. Nogueiras, R., Gallego, R., Gualillo, O., Caminos, J.E., Garcia-Caballero, T., Casanueva, F.F., Dieguez, C., 2003. Resistin is expressed in different rat tissues and is regulated in a tissue- and gender-specific manner. FEBS Lett. 548, 21–27. Nogueiras, R., Barreiro, M.L., Caminos, J.E., Gaytan, F., Suominen, J.S., Navarro, V.M., Casanueva, F.F., Aguilar, E., Toppari, J., Dieguez, C., Tena-Sempere, M., 2004. Novel expression of resistin in rat testis: functional role and regulation by nutritional status and hormonal factors. J. Cell Sci. 117, 3247–3257. Park, S., Hong, S.M., Sung, S.R., Jung, H.K., 2008. Long-term effects of central leptin and resistin on body weight, insulin resistance, and {beta}-cell function and mass by the modulation of hypothalamic leptin and insulin signaling. Endocrinology 149, 445–454. Patel, L., Buckels, A.C., Kinghorn, I.J., Murdock, P.R., Holbrook, J.D., Plumpton, C., Macphee, C.H., Smith, S.A., 2003. Resistin is expressed in human macrophages and directly regulated by PPAR gamma activators. Biochem. Biophys. Res. Commun. 300, 472–476. Qatanani, M., Szwergold, N.R., Greaves, D.R., Ahima, R.S., Lazar, M.A., 2009. Macrophage-derived human resistin exacerbates adipose tissue inflammation and insulin resistance in mice. J. Clin. Invest. Qi, Y., Takahashi, N., Hileman, S.M., Patel, H.R., Berg, A.H., Pajvani, U.B., Scherer, P.E., Ahima, R.S., 2004. Adiponectin acts in the brain to decrease body weight. Nat. Med. 10, 524–529.

Sheng, C.H., Di, J., Jin, Y., Zhang, Y.C., Wu, M., Sun, Y., Zhang, G.Z., 2008. Resistin is expressed in human hepatocytes and induces insulin resistance. Endocrine 33, 135–143. Singhal, N.S., Lazar, M.A., Ahima, R.S., 2007. Central resistin induces hepatic insulin resistance via neuropeptide Y. J. Neurosci. 27, 12924–12932. Steppan, C.M., Bailey, S.T., Bhat, S., Brown, E.J., Banerjee, R.R., Wright, C.M., Patel, H.R., Ahima, R.S., Lazar, M.A., 2001. The hormone resistin links obesity to diabetes. Nature 409, 307–312. Tovar, S., Nogueiras, R., Tung, L.Y., Castaneda, T.R., Vazquez, M.J., Morris, A., Williams, L.M., Dickson, S.L., Dieguez, C., 2005. Central administration of resistin promotes short-term satiety in rats. Eur. J. Endocrinol. 153, R1–R5. Vazquez, M.J., Gonzalez, C.R., Varela, L., Lage, R., Tovar, S., Sangiao-Alvarellos, S., Williams, L.M., Vidal-Puig, A., Nogueiras, R., Lopez, M., Dieguez, C., 2008. Central resistin regulates hypothalamic and peripheral lipid metabolism in a nutritional dependent fashion. Endocrinology 149, 4534–4543. Wen, J.P., Lv, W.S., Yang, J., Nie, A.F., Cheng, X.B., Yang, Y., Ge, Y., Li, X.Y., Ning, G., 2008. Globular adiponectin inhibits GnRH secretion from GT1-7 hypothalamic GnRH neurons by induction of hyperpolarization of membrane potential. Biochem. Biophys. Res. Commun. 371, 756–761. Wilkinson, M., Wilkinson, D., Wiesner, G., Morash, B., Ur, E., 2005. Hypothalamic resistin immunoreactivity is reduced by obesity in the mouse: co-localization with alpha-melanostimulating hormone. Neuroendocrinology 81, 19–30. Wilkinson, M., Brown, R., Imran, S.A., Ur, E., 2007. Adipokine gene expression in brain and pituitary gland. Neuroendocrinology 86, 191–209.